From 09120d4e88b13967d44d46280fb74d3ac4ac2f73 Mon Sep 17 00:00:00 2001
From: Michel Thierry <michel.thierry@intel.com>
Date: Wed, 29 Jul 2015 17:23:45 +0100
Subject: drm/i915: Remove unnecessary gen8_clamp_pd

gen8_clamp_pd clamps to the next page directory boundary, but the macro
gen8_for_each_pde already has a check to stop at the page directory
boundary.

Furthermore, i915_pte_count also restricts to the next page table
boundary.

v2: Rebase after Mika's ppgtt cleanup / scratch merge patch series.

Suggested-by: Akash Goel <akash.goel@intel.com>
Signed-off-by: Michel Thierry <michel.thierry@intel.com>
Reviewed-by: "Akash Goel" <akash.goel@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 96054a5..566466f 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -955,7 +955,7 @@ static int gen8_alloc_va_range(struct i915_address_space *vm,
 	gen8_for_each_pdpe(pd, &ppgtt->pdp, start, length, temp, pdpe) {
 		gen8_pde_t *const page_directory = kmap_px(pd);
 		struct i915_page_table *pt;
-		uint64_t pd_len = gen8_clamp_pd(start, length);
+		uint64_t pd_len = length;
 		uint64_t pd_start = start;
 		uint32_t pde;
 
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.h b/drivers/gpu/drm/i915/i915_gem_gtt.h
index e1cfa29..d5bf953 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.h
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.h
@@ -444,17 +444,6 @@ static inline uint32_t gen6_pde_index(uint32_t addr)
 	     temp = min(temp, length),					\
 	     start += temp, length -= temp)
 
-/* Clamp length to the next page_directory boundary */
-static inline uint64_t gen8_clamp_pd(uint64_t start, uint64_t length)
-{
-	uint64_t next_pd = ALIGN(start + 1, 1 << GEN8_PDPE_SHIFT);
-
-	if (next_pd > (start + length))
-		return length;
-
-	return next_pd - start;
-}
-
 static inline uint32_t gen8_pte_index(uint64_t address)
 {
 	return i915_pte_index(address, GEN8_PDE_SHIFT);
-- 
cgit v0.10.2


From 6ac1850220732f47bc6ae767fa41542009674ad7 Mon Sep 17 00:00:00 2001
From: Michel Thierry <michel.thierry@intel.com>
Date: Wed, 29 Jul 2015 17:23:46 +0100
Subject: drm/i915/gen8: Make pdp allocation more dynamic

This transitional patch doesn't do much for the existing code. However,
it should make upcoming patches to use the full 48b address space a bit
easier.

32-bit ppgtt uses just 4 PDPs, while 48-bit ppgtt will have up-to 512;
this patch prepares the existing functions to query the right number of pdps
at run-time. This also means that used_pdpes should also be allocated during
ppgtt_init, as the bitmap size will depend on the ppgtt address range
selected.

v2: Renamed  pdp_free to be similar to  pd/pt (unmap_and_free_pdp).
v3: To facilitate testing, 48b mode will be available on Broadwell and
GEN9+, when i915.enable_ppgtt = 3.
v4: Rebase after s/page_tables/page_table/, added extra information
about 4-level page table formats and use IS_ENABLED macro.
v5: Check CONFIG_X86_64 instead of CONFIG_64BIT.
v6: Rebase after Mika's ppgtt cleanup / scratch merge patch series, and
follow
his nomenclature in pdp functions (there is no alloc_pdp yet).
v7: Rebase after merged version of Mika's ppgtt cleanup patch series.
v8: Rebase after final merged version of Mika's ppgtt/scratch patches.
v9: Introduce PML4 (and 48-bit checks) until next patch (Akash).
v10: Also use test_bit to detect when pd/pt are already allocated (Akash)

Cc: Akash Goel <akash.goel@intel.com>
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Michel Thierry <michel.thierry@intel.com> (v2+)
Reviewed-by: Akash Goel <akash.goel@intel.com>
[danvet: Amend commit message as suggested by Michel.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 566466f..6724e71 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -522,6 +522,43 @@ static void gen8_initialize_pd(struct i915_address_space *vm,
 	fill_px(vm->dev, pd, scratch_pde);
 }
 
+static int __pdp_init(struct drm_device *dev,
+		      struct i915_page_directory_pointer *pdp)
+{
+	size_t pdpes = I915_PDPES_PER_PDP(dev);
+
+	pdp->used_pdpes = kcalloc(BITS_TO_LONGS(pdpes),
+				  sizeof(unsigned long),
+				  GFP_KERNEL);
+	if (!pdp->used_pdpes)
+		return -ENOMEM;
+
+	pdp->page_directory = kcalloc(pdpes, sizeof(*pdp->page_directory),
+				      GFP_KERNEL);
+	if (!pdp->page_directory) {
+		kfree(pdp->used_pdpes);
+		/* the PDP might be the statically allocated top level. Keep it
+		 * as clean as possible */
+		pdp->used_pdpes = NULL;
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+static void __pdp_fini(struct i915_page_directory_pointer *pdp)
+{
+	kfree(pdp->used_pdpes);
+	kfree(pdp->page_directory);
+	pdp->page_directory = NULL;
+}
+
+static void free_pdp(struct drm_device *dev,
+		     struct i915_page_directory_pointer *pdp)
+{
+	__pdp_fini(pdp);
+}
+
 /* Broadwell Page Directory Pointer Descriptors */
 static int gen8_write_pdp(struct drm_i915_gem_request *req,
 			  unsigned entry,
@@ -720,7 +757,8 @@ static void gen8_ppgtt_cleanup(struct i915_address_space *vm)
 		container_of(vm, struct i915_hw_ppgtt, base);
 	int i;
 
-	for_each_set_bit(i, ppgtt->pdp.used_pdpes, GEN8_LEGACY_PDPES) {
+	for_each_set_bit(i, ppgtt->pdp.used_pdpes,
+			 I915_PDPES_PER_PDP(ppgtt->base.dev)) {
 		if (WARN_ON(!ppgtt->pdp.page_directory[i]))
 			continue;
 
@@ -729,6 +767,7 @@ static void gen8_ppgtt_cleanup(struct i915_address_space *vm)
 		free_pd(ppgtt->base.dev, ppgtt->pdp.page_directory[i]);
 	}
 
+	free_pdp(ppgtt->base.dev, &ppgtt->pdp);
 	gen8_free_scratch(vm);
 }
 
@@ -763,7 +802,7 @@ static int gen8_ppgtt_alloc_pagetabs(struct i915_hw_ppgtt *ppgtt,
 
 	gen8_for_each_pde(pt, pd, start, length, temp, pde) {
 		/* Don't reallocate page tables */
-		if (pt) {
+		if (test_bit(pde, pd->used_pdes)) {
 			/* Scratch is never allocated this way */
 			WARN_ON(pt == ppgtt->base.scratch_pt);
 			continue;
@@ -820,11 +859,12 @@ static int gen8_ppgtt_alloc_page_directories(struct i915_hw_ppgtt *ppgtt,
 	struct i915_page_directory *pd;
 	uint64_t temp;
 	uint32_t pdpe;
+	uint32_t pdpes = I915_PDPES_PER_PDP(dev);
 
-	WARN_ON(!bitmap_empty(new_pds, GEN8_LEGACY_PDPES));
+	WARN_ON(!bitmap_empty(new_pds, pdpes));
 
 	gen8_for_each_pdpe(pd, pdp, start, length, temp, pdpe) {
-		if (pd)
+		if (test_bit(pdpe, pdp->used_pdpes))
 			continue;
 
 		pd = alloc_pd(dev);
@@ -839,18 +879,19 @@ static int gen8_ppgtt_alloc_page_directories(struct i915_hw_ppgtt *ppgtt,
 	return 0;
 
 unwind_out:
-	for_each_set_bit(pdpe, new_pds, GEN8_LEGACY_PDPES)
+	for_each_set_bit(pdpe, new_pds, pdpes)
 		free_pd(dev, pdp->page_directory[pdpe]);
 
 	return -ENOMEM;
 }
 
 static void
-free_gen8_temp_bitmaps(unsigned long *new_pds, unsigned long **new_pts)
+free_gen8_temp_bitmaps(unsigned long *new_pds, unsigned long **new_pts,
+		       uint32_t pdpes)
 {
 	int i;
 
-	for (i = 0; i < GEN8_LEGACY_PDPES; i++)
+	for (i = 0; i < pdpes; i++)
 		kfree(new_pts[i]);
 	kfree(new_pts);
 	kfree(new_pds);
@@ -861,23 +902,24 @@ free_gen8_temp_bitmaps(unsigned long *new_pds, unsigned long **new_pts)
  */
 static
 int __must_check alloc_gen8_temp_bitmaps(unsigned long **new_pds,
-					 unsigned long ***new_pts)
+					 unsigned long ***new_pts,
+					 uint32_t pdpes)
 {
 	int i;
 	unsigned long *pds;
 	unsigned long **pts;
 
-	pds = kcalloc(BITS_TO_LONGS(GEN8_LEGACY_PDPES), sizeof(unsigned long), GFP_KERNEL);
+	pds = kcalloc(BITS_TO_LONGS(pdpes), sizeof(unsigned long), GFP_KERNEL);
 	if (!pds)
 		return -ENOMEM;
 
-	pts = kcalloc(GEN8_LEGACY_PDPES, sizeof(unsigned long *), GFP_KERNEL);
+	pts = kcalloc(pdpes, sizeof(unsigned long *), GFP_KERNEL);
 	if (!pts) {
 		kfree(pds);
 		return -ENOMEM;
 	}
 
-	for (i = 0; i < GEN8_LEGACY_PDPES; i++) {
+	for (i = 0; i < pdpes; i++) {
 		pts[i] = kcalloc(BITS_TO_LONGS(I915_PDES),
 				 sizeof(unsigned long), GFP_KERNEL);
 		if (!pts[i])
@@ -890,7 +932,7 @@ int __must_check alloc_gen8_temp_bitmaps(unsigned long **new_pds,
 	return 0;
 
 err_out:
-	free_gen8_temp_bitmaps(pds, pts);
+	free_gen8_temp_bitmaps(pds, pts, pdpes);
 	return -ENOMEM;
 }
 
@@ -916,6 +958,7 @@ static int gen8_alloc_va_range(struct i915_address_space *vm,
 	const uint64_t orig_length = length;
 	uint64_t temp;
 	uint32_t pdpe;
+	uint32_t pdpes = I915_PDPES_PER_PDP(ppgtt->base.dev);
 	int ret;
 
 	/* Wrap is never okay since we can only represent 48b, and we don't
@@ -927,7 +970,7 @@ static int gen8_alloc_va_range(struct i915_address_space *vm,
 	if (WARN_ON(start + length > ppgtt->base.total))
 		return -ENODEV;
 
-	ret = alloc_gen8_temp_bitmaps(&new_page_dirs, &new_page_tables);
+	ret = alloc_gen8_temp_bitmaps(&new_page_dirs, &new_page_tables, pdpes);
 	if (ret)
 		return ret;
 
@@ -935,7 +978,7 @@ static int gen8_alloc_va_range(struct i915_address_space *vm,
 	ret = gen8_ppgtt_alloc_page_directories(ppgtt, &ppgtt->pdp, start, length,
 					new_page_dirs);
 	if (ret) {
-		free_gen8_temp_bitmaps(new_page_dirs, new_page_tables);
+		free_gen8_temp_bitmaps(new_page_dirs, new_page_tables, pdpes);
 		return ret;
 	}
 
@@ -989,7 +1032,7 @@ static int gen8_alloc_va_range(struct i915_address_space *vm,
 		__set_bit(pdpe, ppgtt->pdp.used_pdpes);
 	}
 
-	free_gen8_temp_bitmaps(new_page_dirs, new_page_tables);
+	free_gen8_temp_bitmaps(new_page_dirs, new_page_tables, pdpes);
 	mark_tlbs_dirty(ppgtt);
 	return 0;
 
@@ -999,10 +1042,10 @@ err_out:
 			free_pt(vm->dev, ppgtt->pdp.page_directory[pdpe]->page_table[temp]);
 	}
 
-	for_each_set_bit(pdpe, new_page_dirs, GEN8_LEGACY_PDPES)
+	for_each_set_bit(pdpe, new_page_dirs, pdpes)
 		free_pd(vm->dev, ppgtt->pdp.page_directory[pdpe]);
 
-	free_gen8_temp_bitmaps(new_page_dirs, new_page_tables);
+	free_gen8_temp_bitmaps(new_page_dirs, new_page_tables, pdpes);
 	mark_tlbs_dirty(ppgtt);
 	return ret;
 }
@@ -1040,7 +1083,16 @@ static int gen8_ppgtt_init(struct i915_hw_ppgtt *ppgtt)
 
 	ppgtt->switch_mm = gen8_mm_switch;
 
+	ret = __pdp_init(false, &ppgtt->pdp);
+
+	if (ret)
+		goto free_scratch;
+
 	return 0;
+
+free_scratch:
+	gen8_free_scratch(&ppgtt->base);
+	return ret;
 }
 
 static void gen6_dump_ppgtt(struct i915_hw_ppgtt *ppgtt, struct seq_file *m)
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.h b/drivers/gpu/drm/i915/i915_gem_gtt.h
index d5bf953..87e389c 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.h
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.h
@@ -98,6 +98,9 @@ typedef uint64_t gen8_pde_t;
 #define GEN8_LEGACY_PDPES		4
 #define GEN8_PTES			I915_PTES(sizeof(gen8_pte_t))
 
+/* FIXME: Next patch will use dev */
+#define I915_PDPES_PER_PDP(dev)		GEN8_LEGACY_PDPES
+
 #define PPAT_UNCACHED_INDEX		(_PAGE_PWT | _PAGE_PCD)
 #define PPAT_CACHED_PDE_INDEX		0 /* WB LLC */
 #define PPAT_CACHED_INDEX		_PAGE_PAT /* WB LLCeLLC */
@@ -241,9 +244,10 @@ struct i915_page_directory {
 };
 
 struct i915_page_directory_pointer {
-	/* struct page *page; */
-	DECLARE_BITMAP(used_pdpes, GEN8_LEGACY_PDPES);
-	struct i915_page_directory *page_directory[GEN8_LEGACY_PDPES];
+	struct i915_page_dma base;
+
+	unsigned long *used_pdpes;
+	struct i915_page_directory **page_directory;
 };
 
 struct i915_address_space {
@@ -436,9 +440,10 @@ static inline uint32_t gen6_pde_index(uint32_t addr)
 	     temp = min(temp, length),					\
 	     start += temp, length -= temp)
 
-#define gen8_for_each_pdpe(pd, pdp, start, length, temp, iter)		\
-	for (iter = gen8_pdpe_index(start);	\
-	     pd = (pdp)->page_directory[iter], length > 0 && iter < GEN8_LEGACY_PDPES;	\
+#define gen8_for_each_pdpe(pd, pdp, start, length, temp, iter)	\
+	for (iter = gen8_pdpe_index(start); \
+	     pd = (pdp)->page_directory[iter], \
+	     length > 0 && (iter < I915_PDPES_PER_PDP(dev)); \
 	     iter++,				\
 	     temp = ALIGN(start+1, 1 << GEN8_PDPE_SHIFT) - start,	\
 	     temp = min(temp, length),					\
-- 
cgit v0.10.2


From d4ec9da0e17cb6a86c0b76c5b254981601d25031 Mon Sep 17 00:00:00 2001
From: Michel Thierry <michel.thierry@intel.com>
Date: Thu, 30 Jul 2015 11:02:03 +0100
Subject: drm/i915/gen8: Abstract PDP usage

Up until now, ppgtt->pdp has always been the root of our page tables.
Legacy 32b addresses acted like it had 1 PDP with 4 PDPEs.

In preparation for 4 level page tables, we need to stop using ppgtt->pdp
directly unless we know it's what we want. The future structure will use
ppgtt->pml4 for the top level, and the pdp is just one of the entries
being pointed to by a pml4e. The temporal pdp local variable will be
removed once the rest of the 4-level code lands.

Also, start passing the vm pointer to the alloc functions, instead of
ppgtt.

v2: Updated after dynamic page allocation changes.
v3: Rebase after s/page_tables/page_table/.
v4: Rebase after changes in "Dynamic page table allocations" patch.
v5: Rebase after Mika's ppgtt cleanup / scratch merge patch series.
v6: Rebase after final merged version of Mika's ppgtt/scratch patches.
v7: Keep pagetable map in-line (and avoid unnecessary for_each_pde
loops), remove redundant ppgtt pointer in _alloc_pagetabs (Akash)
v8: Fix text indentation in _alloc_pagetabs/page_directories (Chris)
v9: Defer gen8_alloc_va_range_4lvl definition until 4lvl is implemented,
clean-up gen8_ppgtt_cleanup [pun intended] (Akash).
v10: Clean-up commit message (Akash).

Cc: Akash Goel <akash.goel@intel.com>
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Michel Thierry <michel.thierry@intel.com> (v2+)
Reviewed-by: "Akash Goel" <akash.goel@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 6724e71..95a4688 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -607,6 +607,7 @@ static void gen8_ppgtt_clear_range(struct i915_address_space *vm,
 {
 	struct i915_hw_ppgtt *ppgtt =
 		container_of(vm, struct i915_hw_ppgtt, base);
+	struct i915_page_directory_pointer *pdp = &ppgtt->pdp; /* FIXME: 48b */
 	gen8_pte_t *pt_vaddr, scratch_pte;
 	unsigned pdpe = start >> GEN8_PDPE_SHIFT & GEN8_PDPE_MASK;
 	unsigned pde = start >> GEN8_PDE_SHIFT & GEN8_PDE_MASK;
@@ -621,10 +622,10 @@ static void gen8_ppgtt_clear_range(struct i915_address_space *vm,
 		struct i915_page_directory *pd;
 		struct i915_page_table *pt;
 
-		if (WARN_ON(!ppgtt->pdp.page_directory[pdpe]))
+		if (WARN_ON(!pdp->page_directory[pdpe]))
 			break;
 
-		pd = ppgtt->pdp.page_directory[pdpe];
+		pd = pdp->page_directory[pdpe];
 
 		if (WARN_ON(!pd->page_table[pde]))
 			break;
@@ -662,6 +663,7 @@ static void gen8_ppgtt_insert_entries(struct i915_address_space *vm,
 {
 	struct i915_hw_ppgtt *ppgtt =
 		container_of(vm, struct i915_hw_ppgtt, base);
+	struct i915_page_directory_pointer *pdp = &ppgtt->pdp; /* FIXME: 48b */
 	gen8_pte_t *pt_vaddr;
 	unsigned pdpe = start >> GEN8_PDPE_SHIFT & GEN8_PDPE_MASK;
 	unsigned pde = start >> GEN8_PDE_SHIFT & GEN8_PDE_MASK;
@@ -675,7 +677,7 @@ static void gen8_ppgtt_insert_entries(struct i915_address_space *vm,
 			break;
 
 		if (pt_vaddr == NULL) {
-			struct i915_page_directory *pd = ppgtt->pdp.page_directory[pdpe];
+			struct i915_page_directory *pd = pdp->page_directory[pdpe];
 			struct i915_page_table *pt = pd->page_table[pde];
 			pt_vaddr = kmap_px(pt);
 		}
@@ -755,28 +757,29 @@ static void gen8_ppgtt_cleanup(struct i915_address_space *vm)
 {
 	struct i915_hw_ppgtt *ppgtt =
 		container_of(vm, struct i915_hw_ppgtt, base);
+	struct i915_page_directory_pointer *pdp = &ppgtt->pdp; /* FIXME: 48b */
+	struct drm_device *dev = ppgtt->base.dev;
 	int i;
 
-	for_each_set_bit(i, ppgtt->pdp.used_pdpes,
-			 I915_PDPES_PER_PDP(ppgtt->base.dev)) {
-		if (WARN_ON(!ppgtt->pdp.page_directory[i]))
+	for_each_set_bit(i, pdp->used_pdpes, I915_PDPES_PER_PDP(dev)) {
+		if (WARN_ON(!pdp->page_directory[i]))
 			continue;
 
-		gen8_free_page_tables(ppgtt->base.dev,
-				      ppgtt->pdp.page_directory[i]);
-		free_pd(ppgtt->base.dev, ppgtt->pdp.page_directory[i]);
+		gen8_free_page_tables(dev, pdp->page_directory[i]);
+		free_pd(dev, pdp->page_directory[i]);
 	}
 
-	free_pdp(ppgtt->base.dev, &ppgtt->pdp);
+	free_pdp(dev, pdp);
+
 	gen8_free_scratch(vm);
 }
 
 /**
  * gen8_ppgtt_alloc_pagetabs() - Allocate page tables for VA range.
- * @ppgtt:	Master ppgtt structure.
- * @pd:		Page directory for this address range.
+ * @vm:	Master vm structure.
+ * @pd:	Page directory for this address range.
  * @start:	Starting virtual address to begin allocations.
- * @length	Size of the allocations.
+ * @length:	Size of the allocations.
  * @new_pts:	Bitmap set by function with new allocations. Likely used by the
  *		caller to free on error.
  *
@@ -789,13 +792,13 @@ static void gen8_ppgtt_cleanup(struct i915_address_space *vm)
  *
  * Return: 0 if success; negative error code otherwise.
  */
-static int gen8_ppgtt_alloc_pagetabs(struct i915_hw_ppgtt *ppgtt,
+static int gen8_ppgtt_alloc_pagetabs(struct i915_address_space *vm,
 				     struct i915_page_directory *pd,
 				     uint64_t start,
 				     uint64_t length,
 				     unsigned long *new_pts)
 {
-	struct drm_device *dev = ppgtt->base.dev;
+	struct drm_device *dev = vm->dev;
 	struct i915_page_table *pt;
 	uint64_t temp;
 	uint32_t pde;
@@ -804,7 +807,7 @@ static int gen8_ppgtt_alloc_pagetabs(struct i915_hw_ppgtt *ppgtt,
 		/* Don't reallocate page tables */
 		if (test_bit(pde, pd->used_pdes)) {
 			/* Scratch is never allocated this way */
-			WARN_ON(pt == ppgtt->base.scratch_pt);
+			WARN_ON(pt == vm->scratch_pt);
 			continue;
 		}
 
@@ -812,7 +815,7 @@ static int gen8_ppgtt_alloc_pagetabs(struct i915_hw_ppgtt *ppgtt,
 		if (IS_ERR(pt))
 			goto unwind_out;
 
-		gen8_initialize_pt(&ppgtt->base, pt);
+		gen8_initialize_pt(vm, pt);
 		pd->page_table[pde] = pt;
 		__set_bit(pde, new_pts);
 	}
@@ -828,11 +831,11 @@ unwind_out:
 
 /**
  * gen8_ppgtt_alloc_page_directories() - Allocate page directories for VA range.
- * @ppgtt:	Master ppgtt structure.
+ * @vm:	Master vm structure.
  * @pdp:	Page directory pointer for this address range.
  * @start:	Starting virtual address to begin allocations.
- * @length	Size of the allocations.
- * @new_pds	Bitmap set by function with new allocations. Likely used by the
+ * @length:	Size of the allocations.
+ * @new_pds:	Bitmap set by function with new allocations. Likely used by the
  *		caller to free on error.
  *
  * Allocate the required number of page directories starting at the pde index of
@@ -849,13 +852,14 @@ unwind_out:
  *
  * Return: 0 if success; negative error code otherwise.
  */
-static int gen8_ppgtt_alloc_page_directories(struct i915_hw_ppgtt *ppgtt,
-				     struct i915_page_directory_pointer *pdp,
-				     uint64_t start,
-				     uint64_t length,
-				     unsigned long *new_pds)
+static int
+gen8_ppgtt_alloc_page_directories(struct i915_address_space *vm,
+				  struct i915_page_directory_pointer *pdp,
+				  uint64_t start,
+				  uint64_t length,
+				  unsigned long *new_pds)
 {
-	struct drm_device *dev = ppgtt->base.dev;
+	struct drm_device *dev = vm->dev;
 	struct i915_page_directory *pd;
 	uint64_t temp;
 	uint32_t pdpe;
@@ -871,7 +875,7 @@ static int gen8_ppgtt_alloc_page_directories(struct i915_hw_ppgtt *ppgtt,
 		if (IS_ERR(pd))
 			goto unwind_out;
 
-		gen8_initialize_pd(&ppgtt->base, pd);
+		gen8_initialize_pd(vm, pd);
 		pdp->page_directory[pdpe] = pd;
 		__set_bit(pdpe, new_pds);
 	}
@@ -947,18 +951,19 @@ static void mark_tlbs_dirty(struct i915_hw_ppgtt *ppgtt)
 }
 
 static int gen8_alloc_va_range(struct i915_address_space *vm,
-			       uint64_t start,
-			       uint64_t length)
+			       uint64_t start, uint64_t length)
 {
 	struct i915_hw_ppgtt *ppgtt =
 		container_of(vm, struct i915_hw_ppgtt, base);
 	unsigned long *new_page_dirs, **new_page_tables;
+	struct drm_device *dev = vm->dev;
+	struct i915_page_directory_pointer *pdp = &ppgtt->pdp; /* FIXME: 48b */
 	struct i915_page_directory *pd;
 	const uint64_t orig_start = start;
 	const uint64_t orig_length = length;
 	uint64_t temp;
 	uint32_t pdpe;
-	uint32_t pdpes = I915_PDPES_PER_PDP(ppgtt->base.dev);
+	uint32_t pdpes = I915_PDPES_PER_PDP(dev);
 	int ret;
 
 	/* Wrap is never okay since we can only represent 48b, and we don't
@@ -967,7 +972,7 @@ static int gen8_alloc_va_range(struct i915_address_space *vm,
 	if (WARN_ON(start + length < start))
 		return -ENODEV;
 
-	if (WARN_ON(start + length > ppgtt->base.total))
+	if (WARN_ON(start + length > vm->total))
 		return -ENODEV;
 
 	ret = alloc_gen8_temp_bitmaps(&new_page_dirs, &new_page_tables, pdpes);
@@ -975,16 +980,16 @@ static int gen8_alloc_va_range(struct i915_address_space *vm,
 		return ret;
 
 	/* Do the allocations first so we can easily bail out */
-	ret = gen8_ppgtt_alloc_page_directories(ppgtt, &ppgtt->pdp, start, length,
-					new_page_dirs);
+	ret = gen8_ppgtt_alloc_page_directories(vm, pdp, start, length,
+						new_page_dirs);
 	if (ret) {
 		free_gen8_temp_bitmaps(new_page_dirs, new_page_tables, pdpes);
 		return ret;
 	}
 
 	/* For every page directory referenced, allocate page tables */
-	gen8_for_each_pdpe(pd, &ppgtt->pdp, start, length, temp, pdpe) {
-		ret = gen8_ppgtt_alloc_pagetabs(ppgtt, pd, start, length,
+	gen8_for_each_pdpe(pd, pdp, start, length, temp, pdpe) {
+		ret = gen8_ppgtt_alloc_pagetabs(vm, pd, start, length,
 						new_page_tables[pdpe]);
 		if (ret)
 			goto err_out;
@@ -995,7 +1000,7 @@ static int gen8_alloc_va_range(struct i915_address_space *vm,
 
 	/* Allocations have completed successfully, so set the bitmaps, and do
 	 * the mappings. */
-	gen8_for_each_pdpe(pd, &ppgtt->pdp, start, length, temp, pdpe) {
+	gen8_for_each_pdpe(pd, pdp, start, length, temp, pdpe) {
 		gen8_pde_t *const page_directory = kmap_px(pd);
 		struct i915_page_table *pt;
 		uint64_t pd_len = length;
@@ -1028,8 +1033,7 @@ static int gen8_alloc_va_range(struct i915_address_space *vm,
 		}
 
 		kunmap_px(ppgtt, page_directory);
-
-		__set_bit(pdpe, ppgtt->pdp.used_pdpes);
+		__set_bit(pdpe, pdp->used_pdpes);
 	}
 
 	free_gen8_temp_bitmaps(new_page_dirs, new_page_tables, pdpes);
@@ -1039,11 +1043,11 @@ static int gen8_alloc_va_range(struct i915_address_space *vm,
 err_out:
 	while (pdpe--) {
 		for_each_set_bit(temp, new_page_tables[pdpe], I915_PDES)
-			free_pt(vm->dev, ppgtt->pdp.page_directory[pdpe]->page_table[temp]);
+			free_pt(dev, pdp->page_directory[pdpe]->page_table[temp]);
 	}
 
 	for_each_set_bit(pdpe, new_page_dirs, pdpes)
-		free_pd(vm->dev, ppgtt->pdp.page_directory[pdpe]);
+		free_pd(dev, pdp->page_directory[pdpe]);
 
 	free_gen8_temp_bitmaps(new_page_dirs, new_page_tables, pdpes);
 	mark_tlbs_dirty(ppgtt);
-- 
cgit v0.10.2


From f9b5b782c4720f6c353088b559af5842ce66aa6f Mon Sep 17 00:00:00 2001
From: Michel Thierry <michel.thierry@intel.com>
Date: Thu, 30 Jul 2015 11:02:49 +0100
Subject: drm/i915/gen8: Generalize PTE writing for GEN8 PPGTT

The insert_entries function was the function used to write PTEs. For the
PPGTT it was "hardcoded" to only understand two level page tables, which
was the case for GEN7. We can reuse this for 4 level page tables, and
remove the concept of insert_entries, which was never viable past 2
level page tables anyway, but it requires a bit of rework to make the
function a bit more generic.

v2: Rebase after Mika's ppgtt cleanup / scratch merge patch series.
v3: Rebase after final merged version of Mika's ppgtt/scratch patches.
v4: Check and warn for NULL value of pdp pointer (Akash).

Cc: Akash Goel <akash.goel@intel.com>
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Michel Thierry <michel.thierry@intel.com> (v2)
Reviewed-by: Akash Goel <akash.goel@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 95a4688..fc58f54 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -600,23 +600,23 @@ static int gen8_mm_switch(struct i915_hw_ppgtt *ppgtt,
 	return 0;
 }
 
-static void gen8_ppgtt_clear_range(struct i915_address_space *vm,
-				   uint64_t start,
-				   uint64_t length,
-				   bool use_scratch)
+static void gen8_ppgtt_clear_pte_range(struct i915_address_space *vm,
+				       struct i915_page_directory_pointer *pdp,
+				       uint64_t start,
+				       uint64_t length,
+				       gen8_pte_t scratch_pte)
 {
 	struct i915_hw_ppgtt *ppgtt =
 		container_of(vm, struct i915_hw_ppgtt, base);
-	struct i915_page_directory_pointer *pdp = &ppgtt->pdp; /* FIXME: 48b */
-	gen8_pte_t *pt_vaddr, scratch_pte;
+	gen8_pte_t *pt_vaddr;
 	unsigned pdpe = start >> GEN8_PDPE_SHIFT & GEN8_PDPE_MASK;
 	unsigned pde = start >> GEN8_PDE_SHIFT & GEN8_PDE_MASK;
 	unsigned pte = start >> GEN8_PTE_SHIFT & GEN8_PTE_MASK;
 	unsigned num_entries = length >> PAGE_SHIFT;
 	unsigned last_pte, i;
 
-	scratch_pte = gen8_pte_encode(px_dma(ppgtt->base.scratch_page),
-				      I915_CACHE_LLC, use_scratch);
+	if (WARN_ON(!pdp))
+		return;
 
 	while (num_entries) {
 		struct i915_page_directory *pd;
@@ -656,14 +656,30 @@ static void gen8_ppgtt_clear_range(struct i915_address_space *vm,
 	}
 }
 
-static void gen8_ppgtt_insert_entries(struct i915_address_space *vm,
-				      struct sg_table *pages,
-				      uint64_t start,
-				      enum i915_cache_level cache_level, u32 unused)
+static void gen8_ppgtt_clear_range(struct i915_address_space *vm,
+				   uint64_t start,
+				   uint64_t length,
+				   bool use_scratch)
 {
 	struct i915_hw_ppgtt *ppgtt =
 		container_of(vm, struct i915_hw_ppgtt, base);
 	struct i915_page_directory_pointer *pdp = &ppgtt->pdp; /* FIXME: 48b */
+
+	gen8_pte_t scratch_pte = gen8_pte_encode(px_dma(vm->scratch_page),
+						 I915_CACHE_LLC, use_scratch);
+
+	gen8_ppgtt_clear_pte_range(vm, pdp, start, length, scratch_pte);
+}
+
+static void
+gen8_ppgtt_insert_pte_entries(struct i915_address_space *vm,
+			      struct i915_page_directory_pointer *pdp,
+			      struct sg_table *pages,
+			      uint64_t start,
+			      enum i915_cache_level cache_level)
+{
+	struct i915_hw_ppgtt *ppgtt =
+		container_of(vm, struct i915_hw_ppgtt, base);
 	gen8_pte_t *pt_vaddr;
 	unsigned pdpe = start >> GEN8_PDPE_SHIFT & GEN8_PDPE_MASK;
 	unsigned pde = start >> GEN8_PDE_SHIFT & GEN8_PDE_MASK;
@@ -700,6 +716,19 @@ static void gen8_ppgtt_insert_entries(struct i915_address_space *vm,
 		kunmap_px(ppgtt, pt_vaddr);
 }
 
+static void gen8_ppgtt_insert_entries(struct i915_address_space *vm,
+				      struct sg_table *pages,
+				      uint64_t start,
+				      enum i915_cache_level cache_level,
+				      u32 unused)
+{
+	struct i915_hw_ppgtt *ppgtt =
+		container_of(vm, struct i915_hw_ppgtt, base);
+	struct i915_page_directory_pointer *pdp = &ppgtt->pdp; /* FIXME: 48b */
+
+	gen8_ppgtt_insert_pte_entries(vm, pdp, pages, start, cache_level);
+}
+
 static void gen8_free_page_tables(struct drm_device *dev,
 				  struct i915_page_directory *pd)
 {
-- 
cgit v0.10.2


From 4c06ec8d13d2b0e57479bb135e49afd56ebe9275 Mon Sep 17 00:00:00 2001
From: Michel Thierry <michel.thierry@intel.com>
Date: Wed, 29 Jul 2015 17:23:49 +0100
Subject: drm/i915/gen8: Add dynamic page trace events

The dynamic page allocation patch series added it for GEN6, this patch
adds them for GEN8.

v2: Consolidate pagetable/page_directory events
v3: Multiple rebases.
v4: Rebase after s/page_tables/page_table/.
v5: Rebase after Mika's ppgtt cleanup / scratch merge patch series.
v6: Rebase after gen8_map_pagetable_range removal.
v7: Use generic page name (px) in DECLARE_EVENT_CLASS (Akash)
v8: Defer define of i915_page_directory_pointer_entry_alloc (Akash)

Cc: Akash Goel <akash.goel@intel.com>
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Michel Thierry <michel.thierry@intel.com> (v3+)
Reviewed-by: Akash Goel <akash.goel@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index fc58f54..72f4777f 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -847,6 +847,7 @@ static int gen8_ppgtt_alloc_pagetabs(struct i915_address_space *vm,
 		gen8_initialize_pt(vm, pt);
 		pd->page_table[pde] = pt;
 		__set_bit(pde, new_pts);
+		trace_i915_page_table_entry_alloc(vm, pde, start, GEN8_PDE_SHIFT);
 	}
 
 	return 0;
@@ -907,6 +908,7 @@ gen8_ppgtt_alloc_page_directories(struct i915_address_space *vm,
 		gen8_initialize_pd(vm, pd);
 		pdp->page_directory[pdpe] = pd;
 		__set_bit(pdpe, new_pds);
+		trace_i915_page_directory_entry_alloc(vm, pdpe, start, GEN8_PDPE_SHIFT);
 	}
 
 	return 0;
@@ -1056,6 +1058,10 @@ static int gen8_alloc_va_range(struct i915_address_space *vm,
 			/* Map the PDE to the page table */
 			page_directory[pde] = gen8_pde_encode(px_dma(pt),
 							      I915_CACHE_LLC);
+			trace_i915_page_table_entry_map(&ppgtt->base, pde, pt,
+							gen8_pte_index(start),
+							gen8_pte_count(start, length),
+							GEN8_PTES);
 
 			/* NB: We haven't yet mapped ptes to pages. At this
 			 * point we're still relying on insert_entries() */
diff --git a/drivers/gpu/drm/i915/i915_trace.h b/drivers/gpu/drm/i915/i915_trace.h
index 2f34c47..f230d76 100644
--- a/drivers/gpu/drm/i915/i915_trace.h
+++ b/drivers/gpu/drm/i915/i915_trace.h
@@ -186,33 +186,41 @@ DEFINE_EVENT(i915_va, i915_va_alloc,
 	     TP_ARGS(vm, start, length, name)
 );
 
-DECLARE_EVENT_CLASS(i915_page_table_entry,
-	TP_PROTO(struct i915_address_space *vm, u32 pde, u64 start, u64 pde_shift),
-	TP_ARGS(vm, pde, start, pde_shift),
+DECLARE_EVENT_CLASS(i915_px_entry,
+	TP_PROTO(struct i915_address_space *vm, u32 px, u64 start, u64 px_shift),
+	TP_ARGS(vm, px, start, px_shift),
 
 	TP_STRUCT__entry(
 		__field(struct i915_address_space *, vm)
-		__field(u32, pde)
+		__field(u32, px)
 		__field(u64, start)
 		__field(u64, end)
 	),
 
 	TP_fast_assign(
 		__entry->vm = vm;
-		__entry->pde = pde;
+		__entry->px = px;
 		__entry->start = start;
-		__entry->end = ((start + (1ULL << pde_shift)) & ~((1ULL << pde_shift)-1)) - 1;
+		__entry->end = ((start + (1ULL << px_shift)) & ~((1ULL << px_shift)-1)) - 1;
 	),
 
 	TP_printk("vm=%p, pde=%d (0x%llx-0x%llx)",
-		  __entry->vm, __entry->pde, __entry->start, __entry->end)
+		  __entry->vm, __entry->px, __entry->start, __entry->end)
 );
 
-DEFINE_EVENT(i915_page_table_entry, i915_page_table_entry_alloc,
+DEFINE_EVENT(i915_px_entry, i915_page_table_entry_alloc,
 	     TP_PROTO(struct i915_address_space *vm, u32 pde, u64 start, u64 pde_shift),
 	     TP_ARGS(vm, pde, start, pde_shift)
 );
 
+DEFINE_EVENT_PRINT(i915_px_entry, i915_page_directory_entry_alloc,
+		   TP_PROTO(struct i915_address_space *vm, u32 pdpe, u64 start, u64 pdpe_shift),
+		   TP_ARGS(vm, pdpe, start, pdpe_shift),
+
+		   TP_printk("vm=%p, pdpe=%d (0x%llx-0x%llx)",
+			     __entry->vm, __entry->px, __entry->start, __entry->end)
+);
+
 /* Avoid extra math because we only support two sizes. The format is defined by
  * bitmap_scnprintf. Each 32 bits is 8 HEX digits followed by comma */
 #define TRACE_PT_SIZE(bits) \
-- 
cgit v0.10.2


From 81ba8aefd03803a8aec3395d18f7b1dda5942105 Mon Sep 17 00:00:00 2001
From: Michel Thierry <michel.thierry@intel.com>
Date: Mon, 3 Aug 2015 09:52:01 +0100
Subject: drm/i915/gen8: Add PML4 structure

Introduces the Page Map Level 4 (PML4), ie. the new top level structure
of the page tables.

To facilitate testing, 48b mode will be available on Broadwell and
GEN9+, when i915.enable_ppgtt = 3.

v2: Remove unnecessary CONFIG_X86_64 checks, ppgtt code is already
32/64-bit safe (Chris).
v3: Add goto free_scratch in temp 48-bit mode init code (Akash).
v4: kfree the pdp until the 4lvl alloc/free patch (Akash).
v5: Postpone 48-bit code in sanitize_enable_ppgtt (Akash).
v6: Keep _insert_pte_entries changes outside this patch (Akash).

Cc: Akash Goel <akash.goel@intel.com>
Signed-off-by: Michel Thierry <michel.thierry@intel.com>
Reviewed-by: Akash Goel <akash.goel@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 4273c28..79d2dcb 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -2510,7 +2510,8 @@ struct drm_i915_cmd_table {
 #define HAS_HW_CONTEXTS(dev)	(INTEL_INFO(dev)->gen >= 6)
 #define HAS_LOGICAL_RING_CONTEXTS(dev)	(INTEL_INFO(dev)->gen >= 8)
 #define USES_PPGTT(dev)		(i915.enable_ppgtt)
-#define USES_FULL_PPGTT(dev)	(i915.enable_ppgtt == 2)
+#define USES_FULL_PPGTT(dev)	(i915.enable_ppgtt >= 2)
+#define USES_FULL_48BIT_PPGTT(dev)	(i915.enable_ppgtt == 3)
 
 #define HAS_OVERLAY(dev)		(INTEL_INFO(dev)->has_overlay)
 #define OVERLAY_NEEDS_PHYSICAL(dev)	(INTEL_INFO(dev)->overlay_needs_physical)
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 72f4777f..da7863b 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -1105,14 +1105,6 @@ static int gen8_ppgtt_init(struct i915_hw_ppgtt *ppgtt)
 		return ret;
 
 	ppgtt->base.start = 0;
-	ppgtt->base.total = 1ULL << 32;
-	if (IS_ENABLED(CONFIG_X86_32))
-		/* While we have a proliferation of size_t variables
-		 * we cannot represent the full ppgtt size on 32bit,
-		 * so limit it to the same size as the GGTT (currently
-		 * 2GiB).
-		 */
-		ppgtt->base.total = to_i915(ppgtt->base.dev)->gtt.base.total;
 	ppgtt->base.cleanup = gen8_ppgtt_cleanup;
 	ppgtt->base.allocate_va_range = gen8_alloc_va_range;
 	ppgtt->base.insert_entries = gen8_ppgtt_insert_entries;
@@ -1122,10 +1114,25 @@ static int gen8_ppgtt_init(struct i915_hw_ppgtt *ppgtt)
 
 	ppgtt->switch_mm = gen8_mm_switch;
 
-	ret = __pdp_init(false, &ppgtt->pdp);
+	if (!USES_FULL_48BIT_PPGTT(ppgtt->base.dev)) {
+		ret = __pdp_init(false, &ppgtt->pdp);
 
-	if (ret)
+		if (ret)
+			goto free_scratch;
+
+		ppgtt->base.total = 1ULL << 32;
+		if (IS_ENABLED(CONFIG_X86_32))
+			/* While we have a proliferation of size_t variables
+			 * we cannot represent the full ppgtt size on 32bit,
+			 * so limit it to the same size as the GGTT (currently
+			 * 2GiB).
+			 */
+			ppgtt->base.total = to_i915(ppgtt->base.dev)->gtt.base.total;
+	} else {
+		ppgtt->base.total = 1ULL << 48;
+		ret = -EPERM; /* Not yet implemented */
 		goto free_scratch;
+	}
 
 	return 0;
 
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.h b/drivers/gpu/drm/i915/i915_gem_gtt.h
index 87e389c..04bc66f 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.h
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.h
@@ -88,9 +88,17 @@ typedef uint64_t gen8_pde_t;
  * PDPE  |  PDE  |  PTE  | offset
  * The difference as compared to normal x86 3 level page table is the PDPEs are
  * programmed via register.
+ *
+ * GEN8 48b legacy style address is defined as a 4 level page table:
+ * 47:39 | 38:30 | 29:21 | 20:12 |  11:0
+ * PML4E | PDPE  |  PDE  |  PTE  | offset
  */
+#define GEN8_PML4ES_PER_PML4		512
+#define GEN8_PML4E_SHIFT		39
 #define GEN8_PDPE_SHIFT			30
-#define GEN8_PDPE_MASK			0x3
+/* NB: GEN8_PDPE_MASK is untrue for 32b platforms, but it has no impact on 32b page
+ * tables */
+#define GEN8_PDPE_MASK			0x1ff
 #define GEN8_PDE_SHIFT			21
 #define GEN8_PDE_MASK			0x1ff
 #define GEN8_PTE_SHIFT			12
@@ -98,8 +106,8 @@ typedef uint64_t gen8_pde_t;
 #define GEN8_LEGACY_PDPES		4
 #define GEN8_PTES			I915_PTES(sizeof(gen8_pte_t))
 
-/* FIXME: Next patch will use dev */
-#define I915_PDPES_PER_PDP(dev)		GEN8_LEGACY_PDPES
+#define I915_PDPES_PER_PDP(dev) (USES_FULL_48BIT_PPGTT(dev) ?\
+				 GEN8_PML4ES_PER_PML4 : GEN8_LEGACY_PDPES)
 
 #define PPAT_UNCACHED_INDEX		(_PAGE_PWT | _PAGE_PCD)
 #define PPAT_CACHED_PDE_INDEX		0 /* WB LLC */
@@ -250,6 +258,13 @@ struct i915_page_directory_pointer {
 	struct i915_page_directory **page_directory;
 };
 
+struct i915_pml4 {
+	struct i915_page_dma base;
+
+	DECLARE_BITMAP(used_pml4es, GEN8_PML4ES_PER_PML4);
+	struct i915_page_directory_pointer *pdps[GEN8_PML4ES_PER_PML4];
+};
+
 struct i915_address_space {
 	struct drm_mm mm;
 	struct drm_device *dev;
@@ -345,8 +360,9 @@ struct i915_hw_ppgtt {
 	struct drm_mm_node node;
 	unsigned long pd_dirty_rings;
 	union {
-		struct i915_page_directory_pointer pdp;
-		struct i915_page_directory pd;
+		struct i915_pml4 pml4;		/* GEN8+ & 48b PPGTT */
+		struct i915_page_directory_pointer pdp;	/* GEN8+ */
+		struct i915_page_directory pd;		/* GEN6-7 */
 	};
 
 	struct drm_i915_file_private *file_priv;
-- 
cgit v0.10.2


From 762d99363dc9bc436f39f8bdc3f8670ea272a5a9 Mon Sep 17 00:00:00 2001
From: Michel Thierry <michel.thierry@intel.com>
Date: Thu, 30 Jul 2015 11:05:29 +0100
Subject: drm/i915/gen8: implement alloc/free for 4lvl

PML4 has no special attributes, and there will always be a PML4.
So simply initialize it at creation, and destroy it at the end.

The code for 4lvl is able to call into the existing 3lvl page table code
to handle all of the lower levels.

v2: Return something at the end of gen8_alloc_va_range_4lvl to keep the
compiler happy. And define ret only in one place.
Updated gen8_ppgtt_unmap_pages and gen8_ppgtt_free to handle 4lvl.
v3: Use i915_dma_unmap_single instead of pci API. Fix a
couple of incorrect checks when unmapping pdp and pd pages (Akash).
v4: Call __pdp_fini also for 32b PPGTT. Clean up alloc_pdp param list.
v5: Prevent (harmless) out of range access in gen8_for_each_pml4e.
v6: Simplify alloc_vma_range_4lvl and gen8_ppgtt_init_common error
paths. (Akash)
v7: Rebase, s/gen8_ppgtt_free_*/gen8_ppgtt_cleanup_*/.
v8: Change location of pml4_init/fini. It will make next patches
cleaner.
v9: Rebase after Mika's ppgtt cleanup / scratch merge patch series, while
trying to reuse as much as possible for pdp alloc. pml4_init/fini
replaced by setup/cleanup_px macros.
v10: Rebase after Mika's merged ppgtt cleanup patch series.
v11: Rebase after final merged version of Mika's ppgtt/scratch
patches.
v12: Fix pdpe start value in trace (Akash)
v13: Define all 4lvl functions in this patch directly, instead of
previous patches, add i915_page_directory_pointer_entry_alloc here,
use test_bit to detect when pdp is already allocated (Akash).
v14: Move pdp allocation into a new gen8_ppgtt_alloc_page_dirpointers
funtion, as we do for pds and pts; move pd and pdp setup functions to
this patch (Akash).
v15: Added kfree(pdp) from previous patch to this (Akash).

Cc: Akash Goel <akash.goel@intel.com>
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Michel Thierry <michel.thierry@intel.com> (v2+)
Reviewed-by: Akash Goel <akash.goel@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index da7863b..229a317 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -204,6 +204,9 @@ static gen8_pde_t gen8_pde_encode(const dma_addr_t addr,
 	return pde;
 }
 
+#define gen8_pdpe_encode gen8_pde_encode
+#define gen8_pml4e_encode gen8_pde_encode
+
 static gen6_pte_t snb_pte_encode(dma_addr_t addr,
 				 enum i915_cache_level level,
 				 bool valid, u32 unused)
@@ -553,10 +556,73 @@ static void __pdp_fini(struct i915_page_directory_pointer *pdp)
 	pdp->page_directory = NULL;
 }
 
+static struct
+i915_page_directory_pointer *alloc_pdp(struct drm_device *dev)
+{
+	struct i915_page_directory_pointer *pdp;
+	int ret = -ENOMEM;
+
+	WARN_ON(!USES_FULL_48BIT_PPGTT(dev));
+
+	pdp = kzalloc(sizeof(*pdp), GFP_KERNEL);
+	if (!pdp)
+		return ERR_PTR(-ENOMEM);
+
+	ret = __pdp_init(dev, pdp);
+	if (ret)
+		goto fail_bitmap;
+
+	ret = setup_px(dev, pdp);
+	if (ret)
+		goto fail_page_m;
+
+	return pdp;
+
+fail_page_m:
+	__pdp_fini(pdp);
+fail_bitmap:
+	kfree(pdp);
+
+	return ERR_PTR(ret);
+}
+
 static void free_pdp(struct drm_device *dev,
 		     struct i915_page_directory_pointer *pdp)
 {
 	__pdp_fini(pdp);
+	if (USES_FULL_48BIT_PPGTT(dev)) {
+		cleanup_px(dev, pdp);
+		kfree(pdp);
+	}
+}
+
+static void
+gen8_setup_page_directory(struct i915_hw_ppgtt *ppgtt,
+			  struct i915_page_directory_pointer *pdp,
+			  struct i915_page_directory *pd,
+			  int index)
+{
+	gen8_ppgtt_pdpe_t *page_directorypo;
+
+	if (!USES_FULL_48BIT_PPGTT(ppgtt->base.dev))
+		return;
+
+	page_directorypo = kmap_px(pdp);
+	page_directorypo[index] = gen8_pdpe_encode(px_dma(pd), I915_CACHE_LLC);
+	kunmap_px(ppgtt, page_directorypo);
+}
+
+static void
+gen8_setup_page_directory_pointer(struct i915_hw_ppgtt *ppgtt,
+				  struct i915_pml4 *pml4,
+				  struct i915_page_directory_pointer *pdp,
+				  int index)
+{
+	gen8_ppgtt_pml4e_t *pagemap = kmap_px(pml4);
+
+	WARN_ON(!USES_FULL_48BIT_PPGTT(ppgtt->base.dev));
+	pagemap[index] = gen8_pml4e_encode(px_dma(pdp), I915_CACHE_LLC);
+	kunmap_px(ppgtt, pagemap);
 }
 
 /* Broadwell Page Directory Pointer Descriptors */
@@ -782,12 +848,9 @@ static void gen8_free_scratch(struct i915_address_space *vm)
 	free_scratch_page(dev, vm->scratch_page);
 }
 
-static void gen8_ppgtt_cleanup(struct i915_address_space *vm)
+static void gen8_ppgtt_cleanup_3lvl(struct drm_device *dev,
+				    struct i915_page_directory_pointer *pdp)
 {
-	struct i915_hw_ppgtt *ppgtt =
-		container_of(vm, struct i915_hw_ppgtt, base);
-	struct i915_page_directory_pointer *pdp = &ppgtt->pdp; /* FIXME: 48b */
-	struct drm_device *dev = ppgtt->base.dev;
 	int i;
 
 	for_each_set_bit(i, pdp->used_pdpes, I915_PDPES_PER_PDP(dev)) {
@@ -799,6 +862,31 @@ static void gen8_ppgtt_cleanup(struct i915_address_space *vm)
 	}
 
 	free_pdp(dev, pdp);
+}
+
+static void gen8_ppgtt_cleanup_4lvl(struct i915_hw_ppgtt *ppgtt)
+{
+	int i;
+
+	for_each_set_bit(i, ppgtt->pml4.used_pml4es, GEN8_PML4ES_PER_PML4) {
+		if (WARN_ON(!ppgtt->pml4.pdps[i]))
+			continue;
+
+		gen8_ppgtt_cleanup_3lvl(ppgtt->base.dev, ppgtt->pml4.pdps[i]);
+	}
+
+	cleanup_px(ppgtt->base.dev, &ppgtt->pml4);
+}
+
+static void gen8_ppgtt_cleanup(struct i915_address_space *vm)
+{
+	struct i915_hw_ppgtt *ppgtt =
+		container_of(vm, struct i915_hw_ppgtt, base);
+
+	if (!USES_FULL_48BIT_PPGTT(ppgtt->base.dev))
+		gen8_ppgtt_cleanup_3lvl(ppgtt->base.dev, &ppgtt->pdp);
+	else
+		gen8_ppgtt_cleanup_4lvl(ppgtt);
 
 	gen8_free_scratch(vm);
 }
@@ -920,6 +1008,60 @@ unwind_out:
 	return -ENOMEM;
 }
 
+/**
+ * gen8_ppgtt_alloc_page_dirpointers() - Allocate pdps for VA range.
+ * @vm:	Master vm structure.
+ * @pml4:	Page map level 4 for this address range.
+ * @start:	Starting virtual address to begin allocations.
+ * @length:	Size of the allocations.
+ * @new_pdps:	Bitmap set by function with new allocations. Likely used by the
+ *		caller to free on error.
+ *
+ * Allocate the required number of page directory pointers. Extremely similar to
+ * gen8_ppgtt_alloc_page_directories() and gen8_ppgtt_alloc_pagetabs().
+ * The main difference is here we are limited by the pml4 boundary (instead of
+ * the page directory pointer).
+ *
+ * Return: 0 if success; negative error code otherwise.
+ */
+static int
+gen8_ppgtt_alloc_page_dirpointers(struct i915_address_space *vm,
+				  struct i915_pml4 *pml4,
+				  uint64_t start,
+				  uint64_t length,
+				  unsigned long *new_pdps)
+{
+	struct drm_device *dev = vm->dev;
+	struct i915_page_directory_pointer *pdp;
+	uint64_t temp;
+	uint32_t pml4e;
+
+	WARN_ON(!bitmap_empty(new_pdps, GEN8_PML4ES_PER_PML4));
+
+	gen8_for_each_pml4e(pdp, pml4, start, length, temp, pml4e) {
+		if (!test_bit(pml4e, pml4->used_pml4es)) {
+			pdp = alloc_pdp(dev);
+			if (IS_ERR(pdp))
+				goto unwind_out;
+
+			pml4->pdps[pml4e] = pdp;
+			__set_bit(pml4e, new_pdps);
+			trace_i915_page_directory_pointer_entry_alloc(vm,
+								      pml4e,
+								      start,
+								      GEN8_PML4E_SHIFT);
+		}
+	}
+
+	return 0;
+
+unwind_out:
+	for_each_set_bit(pml4e, new_pdps, GEN8_PML4ES_PER_PML4)
+		free_pdp(dev, pml4->pdps[pml4e]);
+
+	return -ENOMEM;
+}
+
 static void
 free_gen8_temp_bitmaps(unsigned long *new_pds, unsigned long **new_pts,
 		       uint32_t pdpes)
@@ -981,14 +1123,15 @@ static void mark_tlbs_dirty(struct i915_hw_ppgtt *ppgtt)
 	ppgtt->pd_dirty_rings = INTEL_INFO(ppgtt->base.dev)->ring_mask;
 }
 
-static int gen8_alloc_va_range(struct i915_address_space *vm,
-			       uint64_t start, uint64_t length)
+static int gen8_alloc_va_range_3lvl(struct i915_address_space *vm,
+				    struct i915_page_directory_pointer *pdp,
+				    uint64_t start,
+				    uint64_t length)
 {
 	struct i915_hw_ppgtt *ppgtt =
 		container_of(vm, struct i915_hw_ppgtt, base);
 	unsigned long *new_page_dirs, **new_page_tables;
 	struct drm_device *dev = vm->dev;
-	struct i915_page_directory_pointer *pdp = &ppgtt->pdp; /* FIXME: 48b */
 	struct i915_page_directory *pd;
 	const uint64_t orig_start = start;
 	const uint64_t orig_length = length;
@@ -1069,6 +1212,7 @@ static int gen8_alloc_va_range(struct i915_address_space *vm,
 
 		kunmap_px(ppgtt, page_directory);
 		__set_bit(pdpe, pdp->used_pdpes);
+		gen8_setup_page_directory(ppgtt, pdp, pd, pdpe);
 	}
 
 	free_gen8_temp_bitmaps(new_page_dirs, new_page_tables, pdpes);
@@ -1089,6 +1233,68 @@ err_out:
 	return ret;
 }
 
+static int gen8_alloc_va_range_4lvl(struct i915_address_space *vm,
+				    struct i915_pml4 *pml4,
+				    uint64_t start,
+				    uint64_t length)
+{
+	DECLARE_BITMAP(new_pdps, GEN8_PML4ES_PER_PML4);
+	struct i915_hw_ppgtt *ppgtt =
+			container_of(vm, struct i915_hw_ppgtt, base);
+	struct i915_page_directory_pointer *pdp;
+	uint64_t temp, pml4e;
+	int ret = 0;
+
+	/* Do the pml4 allocations first, so we don't need to track the newly
+	 * allocated tables below the pdp */
+	bitmap_zero(new_pdps, GEN8_PML4ES_PER_PML4);
+
+	/* The pagedirectory and pagetable allocations are done in the shared 3
+	 * and 4 level code. Just allocate the pdps.
+	 */
+	ret = gen8_ppgtt_alloc_page_dirpointers(vm, pml4, start, length,
+						new_pdps);
+	if (ret)
+		return ret;
+
+	WARN(bitmap_weight(new_pdps, GEN8_PML4ES_PER_PML4) > 2,
+	     "The allocation has spanned more than 512GB. "
+	     "It is highly likely this is incorrect.");
+
+	gen8_for_each_pml4e(pdp, pml4, start, length, temp, pml4e) {
+		WARN_ON(!pdp);
+
+		ret = gen8_alloc_va_range_3lvl(vm, pdp, start, length);
+		if (ret)
+			goto err_out;
+
+		gen8_setup_page_directory_pointer(ppgtt, pml4, pdp, pml4e);
+	}
+
+	bitmap_or(pml4->used_pml4es, new_pdps, pml4->used_pml4es,
+		  GEN8_PML4ES_PER_PML4);
+
+	return 0;
+
+err_out:
+	for_each_set_bit(pml4e, new_pdps, GEN8_PML4ES_PER_PML4)
+		gen8_ppgtt_cleanup_3lvl(vm->dev, pml4->pdps[pml4e]);
+
+	return ret;
+}
+
+static int gen8_alloc_va_range(struct i915_address_space *vm,
+			       uint64_t start, uint64_t length)
+{
+	struct i915_hw_ppgtt *ppgtt =
+		container_of(vm, struct i915_hw_ppgtt, base);
+
+	if (USES_FULL_48BIT_PPGTT(vm->dev))
+		return gen8_alloc_va_range_4lvl(vm, &ppgtt->pml4, start, length);
+	else
+		return gen8_alloc_va_range_3lvl(vm, &ppgtt->pdp, start, length);
+}
+
 /*
  * GEN8 legacy ppgtt programming is accomplished through a max 4 PDP registers
  * with a net effect resembling a 2-level page table in normal x86 terms. Each
@@ -1114,9 +1320,14 @@ static int gen8_ppgtt_init(struct i915_hw_ppgtt *ppgtt)
 
 	ppgtt->switch_mm = gen8_mm_switch;
 
-	if (!USES_FULL_48BIT_PPGTT(ppgtt->base.dev)) {
-		ret = __pdp_init(false, &ppgtt->pdp);
+	if (USES_FULL_48BIT_PPGTT(ppgtt->base.dev)) {
+		ret = setup_px(ppgtt->base.dev, &ppgtt->pml4);
+		if (ret)
+			goto free_scratch;
 
+		ppgtt->base.total = 1ULL << 48;
+	} else {
+		ret = __pdp_init(false, &ppgtt->pdp);
 		if (ret)
 			goto free_scratch;
 
@@ -1128,10 +1339,10 @@ static int gen8_ppgtt_init(struct i915_hw_ppgtt *ppgtt)
 			 * 2GiB).
 			 */
 			ppgtt->base.total = to_i915(ppgtt->base.dev)->gtt.base.total;
-	} else {
-		ppgtt->base.total = 1ULL << 48;
-		ret = -EPERM; /* Not yet implemented */
-		goto free_scratch;
+
+		trace_i915_page_directory_pointer_entry_alloc(&ppgtt->base,
+							      0, 0,
+							      GEN8_PML4E_SHIFT);
 	}
 
 	return 0;
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.h b/drivers/gpu/drm/i915/i915_gem_gtt.h
index 04bc66f..11d44b3 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.h
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.h
@@ -39,6 +39,8 @@ struct drm_i915_file_private;
 typedef uint32_t gen6_pte_t;
 typedef uint64_t gen8_pte_t;
 typedef uint64_t gen8_pde_t;
+typedef uint64_t gen8_ppgtt_pdpe_t;
+typedef uint64_t gen8_ppgtt_pml4e_t;
 
 #define gtt_total_entries(gtt) ((gtt).base.total >> PAGE_SHIFT)
 
@@ -95,6 +97,7 @@ typedef uint64_t gen8_pde_t;
  */
 #define GEN8_PML4ES_PER_PML4		512
 #define GEN8_PML4E_SHIFT		39
+#define GEN8_PML4E_MASK			(GEN8_PML4ES_PER_PML4 - 1)
 #define GEN8_PDPE_SHIFT			30
 /* NB: GEN8_PDPE_MASK is untrue for 32b platforms, but it has no impact on 32b page
  * tables */
@@ -465,6 +468,15 @@ static inline uint32_t gen6_pde_index(uint32_t addr)
 	     temp = min(temp, length),					\
 	     start += temp, length -= temp)
 
+#define gen8_for_each_pml4e(pdp, pml4, start, length, temp, iter)	\
+	for (iter = gen8_pml4e_index(start);	\
+	     pdp = (pml4)->pdps[iter], \
+	     length > 0 && iter < GEN8_PML4ES_PER_PML4; \
+	     iter++,				\
+	     temp = ALIGN(start+1, 1ULL << GEN8_PML4E_SHIFT) - start,	\
+	     temp = min(temp, length),					\
+	     start += temp, length -= temp)
+
 static inline uint32_t gen8_pte_index(uint64_t address)
 {
 	return i915_pte_index(address, GEN8_PDE_SHIFT);
@@ -482,8 +494,7 @@ static inline uint32_t gen8_pdpe_index(uint64_t address)
 
 static inline uint32_t gen8_pml4e_index(uint64_t address)
 {
-	WARN_ON(1); /* For 64B */
-	return 0;
+	return (address >> GEN8_PML4E_SHIFT) & GEN8_PML4E_MASK;
 }
 
 static inline size_t gen8_pte_count(uint64_t address, uint64_t length)
diff --git a/drivers/gpu/drm/i915/i915_trace.h b/drivers/gpu/drm/i915/i915_trace.h
index f230d76..e6b5c74 100644
--- a/drivers/gpu/drm/i915/i915_trace.h
+++ b/drivers/gpu/drm/i915/i915_trace.h
@@ -221,6 +221,14 @@ DEFINE_EVENT_PRINT(i915_px_entry, i915_page_directory_entry_alloc,
 			     __entry->vm, __entry->px, __entry->start, __entry->end)
 );
 
+DEFINE_EVENT_PRINT(i915_px_entry, i915_page_directory_pointer_entry_alloc,
+		   TP_PROTO(struct i915_address_space *vm, u32 pml4e, u64 start, u64 pml4e_shift),
+		   TP_ARGS(vm, pml4e, start, pml4e_shift),
+
+		   TP_printk("vm=%p, pml4e=%d (0x%llx-0x%llx)",
+			     __entry->vm, __entry->px, __entry->start, __entry->end)
+);
+
 /* Avoid extra math because we only support two sizes. The format is defined by
  * bitmap_scnprintf. Each 32 bits is 8 HEX digits followed by comma */
 #define TRACE_PT_SIZE(bits) \
-- 
cgit v0.10.2


From 2dba3239f5c7511ffac957887facd0a0c9d003a5 Mon Sep 17 00:00:00 2001
From: Michel Thierry <michel.thierry@intel.com>
Date: Thu, 30 Jul 2015 11:06:23 +0100
Subject: drm/i915/gen8: Add 4 level switching infrastructure and lrc support

In 64b (48bit canonical) PPGTT addressing, the PDP0 register contains
the base address to PML4, while the other PDP registers are ignored.

In LRC, the addressing mode must be specified in every context
descriptor, and the base address to PML4 is stored in the reg state.

v2: PML4 update in legacy context switch is left for historic reasons,
the preferred mode of operation is with lrc context based submission.
v3: s/gen8_map_page_directory/gen8_setup_page_directory and
s/gen8_map_page_directory_pointer/gen8_setup_page_directory_pointer.
Also, clflush will be needed for bxt. (Akash)
v4: Squashed lrc-specific code and use a macro to set PML4 register.
v5: Rebase after Mika's ppgtt cleanup / scratch merge patch series.
PDP update in bb_start is only for legacy 32b mode.
v6: Rebase after final merged version of Mika's ppgtt/scratch
patches.
v7: There is no need to update the pml4 register value in
execlists_update_context. (Akash)
v8: Move pd and pdp setup functions to a previous patch, they do not
belong here. (Akash)
v9: Check USES_FULL_48BIT_PPGTT instead of GEN8_CTX_ADDRESSING_MODE in
gen8_emit_bb_start to check if emit pdps is needed. (Akash)

Cc: Akash Goel <akash.goel@intel.com>
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Michel Thierry <michel.thierry@intel.com> (v2+)
Reviewed-by: Akash Goel <akash.goel@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 229a317..2ab3cef 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -650,8 +650,8 @@ static int gen8_write_pdp(struct drm_i915_gem_request *req,
 	return 0;
 }
 
-static int gen8_mm_switch(struct i915_hw_ppgtt *ppgtt,
-			  struct drm_i915_gem_request *req)
+static int gen8_legacy_mm_switch(struct i915_hw_ppgtt *ppgtt,
+				 struct drm_i915_gem_request *req)
 {
 	int i, ret;
 
@@ -666,6 +666,12 @@ static int gen8_mm_switch(struct i915_hw_ppgtt *ppgtt,
 	return 0;
 }
 
+static int gen8_48b_mm_switch(struct i915_hw_ppgtt *ppgtt,
+			      struct drm_i915_gem_request *req)
+{
+	return gen8_write_pdp(req, 0, px_dma(&ppgtt->pml4));
+}
+
 static void gen8_ppgtt_clear_pte_range(struct i915_address_space *vm,
 				       struct i915_page_directory_pointer *pdp,
 				       uint64_t start,
@@ -1318,14 +1324,13 @@ static int gen8_ppgtt_init(struct i915_hw_ppgtt *ppgtt)
 	ppgtt->base.unbind_vma = ppgtt_unbind_vma;
 	ppgtt->base.bind_vma = ppgtt_bind_vma;
 
-	ppgtt->switch_mm = gen8_mm_switch;
-
 	if (USES_FULL_48BIT_PPGTT(ppgtt->base.dev)) {
 		ret = setup_px(ppgtt->base.dev, &ppgtt->pml4);
 		if (ret)
 			goto free_scratch;
 
 		ppgtt->base.total = 1ULL << 48;
+		ppgtt->switch_mm = gen8_48b_mm_switch;
 	} else {
 		ret = __pdp_init(false, &ppgtt->pdp);
 		if (ret)
@@ -1340,6 +1345,7 @@ static int gen8_ppgtt_init(struct i915_hw_ppgtt *ppgtt)
 			 */
 			ppgtt->base.total = to_i915(ppgtt->base.dev)->gtt.base.total;
 
+		ppgtt->switch_mm = gen8_legacy_mm_switch;
 		trace_i915_page_directory_pointer_entry_alloc(&ppgtt->base,
 							      0, 0,
 							      GEN8_PML4E_SHIFT);
@@ -1537,8 +1543,9 @@ static void gen8_ppgtt_enable(struct drm_device *dev)
 	int j;
 
 	for_each_ring(ring, dev_priv, j) {
+		u32 four_level = USES_FULL_48BIT_PPGTT(dev) ? GEN8_GFX_PPGTT_48B : 0;
 		I915_WRITE(RING_MODE_GEN7(ring),
-			   _MASKED_BIT_ENABLE(GFX_PPGTT_ENABLE));
+			   _MASKED_BIT_ENABLE(GFX_PPGTT_ENABLE | four_level));
 	}
 }
 
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 8e46c34..88f20cc 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -1679,6 +1679,7 @@ enum skl_disp_power_wells {
 #define   GFX_REPLAY_MODE		(1<<11)
 #define   GFX_PSMI_GRANULARITY		(1<<10)
 #define   GFX_PPGTT_ENABLE		(1<<9)
+#define   GEN8_GFX_PPGTT_48B		(1<<7)
 
 #define VLV_DISPLAY_BASE 0x180000
 #define VLV_MIPI_BASE VLV_DISPLAY_BASE
diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
index 5bc0ce1..138964a 100644
--- a/drivers/gpu/drm/i915/intel_lrc.c
+++ b/drivers/gpu/drm/i915/intel_lrc.c
@@ -196,13 +196,21 @@
 	reg_state[CTX_PDP ## n ## _LDW+1] = lower_32_bits(_addr); \
 }
 
+#define ASSIGN_CTX_PML4(ppgtt, reg_state) { \
+	reg_state[CTX_PDP0_UDW + 1] = upper_32_bits(px_dma(&ppgtt->pml4)); \
+	reg_state[CTX_PDP0_LDW + 1] = lower_32_bits(px_dma(&ppgtt->pml4)); \
+}
+
 enum {
 	ADVANCED_CONTEXT = 0,
-	LEGACY_CONTEXT,
+	LEGACY_32B_CONTEXT,
 	ADVANCED_AD_CONTEXT,
 	LEGACY_64B_CONTEXT
 };
-#define GEN8_CTX_MODE_SHIFT 3
+#define GEN8_CTX_ADDRESSING_MODE_SHIFT 3
+#define GEN8_CTX_ADDRESSING_MODE(dev)  (USES_FULL_48BIT_PPGTT(dev) ?\
+		LEGACY_64B_CONTEXT :\
+		LEGACY_32B_CONTEXT)
 enum {
 	FAULT_AND_HANG = 0,
 	FAULT_AND_HALT, /* Debug only */
@@ -273,7 +281,7 @@ static uint64_t execlists_ctx_descriptor(struct drm_i915_gem_request *rq)
 	WARN_ON(lrca & 0xFFFFFFFF00000FFFULL);
 
 	desc = GEN8_CTX_VALID;
-	desc |= LEGACY_CONTEXT << GEN8_CTX_MODE_SHIFT;
+	desc |= GEN8_CTX_ADDRESSING_MODE(dev) << GEN8_CTX_ADDRESSING_MODE_SHIFT;
 	if (IS_GEN8(ctx_obj->base.dev))
 		desc |= GEN8_CTX_L3LLC_COHERENT;
 	desc |= GEN8_CTX_PRIVILEGE;
@@ -348,10 +356,12 @@ static int execlists_update_context(struct drm_i915_gem_request *rq)
 	reg_state[CTX_RING_TAIL+1] = rq->tail;
 	reg_state[CTX_RING_BUFFER_START+1] = i915_gem_obj_ggtt_offset(rb_obj);
 
-	/* True PPGTT with dynamic page allocation: update PDP registers and
-	 * point the unallocated PDPs to the scratch page
-	 */
-	if (ppgtt) {
+	if (ppgtt && !USES_FULL_48BIT_PPGTT(ppgtt->base.dev)) {
+		/* True 32b PPGTT with dynamic page allocation: update PDP
+		 * registers and point the unallocated PDPs to scratch page.
+		 * PML4 is allocated during ppgtt init, so this is not needed
+		 * in 48-bit mode.
+		 */
 		ASSIGN_CTX_PDP(ppgtt, reg_state, 3);
 		ASSIGN_CTX_PDP(ppgtt, reg_state, 2);
 		ASSIGN_CTX_PDP(ppgtt, reg_state, 1);
@@ -1515,12 +1525,15 @@ static int gen8_emit_bb_start(struct drm_i915_gem_request *req,
 	 * Ideally, we should set Force PD Restore in ctx descriptor,
 	 * but we can't. Force Restore would be a second option, but
 	 * it is unsafe in case of lite-restore (because the ctx is
-	 * not idle). */
+	 * not idle). PML4 is allocated during ppgtt init so this is
+	 * not needed in 48-bit.*/
 	if (req->ctx->ppgtt &&
 	    (intel_ring_flag(req->ring) & req->ctx->ppgtt->pd_dirty_rings)) {
-		ret = intel_logical_ring_emit_pdps(req);
-		if (ret)
-			return ret;
+		if (!USES_FULL_48BIT_PPGTT(req->i915)) {
+			ret = intel_logical_ring_emit_pdps(req);
+			if (ret)
+				return ret;
+		}
 
 		req->ctx->ppgtt->pd_dirty_rings &= ~intel_ring_flag(req->ring);
 	}
@@ -2201,13 +2214,24 @@ populate_lr_context(struct intel_context *ctx, struct drm_i915_gem_object *ctx_o
 	reg_state[CTX_PDP0_UDW] = GEN8_RING_PDP_UDW(ring, 0);
 	reg_state[CTX_PDP0_LDW] = GEN8_RING_PDP_LDW(ring, 0);
 
-	/* With dynamic page allocation, PDPs may not be allocated at this point,
-	 * Point the unallocated PDPs to the scratch page
-	 */
-	ASSIGN_CTX_PDP(ppgtt, reg_state, 3);
-	ASSIGN_CTX_PDP(ppgtt, reg_state, 2);
-	ASSIGN_CTX_PDP(ppgtt, reg_state, 1);
-	ASSIGN_CTX_PDP(ppgtt, reg_state, 0);
+	if (USES_FULL_48BIT_PPGTT(ppgtt->base.dev)) {
+		/* 64b PPGTT (48bit canonical)
+		 * PDP0_DESCRIPTOR contains the base address to PML4 and
+		 * other PDP Descriptors are ignored.
+		 */
+		ASSIGN_CTX_PML4(ppgtt, reg_state);
+	} else {
+		/* 32b PPGTT
+		 * PDP*_DESCRIPTOR contains the base address of space supported.
+		 * With dynamic page allocation, PDPs may not be allocated at
+		 * this point. Point the unallocated PDPs to the scratch page
+		 */
+		ASSIGN_CTX_PDP(ppgtt, reg_state, 3);
+		ASSIGN_CTX_PDP(ppgtt, reg_state, 2);
+		ASSIGN_CTX_PDP(ppgtt, reg_state, 1);
+		ASSIGN_CTX_PDP(ppgtt, reg_state, 0);
+	}
+
 	if (ring->id == RCS) {
 		reg_state[CTX_LRI_HEADER_2] = MI_LOAD_REGISTER_IMM(1);
 		reg_state[CTX_R_PWR_CLK_STATE] = GEN8_R_PWR_CLK_STATE;
-- 
cgit v0.10.2


From 3387d433b0bbdda53ad51016ae9d10c6c17046a0 Mon Sep 17 00:00:00 2001
From: Michel Thierry <michel.thierry@intel.com>
Date: Mon, 3 Aug 2015 09:52:47 +0100
Subject: drm/i915/gen8: Pass sg_iter through pte inserts

As a step towards implementing 4 levels, while not discarding the
existing pte insert functions, we need to pass the sg_iter through.
The current function understands to the page directory granularity.
An object's pages may span the page directory, and so using the iter
directly as we write the PTEs allows the iterator to stay coherent
through a VMA insert operation spanning multiple page table levels.

v2: Rebase after s/page_tables/page_table/.
v3: Rebase after Mika's ppgtt cleanup / scratch merge patch series;
updated commit message (s/map/insert).
v4: Rebase.

Reviewed-by: Akash Goel <akash.goel@intel.com> (v3)
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Michel Thierry <michel.thierry@intel.com> (v2+)
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 2ab3cef..4080034 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -746,7 +746,7 @@ static void gen8_ppgtt_clear_range(struct i915_address_space *vm,
 static void
 gen8_ppgtt_insert_pte_entries(struct i915_address_space *vm,
 			      struct i915_page_directory_pointer *pdp,
-			      struct sg_table *pages,
+			      struct sg_page_iter *sg_iter,
 			      uint64_t start,
 			      enum i915_cache_level cache_level)
 {
@@ -756,11 +756,10 @@ gen8_ppgtt_insert_pte_entries(struct i915_address_space *vm,
 	unsigned pdpe = start >> GEN8_PDPE_SHIFT & GEN8_PDPE_MASK;
 	unsigned pde = start >> GEN8_PDE_SHIFT & GEN8_PDE_MASK;
 	unsigned pte = start >> GEN8_PTE_SHIFT & GEN8_PTE_MASK;
-	struct sg_page_iter sg_iter;
 
 	pt_vaddr = NULL;
 
-	for_each_sg_page(pages->sgl, &sg_iter, pages->nents, 0) {
+	while (__sg_page_iter_next(sg_iter)) {
 		if (WARN_ON(pdpe >= GEN8_LEGACY_PDPES))
 			break;
 
@@ -771,7 +770,7 @@ gen8_ppgtt_insert_pte_entries(struct i915_address_space *vm,
 		}
 
 		pt_vaddr[pte] =
-			gen8_pte_encode(sg_page_iter_dma_address(&sg_iter),
+			gen8_pte_encode(sg_page_iter_dma_address(sg_iter),
 					cache_level, true);
 		if (++pte == GEN8_PTES) {
 			kunmap_px(ppgtt, pt_vaddr);
@@ -797,8 +796,10 @@ static void gen8_ppgtt_insert_entries(struct i915_address_space *vm,
 	struct i915_hw_ppgtt *ppgtt =
 		container_of(vm, struct i915_hw_ppgtt, base);
 	struct i915_page_directory_pointer *pdp = &ppgtt->pdp; /* FIXME: 48b */
+	struct sg_page_iter sg_iter;
 
-	gen8_ppgtt_insert_pte_entries(vm, pdp, pages, start, cache_level);
+	__sg_page_iter_start(&sg_iter, pages->sgl, sg_nents(pages->sgl), 0);
+	gen8_ppgtt_insert_pte_entries(vm, pdp, &sg_iter, start, cache_level);
 }
 
 static void gen8_free_page_tables(struct drm_device *dev,
-- 
cgit v0.10.2


From de5ba8eb9cefff78ac4951bd68633d2b7a0fc5c1 Mon Sep 17 00:00:00 2001
From: Michel Thierry <michel.thierry@intel.com>
Date: Mon, 3 Aug 2015 09:53:27 +0100
Subject: drm/i915/gen8: Add 4 level support in insert_entries and clear_range

When 48b is enabled, gen8_ppgtt_insert_entries needs to read the Page Map
Level 4 (PML4), before it selects which Page Directory Pointer (PDP)
it will write to.

Similarly, gen8_ppgtt_clear_range needs to get the correct PDP/PD range.

This patch was inspired by Ben's "Depend exclusively on map and
unmap_vma".

v2: Rebase after s/page_tables/page_table/.
v3: Remove unnecessary pdpe loop in gen8_ppgtt_clear_range_4lvl and use
clamp_pdp in gen8_ppgtt_insert_entries (Akash).
v4: Merge gen8_ppgtt_clear_range_4lvl into gen8_ppgtt_clear_range to
maintain symmetry with gen8_ppgtt_insert_entries (Akash).
v5: Do not mix pages and bytes in insert_entries (Akash).
v6: Prevent overflow in sg_nents << PAGE_SHIFT, when inserting 4GB at
once.
v7: Rebase after Mika's ppgtt cleanup / scratch merge patch series.
Use gen8_px_index functions, and remove unnecessary number of pages
parameter in insert_pte_entries.
v8: Change gen8_ppgtt_clear_pte_range to stop at PDP boundary, instead of
adding and extra clamp function; remove unnecessary pdp_start/pdp_len
variables (Akash).
v9: pages->orig_nents instead of sg_nents(pages->sgl) to get the
length (Akash).
v10: Remove pdp warning check ingen8_ppgtt_insert_pte_entries until this
commit (Akash).

Reviewed-by: Akash Goel <akash.goel@intel.com> (v9)
Cc: Akash Goel <akash.goel@intel.com>
Signed-off-by: Michel Thierry <michel.thierry@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 4080034..032801e 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -681,9 +681,9 @@ static void gen8_ppgtt_clear_pte_range(struct i915_address_space *vm,
 	struct i915_hw_ppgtt *ppgtt =
 		container_of(vm, struct i915_hw_ppgtt, base);
 	gen8_pte_t *pt_vaddr;
-	unsigned pdpe = start >> GEN8_PDPE_SHIFT & GEN8_PDPE_MASK;
-	unsigned pde = start >> GEN8_PDE_SHIFT & GEN8_PDE_MASK;
-	unsigned pte = start >> GEN8_PTE_SHIFT & GEN8_PTE_MASK;
+	unsigned pdpe = gen8_pdpe_index(start);
+	unsigned pde = gen8_pde_index(start);
+	unsigned pte = gen8_pte_index(start);
 	unsigned num_entries = length >> PAGE_SHIFT;
 	unsigned last_pte, i;
 
@@ -722,7 +722,8 @@ static void gen8_ppgtt_clear_pte_range(struct i915_address_space *vm,
 
 		pte = 0;
 		if (++pde == I915_PDES) {
-			pdpe++;
+			if (++pdpe == I915_PDPES_PER_PDP(vm->dev))
+				break;
 			pde = 0;
 		}
 	}
@@ -735,12 +736,21 @@ static void gen8_ppgtt_clear_range(struct i915_address_space *vm,
 {
 	struct i915_hw_ppgtt *ppgtt =
 		container_of(vm, struct i915_hw_ppgtt, base);
-	struct i915_page_directory_pointer *pdp = &ppgtt->pdp; /* FIXME: 48b */
-
 	gen8_pte_t scratch_pte = gen8_pte_encode(px_dma(vm->scratch_page),
 						 I915_CACHE_LLC, use_scratch);
 
-	gen8_ppgtt_clear_pte_range(vm, pdp, start, length, scratch_pte);
+	if (!USES_FULL_48BIT_PPGTT(vm->dev)) {
+		gen8_ppgtt_clear_pte_range(vm, &ppgtt->pdp, start, length,
+					   scratch_pte);
+	} else {
+		uint64_t templ4, pml4e;
+		struct i915_page_directory_pointer *pdp;
+
+		gen8_for_each_pml4e(pdp, &ppgtt->pml4, start, length, templ4, pml4e) {
+			gen8_ppgtt_clear_pte_range(vm, pdp, start, length,
+						   scratch_pte);
+		}
+	}
 }
 
 static void
@@ -753,16 +763,13 @@ gen8_ppgtt_insert_pte_entries(struct i915_address_space *vm,
 	struct i915_hw_ppgtt *ppgtt =
 		container_of(vm, struct i915_hw_ppgtt, base);
 	gen8_pte_t *pt_vaddr;
-	unsigned pdpe = start >> GEN8_PDPE_SHIFT & GEN8_PDPE_MASK;
-	unsigned pde = start >> GEN8_PDE_SHIFT & GEN8_PDE_MASK;
-	unsigned pte = start >> GEN8_PTE_SHIFT & GEN8_PTE_MASK;
+	unsigned pdpe = gen8_pdpe_index(start);
+	unsigned pde = gen8_pde_index(start);
+	unsigned pte = gen8_pte_index(start);
 
 	pt_vaddr = NULL;
 
 	while (__sg_page_iter_next(sg_iter)) {
-		if (WARN_ON(pdpe >= GEN8_LEGACY_PDPES))
-			break;
-
 		if (pt_vaddr == NULL) {
 			struct i915_page_directory *pd = pdp->page_directory[pdpe];
 			struct i915_page_table *pt = pd->page_table[pde];
@@ -776,7 +783,8 @@ gen8_ppgtt_insert_pte_entries(struct i915_address_space *vm,
 			kunmap_px(ppgtt, pt_vaddr);
 			pt_vaddr = NULL;
 			if (++pde == I915_PDES) {
-				pdpe++;
+				if (++pdpe == I915_PDPES_PER_PDP(vm->dev))
+					break;
 				pde = 0;
 			}
 			pte = 0;
@@ -795,11 +803,23 @@ static void gen8_ppgtt_insert_entries(struct i915_address_space *vm,
 {
 	struct i915_hw_ppgtt *ppgtt =
 		container_of(vm, struct i915_hw_ppgtt, base);
-	struct i915_page_directory_pointer *pdp = &ppgtt->pdp; /* FIXME: 48b */
 	struct sg_page_iter sg_iter;
 
 	__sg_page_iter_start(&sg_iter, pages->sgl, sg_nents(pages->sgl), 0);
-	gen8_ppgtt_insert_pte_entries(vm, pdp, &sg_iter, start, cache_level);
+
+	if (!USES_FULL_48BIT_PPGTT(vm->dev)) {
+		gen8_ppgtt_insert_pte_entries(vm, &ppgtt->pdp, &sg_iter, start,
+					      cache_level);
+	} else {
+		struct i915_page_directory_pointer *pdp;
+		uint64_t templ4, pml4e;
+		uint64_t length = (uint64_t)pages->orig_nents << PAGE_SHIFT;
+
+		gen8_for_each_pml4e(pdp, &ppgtt->pml4, start, length, templ4, pml4e) {
+			gen8_ppgtt_insert_pte_entries(vm, pdp, &sg_iter,
+						      start, cache_level);
+		}
+	}
 }
 
 static void gen8_free_page_tables(struct drm_device *dev,
-- 
cgit v0.10.2


From 69ab76fd3d497816992b22dd201d2327cb921c94 Mon Sep 17 00:00:00 2001
From: Michel Thierry <michel.thierry@intel.com>
Date: Wed, 29 Jul 2015 17:23:55 +0100
Subject: drm/i915/gen8: Initialize PDPs and PML4

Similar to PDs, while setting up a page directory pointer, make all entries
of the pdp point to the scratch pd before mapping (and make all its entries
point to the scratch page); this is to be safe in case of out of bound
access or  proactive prefetch.

Also add a scratch pdp, which the PML4 entries point to.

v2: Handle scratch_pdp allocation failure correctly, and keep
initialize_px functions together (Akash)
v3: Rebase after Mika's ppgtt cleanup / scratch merge patch series. Rely on
the added macros to initialize the pdps.
v4: Rebase after final merged version of Mika's ppgtt/scratch patches
(and removed commit message part related to v3).
v5: Update commit message to also mention PML4 table initialization and
the new scratch pdp (Akash).

Suggested-by: Akash Goel <akash.goel@intel.com>
Signed-off-by: Michel Thierry <michel.thierry@intel.com>
Reviewed-by: Akash Goel <akash.goel@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 032801e..a4bdef3 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -596,6 +596,27 @@ static void free_pdp(struct drm_device *dev,
 	}
 }
 
+static void gen8_initialize_pdp(struct i915_address_space *vm,
+				struct i915_page_directory_pointer *pdp)
+{
+	gen8_ppgtt_pdpe_t scratch_pdpe;
+
+	scratch_pdpe = gen8_pdpe_encode(px_dma(vm->scratch_pd), I915_CACHE_LLC);
+
+	fill_px(vm->dev, pdp, scratch_pdpe);
+}
+
+static void gen8_initialize_pml4(struct i915_address_space *vm,
+				 struct i915_pml4 *pml4)
+{
+	gen8_ppgtt_pml4e_t scratch_pml4e;
+
+	scratch_pml4e = gen8_pml4e_encode(px_dma(vm->scratch_pdp),
+					  I915_CACHE_LLC);
+
+	fill_px(vm->dev, pml4, scratch_pml4e);
+}
+
 static void
 gen8_setup_page_directory(struct i915_hw_ppgtt *ppgtt,
 			  struct i915_page_directory_pointer *pdp,
@@ -860,8 +881,20 @@ static int gen8_init_scratch(struct i915_address_space *vm)
 		return PTR_ERR(vm->scratch_pd);
 	}
 
+	if (USES_FULL_48BIT_PPGTT(dev)) {
+		vm->scratch_pdp = alloc_pdp(dev);
+		if (IS_ERR(vm->scratch_pdp)) {
+			free_pd(dev, vm->scratch_pd);
+			free_pt(dev, vm->scratch_pt);
+			free_scratch_page(dev, vm->scratch_page);
+			return PTR_ERR(vm->scratch_pdp);
+		}
+	}
+
 	gen8_initialize_pt(vm, vm->scratch_pt);
 	gen8_initialize_pd(vm, vm->scratch_pd);
+	if (USES_FULL_48BIT_PPGTT(dev))
+		gen8_initialize_pdp(vm, vm->scratch_pdp);
 
 	return 0;
 }
@@ -870,6 +903,8 @@ static void gen8_free_scratch(struct i915_address_space *vm)
 {
 	struct drm_device *dev = vm->dev;
 
+	if (USES_FULL_48BIT_PPGTT(dev))
+		free_pdp(dev, vm->scratch_pdp);
 	free_pd(dev, vm->scratch_pd);
 	free_pt(dev, vm->scratch_pt);
 	free_scratch_page(dev, vm->scratch_page);
@@ -1071,6 +1106,7 @@ gen8_ppgtt_alloc_page_dirpointers(struct i915_address_space *vm,
 			if (IS_ERR(pdp))
 				goto unwind_out;
 
+			gen8_initialize_pdp(vm, pdp);
 			pml4->pdps[pml4e] = pdp;
 			__set_bit(pml4e, new_pdps);
 			trace_i915_page_directory_pointer_entry_alloc(vm,
@@ -1350,6 +1386,8 @@ static int gen8_ppgtt_init(struct i915_hw_ppgtt *ppgtt)
 		if (ret)
 			goto free_scratch;
 
+		gen8_initialize_pml4(&ppgtt->base, &ppgtt->pml4);
+
 		ppgtt->base.total = 1ULL << 48;
 		ppgtt->switch_mm = gen8_48b_mm_switch;
 	} else {
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.h b/drivers/gpu/drm/i915/i915_gem_gtt.h
index 11d44b3..70c50e7 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.h
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.h
@@ -278,6 +278,7 @@ struct i915_address_space {
 	struct i915_page_scratch *scratch_page;
 	struct i915_page_table *scratch_pt;
 	struct i915_page_directory *scratch_pd;
+	struct i915_page_directory_pointer *scratch_pdp; /* GEN8+ & 48b PPGTT */
 
 	/**
 	 * List of objects currently involved in rendering.
-- 
cgit v0.10.2


From e1f123257a1f7d3af36a31a0fb2d4c6f40039fed Mon Sep 17 00:00:00 2001
From: Michel Thierry <michel.thierry@intel.com>
Date: Wed, 29 Jul 2015 17:23:56 +0100
Subject: drm/i915: Expand error state's address width to 64b

v2: For semaphore errors, object is mapped to GGTT and offset will not
be > 4GB, print only lower 32-bits (Akash)
v3: Print gtt_offset in groups of 32-bit (Chris)

Cc: Akash Goel <akash.goel@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Michel Thierry <michel.thierry@intel.com>
Reviewed-by: Akash Goel <akash.goel@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 79d2dcb..91bd778 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -547,7 +547,7 @@ struct drm_i915_error_state {
 
 		struct drm_i915_error_object {
 			int page_count;
-			u32 gtt_offset;
+			u64 gtt_offset;
 			u32 *pages[0];
 		} *ringbuffer, *batchbuffer, *wa_batchbuffer, *ctx, *hws_page;
 
@@ -573,7 +573,7 @@ struct drm_i915_error_state {
 		u32 size;
 		u32 name;
 		u32 rseqno[I915_NUM_RINGS], wseqno;
-		u32 gtt_offset;
+		u64 gtt_offset;
 		u32 read_domains;
 		u32 write_domain;
 		s32 fence_reg:I915_MAX_NUM_FENCE_BITS;
diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
index 41d0739..0d0a7b1 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.c
+++ b/drivers/gpu/drm/i915/i915_gpu_error.c
@@ -197,8 +197,9 @@ static void print_error_buffers(struct drm_i915_error_state_buf *m,
 	err_printf(m, "  %s [%d]:\n", name, count);
 
 	while (count--) {
-		err_printf(m, "    %08x %8u %02x %02x [ ",
-			   err->gtt_offset,
+		err_printf(m, "    %08x_%08x %8u %02x %02x [ ",
+			   upper_32_bits(err->gtt_offset),
+			   lower_32_bits(err->gtt_offset),
 			   err->size,
 			   err->read_domains,
 			   err->write_domain);
@@ -427,15 +428,17 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m,
 				err_printf(m, " (submitted by %s [%d])",
 					   error->ring[i].comm,
 					   error->ring[i].pid);
-			err_printf(m, " --- gtt_offset = 0x%08x\n",
-				   obj->gtt_offset);
+			err_printf(m, " --- gtt_offset = 0x%08x %08x\n",
+				   upper_32_bits(obj->gtt_offset),
+				   lower_32_bits(obj->gtt_offset));
 			print_error_obj(m, obj);
 		}
 
 		obj = error->ring[i].wa_batchbuffer;
 		if (obj) {
 			err_printf(m, "%s (w/a) --- gtt_offset = 0x%08x\n",
-				   dev_priv->ring[i].name, obj->gtt_offset);
+				   dev_priv->ring[i].name,
+				   lower_32_bits(obj->gtt_offset));
 			print_error_obj(m, obj);
 		}
 
@@ -454,14 +457,14 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m,
 		if ((obj = error->ring[i].ringbuffer)) {
 			err_printf(m, "%s --- ringbuffer = 0x%08x\n",
 				   dev_priv->ring[i].name,
-				   obj->gtt_offset);
+				   lower_32_bits(obj->gtt_offset));
 			print_error_obj(m, obj);
 		}
 
 		if ((obj = error->ring[i].hws_page)) {
 			err_printf(m, "%s --- HW Status = 0x%08x\n",
 				   dev_priv->ring[i].name,
-				   obj->gtt_offset);
+				   lower_32_bits(obj->gtt_offset));
 			offset = 0;
 			for (elt = 0; elt < PAGE_SIZE/16; elt += 4) {
 				err_printf(m, "[%04x] %08x %08x %08x %08x\n",
@@ -477,13 +480,14 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m,
 		if ((obj = error->ring[i].ctx)) {
 			err_printf(m, "%s --- HW Context = 0x%08x\n",
 				   dev_priv->ring[i].name,
-				   obj->gtt_offset);
+				   lower_32_bits(obj->gtt_offset));
 			print_error_obj(m, obj);
 		}
 	}
 
 	if ((obj = error->semaphore_obj)) {
-		err_printf(m, "Semaphore page = 0x%08x\n", obj->gtt_offset);
+		err_printf(m, "Semaphore page = 0x%08x\n",
+			   lower_32_bits(obj->gtt_offset));
 		for (elt = 0; elt < PAGE_SIZE/16; elt += 4) {
 			err_printf(m, "[%04x] %08x %08x %08x %08x\n",
 				   elt * 4,
@@ -591,7 +595,7 @@ i915_error_object_create(struct drm_i915_private *dev_priv,
 	int num_pages;
 	bool use_ggtt;
 	int i = 0;
-	u32 reloc_offset;
+	u64 reloc_offset;
 
 	if (src == NULL || src->pages == NULL)
 		return NULL;
-- 
cgit v0.10.2


From ea91e401501a72597554cb5c080fe5671a6ccb84 Mon Sep 17 00:00:00 2001
From: Michel Thierry <michel.thierry@intel.com>
Date: Wed, 29 Jul 2015 17:23:57 +0100
Subject: drm/i915/gen8: Add ppgtt info and debug_dump

v2: Clean up patch after rebases.
v3: gen8_dump_ppgtt for 32b and 48b PPGTT.
v4: Use used_pml4es/pdpes (Akash).
v5: Rebase after Mika's ppgtt cleanup / scratch merge patch series.
v6: Rely on used_px bits instead of null checking (Akash)

Cc: Akash Goel <akash.goel@intel.com>
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Michel Thierry <michel.thierry@intel.com> (v2+)
Reviewed-by: Akash Goel <akash.goel@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index d1c643a..95e7b82 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -2250,7 +2250,6 @@ static void gen6_ppgtt_info(struct seq_file *m, struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_engine_cs *ring;
-	struct drm_file *file;
 	int i;
 
 	if (INTEL_INFO(dev)->gen == 6)
@@ -2273,13 +2272,6 @@ static void gen6_ppgtt_info(struct seq_file *m, struct drm_device *dev)
 		ppgtt->debug_dump(ppgtt, m);
 	}
 
-	list_for_each_entry_reverse(file, &dev->filelist, lhead) {
-		struct drm_i915_file_private *file_priv = file->driver_priv;
-
-		seq_printf(m, "proc: %s\n",
-			   get_pid_task(file->pid, PIDTYPE_PID)->comm);
-		idr_for_each(&file_priv->context_idr, per_file_ctx, m);
-	}
 	seq_printf(m, "ECOCHK: 0x%08x\n", I915_READ(GAM_ECOCHK));
 }
 
@@ -2288,6 +2280,7 @@ static int i915_ppgtt_info(struct seq_file *m, void *data)
 	struct drm_info_node *node = m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_file *file;
 
 	int ret = mutex_lock_interruptible(&dev->struct_mutex);
 	if (ret)
@@ -2299,6 +2292,15 @@ static int i915_ppgtt_info(struct seq_file *m, void *data)
 	else if (INTEL_INFO(dev)->gen >= 6)
 		gen6_ppgtt_info(m, dev);
 
+	list_for_each_entry_reverse(file, &dev->filelist, lhead) {
+		struct drm_i915_file_private *file_priv = file->driver_priv;
+
+		seq_printf(m, "\nproc: %s\n",
+			   get_pid_task(file->pid, PIDTYPE_PID)->comm);
+		idr_for_each(&file_priv->context_idr, per_file_ctx,
+			     (void *)(unsigned long)m);
+	}
+
 	intel_runtime_pm_put(dev_priv);
 	mutex_unlock(&dev->struct_mutex);
 
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index a4bdef3..38439fa 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -1358,6 +1358,89 @@ static int gen8_alloc_va_range(struct i915_address_space *vm,
 		return gen8_alloc_va_range_3lvl(vm, &ppgtt->pdp, start, length);
 }
 
+static void gen8_dump_pdp(struct i915_page_directory_pointer *pdp,
+			  uint64_t start, uint64_t length,
+			  gen8_pte_t scratch_pte,
+			  struct seq_file *m)
+{
+	struct i915_page_directory *pd;
+	uint64_t temp;
+	uint32_t pdpe;
+
+	gen8_for_each_pdpe(pd, pdp, start, length, temp, pdpe) {
+		struct i915_page_table *pt;
+		uint64_t pd_len = length;
+		uint64_t pd_start = start;
+		uint32_t pde;
+
+		if (!test_bit(pdpe, pdp->used_pdpes))
+			continue;
+
+		seq_printf(m, "\tPDPE #%d\n", pdpe);
+		gen8_for_each_pde(pt, pd, pd_start, pd_len, temp, pde) {
+			uint32_t  pte;
+			gen8_pte_t *pt_vaddr;
+
+			if (!test_bit(pde, pd->used_pdes))
+				continue;
+
+			pt_vaddr = kmap_px(pt);
+			for (pte = 0; pte < GEN8_PTES; pte += 4) {
+				uint64_t va =
+					(pdpe << GEN8_PDPE_SHIFT) |
+					(pde << GEN8_PDE_SHIFT) |
+					(pte << GEN8_PTE_SHIFT);
+				int i;
+				bool found = false;
+
+				for (i = 0; i < 4; i++)
+					if (pt_vaddr[pte + i] != scratch_pte)
+						found = true;
+				if (!found)
+					continue;
+
+				seq_printf(m, "\t\t0x%llx [%03d,%03d,%04d]: =", va, pdpe, pde, pte);
+				for (i = 0; i < 4; i++) {
+					if (pt_vaddr[pte + i] != scratch_pte)
+						seq_printf(m, " %llx", pt_vaddr[pte + i]);
+					else
+						seq_puts(m, "  SCRATCH ");
+				}
+				seq_puts(m, "\n");
+			}
+			/* don't use kunmap_px, it could trigger
+			 * an unnecessary flush.
+			 */
+			kunmap_atomic(pt_vaddr);
+		}
+	}
+}
+
+static void gen8_dump_ppgtt(struct i915_hw_ppgtt *ppgtt, struct seq_file *m)
+{
+	struct i915_address_space *vm = &ppgtt->base;
+	uint64_t start = ppgtt->base.start;
+	uint64_t length = ppgtt->base.total;
+	gen8_pte_t scratch_pte = gen8_pte_encode(px_dma(vm->scratch_page),
+						 I915_CACHE_LLC, true);
+
+	if (!USES_FULL_48BIT_PPGTT(vm->dev)) {
+		gen8_dump_pdp(&ppgtt->pdp, start, length, scratch_pte, m);
+	} else {
+		uint64_t templ4, pml4e;
+		struct i915_pml4 *pml4 = &ppgtt->pml4;
+		struct i915_page_directory_pointer *pdp;
+
+		gen8_for_each_pml4e(pdp, pml4, start, length, templ4, pml4e) {
+			if (!test_bit(pml4e, pml4->used_pml4es))
+				continue;
+
+			seq_printf(m, "    PML4E #%llu\n", pml4e);
+			gen8_dump_pdp(pdp, start, length, scratch_pte, m);
+		}
+	}
+}
+
 /*
  * GEN8 legacy ppgtt programming is accomplished through a max 4 PDP registers
  * with a net effect resembling a 2-level page table in normal x86 terms. Each
@@ -1380,6 +1463,7 @@ static int gen8_ppgtt_init(struct i915_hw_ppgtt *ppgtt)
 	ppgtt->base.clear_range = gen8_ppgtt_clear_range;
 	ppgtt->base.unbind_vma = ppgtt_unbind_vma;
 	ppgtt->base.bind_vma = ppgtt_bind_vma;
+	ppgtt->debug_dump = gen8_dump_ppgtt;
 
 	if (USES_FULL_48BIT_PPGTT(ppgtt->base.dev)) {
 		ret = setup_px(ppgtt->base.dev, &ppgtt->pml4);
-- 
cgit v0.10.2


From 65bd342ff2f1f9f3ca4a17691a825935d2aa13c9 Mon Sep 17 00:00:00 2001
From: Michel Thierry <michel.thierry@intel.com>
Date: Wed, 29 Jul 2015 17:23:58 +0100
Subject: drm/i915: object size needs to be u64

In a 48b world, users can try to allocate buffers bigger than 4GB; in
these cases it is important that size is a 64b variable.

v2: Drop the warning about bind with size 0, it shouldn't happen anyway.
Signed-off-by: Michel Thierry <michel.thierry@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 84f91bc..73293b4 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -3355,7 +3355,8 @@ i915_gem_object_bind_to_vm(struct drm_i915_gem_object *obj,
 {
 	struct drm_device *dev = obj->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 size, fence_size, fence_alignment, unfenced_alignment;
+	u32 fence_alignment, unfenced_alignment;
+	u64 size, fence_size;
 	u64 start =
 		flags & PIN_OFFSET_BIAS ? flags & PIN_OFFSET_MASK : 0;
 	u64 end =
@@ -3414,7 +3415,7 @@ i915_gem_object_bind_to_vm(struct drm_i915_gem_object *obj,
 	 * attempt to find space.
 	 */
 	if (size > end) {
-		DRM_DEBUG("Attempting to bind an object (view type=%u) larger than the aperture: size=%u > %s aperture=%llu\n",
+		DRM_DEBUG("Attempting to bind an object (view type=%u) larger than the aperture: size=%llu > %s aperture=%llu\n",
 			  ggtt_view ? ggtt_view->type : 0,
 			  size,
 			  flags & PIN_MAPPABLE ? "mappable" : "total",
-- 
cgit v0.10.2


From af98714e5d05fcbf7638ea4b53d158d96d693bf2 Mon Sep 17 00:00:00 2001
From: Michel Thierry <michel.thierry@intel.com>
Date: Wed, 29 Jul 2015 17:23:59 +0100
Subject: drm/i915: batch_obj vm offset must be u64

Otherwise it can overflow in 48-bit mode, and cause an incorrect
exec_start.

Before commit 5f19e2bffa63a91cd4ac1adcec648e14a44277ce ("drm/i915: Merged
the many do_execbuf() parameters into a structure"), it was already an u64.

Signed-off-by: Michel Thierry <michel.thierry@intel.com>
Reviewed-by: Akash Goel <akash.goel@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 91bd778..cefb6e9 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -1686,7 +1686,7 @@ struct i915_execbuffer_params {
 	struct drm_file                 *file;
 	uint32_t                        dispatch_flags;
 	uint32_t                        args_batch_start_offset;
-	uint32_t                        batch_obj_vm_offset;
+	uint64_t                        batch_obj_vm_offset;
 	struct intel_engine_cs          *ring;
 	struct drm_i915_gem_object      *batch_obj;
 	struct intel_context            *ctx;
-- 
cgit v0.10.2


From c6d576cc576196593671eabc8aad6c1744b32568 Mon Sep 17 00:00:00 2001
From: Michel Thierry <michel.thierry@intel.com>
Date: Wed, 29 Jul 2015 17:24:00 +0100
Subject: drm/i915/userptr: Kill user_size limit check

GTT was only 32b and its max value is 4GB. In order to allow objects
bigger than 4GB in 48b PPGTT, i915_gem_userptr_ioctl we could check
against max 48b range (1ULL << 48).

But since the check no longer applies, just kill the limit.

v2: Use the default ctx to infer the ppgtt max size (Akash).
v3: Just kill the limit, it was only there for early detection of an
error when used for execbuffer (Chris).

Cc: Akash Goel <akash.goel@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Michel Thierry <michel.thierry@intel.com>
Reviewed-by: Akash Goel <akash.goel@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_userptr.c b/drivers/gpu/drm/i915/i915_gem_userptr.c
index 8fd431b..d11901d 100644
--- a/drivers/gpu/drm/i915/i915_gem_userptr.c
+++ b/drivers/gpu/drm/i915/i915_gem_userptr.c
@@ -813,7 +813,6 @@ static const struct drm_i915_gem_object_ops i915_gem_userptr_ops = {
 int
 i915_gem_userptr_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_gem_userptr *args = data;
 	struct drm_i915_gem_object *obj;
 	int ret;
@@ -826,9 +825,6 @@ i915_gem_userptr_ioctl(struct drm_device *dev, void *data, struct drm_file *file
 	if (offset_in_page(args->user_ptr | args->user_size))
 		return -EINVAL;
 
-	if (args->user_size > dev_priv->gtt.base.total)
-		return -E2BIG;
-
 	if (!access_ok(args->flags & I915_USERPTR_READ_ONLY ? VERIFY_READ : VERIFY_WRITE,
 		       (char __user *)(unsigned long)args->user_ptr, args->user_size))
 		return -EFAULT;
-- 
cgit v0.10.2


From e5a1cab5e57d7bd6ef77502f950cfab6e14d92a8 Mon Sep 17 00:00:00 2001
From: Rodrigo Vivi <rodrigo.vivi@intel.com>
Date: Thu, 23 Jul 2015 16:35:48 -0700
Subject: drm/i915: Force sink crc stop before start.

By Vesa DP spec, test counter at DP_TEST_SINK_MISC just reset to 0
when unsetting DP_TEST_SINK_START, so let's force this stop here.

But let's minimize the aux transactions and just do it when we know
it hasn't been properly stoped.

Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Rafael Antognolli <rafael.antognolli@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 016e7bc..1f7cc40 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -3995,22 +3995,30 @@ intel_dp_probe_mst(struct intel_dp *intel_dp)
 	return intel_dp->is_mst;
 }
 
-static void intel_dp_sink_crc_stop(struct intel_dp *intel_dp)
+static int intel_dp_sink_crc_stop(struct intel_dp *intel_dp)
 {
 	struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp);
 	struct intel_crtc *intel_crtc = to_intel_crtc(dig_port->base.base.crtc);
 	u8 buf;
+	int ret = 0;
 
 	if (drm_dp_dpcd_readb(&intel_dp->aux, DP_TEST_SINK, &buf) < 0) {
 		DRM_DEBUG_KMS("Sink CRC couldn't be stopped properly\n");
-		return;
+		ret = -EIO;
+		goto out;
 	}
 
 	if (drm_dp_dpcd_writeb(&intel_dp->aux, DP_TEST_SINK,
-			       buf & ~DP_TEST_SINK_START) < 0)
+			       buf & ~DP_TEST_SINK_START) < 0) {
 		DRM_DEBUG_KMS("Sink CRC couldn't be stopped properly\n");
+		ret = -EIO;
+		goto out;
+	}
 
+	intel_dp->sink_crc_started = false;
+ out:
 	hsw_enable_ips(intel_crtc);
+	return ret;
 }
 
 static int intel_dp_sink_crc_start(struct intel_dp *intel_dp)
@@ -4018,6 +4026,13 @@ static int intel_dp_sink_crc_start(struct intel_dp *intel_dp)
 	struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp);
 	struct intel_crtc *intel_crtc = to_intel_crtc(dig_port->base.base.crtc);
 	u8 buf;
+	int ret;
+
+	if (intel_dp->sink_crc_started) {
+		ret = intel_dp_sink_crc_stop(intel_dp);
+		if (ret)
+			return ret;
+	}
 
 	if (drm_dp_dpcd_readb(&intel_dp->aux, DP_TEST_SINK_MISC, &buf) < 0)
 		return -EIO;
@@ -4036,6 +4051,7 @@ static int intel_dp_sink_crc_start(struct intel_dp *intel_dp)
 		return -EIO;
 	}
 
+	intel_dp->sink_crc_started = true;
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index e31b95b..8858bb1 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -713,6 +713,7 @@ struct intel_dp {
 	/* sink rates as reported by DP_SUPPORTED_LINK_RATES */
 	uint8_t num_sink_rates;
 	int sink_rates[DP_MAX_SUPPORTED_RATES];
+	bool sink_crc_started;
 	struct drm_dp_aux aux;
 	uint8_t train_set[4];
 	int panel_power_up_delay;
-- 
cgit v0.10.2


From 621d4c76fd4ce7f648fa59a26e3eb6b2346cf873 Mon Sep 17 00:00:00 2001
From: Rodrigo Vivi <rodrigo.vivi@intel.com>
Date: Thu, 23 Jul 2015 16:35:49 -0700
Subject: drm/i915: Save latest known sink CRC to compensate delayed counter
 reset.

By Vesa DP 1.2 Spec TEST_CRC_COUNT should be
"reset to 0 when TEST_SINK bit 0 = 0."

However for some strange reason when PSR is enabled in
certain platforms this is not true. At least not immediatelly.

So we face cases like this:

first get_sink_crc operation:
	     count: 0, crc: 000000000000
	     count: 1, crc: c101c101c101
returned expected crc: c101c101c101

secont get_sink_crc operation:
	     count: 1, crc: c101c101c101
	     count: 0, crc: 000000000000
	     count: 1, crc: 0000c1010000
should return expected crc: 0000c1010000

But also the reset to 0 should be faster resulting into:

get_sink_crc operation:
	     count: 1, crc: c101c101c101
	     count: 1, crc: 0000c1010000
should return expected crc: 0000c1010000

So in order to know that the second one is valid one
we need to compare the pair (count, crc) with latest (count, crc).

If the pair changed you have your valid CRC.

Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Rafael Antognolli <rafael.antognolli@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 1f7cc40..abcc2a3 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -4015,7 +4015,7 @@ static int intel_dp_sink_crc_stop(struct intel_dp *intel_dp)
 		goto out;
 	}
 
-	intel_dp->sink_crc_started = false;
+	intel_dp->sink_crc.started = false;
  out:
 	hsw_enable_ips(intel_crtc);
 	return ret;
@@ -4028,7 +4028,7 @@ static int intel_dp_sink_crc_start(struct intel_dp *intel_dp)
 	u8 buf;
 	int ret;
 
-	if (intel_dp->sink_crc_started) {
+	if (intel_dp->sink_crc.started) {
 		ret = intel_dp_sink_crc_stop(intel_dp);
 		if (ret)
 			return ret;
@@ -4040,6 +4040,8 @@ static int intel_dp_sink_crc_start(struct intel_dp *intel_dp)
 	if (!(buf & DP_TEST_CRC_SUPPORTED))
 		return -ENOTTY;
 
+	intel_dp->sink_crc.last_count = buf & DP_TEST_COUNT_MASK;
+
 	if (drm_dp_dpcd_readb(&intel_dp->aux, DP_TEST_SINK, &buf) < 0)
 		return -EIO;
 
@@ -4051,7 +4053,7 @@ static int intel_dp_sink_crc_start(struct intel_dp *intel_dp)
 		return -EIO;
 	}
 
-	intel_dp->sink_crc_started = true;
+	intel_dp->sink_crc.started = true;
 	return 0;
 }
 
@@ -4061,29 +4063,39 @@ int intel_dp_sink_crc(struct intel_dp *intel_dp, u8 *crc)
 	struct drm_device *dev = dig_port->base.base.dev;
 	struct intel_crtc *intel_crtc = to_intel_crtc(dig_port->base.base.crtc);
 	u8 buf;
-	int test_crc_count;
+	int count, ret;
 	int attempts = 6;
-	int ret;
 
 	ret = intel_dp_sink_crc_start(intel_dp);
 	if (ret)
 		return ret;
 
-	if (drm_dp_dpcd_readb(&intel_dp->aux, DP_TEST_SINK_MISC, &buf) < 0) {
-		ret = -EIO;
-		goto stop;
-	}
-
-	test_crc_count = buf & DP_TEST_COUNT_MASK;
-
 	do {
+		intel_wait_for_vblank(dev, intel_crtc->pipe);
+
 		if (drm_dp_dpcd_readb(&intel_dp->aux,
 				      DP_TEST_SINK_MISC, &buf) < 0) {
 			ret = -EIO;
 			goto stop;
 		}
-		intel_wait_for_vblank(dev, intel_crtc->pipe);
-	} while (--attempts && (buf & DP_TEST_COUNT_MASK) == test_crc_count);
+		count = buf & DP_TEST_COUNT_MASK;
+		/*
+		 * Count might be reset during the loop. In this case
+		 * last known count needs to be reset as well.
+		 */
+		if (count == 0)
+			intel_dp->sink_crc.last_count = 0;
+
+		if (drm_dp_dpcd_read(&intel_dp->aux, DP_TEST_CRC_R_CR, crc, 6) < 0) {
+			ret = -EIO;
+			goto stop;
+		}
+	} while (--attempts && (count == 0 || (count == intel_dp->sink_crc.last_count &&
+					       !memcmp(intel_dp->sink_crc.last_crc, crc,
+						       6 * sizeof(u8)))));
+
+	intel_dp->sink_crc.last_count = buf & DP_TEST_COUNT_MASK;
+	memcpy(intel_dp->sink_crc.last_crc, crc, 6 * sizeof(u8));
 
 	if (attempts == 0) {
 		DRM_DEBUG_KMS("Panel is unable to calculate CRC after 6 vblanks\n");
@@ -4091,8 +4103,6 @@ int intel_dp_sink_crc(struct intel_dp *intel_dp, u8 *crc)
 		goto stop;
 	}
 
-	if (drm_dp_dpcd_read(&intel_dp->aux, DP_TEST_CRC_R_CR, crc, 6) < 0)
-		ret = -EIO;
 stop:
 	intel_dp_sink_crc_stop(intel_dp);
 	return ret;
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 8858bb1..6b3e2c6 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -696,6 +696,12 @@ enum link_m_n_set {
 	M2_N2
 };
 
+struct sink_crc {
+	bool started;
+	u8 last_crc[6];
+	int last_count;
+};
+
 struct intel_dp {
 	uint32_t output_reg;
 	uint32_t aux_ch_ctl_reg;
@@ -713,7 +719,7 @@ struct intel_dp {
 	/* sink rates as reported by DP_SUPPORTED_LINK_RATES */
 	uint8_t num_sink_rates;
 	int sink_rates[DP_MAX_SUPPORTED_RATES];
-	bool sink_crc_started;
+	struct sink_crc sink_crc;
 	struct drm_dp_aux aux;
 	uint8_t train_set[4];
 	int panel_power_up_delay;
-- 
cgit v0.10.2


From aabc95dcf200908993573b2019ff5210121bcba9 Mon Sep 17 00:00:00 2001
From: Rodrigo Vivi <rodrigo.vivi@intel.com>
Date: Thu, 23 Jul 2015 16:35:50 -0700
Subject: drm/i915: Dont -ETIMEDOUT on identical new and previous (count, crc).

By Vesa DP 1.2 spec TEST_CRC_COUNT is a "4 bit wrap counter which
increments each time the TEST_CRC_x_x are updated."

However if we are trying to verify the screen hasn't changed we get
same (count, crc) pair twice. Without this patch we would return
-ETIMEOUT in this case.

So, if in 6 vblanks the pair (count, crc) hasn't changed we
return it anyway instead of returning error and let test case decide
if it was right or not.

Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Rafael Antognolli <rafael.antognolli@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index abcc2a3..54e9b53 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -4065,6 +4065,7 @@ int intel_dp_sink_crc(struct intel_dp *intel_dp, u8 *crc)
 	u8 buf;
 	int count, ret;
 	int attempts = 6;
+	bool old_equal_new;
 
 	ret = intel_dp_sink_crc_start(intel_dp);
 	if (ret)
@@ -4079,6 +4080,7 @@ int intel_dp_sink_crc(struct intel_dp *intel_dp, u8 *crc)
 			goto stop;
 		}
 		count = buf & DP_TEST_COUNT_MASK;
+
 		/*
 		 * Count might be reset during the loop. In this case
 		 * last known count needs to be reset as well.
@@ -4090,17 +4092,24 @@ int intel_dp_sink_crc(struct intel_dp *intel_dp, u8 *crc)
 			ret = -EIO;
 			goto stop;
 		}
-	} while (--attempts && (count == 0 || (count == intel_dp->sink_crc.last_count &&
-					       !memcmp(intel_dp->sink_crc.last_crc, crc,
-						       6 * sizeof(u8)))));
+
+		old_equal_new = (count == intel_dp->sink_crc.last_count &&
+				 !memcmp(intel_dp->sink_crc.last_crc, crc,
+					 6 * sizeof(u8)));
+
+	} while (--attempts && (count == 0 || old_equal_new));
 
 	intel_dp->sink_crc.last_count = buf & DP_TEST_COUNT_MASK;
 	memcpy(intel_dp->sink_crc.last_crc, crc, 6 * sizeof(u8));
 
 	if (attempts == 0) {
-		DRM_DEBUG_KMS("Panel is unable to calculate CRC after 6 vblanks\n");
-		ret = -ETIMEDOUT;
-		goto stop;
+		if (old_equal_new) {
+			DRM_DEBUG_KMS("Unreliable Sink CRC counter: Current returned CRC is identical to the previous one\n");
+		} else {
+			DRM_ERROR("Panel is unable to calculate any CRC after 6 vblanks\n");
+			ret = -ETIMEDOUT;
+			goto stop;
+		}
 	}
 
 stop:
-- 
cgit v0.10.2


From 088e0df4020e30c4952f29dc672ceb4742b98e73 Mon Sep 17 00:00:00 2001
From: Michel Thierry <michel.thierry@intel.com>
Date: Fri, 7 Aug 2015 17:40:17 +0100
Subject: drm/i915/gtt: Allow >= 4GB offsets in X86_32

Similar to commit c44ef60e437019b8ca1dab8b4d2e8761fd4ce1e9 ("drm/i915/gtt:
Allow >= 4GB sizes for vm"), i915_gem_obj_offset and i915_gem_obj_ggtt_offset
return an unsigned long, which in only 4-bytes long in 32-bit kernels.

Change return type (and other related offset variables) to u64.

Since Global GTT is always limited to 4GB, this change would not be required
in i915_gem_obj_ggtt_offset, but this is done for consistency.

v2: Remove unnecessary offset variable in do_pin, as we already have
    vma->node.start (Chris).
    Update GGTT offset too (Tvrtko).

Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Signed-off-by: Michel Thierry <michel.thierry@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index cefb6e9..6b85338 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -2977,13 +2977,11 @@ struct drm_gem_object *i915_gem_prime_import(struct drm_device *dev,
 struct dma_buf *i915_gem_prime_export(struct drm_device *dev,
 				struct drm_gem_object *gem_obj, int flags);
 
-unsigned long
-i915_gem_obj_ggtt_offset_view(struct drm_i915_gem_object *o,
-			      const struct i915_ggtt_view *view);
-unsigned long
-i915_gem_obj_offset(struct drm_i915_gem_object *o,
-		    struct i915_address_space *vm);
-static inline unsigned long
+u64 i915_gem_obj_ggtt_offset_view(struct drm_i915_gem_object *o,
+				  const struct i915_ggtt_view *view);
+u64 i915_gem_obj_offset(struct drm_i915_gem_object *o,
+			struct i915_address_space *vm);
+static inline u64
 i915_gem_obj_ggtt_offset(struct drm_i915_gem_object *o)
 {
 	return i915_gem_obj_ggtt_offset_view(o, &i915_ggtt_view_normal);
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 73293b4..6d0f834 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -4012,15 +4012,13 @@ i915_gem_object_do_pin(struct drm_i915_gem_object *obj,
 			return -EBUSY;
 
 		if (i915_vma_misplaced(vma, alignment, flags)) {
-			unsigned long offset;
-			offset = ggtt_view ? i915_gem_obj_ggtt_offset_view(obj, ggtt_view) :
-					     i915_gem_obj_offset(obj, vm);
 			WARN(vma->pin_count,
 			     "bo is already pinned in %s with incorrect alignment:"
-			     " offset=%lx, req.alignment=%x, req.map_and_fenceable=%d,"
+			     " offset=%08x %08x, req.alignment=%x, req.map_and_fenceable=%d,"
 			     " obj->map_and_fenceable=%d\n",
 			     ggtt_view ? "ggtt" : "ppgtt",
-			     offset,
+			     upper_32_bits(vma->node.start),
+			     lower_32_bits(vma->node.start),
 			     alignment,
 			     !!(flags & PIN_MAPPABLE),
 			     obj->map_and_fenceable);
@@ -4975,9 +4973,8 @@ void i915_gem_track_fb(struct drm_i915_gem_object *old,
 }
 
 /* All the new VM stuff */
-unsigned long
-i915_gem_obj_offset(struct drm_i915_gem_object *o,
-		    struct i915_address_space *vm)
+u64 i915_gem_obj_offset(struct drm_i915_gem_object *o,
+			struct i915_address_space *vm)
 {
 	struct drm_i915_private *dev_priv = o->base.dev->dev_private;
 	struct i915_vma *vma;
@@ -4997,9 +4994,8 @@ i915_gem_obj_offset(struct drm_i915_gem_object *o,
 	return -1;
 }
 
-unsigned long
-i915_gem_obj_ggtt_offset_view(struct drm_i915_gem_object *o,
-			      const struct i915_ggtt_view *view)
+u64 i915_gem_obj_ggtt_offset_view(struct drm_i915_gem_object *o,
+				  const struct i915_ggtt_view *view)
 {
 	struct i915_address_space *ggtt = i915_obj_to_ggtt(o);
 	struct i915_vma *vma;
diff --git a/drivers/gpu/drm/i915/i915_gem_fence.c b/drivers/gpu/drm/i915/i915_gem_fence.c
index af1f8c4..6077dff 100644
--- a/drivers/gpu/drm/i915/i915_gem_fence.c
+++ b/drivers/gpu/drm/i915/i915_gem_fence.c
@@ -128,7 +128,7 @@ static void i915_write_fence_reg(struct drm_device *dev, int reg,
 		WARN((i915_gem_obj_ggtt_offset(obj) & ~I915_FENCE_START_MASK) ||
 		     (size & -size) != size ||
 		     (i915_gem_obj_ggtt_offset(obj) & (size - 1)),
-		     "object 0x%08lx [fenceable? %d] not 1M or pot-size (0x%08x) aligned\n",
+		     "object 0x%08llx [fenceable? %d] not 1M or pot-size (0x%08x) aligned\n",
 		     i915_gem_obj_ggtt_offset(obj), obj->map_and_fenceable, size);
 
 		if (obj->tiling_mode == I915_TILING_Y && HAS_128_BYTE_Y_TILING(dev))
@@ -171,7 +171,7 @@ static void i830_write_fence_reg(struct drm_device *dev, int reg,
 		WARN((i915_gem_obj_ggtt_offset(obj) & ~I830_FENCE_START_MASK) ||
 		     (size & -size) != size ||
 		     (i915_gem_obj_ggtt_offset(obj) & (size - 1)),
-		     "object 0x%08lx not 512K or pot-size 0x%08x aligned\n",
+		     "object 0x%08llx not 512K or pot-size 0x%08x aligned\n",
 		     i915_gem_obj_ggtt_offset(obj), size);
 
 		pitch_val = obj->stride / 128;
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 38439fa..7a89e97 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -2543,9 +2543,9 @@ static void i915_gtt_color_adjust(struct drm_mm_node *node,
 }
 
 static int i915_gem_setup_global_gtt(struct drm_device *dev,
-				     unsigned long start,
-				     unsigned long mappable_end,
-				     unsigned long end)
+				     u64 start,
+				     u64 mappable_end,
+				     u64 end)
 {
 	/* Let GEM Manage all of the aperture.
 	 *
@@ -2584,7 +2584,7 @@ static int i915_gem_setup_global_gtt(struct drm_device *dev,
 	list_for_each_entry(obj, &dev_priv->mm.bound_list, global_list) {
 		struct i915_vma *vma = i915_gem_obj_to_vma(obj, ggtt_vm);
 
-		DRM_DEBUG_KMS("reserving preallocated space: %lx + %zx\n",
+		DRM_DEBUG_KMS("reserving preallocated space: %llx + %zx\n",
 			      i915_gem_obj_ggtt_offset(obj), obj->base.size);
 
 		WARN_ON(i915_gem_obj_ggtt_bound(obj));
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.h b/drivers/gpu/drm/i915/i915_gem_gtt.h
index 70c50e7..8275007 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.h
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.h
@@ -149,7 +149,7 @@ struct i915_ggtt_view {
 
 	union {
 		struct {
-			unsigned long offset;
+			u64 offset;
 			unsigned int size;
 		} partial;
 	} params;
diff --git a/drivers/gpu/drm/i915/intel_fbdev.c b/drivers/gpu/drm/i915/intel_fbdev.c
index 7eff33f..6c9351b 100644
--- a/drivers/gpu/drm/i915/intel_fbdev.c
+++ b/drivers/gpu/drm/i915/intel_fbdev.c
@@ -295,7 +295,7 @@ static int intelfb_create(struct drm_fb_helper *helper,
 
 	/* Use default scratch pixmap (info->pixmap.flags = FB_PIXMAP_SYSTEM) */
 
-	DRM_DEBUG_KMS("allocated %dx%d fb: 0x%08lx, bo %p\n",
+	DRM_DEBUG_KMS("allocated %dx%d fb: 0x%08llx, bo %p\n",
 		      fb->width, fb->height,
 		      i915_gem_obj_ggtt_offset(obj), obj);
 
-- 
cgit v0.10.2


From f365f91181ce862181d0d2f4fb3ca0fe2276874b Mon Sep 17 00:00:00 2001
From: Michel Thierry <michel.thierry@intel.com>
Date: Fri, 7 Aug 2015 17:40:18 +0100
Subject: drm/i915: Use complete virtual address range on 32-bit platforms

With the offset length being taken care of in ("drm/i915/gtt: Allow >=
4GB offsets in X86_32"), the code should be finally safe in 32-bit
kernels.

This reverts commit 501fd70fcaebc911b6b96a7b331e6960e5af67e7
Author: Michel Thierry <michel.thierry@intel.com>
Date:   Fri May 29 14:15:05 2015 +0100

    drm/i915: limit PPGTT size to 2GB in 32-bit platforms

Signed-off-by: Michel Thierry <michel.thierry@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 7a89e97..8fda1fb 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -1480,14 +1480,6 @@ static int gen8_ppgtt_init(struct i915_hw_ppgtt *ppgtt)
 			goto free_scratch;
 
 		ppgtt->base.total = 1ULL << 32;
-		if (IS_ENABLED(CONFIG_X86_32))
-			/* While we have a proliferation of size_t variables
-			 * we cannot represent the full ppgtt size on 32bit,
-			 * so limit it to the same size as the GGTT (currently
-			 * 2GiB).
-			 */
-			ppgtt->base.total = to_i915(ppgtt->base.dev)->gtt.base.total;
-
 		ppgtt->switch_mm = gen8_legacy_mm_switch;
 		trace_i915_page_directory_pointer_entry_alloc(&ppgtt->base,
 							      0, 0,
-- 
cgit v0.10.2


From 25f5033761f85a659db729a10b59d80f93a6be0d Mon Sep 17 00:00:00 2001
From: Michel Thierry <michel.thierry@intel.com>
Date: Fri, 7 Aug 2015 17:40:19 +0100
Subject: drm/i915: Always pass dev pointer in pdp_init

And fix 0-DAY kernel test infrastructure warning.

Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Michel Thierry <michel.thierry@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 8fda1fb..4a76807 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -1475,7 +1475,7 @@ static int gen8_ppgtt_init(struct i915_hw_ppgtt *ppgtt)
 		ppgtt->base.total = 1ULL << 48;
 		ppgtt->switch_mm = gen8_48b_mm_switch;
 	} else {
-		ret = __pdp_init(false, &ppgtt->pdp);
+		ret = __pdp_init(ppgtt->base.dev, &ppgtt->pdp);
 		if (ret)
 			goto free_scratch;
 
-- 
cgit v0.10.2


From 7f3561bec7cb4e4804bc246069420d3f2f7f3ebe Mon Sep 17 00:00:00 2001
From: Sonika Jindal <sonika.jindal@intel.com>
Date: Mon, 10 Aug 2015 10:35:35 +0530
Subject: drm/i915/bxt: Add HPD support for DDIA

Also remove redundant comments.

Signed-off-by: Sonika Jindal <sonika.jindal@intel.com>
Reviewed-by: Sivakumar Thulasimani <sivakumar.thulasimani@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 1118c39..a051045 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -90,6 +90,7 @@ static const u32 hpd_status_i915[HPD_NUM_PINS] = {
 
 /* BXT hpd list */
 static const u32 hpd_bxt[HPD_NUM_PINS] = {
+	[HPD_PORT_A] = BXT_DE_PORT_HP_DDIA,
 	[HPD_PORT_B] = BXT_DE_PORT_HP_DDIB,
 	[HPD_PORT_C] = BXT_DE_PORT_HP_DDIC
 };
@@ -3014,30 +3015,25 @@ static void bxt_hpd_irq_setup(struct drm_device *dev)
 	u32 hotplug_port = 0;
 	u32 hotplug_ctrl;
 
-	/* Now, enable HPD */
 	for_each_intel_encoder(dev, intel_encoder) {
 		if (dev_priv->hotplug.stats[intel_encoder->hpd_pin].state
 				== HPD_ENABLED)
 			hotplug_port |= hpd_bxt[intel_encoder->hpd_pin];
 	}
 
-	/* Mask all HPD control bits */
 	hotplug_ctrl = I915_READ(BXT_HOTPLUG_CTL) & ~BXT_HOTPLUG_CTL_MASK;
 
-	/* Enable requested port in hotplug control */
-	/* TODO: implement (short) HPD support on port A */
-	WARN_ON_ONCE(hotplug_port & BXT_DE_PORT_HP_DDIA);
+	if (hotplug_port & BXT_DE_PORT_HP_DDIA)
+		hotplug_ctrl |= BXT_DDIA_HPD_ENABLE;
 	if (hotplug_port & BXT_DE_PORT_HP_DDIB)
 		hotplug_ctrl |= BXT_DDIB_HPD_ENABLE;
 	if (hotplug_port & BXT_DE_PORT_HP_DDIC)
 		hotplug_ctrl |= BXT_DDIC_HPD_ENABLE;
 	I915_WRITE(BXT_HOTPLUG_CTL, hotplug_ctrl);
 
-	/* Unmask DDI hotplug in IMR */
 	hotplug_ctrl = I915_READ(GEN8_DE_PORT_IMR) & ~hotplug_port;
 	I915_WRITE(GEN8_DE_PORT_IMR, hotplug_ctrl);
 
-	/* Enable DDI hotplug in IER */
 	hotplug_ctrl = I915_READ(GEN8_DE_PORT_IER) | hotplug_port;
 	I915_WRITE(GEN8_DE_PORT_IER, hotplug_ctrl);
 	POSTING_READ(GEN8_DE_PORT_IER);
-- 
cgit v0.10.2


From cf1d58833f07afbb4534b15caa3fd48baa313b2c Mon Sep 17 00:00:00 2001
From: Sonika Jindal <sonika.jindal@intel.com>
Date: Mon, 10 Aug 2015 10:35:36 +0530
Subject: drm/i915/bxt: WA for swapped HPD pins in A stepping
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

WA for BXT A0/A1, where DDIB's HPD pin is swapped to DDIA, so enabling
DDIA HPD pin in place of DDIB.

v2: For DP, irq_port is used to determine the encoder instead of
hpd_pin and removing the edp HPD logic because port A HPD is not
present(Imre)
v3: Rebased on top of Imre's patchset for enabling HPD on PORT A.
Added hpd_pin swapping for intel_dp_init_connector, setting encoder
for PORT_A as per the WA in irq_port (Imre)
v4: Dont enable interrupt for edp, also reframe the description (Siva)
v5: Don’t check for PORT_A in intel_ddi_init to update dig_port,
instead avoid setting hpd_pin itself (Imre)

Signed-off-by: Sonika Jindal <sonika.jindal@intel.com>
Reviewed-by: Sivakumar Thulasimani <sivakumar.thulasimani@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
index 6cfe65d..d5c331d 100644
--- a/drivers/gpu/drm/i915/intel_ddi.c
+++ b/drivers/gpu/drm/i915/intel_ddi.c
@@ -3243,7 +3243,15 @@ void intel_ddi_init(struct drm_device *dev, enum port port)
 			goto err;
 
 		intel_dig_port->hpd_pulse = intel_dp_hpd_pulse;
-		dev_priv->hotplug.irq_port[port] = intel_dig_port;
+		/*
+		 * On BXT A0/A1, sw needs to activate DDIA HPD logic and
+		 * interrupts to check the external panel connection.
+		 */
+		if (IS_BROXTON(dev_priv) && (INTEL_REVID(dev) < BXT_REVID_B0)
+					 && port == PORT_B)
+			dev_priv->hotplug.irq_port[PORT_A] = intel_dig_port;
+		else
+			dev_priv->hotplug.irq_port[port] = intel_dig_port;
 	}
 
 	/* In theory we don't need the encoder->type check, but leave it just in
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 54e9b53..a028246 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -5876,6 +5876,8 @@ intel_dp_init_connector(struct intel_digital_port *intel_dig_port,
 		break;
 	case PORT_B:
 		intel_encoder->hpd_pin = HPD_PORT_B;
+		if (IS_BROXTON(dev_priv) && (INTEL_REVID(dev) < BXT_REVID_B0))
+			intel_encoder->hpd_pin = HPD_PORT_A;
 		break;
 	case PORT_C:
 		intel_encoder->hpd_pin = HPD_PORT_C;
diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index 51cbea8..0cfbe85 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -1973,7 +1973,14 @@ void intel_hdmi_init_connector(struct intel_digital_port *intel_dig_port,
 			intel_hdmi->ddc_bus = GMBUS_PIN_1_BXT;
 		else
 			intel_hdmi->ddc_bus = GMBUS_PIN_DPB;
-		intel_encoder->hpd_pin = HPD_PORT_B;
+		/*
+		 * On BXT A0/A1, sw needs to activate DDIA HPD logic and
+		 * interrupts to check the external panel connection.
+		 */
+		if (IS_BROXTON(dev_priv) && (INTEL_REVID(dev) < BXT_REVID_B0))
+			intel_encoder->hpd_pin = HPD_PORT_A;
+		else
+			intel_encoder->hpd_pin = HPD_PORT_B;
 		break;
 	case PORT_C:
 		if (IS_BROXTON(dev_priv))
-- 
cgit v0.10.2


From ed75a55bb36dec17b9e0f878cea7c8ccb037d8f9 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 11 Aug 2015 19:47:10 +0300
Subject: drm/i915: clflush on pin_to_display after pwrite to UC bo in LLC
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Currently we don't clflush on pin_to_display if the bo is already
UC/WT and is not in the CPU write domain. This causes problems with
pwrite since pwrite doesn't change the write domain, and it avoids
clflushing on UC/WT buffers on LLC platforms unless the buffer is
currently being scanned out.

Fix the problem by marking the cache dirty and adjusting
i915_gem_object_set_cache_level() to clflush when the cache is dirty
even if the cache_level doesn't change.

My last attempt [1] at fixing this via write domain frobbing was shot
down, but now with the cache_dirty flag we can do things in a nicer way.

[1] http://lists.freedesktop.org/archives/intel-gfx/2014-November/055390.html

v2: Drop the I915_CACHE_NONE/WT checks from pwrite

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=86422
Testcase: igt/kms_pwrite_crc
Testcase: igt/gem_pwrite_snooped
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 6d0f834..407b6b3 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -1005,12 +1005,14 @@ out:
 		if (!needs_clflush_after &&
 		    obj->base.write_domain != I915_GEM_DOMAIN_CPU) {
 			if (i915_gem_clflush_object(obj, obj->pin_display))
-				i915_gem_chipset_flush(dev);
+				needs_clflush_after = true;
 		}
 	}
 
 	if (needs_clflush_after)
 		i915_gem_chipset_flush(dev);
+	else
+		obj->cache_dirty = true;
 
 	intel_fb_obj_flush(obj, false, ORIGIN_CPU);
 	return ret;
@@ -3639,10 +3641,10 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,
 {
 	struct drm_device *dev = obj->base.dev;
 	struct i915_vma *vma, *next;
-	int ret;
+	int ret = 0;
 
 	if (obj->cache_level == cache_level)
-		return 0;
+		goto out;
 
 	if (i915_gem_obj_is_pinned(obj)) {
 		DRM_DEBUG("can not change the cache level of pinned objects\n");
@@ -3687,6 +3689,7 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,
 		vma->node.color = cache_level;
 	obj->cache_level = cache_level;
 
+out:
 	if (obj->cache_dirty &&
 	    obj->base.write_domain != I915_GEM_DOMAIN_CPU &&
 	    cpu_write_needs_clflush(obj)) {
-- 
cgit v0.10.2


From 908565c20888f790b96b03ae5be9cd175f3d8d72 Mon Sep 17 00:00:00 2001
From: Chris Wilson <chris@chris-wilson.co.uk>
Date: Wed, 12 Aug 2015 13:08:22 +0100
Subject: drm/i915: Do not check or a stalled pageflip prior to it being queued
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

When we queue the command or operation to change the scanout address, we
mark the flip as in progress. We can use this flag to prevent us from
checking for a stalled flip prior to its existence!

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 7f6d5fd..dc3e6b2 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -11236,6 +11236,9 @@ static bool __intel_pageflip_stall_check(struct drm_device *dev,
 	if (atomic_read(&work->pending) >= INTEL_FLIP_COMPLETE)
 		return true;
 
+	if (atomic_read(&work->pending) < INTEL_FLIP_PENDING)
+		return false;
+
 	if (!work->enable_stall_check)
 		return false;
 
-- 
cgit v0.10.2


From 0f2a2a756e862e1936ef599451015ca4c12efde8 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 6 Jul 2015 15:10:00 +0300
Subject: drm/i915: Clean up DP/HDMI limited color range handling
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Currently we treat intel_{dp,hdmi}->color_range as partly user
controller value (via the property) but we also change it during
.compute_config() when using the "Automatic" mode. That is a bit
confusing, so let's just change things so that we store the user
property values in intel_dp, and only change what's stored in
pipe_config during .compute_config().

There should be no functional change.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Sivakumar Thulasimani <sivakumar.thulasimani@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index a028246..df1b663 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -1492,15 +1492,13 @@ found:
 		 * CEA-861-E - 5.1 Default Encoding Parameters
 		 * VESA DisplayPort Ver.1.2a - 5.1.1.1 Video Colorimetry
 		 */
-		if (bpp != 18 && drm_match_cea_mode(adjusted_mode) > 1)
-			intel_dp->color_range = DP_COLOR_RANGE_16_235;
-		else
-			intel_dp->color_range = 0;
+		pipe_config->limited_color_range =
+			bpp != 18 && drm_match_cea_mode(adjusted_mode) > 1;
+	} else {
+		pipe_config->limited_color_range =
+			intel_dp->limited_color_range;
 	}
 
-	if (intel_dp->color_range)
-		pipe_config->limited_color_range = true;
-
 	intel_dp->lane_count = lane_count;
 
 	if (intel_dp->num_sink_rates) {
@@ -1642,8 +1640,9 @@ static void intel_dp_prepare(struct intel_encoder *encoder)
 			trans_dp &= ~TRANS_DP_ENH_FRAMING;
 		I915_WRITE(TRANS_DP_CTL(crtc->pipe), trans_dp);
 	} else {
-		if (!HAS_PCH_SPLIT(dev) && !IS_VALLEYVIEW(dev))
-			intel_dp->DP |= intel_dp->color_range;
+		if (!HAS_PCH_SPLIT(dev) && !IS_VALLEYVIEW(dev) &&
+		    crtc->config->limited_color_range)
+			intel_dp->DP |= DP_COLOR_RANGE_16_235;
 
 		if (adjusted_mode->flags & DRM_MODE_FLAG_PHSYNC)
 			intel_dp->DP |= DP_SYNC_HS_HIGH;
@@ -4751,7 +4750,7 @@ intel_dp_set_property(struct drm_connector *connector,
 
 	if (property == dev_priv->broadcast_rgb_property) {
 		bool old_auto = intel_dp->color_range_auto;
-		uint32_t old_range = intel_dp->color_range;
+		bool old_range = intel_dp->limited_color_range;
 
 		switch (val) {
 		case INTEL_BROADCAST_RGB_AUTO:
@@ -4759,18 +4758,18 @@ intel_dp_set_property(struct drm_connector *connector,
 			break;
 		case INTEL_BROADCAST_RGB_FULL:
 			intel_dp->color_range_auto = false;
-			intel_dp->color_range = 0;
+			intel_dp->limited_color_range = false;
 			break;
 		case INTEL_BROADCAST_RGB_LIMITED:
 			intel_dp->color_range_auto = false;
-			intel_dp->color_range = DP_COLOR_RANGE_16_235;
+			intel_dp->limited_color_range = true;
 			break;
 		default:
 			return -EINVAL;
 		}
 
 		if (old_auto == intel_dp->color_range_auto &&
-		    old_range == intel_dp->color_range)
+		    old_range == intel_dp->limited_color_range)
 			return 0;
 
 		goto done;
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 6b3e2c6..e2456bc 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -657,7 +657,7 @@ struct cxsr_latency {
 struct intel_hdmi {
 	u32 hdmi_reg;
 	int ddc_bus;
-	uint32_t color_range;
+	bool limited_color_range;
 	bool color_range_auto;
 	bool has_hdmi_sink;
 	bool has_audio;
@@ -708,7 +708,7 @@ struct intel_dp {
 	uint32_t DP;
 	bool has_audio;
 	enum hdmi_force_audio force_audio;
-	uint32_t color_range;
+	bool limited_color_range;
 	bool color_range_auto;
 	uint8_t link_bw;
 	uint8_t rate_select;
diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index 0cfbe85..7185062 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -848,8 +848,8 @@ static void intel_hdmi_prepare(struct intel_encoder *encoder)
 	u32 hdmi_val;
 
 	hdmi_val = SDVO_ENCODING_HDMI;
-	if (!HAS_PCH_SPLIT(dev))
-		hdmi_val |= intel_hdmi->color_range;
+	if (!HAS_PCH_SPLIT(dev) && crtc->config->limited_color_range)
+		hdmi_val |= HDMI_COLOR_RANGE_16_235;
 	if (adjusted_mode->flags & DRM_MODE_FLAG_PVSYNC)
 		hdmi_val |= SDVO_VSYNC_ACTIVE_HIGH;
 	if (adjusted_mode->flags & DRM_MODE_FLAG_PHSYNC)
@@ -1260,11 +1260,12 @@ bool intel_hdmi_compute_config(struct intel_encoder *encoder,
 
 	if (intel_hdmi->color_range_auto) {
 		/* See CEA-861-E - 5.1 Default Encoding Parameters */
-		if (pipe_config->has_hdmi_sink &&
-		    drm_match_cea_mode(adjusted_mode) > 1)
-			intel_hdmi->color_range = HDMI_COLOR_RANGE_16_235;
-		else
-			intel_hdmi->color_range = 0;
+		pipe_config->limited_color_range =
+			pipe_config->has_hdmi_sink &&
+			drm_match_cea_mode(adjusted_mode) > 1;
+	} else {
+		pipe_config->limited_color_range =
+			intel_hdmi->limited_color_range;
 	}
 
 	if (adjusted_mode->flags & DRM_MODE_FLAG_DBLCLK) {
@@ -1273,9 +1274,6 @@ bool intel_hdmi_compute_config(struct intel_encoder *encoder,
 		clock_12bpc *= 2;
 	}
 
-	if (intel_hdmi->color_range)
-		pipe_config->limited_color_range = true;
-
 	if (HAS_PCH_SPLIT(dev) && !HAS_DDI(dev))
 		pipe_config->has_pch_encoder = true;
 
@@ -1470,7 +1468,7 @@ intel_hdmi_set_property(struct drm_connector *connector,
 
 	if (property == dev_priv->broadcast_rgb_property) {
 		bool old_auto = intel_hdmi->color_range_auto;
-		uint32_t old_range = intel_hdmi->color_range;
+		bool old_range = intel_hdmi->limited_color_range;
 
 		switch (val) {
 		case INTEL_BROADCAST_RGB_AUTO:
@@ -1478,18 +1476,18 @@ intel_hdmi_set_property(struct drm_connector *connector,
 			break;
 		case INTEL_BROADCAST_RGB_FULL:
 			intel_hdmi->color_range_auto = false;
-			intel_hdmi->color_range = 0;
+			intel_hdmi->limited_color_range = false;
 			break;
 		case INTEL_BROADCAST_RGB_LIMITED:
 			intel_hdmi->color_range_auto = false;
-			intel_hdmi->color_range = HDMI_COLOR_RANGE_16_235;
+			intel_hdmi->limited_color_range = true;
 			break;
 		default:
 			return -EINVAL;
 		}
 
 		if (old_auto == intel_hdmi->color_range_auto &&
-		    old_range == intel_hdmi->color_range)
+		    old_range == intel_hdmi->limited_color_range)
 			return 0;
 
 		goto done;
-- 
cgit v0.10.2


From 840b32b7eda7837db6e0236293f71fc6644dedf8 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 11 Aug 2015 20:21:46 +0300
Subject: drm/i915: Don't use link_bw for PLL setup
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Use port_clock instead of link_bw when picking the PLL parameters for
DP. link_bw may be zero with an eDP 1.4 sink that supports
DP_LINK_RATE_SET so we shouldn't use it for anything other than feed it
to the sink appropriately.

v2: Fix typo in commit message (Sivakumar)

Reviewed-by: Sivakumar Thulasimani <sivakumar.thulasimani@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
index d5c331d..0e2f982 100644
--- a/drivers/gpu/drm/i915/intel_ddi.c
+++ b/drivers/gpu/drm/i915/intel_ddi.c
@@ -1579,17 +1579,14 @@ skl_ddi_pll_select(struct intel_crtc *intel_crtc,
 			 DPLL_CFGCR2_PDIV(wrpll_params.pdiv) |
 			 wrpll_params.central_freq;
 	} else if (intel_encoder->type == INTEL_OUTPUT_DISPLAYPORT) {
-		struct drm_encoder *encoder = &intel_encoder->base;
-		struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
-
-		switch (intel_dp->link_bw) {
-		case DP_LINK_BW_1_62:
+		switch (crtc_state->port_clock / 2) {
+		case 81000:
 			ctrl1 |= DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_810, 0);
 			break;
-		case DP_LINK_BW_2_7:
+		case 135000:
 			ctrl1 |= DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_1350, 0);
 			break;
-		case DP_LINK_BW_5_4:
+		case 270000:
 			ctrl1 |= DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_2700, 0);
 			break;
 		}
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index df1b663..3f885f6 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -48,28 +48,28 @@
 #define INTEL_DP_RESOLUTION_FAILSAFE	(3 << INTEL_DP_RESOLUTION_SHIFT_MASK)
 
 struct dp_link_dpll {
-	int link_bw;
+	int clock;
 	struct dpll dpll;
 };
 
 static const struct dp_link_dpll gen4_dpll[] = {
-	{ DP_LINK_BW_1_62,
+	{ 162000,
 		{ .p1 = 2, .p2 = 10, .n = 2, .m1 = 23, .m2 = 8 } },
-	{ DP_LINK_BW_2_7,
+	{ 270000,
 		{ .p1 = 1, .p2 = 10, .n = 1, .m1 = 14, .m2 = 2 } }
 };
 
 static const struct dp_link_dpll pch_dpll[] = {
-	{ DP_LINK_BW_1_62,
+	{ 162000,
 		{ .p1 = 2, .p2 = 10, .n = 1, .m1 = 12, .m2 = 9 } },
-	{ DP_LINK_BW_2_7,
+	{ 270000,
 		{ .p1 = 1, .p2 = 10, .n = 2, .m1 = 14, .m2 = 8 } }
 };
 
 static const struct dp_link_dpll vlv_dpll[] = {
-	{ DP_LINK_BW_1_62,
+	{ 162000,
 		{ .p1 = 3, .p2 = 2, .n = 5, .m1 = 3, .m2 = 81 } },
-	{ DP_LINK_BW_2_7,
+	{ 270000,
 		{ .p1 = 2, .p2 = 2, .n = 1, .m1 = 2, .m2 = 27 } }
 };
 
@@ -83,11 +83,11 @@ static const struct dp_link_dpll chv_dpll[] = {
 	 * m2 is stored in fixed point format using formula below
 	 * (m2_int << 22) | m2_fraction
 	 */
-	{ DP_LINK_BW_1_62,	/* m2_int = 32, m2_fraction = 1677722 */
+	{ 162000,	/* m2_int = 32, m2_fraction = 1677722 */
 		{ .p1 = 4, .p2 = 2, .n = 1, .m1 = 2, .m2 = 0x819999a } },
-	{ DP_LINK_BW_2_7,	/* m2_int = 27, m2_fraction = 0 */
+	{ 270000,	/* m2_int = 27, m2_fraction = 0 */
 		{ .p1 = 4, .p2 = 1, .n = 1, .m1 = 2, .m2 = 0x6c00000 } },
-	{ DP_LINK_BW_5_4,	/* m2_int = 27, m2_fraction = 0 */
+	{ 540000,	/* m2_int = 27, m2_fraction = 0 */
 		{ .p1 = 2, .p2 = 1, .n = 1, .m1 = 2, .m2 = 0x6c00000 } }
 };
 
@@ -1133,7 +1133,7 @@ intel_dp_connector_unregister(struct intel_connector *intel_connector)
 }
 
 static void
-skl_edp_set_pll_config(struct intel_crtc_state *pipe_config, int link_clock)
+skl_edp_set_pll_config(struct intel_crtc_state *pipe_config)
 {
 	u32 ctrl1;
 
@@ -1145,7 +1145,7 @@ skl_edp_set_pll_config(struct intel_crtc_state *pipe_config, int link_clock)
 	pipe_config->dpll_hw_state.cfgcr2 = 0;
 
 	ctrl1 = DPLL_CTRL1_OVERRIDE(SKL_DPLL0);
-	switch (link_clock / 2) {
+	switch (pipe_config->port_clock / 2) {
 	case 81000:
 		ctrl1 |= DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_810,
 					      SKL_DPLL0);
@@ -1179,19 +1179,19 @@ skl_edp_set_pll_config(struct intel_crtc_state *pipe_config, int link_clock)
 }
 
 static void
-hsw_dp_set_ddi_pll_sel(struct intel_crtc_state *pipe_config, int link_bw)
+hsw_dp_set_ddi_pll_sel(struct intel_crtc_state *pipe_config)
 {
 	memset(&pipe_config->dpll_hw_state, 0,
 	       sizeof(pipe_config->dpll_hw_state));
 
-	switch (link_bw) {
-	case DP_LINK_BW_1_62:
+	switch (pipe_config->port_clock / 2) {
+	case 81000:
 		pipe_config->ddi_pll_sel = PORT_CLK_SEL_LCPLL_810;
 		break;
-	case DP_LINK_BW_2_7:
+	case 135000:
 		pipe_config->ddi_pll_sel = PORT_CLK_SEL_LCPLL_1350;
 		break;
-	case DP_LINK_BW_5_4:
+	case 270000:
 		pipe_config->ddi_pll_sel = PORT_CLK_SEL_LCPLL_2700;
 		break;
 	}
@@ -1238,7 +1238,7 @@ intel_dp_source_rates(struct drm_device *dev, const int **source_rates)
 
 static void
 intel_dp_set_clock(struct intel_encoder *encoder,
-		   struct intel_crtc_state *pipe_config, int link_bw)
+		   struct intel_crtc_state *pipe_config)
 {
 	struct drm_device *dev = encoder->base.dev;
 	const struct dp_link_dpll *divisor = NULL;
@@ -1260,7 +1260,7 @@ intel_dp_set_clock(struct intel_encoder *encoder,
 
 	if (divisor && count) {
 		for (i = 0; i < count; i++) {
-			if (link_bw == divisor[i].link_bw) {
+			if (pipe_config->port_clock == divisor[i].clock) {
 				pipe_config->dpll = divisor[i].dpll;
 				pipe_config->clock_set = true;
 				break;
@@ -1535,13 +1535,13 @@ found:
 	}
 
 	if (IS_SKYLAKE(dev) && is_edp(intel_dp))
-		skl_edp_set_pll_config(pipe_config, common_rates[clock]);
+		skl_edp_set_pll_config(pipe_config);
 	else if (IS_BROXTON(dev))
 		/* handled in ddi */;
 	else if (IS_HASWELL(dev) || IS_BROADWELL(dev))
-		hsw_dp_set_ddi_pll_sel(pipe_config, intel_dp->link_bw);
+		hsw_dp_set_ddi_pll_sel(pipe_config);
 	else
-		intel_dp_set_clock(encoder, pipe_config, intel_dp->link_bw);
+		intel_dp_set_clock(encoder, pipe_config);
 
 	return true;
 }
-- 
cgit v0.10.2


From 96f3f1f905e9c4977220fa07477cd8a8ed4af4da Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 6 Jul 2015 15:10:02 +0300
Subject: drm/i915: Don't pass clock to DDI PLL select functions
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

All the *_ddi_pll_select() functions get passed the port_clock and pipe
config as parameters. We only need to pass the pipe config, and the
functions can dig up the port_clock themselves.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Sivakumar Thulasimani <sivakumar.thulasimani@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
index 0e2f982..b6d1ab2 100644
--- a/drivers/gpu/drm/i915/intel_ddi.c
+++ b/drivers/gpu/drm/i915/intel_ddi.c
@@ -1267,9 +1267,10 @@ hsw_ddi_calculate_wrpll(int clock /* in Hz */,
 static bool
 hsw_ddi_pll_select(struct intel_crtc *intel_crtc,
 		   struct intel_crtc_state *crtc_state,
-		   struct intel_encoder *intel_encoder,
-		   int clock)
+		   struct intel_encoder *intel_encoder)
 {
+	int clock = crtc_state->port_clock;
+
 	if (intel_encoder->type == INTEL_OUTPUT_HDMI) {
 		struct intel_shared_dpll *pll;
 		uint32_t val;
@@ -1548,11 +1549,11 @@ skip_remaining_dividers:
 static bool
 skl_ddi_pll_select(struct intel_crtc *intel_crtc,
 		   struct intel_crtc_state *crtc_state,
-		   struct intel_encoder *intel_encoder,
-		   int clock)
+		   struct intel_encoder *intel_encoder)
 {
 	struct intel_shared_dpll *pll;
 	uint32_t ctrl1, cfgcr1, cfgcr2;
+	int clock = crtc_state->port_clock;
 
 	/*
 	 * See comment in intel_dpll_hw_state to understand why we always use 0
@@ -1640,14 +1641,14 @@ static const struct bxt_clk_div bxt_dp_clk_val[] = {
 static bool
 bxt_ddi_pll_select(struct intel_crtc *intel_crtc,
 		   struct intel_crtc_state *crtc_state,
-		   struct intel_encoder *intel_encoder,
-		   int clock)
+		   struct intel_encoder *intel_encoder)
 {
 	struct intel_shared_dpll *pll;
 	struct bxt_clk_div clk_div = {0};
 	int vco = 0;
 	uint32_t prop_coef, int_coef, gain_ctl, targ_cnt;
 	uint32_t lanestagger;
+	int clock = crtc_state->port_clock;
 
 	if (intel_encoder->type == INTEL_OUTPUT_HDMI) {
 		intel_clock_t best_clock;
@@ -1775,17 +1776,16 @@ bool intel_ddi_pll_select(struct intel_crtc *intel_crtc,
 	struct drm_device *dev = intel_crtc->base.dev;
 	struct intel_encoder *intel_encoder =
 		intel_ddi_get_crtc_new_encoder(crtc_state);
-	int clock = crtc_state->port_clock;
 
 	if (IS_SKYLAKE(dev))
 		return skl_ddi_pll_select(intel_crtc, crtc_state,
-					  intel_encoder, clock);
+					  intel_encoder);
 	else if (IS_BROXTON(dev))
 		return bxt_ddi_pll_select(intel_crtc, crtc_state,
-					  intel_encoder, clock);
+					  intel_encoder);
 	else
 		return hsw_ddi_pll_select(intel_crtc, crtc_state,
-					  intel_encoder, clock);
+					  intel_encoder);
 }
 
 void intel_ddi_set_pipe_settings(struct drm_crtc *crtc)
-- 
cgit v0.10.2


From b81e34c29e67f5b3e5ce28b5055ae2ca59c36f7b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 6 Jul 2015 15:10:03 +0300
Subject: drm/i915: Avoid confusion between DP and TRANS_DP_CTL in DP
 .get_config()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Use a separate variable for the TRANS_DP_CTL value instead of reusing
'tmp' that otherwise contains the DP port register value.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Sivakumar Thulasimani <sivakumar.thulasimani@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 3f885f6..3a9b8e4 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -2282,13 +2282,14 @@ static void intel_dp_get_config(struct intel_encoder *encoder,
 	pipe_config->has_audio = tmp & DP_AUDIO_OUTPUT_ENABLE && port != PORT_A;
 
 	if (HAS_PCH_CPT(dev) && port != PORT_A) {
-		tmp = I915_READ(TRANS_DP_CTL(crtc->pipe));
-		if (tmp & TRANS_DP_HSYNC_ACTIVE_HIGH)
+		u32 trans_dp = I915_READ(TRANS_DP_CTL(crtc->pipe));
+
+		if (trans_dp & TRANS_DP_HSYNC_ACTIVE_HIGH)
 			flags |= DRM_MODE_FLAG_PHSYNC;
 		else
 			flags |= DRM_MODE_FLAG_NHSYNC;
 
-		if (tmp & TRANS_DP_VSYNC_ACTIVE_HIGH)
+		if (trans_dp & TRANS_DP_VSYNC_ACTIVE_HIGH)
 			flags |= DRM_MODE_FLAG_PVSYNC;
 		else
 			flags |= DRM_MODE_FLAG_NVSYNC;
-- 
cgit v0.10.2


From 90a6b7b052b1aa17fbb98b049e9c8b7f729c35a7 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 6 Jul 2015 16:39:15 +0300
Subject: drm/i915: Move intel_dp->lane_count into pipe_config
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Currently we clobber intel_dp->lane_count in compute config, which means
after a rejected modeset we may no longer be able to retrain the current
link. Move lane_count into pipe_config to avoid that.

v2: Add missing ':' to the pipe config debug dump

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Sivakumar Thulasimani <sivakumar.thulasimani@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 88f20cc..f22aca3 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -4108,6 +4108,7 @@ enum skl_disp_power_wells {
 /* How many wires to use. I guess 3 was too hard */
 #define   DP_PORT_WIDTH(width)		(((width) - 1) << 19)
 #define   DP_PORT_WIDTH_MASK		(7 << 19)
+#define   DP_PORT_WIDTH_SHIFT		19
 
 /* Mystic DPCD version 1.1 special mode */
 #define   DP_ENHANCED_FRAMING		(1 << 18)
@@ -7148,6 +7149,8 @@ enum skl_disp_power_wells {
 #define  DDI_BUF_IS_IDLE			(1<<7)
 #define  DDI_A_4_LANES				(1<<4)
 #define  DDI_PORT_WIDTH(width)			(((width) - 1) << 1)
+#define  DDI_PORT_WIDTH_MASK			(7 << 1)
+#define  DDI_PORT_WIDTH_SHIFT			1
 #define  DDI_INIT_DISPLAY_DETECTED		(1<<0)
 
 /* DDI Buffer Translations */
diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
index b6d1ab2..56d778f 100644
--- a/drivers/gpu/drm/i915/intel_ddi.c
+++ b/drivers/gpu/drm/i915/intel_ddi.c
@@ -728,11 +728,11 @@ void intel_ddi_init_dp_buf_reg(struct intel_encoder *encoder)
 	struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base);
 	struct intel_digital_port *intel_dig_port =
 		enc_to_dig_port(&encoder->base);
+	struct intel_crtc *crtc = to_intel_crtc(encoder->base.crtc);
 
 	intel_dp->DP = intel_dig_port->saved_port_bits |
 		DDI_BUF_CTL_ENABLE | DDI_BUF_TRANS_SELECT(0);
-	intel_dp->DP |= DDI_PORT_WIDTH(intel_dp->lane_count);
-
+	intel_dp->DP |= DDI_PORT_WIDTH(crtc->config->lane_count);
 }
 
 static struct intel_encoder *
@@ -1918,7 +1918,7 @@ void intel_ddi_enable_transcoder_func(struct drm_crtc *crtc)
 		} else
 			temp |= TRANS_DDI_MODE_SELECT_DP_SST;
 
-		temp |= DDI_PORT_WIDTH(intel_dp->lane_count);
+		temp |= DDI_PORT_WIDTH(intel_crtc->config->lane_count);
 	} else if (type == INTEL_OUTPUT_DP_MST) {
 		struct intel_dp *intel_dp = &enc_to_mst(encoder)->primary->dp;
 
@@ -1927,7 +1927,7 @@ void intel_ddi_enable_transcoder_func(struct drm_crtc *crtc)
 		} else
 			temp |= TRANS_DDI_MODE_SELECT_DP_SST;
 
-		temp |= DDI_PORT_WIDTH(intel_dp->lane_count);
+		temp |= DDI_PORT_WIDTH(intel_crtc->config->lane_count);
 	} else {
 		WARN(1, "Invalid encoder type %d for pipe %c\n",
 		     intel_encoder->type, pipe_name(pipe));
@@ -3094,6 +3094,8 @@ void intel_ddi_get_config(struct intel_encoder *encoder,
 	case TRANS_DDI_MODE_SELECT_DP_SST:
 	case TRANS_DDI_MODE_SELECT_DP_MST:
 		pipe_config->has_dp_encoder = true;
+		pipe_config->lane_count =
+			((temp & DDI_PORT_WIDTH_MASK) >> DDI_PORT_WIDTH_SHIFT) + 1;
 		intel_dp_get_m_n(intel_crtc, pipe_config);
 		break;
 	default:
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index dc3e6b2..40e70f2 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -11902,14 +11902,16 @@ static void intel_dump_pipe_config(struct intel_crtc *crtc,
 		      pipe_config->fdi_m_n.gmch_m, pipe_config->fdi_m_n.gmch_n,
 		      pipe_config->fdi_m_n.link_m, pipe_config->fdi_m_n.link_n,
 		      pipe_config->fdi_m_n.tu);
-	DRM_DEBUG_KMS("dp: %i, gmch_m: %u, gmch_n: %u, link_m: %u, link_n: %u, tu: %u\n",
+	DRM_DEBUG_KMS("dp: %i, lanes: %i, gmch_m: %u, gmch_n: %u, link_m: %u, link_n: %u, tu: %u\n",
 		      pipe_config->has_dp_encoder,
+		      pipe_config->lane_count,
 		      pipe_config->dp_m_n.gmch_m, pipe_config->dp_m_n.gmch_n,
 		      pipe_config->dp_m_n.link_m, pipe_config->dp_m_n.link_n,
 		      pipe_config->dp_m_n.tu);
 
-	DRM_DEBUG_KMS("dp: %i, gmch_m2: %u, gmch_n2: %u, link_m2: %u, link_n2: %u, tu2: %u\n",
+	DRM_DEBUG_KMS("dp: %i, lanes: %i, gmch_m2: %u, gmch_n2: %u, link_m2: %u, link_n2: %u, tu2: %u\n",
 		      pipe_config->has_dp_encoder,
+		      pipe_config->lane_count,
 		      pipe_config->dp_m2_n2.gmch_m,
 		      pipe_config->dp_m2_n2.gmch_n,
 		      pipe_config->dp_m2_n2.link_m,
@@ -12416,6 +12418,7 @@ intel_pipe_config_compare(struct drm_device *dev,
 	PIPE_CONF_CHECK_M_N(fdi_m_n);
 
 	PIPE_CONF_CHECK_I(has_dp_encoder);
+	PIPE_CONF_CHECK_I(lane_count);
 
 	if (INTEL_INFO(dev)->gen < 8) {
 		PIPE_CONF_CHECK_M_N(dp_m_n);
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 3a9b8e4..3d3a908 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -1499,7 +1499,7 @@ found:
 			intel_dp->limited_color_range;
 	}
 
-	intel_dp->lane_count = lane_count;
+	pipe_config->lane_count = lane_count;
 
 	if (intel_dp->num_sink_rates) {
 		intel_dp->link_bw = 0;
@@ -1515,7 +1515,7 @@ found:
 	pipe_config->port_clock = common_rates[clock];
 
 	DRM_DEBUG_KMS("DP link bw %02x lane count %d clock %d bpp %d\n",
-		      intel_dp->link_bw, intel_dp->lane_count,
+		      intel_dp->link_bw, pipe_config->lane_count,
 		      pipe_config->port_clock, bpp);
 	DRM_DEBUG_KMS("DP link bw required %i available %i\n",
 		      mode_rate, link_avail);
@@ -1610,7 +1610,7 @@ static void intel_dp_prepare(struct intel_encoder *encoder)
 
 	/* Handle DP bits in common between all three register formats */
 	intel_dp->DP |= DP_VOLTAGE_0_4 | DP_PRE_EMPHASIS_0;
-	intel_dp->DP |= DP_PORT_WIDTH(intel_dp->lane_count);
+	intel_dp->DP |= DP_PORT_WIDTH(crtc->config->lane_count);
 
 	if (crtc->config->has_audio)
 		intel_dp->DP |= DP_AUDIO_OUTPUT_ENABLE;
@@ -2313,6 +2313,9 @@ static void intel_dp_get_config(struct intel_encoder *encoder,
 
 	pipe_config->has_dp_encoder = true;
 
+	pipe_config->lane_count =
+		((tmp & DP_PORT_WIDTH_MASK) >> DP_PORT_WIDTH_SHIFT) + 1;
+
 	intel_dp_get_m_n(crtc, pipe_config);
 
 	if (port == PORT_A) {
@@ -3338,13 +3341,15 @@ static void
 intel_get_adjust_train(struct intel_dp *intel_dp,
 		       const uint8_t link_status[DP_LINK_STATUS_SIZE])
 {
+	struct intel_crtc *crtc =
+		to_intel_crtc(dp_to_dig_port(intel_dp)->base.base.crtc);
 	uint8_t v = 0;
 	uint8_t p = 0;
 	int lane;
 	uint8_t voltage_max;
 	uint8_t preemph_max;
 
-	for (lane = 0; lane < intel_dp->lane_count; lane++) {
+	for (lane = 0; lane < crtc->config->lane_count; lane++) {
 		uint8_t this_v = drm_dp_get_adjust_request_voltage(link_status, lane);
 		uint8_t this_p = drm_dp_get_adjust_request_pre_emphasis(link_status, lane);
 
@@ -3513,8 +3518,10 @@ intel_dp_set_link_train(struct intel_dp *intel_dp,
 			uint8_t dp_train_pat)
 {
 	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
-	struct drm_device *dev = intel_dig_port->base.base.dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv =
+		to_i915(intel_dig_port->base.base.dev);
+	struct intel_crtc *crtc =
+		to_intel_crtc(intel_dig_port->base.base.crtc);
 	uint8_t buf[sizeof(intel_dp->train_set) + 1];
 	int ret, len;
 
@@ -3530,8 +3537,8 @@ intel_dp_set_link_train(struct intel_dp *intel_dp,
 		len = 1;
 	} else {
 		/* DP_TRAINING_LANEx_SET follow DP_TRAINING_PATTERN_SET */
-		memcpy(buf + 1, intel_dp->train_set, intel_dp->lane_count);
-		len = intel_dp->lane_count + 1;
+		memcpy(buf + 1, intel_dp->train_set, crtc->config->lane_count);
+		len = crtc->config->lane_count + 1;
 	}
 
 	ret = drm_dp_dpcd_write(&intel_dp->aux, DP_TRAINING_PATTERN_SET,
@@ -3555,8 +3562,10 @@ intel_dp_update_link_train(struct intel_dp *intel_dp, uint32_t *DP,
 			   const uint8_t link_status[DP_LINK_STATUS_SIZE])
 {
 	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
-	struct drm_device *dev = intel_dig_port->base.base.dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_private *dev_priv =
+		to_i915(intel_dig_port->base.base.dev);
+	struct intel_crtc *crtc =
+		to_intel_crtc(intel_dig_port->base.base.crtc);
 	int ret;
 
 	intel_get_adjust_train(intel_dp, link_status);
@@ -3566,9 +3575,9 @@ intel_dp_update_link_train(struct intel_dp *intel_dp, uint32_t *DP,
 	POSTING_READ(intel_dp->output_reg);
 
 	ret = drm_dp_dpcd_write(&intel_dp->aux, DP_TRAINING_LANE0_SET,
-				intel_dp->train_set, intel_dp->lane_count);
+				intel_dp->train_set, crtc->config->lane_count);
 
-	return ret == intel_dp->lane_count;
+	return ret == crtc->config->lane_count;
 }
 
 static void intel_dp_set_idle_link_train(struct intel_dp *intel_dp)
@@ -3607,6 +3616,8 @@ void
 intel_dp_start_link_train(struct intel_dp *intel_dp)
 {
 	struct drm_encoder *encoder = &dp_to_dig_port(intel_dp)->base.base;
+	struct intel_crtc *crtc =
+		to_intel_crtc(dp_to_dig_port(intel_dp)->base.base.crtc);
 	struct drm_device *dev = encoder->dev;
 	int i;
 	uint8_t voltage;
@@ -3619,7 +3630,7 @@ intel_dp_start_link_train(struct intel_dp *intel_dp)
 
 	/* Write the link configuration data */
 	link_config[0] = intel_dp->link_bw;
-	link_config[1] = intel_dp->lane_count;
+	link_config[1] = crtc->config->lane_count;
 	if (drm_dp_enhanced_frame_cap(intel_dp->dpcd))
 		link_config[1] |= DP_LANE_COUNT_ENHANCED_FRAME_EN;
 	drm_dp_dpcd_write(&intel_dp->aux, DP_LINK_BW_SET, link_config, 2);
@@ -3653,7 +3664,7 @@ intel_dp_start_link_train(struct intel_dp *intel_dp)
 			break;
 		}
 
-		if (drm_dp_clock_recovery_ok(link_status, intel_dp->lane_count)) {
+		if (drm_dp_clock_recovery_ok(link_status, crtc->config->lane_count)) {
 			DRM_DEBUG_KMS("clock recovery OK\n");
 			break;
 		}
@@ -3676,10 +3687,10 @@ intel_dp_start_link_train(struct intel_dp *intel_dp)
 		}
 
 		/* Check to see if we've tried the max voltage */
-		for (i = 0; i < intel_dp->lane_count; i++)
+		for (i = 0; i < crtc->config->lane_count; i++)
 			if ((intel_dp->train_set[i] & DP_TRAIN_MAX_SWING_REACHED) == 0)
 				break;
-		if (i == intel_dp->lane_count) {
+		if (i == crtc->config->lane_count) {
 			++loop_tries;
 			if (loop_tries == 5) {
 				DRM_ERROR("too many full retries, give up\n");
@@ -3716,6 +3727,8 @@ intel_dp_start_link_train(struct intel_dp *intel_dp)
 void
 intel_dp_complete_link_train(struct intel_dp *intel_dp)
 {
+	struct intel_crtc *crtc =
+		to_intel_crtc(dp_to_dig_port(intel_dp)->base.base.crtc);
 	bool channel_eq = false;
 	int tries, cr_tries;
 	uint32_t DP = intel_dp->DP;
@@ -3751,7 +3764,8 @@ intel_dp_complete_link_train(struct intel_dp *intel_dp)
 		}
 
 		/* Make sure clock is still ok */
-		if (!drm_dp_clock_recovery_ok(link_status, intel_dp->lane_count)) {
+		if (!drm_dp_clock_recovery_ok(link_status,
+					      crtc->config->lane_count)) {
 			intel_dp->train_set_valid = false;
 			intel_dp_start_link_train(intel_dp);
 			intel_dp_set_link_train(intel_dp, &DP,
@@ -3761,7 +3775,8 @@ intel_dp_complete_link_train(struct intel_dp *intel_dp)
 			continue;
 		}
 
-		if (drm_dp_channel_eq_ok(link_status, intel_dp->lane_count)) {
+		if (drm_dp_channel_eq_ok(link_status,
+					 crtc->config->lane_count)) {
 			channel_eq = true;
 			break;
 		}
@@ -4259,6 +4274,8 @@ update_status:
 static int
 intel_dp_check_mst_status(struct intel_dp *intel_dp)
 {
+	struct intel_crtc *crtc =
+		to_intel_crtc(dp_to_dig_port(intel_dp)->base.base.crtc);
 	bool bret;
 
 	if (intel_dp->is_mst) {
@@ -4271,7 +4288,8 @@ go_again:
 		if (bret == true) {
 
 			/* check link status - esi[10] = 0x200c */
-			if (intel_dp->active_mst_links && !drm_dp_channel_eq_ok(&esi[10], intel_dp->lane_count)) {
+			if (intel_dp->active_mst_links &&
+			    !drm_dp_channel_eq_ok(&esi[10], crtc->config->lane_count)) {
 				DRM_DEBUG_KMS("channel EQ not ok, retraining\n");
 				intel_dp_start_link_train(intel_dp);
 				intel_dp_complete_link_train(intel_dp);
@@ -4326,6 +4344,8 @@ intel_dp_check_link_status(struct intel_dp *intel_dp)
 {
 	struct drm_device *dev = intel_dp_to_dev(intel_dp);
 	struct intel_encoder *intel_encoder = &dp_to_dig_port(intel_dp)->base;
+	struct intel_crtc *crtc =
+		to_intel_crtc(dp_to_dig_port(intel_dp)->base.base.crtc);
 	u8 sink_irq_vector;
 	u8 link_status[DP_LINK_STATUS_SIZE];
 
@@ -4361,7 +4381,7 @@ intel_dp_check_link_status(struct intel_dp *intel_dp)
 			DRM_DEBUG_DRIVER("CP or sink specific irq unhandled\n");
 	}
 
-	if (!drm_dp_channel_eq_ok(link_status, intel_dp->lane_count)) {
+	if (!drm_dp_channel_eq_ok(link_status, crtc->config->lane_count)) {
 		DRM_DEBUG_KMS("%s: channel EQ not ok, retraining\n",
 			      intel_encoder->base.name);
 		intel_dp_start_link_train(intel_dp);
diff --git a/drivers/gpu/drm/i915/intel_dp_mst.c b/drivers/gpu/drm/i915/intel_dp_mst.c
index f4fe118..fdb5e31 100644
--- a/drivers/gpu/drm/i915/intel_dp_mst.c
+++ b/drivers/gpu/drm/i915/intel_dp_mst.c
@@ -65,7 +65,7 @@ static bool intel_dp_mst_compute_config(struct intel_encoder *encoder,
 		intel_dp->rate_select = 0;
 	}
 
-	intel_dp->lane_count = lane_count;
+	pipe_config->lane_count = lane_count;
 
 	pipe_config->pipe_bpp = 24;
 	pipe_config->port_clock = rate;
@@ -276,6 +276,10 @@ static void intel_dp_mst_enc_get_config(struct intel_encoder *encoder,
 		break;
 	}
 	pipe_config->base.adjusted_mode.flags |= flags;
+
+	pipe_config->lane_count =
+		((temp & DDI_PORT_WIDTH_MASK) >> DDI_PORT_WIDTH_SHIFT) + 1;
+
 	intel_dp_get_m_n(crtc, pipe_config);
 
 	intel_ddi_clock_get(&intel_dig_port->base, pipe_config);
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index e2456bc..0a11041 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -423,6 +423,8 @@ struct intel_crtc_state {
 	/* Used by SDVO (and if we ever fix it, HDMI). */
 	unsigned pixel_multiplier;
 
+	uint8_t lane_count;
+
 	/* Panel fitter controls for gen2-gen4 + VLV */
 	struct {
 		u32 control;
@@ -712,7 +714,6 @@ struct intel_dp {
 	bool color_range_auto;
 	uint8_t link_bw;
 	uint8_t rate_select;
-	uint8_t lane_count;
 	uint8_t dpcd[DP_RECEIVER_CAP_SIZE];
 	uint8_t psr_dpcd[EDP_PSR_RECEIVER_CAP_SIZE];
 	uint8_t downstream_ports[DP_MAX_DOWNSTREAM_PORTS];
-- 
cgit v0.10.2


From a79b8165beed02a96dc54737feb20700ddf71127 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 6 Jul 2015 15:10:05 +0300
Subject: drm/i915: Don't use link_bw to select between TP1 and TP3
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

intel_dp->link_bw is going away, so consul the port_clock instead when
choosing between TP1 and TP3.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Sivakumar Thulasimani <sivakumar.thulasimani@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 3d3a908..6231eb1 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -3734,8 +3734,8 @@ intel_dp_complete_link_train(struct intel_dp *intel_dp)
 	uint32_t DP = intel_dp->DP;
 	uint32_t training_pattern = DP_TRAINING_PATTERN_2;
 
-	/* Training Pattern 3 for HBR2 ot 1.2 devices that support it*/
-	if (intel_dp->link_bw == DP_LINK_BW_5_4 || intel_dp->use_tps3)
+	/* Training Pattern 3 for HBR2 or 1.2 devices that support it*/
+	if (crtc->config->port_clock == 540000 || intel_dp->use_tps3)
 		training_pattern = DP_TRAINING_PATTERN_3;
 
 	/* channel equalization */
-- 
cgit v0.10.2


From 04a60f9ffa9a9ac05d5616408e3f11115ed1450a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 6 Jul 2015 15:10:06 +0300
Subject: drm/i915: Kill intel_dp->{link_bw, rate_select}
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We only need the link_bw/rate_select parameters when starting link
training, and they should be computed based on the currently active
config, so throw them out from intel_dp and just compute on demand.

Toss in an extra debug print to see rate_select in addition to link_bw,
as the latter may be 0 for eDP 1.4.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Sivakumar Thulasimani <sivakumar.thulasimani@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 6231eb1..b905c19 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -1376,6 +1376,19 @@ int intel_dp_rate_select(struct intel_dp *intel_dp, int rate)
 	return rate_to_index(rate, intel_dp->sink_rates);
 }
 
+static void intel_dp_compute_rate(struct intel_dp *intel_dp, int port_clock,
+				  uint8_t *link_bw, uint8_t *rate_select)
+{
+	if (intel_dp->num_sink_rates) {
+		*link_bw = 0;
+		*rate_select =
+			intel_dp_rate_select(intel_dp, port_clock);
+	} else {
+		*link_bw = drm_dp_link_rate_to_bw_code(port_clock);
+		*rate_select = 0;
+	}
+}
+
 bool
 intel_dp_compute_config(struct intel_encoder *encoder,
 			struct intel_crtc_state *pipe_config)
@@ -1397,6 +1410,7 @@ intel_dp_compute_config(struct intel_encoder *encoder,
 	int link_avail, link_clock;
 	int common_rates[DP_MAX_SUPPORTED_RATES] = {};
 	int common_len;
+	uint8_t link_bw, rate_select;
 
 	common_len = intel_dp_common_rates(intel_dp, common_rates);
 
@@ -1501,21 +1515,14 @@ found:
 
 	pipe_config->lane_count = lane_count;
 
-	if (intel_dp->num_sink_rates) {
-		intel_dp->link_bw = 0;
-		intel_dp->rate_select =
-			intel_dp_rate_select(intel_dp, common_rates[clock]);
-	} else {
-		intel_dp->link_bw =
-			drm_dp_link_rate_to_bw_code(common_rates[clock]);
-		intel_dp->rate_select = 0;
-	}
-
 	pipe_config->pipe_bpp = bpp;
 	pipe_config->port_clock = common_rates[clock];
 
-	DRM_DEBUG_KMS("DP link bw %02x lane count %d clock %d bpp %d\n",
-		      intel_dp->link_bw, pipe_config->lane_count,
+	intel_dp_compute_rate(intel_dp, pipe_config->port_clock,
+			      &link_bw, &rate_select);
+
+	DRM_DEBUG_KMS("DP link bw %02x rate select %02x lane count %d clock %d bpp %d\n",
+		      link_bw, rate_select, pipe_config->lane_count,
 		      pipe_config->port_clock, bpp);
 	DRM_DEBUG_KMS("DP link bw required %i available %i\n",
 		      mode_rate, link_avail);
@@ -3624,19 +3631,23 @@ intel_dp_start_link_train(struct intel_dp *intel_dp)
 	int voltage_tries, loop_tries;
 	uint32_t DP = intel_dp->DP;
 	uint8_t link_config[2];
+	uint8_t link_bw, rate_select;
 
 	if (HAS_DDI(dev))
 		intel_ddi_prepare_link_retrain(encoder);
 
+	intel_dp_compute_rate(intel_dp, crtc->config->port_clock,
+			      &link_bw, &rate_select);
+
 	/* Write the link configuration data */
-	link_config[0] = intel_dp->link_bw;
+	link_config[0] = link_bw;
 	link_config[1] = crtc->config->lane_count;
 	if (drm_dp_enhanced_frame_cap(intel_dp->dpcd))
 		link_config[1] |= DP_LANE_COUNT_ENHANCED_FRAME_EN;
 	drm_dp_dpcd_write(&intel_dp->aux, DP_LINK_BW_SET, link_config, 2);
 	if (intel_dp->num_sink_rates)
 		drm_dp_dpcd_write(&intel_dp->aux, DP_LINK_RATE_SET,
-				&intel_dp->rate_select, 1);
+				  &rate_select, 1);
 
 	link_config[0] = 0;
 	link_config[1] = DP_SET_ANSI_8B10B;
diff --git a/drivers/gpu/drm/i915/intel_dp_mst.c b/drivers/gpu/drm/i915/intel_dp_mst.c
index fdb5e31..1dd2b37 100644
--- a/drivers/gpu/drm/i915/intel_dp_mst.c
+++ b/drivers/gpu/drm/i915/intel_dp_mst.c
@@ -38,7 +38,7 @@ static bool intel_dp_mst_compute_config(struct intel_encoder *encoder,
 	struct intel_dp *intel_dp = &intel_dig_port->dp;
 	struct drm_atomic_state *state;
 	int bpp, i;
-	int lane_count, slots, rate;
+	int lane_count, slots;
 	struct drm_display_mode *adjusted_mode = &pipe_config->base.adjusted_mode;
 	struct drm_connector *drm_connector;
 	struct intel_connector *connector, *found = NULL;
@@ -55,20 +55,11 @@ static bool intel_dp_mst_compute_config(struct intel_encoder *encoder,
 	 */
 	lane_count = drm_dp_max_lane_count(intel_dp->dpcd);
 
-	rate = intel_dp_max_link_rate(intel_dp);
-
-	if (intel_dp->num_sink_rates) {
-		intel_dp->link_bw = 0;
-		intel_dp->rate_select = intel_dp_rate_select(intel_dp, rate);
-	} else {
-		intel_dp->link_bw = drm_dp_link_rate_to_bw_code(rate);
-		intel_dp->rate_select = 0;
-	}
 
 	pipe_config->lane_count = lane_count;
 
 	pipe_config->pipe_bpp = 24;
-	pipe_config->port_clock = rate;
+	pipe_config->port_clock = intel_dp_max_link_rate(intel_dp);
 
 	state = pipe_config->base.state;
 
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 0a11041..22dc8b6 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -712,8 +712,6 @@ struct intel_dp {
 	enum hdmi_force_audio force_audio;
 	bool limited_color_range;
 	bool color_range_auto;
-	uint8_t link_bw;
-	uint8_t rate_select;
 	uint8_t dpcd[DP_RECEIVER_CAP_SIZE];
 	uint8_t psr_dpcd[EDP_PSR_RECEIVER_CAP_SIZE];
 	uint8_t downstream_ports[DP_MAX_DOWNSTREAM_PORTS];
-- 
cgit v0.10.2


From 33a732f407fed464df687370d7bb4d64533f9920 Mon Sep 17 00:00:00 2001
From: Alex Dai <yu.dai@intel.com>
Date: Wed, 12 Aug 2015 15:43:36 +0100
Subject: drm/i915: GuC-specific firmware loader

This fetches the required firmware image from the filesystem,
then loads it into the GuC's memory via a dedicated DMA engine.

This patch is derived from GuC loading work originally done by
Vinit Azad and Ben Widawsky.

v2:
    Various improvements per review comments by Chris Wilson

v3:
    Removed 'wait' parameter to intel_guc_ucode_load() as firmware
        prefetch is no longer supported in the common firmware loader,
	per Daniel Vetter's request.
    Firmware checker callback fn now returns errno rather than bool.

v4:
    Squash uC-independent code into GuC-specifc loader [Daniel Vetter]
    Don't keep the driver working (by falling back to execlist mode)
        if GuC firmware loading fails [Daniel Vetter]

v5:
    Clarify WOPCM-related #defines [Tom O'Rourke]
    Delete obsolete code no longer required with current h/w & f/w
        [Tom O'Rourke]
    Move the call to intel_guc_ucode_init() later, so that it can
        allocate GEM objects, and have it fetch the firmware; then
	intel_guc_ucode_load() doesn't need to fetch it later.
        [Daniel Vetter].

v6:
    Update comment describing intel_guc_ucode_load() [Tom O'Rourke]

Issue: VIZ-4884
Signed-off-by: Alex Dai <yu.dai@intel.com>
Signed-off-by: Dave Gordon <david.s.gordon@intel.com>
Reviewed-by: Tom O'Rourke <Tom.O'Rourke@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile
index 41fb8a9..cc359e0 100644
--- a/drivers/gpu/drm/i915/Makefile
+++ b/drivers/gpu/drm/i915/Makefile
@@ -40,6 +40,9 @@ i915-y += i915_cmd_parser.o \
 	  intel_ringbuffer.o \
 	  intel_uncore.o
 
+# general-purpose microcontroller (GuC) support
+i915-y += intel_guc_loader.o
+
 # autogenerated null render state
 i915-y += intel_renderstate_gen6.o \
 	  intel_renderstate_gen7.o \
diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index ab37d11..2193cc2 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -435,6 +435,11 @@ static int i915_load_modeset_init(struct drm_device *dev)
 	 * working irqs for e.g. gmbus and dp aux transfers. */
 	intel_modeset_init(dev);
 
+	/* intel_guc_ucode_init() needs the mutex to allocate GEM objects */
+	mutex_lock(&dev->struct_mutex);
+	intel_guc_ucode_init(dev);
+	mutex_unlock(&dev->struct_mutex);
+
 	ret = i915_gem_init(dev);
 	if (ret)
 		goto cleanup_irq;
@@ -476,6 +481,9 @@ cleanup_gem:
 	i915_gem_context_fini(dev);
 	mutex_unlock(&dev->struct_mutex);
 cleanup_irq:
+	mutex_lock(&dev->struct_mutex);
+	intel_guc_ucode_fini(dev);
+	mutex_unlock(&dev->struct_mutex);
 	drm_irq_uninstall(dev);
 cleanup_gem_stolen:
 	i915_gem_cleanup_stolen(dev);
@@ -1128,6 +1136,7 @@ int i915_driver_unload(struct drm_device *dev)
 	flush_workqueue(dev_priv->wq);
 
 	mutex_lock(&dev->struct_mutex);
+	intel_guc_ucode_fini(dev);
 	i915_gem_cleanup_ringbuffer(dev);
 	i915_gem_context_fini(dev);
 	mutex_unlock(&dev->struct_mutex);
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 6b85338..6124445 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -50,6 +50,7 @@
 #include <linux/intel-iommu.h>
 #include <linux/kref.h>
 #include <linux/pm_qos.h>
+#include "intel_guc.h"
 
 /* General customization:
  */
@@ -1709,6 +1710,8 @@ struct drm_i915_private {
 
 	struct i915_virtual_gpu vgpu;
 
+	struct intel_guc guc;
+
 	struct intel_csr csr;
 
 	/* Display CSR-related protection */
@@ -1953,6 +1956,11 @@ static inline struct drm_i915_private *dev_to_i915(struct device *dev)
 	return to_i915(dev_get_drvdata(dev));
 }
 
+static inline struct drm_i915_private *guc_to_i915(struct intel_guc *guc)
+{
+	return container_of(guc, struct drm_i915_private, guc);
+}
+
 /* Iterate over initialised rings */
 #define for_each_ring(ring__, dev_priv__, i__) \
 	for ((i__) = 0; (i__) < I915_NUM_RINGS; (i__)++) \
@@ -2557,6 +2565,9 @@ struct drm_i915_cmd_table {
 
 #define HAS_CSR(dev)	(IS_SKYLAKE(dev))
 
+#define HAS_GUC_UCODE(dev)	(IS_GEN9(dev))
+#define HAS_GUC_SCHED(dev)	(IS_GEN9(dev))
+
 #define HAS_RESOURCE_STREAMER(dev) (IS_HASWELL(dev) || \
 				    INTEL_INFO(dev)->gen >= 8)
 
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 407b6b3..1ac57ec 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -4681,6 +4681,22 @@ i915_gem_init_hw(struct drm_device *dev)
 			goto out;
 	}
 
+	/* We can't enable contexts until all firmware is loaded */
+	ret = intel_guc_ucode_load(dev);
+	if (ret) {
+		/*
+		 * If we got an error and GuC submission is enabled, map
+		 * the error to -EIO so the GPU will be declared wedged.
+		 * OTOH, if we didn't intend to use the GuC anyway, just
+		 * discard the error and carry on.
+		 */
+		DRM_ERROR("Failed to initialize GuC, error %d%s\n", ret,
+			i915.enable_guc_submission ? "" : " (ignored)");
+		ret = i915.enable_guc_submission ? -EIO : 0;
+		if (ret)
+			goto out;
+	}
+
 	/* Now it is safe to go back round and do everything else: */
 	for_each_ring(ring, dev_priv, i) {
 		struct drm_i915_gem_request *req;
diff --git a/drivers/gpu/drm/i915/i915_guc_reg.h b/drivers/gpu/drm/i915/i915_guc_reg.h
index ccdc6c8..8c8e574 100644
--- a/drivers/gpu/drm/i915/i915_guc_reg.h
+++ b/drivers/gpu/drm/i915/i915_guc_reg.h
@@ -38,10 +38,6 @@
 #define   GS_MIA_SHIFT			16
 #define   GS_MIA_MASK			  (0x07 << GS_MIA_SHIFT)
 
-#define GUC_WOPCM_SIZE			0xc050
-#define   GUC_WOPCM_SIZE_VALUE  	  (0x80 << 12)	/* 512KB */
-#define GUC_WOPCM_OFFSET		0x80000		/* 512KB */
-
 #define SOFT_SCRATCH(n)			(0xc180 + ((n) * 4))
 
 #define UOS_RSA_SCRATCH_0		0xc200
@@ -56,10 +52,18 @@
 #define   UOS_MOVE			  (1<<4)
 #define   START_DMA			  (1<<0)
 #define DMA_GUC_WOPCM_OFFSET		0xc340
+#define   GUC_WOPCM_OFFSET_VALUE	  0x80000	/* 512KB */
+
+#define GUC_WOPCM_SIZE			0xc050
+#define   GUC_WOPCM_SIZE_VALUE  	  (0x80 << 12)	/* 512KB */
+
+/* GuC addresses below GUC_WOPCM_TOP don't map through the GTT */
+#define	GUC_WOPCM_TOP			(GUC_WOPCM_SIZE_VALUE)
 
 #define GEN8_GT_PM_CONFIG		0x138140
+#define GEN9LP_GT_PM_CONFIG		0x138140
 #define GEN9_GT_PM_CONFIG		0x13816c
-#define   GEN8_GT_DOORBELL_ENABLE	  (1<<0)
+#define   GT_DOORBELL_ENABLE		  (1<<0)
 
 #define GEN8_GTCR			0x4274
 #define   GEN8_GTCR_INVALIDATE		  (1<<0)
@@ -80,7 +84,8 @@
 				 GUC_ENABLE_READ_CACHE_LOGIC		| \
 				 GUC_ENABLE_MIA_CACHING			| \
 				 GUC_ENABLE_READ_CACHE_FOR_SRAM_DATA	| \
-				 GUC_ENABLE_READ_CACHE_FOR_WOPCM_DATA)
+				 GUC_ENABLE_READ_CACHE_FOR_WOPCM_DATA	| \
+				 GUC_ENABLE_MIA_CLOCK_GATING)
 
 #define HOST2GUC_INTERRUPT		0xc4c8
 #define   HOST2GUC_TRIGGER		  (1<<0)
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index f22aca3..651d0dd 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -6860,7 +6860,9 @@ enum skl_disp_power_wells {
 #define   GEN9_PGCTL_SSB_EU311_ACK	(1 << 14)
 
 #define GEN7_MISCCPCTL			(0x9424)
-#define   GEN7_DOP_CLOCK_GATE_ENABLE	(1<<0)
+#define   GEN7_DOP_CLOCK_GATE_ENABLE		(1<<0)
+#define   GEN8_DOP_CLOCK_GATE_CFCLK_ENABLE	(1<<2)
+#define   GEN8_DOP_CLOCK_GATE_GUC_ENABLE	(1<<4)
 
 #define GEN8_GARBCNTL                   0xB004
 #define   GEN9_GAPS_TSV_CREDIT_DISABLE  (1<<7)
diff --git a/drivers/gpu/drm/i915/intel_guc.h b/drivers/gpu/drm/i915/intel_guc.h
new file mode 100644
index 0000000..2846b6d
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_guc.h
@@ -0,0 +1,67 @@
+/*
+ * Copyright © 2014 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ *
+ */
+#ifndef _INTEL_GUC_H_
+#define _INTEL_GUC_H_
+
+#include "intel_guc_fwif.h"
+#include "i915_guc_reg.h"
+
+enum intel_guc_fw_status {
+	GUC_FIRMWARE_FAIL = -1,
+	GUC_FIRMWARE_NONE = 0,
+	GUC_FIRMWARE_PENDING,
+	GUC_FIRMWARE_SUCCESS
+};
+
+/*
+ * This structure encapsulates all the data needed during the process
+ * of fetching, caching, and loading the firmware image into the GuC.
+ */
+struct intel_guc_fw {
+	struct drm_device *		guc_dev;
+	const char *			guc_fw_path;
+	size_t				guc_fw_size;
+	struct drm_i915_gem_object *	guc_fw_obj;
+	enum intel_guc_fw_status	guc_fw_fetch_status;
+	enum intel_guc_fw_status	guc_fw_load_status;
+
+	uint16_t			guc_fw_major_wanted;
+	uint16_t			guc_fw_minor_wanted;
+	uint16_t			guc_fw_major_found;
+	uint16_t			guc_fw_minor_found;
+};
+
+struct intel_guc {
+	struct intel_guc_fw guc_fw;
+
+	uint32_t log_flags;
+};
+
+/* intel_guc_loader.c */
+extern void intel_guc_ucode_init(struct drm_device *dev);
+extern int intel_guc_ucode_load(struct drm_device *dev);
+extern void intel_guc_ucode_fini(struct drm_device *dev);
+extern const char *intel_guc_fw_status_repr(enum intel_guc_fw_status status);
+
+#endif
diff --git a/drivers/gpu/drm/i915/intel_guc_fwif.h b/drivers/gpu/drm/i915/intel_guc_fwif.h
index 18d7f20..06aad6f 100644
--- a/drivers/gpu/drm/i915/intel_guc_fwif.h
+++ b/drivers/gpu/drm/i915/intel_guc_fwif.h
@@ -32,9 +32,8 @@
  * EDITING THIS FILE IS THEREFORE NOT RECOMMENDED - YOUR CHANGES MAY BE LOST.
  */
 
-#define GFXCORE_FAMILY_GEN8		11
 #define GFXCORE_FAMILY_GEN9		12
-#define GFXCORE_FAMILY_FORCE_ULONG	0x7fffffff
+#define GFXCORE_FAMILY_UNKNOWN		0x7fffffff
 
 #define GUC_CTX_PRIORITY_CRITICAL	0
 #define GUC_CTX_PRIORITY_HIGH		1
diff --git a/drivers/gpu/drm/i915/intel_guc_loader.c b/drivers/gpu/drm/i915/intel_guc_loader.c
new file mode 100644
index 0000000..dd62c31
--- /dev/null
+++ b/drivers/gpu/drm/i915/intel_guc_loader.c
@@ -0,0 +1,529 @@
+/*
+ * Copyright © 2014 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ *
+ * Authors:
+ *    Vinit Azad <vinit.azad@intel.com>
+ *    Ben Widawsky <ben@bwidawsk.net>
+ *    Dave Gordon <david.s.gordon@intel.com>
+ *    Alex Dai <yu.dai@intel.com>
+ */
+#include <linux/firmware.h>
+#include "i915_drv.h"
+#include "intel_guc.h"
+
+/**
+ * DOC: GuC
+ *
+ * intel_guc:
+ * Top level structure of guc. It handles firmware loading and manages client
+ * pool and doorbells. intel_guc owns a i915_guc_client to replace the legacy
+ * ExecList submission.
+ *
+ * Firmware versioning:
+ * The firmware build process will generate a version header file with major and
+ * minor version defined. The versions are built into CSS header of firmware.
+ * i915 kernel driver set the minimal firmware version required per platform.
+ * The firmware installation package will install (symbolic link) proper version
+ * of firmware.
+ *
+ * GuC address space:
+ * GuC does not allow any gfx GGTT address that falls into range [0, WOPCM_TOP),
+ * which is reserved for Boot ROM, SRAM and WOPCM. Currently this top address is
+ * 512K. In order to exclude 0-512K address space from GGTT, all gfx objects
+ * used by GuC is pinned with PIN_OFFSET_BIAS along with size of WOPCM.
+ *
+ * Firmware log:
+ * Firmware log is enabled by setting i915.guc_log_level to non-negative level.
+ * Log data is printed out via reading debugfs i915_guc_log_dump. Reading from
+ * i915_guc_load_status will print out firmware loading status and scratch
+ * registers value.
+ *
+ */
+
+#define I915_SKL_GUC_UCODE "i915/skl_guc_ver3.bin"
+MODULE_FIRMWARE(I915_SKL_GUC_UCODE);
+
+/* User-friendly representation of an enum */
+const char *intel_guc_fw_status_repr(enum intel_guc_fw_status status)
+{
+	switch (status) {
+	case GUC_FIRMWARE_FAIL:
+		return "FAIL";
+	case GUC_FIRMWARE_NONE:
+		return "NONE";
+	case GUC_FIRMWARE_PENDING:
+		return "PENDING";
+	case GUC_FIRMWARE_SUCCESS:
+		return "SUCCESS";
+	default:
+		return "UNKNOWN!";
+	}
+};
+
+static u32 get_gttype(struct drm_i915_private *dev_priv)
+{
+	/* XXX: GT type based on PCI device ID? field seems unused by fw */
+	return 0;
+}
+
+static u32 get_core_family(struct drm_i915_private *dev_priv)
+{
+	switch (INTEL_INFO(dev_priv)->gen) {
+	case 9:
+		return GFXCORE_FAMILY_GEN9;
+
+	default:
+		DRM_ERROR("GUC: unsupported core family\n");
+		return GFXCORE_FAMILY_UNKNOWN;
+	}
+}
+
+static void set_guc_init_params(struct drm_i915_private *dev_priv)
+{
+	struct intel_guc *guc = &dev_priv->guc;
+	u32 params[GUC_CTL_MAX_DWORDS];
+	int i;
+
+	memset(&params, 0, sizeof(params));
+
+	params[GUC_CTL_DEVICE_INFO] |=
+		(get_gttype(dev_priv) << GUC_CTL_GTTYPE_SHIFT) |
+		(get_core_family(dev_priv) << GUC_CTL_COREFAMILY_SHIFT);
+
+	/*
+	 * GuC ARAT increment is 10 ns. GuC default scheduler quantum is one
+	 * second. This ARAR is calculated by:
+	 * Scheduler-Quantum-in-ns / ARAT-increment-in-ns = 1000000000 / 10
+	 */
+	params[GUC_CTL_ARAT_HIGH] = 0;
+	params[GUC_CTL_ARAT_LOW] = 100000000;
+
+	params[GUC_CTL_WA] |= GUC_CTL_WA_UK_BY_DRIVER;
+
+	params[GUC_CTL_FEATURE] |= GUC_CTL_DISABLE_SCHEDULER |
+			GUC_CTL_VCS2_ENABLED;
+
+	if (i915.guc_log_level >= 0) {
+		params[GUC_CTL_LOG_PARAMS] = guc->log_flags;
+		params[GUC_CTL_DEBUG] =
+			i915.guc_log_level << GUC_LOG_VERBOSITY_SHIFT;
+	}
+
+	I915_WRITE(SOFT_SCRATCH(0), 0);
+
+	for (i = 0; i < GUC_CTL_MAX_DWORDS; i++)
+		I915_WRITE(SOFT_SCRATCH(1 + i), params[i]);
+}
+
+/*
+ * Read the GuC status register (GUC_STATUS) and store it in the
+ * specified location; then return a boolean indicating whether
+ * the value matches either of two values representing completion
+ * of the GuC boot process.
+ *
+ * This is used for polling the GuC status in a wait_for_atomic()
+ * loop below.
+ */
+static inline bool guc_ucode_response(struct drm_i915_private *dev_priv,
+				      u32 *status)
+{
+	u32 val = I915_READ(GUC_STATUS);
+	*status = val;
+	return ((val & GS_UKERNEL_MASK) == GS_UKERNEL_READY ||
+		(val & GS_UKERNEL_MASK) == GS_UKERNEL_LAPIC_DONE);
+}
+
+/*
+ * Transfer the firmware image to RAM for execution by the microcontroller.
+ *
+ * GuC Firmware layout:
+ * +-------------------------------+  ----
+ * |          CSS header           |  128B
+ * | contains major/minor version  |
+ * +-------------------------------+  ----
+ * |             uCode             |
+ * +-------------------------------+  ----
+ * |         RSA signature         |  256B
+ * +-------------------------------+  ----
+ * |         RSA public Key        |  256B
+ * +-------------------------------+  ----
+ * |       Public key modulus      |    4B
+ * +-------------------------------+  ----
+ *
+ * Architecturally, the DMA engine is bidirectional, and can potentially even
+ * transfer between GTT locations. This functionality is left out of the API
+ * for now as there is no need for it.
+ *
+ * Note that GuC needs the CSS header plus uKernel code to be copied by the
+ * DMA engine in one operation, whereas the RSA signature is loaded via MMIO.
+ */
+
+#define UOS_CSS_HEADER_OFFSET		0
+#define UOS_VER_MINOR_OFFSET		0x44
+#define UOS_VER_MAJOR_OFFSET		0x46
+#define UOS_CSS_HEADER_SIZE		0x80
+#define UOS_RSA_SIG_SIZE		0x100
+#define UOS_CSS_SIGNING_SIZE		0x204
+
+static int guc_ucode_xfer_dma(struct drm_i915_private *dev_priv)
+{
+	struct intel_guc_fw *guc_fw = &dev_priv->guc.guc_fw;
+	struct drm_i915_gem_object *fw_obj = guc_fw->guc_fw_obj;
+	unsigned long offset;
+	struct sg_table *sg = fw_obj->pages;
+	u32 status, ucode_size, rsa[UOS_RSA_SIG_SIZE / sizeof(u32)];
+	int i, ret = 0;
+
+	/* uCode size, also is where RSA signature starts */
+	offset = ucode_size = guc_fw->guc_fw_size - UOS_CSS_SIGNING_SIZE;
+	I915_WRITE(DMA_COPY_SIZE, ucode_size);
+
+	/* Copy RSA signature from the fw image to HW for verification */
+	sg_pcopy_to_buffer(sg->sgl, sg->nents, rsa, UOS_RSA_SIG_SIZE, offset);
+	for (i = 0; i < UOS_RSA_SIG_SIZE / sizeof(u32); i++)
+		I915_WRITE(UOS_RSA_SCRATCH_0 + i * sizeof(u32), rsa[i]);
+
+	/* Set the source address for the new blob */
+	offset = i915_gem_obj_ggtt_offset(fw_obj);
+	I915_WRITE(DMA_ADDR_0_LOW, lower_32_bits(offset));
+	I915_WRITE(DMA_ADDR_0_HIGH, upper_32_bits(offset) & 0xFFFF);
+
+	/*
+	 * Set the DMA destination. Current uCode expects the code to be
+	 * loaded at 8k; locations below this are used for the stack.
+	 */
+	I915_WRITE(DMA_ADDR_1_LOW, 0x2000);
+	I915_WRITE(DMA_ADDR_1_HIGH, DMA_ADDRESS_SPACE_WOPCM);
+
+	/* Finally start the DMA */
+	I915_WRITE(DMA_CTRL, _MASKED_BIT_ENABLE(UOS_MOVE | START_DMA));
+
+	/*
+	 * Spin-wait for the DMA to complete & the GuC to start up.
+	 * NB: Docs recommend not using the interrupt for completion.
+	 * Measurements indicate this should take no more than 20ms, so a
+	 * timeout here indicates that the GuC has failed and is unusable.
+	 * (Higher levels of the driver will attempt to fall back to
+	 * execlist mode if this happens.)
+	 */
+	ret = wait_for_atomic(guc_ucode_response(dev_priv, &status), 100);
+
+	DRM_DEBUG_DRIVER("DMA status 0x%x, GuC status 0x%x\n",
+			I915_READ(DMA_CTRL), status);
+
+	if ((status & GS_BOOTROM_MASK) == GS_BOOTROM_RSA_FAILED) {
+		DRM_ERROR("GuC firmware signature verification failed\n");
+		ret = -ENOEXEC;
+	}
+
+	DRM_DEBUG_DRIVER("returning %d\n", ret);
+
+	return ret;
+}
+
+/*
+ * Load the GuC firmware blob into the MinuteIA.
+ */
+static int guc_ucode_xfer(struct drm_i915_private *dev_priv)
+{
+	struct intel_guc_fw *guc_fw = &dev_priv->guc.guc_fw;
+	struct drm_device *dev = dev_priv->dev;
+	int ret;
+
+	ret = i915_gem_object_set_to_gtt_domain(guc_fw->guc_fw_obj, false);
+	if (ret) {
+		DRM_DEBUG_DRIVER("set-domain failed %d\n", ret);
+		return ret;
+	}
+
+	ret = i915_gem_obj_ggtt_pin(guc_fw->guc_fw_obj, 0, 0);
+	if (ret) {
+		DRM_DEBUG_DRIVER("pin failed %d\n", ret);
+		return ret;
+	}
+
+	/* Invalidate GuC TLB to let GuC take the latest updates to GTT. */
+	I915_WRITE(GEN8_GTCR, GEN8_GTCR_INVALIDATE);
+
+	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
+
+	/* init WOPCM */
+	I915_WRITE(GUC_WOPCM_SIZE, GUC_WOPCM_SIZE_VALUE);
+	I915_WRITE(DMA_GUC_WOPCM_OFFSET, GUC_WOPCM_OFFSET_VALUE);
+
+	/* Enable MIA caching. GuC clock gating is disabled. */
+	I915_WRITE(GUC_SHIM_CONTROL, GUC_SHIM_CONTROL_VALUE);
+
+	/* WaC6DisallowByGfxPause*/
+	I915_WRITE(GEN6_GFXPAUSE, 0x30FFF);
+
+	if (IS_BROXTON(dev))
+		I915_WRITE(GEN9LP_GT_PM_CONFIG, GT_DOORBELL_ENABLE);
+	else
+		I915_WRITE(GEN9_GT_PM_CONFIG, GT_DOORBELL_ENABLE);
+
+	if (IS_GEN9(dev)) {
+		/* DOP Clock Gating Enable for GuC clocks */
+		I915_WRITE(GEN7_MISCCPCTL, (GEN8_DOP_CLOCK_GATE_GUC_ENABLE |
+					    I915_READ(GEN7_MISCCPCTL)));
+
+		/* allows for 5us before GT can go to RC6 */
+		I915_WRITE(GUC_ARAT_C6DIS, 0x1FF);
+	}
+
+	set_guc_init_params(dev_priv);
+
+	ret = guc_ucode_xfer_dma(dev_priv);
+
+	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+
+	/*
+	 * We keep the object pages for reuse during resume. But we can unpin it
+	 * now that DMA has completed, so it doesn't continue to take up space.
+	 */
+	i915_gem_object_ggtt_unpin(guc_fw->guc_fw_obj);
+
+	return ret;
+}
+
+/**
+ * intel_guc_ucode_load() - load GuC uCode into the device
+ * @dev:	drm device
+ *
+ * Called from gem_init_hw() during driver loading and also after a GPU reset.
+ *
+ * The firmware image should have already been fetched into memory by the
+ * earlier call to intel_guc_ucode_init(), so here we need only check that
+ * is succeeded, and then transfer the image to the h/w.
+ *
+ * Return:	non-zero code on error
+ */
+int intel_guc_ucode_load(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_guc_fw *guc_fw = &dev_priv->guc.guc_fw;
+	int err = 0;
+
+	DRM_DEBUG_DRIVER("GuC fw status: fetch %s, load %s\n",
+		intel_guc_fw_status_repr(guc_fw->guc_fw_fetch_status),
+		intel_guc_fw_status_repr(guc_fw->guc_fw_load_status));
+
+	if (guc_fw->guc_fw_fetch_status == GUC_FIRMWARE_NONE)
+		return 0;
+
+	if (guc_fw->guc_fw_fetch_status == GUC_FIRMWARE_SUCCESS &&
+	    guc_fw->guc_fw_load_status == GUC_FIRMWARE_FAIL)
+		return -ENOEXEC;
+
+	guc_fw->guc_fw_load_status = GUC_FIRMWARE_PENDING;
+
+	DRM_DEBUG_DRIVER("GuC fw fetch status %s\n",
+		intel_guc_fw_status_repr(guc_fw->guc_fw_fetch_status));
+
+	switch (guc_fw->guc_fw_fetch_status) {
+	case GUC_FIRMWARE_FAIL:
+		/* something went wrong :( */
+		err = -EIO;
+		goto fail;
+
+	case GUC_FIRMWARE_NONE:
+	case GUC_FIRMWARE_PENDING:
+	default:
+		/* "can't happen" */
+		WARN_ONCE(1, "GuC fw %s invalid guc_fw_fetch_status %s [%d]\n",
+			guc_fw->guc_fw_path,
+			intel_guc_fw_status_repr(guc_fw->guc_fw_fetch_status),
+			guc_fw->guc_fw_fetch_status);
+		err = -ENXIO;
+		goto fail;
+
+	case GUC_FIRMWARE_SUCCESS:
+		break;
+	}
+
+	err = guc_ucode_xfer(dev_priv);
+	if (err)
+		goto fail;
+
+	guc_fw->guc_fw_load_status = GUC_FIRMWARE_SUCCESS;
+
+	DRM_DEBUG_DRIVER("GuC fw status: fetch %s, load %s\n",
+		intel_guc_fw_status_repr(guc_fw->guc_fw_fetch_status),
+		intel_guc_fw_status_repr(guc_fw->guc_fw_load_status));
+
+	return 0;
+
+fail:
+	if (guc_fw->guc_fw_load_status == GUC_FIRMWARE_PENDING)
+		guc_fw->guc_fw_load_status = GUC_FIRMWARE_FAIL;
+
+	return err;
+}
+
+static void guc_fw_fetch(struct drm_device *dev, struct intel_guc_fw *guc_fw)
+{
+	struct drm_i915_gem_object *obj;
+	const struct firmware *fw;
+	const u8 *css_header;
+	const size_t minsize = UOS_CSS_HEADER_SIZE + UOS_CSS_SIGNING_SIZE;
+	const size_t maxsize = GUC_WOPCM_SIZE_VALUE + UOS_CSS_SIGNING_SIZE
+			- 0x8000; /* 32k reserved (8K stack + 24k context) */
+	int err;
+
+	DRM_DEBUG_DRIVER("before requesting firmware: GuC fw fetch status %s\n",
+		intel_guc_fw_status_repr(guc_fw->guc_fw_fetch_status));
+
+	err = request_firmware(&fw, guc_fw->guc_fw_path, &dev->pdev->dev);
+	if (err)
+		goto fail;
+	if (!fw)
+		goto fail;
+
+	DRM_DEBUG_DRIVER("fetch GuC fw from %s succeeded, fw %p\n",
+		guc_fw->guc_fw_path, fw);
+	DRM_DEBUG_DRIVER("firmware file size %zu (minimum %zu, maximum %zu)\n",
+		fw->size, minsize, maxsize);
+
+	/* Check the size of the blob befoe examining buffer contents */
+	if (fw->size < minsize || fw->size > maxsize)
+		goto fail;
+
+	/*
+	 * The GuC firmware image has the version number embedded at a well-known
+	 * offset within the firmware blob; note that major / minor version are
+	 * TWO bytes each (i.e. u16), although all pointers and offsets are defined
+	 * in terms of bytes (u8).
+	 */
+	css_header = fw->data + UOS_CSS_HEADER_OFFSET;
+	guc_fw->guc_fw_major_found = *(u16 *)(css_header + UOS_VER_MAJOR_OFFSET);
+	guc_fw->guc_fw_minor_found = *(u16 *)(css_header + UOS_VER_MINOR_OFFSET);
+
+	if (guc_fw->guc_fw_major_found != guc_fw->guc_fw_major_wanted ||
+	    guc_fw->guc_fw_minor_found < guc_fw->guc_fw_minor_wanted) {
+		DRM_ERROR("GuC firmware version %d.%d, required %d.%d\n",
+			guc_fw->guc_fw_major_found, guc_fw->guc_fw_minor_found,
+			guc_fw->guc_fw_major_wanted, guc_fw->guc_fw_minor_wanted);
+		err = -ENOEXEC;
+		goto fail;
+	}
+
+	DRM_DEBUG_DRIVER("firmware version %d.%d OK (minimum %d.%d)\n",
+			guc_fw->guc_fw_major_found, guc_fw->guc_fw_minor_found,
+			guc_fw->guc_fw_major_wanted, guc_fw->guc_fw_minor_wanted);
+
+	obj = i915_gem_object_create_from_data(dev, fw->data, fw->size);
+	if (IS_ERR_OR_NULL(obj)) {
+		err = obj ? PTR_ERR(obj) : -ENOMEM;
+		goto fail;
+	}
+
+	guc_fw->guc_fw_obj = obj;
+	guc_fw->guc_fw_size = fw->size;
+
+	DRM_DEBUG_DRIVER("GuC fw fetch status SUCCESS, obj %p\n",
+			guc_fw->guc_fw_obj);
+
+	release_firmware(fw);
+	guc_fw->guc_fw_fetch_status = GUC_FIRMWARE_SUCCESS;
+	return;
+
+fail:
+	DRM_DEBUG_DRIVER("GuC fw fetch status FAIL; err %d, fw %p, obj %p\n",
+		err, fw, guc_fw->guc_fw_obj);
+	DRM_ERROR("Failed to fetch GuC firmware from %s (error %d)\n",
+		  guc_fw->guc_fw_path, err);
+
+	obj = guc_fw->guc_fw_obj;
+	if (obj)
+		drm_gem_object_unreference(&obj->base);
+	guc_fw->guc_fw_obj = NULL;
+
+	release_firmware(fw);		/* OK even if fw is NULL */
+	guc_fw->guc_fw_fetch_status = GUC_FIRMWARE_FAIL;
+}
+
+/**
+ * intel_guc_ucode_init() - define parameters and fetch firmware
+ * @dev:	drm device
+ *
+ * Called early during driver load, but after GEM is initialised.
+ * The device struct_mutex must be held by the caller, as we're
+ * going to allocate a GEM object to hold the firmware image.
+ *
+ * The firmware will be transferred to the GuC's memory later,
+ * when intel_guc_ucode_load() is called.
+ */
+void intel_guc_ucode_init(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_guc_fw *guc_fw = &dev_priv->guc.guc_fw;
+	const char *fw_path;
+
+	if (!HAS_GUC_SCHED(dev))
+		i915.enable_guc_submission = false;
+
+	if (!HAS_GUC_UCODE(dev)) {
+		fw_path = NULL;
+	} else if (IS_SKYLAKE(dev)) {
+		fw_path = I915_SKL_GUC_UCODE;
+		guc_fw->guc_fw_major_wanted = 3;
+		guc_fw->guc_fw_minor_wanted = 0;
+	} else {
+		i915.enable_guc_submission = false;
+		fw_path = "";	/* unknown device */
+	}
+
+	guc_fw->guc_dev = dev;
+	guc_fw->guc_fw_path = fw_path;
+	guc_fw->guc_fw_fetch_status = GUC_FIRMWARE_NONE;
+	guc_fw->guc_fw_load_status = GUC_FIRMWARE_NONE;
+
+	if (fw_path == NULL)
+		return;
+
+	if (*fw_path == '\0') {
+		DRM_ERROR("No GuC firmware known for this platform\n");
+		guc_fw->guc_fw_fetch_status = GUC_FIRMWARE_FAIL;
+		return;
+	}
+
+	guc_fw->guc_fw_fetch_status = GUC_FIRMWARE_PENDING;
+	DRM_DEBUG_DRIVER("GuC firmware pending, path %s\n", fw_path);
+	guc_fw_fetch(dev, guc_fw);
+	/* status must now be FAIL or SUCCESS */
+}
+
+/**
+ * intel_guc_ucode_fini() - clean up all allocated resources
+ * @dev:	drm device
+ */
+void intel_guc_ucode_fini(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_guc_fw *guc_fw = &dev_priv->guc.guc_fw;
+
+	if (guc_fw->guc_fw_obj)
+		drm_gem_object_unreference(&guc_fw->guc_fw_obj->base);
+	guc_fw->guc_fw_obj = NULL;
+
+	guc_fw->guc_fw_fetch_status = GUC_FIRMWARE_NONE;
+}
-- 
cgit v0.10.2


From fdf5d3572fd1f2103a6bda0dba643e76a3ea978e Mon Sep 17 00:00:00 2001
From: Alex Dai <yu.dai@intel.com>
Date: Wed, 12 Aug 2015 15:43:37 +0100
Subject: drm/i915: Debugfs interface to read GuC load status

The new node provides access to the status of the GuC-specific loader;
also the scratch registers used for communication between the i915
driver and the GuC firmware.

v2:
    Changes to output formats per Chris Wilson's suggestions

v6:
    Rebased

Issue: VIZ-4884
Signed-off-by: Alex Dai <yu.dai@intel.com>
Signed-off-by: Dave Gordon <david.s.gordon@intel.com>
Reviewed-by: Tom O'Rourke <Tom.O'Rourke@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index 95e7b82..1946411 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -2374,6 +2374,44 @@ static int i915_llc(struct seq_file *m, void *data)
 	return 0;
 }
 
+static int i915_guc_load_status_info(struct seq_file *m, void *data)
+{
+	struct drm_info_node *node = m->private;
+	struct drm_i915_private *dev_priv = node->minor->dev->dev_private;
+	struct intel_guc_fw *guc_fw = &dev_priv->guc.guc_fw;
+	u32 tmp, i;
+
+	if (!HAS_GUC_UCODE(dev_priv->dev))
+		return 0;
+
+	seq_printf(m, "GuC firmware status:\n");
+	seq_printf(m, "\tpath: %s\n",
+		guc_fw->guc_fw_path);
+	seq_printf(m, "\tfetch: %s\n",
+		intel_guc_fw_status_repr(guc_fw->guc_fw_fetch_status));
+	seq_printf(m, "\tload: %s\n",
+		intel_guc_fw_status_repr(guc_fw->guc_fw_load_status));
+	seq_printf(m, "\tversion wanted: %d.%d\n",
+		guc_fw->guc_fw_major_wanted, guc_fw->guc_fw_minor_wanted);
+	seq_printf(m, "\tversion found: %d.%d\n",
+		guc_fw->guc_fw_major_found, guc_fw->guc_fw_minor_found);
+
+	tmp = I915_READ(GUC_STATUS);
+
+	seq_printf(m, "\nGuC status 0x%08x:\n", tmp);
+	seq_printf(m, "\tBootrom status = 0x%x\n",
+		(tmp & GS_BOOTROM_MASK) >> GS_BOOTROM_SHIFT);
+	seq_printf(m, "\tuKernel status = 0x%x\n",
+		(tmp & GS_UKERNEL_MASK) >> GS_UKERNEL_SHIFT);
+	seq_printf(m, "\tMIA Core status = 0x%x\n",
+		(tmp & GS_MIA_MASK) >> GS_MIA_SHIFT);
+	seq_puts(m, "\nScratch registers:\n");
+	for (i = 0; i < 16; i++)
+		seq_printf(m, "\t%2d: \t0x%x\n", i, I915_READ(SOFT_SCRATCH(i)));
+
+	return 0;
+}
+
 static int i915_edp_psr_status(struct seq_file *m, void *data)
 {
 	struct drm_info_node *node = m->private;
@@ -5033,6 +5071,7 @@ static const struct drm_info_list i915_debugfs_list[] = {
 	{"i915_gem_hws_bsd", i915_hws_info, 0, (void *)VCS},
 	{"i915_gem_hws_vebox", i915_hws_info, 0, (void *)VECS},
 	{"i915_gem_batch_pool", i915_gem_batch_pool_info, 0},
+	{"i915_guc_load_status", i915_guc_load_status_info, 0},
 	{"i915_frequency_info", i915_frequency_info, 0},
 	{"i915_hangcheck_info", i915_hangcheck_info, 0},
 	{"i915_drpc_info", i915_drpc_info, 0},
-- 
cgit v0.10.2


From 919f1f55d90b5487a9f38e94842e486509474f09 Mon Sep 17 00:00:00 2001
From: Dave Gordon <david.s.gordon@intel.com>
Date: Wed, 12 Aug 2015 15:43:38 +0100
Subject: drm/i915: Expose one LRC function for GuC submission mode

GuC submission is basically execlist submission, but with the GuC
handling the actual writes to the ELSP and the resulting context
switch interrupts.  So to describe a context for submission via
the GuC, we need one of the same functions used in execlist mode.
This commit exposes one such function, changing its name to better
describe what it does (it's related to logical ring contexts rather
than to execlists per se).

v2:
    Replaces previous "drm/i915: Move execlists defines from .c to .h"

v3:
    Incorporates a change to one of the functions exposed here that was
        previously part of an internal patch, but which was omitted from
        the version recently committed to drm-intel-nightly:
	    7a01a0a drm/i915/lrc: Update PDPx registers with lri commands
        So we reinstate this change here.

v4:
    Drop v3 change, update function parameters due to collision with
        8ee3615 drm/i915: Convert execlists_ctx_descriptor() for requests

v5:
    Don't expose execlists_update_context() after all. The current
        version is no longer compatible with GuC submission; trying to
        share the execlist version of this function results in both GuC
        and CPU updating TAIL in the context image, with bad results when
        they get out of step. The GuC submission path now has its own
        private version that just updates the ringbuffer start address,
        and not TAIL or PDPx.

v6:
    Rebased

Issue: VIZ-4884
Signed-off-by: Dave Gordon <david.s.gordon@intel.com>
Reviewed-by: Tom O'Rourke <Tom.O'Rourke@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
index 138964a..f3411f8 100644
--- a/drivers/gpu/drm/i915/intel_lrc.c
+++ b/drivers/gpu/drm/i915/intel_lrc.c
@@ -270,11 +270,11 @@ u32 intel_execlists_ctx_id(struct drm_i915_gem_object *ctx_obj)
 	return lrca >> 12;
 }
 
-static uint64_t execlists_ctx_descriptor(struct drm_i915_gem_request *rq)
+uint64_t intel_lr_context_descriptor(struct intel_context *ctx,
+				     struct intel_engine_cs *ring)
 {
-	struct intel_engine_cs *ring = rq->ring;
 	struct drm_device *dev = ring->dev;
-	struct drm_i915_gem_object *ctx_obj = rq->ctx->engine[ring->id].state;
+	struct drm_i915_gem_object *ctx_obj = ctx->engine[ring->id].state;
 	uint64_t desc;
 	uint64_t lrca = i915_gem_obj_ggtt_offset(ctx_obj);
 
@@ -312,13 +312,13 @@ static void execlists_elsp_write(struct drm_i915_gem_request *rq0,
 	uint64_t desc[2];
 
 	if (rq1) {
-		desc[1] = execlists_ctx_descriptor(rq1);
+		desc[1] = intel_lr_context_descriptor(rq1->ctx, rq1->ring);
 		rq1->elsp_submitted++;
 	} else {
 		desc[1] = 0;
 	}
 
-	desc[0] = execlists_ctx_descriptor(rq0);
+	desc[0] = intel_lr_context_descriptor(rq0->ctx, rq0->ring);
 	rq0->elsp_submitted++;
 
 	/* You must always write both descriptors in the order below. */
diff --git a/drivers/gpu/drm/i915/intel_lrc.h b/drivers/gpu/drm/i915/intel_lrc.h
index 64f89f99..5e5788c 100644
--- a/drivers/gpu/drm/i915/intel_lrc.h
+++ b/drivers/gpu/drm/i915/intel_lrc.h
@@ -74,6 +74,8 @@ int intel_lr_context_deferred_create(struct intel_context *ctx,
 void intel_lr_context_unpin(struct drm_i915_gem_request *req);
 void intel_lr_context_reset(struct drm_device *dev,
 			struct intel_context *ctx);
+uint64_t intel_lr_context_descriptor(struct intel_context *ctx,
+				     struct intel_engine_cs *ring);
 
 /* Execlists */
 int intel_sanitize_enable_execlists(struct drm_device *dev, int enable_execlists);
-- 
cgit v0.10.2


From bac427f8abfc8c11e10274d5edd9db4e4b8e74d3 Mon Sep 17 00:00:00 2001
From: Alex Dai <yu.dai@intel.com>
Date: Wed, 12 Aug 2015 15:43:39 +0100
Subject: drm/i915: Prepare for GuC-based command submission

This adds the first of the data structures used to communicate with the
GuC (the pool of guc_context structures).

We create a GuC-specific wrapper round the GEM object allocator as all
GEM objects shared with the GuC must be pinned into GGTT space at an
address that is NOT in the range [0..WOPCM_TOP), as that range of GGTT
addresses is not accessible to the GuC (from the GuC's point of view,
it's permanently reserved for other objects such as the BootROM & SRAM).

Later, we will need to allocate additional GuC-sharable objects for the
submission client(s) and the GuC's debug log.

v2:
    Remove redundant initialisation [Chris Wilson]
    Defer adding struct members until needed [Chris Wilson]
    Local functions should pass dev_priv rather than dev [Chris Wilson]

v5:
    Invalidate GuC TLB after allocating and pinning a new object

v6:
    Rebased

Issue: VIZ-4884
Signed-off-by: Alex Dai <yu.dai@intel.com>
Signed-off-by: Dave Gordon <david.s.gordon@intel.com>
Reviewed-by: Tom O'Rourke <Tom.O'Rourke@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile
index cc359e0..ddb69f3 100644
--- a/drivers/gpu/drm/i915/Makefile
+++ b/drivers/gpu/drm/i915/Makefile
@@ -41,7 +41,8 @@ i915-y += i915_cmd_parser.o \
 	  intel_uncore.o
 
 # general-purpose microcontroller (GuC) support
-i915-y += intel_guc_loader.o
+i915-y += intel_guc_loader.o \
+	  i915_guc_submission.o
 
 # autogenerated null render state
 i915-y += intel_renderstate_gen6.o \
diff --git a/drivers/gpu/drm/i915/i915_guc_submission.c b/drivers/gpu/drm/i915/i915_guc_submission.c
new file mode 100644
index 0000000..8ff59aa
--- /dev/null
+++ b/drivers/gpu/drm/i915/i915_guc_submission.c
@@ -0,0 +1,118 @@
+/*
+ * Copyright © 2014 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ *
+ */
+#include <linux/firmware.h>
+#include <linux/circ_buf.h>
+#include "i915_drv.h"
+#include "intel_guc.h"
+
+/**
+ * gem_allocate_guc_obj() - Allocate gem object for GuC usage
+ * @dev:	drm device
+ * @size:	size of object
+ *
+ * This is a wrapper to create a gem obj. In order to use it inside GuC, the
+ * object needs to be pinned lifetime. Also we must pin it to gtt space other
+ * than [0, GUC_WOPCM_TOP) because this range is reserved inside GuC.
+ *
+ * Return:	A drm_i915_gem_object if successful, otherwise NULL.
+ */
+static struct drm_i915_gem_object *gem_allocate_guc_obj(struct drm_device *dev,
+							u32 size)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_gem_object *obj;
+
+	obj = i915_gem_alloc_object(dev, size);
+	if (!obj)
+		return NULL;
+
+	if (i915_gem_object_get_pages(obj)) {
+		drm_gem_object_unreference(&obj->base);
+		return NULL;
+	}
+
+	if (i915_gem_obj_ggtt_pin(obj, PAGE_SIZE,
+			PIN_OFFSET_BIAS | GUC_WOPCM_TOP)) {
+		drm_gem_object_unreference(&obj->base);
+		return NULL;
+	}
+
+	/* Invalidate GuC TLB to let GuC take the latest updates to GTT. */
+	I915_WRITE(GEN8_GTCR, GEN8_GTCR_INVALIDATE);
+
+	return obj;
+}
+
+/**
+ * gem_release_guc_obj() - Release gem object allocated for GuC usage
+ * @obj:	gem obj to be released
+  */
+static void gem_release_guc_obj(struct drm_i915_gem_object *obj)
+{
+	if (!obj)
+		return;
+
+	if (i915_gem_obj_is_pinned(obj))
+		i915_gem_object_ggtt_unpin(obj);
+
+	drm_gem_object_unreference(&obj->base);
+}
+
+/*
+ * Set up the memory resources to be shared with the GuC.  At this point,
+ * we require just one object that can be mapped through the GGTT.
+ */
+int i915_guc_submission_init(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	const size_t ctxsize = sizeof(struct guc_context_desc);
+	const size_t poolsize = GUC_MAX_GPU_CONTEXTS * ctxsize;
+	const size_t gemsize = round_up(poolsize, PAGE_SIZE);
+	struct intel_guc *guc = &dev_priv->guc;
+
+	if (!i915.enable_guc_submission)
+		return 0; /* not enabled  */
+
+	if (guc->ctx_pool_obj)
+		return 0; /* already allocated */
+
+	guc->ctx_pool_obj = gem_allocate_guc_obj(dev_priv->dev, gemsize);
+	if (!guc->ctx_pool_obj)
+		return -ENOMEM;
+
+	ida_init(&guc->ctx_ids);
+
+	return 0;
+}
+
+void i915_guc_submission_fini(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_guc *guc = &dev_priv->guc;
+
+	if (guc->ctx_pool_obj)
+		ida_destroy(&guc->ctx_ids);
+	gem_release_guc_obj(guc->ctx_pool_obj);
+	guc->ctx_pool_obj = NULL;
+}
diff --git a/drivers/gpu/drm/i915/intel_guc.h b/drivers/gpu/drm/i915/intel_guc.h
index 2846b6d..be3cad8 100644
--- a/drivers/gpu/drm/i915/intel_guc.h
+++ b/drivers/gpu/drm/i915/intel_guc.h
@@ -56,6 +56,9 @@ struct intel_guc {
 	struct intel_guc_fw guc_fw;
 
 	uint32_t log_flags;
+
+	struct drm_i915_gem_object *ctx_pool_obj;
+	struct ida ctx_ids;
 };
 
 /* intel_guc_loader.c */
@@ -64,4 +67,8 @@ extern int intel_guc_ucode_load(struct drm_device *dev);
 extern void intel_guc_ucode_fini(struct drm_device *dev);
 extern const char *intel_guc_fw_status_repr(enum intel_guc_fw_status status);
 
+/* i915_guc_submission.c */
+int i915_guc_submission_init(struct drm_device *dev);
+void i915_guc_submission_fini(struct drm_device *dev);
+
 #endif
diff --git a/drivers/gpu/drm/i915/intel_guc_loader.c b/drivers/gpu/drm/i915/intel_guc_loader.c
index dd62c31..6ff7fea 100644
--- a/drivers/gpu/drm/i915/intel_guc_loader.c
+++ b/drivers/gpu/drm/i915/intel_guc_loader.c
@@ -128,6 +128,21 @@ static void set_guc_init_params(struct drm_i915_private *dev_priv)
 			i915.guc_log_level << GUC_LOG_VERBOSITY_SHIFT;
 	}
 
+	/* If GuC submission is enabled, set up additional parameters here */
+	if (i915.enable_guc_submission) {
+		u32 pgs = i915_gem_obj_ggtt_offset(dev_priv->guc.ctx_pool_obj);
+		u32 ctx_in_16 = GUC_MAX_GPU_CONTEXTS / 16;
+
+		pgs >>= PAGE_SHIFT;
+		params[GUC_CTL_CTXINFO] = (pgs << GUC_CTL_BASE_ADDR_SHIFT) |
+			(ctx_in_16 << GUC_CTL_CTXNUM_IN16_SHIFT);
+
+		params[GUC_CTL_FEATURE] |= GUC_CTL_KERNEL_SUBMISSIONS;
+
+		/* Unmask this bit to enable the GuC's internal scheduler */
+		params[GUC_CTL_FEATURE] &= ~GUC_CTL_DISABLE_SCHEDULER;
+	}
+
 	I915_WRITE(SOFT_SCRATCH(0), 0);
 
 	for (i = 0; i < GUC_CTL_MAX_DWORDS; i++)
@@ -360,6 +375,10 @@ int intel_guc_ucode_load(struct drm_device *dev)
 		break;
 	}
 
+	err = i915_guc_submission_init(dev);
+	if (err)
+		goto fail;
+
 	err = guc_ucode_xfer(dev_priv);
 	if (err)
 		goto fail;
@@ -521,6 +540,8 @@ void intel_guc_ucode_fini(struct drm_device *dev)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_guc_fw *guc_fw = &dev_priv->guc.guc_fw;
 
+	i915_guc_submission_fini(dev);
+
 	if (guc_fw->guc_fw_obj)
 		drm_gem_object_unreference(&guc_fw->guc_fw_obj->base);
 	guc_fw->guc_fw_obj = NULL;
-- 
cgit v0.10.2


From 4c7e77fc1007b1e12d25c6e5083369863346f5dd Mon Sep 17 00:00:00 2001
From: Alex Dai <yu.dai@intel.com>
Date: Wed, 12 Aug 2015 15:43:40 +0100
Subject: drm/i915: Enable GuC firmware log

Allocate a GEM object to hold GuC log data. A debugfs interface
(i915_guc_log_dump) is provided to print out the log content.

v2:
    Add struct members at point of use [Chris Wilson]

v6:
    Rebased

Issue: VIZ-4884
Signed-off-by: Alex Dai <yu.dai@intel.com>
Signed-off-by: Dave Gordon <david.s.gordon@intel.com>
Reviewed-by: Tom O'Rourke <Tom.O'Rourke@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index 1946411..1123b8e 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -2412,6 +2412,34 @@ static int i915_guc_load_status_info(struct seq_file *m, void *data)
 	return 0;
 }
 
+static int i915_guc_log_dump(struct seq_file *m, void *data)
+{
+	struct drm_info_node *node = m->private;
+	struct drm_device *dev = node->minor->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_i915_gem_object *log_obj = dev_priv->guc.log_obj;
+	u32 *log;
+	int i = 0, pg;
+
+	if (!log_obj)
+		return 0;
+
+	for (pg = 0; pg < log_obj->base.size / PAGE_SIZE; pg++) {
+		log = kmap_atomic(i915_gem_object_get_page(log_obj, pg));
+
+		for (i = 0; i < PAGE_SIZE / sizeof(u32); i += 4)
+			seq_printf(m, "0x%08x 0x%08x 0x%08x 0x%08x\n",
+				   *(log + i), *(log + i + 1),
+				   *(log + i + 2), *(log + i + 3));
+
+		kunmap_atomic(log);
+	}
+
+	seq_putc(m, '\n');
+
+	return 0;
+}
+
 static int i915_edp_psr_status(struct seq_file *m, void *data)
 {
 	struct drm_info_node *node = m->private;
@@ -5072,6 +5100,7 @@ static const struct drm_info_list i915_debugfs_list[] = {
 	{"i915_gem_hws_vebox", i915_hws_info, 0, (void *)VECS},
 	{"i915_gem_batch_pool", i915_gem_batch_pool_info, 0},
 	{"i915_guc_load_status", i915_guc_load_status_info, 0},
+	{"i915_guc_log_dump", i915_guc_log_dump, 0},
 	{"i915_frequency_info", i915_frequency_info, 0},
 	{"i915_hangcheck_info", i915_hangcheck_info, 0},
 	{"i915_drpc_info", i915_drpc_info, 0},
diff --git a/drivers/gpu/drm/i915/i915_guc_submission.c b/drivers/gpu/drm/i915/i915_guc_submission.c
index 8ff59aa..669c889 100644
--- a/drivers/gpu/drm/i915/i915_guc_submission.c
+++ b/drivers/gpu/drm/i915/i915_guc_submission.c
@@ -79,6 +79,47 @@ static void gem_release_guc_obj(struct drm_i915_gem_object *obj)
 	drm_gem_object_unreference(&obj->base);
 }
 
+static void guc_create_log(struct intel_guc *guc)
+{
+	struct drm_i915_private *dev_priv = guc_to_i915(guc);
+	struct drm_i915_gem_object *obj;
+	unsigned long offset;
+	uint32_t size, flags;
+
+	if (i915.guc_log_level < GUC_LOG_VERBOSITY_MIN)
+		return;
+
+	if (i915.guc_log_level > GUC_LOG_VERBOSITY_MAX)
+		i915.guc_log_level = GUC_LOG_VERBOSITY_MAX;
+
+	/* The first page is to save log buffer state. Allocate one
+	 * extra page for others in case for overlap */
+	size = (1 + GUC_LOG_DPC_PAGES + 1 +
+		GUC_LOG_ISR_PAGES + 1 +
+		GUC_LOG_CRASH_PAGES + 1) << PAGE_SHIFT;
+
+	obj = guc->log_obj;
+	if (!obj) {
+		obj = gem_allocate_guc_obj(dev_priv->dev, size);
+		if (!obj) {
+			/* logging will be off */
+			i915.guc_log_level = -1;
+			return;
+		}
+
+		guc->log_obj = obj;
+	}
+
+	/* each allocated unit is a page */
+	flags = GUC_LOG_VALID | GUC_LOG_NOTIFY_ON_HALF_FULL |
+		(GUC_LOG_DPC_PAGES << GUC_LOG_DPC_SHIFT) |
+		(GUC_LOG_ISR_PAGES << GUC_LOG_ISR_SHIFT) |
+		(GUC_LOG_CRASH_PAGES << GUC_LOG_CRASH_SHIFT);
+
+	offset = i915_gem_obj_ggtt_offset(obj) >> PAGE_SHIFT; /* in pages */
+	guc->log_flags = (offset << GUC_LOG_BUF_ADDR_SHIFT) | flags;
+}
+
 /*
  * Set up the memory resources to be shared with the GuC.  At this point,
  * we require just one object that can be mapped through the GGTT.
@@ -103,6 +144,8 @@ int i915_guc_submission_init(struct drm_device *dev)
 
 	ida_init(&guc->ctx_ids);
 
+	guc_create_log(guc);
+
 	return 0;
 }
 
@@ -111,6 +154,9 @@ void i915_guc_submission_fini(struct drm_device *dev)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_guc *guc = &dev_priv->guc;
 
+	gem_release_guc_obj(dev_priv->guc.log_obj);
+	guc->log_obj = NULL;
+
 	if (guc->ctx_pool_obj)
 		ida_destroy(&guc->ctx_ids);
 	gem_release_guc_obj(guc->ctx_pool_obj);
diff --git a/drivers/gpu/drm/i915/intel_guc.h b/drivers/gpu/drm/i915/intel_guc.h
index be3cad8..5b51b05 100644
--- a/drivers/gpu/drm/i915/intel_guc.h
+++ b/drivers/gpu/drm/i915/intel_guc.h
@@ -56,6 +56,7 @@ struct intel_guc {
 	struct intel_guc_fw guc_fw;
 
 	uint32_t log_flags;
+	struct drm_i915_gem_object *log_obj;
 
 	struct drm_i915_gem_object *ctx_pool_obj;
 	struct ida ctx_ids;
-- 
cgit v0.10.2


From 44a28b1d36762499de6fd701fcce6814eefe31d7 Mon Sep 17 00:00:00 2001
From: Dave Gordon <david.s.gordon@intel.com>
Date: Wed, 12 Aug 2015 15:43:41 +0100
Subject: drm/i915: Implementation of GuC submission client

A GuC client has its own doorbell and workqueue. It maintains the
doorbell cache line, process description object and work queue item.

A default guc_client is created for the i915 driver to use for
normal-priority in-order submission.

Note that the created client is not yet ready for use; doorbell
allocation will fail as we haven't yet linked the GuC's context
descriptor to the default contexts for each ring (see later patch).

v2:
    Defer adding structure members until needed [Chris Wilson]
    Rationalise type declarations [Chris Wilson]

v5:
    Add GuC per-engine submission & seqno statistics.
    Move wq locking to encompass both get_space() and add_item().
    Take forcewake lock in host2guc_action() [Tom O'Rourke]

v6:
    Fix GuC doorbell cacheline selection code (the
        cacheline-within-page calculation was wrong).
    Rename GuC priorities to make them closer to the names used in
        the GuC firmware source, matching what the autogenerated
        versions will (probably) be.
    Add per-ring statistics to client.

Issue: VIZ-4884
Signed-off-by: Alex Dai <yu.dai@intel.com>
Signed-off-by: Dave Gordon <david.s.gordon@intel.com>
Reviewed-by: Tom O'Rourke <Tom.O'Rourke@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_guc_submission.c b/drivers/gpu/drm/i915/i915_guc_submission.c
index 669c889..3352b85 100644
--- a/drivers/gpu/drm/i915/i915_guc_submission.c
+++ b/drivers/gpu/drm/i915/i915_guc_submission.c
@@ -27,6 +27,529 @@
 #include "intel_guc.h"
 
 /**
+ * DOC: GuC Client
+ *
+ * i915_guc_client:
+ * We use the term client to avoid confusion with contexts. A i915_guc_client is
+ * equivalent to GuC object guc_context_desc. This context descriptor is
+ * allocated from a pool of 1024 entries. Kernel driver will allocate doorbell
+ * and workqueue for it. Also the process descriptor (guc_process_desc), which
+ * is mapped to client space. So the client can write Work Item then ring the
+ * doorbell.
+ *
+ * To simplify the implementation, we allocate one gem object that contains all
+ * pages for doorbell, process descriptor and workqueue.
+ *
+ * The Scratch registers:
+ * There are 16 MMIO-based registers start from 0xC180. The kernel driver writes
+ * a value to the action register (SOFT_SCRATCH_0) along with any data. It then
+ * triggers an interrupt on the GuC via another register write (0xC4C8).
+ * Firmware writes a success/fail code back to the action register after
+ * processes the request. The kernel driver polls waiting for this update and
+ * then proceeds.
+ * See host2guc_action()
+ *
+ * Doorbells:
+ * Doorbells are interrupts to uKernel. A doorbell is a single cache line (QW)
+ * mapped into process space.
+ *
+ * Work Items:
+ * There are several types of work items that the host may place into a
+ * workqueue, each with its own requirements and limitations. Currently only
+ * WQ_TYPE_INORDER is needed to support legacy submission via GuC, which
+ * represents in-order queue. The kernel driver packs ring tail pointer and an
+ * ELSP context descriptor dword into Work Item.
+ * See guc_add_workqueue_item()
+ *
+ */
+
+/*
+ * Read GuC command/status register (SOFT_SCRATCH_0)
+ * Return true if it contains a response rather than a command
+ */
+static inline bool host2guc_action_response(struct drm_i915_private *dev_priv,
+					    u32 *status)
+{
+	u32 val = I915_READ(SOFT_SCRATCH(0));
+	*status = val;
+	return GUC2HOST_IS_RESPONSE(val);
+}
+
+static int host2guc_action(struct intel_guc *guc, u32 *data, u32 len)
+{
+	struct drm_i915_private *dev_priv = guc_to_i915(guc);
+	u32 status;
+	int i;
+	int ret;
+
+	if (WARN_ON(len < 1 || len > 15))
+		return -EINVAL;
+
+	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
+	spin_lock(&dev_priv->guc.host2guc_lock);
+
+	dev_priv->guc.action_count += 1;
+	dev_priv->guc.action_cmd = data[0];
+
+	for (i = 0; i < len; i++)
+		I915_WRITE(SOFT_SCRATCH(i), data[i]);
+
+	POSTING_READ(SOFT_SCRATCH(i - 1));
+
+	I915_WRITE(HOST2GUC_INTERRUPT, HOST2GUC_TRIGGER);
+
+	/* No HOST2GUC command should take longer than 10ms */
+	ret = wait_for_atomic(host2guc_action_response(dev_priv, &status), 10);
+	if (status != GUC2HOST_STATUS_SUCCESS) {
+		/*
+		 * Either the GuC explicitly returned an error (which
+		 * we convert to -EIO here) or no response at all was
+		 * received within the timeout limit (-ETIMEDOUT)
+		 */
+		if (ret != -ETIMEDOUT)
+			ret = -EIO;
+
+		DRM_ERROR("GUC: host2guc action 0x%X failed. ret=%d "
+				"status=0x%08X response=0x%08X\n",
+				data[0], ret, status,
+				I915_READ(SOFT_SCRATCH(15)));
+
+		dev_priv->guc.action_fail += 1;
+		dev_priv->guc.action_err = ret;
+	}
+	dev_priv->guc.action_status = status;
+
+	spin_unlock(&dev_priv->guc.host2guc_lock);
+	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+
+	return ret;
+}
+
+/*
+ * Tell the GuC to allocate or deallocate a specific doorbell
+ */
+
+static int host2guc_allocate_doorbell(struct intel_guc *guc,
+				      struct i915_guc_client *client)
+{
+	u32 data[2];
+
+	data[0] = HOST2GUC_ACTION_ALLOCATE_DOORBELL;
+	data[1] = client->ctx_index;
+
+	return host2guc_action(guc, data, 2);
+}
+
+static int host2guc_release_doorbell(struct intel_guc *guc,
+				     struct i915_guc_client *client)
+{
+	u32 data[2];
+
+	data[0] = HOST2GUC_ACTION_DEALLOCATE_DOORBELL;
+	data[1] = client->ctx_index;
+
+	return host2guc_action(guc, data, 2);
+}
+
+/*
+ * Initialise, update, or clear doorbell data shared with the GuC
+ *
+ * These functions modify shared data and so need access to the mapped
+ * client object which contains the page being used for the doorbell
+ */
+
+static void guc_init_doorbell(struct intel_guc *guc,
+			      struct i915_guc_client *client)
+{
+	struct guc_doorbell_info *doorbell;
+	void *base;
+
+	base = kmap_atomic(i915_gem_object_get_page(client->client_obj, 0));
+	doorbell = base + client->doorbell_offset;
+
+	doorbell->db_status = 1;
+	doorbell->cookie = 0;
+
+	kunmap_atomic(base);
+}
+
+static int guc_ring_doorbell(struct i915_guc_client *gc)
+{
+	struct guc_process_desc *desc;
+	union guc_doorbell_qw db_cmp, db_exc, db_ret;
+	union guc_doorbell_qw *db;
+	void *base;
+	int attempt = 2, ret = -EAGAIN;
+
+	base = kmap_atomic(i915_gem_object_get_page(gc->client_obj, 0));
+	desc = base + gc->proc_desc_offset;
+
+	/* Update the tail so it is visible to GuC */
+	desc->tail = gc->wq_tail;
+
+	/* current cookie */
+	db_cmp.db_status = GUC_DOORBELL_ENABLED;
+	db_cmp.cookie = gc->cookie;
+
+	/* cookie to be updated */
+	db_exc.db_status = GUC_DOORBELL_ENABLED;
+	db_exc.cookie = gc->cookie + 1;
+	if (db_exc.cookie == 0)
+		db_exc.cookie = 1;
+
+	/* pointer of current doorbell cacheline */
+	db = base + gc->doorbell_offset;
+
+	while (attempt--) {
+		/* lets ring the doorbell */
+		db_ret.value_qw = atomic64_cmpxchg((atomic64_t *)db,
+			db_cmp.value_qw, db_exc.value_qw);
+
+		/* if the exchange was successfully executed */
+		if (db_ret.value_qw == db_cmp.value_qw) {
+			/* db was successfully rung */
+			gc->cookie = db_exc.cookie;
+			ret = 0;
+			break;
+		}
+
+		/* XXX: doorbell was lost and need to acquire it again */
+		if (db_ret.db_status == GUC_DOORBELL_DISABLED)
+			break;
+
+		DRM_ERROR("Cookie mismatch. Expected %d, returned %d\n",
+			  db_cmp.cookie, db_ret.cookie);
+
+		/* update the cookie to newly read cookie from GuC */
+		db_cmp.cookie = db_ret.cookie;
+		db_exc.cookie = db_ret.cookie + 1;
+		if (db_exc.cookie == 0)
+			db_exc.cookie = 1;
+	}
+
+	kunmap_atomic(base);
+	return ret;
+}
+
+static void guc_disable_doorbell(struct intel_guc *guc,
+				 struct i915_guc_client *client)
+{
+	struct drm_i915_private *dev_priv = guc_to_i915(guc);
+	struct guc_doorbell_info *doorbell;
+	void *base;
+	int drbreg = GEN8_DRBREGL(client->doorbell_id);
+	int value;
+
+	base = kmap_atomic(i915_gem_object_get_page(client->client_obj, 0));
+	doorbell = base + client->doorbell_offset;
+
+	doorbell->db_status = 0;
+
+	kunmap_atomic(base);
+
+	I915_WRITE(drbreg, I915_READ(drbreg) & ~GEN8_DRB_VALID);
+
+	value = I915_READ(drbreg);
+	WARN_ON((value & GEN8_DRB_VALID) != 0);
+
+	I915_WRITE(GEN8_DRBREGU(client->doorbell_id), 0);
+	I915_WRITE(drbreg, 0);
+
+	/* XXX: wait for any interrupts */
+	/* XXX: wait for workqueue to drain */
+}
+
+/*
+ * Select, assign and relase doorbell cachelines
+ *
+ * These functions track which doorbell cachelines are in use.
+ * The data they manipulate is protected by the host2guc lock.
+ */
+
+static uint32_t select_doorbell_cacheline(struct intel_guc *guc)
+{
+	const uint32_t cacheline_size = cache_line_size();
+	uint32_t offset;
+
+	spin_lock(&guc->host2guc_lock);
+
+	/* Doorbell uses a single cache line within a page */
+	offset = offset_in_page(guc->db_cacheline);
+
+	/* Moving to next cache line to reduce contention */
+	guc->db_cacheline += cacheline_size;
+
+	spin_unlock(&guc->host2guc_lock);
+
+	DRM_DEBUG_DRIVER("selected doorbell cacheline 0x%x, next 0x%x, linesize %u\n",
+			offset, guc->db_cacheline, cacheline_size);
+
+	return offset;
+}
+
+static uint16_t assign_doorbell(struct intel_guc *guc, uint32_t priority)
+{
+	/*
+	 * The bitmap is split into two halves; the first half is used for
+	 * normal priority contexts, the second half for high-priority ones.
+	 * Note that logically higher priorities are numerically less than
+	 * normal ones, so the test below means "is it high-priority?"
+	 */
+	const bool hi_pri = (priority <= GUC_CTX_PRIORITY_HIGH);
+	const uint16_t half = GUC_MAX_DOORBELLS / 2;
+	const uint16_t start = hi_pri ? half : 0;
+	const uint16_t end = start + half;
+	uint16_t id;
+
+	spin_lock(&guc->host2guc_lock);
+	id = find_next_zero_bit(guc->doorbell_bitmap, end, start);
+	if (id == end)
+		id = GUC_INVALID_DOORBELL_ID;
+	else
+		bitmap_set(guc->doorbell_bitmap, id, 1);
+	spin_unlock(&guc->host2guc_lock);
+
+	DRM_DEBUG_DRIVER("assigned %s priority doorbell id 0x%x\n",
+			hi_pri ? "high" : "normal", id);
+
+	return id;
+}
+
+static void release_doorbell(struct intel_guc *guc, uint16_t id)
+{
+	spin_lock(&guc->host2guc_lock);
+	bitmap_clear(guc->doorbell_bitmap, id, 1);
+	spin_unlock(&guc->host2guc_lock);
+}
+
+/*
+ * Initialise the process descriptor shared with the GuC firmware.
+ */
+static void guc_init_proc_desc(struct intel_guc *guc,
+			       struct i915_guc_client *client)
+{
+	struct guc_process_desc *desc;
+	void *base;
+
+	base = kmap_atomic(i915_gem_object_get_page(client->client_obj, 0));
+	desc = base + client->proc_desc_offset;
+
+	memset(desc, 0, sizeof(*desc));
+
+	/*
+	 * XXX: pDoorbell and WQVBaseAddress are pointers in process address
+	 * space for ring3 clients (set them as in mmap_ioctl) or kernel
+	 * space for kernel clients (map on demand instead? May make debug
+	 * easier to have it mapped).
+	 */
+	desc->wq_base_addr = 0;
+	desc->db_base_addr = 0;
+
+	desc->context_id = client->ctx_index;
+	desc->wq_size_bytes = client->wq_size;
+	desc->wq_status = WQ_STATUS_ACTIVE;
+	desc->priority = client->priority;
+
+	kunmap_atomic(base);
+}
+
+/*
+ * Initialise/clear the context descriptor shared with the GuC firmware.
+ *
+ * This descriptor tells the GuC where (in GGTT space) to find the important
+ * data structures relating to this client (doorbell, process descriptor,
+ * write queue, etc).
+ */
+
+static void guc_init_ctx_desc(struct intel_guc *guc,
+			      struct i915_guc_client *client)
+{
+	struct guc_context_desc desc;
+	struct sg_table *sg;
+
+	memset(&desc, 0, sizeof(desc));
+
+	desc.attribute = GUC_CTX_DESC_ATTR_ACTIVE | GUC_CTX_DESC_ATTR_KERNEL;
+	desc.context_id = client->ctx_index;
+	desc.priority = client->priority;
+	desc.engines_used = (1 << RCS) | (1 << VCS) | (1 << BCS) |
+			    (1 << VECS) | (1 << VCS2); /* all engines */
+	desc.db_id = client->doorbell_id;
+
+	/*
+	 * The CPU address is only needed at certain points, so kmap_atomic on
+	 * demand instead of storing it in the ctx descriptor.
+	 * XXX: May make debug easier to have it mapped
+	 */
+	desc.db_trigger_cpu = 0;
+	desc.db_trigger_uk = client->doorbell_offset +
+		i915_gem_obj_ggtt_offset(client->client_obj);
+	desc.db_trigger_phy = client->doorbell_offset +
+		sg_dma_address(client->client_obj->pages->sgl);
+
+	desc.process_desc = client->proc_desc_offset +
+		i915_gem_obj_ggtt_offset(client->client_obj);
+
+	desc.wq_addr = client->wq_offset +
+		i915_gem_obj_ggtt_offset(client->client_obj);
+
+	desc.wq_size = client->wq_size;
+
+	/*
+	 * XXX: Take LRCs from an existing intel_context if this is not an
+	 * IsKMDCreatedContext client
+	 */
+	desc.desc_private = (uintptr_t)client;
+
+	/* Pool context is pinned already */
+	sg = guc->ctx_pool_obj->pages;
+	sg_pcopy_from_buffer(sg->sgl, sg->nents, &desc, sizeof(desc),
+			     sizeof(desc) * client->ctx_index);
+}
+
+static void guc_fini_ctx_desc(struct intel_guc *guc,
+			      struct i915_guc_client *client)
+{
+	struct guc_context_desc desc;
+	struct sg_table *sg;
+
+	memset(&desc, 0, sizeof(desc));
+
+	sg = guc->ctx_pool_obj->pages;
+	sg_pcopy_from_buffer(sg->sgl, sg->nents, &desc, sizeof(desc),
+			     sizeof(desc) * client->ctx_index);
+}
+
+/* Get valid workqueue item and return it back to offset */
+static int guc_get_workqueue_space(struct i915_guc_client *gc, u32 *offset)
+{
+	struct guc_process_desc *desc;
+	void *base;
+	u32 size = sizeof(struct guc_wq_item);
+	int ret = 0, timeout_counter = 200;
+
+	base = kmap_atomic(i915_gem_object_get_page(gc->client_obj, 0));
+	desc = base + gc->proc_desc_offset;
+
+	while (timeout_counter-- > 0) {
+		ret = wait_for_atomic(CIRC_SPACE(gc->wq_tail, desc->head,
+				gc->wq_size) >= size, 1);
+
+		if (!ret) {
+			*offset = gc->wq_tail;
+
+			/* advance the tail for next workqueue item */
+			gc->wq_tail += size;
+			gc->wq_tail &= gc->wq_size - 1;
+
+			/* this will break the loop */
+			timeout_counter = 0;
+		}
+	};
+
+	kunmap_atomic(base);
+
+	return ret;
+}
+
+static int guc_add_workqueue_item(struct i915_guc_client *gc,
+				  struct drm_i915_gem_request *rq)
+{
+	enum intel_ring_id ring_id = rq->ring->id;
+	struct guc_wq_item *wqi;
+	void *base;
+	u32 tail, wq_len, wq_off = 0;
+	int ret;
+
+	ret = guc_get_workqueue_space(gc, &wq_off);
+	if (ret)
+		return ret;
+
+	/* For now workqueue item is 4 DWs; workqueue buffer is 2 pages. So we
+	 * should not have the case where structure wqi is across page, neither
+	 * wrapped to the beginning. This simplifies the implementation below.
+	 *
+	 * XXX: if not the case, we need save data to a temp wqi and copy it to
+	 * workqueue buffer dw by dw.
+	 */
+	WARN_ON(sizeof(struct guc_wq_item) != 16);
+	WARN_ON(wq_off & 3);
+
+	/* wq starts from the page after doorbell / process_desc */
+	base = kmap_atomic(i915_gem_object_get_page(gc->client_obj,
+			(wq_off + GUC_DB_SIZE) >> PAGE_SHIFT));
+	wq_off &= PAGE_SIZE - 1;
+	wqi = (struct guc_wq_item *)((char *)base + wq_off);
+
+	/* len does not include the header */
+	wq_len = sizeof(struct guc_wq_item) / sizeof(u32) - 1;
+	wqi->header = WQ_TYPE_INORDER |
+			(wq_len << WQ_LEN_SHIFT) |
+			(ring_id << WQ_TARGET_SHIFT) |
+			WQ_NO_WCFLUSH_WAIT;
+
+	/* The GuC wants only the low-order word of the context descriptor */
+	wqi->context_desc = (u32)intel_lr_context_descriptor(rq->ctx, rq->ring);
+
+	/* The GuC firmware wants the tail index in QWords, not bytes */
+	tail = rq->ringbuf->tail >> 3;
+	wqi->ring_tail = tail << WQ_RING_TAIL_SHIFT;
+	wqi->fence_id = 0; /*XXX: what fence to be here */
+
+	kunmap_atomic(base);
+
+	return 0;
+}
+
+/**
+ * i915_guc_submit() - Submit commands through GuC
+ * @client:	the guc client where commands will go through
+ * @ctx:	LRC where commands come from
+ * @ring:	HW engine that will excute the commands
+ *
+ * Return:	0 if succeed
+ */
+int i915_guc_submit(struct i915_guc_client *client,
+		    struct drm_i915_gem_request *rq)
+{
+	struct intel_guc *guc = client->guc;
+	enum intel_ring_id ring_id = rq->ring->id;
+	unsigned long flags;
+	int q_ret, b_ret;
+
+	spin_lock_irqsave(&client->wq_lock, flags);
+
+	q_ret = guc_add_workqueue_item(client, rq);
+	if (q_ret == 0)
+		b_ret = guc_ring_doorbell(client);
+
+	client->submissions[ring_id] += 1;
+	if (q_ret) {
+		client->q_fail += 1;
+		client->retcode = q_ret;
+	} else if (b_ret) {
+		client->b_fail += 1;
+		client->retcode = q_ret = b_ret;
+	} else {
+		client->retcode = 0;
+	}
+	spin_unlock_irqrestore(&client->wq_lock, flags);
+
+	spin_lock(&guc->host2guc_lock);
+	guc->submissions[ring_id] += 1;
+	guc->last_seqno[ring_id] = rq->seqno;
+	spin_unlock(&guc->host2guc_lock);
+
+	return q_ret;
+}
+
+/*
+ * Everything below here is concerned with setup & teardown, and is
+ * therefore not part of the somewhat time-critical batch-submission
+ * path of i915_guc_submit() above.
+ */
+
+/**
  * gem_allocate_guc_obj() - Allocate gem object for GuC usage
  * @dev:	drm device
  * @size:	size of object
@@ -79,6 +602,121 @@ static void gem_release_guc_obj(struct drm_i915_gem_object *obj)
 	drm_gem_object_unreference(&obj->base);
 }
 
+static void guc_client_free(struct drm_device *dev,
+			    struct i915_guc_client *client)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_guc *guc = &dev_priv->guc;
+
+	if (!client)
+		return;
+
+	if (client->doorbell_id != GUC_INVALID_DOORBELL_ID) {
+		/*
+		 * First disable the doorbell, then tell the GuC we've
+		 * finished with it, finally deallocate it in our bitmap
+		 */
+		guc_disable_doorbell(guc, client);
+		host2guc_release_doorbell(guc, client);
+		release_doorbell(guc, client->doorbell_id);
+	}
+
+	/*
+	 * XXX: wait for any outstanding submissions before freeing memory.
+	 * Be sure to drop any locks
+	 */
+
+	gem_release_guc_obj(client->client_obj);
+
+	if (client->ctx_index != GUC_INVALID_CTX_ID) {
+		guc_fini_ctx_desc(guc, client);
+		ida_simple_remove(&guc->ctx_ids, client->ctx_index);
+	}
+
+	kfree(client);
+}
+
+/**
+ * guc_client_alloc() - Allocate an i915_guc_client
+ * @dev:	drm device
+ * @priority:	four levels priority _CRITICAL, _HIGH, _NORMAL and _LOW
+ * 		The kernel client to replace ExecList submission is created with
+ * 		NORMAL priority. Priority of a client for scheduler can be HIGH,
+ * 		while a preemption context can use CRITICAL.
+ *
+ * Return:	An i915_guc_client object if success.
+ */
+static struct i915_guc_client *guc_client_alloc(struct drm_device *dev,
+						uint32_t priority)
+{
+	struct i915_guc_client *client;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_guc *guc = &dev_priv->guc;
+	struct drm_i915_gem_object *obj;
+
+	client = kzalloc(sizeof(*client), GFP_KERNEL);
+	if (!client)
+		return NULL;
+
+	client->doorbell_id = GUC_INVALID_DOORBELL_ID;
+	client->priority = priority;
+	client->guc = guc;
+
+	client->ctx_index = (uint32_t)ida_simple_get(&guc->ctx_ids, 0,
+			GUC_MAX_GPU_CONTEXTS, GFP_KERNEL);
+	if (client->ctx_index >= GUC_MAX_GPU_CONTEXTS) {
+		client->ctx_index = GUC_INVALID_CTX_ID;
+		goto err;
+	}
+
+	/* The first page is doorbell/proc_desc. Two followed pages are wq. */
+	obj = gem_allocate_guc_obj(dev, GUC_DB_SIZE + GUC_WQ_SIZE);
+	if (!obj)
+		goto err;
+
+	client->client_obj = obj;
+	client->wq_offset = GUC_DB_SIZE;
+	client->wq_size = GUC_WQ_SIZE;
+	spin_lock_init(&client->wq_lock);
+
+	client->doorbell_offset = select_doorbell_cacheline(guc);
+
+	/*
+	 * Since the doorbell only requires a single cacheline, we can save
+	 * space by putting the application process descriptor in the same
+	 * page. Use the half of the page that doesn't include the doorbell.
+	 */
+	if (client->doorbell_offset >= (GUC_DB_SIZE / 2))
+		client->proc_desc_offset = 0;
+	else
+		client->proc_desc_offset = (GUC_DB_SIZE / 2);
+
+	client->doorbell_id = assign_doorbell(guc, client->priority);
+	if (client->doorbell_id == GUC_INVALID_DOORBELL_ID)
+		/* XXX: evict a doorbell instead */
+		goto err;
+
+	guc_init_proc_desc(guc, client);
+	guc_init_ctx_desc(guc, client);
+	guc_init_doorbell(guc, client);
+
+	/* XXX: Any cache flushes needed? General domain mgmt calls? */
+
+	if (host2guc_allocate_doorbell(guc, client))
+		goto err;
+
+	DRM_DEBUG_DRIVER("new priority %u client %p: ctx_index %u db_id %u\n",
+		priority, client, client->ctx_index, client->doorbell_id);
+
+	return client;
+
+err:
+	DRM_ERROR("FAILED to create priority %u GuC client!\n", priority);
+
+	guc_client_free(dev, client);
+	return NULL;
+}
+
 static void guc_create_log(struct intel_guc *guc)
 {
 	struct drm_i915_private *dev_priv = guc_to_i915(guc);
@@ -142,6 +780,8 @@ int i915_guc_submission_init(struct drm_device *dev)
 	if (!guc->ctx_pool_obj)
 		return -ENOMEM;
 
+	spin_lock_init(&dev_priv->guc.host2guc_lock);
+
 	ida_init(&guc->ctx_ids);
 
 	guc_create_log(guc);
@@ -149,6 +789,32 @@ int i915_guc_submission_init(struct drm_device *dev)
 	return 0;
 }
 
+int i915_guc_submission_enable(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_guc *guc = &dev_priv->guc;
+	struct i915_guc_client *client;
+
+	/* client for execbuf submission */
+	client = guc_client_alloc(dev, GUC_CTX_PRIORITY_KMD_NORMAL);
+	if (!client) {
+		DRM_ERROR("Failed to create execbuf guc_client\n");
+		return -ENOMEM;
+	}
+
+	guc->execbuf_client = client;
+	return 0;
+}
+
+void i915_guc_submission_disable(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_guc *guc = &dev_priv->guc;
+
+	guc_client_free(dev, guc->execbuf_client);
+	guc->execbuf_client = NULL;
+}
+
 void i915_guc_submission_fini(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
diff --git a/drivers/gpu/drm/i915/intel_guc.h b/drivers/gpu/drm/i915/intel_guc.h
index 5b51b05..92658ce 100644
--- a/drivers/gpu/drm/i915/intel_guc.h
+++ b/drivers/gpu/drm/i915/intel_guc.h
@@ -27,6 +27,31 @@
 #include "intel_guc_fwif.h"
 #include "i915_guc_reg.h"
 
+struct i915_guc_client {
+	struct drm_i915_gem_object *client_obj;
+	struct intel_guc *guc;
+	uint32_t priority;
+	uint32_t ctx_index;
+
+	uint32_t proc_desc_offset;
+	uint32_t doorbell_offset;
+	uint32_t cookie;
+	uint16_t doorbell_id;
+	uint16_t padding;		/* Maintain alignment		*/
+
+	uint32_t wq_offset;
+	uint32_t wq_size;
+
+	spinlock_t wq_lock;		/* Protects all data below	*/
+	uint32_t wq_tail;
+
+	/* GuC submission statistics & status */
+	uint64_t submissions[I915_NUM_RINGS];
+	uint32_t q_fail;
+	uint32_t b_fail;
+	int retcode;
+};
+
 enum intel_guc_fw_status {
 	GUC_FIRMWARE_FAIL = -1,
 	GUC_FIRMWARE_NONE = 0,
@@ -60,6 +85,23 @@ struct intel_guc {
 
 	struct drm_i915_gem_object *ctx_pool_obj;
 	struct ida ctx_ids;
+
+	struct i915_guc_client *execbuf_client;
+
+	spinlock_t host2guc_lock;	/* Protects all data below	*/
+
+	DECLARE_BITMAP(doorbell_bitmap, GUC_MAX_DOORBELLS);
+	uint32_t db_cacheline;		/* Cyclic counter mod pagesize	*/
+
+	/* Action status & statistics */
+	uint64_t action_count;		/* Total commands issued	*/
+	uint32_t action_cmd;		/* Last command word		*/
+	uint32_t action_status;		/* Last return status		*/
+	uint32_t action_fail;		/* Total number of failures	*/
+	int32_t action_err;		/* Last error code		*/
+
+	uint64_t submissions[I915_NUM_RINGS];
+	uint32_t last_seqno[I915_NUM_RINGS];
 };
 
 /* intel_guc_loader.c */
@@ -70,6 +112,10 @@ extern const char *intel_guc_fw_status_repr(enum intel_guc_fw_status status);
 
 /* i915_guc_submission.c */
 int i915_guc_submission_init(struct drm_device *dev);
+int i915_guc_submission_enable(struct drm_device *dev);
+int i915_guc_submit(struct i915_guc_client *client,
+		    struct drm_i915_gem_request *rq);
+void i915_guc_submission_disable(struct drm_device *dev);
 void i915_guc_submission_fini(struct drm_device *dev);
 
 #endif
diff --git a/drivers/gpu/drm/i915/intel_guc_fwif.h b/drivers/gpu/drm/i915/intel_guc_fwif.h
index 06aad6f..950c7e7 100644
--- a/drivers/gpu/drm/i915/intel_guc_fwif.h
+++ b/drivers/gpu/drm/i915/intel_guc_fwif.h
@@ -35,10 +35,10 @@
 #define GFXCORE_FAMILY_GEN9		12
 #define GFXCORE_FAMILY_UNKNOWN		0x7fffffff
 
-#define GUC_CTX_PRIORITY_CRITICAL	0
+#define GUC_CTX_PRIORITY_KMD_HIGH	0
 #define GUC_CTX_PRIORITY_HIGH		1
-#define GUC_CTX_PRIORITY_NORMAL		2
-#define GUC_CTX_PRIORITY_LOW		3
+#define GUC_CTX_PRIORITY_KMD_NORMAL	2
+#define GUC_CTX_PRIORITY_NORMAL		3
 
 #define GUC_MAX_GPU_CONTEXTS		1024
 #define	GUC_INVALID_CTX_ID		(GUC_MAX_GPU_CONTEXTS + 1)
diff --git a/drivers/gpu/drm/i915/intel_guc_loader.c b/drivers/gpu/drm/i915/intel_guc_loader.c
index 6ff7fea..8b4b057 100644
--- a/drivers/gpu/drm/i915/intel_guc_loader.c
+++ b/drivers/gpu/drm/i915/intel_guc_loader.c
@@ -342,6 +342,8 @@ int intel_guc_ucode_load(struct drm_device *dev)
 		intel_guc_fw_status_repr(guc_fw->guc_fw_fetch_status),
 		intel_guc_fw_status_repr(guc_fw->guc_fw_load_status));
 
+	i915_guc_submission_disable(dev);
+
 	if (guc_fw->guc_fw_fetch_status == GUC_FIRMWARE_NONE)
 		return 0;
 
@@ -389,12 +391,20 @@ int intel_guc_ucode_load(struct drm_device *dev)
 		intel_guc_fw_status_repr(guc_fw->guc_fw_fetch_status),
 		intel_guc_fw_status_repr(guc_fw->guc_fw_load_status));
 
+	if (i915.enable_guc_submission) {
+		err = i915_guc_submission_enable(dev);
+		if (err)
+			goto fail;
+	}
+
 	return 0;
 
 fail:
 	if (guc_fw->guc_fw_load_status == GUC_FIRMWARE_PENDING)
 		guc_fw->guc_fw_load_status = GUC_FIRMWARE_FAIL;
 
+	i915_guc_submission_disable(dev);
+
 	return err;
 }
 
-- 
cgit v0.10.2


From 4df001d3989ed319bc5abf2058dc5173bb02c53b Mon Sep 17 00:00:00 2001
From: Dave Gordon <david.s.gordon@intel.com>
Date: Wed, 12 Aug 2015 15:43:42 +0100
Subject: drm/i915: Interrupt routing for GuC submission

Turn on interrupt steering to route necessary interrupts to GuC.

v6:
    Rebased

Issue: VIZ-4884
Signed-off-by: Alex Dai <yu.dai@intel.com>
Signed-off-by: Dave Gordon <david.s.gordon@intel.com>
Reviewed-by: Tom O'Rourke <Tom.O'Rourke@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 651d0dd..5cdb328 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -1674,6 +1674,7 @@ enum skl_disp_power_wells {
 #define GFX_MODE_GEN7	0x0229c
 #define RING_MODE_GEN7(ring)	((ring)->mmio_base+0x29c)
 #define   GFX_RUN_LIST_ENABLE		(1<<15)
+#define   GFX_INTERRUPT_STEERING	(1<<14)
 #define   GFX_TLB_INVALIDATE_EXPLICIT	(1<<13)
 #define   GFX_SURFACE_FAULT_ENABLE	(1<<12)
 #define   GFX_REPLAY_MODE		(1<<11)
@@ -1681,6 +1682,11 @@ enum skl_disp_power_wells {
 #define   GFX_PPGTT_ENABLE		(1<<9)
 #define   GEN8_GFX_PPGTT_48B		(1<<7)
 
+#define   GFX_FORWARD_VBLANK_MASK	(3<<5)
+#define   GFX_FORWARD_VBLANK_NEVER	(0<<5)
+#define   GFX_FORWARD_VBLANK_ALWAYS	(1<<5)
+#define   GFX_FORWARD_VBLANK_COND	(2<<5)
+
 #define VLV_DISPLAY_BASE 0x180000
 #define VLV_MIPI_BASE VLV_DISPLAY_BASE
 
@@ -5695,11 +5701,12 @@ enum skl_disp_power_wells {
 #define GEN8_GT_IIR(which) (0x44308 + (0x10 * (which)))
 #define GEN8_GT_IER(which) (0x4430c + (0x10 * (which)))
 
-#define GEN8_BCS_IRQ_SHIFT 16
 #define GEN8_RCS_IRQ_SHIFT 0
-#define GEN8_VCS2_IRQ_SHIFT 16
+#define GEN8_BCS_IRQ_SHIFT 16
 #define GEN8_VCS1_IRQ_SHIFT 0
+#define GEN8_VCS2_IRQ_SHIFT 16
 #define GEN8_VECS_IRQ_SHIFT 0
+#define GEN8_WD_IRQ_SHIFT 16
 
 #define GEN8_DE_PIPE_ISR(pipe) (0x44400 + (0x10 * (pipe)))
 #define GEN8_DE_PIPE_IMR(pipe) (0x44404 + (0x10 * (pipe)))
diff --git a/drivers/gpu/drm/i915/intel_guc_loader.c b/drivers/gpu/drm/i915/intel_guc_loader.c
index 8b4b057..13e75f6 100644
--- a/drivers/gpu/drm/i915/intel_guc_loader.c
+++ b/drivers/gpu/drm/i915/intel_guc_loader.c
@@ -79,6 +79,53 @@ const char *intel_guc_fw_status_repr(enum intel_guc_fw_status status)
 	}
 };
 
+static void direct_interrupts_to_host(struct drm_i915_private *dev_priv)
+{
+	struct intel_engine_cs *ring;
+	int i, irqs;
+
+	/* tell all command streamers NOT to forward interrupts and vblank to GuC */
+	irqs = _MASKED_FIELD(GFX_FORWARD_VBLANK_MASK, GFX_FORWARD_VBLANK_NEVER);
+	irqs |= _MASKED_BIT_DISABLE(GFX_INTERRUPT_STEERING);
+	for_each_ring(ring, dev_priv, i)
+		I915_WRITE(RING_MODE_GEN7(ring), irqs);
+
+	/* tell DE to send nothing to GuC */
+	I915_WRITE(DE_GUCRMR, ~0);
+
+	/* route all GT interrupts to the host */
+	I915_WRITE(GUC_BCS_RCS_IER, 0);
+	I915_WRITE(GUC_VCS2_VCS1_IER, 0);
+	I915_WRITE(GUC_WD_VECS_IER, 0);
+}
+
+static void direct_interrupts_to_guc(struct drm_i915_private *dev_priv)
+{
+	struct intel_engine_cs *ring;
+	int i, irqs;
+
+	/* tell all command streamers to forward interrupts and vblank to GuC */
+	irqs = _MASKED_FIELD(GFX_FORWARD_VBLANK_MASK, GFX_FORWARD_VBLANK_ALWAYS);
+	irqs |= _MASKED_BIT_ENABLE(GFX_INTERRUPT_STEERING);
+	for_each_ring(ring, dev_priv, i)
+		I915_WRITE(RING_MODE_GEN7(ring), irqs);
+
+	/* tell DE to send (all) flip_done to GuC */
+	irqs = DERRMR_PIPEA_PRI_FLIP_DONE | DERRMR_PIPEA_SPR_FLIP_DONE |
+	       DERRMR_PIPEB_PRI_FLIP_DONE | DERRMR_PIPEB_SPR_FLIP_DONE |
+	       DERRMR_PIPEC_PRI_FLIP_DONE | DERRMR_PIPEC_SPR_FLIP_DONE;
+	/* Unmasked bits will cause GuC response message to be sent */
+	I915_WRITE(DE_GUCRMR, ~irqs);
+
+	/* route USER_INTERRUPT to Host, all others are sent to GuC. */
+	irqs = GT_RENDER_USER_INTERRUPT << GEN8_RCS_IRQ_SHIFT |
+	       GT_RENDER_USER_INTERRUPT << GEN8_BCS_IRQ_SHIFT;
+	/* These three registers have the same bit definitions */
+	I915_WRITE(GUC_BCS_RCS_IER, ~irqs);
+	I915_WRITE(GUC_VCS2_VCS1_IER, ~irqs);
+	I915_WRITE(GUC_WD_VECS_IER, ~irqs);
+}
+
 static u32 get_gttype(struct drm_i915_private *dev_priv)
 {
 	/* XXX: GT type based on PCI device ID? field seems unused by fw */
@@ -342,6 +389,7 @@ int intel_guc_ucode_load(struct drm_device *dev)
 		intel_guc_fw_status_repr(guc_fw->guc_fw_fetch_status),
 		intel_guc_fw_status_repr(guc_fw->guc_fw_load_status));
 
+	direct_interrupts_to_host(dev_priv);
 	i915_guc_submission_disable(dev);
 
 	if (guc_fw->guc_fw_fetch_status == GUC_FIRMWARE_NONE)
@@ -395,6 +443,7 @@ int intel_guc_ucode_load(struct drm_device *dev)
 		err = i915_guc_submission_enable(dev);
 		if (err)
 			goto fail;
+		direct_interrupts_to_guc(dev_priv);
 	}
 
 	return 0;
@@ -403,6 +452,7 @@ fail:
 	if (guc_fw->guc_fw_load_status == GUC_FIRMWARE_PENDING)
 		guc_fw->guc_fw_load_status = GUC_FIRMWARE_FAIL;
 
+	direct_interrupts_to_host(dev_priv);
 	i915_guc_submission_disable(dev);
 
 	return err;
@@ -550,6 +600,7 @@ void intel_guc_ucode_fini(struct drm_device *dev)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_guc_fw *guc_fw = &dev_priv->guc.guc_fw;
 
+	direct_interrupts_to_host(dev_priv);
 	i915_guc_submission_fini(dev);
 
 	if (guc_fw->guc_fw_obj)
-- 
cgit v0.10.2


From d1675198ed1f21aec6e036336e4340c40b726497 Mon Sep 17 00:00:00 2001
From: Alex Dai <yu.dai@intel.com>
Date: Wed, 12 Aug 2015 15:43:43 +0100
Subject: drm/i915: Integrate GuC-based command submission

GuC-based submission is mostly the same as execlist mode, up to
intel_logical_ring_advance_and_submit(), where the context being
dispatched would be added to the execlist queue; at this point
we submit the context to the GuC backend instead.

There are, however, a few other changes also required, notably:
1.  Contexts must be pinned at GGTT addresses accessible by the GuC
    i.e. NOT in the range [0..WOPCM_SIZE), so we have to add the
    PIN_OFFSET_BIAS flag to the relevant GGTT-pinning calls.

2.  The GuC's TLB must be invalidated after a context is pinned at
    a new GGTT address.

3.  GuC firmware uses the one page before Ring Context as shared data.
    Therefore, whenever driver wants to get base address of LRC, we
    will offset one page for it. LRC_PPHWSP_PN is defined as the page
    number of LRCA.

4.  In the work queue used to pass requests to the GuC, the GuC
    firmware requires the ring-tail-offset to be represented as an
    11-bit value, expressed in QWords. Therefore, the ringbuffer
    size must be reduced to the representable range (4 pages).

v2:
    Defer adding #defines until needed [Chris Wilson]
    Rationalise type declarations [Chris Wilson]

v4:
    Squashed kerneldoc patch into here [Daniel Vetter]

v5:
    Update request->tail in code common to both GuC and execlist modes.
    Add a private version of lr_context_update(), as sharing the
        execlist version leads to race conditions when the CPU and
        the GuC both update TAIL in the context image.
    Conversion of error-captured HWS page to string must account
        for offset from start of object to actual HWS (LRC_PPHWSP_PN).

Issue: VIZ-4884
Signed-off-by: Alex Dai <yu.dai@intel.com>
Signed-off-by: Dave Gordon <david.s.gordon@intel.com>
Reviewed-by: Tom O'Rourke <Tom.O'Rourke@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/Documentation/DocBook/drm.tmpl b/Documentation/DocBook/drm.tmpl
index 9ddf8c6..069958c 100644
--- a/Documentation/DocBook/drm.tmpl
+++ b/Documentation/DocBook/drm.tmpl
@@ -4238,6 +4238,20 @@ int num_ioctls;</synopsis>
       </sect2>
     </sect1>
     <sect1>
+      <title>GuC-based Command Submission</title>
+      <sect2>
+        <title>GuC</title>
+!Pdrivers/gpu/drm/i915/intel_guc_loader.c GuC-specific firmware loader
+!Idrivers/gpu/drm/i915/intel_guc_loader.c
+      </sect2>
+      <sect2>
+        <title>GuC Client</title>
+!Pdrivers/gpu/drm/i915/intel_guc_submission.c GuC-based command submissison
+!Idrivers/gpu/drm/i915/intel_guc_submission.c
+      </sect2>
+    </sect1>
+
+    <sect1>
       <title> Tracing </title>
       <para>
     This sections covers all things related to the tracepoints implemented in
diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index 1123b8e..a815fcc 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -1995,7 +1995,7 @@ static void i915_dump_lrc_obj(struct seq_file *m,
 		return;
 	}
 
-	page = i915_gem_object_get_page(ctx_obj, 1);
+	page = i915_gem_object_get_page(ctx_obj, LRC_STATE_PN);
 	if (!WARN_ON(page == NULL)) {
 		reg_state = kmap_atomic(page);
 
diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
index 0d0a7b1..493e9b2 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.c
+++ b/drivers/gpu/drm/i915/i915_gpu_error.c
@@ -462,17 +462,17 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m,
 		}
 
 		if ((obj = error->ring[i].hws_page)) {
-			err_printf(m, "%s --- HW Status = 0x%08x\n",
-				   dev_priv->ring[i].name,
-				   lower_32_bits(obj->gtt_offset));
+			err_printf(m, "%s --- HW Status = 0x%08llx\n",
+				dev_priv->ring[i].name,
+				obj->gtt_offset + LRC_PPHWSP_PN * PAGE_SIZE);
 			offset = 0;
 			for (elt = 0; elt < PAGE_SIZE/16; elt += 4) {
 				err_printf(m, "[%04x] %08x %08x %08x %08x\n",
 					   offset,
-					   obj->pages[0][elt],
-					   obj->pages[0][elt+1],
-					   obj->pages[0][elt+2],
-					   obj->pages[0][elt+3]);
+					   obj->pages[LRC_PPHWSP_PN][elt],
+					   obj->pages[LRC_PPHWSP_PN][elt+1],
+					   obj->pages[LRC_PPHWSP_PN][elt+2],
+					   obj->pages[LRC_PPHWSP_PN][elt+3]);
 					offset += 16;
 			}
 		}
diff --git a/drivers/gpu/drm/i915/i915_guc_submission.c b/drivers/gpu/drm/i915/i915_guc_submission.c
index 3352b85..ec70393 100644
--- a/drivers/gpu/drm/i915/i915_guc_submission.c
+++ b/drivers/gpu/drm/i915/i915_guc_submission.c
@@ -364,18 +364,58 @@ static void guc_init_proc_desc(struct intel_guc *guc,
 static void guc_init_ctx_desc(struct intel_guc *guc,
 			      struct i915_guc_client *client)
 {
+	struct intel_context *ctx = client->owner;
 	struct guc_context_desc desc;
 	struct sg_table *sg;
+	int i;
 
 	memset(&desc, 0, sizeof(desc));
 
 	desc.attribute = GUC_CTX_DESC_ATTR_ACTIVE | GUC_CTX_DESC_ATTR_KERNEL;
 	desc.context_id = client->ctx_index;
 	desc.priority = client->priority;
-	desc.engines_used = (1 << RCS) | (1 << VCS) | (1 << BCS) |
-			    (1 << VECS) | (1 << VCS2); /* all engines */
 	desc.db_id = client->doorbell_id;
 
+	for (i = 0; i < I915_NUM_RINGS; i++) {
+		struct guc_execlist_context *lrc = &desc.lrc[i];
+		struct intel_ringbuffer *ringbuf = ctx->engine[i].ringbuf;
+		struct intel_engine_cs *ring;
+		struct drm_i915_gem_object *obj;
+		uint64_t ctx_desc;
+
+		/* TODO: We have a design issue to be solved here. Only when we
+		 * receive the first batch, we know which engine is used by the
+		 * user. But here GuC expects the lrc and ring to be pinned. It
+		 * is not an issue for default context, which is the only one
+		 * for now who owns a GuC client. But for future owner of GuC
+		 * client, need to make sure lrc is pinned prior to enter here.
+		 */
+		obj = ctx->engine[i].state;
+		if (!obj)
+			break;	/* XXX: continue? */
+
+		ring = ringbuf->ring;
+		ctx_desc = intel_lr_context_descriptor(ctx, ring);
+		lrc->context_desc = (u32)ctx_desc;
+
+		/* The state page is after PPHWSP */
+		lrc->ring_lcra = i915_gem_obj_ggtt_offset(obj) +
+				LRC_STATE_PN * PAGE_SIZE;
+		lrc->context_id = (client->ctx_index << GUC_ELC_CTXID_OFFSET) |
+				(ring->id << GUC_ELC_ENGINE_OFFSET);
+
+		obj = ringbuf->obj;
+
+		lrc->ring_begin = i915_gem_obj_ggtt_offset(obj);
+		lrc->ring_end = lrc->ring_begin + obj->base.size - 1;
+		lrc->ring_next_free_location = lrc->ring_begin;
+		lrc->ring_current_tail_pointer_value = 0;
+
+		desc.engines_used |= (1 << ring->id);
+	}
+
+	WARN_ON(desc.engines_used == 0);
+
 	/*
 	 * The CPU address is only needed at certain points, so kmap_atomic on
 	 * demand instead of storing it in the ctx descriptor.
@@ -501,6 +541,29 @@ static int guc_add_workqueue_item(struct i915_guc_client *gc,
 	return 0;
 }
 
+#define CTX_RING_BUFFER_START		0x08
+
+/* Update the ringbuffer pointer in a saved context image */
+static void lr_context_update(struct drm_i915_gem_request *rq)
+{
+	enum intel_ring_id ring_id = rq->ring->id;
+	struct drm_i915_gem_object *ctx_obj = rq->ctx->engine[ring_id].state;
+	struct drm_i915_gem_object *rb_obj = rq->ringbuf->obj;
+	struct page *page;
+	uint32_t *reg_state;
+
+	BUG_ON(!ctx_obj);
+	WARN_ON(!i915_gem_obj_is_pinned(ctx_obj));
+	WARN_ON(!i915_gem_obj_is_pinned(rb_obj));
+
+	page = i915_gem_object_get_page(ctx_obj, LRC_STATE_PN);
+	reg_state = kmap_atomic(page);
+
+	reg_state[CTX_RING_BUFFER_START+1] = i915_gem_obj_ggtt_offset(rb_obj);
+
+	kunmap_atomic(reg_state);
+}
+
 /**
  * i915_guc_submit() - Submit commands through GuC
  * @client:	the guc client where commands will go through
@@ -517,6 +580,10 @@ int i915_guc_submit(struct i915_guc_client *client,
 	unsigned long flags;
 	int q_ret, b_ret;
 
+	/* Need this because of the deferred pin ctx and ring */
+	/* Shall we move this right after ring is pinned? */
+	lr_context_update(rq);
+
 	spin_lock_irqsave(&client->wq_lock, flags);
 
 	q_ret = guc_add_workqueue_item(client, rq);
@@ -643,11 +710,13 @@ static void guc_client_free(struct drm_device *dev,
  * 		The kernel client to replace ExecList submission is created with
  * 		NORMAL priority. Priority of a client for scheduler can be HIGH,
  * 		while a preemption context can use CRITICAL.
+ * @ctx		the context to own the client (we use the default render context)
  *
  * Return:	An i915_guc_client object if success.
  */
 static struct i915_guc_client *guc_client_alloc(struct drm_device *dev,
-						uint32_t priority)
+						uint32_t priority,
+						struct intel_context *ctx)
 {
 	struct i915_guc_client *client;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -660,6 +729,7 @@ static struct i915_guc_client *guc_client_alloc(struct drm_device *dev,
 
 	client->doorbell_id = GUC_INVALID_DOORBELL_ID;
 	client->priority = priority;
+	client->owner = ctx;
 	client->guc = guc;
 
 	client->ctx_index = (uint32_t)ida_simple_get(&guc->ctx_ids, 0,
@@ -793,10 +863,11 @@ int i915_guc_submission_enable(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_guc *guc = &dev_priv->guc;
+	struct intel_context *ctx = dev_priv->ring[RCS].default_context;
 	struct i915_guc_client *client;
 
 	/* client for execbuf submission */
-	client = guc_client_alloc(dev, GUC_CTX_PRIORITY_KMD_NORMAL);
+	client = guc_client_alloc(dev, GUC_CTX_PRIORITY_KMD_NORMAL, ctx);
 	if (!client) {
 		DRM_ERROR("Failed to create execbuf guc_client\n");
 		return -ENOMEM;
diff --git a/drivers/gpu/drm/i915/intel_guc.h b/drivers/gpu/drm/i915/intel_guc.h
index 92658ce..4ec2d27 100644
--- a/drivers/gpu/drm/i915/intel_guc.h
+++ b/drivers/gpu/drm/i915/intel_guc.h
@@ -29,6 +29,7 @@
 
 struct i915_guc_client {
 	struct drm_i915_gem_object *client_obj;
+	struct intel_context *owner;
 	struct intel_guc *guc;
 	uint32_t priority;
 	uint32_t ctx_index;
diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
index f3411f8..e77b6b0 100644
--- a/drivers/gpu/drm/i915/intel_lrc.c
+++ b/drivers/gpu/drm/i915/intel_lrc.c
@@ -263,7 +263,8 @@ int intel_sanitize_enable_execlists(struct drm_device *dev, int enable_execlists
  */
 u32 intel_execlists_ctx_id(struct drm_i915_gem_object *ctx_obj)
 {
-	u32 lrca = i915_gem_obj_ggtt_offset(ctx_obj);
+	u32 lrca = i915_gem_obj_ggtt_offset(ctx_obj) +
+			LRC_PPHWSP_PN * PAGE_SIZE;
 
 	/* LRCA is required to be 4K aligned so the more significant 20 bits
 	 * are globally unique */
@@ -276,7 +277,8 @@ uint64_t intel_lr_context_descriptor(struct intel_context *ctx,
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_gem_object *ctx_obj = ctx->engine[ring->id].state;
 	uint64_t desc;
-	uint64_t lrca = i915_gem_obj_ggtt_offset(ctx_obj);
+	uint64_t lrca = i915_gem_obj_ggtt_offset(ctx_obj) +
+			LRC_PPHWSP_PN * PAGE_SIZE;
 
 	WARN_ON(lrca & 0xFFFFFFFF00000FFFULL);
 
@@ -350,7 +352,7 @@ static int execlists_update_context(struct drm_i915_gem_request *rq)
 	WARN_ON(!i915_gem_obj_is_pinned(ctx_obj));
 	WARN_ON(!i915_gem_obj_is_pinned(rb_obj));
 
-	page = i915_gem_object_get_page(ctx_obj, 1);
+	page = i915_gem_object_get_page(ctx_obj, LRC_STATE_PN);
 	reg_state = kmap_atomic(page);
 
 	reg_state[CTX_RING_TAIL+1] = rq->tail;
@@ -548,8 +550,6 @@ static int execlists_context_queue(struct drm_i915_gem_request *request)
 
 	i915_gem_request_reference(request);
 
-	request->tail = request->ringbuf->tail;
-
 	spin_lock_irq(&ring->execlist_lock);
 
 	list_for_each_entry(cursor, &ring->execlist_queue, execlist_link)
@@ -702,13 +702,19 @@ static void
 intel_logical_ring_advance_and_submit(struct drm_i915_gem_request *request)
 {
 	struct intel_engine_cs *ring = request->ring;
+	struct drm_i915_private *dev_priv = request->i915;
 
 	intel_logical_ring_advance(request->ringbuf);
 
+	request->tail = request->ringbuf->tail;
+
 	if (intel_ring_stopped(ring))
 		return;
 
-	execlists_context_queue(request);
+	if (dev_priv->guc.execbuf_client)
+		i915_guc_submit(dev_priv->guc.execbuf_client, request);
+	else
+		execlists_context_queue(request);
 }
 
 static void __wrap_ring_buffer(struct intel_ringbuffer *ringbuf)
@@ -998,6 +1004,7 @@ int logical_ring_flush_all_caches(struct drm_i915_gem_request *req)
 
 static int intel_lr_context_pin(struct drm_i915_gem_request *rq)
 {
+	struct drm_i915_private *dev_priv = rq->i915;
 	struct intel_engine_cs *ring = rq->ring;
 	struct drm_i915_gem_object *ctx_obj = rq->ctx->engine[ring->id].state;
 	struct intel_ringbuffer *ringbuf = rq->ringbuf;
@@ -1005,14 +1012,18 @@ static int intel_lr_context_pin(struct drm_i915_gem_request *rq)
 
 	WARN_ON(!mutex_is_locked(&ring->dev->struct_mutex));
 	if (rq->ctx->engine[ring->id].pin_count++ == 0) {
-		ret = i915_gem_obj_ggtt_pin(ctx_obj,
-				GEN8_LR_CONTEXT_ALIGN, 0);
+		ret = i915_gem_obj_ggtt_pin(ctx_obj, GEN8_LR_CONTEXT_ALIGN,
+				PIN_OFFSET_BIAS | GUC_WOPCM_TOP);
 		if (ret)
 			goto reset_pin_count;
 
 		ret = intel_pin_and_map_ringbuffer_obj(ring->dev, ringbuf);
 		if (ret)
 			goto unpin_ctx_obj;
+
+		/* Invalidate GuC TLB. */
+		if (i915.enable_guc_submission)
+			I915_WRITE(GEN8_GTCR, GEN8_GTCR_INVALIDATE);
 	}
 
 	return ret;
@@ -2137,7 +2148,7 @@ populate_lr_context(struct intel_context *ctx, struct drm_i915_gem_object *ctx_o
 
 	/* The second page of the context object contains some fields which must
 	 * be set up prior to the first execution. */
-	page = i915_gem_object_get_page(ctx_obj, 1);
+	page = i915_gem_object_get_page(ctx_obj, LRC_STATE_PN);
 	reg_state = kmap_atomic(page);
 
 	/* A context is actually a big batch buffer with several MI_LOAD_REGISTER_IMM
@@ -2307,12 +2318,13 @@ static void lrc_setup_hardware_status_page(struct intel_engine_cs *ring,
 		struct drm_i915_gem_object *default_ctx_obj)
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
+	struct page *page;
 
-	/* The status page is offset 0 from the default context object
-	 * in LRC mode. */
-	ring->status_page.gfx_addr = i915_gem_obj_ggtt_offset(default_ctx_obj);
-	ring->status_page.page_addr =
-			kmap(sg_page(default_ctx_obj->pages->sgl));
+	/* The HWSP is part of the default context object in LRC mode. */
+	ring->status_page.gfx_addr = i915_gem_obj_ggtt_offset(default_ctx_obj)
+			+ LRC_PPHWSP_PN * PAGE_SIZE;
+	page = i915_gem_object_get_page(default_ctx_obj, LRC_PPHWSP_PN);
+	ring->status_page.page_addr = kmap(page);
 	ring->status_page.obj = default_ctx_obj;
 
 	I915_WRITE(RING_HWS_PGA(ring->mmio_base),
@@ -2338,6 +2350,7 @@ int intel_lr_context_deferred_create(struct intel_context *ctx,
 {
 	const bool is_global_default_ctx = (ctx == ring->default_context);
 	struct drm_device *dev = ring->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_gem_object *ctx_obj;
 	uint32_t context_size;
 	struct intel_ringbuffer *ringbuf;
@@ -2348,6 +2361,9 @@ int intel_lr_context_deferred_create(struct intel_context *ctx,
 
 	context_size = round_up(get_lr_context_size(ring), 4096);
 
+	/* One extra page as the sharing data between driver and GuC */
+	context_size += PAGE_SIZE * LRC_PPHWSP_PN;
+
 	ctx_obj = i915_gem_alloc_object(dev, context_size);
 	if (!ctx_obj) {
 		DRM_DEBUG_DRIVER("Alloc LRC backing obj failed.\n");
@@ -2355,13 +2371,18 @@ int intel_lr_context_deferred_create(struct intel_context *ctx,
 	}
 
 	if (is_global_default_ctx) {
-		ret = i915_gem_obj_ggtt_pin(ctx_obj, GEN8_LR_CONTEXT_ALIGN, 0);
+		ret = i915_gem_obj_ggtt_pin(ctx_obj, GEN8_LR_CONTEXT_ALIGN,
+				PIN_OFFSET_BIAS | GUC_WOPCM_TOP);
 		if (ret) {
 			DRM_DEBUG_DRIVER("Pin LRC backing obj failed: %d\n",
 					ret);
 			drm_gem_object_unreference(&ctx_obj->base);
 			return ret;
 		}
+
+		/* Invalidate GuC TLB. */
+		if (i915.enable_guc_submission)
+			I915_WRITE(GEN8_GTCR, GEN8_GTCR_INVALIDATE);
 	}
 
 	ringbuf = kzalloc(sizeof(*ringbuf), GFP_KERNEL);
@@ -2374,7 +2395,7 @@ int intel_lr_context_deferred_create(struct intel_context *ctx,
 
 	ringbuf->ring = ring;
 
-	ringbuf->size = 32 * PAGE_SIZE;
+	ringbuf->size = 4 * PAGE_SIZE;
 	ringbuf->effective_size = ringbuf->size;
 	ringbuf->head = 0;
 	ringbuf->tail = 0;
@@ -2474,7 +2495,7 @@ void intel_lr_context_reset(struct drm_device *dev,
 			WARN(1, "Failed get_pages for context obj\n");
 			continue;
 		}
-		page = i915_gem_object_get_page(ctx_obj, 1);
+		page = i915_gem_object_get_page(ctx_obj, LRC_STATE_PN);
 		reg_state = kmap_atomic(page);
 
 		reg_state[CTX_RING_HEAD+1] = 0;
diff --git a/drivers/gpu/drm/i915/intel_lrc.h b/drivers/gpu/drm/i915/intel_lrc.h
index 5e5788c..4cc54b3 100644
--- a/drivers/gpu/drm/i915/intel_lrc.h
+++ b/drivers/gpu/drm/i915/intel_lrc.h
@@ -68,6 +68,12 @@ static inline void intel_logical_ring_emit(struct intel_ringbuffer *ringbuf,
 }
 
 /* Logical Ring Contexts */
+
+/* One extra page is added before LRC for GuC as shared data */
+#define LRC_GUCSHR_PN	(0)
+#define LRC_PPHWSP_PN	(LRC_GUCSHR_PN + 1)
+#define LRC_STATE_PN	(LRC_PPHWSP_PN + 1)
+
 void intel_lr_context_free(struct intel_context *ctx);
 int intel_lr_context_deferred_create(struct intel_context *ctx,
 				     struct intel_engine_cs *ring);
-- 
cgit v0.10.2


From 8b417c266b715b3797cd3e65342149372b9ac0c8 Mon Sep 17 00:00:00 2001
From: Dave Gordon <david.s.gordon@intel.com>
Date: Wed, 12 Aug 2015 15:43:44 +0100
Subject: drm/i915: Debugfs interface for GuC submission statistics

This provides a means of reading status and counts relating
to GuC actions and submissions.

v2:
    Remove surplus blank line in output [Chris Wilson]

v5:
    Added GuC per-engine submission & seqno statistics

v6:
    Add per-ring statistics to client, refactor client-dumper.

Signed-off-by: Dave Gordon <david.s.gordon@intel.com>
Signed-off-by: Alex Dai <yu.dai@intel.com>
Reviewed-by: Tom O'Rourke <Tom.O'Rourke@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index a815fcc..e192f02 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -2412,6 +2412,81 @@ static int i915_guc_load_status_info(struct seq_file *m, void *data)
 	return 0;
 }
 
+static void i915_guc_client_info(struct seq_file *m,
+				 struct drm_i915_private *dev_priv,
+				 struct i915_guc_client *client)
+{
+	struct intel_engine_cs *ring;
+	uint64_t tot = 0;
+	uint32_t i;
+
+	seq_printf(m, "\tPriority %d, GuC ctx index: %u, PD offset 0x%x\n",
+		client->priority, client->ctx_index, client->proc_desc_offset);
+	seq_printf(m, "\tDoorbell id %d, offset: 0x%x, cookie 0x%x\n",
+		client->doorbell_id, client->doorbell_offset, client->cookie);
+	seq_printf(m, "\tWQ size %d, offset: 0x%x, tail %d\n",
+		client->wq_size, client->wq_offset, client->wq_tail);
+
+	seq_printf(m, "\tFailed to queue: %u\n", client->q_fail);
+	seq_printf(m, "\tFailed doorbell: %u\n", client->b_fail);
+	seq_printf(m, "\tLast submission result: %d\n", client->retcode);
+
+	for_each_ring(ring, dev_priv, i) {
+		seq_printf(m, "\tSubmissions: %llu %s\n",
+				client->submissions[i],
+				ring->name);
+		tot += client->submissions[i];
+	}
+	seq_printf(m, "\tTotal: %llu\n", tot);
+}
+
+static int i915_guc_info(struct seq_file *m, void *data)
+{
+	struct drm_info_node *node = m->private;
+	struct drm_device *dev = node->minor->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_guc guc;
+	struct i915_guc_client client = { .client_obj = 0 };
+	struct intel_engine_cs *ring;
+	enum intel_ring_id i;
+	u64 total = 0;
+
+	if (!HAS_GUC_SCHED(dev_priv->dev))
+		return 0;
+
+	/* Take a local copy of the GuC data, so we can dump it at leisure */
+	spin_lock(&dev_priv->guc.host2guc_lock);
+	guc = dev_priv->guc;
+	if (guc.execbuf_client) {
+		spin_lock(&guc.execbuf_client->wq_lock);
+		client = *guc.execbuf_client;
+		spin_unlock(&guc.execbuf_client->wq_lock);
+	}
+	spin_unlock(&dev_priv->guc.host2guc_lock);
+
+	seq_printf(m, "GuC total action count: %llu\n", guc.action_count);
+	seq_printf(m, "GuC action failure count: %u\n", guc.action_fail);
+	seq_printf(m, "GuC last action command: 0x%x\n", guc.action_cmd);
+	seq_printf(m, "GuC last action status: 0x%x\n", guc.action_status);
+	seq_printf(m, "GuC last action error code: %d\n", guc.action_err);
+
+	seq_printf(m, "\nGuC submissions:\n");
+	for_each_ring(ring, dev_priv, i) {
+		seq_printf(m, "\t%-24s: %10llu, last seqno 0x%08x %9d\n",
+			ring->name, guc.submissions[i],
+			guc.last_seqno[i], guc.last_seqno[i]);
+		total += guc.submissions[i];
+	}
+	seq_printf(m, "\t%s: %llu\n", "Total", total);
+
+	seq_printf(m, "\nGuC execbuf client @ %p:\n", guc.execbuf_client);
+	i915_guc_client_info(m, dev_priv, &client);
+
+	/* Add more as required ... */
+
+	return 0;
+}
+
 static int i915_guc_log_dump(struct seq_file *m, void *data)
 {
 	struct drm_info_node *node = m->private;
@@ -5099,6 +5174,7 @@ static const struct drm_info_list i915_debugfs_list[] = {
 	{"i915_gem_hws_bsd", i915_hws_info, 0, (void *)VCS},
 	{"i915_gem_hws_vebox", i915_hws_info, 0, (void *)VECS},
 	{"i915_gem_batch_pool", i915_gem_batch_pool_info, 0},
+	{"i915_guc_info", i915_guc_info, 0},
 	{"i915_guc_load_status", i915_guc_load_status_info, 0},
 	{"i915_guc_log_dump", i915_guc_log_dump, 0},
 	{"i915_frequency_info", i915_frequency_info, 0},
-- 
cgit v0.10.2


From 66e2806656b12bbb98ff36b7beb1538b87cb2f2f Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 14 Aug 2015 15:59:53 +0300
Subject: drm/i915: remove excessive scaler debugging messages

There's so much scaler debugging messages that it makes other debugging
hard. Remove them.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_atomic.c b/drivers/gpu/drm/i915/intel_atomic.c
index e2531cf..9336e80 100644
--- a/drivers/gpu/drm/i915/intel_atomic.c
+++ b/drivers/gpu/drm/i915/intel_atomic.c
@@ -149,9 +149,6 @@ int intel_atomic_setup_scalers(struct drm_device *dev,
 	int i, j;
 
 	num_scalers_need = hweight32(scaler_state->scaler_users);
-	DRM_DEBUG_KMS("crtc_state = %p need = %d avail = %d scaler_users = 0x%x\n",
-		crtc_state, num_scalers_need, intel_crtc->num_scalers,
-		scaler_state->scaler_users);
 
 	/*
 	 * High level flow:
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 40e70f2..364104d 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -2933,8 +2933,6 @@ static void skl_detach_scaler(struct intel_crtc *intel_crtc, int id)
 	I915_WRITE(SKL_PS_CTRL(intel_crtc->pipe, id), 0);
 	I915_WRITE(SKL_PS_WIN_POS(intel_crtc->pipe, id), 0);
 	I915_WRITE(SKL_PS_WIN_SZ(intel_crtc->pipe, id), 0);
-	DRM_DEBUG_KMS("CRTC:%d Disabled scaler id %u.%u\n",
-		intel_crtc->base.base.id, intel_crtc->pipe, id);
 }
 
 /*
-- 
cgit v0.10.2


From cff4f55bf4fb1e32483885bf0e571847668d9d67 Mon Sep 17 00:00:00 2001
From: Graham Whaley <graham.whaley@linux.intel.com>
Date: Mon, 24 Aug 2015 14:41:21 +0100
Subject: doc: drm: Fix mis-spelling of i915_guc_submission includes

In commit
 d1675198e: drm/i915: Integrate GuC-based command submission

the drm.tmpl include lines reference the intel_guc_submission.c but the
patch adds the file i915_guc_submission.c. drm.tmpl fails to build with:
 docproc: .//drivers/gpu/drm/i915/intel_guc_submission.c:
  No such file or directory

Change the file reference to the actual file.

Signed-off-by: Graham Whaley <graham.whaley@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/Documentation/DocBook/drm.tmpl b/Documentation/DocBook/drm.tmpl
index 069958c..1f9dc3e 100644
--- a/Documentation/DocBook/drm.tmpl
+++ b/Documentation/DocBook/drm.tmpl
@@ -4246,8 +4246,8 @@ int num_ioctls;</synopsis>
       </sect2>
       <sect2>
         <title>GuC Client</title>
-!Pdrivers/gpu/drm/i915/intel_guc_submission.c GuC-based command submissison
-!Idrivers/gpu/drm/i915/intel_guc_submission.c
+!Pdrivers/gpu/drm/i915/i915_guc_submission.c GuC-based command submissison
+!Idrivers/gpu/drm/i915/i915_guc_submission.c
       </sect2>
     </sect1>
 
-- 
cgit v0.10.2


From f1afe24f0e736b9d7f2275e2b1504af3fe612f2a Mon Sep 17 00:00:00 2001
From: Arun Siluvery <arun.siluvery@linux.intel.com>
Date: Tue, 4 Aug 2015 16:22:20 +0100
Subject: drm/i915: Change SRM, LRM instructions to use correct length

MI_STORE_REGISTER_MEM, MI_LOAD_REGISTER_MEM instructions are not really
variable length instructions unlike MI_LOAD_REGISTER_IMM where it expects
(reg, addr) pairs so use fixed length for these instructions.

v2: rebase

Cc: Dave Gordon <david.s.gordon@intel.com>
Signed-off-by: Arun Siluvery <arun.siluvery@linux.intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
[danvet: Appease checkpatch as Mika spotted in i915_reg.h - it seems
terminally unhappy about i915_cmd_parser.c so that would be a separate
patch.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_cmd_parser.c b/drivers/gpu/drm/i915/i915_cmd_parser.c
index 237ff68..3977bf3 100644
--- a/drivers/gpu/drm/i915/i915_cmd_parser.c
+++ b/drivers/gpu/drm/i915/i915_cmd_parser.c
@@ -94,7 +94,7 @@
 #define CMD(op, opm, f, lm, fl, ...)				\
 	{							\
 		.flags = (fl) | ((f) ? CMD_DESC_FIXED : 0),	\
-		.cmd = { (op), (opm) }, 			\
+		.cmd = { (op), (opm) },				\
 		.length = { (lm) },				\
 		__VA_ARGS__					\
 	}
@@ -124,14 +124,14 @@ static const struct drm_i915_cmd_descriptor common_cmds[] = {
 	CMD(  MI_STORE_DWORD_INDEX,             SMI,   !F,  0xFF,   R  ),
 	CMD(  MI_LOAD_REGISTER_IMM(1),          SMI,   !F,  0xFF,   W,
 	      .reg = { .offset = 1, .mask = 0x007FFFFC, .step = 2 }    ),
-	CMD(  MI_STORE_REGISTER_MEM(1),         SMI,   !F,  0xFF,   W | B,
+	CMD(  MI_STORE_REGISTER_MEM,            SMI,    F,  1,     W | B,
 	      .reg = { .offset = 1, .mask = 0x007FFFFC },
 	      .bits = {{
 			.offset = 0,
 			.mask = MI_GLOBAL_GTT,
 			.expected = 0,
 	      }},						       ),
-	CMD(  MI_LOAD_REGISTER_MEM(1),             SMI,   !F,  0xFF,   W | B,
+	CMD(  MI_LOAD_REGISTER_MEM,             SMI,    F,  1,     W | B,
 	      .reg = { .offset = 1, .mask = 0x007FFFFC },
 	      .bits = {{
 			.offset = 0,
@@ -1021,7 +1021,7 @@ static bool check_cmd(const struct intel_engine_cs *ring,
 			 * only MI_LOAD_REGISTER_IMM commands.
 			 */
 			if (reg_addr == OACONTROL) {
-				if (desc->cmd.value == MI_LOAD_REGISTER_MEM(1)) {
+				if (desc->cmd.value == MI_LOAD_REGISTER_MEM) {
 					DRM_DEBUG_DRIVER("CMD: Rejected LRM to OACONTROL\n");
 					return false;
 				}
@@ -1035,7 +1035,7 @@ static bool check_cmd(const struct intel_engine_cs *ring,
 			 * allowed mask/value pair given in the whitelist entry.
 			 */
 			if (reg->mask) {
-				if (desc->cmd.value == MI_LOAD_REGISTER_MEM(1)) {
+				if (desc->cmd.value == MI_LOAD_REGISTER_MEM) {
 					DRM_DEBUG_DRIVER("CMD: Rejected LRM to masked register 0x%08X\n",
 							 reg_addr);
 					return false;
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 5cdb328..211b73a 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -352,8 +352,8 @@
  */
 #define MI_LOAD_REGISTER_IMM(x)	MI_INSTR(0x22, 2*(x)-1)
 #define   MI_LRI_FORCE_POSTED		(1<<12)
-#define MI_STORE_REGISTER_MEM(x) MI_INSTR(0x24, 2*(x)-1)
-#define MI_STORE_REGISTER_MEM_GEN8(x) MI_INSTR(0x24, 3*(x)-1)
+#define MI_STORE_REGISTER_MEM        MI_INSTR(0x24, 1)
+#define MI_STORE_REGISTER_MEM_GEN8   MI_INSTR(0x24, 2)
 #define   MI_SRM_LRM_GLOBAL_GTT		(1<<22)
 #define MI_FLUSH_DW		MI_INSTR(0x26, 1) /* for GEN6 */
 #define   MI_FLUSH_DW_STORE_INDEX	(1<<21)
@@ -364,8 +364,8 @@
 #define   MI_INVALIDATE_BSD		(1<<7)
 #define   MI_FLUSH_DW_USE_GTT		(1<<2)
 #define   MI_FLUSH_DW_USE_PPGTT		(0<<2)
-#define MI_LOAD_REGISTER_MEM(x) MI_INSTR(0x29, 2*(x)-1)
-#define MI_LOAD_REGISTER_MEM_GEN8(x) MI_INSTR(0x29, 3*(x)-1)
+#define MI_LOAD_REGISTER_MEM	   MI_INSTR(0x29, 1)
+#define MI_LOAD_REGISTER_MEM_GEN8  MI_INSTR(0x29, 2)
 #define MI_BATCH_BUFFER		MI_INSTR(0x30, 1)
 #define   MI_BATCH_NON_SECURE		(1)
 /* for snb/ivb/vlv this also means "batch in ppgtt" when ppgtt is enabled. */
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 364104d..542fb8c 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -11031,10 +11031,10 @@ static int intel_gen7_queue_flip(struct drm_device *dev,
 					DERRMR_PIPEB_PRI_FLIP_DONE |
 					DERRMR_PIPEC_PRI_FLIP_DONE));
 		if (IS_GEN8(dev))
-			intel_ring_emit(ring, MI_STORE_REGISTER_MEM_GEN8(1) |
+			intel_ring_emit(ring, MI_STORE_REGISTER_MEM_GEN8 |
 					      MI_SRM_LRM_GLOBAL_GTT);
 		else
-			intel_ring_emit(ring, MI_STORE_REGISTER_MEM(1) |
+			intel_ring_emit(ring, MI_STORE_REGISTER_MEM |
 					      MI_SRM_LRM_GLOBAL_GTT);
 		intel_ring_emit(ring, DERRMR);
 		intel_ring_emit(ring, ring->scratch.gtt_offset + 256);
diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
index e77b6b0..8e9a85b 100644
--- a/drivers/gpu/drm/i915/intel_lrc.c
+++ b/drivers/gpu/drm/i915/intel_lrc.c
@@ -1130,7 +1130,7 @@ static inline int gen8_emit_flush_coherentl3_wa(struct intel_engine_cs *ring,
 	if (IS_SKYLAKE(ring->dev) && INTEL_REVID(ring->dev) <= SKL_REVID_E0)
 		l3sqc4_flush |= GEN8_LQSC_RO_PERF_DIS;
 
-	wa_ctx_emit(batch, index, (MI_STORE_REGISTER_MEM_GEN8(1) |
+	wa_ctx_emit(batch, index, (MI_STORE_REGISTER_MEM_GEN8 |
 				   MI_SRM_LRM_GLOBAL_GTT));
 	wa_ctx_emit(batch, index, GEN8_L3SQCREG4);
 	wa_ctx_emit(batch, index, ring->scratch.gtt_offset + 256);
@@ -1148,7 +1148,7 @@ static inline int gen8_emit_flush_coherentl3_wa(struct intel_engine_cs *ring,
 	wa_ctx_emit(batch, index, 0);
 	wa_ctx_emit(batch, index, 0);
 
-	wa_ctx_emit(batch, index, (MI_LOAD_REGISTER_MEM_GEN8(1) |
+	wa_ctx_emit(batch, index, (MI_LOAD_REGISTER_MEM_GEN8 |
 				   MI_SRM_LRM_GLOBAL_GTT));
 	wa_ctx_emit(batch, index, GEN8_L3SQCREG4);
 	wa_ctx_emit(batch, index, ring->scratch.gtt_offset + 256);
-- 
cgit v0.10.2


From 8be6ca8537e1230da8e92c753df4125151a3f6b1 Mon Sep 17 00:00:00 2001
From: Rodrigo Vivi <rodrigo.vivi@intel.com>
Date: Mon, 24 Aug 2015 16:38:23 -0700
Subject: drm/i915: Also call frontbuffer flip when disabling planes.

We also need to call the frontbuffer flip to trigger proper
invalidations when disabling planes. Otherwise we will miss
screen updates when disabling sprites or cursor.

On core platforms where HW tracking also works, this issue
is totally masked because HW tracking triggers PSR exit
however on VLV/CHV that has only SW tracking we miss screen
updates when disabling planes.

It was caught with kms_psr_sink_crc sprite_plane_onoff
and cursor_plane_onoff subtests running on VLV/CHV.

This is probably a regression since I can also get this
with the manual test case, but with so many changes on atomic
modeset I couldn't track exactly when this was introduced.

Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 542fb8c..cf8dfa5 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -11627,7 +11627,7 @@ int intel_plane_atomic_calc_changes(struct drm_crtc_state *crtc_state,
 		intel_crtc->atomic.update_wm_pre = true;
 	}
 
-	if (visible)
+	if (visible || was_visible)
 		intel_crtc->atomic.fb_bits |=
 			to_intel_plane(plane)->frontbuffer_bit;
 
-- 
cgit v0.10.2


From 319404df2f682ddabf98cbf6421bb43b951ca978 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Fri, 14 Aug 2015 18:35:27 +0300
Subject: drm/i915/bxt: work around HW coherency issue when accessing GPU seqno

By running igt/store_dword_loop_render on BXT we can hit a coherency
problem where the seqno written at GPU command completion time is not
seen by the CPU. This results in __i915_wait_request seeing the stale
seqno and not completing the request (not considering the lost
interrupt/GPU reset mechanism). I also verified that this isn't a case
of a lost interrupt, or that the command didn't complete somehow: when
the coherency issue occured I read the seqno via an uncached GTT mapping
too. While the cached version of the seqno still showed the stale value
the one read via the uncached mapping was the correct one.

Work around this issue by clflushing the corresponding CPU cacheline
following any store of the seqno and preceding any reading of it. When
reading it do this only when the caller expects a coherent view.

v2:
- fix using the proper logical && instead of a bitwise & (Jani, Mika)
- limit the workaround to A stepping, on later steppings this HW issue
  is fixed
v3:
- use a separate get_seqno/set_seqno vfunc (Chris)

Testcase: igt/store_dword_loop_render
Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
index 8e9a85b..1af84c5 100644
--- a/drivers/gpu/drm/i915/intel_lrc.c
+++ b/drivers/gpu/drm/i915/intel_lrc.c
@@ -1710,6 +1710,34 @@ static void gen8_set_seqno(struct intel_engine_cs *ring, u32 seqno)
 	intel_write_status_page(ring, I915_GEM_HWS_INDEX, seqno);
 }
 
+static u32 bxt_a_get_seqno(struct intel_engine_cs *ring, bool lazy_coherency)
+{
+
+	/*
+	 * On BXT A steppings there is a HW coherency issue whereby the
+	 * MI_STORE_DATA_IMM storing the completed request's seqno
+	 * occasionally doesn't invalidate the CPU cache. Work around this by
+	 * clflushing the corresponding cacheline whenever the caller wants
+	 * the coherency to be guaranteed. Note that this cacheline is known
+	 * to be clean at this point, since we only write it in
+	 * bxt_a_set_seqno(), where we also do a clflush after the write. So
+	 * this clflush in practice becomes an invalidate operation.
+	 */
+
+	if (!lazy_coherency)
+		intel_flush_status_page(ring, I915_GEM_HWS_INDEX);
+
+	return intel_read_status_page(ring, I915_GEM_HWS_INDEX);
+}
+
+static void bxt_a_set_seqno(struct intel_engine_cs *ring, u32 seqno)
+{
+	intel_write_status_page(ring, I915_GEM_HWS_INDEX, seqno);
+
+	/* See bxt_a_get_seqno() explaining the reason for the clflush. */
+	intel_flush_status_page(ring, I915_GEM_HWS_INDEX);
+}
+
 static int gen8_emit_request(struct drm_i915_gem_request *request)
 {
 	struct intel_ringbuffer *ringbuf = request->ringbuf;
@@ -1879,8 +1907,13 @@ static int logical_render_ring_init(struct drm_device *dev)
 		ring->init_hw = gen8_init_render_ring;
 	ring->init_context = gen8_init_rcs_context;
 	ring->cleanup = intel_fini_pipe_control;
-	ring->get_seqno = gen8_get_seqno;
-	ring->set_seqno = gen8_set_seqno;
+	if (IS_BROXTON(dev) && INTEL_REVID(dev) < BXT_REVID_B0) {
+		ring->get_seqno = bxt_a_get_seqno;
+		ring->set_seqno = bxt_a_set_seqno;
+	} else {
+		ring->get_seqno = gen8_get_seqno;
+		ring->set_seqno = gen8_set_seqno;
+	}
 	ring->emit_request = gen8_emit_request;
 	ring->emit_flush = gen8_emit_flush_render;
 	ring->irq_get = gen8_logical_ring_get_irq;
@@ -1926,8 +1959,13 @@ static int logical_bsd_ring_init(struct drm_device *dev)
 		GT_CONTEXT_SWITCH_INTERRUPT << GEN8_VCS1_IRQ_SHIFT;
 
 	ring->init_hw = gen8_init_common_ring;
-	ring->get_seqno = gen8_get_seqno;
-	ring->set_seqno = gen8_set_seqno;
+	if (IS_BROXTON(dev) && INTEL_REVID(dev) < BXT_REVID_B0) {
+		ring->get_seqno = bxt_a_get_seqno;
+		ring->set_seqno = bxt_a_set_seqno;
+	} else {
+		ring->get_seqno = gen8_get_seqno;
+		ring->set_seqno = gen8_set_seqno;
+	}
 	ring->emit_request = gen8_emit_request;
 	ring->emit_flush = gen8_emit_flush;
 	ring->irq_get = gen8_logical_ring_get_irq;
@@ -1976,8 +2014,13 @@ static int logical_blt_ring_init(struct drm_device *dev)
 		GT_CONTEXT_SWITCH_INTERRUPT << GEN8_BCS_IRQ_SHIFT;
 
 	ring->init_hw = gen8_init_common_ring;
-	ring->get_seqno = gen8_get_seqno;
-	ring->set_seqno = gen8_set_seqno;
+	if (IS_BROXTON(dev) && INTEL_REVID(dev) < BXT_REVID_B0) {
+		ring->get_seqno = bxt_a_get_seqno;
+		ring->set_seqno = bxt_a_set_seqno;
+	} else {
+		ring->get_seqno = gen8_get_seqno;
+		ring->set_seqno = gen8_set_seqno;
+	}
 	ring->emit_request = gen8_emit_request;
 	ring->emit_flush = gen8_emit_flush;
 	ring->irq_get = gen8_logical_ring_get_irq;
@@ -2001,8 +2044,13 @@ static int logical_vebox_ring_init(struct drm_device *dev)
 		GT_CONTEXT_SWITCH_INTERRUPT << GEN8_VECS_IRQ_SHIFT;
 
 	ring->init_hw = gen8_init_common_ring;
-	ring->get_seqno = gen8_get_seqno;
-	ring->set_seqno = gen8_set_seqno;
+	if (IS_BROXTON(dev) && INTEL_REVID(dev) < BXT_REVID_B0) {
+		ring->get_seqno = bxt_a_get_seqno;
+		ring->set_seqno = bxt_a_set_seqno;
+	} else {
+		ring->get_seqno = gen8_get_seqno;
+		ring->set_seqno = gen8_set_seqno;
+	}
 	ring->emit_request = gen8_emit_request;
 	ring->emit_flush = gen8_emit_flush;
 	ring->irq_get = gen8_logical_ring_get_irq;
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h
index 2e85fda..95b0b4b 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.h
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.h
@@ -377,6 +377,13 @@ intel_ring_sync_index(struct intel_engine_cs *ring,
 	return idx;
 }
 
+static inline void
+intel_flush_status_page(struct intel_engine_cs *ring, int reg)
+{
+	drm_clflush_virt_range(&ring->status_page.page_addr[reg],
+			       sizeof(uint32_t));
+}
+
 static inline u32
 intel_read_status_page(struct intel_engine_cs *ring,
 		       int reg)
-- 
cgit v0.10.2


From e5756c10d841ddb448293c849392f3d6b809561f Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Fri, 14 Aug 2015 18:43:30 +0300
Subject: drm/i915/bxt: don't allow cached GEM mappings on A stepping

Due to a coherency issue on BXT A steppings we can't guarantee a
coherent view of cached (CPU snooped) GPU mappings, so fail such
requests. User space is supposed to fall back to uncached mappings in
this case.

v2:
- limit the WA to A steppings, on later stepping this HW issue is fixed
v3:
- return error instead of trying to work around the issue in kernel,
  since that could confuse user space (Chris)

Testcast: igt/gem_store_dword_batches_loop/cached-mapping
Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 1ac57ec..41263cd 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -3742,6 +3742,15 @@ int i915_gem_set_caching_ioctl(struct drm_device *dev, void *data,
 		level = I915_CACHE_NONE;
 		break;
 	case I915_CACHING_CACHED:
+		/*
+		 * Due to a HW issue on BXT A stepping, GPU stores via a
+		 * snooped mapping may leave stale data in a corresponding CPU
+		 * cacheline, whereas normally such cachelines would get
+		 * invalidated.
+		 */
+		if (IS_BROXTON(dev) && INTEL_REVID(dev) < BXT_REVID_B0)
+			return -ENODEV;
+
 		level = I915_CACHE_LLC;
 		break;
 	case I915_CACHING_DISPLAY:
-- 
cgit v0.10.2


From 901c2daf05c8ae6c3f85370fc96b9b6892f5da2d Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 17 Aug 2015 18:05:12 +0300
Subject: drm/i915: Put back lane_count into intel_dp and add link_rate too
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

With MST there won't be a crtc assigned to the main link encoder, so
trying to dig up the pipe_config from there is a recipe for an oops.

Instead store the parameters (lane_count and link_rate) in the encoder,
and use those values during link training etc. Since those parameters
are now assigned only when the link is actually enabled,
.compute_config() won't clobber them as it did before.

Hardware state readout is still bonkers though as we don't transfer the
link parameters from pipe_config intel_dp. We should do that during
encoder sanitation. But since we don't even do a proper job of reading
out the main link encoder state for MST there's littel point in
worrying about this now.

Fixes a regression with MST caused by:
 commit 90a6b7b052b1aa17fbb98b049e9c8b7f729c35a7
 Author: Ville Syrjälä <ville.syrjala@linux.intel.com>
 Date:   Mon Jul 6 16:39:15 2015 +0300

    drm/i915: Move intel_dp->lane_count into pipe_config

v2: Different apporoach that should keep intel_dp_check_mst_status()
    somewhat less oopsy

Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reported-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Tested-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
index 56d778f..5dff8b7 100644
--- a/drivers/gpu/drm/i915/intel_ddi.c
+++ b/drivers/gpu/drm/i915/intel_ddi.c
@@ -728,11 +728,10 @@ void intel_ddi_init_dp_buf_reg(struct intel_encoder *encoder)
 	struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base);
 	struct intel_digital_port *intel_dig_port =
 		enc_to_dig_port(&encoder->base);
-	struct intel_crtc *crtc = to_intel_crtc(encoder->base.crtc);
 
 	intel_dp->DP = intel_dig_port->saved_port_bits |
 		DDI_BUF_CTL_ENABLE | DDI_BUF_TRANS_SELECT(0);
-	intel_dp->DP |= DDI_PORT_WIDTH(crtc->config->lane_count);
+	intel_dp->DP |= DDI_PORT_WIDTH(intel_dp->lane_count);
 }
 
 static struct intel_encoder *
@@ -2314,6 +2313,8 @@ static void intel_ddi_pre_enable(struct intel_encoder *intel_encoder)
 	if (type == INTEL_OUTPUT_DISPLAYPORT || type == INTEL_OUTPUT_EDP) {
 		struct intel_dp *intel_dp = enc_to_intel_dp(encoder);
 
+		intel_dp_set_link_params(intel_dp, crtc->config);
+
 		intel_ddi_init_dp_buf_reg(intel_encoder);
 
 		intel_dp_sink_dpms(intel_dp, DRM_MODE_DPMS_ON);
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index b905c19..dcda86a 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -1584,6 +1584,13 @@ static void ironlake_set_pll_cpu_edp(struct intel_dp *intel_dp)
 	udelay(500);
 }
 
+void intel_dp_set_link_params(struct intel_dp *intel_dp,
+			      const struct intel_crtc_state *pipe_config)
+{
+	intel_dp->link_rate = pipe_config->port_clock;
+	intel_dp->lane_count = pipe_config->lane_count;
+}
+
 static void intel_dp_prepare(struct intel_encoder *encoder)
 {
 	struct drm_device *dev = encoder->base.dev;
@@ -1593,6 +1600,8 @@ static void intel_dp_prepare(struct intel_encoder *encoder)
 	struct intel_crtc *crtc = to_intel_crtc(encoder->base.crtc);
 	struct drm_display_mode *adjusted_mode = &crtc->config->base.adjusted_mode;
 
+	intel_dp_set_link_params(intel_dp, crtc->config);
+
 	/*
 	 * There are four kinds of DP registers:
 	 *
@@ -3348,15 +3357,13 @@ static void
 intel_get_adjust_train(struct intel_dp *intel_dp,
 		       const uint8_t link_status[DP_LINK_STATUS_SIZE])
 {
-	struct intel_crtc *crtc =
-		to_intel_crtc(dp_to_dig_port(intel_dp)->base.base.crtc);
 	uint8_t v = 0;
 	uint8_t p = 0;
 	int lane;
 	uint8_t voltage_max;
 	uint8_t preemph_max;
 
-	for (lane = 0; lane < crtc->config->lane_count; lane++) {
+	for (lane = 0; lane < intel_dp->lane_count; lane++) {
 		uint8_t this_v = drm_dp_get_adjust_request_voltage(link_status, lane);
 		uint8_t this_p = drm_dp_get_adjust_request_pre_emphasis(link_status, lane);
 
@@ -3527,8 +3534,6 @@ intel_dp_set_link_train(struct intel_dp *intel_dp,
 	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
 	struct drm_i915_private *dev_priv =
 		to_i915(intel_dig_port->base.base.dev);
-	struct intel_crtc *crtc =
-		to_intel_crtc(intel_dig_port->base.base.crtc);
 	uint8_t buf[sizeof(intel_dp->train_set) + 1];
 	int ret, len;
 
@@ -3544,8 +3549,8 @@ intel_dp_set_link_train(struct intel_dp *intel_dp,
 		len = 1;
 	} else {
 		/* DP_TRAINING_LANEx_SET follow DP_TRAINING_PATTERN_SET */
-		memcpy(buf + 1, intel_dp->train_set, crtc->config->lane_count);
-		len = crtc->config->lane_count + 1;
+		memcpy(buf + 1, intel_dp->train_set, intel_dp->lane_count);
+		len = intel_dp->lane_count + 1;
 	}
 
 	ret = drm_dp_dpcd_write(&intel_dp->aux, DP_TRAINING_PATTERN_SET,
@@ -3571,8 +3576,6 @@ intel_dp_update_link_train(struct intel_dp *intel_dp, uint32_t *DP,
 	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
 	struct drm_i915_private *dev_priv =
 		to_i915(intel_dig_port->base.base.dev);
-	struct intel_crtc *crtc =
-		to_intel_crtc(intel_dig_port->base.base.crtc);
 	int ret;
 
 	intel_get_adjust_train(intel_dp, link_status);
@@ -3582,9 +3585,9 @@ intel_dp_update_link_train(struct intel_dp *intel_dp, uint32_t *DP,
 	POSTING_READ(intel_dp->output_reg);
 
 	ret = drm_dp_dpcd_write(&intel_dp->aux, DP_TRAINING_LANE0_SET,
-				intel_dp->train_set, crtc->config->lane_count);
+				intel_dp->train_set, intel_dp->lane_count);
 
-	return ret == crtc->config->lane_count;
+	return ret == intel_dp->lane_count;
 }
 
 static void intel_dp_set_idle_link_train(struct intel_dp *intel_dp)
@@ -3623,8 +3626,6 @@ void
 intel_dp_start_link_train(struct intel_dp *intel_dp)
 {
 	struct drm_encoder *encoder = &dp_to_dig_port(intel_dp)->base.base;
-	struct intel_crtc *crtc =
-		to_intel_crtc(dp_to_dig_port(intel_dp)->base.base.crtc);
 	struct drm_device *dev = encoder->dev;
 	int i;
 	uint8_t voltage;
@@ -3636,12 +3637,12 @@ intel_dp_start_link_train(struct intel_dp *intel_dp)
 	if (HAS_DDI(dev))
 		intel_ddi_prepare_link_retrain(encoder);
 
-	intel_dp_compute_rate(intel_dp, crtc->config->port_clock,
+	intel_dp_compute_rate(intel_dp, intel_dp->link_rate,
 			      &link_bw, &rate_select);
 
 	/* Write the link configuration data */
 	link_config[0] = link_bw;
-	link_config[1] = crtc->config->lane_count;
+	link_config[1] = intel_dp->lane_count;
 	if (drm_dp_enhanced_frame_cap(intel_dp->dpcd))
 		link_config[1] |= DP_LANE_COUNT_ENHANCED_FRAME_EN;
 	drm_dp_dpcd_write(&intel_dp->aux, DP_LINK_BW_SET, link_config, 2);
@@ -3675,7 +3676,7 @@ intel_dp_start_link_train(struct intel_dp *intel_dp)
 			break;
 		}
 
-		if (drm_dp_clock_recovery_ok(link_status, crtc->config->lane_count)) {
+		if (drm_dp_clock_recovery_ok(link_status, intel_dp->lane_count)) {
 			DRM_DEBUG_KMS("clock recovery OK\n");
 			break;
 		}
@@ -3698,10 +3699,10 @@ intel_dp_start_link_train(struct intel_dp *intel_dp)
 		}
 
 		/* Check to see if we've tried the max voltage */
-		for (i = 0; i < crtc->config->lane_count; i++)
+		for (i = 0; i < intel_dp->lane_count; i++)
 			if ((intel_dp->train_set[i] & DP_TRAIN_MAX_SWING_REACHED) == 0)
 				break;
-		if (i == crtc->config->lane_count) {
+		if (i == intel_dp->lane_count) {
 			++loop_tries;
 			if (loop_tries == 5) {
 				DRM_ERROR("too many full retries, give up\n");
@@ -3738,15 +3739,13 @@ intel_dp_start_link_train(struct intel_dp *intel_dp)
 void
 intel_dp_complete_link_train(struct intel_dp *intel_dp)
 {
-	struct intel_crtc *crtc =
-		to_intel_crtc(dp_to_dig_port(intel_dp)->base.base.crtc);
 	bool channel_eq = false;
 	int tries, cr_tries;
 	uint32_t DP = intel_dp->DP;
 	uint32_t training_pattern = DP_TRAINING_PATTERN_2;
 
 	/* Training Pattern 3 for HBR2 or 1.2 devices that support it*/
-	if (crtc->config->port_clock == 540000 || intel_dp->use_tps3)
+	if (intel_dp->link_rate == 540000 || intel_dp->use_tps3)
 		training_pattern = DP_TRAINING_PATTERN_3;
 
 	/* channel equalization */
@@ -3776,7 +3775,7 @@ intel_dp_complete_link_train(struct intel_dp *intel_dp)
 
 		/* Make sure clock is still ok */
 		if (!drm_dp_clock_recovery_ok(link_status,
-					      crtc->config->lane_count)) {
+					      intel_dp->lane_count)) {
 			intel_dp->train_set_valid = false;
 			intel_dp_start_link_train(intel_dp);
 			intel_dp_set_link_train(intel_dp, &DP,
@@ -3787,7 +3786,7 @@ intel_dp_complete_link_train(struct intel_dp *intel_dp)
 		}
 
 		if (drm_dp_channel_eq_ok(link_status,
-					 crtc->config->lane_count)) {
+					 intel_dp->lane_count)) {
 			channel_eq = true;
 			break;
 		}
@@ -4285,8 +4284,6 @@ update_status:
 static int
 intel_dp_check_mst_status(struct intel_dp *intel_dp)
 {
-	struct intel_crtc *crtc =
-		to_intel_crtc(dp_to_dig_port(intel_dp)->base.base.crtc);
 	bool bret;
 
 	if (intel_dp->is_mst) {
@@ -4300,7 +4297,7 @@ go_again:
 
 			/* check link status - esi[10] = 0x200c */
 			if (intel_dp->active_mst_links &&
-			    !drm_dp_channel_eq_ok(&esi[10], crtc->config->lane_count)) {
+			    !drm_dp_channel_eq_ok(&esi[10], intel_dp->lane_count)) {
 				DRM_DEBUG_KMS("channel EQ not ok, retraining\n");
 				intel_dp_start_link_train(intel_dp);
 				intel_dp_complete_link_train(intel_dp);
@@ -4355,8 +4352,6 @@ intel_dp_check_link_status(struct intel_dp *intel_dp)
 {
 	struct drm_device *dev = intel_dp_to_dev(intel_dp);
 	struct intel_encoder *intel_encoder = &dp_to_dig_port(intel_dp)->base;
-	struct intel_crtc *crtc =
-		to_intel_crtc(dp_to_dig_port(intel_dp)->base.base.crtc);
 	u8 sink_irq_vector;
 	u8 link_status[DP_LINK_STATUS_SIZE];
 
@@ -4392,7 +4387,7 @@ intel_dp_check_link_status(struct intel_dp *intel_dp)
 			DRM_DEBUG_DRIVER("CP or sink specific irq unhandled\n");
 	}
 
-	if (!drm_dp_channel_eq_ok(link_status, crtc->config->lane_count)) {
+	if (!drm_dp_channel_eq_ok(link_status, intel_dp->lane_count)) {
 		DRM_DEBUG_KMS("%s: channel EQ not ok, retraining\n",
 			      intel_encoder->base.name);
 		intel_dp_start_link_train(intel_dp);
diff --git a/drivers/gpu/drm/i915/intel_dp_mst.c b/drivers/gpu/drm/i915/intel_dp_mst.c
index 1dd2b37..dd291d1 100644
--- a/drivers/gpu/drm/i915/intel_dp_mst.c
+++ b/drivers/gpu/drm/i915/intel_dp_mst.c
@@ -165,6 +165,8 @@ static void intel_mst_pre_enable_dp(struct intel_encoder *encoder)
 	if (intel_dp->active_mst_links == 0) {
 		enum port port = intel_ddi_get_encoder_port(encoder);
 
+		intel_dp_set_link_params(intel_dp, intel_crtc->config);
+
 		/* FIXME: add support for SKL */
 		if (INTEL_INFO(dev)->gen < 9)
 			I915_WRITE(PORT_CLK_SEL(port),
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 22dc8b6..8383442 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -708,6 +708,8 @@ struct intel_dp {
 	uint32_t output_reg;
 	uint32_t aux_ch_ctl_reg;
 	uint32_t DP;
+	int link_rate;
+	uint8_t lane_count;
 	bool has_audio;
 	enum hdmi_force_audio force_audio;
 	bool limited_color_range;
@@ -1161,6 +1163,8 @@ void assert_csr_loaded(struct drm_i915_private *dev_priv);
 void intel_dp_init(struct drm_device *dev, int output_reg, enum port port);
 bool intel_dp_init_connector(struct intel_digital_port *intel_dig_port,
 			     struct intel_connector *intel_connector);
+void intel_dp_set_link_params(struct intel_dp *intel_dp,
+			      const struct intel_crtc_state *pipe_config);
 void intel_dp_start_link_train(struct intel_dp *intel_dp);
 void intel_dp_complete_link_train(struct intel_dp *intel_dp);
 void intel_dp_stop_link_train(struct intel_dp *intel_dp);
-- 
cgit v0.10.2


From 4eee4920f05e39e22571deb57f1c314ce7a46e7f Mon Sep 17 00:00:00 2001
From: Dave Gordon <david.s.gordon@intel.com>
Date: Mon, 17 Aug 2015 17:30:52 +0100
Subject: drm/i915: fix driver's versions of WARN_ON & WARN_ON_ONCE

The current versions of these two macros don't work correctly if the
argument expression happens to contain a modulo operator (%) -- when
stringified, it gets interpreted as a printf formatting character!
With a specifically crafted parameter, this could probably cause a
kernel OOPS; consider WARN_ON(p%s) or WARN_ON(f %*pEp).

Instead, we should use an explicit "%s" format, with the stringified
expression as the coresponding literal-string argument.

Signed-off-by: Dave Gordon <david.s.gordon@intel.com>
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 6124445..ce41d24 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -68,11 +68,11 @@
 		BUILD_BUG_ON(__i915_warn_cond); \
 	WARN(__i915_warn_cond, "WARN_ON(" #x ")"); })
 #else
-#define WARN_ON(x) WARN((x), "WARN_ON(" #x ")")
+#define WARN_ON(x) WARN((x), "WARN_ON(%s)", #x )
 #endif
 
 #undef WARN_ON_ONCE
-#define WARN_ON_ONCE(x) WARN_ONCE((x), "WARN_ON_ONCE(" #x ")")
+#define WARN_ON_ONCE(x) WARN_ONCE((x), "WARN_ON_ONCE(%s)", #x )
 
 #define MISSING_CASE(x) WARN(1, "Missing switch case (%lu) in %s\n", \
 			     (long) (x), __func__);
-- 
cgit v0.10.2


From 25a25dfce4e678bb18c4d63937d240839ecd19cd Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 8 Jul 2015 23:45:47 +0300
Subject: drm/i915: Always program m2 fractional value on CHV
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

When fractional m2 divider isn't used on CHV the fractional part
is ignore by the hardware. Despite that, program the fractional
value (0 in this case) to the hardware register just to keep
things a bit more consistent. Might at least make register dumps
a bit less confusing when there isn't some stale fractional part
hanging around.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Deepak S <deepak.s@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index cf8dfa5..a64da67 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -7374,8 +7374,7 @@ static void chv_prepare_pll(struct intel_crtc *crtc,
 			1 << DPIO_CHV_N_DIV_SHIFT);
 
 	/* M2 fraction division */
-	if (bestm2_frac)
-		vlv_dpio_write(dev_priv, pipe, CHV_PLL_DW2(port), bestm2_frac);
+	vlv_dpio_write(dev_priv, pipe, CHV_PLL_DW2(port), bestm2_frac);
 
 	/* M2 fraction division enable */
 	dpio_val = vlv_dpio_read(dev_priv, pipe, CHV_PLL_DW3(port));
-- 
cgit v0.10.2


From 67fa24b4041436f1a21f7640d1d7ab56d023d7e9 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 8 Jul 2015 23:45:48 +0300
Subject: drm/i915: Always program unique transition scale for CHV
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The docs give you the impression that the unique transition scale
value shouldn't matter when unique transition scale is enabled. But
as Imre found on BXT (and I verfied also on BSW) the value does
matter. So from now on just program the same value 0x9a always.

Cc: Imre Deak <imre.deak@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Deepak S <deepak.s@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index dcda86a..da7d1ae 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -3179,6 +3179,12 @@ static uint32_t vlv_signal_levels(struct intel_dp *intel_dp)
 	return 0;
 }
 
+static bool chv_need_uniq_trans_scale(uint8_t train_set)
+{
+	return (train_set & DP_TRAIN_PRE_EMPHASIS_MASK) == DP_TRAIN_PRE_EMPH_LEVEL_0 &&
+		(train_set & DP_TRAIN_VOLTAGE_SWING_MASK) == DP_TRAIN_VOLTAGE_SWING_LEVEL_3;
+}
+
 static uint32_t chv_signal_levels(struct intel_dp *intel_dp)
 {
 	struct drm_device *dev = intel_dp_to_dev(intel_dp);
@@ -3297,41 +3303,34 @@ static uint32_t chv_signal_levels(struct intel_dp *intel_dp)
 	/* Program swing margin */
 	for (i = 0; i < 4; i++) {
 		val = vlv_dpio_read(dev_priv, pipe, CHV_TX_DW2(ch, i));
+
 		val &= ~DPIO_SWING_MARGIN000_MASK;
 		val |= margin_reg_value << DPIO_SWING_MARGIN000_SHIFT;
+
+		/*
+		 * Supposedly this value shouldn't matter when unique transition
+		 * scale is disabled, but in fact it does matter. Let's just
+		 * always program the same value and hope it's OK.
+		 */
+		val &= ~(0xff << DPIO_UNIQ_TRANS_SCALE_SHIFT);
+		val |= 0x9a << DPIO_UNIQ_TRANS_SCALE_SHIFT;
+
 		vlv_dpio_write(dev_priv, pipe, CHV_TX_DW2(ch, i), val);
 	}
 
-	/* Disable unique transition scale */
+	/*
+	 * The document said it needs to set bit 27 for ch0 and bit 26
+	 * for ch1. Might be a typo in the doc.
+	 * For now, for this unique transition scale selection, set bit
+	 * 27 for ch0 and ch1.
+	 */
 	for (i = 0; i < 4; i++) {
 		val = vlv_dpio_read(dev_priv, pipe, CHV_TX_DW3(ch, i));
-		val &= ~DPIO_TX_UNIQ_TRANS_SCALE_EN;
-		vlv_dpio_write(dev_priv, pipe, CHV_TX_DW3(ch, i), val);
-	}
-
-	if (((train_set & DP_TRAIN_PRE_EMPHASIS_MASK)
-			== DP_TRAIN_PRE_EMPH_LEVEL_0) &&
-		((train_set & DP_TRAIN_VOLTAGE_SWING_MASK)
-			== DP_TRAIN_VOLTAGE_SWING_LEVEL_3)) {
-
-		/*
-		 * The document said it needs to set bit 27 for ch0 and bit 26
-		 * for ch1. Might be a typo in the doc.
-		 * For now, for this unique transition scale selection, set bit
-		 * 27 for ch0 and ch1.
-		 */
-		for (i = 0; i < 4; i++) {
-			val = vlv_dpio_read(dev_priv, pipe, CHV_TX_DW3(ch, i));
+		if (chv_need_uniq_trans_scale(train_set))
 			val |= DPIO_TX_UNIQ_TRANS_SCALE_EN;
-			vlv_dpio_write(dev_priv, pipe, CHV_TX_DW3(ch, i), val);
-		}
-
-		for (i = 0; i < 4; i++) {
-			val = vlv_dpio_read(dev_priv, pipe, CHV_TX_DW2(ch, i));
-			val &= ~(0xff << DPIO_UNIQ_TRANS_SCALE_SHIFT);
-			val |= (0x9a << DPIO_UNIQ_TRANS_SCALE_SHIFT);
-			vlv_dpio_write(dev_priv, pipe, CHV_TX_DW2(ch, i), val);
-		}
+		else
+			val &= ~DPIO_TX_UNIQ_TRANS_SCALE_EN;
+		vlv_dpio_write(dev_priv, pipe, CHV_TX_DW3(ch, i), val);
 	}
 
 	/* Start swing calculation */
diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index 7185062..c25835f 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -1849,31 +1849,33 @@ static void chv_hdmi_pre_enable(struct intel_encoder *encoder)
 
 	for (i = 0; i < 4; i++) {
 		val = vlv_dpio_read(dev_priv, pipe, CHV_TX_DW2(ch, i));
+
 		val &= ~DPIO_SWING_MARGIN000_MASK;
 		val |= 102 << DPIO_SWING_MARGIN000_SHIFT;
+
+		/*
+		 * Supposedly this value shouldn't matter when unique transition
+		 * scale is disabled, but in fact it does matter. Let's just
+		 * always program the same value and hope it's OK.
+		 */
+		val &= ~(0xff << DPIO_UNIQ_TRANS_SCALE_SHIFT);
+		val |= 0x9a << DPIO_UNIQ_TRANS_SCALE_SHIFT;
+
 		vlv_dpio_write(dev_priv, pipe, CHV_TX_DW2(ch, i), val);
 	}
 
-	/* Disable unique transition scale */
+	/*
+	 * The document said it needs to set bit 27 for ch0 and bit 26
+	 * for ch1. Might be a typo in the doc.
+	 * For now, for this unique transition scale selection, set bit
+	 * 27 for ch0 and ch1.
+	 */
 	for (i = 0; i < 4; i++) {
 		val = vlv_dpio_read(dev_priv, pipe, CHV_TX_DW3(ch, i));
 		val &= ~DPIO_TX_UNIQ_TRANS_SCALE_EN;
 		vlv_dpio_write(dev_priv, pipe, CHV_TX_DW3(ch, i), val);
 	}
 
-	/* Additional steps for 1200mV-0dB */
-#if 0
-	val = vlv_dpio_read(dev_priv, pipe, VLV_TX_DW3(ch));
-	if (ch)
-		val |= DPIO_TX_UNIQ_TRANS_SCALE_CH1;
-	else
-		val |= DPIO_TX_UNIQ_TRANS_SCALE_CH0;
-	vlv_dpio_write(dev_priv, pipe, VLV_TX_DW3(ch), val);
-
-	vlv_dpio_write(dev_priv, pipe, VLV_TX_DW2(ch),
-			vlv_dpio_read(dev_priv, pipe, VLV_TX_DW2(ch)) |
-				(0x9a << DPIO_UNIQ_TRANS_SCALE_SHIFT));
-#endif
 	/* Start swing calculation */
 	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW10(ch));
 	val |= DPIO_PCS_SWING_CALC_TX0_TX2 | DPIO_PCS_SWING_CALC_TX1_TX3;
-- 
cgit v0.10.2


From d6db995fe36c7460a30c8c6250bb47937d38c482 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 8 Jul 2015 23:45:49 +0300
Subject: drm/i915: Add encoder->post_pll_disable() hooks and move CHV clock
 buffer disables there
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Move the CHV clock buffer disable from chv_disable_pll() to the new
encoder .post_pll_disable() hook. This is more symmetric since the
clock buffer enable happens from the .pre_pll_enable() hook.

We'll have more use for the new hook soon.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Deepak S <deepak.s@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index a64da67..328d1ee 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -1828,17 +1828,6 @@ static void chv_disable_pll(struct drm_i915_private *dev_priv, enum pipe pipe)
 	val &= ~DPIO_DCLKP_EN;
 	vlv_dpio_write(dev_priv, pipe, CHV_CMN_DW14(port), val);
 
-	/* disable left/right clock distribution */
-	if (pipe != PIPE_B) {
-		val = vlv_dpio_read(dev_priv, pipe, _CHV_CMN_DW5_CH0);
-		val &= ~(CHV_BUFLEFTENA1_MASK | CHV_BUFRIGHTENA1_MASK);
-		vlv_dpio_write(dev_priv, pipe, _CHV_CMN_DW5_CH0, val);
-	} else {
-		val = vlv_dpio_read(dev_priv, pipe, _CHV_CMN_DW1_CH1);
-		val &= ~(CHV_BUFLEFTENA2_MASK | CHV_BUFRIGHTENA2_MASK);
-		vlv_dpio_write(dev_priv, pipe, _CHV_CMN_DW1_CH1, val);
-	}
-
 	mutex_unlock(&dev_priv->sb_lock);
 }
 
@@ -6193,6 +6182,10 @@ static void i9xx_crtc_disable(struct drm_crtc *crtc)
 			i9xx_disable_pll(intel_crtc);
 	}
 
+	for_each_encoder_on_crtc(dev, crtc, encoder)
+		if (encoder->post_pll_disable)
+			encoder->post_pll_disable(encoder);
+
 	if (!IS_GEN2(dev))
 		intel_set_cpu_fifo_underrun_reporting(dev_priv, pipe, false);
 
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index da7d1ae..4e5bd87 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -2943,6 +2943,28 @@ static void chv_dp_pre_pll_enable(struct intel_encoder *encoder)
 	mutex_unlock(&dev_priv->sb_lock);
 }
 
+static void chv_dp_post_pll_disable(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	enum pipe pipe = to_intel_crtc(encoder->base.crtc)->pipe;
+	u32 val;
+
+	mutex_lock(&dev_priv->sb_lock);
+
+	/* disable left/right clock distribution */
+	if (pipe != PIPE_B) {
+		val = vlv_dpio_read(dev_priv, pipe, _CHV_CMN_DW5_CH0);
+		val &= ~(CHV_BUFLEFTENA1_MASK | CHV_BUFRIGHTENA1_MASK);
+		vlv_dpio_write(dev_priv, pipe, _CHV_CMN_DW5_CH0, val);
+	} else {
+		val = vlv_dpio_read(dev_priv, pipe, _CHV_CMN_DW1_CH1);
+		val &= ~(CHV_BUFLEFTENA2_MASK | CHV_BUFRIGHTENA2_MASK);
+		vlv_dpio_write(dev_priv, pipe, _CHV_CMN_DW1_CH1, val);
+	}
+
+	mutex_unlock(&dev_priv->sb_lock);
+}
+
 /*
  * Native read with retry for link status and receiver capability reads for
  * cases where the sink may still be asleep.
@@ -6000,6 +6022,7 @@ intel_dp_init(struct drm_device *dev, int output_reg, enum port port)
 		intel_encoder->pre_enable = chv_pre_enable_dp;
 		intel_encoder->enable = vlv_enable_dp;
 		intel_encoder->post_disable = chv_post_disable_dp;
+		intel_encoder->post_pll_disable = chv_dp_post_pll_disable;
 	} else if (IS_VALLEYVIEW(dev)) {
 		intel_encoder->pre_pll_enable = vlv_dp_pre_pll_enable;
 		intel_encoder->pre_enable = vlv_pre_enable_dp;
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 8383442..525e457 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -142,6 +142,7 @@ struct intel_encoder {
 	void (*mode_set)(struct intel_encoder *intel_encoder);
 	void (*disable)(struct intel_encoder *);
 	void (*post_disable)(struct intel_encoder *);
+	void (*post_pll_disable)(struct intel_encoder *);
 	/* Read out the current hw state of this connector, returning true if
 	 * the encoder is active. If the encoder is enabled it also set the pipe
 	 * it is connected to in the pipe parameter. */
diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index c25835f..269937a 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -1681,6 +1681,28 @@ static void chv_hdmi_pre_pll_enable(struct intel_encoder *encoder)
 	mutex_unlock(&dev_priv->sb_lock);
 }
 
+static void chv_hdmi_post_pll_disable(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	enum pipe pipe = to_intel_crtc(encoder->base.crtc)->pipe;
+	u32 val;
+
+	mutex_lock(&dev_priv->sb_lock);
+
+	/* disable left/right clock distribution */
+	if (pipe != PIPE_B) {
+		val = vlv_dpio_read(dev_priv, pipe, _CHV_CMN_DW5_CH0);
+		val &= ~(CHV_BUFLEFTENA1_MASK | CHV_BUFRIGHTENA1_MASK);
+		vlv_dpio_write(dev_priv, pipe, _CHV_CMN_DW5_CH0, val);
+	} else {
+		val = vlv_dpio_read(dev_priv, pipe, _CHV_CMN_DW1_CH1);
+		val &= ~(CHV_BUFLEFTENA2_MASK | CHV_BUFRIGHTENA2_MASK);
+		vlv_dpio_write(dev_priv, pipe, _CHV_CMN_DW1_CH1, val);
+	}
+
+	mutex_unlock(&dev_priv->sb_lock);
+}
+
 static void vlv_hdmi_post_disable(struct intel_encoder *encoder)
 {
 	struct intel_digital_port *dport = enc_to_dig_port(&encoder->base);
@@ -2083,6 +2105,7 @@ void intel_hdmi_init(struct drm_device *dev, int hdmi_reg, enum port port)
 		intel_encoder->pre_enable = chv_hdmi_pre_enable;
 		intel_encoder->enable = vlv_enable_hdmi;
 		intel_encoder->post_disable = chv_hdmi_post_disable;
+		intel_encoder->post_pll_disable = chv_hdmi_post_pll_disable;
 	} else if (IS_VALLEYVIEW(dev)) {
 		intel_encoder->pre_pll_enable = vlv_hdmi_pre_pll_enable;
 		intel_encoder->pre_enable = vlv_hdmi_pre_enable;
-- 
cgit v0.10.2


From e27f299ec3ddd03373bb07db23dd19747eecef72 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 8 Jul 2015 23:45:50 +0300
Subject: drm/i915: Move DPIO port init earlier
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

To implement DPIO lane power gating on CHV we're going to need to access
DPIO registers from the cmn power well enable hook. That gets called
rather early, so we need to move the DPIO port IOSF sideband port
assignment earlier as well.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Deepak S <deepak.s@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index 2193cc2..097d4ba 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -799,6 +799,24 @@ static void intel_device_info_runtime_init(struct drm_device *dev)
 			 info->has_eu_pg ? "y" : "n");
 }
 
+static void intel_init_dpio(struct drm_i915_private *dev_priv)
+{
+	if (!IS_VALLEYVIEW(dev_priv))
+		return;
+
+	/*
+	 * IOSF_PORT_DPIO is used for VLV x2 PHY (DP/HDMI B and C),
+	 * CHV x1 PHY (DP/HDMI D)
+	 * IOSF_PORT_DPIO_2 is used for CHV x2 PHY (DP/HDMI B and C)
+	 */
+	if (IS_CHERRYVIEW(dev_priv)) {
+		DPIO_PHY_IOSF_PORT(DPIO_PHY0) = IOSF_PORT_DPIO_2;
+		DPIO_PHY_IOSF_PORT(DPIO_PHY1) = IOSF_PORT_DPIO;
+	} else {
+		DPIO_PHY_IOSF_PORT(DPIO_PHY0) = IOSF_PORT_DPIO;
+	}
+}
+
 /**
  * i915_driver_load - setup chip and create an initial config
  * @dev: DRM device
@@ -999,6 +1017,8 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 
 	intel_device_info_runtime_init(dev);
 
+	intel_init_dpio(dev_priv);
+
 	if (INTEL_INFO(dev)->num_pipes) {
 		ret = drm_vblank_init(dev, INTEL_INFO(dev)->num_pipes);
 		if (ret)
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 328d1ee..3c2c76b 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -1582,26 +1582,6 @@ static void assert_pch_ports_disabled(struct drm_i915_private *dev_priv,
 	assert_pch_hdmi_disabled(dev_priv, pipe, PCH_HDMID);
 }
 
-static void intel_init_dpio(struct drm_device *dev)
-{
-	struct drm_i915_private *dev_priv = dev->dev_private;
-
-	if (!IS_VALLEYVIEW(dev))
-		return;
-
-	/*
-	 * IOSF_PORT_DPIO is used for VLV x2 PHY (DP/HDMI B and C),
-	 * CHV x1 PHY (DP/HDMI D)
-	 * IOSF_PORT_DPIO_2 is used for CHV x2 PHY (DP/HDMI B and C)
-	 */
-	if (IS_CHERRYVIEW(dev)) {
-		DPIO_PHY_IOSF_PORT(DPIO_PHY0) = IOSF_PORT_DPIO_2;
-		DPIO_PHY_IOSF_PORT(DPIO_PHY1) = IOSF_PORT_DPIO;
-	} else {
-		DPIO_PHY_IOSF_PORT(DPIO_PHY0) = IOSF_PORT_DPIO;
-	}
-}
-
 static void vlv_enable_pll(struct intel_crtc *crtc,
 			   const struct intel_crtc_state *pipe_config)
 {
@@ -14767,8 +14747,6 @@ void intel_modeset_init(struct drm_device *dev)
 		}
 	}
 
-	intel_init_dpio(dev);
-
 	intel_shared_dpll_init(dev);
 
 	/* Just disable it once at startup */
-- 
cgit v0.10.2


From 770effb19fbdcb44c6bdacf4a78571d28393f48f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 8 Jul 2015 23:45:51 +0300
Subject: drm/i915: Add locking around chv_phy_control_init()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

dev_priv->chv_phy_control is protected by the power_domains->lock
elsewhere, so also grab it when initializing chv_phy_control.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Deepak S <deepak.s@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c
index 821644d..974c4cb 100644
--- a/drivers/gpu/drm/i915/intel_runtime_pm.c
+++ b/drivers/gpu/drm/i915/intel_runtime_pm.c
@@ -1685,7 +1685,9 @@ void intel_power_domains_init_hw(struct drm_i915_private *dev_priv)
 	power_domains->initializing = true;
 
 	if (IS_CHERRYVIEW(dev)) {
+		mutex_lock(&power_domains->lock);
 		chv_phy_control_init(dev_priv);
+		mutex_unlock(&power_domains->lock);
 	} else if (IS_VALLEYVIEW(dev)) {
 		mutex_lock(&power_domains->lock);
 		vlv_cmnlane_wa(dev_priv);
-- 
cgit v0.10.2


From c0b4c660311c73c8ee4ed0e58b65ae9b4b75276d Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 8 Jul 2015 23:45:52 +0300
Subject: drm/i915: Move VLV/CHV prepare_pll later
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

With DPIO powergating active on CHV, we can't even access the DPIO PLL
registers until the lane power state overrides have been enabled. That
will happen from the encoder .pre_pll_enable() hook, so move
chv_prepare_pll() to happen after that point, which puts it just before
chv_enable_pll() actually.

Do the same for VLV to avoid accumulating weird differences between the
platforms. Both platforms seem happy with the new arrangement.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Deepak S <deepak.s@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 3c2c76b..50cbc7e 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -6001,13 +6001,6 @@ static void valleyview_crtc_enable(struct drm_crtc *crtc)
 
 	is_dsi = intel_pipe_has_type(intel_crtc, INTEL_OUTPUT_DSI);
 
-	if (!is_dsi) {
-		if (IS_CHERRYVIEW(dev))
-			chv_prepare_pll(intel_crtc, intel_crtc->config);
-		else
-			vlv_prepare_pll(intel_crtc, intel_crtc->config);
-	}
-
 	if (intel_crtc->config->has_dp_encoder)
 		intel_dp_set_m_n(intel_crtc, M1_N1);
 
@@ -6031,10 +6024,13 @@ static void valleyview_crtc_enable(struct drm_crtc *crtc)
 			encoder->pre_pll_enable(encoder);
 
 	if (!is_dsi) {
-		if (IS_CHERRYVIEW(dev))
+		if (IS_CHERRYVIEW(dev)) {
+			chv_prepare_pll(intel_crtc, intel_crtc->config);
 			chv_enable_pll(intel_crtc, intel_crtc->config);
-		else
+		} else {
+			vlv_prepare_pll(intel_crtc, intel_crtc->config);
 			vlv_enable_pll(intel_crtc, intel_crtc->config);
+		}
 	}
 
 	for_each_encoder_on_crtc(dev, crtc, encoder)
-- 
cgit v0.10.2


From 65d64cc5bb7aee1f5a8d6717f4d421623c58ea30 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 8 Jul 2015 23:45:53 +0300
Subject: drm/i915: Add vlv_dport_to_phy()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add vlv_dport_to_phy() and fix up the return values of
vlv_dport_to_channel() and vlv_pipe_to_channel() to use
the appropriate enums.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Deepak S <deepak.s@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 525e457..49d1b61 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -788,7 +788,7 @@ struct intel_dp_mst_encoder {
 	void *port; /* store this opaque as its illegal to dereference it */
 };
 
-static inline int
+static inline enum dpio_channel
 vlv_dport_to_channel(struct intel_digital_port *dport)
 {
 	switch (dport->port) {
@@ -802,7 +802,21 @@ vlv_dport_to_channel(struct intel_digital_port *dport)
 	}
 }
 
-static inline int
+static inline enum dpio_phy
+vlv_dport_to_phy(struct intel_digital_port *dport)
+{
+	switch (dport->port) {
+	case PORT_B:
+	case PORT_C:
+		return DPIO_PHY0;
+	case PORT_D:
+		return DPIO_PHY1;
+	default:
+		BUG();
+	}
+}
+
+static inline enum dpio_channel
 vlv_pipe_to_channel(enum pipe pipe)
 {
 	switch (pipe) {
-- 
cgit v0.10.2


From adafdc6fcb66ea6541a88884ec522273f857c145 Mon Sep 17 00:00:00 2001
From: Mika Kahola <mika.kahola@intel.com>
Date: Tue, 18 Aug 2015 14:36:59 +0300
Subject: drm/i915: Store max dotclock
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Store max dotclock into dev_priv structure so we are able
to filter out the modes that are not supported by our
platforms.

V2:
- limit the max dot clock frequency to max CD clock frequency
  for the gen9 and above
- limit the max dot clock frequency to 90% of the max CD clock
  frequency for the older gens
- for Cherryview the max dot clock frequency is limited to 95%
  of the max CD clock frequency
- for gen2 and gen3 the max dot clock limit is set to 90% of the
  2X max CD clock frequency

V3:
- max_dotclk variable renamed as max_dotclk_freq in i915_drv.h
- in intel_compute_max_dotclk() the rounding method changed from
  round up to round down when computing max dotclock

V4:
- Haswell and Broadwell supports now dot clocks up to max CD clock
  frequency

Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index ce41d24..705f208 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -1792,6 +1792,7 @@ struct drm_i915_private {
 	unsigned int fsb_freq, mem_freq, is_ddr3;
 	unsigned int skl_boot_cdclk;
 	unsigned int cdclk_freq, max_cdclk_freq;
+	unsigned int max_dotclk_freq;
 	unsigned int hpll_freq;
 
 	/**
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 50cbc7e..03e84ac 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -5240,6 +5240,21 @@ static void modeset_update_crtc_power_domains(struct drm_atomic_state *state)
 			modeset_put_power_domains(dev_priv, put_domains[i]);
 }
 
+static int intel_compute_max_dotclk(struct drm_i915_private *dev_priv)
+{
+	int max_cdclk_freq = dev_priv->max_cdclk_freq;
+
+	if (INTEL_INFO(dev_priv)->gen >= 9 ||
+	    IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv))
+		return max_cdclk_freq;
+	else if (IS_CHERRYVIEW(dev_priv))
+		return max_cdclk_freq*95/100;
+	else if (INTEL_INFO(dev_priv)->gen < 4)
+		return 2*max_cdclk_freq*90/100;
+	else
+		return max_cdclk_freq*90/100;
+}
+
 static void intel_update_max_cdclk(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -5279,8 +5294,13 @@ static void intel_update_max_cdclk(struct drm_device *dev)
 		dev_priv->max_cdclk_freq = dev_priv->cdclk_freq;
 	}
 
+	dev_priv->max_dotclk_freq = intel_compute_max_dotclk(dev_priv);
+
 	DRM_DEBUG_DRIVER("Max CD clock rate: %d kHz\n",
 			 dev_priv->max_cdclk_freq);
+
+	DRM_DEBUG_DRIVER("Max dotclock rate: %d kHz\n",
+			 dev_priv->max_dotclk_freq);
 }
 
 static void intel_update_cdclk(struct drm_device *dev)
-- 
cgit v0.10.2


From 7f7b58cc61883bd16dfd4cae67d7404f7905d0a7 Mon Sep 17 00:00:00 2001
From: Mika Kahola <mika.kahola@intel.com>
Date: Tue, 18 Aug 2015 14:37:00 +0300
Subject: drm/i915: LVDS pixel clock check
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

It is possible the we request to have a mode that has
higher pixel clock than our HW can support. This patch
checks if requested pixel clock is lower than the one
supported by the HW. The requested mode is discarded
if we cannot support the requested pixel clock.

This patch applies to LVDS.

V2:
- removed computation for max pixel clock

V3:
- cleanup by removing unnecessary lines

V4:
- moved supported dotclock check from mode_valid() to intel_lvds_init()

V5:
- dotclock check moved back to mode_valid() function
- dotclock check for fixed mode

Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c
index 881b5d1..0794dc8 100644
--- a/drivers/gpu/drm/i915/intel_lvds.c
+++ b/drivers/gpu/drm/i915/intel_lvds.c
@@ -289,11 +289,14 @@ intel_lvds_mode_valid(struct drm_connector *connector,
 {
 	struct intel_connector *intel_connector = to_intel_connector(connector);
 	struct drm_display_mode *fixed_mode = intel_connector->panel.fixed_mode;
+	int max_pixclk = to_i915(connector->dev)->max_dotclk_freq;
 
 	if (mode->hdisplay > fixed_mode->hdisplay)
 		return MODE_PANEL;
 	if (mode->vdisplay > fixed_mode->vdisplay)
 		return MODE_PANEL;
+	if (fixed_mode->clock > max_pixclk)
+		return MODE_CLOCK_HIGH;
 
 	return MODE_OK;
 }
-- 
cgit v0.10.2


From 759a1e98213f17d51926495e6754fdaced5f7d81 Mon Sep 17 00:00:00 2001
From: Mika Kahola <mika.kahola@intel.com>
Date: Tue, 18 Aug 2015 14:37:01 +0300
Subject: drm/i915: DSI pixel clock check
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

It is possible the we request to have a mode that has
higher pixel clock than our HW can support. This patch
checks if requested pixel clock is lower than the one
supported by the HW. The requested mode is discarded
if we cannot support the requested pixel clock.

This patch applies to DSI.

V2:
- removed computation for max pixel clock

V3:
- cleanup by removing unnecessary lines

V4:
- max_pixclk variable renamed as max_dotclk
- moved dot clock checking inside 'if (fixed_mode)'

V5:
- dot clock checked against fixed_mode clock

Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dsi.c b/drivers/gpu/drm/i915/intel_dsi.c
index 4a601cf..781c267 100644
--- a/drivers/gpu/drm/i915/intel_dsi.c
+++ b/drivers/gpu/drm/i915/intel_dsi.c
@@ -654,6 +654,7 @@ intel_dsi_mode_valid(struct drm_connector *connector,
 {
 	struct intel_connector *intel_connector = to_intel_connector(connector);
 	struct drm_display_mode *fixed_mode = intel_connector->panel.fixed_mode;
+	int max_dotclk = to_i915(connector->dev)->max_dotclk_freq;
 
 	DRM_DEBUG_KMS("\n");
 
@@ -667,6 +668,8 @@ intel_dsi_mode_valid(struct drm_connector *connector,
 			return MODE_PANEL;
 		if (mode->vdisplay > fixed_mode->vdisplay)
 			return MODE_PANEL;
+		if (fixed_mode->clock > max_dotclk)
+			return MODE_CLOCK_HIGH;
 	}
 
 	return MODE_OK;
-- 
cgit v0.10.2


From 26a91555bd27b0672ab4efaf4de04fe7321540dc Mon Sep 17 00:00:00 2001
From: Mika Kahola <mika.kahola@intel.com>
Date: Tue, 18 Aug 2015 14:37:02 +0300
Subject: drm/i915: DVO pixel clock check
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

It is possible the we request to have a mode that has
higher pixel clock than our HW can support. This patch
checks if requested pixel clock is lower than the one
supported by the HW. The requested mode is discarded
if we cannot support the requested pixel clock.

This patch applies to DVO.

V2:
- removed computation for max pixel clock

V3:
- cleanup by removing unnecessary lines

V4:
- clock check against max dotclock moved inside 'if (fixed_mode)'

V5:
- dot clock check against fixed_mode clock when available

Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dvo.c b/drivers/gpu/drm/i915/intel_dvo.c
index dc532bb..c80fe1f 100644
--- a/drivers/gpu/drm/i915/intel_dvo.c
+++ b/drivers/gpu/drm/i915/intel_dvo.c
@@ -201,6 +201,8 @@ intel_dvo_mode_valid(struct drm_connector *connector,
 		     struct drm_display_mode *mode)
 {
 	struct intel_dvo *intel_dvo = intel_attached_dvo(connector);
+	int max_dotclk = to_i915(connector->dev)->max_dotclk_freq;
+	int target_clock = mode->clock;
 
 	if (mode->flags & DRM_MODE_FLAG_DBLSCAN)
 		return MODE_NO_DBLESCAN;
@@ -212,8 +214,13 @@ intel_dvo_mode_valid(struct drm_connector *connector,
 			return MODE_PANEL;
 		if (mode->vdisplay > intel_dvo->panel_fixed_mode->vdisplay)
 			return MODE_PANEL;
+
+		target_clock = intel_dvo->panel_fixed_mode->clock;
 	}
 
+	if (target_clock > max_dotclk)
+		return MODE_CLOCK_HIGH;
+
 	return intel_dvo->dev.dev_ops->mode_valid(&intel_dvo->dev, mode);
 }
 
-- 
cgit v0.10.2


From b93433ccf64846820b9448f5ff5dd4348b58a8ed Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 20 Aug 2015 10:47:36 +0300
Subject: drm/i915: move ibx_digital_port_connected to intel_dp.c

The function can be made static there. No functional changes.

Reviewed-by: Durgadoss R <durgadoss.r@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 03e84ac..f32206d 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -1061,51 +1061,6 @@ static void intel_wait_for_pipe_off(struct intel_crtc *crtc)
 	}
 }
 
-/*
- * ibx_digital_port_connected - is the specified port connected?
- * @dev_priv: i915 private structure
- * @port: the port to test
- *
- * Returns true if @port is connected, false otherwise.
- */
-bool ibx_digital_port_connected(struct drm_i915_private *dev_priv,
-				struct intel_digital_port *port)
-{
-	u32 bit;
-
-	if (HAS_PCH_IBX(dev_priv->dev)) {
-		switch (port->port) {
-		case PORT_B:
-			bit = SDE_PORTB_HOTPLUG;
-			break;
-		case PORT_C:
-			bit = SDE_PORTC_HOTPLUG;
-			break;
-		case PORT_D:
-			bit = SDE_PORTD_HOTPLUG;
-			break;
-		default:
-			return true;
-		}
-	} else {
-		switch (port->port) {
-		case PORT_B:
-			bit = SDE_PORTB_HOTPLUG_CPT;
-			break;
-		case PORT_C:
-			bit = SDE_PORTC_HOTPLUG_CPT;
-			break;
-		case PORT_D:
-			bit = SDE_PORTD_HOTPLUG_CPT;
-			break;
-		default:
-			return true;
-		}
-	}
-
-	return I915_READ(SDEISR) & bit;
-}
-
 static const char *state_string(bool enabled)
 {
 	return enabled ? "on" : "off";
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 4e5bd87..307a45b 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -4480,17 +4480,49 @@ edp_detect(struct intel_dp *intel_dp)
 	return status;
 }
 
-static enum drm_connector_status
-ironlake_dp_detect(struct intel_dp *intel_dp)
+/*
+ * ibx_digital_port_connected - is the specified port connected?
+ * @dev_priv: i915 private structure
+ * @port: the port to test
+ *
+ * Returns true if @port is connected, false otherwise.
+ */
+static bool ibx_digital_port_connected(struct drm_i915_private *dev_priv,
+				       struct intel_digital_port *port)
 {
-	struct drm_device *dev = intel_dp_to_dev(intel_dp);
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
+	u32 bit;
 
-	if (!ibx_digital_port_connected(dev_priv, intel_dig_port))
-		return connector_status_disconnected;
+	if (HAS_PCH_IBX(dev_priv->dev)) {
+		switch (port->port) {
+		case PORT_B:
+			bit = SDE_PORTB_HOTPLUG;
+			break;
+		case PORT_C:
+			bit = SDE_PORTC_HOTPLUG;
+			break;
+		case PORT_D:
+			bit = SDE_PORTD_HOTPLUG;
+			break;
+		default:
+			return true;
+		}
+	} else {
+		switch (port->port) {
+		case PORT_B:
+			bit = SDE_PORTB_HOTPLUG_CPT;
+			break;
+		case PORT_C:
+			bit = SDE_PORTC_HOTPLUG_CPT;
+			break;
+		case PORT_D:
+			bit = SDE_PORTD_HOTPLUG_CPT;
+			break;
+		default:
+			return true;
+		}
+	}
 
-	return intel_dp_detect_dpcd(intel_dp);
+	return I915_READ(SDEISR) & bit;
 }
 
 static int g4x_digital_port_connected(struct drm_device *dev,
@@ -4535,6 +4567,19 @@ static int g4x_digital_port_connected(struct drm_device *dev,
 }
 
 static enum drm_connector_status
+ironlake_dp_detect(struct intel_dp *intel_dp)
+{
+	struct drm_device *dev = intel_dp_to_dev(intel_dp);
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
+
+	if (!ibx_digital_port_connected(dev_priv, intel_dig_port))
+		return connector_status_disconnected;
+
+	return intel_dp_detect_dpcd(intel_dp);
+}
+
+static enum drm_connector_status
 g4x_dp_detect(struct intel_dp *intel_dp)
 {
 	struct drm_device *dev = intel_dp_to_dev(intel_dp);
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 49d1b61..aa79717 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -1018,8 +1018,6 @@ void intel_encoder_destroy(struct drm_encoder *encoder);
 int intel_connector_init(struct intel_connector *);
 struct intel_connector *intel_connector_alloc(void);
 bool intel_connector_get_hw_state(struct intel_connector *connector);
-bool ibx_digital_port_connected(struct drm_i915_private *dev_priv,
-				struct intel_digital_port *port);
 void intel_connector_attach_encoder(struct intel_connector *connector,
 				    struct intel_encoder *encoder);
 struct drm_encoder *intel_best_encoder(struct drm_connector *connector);
-- 
cgit v0.10.2


From 1d245987754f8cd0bf21c77ecda2023ca13ff8c1 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 20 Aug 2015 10:47:37 +0300
Subject: drm/i915: make g4x_digital_port_connected return boolean status

We should not be hitting any of the default cases in
g4x_digital_port_connected, so add MISSING_CASE annotation and return
boolean status. The current behaviour is just cargo culting from the
days of yonder when the display port support was added to i915.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Durgadoss R <durgadoss.r@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 307a45b..be53e36 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -4525,14 +4525,14 @@ static bool ibx_digital_port_connected(struct drm_i915_private *dev_priv,
 	return I915_READ(SDEISR) & bit;
 }
 
-static int g4x_digital_port_connected(struct drm_device *dev,
-				       struct intel_digital_port *intel_dig_port)
+static bool g4x_digital_port_connected(struct drm_device *dev,
+				       struct intel_digital_port *port)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	uint32_t bit;
 
 	if (IS_VALLEYVIEW(dev)) {
-		switch (intel_dig_port->port) {
+		switch (port->port) {
 		case PORT_B:
 			bit = PORTB_HOTPLUG_LIVE_STATUS_VLV;
 			break;
@@ -4543,10 +4543,11 @@ static int g4x_digital_port_connected(struct drm_device *dev,
 			bit = PORTD_HOTPLUG_LIVE_STATUS_VLV;
 			break;
 		default:
-			return -EINVAL;
+			MISSING_CASE(port->port);
+			return false;
 		}
 	} else {
-		switch (intel_dig_port->port) {
+		switch (port->port) {
 		case PORT_B:
 			bit = PORTB_HOTPLUG_LIVE_STATUS_G4X;
 			break;
@@ -4557,13 +4558,12 @@ static int g4x_digital_port_connected(struct drm_device *dev,
 			bit = PORTD_HOTPLUG_LIVE_STATUS_G4X;
 			break;
 		default:
-			return -EINVAL;
+			MISSING_CASE(port->port);
+			return false;
 		}
 	}
 
-	if ((I915_READ(PORT_HOTPLUG_STAT) & bit) == 0)
-		return 0;
-	return 1;
+	return I915_READ(PORT_HOTPLUG_STAT) & bit;
 }
 
 static enum drm_connector_status
@@ -4584,7 +4584,6 @@ g4x_dp_detect(struct intel_dp *intel_dp)
 {
 	struct drm_device *dev = intel_dp_to_dev(intel_dp);
 	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
-	int ret;
 
 	/* Can't disconnect eDP, but you can close the lid... */
 	if (is_edp(intel_dp)) {
@@ -4596,10 +4595,7 @@ g4x_dp_detect(struct intel_dp *intel_dp)
 		return status;
 	}
 
-	ret = g4x_digital_port_connected(dev, intel_dig_port);
-	if (ret == -EINVAL)
-		return connector_status_unknown;
-	else if (ret == 0)
+	if (!g4x_digital_port_connected(dev, intel_dig_port))
 		return connector_status_disconnected;
 
 	return intel_dp_detect_dpcd(intel_dp);
@@ -5066,7 +5062,7 @@ intel_dp_hpd_pulse(struct intel_digital_port *intel_dig_port, bool long_hpd)
 			if (!ibx_digital_port_connected(dev_priv, intel_dig_port))
 				goto mst_fail;
 		} else {
-			if (g4x_digital_port_connected(dev, intel_dig_port) != 1)
+			if (!g4x_digital_port_connected(dev, intel_dig_port))
 				goto mst_fail;
 		}
 
-- 
cgit v0.10.2


From 196cabd4a39496038ee81c2755496955e7355bd5 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 20 Aug 2015 10:47:38 +0300
Subject: drm/i915: add MISSING_CASE annotation to ibx_digital_port_connected

With the case added for eDP on port A (always connected from this
function's point of view), we should not be hitting any of the default
cases in ibx_digital_port_connected, so add MISSING_CASE annotation.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Durgadoss R <durgadoss.r@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index be53e36..8c8070e 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -4494,6 +4494,8 @@ static bool ibx_digital_port_connected(struct drm_i915_private *dev_priv,
 
 	if (HAS_PCH_IBX(dev_priv->dev)) {
 		switch (port->port) {
+		case PORT_A:
+			return true;
 		case PORT_B:
 			bit = SDE_PORTB_HOTPLUG;
 			break;
@@ -4504,10 +4506,13 @@ static bool ibx_digital_port_connected(struct drm_i915_private *dev_priv,
 			bit = SDE_PORTD_HOTPLUG;
 			break;
 		default:
-			return true;
+			MISSING_CASE(port->port);
+			return false;
 		}
 	} else {
 		switch (port->port) {
+		case PORT_A:
+			return true;
 		case PORT_B:
 			bit = SDE_PORTB_HOTPLUG_CPT;
 			break;
@@ -4518,7 +4523,8 @@ static bool ibx_digital_port_connected(struct drm_i915_private *dev_priv,
 			bit = SDE_PORTD_HOTPLUG_CPT;
 			break;
 		default:
-			return true;
+			MISSING_CASE(port->port);
+			return false;
 		}
 	}
 
-- 
cgit v0.10.2


From 7e66bcf2656ad2e53de939c5a5f211288a332ede Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 20 Aug 2015 10:47:39 +0300
Subject: drm/i915: add common intel_digital_port_connected function

Add a common intel_digital_port_connected() that splits out to functions
for different platforms. No functional changes.

v2: make the function return a boolean

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Durgadoss R <durgadoss.r@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 8c8070e..1db7e1e 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -4480,13 +4480,6 @@ edp_detect(struct intel_dp *intel_dp)
 	return status;
 }
 
-/*
- * ibx_digital_port_connected - is the specified port connected?
- * @dev_priv: i915 private structure
- * @port: the port to test
- *
- * Returns true if @port is connected, false otherwise.
- */
 static bool ibx_digital_port_connected(struct drm_i915_private *dev_priv,
 				       struct intel_digital_port *port)
 {
@@ -4531,13 +4524,12 @@ static bool ibx_digital_port_connected(struct drm_i915_private *dev_priv,
 	return I915_READ(SDEISR) & bit;
 }
 
-static bool g4x_digital_port_connected(struct drm_device *dev,
+static bool g4x_digital_port_connected(struct drm_i915_private *dev_priv,
 				       struct intel_digital_port *port)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	uint32_t bit;
 
-	if (IS_VALLEYVIEW(dev)) {
+	if (IS_VALLEYVIEW(dev_priv)) {
 		switch (port->port) {
 		case PORT_B:
 			bit = PORTB_HOTPLUG_LIVE_STATUS_VLV;
@@ -4572,6 +4564,22 @@ static bool g4x_digital_port_connected(struct drm_device *dev,
 	return I915_READ(PORT_HOTPLUG_STAT) & bit;
 }
 
+/*
+ * intel_digital_port_connected - is the specified port connected?
+ * @dev_priv: i915 private structure
+ * @port: the port to test
+ *
+ * Return %true if @port is connected, %false otherwise.
+ */
+static bool intel_digital_port_connected(struct drm_i915_private *dev_priv,
+					 struct intel_digital_port *port)
+{
+	if (HAS_PCH_SPLIT(dev_priv))
+		return ibx_digital_port_connected(dev_priv, port);
+	else
+		return g4x_digital_port_connected(dev_priv, port);
+}
+
 static enum drm_connector_status
 ironlake_dp_detect(struct intel_dp *intel_dp)
 {
@@ -4579,7 +4587,7 @@ ironlake_dp_detect(struct intel_dp *intel_dp)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp);
 
-	if (!ibx_digital_port_connected(dev_priv, intel_dig_port))
+	if (!intel_digital_port_connected(dev_priv, intel_dig_port))
 		return connector_status_disconnected;
 
 	return intel_dp_detect_dpcd(intel_dp);
@@ -4601,7 +4609,7 @@ g4x_dp_detect(struct intel_dp *intel_dp)
 		return status;
 	}
 
-	if (!g4x_digital_port_connected(dev, intel_dig_port))
+	if (!intel_digital_port_connected(dev->dev_private, intel_dig_port))
 		return connector_status_disconnected;
 
 	return intel_dp_detect_dpcd(intel_dp);
@@ -5064,13 +5072,8 @@ intel_dp_hpd_pulse(struct intel_digital_port *intel_dig_port, bool long_hpd)
 		/* indicate that we need to restart link training */
 		intel_dp->train_set_valid = false;
 
-		if (HAS_PCH_SPLIT(dev)) {
-			if (!ibx_digital_port_connected(dev_priv, intel_dig_port))
-				goto mst_fail;
-		} else {
-			if (!g4x_digital_port_connected(dev, intel_dig_port))
-				goto mst_fail;
-		}
+		if (!intel_digital_port_connected(dev_priv, intel_dig_port))
+			goto mst_fail;
 
 		if (!intel_dp_get_dpcd(intel_dp)) {
 			goto mst_fail;
-- 
cgit v0.10.2


From 0df53b77287198502ff2eb51120c041ba317f059 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 20 Aug 2015 10:47:40 +0300
Subject: drm/i915: split ibx_digital_port_connected to ibx and cpt variants

Choose the right function at the intel_digital_port_connected level.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Durgadoss R <durgadoss.r@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 1db7e1e..689dfe6 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -4485,40 +4485,46 @@ static bool ibx_digital_port_connected(struct drm_i915_private *dev_priv,
 {
 	u32 bit;
 
-	if (HAS_PCH_IBX(dev_priv->dev)) {
-		switch (port->port) {
-		case PORT_A:
-			return true;
-		case PORT_B:
-			bit = SDE_PORTB_HOTPLUG;
-			break;
-		case PORT_C:
-			bit = SDE_PORTC_HOTPLUG;
-			break;
-		case PORT_D:
-			bit = SDE_PORTD_HOTPLUG;
-			break;
-		default:
-			MISSING_CASE(port->port);
-			return false;
-		}
-	} else {
-		switch (port->port) {
-		case PORT_A:
-			return true;
-		case PORT_B:
-			bit = SDE_PORTB_HOTPLUG_CPT;
-			break;
-		case PORT_C:
-			bit = SDE_PORTC_HOTPLUG_CPT;
-			break;
-		case PORT_D:
-			bit = SDE_PORTD_HOTPLUG_CPT;
-			break;
-		default:
-			MISSING_CASE(port->port);
-			return false;
-		}
+	switch (port->port) {
+	case PORT_A:
+		return true;
+	case PORT_B:
+		bit = SDE_PORTB_HOTPLUG;
+		break;
+	case PORT_C:
+		bit = SDE_PORTC_HOTPLUG;
+		break;
+	case PORT_D:
+		bit = SDE_PORTD_HOTPLUG;
+		break;
+	default:
+		MISSING_CASE(port->port);
+		return false;
+	}
+
+	return I915_READ(SDEISR) & bit;
+}
+
+static bool cpt_digital_port_connected(struct drm_i915_private *dev_priv,
+				       struct intel_digital_port *port)
+{
+	u32 bit;
+
+	switch (port->port) {
+	case PORT_A:
+		return true;
+	case PORT_B:
+		bit = SDE_PORTB_HOTPLUG_CPT;
+		break;
+	case PORT_C:
+		bit = SDE_PORTC_HOTPLUG_CPT;
+		break;
+	case PORT_D:
+		bit = SDE_PORTD_HOTPLUG_CPT;
+		break;
+	default:
+		MISSING_CASE(port->port);
+		return false;
 	}
 
 	return I915_READ(SDEISR) & bit;
@@ -4574,8 +4580,10 @@ static bool g4x_digital_port_connected(struct drm_i915_private *dev_priv,
 static bool intel_digital_port_connected(struct drm_i915_private *dev_priv,
 					 struct intel_digital_port *port)
 {
-	if (HAS_PCH_SPLIT(dev_priv))
+	if (HAS_PCH_IBX(dev_priv))
 		return ibx_digital_port_connected(dev_priv, port);
+	if (HAS_PCH_SPLIT(dev_priv))
+		return cpt_digital_port_connected(dev_priv, port);
 	else
 		return g4x_digital_port_connected(dev_priv, port);
 }
-- 
cgit v0.10.2


From 9642c81c225a43982536fcb7a3fa212dcfaa67ea Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 20 Aug 2015 10:47:41 +0300
Subject: drm/i915: split g4x_digital_port_connected to g4x and vlv variants

Choose the right function at the intel_digital_port_connected level.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Durgadoss R <durgadoss.r@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 689dfe6..e7ebacd5 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -4533,38 +4533,44 @@ static bool cpt_digital_port_connected(struct drm_i915_private *dev_priv,
 static bool g4x_digital_port_connected(struct drm_i915_private *dev_priv,
 				       struct intel_digital_port *port)
 {
-	uint32_t bit;
+	u32 bit;
 
-	if (IS_VALLEYVIEW(dev_priv)) {
-		switch (port->port) {
-		case PORT_B:
-			bit = PORTB_HOTPLUG_LIVE_STATUS_VLV;
-			break;
-		case PORT_C:
-			bit = PORTC_HOTPLUG_LIVE_STATUS_VLV;
-			break;
-		case PORT_D:
-			bit = PORTD_HOTPLUG_LIVE_STATUS_VLV;
-			break;
-		default:
-			MISSING_CASE(port->port);
-			return false;
-		}
-	} else {
-		switch (port->port) {
-		case PORT_B:
-			bit = PORTB_HOTPLUG_LIVE_STATUS_G4X;
-			break;
-		case PORT_C:
-			bit = PORTC_HOTPLUG_LIVE_STATUS_G4X;
-			break;
-		case PORT_D:
-			bit = PORTD_HOTPLUG_LIVE_STATUS_G4X;
-			break;
-		default:
-			MISSING_CASE(port->port);
-			return false;
-		}
+	switch (port->port) {
+	case PORT_B:
+		bit = PORTB_HOTPLUG_LIVE_STATUS_G4X;
+		break;
+	case PORT_C:
+		bit = PORTC_HOTPLUG_LIVE_STATUS_G4X;
+		break;
+	case PORT_D:
+		bit = PORTD_HOTPLUG_LIVE_STATUS_G4X;
+		break;
+	default:
+		MISSING_CASE(port->port);
+		return false;
+	}
+
+	return I915_READ(PORT_HOTPLUG_STAT) & bit;
+}
+
+static bool vlv_digital_port_connected(struct drm_i915_private *dev_priv,
+				       struct intel_digital_port *port)
+{
+	u32 bit;
+
+	switch (port->port) {
+	case PORT_B:
+		bit = PORTB_HOTPLUG_LIVE_STATUS_VLV;
+		break;
+	case PORT_C:
+		bit = PORTC_HOTPLUG_LIVE_STATUS_VLV;
+		break;
+	case PORT_D:
+		bit = PORTD_HOTPLUG_LIVE_STATUS_VLV;
+		break;
+	default:
+		MISSING_CASE(port->port);
+		return false;
 	}
 
 	return I915_READ(PORT_HOTPLUG_STAT) & bit;
@@ -4584,6 +4590,8 @@ static bool intel_digital_port_connected(struct drm_i915_private *dev_priv,
 		return ibx_digital_port_connected(dev_priv, port);
 	if (HAS_PCH_SPLIT(dev_priv))
 		return cpt_digital_port_connected(dev_priv, port);
+	else if (IS_VALLEYVIEW(dev_priv))
+		return vlv_digital_port_connected(dev_priv, port);
 	else
 		return g4x_digital_port_connected(dev_priv, port);
 }
-- 
cgit v0.10.2


From e464bfdeda3da3b0ef29ca84960715993ccbb1fc Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 20 Aug 2015 10:47:42 +0300
Subject: drm/i915/bxt: Use correct live status register for BXT platform

BXT platform uses live status bits from 0x44440 register to obtain DP
status on hotplug. The existing g4x_digital_port_connected() uses a
different register and hence misses DP hotplug events on BXT
platform. This patch fixes it by using the appropriate register(0x44440)
and live status bits(3:5).

Based on a patch by Durgadoss R <durgadoss.r@intel.com>, from whom the
commit message is shamelessly copy pasted.

Reported-by: Durgadoss R <durgadoss.r@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Durgadoss R <durgadoss.r@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index e7ebacd5..1ff063e 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -4576,6 +4576,29 @@ static bool vlv_digital_port_connected(struct drm_i915_private *dev_priv,
 	return I915_READ(PORT_HOTPLUG_STAT) & bit;
 }
 
+static bool bxt_digital_port_connected(struct drm_i915_private *dev_priv,
+				       struct intel_digital_port *port)
+{
+	u32 bit;
+
+	switch (port->port) {
+	case PORT_A:
+		bit = BXT_DE_PORT_HP_DDIA;
+		break;
+	case PORT_B:
+		bit = BXT_DE_PORT_HP_DDIB;
+		break;
+	case PORT_C:
+		bit = BXT_DE_PORT_HP_DDIC;
+		break;
+	default:
+		MISSING_CASE(port->port);
+		return false;
+	}
+
+	return I915_READ(GEN8_DE_PORT_ISR) & bit;
+}
+
 /*
  * intel_digital_port_connected - is the specified port connected?
  * @dev_priv: i915 private structure
@@ -4590,6 +4613,8 @@ static bool intel_digital_port_connected(struct drm_i915_private *dev_priv,
 		return ibx_digital_port_connected(dev_priv, port);
 	if (HAS_PCH_SPLIT(dev_priv))
 		return cpt_digital_port_connected(dev_priv, port);
+	else if (IS_BROXTON(dev_priv))
+		return bxt_digital_port_connected(dev_priv, port);
 	else if (IS_VALLEYVIEW(dev_priv))
 		return vlv_digital_port_connected(dev_priv, port);
 	else
-- 
cgit v0.10.2


From 0a0b457fbf765d0b98fa48495b9bb7901eb8b091 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 21 Aug 2015 20:45:27 +0300
Subject: drm/i915: Fix some gcc warnings
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Simple one:
drivers/gpu/drm/i915/i915_debugfs.c:2449:57: warning: Using plain integer as NULL pointer

And something a bit more peculiar:
drivers/gpu/drm/i915/i915_debugfs.c:4953:18: warning: Variable length array is used.
drivers/gpu/drm/i915/i915_debugfs.c:4953:32: warning: Variable length array is used.

We pass a 'const int' as the array size which results in the warning,
dropping the const gets rid of the warning. Weird, but I think getting
rid of the warnings is better than holding on to the const.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index e192f02..624024c 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -2446,7 +2446,7 @@ static int i915_guc_info(struct seq_file *m, void *data)
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_guc guc;
-	struct i915_guc_client client = { .client_obj = 0 };
+	struct i915_guc_client client = {};
 	struct intel_engine_cs *ring;
 	enum intel_ring_id i;
 	u64 total = 0;
@@ -4948,7 +4948,7 @@ static void cherryview_sseu_device_status(struct drm_device *dev,
 					  struct sseu_dev_status *stat)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	const int ss_max = 2;
+	int ss_max = 2;
 	int ss;
 	u32 sig1[ss_max], sig2[ss_max];
 
-- 
cgit v0.10.2


From 53abb6794ac022e6c0b5f239fc26d99ea054ee86 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 21 Aug 2015 20:45:28 +0300
Subject: drm/i915: Use ARRAY_SIZE() instead of hand rolling it
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

A couple of hand rolled ARRAY_SIZE()s caught my eye. Get rid of them.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c
index c98098e..25d74d2 100644
--- a/drivers/gpu/drm/i915/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/intel_sdvo.c
@@ -63,7 +63,7 @@ static const char *tv_format_names[] = {
 	"SECAM_60"
 };
 
-#define TV_FORMAT_NUM  (sizeof(tv_format_names) / sizeof(*tv_format_names))
+#define TV_FORMAT_NUM  ARRAY_SIZE(tv_format_names)
 
 struct intel_sdvo {
 	struct intel_encoder base;
diff --git a/drivers/gpu/drm/i915/intel_tv.c b/drivers/gpu/drm/i915/intel_tv.c
index 0568ae6..cbe39dc 100644
--- a/drivers/gpu/drm/i915/intel_tv.c
+++ b/drivers/gpu/drm/i915/intel_tv.c
@@ -1291,7 +1291,7 @@ static void intel_tv_find_better_format(struct drm_connector *connector)
 		return;
 
 
-	for (i = 0; i < sizeof(tv_modes) / sizeof(*tv_modes); i++) {
+	for (i = 0; i < ARRAY_SIZE(tv_modes); i++) {
 		tv_mode = tv_modes + i;
 
 		if ((intel_tv->type == DRM_MODE_CONNECTOR_Component) ==
diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c
index 9d3c2e4..ad4e421 100644
--- a/drivers/gpu/drm/i915/intel_uncore.c
+++ b/drivers/gpu/drm/i915/intel_uncore.c
@@ -52,8 +52,7 @@ static const char * const forcewake_domain_names[] = {
 const char *
 intel_uncore_forcewake_domain_to_str(const enum forcewake_domain_id id)
 {
-	BUILD_BUG_ON((sizeof(forcewake_domain_names)/sizeof(const char *)) !=
-		     FW_DOMAIN_ID_COUNT);
+	BUILD_BUG_ON(ARRAY_SIZE(forcewake_domain_names) != FW_DOMAIN_ID_COUNT);
 
 	if (id >= 0 && id < FW_DOMAIN_ID_COUNT)
 		return forcewake_domain_names[id];
-- 
cgit v0.10.2


From 4d9194dec37a9bf22354f6a5872e285e1bb8c1da Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 21 Aug 2015 20:45:29 +0300
Subject: drm/i915: Make some string arrays const
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Most of our char* arrays are markes as const already, but a few slipped
through the cracks. Fix it.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c
index 25d74d2..ca3dd7c 100644
--- a/drivers/gpu/drm/i915/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/intel_sdvo.c
@@ -53,7 +53,7 @@
 #define IS_DIGITAL(c) (c->output_flag & (SDVO_TMDS_MASK | SDVO_LVDS_MASK))
 
 
-static const char *tv_format_names[] = {
+static const char * const tv_format_names[] = {
 	"NTSC_M"   , "NTSC_J"  , "NTSC_443",
 	"PAL_B"    , "PAL_D"   , "PAL_G"   ,
 	"PAL_H"    , "PAL_I"   , "PAL_M"   ,
@@ -452,7 +452,7 @@ static void intel_sdvo_debug_write(struct intel_sdvo *intel_sdvo, u8 cmd,
 	DRM_DEBUG_KMS("%s: W: %02X %s\n", SDVO_NAME(intel_sdvo), cmd, buffer);
 }
 
-static const char *cmd_status_names[] = {
+static const char * const cmd_status_names[] = {
 	"Power on",
 	"Success",
 	"Not supported",
-- 
cgit v0.10.2


From 5a8fbb7d192b96de3d258164e5fc95b769d698c3 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 29 Jun 2015 15:25:53 +0300
Subject: drm/i915: Move DPLL ref/cri/VGA mode frobbing to the disp2d well
 enable
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Bunch of stuff needs the DPLL ref/cri clocks on both VLV and CHV,
and having VGA mode enabled causes some problems for CHV. So let's just
pull the code to configure those bits into the disp2d well enable hook.
With the DPLL disable code also fixed to leave those bits alone we
should now have a consistent DPLL state all the time even if the DPLL
is disabled.

This also neatly removes some duplicated code between the VLV and
CHV codepaths.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Sivakumar Thulasimani <sivakumar.thulasimani@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c
index 974c4cb..0588baa 100644
--- a/drivers/gpu/drm/i915/intel_runtime_pm.c
+++ b/drivers/gpu/drm/i915/intel_runtime_pm.c
@@ -853,6 +853,25 @@ static bool vlv_power_well_enabled(struct drm_i915_private *dev_priv,
 
 static void vlv_display_power_well_init(struct drm_i915_private *dev_priv)
 {
+	enum pipe pipe;
+
+	/*
+	 * Enable the CRI clock source so we can get at the
+	 * display and the reference clock for VGA
+	 * hotplug / manual detection. Supposedly DSI also
+	 * needs the ref clock up and running.
+	 *
+	 * CHV DPLL B/C have some issues if VGA mode is enabled.
+	 */
+	for_each_pipe(dev_priv->dev, pipe) {
+		u32 val = I915_READ(DPLL(pipe));
+
+		val |= DPLL_REF_CLK_ENABLE_VLV | DPLL_VGA_MODE_DIS;
+		if (pipe != PIPE_A)
+			val |= DPLL_INTEGRATED_CRI_CLK_VLV;
+
+		I915_WRITE(DPLL(pipe), val);
+	}
 
 	spin_lock_irq(&dev_priv->irq_lock);
 	valleyview_enable_display_irqs(dev_priv);
@@ -904,13 +923,7 @@ static void vlv_dpio_cmn_power_well_enable(struct drm_i915_private *dev_priv,
 {
 	WARN_ON_ONCE(power_well->data != PUNIT_POWER_WELL_DPIO_CMN_BC);
 
-	/*
-	 * Enable the CRI clock source so we can get at the
-	 * display and the reference clock for VGA
-	 * hotplug / manual detection.
-	 */
-	I915_WRITE(DPLL(PIPE_B), I915_READ(DPLL(PIPE_B)) | DPLL_VGA_MODE_DIS |
-		   DPLL_REF_CLK_ENABLE_VLV | DPLL_INTEGRATED_CRI_CLK_VLV);
+	/* since ref/cri clock was enabled */
 	udelay(1); /* >10ns for cmnreset, >0ns for sidereset */
 
 	vlv_set_power_well(dev_priv, power_well, true);
@@ -953,22 +966,12 @@ static void chv_dpio_cmn_power_well_enable(struct drm_i915_private *dev_priv,
 	WARN_ON_ONCE(power_well->data != PUNIT_POWER_WELL_DPIO_CMN_BC &&
 		     power_well->data != PUNIT_POWER_WELL_DPIO_CMN_D);
 
-	/*
-	 * Enable the CRI clock source so we can get at the
-	 * display and the reference clock for VGA
-	 * hotplug / manual detection.
-	 */
-	if (power_well->data == PUNIT_POWER_WELL_DPIO_CMN_BC) {
+	if (power_well->data == PUNIT_POWER_WELL_DPIO_CMN_BC)
 		phy = DPIO_PHY0;
-		I915_WRITE(DPLL(PIPE_B), I915_READ(DPLL(PIPE_B)) | DPLL_VGA_MODE_DIS |
-			   DPLL_REF_CLK_ENABLE_VLV);
-		I915_WRITE(DPLL(PIPE_B), I915_READ(DPLL(PIPE_B)) | DPLL_VGA_MODE_DIS |
-			   DPLL_REF_CLK_ENABLE_VLV | DPLL_INTEGRATED_CRI_CLK_VLV);
-	} else {
+	else
 		phy = DPIO_PHY1;
-		I915_WRITE(DPLL(PIPE_C), I915_READ(DPLL(PIPE_C)) | DPLL_VGA_MODE_DIS |
-			   DPLL_REF_CLK_ENABLE_VLV | DPLL_INTEGRATED_CRI_CLK_VLV);
-	}
+
+	/* since ref/cri clock was enabled */
 	udelay(1); /* >10ns for cmnreset, >0ns for sidereset */
 	vlv_set_power_well(dev_priv, power_well, true);
 
-- 
cgit v0.10.2


From e0fce78f041014846d77940d3a350a4cffe4ab2b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 8 Jul 2015 23:45:54 +0300
Subject: drm/i915: Implement PHY lane power gating for CHV
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Powergate the PHY lanes when they're not needed. For HDMI all four lanes
are needed always, but for DP we can enable only the needed lanes. To
power down the unused lanes we use some power down override bits in the
DISPLAY_PHY_CONTROL register. Without the overrides it appears that the
hardware always powers on all the lanes. When the port is disabled the
power down override is not needed and the lanes will shut off on their
own. That also means the override is critical to actually be able to
access the DPIO registers before the port is actually enabled.

Additionally the common lanes will power down when not needed. CL1
remains on as long as anything else is on, CL2 will shut down when
all the lanes in the same channel will shut down. There is one exception
for CL2 that will be dealt in a separate patch for clarity.

With potentially some lanes powered down, the DP code now has to check
the number of active lanes before accessing PCS/TX registers. All
registers in powered down blocks will reads as 0xffffffff, and soe we
would drown in warnings from vlv_dpio_read() if we allowed the code
to access all those registers.

Another important detail in the DP code is the "TX latency optimal"
setting. Normally the second TX lane acts as some kind of reset master,
with the other lanes as slaves. But when only a single lane is enabled,
that single lane obviously has to be the master.

A bit of extra care is needed to reconstruct the initial state of the
DISPLAY_PHY_CONTROL register since it can't be read safely. So instead
read the actual lane status from the DPLL/PHY_STATUS registers and
use that to determine which lanes ought to be powergated initially.

We also need to switch the PHY power modes to "deep PSR" to avoid
a hard system hang when powering down the single channel PHY.

Also sprinkle a few debug prints around so that we can monitor the
DISPLAY_PHY_STATUS changes without having to read it and risk
corrupting it.

v2: Add locking to chv_powergate_phy_lanes()
v3: Actually enable dynamic powerdown in the PHY and deal with the
    fallout

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Deepak S <deepak.s@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 211b73a..6918ffa 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -1135,9 +1135,15 @@ enum skl_disp_power_wells {
 
 #define _CHV_CMN_DW19_CH0		0x814c
 #define _CHV_CMN_DW6_CH1		0x8098
+#define   DPIO_DYNPWRDOWNEN_CH1		(1 << 28) /* CL2 DW6 only */
 #define   CHV_CMN_USEDCLKCHANNEL	(1 << 13)
+
 #define CHV_CMN_DW19(ch) _PIPE(ch, _CHV_CMN_DW19_CH0, _CHV_CMN_DW6_CH1)
 
+#define CHV_CMN_DW28			0x8170
+#define   DPIO_CL1POWERDOWNEN		(1 << 23)
+#define   DPIO_DYNPWRDOWNEN_CH0		(1 << 22)
+
 #define CHV_CMN_DW30			0x8178
 #define   DPIO_LRC_BYPASS		(1 << 3)
 
@@ -2192,10 +2198,12 @@ enum skl_disp_power_wells {
 #define DPIO_PHY_STATUS			(VLV_DISPLAY_BASE + 0x6240)
 #define   DPLL_PORTD_READY_MASK		(0xf)
 #define DISPLAY_PHY_CONTROL (VLV_DISPLAY_BASE + 0x60100)
+#define   PHY_CH_POWER_DOWN_OVRD_EN(phy, ch)	(1 << (2*(phy)+(ch)+27))
 #define   PHY_LDO_DELAY_0NS			0x0
 #define   PHY_LDO_DELAY_200NS			0x1
 #define   PHY_LDO_DELAY_600NS			0x2
 #define   PHY_LDO_SEQ_DELAY(delay, phy)		((delay) << (2*(phy)+23))
+#define   PHY_CH_POWER_DOWN_OVRD(mask, phy, ch)	((mask) << (8*(phy)+4*(ch)+11))
 #define   PHY_CH_SU_PSR				0x1
 #define   PHY_CH_DEEP_PSR			0x7
 #define   PHY_CH_POWER_MODE(mode, phy, ch)	((mode) << (6*(phy)+3*(ch)+2))
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 1ff063e..c0eaf13 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -133,6 +133,11 @@ static void vlv_init_panel_power_sequencer(struct intel_dp *intel_dp);
 static void vlv_steal_power_sequencer(struct drm_device *dev,
 				      enum pipe pipe);
 
+static unsigned int intel_dp_unused_lane_mask(int lane_count)
+{
+	return ~((1 << lane_count) - 1) & 0xf;
+}
+
 static int
 intel_dp_max_link_bw(struct intel_dp  *intel_dp)
 {
@@ -2432,17 +2437,21 @@ static void chv_post_disable_dp(struct intel_encoder *encoder)
 	val |= CHV_PCS_REQ_SOFTRESET_EN;
 	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW1(ch), val);
 
-	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW1(ch));
-	val |= CHV_PCS_REQ_SOFTRESET_EN;
-	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW1(ch), val);
+	if (intel_crtc->config->lane_count > 2) {
+		val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW1(ch));
+		val |= CHV_PCS_REQ_SOFTRESET_EN;
+		vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW1(ch), val);
+	}
 
 	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW0(ch));
 	val &= ~(DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
 	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW0(ch), val);
 
-	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW0(ch));
-	val &= ~(DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
-	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW0(ch), val);
+	if (intel_crtc->config->lane_count > 2) {
+		val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW0(ch));
+		val &= ~(DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
+		vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW0(ch), val);
+	}
 
 	mutex_unlock(&dev_priv->sb_lock);
 }
@@ -2562,7 +2571,6 @@ static void intel_enable_dp(struct intel_encoder *encoder)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *crtc = to_intel_crtc(encoder->base.crtc);
 	uint32_t dp_reg = I915_READ(intel_dp->output_reg);
-	unsigned int lane_mask = 0x0;
 
 	if (WARN_ON(dp_reg & DP_PORT_EN))
 		return;
@@ -2580,9 +2588,15 @@ static void intel_enable_dp(struct intel_encoder *encoder)
 
 	pps_unlock(intel_dp);
 
-	if (IS_VALLEYVIEW(dev))
+	if (IS_VALLEYVIEW(dev)) {
+		unsigned int lane_mask = 0x0;
+
+		if (IS_CHERRYVIEW(dev))
+			lane_mask = intel_dp_unused_lane_mask(crtc->config->lane_count);
+
 		vlv_wait_port_ready(dev_priv, dp_to_dig_port(intel_dp),
 				    lane_mask);
+	}
 
 	intel_dp_sink_dpms(intel_dp, DRM_MODE_DPMS_ON);
 	intel_dp_start_link_train(intel_dp);
@@ -2809,31 +2823,40 @@ static void chv_pre_enable_dp(struct intel_encoder *encoder)
 	val &= ~DPIO_LANEDESKEW_STRAP_OVRD;
 	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW11(ch), val);
 
-	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW11(ch));
-	val &= ~DPIO_LANEDESKEW_STRAP_OVRD;
-	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW11(ch), val);
+	if (intel_crtc->config->lane_count > 2) {
+		val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW11(ch));
+		val &= ~DPIO_LANEDESKEW_STRAP_OVRD;
+		vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW11(ch), val);
+	}
 
 	/* Deassert soft data lane reset*/
 	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW1(ch));
 	val |= CHV_PCS_REQ_SOFTRESET_EN;
 	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW1(ch), val);
 
-	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW1(ch));
-	val |= CHV_PCS_REQ_SOFTRESET_EN;
-	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW1(ch), val);
+	if (intel_crtc->config->lane_count > 2) {
+		val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW1(ch));
+		val |= CHV_PCS_REQ_SOFTRESET_EN;
+		vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW1(ch), val);
+	}
 
 	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW0(ch));
 	val |= (DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
 	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW0(ch), val);
 
-	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW0(ch));
-	val |= (DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
-	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW0(ch), val);
+	if (intel_crtc->config->lane_count > 2) {
+		val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW0(ch));
+		val |= (DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
+		vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW0(ch), val);
+	}
 
 	/* Program Tx lane latency optimal setting*/
-	for (i = 0; i < 4; i++) {
+	for (i = 0; i < intel_crtc->config->lane_count; i++) {
 		/* Set the upar bit */
-		data = (i == 1) ? 0x0 : 0x1;
+		if (intel_crtc->config->lane_count == 1)
+			data = 0x0;
+		else
+			data = (i == 1) ? 0x0 : 0x1;
 		vlv_dpio_write(dev_priv, pipe, CHV_TX_DW14(ch, i),
 				data << DPIO_UPAR_SHIFT);
 	}
@@ -2854,9 +2877,11 @@ static void chv_pre_enable_dp(struct intel_encoder *encoder)
 	val |= DPIO_TX2_STAGGER_MASK(0x1f);
 	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW11(ch), val);
 
-	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW11(ch));
-	val |= DPIO_TX2_STAGGER_MASK(0x1f);
-	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW11(ch), val);
+	if (intel_crtc->config->lane_count > 2) {
+		val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW11(ch));
+		val |= DPIO_TX2_STAGGER_MASK(0x1f);
+		vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW11(ch), val);
+	}
 
 	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW12(ch),
 		       DPIO_LANESTAGGER_STRAP(stagger) |
@@ -2865,12 +2890,14 @@ static void chv_pre_enable_dp(struct intel_encoder *encoder)
 		       DPIO_TX1_STAGGER_MULT(6) |
 		       DPIO_TX2_STAGGER_MULT(0));
 
-	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW12(ch),
-		       DPIO_LANESTAGGER_STRAP(stagger) |
-		       DPIO_LANESTAGGER_STRAP_OVRD |
-		       DPIO_TX1_STAGGER_MASK(0x1f) |
-		       DPIO_TX1_STAGGER_MULT(7) |
-		       DPIO_TX2_STAGGER_MULT(5));
+	if (intel_crtc->config->lane_count > 2) {
+		vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW12(ch),
+			       DPIO_LANESTAGGER_STRAP(stagger) |
+			       DPIO_LANESTAGGER_STRAP_OVRD |
+			       DPIO_TX1_STAGGER_MASK(0x1f) |
+			       DPIO_TX1_STAGGER_MULT(7) |
+			       DPIO_TX2_STAGGER_MULT(5));
+	}
 
 	mutex_unlock(&dev_priv->sb_lock);
 
@@ -2886,10 +2913,14 @@ static void chv_dp_pre_pll_enable(struct intel_encoder *encoder)
 		to_intel_crtc(encoder->base.crtc);
 	enum dpio_channel ch = vlv_dport_to_channel(dport);
 	enum pipe pipe = intel_crtc->pipe;
+	unsigned int lane_mask =
+		intel_dp_unused_lane_mask(intel_crtc->config->lane_count);
 	u32 val;
 
 	intel_dp_prepare(encoder);
 
+	chv_phy_powergate_lanes(encoder, true, lane_mask);
+
 	mutex_lock(&dev_priv->sb_lock);
 
 	/* program left/right clock distribution */
@@ -2920,13 +2951,15 @@ static void chv_dp_pre_pll_enable(struct intel_encoder *encoder)
 		val |= CHV_PCS_USEDCLKCHANNEL;
 	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW8(ch), val);
 
-	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW8(ch));
-	val |= CHV_PCS_USEDCLKCHANNEL_OVRRIDE;
-	if (pipe != PIPE_B)
-		val &= ~CHV_PCS_USEDCLKCHANNEL;
-	else
-		val |= CHV_PCS_USEDCLKCHANNEL;
-	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW8(ch), val);
+	if (intel_crtc->config->lane_count > 2) {
+		val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW8(ch));
+		val |= CHV_PCS_USEDCLKCHANNEL_OVRRIDE;
+		if (pipe != PIPE_B)
+			val &= ~CHV_PCS_USEDCLKCHANNEL;
+		else
+			val |= CHV_PCS_USEDCLKCHANNEL;
+		vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW8(ch), val);
+	}
 
 	/*
 	 * This a a bit weird since generally CL
@@ -2963,6 +2996,8 @@ static void chv_dp_post_pll_disable(struct intel_encoder *encoder)
 	}
 
 	mutex_unlock(&dev_priv->sb_lock);
+
+	chv_phy_powergate_lanes(encoder, false, 0x0);
 }
 
 /*
@@ -3298,24 +3333,28 @@ static uint32_t chv_signal_levels(struct intel_dp *intel_dp)
 	val |= DPIO_PCS_TX1DEEMP_9P5 | DPIO_PCS_TX2DEEMP_9P5;
 	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW10(ch), val);
 
-	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW10(ch));
-	val &= ~(DPIO_PCS_SWING_CALC_TX0_TX2 | DPIO_PCS_SWING_CALC_TX1_TX3);
-	val &= ~(DPIO_PCS_TX1DEEMP_MASK | DPIO_PCS_TX2DEEMP_MASK);
-	val |= DPIO_PCS_TX1DEEMP_9P5 | DPIO_PCS_TX2DEEMP_9P5;
-	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW10(ch), val);
+	if (intel_crtc->config->lane_count > 2) {
+		val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW10(ch));
+		val &= ~(DPIO_PCS_SWING_CALC_TX0_TX2 | DPIO_PCS_SWING_CALC_TX1_TX3);
+		val &= ~(DPIO_PCS_TX1DEEMP_MASK | DPIO_PCS_TX2DEEMP_MASK);
+		val |= DPIO_PCS_TX1DEEMP_9P5 | DPIO_PCS_TX2DEEMP_9P5;
+		vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW10(ch), val);
+	}
 
 	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW9(ch));
 	val &= ~(DPIO_PCS_TX1MARGIN_MASK | DPIO_PCS_TX2MARGIN_MASK);
 	val |= DPIO_PCS_TX1MARGIN_000 | DPIO_PCS_TX2MARGIN_000;
 	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW9(ch), val);
 
-	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW9(ch));
-	val &= ~(DPIO_PCS_TX1MARGIN_MASK | DPIO_PCS_TX2MARGIN_MASK);
-	val |= DPIO_PCS_TX1MARGIN_000 | DPIO_PCS_TX2MARGIN_000;
-	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW9(ch), val);
+	if (intel_crtc->config->lane_count > 2) {
+		val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW9(ch));
+		val &= ~(DPIO_PCS_TX1MARGIN_MASK | DPIO_PCS_TX2MARGIN_MASK);
+		val |= DPIO_PCS_TX1MARGIN_000 | DPIO_PCS_TX2MARGIN_000;
+		vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW9(ch), val);
+	}
 
 	/* Program swing deemph */
-	for (i = 0; i < 4; i++) {
+	for (i = 0; i < intel_crtc->config->lane_count; i++) {
 		val = vlv_dpio_read(dev_priv, pipe, CHV_TX_DW4(ch, i));
 		val &= ~DPIO_SWING_DEEMPH9P5_MASK;
 		val |= deemph_reg_value << DPIO_SWING_DEEMPH9P5_SHIFT;
@@ -3323,7 +3362,7 @@ static uint32_t chv_signal_levels(struct intel_dp *intel_dp)
 	}
 
 	/* Program swing margin */
-	for (i = 0; i < 4; i++) {
+	for (i = 0; i < intel_crtc->config->lane_count; i++) {
 		val = vlv_dpio_read(dev_priv, pipe, CHV_TX_DW2(ch, i));
 
 		val &= ~DPIO_SWING_MARGIN000_MASK;
@@ -3346,7 +3385,7 @@ static uint32_t chv_signal_levels(struct intel_dp *intel_dp)
 	 * For now, for this unique transition scale selection, set bit
 	 * 27 for ch0 and ch1.
 	 */
-	for (i = 0; i < 4; i++) {
+	for (i = 0; i < intel_crtc->config->lane_count; i++) {
 		val = vlv_dpio_read(dev_priv, pipe, CHV_TX_DW3(ch, i));
 		if (chv_need_uniq_trans_scale(train_set))
 			val |= DPIO_TX_UNIQ_TRANS_SCALE_EN;
@@ -3360,9 +3399,11 @@ static uint32_t chv_signal_levels(struct intel_dp *intel_dp)
 	val |= DPIO_PCS_SWING_CALC_TX0_TX2 | DPIO_PCS_SWING_CALC_TX1_TX3;
 	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW10(ch), val);
 
-	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW10(ch));
-	val |= DPIO_PCS_SWING_CALC_TX0_TX2 | DPIO_PCS_SWING_CALC_TX1_TX3;
-	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW10(ch), val);
+	if (intel_crtc->config->lane_count > 2) {
+		val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW10(ch));
+		val |= DPIO_PCS_SWING_CALC_TX0_TX2 | DPIO_PCS_SWING_CALC_TX1_TX3;
+		vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW10(ch), val);
+	}
 
 	/* LRC Bypass */
 	val = vlv_dpio_read(dev_priv, pipe, CHV_CMN_DW30);
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index aa79717..df420d8 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -1361,6 +1361,10 @@ void intel_runtime_pm_put(struct drm_i915_private *dev_priv);
 
 void intel_display_set_init_power(struct drm_i915_private *dev, bool enable);
 
+void chv_phy_powergate_lanes(struct intel_encoder *encoder,
+			     bool override, unsigned int mask);
+
+
 /* intel_pm.c */
 void intel_init_clock_gating(struct drm_device *dev);
 void intel_suspend_hw(struct drm_device *dev);
diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index 269937a..f1ecc2c 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -1628,6 +1628,8 @@ static void chv_hdmi_pre_pll_enable(struct intel_encoder *encoder)
 
 	intel_hdmi_prepare(encoder);
 
+	chv_phy_powergate_lanes(encoder, true, 0x0);
+
 	mutex_lock(&dev_priv->sb_lock);
 
 	/* program left/right clock distribution */
@@ -1701,6 +1703,8 @@ static void chv_hdmi_post_pll_disable(struct intel_encoder *encoder)
 	}
 
 	mutex_unlock(&dev_priv->sb_lock);
+
+	chv_phy_powergate_lanes(encoder, false, 0x0);
 }
 
 static void vlv_hdmi_post_disable(struct intel_encoder *encoder)
diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c
index 0588baa..a159771 100644
--- a/drivers/gpu/drm/i915/intel_runtime_pm.c
+++ b/drivers/gpu/drm/i915/intel_runtime_pm.c
@@ -962,14 +962,19 @@ static void chv_dpio_cmn_power_well_enable(struct drm_i915_private *dev_priv,
 					   struct i915_power_well *power_well)
 {
 	enum dpio_phy phy;
+	enum pipe pipe;
+	uint32_t tmp;
 
 	WARN_ON_ONCE(power_well->data != PUNIT_POWER_WELL_DPIO_CMN_BC &&
 		     power_well->data != PUNIT_POWER_WELL_DPIO_CMN_D);
 
-	if (power_well->data == PUNIT_POWER_WELL_DPIO_CMN_BC)
+	if (power_well->data == PUNIT_POWER_WELL_DPIO_CMN_BC) {
+		pipe = PIPE_A;
 		phy = DPIO_PHY0;
-	else
+	} else {
+		pipe = PIPE_C;
 		phy = DPIO_PHY1;
+	}
 
 	/* since ref/cri clock was enabled */
 	udelay(1); /* >10ns for cmnreset, >0ns for sidereset */
@@ -979,8 +984,26 @@ static void chv_dpio_cmn_power_well_enable(struct drm_i915_private *dev_priv,
 	if (wait_for(I915_READ(DISPLAY_PHY_STATUS) & PHY_POWERGOOD(phy), 1))
 		DRM_ERROR("Display PHY %d is not power up\n", phy);
 
+	mutex_lock(&dev_priv->sb_lock);
+
+	/* Enable dynamic power down */
+	tmp = vlv_dpio_read(dev_priv, pipe, CHV_CMN_DW28);
+	tmp |= DPIO_DYNPWRDOWNEN_CH0 | DPIO_CL1POWERDOWNEN;
+	vlv_dpio_write(dev_priv, pipe, CHV_CMN_DW28, tmp);
+
+	if (power_well->data == PUNIT_POWER_WELL_DPIO_CMN_BC) {
+		tmp = vlv_dpio_read(dev_priv, pipe, _CHV_CMN_DW6_CH1);
+		tmp |= DPIO_DYNPWRDOWNEN_CH1;
+		vlv_dpio_write(dev_priv, pipe, _CHV_CMN_DW6_CH1, tmp);
+	}
+
+	mutex_unlock(&dev_priv->sb_lock);
+
 	dev_priv->chv_phy_control |= PHY_COM_LANE_RESET_DEASSERT(phy);
 	I915_WRITE(DISPLAY_PHY_CONTROL, dev_priv->chv_phy_control);
+
+	DRM_DEBUG_KMS("Enabled DPIO PHY%d (PHY_CONTROL=0x%08x)\n",
+		      phy, dev_priv->chv_phy_control);
 }
 
 static void chv_dpio_cmn_power_well_disable(struct drm_i915_private *dev_priv,
@@ -1004,6 +1027,35 @@ static void chv_dpio_cmn_power_well_disable(struct drm_i915_private *dev_priv,
 	I915_WRITE(DISPLAY_PHY_CONTROL, dev_priv->chv_phy_control);
 
 	vlv_set_power_well(dev_priv, power_well, false);
+
+	DRM_DEBUG_KMS("Disabled DPIO PHY%d (PHY_CONTROL=0x%08x)\n",
+		      phy, dev_priv->chv_phy_control);
+}
+
+void chv_phy_powergate_lanes(struct intel_encoder *encoder,
+			     bool override, unsigned int mask)
+{
+	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	struct i915_power_domains *power_domains = &dev_priv->power_domains;
+	enum dpio_phy phy = vlv_dport_to_phy(enc_to_dig_port(&encoder->base));
+	enum dpio_channel ch = vlv_dport_to_channel(enc_to_dig_port(&encoder->base));
+
+	mutex_lock(&power_domains->lock);
+
+	dev_priv->chv_phy_control &= ~PHY_CH_POWER_DOWN_OVRD(0xf, phy, ch);
+	dev_priv->chv_phy_control |= PHY_CH_POWER_DOWN_OVRD(mask, phy, ch);
+
+	if (override)
+		dev_priv->chv_phy_control |= PHY_CH_POWER_DOWN_OVRD_EN(phy, ch);
+	else
+		dev_priv->chv_phy_control &= ~PHY_CH_POWER_DOWN_OVRD_EN(phy, ch);
+
+	I915_WRITE(DISPLAY_PHY_CONTROL, dev_priv->chv_phy_control);
+
+	DRM_DEBUG_KMS("Power gating DPIO PHY%d CH%d lanes 0x%x (PHY_CONTROL=0x%08x)\n",
+		      phy, ch, mask, dev_priv->chv_phy_control);
+
+	mutex_unlock(&power_domains->lock);
 }
 
 static bool chv_pipe_power_well_enabled(struct drm_i915_private *dev_priv,
@@ -1630,19 +1682,72 @@ static void chv_phy_control_init(struct drm_i915_private *dev_priv)
 	 * DISPLAY_PHY_CONTROL can get corrupted if read. As a
 	 * workaround never ever read DISPLAY_PHY_CONTROL, and
 	 * instead maintain a shadow copy ourselves. Use the actual
-	 * power well state to reconstruct the expected initial
-	 * value.
+	 * power well state and lane status to reconstruct the
+	 * expected initial value.
 	 */
 	dev_priv->chv_phy_control =
 		PHY_LDO_SEQ_DELAY(PHY_LDO_DELAY_600NS, DPIO_PHY0) |
 		PHY_LDO_SEQ_DELAY(PHY_LDO_DELAY_600NS, DPIO_PHY1) |
-		PHY_CH_POWER_MODE(PHY_CH_SU_PSR, DPIO_PHY0, DPIO_CH0) |
-		PHY_CH_POWER_MODE(PHY_CH_SU_PSR, DPIO_PHY0, DPIO_CH1) |
-		PHY_CH_POWER_MODE(PHY_CH_SU_PSR, DPIO_PHY1, DPIO_CH0);
-	if (cmn_bc->ops->is_enabled(dev_priv, cmn_bc))
+		PHY_CH_POWER_MODE(PHY_CH_DEEP_PSR, DPIO_PHY0, DPIO_CH0) |
+		PHY_CH_POWER_MODE(PHY_CH_DEEP_PSR, DPIO_PHY0, DPIO_CH1) |
+		PHY_CH_POWER_MODE(PHY_CH_DEEP_PSR, DPIO_PHY1, DPIO_CH0);
+
+	/*
+	 * If all lanes are disabled we leave the override disabled
+	 * with all power down bits cleared to match the state we
+	 * would use after disabling the port. Otherwise enable the
+	 * override and set the lane powerdown bits accding to the
+	 * current lane status.
+	 */
+	if (cmn_bc->ops->is_enabled(dev_priv, cmn_bc)) {
+		uint32_t status = I915_READ(DPLL(PIPE_A));
+		unsigned int mask;
+
+		mask = status & DPLL_PORTB_READY_MASK;
+		if (mask == 0xf)
+			mask = 0x0;
+		else
+			dev_priv->chv_phy_control |=
+				PHY_CH_POWER_DOWN_OVRD_EN(DPIO_PHY0, DPIO_CH0);
+
+		dev_priv->chv_phy_control |=
+			PHY_CH_POWER_DOWN_OVRD(mask, DPIO_PHY0, DPIO_CH0);
+
+		mask = (status & DPLL_PORTC_READY_MASK) >> 4;
+		if (mask == 0xf)
+			mask = 0x0;
+		else
+			dev_priv->chv_phy_control |=
+				PHY_CH_POWER_DOWN_OVRD_EN(DPIO_PHY0, DPIO_CH1);
+
+		dev_priv->chv_phy_control |=
+			PHY_CH_POWER_DOWN_OVRD(mask, DPIO_PHY0, DPIO_CH1);
+
 		dev_priv->chv_phy_control |= PHY_COM_LANE_RESET_DEASSERT(DPIO_PHY0);
-	if (cmn_d->ops->is_enabled(dev_priv, cmn_d))
+	}
+
+	if (cmn_d->ops->is_enabled(dev_priv, cmn_d)) {
+		uint32_t status = I915_READ(DPIO_PHY_STATUS);
+		unsigned int mask;
+
+		mask = status & DPLL_PORTD_READY_MASK;
+
+		if (mask == 0xf)
+			mask = 0x0;
+		else
+			dev_priv->chv_phy_control |=
+				PHY_CH_POWER_DOWN_OVRD_EN(DPIO_PHY1, DPIO_CH0);
+
+		dev_priv->chv_phy_control |=
+			PHY_CH_POWER_DOWN_OVRD(mask, DPIO_PHY1, DPIO_CH0);
+
 		dev_priv->chv_phy_control |= PHY_COM_LANE_RESET_DEASSERT(DPIO_PHY1);
+	}
+
+	I915_WRITE(DISPLAY_PHY_CONTROL, dev_priv->chv_phy_control);
+
+	DRM_DEBUG_KMS("Initial PHY_CONTROL=0x%08x\n",
+		      dev_priv->chv_phy_control);
 }
 
 static void vlv_cmnlane_wa(struct drm_i915_private *dev_priv)
-- 
cgit v0.10.2


From b0b3384612bd4ce608c5d95626149311bb43f121 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 8 Jul 2015 23:45:55 +0300
Subject: drm/i915: Trick CL2 into life on CHV when using pipe B with port B
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Normmally the common lane in a PHY channel gets powered up when some
of the data lanes get powered up. But when we're driving port B with
pipe B we don't want to enabled any of the data lanes, and just want
the DPLL in the common lane to be active.

To make that happens we have to temporarily enable some data lanes
after which we can access the DPLL registers in the common lane. Once
the pipe is up and running we can drop the power override on the data
lanes allowing them to shut down. From this point forward the common
lane will in fact stay powered on until the data lanes in the other
channel get powered down.

Ville's extended explanation from the review thread:

On Wed, Aug 19, 2015 at 07:47:41AM +0530, Deepak wrote:
> One Q, why only for port B? Port C is also in same common lane right?

Port B is in the first PHY channel which also houses CL1. CL1 always
powers up whenever any lanes in either PHY channel are powered up.
CL2 only powers up if lanes in the second channel (ie. the one with
port C) powers up.

So in this scenario (pipe B->port B) we want the DPLL from CL2, but
ideally we only want to power up the lanes for port B. Powering up
port B lanes will only power up CL1, but as we need CL2 instead we
need to, temporarily, power up some lanes in port C as well.

Crossing the streams the other way (pipe A->port C) is not a problem
since CL1 powers up whenever anything else powers up. So powering up
some port C lanes is enough on its own to make the CL1 DPLL
operational, even though CL1 and the lanes live in separate channels.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Deepak S <deepak.s@linux.intel.com>
[danvet: Amend commit message with extended explanation.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index c0eaf13..b032eef 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -2902,6 +2902,12 @@ static void chv_pre_enable_dp(struct intel_encoder *encoder)
 	mutex_unlock(&dev_priv->sb_lock);
 
 	intel_enable_dp(encoder);
+
+	/* Second common lane will stay alive on its own now */
+	if (dport->release_cl2_override) {
+		chv_phy_powergate_ch(dev_priv, DPIO_PHY0, DPIO_CH1, false);
+		dport->release_cl2_override = false;
+	}
 }
 
 static void chv_dp_pre_pll_enable(struct intel_encoder *encoder)
@@ -2919,6 +2925,14 @@ static void chv_dp_pre_pll_enable(struct intel_encoder *encoder)
 
 	intel_dp_prepare(encoder);
 
+	/*
+	 * Must trick the second common lane into life.
+	 * Otherwise we can't even access the PLL.
+	 */
+	if (ch == DPIO_CH0 && pipe == PIPE_B)
+		dport->release_cl2_override =
+			!chv_phy_powergate_ch(dev_priv, DPIO_PHY0, DPIO_CH1, true);
+
 	chv_phy_powergate_lanes(encoder, true, lane_mask);
 
 	mutex_lock(&dev_priv->sb_lock);
@@ -2997,6 +3011,15 @@ static void chv_dp_post_pll_disable(struct intel_encoder *encoder)
 
 	mutex_unlock(&dev_priv->sb_lock);
 
+	/*
+	 * Leave the power down bit cleared for at least one
+	 * lane so that chv_powergate_phy_ch() will power
+	 * on something when the channel is otherwise unused.
+	 * When the port is off and the override is removed
+	 * the lanes power down anyway, so otherwise it doesn't
+	 * really matter what the state of power down bits is
+	 * after this.
+	 */
 	chv_phy_powergate_lanes(encoder, false, 0x0);
 }
 
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index df420d8..72c1181 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -779,6 +779,7 @@ struct intel_digital_port {
 	struct intel_dp dp;
 	struct intel_hdmi hdmi;
 	enum irqreturn (*hpd_pulse)(struct intel_digital_port *, bool);
+	bool release_cl2_override;
 };
 
 struct intel_dp_mst_encoder {
@@ -1363,6 +1364,8 @@ void intel_display_set_init_power(struct drm_i915_private *dev, bool enable);
 
 void chv_phy_powergate_lanes(struct intel_encoder *encoder,
 			     bool override, unsigned int mask);
+bool chv_phy_powergate_ch(struct drm_i915_private *dev_priv, enum dpio_phy phy,
+			  enum dpio_channel ch, bool override);
 
 
 /* intel_pm.c */
diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index f1ecc2c..7c56053 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -1628,6 +1628,14 @@ static void chv_hdmi_pre_pll_enable(struct intel_encoder *encoder)
 
 	intel_hdmi_prepare(encoder);
 
+	/*
+	 * Must trick the second common lane into life.
+	 * Otherwise we can't even access the PLL.
+	 */
+	if (ch == DPIO_CH0 && pipe == PIPE_B)
+		dport->release_cl2_override =
+			!chv_phy_powergate_ch(dev_priv, DPIO_PHY0, DPIO_CH1, true);
+
 	chv_phy_powergate_lanes(encoder, true, 0x0);
 
 	mutex_lock(&dev_priv->sb_lock);
@@ -1704,6 +1712,15 @@ static void chv_hdmi_post_pll_disable(struct intel_encoder *encoder)
 
 	mutex_unlock(&dev_priv->sb_lock);
 
+	/*
+	 * Leave the power down bit cleared for at least one
+	 * lane so that chv_powergate_phy_ch() will power
+	 * on something when the channel is otherwise unused.
+	 * When the port is off and the override is removed
+	 * the lanes power down anyway, so otherwise it doesn't
+	 * really matter what the state of power down bits is
+	 * after this.
+	 */
 	chv_phy_powergate_lanes(encoder, false, 0x0);
 }
 
@@ -1925,6 +1942,12 @@ static void chv_hdmi_pre_enable(struct intel_encoder *encoder)
 	g4x_enable_hdmi(encoder);
 
 	vlv_wait_port_ready(dev_priv, dport, 0x0);
+
+	/* Second common lane will stay alive on its own now */
+	if (dport->release_cl2_override) {
+		chv_phy_powergate_ch(dev_priv, DPIO_PHY0, DPIO_CH1, false);
+		dport->release_cl2_override = false;
+	}
 }
 
 static void intel_hdmi_destroy(struct drm_connector *connector)
diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c
index a159771..4a43885 100644
--- a/drivers/gpu/drm/i915/intel_runtime_pm.c
+++ b/drivers/gpu/drm/i915/intel_runtime_pm.c
@@ -1032,6 +1032,35 @@ static void chv_dpio_cmn_power_well_disable(struct drm_i915_private *dev_priv,
 		      phy, dev_priv->chv_phy_control);
 }
 
+bool chv_phy_powergate_ch(struct drm_i915_private *dev_priv, enum dpio_phy phy,
+			  enum dpio_channel ch, bool override)
+{
+	struct i915_power_domains *power_domains = &dev_priv->power_domains;
+	bool was_override;
+
+	mutex_lock(&power_domains->lock);
+
+	was_override = dev_priv->chv_phy_control & PHY_CH_POWER_DOWN_OVRD_EN(phy, ch);
+
+	if (override == was_override)
+		goto out;
+
+	if (override)
+		dev_priv->chv_phy_control |= PHY_CH_POWER_DOWN_OVRD_EN(phy, ch);
+	else
+		dev_priv->chv_phy_control &= ~PHY_CH_POWER_DOWN_OVRD_EN(phy, ch);
+
+	I915_WRITE(DISPLAY_PHY_CONTROL, dev_priv->chv_phy_control);
+
+	DRM_DEBUG_KMS("Power gating DPIO PHY%d CH%d (DPIO_PHY_CONTROL=0x%08x)\n",
+		      phy, ch, dev_priv->chv_phy_control);
+
+out:
+	mutex_unlock(&power_domains->lock);
+
+	return was_override;
+}
+
 void chv_phy_powergate_lanes(struct intel_encoder *encoder,
 			     bool override, unsigned int mask)
 {
-- 
cgit v0.10.2


From 0047eedc48869f8c7797dd10f0cf976ac34c1d33 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 10 Jul 2015 10:56:24 +0300
Subject: drm/i915: Force common lane on for the PPS kick on CHV
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

With DPIO powergating active the DPLL can't be accessed unless
something else is keeping the common lane in the channel on.
That means the PPS kick procedure could fail to enable the PLL.

Power up some data lanes to force the common lane to power up
so that the PLL can be enabled temporarily.

v2: Avoid gcc uninitilized variable warning

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Deepak S <deepak.s@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index b032eef..ea11469 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -341,7 +341,9 @@ vlv_power_sequencer_kick(struct intel_dp *intel_dp)
 	struct drm_device *dev = intel_dig_port->base.base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	enum pipe pipe = intel_dp->pps_pipe;
-	bool pll_enabled;
+	bool pll_enabled, release_cl_override = false;
+	enum dpio_phy phy = DPIO_PHY(pipe);
+	enum dpio_channel ch = vlv_pipe_to_channel(pipe);
 	uint32_t DP;
 
 	if (WARN(I915_READ(intel_dp->output_reg) & DP_PORT_EN,
@@ -371,9 +373,13 @@ vlv_power_sequencer_kick(struct intel_dp *intel_dp)
 	 * The DPLL for the pipe must be enabled for this to work.
 	 * So enable temporarily it if it's not already enabled.
 	 */
-	if (!pll_enabled)
+	if (!pll_enabled) {
+		release_cl_override = IS_CHERRYVIEW(dev) &&
+			!chv_phy_powergate_ch(dev_priv, phy, ch, true);
+
 		vlv_force_pll_on(dev, pipe, IS_CHERRYVIEW(dev) ?
 				 &chv_dpll[0].dpll : &vlv_dpll[0].dpll);
+	}
 
 	/*
 	 * Similar magic as in intel_dp_enable_port().
@@ -390,8 +396,12 @@ vlv_power_sequencer_kick(struct intel_dp *intel_dp)
 	I915_WRITE(intel_dp->output_reg, DP & ~DP_PORT_EN);
 	POSTING_READ(intel_dp->output_reg);
 
-	if (!pll_enabled)
+	if (!pll_enabled) {
 		vlv_force_pll_off(dev, pipe);
+
+		if (release_cl_override)
+			chv_phy_powergate_ch(dev_priv, phy, ch, false);
+	}
 }
 
 static enum pipe
-- 
cgit v0.10.2


From ee27921824e6ad0ca2d8e5abfa12cf4d853ded6c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 8 Jul 2015 23:45:57 +0300
Subject: drm/i915: Enable DPIO SUS clock gating on CHV
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

CHV has supports some form of automagic clock gating for the
DPIO SUS clock. We can simply enable the magic bits and the
hardware should take care of the rest.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Deepak S <deepak.s@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 6918ffa..6ed7141 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -1143,6 +1143,10 @@ enum skl_disp_power_wells {
 #define CHV_CMN_DW28			0x8170
 #define   DPIO_CL1POWERDOWNEN		(1 << 23)
 #define   DPIO_DYNPWRDOWNEN_CH0		(1 << 22)
+#define   DPIO_SUS_CLK_CONFIG_ON		(0 << 0)
+#define   DPIO_SUS_CLK_CONFIG_CLKREQ		(1 << 0)
+#define   DPIO_SUS_CLK_CONFIG_GATE		(2 << 0)
+#define   DPIO_SUS_CLK_CONFIG_GATE_CLKREQ	(3 << 0)
 
 #define CHV_CMN_DW30			0x8178
 #define   DPIO_LRC_BYPASS		(1 << 3)
diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c
index 4a43885..ef043b2 100644
--- a/drivers/gpu/drm/i915/intel_runtime_pm.c
+++ b/drivers/gpu/drm/i915/intel_runtime_pm.c
@@ -988,7 +988,8 @@ static void chv_dpio_cmn_power_well_enable(struct drm_i915_private *dev_priv,
 
 	/* Enable dynamic power down */
 	tmp = vlv_dpio_read(dev_priv, pipe, CHV_CMN_DW28);
-	tmp |= DPIO_DYNPWRDOWNEN_CH0 | DPIO_CL1POWERDOWNEN;
+	tmp |= DPIO_DYNPWRDOWNEN_CH0 | DPIO_CL1POWERDOWNEN |
+		DPIO_SUS_CLK_CONFIG_GATE_CLKREQ;
 	vlv_dpio_write(dev_priv, pipe, CHV_CMN_DW28, tmp);
 
 	if (power_well->data == PUNIT_POWER_WELL_DPIO_CMN_BC) {
-- 
cgit v0.10.2


From 3e28878635cc3bb3159445dc9cfbdc3d34eb8daf Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 8 Jul 2015 23:45:58 +0300
Subject: drm/i915: Force CL2 off in CHV x1 PHY
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We can choose to leave the display PHY CL2 powerdown up to some hardware
signals, or we can force it. The BXT code forces the nonexistent CL2 in
the x1 PHY to power down. Follow suit on CHV. Maybe it can still save
some extra power by disabling some extra logic in CL1, or something.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Deepak S <deepak.s@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 6ed7141..439bb27 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -1149,6 +1149,7 @@ enum skl_disp_power_wells {
 #define   DPIO_SUS_CLK_CONFIG_GATE_CLKREQ	(3 << 0)
 
 #define CHV_CMN_DW30			0x8178
+#define   DPIO_CL2_LDOFUSE_PWRENB	(1 << 6)
 #define   DPIO_LRC_BYPASS		(1 << 3)
 
 #define _TXLANE(ch, lane, offset) ((ch ? 0x2400 : 0) + \
diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c
index ef043b2..a0df156 100644
--- a/drivers/gpu/drm/i915/intel_runtime_pm.c
+++ b/drivers/gpu/drm/i915/intel_runtime_pm.c
@@ -996,6 +996,15 @@ static void chv_dpio_cmn_power_well_enable(struct drm_i915_private *dev_priv,
 		tmp = vlv_dpio_read(dev_priv, pipe, _CHV_CMN_DW6_CH1);
 		tmp |= DPIO_DYNPWRDOWNEN_CH1;
 		vlv_dpio_write(dev_priv, pipe, _CHV_CMN_DW6_CH1, tmp);
+	} else {
+		/*
+		 * Force the non-existing CL2 off. BXT does this
+		 * too, so maybe it saves some power even though
+		 * CL2 doesn't exist?
+		 */
+		tmp = vlv_dpio_read(dev_priv, pipe, CHV_CMN_DW30);
+		tmp |= DPIO_CL2_LDOFUSE_PWRENB;
+		vlv_dpio_write(dev_priv, pipe, CHV_CMN_DW30, tmp);
 	}
 
 	mutex_unlock(&dev_priv->sb_lock);
-- 
cgit v0.10.2


From 6c908bf43df417b10917c3a463d3bd413eb476fd Mon Sep 17 00:00:00 2001
From: Paulo Zanoni <paulo.r.zanoni@intel.com>
Date: Tue, 25 Aug 2015 19:03:41 -0300
Subject: drm/i915: gen 9 can check for unclaimed registers too

Dear git bisect user,

Even though this is the patch that introduced the WARN() you're
bisecting, please notice that it's very likely that the problem you're
facing was already present before this commit. In other words: this
commit adds code to detect errors and give WARN()s about them, but the
errors were already there.

In order to continue your debug, please use the i915.mmio_debug
option, check the backtraces and try to discover which read or write
operation is causing the error message. Then check if this is
happening because the register does not exist or because its power
well is down when the operation is being done.

On my SKL machine, if I use i915.mmio_debug=999, this patch triggers
42 WARNs just by booting. I didn't investigate them yet. Normal users
are only going to get a single WARN due to the default i915.mmio_debug
setting.

Thank you for your comprehension,
Paulo

Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 1d88745..6e03e11 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -362,6 +362,7 @@ static const struct intel_device_info intel_skylake_info = {
 	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING,
 	.has_llc = 1,
 	.has_ddi = 1,
+	.has_fpga_dbg = 1,
 	.has_fbc = 1,
 	GEN_DEFAULT_PIPEOFFSETS,
 	IVB_CURSOR_OFFSETS,
@@ -374,6 +375,7 @@ static const struct intel_device_info intel_skylake_gt3_info = {
 	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING | BSD2_RING,
 	.has_llc = 1,
 	.has_ddi = 1,
+	.has_fpga_dbg = 1,
 	.has_fbc = 1,
 	GEN_DEFAULT_PIPEOFFSETS,
 	IVB_CURSOR_OFFSETS,
@@ -386,6 +388,7 @@ static const struct intel_device_info intel_broxton_info = {
 	.ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING,
 	.num_pipes = 3,
 	.has_ddi = 1,
+	.has_fpga_dbg = 1,
 	.has_fbc = 1,
 	GEN_DEFAULT_PIPEOFFSETS,
 	IVB_CURSOR_OFFSETS,
diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c
index ad4e421..dec20d6 100644
--- a/drivers/gpu/drm/i915/intel_uncore.c
+++ b/drivers/gpu/drm/i915/intel_uncore.c
@@ -769,6 +769,7 @@ static u##x \
 gen9_read##x(struct drm_i915_private *dev_priv, off_t reg, bool trace) { \
 	enum forcewake_domains fw_engine; \
 	GEN6_READ_HEADER(x); \
+	hsw_unclaimed_reg_debug(dev_priv, reg, true, true); \
 	if (!SKL_NEEDS_FORCE_WAKE((dev_priv), (reg)))	\
 		fw_engine = 0; \
 	else if (FORCEWAKE_GEN9_RENDER_RANGE_OFFSET(reg))	\
@@ -782,6 +783,7 @@ gen9_read##x(struct drm_i915_private *dev_priv, off_t reg, bool trace) { \
 	if (fw_engine) \
 		__force_wake_get(dev_priv, fw_engine); \
 	val = __raw_i915_read##x(dev_priv, reg); \
+	hsw_unclaimed_reg_debug(dev_priv, reg, true, false); \
 	GEN6_READ_FOOTER; \
 }
 
@@ -982,6 +984,7 @@ gen9_write##x(struct drm_i915_private *dev_priv, off_t reg, u##x val, \
 		bool trace) { \
 	enum forcewake_domains fw_engine; \
 	GEN6_WRITE_HEADER; \
+	hsw_unclaimed_reg_debug(dev_priv, reg, false, true); \
 	if (!SKL_NEEDS_FORCE_WAKE((dev_priv), (reg)) ||	\
 	    is_gen9_shadowed(dev_priv, reg)) \
 		fw_engine = 0; \
@@ -996,6 +999,8 @@ gen9_write##x(struct drm_i915_private *dev_priv, off_t reg, u##x val, \
 	if (fw_engine) \
 		__force_wake_get(dev_priv, fw_engine); \
 	__raw_i915_write##x(dev_priv, reg, val); \
+	hsw_unclaimed_reg_debug(dev_priv, reg, false, false); \
+	hsw_unclaimed_reg_detect(dev_priv); \
 	GEN6_WRITE_FOOTER; \
 }
 
-- 
cgit v0.10.2


From c5b852f33e7844f4664794a3b3c738d07b09ce0a Mon Sep 17 00:00:00 2001
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Date: Wed, 26 Aug 2015 09:29:56 +0200
Subject: Partially revert "drm/i915: Use full atomic modeset."

This partially reverts commit 74c090b1bdc57b1c9f1361908cca5a3d8a80fb08.

The DRIVER_ATOMIC cap cannot yet be exported because i915 lacks async
support.

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 6e03e11..ce3bd0c 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -1726,7 +1726,7 @@ static int __init i915_init(void)
 	 * to the atomic ioctl and the atomic properties.  Only plane operations on
 	 * a single CRTC will actually work.
 	 */
-	if (driver.driver_features & DRIVER_MODESET)
+	if (i915.nuclear_pageflip)
 		driver.driver_features |= DRIVER_ATOMIC;
 
 	return drm_pci_init(&driver, &i915_pci_driver);
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 705f208..7cd1555 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -2640,6 +2640,7 @@ struct i915_params {
 	int use_mmio_flip;
 	int mmio_debug;
 	bool verbose_state_checks;
+	bool nuclear_pageflip;
 	int edp_vswing;
 };
 extern struct i915_params i915 __read_mostly;
diff --git a/drivers/gpu/drm/i915/i915_params.c b/drivers/gpu/drm/i915/i915_params.c
index 5ae4b0a..05053e2 100644
--- a/drivers/gpu/drm/i915/i915_params.c
+++ b/drivers/gpu/drm/i915/i915_params.c
@@ -51,6 +51,7 @@ struct i915_params i915 __read_mostly = {
 	.use_mmio_flip = 0,
 	.mmio_debug = 0,
 	.verbose_state_checks = 1,
+	.nuclear_pageflip = 0,
 	.edp_vswing = 0,
 	.enable_guc_submission = false,
 	.guc_log_level = -1,
@@ -177,6 +178,10 @@ module_param_named(verbose_state_checks, i915.verbose_state_checks, bool, 0600);
 MODULE_PARM_DESC(verbose_state_checks,
 	"Enable verbose logs (ie. WARN_ON()) in case of unexpected hw state conditions.");
 
+module_param_named_unsafe(nuclear_pageflip, i915.nuclear_pageflip, bool, 0600);
+MODULE_PARM_DESC(nuclear_pageflip,
+		 "Force atomic modeset functionality; asynchronous mode is not yet supported. (default: false).");
+
 /* WA to get away with the default setting in VBT for early platforms.Will be removed */
 module_param_named_unsafe(edp_vswing, i915.edp_vswing, int, 0400);
 MODULE_PARM_DESC(edp_vswing,
-- 
cgit v0.10.2


From 01302d4d17ed524e7d855f5fbbac5822f14f9c5e Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Fri, 28 Aug 2015 18:05:26 +0200
Subject: drm/i915: Update DRIVER_DATE to 20150828

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 7cd1555..e1bbe46 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -57,7 +57,7 @@
 
 #define DRIVER_NAME		"i915"
 #define DRIVER_DESC		"Intel Graphics"
-#define DRIVER_DATE		"20150731"
+#define DRIVER_DATE		"20150828"
 
 #undef WARN_ON
 /* Many gcc seem to no see through this and fall over :( */
-- 
cgit v0.10.2


From 7cc53cf01e61bd1a774d2ba492bbe3e93e58ca79 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Wed, 26 Aug 2015 14:33:31 +0300
Subject: drm/dp: add drm_dp_tps3_supported helper

Cc: Thierry Reding <thierry.reding@gmail.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/include/drm/drm_dp_helper.h b/include/drm/drm_dp_helper.h
index 94898f6..6aa59b9 100644
--- a/include/drm/drm_dp_helper.h
+++ b/include/drm/drm_dp_helper.h
@@ -633,6 +633,13 @@ drm_dp_enhanced_frame_cap(const u8 dpcd[DP_RECEIVER_CAP_SIZE])
 		(dpcd[DP_MAX_LANE_COUNT] & DP_ENHANCED_FRAME_CAP);
 }
 
+static inline bool
+drm_dp_tps3_supported(const u8 dpcd[DP_RECEIVER_CAP_SIZE])
+{
+	return dpcd[DP_DPCD_REV] >= 0x12 &&
+		dpcd[DP_MAX_LANE_COUNT] & DP_TPS3_SUPPORTED;
+}
+
 /*
  * DisplayPort AUX channel
  */
-- 
cgit v0.10.2


From caa860d919865ade32e029631bd0d531eead9212 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 27 Aug 2015 13:25:37 +0300
Subject: drm/i915/dp: use the drm dp helper for determining sink tps3 support
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

No functional changes.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
[danvet: s/intel_dp_tps/drm_dp_tps/.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index ea11469..b794439 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -4023,8 +4023,7 @@ intel_dp_get_dpcd(struct intel_dp *intel_dp)
 	}
 
 	/* Training Pattern 3 support, both source and sink */
-	if (intel_dp->dpcd[DP_DPCD_REV] >= 0x12 &&
-	    intel_dp->dpcd[DP_MAX_LANE_COUNT] & DP_TPS3_SUPPORTED &&
+	if (drm_dp_tps3_supported(intel_dp->dpcd) &&
 	    (IS_HASWELL(dev_priv) || INTEL_INFO(dev_priv)->gen >= 8)) {
 		intel_dp->use_tps3 = true;
 		DRM_DEBUG_KMS("Displayport TPS3 supported\n");
-- 
cgit v0.10.2


From 2bbe6bbb0dc94fd4ce287bdac9e1bd184e23057b Mon Sep 17 00:00:00 2001
From: Francisco Jerez <currojerez@riseup.net>
Date: Mon, 15 Jun 2015 14:03:29 +0300
Subject: drm/i915: Bump command parser version number.

This was forgotten in

commit d351f6d94893f3ba98b1b20c5ef44c35fc1da124
Author: Francisco Jerez <currojerez@riseup.net>
Date:   Fri May 29 16:44:15 2015 +0300

    drm/i915: Add SCRATCH1 and ROW_CHICKEN3 to the register whitelist.

Signed-off-by: Francisco Jerez <currojerez@riseup.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_cmd_parser.c b/drivers/gpu/drm/i915/i915_cmd_parser.c
index 3977bf3..ad7d7ab 100644
--- a/drivers/gpu/drm/i915/i915_cmd_parser.c
+++ b/drivers/gpu/drm/i915/i915_cmd_parser.c
@@ -1213,6 +1213,7 @@ int i915_cmd_parser_get_version(void)
 	 * 2. Allow access to the MI_PREDICATE_SRC0 and
 	 *    MI_PREDICATE_SRC1 registers.
 	 * 3. Allow access to the GPGPU_THREADS_DISPATCHED register.
+	 * 4. L3 atomic chicken bits of HSW_SCRATCH1 and HSW_ROW_CHICKEN3.
 	 */
-	return 3;
+	return 4;
 }
-- 
cgit v0.10.2


From a8f327fb8464875e5b764008a1c50dbc0b384c17 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 9 Jul 2015 20:14:11 +0300
Subject: drm/i915: Clean up CHV lane soft reset programming
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Currently we release the lane soft reset before lane stagger settings
have been programmed. I believe that means we don't actually do lane
staggering. So move the soft reset deassert to happen after lane
staggering has been programmed.

The one confusing thing in this is that when we remove the power down
override from the lanes, they power up with defaul register values,
which do not have the soft reset overrides enabled. And according to
some docs by default the data lane resets are tied to cmnreset. So that
would mean that lanes would come out of reset without staggering as
soon as the power down overrides are removed. But since we can't access
either the lane stagger register nor the soft reset override registers
until the lanes are powered on, we can't really do anything about it.
So let's just set the soft reset overrides as soon as the lane is
powered on and hope for the best.

v2: Fix typos in commit message (Daniel)

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Deepak S <deepak.s@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index b794439..3a9c43d 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -2426,42 +2426,62 @@ static void vlv_post_disable_dp(struct intel_encoder *encoder)
 	intel_dp_link_down(intel_dp);
 }
 
-static void chv_post_disable_dp(struct intel_encoder *encoder)
+static void chv_data_lane_soft_reset(struct intel_encoder *encoder,
+				     bool reset)
 {
-	struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base);
-	struct intel_digital_port *dport = dp_to_dig_port(intel_dp);
-	struct drm_device *dev = encoder->base.dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_crtc *intel_crtc =
-		to_intel_crtc(encoder->base.crtc);
-	enum dpio_channel ch = vlv_dport_to_channel(dport);
-	enum pipe pipe = intel_crtc->pipe;
-	u32 val;
+	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	enum dpio_channel ch = vlv_dport_to_channel(enc_to_dig_port(&encoder->base));
+	struct intel_crtc *crtc = to_intel_crtc(encoder->base.crtc);
+	enum pipe pipe = crtc->pipe;
+	uint32_t val;
 
-	intel_dp_link_down(intel_dp);
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW0(ch));
+	if (reset)
+		val &= ~(DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
+	else
+		val |= DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET;
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW0(ch), val);
 
-	mutex_lock(&dev_priv->sb_lock);
+	if (crtc->config->lane_count > 2) {
+		val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW0(ch));
+		if (reset)
+			val &= ~(DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
+		else
+			val |= DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET;
+		vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW0(ch), val);
+	}
 
-	/* Propagate soft reset to data lane reset */
 	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW1(ch));
 	val |= CHV_PCS_REQ_SOFTRESET_EN;
+	if (reset)
+		val &= ~DPIO_PCS_CLK_SOFT_RESET;
+	else
+		val |= DPIO_PCS_CLK_SOFT_RESET;
 	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW1(ch), val);
 
-	if (intel_crtc->config->lane_count > 2) {
+	if (crtc->config->lane_count > 2) {
 		val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW1(ch));
 		val |= CHV_PCS_REQ_SOFTRESET_EN;
+		if (reset)
+			val &= ~DPIO_PCS_CLK_SOFT_RESET;
+		else
+			val |= DPIO_PCS_CLK_SOFT_RESET;
 		vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW1(ch), val);
 	}
+}
 
-	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW0(ch));
-	val &= ~(DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
-	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW0(ch), val);
+static void chv_post_disable_dp(struct intel_encoder *encoder)
+{
+	struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base);
+	struct drm_device *dev = encoder->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	if (intel_crtc->config->lane_count > 2) {
-		val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW0(ch));
-		val &= ~(DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
-		vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW0(ch), val);
-	}
+	intel_dp_link_down(intel_dp);
+
+	mutex_lock(&dev_priv->sb_lock);
+
+	/* Assert data lane reset */
+	chv_data_lane_soft_reset(encoder, true);
 
 	mutex_unlock(&dev_priv->sb_lock);
 }
@@ -2839,27 +2859,6 @@ static void chv_pre_enable_dp(struct intel_encoder *encoder)
 		vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW11(ch), val);
 	}
 
-	/* Deassert soft data lane reset*/
-	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW1(ch));
-	val |= CHV_PCS_REQ_SOFTRESET_EN;
-	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW1(ch), val);
-
-	if (intel_crtc->config->lane_count > 2) {
-		val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW1(ch));
-		val |= CHV_PCS_REQ_SOFTRESET_EN;
-		vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW1(ch), val);
-	}
-
-	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW0(ch));
-	val |= (DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
-	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW0(ch), val);
-
-	if (intel_crtc->config->lane_count > 2) {
-		val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW0(ch));
-		val |= (DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
-		vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW0(ch), val);
-	}
-
 	/* Program Tx lane latency optimal setting*/
 	for (i = 0; i < intel_crtc->config->lane_count; i++) {
 		/* Set the upar bit */
@@ -2909,6 +2908,9 @@ static void chv_pre_enable_dp(struct intel_encoder *encoder)
 			       DPIO_TX2_STAGGER_MULT(5));
 	}
 
+	/* Deassert data lane reset */
+	chv_data_lane_soft_reset(encoder, false);
+
 	mutex_unlock(&dev_priv->sb_lock);
 
 	intel_enable_dp(encoder);
@@ -2947,6 +2949,9 @@ static void chv_dp_pre_pll_enable(struct intel_encoder *encoder)
 
 	mutex_lock(&dev_priv->sb_lock);
 
+	/* Assert data lane reset */
+	chv_data_lane_soft_reset(encoder, true);
+
 	/* program left/right clock distribution */
 	if (pipe != PIPE_B) {
 		val = vlv_dpio_read(dev_priv, pipe, _CHV_CMN_DW5_CH0);
diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index 7c56053..4da737c 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -1615,6 +1615,50 @@ static void vlv_hdmi_pre_pll_enable(struct intel_encoder *encoder)
 	mutex_unlock(&dev_priv->sb_lock);
 }
 
+static void chv_data_lane_soft_reset(struct intel_encoder *encoder,
+				     bool reset)
+{
+	struct drm_i915_private *dev_priv = to_i915(encoder->base.dev);
+	enum dpio_channel ch = vlv_dport_to_channel(enc_to_dig_port(&encoder->base));
+	struct intel_crtc *crtc = to_intel_crtc(encoder->base.crtc);
+	enum pipe pipe = crtc->pipe;
+	uint32_t val;
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW0(ch));
+	if (reset)
+		val &= ~(DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
+	else
+		val |= DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET;
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW0(ch), val);
+
+	if (crtc->config->lane_count > 2) {
+		val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW0(ch));
+		if (reset)
+			val &= ~(DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
+		else
+			val |= DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET;
+		vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW0(ch), val);
+	}
+
+	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW1(ch));
+	val |= CHV_PCS_REQ_SOFTRESET_EN;
+	if (reset)
+		val &= ~DPIO_PCS_CLK_SOFT_RESET;
+	else
+		val |= DPIO_PCS_CLK_SOFT_RESET;
+	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW1(ch), val);
+
+	if (crtc->config->lane_count > 2) {
+		val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW1(ch));
+		val |= CHV_PCS_REQ_SOFTRESET_EN;
+		if (reset)
+			val &= ~DPIO_PCS_CLK_SOFT_RESET;
+		else
+			val |= DPIO_PCS_CLK_SOFT_RESET;
+		vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW1(ch), val);
+	}
+}
+
 static void chv_hdmi_pre_pll_enable(struct intel_encoder *encoder)
 {
 	struct intel_digital_port *dport = enc_to_dig_port(&encoder->base);
@@ -1640,6 +1684,9 @@ static void chv_hdmi_pre_pll_enable(struct intel_encoder *encoder)
 
 	mutex_lock(&dev_priv->sb_lock);
 
+	/* Assert data lane reset */
+	chv_data_lane_soft_reset(encoder, true);
+
 	/* program left/right clock distribution */
 	if (pipe != PIPE_B) {
 		val = vlv_dpio_read(dev_priv, pipe, _CHV_CMN_DW5_CH0);
@@ -1742,33 +1789,13 @@ static void vlv_hdmi_post_disable(struct intel_encoder *encoder)
 
 static void chv_hdmi_post_disable(struct intel_encoder *encoder)
 {
-	struct intel_digital_port *dport = enc_to_dig_port(&encoder->base);
 	struct drm_device *dev = encoder->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_crtc *intel_crtc =
-		to_intel_crtc(encoder->base.crtc);
-	enum dpio_channel ch = vlv_dport_to_channel(dport);
-	enum pipe pipe = intel_crtc->pipe;
-	u32 val;
 
 	mutex_lock(&dev_priv->sb_lock);
 
-	/* Propagate soft reset to data lane reset */
-	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW1(ch));
-	val |= CHV_PCS_REQ_SOFTRESET_EN;
-	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW1(ch), val);
-
-	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW1(ch));
-	val |= CHV_PCS_REQ_SOFTRESET_EN;
-	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW1(ch), val);
-
-	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW0(ch));
-	val &= ~(DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
-	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW0(ch), val);
-
-	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW0(ch));
-	val &= ~(DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
-	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW0(ch), val);
+	/* Assert data lane reset */
+	chv_data_lane_soft_reset(encoder, true);
 
 	mutex_unlock(&dev_priv->sb_lock);
 }
@@ -1799,23 +1826,6 @@ static void chv_hdmi_pre_enable(struct intel_encoder *encoder)
 	val &= ~DPIO_LANEDESKEW_STRAP_OVRD;
 	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW11(ch), val);
 
-	/* Deassert soft data lane reset*/
-	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW1(ch));
-	val |= CHV_PCS_REQ_SOFTRESET_EN;
-	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW1(ch), val);
-
-	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW1(ch));
-	val |= CHV_PCS_REQ_SOFTRESET_EN;
-	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW1(ch), val);
-
-	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW0(ch));
-	val |= (DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
-	vlv_dpio_write(dev_priv, pipe, VLV_PCS01_DW0(ch), val);
-
-	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS23_DW0(ch));
-	val |= (DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET);
-	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW0(ch), val);
-
 	/* Program Tx latency optimal setting */
 	for (i = 0; i < 4; i++) {
 		/* Set the upar bit */
@@ -1858,6 +1868,9 @@ static void chv_hdmi_pre_enable(struct intel_encoder *encoder)
 		       DPIO_TX1_STAGGER_MULT(7) |
 		       DPIO_TX2_STAGGER_MULT(5));
 
+	/* Deassert data lane reset */
+	chv_data_lane_soft_reset(encoder, false);
+
 	/* Clear calc init */
 	val = vlv_dpio_read(dev_priv, pipe, VLV_PCS01_DW10(ch));
 	val &= ~(DPIO_PCS_SWING_CALC_TX0_TX2 | DPIO_PCS_SWING_CALC_TX1_TX3);
-- 
cgit v0.10.2


From 6669e39f95b5530ca8cb9137703ceb5e83e5d648 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 8 Jul 2015 23:46:00 +0300
Subject: drm/i915: Add some CHV DPIO lane power state asserts
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add some checks that the state of the DPIO lanes is more or less what we
expect based on the overrides.

The hardware only provides two bits per channel indicating whether all
or some of the lanes are powered down, so we can't do an exact check.

Additionally, CL2 powering down before we can check it adds another
twist. To work around this we simply check for the 0 value of the
CL2 register (which is what we get when it's powered down) and
adjust our expectations.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Deepak S <deepak.s@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 439bb27..8a18023 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -1099,6 +1099,12 @@ enum skl_disp_power_wells {
 #define  DPIO_CHV_INT_LOCK_THRESHOLD_SEL_COARSE	1 /* 1: coarse & 0 : fine  */
 #define CHV_PLL_DW9(ch) _PIPE(ch, _CHV_PLL_DW9_CH0, _CHV_PLL_DW9_CH1)
 
+#define _CHV_CMN_DW0_CH0               0x8100
+#define   DPIO_ALLDL_POWERDOWN_SHIFT_CH0	19
+#define   DPIO_ANYDL_POWERDOWN_SHIFT_CH0	18
+#define   DPIO_ALLDL_POWERDOWN			(1 << 1)
+#define   DPIO_ANYDL_POWERDOWN			(1 << 0)
+
 #define _CHV_CMN_DW5_CH0               0x8114
 #define   CHV_BUFRIGHTENA1_DISABLE	(0 << 20)
 #define   CHV_BUFRIGHTENA1_NORMAL	(1 << 20)
@@ -1135,6 +1141,8 @@ enum skl_disp_power_wells {
 
 #define _CHV_CMN_DW19_CH0		0x814c
 #define _CHV_CMN_DW6_CH1		0x8098
+#define   DPIO_ALLDL_POWERDOWN_SHIFT_CH1	30 /* CL2 DW6 only */
+#define   DPIO_ANYDL_POWERDOWN_SHIFT_CH1	29 /* CL2 DW6 only */
 #define   DPIO_DYNPWRDOWNEN_CH1		(1 << 28) /* CL2 DW6 only */
 #define   CHV_CMN_USEDCLKCHANNEL	(1 << 13)
 
diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c
index a0df156..7991eff 100644
--- a/drivers/gpu/drm/i915/intel_runtime_pm.c
+++ b/drivers/gpu/drm/i915/intel_runtime_pm.c
@@ -1042,6 +1042,58 @@ static void chv_dpio_cmn_power_well_disable(struct drm_i915_private *dev_priv,
 		      phy, dev_priv->chv_phy_control);
 }
 
+static void assert_chv_phy_powergate(struct drm_i915_private *dev_priv, enum dpio_phy phy,
+				     enum dpio_channel ch, bool override, unsigned int mask)
+{
+	enum pipe pipe = phy == DPIO_PHY0 ? PIPE_A : PIPE_C;
+	u32 reg, val, expected, actual;
+
+	if (ch == DPIO_CH0)
+		reg = _CHV_CMN_DW0_CH0;
+	else
+		reg = _CHV_CMN_DW6_CH1;
+
+	mutex_lock(&dev_priv->sb_lock);
+	val = vlv_dpio_read(dev_priv, pipe, reg);
+	mutex_unlock(&dev_priv->sb_lock);
+
+	/*
+	 * This assumes !override is only used when the port is disabled.
+	 * All lanes should power down even without the override when
+	 * the port is disabled.
+	 */
+	if (!override || mask == 0xf) {
+		expected = DPIO_ALLDL_POWERDOWN | DPIO_ANYDL_POWERDOWN;
+		/*
+		 * If CH1 common lane is not active anymore
+		 * (eg. for pipe B DPLL) the entire channel will
+		 * shut down, which causes the common lane registers
+		 * to read as 0. That means we can't actually check
+		 * the lane power down status bits, but as the entire
+		 * register reads as 0 it's a good indication that the
+		 * channel is indeed entirely powered down.
+		 */
+		if (ch == DPIO_CH1 && val == 0)
+			expected = 0;
+	} else if (mask != 0x0) {
+		expected = DPIO_ANYDL_POWERDOWN;
+	} else {
+		expected = 0;
+	}
+
+	if (ch == DPIO_CH0)
+		actual = val >> DPIO_ANYDL_POWERDOWN_SHIFT_CH0;
+	else
+		actual = val >> DPIO_ANYDL_POWERDOWN_SHIFT_CH1;
+	actual &= DPIO_ALLDL_POWERDOWN | DPIO_ANYDL_POWERDOWN;
+
+	WARN(actual != expected,
+	     "Unexpected DPIO lane power down: all %d, any %d. Expected: all %d, any %d. (0x%x = 0x%08x)\n",
+	     !!(actual & DPIO_ALLDL_POWERDOWN), !!(actual & DPIO_ANYDL_POWERDOWN),
+	     !!(expected & DPIO_ALLDL_POWERDOWN), !!(expected & DPIO_ANYDL_POWERDOWN),
+	     reg, val);
+}
+
 bool chv_phy_powergate_ch(struct drm_i915_private *dev_priv, enum dpio_phy phy,
 			  enum dpio_channel ch, bool override)
 {
@@ -1094,6 +1146,8 @@ void chv_phy_powergate_lanes(struct intel_encoder *encoder,
 	DRM_DEBUG_KMS("Power gating DPIO PHY%d CH%d lanes 0x%x (PHY_CONTROL=0x%08x)\n",
 		      phy, ch, mask, dev_priv->chv_phy_control);
 
+	assert_chv_phy_powergate(dev_priv, phy, ch, override, mask);
+
 	mutex_unlock(&power_domains->lock);
 }
 
-- 
cgit v0.10.2


From 30142273a3e83936fd7b45aa5339311a9295ca51 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 8 Jul 2015 23:46:01 +0300
Subject: drm/i915: Add CHV PHY LDO power sanity checks
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

At various points when changing the DPIO lane/phy power states,
construct an expected value of the DISPLAY_PHY_STATUS register
and compare it with the real thing.

To construct the expected value we look at our shadow PHY_CONTROL
register value (which should match what we've just written to the
hardware), and we also need to look at the actual state of the cmn
power wells as a disabled power well causes the relevant LDO status
to be reported as 'on' in DISPLAY_PHY_STATUS.

When initially powering up the PHY it performs various internal
calibrations for which it fully powers up. That means that if we check
for the expetected power state immediately upon releasing cmnreset we
would get the occasional false positive. But we can of course
poll until the expected value appears. It shouldn't be too long so
this shouldn't make modesets substantially longer.

One extra complication is introduced when we cross the streams, ie.
drive port B with pipe B. In this case we trick CL2 (where the DPLL lives)
into life by temporaily powering up the lanes in the second channel,
and once the pipe is up and runnign we release the lane power override.
At that point the power state of CL2 has somehow gotten entangled with
the power state of the first channel. That means that constructing the
expected DISPLAY_PHY_STATUS value is a bit tricky since based on the
lane power states in the second channel, CL2 should also be powered
down. But we can use the DPLL enable bit to determine when CL2 should
be alive even if the lanes are powered down. However the power state
of CL2 isn't actually tied in with the DPLL state, but to the state
of the lanes in first channel, so we have to avoid checking the
expected state between shutting down the DPLL and powering down
the lanes in the first channel. So no calling assert_chv_phy_status()
before the DISPLAY_PHY_CONTROL write in chv_phy_powergate_lanes(),
but after the write is a safe time to check.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Deepak S <deepak.s@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 8a18023..c82db2a 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -2223,6 +2223,8 @@ enum skl_disp_power_wells {
 #define   PHY_COM_LANE_RESET_DEASSERT(phy)	(1 << (phy))
 #define DISPLAY_PHY_STATUS (VLV_DISPLAY_BASE + 0x60104)
 #define   PHY_POWERGOOD(phy)	(((phy) == DPIO_PHY0) ? (1<<31) : (1<<30))
+#define   PHY_STATUS_CMN_LDO(phy, ch)                   (1 << (6-(6*(phy)+3*(ch))))
+#define   PHY_STATUS_SPLINE_LDO(phy, ch, spline)        (1 << (8-(6*(phy)+3*(ch)+(spline))))
 
 /*
  * The i830 generation, in LVDS mode, defines P1 as the bit number set within
diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c
index 7991eff..b1bd25e 100644
--- a/drivers/gpu/drm/i915/intel_runtime_pm.c
+++ b/drivers/gpu/drm/i915/intel_runtime_pm.c
@@ -958,6 +958,107 @@ static void vlv_dpio_cmn_power_well_disable(struct drm_i915_private *dev_priv,
 	vlv_set_power_well(dev_priv, power_well, false);
 }
 
+#define POWER_DOMAIN_MASK (BIT(POWER_DOMAIN_NUM) - 1)
+
+static struct i915_power_well *lookup_power_well(struct drm_i915_private *dev_priv,
+						 int power_well_id)
+{
+	struct i915_power_domains *power_domains = &dev_priv->power_domains;
+	struct i915_power_well *power_well;
+	int i;
+
+	for_each_power_well(i, power_well, POWER_DOMAIN_MASK, power_domains) {
+		if (power_well->data == power_well_id)
+			return power_well;
+	}
+
+	return NULL;
+}
+
+#define BITS_SET(val, bits) (((val) & (bits)) == (bits))
+
+static void assert_chv_phy_status(struct drm_i915_private *dev_priv)
+{
+	struct i915_power_well *cmn_bc =
+		lookup_power_well(dev_priv, PUNIT_POWER_WELL_DPIO_CMN_BC);
+	struct i915_power_well *cmn_d =
+		lookup_power_well(dev_priv, PUNIT_POWER_WELL_DPIO_CMN_D);
+	u32 phy_control = dev_priv->chv_phy_control;
+	u32 phy_status = 0;
+	u32 tmp;
+
+	if (cmn_bc->ops->is_enabled(dev_priv, cmn_bc)) {
+		phy_status |= PHY_POWERGOOD(DPIO_PHY0);
+
+		/* this assumes override is only used to enable lanes */
+		if ((phy_control & PHY_CH_POWER_DOWN_OVRD_EN(DPIO_PHY0, DPIO_CH0)) == 0)
+			phy_control |= PHY_CH_POWER_DOWN_OVRD(0xf, DPIO_PHY0, DPIO_CH0);
+
+		if ((phy_control & PHY_CH_POWER_DOWN_OVRD_EN(DPIO_PHY0, DPIO_CH1)) == 0)
+			phy_control |= PHY_CH_POWER_DOWN_OVRD(0xf, DPIO_PHY0, DPIO_CH1);
+
+		/* CL1 is on whenever anything is on in either channel */
+		if (BITS_SET(phy_control,
+			     PHY_CH_POWER_DOWN_OVRD(0xf, DPIO_PHY0, DPIO_CH0) |
+			     PHY_CH_POWER_DOWN_OVRD(0xf, DPIO_PHY0, DPIO_CH1)))
+			phy_status |= PHY_STATUS_CMN_LDO(DPIO_PHY0, DPIO_CH0);
+
+		/*
+		 * The DPLLB check accounts for the pipe B + port A usage
+		 * with CL2 powered up but all the lanes in the second channel
+		 * powered down.
+		 */
+		if (BITS_SET(phy_control,
+			     PHY_CH_POWER_DOWN_OVRD(0xf, DPIO_PHY0, DPIO_CH1)) &&
+		    (I915_READ(DPLL(PIPE_B)) & DPLL_VCO_ENABLE) == 0)
+			phy_status |= PHY_STATUS_CMN_LDO(DPIO_PHY0, DPIO_CH1);
+
+		if (BITS_SET(phy_control,
+			     PHY_CH_POWER_DOWN_OVRD(0x3, DPIO_PHY0, DPIO_CH0)))
+			phy_status |= PHY_STATUS_SPLINE_LDO(DPIO_PHY0, DPIO_CH0, 0);
+		if (BITS_SET(phy_control,
+			     PHY_CH_POWER_DOWN_OVRD(0xc, DPIO_PHY0, DPIO_CH0)))
+			phy_status |= PHY_STATUS_SPLINE_LDO(DPIO_PHY0, DPIO_CH0, 1);
+
+		if (BITS_SET(phy_control,
+			     PHY_CH_POWER_DOWN_OVRD(0x3, DPIO_PHY0, DPIO_CH1)))
+			phy_status |= PHY_STATUS_SPLINE_LDO(DPIO_PHY0, DPIO_CH1, 0);
+		if (BITS_SET(phy_control,
+			     PHY_CH_POWER_DOWN_OVRD(0xc, DPIO_PHY0, DPIO_CH1)))
+			phy_status |= PHY_STATUS_SPLINE_LDO(DPIO_PHY0, DPIO_CH1, 1);
+	}
+
+	if (cmn_d->ops->is_enabled(dev_priv, cmn_d)) {
+		phy_status |= PHY_POWERGOOD(DPIO_PHY1);
+
+		/* this assumes override is only used to enable lanes */
+		if ((phy_control & PHY_CH_POWER_DOWN_OVRD_EN(DPIO_PHY1, DPIO_CH0)) == 0)
+			phy_control |= PHY_CH_POWER_DOWN_OVRD(0xf, DPIO_PHY1, DPIO_CH0);
+
+		if (BITS_SET(phy_control,
+			     PHY_CH_POWER_DOWN_OVRD(0xf, DPIO_PHY1, DPIO_CH0)))
+			phy_status |= PHY_STATUS_CMN_LDO(DPIO_PHY1, DPIO_CH0);
+
+		if (BITS_SET(phy_control,
+			     PHY_CH_POWER_DOWN_OVRD(0x3, DPIO_PHY1, DPIO_CH0)))
+			phy_status |= PHY_STATUS_SPLINE_LDO(DPIO_PHY1, DPIO_CH0, 0);
+		if (BITS_SET(phy_control,
+			     PHY_CH_POWER_DOWN_OVRD(0xc, DPIO_PHY1, DPIO_CH0)))
+			phy_status |= PHY_STATUS_SPLINE_LDO(DPIO_PHY1, DPIO_CH0, 1);
+	}
+
+	/*
+	 * The PHY may be busy with some initial calibration and whatnot,
+	 * so the power state can take a while to actually change.
+	 */
+	if (wait_for((tmp = I915_READ(DISPLAY_PHY_STATUS)) == phy_status, 10))
+		WARN(phy_status != tmp,
+		     "Unexpected PHY_STATUS 0x%08x, expected 0x%08x (PHY_CONTROL=0x%08x)\n",
+		     tmp, phy_status, dev_priv->chv_phy_control);
+}
+
+#undef BITS_SET
+
 static void chv_dpio_cmn_power_well_enable(struct drm_i915_private *dev_priv,
 					   struct i915_power_well *power_well)
 {
@@ -1014,6 +1115,8 @@ static void chv_dpio_cmn_power_well_enable(struct drm_i915_private *dev_priv,
 
 	DRM_DEBUG_KMS("Enabled DPIO PHY%d (PHY_CONTROL=0x%08x)\n",
 		      phy, dev_priv->chv_phy_control);
+
+	assert_chv_phy_status(dev_priv);
 }
 
 static void chv_dpio_cmn_power_well_disable(struct drm_i915_private *dev_priv,
@@ -1040,6 +1143,8 @@ static void chv_dpio_cmn_power_well_disable(struct drm_i915_private *dev_priv,
 
 	DRM_DEBUG_KMS("Disabled DPIO PHY%d (PHY_CONTROL=0x%08x)\n",
 		      phy, dev_priv->chv_phy_control);
+
+	assert_chv_phy_status(dev_priv);
 }
 
 static void assert_chv_phy_powergate(struct drm_i915_private *dev_priv, enum dpio_phy phy,
@@ -1117,6 +1222,8 @@ bool chv_phy_powergate_ch(struct drm_i915_private *dev_priv, enum dpio_phy phy,
 	DRM_DEBUG_KMS("Power gating DPIO PHY%d CH%d (DPIO_PHY_CONTROL=0x%08x)\n",
 		      phy, ch, dev_priv->chv_phy_control);
 
+	assert_chv_phy_status(dev_priv);
+
 out:
 	mutex_unlock(&power_domains->lock);
 
@@ -1146,6 +1253,8 @@ void chv_phy_powergate_lanes(struct intel_encoder *encoder,
 	DRM_DEBUG_KMS("Power gating DPIO PHY%d CH%d lanes 0x%x (PHY_CONTROL=0x%08x)\n",
 		      phy, ch, mask, dev_priv->chv_phy_control);
 
+	assert_chv_phy_status(dev_priv);
+
 	assert_chv_phy_powergate(dev_priv, phy, ch, override, mask);
 
 	mutex_unlock(&power_domains->lock);
@@ -1312,8 +1421,6 @@ void intel_display_power_put(struct drm_i915_private *dev_priv,
 	intel_runtime_pm_put(dev_priv);
 }
 
-#define POWER_DOMAIN_MASK (BIT(POWER_DOMAIN_NUM) - 1)
-
 #define HSW_ALWAYS_ON_POWER_DOMAINS (			\
 	BIT(POWER_DOMAIN_PIPE_A) |			\
 	BIT(POWER_DOMAIN_TRANSCODER_EDP) |		\
@@ -1575,21 +1682,6 @@ static struct i915_power_well chv_power_wells[] = {
 	},
 };
 
-static struct i915_power_well *lookup_power_well(struct drm_i915_private *dev_priv,
-						 int power_well_id)
-{
-	struct i915_power_domains *power_domains = &dev_priv->power_domains;
-	struct i915_power_well *power_well;
-	int i;
-
-	for_each_power_well(i, power_well, POWER_DOMAIN_MASK, power_domains) {
-		if (power_well->data == power_well_id)
-			return power_well;
-	}
-
-	return NULL;
-}
-
 bool intel_display_power_well_is_enabled(struct drm_i915_private *dev_priv,
 				    int power_well_id)
 {
-- 
cgit v0.10.2


From 0f64614dde172e4f4b2734a55091a4d307863cbb Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 26 Aug 2015 19:39:18 +0300
Subject: drm/i915: Fix clock readout when pipes are enabled w/o ports
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The BIOS sometimes likes to enable pipes w/o any ports, at least on
older machines. Currently we fail to assign anything sensible to
crtc->hwmode.crtc_clock which leads to complaints from the vblank code.
Deal with active pipes w/o ports and assign something sensible to
crtc_clock in i9xx_get_pipe_config(). The encoder .get_config() will
override this if the port is enabled.

Gets rid of rest of these on my gen4:
[drm:drm_calc_timestamping_constants [drm]] *ERROR* crtc 24: Can't calculate constants, dotclock = 0!
[drm:i915_get_vblank_timestamp] crtc 1 is disabled

v2: Fill out crtc_clock already in i9xx_get_pipe_config() (Maarten)

Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index f32206d..ba0c85e 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -8059,6 +8059,14 @@ static bool i9xx_get_pipe_config(struct intel_crtc *crtc,
 	else
 		i9xx_crtc_clock_get(crtc, pipe_config);
 
+	/*
+	 * Normally the dotclock is filled in by the encoder .get_config()
+	 * but in case the pipe is enabled w/o any ports we need a sane
+	 * default.
+	 */
+	pipe_config->base.adjusted_mode.crtc_clock =
+		pipe_config->port_clock / pipe_config->pixel_multiplier;
+
 	return true;
 }
 
-- 
cgit v0.10.2


From 02e93c3537d5b496bbf4d6a97c0d02becc501b5c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 26 Aug 2015 19:39:19 +0300
Subject: drm/i915: Factor out intel_crtc_has_encoders()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Make the code mode readable by pulling the "does this crtc have any
encoders?" deduction into a separate function.

Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index ba0c85e..439e66e 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -14807,13 +14807,22 @@ intel_check_plane_mapping(struct intel_crtc *crtc)
 	return true;
 }
 
+static bool intel_crtc_has_encoders(struct intel_crtc *crtc)
+{
+	struct drm_device *dev = crtc->base.dev;
+	struct intel_encoder *encoder;
+
+	for_each_encoder_on_crtc(dev, &crtc->base, encoder)
+		return true;
+
+	return false;
+}
+
 static void intel_sanitize_crtc(struct intel_crtc *crtc)
 {
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_encoder *encoder;
 	u32 reg;
-	bool enable;
 
 	/* Clear any frame start delays used for debugging left by the BIOS */
 	reg = PIPECONF(crtc->config->cpu_transcoder);
@@ -14857,16 +14866,11 @@ static void intel_sanitize_crtc(struct intel_crtc *crtc)
 
 	/* Adjust the state of the output pipe according to whether we
 	 * have active connectors/encoders. */
-	enable = false;
-	for_each_encoder_on_crtc(dev, &crtc->base, encoder) {
-		enable = true;
-		break;
-	}
-
-	if (!enable)
+	if (!intel_crtc_has_encoders(crtc))
 		intel_crtc_disable_noatomic(&crtc->base);
 
 	if (crtc->active != crtc->base.state->active) {
+		struct intel_encoder *encoder;
 
 		/* This can happen either due to bugs in the get_hw_state
 		 * functions or because of calls to intel_crtc_disable_noatomic,
-- 
cgit v0.10.2


From 1751fcf9f92ebfd230260230956535b19b4ec675 Mon Sep 17 00:00:00 2001
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Date: Thu, 27 Aug 2015 15:15:15 +0200
Subject: drm/i915: Fix module initialisation, v2.

The driver doesn't support UMS any more, so set DRIVER_MODESET by default,
remove the legacy s/r callbacks, and rename the s/r functions to make it more clear
they're only in use by switcheroo now.

Also remove an obsolete comment about atomic. Normal updates are supported only
async updates aren't yet.

v2: Don't unconditionally set DRIVER_ATOMIC, we're not yet there.

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index 097d4ba..f0eaa6f 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -364,12 +364,12 @@ static void i915_switcheroo_set_state(struct pci_dev *pdev, enum vga_switcheroo_
 		dev->switch_power_state = DRM_SWITCH_POWER_CHANGING;
 		/* i915 resume handler doesn't set to D0 */
 		pci_set_power_state(dev->pdev, PCI_D0);
-		i915_resume_legacy(dev);
+		i915_resume_switcheroo(dev);
 		dev->switch_power_state = DRM_SWITCH_POWER_ON;
 	} else {
 		pr_err("switched off\n");
 		dev->switch_power_state = DRM_SWITCH_POWER_CHANGING;
-		i915_suspend_legacy(dev, pmm);
+		i915_suspend_switcheroo(dev, pmm);
 		dev->switch_power_state = DRM_SWITCH_POWER_OFF;
 	}
 }
diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index ce3bd0c..4646fe1 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -679,7 +679,7 @@ static int i915_drm_suspend_late(struct drm_device *drm_dev, bool hibernation)
 	return 0;
 }
 
-int i915_suspend_legacy(struct drm_device *dev, pm_message_t state)
+int i915_suspend_switcheroo(struct drm_device *dev, pm_message_t state)
 {
 	int error;
 
@@ -812,7 +812,7 @@ static int i915_drm_resume_early(struct drm_device *dev)
 	return ret;
 }
 
-int i915_resume_legacy(struct drm_device *dev)
+int i915_resume_switcheroo(struct drm_device *dev)
 {
 	int ret;
 
@@ -1649,7 +1649,7 @@ static struct drm_driver driver = {
 	 */
 	.driver_features =
 	    DRIVER_HAVE_IRQ | DRIVER_IRQ_SHARED | DRIVER_GEM | DRIVER_PRIME |
-	    DRIVER_RENDER,
+	    DRIVER_RENDER | DRIVER_MODESET,
 	.load = i915_driver_load,
 	.unload = i915_driver_unload,
 	.open = i915_driver_open,
@@ -1658,10 +1658,6 @@ static struct drm_driver driver = {
 	.postclose = i915_driver_postclose,
 	.set_busid = drm_pci_set_busid,
 
-	/* Used in place of i915_pm_ops for non-DRIVER_MODESET */
-	.suspend = i915_suspend_legacy,
-	.resume = i915_resume_legacy,
-
 #if defined(CONFIG_DEBUG_FS)
 	.debugfs_init = i915_debugfs_init,
 	.debugfs_cleanup = i915_debugfs_cleanup,
@@ -1704,7 +1700,6 @@ static int __init i915_init(void)
 	 * either the i915.modeset prarameter or by the
 	 * vga_text_mode_force boot option.
 	 */
-	driver.driver_features |= DRIVER_MODESET;
 
 	if (i915.modeset == 0)
 		driver.driver_features &= ~DRIVER_MODESET;
@@ -1715,17 +1710,11 @@ static int __init i915_init(void)
 #endif
 
 	if (!(driver.driver_features & DRIVER_MODESET)) {
-		driver.get_vblank_timestamp = NULL;
 		/* Silently fail loading to not upset userspace. */
 		DRM_DEBUG_DRIVER("KMS and UMS disabled.\n");
 		return 0;
 	}
 
-	/*
-	 * FIXME: Note that we're lying to the DRM core here so that we can get access
-	 * to the atomic ioctl and the atomic properties.  Only plane operations on
-	 * a single CRTC will actually work.
-	 */
 	if (i915.nuclear_pageflip)
 		driver.driver_features |= DRIVER_ATOMIC;
 
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index e1bbe46..d2c4449 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -2606,8 +2606,8 @@ struct drm_i915_cmd_table {
 extern const struct drm_ioctl_desc i915_ioctls[];
 extern int i915_max_ioctl;
 
-extern int i915_suspend_legacy(struct drm_device *dev, pm_message_t state);
-extern int i915_resume_legacy(struct drm_device *dev);
+extern int i915_suspend_switcheroo(struct drm_device *dev, pm_message_t state);
+extern int i915_resume_switcheroo(struct drm_device *dev);
 
 /* i915_params.c */
 struct i915_params {
-- 
cgit v0.10.2


From aa557ab01576531c39fa3e7ceaa33314f65164fd Mon Sep 17 00:00:00 2001
From: Alex Dai <yu.dai@intel.com>
Date: Tue, 18 Aug 2015 14:32:35 -0700
Subject: drm/i915/guc: Support GuC version 4.3

The firmware layout changes that now it only has css header +
uCode + RSA signature. Plus, other trivial changes to support
GuC V4.3.

Signed-off-by: Alex Dai <yu.dai@intel.com>
Reviewed-by: Dave Gordon <david.s.gordon@intel.com>
Reviewed-by: Dave Gordon <david.s.gordon@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_guc_fwif.h b/drivers/gpu/drm/i915/intel_guc_fwif.h
index 950c7e7..e1f47ba 100644
--- a/drivers/gpu/drm/i915/intel_guc_fwif.h
+++ b/drivers/gpu/drm/i915/intel_guc_fwif.h
@@ -41,7 +41,7 @@
 #define GUC_CTX_PRIORITY_NORMAL		3
 
 #define GUC_MAX_GPU_CONTEXTS		1024
-#define	GUC_INVALID_CTX_ID		(GUC_MAX_GPU_CONTEXTS + 1)
+#define	GUC_INVALID_CTX_ID		GUC_MAX_GPU_CONTEXTS
 
 /* Work queue item header definitions */
 #define WQ_STATUS_ACTIVE		1
@@ -75,6 +75,7 @@
 #define GUC_CTX_DESC_ATTR_RESET		(1 << 4)
 #define GUC_CTX_DESC_ATTR_WQLOCKED	(1 << 5)
 #define GUC_CTX_DESC_ATTR_PCH		(1 << 6)
+#define GUC_CTX_DESC_ATTR_TERMINATED	(1 << 7)
 
 /* The guc control data is 10 DWORDs */
 #define GUC_CTL_CTXINFO			0
@@ -107,6 +108,7 @@
 #define   GUC_CTL_DISABLE_SCHEDULER	(1 << 4)
 #define   GUC_CTL_PREEMPTION_LOG	(1 << 5)
 #define   GUC_CTL_ENABLE_SLPC		(1 << 7)
+#define   GUC_CTL_RESET_ON_PREMPT_FAILURE	(1 << 8)
 #define GUC_CTL_DEBUG			8
 #define   GUC_LOG_VERBOSITY_SHIFT	0
 #define   GUC_LOG_VERBOSITY_LOW		(0 << GUC_LOG_VERBOSITY_SHIFT)
@@ -116,8 +118,9 @@
 /* Verbosity range-check limits, without the shift */
 #define	  GUC_LOG_VERBOSITY_MIN		0
 #define	  GUC_LOG_VERBOSITY_MAX		3
+#define GUC_CTL_RSRVD			9
 
-#define GUC_CTL_MAX_DWORDS		(GUC_CTL_DEBUG + 1)
+#define GUC_CTL_MAX_DWORDS		(GUC_CTL_RSRVD + 1)
 
 struct guc_doorbell_info {
 	u32 db_status;
@@ -207,7 +210,9 @@ struct guc_context_desc {
 
 	u32 engine_presence;
 
-	u32 reserved0[1];
+	u8 engine_suspended;
+
+	u8 reserved0[3];
 	u64 reserved1[1];
 
 	u64 desc_private;
diff --git a/drivers/gpu/drm/i915/intel_guc_loader.c b/drivers/gpu/drm/i915/intel_guc_loader.c
index 13e75f6..5eafd31 100644
--- a/drivers/gpu/drm/i915/intel_guc_loader.c
+++ b/drivers/gpu/drm/i915/intel_guc_loader.c
@@ -59,7 +59,7 @@
  *
  */
 
-#define I915_SKL_GUC_UCODE "i915/skl_guc_ver3.bin"
+#define I915_SKL_GUC_UCODE "i915/skl_guc_ver4.bin"
 MODULE_FIRMWARE(I915_SKL_GUC_UCODE);
 
 /* User-friendly representation of an enum */
@@ -226,10 +226,6 @@ static inline bool guc_ucode_response(struct drm_i915_private *dev_priv,
  * +-------------------------------+  ----
  * |         RSA signature         |  256B
  * +-------------------------------+  ----
- * |         RSA public Key        |  256B
- * +-------------------------------+  ----
- * |       Public key modulus      |    4B
- * +-------------------------------+  ----
  *
  * Architecturally, the DMA engine is bidirectional, and can potentially even
  * transfer between GTT locations. This functionality is left out of the API
@@ -244,7 +240,6 @@ static inline bool guc_ucode_response(struct drm_i915_private *dev_priv,
 #define UOS_VER_MAJOR_OFFSET		0x46
 #define UOS_CSS_HEADER_SIZE		0x80
 #define UOS_RSA_SIG_SIZE		0x100
-#define UOS_CSS_SIGNING_SIZE		0x204
 
 static int guc_ucode_xfer_dma(struct drm_i915_private *dev_priv)
 {
@@ -256,7 +251,7 @@ static int guc_ucode_xfer_dma(struct drm_i915_private *dev_priv)
 	int i, ret = 0;
 
 	/* uCode size, also is where RSA signature starts */
-	offset = ucode_size = guc_fw->guc_fw_size - UOS_CSS_SIGNING_SIZE;
+	offset = ucode_size = guc_fw->guc_fw_size - UOS_RSA_SIG_SIZE;
 	I915_WRITE(DMA_COPY_SIZE, ucode_size);
 
 	/* Copy RSA signature from the fw image to HW for verification */
@@ -463,8 +458,8 @@ static void guc_fw_fetch(struct drm_device *dev, struct intel_guc_fw *guc_fw)
 	struct drm_i915_gem_object *obj;
 	const struct firmware *fw;
 	const u8 *css_header;
-	const size_t minsize = UOS_CSS_HEADER_SIZE + UOS_CSS_SIGNING_SIZE;
-	const size_t maxsize = GUC_WOPCM_SIZE_VALUE + UOS_CSS_SIGNING_SIZE
+	const size_t minsize = UOS_CSS_HEADER_SIZE + UOS_RSA_SIG_SIZE;
+	const size_t maxsize = GUC_WOPCM_SIZE_VALUE + UOS_RSA_SIG_SIZE
 			- 0x8000; /* 32k reserved (8K stack + 24k context) */
 	int err;
 
@@ -564,8 +559,8 @@ void intel_guc_ucode_init(struct drm_device *dev)
 		fw_path = NULL;
 	} else if (IS_SKYLAKE(dev)) {
 		fw_path = I915_SKL_GUC_UCODE;
-		guc_fw->guc_fw_major_wanted = 3;
-		guc_fw->guc_fw_minor_wanted = 0;
+		guc_fw->guc_fw_major_wanted = 4;
+		guc_fw->guc_fw_minor_wanted = 3;
 	} else {
 		i915.enable_guc_submission = false;
 		fw_path = "";	/* unknown device */
-- 
cgit v0.10.2


From f5d3c3eaab1d9e38854deb0251e659d04e3bc771 Mon Sep 17 00:00:00 2001
From: Alex Dai <yu.dai@intel.com>
Date: Tue, 18 Aug 2015 14:34:47 -0700
Subject: drm/i915: Notify GuC rc6 state

If rc6 is enabled, notify GuC so it can do proper forcewake before
command submission.

Signed-off-by: Alex Dai <yu.dai@intel.com>
Reviewed-by: Tom O'Rourke <Tom.O'Rourke@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_guc_submission.c b/drivers/gpu/drm/i915/i915_guc_submission.c
index ec70393..792d0b9 100644
--- a/drivers/gpu/drm/i915/i915_guc_submission.c
+++ b/drivers/gpu/drm/i915/i915_guc_submission.c
@@ -151,6 +151,18 @@ static int host2guc_release_doorbell(struct intel_guc *guc,
 	return host2guc_action(guc, data, 2);
 }
 
+static int host2guc_sample_forcewake(struct intel_guc *guc,
+				     struct i915_guc_client *client)
+{
+	struct drm_i915_private *dev_priv = guc_to_i915(guc);
+	u32 data[2];
+
+	data[0] = HOST2GUC_ACTION_SAMPLE_FORCEWAKE;
+	data[1] = (intel_enable_rc6(dev_priv->dev)) ? 1 : 0;
+
+	return host2guc_action(guc, data, 2);
+}
+
 /*
  * Initialise, update, or clear doorbell data shared with the GuC
  *
@@ -874,6 +886,9 @@ int i915_guc_submission_enable(struct drm_device *dev)
 	}
 
 	guc->execbuf_client = client;
+
+	host2guc_sample_forcewake(guc, client);
+
 	return 0;
 }
 
-- 
cgit v0.10.2


From 79e50a4f7228e5f27d2e9785fad5509d1392a0d3 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Wed, 26 Aug 2015 10:58:20 +0300
Subject: drm/i915: move intel_hrawclk() to intel_display.c

Make it available outside of intel_dp.c.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Clint Taylor <Clinton.A.Taylor@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 439e66e..3f60cf4 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -135,6 +135,39 @@ intel_pch_rawclk(struct drm_device *dev)
 	return I915_READ(PCH_RAWCLK_FREQ) & RAWCLK_FREQ_MASK;
 }
 
+/* hrawclock is 1/4 the FSB frequency */
+int intel_hrawclk(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	uint32_t clkcfg;
+
+	/* There is no CLKCFG reg in Valleyview. VLV hrawclk is 200 MHz */
+	if (IS_VALLEYVIEW(dev))
+		return 200;
+
+	clkcfg = I915_READ(CLKCFG);
+	switch (clkcfg & CLKCFG_FSB_MASK) {
+	case CLKCFG_FSB_400:
+		return 100;
+	case CLKCFG_FSB_533:
+		return 133;
+	case CLKCFG_FSB_667:
+		return 166;
+	case CLKCFG_FSB_800:
+		return 200;
+	case CLKCFG_FSB_1067:
+		return 266;
+	case CLKCFG_FSB_1333:
+		return 333;
+	/* these two are just a guess; one of them might be right */
+	case CLKCFG_FSB_1600:
+	case CLKCFG_FSB_1600_ALT:
+		return 400;
+	default:
+		return 133;
+	}
+}
+
 static inline u32 /* units of 100MHz */
 intel_fdi_link_freq(struct drm_device *dev)
 {
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 3a9c43d..f45872c 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -261,40 +261,6 @@ static void intel_dp_unpack_aux(uint32_t src, uint8_t *dst, int dst_bytes)
 		dst[i] = src >> ((3-i) * 8);
 }
 
-/* hrawclock is 1/4 the FSB frequency */
-static int
-intel_hrawclk(struct drm_device *dev)
-{
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	uint32_t clkcfg;
-
-	/* There is no CLKCFG reg in Valleyview. VLV hrawclk is 200 MHz */
-	if (IS_VALLEYVIEW(dev))
-		return 200;
-
-	clkcfg = I915_READ(CLKCFG);
-	switch (clkcfg & CLKCFG_FSB_MASK) {
-	case CLKCFG_FSB_400:
-		return 100;
-	case CLKCFG_FSB_533:
-		return 133;
-	case CLKCFG_FSB_667:
-		return 166;
-	case CLKCFG_FSB_800:
-		return 200;
-	case CLKCFG_FSB_1067:
-		return 266;
-	case CLKCFG_FSB_1333:
-		return 333;
-	/* these two are just a guess; one of them might be right */
-	case CLKCFG_FSB_1600:
-	case CLKCFG_FSB_1600_ALT:
-		return 400;
-	default:
-		return 133;
-	}
-}
-
 static void
 intel_dp_init_panel_power_sequencer(struct drm_device *dev,
 				    struct intel_dp *intel_dp);
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 72c1181..30012c8 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -1011,6 +1011,7 @@ void i915_audio_component_cleanup(struct drm_i915_private *dev_priv);
 extern const struct drm_plane_funcs intel_plane_funcs;
 bool intel_has_pending_fb_unpin(struct drm_device *dev);
 int intel_pch_rawclk(struct drm_device *dev);
+int intel_hrawclk(struct drm_device *dev);
 void intel_mark_busy(struct drm_device *dev);
 void intel_mark_idle(struct drm_device *dev);
 void intel_crtc_restore_mode(struct drm_crtc *crtc);
-- 
cgit v0.10.2


From 42a8ca4cb4a48ddbf40e8edb291425e76bcdc230 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 27 Aug 2015 16:23:30 +0300
Subject: drm/i915: add yesno utility function

Add a common function to return "yes" or "no" string based on the
argument, and drop the local versions of it.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index 624024c..2f1b693 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -46,11 +46,6 @@ enum {
 	PINNED_LIST,
 };
 
-static const char *yesno(int v)
-{
-	return v ? "yes" : "no";
-}
-
 /* As the drm_debugfs_init() routines are called before dev->dev_private is
  * allocated we need to hook into the minor for release. */
 static int
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index d2c4449..1d99402 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -106,6 +106,11 @@
 	unlikely(__ret_warn_on);					\
 })
 
+static inline const char *yesno(bool v)
+{
+	return v ? "yes" : "no";
+}
+
 enum pipe {
 	INVALID_PIPE = -1,
 	PIPE_A = 0,
diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
index 493e9b2..3379f9c 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.c
+++ b/drivers/gpu/drm/i915/i915_gpu_error.c
@@ -30,11 +30,6 @@
 #include <generated/utsrelease.h>
 #include "i915_drv.h"
 
-static const char *yesno(int v)
-{
-	return v ? "yes" : "no";
-}
-
 static const char *ring_str(int ring)
 {
 	switch (ring) {
-- 
cgit v0.10.2


From 331f38e77d4519d2ca60cea232c7f2bfad5e4d28 Mon Sep 17 00:00:00 2001
From: Zhiyuan Lv <zhiyuan.lv@intel.com>
Date: Fri, 28 Aug 2015 15:41:14 +0800
Subject: drm/i915: preallocate pdps for 32 bit vgpu

This is based on Mika Kuoppala's patch below:
http://article.gmane.org/gmane.comp.freedesktop.xorg.drivers.intel/61104/match=workaround+hw+preload

The patch will preallocate the page directories for 32-bit PPGTT when
i915 runs inside a virtual machine with Intel GVT-g. With this change,
the root pointers in EXECLIST context will always keep the same.

The change is needed for vGPU because Intel GVT-g will do page table
shadowing, and needs to track all the page table changes from guest
i915 driver. However, if guest PPGTT is modified through GPU commands
like LRI, it is not possible to trap the operations in the right time,
so it will be hard to make shadow PPGTT to work correctly.

Shadow PPGTT could be much simpler with this change. Meanwhile
hypervisor could simply prohibit any attempt of PPGTT modification
through GPU command for security.

The function gen8_preallocate_top_level_pdps() in the patch is from
Mika, with only one change to set "used_pdpes" to avoid duplicated
allocation later.

Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Cc: Dave Gordon <david.s.gordon@intel.com>
Cc: Michel Thierry <michel.thierry@intel.com>
Signed-off-by: Zhiyuan Lv <zhiyuan.lv@intel.com>
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 4a76807..ed10e77 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -1441,6 +1441,33 @@ static void gen8_dump_ppgtt(struct i915_hw_ppgtt *ppgtt, struct seq_file *m)
 	}
 }
 
+static int gen8_preallocate_top_level_pdps(struct i915_hw_ppgtt *ppgtt)
+{
+	unsigned long *new_page_dirs, **new_page_tables;
+	uint32_t pdpes = I915_PDPES_PER_PDP(dev);
+	int ret;
+
+	/* We allocate temp bitmap for page tables for no gain
+	 * but as this is for init only, lets keep the things simple
+	 */
+	ret = alloc_gen8_temp_bitmaps(&new_page_dirs, &new_page_tables, pdpes);
+	if (ret)
+		return ret;
+
+	/* Allocate for all pdps regardless of how the ppgtt
+	 * was defined.
+	 */
+	ret = gen8_ppgtt_alloc_page_directories(&ppgtt->base, &ppgtt->pdp,
+						0, 1ULL << 32,
+						new_page_dirs);
+	if (!ret)
+		*ppgtt->pdp.used_pdpes = *new_page_dirs;
+
+	free_gen8_temp_bitmaps(new_page_dirs, new_page_tables, pdpes);
+
+	return ret;
+}
+
 /*
  * GEN8 legacy ppgtt programming is accomplished through a max 4 PDP registers
  * with a net effect resembling a 2-level page table in normal x86 terms. Each
@@ -1484,6 +1511,12 @@ static int gen8_ppgtt_init(struct i915_hw_ppgtt *ppgtt)
 		trace_i915_page_directory_pointer_entry_alloc(&ppgtt->base,
 							      0, 0,
 							      GEN8_PML4E_SHIFT);
+
+		if (intel_vgpu_active(ppgtt->base.dev)) {
+			ret = gen8_preallocate_top_level_pdps(ppgtt);
+			if (ret)
+				goto free_scratch;
+		}
 	}
 
 	return 0;
diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
index 1af84c5..258af9b 100644
--- a/drivers/gpu/drm/i915/intel_lrc.c
+++ b/drivers/gpu/drm/i915/intel_lrc.c
@@ -1540,7 +1540,8 @@ static int gen8_emit_bb_start(struct drm_i915_gem_request *req,
 	 * not needed in 48-bit.*/
 	if (req->ctx->ppgtt &&
 	    (intel_ring_flag(req->ring) & req->ctx->ppgtt->pd_dirty_rings)) {
-		if (!USES_FULL_48BIT_PPGTT(req->i915)) {
+		if (!USES_FULL_48BIT_PPGTT(req->i915) &&
+		    !intel_vgpu_active(req->i915->dev)) {
 			ret = intel_logical_ring_emit_pdps(req);
 			if (ret)
 				return ret;
-- 
cgit v0.10.2


From a0bd6c3183a164970e934ee4f514956dd577a628 Mon Sep 17 00:00:00 2001
From: Zhiyuan Lv <zhiyuan.lv@intel.com>
Date: Fri, 28 Aug 2015 15:41:16 +0800
Subject: drm/i915: Always enable execlists on BDW for vgpu

Broadwell hardware supports both ring buffer mode and execlist mode.
When i915 runs inside a VM with Intel GVT-g, we allow execlist mode
only.

The main reason of EXECLIST only is that GVT-g does not support the
dynamic mode switch between ring buffer mode and execlist mode when
running multiple virtual machines.

v2:
- Adjust the position of vgpu check in sanitize function (Joonas)
- Add vgpu error check in context initialization. (Joonas, Daniel)

Signed-off-by: Zhiyuan Lv <zhiyuan.lv@intel.com>
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c
index 8e893b3..74aa0c9 100644
--- a/drivers/gpu/drm/i915/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/i915_gem_context.c
@@ -332,6 +332,13 @@ int i915_gem_context_init(struct drm_device *dev)
 	if (WARN_ON(dev_priv->ring[RCS].default_context))
 		return 0;
 
+	if (intel_vgpu_active(dev) && HAS_LOGICAL_RING_CONTEXTS(dev)) {
+		if (!i915.enable_execlists) {
+			DRM_INFO("Only EXECLIST mode is supported in vgpu.\n");
+			return -EINVAL;
+		}
+	}
+
 	if (i915.enable_execlists) {
 		/* NB: intentionally left blank. We will allocate our own
 		 * backing objects as we need them, thank you very much */
diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
index 258af9b..e9520af 100644
--- a/drivers/gpu/drm/i915/intel_lrc.c
+++ b/drivers/gpu/drm/i915/intel_lrc.c
@@ -236,6 +236,12 @@ int intel_sanitize_enable_execlists(struct drm_device *dev, int enable_execlists
 {
 	WARN_ON(i915.enable_ppgtt == -1);
 
+	/* On platforms with execlist available, vGPU will only
+	 * support execlist mode, no ring buffer mode.
+	 */
+	if (HAS_LOGICAL_RING_CONTEXTS(dev) && intel_vgpu_active(dev))
+		return 1;
+
 	if (INTEL_INFO(dev)->gen >= 9)
 		return 1;
 
-- 
cgit v0.10.2


From 532beabf343d16d269491c63ed4590e5a675ea2b Mon Sep 17 00:00:00 2001
From: Zhiyuan Lv <zhiyuan.lv@intel.com>
Date: Fri, 28 Aug 2015 15:41:17 +0800
Subject: drm/i915: Update PV INFO page definition for Intel GVT-g

Some more definitions in the PV info page are added. They are mainly
for the guest notification to Intel GVT-g device model. They are used
for Broadwell enabling.

The notification of PPGTT page table creation/destroy is to notify
GVT-g device model the life cycle of guest page tables. Then device
model will implement shadow page table for guests.

The notification of context create/destroy is optional. If it is used,
the device model will create/destroy shadow context corresponding to
the context's life cycle. Guest driver needs to make sure that the
context's LRCA and backing storage address unchanged. If it is not
used, the device model will perform the context shadow work in the
context scheduling time.

Signed-off-by: Zhiyuan Lv <zhiyuan.lv@intel.com>
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_vgpu.h b/drivers/gpu/drm/i915/i915_vgpu.h
index 97a88b5..21c97f4 100644
--- a/drivers/gpu/drm/i915/i915_vgpu.h
+++ b/drivers/gpu/drm/i915/i915_vgpu.h
@@ -40,6 +40,19 @@
 #define INTEL_VGT_IF_VERSION \
 	INTEL_VGT_IF_VERSION_ENCODE(VGT_VERSION_MAJOR, VGT_VERSION_MINOR)
 
+/*
+ * notifications from guest to vgpu device model
+ */
+enum vgt_g2v_type {
+	VGT_G2V_PPGTT_L3_PAGE_TABLE_CREATE = 2,
+	VGT_G2V_PPGTT_L3_PAGE_TABLE_DESTROY,
+	VGT_G2V_PPGTT_L4_PAGE_TABLE_CREATE,
+	VGT_G2V_PPGTT_L4_PAGE_TABLE_DESTROY,
+	VGT_G2V_EXECLIST_CONTEXT_CREATE,
+	VGT_G2V_EXECLIST_CONTEXT_DESTROY,
+	VGT_G2V_MAX,
+};
+
 struct vgt_if {
 	uint64_t magic;		/* VGT_MAGIC */
 	uint16_t version_major;
@@ -70,11 +83,28 @@ struct vgt_if {
 	uint32_t rsv3[0x200 - 24];	/* pad to half page */
 	/*
 	 * The bottom half page is for response from Gfx driver to hypervisor.
-	 * Set to reserved fields temporarily by now.
 	 */
 	uint32_t rsv4;
 	uint32_t display_ready;	/* ready for display owner switch */
-	uint32_t rsv5[0x200 - 2];	/* pad to one page */
+
+	uint32_t rsv5[4];
+
+	uint32_t g2v_notify;
+	uint32_t rsv6[7];
+
+	uint32_t pdp0_lo;
+	uint32_t pdp0_hi;
+	uint32_t pdp1_lo;
+	uint32_t pdp1_hi;
+	uint32_t pdp2_lo;
+	uint32_t pdp2_hi;
+	uint32_t pdp3_lo;
+	uint32_t pdp3_hi;
+
+	uint32_t execlist_context_descriptor_lo;
+	uint32_t execlist_context_descriptor_hi;
+
+	uint32_t  rsv7[0x200 - 24];    /* pad to one page */
 } __packed;
 
 #define vgtif_reg(x) \
-- 
cgit v0.10.2


From 650da34c4af12921e0148bbc0dba7419d7776189 Mon Sep 17 00:00:00 2001
From: Zhiyuan Lv <zhiyuan.lv@intel.com>
Date: Fri, 28 Aug 2015 15:41:18 +0800
Subject: drm/i915: guest i915 notification for Intel GVT-g

When i915 drivers run inside a VM with Intel GVT-g, some explicit
notifications are needed from guest to host device model through PV
INFO page write. The notifications include:

	PPGTT create
	PPGTT destroy

They are used for the shadow implementation of PPGTT. Intel GVT-g
needs to write-protect the guest pages of PPGTT, and clear the write
protection when they end their life cycle.

v2:
- Use lower_32_bits()/upper_32_bits() for qword operations;
- Remove the notification of guest context creation/destroy;

Signed-off-by: Zhiyuan Lv <zhiyuan.lv@intel.com>
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index ed10e77..bdb7adc 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -899,6 +899,41 @@ static int gen8_init_scratch(struct i915_address_space *vm)
 	return 0;
 }
 
+static int gen8_ppgtt_notify_vgt(struct i915_hw_ppgtt *ppgtt, bool create)
+{
+	enum vgt_g2v_type msg;
+	struct drm_device *dev = ppgtt->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	unsigned int offset = vgtif_reg(pdp0_lo);
+	int i;
+
+	if (USES_FULL_48BIT_PPGTT(dev)) {
+		u64 daddr = px_dma(&ppgtt->pml4);
+
+		I915_WRITE(offset, lower_32_bits(daddr));
+		I915_WRITE(offset + 4, upper_32_bits(daddr));
+
+		msg = (create ? VGT_G2V_PPGTT_L4_PAGE_TABLE_CREATE :
+				VGT_G2V_PPGTT_L4_PAGE_TABLE_DESTROY);
+	} else {
+		for (i = 0; i < GEN8_LEGACY_PDPES; i++) {
+			u64 daddr = i915_page_dir_dma_addr(ppgtt, i);
+
+			I915_WRITE(offset, lower_32_bits(daddr));
+			I915_WRITE(offset + 4, upper_32_bits(daddr));
+
+			offset += 8;
+		}
+
+		msg = (create ? VGT_G2V_PPGTT_L3_PAGE_TABLE_CREATE :
+				VGT_G2V_PPGTT_L3_PAGE_TABLE_DESTROY);
+	}
+
+	I915_WRITE(vgtif_reg(g2v_notify), msg);
+
+	return 0;
+}
+
 static void gen8_free_scratch(struct i915_address_space *vm)
 {
 	struct drm_device *dev = vm->dev;
@@ -945,6 +980,9 @@ static void gen8_ppgtt_cleanup(struct i915_address_space *vm)
 	struct i915_hw_ppgtt *ppgtt =
 		container_of(vm, struct i915_hw_ppgtt, base);
 
+	if (intel_vgpu_active(vm->dev))
+		gen8_ppgtt_notify_vgt(ppgtt, false);
+
 	if (!USES_FULL_48BIT_PPGTT(ppgtt->base.dev))
 		gen8_ppgtt_cleanup_3lvl(ppgtt->base.dev, &ppgtt->pdp);
 	else
@@ -1519,6 +1557,9 @@ static int gen8_ppgtt_init(struct i915_hw_ppgtt *ppgtt)
 		}
 	}
 
+	if (intel_vgpu_active(ppgtt->base.dev))
+		gen8_ppgtt_notify_vgt(ppgtt, true);
+
 	return 0;
 
 free_scratch:
-- 
cgit v0.10.2


From 34e0adbb4508a318e8039ee3959383e4bf080f04 Mon Sep 17 00:00:00 2001
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Date: Mon, 31 Aug 2015 13:04:25 +0200
Subject: drm/i915: Remove start frame argument to pipe_update_begin/end.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

There's already a per crtc member that can be used for it.

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 3f60cf4..d38ceb0 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -11142,11 +11142,10 @@ static void ilk_do_mmio_flip(struct intel_crtc *intel_crtc)
 static void intel_do_mmio_flip(struct intel_crtc *intel_crtc)
 {
 	struct drm_device *dev = intel_crtc->base.dev;
-	u32 start_vbl_count;
 
 	intel_mark_page_flip_active(intel_crtc);
 
-	intel_pipe_update_start(intel_crtc, &start_vbl_count);
+	intel_pipe_update_start(intel_crtc);
 
 	if (INTEL_INFO(dev)->gen >= 9)
 		skl_do_mmio_flip(intel_crtc);
@@ -11154,7 +11153,7 @@ static void intel_do_mmio_flip(struct intel_crtc *intel_crtc)
 		/* use_mmio_flip() retricts MMIO flips to ilk+ */
 		ilk_do_mmio_flip(intel_crtc);
 
-	intel_pipe_update_end(intel_crtc, start_vbl_count);
+	intel_pipe_update_end(intel_crtc);
 }
 
 static void intel_mmio_flip_work_func(struct work_struct *work)
@@ -13463,7 +13462,7 @@ static void intel_begin_crtc_commit(struct drm_crtc *crtc,
 
 	/* Perform vblank evasion around commit operation */
 	if (crtc->state->active)
-		intel_pipe_update_start(intel_crtc, &intel_crtc->start_vbl_count);
+		intel_pipe_update_start(intel_crtc);
 
 	if (!needs_modeset(crtc->state) && INTEL_INFO(dev)->gen >= 9)
 		skl_detach_scalers(intel_crtc);
@@ -13475,7 +13474,7 @@ static void intel_finish_crtc_commit(struct drm_crtc *crtc,
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 
 	if (crtc->state->active)
-		intel_pipe_update_end(intel_crtc, intel_crtc->start_vbl_count);
+		intel_pipe_update_end(intel_crtc);
 }
 
 /**
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 30012c8..8f8d320 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -1414,9 +1414,8 @@ bool intel_sdvo_init(struct drm_device *dev, uint32_t sdvo_reg, bool is_sdvob);
 int intel_plane_init(struct drm_device *dev, enum pipe pipe, int plane);
 int intel_sprite_set_colorkey(struct drm_device *dev, void *data,
 			      struct drm_file *file_priv);
-void intel_pipe_update_start(struct intel_crtc *crtc,
-			     uint32_t *start_vbl_count);
-void intel_pipe_update_end(struct intel_crtc *crtc, u32 start_vbl_count);
+void intel_pipe_update_start(struct intel_crtc *crtc);
+void intel_pipe_update_end(struct intel_crtc *crtc);
 
 /* intel_tv.c */
 void intel_tv_init(struct drm_device *dev);
diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c
index 9d8af2f..c35c233 100644
--- a/drivers/gpu/drm/i915/intel_sprite.c
+++ b/drivers/gpu/drm/i915/intel_sprite.c
@@ -76,7 +76,7 @@ static int usecs_to_scanlines(const struct drm_display_mode *mode, int usecs)
  * avoid random delays. The value written to @start_vbl_count should be
  * supplied to intel_pipe_update_end() for error checking.
  */
-void intel_pipe_update_start(struct intel_crtc *crtc, uint32_t *start_vbl_count)
+void intel_pipe_update_start(struct intel_crtc *crtc)
 {
 	struct drm_device *dev = crtc->base.dev;
 	const struct drm_display_mode *mode = &crtc->config->base.adjusted_mode;
@@ -95,7 +95,7 @@ void intel_pipe_update_start(struct intel_crtc *crtc, uint32_t *start_vbl_count)
 	max = vblank_start - 1;
 
 	local_irq_disable();
-	*start_vbl_count = 0;
+	crtc->start_vbl_count = 0;
 
 	if (min <= 0 || max <= 0)
 		return;
@@ -134,9 +134,10 @@ void intel_pipe_update_start(struct intel_crtc *crtc, uint32_t *start_vbl_count)
 
 	drm_crtc_vblank_put(&crtc->base);
 
-	*start_vbl_count = dev->driver->get_vblank_counter(dev, pipe);
+	crtc->start_vbl_count = dev->driver->get_vblank_counter(dev, pipe);
 
-	trace_i915_pipe_update_vblank_evaded(crtc, min, max, *start_vbl_count);
+	trace_i915_pipe_update_vblank_evaded(crtc, min, max,
+					     crtc->start_vbl_count);
 }
 
 /**
@@ -148,7 +149,7 @@ void intel_pipe_update_start(struct intel_crtc *crtc, uint32_t *start_vbl_count)
  * re-enables interrupts and verifies the update was actually completed
  * before a vblank using the value of @start_vbl_count.
  */
-void intel_pipe_update_end(struct intel_crtc *crtc, u32 start_vbl_count)
+void intel_pipe_update_end(struct intel_crtc *crtc)
 {
 	struct drm_device *dev = crtc->base.dev;
 	enum pipe pipe = crtc->pipe;
@@ -158,9 +159,9 @@ void intel_pipe_update_end(struct intel_crtc *crtc, u32 start_vbl_count)
 
 	local_irq_enable();
 
-	if (start_vbl_count && start_vbl_count != end_vbl_count)
+	if (crtc->start_vbl_count && crtc->start_vbl_count != end_vbl_count)
 		DRM_ERROR("Atomic update failure on pipe %c (start=%u end=%u)\n",
-			  pipe_name(pipe), start_vbl_count, end_vbl_count);
+			  pipe_name(pipe), crtc->start_vbl_count, end_vbl_count);
 }
 
 static void
-- 
cgit v0.10.2


From 85a62bf9d8ef8d533635270ae985281c58e8c974 Mon Sep 17 00:00:00 2001
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Date: Tue, 1 Sep 2015 12:15:33 +0200
Subject: drm/i915: Also record time difference if vblank evasion fails, v2.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This makes the error message slightly more useful.

Changes since v1:
- Use ktime_get() while irqs are still disabled. (vsyrjala)

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 8f8d320..090d67b 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -564,6 +564,8 @@ struct intel_crtc {
 	int scanline_offset;
 
 	unsigned start_vbl_count;
+	ktime_t start_vbl_time;
+
 	struct intel_crtc_atomic_commit atomic;
 
 	/* scalers available on this crtc */
diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c
index c35c233..ca7e264 100644
--- a/drivers/gpu/drm/i915/intel_sprite.c
+++ b/drivers/gpu/drm/i915/intel_sprite.c
@@ -134,6 +134,7 @@ void intel_pipe_update_start(struct intel_crtc *crtc)
 
 	drm_crtc_vblank_put(&crtc->base);
 
+	crtc->start_vbl_time = ktime_get();
 	crtc->start_vbl_count = dev->driver->get_vblank_counter(dev, pipe);
 
 	trace_i915_pipe_update_vblank_evaded(crtc, min, max,
@@ -154,14 +155,16 @@ void intel_pipe_update_end(struct intel_crtc *crtc)
 	struct drm_device *dev = crtc->base.dev;
 	enum pipe pipe = crtc->pipe;
 	u32 end_vbl_count = dev->driver->get_vblank_counter(dev, pipe);
+	ktime_t end_vbl_time = ktime_get();
 
 	trace_i915_pipe_update_end(crtc, end_vbl_count);
 
 	local_irq_enable();
 
 	if (crtc->start_vbl_count && crtc->start_vbl_count != end_vbl_count)
-		DRM_ERROR("Atomic update failure on pipe %c (start=%u end=%u)\n",
-			  pipe_name(pipe), crtc->start_vbl_count, end_vbl_count);
+		DRM_ERROR("Atomic update failure on pipe %c (start=%u end=%u) time %lld us\n",
+			  pipe_name(pipe), crtc->start_vbl_count, end_vbl_count,
+			  ktime_us_delta(end_vbl_time, crtc->start_vbl_time));
 }
 
 static void
-- 
cgit v0.10.2


From 62f0ace531b2ca69a1defe5a01a85539fd28353f Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Wed, 26 Aug 2015 18:57:26 +0200
Subject: drm/i915: Update comments around base bpp

Forgot to do that in

commit d328c9d78d64ca11e744fe227096990430a88477
Author: Daniel Vetter <daniel.vetter@ffwll.ch>
Date:   Fri Apr 10 16:22:37 2015 +0200

    drm/i915: Select starting pipe bpp irrespective or the primary plane

and it's confusing. Fix it.

Cc: Jesse Barnes <jbarnes@virtuousgeek.org>
Acked-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index deba333..ae485e7 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -12102,10 +12102,6 @@ intel_modeset_pipe_config(struct drm_crtc *crtc,
 	      (DRM_MODE_FLAG_PVSYNC | DRM_MODE_FLAG_NVSYNC)))
 		pipe_config->base.adjusted_mode.flags |= DRM_MODE_FLAG_NVSYNC;
 
-	/* Compute a starting value for pipe_config->pipe_bpp taking the source
-	 * plane pixel format and any sink constraints into account. Returns the
-	 * source plane bpp so that dithering can be selected on mismatches
-	 * after encoders and crtc also have had their say. */
 	base_bpp = compute_baseline_pipe_bpp(to_intel_crtc(crtc),
 					     pipe_config);
 	if (base_bpp < 0)
@@ -12174,7 +12170,7 @@ encoder_retry:
 	/* Dithering seems to not pass-through bits correctly when it should, so
 	 * only enable it on 6bpc panels. */
 	pipe_config->dither = pipe_config->pipe_bpp == 6*3;
-	DRM_DEBUG_KMS("plane bpp: %i, pipe bpp: %i, dithering: %i\n",
+	DRM_DEBUG_KMS("hw max bpp: %i, pipe bpp: %i, dithering: %i\n",
 		      base_bpp, pipe_config->pipe_bpp, pipe_config->dither);
 
 fail:
-- 
cgit v0.10.2


From 40bfd7a3303b7c383493c80a250c59b61d812ce5 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 27 Aug 2015 23:55:56 +0300
Subject: drm/i915: Clean up various HPD defines
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Indent the PORTx_HOTPLUG_... defines appropriately, and fix some space
vs. tab issues.

v2: Document pre-HSW/LPT bits, and order another tab (Paulo)

Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index e7c9dc8..e36baca 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -5394,15 +5394,17 @@ enum skl_disp_power_wells {
 
 #define CPU_VGACNTRL	0x41000
 
-#define DIGITAL_PORT_HOTPLUG_CNTRL      0x44030
-#define  DIGITAL_PORTA_HOTPLUG_ENABLE           (1 << 4)
-#define  DIGITAL_PORTA_SHORT_PULSE_2MS          (0 << 2)
-#define  DIGITAL_PORTA_SHORT_PULSE_4_5MS        (1 << 2)
-#define  DIGITAL_PORTA_SHORT_PULSE_6MS          (2 << 2)
-#define  DIGITAL_PORTA_SHORT_PULSE_100MS        (3 << 2)
-#define  DIGITAL_PORTA_NO_DETECT                (0 << 0)
-#define  DIGITAL_PORTA_LONG_PULSE_DETECT_MASK   (1 << 1)
-#define  DIGITAL_PORTA_SHORT_PULSE_DETECT_MASK  (1 << 0)
+#define DIGITAL_PORT_HOTPLUG_CNTRL	0x44030
+#define  DIGITAL_PORTA_HOTPLUG_ENABLE		(1 << 4)
+#define  DIGITAL_PORTA_PULSE_DURATION_2ms	(0 << 2) /* pre-HSW */
+#define  DIGITAL_PORTA_PULSE_DURATION_4_5ms	(1 << 2) /* pre-HSW */
+#define  DIGITAL_PORTA_PULSE_DURATION_6ms	(2 << 2) /* pre-HSW */
+#define  DIGITAL_PORTA_PULSE_DURATION_100ms	(3 << 2) /* pre-HSW */
+#define  DIGITAL_PORTA_PULSE_DURATION_MASK	(3 << 2) /* pre-HSW */
+#define  DIGITAL_PORTA_HOTPLUG_STATUS_MASK	(3 << 0)
+#define  DIGITAL_PORTA_HOTPLUG_NO_DETECT	(0 << 0)
+#define  DIGITAL_PORTA_HOTPLUG_SHORT_DETECT	(1 << 0)
+#define  DIGITAL_PORTA_HOTPLUG_LONG_DETECT	(2 << 0)
 
 /* refresh rate hardware control */
 #define RR_HW_CTL       0x45300
@@ -6030,46 +6032,46 @@ enum skl_disp_power_wells {
 #define  SERR_INT_TRANS_FIFO_UNDERRUN(pipe)	(1<<(pipe*3))
 
 /* digital port hotplug */
-#define PCH_PORT_HOTPLUG        0xc4030		/* SHOTPLUG_CTL */
-#define BXT_PORTA_HOTPLUG_ENABLE	(1 << 28)
-#define BXT_PORTA_HOTPLUG_STATUS_MASK	(0x3 << 24)
+#define PCH_PORT_HOTPLUG		0xc4030	/* SHOTPLUG_CTL */
+#define  BXT_PORTA_HOTPLUG_ENABLE	(1 << 28)
+#define  BXT_PORTA_HOTPLUG_STATUS_MASK	(3 << 24)
 #define  BXT_PORTA_HOTPLUG_NO_DETECT	(0 << 24)
 #define  BXT_PORTA_HOTPLUG_SHORT_DETECT	(1 << 24)
 #define  BXT_PORTA_HOTPLUG_LONG_DETECT	(2 << 24)
-#define PORTD_HOTPLUG_ENABLE            (1 << 20)
-#define PORTD_PULSE_DURATION_2ms        (0)
-#define PORTD_PULSE_DURATION_4_5ms      (1 << 18)
-#define PORTD_PULSE_DURATION_6ms        (2 << 18)
-#define PORTD_PULSE_DURATION_100ms      (3 << 18)
-#define PORTD_PULSE_DURATION_MASK	(3 << 18)
-#define PORTD_HOTPLUG_STATUS_MASK	(0x3 << 16)
+#define  PORTD_HOTPLUG_ENABLE		(1 << 20)
+#define  PORTD_PULSE_DURATION_2ms	(0 << 18) /* pre-LPT */
+#define  PORTD_PULSE_DURATION_4_5ms	(1 << 18) /* pre-LPT */
+#define  PORTD_PULSE_DURATION_6ms	(2 << 18) /* pre-LPT */
+#define  PORTD_PULSE_DURATION_100ms	(3 << 18) /* pre-LPT */
+#define  PORTD_PULSE_DURATION_MASK	(3 << 18) /* pre-LPT */
+#define  PORTD_HOTPLUG_STATUS_MASK	(3 << 16)
 #define  PORTD_HOTPLUG_NO_DETECT	(0 << 16)
 #define  PORTD_HOTPLUG_SHORT_DETECT	(1 << 16)
 #define  PORTD_HOTPLUG_LONG_DETECT	(2 << 16)
-#define PORTC_HOTPLUG_ENABLE            (1 << 12)
-#define PORTC_PULSE_DURATION_2ms        (0)
-#define PORTC_PULSE_DURATION_4_5ms      (1 << 10)
-#define PORTC_PULSE_DURATION_6ms        (2 << 10)
-#define PORTC_PULSE_DURATION_100ms      (3 << 10)
-#define PORTC_PULSE_DURATION_MASK	(3 << 10)
-#define PORTC_HOTPLUG_STATUS_MASK	(0x3 << 8)
+#define  PORTC_HOTPLUG_ENABLE		(1 << 12)
+#define  PORTC_PULSE_DURATION_2ms	(0 << 10) /* pre-LPT */
+#define  PORTC_PULSE_DURATION_4_5ms	(1 << 10) /* pre-LPT */
+#define  PORTC_PULSE_DURATION_6ms	(2 << 10) /* pre-LPT */
+#define  PORTC_PULSE_DURATION_100ms	(3 << 10) /* pre-LPT */
+#define  PORTC_PULSE_DURATION_MASK	(3 << 10) /* pre-LPT */
+#define  PORTC_HOTPLUG_STATUS_MASK	(3 << 8)
 #define  PORTC_HOTPLUG_NO_DETECT	(0 << 8)
 #define  PORTC_HOTPLUG_SHORT_DETECT	(1 << 8)
 #define  PORTC_HOTPLUG_LONG_DETECT	(2 << 8)
-#define PORTB_HOTPLUG_ENABLE            (1 << 4)
-#define PORTB_PULSE_DURATION_2ms        (0)
-#define PORTB_PULSE_DURATION_4_5ms      (1 << 2)
-#define PORTB_PULSE_DURATION_6ms        (2 << 2)
-#define PORTB_PULSE_DURATION_100ms      (3 << 2)
-#define PORTB_PULSE_DURATION_MASK	(3 << 2)
-#define PORTB_HOTPLUG_STATUS_MASK	(0x3 << 0)
+#define  PORTB_HOTPLUG_ENABLE		(1 << 4)
+#define  PORTB_PULSE_DURATION_2ms	(0 << 2) /* pre-LPT */
+#define  PORTB_PULSE_DURATION_4_5ms	(1 << 2) /* pre-LPT */
+#define  PORTB_PULSE_DURATION_6ms	(2 << 2) /* pre-LPT */
+#define  PORTB_PULSE_DURATION_100ms	(3 << 2) /* pre-LPT */
+#define  PORTB_PULSE_DURATION_MASK	(3 << 2) /* pre-LPT */
+#define  PORTB_HOTPLUG_STATUS_MASK	(3 << 0)
 #define  PORTB_HOTPLUG_NO_DETECT	(0 << 0)
 #define  PORTB_HOTPLUG_SHORT_DETECT	(1 << 0)
 #define  PORTB_HOTPLUG_LONG_DETECT	(2 << 0)
 
-#define PCH_PORT_HOTPLUG2        0xc403C		/* SHOTPLUG_CTL2 */
-#define PORTE_HOTPLUG_ENABLE            (1 << 4)
-#define PORTE_HOTPLUG_STATUS_MASK	(0x3 << 0)
+#define PCH_PORT_HOTPLUG2		0xc403C	/* SHOTPLUG_CTL2 SPT+ */
+#define  PORTE_HOTPLUG_ENABLE		(1 << 4)
+#define  PORTE_HOTPLUG_STATUS_MASK	(3 << 0)
 #define  PORTE_HOTPLUG_NO_DETECT	(0 << 0)
 #define  PORTE_HOTPLUG_SHORT_DETECT	(1 << 0)
 #define  PORTE_HOTPLUG_LONG_DETECT	(2 << 0)
-- 
cgit v0.10.2


From 87a0210665eeca7c5fa237592da025650b216831 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 27 Aug 2015 23:55:57 +0300
Subject: drm/i915: Extract intel_hpd_enabled_irqs()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Eliminate a bunch of duplicated code that calculates the currently
enabled HPD interrupt bits.

v2: s/;/:/ in patch subject (Paulo)

Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 8485bea..de0edbd 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -3002,27 +3002,34 @@ static void cherryview_irq_preinstall(struct drm_device *dev)
 	vlv_display_irq_reset(dev_priv);
 }
 
+static u32 intel_hpd_enabled_irqs(struct drm_device *dev,
+				  const u32 hpd[HPD_NUM_PINS])
+{
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct intel_encoder *encoder;
+	u32 enabled_irqs = 0;
+
+	for_each_intel_encoder(dev, encoder)
+		if (dev_priv->hotplug.stats[encoder->hpd_pin].state == HPD_ENABLED)
+			enabled_irqs |= hpd[encoder->hpd_pin];
+
+	return enabled_irqs;
+}
+
 static void ibx_hpd_irq_setup(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_encoder *intel_encoder;
-	u32 hotplug_irqs, hotplug, enabled_irqs = 0;
+	u32 hotplug_irqs, hotplug, enabled_irqs;
 
 	if (HAS_PCH_IBX(dev)) {
 		hotplug_irqs = SDE_HOTPLUG_MASK;
-		for_each_intel_encoder(dev, intel_encoder)
-			if (dev_priv->hotplug.stats[intel_encoder->hpd_pin].state == HPD_ENABLED)
-				enabled_irqs |= hpd_ibx[intel_encoder->hpd_pin];
+		enabled_irqs = intel_hpd_enabled_irqs(dev, hpd_ibx);
 	} else if (HAS_PCH_SPT(dev)) {
 		hotplug_irqs = SDE_HOTPLUG_MASK_SPT;
-		for_each_intel_encoder(dev, intel_encoder)
-			if (dev_priv->hotplug.stats[intel_encoder->hpd_pin].state == HPD_ENABLED)
-				enabled_irqs |= hpd_spt[intel_encoder->hpd_pin];
+		enabled_irqs = intel_hpd_enabled_irqs(dev, hpd_spt);
 	} else {
 		hotplug_irqs = SDE_HOTPLUG_MASK_CPT;
-		for_each_intel_encoder(dev, intel_encoder)
-			if (dev_priv->hotplug.stats[intel_encoder->hpd_pin].state == HPD_ENABLED)
-				enabled_irqs |= hpd_cpt[intel_encoder->hpd_pin];
+		enabled_irqs = intel_hpd_enabled_irqs(dev, hpd_cpt);
 	}
 
 	ibx_display_interrupt_update(dev_priv, hotplug_irqs, enabled_irqs);
@@ -3051,15 +3058,10 @@ static void ibx_hpd_irq_setup(struct drm_device *dev)
 static void bxt_hpd_irq_setup(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_encoder *intel_encoder;
-	u32 hotplug_port = 0;
+	u32 hotplug_port;
 	u32 hotplug_ctrl;
 
-	for_each_intel_encoder(dev, intel_encoder) {
-		if (dev_priv->hotplug.stats[intel_encoder->hpd_pin].state
-				== HPD_ENABLED)
-			hotplug_port |= hpd_bxt[intel_encoder->hpd_pin];
-	}
+	hotplug_port = intel_hpd_enabled_irqs(dev, hpd_bxt);
 
 	hotplug_ctrl = I915_READ(BXT_HOTPLUG_CTL) & ~BXT_HOTPLUG_CTL_MASK;
 
@@ -3935,7 +3937,6 @@ static int i965_irq_postinstall(struct drm_device *dev)
 static void i915_hpd_irq_setup(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_encoder *intel_encoder;
 	u32 hotplug_en;
 
 	assert_spin_locked(&dev_priv->irq_lock);
@@ -3944,9 +3945,7 @@ static void i915_hpd_irq_setup(struct drm_device *dev)
 	hotplug_en &= ~HOTPLUG_INT_EN_MASK;
 	/* Note HDMI and DP share hotplug bits */
 	/* enable bits are the same for all generations */
-	for_each_intel_encoder(dev, intel_encoder)
-		if (dev_priv->hotplug.stats[intel_encoder->hpd_pin].state == HPD_ENABLED)
-			hotplug_en |= hpd_mask_i915[intel_encoder->hpd_pin];
+	hotplug_en |= intel_hpd_enabled_irqs(dev, hpd_mask_i915);
 	/* Programming the CRT detection parameters tends
 	   to generate a spurious hotplug event about three
 	   seconds later.  So just do it once.
-- 
cgit v0.10.2


From d9dc34f1a1e9649f6e1e3709f6177f7883ce75ce Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 27 Aug 2015 23:55:58 +0300
Subject: drm/i915: Factor out ilk_update_display_irq()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Extract the core of ironlake_{enable,disable}_display_irq() into a new
function. We'll have further use for it later.

v2: Warn about invalid mask vs. enable bits (Paulo)

Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index de0edbd..baff963 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -154,35 +154,46 @@ static const u32 hpd_bxt[HPD_NUM_PINS] = {
 
 static void gen6_rps_irq_handler(struct drm_i915_private *dev_priv, u32 pm_iir);
 
-/* For display hotplug interrupt */
-void
-ironlake_enable_display_irq(struct drm_i915_private *dev_priv, u32 mask)
+/**
+ * ilk_update_display_irq - update DEIMR
+ * @dev_priv: driver private
+ * @interrupt_mask: mask of interrupt bits to update
+ * @enabled_irq_mask: mask of interrupt bits to enable
+ */
+static void ilk_update_display_irq(struct drm_i915_private *dev_priv,
+				   uint32_t interrupt_mask,
+				   uint32_t enabled_irq_mask)
 {
+	uint32_t new_val;
+
 	assert_spin_locked(&dev_priv->irq_lock);
 
+	WARN_ON(enabled_irq_mask & ~interrupt_mask);
+
 	if (WARN_ON(!intel_irqs_enabled(dev_priv)))
 		return;
 
-	if ((dev_priv->irq_mask & mask) != 0) {
-		dev_priv->irq_mask &= ~mask;
+	new_val = dev_priv->irq_mask;
+	new_val &= ~interrupt_mask;
+	new_val |= (~enabled_irq_mask & interrupt_mask);
+
+	if (new_val != dev_priv->irq_mask) {
+		dev_priv->irq_mask = new_val;
 		I915_WRITE(DEIMR, dev_priv->irq_mask);
 		POSTING_READ(DEIMR);
 	}
 }
 
 void
-ironlake_disable_display_irq(struct drm_i915_private *dev_priv, u32 mask)
+ironlake_enable_display_irq(struct drm_i915_private *dev_priv, u32 mask)
 {
-	assert_spin_locked(&dev_priv->irq_lock);
-
-	if (WARN_ON(!intel_irqs_enabled(dev_priv)))
-		return;
+	ilk_update_display_irq(dev_priv, mask, mask);
+}
 
-	if ((dev_priv->irq_mask & mask) != mask) {
-		dev_priv->irq_mask |= mask;
-		I915_WRITE(DEIMR, dev_priv->irq_mask);
-		POSTING_READ(DEIMR);
-	}
+void
+ironlake_disable_display_irq(struct drm_i915_private *dev_priv, u32 mask)
+{
+	ilk_update_display_irq(dev_priv, mask, 0);
 }
 
 /**
-- 
cgit v0.10.2


From c2699524d666bd606a2de0aa68691c984efb1fec Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 27 Aug 2015 23:55:59 +0300
Subject: drm/i915: Add HAS_PCH_LPT_LP() macro
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Make LPT:LP checks look neater by wrapping the details in a
new HAS_PCH_LPT_LP() macro.

Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 4eabe19..482ddb7 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -2599,6 +2599,7 @@ struct drm_i915_cmd_table {
 #define INTEL_PCH_TYPE(dev) (__I915__(dev)->pch_type)
 #define HAS_PCH_SPT(dev) (INTEL_PCH_TYPE(dev) == PCH_SPT)
 #define HAS_PCH_LPT(dev) (INTEL_PCH_TYPE(dev) == PCH_LPT)
+#define HAS_PCH_LPT_LP(dev) (__I915__(dev)->pch_id == INTEL_PCH_LPT_LP_DEVICE_ID_TYPE)
 #define HAS_PCH_CPT(dev) (INTEL_PCH_TYPE(dev) == PCH_CPT)
 #define HAS_PCH_IBX(dev) (INTEL_PCH_TYPE(dev) == PCH_IBX)
 #define HAS_PCH_NOP(dev) (INTEL_PCH_TYPE(dev) == PCH_NOP)
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index ae485e7..888175c 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -8359,8 +8359,7 @@ static void lpt_enable_clkout_dp(struct drm_device *dev, bool with_spread,
 
 	if (WARN(with_fdi && !with_spread, "FDI requires downspread\n"))
 		with_spread = true;
-	if (WARN(dev_priv->pch_id == INTEL_PCH_LPT_LP_DEVICE_ID_TYPE &&
-		 with_fdi, "LP PCH doesn't have FDI\n"))
+	if (WARN(HAS_PCH_LPT_LP(dev) && with_fdi, "LP PCH doesn't have FDI\n"))
 		with_fdi = false;
 
 	mutex_lock(&dev_priv->sb_lock);
@@ -8383,8 +8382,7 @@ static void lpt_enable_clkout_dp(struct drm_device *dev, bool with_spread,
 		}
 	}
 
-	reg = (dev_priv->pch_id == INTEL_PCH_LPT_LP_DEVICE_ID_TYPE) ?
-	       SBI_GEN0 : SBI_DBUFF0;
+	reg = HAS_PCH_LPT_LP(dev) ? SBI_GEN0 : SBI_DBUFF0;
 	tmp = intel_sbi_read(dev_priv, reg, SBI_ICLK);
 	tmp |= SBI_GEN0_CFG_BUFFENABLE_DISABLE;
 	intel_sbi_write(dev_priv, reg, tmp, SBI_ICLK);
@@ -8400,8 +8398,7 @@ static void lpt_disable_clkout_dp(struct drm_device *dev)
 
 	mutex_lock(&dev_priv->sb_lock);
 
-	reg = (dev_priv->pch_id == INTEL_PCH_LPT_LP_DEVICE_ID_TYPE) ?
-	       SBI_GEN0 : SBI_DBUFF0;
+	reg = HAS_PCH_LPT_LP(dev) ? SBI_GEN0 : SBI_DBUFF0;
 	tmp = intel_sbi_read(dev_priv, reg, SBI_ICLK);
 	tmp &= ~SBI_GEN0_CFG_BUFFENABLE_DISABLE;
 	intel_sbi_write(dev_priv, reg, tmp, SBI_ICLK);
@@ -9413,7 +9410,7 @@ void hsw_enable_pc8(struct drm_i915_private *dev_priv)
 
 	DRM_DEBUG_KMS("Enabling package C8+\n");
 
-	if (dev_priv->pch_id == INTEL_PCH_LPT_LP_DEVICE_ID_TYPE) {
+	if (HAS_PCH_LPT_LP(dev)) {
 		val = I915_READ(SOUTH_DSPCLK_GATE_D);
 		val &= ~PCH_LP_PARTITION_LEVEL_DISABLE;
 		I915_WRITE(SOUTH_DSPCLK_GATE_D, val);
@@ -9433,7 +9430,7 @@ void hsw_disable_pc8(struct drm_i915_private *dev_priv)
 	hsw_restore_lcpll(dev_priv);
 	lpt_init_pch_refclk(dev);
 
-	if (dev_priv->pch_id == INTEL_PCH_LPT_LP_DEVICE_ID_TYPE) {
+	if (HAS_PCH_LPT_LP(dev)) {
 		val = I915_READ(SOUTH_DSPCLK_GATE_D);
 		val |= PCH_LP_PARTITION_LEVEL_DISABLE;
 		I915_WRITE(SOUTH_DSPCLK_GATE_D, val);
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index fff0c22..ea49661 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -6588,7 +6588,7 @@ static void lpt_init_clock_gating(struct drm_device *dev)
 	 * TODO: this bit should only be enabled when really needed, then
 	 * disabled when not needed anymore in order to save power.
 	 */
-	if (dev_priv->pch_id == INTEL_PCH_LPT_LP_DEVICE_ID_TYPE)
+	if (HAS_PCH_LPT_LP(dev))
 		I915_WRITE(SOUTH_DSPCLK_GATE_D,
 			   I915_READ(SOUTH_DSPCLK_GATE_D) |
 			   PCH_LP_PARTITION_LEVEL_DISABLE);
@@ -6603,7 +6603,7 @@ static void lpt_suspend_hw(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	if (dev_priv->pch_id == INTEL_PCH_LPT_LP_DEVICE_ID_TYPE) {
+	if (HAS_PCH_LPT_LP(dev)) {
 		uint32_t val = I915_READ(SOUTH_DSPCLK_GATE_D);
 
 		val &= ~PCH_LP_PARTITION_LEVEL_DISABLE;
-- 
cgit v0.10.2


From 195baa0673345c70f04d19e9e18470c9cbf88bcf Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 27 Aug 2015 23:56:00 +0300
Subject: drm/i915: Rename BXT PORTA HPD defines
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The PORTA HPD defines are not BXT specific. They also exist on SPT,
and partially already on LPT:LP.

Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index baff963..3388b64 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -1250,7 +1250,7 @@ static bool bxt_port_hotplug_long_detect(enum port port, u32 val)
 {
 	switch (port) {
 	case PORT_A:
-		return val & BXT_PORTA_HOTPLUG_LONG_DETECT;
+		return val & PORTA_HOTPLUG_LONG_DETECT;
 	case PORT_B:
 		return val & PORTB_HOTPLUG_LONG_DETECT;
 	case PORT_C:
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index e36baca..7eb76b7 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -6033,11 +6033,11 @@ enum skl_disp_power_wells {
 
 /* digital port hotplug */
 #define PCH_PORT_HOTPLUG		0xc4030	/* SHOTPLUG_CTL */
-#define  BXT_PORTA_HOTPLUG_ENABLE	(1 << 28)
-#define  BXT_PORTA_HOTPLUG_STATUS_MASK	(3 << 24)
-#define  BXT_PORTA_HOTPLUG_NO_DETECT	(0 << 24)
-#define  BXT_PORTA_HOTPLUG_SHORT_DETECT	(1 << 24)
-#define  BXT_PORTA_HOTPLUG_LONG_DETECT	(2 << 24)
+#define  PORTA_HOTPLUG_ENABLE		(1 << 28) /* LPT:LP+ & BXT */
+#define  PORTA_HOTPLUG_STATUS_MASK	(3 << 24) /* SPT+ & BXT */
+#define  PORTA_HOTPLUG_NO_DETECT	(0 << 24) /* SPT+ & BXT */
+#define  PORTA_HOTPLUG_SHORT_DETECT	(1 << 24) /* SPT+ & BXT */
+#define  PORTA_HOTPLUG_LONG_DETECT	(2 << 24) /* SPT+ & BXT */
 #define  PORTD_HOTPLUG_ENABLE		(1 << 20)
 #define  PORTD_PULSE_DURATION_2ms	(0 << 18) /* pre-LPT */
 #define  PORTD_PULSE_DURATION_4_5ms	(1 << 18) /* pre-LPT */
-- 
cgit v0.10.2


From 42db67d664af417e946211c7f1f4178c782a582a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 28 Aug 2015 21:26:27 +0300
Subject: drm/i915: Move {pin, long}_mask initialization to caller from
 intel_get_hpd_pins()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Move the 0 initialization of pin_mask and long_mask from
intel_get_hpd_pins() into each caller. This we we can call
intel_get_hpd_pins() multiple times to accumulate more pins from several
sources.

v2: Add a comment explaining the dangers of intel_get_hpd_pins() (Paulo)

Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 3388b64..7129389 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -1292,7 +1292,13 @@ static bool i9xx_port_hotplug_long_detect(enum port port, u32 val)
 	}
 }
 
-/* Get a bit mask of pins that have triggered, and which ones may be long. */
+/*
+ * Get a bit mask of pins that have triggered, and which ones may be long.
+ * This can be called multiple times with the same masks to accumulate
+ * hotplug detection results from several registers.
+ *
+ * Note that the caller is expected to zero out the masks initially.
+ */
 static void intel_get_hpd_pins(u32 *pin_mask, u32 *long_mask,
 			     u32 hotplug_trigger, u32 dig_hotplug_reg,
 			     const u32 hpd[HPD_NUM_PINS],
@@ -1301,9 +1307,6 @@ static void intel_get_hpd_pins(u32 *pin_mask, u32 *long_mask,
 	enum port port;
 	int i;
 
-	*pin_mask = 0;
-	*long_mask = 0;
-
 	for_each_hpd_pin(i) {
 		if ((hpd[i] & hotplug_trigger) == 0)
 			continue;
@@ -1544,7 +1547,7 @@ static void i9xx_hpd_irq_handler(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 hotplug_status = I915_READ(PORT_HOTPLUG_STAT);
-	u32 pin_mask, long_mask;
+	u32 pin_mask = 0, long_mask = 0;
 
 	if (!hotplug_status)
 		return;
@@ -1673,7 +1676,7 @@ static void ibx_irq_handler(struct drm_device *dev, u32 pch_iir)
 	u32 hotplug_trigger = pch_iir & SDE_HOTPLUG_MASK;
 
 	if (hotplug_trigger) {
-		u32 dig_hotplug_reg, pin_mask, long_mask;
+		u32 dig_hotplug_reg, pin_mask = 0, long_mask = 0;
 
 		dig_hotplug_reg = I915_READ(PCH_PORT_HOTPLUG);
 		I915_WRITE(PCH_PORT_HOTPLUG, dig_hotplug_reg);
@@ -1781,7 +1784,7 @@ static void cpt_irq_handler(struct drm_device *dev, u32 pch_iir)
 		hotplug_trigger = pch_iir & SDE_HOTPLUG_MASK_CPT;
 
 	if (hotplug_trigger) {
-		u32 dig_hotplug_reg, pin_mask, long_mask;
+		u32 dig_hotplug_reg, pin_mask = 0, long_mask = 0;
 
 		dig_hotplug_reg = I915_READ(PCH_PORT_HOTPLUG);
 		I915_WRITE(PCH_PORT_HOTPLUG, dig_hotplug_reg);
@@ -2004,7 +2007,7 @@ static void bxt_hpd_handler(struct drm_device *dev, uint32_t iir_status)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 hp_control, hp_trigger;
-	u32 pin_mask, long_mask;
+	u32 pin_mask = 0, long_mask = 0;
 
 	/* Get the status */
 	hp_trigger = iir_status & BXT_DE_PORT_HOTPLUG_MASK;
-- 
cgit v0.10.2


From 6dbf30ce234925e4770f0b52950b601e816b6586 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 27 Aug 2015 23:56:02 +0300
Subject: drm/i915: Introduce spt_irq_handler()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Starting from SPT the only interrupts living in the south are GMBUS and
HPD. What's worse some of the SPT specific new bits conflict with some
other bits on earlier PCH generations. So better not use the
cpt_irq_handler() for SPT+ anymore.

Also kill the hand rolled port E handling with something more
standardish. This also avoids accidentally confusing port B and port E
long pulses since the bits occupy the same positions, just in different
registers.

Also add a comment noting that the short pulse duration bits are
reserved on LPT+. The 2ms value we program is 0, so no issue wrt. the
MBZ in the spec.

v2: Call intel_hpd_irq_handler() only once (Paulo)

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 7129389..755d9ce 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -1262,6 +1262,16 @@ static bool bxt_port_hotplug_long_detect(enum port port, u32 val)
 	}
 }
 
+static bool spt_port_hotplug2_long_detect(enum port port, u32 val)
+{
+	switch (port) {
+	case PORT_E:
+		return val & PORTE_HOTPLUG_LONG_DETECT;
+	default:
+		return false;
+	}
+}
+
 static bool pch_port_hotplug_long_detect(enum port port, u32 val)
 {
 	switch (port) {
@@ -1271,8 +1281,6 @@ static bool pch_port_hotplug_long_detect(enum port port, u32 val)
 		return val & PORTC_HOTPLUG_LONG_DETECT;
 	case PORT_D:
 		return val & PORTD_HOTPLUG_LONG_DETECT;
-	case PORT_E:
-		return val & PORTE_HOTPLUG_LONG_DETECT;
 	default:
 		return false;
 	}
@@ -1776,12 +1784,7 @@ static void cpt_irq_handler(struct drm_device *dev, u32 pch_iir)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int pipe;
-	u32 hotplug_trigger;
-
-	if (HAS_PCH_SPT(dev))
-		hotplug_trigger = pch_iir & SDE_HOTPLUG_MASK_SPT;
-	else
-		hotplug_trigger = pch_iir & SDE_HOTPLUG_MASK_CPT;
+	u32 hotplug_trigger = pch_iir & SDE_HOTPLUG_MASK_CPT;
 
 	if (hotplug_trigger) {
 		u32 dig_hotplug_reg, pin_mask = 0, long_mask = 0;
@@ -1789,22 +1792,10 @@ static void cpt_irq_handler(struct drm_device *dev, u32 pch_iir)
 		dig_hotplug_reg = I915_READ(PCH_PORT_HOTPLUG);
 		I915_WRITE(PCH_PORT_HOTPLUG, dig_hotplug_reg);
 
-		if (HAS_PCH_SPT(dev)) {
-			intel_get_hpd_pins(&pin_mask, &long_mask,
-					   hotplug_trigger,
-					   dig_hotplug_reg, hpd_spt,
-					   pch_port_hotplug_long_detect);
-
-			/* detect PORTE HP event */
-			dig_hotplug_reg = I915_READ(PCH_PORT_HOTPLUG2);
-			if (pch_port_hotplug_long_detect(PORT_E,
-							 dig_hotplug_reg))
-				long_mask |= 1 << HPD_PORT_E;
-		} else
-			intel_get_hpd_pins(&pin_mask, &long_mask,
-					   hotplug_trigger,
-					   dig_hotplug_reg, hpd_cpt,
-					   pch_port_hotplug_long_detect);
+		intel_get_hpd_pins(&pin_mask, &long_mask,
+				   hotplug_trigger,
+				   dig_hotplug_reg, hpd_cpt,
+				   pch_port_hotplug_long_detect);
 
 		intel_hpd_irq_handler(dev, pin_mask, long_mask);
 	}
@@ -1838,6 +1829,43 @@ static void cpt_irq_handler(struct drm_device *dev, u32 pch_iir)
 		cpt_serr_int_handler(dev);
 }
 
+static void spt_irq_handler(struct drm_device *dev, u32 pch_iir)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	u32 hotplug_trigger = pch_iir & SDE_HOTPLUG_MASK_SPT &
+		~SDE_PORTE_HOTPLUG_SPT;
+	u32 hotplug2_trigger = pch_iir & SDE_PORTE_HOTPLUG_SPT;
+	u32 pin_mask = 0, long_mask = 0;
+
+	if (hotplug_trigger) {
+		u32 dig_hotplug_reg;
+
+		dig_hotplug_reg = I915_READ(PCH_PORT_HOTPLUG);
+		I915_WRITE(PCH_PORT_HOTPLUG, dig_hotplug_reg);
+
+		intel_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger,
+				   dig_hotplug_reg, hpd_spt,
+				   pch_port_hotplug_long_detect);
+	}
+
+	if (hotplug2_trigger) {
+		u32 dig_hotplug_reg;
+
+		dig_hotplug_reg = I915_READ(PCH_PORT_HOTPLUG2);
+		I915_WRITE(PCH_PORT_HOTPLUG2, dig_hotplug_reg);
+
+		intel_get_hpd_pins(&pin_mask, &long_mask, hotplug2_trigger,
+				   dig_hotplug_reg, hpd_spt,
+				   spt_port_hotplug2_long_detect);
+	}
+
+	if (pin_mask)
+		intel_hpd_irq_handler(dev, pin_mask, long_mask);
+
+	if (pch_iir & SDE_GMBUS_CPT)
+		gmbus_irq_handler(dev);
+}
+
 static void ilk_display_irq_handler(struct drm_device *dev, u32 de_iir)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -2156,7 +2184,11 @@ static irqreturn_t gen8_irq_handler(int irq, void *arg)
 		if (pch_iir) {
 			I915_WRITE(SDEIIR, pch_iir);
 			ret = IRQ_HANDLED;
-			cpt_irq_handler(dev, pch_iir);
+
+			if (HAS_PCH_SPT(dev_priv))
+				spt_irq_handler(dev, pch_iir);
+			else
+				cpt_irq_handler(dev, pch_iir);
 		} else
 			DRM_ERROR("The master control interrupt lied (SDE)!\n");
 
@@ -3038,9 +3070,6 @@ static void ibx_hpd_irq_setup(struct drm_device *dev)
 	if (HAS_PCH_IBX(dev)) {
 		hotplug_irqs = SDE_HOTPLUG_MASK;
 		enabled_irqs = intel_hpd_enabled_irqs(dev, hpd_ibx);
-	} else if (HAS_PCH_SPT(dev)) {
-		hotplug_irqs = SDE_HOTPLUG_MASK_SPT;
-		enabled_irqs = intel_hpd_enabled_irqs(dev, hpd_spt);
 	} else {
 		hotplug_irqs = SDE_HOTPLUG_MASK_CPT;
 		enabled_irqs = intel_hpd_enabled_irqs(dev, hpd_cpt);
@@ -3050,9 +3079,8 @@ static void ibx_hpd_irq_setup(struct drm_device *dev)
 
 	/*
 	 * Enable digital hotplug on the PCH, and configure the DP short pulse
-	 * duration to 2ms (which is the minimum in the Display Port spec)
-	 *
-	 * This register is the same on all known PCH chips.
+	 * duration to 2ms (which is the minimum in the Display Port spec).
+	 * The pulse duration bits are reserved on LPT+.
 	 */
 	hotplug = I915_READ(PCH_PORT_HOTPLUG);
 	hotplug &= ~(PORTD_PULSE_DURATION_MASK|PORTC_PULSE_DURATION_MASK|PORTB_PULSE_DURATION_MASK);
@@ -3060,13 +3088,27 @@ static void ibx_hpd_irq_setup(struct drm_device *dev)
 	hotplug |= PORTC_HOTPLUG_ENABLE | PORTC_PULSE_DURATION_2ms;
 	hotplug |= PORTB_HOTPLUG_ENABLE | PORTB_PULSE_DURATION_2ms;
 	I915_WRITE(PCH_PORT_HOTPLUG, hotplug);
+}
 
-	/* enable SPT PORTE hot plug */
-	if (HAS_PCH_SPT(dev)) {
-		hotplug = I915_READ(PCH_PORT_HOTPLUG2);
-		hotplug |= PORTE_HOTPLUG_ENABLE;
-		I915_WRITE(PCH_PORT_HOTPLUG2, hotplug);
-	}
+static void spt_hpd_irq_setup(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	u32 hotplug_irqs, hotplug, enabled_irqs;
+
+	hotplug_irqs = SDE_HOTPLUG_MASK_SPT;
+	enabled_irqs = intel_hpd_enabled_irqs(dev, hpd_spt);
+
+	ibx_display_interrupt_update(dev_priv, hotplug_irqs, enabled_irqs);
+
+	/* Enable digital hotplug on the PCH */
+	hotplug = I915_READ(PCH_PORT_HOTPLUG);
+	hotplug |= PORTD_HOTPLUG_ENABLE | PORTC_HOTPLUG_ENABLE |
+		PORTB_HOTPLUG_ENABLE;
+	I915_WRITE(PCH_PORT_HOTPLUG, hotplug);
+
+	hotplug = I915_READ(PCH_PORT_HOTPLUG2);
+	hotplug |= PORTE_HOTPLUG_ENABLE;
+	I915_WRITE(PCH_PORT_HOTPLUG2, hotplug);
 }
 
 static void bxt_hpd_irq_setup(struct drm_device *dev)
@@ -4171,10 +4213,12 @@ void intel_irq_init(struct drm_i915_private *dev_priv)
 		dev->driver->irq_uninstall = gen8_irq_uninstall;
 		dev->driver->enable_vblank = gen8_enable_vblank;
 		dev->driver->disable_vblank = gen8_disable_vblank;
-		if (HAS_PCH_SPLIT(dev))
-			dev_priv->display.hpd_irq_setup = ibx_hpd_irq_setup;
-		else
+		if (IS_BROXTON(dev))
 			dev_priv->display.hpd_irq_setup = bxt_hpd_irq_setup;
+		else if (HAS_PCH_SPT(dev))
+			dev_priv->display.hpd_irq_setup = spt_hpd_irq_setup;
+		else
+			dev_priv->display.hpd_irq_setup = ibx_hpd_irq_setup;
 	} else if (HAS_PCH_SPLIT(dev)) {
 		dev->driver->irq_handler = ironlake_irq_handler;
 		dev->driver->irq_preinstall = ironlake_irq_reset;
-- 
cgit v0.10.2


From e4ce95aa661e411d7da554eb8e6c7f81504acae6 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 27 Aug 2015 23:56:03 +0300
Subject: drm/i915: Add port A HPD support for ILK/SNB
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

ILK/SNB support port A HPD. While HPD is optional on eDP let's at least
try to wite it up so that we might notice if the link has issues.

The eDP spec suggests that if HPD is not wired up, one should poll the
link status instead. We don't even do that currently.

v2: 0 initialize pin_mask/long_mask due to intel_get_hpd_pins() changes

Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 755d9ce..57d23f0 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -45,6 +45,10 @@
  * and related files, but that will be described in separate chapters.
  */
 
+static const u32 hpd_ilk[HPD_NUM_PINS] = {
+	[HPD_PORT_A] = DE_DP_A_HOTPLUG,
+};
+
 static const u32 hpd_ibx[HPD_NUM_PINS] = {
 	[HPD_CRT] = SDE_CRT_HOTPLUG,
 	[HPD_SDVO_B] = SDE_SDVOB_HOTPLUG,
@@ -1272,6 +1276,16 @@ static bool spt_port_hotplug2_long_detect(enum port port, u32 val)
 	}
 }
 
+static bool ilk_port_hotplug_long_detect(enum port port, u32 val)
+{
+	switch (port) {
+	case PORT_A:
+		return val & DIGITAL_PORTA_HOTPLUG_LONG_DETECT;
+	default:
+		return false;
+	}
+}
+
 static bool pch_port_hotplug_long_detect(enum port port, u32 val)
 {
 	switch (port) {
@@ -1870,6 +1884,19 @@ static void ilk_display_irq_handler(struct drm_device *dev, u32 de_iir)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	enum pipe pipe;
+	u32 hotplug_trigger = de_iir & DE_DP_A_HOTPLUG;
+
+	if (hotplug_trigger) {
+		u32 dig_hotplug_reg, pin_mask = 0, long_mask = 0;
+
+		dig_hotplug_reg = I915_READ(DIGITAL_PORT_HOTPLUG_CNTRL);
+		I915_WRITE(DIGITAL_PORT_HOTPLUG_CNTRL, dig_hotplug_reg);
+
+		intel_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger,
+				   dig_hotplug_reg, hpd_ilk,
+				   ilk_port_hotplug_long_detect);
+		intel_hpd_irq_handler(dev, pin_mask, long_mask);
+	}
 
 	if (de_iir & DE_AUX_CHANNEL_A)
 		dp_aux_irq_handler(dev);
@@ -3111,6 +3138,28 @@ static void spt_hpd_irq_setup(struct drm_device *dev)
 	I915_WRITE(PCH_PORT_HOTPLUG2, hotplug);
 }
 
+static void ilk_hpd_irq_setup(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	u32 hotplug_irqs, hotplug, enabled_irqs;
+
+	hotplug_irqs = DE_DP_A_HOTPLUG;
+	enabled_irqs = intel_hpd_enabled_irqs(dev, hpd_ilk);
+
+	ilk_update_display_irq(dev_priv, hotplug_irqs, enabled_irqs);
+
+	/*
+	 * Enable digital hotplug on the CPU, and configure the DP short pulse
+	 * duration to 2ms (which is the minimum in the Display Port spec)
+	 */
+	hotplug = I915_READ(DIGITAL_PORT_HOTPLUG_CNTRL);
+	hotplug &= ~DIGITAL_PORTA_PULSE_DURATION_MASK;
+	hotplug |= DIGITAL_PORTA_HOTPLUG_ENABLE | DIGITAL_PORTA_PULSE_DURATION_2ms;
+	I915_WRITE(DIGITAL_PORT_HOTPLUG_CNTRL, hotplug);
+
+	ibx_hpd_irq_setup(dev);
+}
+
 static void bxt_hpd_irq_setup(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -3209,8 +3258,9 @@ static int ironlake_irq_postinstall(struct drm_device *dev)
 				DE_AUX_CHANNEL_A |
 				DE_PIPEB_CRC_DONE | DE_PIPEA_CRC_DONE |
 				DE_POISON);
-		extra_mask = DE_PIPEA_VBLANK | DE_PIPEB_VBLANK | DE_PCU_EVENT |
-				DE_PIPEB_FIFO_UNDERRUN | DE_PIPEA_FIFO_UNDERRUN;
+		extra_mask = (DE_PIPEA_VBLANK | DE_PIPEB_VBLANK | DE_PCU_EVENT |
+			      DE_PIPEB_FIFO_UNDERRUN | DE_PIPEA_FIFO_UNDERRUN |
+			      DE_DP_A_HOTPLUG);
 	}
 
 	dev_priv->irq_mask = ~display_mask;
@@ -4226,7 +4276,10 @@ void intel_irq_init(struct drm_i915_private *dev_priv)
 		dev->driver->irq_uninstall = ironlake_irq_uninstall;
 		dev->driver->enable_vblank = ironlake_enable_vblank;
 		dev->driver->disable_vblank = ironlake_disable_vblank;
-		dev_priv->display.hpd_irq_setup = ibx_hpd_irq_setup;
+		if (INTEL_INFO(dev)->gen >= 7)
+			dev_priv->display.hpd_irq_setup = ibx_hpd_irq_setup;
+		else
+			dev_priv->display.hpd_irq_setup = ilk_hpd_irq_setup;
 	} else {
 		if (INTEL_INFO(dev_priv)->gen == 2) {
 			dev->driver->irq_preinstall = i8xx_irq_preinstall;
-- 
cgit v0.10.2


From 23bb4cb5122c2c6379be38aaa5c0fd6786ae6c3a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 27 Aug 2015 23:56:04 +0300
Subject: drm/i915: Add port A HPD support for IVB/HSW
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

As with ILK/SNB wire up the port A HPD on IVB/HSW.

This might be more important on HSW with PSR. BSpec tells us that if the
automagic link training performed by the hardware fails for some reason,
we're going to get a short HPD and are supposed to re-train the link
manyally.

v2: 0 initialize pin_mask/long_mask due to intel_get_hpd_pins() changes
    Add a comment about the pulse duration bits being reserved on HSW+
    like we have for LPT+ in ibx_hpd_irq_setup()

Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 57d23f0..8864aaf 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -49,6 +49,10 @@ static const u32 hpd_ilk[HPD_NUM_PINS] = {
 	[HPD_PORT_A] = DE_DP_A_HOTPLUG,
 };
 
+static const u32 hpd_ivb[HPD_NUM_PINS] = {
+	[HPD_PORT_A] = DE_DP_A_HOTPLUG_IVB,
+};
+
 static const u32 hpd_ibx[HPD_NUM_PINS] = {
 	[HPD_CRT] = SDE_CRT_HOTPLUG,
 	[HPD_SDVO_B] = SDE_SDVOB_HOTPLUG,
@@ -1946,6 +1950,19 @@ static void ivb_display_irq_handler(struct drm_device *dev, u32 de_iir)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	enum pipe pipe;
+	u32 hotplug_trigger = de_iir & DE_DP_A_HOTPLUG_IVB;
+
+	if (hotplug_trigger) {
+		u32 dig_hotplug_reg, pin_mask = 0, long_mask = 0;
+
+		dig_hotplug_reg = I915_READ(DIGITAL_PORT_HOTPLUG_CNTRL);
+		I915_WRITE(DIGITAL_PORT_HOTPLUG_CNTRL, dig_hotplug_reg);
+
+		intel_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger,
+				   dig_hotplug_reg, hpd_ivb,
+				   ilk_port_hotplug_long_detect);
+		intel_hpd_irq_handler(dev, pin_mask, long_mask);
+	}
 
 	if (de_iir & DE_ERR_INT_IVB)
 		ivb_err_int_handler(dev);
@@ -3143,14 +3160,20 @@ static void ilk_hpd_irq_setup(struct drm_device *dev)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 hotplug_irqs, hotplug, enabled_irqs;
 
-	hotplug_irqs = DE_DP_A_HOTPLUG;
-	enabled_irqs = intel_hpd_enabled_irqs(dev, hpd_ilk);
+	if (INTEL_INFO(dev)->gen >= 7) {
+		hotplug_irqs = DE_DP_A_HOTPLUG_IVB;
+		enabled_irqs = intel_hpd_enabled_irqs(dev, hpd_ivb);
+	} else {
+		hotplug_irqs = DE_DP_A_HOTPLUG;
+		enabled_irqs = intel_hpd_enabled_irqs(dev, hpd_ilk);
+	}
 
 	ilk_update_display_irq(dev_priv, hotplug_irqs, enabled_irqs);
 
 	/*
 	 * Enable digital hotplug on the CPU, and configure the DP short pulse
 	 * duration to 2ms (which is the minimum in the Display Port spec)
+	 * The pulse duration bits are reserved on HSW+.
 	 */
 	hotplug = I915_READ(DIGITAL_PORT_HOTPLUG_CNTRL);
 	hotplug &= ~DIGITAL_PORTA_PULSE_DURATION_MASK;
@@ -3251,7 +3274,8 @@ static int ironlake_irq_postinstall(struct drm_device *dev)
 				DE_PLANEB_FLIP_DONE_IVB |
 				DE_PLANEA_FLIP_DONE_IVB | DE_AUX_CHANNEL_A_IVB);
 		extra_mask = (DE_PIPEC_VBLANK_IVB | DE_PIPEB_VBLANK_IVB |
-			      DE_PIPEA_VBLANK_IVB | DE_ERR_INT_IVB);
+			      DE_PIPEA_VBLANK_IVB | DE_ERR_INT_IVB |
+			      DE_DP_A_HOTPLUG_IVB);
 	} else {
 		display_mask = (DE_MASTER_IRQ_CONTROL | DE_GSE | DE_PCH_EVENT |
 				DE_PLANEA_FLIP_DONE | DE_PLANEB_FLIP_DONE |
@@ -4276,10 +4300,7 @@ void intel_irq_init(struct drm_i915_private *dev_priv)
 		dev->driver->irq_uninstall = ironlake_irq_uninstall;
 		dev->driver->enable_vblank = ironlake_enable_vblank;
 		dev->driver->disable_vblank = ironlake_disable_vblank;
-		if (INTEL_INFO(dev)->gen >= 7)
-			dev_priv->display.hpd_irq_setup = ibx_hpd_irq_setup;
-		else
-			dev_priv->display.hpd_irq_setup = ilk_hpd_irq_setup;
+		dev_priv->display.hpd_irq_setup = ilk_hpd_irq_setup;
 	} else {
 		if (INTEL_INFO(dev_priv)->gen == 2) {
 			dev->driver->irq_preinstall = i8xx_irq_preinstall;
-- 
cgit v0.10.2


From 0b2eb33e5e2b4aaeaa786d551a1b5f08904a9bfc Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 27 Aug 2015 23:56:05 +0300
Subject: drm/i915: LPT:LP needs port A HPD enabled in both north and south
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

If the CPU and PCH are on the same package we must enabled the port A
HPD also in the south hotplug register. To identify the package type
we simply look at the PCH type: LPT-H means separate package, and
LPT-LP means multi chip package (MCP).

v2: Add comment and pimp commit message

Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 8864aaf..0aa87d0 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -3131,6 +3131,12 @@ static void ibx_hpd_irq_setup(struct drm_device *dev)
 	hotplug |= PORTD_HOTPLUG_ENABLE | PORTD_PULSE_DURATION_2ms;
 	hotplug |= PORTC_HOTPLUG_ENABLE | PORTC_PULSE_DURATION_2ms;
 	hotplug |= PORTB_HOTPLUG_ENABLE | PORTB_PULSE_DURATION_2ms;
+	/*
+	 * When CPU and PCH are on the same package, port A
+	 * HPD must be enabled in both north and south.
+	 */
+	if (HAS_PCH_LPT_LP(dev))
+		hotplug |= PORTA_HOTPLUG_ENABLE;
 	I915_WRITE(PCH_PORT_HOTPLUG, hotplug);
 }
 
-- 
cgit v0.10.2


From 3a3b3c7da37b35955c7c6a36a1f059d02f17eafb Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 27 Aug 2015 23:56:06 +0300
Subject: drm/i915: Add port A HPD support for BDW
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Wire up the port A HPD for BDW. Compared to earlier platforms the
interrupt setup is a bit different, but basically everything else
looks the same.

v2: 0 initialize pin_mask/long_mask due to intel_get_hpd_pins() changes
    Check for BDW before processing the HPD to not break BXT
    Set found=true when processing port A HPD
    Sort out the mess I made of the irq setup in v1
    Warn about bad irq mask vs. enable bits in bdw_update_port_irq()  (Paulo)

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 0aa87d0..03c79f6 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -53,6 +53,10 @@ static const u32 hpd_ivb[HPD_NUM_PINS] = {
 	[HPD_PORT_A] = DE_DP_A_HOTPLUG_IVB,
 };
 
+static const u32 hpd_bdw[HPD_NUM_PINS] = {
+	[HPD_PORT_A] = GEN8_PORT_DP_A_HOTPLUG,
+};
+
 static const u32 hpd_ibx[HPD_NUM_PINS] = {
 	[HPD_CRT] = SDE_CRT_HOTPLUG,
 	[HPD_SDVO_B] = SDE_SDVOB_HOTPLUG,
@@ -371,6 +375,38 @@ void gen6_disable_rps_interrupts(struct drm_device *dev)
 }
 
 /**
+  * bdw_update_port_irq - update DE port interrupt
+  * @dev_priv: driver private
+  * @interrupt_mask: mask of interrupt bits to update
+  * @enabled_irq_mask: mask of interrupt bits to enable
+  */
+static void bdw_update_port_irq(struct drm_i915_private *dev_priv,
+				uint32_t interrupt_mask,
+				uint32_t enabled_irq_mask)
+{
+	uint32_t new_val;
+	uint32_t old_val;
+
+	assert_spin_locked(&dev_priv->irq_lock);
+
+	WARN_ON(enabled_irq_mask & ~interrupt_mask);
+
+	if (WARN_ON(!intel_irqs_enabled(dev_priv)))
+		return;
+
+	old_val = I915_READ(GEN8_DE_PORT_IMR);
+
+	new_val = old_val;
+	new_val &= ~interrupt_mask;
+	new_val |= (~enabled_irq_mask & interrupt_mask);
+
+	if (new_val != old_val) {
+		I915_WRITE(GEN8_DE_PORT_IMR, new_val);
+		POSTING_READ(GEN8_DE_PORT_IMR);
+	}
+}
+
+/**
  * ibx_display_interrupt_update - update SDEIMR
  * @dev_priv: driver private
  * @interrupt_mask: mask of interrupt bits to update
@@ -2145,10 +2181,24 @@ static irqreturn_t gen8_irq_handler(int irq, void *arg)
 		tmp = I915_READ(GEN8_DE_PORT_IIR);
 		if (tmp) {
 			bool found = false;
+			u32 hotplug_trigger = tmp & GEN8_PORT_DP_A_HOTPLUG;
 
 			I915_WRITE(GEN8_DE_PORT_IIR, tmp);
 			ret = IRQ_HANDLED;
 
+			if (IS_BROADWELL(dev) && hotplug_trigger) {
+				u32 dig_hotplug_reg, pin_mask = 0, long_mask = 0;
+
+				dig_hotplug_reg = I915_READ(DIGITAL_PORT_HOTPLUG_CNTRL);
+				I915_WRITE(DIGITAL_PORT_HOTPLUG_CNTRL, dig_hotplug_reg);
+
+				intel_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger,
+						   dig_hotplug_reg, hpd_bdw,
+						   ilk_port_hotplug_long_detect);
+				intel_hpd_irq_handler(dev, pin_mask, long_mask);
+				found = true;
+			}
+
 			if (tmp & aux_mask) {
 				dp_aux_irq_handler(dev);
 				found = true;
@@ -3166,15 +3216,22 @@ static void ilk_hpd_irq_setup(struct drm_device *dev)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 hotplug_irqs, hotplug, enabled_irqs;
 
-	if (INTEL_INFO(dev)->gen >= 7) {
+	if (INTEL_INFO(dev)->gen >= 8) {
+		hotplug_irqs = GEN8_PORT_DP_A_HOTPLUG;
+		enabled_irqs = intel_hpd_enabled_irqs(dev, hpd_bdw);
+
+		bdw_update_port_irq(dev_priv, hotplug_irqs, enabled_irqs);
+	} else if (INTEL_INFO(dev)->gen >= 7) {
 		hotplug_irqs = DE_DP_A_HOTPLUG_IVB;
 		enabled_irqs = intel_hpd_enabled_irqs(dev, hpd_ivb);
+
+		ilk_update_display_irq(dev_priv, hotplug_irqs, enabled_irqs);
 	} else {
 		hotplug_irqs = DE_DP_A_HOTPLUG;
 		enabled_irqs = intel_hpd_enabled_irqs(dev, hpd_ilk);
-	}
 
-	ilk_update_display_irq(dev_priv, hotplug_irqs, enabled_irqs);
+		ilk_update_display_irq(dev_priv, hotplug_irqs, enabled_irqs);
+	}
 
 	/*
 	 * Enable digital hotplug on the CPU, and configure the DP short pulse
@@ -3486,24 +3543,29 @@ static void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv)
 {
 	uint32_t de_pipe_masked = GEN8_PIPE_CDCLK_CRC_DONE;
 	uint32_t de_pipe_enables;
-	int pipe;
-	u32 de_port_en = GEN8_AUX_CHANNEL_A;
+	u32 de_port_masked = GEN8_AUX_CHANNEL_A;
+	u32 de_port_enables;
+	enum pipe pipe;
 
 	if (IS_GEN9(dev_priv)) {
 		de_pipe_masked |= GEN9_PIPE_PLANE1_FLIP_DONE |
 				  GEN9_DE_PIPE_IRQ_FAULT_ERRORS;
-		de_port_en |= GEN9_AUX_CHANNEL_B | GEN9_AUX_CHANNEL_C |
-			GEN9_AUX_CHANNEL_D;
-
+		de_port_masked |= GEN9_AUX_CHANNEL_B | GEN9_AUX_CHANNEL_C |
+				  GEN9_AUX_CHANNEL_D;
 		if (IS_BROXTON(dev_priv))
-			de_port_en |= BXT_DE_PORT_GMBUS;
-	} else
+			de_port_masked |= BXT_DE_PORT_GMBUS;
+	} else {
 		de_pipe_masked |= GEN8_PIPE_PRIMARY_FLIP_DONE |
 				  GEN8_DE_PIPE_IRQ_FAULT_ERRORS;
+	}
 
 	de_pipe_enables = de_pipe_masked | GEN8_PIPE_VBLANK |
 					   GEN8_PIPE_FIFO_UNDERRUN;
 
+	de_port_enables = de_port_masked;
+	if (IS_BROADWELL(dev_priv))
+		de_port_enables |= GEN8_PORT_DP_A_HOTPLUG;
+
 	dev_priv->de_irq_mask[PIPE_A] = ~de_pipe_masked;
 	dev_priv->de_irq_mask[PIPE_B] = ~de_pipe_masked;
 	dev_priv->de_irq_mask[PIPE_C] = ~de_pipe_masked;
@@ -3515,7 +3577,7 @@ static void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv)
 					  dev_priv->de_irq_mask[pipe],
 					  de_pipe_enables);
 
-	GEN5_IRQ_INIT(GEN8_DE_PORT_, ~de_port_en, de_port_en);
+	GEN5_IRQ_INIT(GEN8_DE_PORT_, ~de_port_masked, de_port_enables);
 }
 
 static int gen8_irq_postinstall(struct drm_device *dev)
@@ -4298,7 +4360,7 @@ void intel_irq_init(struct drm_i915_private *dev_priv)
 		else if (HAS_PCH_SPT(dev))
 			dev_priv->display.hpd_irq_setup = spt_hpd_irq_setup;
 		else
-			dev_priv->display.hpd_irq_setup = ibx_hpd_irq_setup;
+			dev_priv->display.hpd_irq_setup = ilk_hpd_irq_setup;
 	} else if (HAS_PCH_SPLIT(dev)) {
 		dev->driver->irq_handler = ironlake_irq_handler;
 		dev->driver->irq_preinstall = ironlake_irq_reset;
-- 
cgit v0.10.2


From 74c0b395fce4c807a49c404fb59462b22069b001 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 27 Aug 2015 23:56:07 +0300
Subject: drm/i915: Add port A HPD support for SPT
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

On SKL the port A HPD has moved to the PCH. Hook it up.

Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 03c79f6..c3edcce 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -74,6 +74,7 @@ static const u32 hpd_cpt[HPD_NUM_PINS] = {
 };
 
 static const u32 hpd_spt[HPD_NUM_PINS] = {
+	[HPD_PORT_A] = SDE_PORTA_HOTPLUG_SPT,
 	[HPD_PORT_B] = SDE_PORTB_HOTPLUG_CPT,
 	[HPD_PORT_C] = SDE_PORTC_HOTPLUG_CPT,
 	[HPD_PORT_D] = SDE_PORTD_HOTPLUG_CPT,
@@ -1316,6 +1317,22 @@ static bool spt_port_hotplug2_long_detect(enum port port, u32 val)
 	}
 }
 
+static bool spt_port_hotplug_long_detect(enum port port, u32 val)
+{
+	switch (port) {
+	case PORT_A:
+		return val & PORTA_HOTPLUG_LONG_DETECT;
+	case PORT_B:
+		return val & PORTB_HOTPLUG_LONG_DETECT;
+	case PORT_C:
+		return val & PORTC_HOTPLUG_LONG_DETECT;
+	case PORT_D:
+		return val & PORTD_HOTPLUG_LONG_DETECT;
+	default:
+		return false;
+	}
+}
+
 static bool ilk_port_hotplug_long_detect(enum port port, u32 val)
 {
 	switch (port) {
@@ -1899,7 +1916,7 @@ static void spt_irq_handler(struct drm_device *dev, u32 pch_iir)
 
 		intel_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger,
 				   dig_hotplug_reg, hpd_spt,
-				   pch_port_hotplug_long_detect);
+				   spt_port_hotplug_long_detect);
 	}
 
 	if (hotplug2_trigger) {
@@ -3203,7 +3220,7 @@ static void spt_hpd_irq_setup(struct drm_device *dev)
 	/* Enable digital hotplug on the PCH */
 	hotplug = I915_READ(PCH_PORT_HOTPLUG);
 	hotplug |= PORTD_HOTPLUG_ENABLE | PORTC_HOTPLUG_ENABLE |
-		PORTB_HOTPLUG_ENABLE;
+		PORTB_HOTPLUG_ENABLE | PORTA_HOTPLUG_ENABLE;
 	I915_WRITE(PCH_PORT_HOTPLUG, hotplug);
 
 	hotplug = I915_READ(PCH_PORT_HOTPLUG2);
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 7eb76b7..689b428 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -5984,6 +5984,7 @@ enum skl_disp_power_wells {
 #define SDE_AUXB_CPT		(1 << 25)
 #define SDE_AUX_MASK_CPT	(7 << 25)
 #define SDE_PORTE_HOTPLUG_SPT	(1 << 25)
+#define SDE_PORTA_HOTPLUG_SPT	(1 << 24)
 #define SDE_PORTD_HOTPLUG_CPT	(1 << 23)
 #define SDE_PORTC_HOTPLUG_CPT	(1 << 22)
 #define SDE_PORTB_HOTPLUG_CPT	(1 << 21)
@@ -5997,7 +5998,8 @@ enum skl_disp_power_wells {
 #define SDE_HOTPLUG_MASK_SPT	(SDE_PORTE_HOTPLUG_SPT |	\
 				 SDE_PORTD_HOTPLUG_CPT |	\
 				 SDE_PORTC_HOTPLUG_CPT |	\
-				 SDE_PORTB_HOTPLUG_CPT)
+				 SDE_PORTB_HOTPLUG_CPT |	\
+				 SDE_PORTA_HOTPLUG_SPT)
 #define SDE_GMBUS_CPT		(1 << 17)
 #define SDE_ERROR_CPT		(1 << 16)
 #define SDE_AUDIO_CP_REQ_C_CPT	(1 << 10)
-- 
cgit v0.10.2


From 08d8a232aa5e145f995a98ac7f3de55e2fe6832b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 27 Aug 2015 23:56:08 +0300
Subject: drm/i915: Reinitialize HPD after runtime D3
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Runtime suspends disabled all interrupts, so in order to get them back
fully we need to also do the HPD irq setup on runtime resume. Except
on VLV/CHV where the display interrupt initialization is part of the
display power well powerup.

Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 4737d15..bdec64c 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -1555,6 +1555,15 @@ static int intel_runtime_resume(struct device *device)
 	gen6_update_ring_freq(dev);
 
 	intel_runtime_pm_enable_interrupts(dev_priv);
+
+	/*
+	 * On VLV/CHV display interrupts are part of the display
+	 * power well, so hpd is reinitialized from there. For
+	 * everyone else do it here.
+	 */
+	if (!IS_VALLEYVIEW(dev_priv))
+		intel_hpd_init(dev_priv);
+
 	intel_enable_gt_powersave(dev);
 
 	if (ret)
-- 
cgit v0.10.2


From cebd87a07748b649f22356efd56f478248aec300 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 27 Aug 2015 23:56:09 +0300
Subject: drm/i915: Rewrite bxt_hpd_handler() to look like everyone else
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

bxt_hpd_handler() looks different to everyone else for no good reason.
Rewrite it to use the standard variable namees etc.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index c3edcce..f0dedff 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -2128,27 +2128,17 @@ static irqreturn_t ironlake_irq_handler(int irq, void *arg)
 	return ret;
 }
 
-static void bxt_hpd_handler(struct drm_device *dev, uint32_t iir_status)
+static void bxt_hpd_irq_handler(struct drm_device *dev, u32 hotplug_trigger)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 hp_control, hp_trigger;
-	u32 pin_mask = 0, long_mask = 0;
-
-	/* Get the status */
-	hp_trigger = iir_status & BXT_DE_PORT_HOTPLUG_MASK;
-	hp_control = I915_READ(BXT_HOTPLUG_CTL);
-
-	/* Hotplug not enabled ? */
-	if (!(hp_control & BXT_HOTPLUG_CTL_MASK)) {
-		DRM_ERROR("Interrupt when HPD disabled\n");
-		return;
-	}
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	u32 dig_hotplug_reg, pin_mask = 0, long_mask = 0;
 
-	/* Clear sticky bits in hpd status */
-	I915_WRITE(BXT_HOTPLUG_CTL, hp_control);
+	dig_hotplug_reg = I915_READ(BXT_HOTPLUG_CTL);
+	I915_WRITE(BXT_HOTPLUG_CTL, dig_hotplug_reg);
 
-	intel_get_hpd_pins(&pin_mask, &long_mask, hp_trigger, hp_control,
-			   hpd_bxt, bxt_port_hotplug_long_detect);
+	intel_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger,
+			   dig_hotplug_reg, hpd_bxt,
+			   bxt_port_hotplug_long_detect);
 	intel_hpd_irq_handler(dev, pin_mask, long_mask);
 }
 
@@ -2198,7 +2188,12 @@ static irqreturn_t gen8_irq_handler(int irq, void *arg)
 		tmp = I915_READ(GEN8_DE_PORT_IIR);
 		if (tmp) {
 			bool found = false;
-			u32 hotplug_trigger = tmp & GEN8_PORT_DP_A_HOTPLUG;
+			u32 hotplug_trigger = 0;
+
+			if (IS_BROXTON(dev_priv))
+				hotplug_trigger = tmp & BXT_DE_PORT_HOTPLUG_MASK;
+			else if (IS_BROADWELL(dev_priv))
+				hotplug_trigger = tmp & GEN8_PORT_DP_A_HOTPLUG;
 
 			I915_WRITE(GEN8_DE_PORT_IIR, tmp);
 			ret = IRQ_HANDLED;
@@ -2221,8 +2216,8 @@ static irqreturn_t gen8_irq_handler(int irq, void *arg)
 				found = true;
 			}
 
-			if (IS_BROXTON(dev) && tmp & BXT_DE_PORT_HOTPLUG_MASK) {
-				bxt_hpd_handler(dev, tmp);
+			if (IS_BROXTON(dev) && hotplug_trigger) {
+				bxt_hpd_irq_handler(dev, hotplug_trigger);
 				found = true;
 			}
 
-- 
cgit v0.10.2


From 40e5641086f6f64acb13ff6893cf82c4fb30a6d7 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 27 Aug 2015 23:56:10 +0300
Subject: drm/i915: Refactor the hpd irq handling functions
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

A lot of the hpd irq handling is duplicated code, so refactor it a bit
by observing that in several places the only difference is the hpd[]
array. So pull the code to a few functions and pass in the hpd[] array
from the caller. Another option would be to determine the correct array
to use within the functions themselves, but somehow passing it in felt
nicer.

Further code reduction could be achieved by passing in the hotplug
register offset, and the long pulse detection function pointer. But that
didn't feel as good for some reason, so I left it at the middle ground.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index f0dedff..d7a37a5 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -1748,23 +1748,30 @@ static irqreturn_t cherryview_irq_handler(int irq, void *arg)
 	return ret;
 }
 
+static void ibx_hpd_irq_handler(struct drm_device *dev, u32 hotplug_trigger,
+				const u32 hpd[HPD_NUM_PINS])
+{
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	u32 dig_hotplug_reg, pin_mask = 0, long_mask = 0;
+
+	dig_hotplug_reg = I915_READ(PCH_PORT_HOTPLUG);
+	I915_WRITE(PCH_PORT_HOTPLUG, dig_hotplug_reg);
+
+	intel_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger,
+			   dig_hotplug_reg, hpd,
+			   pch_port_hotplug_long_detect);
+
+	intel_hpd_irq_handler(dev, pin_mask, long_mask);
+}
+
 static void ibx_irq_handler(struct drm_device *dev, u32 pch_iir)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int pipe;
 	u32 hotplug_trigger = pch_iir & SDE_HOTPLUG_MASK;
 
-	if (hotplug_trigger) {
-		u32 dig_hotplug_reg, pin_mask = 0, long_mask = 0;
-
-		dig_hotplug_reg = I915_READ(PCH_PORT_HOTPLUG);
-		I915_WRITE(PCH_PORT_HOTPLUG, dig_hotplug_reg);
-
-		intel_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger,
-				   dig_hotplug_reg, hpd_ibx,
-				   pch_port_hotplug_long_detect);
-		intel_hpd_irq_handler(dev, pin_mask, long_mask);
-	}
+	if (hotplug_trigger)
+		ibx_hpd_irq_handler(dev, hotplug_trigger, hpd_ibx);
 
 	if (pch_iir & SDE_AUDIO_POWER_MASK) {
 		int port = ffs((pch_iir & SDE_AUDIO_POWER_MASK) >>
@@ -1857,19 +1864,8 @@ static void cpt_irq_handler(struct drm_device *dev, u32 pch_iir)
 	int pipe;
 	u32 hotplug_trigger = pch_iir & SDE_HOTPLUG_MASK_CPT;
 
-	if (hotplug_trigger) {
-		u32 dig_hotplug_reg, pin_mask = 0, long_mask = 0;
-
-		dig_hotplug_reg = I915_READ(PCH_PORT_HOTPLUG);
-		I915_WRITE(PCH_PORT_HOTPLUG, dig_hotplug_reg);
-
-		intel_get_hpd_pins(&pin_mask, &long_mask,
-				   hotplug_trigger,
-				   dig_hotplug_reg, hpd_cpt,
-				   pch_port_hotplug_long_detect);
-
-		intel_hpd_irq_handler(dev, pin_mask, long_mask);
-	}
+	if (hotplug_trigger)
+		ibx_hpd_irq_handler(dev, hotplug_trigger, hpd_cpt);
 
 	if (pch_iir & SDE_AUDIO_POWER_MASK_CPT) {
 		int port = ffs((pch_iir & SDE_AUDIO_POWER_MASK_CPT) >>
@@ -1937,23 +1933,30 @@ static void spt_irq_handler(struct drm_device *dev, u32 pch_iir)
 		gmbus_irq_handler(dev);
 }
 
+static void ilk_hpd_irq_handler(struct drm_device *dev, u32 hotplug_trigger,
+				const u32 hpd[HPD_NUM_PINS])
+{
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	u32 dig_hotplug_reg, pin_mask = 0, long_mask = 0;
+
+	dig_hotplug_reg = I915_READ(DIGITAL_PORT_HOTPLUG_CNTRL);
+	I915_WRITE(DIGITAL_PORT_HOTPLUG_CNTRL, dig_hotplug_reg);
+
+	intel_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger,
+			   dig_hotplug_reg, hpd,
+			   ilk_port_hotplug_long_detect);
+
+	intel_hpd_irq_handler(dev, pin_mask, long_mask);
+}
+
 static void ilk_display_irq_handler(struct drm_device *dev, u32 de_iir)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	enum pipe pipe;
 	u32 hotplug_trigger = de_iir & DE_DP_A_HOTPLUG;
 
-	if (hotplug_trigger) {
-		u32 dig_hotplug_reg, pin_mask = 0, long_mask = 0;
-
-		dig_hotplug_reg = I915_READ(DIGITAL_PORT_HOTPLUG_CNTRL);
-		I915_WRITE(DIGITAL_PORT_HOTPLUG_CNTRL, dig_hotplug_reg);
-
-		intel_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger,
-				   dig_hotplug_reg, hpd_ilk,
-				   ilk_port_hotplug_long_detect);
-		intel_hpd_irq_handler(dev, pin_mask, long_mask);
-	}
+	if (hotplug_trigger)
+		ilk_hpd_irq_handler(dev, hotplug_trigger, hpd_ilk);
 
 	if (de_iir & DE_AUX_CHANNEL_A)
 		dp_aux_irq_handler(dev);
@@ -2005,17 +2008,8 @@ static void ivb_display_irq_handler(struct drm_device *dev, u32 de_iir)
 	enum pipe pipe;
 	u32 hotplug_trigger = de_iir & DE_DP_A_HOTPLUG_IVB;
 
-	if (hotplug_trigger) {
-		u32 dig_hotplug_reg, pin_mask = 0, long_mask = 0;
-
-		dig_hotplug_reg = I915_READ(DIGITAL_PORT_HOTPLUG_CNTRL);
-		I915_WRITE(DIGITAL_PORT_HOTPLUG_CNTRL, dig_hotplug_reg);
-
-		intel_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger,
-				   dig_hotplug_reg, hpd_ivb,
-				   ilk_port_hotplug_long_detect);
-		intel_hpd_irq_handler(dev, pin_mask, long_mask);
-	}
+	if (hotplug_trigger)
+		ilk_hpd_irq_handler(dev, hotplug_trigger, hpd_ivb);
 
 	if (de_iir & DE_ERR_INT_IVB)
 		ivb_err_int_handler(dev);
@@ -2128,7 +2122,8 @@ static irqreturn_t ironlake_irq_handler(int irq, void *arg)
 	return ret;
 }
 
-static void bxt_hpd_irq_handler(struct drm_device *dev, u32 hotplug_trigger)
+static void bxt_hpd_irq_handler(struct drm_device *dev, u32 hotplug_trigger,
+				const u32 hpd[HPD_NUM_PINS])
 {
 	struct drm_i915_private *dev_priv = to_i915(dev);
 	u32 dig_hotplug_reg, pin_mask = 0, long_mask = 0;
@@ -2137,8 +2132,9 @@ static void bxt_hpd_irq_handler(struct drm_device *dev, u32 hotplug_trigger)
 	I915_WRITE(BXT_HOTPLUG_CTL, dig_hotplug_reg);
 
 	intel_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger,
-			   dig_hotplug_reg, hpd_bxt,
+			   dig_hotplug_reg, hpd,
 			   bxt_port_hotplug_long_detect);
+
 	intel_hpd_irq_handler(dev, pin_mask, long_mask);
 }
 
@@ -2198,26 +2194,16 @@ static irqreturn_t gen8_irq_handler(int irq, void *arg)
 			I915_WRITE(GEN8_DE_PORT_IIR, tmp);
 			ret = IRQ_HANDLED;
 
-			if (IS_BROADWELL(dev) && hotplug_trigger) {
-				u32 dig_hotplug_reg, pin_mask = 0, long_mask = 0;
-
-				dig_hotplug_reg = I915_READ(DIGITAL_PORT_HOTPLUG_CNTRL);
-				I915_WRITE(DIGITAL_PORT_HOTPLUG_CNTRL, dig_hotplug_reg);
-
-				intel_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger,
-						   dig_hotplug_reg, hpd_bdw,
-						   ilk_port_hotplug_long_detect);
-				intel_hpd_irq_handler(dev, pin_mask, long_mask);
-				found = true;
-			}
-
 			if (tmp & aux_mask) {
 				dp_aux_irq_handler(dev);
 				found = true;
 			}
 
-			if (IS_BROXTON(dev) && hotplug_trigger) {
-				bxt_hpd_irq_handler(dev, hotplug_trigger);
+			if (hotplug_trigger) {
+				if (IS_BROXTON(dev))
+					bxt_hpd_irq_handler(dev, hotplug_trigger, hpd_bxt);
+				else
+					ilk_hpd_irq_handler(dev, hotplug_trigger, hpd_bdw);
 				found = true;
 			}
 
-- 
cgit v0.10.2


From a52bb15bfad09115cff2433e76b164be94c42af5 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 27 Aug 2015 23:56:11 +0300
Subject: drm/i915: Rewrite BXT HPD code to conform to pre-existing style
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Rewrite the BXT hpd setup to match the way we do it on other platforms:
- Throw out BXT_HOTPLUG_CTL since it's the same as PCH_PORT_HOTPLUG
- Enable the HPD bits in the DE port IER in gen8_de_irq_postinstall()
- Update DE port IMR using bdw_update_port_irq()

Also throw out port D from bxt_port_hotplug_long_detect() since BXT only
goes up to C.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index d7a37a5..218fe12 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -1300,8 +1300,6 @@ static bool bxt_port_hotplug_long_detect(enum port port, u32 val)
 		return val & PORTB_HOTPLUG_LONG_DETECT;
 	case PORT_C:
 		return val & PORTC_HOTPLUG_LONG_DETECT;
-	case PORT_D:
-		return val & PORTD_HOTPLUG_LONG_DETECT;
 	default:
 		return false;
 	}
@@ -2128,8 +2126,8 @@ static void bxt_hpd_irq_handler(struct drm_device *dev, u32 hotplug_trigger,
 	struct drm_i915_private *dev_priv = to_i915(dev);
 	u32 dig_hotplug_reg, pin_mask = 0, long_mask = 0;
 
-	dig_hotplug_reg = I915_READ(BXT_HOTPLUG_CTL);
-	I915_WRITE(BXT_HOTPLUG_CTL, dig_hotplug_reg);
+	dig_hotplug_reg = I915_READ(PCH_PORT_HOTPLUG);
+	I915_WRITE(PCH_PORT_HOTPLUG, dig_hotplug_reg);
 
 	intel_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger,
 			   dig_hotplug_reg, hpd,
@@ -3247,27 +3245,17 @@ static void ilk_hpd_irq_setup(struct drm_device *dev)
 static void bxt_hpd_irq_setup(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 hotplug_port;
-	u32 hotplug_ctrl;
-
-	hotplug_port = intel_hpd_enabled_irqs(dev, hpd_bxt);
-
-	hotplug_ctrl = I915_READ(BXT_HOTPLUG_CTL) & ~BXT_HOTPLUG_CTL_MASK;
+	u32 hotplug_irqs, hotplug, enabled_irqs;
 
-	if (hotplug_port & BXT_DE_PORT_HP_DDIA)
-		hotplug_ctrl |= BXT_DDIA_HPD_ENABLE;
-	if (hotplug_port & BXT_DE_PORT_HP_DDIB)
-		hotplug_ctrl |= BXT_DDIB_HPD_ENABLE;
-	if (hotplug_port & BXT_DE_PORT_HP_DDIC)
-		hotplug_ctrl |= BXT_DDIC_HPD_ENABLE;
-	I915_WRITE(BXT_HOTPLUG_CTL, hotplug_ctrl);
+	enabled_irqs = intel_hpd_enabled_irqs(dev, hpd_bxt);
+	hotplug_irqs = BXT_DE_PORT_HOTPLUG_MASK;
 
-	hotplug_ctrl = I915_READ(GEN8_DE_PORT_IMR) & ~hotplug_port;
-	I915_WRITE(GEN8_DE_PORT_IMR, hotplug_ctrl);
+	bdw_update_port_irq(dev_priv, hotplug_irqs, enabled_irqs);
 
-	hotplug_ctrl = I915_READ(GEN8_DE_PORT_IER) | hotplug_port;
-	I915_WRITE(GEN8_DE_PORT_IER, hotplug_ctrl);
-	POSTING_READ(GEN8_DE_PORT_IER);
+	hotplug = I915_READ(PCH_PORT_HOTPLUG);
+	hotplug |= PORTC_HOTPLUG_ENABLE | PORTB_HOTPLUG_ENABLE |
+		PORTA_HOTPLUG_ENABLE;
+	I915_WRITE(PCH_PORT_HOTPLUG, hotplug);
 }
 
 static void ibx_irq_postinstall(struct drm_device *dev)
@@ -3561,7 +3549,9 @@ static void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv)
 					   GEN8_PIPE_FIFO_UNDERRUN;
 
 	de_port_enables = de_port_masked;
-	if (IS_BROADWELL(dev_priv))
+	if (IS_BROXTON(dev_priv))
+		de_port_enables |= BXT_DE_PORT_HOTPLUG_MASK;
+	else if (IS_BROADWELL(dev_priv))
 		de_port_enables |= GEN8_PORT_DP_A_HOTPLUG;
 
 	dev_priv->de_irq_mask[PIPE_A] = ~de_pipe_masked;
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 689b428..5811eeb 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -5797,21 +5797,6 @@ enum skl_disp_power_wells {
 #define GEN8_PCU_IIR 0x444e8
 #define GEN8_PCU_IER 0x444ec
 
-/* BXT hotplug control */
-#define BXT_HOTPLUG_CTL			0xC4030
-#define   BXT_DDIA_HPD_ENABLE		(1 << 28)
-#define   BXT_DDIA_HPD_STATUS		(3 << 24)
-#define   BXT_DDIC_HPD_ENABLE		(1 << 12)
-#define   BXT_DDIC_HPD_STATUS		(3 << 8)
-#define   BXT_DDIB_HPD_ENABLE		(1 << 4)
-#define   BXT_DDIB_HPD_STATUS		(3 << 0)
-#define   BXT_HOTPLUG_CTL_MASK		(BXT_DDIA_HPD_ENABLE | \
-					 BXT_DDIB_HPD_ENABLE | \
-					 BXT_DDIC_HPD_ENABLE)
-#define   BXT_HPD_STATUS_MASK		(BXT_DDIA_HPD_STATUS | \
-					 BXT_DDIB_HPD_STATUS | \
-					 BXT_DDIC_HPD_STATUS)
-
 #define ILK_DISPLAY_CHICKEN2	0x42004
 /* Required on all Ironlake and Sandybridge according to the B-Spec. */
 #define  ILK_ELPIN_409_SELECT	(1 << 25)
-- 
cgit v0.10.2


From 58f2cf241fb98414e2e35fc3d4d0494eb30b3245 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 28 Aug 2015 22:59:08 +0300
Subject: drm/i915: Don't call intel_get_hpd_pins() when there's no hotplug
 interrupt
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

On GMCH plaforms we are now getting the following spew on aux
interrupts:
[drm:intel_get_hpd_pins] hotplug event received, stat 0x00000000, dig 0x00000000, pins 0x00000000
[drm:intel_get_hpd_pins] hotplug event received, stat 0x00000000, dig 0x00000000, pins 0x00000000
[drm:intel_get_hpd_pins] hotplug event received, stat 0x00000000, dig 0x00000000, pins 0x00000000
[drm:intel_get_hpd_pins] hotplug event received, stat 0x00000000, dig 0x00000000, pins 0x00000000
[drm:intel_get_hpd_pins] hotplug event received, stat 0x00000000, dig 0x00000000, pins 0x00000000
[drm:intel_dp_aux_ch] dp_aux_ch timeout status 0x71450064

Prevent it by not calling intel_get_hpd_pins() unless one of the HPD
interrupt bits are actually set.

I already fixed similar annoyance once with
4bca26d0a6518d51a9abe64fbde4b12f04c74053 drm/i915: Use HOTPLUG_INT_STATUS_G4X on VLV/CHV

but another source for it got added in
fd63e2a972c670887e5e8a08440111d3812c0996 drm/i915: combine i9xx_get_hpd_pins and pch_get_hpd_pins

due to pch_get_hpd_pins() being chosen over i9xx_get_hpd_pins() to
serve as the new unified piece of code. pch_get_hpd_pins() had the debug
print, and i9xx_get_hpd_pins() didn't.

Cc: Imre Deak <imre.deak@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 218fe12..ff543a4 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -1639,20 +1639,26 @@ static void i9xx_hpd_irq_handler(struct drm_device *dev)
 	if (IS_G4X(dev) || IS_VALLEYVIEW(dev)) {
 		u32 hotplug_trigger = hotplug_status & HOTPLUG_INT_STATUS_G4X;
 
-		intel_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger,
-				   hotplug_trigger, hpd_status_g4x,
-				   i9xx_port_hotplug_long_detect);
-		intel_hpd_irq_handler(dev, pin_mask, long_mask);
+		if (hotplug_trigger) {
+			intel_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger,
+					   hotplug_trigger, hpd_status_g4x,
+					   i9xx_port_hotplug_long_detect);
+
+			intel_hpd_irq_handler(dev, pin_mask, long_mask);
+		}
 
 		if (hotplug_status & DP_AUX_CHANNEL_MASK_INT_STATUS_G4X)
 			dp_aux_irq_handler(dev);
 	} else {
 		u32 hotplug_trigger = hotplug_status & HOTPLUG_INT_STATUS_I915;
 
-		intel_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger,
-				   hotplug_trigger, hpd_status_g4x,
-				   i9xx_port_hotplug_long_detect);
-		intel_hpd_irq_handler(dev, pin_mask, long_mask);
+		if (hotplug_trigger) {
+			intel_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger,
+					   hotplug_trigger, hpd_status_g4x,
+					   i9xx_port_hotplug_long_detect);
+
+			intel_hpd_irq_handler(dev, pin_mask, long_mask);
+		}
 	}
 }
 
-- 
cgit v0.10.2


From 395ab7541a0e141ffbde127b534e44f28d39a050 Mon Sep 17 00:00:00 2001
From: "Kumar, Mahesh" <mahesh1.kumar@intel.com>
Date: Thu, 3 Sep 2015 16:17:08 +0530
Subject: drm/i915/skl: Avoid using un-initialized bits_per_pixel

Don't rely on fb->bits_per_pixel as intel_framebuffer_init is not
filling bits_per_pixel field of fb-struct for YUV pixel format.
This leads to divide by zero error during watermark calculation.

Signed-off-by: Kumar, Mahesh <mahesh1.kumar@intel.com>
Cc: Konduru, Chandra <chandra.konduru@intel.com>
Reviewed-by: Sonika Jindal <sonika.jindal@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index ea49661..1b90f03 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -3170,7 +3170,8 @@ static void skl_compute_wm_pipe_parameters(struct drm_crtc *crtc,
 		if (fb) {
 			p->plane[0].enabled = true;
 			p->plane[0].bytes_per_pixel = fb->pixel_format == DRM_FORMAT_NV12 ?
-				drm_format_plane_cpp(fb->pixel_format, 1) : fb->bits_per_pixel / 8;
+				drm_format_plane_cpp(fb->pixel_format, 1) :
+				drm_format_plane_cpp(fb->pixel_format, 0);
 			p->plane[0].y_bytes_per_pixel = fb->pixel_format == DRM_FORMAT_NV12 ?
 				drm_format_plane_cpp(fb->pixel_format, 0) : 0;
 			p->plane[0].tiling = fb->modifier[0];
-- 
cgit v0.10.2


From ea916ea057299f66e774c9131e269df070a84779 Mon Sep 17 00:00:00 2001
From: "Kumar, Mahesh" <mahesh1.kumar@intel.com>
Date: Thu, 3 Sep 2015 16:17:09 +0530
Subject: drm/i915/skl+: Add YUV pixel format in Capability list

GEN >= 9 supports YUV format for all planes, but it's not exported in
Capability list of primary plane. Add YUV formats in skl_primary_formats
list.

Testcase: igt/kms_universal_plane.c

Signed-off-by: Kumar, Mahesh <mahesh1.kumar@intel.com>
Cc: Konduru, Chandra <chandra.konduru@intel.com>
Reviewed-by: Sonika Jindal <sonika.jindal@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 888175c..6985f6f 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -72,6 +72,10 @@ static const uint32_t skl_primary_formats[] = {
 	DRM_FORMAT_ABGR8888,
 	DRM_FORMAT_XRGB2101010,
 	DRM_FORMAT_XBGR2101010,
+	DRM_FORMAT_YUYV,
+	DRM_FORMAT_YVYU,
+	DRM_FORMAT_UYVY,
+	DRM_FORMAT_VYUY,
 };
 
 /* Cursor formats */
-- 
cgit v0.10.2


From 3a41a05d70b4fb455256fd3a8348e186f156c310 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Micha=C5=82=20Winiarski?= <michal.winiarski@intel.com>
Date: Thu, 3 Sep 2015 19:22:18 +0200
Subject: drm/i915/gtt: Avoid calling kcalloc in a loop when allocating temp
 bitmaps
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

On each call to gen8_alloc_va_range_3lvl we're allocating temporary
bitmaps needed for error handling. Unfortunately, when we increase
address space size (48b ppgtt) we do additional (512 - 4) calls to
kcalloc, increasing latency between exec and actual start of execution
on the GPU. Let's just do a single kcalloc, we can also drop the size
from free_gen8_temp_bitmaps since it's no longer used.

v2: Use GFP_TEMPORARY to make the allocations reclaimable.
v3: Drop the 2D array, just allocate a single block.
v4: Rebase to handle gen8_preallocate_top_level_pdps.
v5: Align misaligned bracket.

Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Michel Thierry <michel.thierry@intel.com>
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
[danvet: Correct kcalloc arguments as suggested by Chris.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index bdb7adc..8786281 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -1164,13 +1164,8 @@ unwind_out:
 }
 
 static void
-free_gen8_temp_bitmaps(unsigned long *new_pds, unsigned long **new_pts,
-		       uint32_t pdpes)
+free_gen8_temp_bitmaps(unsigned long *new_pds, unsigned long *new_pts)
 {
-	int i;
-
-	for (i = 0; i < pdpes; i++)
-		kfree(new_pts[i]);
 	kfree(new_pts);
 	kfree(new_pds);
 }
@@ -1180,29 +1175,20 @@ free_gen8_temp_bitmaps(unsigned long *new_pds, unsigned long **new_pts,
  */
 static
 int __must_check alloc_gen8_temp_bitmaps(unsigned long **new_pds,
-					 unsigned long ***new_pts,
+					 unsigned long **new_pts,
 					 uint32_t pdpes)
 {
-	int i;
 	unsigned long *pds;
-	unsigned long **pts;
+	unsigned long *pts;
 
-	pds = kcalloc(BITS_TO_LONGS(pdpes), sizeof(unsigned long), GFP_KERNEL);
+	pds = kcalloc(BITS_TO_LONGS(pdpes), sizeof(unsigned long), GFP_TEMPORARY);
 	if (!pds)
 		return -ENOMEM;
 
-	pts = kcalloc(pdpes, sizeof(unsigned long *), GFP_KERNEL);
-	if (!pts) {
-		kfree(pds);
-		return -ENOMEM;
-	}
-
-	for (i = 0; i < pdpes; i++) {
-		pts[i] = kcalloc(BITS_TO_LONGS(I915_PDES),
-				 sizeof(unsigned long), GFP_KERNEL);
-		if (!pts[i])
-			goto err_out;
-	}
+	pts = kcalloc(pdpes, BITS_TO_LONGS(I915_PDES) * sizeof(unsigned long),
+		      GFP_TEMPORARY);
+	if (!pts)
+		goto err_out;
 
 	*new_pds = pds;
 	*new_pts = pts;
@@ -1210,7 +1196,7 @@ int __must_check alloc_gen8_temp_bitmaps(unsigned long **new_pds,
 	return 0;
 
 err_out:
-	free_gen8_temp_bitmaps(pds, pts, pdpes);
+	free_gen8_temp_bitmaps(pds, pts);
 	return -ENOMEM;
 }
 
@@ -1231,7 +1217,7 @@ static int gen8_alloc_va_range_3lvl(struct i915_address_space *vm,
 {
 	struct i915_hw_ppgtt *ppgtt =
 		container_of(vm, struct i915_hw_ppgtt, base);
-	unsigned long *new_page_dirs, **new_page_tables;
+	unsigned long *new_page_dirs, *new_page_tables;
 	struct drm_device *dev = vm->dev;
 	struct i915_page_directory *pd;
 	const uint64_t orig_start = start;
@@ -1258,14 +1244,14 @@ static int gen8_alloc_va_range_3lvl(struct i915_address_space *vm,
 	ret = gen8_ppgtt_alloc_page_directories(vm, pdp, start, length,
 						new_page_dirs);
 	if (ret) {
-		free_gen8_temp_bitmaps(new_page_dirs, new_page_tables, pdpes);
+		free_gen8_temp_bitmaps(new_page_dirs, new_page_tables);
 		return ret;
 	}
 
 	/* For every page directory referenced, allocate page tables */
 	gen8_for_each_pdpe(pd, pdp, start, length, temp, pdpe) {
 		ret = gen8_ppgtt_alloc_pagetabs(vm, pd, start, length,
-						new_page_tables[pdpe]);
+						new_page_tables + pdpe * BITS_TO_LONGS(I915_PDES));
 		if (ret)
 			goto err_out;
 	}
@@ -1316,20 +1302,21 @@ static int gen8_alloc_va_range_3lvl(struct i915_address_space *vm,
 		gen8_setup_page_directory(ppgtt, pdp, pd, pdpe);
 	}
 
-	free_gen8_temp_bitmaps(new_page_dirs, new_page_tables, pdpes);
+	free_gen8_temp_bitmaps(new_page_dirs, new_page_tables);
 	mark_tlbs_dirty(ppgtt);
 	return 0;
 
 err_out:
 	while (pdpe--) {
-		for_each_set_bit(temp, new_page_tables[pdpe], I915_PDES)
+		for_each_set_bit(temp, new_page_tables + pdpe *
+				BITS_TO_LONGS(I915_PDES), I915_PDES)
 			free_pt(dev, pdp->page_directory[pdpe]->page_table[temp]);
 	}
 
 	for_each_set_bit(pdpe, new_page_dirs, pdpes)
 		free_pd(dev, pdp->page_directory[pdpe]);
 
-	free_gen8_temp_bitmaps(new_page_dirs, new_page_tables, pdpes);
+	free_gen8_temp_bitmaps(new_page_dirs, new_page_tables);
 	mark_tlbs_dirty(ppgtt);
 	return ret;
 }
@@ -1481,7 +1468,7 @@ static void gen8_dump_ppgtt(struct i915_hw_ppgtt *ppgtt, struct seq_file *m)
 
 static int gen8_preallocate_top_level_pdps(struct i915_hw_ppgtt *ppgtt)
 {
-	unsigned long *new_page_dirs, **new_page_tables;
+	unsigned long *new_page_dirs, *new_page_tables;
 	uint32_t pdpes = I915_PDPES_PER_PDP(dev);
 	int ret;
 
@@ -1501,7 +1488,7 @@ static int gen8_preallocate_top_level_pdps(struct i915_hw_ppgtt *ppgtt)
 	if (!ret)
 		*ppgtt->pdp.used_pdpes = *new_page_dirs;
 
-	free_gen8_temp_bitmaps(new_page_dirs, new_page_tables, pdpes);
+	free_gen8_temp_bitmaps(new_page_dirs, new_page_tables);
 
 	return ret;
 }
-- 
cgit v0.10.2


From b4834a508a1c25773c929c3d8eb7a29ad1041e19 Mon Sep 17 00:00:00 2001
From: Rodrigo Vivi <rodrigo.vivi@intel.com>
Date: Wed, 2 Sep 2015 15:19:24 -0700
Subject: drm/i915: Future proof interrupt handler.

These functions are already being called for gen >= 9,
so let's be sure when this happens we use whatever is
there already for the latest platform.

No functional change.

Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index ff543a4..ca6d806 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -2155,7 +2155,7 @@ static irqreturn_t gen8_irq_handler(int irq, void *arg)
 	if (!intel_irqs_enabled(dev_priv))
 		return IRQ_NONE;
 
-	if (IS_GEN9(dev))
+	if (INTEL_INFO(dev_priv)->gen >= 9)
 		aux_mask |=  GEN9_AUX_CHANNEL_B | GEN9_AUX_CHANNEL_C |
 			GEN9_AUX_CHANNEL_D;
 
@@ -2238,7 +2238,7 @@ static irqreturn_t gen8_irq_handler(int irq, void *arg)
 			    intel_pipe_handle_vblank(dev, pipe))
 				intel_check_page_flip(dev, pipe);
 
-			if (IS_GEN9(dev))
+			if (INTEL_INFO(dev_priv)->gen >= 9)
 				flip_done = pipe_iir & GEN9_PIPE_PLANE1_FLIP_DONE;
 			else
 				flip_done = pipe_iir & GEN8_PIPE_PRIMARY_FLIP_DONE;
@@ -2256,7 +2256,7 @@ static irqreturn_t gen8_irq_handler(int irq, void *arg)
 								    pipe);
 
 
-			if (IS_GEN9(dev))
+			if (INTEL_INFO(dev_priv)->gen >= 9)
 				fault_errors = pipe_iir & GEN9_DE_PIPE_IRQ_FAULT_ERRORS;
 			else
 				fault_errors = pipe_iir & GEN8_DE_PIPE_IRQ_FAULT_ERRORS;
@@ -3539,7 +3539,7 @@ static void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv)
 	u32 de_port_enables;
 	enum pipe pipe;
 
-	if (IS_GEN9(dev_priv)) {
+	if (INTEL_INFO(dev_priv)->gen >= 9) {
 		de_pipe_masked |= GEN9_PIPE_PLANE1_FLIP_DONE |
 				  GEN9_DE_PIPE_IRQ_FAULT_ERRORS;
 		de_port_masked |= GEN9_AUX_CHANNEL_B | GEN9_AUX_CHANNEL_C |
-- 
cgit v0.10.2


From 22cc27bb7a47cf6ad33b983539b14bc46e65f9b2 Mon Sep 17 00:00:00 2001
From: Rodrigo Vivi <rodrigo.vivi@intel.com>
Date: Thu, 3 Sep 2015 14:38:00 -0700
Subject: drm/i915: Future proof uncore_init.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Unless future specs tells otherwise we can assume future gens
inherit some stuff from the previous so let's handle
missed cases when we know tehy should't be there and assume
default equals newest one.

No functional changes.

v2: Remove useless case as pointed out by Ville.

Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c
index dec20d6..440e2a5 100644
--- a/drivers/gpu/drm/i915/intel_uncore.c
+++ b/drivers/gpu/drm/i915/intel_uncore.c
@@ -1202,8 +1202,6 @@ void intel_uncore_init(struct drm_device *dev)
 
 	switch (INTEL_INFO(dev)->gen) {
 	default:
-		MISSING_CASE(INTEL_INFO(dev)->gen);
-		return;
 	case 9:
 		ASSIGN_WRITE_MMIO_VFUNCS(gen9);
 		ASSIGN_READ_MMIO_VFUNCS(gen9);
-- 
cgit v0.10.2


From 1c132b44f0eabaaf801e79a3e4b98770431a4274 Mon Sep 17 00:00:00 2001
From: Rodrigo Vivi <rodrigo.vivi@intel.com>
Date: Wed, 2 Sep 2015 15:19:26 -0700
Subject: drm/i915: Future proof panel fitter.

This is another case where we can consider the default is the
newest available and not actually a missed case.

No functional change.

Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 6985f6f..b59717a 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -4919,12 +4919,10 @@ static void haswell_crtc_enable(struct drm_crtc *crtc)
 
 	intel_ddi_enable_pipe_clock(intel_crtc);
 
-	if (INTEL_INFO(dev)->gen == 9)
+	if (INTEL_INFO(dev)->gen >= 9)
 		skylake_pfit_enable(intel_crtc);
-	else if (INTEL_INFO(dev)->gen < 9)
-		ironlake_pfit_enable(intel_crtc);
 	else
-		MISSING_CASE(INTEL_INFO(dev)->gen);
+		ironlake_pfit_enable(intel_crtc);
 
 	/*
 	 * On ILK+ LUT must be loaded before the pipe is running but with
@@ -5056,12 +5054,10 @@ static void haswell_crtc_disable(struct drm_crtc *crtc)
 
 	intel_ddi_disable_transcoder_func(dev_priv, cpu_transcoder);
 
-	if (INTEL_INFO(dev)->gen == 9)
+	if (INTEL_INFO(dev)->gen >= 9)
 		skylake_scaler_disable(intel_crtc);
-	else if (INTEL_INFO(dev)->gen < 9)
-		ironlake_pfit_disable(intel_crtc);
 	else
-		MISSING_CASE(INTEL_INFO(dev)->gen);
+		ironlake_pfit_disable(intel_crtc);
 
 	intel_ddi_disable_pipe_clock(intel_crtc);
 
@@ -9784,12 +9780,10 @@ static bool haswell_get_pipe_config(struct intel_crtc *crtc,
 	}
 
 	if (intel_display_power_is_enabled(dev_priv, pfit_domain)) {
-		if (INTEL_INFO(dev)->gen == 9)
+		if (INTEL_INFO(dev)->gen >= 9)
 			skylake_get_pfit_config(crtc, pipe_config);
-		else if (INTEL_INFO(dev)->gen < 9)
-			ironlake_get_pfit_config(crtc, pipe_config);
 		else
-			MISSING_CASE(INTEL_INFO(dev)->gen);
+			ironlake_get_pfit_config(crtc, pipe_config);
 	}
 
 	if (IS_HASWELL(dev))
-- 
cgit v0.10.2


From bc5133d595a0ddfaaa5edb4b077f44d92cc8fb0d Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 3 Sep 2015 11:16:07 +0300
Subject: drm/i915/dp: move TPS3 logic to where it's used
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

There is no need to have a separate flag for tps3 as the information is
only used at one location. Move the logic there to make it easier to
follow.

Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index f8f4d99..9a6fb5d 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -3812,13 +3812,25 @@ intel_dp_start_link_train(struct intel_dp *intel_dp)
 void
 intel_dp_complete_link_train(struct intel_dp *intel_dp)
 {
+	struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp);
+	struct drm_device *dev = dig_port->base.base.dev;
 	bool channel_eq = false;
 	int tries, cr_tries;
 	uint32_t DP = intel_dp->DP;
 	uint32_t training_pattern = DP_TRAINING_PATTERN_2;
 
-	/* Training Pattern 3 for HBR2 or 1.2 devices that support it*/
-	if (intel_dp->link_rate == 540000 || intel_dp->use_tps3)
+	/*
+	 * Training Pattern 3 for HBR2 or 1.2 devices that support it.
+	 *
+	 * Intel platforms that support HBR2 also support TPS3. TPS3 support is
+	 * also mandatory for downstream devices that support HBR2.
+	 *
+	 * Due to WaDisableHBR2 SKL < B0 is the only exception where TPS3 is
+	 * supported but still not enabled.
+	 */
+	if (intel_dp->link_rate == 540000 ||
+	    (intel_dp_source_supports_hbr2(dev) &&
+	     drm_dp_tps3_supported(intel_dp->dpcd)))
 		training_pattern = DP_TRAINING_PATTERN_3;
 
 	/* channel equalization */
@@ -4000,18 +4012,9 @@ intel_dp_get_dpcd(struct intel_dp *intel_dp)
 		}
 	}
 
-	/* Training Pattern 3 support, Intel platforms that support HBR2 alone
-	 * have support for TP3 hence that check is used along with dpcd check
-	 * to ensure TP3 can be enabled.
-	 * SKL < B0: due it's WaDisableHBR2 is the only exception where TP3 is
-	 * supported but still not enabled.
-	 */
-	if (drm_dp_tps3_supported(intel_dp->dpcd) &&
-	    intel_dp_source_supports_hbr2(dev)) {
-		intel_dp->use_tps3 = true;
-		DRM_DEBUG_KMS("Displayport TPS3 supported\n");
-	} else
-		intel_dp->use_tps3 = false;
+	DRM_DEBUG_KMS("Display Port TPS3 support: source %s, sink %s\n",
+		      intel_dp_source_supports_hbr2(dev) ? "yes" : "no",
+		      drm_dp_tps3_supported(intel_dp->dpcd) ? "yes" : "no");
 
 	/* Intermediate frequency support */
 	if (is_edp(intel_dp) &&
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 40e825d..2bdd544 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -746,7 +746,6 @@ struct intel_dp {
 	enum pipe pps_pipe;
 	struct edp_power_seq pps_delays;
 
-	bool use_tps3;
 	bool can_mst; /* this port supports mst */
 	bool is_mst;
 	int active_mst_links;
-- 
cgit v0.10.2


From 1da7d7131c35cde83f1bab8ec732b57b69bef814 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 3 Sep 2015 11:16:08 +0300
Subject: drm/i915: ignore link rate in TPS3 selection

TPS3 is mandatory for downstream devices that support HBR2, and Intel
platforms that support HBR2 also support TPS3. Whenever TPS3 is
supported by both the source and sink, it should be used. In other
words, whenever the source and sink are capable of 5.4 Gbps link, we
should anyway go for TPS3, regardless of the link rate being selected.

Log an error if the sink has advertized HBR2 capability without TPS3
capability.

Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 9a6fb5d..44e753d 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -3828,10 +3828,11 @@ intel_dp_complete_link_train(struct intel_dp *intel_dp)
 	 * Due to WaDisableHBR2 SKL < B0 is the only exception where TPS3 is
 	 * supported but still not enabled.
 	 */
-	if (intel_dp->link_rate == 540000 ||
-	    (intel_dp_source_supports_hbr2(dev) &&
-	     drm_dp_tps3_supported(intel_dp->dpcd)))
+	if (intel_dp_source_supports_hbr2(dev) &&
+	    drm_dp_tps3_supported(intel_dp->dpcd))
 		training_pattern = DP_TRAINING_PATTERN_3;
+	else if (intel_dp->link_rate == 540000)
+		DRM_ERROR("5.4 Gbps link rate without HBR2/TPS3 support\n");
 
 	/* channel equalization */
 	if (!intel_dp_set_link_train(intel_dp, &DP,
-- 
cgit v0.10.2


From 742f491d2c204204086d2bc85cc5100daa6ff336 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 3 Sep 2015 11:16:09 +0300
Subject: drm/i915: use the yesno helper for logging

Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index 4563f8b..41629fa 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -1382,17 +1382,16 @@ static int ironlake_drpc_info(struct seq_file *m)
 	intel_runtime_pm_put(dev_priv);
 	mutex_unlock(&dev->struct_mutex);
 
-	seq_printf(m, "HD boost: %s\n", (rgvmodectl & MEMMODE_BOOST_EN) ?
-		   "yes" : "no");
+	seq_printf(m, "HD boost: %s\n", yesno(rgvmodectl & MEMMODE_BOOST_EN));
 	seq_printf(m, "Boost freq: %d\n",
 		   (rgvmodectl & MEMMODE_BOOST_FREQ_MASK) >>
 		   MEMMODE_BOOST_FREQ_SHIFT);
 	seq_printf(m, "HW control enabled: %s\n",
-		   rgvmodectl & MEMMODE_HWIDLE_EN ? "yes" : "no");
+		   yesno(rgvmodectl & MEMMODE_HWIDLE_EN));
 	seq_printf(m, "SW control enabled: %s\n",
-		   rgvmodectl & MEMMODE_SWMODE_EN ? "yes" : "no");
+		   yesno(rgvmodectl & MEMMODE_SWMODE_EN));
 	seq_printf(m, "Gated voltage change: %s\n",
-		   rgvmodectl & MEMMODE_RCLK_GATE ? "yes" : "no");
+		   yesno(rgvmodectl & MEMMODE_RCLK_GATE));
 	seq_printf(m, "Starting frequency: P%d\n",
 		   (rgvmodectl & MEMMODE_FSTART_MASK) >> MEMMODE_FSTART_SHIFT);
 	seq_printf(m, "Max P-state: P%d\n",
@@ -1401,7 +1400,7 @@ static int ironlake_drpc_info(struct seq_file *m)
 	seq_printf(m, "RS1 VID: %d\n", (crstandvid & 0x3f));
 	seq_printf(m, "RS2 VID: %d\n", ((crstandvid >> 8) & 0x3f));
 	seq_printf(m, "Render standby enabled: %s\n",
-		   (rstdbyctl & RCX_SW_EXIT) ? "no" : "yes");
+		   yesno(!(rstdbyctl & RCX_SW_EXIT)));
 	seq_puts(m, "Current RS state: ");
 	switch (rstdbyctl & RSX_STATUS_MASK) {
 	case RSX_STATUS_ON:
@@ -2844,8 +2843,7 @@ static void intel_dp_info(struct seq_file *m,
 	struct intel_dp *intel_dp = enc_to_intel_dp(&intel_encoder->base);
 
 	seq_printf(m, "\tDPCD rev: %x\n", intel_dp->dpcd[DP_DPCD_REV]);
-	seq_printf(m, "\taudio support: %s\n", intel_dp->has_audio ? "yes" :
-		   "no");
+	seq_printf(m, "\taudio support: %s\n", yesno(intel_dp->has_audio));
 	if (intel_encoder->type == INTEL_OUTPUT_EDP)
 		intel_panel_info(m, &intel_connector->panel);
 }
@@ -2856,8 +2854,7 @@ static void intel_hdmi_info(struct seq_file *m,
 	struct intel_encoder *intel_encoder = intel_connector->encoder;
 	struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(&intel_encoder->base);
 
-	seq_printf(m, "\taudio support: %s\n", intel_hdmi->has_audio ? "yes" :
-		   "no");
+	seq_printf(m, "\taudio support: %s\n", yesno(intel_hdmi->has_audio));
 }
 
 static void intel_lvds_info(struct seq_file *m,
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 44e753d..45ab25e 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -4014,8 +4014,8 @@ intel_dp_get_dpcd(struct intel_dp *intel_dp)
 	}
 
 	DRM_DEBUG_KMS("Display Port TPS3 support: source %s, sink %s\n",
-		      intel_dp_source_supports_hbr2(dev) ? "yes" : "no",
-		      drm_dp_tps3_supported(intel_dp->dpcd) ? "yes" : "no");
+		      yesno(intel_dp_source_supports_hbr2(dev)),
+		      yesno(drm_dp_tps3_supported(intel_dp->dpcd)));
 
 	/* Intermediate frequency support */
 	if (is_edp(intel_dp) &&
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index 1b90f03..79cd987 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -5550,7 +5550,7 @@ static void cherryview_enable_rps(struct drm_device *dev)
 	/* RPS code assumes GPLL is used */
 	WARN_ONCE((val & GPLLENABLE) == 0, "GPLL not enabled\n");
 
-	DRM_DEBUG_DRIVER("GPLL enabled? %s\n", val & GPLLENABLE ? "yes" : "no");
+	DRM_DEBUG_DRIVER("GPLL enabled? %s\n", yesno(val & GPLLENABLE));
 	DRM_DEBUG_DRIVER("GPU status: 0x%08x\n", val);
 
 	dev_priv->rps.cur_freq = (val >> 8) & 0xff;
@@ -5640,7 +5640,7 @@ static void valleyview_enable_rps(struct drm_device *dev)
 	/* RPS code assumes GPLL is used */
 	WARN_ONCE((val & GPLLENABLE) == 0, "GPLL not enabled\n");
 
-	DRM_DEBUG_DRIVER("GPLL enabled? %s\n", val & GPLLENABLE ? "yes" : "no");
+	DRM_DEBUG_DRIVER("GPLL enabled? %s\n", yesno(val & GPLLENABLE));
 	DRM_DEBUG_DRIVER("GPU status: 0x%08x\n", val);
 
 	dev_priv->rps.cur_freq = (val >> 8) & 0xff;
-- 
cgit v0.10.2


From 01101fa7cc85fffc95798d1f67970dad96063fc6 Mon Sep 17 00:00:00 2001
From: Chris Wilson <chris@chris-wilson.co.uk>
Date: Thu, 3 Sep 2015 13:01:39 +0100
Subject: drm/i915: Refactor common ringbuffer allocation code

A small, very small, step to sharing the duplicate code between
execlists and legacy submission engines, starting with the ringbuffer
allocation code.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Arun Siluvery <arun.siluvery@linux.intel.com>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Cc: Dave Gordon <david.s.gordon@intel.com>
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
index 40cbba4..28a712e 100644
--- a/drivers/gpu/drm/i915/intel_lrc.c
+++ b/drivers/gpu/drm/i915/intel_lrc.c
@@ -2340,8 +2340,7 @@ void intel_lr_context_free(struct intel_context *ctx)
 				i915_gem_object_ggtt_unpin(ctx_obj);
 			}
 			WARN_ON(ctx->engine[ring->id].pin_count);
-			intel_destroy_ringbuffer_obj(ringbuf);
-			kfree(ringbuf);
+			intel_ringbuffer_free(ringbuf);
 			drm_gem_object_unreference(&ctx_obj->base);
 		}
 	}
@@ -2442,42 +2441,20 @@ int intel_lr_context_deferred_create(struct intel_context *ctx,
 			I915_WRITE(GEN8_GTCR, GEN8_GTCR_INVALIDATE);
 	}
 
-	ringbuf = kzalloc(sizeof(*ringbuf), GFP_KERNEL);
-	if (!ringbuf) {
-		DRM_DEBUG_DRIVER("Failed to allocate ringbuffer %s\n",
-				ring->name);
-		ret = -ENOMEM;
+	ringbuf = intel_engine_create_ringbuffer(ring, 4 * PAGE_SIZE);
+	if (IS_ERR(ringbuf)) {
+		ret = PTR_ERR(ringbuf);
 		goto error_unpin_ctx;
 	}
 
-	ringbuf->ring = ring;
-
-	ringbuf->size = 4 * PAGE_SIZE;
-	ringbuf->effective_size = ringbuf->size;
-	ringbuf->head = 0;
-	ringbuf->tail = 0;
-	ringbuf->last_retired_head = -1;
-	intel_ring_update_space(ringbuf);
-
-	if (ringbuf->obj == NULL) {
-		ret = intel_alloc_ringbuffer_obj(dev, ringbuf);
+	if (is_global_default_ctx) {
+		ret = intel_pin_and_map_ringbuffer_obj(dev, ringbuf);
 		if (ret) {
-			DRM_DEBUG_DRIVER(
-				"Failed to allocate ringbuffer obj %s: %d\n",
-				ring->name, ret);
-			goto error_free_rbuf;
+			DRM_ERROR(
+				  "Failed to pin and map ringbuffer %s: %d\n",
+				  ring->name, ret);
+			goto error_ringbuf;
 		}
-
-		if (is_global_default_ctx) {
-			ret = intel_pin_and_map_ringbuffer_obj(dev, ringbuf);
-			if (ret) {
-				DRM_ERROR(
-					"Failed to pin and map ringbuffer %s: %d\n",
-					ring->name, ret);
-				goto error_destroy_rbuf;
-			}
-		}
-
 	}
 
 	ret = populate_lr_context(ctx, ctx_obj, ring, ringbuf);
@@ -2519,10 +2496,8 @@ int intel_lr_context_deferred_create(struct intel_context *ctx,
 error:
 	if (is_global_default_ctx)
 		intel_unpin_ringbuffer_obj(ringbuf);
-error_destroy_rbuf:
-	intel_destroy_ringbuffer_obj(ringbuf);
-error_free_rbuf:
-	kfree(ringbuf);
+error_ringbuf:
+	intel_ringbuffer_free(ringbuf);
 error_unpin_ctx:
 	if (is_global_default_ctx)
 		i915_gem_object_ggtt_unpin(ctx_obj);
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index 6e6b8db..20a75bb 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -1996,14 +1996,14 @@ int intel_pin_and_map_ringbuffer_obj(struct drm_device *dev,
 	return 0;
 }
 
-void intel_destroy_ringbuffer_obj(struct intel_ringbuffer *ringbuf)
+static void intel_destroy_ringbuffer_obj(struct intel_ringbuffer *ringbuf)
 {
 	drm_gem_object_unreference(&ringbuf->obj->base);
 	ringbuf->obj = NULL;
 }
 
-int intel_alloc_ringbuffer_obj(struct drm_device *dev,
-			       struct intel_ringbuffer *ringbuf)
+static int intel_alloc_ringbuffer_obj(struct drm_device *dev,
+				      struct intel_ringbuffer *ringbuf)
 {
 	struct drm_i915_gem_object *obj;
 
@@ -2023,6 +2023,48 @@ int intel_alloc_ringbuffer_obj(struct drm_device *dev,
 	return 0;
 }
 
+struct intel_ringbuffer *
+intel_engine_create_ringbuffer(struct intel_engine_cs *engine, int size)
+{
+	struct intel_ringbuffer *ring;
+	int ret;
+
+	ring = kzalloc(sizeof(*ring), GFP_KERNEL);
+	if (ring == NULL)
+		return ERR_PTR(-ENOMEM);
+
+	ring->ring = engine;
+
+	ring->size = size;
+	/* Workaround an erratum on the i830 which causes a hang if
+	 * the TAIL pointer points to within the last 2 cachelines
+	 * of the buffer.
+	 */
+	ring->effective_size = size;
+	if (IS_I830(engine->dev) || IS_845G(engine->dev))
+		ring->effective_size -= 2 * CACHELINE_BYTES;
+
+	ring->last_retired_head = -1;
+	intel_ring_update_space(ring);
+
+	ret = intel_alloc_ringbuffer_obj(engine->dev, ring);
+	if (ret) {
+		DRM_ERROR("Failed to allocate ringbuffer %s: %d\n",
+			  engine->name, ret);
+		kfree(ring);
+		return ERR_PTR(ret);
+	}
+
+	return ring;
+}
+
+void
+intel_ringbuffer_free(struct intel_ringbuffer *ring)
+{
+	intel_destroy_ringbuffer_obj(ring);
+	kfree(ring);
+}
+
 static int intel_init_ring_buffer(struct drm_device *dev,
 				  struct intel_engine_cs *ring)
 {
@@ -2031,22 +2073,20 @@ static int intel_init_ring_buffer(struct drm_device *dev,
 
 	WARN_ON(ring->buffer);
 
-	ringbuf = kzalloc(sizeof(*ringbuf), GFP_KERNEL);
-	if (!ringbuf)
-		return -ENOMEM;
-	ring->buffer = ringbuf;
-
 	ring->dev = dev;
 	INIT_LIST_HEAD(&ring->active_list);
 	INIT_LIST_HEAD(&ring->request_list);
 	INIT_LIST_HEAD(&ring->execlist_queue);
 	i915_gem_batch_pool_init(dev, &ring->batch_pool);
-	ringbuf->size = 32 * PAGE_SIZE;
-	ringbuf->ring = ring;
 	memset(ring->semaphore.sync_seqno, 0, sizeof(ring->semaphore.sync_seqno));
 
 	init_waitqueue_head(&ring->irq_queue);
 
+	ringbuf = intel_engine_create_ringbuffer(ring, 32 * PAGE_SIZE);
+	if (IS_ERR(ringbuf))
+		return PTR_ERR(ringbuf);
+	ring->buffer = ringbuf;
+
 	if (I915_NEED_GFX_HWS(dev)) {
 		ret = init_status_page(ring);
 		if (ret)
@@ -2058,15 +2098,6 @@ static int intel_init_ring_buffer(struct drm_device *dev,
 			goto error;
 	}
 
-	WARN_ON(ringbuf->obj);
-
-	ret = intel_alloc_ringbuffer_obj(dev, ringbuf);
-	if (ret) {
-		DRM_ERROR("Failed to allocate ringbuffer %s: %d\n",
-				ring->name, ret);
-		goto error;
-	}
-
 	ret = intel_pin_and_map_ringbuffer_obj(dev, ringbuf);
 	if (ret) {
 		DRM_ERROR("Failed to pin and map ringbuffer %s: %d\n",
@@ -2075,14 +2106,6 @@ static int intel_init_ring_buffer(struct drm_device *dev,
 		goto error;
 	}
 
-	/* Workaround an erratum on the i830 which causes a hang if
-	 * the TAIL pointer points to within the last 2 cachelines
-	 * of the buffer.
-	 */
-	ringbuf->effective_size = ringbuf->size;
-	if (IS_I830(dev) || IS_845G(dev))
-		ringbuf->effective_size -= 2 * CACHELINE_BYTES;
-
 	ret = i915_cmd_parser_init_ring(ring);
 	if (ret)
 		goto error;
@@ -2090,7 +2113,7 @@ static int intel_init_ring_buffer(struct drm_device *dev,
 	return 0;
 
 error:
-	kfree(ringbuf);
+	intel_ringbuffer_free(ringbuf);
 	ring->buffer = NULL;
 	return ret;
 }
@@ -2098,19 +2121,18 @@ error:
 void intel_cleanup_ring_buffer(struct intel_engine_cs *ring)
 {
 	struct drm_i915_private *dev_priv;
-	struct intel_ringbuffer *ringbuf;
 
 	if (!intel_ring_initialized(ring))
 		return;
 
 	dev_priv = to_i915(ring->dev);
-	ringbuf = ring->buffer;
 
 	intel_stop_ring_buffer(ring);
 	WARN_ON(!IS_GEN2(ring->dev) && (I915_READ_MODE(ring) & MODE_IDLE) == 0);
 
-	intel_unpin_ringbuffer_obj(ringbuf);
-	intel_destroy_ringbuffer_obj(ringbuf);
+	intel_unpin_ringbuffer_obj(ring->buffer);
+	intel_ringbuffer_free(ring->buffer);
+	ring->buffer = NULL;
 
 	if (ring->cleanup)
 		ring->cleanup(ring);
@@ -2119,9 +2141,6 @@ void intel_cleanup_ring_buffer(struct intel_engine_cs *ring)
 
 	i915_cmd_parser_fini_ring(ring);
 	i915_gem_batch_pool_fini(&ring->batch_pool);
-
-	kfree(ringbuf);
-	ring->buffer = NULL;
 }
 
 static int ring_wait_for_space(struct intel_engine_cs *ring, int n)
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h
index 95b0b4b..49fa41d 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.h
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.h
@@ -420,12 +420,12 @@ intel_write_status_page(struct intel_engine_cs *ring,
 #define I915_GEM_HWS_SCRATCH_INDEX	0x40
 #define I915_GEM_HWS_SCRATCH_ADDR (I915_GEM_HWS_SCRATCH_INDEX << MI_STORE_DWORD_INDEX_SHIFT)
 
-void intel_unpin_ringbuffer_obj(struct intel_ringbuffer *ringbuf);
+struct intel_ringbuffer *
+intel_engine_create_ringbuffer(struct intel_engine_cs *engine, int size);
 int intel_pin_and_map_ringbuffer_obj(struct drm_device *dev,
 				     struct intel_ringbuffer *ringbuf);
-void intel_destroy_ringbuffer_obj(struct intel_ringbuffer *ringbuf);
-int intel_alloc_ringbuffer_obj(struct drm_device *dev,
-			       struct intel_ringbuffer *ringbuf);
+void intel_unpin_ringbuffer_obj(struct intel_ringbuffer *ringbuf);
+void intel_ringbuffer_free(struct intel_ringbuffer *ring);
 
 void intel_stop_ring_buffer(struct intel_engine_cs *ring);
 void intel_cleanup_ring_buffer(struct intel_engine_cs *ring);
-- 
cgit v0.10.2


From c5796b7126b2497513bfa9a9341d02cb833d83d6 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Thu, 3 Sep 2015 16:24:35 +0300
Subject: drm/i915: access the PP_CONTROL reg only pre GEN5
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This register exists only pre GEN5, but atm we also access it on
VLV/BXT/CHV. Prevent accessing it on these latter platforms.

Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c
index 0794dc8..a16308a 100644
--- a/drivers/gpu/drm/i915/intel_lvds.c
+++ b/drivers/gpu/drm/i915/intel_lvds.c
@@ -955,7 +955,7 @@ void intel_lvds_init(struct drm_device *dev)
 	if (HAS_PCH_SPLIT(dev)) {
 		I915_WRITE(PCH_PP_CONTROL,
 			   I915_READ(PCH_PP_CONTROL) | PANEL_UNLOCK_REGS);
-	} else {
+	} else if (INTEL_INFO(dev_priv)->gen < 5) {
 		I915_WRITE(PP_CONTROL,
 			   I915_READ(PP_CONTROL) | PANEL_UNLOCK_REGS);
 	}
-- 
cgit v0.10.2


From 96d12cbdbd03a9ce87a75520512ef9ad8c585e92 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Thu, 3 Sep 2015 16:24:36 +0300
Subject: drm/i915: access the PP_ON_DELAYS/PP_OFF_DELAYS regs only pre GEN5
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

These registers exist only before GEN5, so currently we may access
undefined registers on VLV/CHV and BXT. Apply the workaround only pre
GEN5.

Since the workaround is relevant only when LVDS is present, for clarity
apply it only if this is the case.

This triggered an unclaimed register access warning on BXT.

v2: (Ville)
- move the workaround to the LVDS init code
- print a debug note about the workaround

Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index f0eaa6f..066a0ef 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -997,8 +997,6 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 	intel_setup_gmbus(dev);
 	intel_opregion_setup(dev);
 
-	intel_setup_bios(dev);
-
 	i915_gem_load(dev);
 
 	/* On the 945G/GM, the chipset reports the MSI capability on the
diff --git a/drivers/gpu/drm/i915/intel_bios.c b/drivers/gpu/drm/i915/intel_bios.c
index b3e437b..c8acc29 100644
--- a/drivers/gpu/drm/i915/intel_bios.c
+++ b/drivers/gpu/drm/i915/intel_bios.c
@@ -1340,21 +1340,3 @@ intel_parse_bios(struct drm_device *dev)
 
 	return 0;
 }
-
-/* Ensure that vital registers have been initialised, even if the BIOS
- * is absent or just failing to do its job.
- */
-void intel_setup_bios(struct drm_device *dev)
-{
-	struct drm_i915_private *dev_priv = dev->dev_private;
-
-	 /* Set the Panel Power On/Off timings if uninitialized. */
-	if (!HAS_PCH_SPLIT(dev) &&
-	    I915_READ(PP_ON_DELAYS) == 0 && I915_READ(PP_OFF_DELAYS) == 0) {
-		/* Set T2 to 40ms and T5 to 200ms */
-		I915_WRITE(PP_ON_DELAYS, 0x019007d0);
-
-		/* Set T3 to 35ms and Tx to 200ms */
-		I915_WRITE(PP_OFF_DELAYS, 0x015e07d0);
-	}
-}
diff --git a/drivers/gpu/drm/i915/intel_bios.h b/drivers/gpu/drm/i915/intel_bios.h
index 46cd5c7..1b7417e 100644
--- a/drivers/gpu/drm/i915/intel_bios.h
+++ b/drivers/gpu/drm/i915/intel_bios.h
@@ -588,7 +588,6 @@ struct bdb_psr {
 	struct psr_table psr_table[16];
 } __packed;
 
-void intel_setup_bios(struct drm_device *dev);
 int intel_parse_bios(struct drm_device *dev);
 
 /*
diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c
index a16308a..2c2d1f0 100644
--- a/drivers/gpu/drm/i915/intel_lvds.c
+++ b/drivers/gpu/drm/i915/intel_lvds.c
@@ -985,6 +985,18 @@ void intel_lvds_init(struct drm_device *dev)
 		DRM_DEBUG_KMS("LVDS is not present in VBT, but enabled anyway\n");
 	}
 
+	 /* Set the Panel Power On/Off timings if uninitialized. */
+	if (INTEL_INFO(dev_priv)->gen < 5 &&
+	    I915_READ(PP_ON_DELAYS) == 0 && I915_READ(PP_OFF_DELAYS) == 0) {
+		/* Set T2 to 40ms and T5 to 200ms */
+		I915_WRITE(PP_ON_DELAYS, 0x019007d0);
+
+		/* Set T3 to 35ms and Tx to 200ms */
+		I915_WRITE(PP_OFF_DELAYS, 0x015e07d0);
+
+		DRM_DEBUG_KMS("Panel power timings uninitialized, setting defaults\n");
+	}
+
 	lvds_encoder = kzalloc(sizeof(*lvds_encoder), GFP_KERNEL);
 	if (!lvds_encoder)
 		return;
-- 
cgit v0.10.2


From 333b8ca8cfcafc8c556d4333fe660283fea691d4 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 3 Sep 2015 21:50:16 +0300
Subject: drm/i915: Dump pfit state as hex
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The pfit state is stored as register values, so dump them as hex instead
of decimal to make some sense of the error messages.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index b59717a..110b8dc 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -12439,16 +12439,16 @@ intel_pipe_config_compare(struct drm_device *dev,
 	PIPE_CONF_CHECK_I(pipe_src_w);
 	PIPE_CONF_CHECK_I(pipe_src_h);
 
-	PIPE_CONF_CHECK_I(gmch_pfit.control);
+	PIPE_CONF_CHECK_X(gmch_pfit.control);
 	/* pfit ratios are autocomputed by the hw on gen4+ */
 	if (INTEL_INFO(dev)->gen < 4)
 		PIPE_CONF_CHECK_I(gmch_pfit.pgm_ratios);
-	PIPE_CONF_CHECK_I(gmch_pfit.lvds_border_bits);
+	PIPE_CONF_CHECK_X(gmch_pfit.lvds_border_bits);
 
 	PIPE_CONF_CHECK_I(pch_pfit.enabled);
 	if (current_config->pch_pfit.enabled) {
-		PIPE_CONF_CHECK_I(pch_pfit.pos);
-		PIPE_CONF_CHECK_I(pch_pfit.size);
+		PIPE_CONF_CHECK_X(pch_pfit.pos);
+		PIPE_CONF_CHECK_X(pch_pfit.size);
 	}
 
 	PIPE_CONF_CHECK_I(scaler_state.scaler_id);
-- 
cgit v0.10.2


From 614f4ad7984f930623bf0d25adba707d0d36c6dc Mon Sep 17 00:00:00 2001
From: Chris Wilson <chris@chris-wilson.co.uk>
Date: Wed, 2 Sep 2015 12:29:40 +0100
Subject: drm/i915: Fix cmdparser STORE/LOAD command descriptors

Fixes regression from
commit f1afe24f0e736b9d7f2275e2b1504af3fe612f2a
Author: Arun Siluvery <arun.siluvery@linux.intel.com>
Date:   Tue Aug 4 16:22:20 2015 +0100

    drm/i915: Change SRM, LRM instructions to use correct length

which forgot to account for the length bias when declaring the fixed
length.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=91844
Reported-by: Andreas Reis <andreas.reis@gmail.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Dave Gordon <david.s.gordon@intel.com>
Cc: Arun Siluvery <arun.siluvery@linux.intel.com>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Arun Siluvery <arun.siluvery@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_cmd_parser.c b/drivers/gpu/drm/i915/i915_cmd_parser.c
index ad7d7ab..09932ca 100644
--- a/drivers/gpu/drm/i915/i915_cmd_parser.c
+++ b/drivers/gpu/drm/i915/i915_cmd_parser.c
@@ -124,14 +124,14 @@ static const struct drm_i915_cmd_descriptor common_cmds[] = {
 	CMD(  MI_STORE_DWORD_INDEX,             SMI,   !F,  0xFF,   R  ),
 	CMD(  MI_LOAD_REGISTER_IMM(1),          SMI,   !F,  0xFF,   W,
 	      .reg = { .offset = 1, .mask = 0x007FFFFC, .step = 2 }    ),
-	CMD(  MI_STORE_REGISTER_MEM,            SMI,    F,  1,     W | B,
+	CMD(  MI_STORE_REGISTER_MEM,            SMI,    F,  3,     W | B,
 	      .reg = { .offset = 1, .mask = 0x007FFFFC },
 	      .bits = {{
 			.offset = 0,
 			.mask = MI_GLOBAL_GTT,
 			.expected = 0,
 	      }},						       ),
-	CMD(  MI_LOAD_REGISTER_MEM,             SMI,    F,  1,     W | B,
+	CMD(  MI_LOAD_REGISTER_MEM,             SMI,    F,  3,     W | B,
 	      .reg = { .offset = 1, .mask = 0x007FFFFC },
 	      .bits = {{
 			.offset = 0,
-- 
cgit v0.10.2


From a7546159718171ba9f2a4df37b574c64b1589b5d Mon Sep 17 00:00:00 2001
From: Nick Hoath <nicholas.hoath@intel.com>
Date: Mon, 29 Jun 2015 14:07:32 +0100
Subject: drm/i915/bxt: Clean up bxt_init_clock_gating

Add stepping check for A0 workarounds, and remove the associated
FIXME tags.
Split out unrelated WAs for later condition checking.

v2: Fixed format (PeterL)
v3: Corrected stepping check for WaDisableSDEUnitClockGating
    - Ignoring comment, following hardware spec instead. (ChrisH)
    Added description for TILECTL setting (JonB)

Cc: Peter Lawthers <peter.lawthers@intel.com>
Cc: Chris Harris <chris.harris@intel.com>
Cc: Jon Bloomfield <jon.bloomfield@intel.com>
Signed-off-by: Nick Hoath <nicholas.hoath@intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index 79cd987..64bc77e 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -116,18 +116,24 @@ static void bxt_init_clock_gating(struct drm_device *dev)
 
 	gen9_init_clock_gating(dev);
 
+	/* WaDisableSDEUnitClockGating:bxt */
+	I915_WRITE(GEN8_UCGCTL6, I915_READ(GEN8_UCGCTL6) |
+		   GEN8_SDEUNIT_CLOCK_GATE_DISABLE);
+
 	/*
 	 * FIXME:
-	 * GEN8_SDEUNIT_CLOCK_GATE_DISABLE applies on A0 only.
 	 * GEN8_HDCUNIT_CLOCK_GATE_DISABLE_HDCREQ applies on 3x6 GT SKUs only.
 	 */
-	 /* WaDisableSDEUnitClockGating:bxt */
 	I915_WRITE(GEN8_UCGCTL6, I915_READ(GEN8_UCGCTL6) |
-		   GEN8_SDEUNIT_CLOCK_GATE_DISABLE |
 		   GEN8_HDCUNIT_CLOCK_GATE_DISABLE_HDCREQ);
 
-	/* FIXME: apply on A0 only */
-	I915_WRITE(TILECTL, I915_READ(TILECTL) | TILECTL_TLBPF);
+	if (INTEL_REVID(dev) == BXT_REVID_A0) {
+		/*
+		 * Hardware specification requires this bit to be
+		 * set to 1 for A0
+		 */
+		I915_WRITE(TILECTL, I915_READ(TILECTL) | TILECTL_TLBPF);
+	}
 }
 
 static void i915_pineview_get_mem_freq(struct drm_device *dev)
-- 
cgit v0.10.2


From 437b15b8017e0d946453c10794b0c5d4591cf180 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 4 Sep 2015 16:55:13 +0300
Subject: drm/i915: use pch backlight override on hsw too

Currently the difference between backlight control on HSW vs. BDW/SKL is
that on HSW we modify the duty cycle on the CPU register, and have the
hardware pass the changes on to the PCH registers. We still drive the
PCH PWM on both. While HSW and BDW use the same LPT PCH, BDW does not
pass these messages on to the PCH. Therefore on BDW we need to enable
the PCH override bit, and program the PCH directly. (On SPT PCH, this
mode is the default.) We could as well do this on HSW too, and in fact
I've been told this is what a certain other operating system does. So
use PCH backlight override on HSW too.

This simplifies some follow-up code, but it does have the danger of
breaking backlight on HSW machines. It should work, but mysterious are
the ways of backlight.

While at it, name the related backlight hooks according to the PCH
rather than the CPU for clarity.

Cc: Clint Taylor <clinton.a.taylor@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Clint Taylor <Clinton.A.Taylor@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c
index e2ab3f6..681686e 100644
--- a/drivers/gpu/drm/i915/intel_panel.c
+++ b/drivers/gpu/drm/i915/intel_panel.c
@@ -484,7 +484,7 @@ static u32 intel_panel_compute_brightness(struct intel_connector *connector,
 	return val;
 }
 
-static u32 bdw_get_backlight(struct intel_connector *connector)
+static u32 lpt_get_backlight(struct intel_connector *connector)
 {
 	struct drm_device *dev = connector->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -576,7 +576,7 @@ static u32 intel_panel_get_backlight(struct intel_connector *connector)
 	return val;
 }
 
-static void bdw_set_backlight(struct intel_connector *connector, u32 level)
+static void lpt_set_backlight(struct intel_connector *connector, u32 level)
 {
 	struct drm_device *dev = connector->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -729,6 +729,18 @@ void intel_panel_set_backlight_acpi(struct intel_connector *connector,
 	mutex_unlock(&dev_priv->backlight_lock);
 }
 
+static void lpt_disable_backlight(struct intel_connector *connector)
+{
+	struct drm_device *dev = connector->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	u32 tmp;
+
+	intel_panel_actually_set_backlight(connector, 0);
+
+	tmp = I915_READ(BLC_PWM_PCH_CTL1);
+	I915_WRITE(BLC_PWM_PCH_CTL1, tmp & ~BLM_PCH_PWM_ENABLE);
+}
+
 static void pch_disable_backlight(struct intel_connector *connector)
 {
 	struct drm_device *dev = connector->base.dev;
@@ -829,7 +841,7 @@ void intel_panel_disable_backlight(struct intel_connector *connector)
 	mutex_unlock(&dev_priv->backlight_lock);
 }
 
-static void bdw_enable_backlight(struct intel_connector *connector)
+static void lpt_enable_backlight(struct intel_connector *connector)
 {
 	struct drm_device *dev = connector->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -1243,7 +1255,7 @@ static u32 get_backlight_min_vbt(struct intel_connector *connector)
 	return scale(min, 0, 255, 0, panel->backlight.max);
 }
 
-static int bdw_setup_backlight(struct intel_connector *connector, enum pipe unused)
+static int lpt_setup_backlight(struct intel_connector *connector, enum pipe unused)
 {
 	struct drm_device *dev = connector->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -1260,7 +1272,7 @@ static int bdw_setup_backlight(struct intel_connector *connector, enum pipe unus
 
 	panel->backlight.min = get_backlight_min_vbt(connector);
 
-	val = bdw_get_backlight(connector);
+	val = lpt_get_backlight(connector);
 	panel->backlight.level = intel_panel_compute_brightness(connector, val);
 
 	panel->backlight.enabled = (pch_ctl1 & BLM_PCH_PWM_ENABLE) &&
@@ -1519,12 +1531,12 @@ void intel_panel_init_backlight_funcs(struct drm_device *dev)
 		dev_priv->display.disable_backlight = bxt_disable_backlight;
 		dev_priv->display.set_backlight = bxt_set_backlight;
 		dev_priv->display.get_backlight = bxt_get_backlight;
-	} else if (IS_BROADWELL(dev) || IS_SKYLAKE(dev)) {
-		dev_priv->display.setup_backlight = bdw_setup_backlight;
-		dev_priv->display.enable_backlight = bdw_enable_backlight;
-		dev_priv->display.disable_backlight = pch_disable_backlight;
-		dev_priv->display.set_backlight = bdw_set_backlight;
-		dev_priv->display.get_backlight = bdw_get_backlight;
+	} else if (HAS_PCH_LPT(dev) || HAS_PCH_SPT(dev)) {
+		dev_priv->display.setup_backlight = lpt_setup_backlight;
+		dev_priv->display.enable_backlight = lpt_enable_backlight;
+		dev_priv->display.disable_backlight = lpt_disable_backlight;
+		dev_priv->display.set_backlight = lpt_set_backlight;
+		dev_priv->display.get_backlight = lpt_get_backlight;
 	} else if (HAS_PCH_SPLIT(dev)) {
 		dev_priv->display.setup_backlight = pch_setup_backlight;
 		dev_priv->display.enable_backlight = pch_enable_backlight;
-- 
cgit v0.10.2


From aa17cdb4f8367875d3a53c41e004fc02e983e317 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 4 Sep 2015 16:55:14 +0300
Subject: drm/i915: initialize backlight max from VBT

Normally we determine the backlight PWM modulation frequency (which we
also use as backlight max value) from the backlight registers at module
load time, expecting the registers have been initialized by the BIOS. If
this is not the case, we fail.

The VBT contains the backlight modulation frequency in Hz. Add platform
specific functions to convert the frequency in Hz to backlight PWM
modulation frequency, and use them to initialize the backlight when the
registers are not initialized by the BIOS.

v2: Fix SPT and VLV. Thanks to Clint for the VLV code.

Cc: Clint Taylor <clinton.a.taylor@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Clint Taylor <Clinton.A.Taylor@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 482ddb7..3e46187 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -671,6 +671,8 @@ struct drm_i915_display_funcs {
 			      uint32_t level);
 	void (*disable_backlight)(struct intel_connector *connector);
 	void (*enable_backlight)(struct intel_connector *connector);
+	uint32_t (*backlight_hz_to_pwm)(struct intel_connector *connector,
+					uint32_t hz);
 };
 
 enum forcewake_domain_id {
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 5811eeb..65b5682 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -4648,6 +4648,7 @@ enum skl_disp_power_wells {
 
 #define CBR1_VLV			(VLV_DISPLAY_BASE + 0x70400)
 #define  CBR_PND_DEADLINE_DISABLE	(1<<31)
+#define  CBR_PWM_CLOCK_MUX_SELECT	(1<<30)
 
 /* FIFO watermark sizes etc */
 #define G4X_FIFO_LINE_SIZE	64
@@ -6325,9 +6326,11 @@ enum skl_disp_power_wells {
 #define  FDI_PHASE_SYNC_OVR(pipe) (1<<(FDIA_PHASE_SYNC_SHIFT_OVR - ((pipe) * 2)))
 #define  FDI_PHASE_SYNC_EN(pipe) (1<<(FDIA_PHASE_SYNC_SHIFT_EN - ((pipe) * 2)))
 #define  FDI_BC_BIFURCATION_SELECT	(1 << 12)
+#define  SPT_PWM_GRANULARITY		(1<<0)
 #define SOUTH_CHICKEN2		0xc2004
 #define  FDI_MPHY_IOSFSB_RESET_STATUS	(1<<13)
 #define  FDI_MPHY_IOSFSB_RESET_CTL	(1<<12)
+#define  LPT_PWM_GRANULARITY		(1<<5)
 #define  DPLS_EDP_PPS_FIX_DIS		(1<<0)
 
 #define _FDI_RXA_CHICKEN         0xc200c
diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c
index 681686e..d3a312c 100644
--- a/drivers/gpu/drm/i915/intel_panel.c
+++ b/drivers/gpu/drm/i915/intel_panel.c
@@ -1224,10 +1224,149 @@ static void intel_backlight_device_unregister(struct intel_connector *connector)
 #endif /* CONFIG_BACKLIGHT_CLASS_DEVICE */
 
 /*
- * Note: The setup hooks can't assume pipe is set!
+ * SPT: This value represents the period of the PWM stream in clock periods
+ * multiplied by 16 (default increment) or 128 (alternate increment selected in
+ * SCHICKEN_1 bit 0). PWM clock is 24 MHz.
+ */
+static u32 spt_hz_to_pwm(struct intel_connector *connector, u32 pwm_freq_hz)
+{
+	struct drm_device *dev = connector->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	u32 mul, clock;
+
+	if (I915_READ(SOUTH_CHICKEN1) & SPT_PWM_GRANULARITY)
+		mul = 128;
+	else
+		mul = 16;
+
+	clock = MHz(24);
+
+	return clock / (pwm_freq_hz * mul);
+}
+
+/*
+ * LPT: This value represents the period of the PWM stream in clock periods
+ * multiplied by 128 (default increment) or 16 (alternate increment, selected in
+ * LPT SOUTH_CHICKEN2 register bit 5).
+ */
+static u32 lpt_hz_to_pwm(struct intel_connector *connector, u32 pwm_freq_hz)
+{
+	struct drm_device *dev = connector->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	u32 mul, clock;
+
+	if (I915_READ(SOUTH_CHICKEN2) & LPT_PWM_GRANULARITY)
+		mul = 16;
+	else
+		mul = 128;
+
+	if (dev_priv->pch_id == INTEL_PCH_LPT_DEVICE_ID_TYPE)
+		clock = MHz(135); /* LPT:H */
+	else
+		clock = MHz(24); /* LPT:LP */
+
+	return clock / (pwm_freq_hz * mul);
+}
+
+/*
+ * ILK/SNB/IVB: This value represents the period of the PWM stream in PCH
+ * display raw clocks multiplied by 128.
+ */
+static u32 pch_hz_to_pwm(struct intel_connector *connector, u32 pwm_freq_hz)
+{
+	struct drm_device *dev = connector->base.dev;
+	int clock = MHz(intel_pch_rawclk(dev));
+
+	return clock / (pwm_freq_hz * 128);
+}
+
+/*
+ * Gen2: This field determines the number of time base events (display core
+ * clock frequency/32) in total for a complete cycle of modulated backlight
+ * control.
  *
- * XXX: Query mode clock or hardware clock and program PWM modulation frequency
- * appropriately when it's 0. Use VBT and/or sane defaults.
+ * Gen3: A time base event equals the display core clock ([DevPNV] HRAW clock)
+ * divided by 32.
+ */
+static u32 i9xx_hz_to_pwm(struct intel_connector *connector, u32 pwm_freq_hz)
+{
+	struct drm_device *dev = connector->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int clock;
+
+	if (IS_PINEVIEW(dev))
+		clock = intel_hrawclk(dev);
+	else
+		clock = 1000 * dev_priv->display.get_display_clock_speed(dev);
+
+	return clock / (pwm_freq_hz * 32);
+}
+
+/*
+ * Gen4: This value represents the period of the PWM stream in display core
+ * clocks multiplied by 128.
+ */
+static u32 i965_hz_to_pwm(struct intel_connector *connector, u32 pwm_freq_hz)
+{
+	struct drm_device *dev = connector->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int clock = 1000 * dev_priv->display.get_display_clock_speed(dev);
+
+	return clock / (pwm_freq_hz * 128);
+}
+
+/*
+ * VLV: This value represents the period of the PWM stream in display core
+ * clocks ([DevCTG] 200MHz HRAW clocks) multiplied by 128 or 25MHz S0IX clocks
+ * multiplied by 16. CHV uses a 19.2MHz S0IX clock.
+ */
+static u32 vlv_hz_to_pwm(struct intel_connector *connector, u32 pwm_freq_hz)
+{
+	struct drm_device *dev = connector->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int clock;
+
+	if ((I915_READ(CBR1_VLV) & CBR_PWM_CLOCK_MUX_SELECT) == 0) {
+		if (IS_CHERRYVIEW(dev))
+			return KHz(19200) / (pwm_freq_hz * 16);
+		else
+			return MHz(25) / (pwm_freq_hz * 16);
+	} else {
+		clock = intel_hrawclk(dev);
+		return MHz(clock) / (pwm_freq_hz * 128);
+	}
+}
+
+static u32 get_backlight_max_vbt(struct intel_connector *connector)
+{
+	struct drm_device *dev = connector->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	u16 pwm_freq_hz = dev_priv->vbt.backlight.pwm_freq_hz;
+	u32 pwm;
+
+	if (!pwm_freq_hz) {
+		DRM_DEBUG_KMS("backlight frequency not specified in VBT\n");
+		return 0;
+	}
+
+	if (!dev_priv->display.backlight_hz_to_pwm) {
+		DRM_DEBUG_KMS("backlight frequency setting from VBT currently not supported on this platform\n");
+		return 0;
+	}
+
+	pwm = dev_priv->display.backlight_hz_to_pwm(connector, pwm_freq_hz);
+	if (!pwm) {
+		DRM_DEBUG_KMS("backlight frequency conversion failed\n");
+		return 0;
+	}
+
+	DRM_DEBUG_KMS("backlight frequency %u Hz from VBT\n", pwm_freq_hz);
+
+	return pwm;
+}
+
+/*
+ * Note: The setup hooks can't assume pipe is set!
  */
 static u32 get_backlight_min_vbt(struct intel_connector *connector)
 {
@@ -1267,6 +1406,10 @@ static int lpt_setup_backlight(struct intel_connector *connector, enum pipe unus
 
 	pch_ctl2 = I915_READ(BLC_PWM_PCH_CTL2);
 	panel->backlight.max = pch_ctl2 >> 16;
+
+	if (!panel->backlight.max)
+		panel->backlight.max = get_backlight_max_vbt(connector);
+
 	if (!panel->backlight.max)
 		return -ENODEV;
 
@@ -1293,6 +1436,10 @@ static int pch_setup_backlight(struct intel_connector *connector, enum pipe unus
 
 	pch_ctl2 = I915_READ(BLC_PWM_PCH_CTL2);
 	panel->backlight.max = pch_ctl2 >> 16;
+
+	if (!panel->backlight.max)
+		panel->backlight.max = get_backlight_max_vbt(connector);
+
 	if (!panel->backlight.max)
 		return -ENODEV;
 
@@ -1324,12 +1471,18 @@ static int i9xx_setup_backlight(struct intel_connector *connector, enum pipe unu
 		panel->backlight.active_low_pwm = ctl & BLM_POLARITY_PNV;
 
 	panel->backlight.max = ctl >> 17;
-	if (panel->backlight.combination_mode)
-		panel->backlight.max *= 0xff;
+
+	if (!panel->backlight.max) {
+		panel->backlight.max = get_backlight_max_vbt(connector);
+		panel->backlight.max >>= 1;
+	}
 
 	if (!panel->backlight.max)
 		return -ENODEV;
 
+	if (panel->backlight.combination_mode)
+		panel->backlight.max *= 0xff;
+
 	panel->backlight.min = get_backlight_min_vbt(connector);
 
 	val = i9xx_get_backlight(connector);
@@ -1353,12 +1506,16 @@ static int i965_setup_backlight(struct intel_connector *connector, enum pipe unu
 
 	ctl = I915_READ(BLC_PWM_CTL);
 	panel->backlight.max = ctl >> 16;
-	if (panel->backlight.combination_mode)
-		panel->backlight.max *= 0xff;
+
+	if (!panel->backlight.max)
+		panel->backlight.max = get_backlight_max_vbt(connector);
 
 	if (!panel->backlight.max)
 		return -ENODEV;
 
+	if (panel->backlight.combination_mode)
+		panel->backlight.max *= 0xff;
+
 	panel->backlight.min = get_backlight_min_vbt(connector);
 
 	val = i9xx_get_backlight(connector);
@@ -1398,6 +1555,10 @@ static int vlv_setup_backlight(struct intel_connector *connector, enum pipe pipe
 
 	ctl = I915_READ(VLV_BLC_PWM_CTL(pipe));
 	panel->backlight.max = ctl >> 16;
+
+	if (!panel->backlight.max)
+		panel->backlight.max = get_backlight_max_vbt(connector);
+
 	if (!panel->backlight.max)
 		return -ENODEV;
 
@@ -1424,6 +1585,10 @@ bxt_setup_backlight(struct intel_connector *connector, enum pipe unused)
 	panel->backlight.active_low_pwm = pwm_ctl & BXT_BLC_PWM_POLARITY;
 
 	panel->backlight.max = I915_READ(BXT_BLC_PWM_FREQ1);
+
+	if (!panel->backlight.max)
+		panel->backlight.max = get_backlight_max_vbt(connector);
+
 	if (!panel->backlight.max)
 		return -ENODEV;
 
@@ -1537,12 +1702,17 @@ void intel_panel_init_backlight_funcs(struct drm_device *dev)
 		dev_priv->display.disable_backlight = lpt_disable_backlight;
 		dev_priv->display.set_backlight = lpt_set_backlight;
 		dev_priv->display.get_backlight = lpt_get_backlight;
+		if (HAS_PCH_LPT(dev))
+			dev_priv->display.backlight_hz_to_pwm = lpt_hz_to_pwm;
+		else
+			dev_priv->display.backlight_hz_to_pwm = spt_hz_to_pwm;
 	} else if (HAS_PCH_SPLIT(dev)) {
 		dev_priv->display.setup_backlight = pch_setup_backlight;
 		dev_priv->display.enable_backlight = pch_enable_backlight;
 		dev_priv->display.disable_backlight = pch_disable_backlight;
 		dev_priv->display.set_backlight = pch_set_backlight;
 		dev_priv->display.get_backlight = pch_get_backlight;
+		dev_priv->display.backlight_hz_to_pwm = pch_hz_to_pwm;
 	} else if (IS_VALLEYVIEW(dev)) {
 		if (dev_priv->vbt.has_mipi) {
 			dev_priv->display.setup_backlight = pwm_setup_backlight;
@@ -1556,6 +1726,7 @@ void intel_panel_init_backlight_funcs(struct drm_device *dev)
 			dev_priv->display.disable_backlight = vlv_disable_backlight;
 			dev_priv->display.set_backlight = vlv_set_backlight;
 			dev_priv->display.get_backlight = vlv_get_backlight;
+			dev_priv->display.backlight_hz_to_pwm = vlv_hz_to_pwm;
 		}
 	} else if (IS_GEN4(dev)) {
 		dev_priv->display.setup_backlight = i965_setup_backlight;
@@ -1563,12 +1734,14 @@ void intel_panel_init_backlight_funcs(struct drm_device *dev)
 		dev_priv->display.disable_backlight = i965_disable_backlight;
 		dev_priv->display.set_backlight = i9xx_set_backlight;
 		dev_priv->display.get_backlight = i9xx_get_backlight;
+		dev_priv->display.backlight_hz_to_pwm = i965_hz_to_pwm;
 	} else {
 		dev_priv->display.setup_backlight = i9xx_setup_backlight;
 		dev_priv->display.enable_backlight = i9xx_enable_backlight;
 		dev_priv->display.disable_backlight = i9xx_disable_backlight;
 		dev_priv->display.set_backlight = i9xx_set_backlight;
 		dev_priv->display.get_backlight = i9xx_get_backlight;
+		dev_priv->display.backlight_hz_to_pwm = i9xx_hz_to_pwm;
 	}
 }
 
-- 
cgit v0.10.2


From 4a999d253195db87f18215049cf7a44f795f4ca0 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 4 Sep 2015 16:55:15 +0300
Subject: drm/i915: don't hard code vlv backlight frequency if unset

Fall back to VBT based backlight modulation frequency if it's not
set. Do not hard code.

This could be a problem if there is no VBT.

Cc: Clint Taylor <clinton.a.taylor@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Tested-by: Clint Taylor <Clinton.A.Taylor@intel.com>
Reviewed-by: Clint Taylor <Clinton.A.Taylor@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c
index d3a312c..2034438a 100644
--- a/drivers/gpu/drm/i915/intel_panel.c
+++ b/drivers/gpu/drm/i915/intel_panel.c
@@ -1532,21 +1532,8 @@ static int vlv_setup_backlight(struct intel_connector *connector, enum pipe pipe
 	struct drm_device *dev = connector->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_panel *panel = &connector->panel;
-	enum pipe p;
 	u32 ctl, ctl2, val;
 
-	for_each_pipe(dev_priv, p) {
-		u32 cur_val = I915_READ(VLV_BLC_PWM_CTL(p));
-
-		/* Skip if the modulation freq is already set */
-		if (cur_val & ~BACKLIGHT_DUTY_CYCLE_MASK)
-			continue;
-
-		cur_val &= BACKLIGHT_DUTY_CYCLE_MASK;
-		I915_WRITE(VLV_BLC_PWM_CTL(p), (0xf42 << 16) |
-			   cur_val);
-	}
-
 	if (WARN_ON(pipe != PIPE_A && pipe != PIPE_B))
 		return -ENODEV;
 
-- 
cgit v0.10.2


From bcfe0c0954f3336c44993e5ce444e09ad6087637 Mon Sep 17 00:00:00 2001
From: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Date: Thu, 13 Aug 2015 17:06:39 -0300
Subject: drm: WARN_ON if a modeset driver uses legacy suspend/resume helpers

Legacy s/r hooks are only used for shadow-attaching drivers, warn
when a KMS driver tries to use them.

Signed-off-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c
index 53d09a1..dc93c88 100644
--- a/drivers/gpu/drm/drm_drv.c
+++ b/drivers/gpu/drm/drm_drv.c
@@ -566,6 +566,8 @@ struct drm_device *drm_dev_alloc(struct drm_driver *driver,
 		ret = drm_minor_alloc(dev, DRM_MINOR_CONTROL);
 		if (ret)
 			goto err_minors;
+
+		WARN_ON(driver->suspend || driver->resume);
 	}
 
 	if (drm_core_check_feature(dev, DRIVER_RENDER)) {
-- 
cgit v0.10.2


From fafecc01bcce4a21c0644c79de833e659882d33a Mon Sep 17 00:00:00 2001
From: Michel Thierry <michel.thierry@intel.com>
Date: Sun, 16 Aug 2015 04:02:28 +0100
Subject: drm/mm: Do DRM_MM_CREATE_TOP adj_start calculation after color_adjust

The adj_start calculation for DRM_MM_CREATE_TOP should happen after
mm->color_adjust. There was an inconsistency between
drm_mm_insert_helper_range
and drm_mm_insert_helper, as the later was already updating after
color_adjust.

Didn't spot it before, as color_adjust is only done in systems without
LLC. But I'm not aware of anybody using this test case yet.

Signed-off-by: Michel Thierry <michel.thierry@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_mm.c b/drivers/gpu/drm/drm_mm.c
index 3427b11..04de6fd 100644
--- a/drivers/gpu/drm/drm_mm.c
+++ b/drivers/gpu/drm/drm_mm.c
@@ -267,12 +267,12 @@ static void drm_mm_insert_helper_range(struct drm_mm_node *hole_node,
 	if (adj_end > end)
 		adj_end = end;
 
-	if (flags & DRM_MM_CREATE_TOP)
-		adj_start = adj_end - size;
-
 	if (mm->color_adjust)
 		mm->color_adjust(hole_node, color, &adj_start, &adj_end);
 
+	if (flags & DRM_MM_CREATE_TOP)
+		adj_start = adj_end - size;
+
 	if (alignment) {
 		u64 tmp = adj_start;
 		unsigned rem;
-- 
cgit v0.10.2


From bbda9c1f170e4b31f5d827372ee9a3fb8389f7fa Mon Sep 17 00:00:00 2001
From: Rob Clark <robdclark@gmail.com>
Date: Wed, 19 Aug 2015 19:21:14 -0400
Subject: drm: cleanup modesetting ioctls, one param per line

Since this already confused me once when adding addfb2.1, let's clean up
the header to split params one per line.

Signed-off-by: Rob Clark <robdclark@gmail.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/include/uapi/drm/drm_mode.h b/include/uapi/drm/drm_mode.h
index 359107a..6c11ca4 100644
--- a/include/uapi/drm/drm_mode.h
+++ b/include/uapi/drm/drm_mode.h
@@ -105,8 +105,16 @@
 
 struct drm_mode_modeinfo {
 	__u32 clock;
-	__u16 hdisplay, hsync_start, hsync_end, htotal, hskew;
-	__u16 vdisplay, vsync_start, vsync_end, vtotal, vscan;
+	__u16 hdisplay;
+	__u16 hsync_start;
+	__u16 hsync_end;
+	__u16 htotal;
+	__u16 hskew;
+	__u16 vdisplay;
+	__u16 vsync_start;
+	__u16 vsync_end;
+	__u16 vtotal;
+	__u16 vscan;
 
 	__u32 vrefresh;
 
@@ -124,8 +132,10 @@ struct drm_mode_card_res {
 	__u32 count_crtcs;
 	__u32 count_connectors;
 	__u32 count_encoders;
-	__u32 min_width, max_width;
-	__u32 min_height, max_height;
+	__u32 min_width;
+	__u32 max_width;
+	__u32 min_height;
+	__u32 max_height;
 };
 
 struct drm_mode_crtc {
@@ -135,7 +145,8 @@ struct drm_mode_crtc {
 	__u32 crtc_id; /**< Id */
 	__u32 fb_id; /**< Id of framebuffer */
 
-	__u32 x, y; /**< Position on the frameuffer */
+	__u32 x; /**< x Position on the framebuffer */
+	__u32 y; /**< y Position on the framebuffer */
 
 	__u32 gamma_size;
 	__u32 mode_valid;
@@ -153,12 +164,16 @@ struct drm_mode_set_plane {
 	__u32 flags; /* see above flags */
 
 	/* Signed dest location allows it to be partially off screen */
-	__s32 crtc_x, crtc_y;
-	__u32 crtc_w, crtc_h;
+	__s32 crtc_x;
+	__s32 crtc_y;
+	__u32 crtc_w;
+	__u32 crtc_h;
 
 	/* Source values are 16.16 fixed point */
-	__u32 src_x, src_y;
-	__u32 src_h, src_w;
+	__u32 src_x;
+	__u32 src_y;
+	__u32 src_h;
+	__u32 src_w;
 };
 
 struct drm_mode_get_plane {
@@ -244,7 +259,8 @@ struct drm_mode_get_connector {
 	__u32 connector_type_id;
 
 	__u32 connection;
-	__u32 mm_width, mm_height; /**< HxW in millimeters */
+	__u32 mm_width;  /**< width in millimeters */
+	__u32 mm_height; /**< height in millimeters */
 	__u32 subpixel;
 
 	__u32 pad;
@@ -327,7 +343,8 @@ struct drm_mode_get_blob {
 
 struct drm_mode_fb_cmd {
 	__u32 fb_id;
-	__u32 width, height;
+	__u32 width;
+	__u32 height;
 	__u32 pitch;
 	__u32 bpp;
 	__u32 depth;
@@ -340,7 +357,8 @@ struct drm_mode_fb_cmd {
 
 struct drm_mode_fb_cmd2 {
 	__u32 fb_id;
-	__u32 width, height;
+	__u32 width;
+	__u32 height;
 	__u32 pixel_format; /* fourcc code from drm_fourcc.h */
 	__u32 flags; /* see above flags */
 
-- 
cgit v0.10.2


From fe8660acd8853ad9f95b94a04cce384949ddecfe Mon Sep 17 00:00:00 2001
From: Danilo Cesar Lemes de Paula <danilo.cesar@collabora.co.uk>
Date: Fri, 21 Aug 2015 16:46:14 -0300
Subject: drm/doc: Fixing xml documentation warning

"/**" should be used for kernel-doc documentation only.
It causes a warning with the new "in struct body" format.

Signed-off-by: Danilo Cesar Lemes de Paula <danilo.cesar@collabora.co.uk>
Cc: Randy Dunlap <rdunlap@infradead.org>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Cc: Stephan Mueller <smueller@chronox.de>
Cc: Michal Marek <mmarek@suse.cz>
Cc: linux-kernel@vger.kernel.org
Cc: linux-doc@vger.kernel.org
Cc: intel-gfx <intel-gfx@lists.freedesktop.org>
Cc: dri-devel <dri-devel@lists.freedesktop.org>
Cc: Graham Whaley <graham.whaley@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/include/drm/drm_modeset_lock.h b/include/drm/drm_modeset_lock.h
index 5dd18bf..94938d8 100644
--- a/include/drm/drm_modeset_lock.h
+++ b/include/drm/drm_modeset_lock.h
@@ -43,19 +43,19 @@ struct drm_modeset_acquire_ctx {
 
 	struct ww_acquire_ctx ww_ctx;
 
-	/**
+	/*
 	 * Contended lock: if a lock is contended you should only call
 	 * drm_modeset_backoff() which drops locks and slow-locks the
 	 * contended lock.
 	 */
 	struct drm_modeset_lock *contended;
 
-	/**
+	/*
 	 * list of held locks (drm_modeset_lock)
 	 */
 	struct list_head locked;
 
-	/**
+	/*
 	 * Trylock mode, use only for panic handlers!
 	 */
 	bool trylock_only;
@@ -70,12 +70,12 @@ struct drm_modeset_acquire_ctx {
  * Used for locking CRTCs and other modeset resources.
  */
 struct drm_modeset_lock {
-	/**
+	/*
 	 * modeset lock
 	 */
 	struct ww_mutex mutex;
 
-	/**
+	/*
 	 * Resources that are locked as part of an atomic update are added
 	 * to a list (so we know what to unlock at the end).
 	 */
-- 
cgit v0.10.2


From 26196f7e11f4b2494a2c14eb789fc61a91207e79 Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Tue, 25 Aug 2015 16:26:03 +0200
Subject: drm/atomic-helper: properly annotate functions in kerneldoc

Without the () the markup and more important hyperlinking wont happen.

v2: Also fix nearby type Laurent spotted.

v3: Actually git add. Argh!

Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c
index d432348..a2629ee 100644
--- a/drivers/gpu/drm/drm_atomic_helper.c
+++ b/drivers/gpu/drm/drm_atomic_helper.c
@@ -42,14 +42,14 @@
  * add their own additional internal state.
  *
  * This library also provides default implementations for the check callback in
- * drm_atomic_helper_check and for the commit callback with
- * drm_atomic_helper_commit. But the individual stages and callbacks are expose
- * to allow drivers to mix and match and e.g. use the plane helpers only
+ * drm_atomic_helper_check() and for the commit callback with
+ * drm_atomic_helper_commit(). But the individual stages and callbacks are
+ * exposed to allow drivers to mix and match and e.g. use the plane helpers only
  * together with a driver private modeset implementation.
  *
  * This library also provides implementations for all the legacy driver
- * interfaces on top of the atomic interface. See drm_atomic_helper_set_config,
- * drm_atomic_helper_disable_plane, drm_atomic_helper_disable_plane and the
+ * interfaces on top of the atomic interface. See drm_atomic_helper_set_config(),
+ * drm_atomic_helper_disable_plane(), drm_atomic_helper_disable_plane() and the
  * various functions to implement set_property callbacks. New drivers must not
  * implement these functions themselves but must use the provided helpers.
  */
@@ -1077,7 +1077,7 @@ EXPORT_SYMBOL(drm_atomic_helper_commit);
  * work item, which allows nice concurrent updates on disjoint sets of crtcs.
  *
  * 3. The software state is updated synchronously with
- * drm_atomic_helper_swap_state. Doing this under the protection of all modeset
+ * drm_atomic_helper_swap_state(). Doing this under the protection of all modeset
  * locks means concurrent callers never see inconsistent state. And doing this
  * while it's guaranteed that no relevant async worker runs means that async
  * workers do not need grab any locks. Actually they must not grab locks, for
@@ -1334,7 +1334,7 @@ EXPORT_SYMBOL(drm_atomic_helper_cleanup_planes);
  *
  * 4. Actually commit the hardware state.
  *
- * 5. Call drm_atomic_helper_cleanup_planes with @state, which since step 3
+ * 5. Call drm_atomic_helper_cleanup_planes() with @state, which since step 3
  * contains the old state. Also do any other cleanup required with that state.
  */
 void drm_atomic_helper_swap_state(struct drm_device *dev,
-- 
cgit v0.10.2


From b7bdf0a87add0fead1550533e328a290d8f9da6c Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Tue, 25 Aug 2015 17:20:28 +0200
Subject: drm/fb-helper: Use -errno return in restore_mode_unlocked

Using bool and returning true upon error is very uncommon. Also an int
return value is actually what all the callers which did check it seem
to have expected.

v2: Restore hunk misplaced in a rebase, spotted by Rob.

Cc: Rob Clark <robdclark@gmail.com>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c
index 418d299..859134e 100644
--- a/drivers/gpu/drm/drm_fb_helper.c
+++ b/drivers/gpu/drm/drm_fb_helper.c
@@ -320,11 +320,10 @@ int drm_fb_helper_debug_leave(struct fb_info *info)
 }
 EXPORT_SYMBOL(drm_fb_helper_debug_leave);
 
-static bool restore_fbdev_mode(struct drm_fb_helper *fb_helper)
+static int restore_fbdev_mode(struct drm_fb_helper *fb_helper)
 {
 	struct drm_device *dev = fb_helper->dev;
 	struct drm_plane *plane;
-	bool error = false;
 	int i;
 
 	drm_warn_on_modeset_not_all_locked(dev);
@@ -348,14 +347,15 @@ static bool restore_fbdev_mode(struct drm_fb_helper *fb_helper)
 		if (crtc->funcs->cursor_set) {
 			ret = crtc->funcs->cursor_set(crtc, NULL, 0, 0, 0);
 			if (ret)
-				error = true;
+				return ret;
 		}
 
 		ret = drm_mode_set_config_internal(mode_set);
 		if (ret)
-			error = true;
+			return ret;
 	}
-	return error;
+
+	return 0;
 }
 
 /**
@@ -365,12 +365,15 @@ static bool restore_fbdev_mode(struct drm_fb_helper *fb_helper)
  * This should be called from driver's drm ->lastclose callback
  * when implementing an fbcon on top of kms using this helper. This ensures that
  * the user isn't greeted with a black screen when e.g. X dies.
+ *
+ * RETURNS:
+ * Zero if everything went ok, negative error code otherwise.
  */
-bool drm_fb_helper_restore_fbdev_mode_unlocked(struct drm_fb_helper *fb_helper)
+int drm_fb_helper_restore_fbdev_mode_unlocked(struct drm_fb_helper *fb_helper)
 {
 	struct drm_device *dev = fb_helper->dev;
-	bool ret;
-	bool do_delayed = false;
+	bool do_delayed;
+	int ret;
 
 	drm_modeset_lock_all(dev);
 	ret = restore_fbdev_mode(fb_helper);
diff --git a/include/drm/drm_fb_helper.h b/include/drm/drm_fb_helper.h
index dbab462..67de1f1 100644
--- a/include/drm/drm_fb_helper.h
+++ b/include/drm/drm_fb_helper.h
@@ -136,7 +136,7 @@ int drm_fb_helper_set_par(struct fb_info *info);
 int drm_fb_helper_check_var(struct fb_var_screeninfo *var,
 			    struct fb_info *info);
 
-bool drm_fb_helper_restore_fbdev_mode_unlocked(struct drm_fb_helper *fb_helper);
+int drm_fb_helper_restore_fbdev_mode_unlocked(struct drm_fb_helper *fb_helper);
 
 struct fb_info *drm_fb_helper_alloc_fbi(struct drm_fb_helper *fb_helper);
 void drm_fb_helper_unregister_fbi(struct drm_fb_helper *fb_helper);
@@ -226,10 +226,10 @@ static inline int drm_fb_helper_check_var(struct fb_var_screeninfo *var,
 	return 0;
 }
 
-static inline bool
+static inline int
 drm_fb_helper_restore_fbdev_mode_unlocked(struct drm_fb_helper *fb_helper)
 {
-	return true;
+	return 0;
 }
 
 static inline struct fb_info *
-- 
cgit v0.10.2


From a39a357c6ca3c88ad64e9b51bd7607c06cd6d35d Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Tue, 25 Aug 2015 15:45:11 +0200
Subject: drm: Make drm_fb_unregister/remove accept NULL fb

These functions are used by drivers to release fbdev emulation
buffers. We need to make them resilient to NULL pointers to
make the fbdev compile/runtime knobs not cause Oopses on module
unload.

Cc: Archit Taneja <architt@codeaurora.org>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c
index 33d877c..884690c 100644
--- a/drivers/gpu/drm/drm_crtc.c
+++ b/drivers/gpu/drm/drm_crtc.c
@@ -538,7 +538,12 @@ EXPORT_SYMBOL(drm_framebuffer_reference);
  */
 void drm_framebuffer_unregister_private(struct drm_framebuffer *fb)
 {
-	struct drm_device *dev = fb->dev;
+	struct drm_device *dev;
+
+	if (!fb)
+		return;
+
+	dev = fb->dev;
 
 	mutex_lock(&dev->mode_config.fb_lock);
 	/* Mark fb as reaped and drop idr ref. */
@@ -589,12 +594,17 @@ EXPORT_SYMBOL(drm_framebuffer_cleanup);
  */
 void drm_framebuffer_remove(struct drm_framebuffer *fb)
 {
-	struct drm_device *dev = fb->dev;
+	struct drm_device *dev;
 	struct drm_crtc *crtc;
 	struct drm_plane *plane;
 	struct drm_mode_set set;
 	int ret;
 
+	if (!fb)
+		return;
+
+	dev = fb->dev;
+
 	WARN_ON(!list_empty(&fb->filp_head));
 
 	/*
-- 
cgit v0.10.2


From f64c5573d253ab3f9c01b4056c5f75a889d18502 Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Tue, 25 Aug 2015 15:45:13 +0200
Subject: drm/fb-helper: Add module option to disable fbdev emulation

Faster than recompiling.

Note that restore_fbdev_mode_unlocked is a bit special and the only
one which returns an error code when fbdev isn't there - i915 needs
that one to not fall over with some additional fbcon related restore
code. Everyone else just ignores the return value or only prints a
DRM_DEBUG level message.

Reviewed-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c
index 859134e..ba12f51 100644
--- a/drivers/gpu/drm/drm_fb_helper.c
+++ b/drivers/gpu/drm/drm_fb_helper.c
@@ -39,6 +39,11 @@
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_crtc_helper.h>
 
+static bool drm_fbdev_emulation = true;
+module_param_named(fbdev_emulation, drm_fbdev_emulation, bool, 0600);
+MODULE_PARM_DESC(fbdev_emulation,
+		 "Enable legacy fbdev emulation [default=true]");
+
 static LIST_HEAD(kernel_fb_helper_list);
 
 /**
@@ -99,6 +104,9 @@ int drm_fb_helper_single_add_all_connectors(struct drm_fb_helper *fb_helper)
 	struct drm_connector *connector;
 	int i;
 
+	if (!drm_fbdev_emulation)
+		return 0;
+
 	mutex_lock(&dev->mode_config.mutex);
 	drm_for_each_connector(connector, dev) {
 		struct drm_fb_helper_connector *fb_helper_connector;
@@ -129,6 +137,9 @@ int drm_fb_helper_add_one_connector(struct drm_fb_helper *fb_helper, struct drm_
 	struct drm_fb_helper_connector **temp;
 	struct drm_fb_helper_connector *fb_helper_connector;
 
+	if (!drm_fbdev_emulation)
+		return 0;
+
 	WARN_ON(!mutex_is_locked(&fb_helper->dev->mode_config.mutex));
 	if (fb_helper->connector_count + 1 > fb_helper->connector_info_alloc_count) {
 		temp = krealloc(fb_helper->connector_info, sizeof(struct drm_fb_helper_connector *) * (fb_helper->connector_count + 1), GFP_KERNEL);
@@ -184,6 +195,9 @@ int drm_fb_helper_remove_one_connector(struct drm_fb_helper *fb_helper,
 	struct drm_fb_helper_connector *fb_helper_connector;
 	int i, j;
 
+	if (!drm_fbdev_emulation)
+		return 0;
+
 	WARN_ON(!mutex_is_locked(&fb_helper->dev->mode_config.mutex));
 
 	for (i = 0; i < fb_helper->connector_count; i++) {
@@ -375,6 +389,9 @@ int drm_fb_helper_restore_fbdev_mode_unlocked(struct drm_fb_helper *fb_helper)
 	bool do_delayed;
 	int ret;
 
+	if (!drm_fbdev_emulation)
+		return -ENODEV;
+
 	drm_modeset_lock_all(dev);
 	ret = restore_fbdev_mode(fb_helper);
 
@@ -591,6 +608,9 @@ int drm_fb_helper_init(struct drm_device *dev,
 	struct drm_crtc *crtc;
 	int i;
 
+	if (!drm_fbdev_emulation)
+		return 0;
+
 	if (!max_conn_count)
 		return -EINVAL;
 
@@ -713,6 +733,9 @@ EXPORT_SYMBOL(drm_fb_helper_release_fbi);
 
 void drm_fb_helper_fini(struct drm_fb_helper *fb_helper)
 {
+	if (!drm_fbdev_emulation)
+		return;
+
 	if (!list_empty(&fb_helper->kernel_fb_list)) {
 		list_del(&fb_helper->kernel_fb_list);
 		if (list_empty(&kernel_fb_helper_list)) {
@@ -1933,6 +1956,9 @@ int drm_fb_helper_initial_config(struct drm_fb_helper *fb_helper, int bpp_sel)
 	struct drm_device *dev = fb_helper->dev;
 	int count = 0;
 
+	if (!drm_fbdev_emulation)
+		return 0;
+
 	mutex_lock(&dev->mode_config.mutex);
 	count = drm_fb_helper_probe_connector_modes(fb_helper,
 						    dev->mode_config.max_width,
@@ -1976,6 +2002,9 @@ int drm_fb_helper_hotplug_event(struct drm_fb_helper *fb_helper)
 	struct drm_device *dev = fb_helper->dev;
 	u32 max_width, max_height;
 
+	if (!drm_fbdev_emulation)
+		return 0;
+
 	mutex_lock(&fb_helper->dev->mode_config.mutex);
 	if (!fb_helper->fb || !drm_fb_helper_is_bound(fb_helper)) {
 		fb_helper->delayed_hotplug = true;
-- 
cgit v0.10.2


From f8aeb41c4b7e9de0a4df4ed1ba78cd6ee5c87281 Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Wed, 26 Aug 2015 21:49:42 +0200
Subject: drm/atomic: refuse changing CRTC for planes directly

Very strictly speaking this is possible if you have special hw and
genlocked CRTCs. In general switching a plane between two active CRTC
just won't work so well and is probably not tested at all. Just forbid
it.

I've put this into the core since right now no helper or driver copes
with it, no userspace has code for it and no one asks for it. Yes
there's piles of corner-cases where this would be possible to do this
like:
- switch from inactive crtc to active crtc
- switch from active crtc to inactive crtc
- genlocked display
- invisible plane (to do whatever)
- idle plane hw due to dsi cmd mode/psr
- whatever
but looking at details it's not that easy to implement this correctly.
Hence just put it into the core and add a comment, since the only
userspace we have right now for atomic (weston) doesn't want to use
direct plane switching either.

v2: don't bother with complexity and just outright disallow plane
switching without the intermediate OFF state. Simplifies drivers, we
don't have any hw that could do it anyway and current atomic userspace
(weston) works like this already anyway.

v3: Bikeshed function name (Ville) and add comment (Rob).

v4: Also bikeshed commit message (Rob).

v5: Fix compile warnings reported by 0-day.

Cc: Thierry Reding <thierry.reding@gmail.com>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Daniel Stone <daniels@collabora.com>
Acked-by: Daniel Stone <daniels@collabora.com>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Acked-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

diff --git a/drivers/gpu/drm/drm_atomic.c b/drivers/gpu/drm/drm_atomic.c
index 4349154..e515261 100644
--- a/drivers/gpu/drm/drm_atomic.c
+++ b/drivers/gpu/drm/drm_atomic.c
@@ -663,6 +663,25 @@ drm_atomic_plane_get_property(struct drm_plane *plane,
 	return 0;
 }
 
+static bool
+plane_switching_crtc(struct drm_atomic_state *state,
+		     struct drm_plane *plane,
+		     struct drm_plane_state *plane_state)
+{
+	if (!plane->state->crtc || !plane_state->crtc)
+		return false;
+
+	if (plane->state->crtc == plane_state->crtc)
+		return false;
+
+	/* This could be refined, but currently there's no helper or driver code
+	 * to implement direct switching of active planes nor userspace to take
+	 * advantage of more direct plane switching without the intermediate
+	 * full OFF state.
+	 */
+	return true;
+}
+
 /**
  * drm_atomic_plane_check - check plane state
  * @plane: plane to check
@@ -734,6 +753,12 @@ static int drm_atomic_plane_check(struct drm_plane *plane,
 		return -ENOSPC;
 	}
 
+	if (plane_switching_crtc(state->state, plane, state)) {
+		DRM_DEBUG_ATOMIC("[PLANE:%d] switching CRTC directly\n",
+				 plane->base.id);
+		return -EINVAL;
+	}
+
 	return 0;
 }
 
-- 
cgit v0.10.2


From 96206e2922c114b13cadefa03b9f340b58fee13c Mon Sep 17 00:00:00 2001
From: Bob Paauwe <bob.j.paauwe@intel.com>
Date: Thu, 27 Aug 2015 10:04:13 -0700
Subject: dtrm/edid: Allow comma separated edid binaries. (v3)

Allow comma separated filenames in the edid_firmware parameter.

For example:

edid_firmware=eDP-1:edid/1280x480.bin,DP-2:edid/1920x1080.bin

v2: Use strsep() to simplify parsing of comma seperated string. (Matt)
    Move initial bail before strdup. (Matt)
v3: Changed conditionals after while loop to make more readable (Jani)
    Updated kernel-parameters.txt to reflect changes (Jani)

Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Bob Paauwe <bob.j.paauwe@intel.com>
[danvet: Flatten else control flow and appease checkpatch.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 1d6f045..caf0fd4 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -927,11 +927,11 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
 			The filter can be disabled or changed to another
 			driver later using sysfs.
 
-	drm_kms_helper.edid_firmware=[<connector>:]<file>
-			Broken monitors, graphic adapters and KVMs may
-			send no or incorrect EDID data sets. This parameter
-			allows to specify an EDID data set in the
-			/lib/firmware directory that is used instead.
+	drm_kms_helper.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>]
+			Broken monitors, graphic adapters, KVMs and EDIDless
+			panels may send no or incorrect EDID data sets.
+			This parameter allows to specify an EDID data sets
+			in the /lib/firmware directory that are used instead.
 			Generic built-in EDID data sets are used, if one of
 			edid/1024x768.bin, edid/1280x1024.bin,
 			edid/1680x1050.bin, or edid/1920x1080.bin is given
@@ -940,7 +940,10 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
 			available in Documentation/EDID/HOWTO.txt. An EDID
 			data set will only be used for a particular connector,
 			if its name and a colon are prepended to the EDID
-			name.
+			name. Each connector may use a unique EDID data
+			set by separating the files with a comma.  An EDID
+			data set with no connector name will be used for
+			any connectors not explicitly specified.
 
 	dscc4.setup=	[NET]
 
diff --git a/drivers/gpu/drm/drm_edid_load.c b/drivers/gpu/drm/drm_edid_load.c
index c5605fe..1f445e9 100644
--- a/drivers/gpu/drm/drm_edid_load.c
+++ b/drivers/gpu/drm/drm_edid_load.c
@@ -264,20 +264,43 @@ out:
 int drm_load_edid_firmware(struct drm_connector *connector)
 {
 	const char *connector_name = connector->name;
-	char *edidname = edid_firmware, *last, *colon;
+	char *edidname, *last, *colon, *fwstr, *edidstr, *fallback = NULL;
 	int ret;
 	struct edid *edid;
 
-	if (*edidname == '\0')
+	if (edid_firmware[0] == '\0')
 		return 0;
 
-	colon = strchr(edidname, ':');
-	if (colon != NULL) {
-		if (strncmp(connector_name, edidname, colon - edidname))
-			return 0;
-		edidname = colon + 1;
-		if (*edidname == '\0')
+	/*
+	 * If there are multiple edid files specified and separated
+	 * by commas, search through the list looking for one that
+	 * matches the connector.
+	 *
+	 * If there's one or more that don't't specify a connector, keep
+	 * the last one found one as a fallback.
+	 */
+	fwstr = kstrdup(edid_firmware, GFP_KERNEL);
+	edidstr = fwstr;
+
+	while ((edidname = strsep(&edidstr, ","))) {
+		colon = strchr(edidname, ':');
+		if (colon != NULL) {
+			if (strncmp(connector_name, edidname, colon - edidname))
+				continue;
+			edidname = colon + 1;
+			break;
+		}
+
+		if (*edidname != '\0') /* corner case: multiple ',' */
+			fallback = edidname;
+	}
+
+	if (!edidname) {
+		if (!fallback) {
+			kfree(fwstr);
 			return 0;
+		}
+		edidname = fallback;
 	}
 
 	last = edidname + strlen(edidname) - 1;
@@ -285,6 +308,8 @@ int drm_load_edid_firmware(struct drm_connector *connector)
 		*last = '\0';
 
 	edid = edid_load(connector, edidname, connector_name);
+	kfree(fwstr);
+
 	if (IS_ERR_OR_NULL(edid))
 		return 0;
 
-- 
cgit v0.10.2


From a5b62374618f629f5d9d6d86f3bef9b552fc980b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 31 Aug 2015 15:09:25 +0300
Subject: drm: Constify generic_edid_names[]
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Make generic_edid_names[] const since it's supposed to be immutable.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_edid_load.c b/drivers/gpu/drm/drm_edid_load.c
index 1f445e9..698b8c3 100644
--- a/drivers/gpu/drm/drm_edid_load.c
+++ b/drivers/gpu/drm/drm_edid_load.c
@@ -32,7 +32,7 @@ MODULE_PARM_DESC(edid_firmware, "Do not probe monitor, use specified EDID blob "
 	"from built-in data or /lib/firmware instead. ");
 
 #define GENERIC_EDIDS 6
-static const char *generic_edid_name[GENERIC_EDIDS] = {
+static const char * const generic_edid_name[GENERIC_EDIDS] = {
 	"edid/800x600.bin",
 	"edid/1024x768.bin",
 	"edid/1280x1024.bin",
-- 
cgit v0.10.2


From b7c914b3d94e93bd9b442226231b0bba84c9fa2c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 31 Aug 2015 15:09:26 +0300
Subject: drm: Constify TV mode names
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Make the mode names passed to drm_mode_create_tv_properties() const.

 drivers/gpu/drm/i2c/ch7006.ko:
-.rodata                       596
+.rodata                       664
-.data                        7064
+.data                        6992

 drivers/gpu/drm/nouveau/nouveau.ko:
-.rodata                      146808
+.rodata                      146904
-.data                        178624
+.data                        178528

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c
index 884690c..474f328 100644
--- a/drivers/gpu/drm/drm_crtc.c
+++ b/drivers/gpu/drm/drm_crtc.c
@@ -1519,7 +1519,7 @@ EXPORT_SYMBOL(drm_mode_create_dvi_i_properties);
  */
 int drm_mode_create_tv_properties(struct drm_device *dev,
 				  unsigned int num_modes,
-				  char *modes[])
+				  const char * const modes[])
 {
 	struct drm_property *tv_selector;
 	struct drm_property *tv_subconnector;
diff --git a/drivers/gpu/drm/i2c/ch7006_drv.c b/drivers/gpu/drm/i2c/ch7006_drv.c
index 51fa323..d9a72c9 100644
--- a/drivers/gpu/drm/i2c/ch7006_drv.c
+++ b/drivers/gpu/drm/i2c/ch7006_drv.c
@@ -119,8 +119,8 @@ static void ch7006_encoder_mode_set(struct drm_encoder *encoder,
 	struct ch7006_encoder_params *params = &priv->params;
 	struct ch7006_state *state = &priv->state;
 	uint8_t *regs = state->regs;
-	struct ch7006_mode *mode = priv->mode;
-	struct ch7006_tv_norm_info *norm = &ch7006_tv_norms[priv->norm];
+	const struct ch7006_mode *mode = priv->mode;
+	const struct ch7006_tv_norm_info *norm = &ch7006_tv_norms[priv->norm];
 	int start_active;
 
 	ch7006_dbg(client, "\n");
@@ -226,7 +226,7 @@ static int ch7006_encoder_get_modes(struct drm_encoder *encoder,
 				    struct drm_connector *connector)
 {
 	struct ch7006_priv *priv = to_ch7006_priv(encoder);
-	struct ch7006_mode *mode;
+	const struct ch7006_mode *mode;
 	int n = 0;
 
 	for (mode = ch7006_modes; mode->mode.clock; mode++) {
diff --git a/drivers/gpu/drm/i2c/ch7006_mode.c b/drivers/gpu/drm/i2c/ch7006_mode.c
index 9b83574..f4dca53 100644
--- a/drivers/gpu/drm/i2c/ch7006_mode.c
+++ b/drivers/gpu/drm/i2c/ch7006_mode.c
@@ -26,7 +26,7 @@
 
 #include "ch7006_priv.h"
 
-char *ch7006_tv_norm_names[] = {
+const char * const ch7006_tv_norm_names[] = {
 	[TV_NORM_PAL] = "PAL",
 	[TV_NORM_PAL_M] = "PAL-M",
 	[TV_NORM_PAL_N] = "PAL-N",
@@ -202,7 +202,7 @@ void ch7006_setup_levels(struct drm_encoder *encoder)
 	struct i2c_client *client = drm_i2c_encoder_get_client(encoder);
 	struct ch7006_priv *priv = to_ch7006_priv(encoder);
 	uint8_t *regs = priv->state.regs;
-	struct ch7006_tv_norm_info *norm = &ch7006_tv_norms[priv->norm];
+	const struct ch7006_tv_norm_info *norm = &ch7006_tv_norms[priv->norm];
 	int gain;
 	int black_level;
 
diff --git a/drivers/gpu/drm/i2c/ch7006_priv.h b/drivers/gpu/drm/i2c/ch7006_priv.h
index ce57784..cef6ce7 100644
--- a/drivers/gpu/drm/i2c/ch7006_priv.h
+++ b/drivers/gpu/drm/i2c/ch7006_priv.h
@@ -106,7 +106,7 @@ extern int ch7006_debug;
 extern char *ch7006_tv_norm;
 extern int ch7006_scale;
 
-extern char *ch7006_tv_norm_names[];
+extern const char * const ch7006_tv_norm_names[];
 extern struct ch7006_tv_norm_info ch7006_tv_norms[];
 extern struct ch7006_mode ch7006_modes[];
 
diff --git a/drivers/gpu/drm/i915/intel_tv.c b/drivers/gpu/drm/i915/intel_tv.c
index 0568ae6..590ceab 100644
--- a/drivers/gpu/drm/i915/intel_tv.c
+++ b/drivers/gpu/drm/i915/intel_tv.c
@@ -1579,7 +1579,7 @@ intel_tv_init(struct drm_device *dev)
 	struct intel_encoder *intel_encoder;
 	struct intel_connector *intel_connector;
 	u32 tv_dac_on, tv_dac_off, save_tv_dac;
-	char *tv_format_names[ARRAY_SIZE(tv_modes)];
+	const char *tv_format_names[ARRAY_SIZE(tv_modes)];
 	int i, initial_mode = 0;
 
 	if ((I915_READ(TV_CTL) & TV_FUSE_STATE_MASK) == TV_FUSE_STATE_DISABLED)
@@ -1677,7 +1677,7 @@ intel_tv_init(struct drm_device *dev)
 
 	/* Create TV properties then attach current values */
 	for (i = 0; i < ARRAY_SIZE(tv_modes); i++)
-		tv_format_names[i] = (char *)tv_modes[i].name;
+		tv_format_names[i] = tv_modes[i].name;
 	drm_mode_create_tv_properties(dev,
 				      ARRAY_SIZE(tv_modes),
 				      tv_format_names);
diff --git a/drivers/gpu/drm/nouveau/dispnv04/tvmodesnv17.c b/drivers/gpu/drm/nouveau/dispnv04/tvmodesnv17.c
index 08c6f5e..903c473 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/tvmodesnv17.c
+++ b/drivers/gpu/drm/nouveau/dispnv04/tvmodesnv17.c
@@ -32,7 +32,7 @@
 #include "hw.h"
 #include "tvnv17.h"
 
-char *nv17_tv_norm_names[NUM_TV_NORMS] = {
+const char * const nv17_tv_norm_names[NUM_TV_NORMS] = {
 	[TV_NORM_PAL] = "PAL",
 	[TV_NORM_PAL_M] = "PAL-M",
 	[TV_NORM_PAL_N] = "PAL-N",
diff --git a/drivers/gpu/drm/nouveau/dispnv04/tvnv17.h b/drivers/gpu/drm/nouveau/dispnv04/tvnv17.h
index 459910b..1b07521c 100644
--- a/drivers/gpu/drm/nouveau/dispnv04/tvnv17.h
+++ b/drivers/gpu/drm/nouveau/dispnv04/tvnv17.h
@@ -85,7 +85,7 @@ struct nv17_tv_encoder {
 #define to_tv_enc(x) container_of(nouveau_encoder(x),		\
 				  struct nv17_tv_encoder, base)
 
-extern char *nv17_tv_norm_names[NUM_TV_NORMS];
+extern const char * const nv17_tv_norm_names[NUM_TV_NORMS];
 
 extern struct nv17_tv_norm_params {
 	enum {
diff --git a/include/drm/drm_crtc.h b/include/drm/drm_crtc.h
index faaeff7..75f49c1 100644
--- a/include/drm/drm_crtc.h
+++ b/include/drm/drm_crtc.h
@@ -1390,7 +1390,7 @@ extern int drm_property_add_enum(struct drm_property *property, int index,
 extern int drm_mode_create_dvi_i_properties(struct drm_device *dev);
 extern int drm_mode_create_tv_properties(struct drm_device *dev,
 					 unsigned int num_modes,
-					 char *modes[]);
+					 const char * const modes[]);
 extern int drm_mode_create_scaling_mode_property(struct drm_device *dev);
 extern int drm_mode_create_aspect_ratio_property(struct drm_device *dev);
 extern int drm_mode_create_dirty_info_property(struct drm_device *dev);
-- 
cgit v0.10.2


From 53edb2c62e1a884f53a74899cc7e9558fd89051a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 31 Aug 2015 15:09:27 +0300
Subject: drm/i2c/ch7006: Constify ch7006_tv_norms[] and ch7006_modes[]
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

 drivers/gpu/drm/i2c/ch7006.ko:
-.text                        5913
+.text                        5897
-.rodata                       664
+.rodata                      7256
-.data                        6992
+.data                         416

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i2c/ch7006_mode.c b/drivers/gpu/drm/i2c/ch7006_mode.c
index f4dca53..bb5f67f 100644
--- a/drivers/gpu/drm/i2c/ch7006_mode.c
+++ b/drivers/gpu/drm/i2c/ch7006_mode.c
@@ -46,7 +46,7 @@ const char * const ch7006_tv_norm_names[] = {
 		.vtotal = 625,					\
 		.hvirtual = 810
 
-struct ch7006_tv_norm_info ch7006_tv_norms[] = {
+const struct ch7006_tv_norm_info ch7006_tv_norms[] = {
 	[TV_NORM_NTSC_M] = {
 		NTSC_LIKE_TIMINGS,
 		.black_level = 0.339 * fixed1,
@@ -142,7 +142,7 @@ struct ch7006_tv_norm_info ch7006_tv_norms[] = {
 
 #define PAL_LIKE (1 << TV_NORM_PAL | 1 << TV_NORM_PAL_N | 1 << TV_NORM_PAL_NC)
 
-struct ch7006_mode ch7006_modes[] = {
+const struct ch7006_mode ch7006_modes[] = {
 	MODE(21000, 512, 384, 840, 500, N, N, 181.797557582, 5_4, 0x6, PAL_LIKE),
 	MODE(26250, 512, 384, 840, 625, N, N, 145.438046066, 1_1, 0x1, PAL_LIKE),
 	MODE(20140, 512, 384, 800, 420, N, N, 213.257083791, 5_4, 0x4, NTSC_LIKE),
@@ -171,11 +171,11 @@ struct ch7006_mode ch7006_modes[] = {
 	{}
 };
 
-struct ch7006_mode *ch7006_lookup_mode(struct drm_encoder *encoder,
-				       const struct drm_display_mode *drm_mode)
+const struct ch7006_mode *ch7006_lookup_mode(struct drm_encoder *encoder,
+					     const struct drm_display_mode *drm_mode)
 {
 	struct ch7006_priv *priv = to_ch7006_priv(encoder);
-	struct ch7006_mode *mode;
+	const struct ch7006_mode *mode;
 
 	for (mode = ch7006_modes; mode->mode.clock; mode++) {
 
@@ -233,8 +233,8 @@ void ch7006_setup_subcarrier(struct drm_encoder *encoder)
 	struct i2c_client *client = drm_i2c_encoder_get_client(encoder);
 	struct ch7006_priv *priv = to_ch7006_priv(encoder);
 	struct ch7006_state *state = &priv->state;
-	struct ch7006_tv_norm_info *norm = &ch7006_tv_norms[priv->norm];
-	struct ch7006_mode *mode = priv->mode;
+	const struct ch7006_tv_norm_info *norm = &ch7006_tv_norms[priv->norm];
+	const struct ch7006_mode *mode = priv->mode;
 	uint32_t subc_inc;
 
 	subc_inc = round_fixed((mode->subc_coeff >> 8)
@@ -257,7 +257,7 @@ void ch7006_setup_pll(struct drm_encoder *encoder)
 	struct i2c_client *client = drm_i2c_encoder_get_client(encoder);
 	struct ch7006_priv *priv = to_ch7006_priv(encoder);
 	uint8_t *regs = priv->state.regs;
-	struct ch7006_mode *mode = priv->mode;
+	const struct ch7006_mode *mode = priv->mode;
 	int n, best_n = 0;
 	int m, best_m = 0;
 	int freq, best_freq = 0;
@@ -328,9 +328,9 @@ void ch7006_setup_properties(struct drm_encoder *encoder)
 	struct i2c_client *client = drm_i2c_encoder_get_client(encoder);
 	struct ch7006_priv *priv = to_ch7006_priv(encoder);
 	struct ch7006_state *state = &priv->state;
-	struct ch7006_tv_norm_info *norm = &ch7006_tv_norms[priv->norm];
-	struct ch7006_mode *ch_mode = priv->mode;
-	struct drm_display_mode *mode = &ch_mode->mode;
+	const struct ch7006_tv_norm_info *norm = &ch7006_tv_norms[priv->norm];
+	const struct ch7006_mode *ch_mode = priv->mode;
+	const struct drm_display_mode *mode = &ch_mode->mode;
 	uint8_t *regs = state->regs;
 	int flicker, contrast, hpos, vpos;
 	uint64_t scale, aspect;
diff --git a/drivers/gpu/drm/i2c/ch7006_priv.h b/drivers/gpu/drm/i2c/ch7006_priv.h
index cef6ce7..dc6414a 100644
--- a/drivers/gpu/drm/i2c/ch7006_priv.h
+++ b/drivers/gpu/drm/i2c/ch7006_priv.h
@@ -78,7 +78,7 @@ struct ch7006_state {
 
 struct ch7006_priv {
 	struct ch7006_encoder_params params;
-	struct ch7006_mode *mode;
+	const struct ch7006_mode *mode;
 
 	struct ch7006_state state;
 	struct ch7006_state saved_state;
@@ -107,11 +107,11 @@ extern char *ch7006_tv_norm;
 extern int ch7006_scale;
 
 extern const char * const ch7006_tv_norm_names[];
-extern struct ch7006_tv_norm_info ch7006_tv_norms[];
-extern struct ch7006_mode ch7006_modes[];
+extern const struct ch7006_tv_norm_info ch7006_tv_norms[];
+extern const struct ch7006_mode ch7006_modes[];
 
-struct ch7006_mode *ch7006_lookup_mode(struct drm_encoder *encoder,
-				       const struct drm_display_mode *drm_mode);
+const struct ch7006_mode *ch7006_lookup_mode(struct drm_encoder *encoder,
+					     const struct drm_display_mode *drm_mode);
 
 void ch7006_setup_levels(struct drm_encoder *encoder);
 void ch7006_setup_subcarrier(struct drm_encoder *encoder);
-- 
cgit v0.10.2


From 825926d8e0bfa6f933706236603bb74a8cea002c Mon Sep 17 00:00:00 2001
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Date: Thu, 27 Aug 2015 13:58:09 +0200
Subject: drm/atomic: Make sure lock is held in trylock contexts.

This will make sure we get a lockdep spat in all cases
even if the context is a complete garbage pointer.

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_modeset_lock.c b/drivers/gpu/drm/drm_modeset_lock.c
index fba321c..6675b14 100644
--- a/drivers/gpu/drm/drm_modeset_lock.c
+++ b/drivers/gpu/drm/drm_modeset_lock.c
@@ -307,6 +307,8 @@ static inline int modeset_lock(struct drm_modeset_lock *lock,
 	WARN_ON(ctx->contended);
 
 	if (ctx->trylock_only) {
+		lockdep_assert_held(&ctx->ww_ctx);
+
 		if (!ww_mutex_trylock(&lock->mutex))
 			return -EBUSY;
 		else
-- 
cgit v0.10.2


From 844f9111f6f54f88eb2f0fac121b82ce77193866 Mon Sep 17 00:00:00 2001
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Date: Wed, 2 Sep 2015 10:42:40 +0200
Subject: drm/atomic: Make prepare_fb/cleanup_fb only take state, v3.

This removes the need to separately track fb changes i915.
That will be done as a separate commit, however.

Changes since v1:
- Add dri-devel to cc.
- Fix a check in intel's prepare and cleanup fb to take rotation
  into account.
Changes since v2:
- Split out i915 changes to a separate commit.

Cc: dri-devel@lists.freedesktop.org
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>
[danvet: Squash in msm fixup from Maarten.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_plane.c b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_plane.c
index be9fa82..36fda86 100644
--- a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_plane.c
+++ b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_plane.c
@@ -712,11 +712,13 @@ static int atmel_hlcdc_plane_atomic_check(struct drm_plane *p,
 }
 
 static int atmel_hlcdc_plane_prepare_fb(struct drm_plane *p,
-					struct drm_framebuffer *fb,
 					const struct drm_plane_state *new_state)
 {
 	struct atmel_hlcdc_plane *plane = drm_plane_to_atmel_hlcdc_plane(p);
 
+	if (!new_state->fb)
+		return 0;
+
 	return atmel_hlcdc_layer_update_start(&plane->layer);
 }
 
diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c
index a2629ee..9b0c476 100644
--- a/drivers/gpu/drm/drm_atomic_helper.c
+++ b/drivers/gpu/drm/drm_atomic_helper.c
@@ -1111,17 +1111,14 @@ int drm_atomic_helper_prepare_planes(struct drm_device *dev,
 		const struct drm_plane_helper_funcs *funcs;
 		struct drm_plane *plane = state->planes[i];
 		struct drm_plane_state *plane_state = state->plane_states[i];
-		struct drm_framebuffer *fb;
 
 		if (!plane)
 			continue;
 
 		funcs = plane->helper_private;
 
-		fb = plane_state->fb;
-
-		if (fb && funcs->prepare_fb) {
-			ret = funcs->prepare_fb(plane, fb, plane_state);
+		if (funcs->prepare_fb) {
+			ret = funcs->prepare_fb(plane, plane_state);
 			if (ret)
 				goto fail;
 		}
@@ -1134,17 +1131,14 @@ fail:
 		const struct drm_plane_helper_funcs *funcs;
 		struct drm_plane *plane = state->planes[i];
 		struct drm_plane_state *plane_state = state->plane_states[i];
-		struct drm_framebuffer *fb;
 
 		if (!plane)
 			continue;
 
 		funcs = plane->helper_private;
 
-		fb = state->plane_states[i]->fb;
-
-		if (fb && funcs->cleanup_fb)
-			funcs->cleanup_fb(plane, fb, plane_state);
+		if (funcs->cleanup_fb)
+			funcs->cleanup_fb(plane, plane_state);
 
 	}
 
@@ -1300,14 +1294,11 @@ void drm_atomic_helper_cleanup_planes(struct drm_device *dev,
 
 	for_each_plane_in_state(old_state, plane, plane_state, i) {
 		const struct drm_plane_helper_funcs *funcs;
-		struct drm_framebuffer *old_fb;
 
 		funcs = plane->helper_private;
 
-		old_fb = plane_state->fb;
-
-		if (old_fb && funcs->cleanup_fb)
-			funcs->cleanup_fb(plane, old_fb, plane_state);
+		if (funcs->cleanup_fb)
+			funcs->cleanup_fb(plane, plane_state);
 	}
 }
 EXPORT_SYMBOL(drm_atomic_helper_cleanup_planes);
diff --git a/drivers/gpu/drm/drm_plane_helper.c b/drivers/gpu/drm/drm_plane_helper.c
index 5e5a07a..d384ebc 100644
--- a/drivers/gpu/drm/drm_plane_helper.c
+++ b/drivers/gpu/drm/drm_plane_helper.c
@@ -426,7 +426,7 @@ int drm_plane_helper_commit(struct drm_plane *plane,
 
 	if (plane_funcs->prepare_fb && plane_state->fb &&
 	    plane_state->fb != old_fb) {
-		ret = plane_funcs->prepare_fb(plane, plane_state->fb,
+		ret = plane_funcs->prepare_fb(plane,
 					      plane_state);
 		if (ret)
 			goto out;
@@ -479,8 +479,8 @@ int drm_plane_helper_commit(struct drm_plane *plane,
 		ret = 0;
 	}
 
-	if (plane_funcs->cleanup_fb && old_fb)
-		plane_funcs->cleanup_fb(plane, old_fb, plane_state);
+	if (plane_funcs->cleanup_fb)
+		plane_funcs->cleanup_fb(plane, plane_state);
 out:
 	if (plane_state) {
 		if (plane->funcs->atomic_destroy_state)
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index ca9278b..4eb03b4 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -13313,10 +13313,10 @@ static void intel_shared_dpll_init(struct drm_device *dev)
  */
 int
 intel_prepare_plane_fb(struct drm_plane *plane,
-		       struct drm_framebuffer *fb,
 		       const struct drm_plane_state *new_state)
 {
 	struct drm_device *dev = plane->dev;
+	struct drm_framebuffer *fb = new_state->fb;
 	struct intel_plane *intel_plane = to_intel_plane(plane);
 	struct drm_i915_gem_object *obj = intel_fb_obj(fb);
 	struct drm_i915_gem_object *old_obj = intel_fb_obj(plane->fb);
@@ -13354,19 +13354,18 @@ intel_prepare_plane_fb(struct drm_plane *plane,
  */
 void
 intel_cleanup_plane_fb(struct drm_plane *plane,
-		       struct drm_framebuffer *fb,
 		       const struct drm_plane_state *old_state)
 {
 	struct drm_device *dev = plane->dev;
-	struct drm_i915_gem_object *obj = intel_fb_obj(fb);
+	struct drm_i915_gem_object *obj = intel_fb_obj(old_state->fb);
 
-	if (WARN_ON(!obj))
+	if (!obj)
 		return;
 
 	if (plane->type != DRM_PLANE_TYPE_CURSOR ||
 	    !INTEL_INFO(dev)->cursor_needs_physical) {
 		mutex_lock(&dev->struct_mutex);
-		intel_unpin_fb_obj(fb, old_state);
+		intel_unpin_fb_obj(old_state->fb, old_state);
 		mutex_unlock(&dev->struct_mutex);
 	}
 }
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 2b9e6f9..bfd1204 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -1038,10 +1038,8 @@ void intel_finish_page_flip(struct drm_device *dev, int pipe);
 void intel_finish_page_flip_plane(struct drm_device *dev, int plane);
 void intel_check_page_flip(struct drm_device *dev, int pipe);
 int intel_prepare_plane_fb(struct drm_plane *plane,
-			   struct drm_framebuffer *fb,
 			   const struct drm_plane_state *new_state);
 void intel_cleanup_plane_fb(struct drm_plane *plane,
-			    struct drm_framebuffer *fb,
 			    const struct drm_plane_state *old_state);
 int intel_plane_atomic_get_property(struct drm_plane *plane,
 				    const struct drm_plane_state *state,
diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_plane.c b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_plane.c
index e9dee36..30d57e7 100644
--- a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_plane.c
+++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_plane.c
@@ -99,22 +99,28 @@ static const struct drm_plane_funcs mdp4_plane_funcs = {
 };
 
 static int mdp4_plane_prepare_fb(struct drm_plane *plane,
-		struct drm_framebuffer *fb,
 		const struct drm_plane_state *new_state)
 {
 	struct mdp4_plane *mdp4_plane = to_mdp4_plane(plane);
 	struct mdp4_kms *mdp4_kms = get_kms(plane);
+	struct drm_framebuffer *fb = new_state->fb;
+
+	if (!fb)
+		return 0;
 
 	DBG("%s: prepare: FB[%u]", mdp4_plane->name, fb->base.id);
 	return msm_framebuffer_prepare(fb, mdp4_kms->id);
 }
 
 static void mdp4_plane_cleanup_fb(struct drm_plane *plane,
-		struct drm_framebuffer *fb,
 		const struct drm_plane_state *old_state)
 {
 	struct mdp4_plane *mdp4_plane = to_mdp4_plane(plane);
 	struct mdp4_kms *mdp4_kms = get_kms(plane);
+	struct drm_framebuffer *fb = old_state->fb;
+
+	if (!fb)
+		return;
 
 	DBG("%s: cleanup: FB[%u]", mdp4_plane->name, fb->base.id);
 	msm_framebuffer_cleanup(fb, mdp4_kms->id);
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c
index 07fb62f..a0f5ff0 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c
@@ -250,22 +250,28 @@ static const struct drm_plane_funcs mdp5_plane_funcs = {
 };
 
 static int mdp5_plane_prepare_fb(struct drm_plane *plane,
-		struct drm_framebuffer *fb,
 		const struct drm_plane_state *new_state)
 {
 	struct mdp5_plane *mdp5_plane = to_mdp5_plane(plane);
 	struct mdp5_kms *mdp5_kms = get_kms(plane);
+	struct drm_framebuffer *fb = new_state->fb;
+
+	if (!new_state->fb)
+		return 0;
 
 	DBG("%s: prepare: FB[%u]", mdp5_plane->name, fb->base.id);
 	return msm_framebuffer_prepare(fb, mdp5_kms->id);
 }
 
 static void mdp5_plane_cleanup_fb(struct drm_plane *plane,
-		struct drm_framebuffer *fb,
 		const struct drm_plane_state *old_state)
 {
 	struct mdp5_plane *mdp5_plane = to_mdp5_plane(plane);
 	struct mdp5_kms *mdp5_kms = get_kms(plane);
+	struct drm_framebuffer *fb = old_state->fb;
+
+	if (!fb)
+		return;
 
 	DBG("%s: cleanup: FB[%u]", mdp5_plane->name, fb->base.id);
 	msm_framebuffer_cleanup(fb, mdp5_kms->id);
diff --git a/drivers/gpu/drm/omapdrm/omap_plane.c b/drivers/gpu/drm/omapdrm/omap_plane.c
index 0989046..09e363b 100644
--- a/drivers/gpu/drm/omapdrm/omap_plane.c
+++ b/drivers/gpu/drm/omapdrm/omap_plane.c
@@ -60,17 +60,19 @@ to_omap_plane_state(struct drm_plane_state *state)
 }
 
 static int omap_plane_prepare_fb(struct drm_plane *plane,
-				 struct drm_framebuffer *fb,
 				 const struct drm_plane_state *new_state)
 {
-	return omap_framebuffer_pin(fb);
+	if (!new_state->fb)
+		return 0;
+
+	return omap_framebuffer_pin(new_state->fb);
 }
 
 static void omap_plane_cleanup_fb(struct drm_plane *plane,
-				  struct drm_framebuffer *fb,
 				  const struct drm_plane_state *old_state)
 {
-	omap_framebuffer_unpin(fb);
+	if (old_state->fb)
+		omap_framebuffer_unpin(old_state->fb);
 }
 
 static void omap_plane_atomic_update(struct drm_plane *plane,
diff --git a/drivers/gpu/drm/tegra/dc.c b/drivers/gpu/drm/tegra/dc.c
index ddefb85..b4af4ab 100644
--- a/drivers/gpu/drm/tegra/dc.c
+++ b/drivers/gpu/drm/tegra/dc.c
@@ -480,14 +480,12 @@ static const struct drm_plane_funcs tegra_primary_plane_funcs = {
 };
 
 static int tegra_plane_prepare_fb(struct drm_plane *plane,
-				  struct drm_framebuffer *fb,
 				  const struct drm_plane_state *new_state)
 {
 	return 0;
 }
 
 static void tegra_plane_cleanup_fb(struct drm_plane *plane,
-				   struct drm_framebuffer *fb,
 				   const struct drm_plane_state *old_fb)
 {
 }
diff --git a/include/drm/drm_plane_helper.h b/include/drm/drm_plane_helper.h
index dda401b..5a7f9d4 100644
--- a/include/drm/drm_plane_helper.h
+++ b/include/drm/drm_plane_helper.h
@@ -58,10 +58,8 @@ int drm_crtc_init(struct drm_device *dev, struct drm_crtc *crtc,
  */
 struct drm_plane_helper_funcs {
 	int (*prepare_fb)(struct drm_plane *plane,
-			  struct drm_framebuffer *fb,
 			  const struct drm_plane_state *new_state);
 	void (*cleanup_fb)(struct drm_plane *plane,
-			   struct drm_framebuffer *fb,
 			   const struct drm_plane_state *old_state);
 
 	int (*atomic_check)(struct drm_plane *plane,
-- 
cgit v0.10.2


From aef9dbb8f779ae0ffb46313b07700cac72b58ff4 Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Tue, 8 Sep 2015 12:02:07 +0200
Subject: drm/atomic-helper: Add option to update planes only on active crtc

With drivers supporting runtime pm it's generally not a good idea to
touch the hardware when it's off. Add an option to the commit_planes
helper to support this case.

Note that the helpers already add all planes on a crtc when a modeset
happens, hence plane updates will not be lost if drivers set this to
true.

v2: Check for NULL state->crtc before chasing the pointer. Also check
both old and new crtc if there's a switch. Finally just outright
disallow switching crtcs for a plane if the plane is in active use, on
most hardware that doesn't make sense.

v3: Since commit_planes(active_only = true) is for enabling things
only after all the crtc are on we should only look at the new crtc to
decide whether to call the plane hooks - if the current CRTC isn't on
then skip. If the old crtc (when moving a plane) went down then the
plane should have been disabled as part of the pipe shutdown work
already. For which there's currently no helper really unfortunately.
Also move the check for wether a plane gets a new CRTC assigned while
still in active use out of this patch.

v4: Rebase over exynos changes.

Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Thierry Reding <treding@nvidia.com>
Cc: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Reviewed-by: Thierry Reding <treding@nvidia.com>
Tested-by: Thierry Reding <treding@nvidia.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c
index 9b0c476..12c25c5 100644
--- a/drivers/gpu/drm/drm_atomic_helper.c
+++ b/drivers/gpu/drm/drm_atomic_helper.c
@@ -1037,7 +1037,7 @@ int drm_atomic_helper_commit(struct drm_device *dev,
 
 	drm_atomic_helper_commit_modeset_disables(dev, state);
 
-	drm_atomic_helper_commit_planes(dev, state);
+	drm_atomic_helper_commit_planes(dev, state, false);
 
 	drm_atomic_helper_commit_modeset_enables(dev, state);
 
@@ -1146,10 +1146,16 @@ fail:
 }
 EXPORT_SYMBOL(drm_atomic_helper_prepare_planes);
 
+bool plane_crtc_active(struct drm_plane_state *state)
+{
+	return state->crtc && state->crtc->state->active;
+}
+
 /**
  * drm_atomic_helper_commit_planes - commit plane state
  * @dev: DRM device
  * @old_state: atomic state object with old state structures
+ * @active_only: Only commit on active CRTC if set
  *
  * This function commits the new plane state using the plane and atomic helper
  * functions for planes and crtcs. It assumes that the atomic state has already
@@ -1164,7 +1170,8 @@ EXPORT_SYMBOL(drm_atomic_helper_prepare_planes);
  * drm_atomic_helper_commit_planes_on_crtc() instead.
  */
 void drm_atomic_helper_commit_planes(struct drm_device *dev,
-				     struct drm_atomic_state *old_state)
+				     struct drm_atomic_state *old_state,
+				     bool active_only)
 {
 	struct drm_crtc *crtc;
 	struct drm_crtc_state *old_crtc_state;
@@ -1180,6 +1187,9 @@ void drm_atomic_helper_commit_planes(struct drm_device *dev,
 		if (!funcs || !funcs->atomic_begin)
 			continue;
 
+		if (active_only && !crtc->state->active)
+			continue;
+
 		funcs->atomic_begin(crtc, old_crtc_state);
 	}
 
@@ -1191,6 +1201,9 @@ void drm_atomic_helper_commit_planes(struct drm_device *dev,
 		if (!funcs)
 			continue;
 
+		if (active_only && !plane_crtc_active(plane->state))
+			continue;
+
 		/*
 		 * Special-case disabling the plane if drivers support it.
 		 */
@@ -1210,6 +1223,9 @@ void drm_atomic_helper_commit_planes(struct drm_device *dev,
 		if (!funcs || !funcs->atomic_flush)
 			continue;
 
+		if (active_only && !crtc->state->active)
+			continue;
+
 		funcs->atomic_flush(crtc, old_crtc_state);
 	}
 }
diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.c b/drivers/gpu/drm/exynos/exynos_drm_drv.c
index 831d2e4..f0a5839 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_drv.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_drv.c
@@ -105,7 +105,7 @@ static void exynos_atomic_commit_complete(struct exynos_atomic_commit *commit)
 		atomic_inc(&exynos_crtc->pending_update);
 	}
 
-	drm_atomic_helper_commit_planes(dev, state);
+	drm_atomic_helper_commit_planes(dev, state, false);
 
 	exynos_atomic_wait_for_commit(state);
 
diff --git a/drivers/gpu/drm/msm/msm_atomic.c b/drivers/gpu/drm/msm/msm_atomic.c
index 1ceb4f2..7eb253b 100644
--- a/drivers/gpu/drm/msm/msm_atomic.c
+++ b/drivers/gpu/drm/msm/msm_atomic.c
@@ -125,7 +125,7 @@ static void complete_commit(struct msm_commit *c)
 
 	drm_atomic_helper_commit_modeset_disables(dev, state);
 
-	drm_atomic_helper_commit_planes(dev, state);
+	drm_atomic_helper_commit_planes(dev, state, false);
 
 	drm_atomic_helper_commit_modeset_enables(dev, state);
 
diff --git a/drivers/gpu/drm/omapdrm/omap_drv.c b/drivers/gpu/drm/omapdrm/omap_drv.c
index 419c2e4..a5f9d8b 100644
--- a/drivers/gpu/drm/omapdrm/omap_drv.c
+++ b/drivers/gpu/drm/omapdrm/omap_drv.c
@@ -96,7 +96,7 @@ static void omap_atomic_complete(struct omap_atomic_state_commit *commit)
 	dispc_runtime_get();
 
 	drm_atomic_helper_commit_modeset_disables(dev, old_state);
-	drm_atomic_helper_commit_planes(dev, old_state);
+	drm_atomic_helper_commit_planes(dev, old_state, false);
 	drm_atomic_helper_commit_modeset_enables(dev, old_state);
 
 	omap_atomic_wait_for_completion(dev, old_state);
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_kms.c b/drivers/gpu/drm/rcar-du/rcar_du_kms.c
index 56518eb..ca12e8c 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_kms.c
+++ b/drivers/gpu/drm/rcar-du/rcar_du_kms.c
@@ -456,7 +456,7 @@ static void rcar_du_atomic_complete(struct rcar_du_commit *commit)
 	/* Apply the atomic update. */
 	drm_atomic_helper_commit_modeset_disables(dev, old_state);
 	drm_atomic_helper_commit_modeset_enables(dev, old_state);
-	drm_atomic_helper_commit_planes(dev, old_state);
+	drm_atomic_helper_commit_planes(dev, old_state, false);
 
 	drm_atomic_helper_wait_for_vblanks(dev, old_state);
 
diff --git a/drivers/gpu/drm/sti/sti_drv.c b/drivers/gpu/drm/sti/sti_drv.c
index 6f4af6a..9f85988 100644
--- a/drivers/gpu/drm/sti/sti_drv.c
+++ b/drivers/gpu/drm/sti/sti_drv.c
@@ -59,7 +59,7 @@ static void sti_atomic_complete(struct sti_private *private,
 	 */
 
 	drm_atomic_helper_commit_modeset_disables(drm, state);
-	drm_atomic_helper_commit_planes(drm, state);
+	drm_atomic_helper_commit_planes(drm, state, false);
 	drm_atomic_helper_commit_modeset_enables(drm, state);
 
 	drm_atomic_helper_wait_for_vblanks(drm, state);
diff --git a/drivers/gpu/drm/tegra/drm.c b/drivers/gpu/drm/tegra/drm.c
index 6d88cf1..2486bc2 100644
--- a/drivers/gpu/drm/tegra/drm.c
+++ b/drivers/gpu/drm/tegra/drm.c
@@ -56,7 +56,7 @@ static void tegra_atomic_complete(struct tegra_drm *tegra,
 	 */
 
 	drm_atomic_helper_commit_modeset_disables(drm, state);
-	drm_atomic_helper_commit_planes(drm, state);
+	drm_atomic_helper_commit_planes(drm, state, false);
 	drm_atomic_helper_commit_modeset_enables(drm, state);
 
 	drm_atomic_helper_wait_for_vblanks(drm, state);
diff --git a/include/drm/drm_atomic_helper.h b/include/drm/drm_atomic_helper.h
index 11266d1..4ffe9dc 100644
--- a/include/drm/drm_atomic_helper.h
+++ b/include/drm/drm_atomic_helper.h
@@ -55,7 +55,8 @@ void drm_atomic_helper_commit_modeset_enables(struct drm_device *dev,
 int drm_atomic_helper_prepare_planes(struct drm_device *dev,
 				     struct drm_atomic_state *state);
 void drm_atomic_helper_commit_planes(struct drm_device *dev,
-				     struct drm_atomic_state *state);
+				     struct drm_atomic_state *state,
+				     bool active_only);
 void drm_atomic_helper_cleanup_planes(struct drm_device *dev,
 				      struct drm_atomic_state *old_state);
 void drm_atomic_helper_commit_planes_on_crtc(struct drm_crtc_state *old_crtc_state);
-- 
cgit v0.10.2


From 6e48ae3269e3b89d8014d0eb2e35678b0d242b3d Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Tue, 8 Sep 2015 13:52:45 +0200
Subject: drm/atomic-helper: Pimp docs with recommendations for rpm drivers

Requested by Laurent.

Note that this uses the new markdown support which will only land in
kernel 4.4 (for the code snippet).

v2: A few spelling fixes I spotted myself.

v3: Big reword for commit_planes() kerneldoc based on a text from
Laurent.

Cc: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Thierry Reding <treding@nvidia.com> (v1 on irc)
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c
index 12c25c5..77d55a9 100644
--- a/drivers/gpu/drm/drm_atomic_helper.c
+++ b/drivers/gpu/drm/drm_atomic_helper.c
@@ -993,6 +993,22 @@ EXPORT_SYMBOL(drm_atomic_helper_wait_for_vblanks);
  * object. This can still fail when e.g. the framebuffer reservation fails. For
  * now this doesn't implement asynchronous commits.
  *
+ * Note that right now this function does not support async commits, and hence
+ * driver writers must implement their own version for now. Also note that the
+ * default ordering of how the various stages are called is to match the legacy
+ * modeset helper library closest. One peculiarity of that is that it doesn't
+ * mesh well with runtime PM at all.
+ *
+ * For drivers supporting runtime PM the recommended sequence is
+ *
+ *     drm_atomic_helper_commit_modeset_disables(dev, state);
+ *
+ *     drm_atomic_helper_commit_modeset_enables(dev, state);
+ *
+ *     drm_atomic_helper_commit_planes(dev, state, true);
+ *
+ * See the kerneldoc entries for these three functions for more details.
+ *
  * RETURNS
  * Zero for success or -errno.
  */
@@ -1168,6 +1184,22 @@ bool plane_crtc_active(struct drm_plane_state *state)
  * Note that this function does all plane updates across all CRTCs in one step.
  * If the hardware can't support this approach look at
  * drm_atomic_helper_commit_planes_on_crtc() instead.
+ *
+ * Plane parameters can be updated by applications while the associated CRTC is
+ * disabled. The DRM/KMS core will store the parameters in the plane state,
+ * which will be available to the driver when the CRTC is turned on. As a result
+ * most drivers don't need to be immediately notified of plane updates for a
+ * disabled CRTC.
+ *
+ * Unless otherwise needed, drivers are advised to set the @active_only
+ * parameters to true in order not to receive plane update notifications related
+ * to a disabled CRTC. This avoids the need to manually ignore plane updates in
+ * driver code when the driver and/or hardware can't or just don't need to deal
+ * with updates on disabled CRTCs, for example when supporting runtime PM.
+ *
+ * The drm_atomic_helper_commit() default implementation only sets @active_only
+ * to false to most closely match the behaviour of the legacy helpers. This should
+ * not be copied blindly by drivers.
  */
 void drm_atomic_helper_commit_planes(struct drm_device *dev,
 				     struct drm_atomic_state *old_state,
-- 
cgit v0.10.2


From 397fd77c0491ceb0ed4783eb88fc05d0222e2030 Mon Sep 17 00:00:00 2001
From: Thierry Reding <treding@nvidia.com>
Date: Tue, 8 Sep 2015 15:00:45 +0200
Subject: drm/atomic-helper: Implement drm_atomic_helper_duplicate_state()

This function can be used to duplicate an atomic state object. This is
useful for example to implement suspend/resume, where the state before
suspend can be saved and restored upon resume.

v2: move locking to caller, be more explicit about prerequisites
v3: explicitly pass lock acquisition context, improve kerneldoc

Signed-off-by: Thierry Reding <treding@nvidia.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c
index 77d55a9..9941167 100644
--- a/drivers/gpu/drm/drm_atomic_helper.c
+++ b/drivers/gpu/drm/drm_atomic_helper.c
@@ -2372,6 +2372,84 @@ drm_atomic_helper_connector_duplicate_state(struct drm_connector *connector)
 EXPORT_SYMBOL(drm_atomic_helper_connector_duplicate_state);
 
 /**
+ * drm_atomic_helper_duplicate_state - duplicate an atomic state object
+ * @dev: DRM device
+ * @ctx: lock acquisition context
+ *
+ * Makes a copy of the current atomic state by looping over all objects and
+ * duplicating their respective states.
+ *
+ * Note that this treats atomic state as persistent between save and restore.
+ * Drivers must make sure that this is possible and won't result in confusion
+ * or erroneous behaviour.
+ *
+ * Note that if callers haven't already acquired all modeset locks this might
+ * return -EDEADLK, which must be handled by calling drm_modeset_backoff().
+ *
+ * Returns:
+ * A pointer to the copy of the atomic state object on success or an
+ * ERR_PTR()-encoded error code on failure.
+ */
+struct drm_atomic_state *
+drm_atomic_helper_duplicate_state(struct drm_device *dev,
+				  struct drm_modeset_acquire_ctx *ctx)
+{
+	struct drm_atomic_state *state;
+	struct drm_connector *conn;
+	struct drm_plane *plane;
+	struct drm_crtc *crtc;
+	int err = 0;
+
+	state = drm_atomic_state_alloc(dev);
+	if (!state)
+		return ERR_PTR(-ENOMEM);
+
+	state->acquire_ctx = ctx;
+
+	drm_for_each_crtc(crtc, dev) {
+		struct drm_crtc_state *crtc_state;
+
+		crtc_state = drm_atomic_get_crtc_state(state, crtc);
+		if (IS_ERR(crtc_state)) {
+			err = PTR_ERR(crtc_state);
+			goto free;
+		}
+	}
+
+	drm_for_each_plane(plane, dev) {
+		struct drm_plane_state *plane_state;
+
+		plane_state = drm_atomic_get_plane_state(state, plane);
+		if (IS_ERR(plane_state)) {
+			err = PTR_ERR(plane_state);
+			goto free;
+		}
+	}
+
+	drm_for_each_connector(conn, dev) {
+		struct drm_connector_state *conn_state;
+
+		conn_state = drm_atomic_get_connector_state(state, conn);
+		if (IS_ERR(conn_state)) {
+			err = PTR_ERR(conn_state);
+			goto free;
+		}
+	}
+
+	/* clear the acquire context so that it isn't accidentally reused */
+	state->acquire_ctx = NULL;
+
+free:
+	if (err < 0) {
+		drm_atomic_state_free(state);
+		state = ERR_PTR(err);
+	}
+
+	return state;
+}
+EXPORT_SYMBOL(drm_atomic_helper_duplicate_state);
+
+/**
  * __drm_atomic_helper_connector_destroy_state - release connector state
  * @connector: connector object
  * @state: connector state object to release
diff --git a/include/drm/drm_atomic_helper.h b/include/drm/drm_atomic_helper.h
index 4ffe9dc..1547eb4 100644
--- a/include/drm/drm_atomic_helper.h
+++ b/include/drm/drm_atomic_helper.h
@@ -118,6 +118,9 @@ __drm_atomic_helper_connector_duplicate_state(struct drm_connector *connector,
 					   struct drm_connector_state *state);
 struct drm_connector_state *
 drm_atomic_helper_connector_duplicate_state(struct drm_connector *connector);
+struct drm_atomic_state *
+drm_atomic_helper_duplicate_state(struct drm_device *dev,
+				  struct drm_modeset_acquire_ctx *ctx);
 void
 __drm_atomic_helper_connector_destroy_state(struct drm_connector *connector,
 					    struct drm_connector_state *state);
-- 
cgit v0.10.2


From 2b712be72fddc74ac12c2857af24a20a93d9e9c0 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 27 Aug 2015 17:23:26 +0300
Subject: drm/dp: s/I2C_STATUS/I2C_WRITE_STATUS_UPDATE/
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Rename the I2C_STATUS request to I2C_WRITE_STATUS_UPDATE to match the
spec.

Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

diff --git a/drivers/gpu/drm/tegra/dpaux.c b/drivers/gpu/drm/tegra/dpaux.c
index 224a7dc..1cc09ff 100644
--- a/drivers/gpu/drm/tegra/dpaux.c
+++ b/drivers/gpu/drm/tegra/dpaux.c
@@ -149,7 +149,7 @@ static ssize_t tegra_dpaux_transfer(struct drm_dp_aux *aux,
 
 		break;
 
-	case DP_AUX_I2C_STATUS:
+	case DP_AUX_I2C_WRITE_STATUS_UPDATE:
 		if (msg->request & DP_AUX_I2C_MOT)
 			value |= DPAUX_DP_AUXCTL_CMD_MOT_RQ;
 		else
diff --git a/include/drm/drm_dp_helper.h b/include/drm/drm_dp_helper.h
index 499e9f6..d0c8810 100644
--- a/include/drm/drm_dp_helper.h
+++ b/include/drm/drm_dp_helper.h
@@ -46,7 +46,7 @@
 
 #define DP_AUX_I2C_WRITE		0x0
 #define DP_AUX_I2C_READ			0x1
-#define DP_AUX_I2C_STATUS		0x2
+#define DP_AUX_I2C_WRITE_STATUS_UPDATE	0x2
 #define DP_AUX_I2C_MOT			0x4
 #define DP_AUX_NATIVE_WRITE		0x8
 #define DP_AUX_NATIVE_READ		0x9
-- 
cgit v0.10.2


From c1e74122fb029d30b66d9364122aef50265354aa Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 27 Aug 2015 17:23:27 +0300
Subject: drm/i915: Handle DP_AUX_I2C_WRITE_STATUS_UPDATE
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

When we get an i2c defer or short ack for i2c-over-aux write we need
to switch to WRITE_STATUS_UPDATE to poll for the completion of the
original request.

i915 doesn't try to interpret wht request type apart from separating
reads from writes, and so we should be able to treat this the same as
a normal i2c write.

Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 0a2e33f..7bb96d5 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -974,6 +974,7 @@ intel_dp_aux_transfer(struct drm_dp_aux *aux, struct drm_dp_aux_msg *msg)
 	switch (msg->request & ~DP_AUX_I2C_MOT) {
 	case DP_AUX_NATIVE_WRITE:
 	case DP_AUX_I2C_WRITE:
+	case DP_AUX_I2C_WRITE_STATUS_UPDATE:
 		txsize = msg->size ? HEADER_SIZE + msg->size : BARE_ADDRESS_SIZE;
 		rxsize = 2; /* 0 or 1 data bytes */
 
-- 
cgit v0.10.2


From 1f75b29d3fc9abb06b095860f9312f8190e6015b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 27 Aug 2015 17:23:28 +0300
Subject: drm/radeon: Handle DP_AUX_I2C_WRITE_STATUS_UPDATE
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

When we get an i2c defer or short ack for i2c-over-aux write we need
to switch to WRITE_STATUS_UPDATE to poll for the completion of the
original request.

Looks like radeon doesn't do anything special with the request type,
so hopefully just treating it the same as a i2c write is enough.

Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: "Christian König" <christian.koenig@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/radeon/atombios_dp.c b/drivers/gpu/drm/radeon/atombios_dp.c
index 9cd49c5..bd73b40 100644
--- a/drivers/gpu/drm/radeon/atombios_dp.c
+++ b/drivers/gpu/drm/radeon/atombios_dp.c
@@ -179,6 +179,7 @@ radeon_dp_aux_transfer_atom(struct drm_dp_aux *aux, struct drm_dp_aux_msg *msg)
 	switch (msg->request & ~DP_AUX_I2C_MOT) {
 	case DP_AUX_NATIVE_WRITE:
 	case DP_AUX_I2C_WRITE:
+	case DP_AUX_I2C_WRITE_STATUS_UPDATE:
 		/* The atom implementation only supports writes with a max payload of
 		 * 12 bytes since it uses 4 bits for the total count (header + payload)
 		 * in the parameter space.  The atom interface supports 16 byte
-- 
cgit v0.10.2


From f993406182b3ffad7c53ffc180b65e2b7e3d8986 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 27 Aug 2015 17:23:29 +0300
Subject: drm/tegra: Handle I2C_WRITE_STATUS_UPDATE for address only writes
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

A address-only I2C_WRITE can't be replied with a short i2c ack, but I
suppose it could be replied with an i2c defer. So the code should be
prepared for an address-only I2C_WRITE_STATUS_UPDATE.

Cc: Thierry Reding <thierry.reding@gmail.com>
Cc: "Terje Bergström" <tbergstrom@nvidia.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Acked-by: Thierry Reding <treding@nvidia.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/tegra/dpaux.c b/drivers/gpu/drm/tegra/dpaux.c
index 1cc09ff..6aecb66 100644
--- a/drivers/gpu/drm/tegra/dpaux.c
+++ b/drivers/gpu/drm/tegra/dpaux.c
@@ -119,6 +119,7 @@ static ssize_t tegra_dpaux_transfer(struct drm_dp_aux *aux,
 	 */
 	if (msg->size < 1) {
 		switch (msg->request & ~DP_AUX_I2C_MOT) {
+		case DP_AUX_I2C_WRITE_STATUS_UPDATE:
 		case DP_AUX_I2C_WRITE:
 		case DP_AUX_I2C_READ:
 			value = DPAUX_DP_AUXCTL_CMD_ADDRESS_ONLY;
-- 
cgit v0.10.2


From 68ec2a2a24815c0ee359b1327b60a276cd2280d8 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 27 Aug 2015 17:23:30 +0300
Subject: drm/dp: Use I2C_WRITE_STATUS_UPDATE to drain partial I2C_WRITE
 requests
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

When an i2c WRITE gets an i2c defer or short i2c ack reply, we are
supposed to switch the request from I2C_WRITE to I2C_WRITE_STATUS_UPDATE
when we continue to poll for the completion of the request.

v2: Don't assume DP_AUX_I2C_WRITE is 0 even though it is, to make the
    code more obvious to the casual reader (Jani)

Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
[danvet: Resolve conflict due to changed context.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_dp_helper.c b/drivers/gpu/drm/drm_dp_helper.c
index 80a02a4..5a55d90 100644
--- a/drivers/gpu/drm/drm_dp_helper.c
+++ b/drivers/gpu/drm/drm_dp_helper.c
@@ -422,6 +422,19 @@ static u32 drm_dp_i2c_functionality(struct i2c_adapter *adapter)
 	       I2C_FUNC_10BIT_ADDR;
 }
 
+static void drm_dp_i2c_msg_write_status_update(struct drm_dp_aux_msg *msg)
+{
+	/*
+	 * In case of i2c defer or short i2c ack reply to a write,
+	 * we need to switch to WRITE_STATUS_UPDATE to drain the
+	 * rest of the message
+	 */
+	if ((msg->request & ~DP_AUX_I2C_MOT) == DP_AUX_I2C_WRITE) {
+		msg->request &= DP_AUX_I2C_MOT;
+		msg->request |= DP_AUX_I2C_WRITE_STATUS_UPDATE;
+	}
+}
+
 /*
  * Transfer a single I2C-over-AUX message and handle various error conditions,
  * retrying the transaction as appropriate.  It is assumed that the
@@ -490,6 +503,8 @@ static int drm_dp_i2c_do_msg(struct drm_dp_aux *aux, struct drm_dp_aux_msg *msg)
 			 * Both native ACK and I2C ACK replies received. We
 			 * can assume the transfer was successful.
 			 */
+			if (ret != msg->size)
+				drm_dp_i2c_msg_write_status_update(msg);
 			return ret;
 
 		case DP_AUX_I2C_REPLY_NACK:
@@ -507,6 +522,7 @@ static int drm_dp_i2c_do_msg(struct drm_dp_aux *aux, struct drm_dp_aux_msg *msg)
 			if (defer_i2c < 7)
 				defer_i2c++;
 			usleep_range(400, 500);
+			drm_dp_i2c_msg_write_status_update(msg);
 			continue;
 
 		default:
@@ -519,6 +535,14 @@ static int drm_dp_i2c_do_msg(struct drm_dp_aux *aux, struct drm_dp_aux_msg *msg)
 	return -EREMOTEIO;
 }
 
+static void drm_dp_i2c_msg_set_request(struct drm_dp_aux_msg *msg,
+				       const struct i2c_msg *i2c_msg)
+{
+	msg->request = (i2c_msg->flags & I2C_M_RD) ?
+		DP_AUX_I2C_READ : DP_AUX_I2C_WRITE;
+	msg->request |= DP_AUX_I2C_MOT;
+}
+
 /*
  * Keep retrying drm_dp_i2c_do_msg until all data has been transferred.
  *
@@ -572,10 +596,7 @@ static int drm_dp_i2c_xfer(struct i2c_adapter *adapter, struct i2c_msg *msgs,
 
 	for (i = 0; i < num; i++) {
 		msg.address = msgs[i].addr;
-		msg.request = (msgs[i].flags & I2C_M_RD) ?
-			DP_AUX_I2C_READ :
-			DP_AUX_I2C_WRITE;
-		msg.request |= DP_AUX_I2C_MOT;
+		drm_dp_i2c_msg_set_request(&msg, &msgs[i]);
 		/* Send a bare address packet to start the transaction.
 		 * Zero sized messages specify an address only (bare
 		 * address) transaction.
@@ -583,6 +604,13 @@ static int drm_dp_i2c_xfer(struct i2c_adapter *adapter, struct i2c_msg *msgs,
 		msg.buffer = NULL;
 		msg.size = 0;
 		err = drm_dp_i2c_do_msg(aux, &msg);
+
+		/*
+		 * Reset msg.request in case in case it got
+		 * changed into a WRITE_STATUS_UPDATE.
+		 */
+		drm_dp_i2c_msg_set_request(&msg, &msgs[i]);
+
 		if (err < 0)
 			break;
 		/* We want each transaction to be as large as possible, but
@@ -595,6 +623,13 @@ static int drm_dp_i2c_xfer(struct i2c_adapter *adapter, struct i2c_msg *msgs,
 			msg.size = min(transfer_size, msgs[i].len - j);
 
 			err = drm_dp_i2c_drain_msg(aux, &msg);
+
+			/*
+			 * Reset msg.request in case in case it got
+			 * changed into a WRITE_STATUS_UPDATE.
+			 */
+			drm_dp_i2c_msg_set_request(&msg, &msgs[i]);
+
 			if (err < 0)
 				break;
 			transfer_size = err;
-- 
cgit v0.10.2


From 85f8fcd619d161d65c53b067e2d99590c0d7bbad Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 7 Sep 2015 18:22:56 +0300
Subject: drm: Make some modes const when iterating through them
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

valid_inferred_mode() don't change the modes over which it iterates,
so make the iterator const.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c
index 05bb731..9afb1fc 100644
--- a/drivers/gpu/drm/drm_edid.c
+++ b/drivers/gpu/drm/drm_edid.c
@@ -2044,7 +2044,7 @@ mode_in_range(const struct drm_display_mode *mode, struct edid *edid,
 static bool valid_inferred_mode(const struct drm_connector *connector,
 				const struct drm_display_mode *mode)
 {
-	struct drm_display_mode *m;
+	const struct drm_display_mode *m;
 	bool ok = false;
 
 	list_for_each_entry(m, &connector->probed_modes, head) {
-- 
cgit v0.10.2


From 9e5a3b529e8419db1dd2b32c86a1fb42fc07347d Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 7 Sep 2015 18:22:57 +0300
Subject: drm: Remove the 'mode' argument from drm_select_eld()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

drm_select_eld() doesn't look at the passed in mode, so don't pass it
in.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c
index 9afb1fc..e32218f 100644
--- a/drivers/gpu/drm/drm_edid.c
+++ b/drivers/gpu/drm/drm_edid.c
@@ -3396,7 +3396,6 @@ EXPORT_SYMBOL(drm_av_sync_delay);
 /**
  * drm_select_eld - select one ELD from multiple HDMI/DP sinks
  * @encoder: the encoder just changed display mode
- * @mode: the adjusted display mode
  *
  * It's possible for one encoder to be associated with multiple HDMI/DP sinks.
  * The policy is now hard coded to simply use the first HDMI/DP sink's ELD.
@@ -3404,8 +3403,7 @@ EXPORT_SYMBOL(drm_av_sync_delay);
  * Return: The connector associated with the first HDMI/DP sink that has ELD
  * attached to it.
  */
-struct drm_connector *drm_select_eld(struct drm_encoder *encoder,
-				     struct drm_display_mode *mode)
+struct drm_connector *drm_select_eld(struct drm_encoder *encoder)
 {
 	struct drm_connector *connector;
 	struct drm_device *dev = encoder->dev;
diff --git a/drivers/gpu/drm/i915/intel_audio.c b/drivers/gpu/drm/i915/intel_audio.c
index dc32cf4..1314ebb 100644
--- a/drivers/gpu/drm/i915/intel_audio.c
+++ b/drivers/gpu/drm/i915/intel_audio.c
@@ -401,7 +401,7 @@ void intel_audio_codec_enable(struct intel_encoder *intel_encoder)
 	struct drm_device *dev = encoder->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	connector = drm_select_eld(encoder, mode);
+	connector = drm_select_eld(encoder);
 	if (!connector)
 		return;
 
diff --git a/include/drm/drm_edid.h b/include/drm/drm_edid.h
index 53c53c4..31528d9 100644
--- a/include/drm/drm_edid.h
+++ b/include/drm/drm_edid.h
@@ -327,8 +327,7 @@ int drm_edid_to_sad(struct edid *edid, struct cea_sad **sads);
 int drm_edid_to_speaker_allocation(struct edid *edid, u8 **sadb);
 int drm_av_sync_delay(struct drm_connector *connector,
 		      struct drm_display_mode *mode);
-struct drm_connector *drm_select_eld(struct drm_encoder *encoder,
-				     struct drm_display_mode *mode);
+struct drm_connector *drm_select_eld(struct drm_encoder *encoder);
 int drm_load_edid_firmware(struct drm_connector *connector);
 
 int
-- 
cgit v0.10.2


From 3a818d350f6b5ad542175ab1f71c027787ce952e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 7 Sep 2015 18:22:58 +0300
Subject: drm: Make drm_av_sync_delay() 'mode' argument const
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

drm_av_sync_delay() doesn't change the passed in mode, so make it const.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c
index e32218f..d895556 100644
--- a/drivers/gpu/drm/drm_edid.c
+++ b/drivers/gpu/drm/drm_edid.c
@@ -3361,7 +3361,7 @@ EXPORT_SYMBOL(drm_edid_to_speaker_allocation);
  * the sink doesn't support audio or video.
  */
 int drm_av_sync_delay(struct drm_connector *connector,
-		      struct drm_display_mode *mode)
+		      const struct drm_display_mode *mode)
 {
 	int i = !!(mode->flags & DRM_MODE_FLAG_INTERLACE);
 	int a, v;
diff --git a/include/drm/drm_edid.h b/include/drm/drm_edid.h
index 31528d9..2af9769 100644
--- a/include/drm/drm_edid.h
+++ b/include/drm/drm_edid.h
@@ -326,7 +326,7 @@ void drm_edid_to_eld(struct drm_connector *connector, struct edid *edid);
 int drm_edid_to_sad(struct edid *edid, struct cea_sad **sads);
 int drm_edid_to_speaker_allocation(struct edid *edid, u8 **sadb);
 int drm_av_sync_delay(struct drm_connector *connector,
-		      struct drm_display_mode *mode);
+		      const struct drm_display_mode *mode);
 struct drm_connector *drm_select_eld(struct drm_encoder *encoder);
 int drm_load_edid_firmware(struct drm_connector *connector);
 
-- 
cgit v0.10.2


From 26b91ae4732be89228d207c76827071c6aecc4d8 Mon Sep 17 00:00:00 2001
From: David Herrmann <dh.herrmann@gmail.com>
Date: Wed, 9 Sep 2015 14:21:29 +0200
Subject: drm: simplify drm_sysfs_destroy() via IS_ERR_OR_NULL()

Simplify `foo == NULL || IS_ERR(foo)` via IS_ERR_OR_NULL(). This is
pretty commonly used all over the kernel, especially for debugfs/sysfs
cleanup paths.

Signed-off-by: David Herrmann <dh.herrmann@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_sysfs.c b/drivers/gpu/drm/drm_sysfs.c
index 0f6cd33..3f66cb0 100644
--- a/drivers/gpu/drm/drm_sysfs.c
+++ b/drivers/gpu/drm/drm_sysfs.c
@@ -156,7 +156,7 @@ err_out:
  */
 void drm_sysfs_destroy(void)
 {
-	if ((drm_class == NULL) || (IS_ERR(drm_class)))
+	if (IS_ERR_OR_NULL(drm_class))
 		return;
 	class_remove_file(drm_class, &class_attr_version.attr);
 	class_destroy(drm_class);
-- 
cgit v0.10.2


From fcc9021343212f6a4a52a085b3d383ab29a9ac0a Mon Sep 17 00:00:00 2001
From: David Herrmann <dh.herrmann@gmail.com>
Date: Wed, 9 Sep 2015 14:21:30 +0200
Subject: drm: move drm_class into drm_sysfs.c

Right now, drm_sysfs_create() returns the newly allocated "struct class"
to the caller (which is drm_core_init()), which then has to set the
global variable 'drm_class'. During cleanup, though, we call
drm_sysfs_destroy() which implicitly uses the global 'drm_class'. This is
confusing, as ownership of the global 'drm_class' is non-obvious.

This patch changes drm_sysfs_create() to drm_sysfs_init() and makes it
initialize the 'drm_class' object directly, rather than returning it.
This way, both drm_sysfs_init() and drm_sysfs_destroy() work in a similar
fashion and manage the global drm class.

Signed-off-by: David Herrmann <dh.herrmann@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c
index dc93c88..9ad823f 100644
--- a/drivers/gpu/drm/drm_drv.c
+++ b/drivers/gpu/drm/drm_drv.c
@@ -55,7 +55,6 @@ module_param_named(debug, drm_debug, int, 0600);
 static DEFINE_SPINLOCK(drm_minor_lock);
 static struct idr drm_minors_idr;
 
-struct class *drm_class;
 static struct dentry *drm_debugfs_root;
 
 void drm_err(const char *format, ...)
@@ -841,10 +840,9 @@ static int __init drm_core_init(void)
 	if (register_chrdev(DRM_MAJOR, "drm", &drm_stub_fops))
 		goto err_p1;
 
-	drm_class = drm_sysfs_create(THIS_MODULE, "drm");
-	if (IS_ERR(drm_class)) {
+	ret = drm_sysfs_init();
+	if (ret < 0) {
 		printk(KERN_ERR "DRM: Error creating drm class.\n");
-		ret = PTR_ERR(drm_class);
 		goto err_p2;
 	}
 
diff --git a/drivers/gpu/drm/drm_internal.h b/drivers/gpu/drm/drm_internal.h
index 059af01..43cbda3 100644
--- a/drivers/gpu/drm/drm_internal.h
+++ b/drivers/gpu/drm/drm_internal.h
@@ -73,7 +73,7 @@ int drm_authmagic(struct drm_device *dev, void *data,
 /* drm_sysfs.c */
 extern struct class *drm_class;
 
-struct class *drm_sysfs_create(struct module *owner, char *name);
+int drm_sysfs_init(void);
 void drm_sysfs_destroy(void);
 struct device *drm_sysfs_minor_alloc(struct drm_minor *minor);
 int drm_sysfs_connector_add(struct drm_connector *connector);
diff --git a/drivers/gpu/drm/drm_sysfs.c b/drivers/gpu/drm/drm_sysfs.c
index 3f66cb0..f08873f 100644
--- a/drivers/gpu/drm/drm_sysfs.c
+++ b/drivers/gpu/drm/drm_sysfs.c
@@ -30,6 +30,8 @@ static struct device_type drm_sysfs_device_minor = {
 	.name = "drm_minor"
 };
 
+struct class *drm_class;
+
 /**
  * __drm_class_suspend - internal DRM class suspend routine
  * @dev: Linux device to suspend
@@ -112,41 +114,34 @@ static CLASS_ATTR_STRING(version, S_IRUGO,
 		CORE_DATE);
 
 /**
- * drm_sysfs_create - create a struct drm_sysfs_class structure
- * @owner: pointer to the module that is to "own" this struct drm_sysfs_class
- * @name: pointer to a string for the name of this class.
+ * drm_sysfs_init - initialize sysfs helpers
+ *
+ * This is used to create the DRM class, which is the implicit parent of any
+ * other top-level DRM sysfs objects.
  *
- * This is used to create DRM class pointer that can then be used
- * in calls to drm_sysfs_device_add().
+ * You must call drm_sysfs_destroy() to release the allocated resources.
  *
- * Note, the pointer created here is to be destroyed when finished by making a
- * call to drm_sysfs_destroy().
+ * Return: 0 on success, negative error code on failure.
  */
-struct class *drm_sysfs_create(struct module *owner, char *name)
+int drm_sysfs_init(void)
 {
-	struct class *class;
 	int err;
 
-	class = class_create(owner, name);
-	if (IS_ERR(class)) {
-		err = PTR_ERR(class);
-		goto err_out;
-	}
-
-	class->pm = &drm_class_dev_pm_ops;
+	drm_class = class_create(THIS_MODULE, "drm");
+	if (IS_ERR(drm_class))
+		return PTR_ERR(drm_class);
 
-	err = class_create_file(class, &class_attr_version.attr);
-	if (err)
-		goto err_out_class;
+	drm_class->pm = &drm_class_dev_pm_ops;
 
-	class->devnode = drm_devnode;
-
-	return class;
+	err = class_create_file(drm_class, &class_attr_version.attr);
+	if (err) {
+		class_destroy(drm_class);
+		drm_class = NULL;
+		return err;
+	}
 
-err_out_class:
-	class_destroy(class);
-err_out:
-	return ERR_PTR(err);
+	drm_class->devnode = drm_devnode;
+	return 0;
 }
 
 /**
-- 
cgit v0.10.2


From c099b55a6fa6f9ec2f26105c76df462d7c7c7d5b Mon Sep 17 00:00:00 2001
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Date: Wed, 9 Sep 2015 13:46:21 +0200
Subject: drm/core: Do not call drm_framebuffer_remove internally during
 teardown.

This may cause issues because encoders are already destroyed so removing
active primaries may use freed memory. Instead free the fb directly,
ignoring refcount.

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c
index 474f328..9b9c4b4 100644
--- a/drivers/gpu/drm/drm_crtc.c
+++ b/drivers/gpu/drm/drm_crtc.c
@@ -5742,7 +5742,7 @@ void drm_mode_config_cleanup(struct drm_device *dev)
 	 */
 	WARN_ON(!list_empty(&dev->mode_config.fb_list));
 	list_for_each_entry_safe(fb, fbt, &dev->mode_config.fb_list, head) {
-		drm_framebuffer_remove(fb);
+		drm_framebuffer_free(&fb->refcount);
 	}
 
 	list_for_each_entry_safe(plane, plt, &dev->mode_config.plane_list,
-- 
cgit v0.10.2


From d8b4c43a4baccecee7106c1432baa986fab9593a Mon Sep 17 00:00:00 2001
From: Shashank Sharma <shashank.sharma@intel.com>
Date: Fri, 4 Sep 2015 18:56:11 +0530
Subject: drm/i915: add attached connector to hdmi container

This patch adds the intel_connector initialized to intel_hdmi
display, during the init phase, just like the other encoders do.
This attachment is very useful when we need to extract the connector
pointer during the hotplug handler function

Signed-off-by: Shashank Sharma <shashank.sharma@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 2bdd544..129d99e 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -669,6 +669,7 @@ struct intel_hdmi {
 	enum hdmi_force_audio force_audio;
 	bool rgb_quant_range_selectable;
 	enum hdmi_picture_aspect aspect_ratio;
+	struct intel_connector *attached_connector;
 	void (*write_infoframe)(struct drm_encoder *encoder,
 				enum hdmi_infoframe_type type,
 				const void *frame, ssize_t len);
diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index feb31d8..e978c59 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -2120,6 +2120,7 @@ void intel_hdmi_init_connector(struct intel_digital_port *intel_dig_port,
 
 	intel_connector_attach_encoder(intel_connector, intel_encoder);
 	drm_connector_register(connector);
+	intel_hdmi->attached_connector = intel_connector;
 
 	/* For G4X desktop chip, PEG_BAND_GAP_DATA 3:0 must first be written
 	 * 0xd.  Failure to do so will result in spurious interrupts being
-- 
cgit v0.10.2


From d4b08630fb6e3fe9948456b09feb223fba2b48a6 Mon Sep 17 00:00:00 2001
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Date: Thu, 10 Sep 2015 16:07:56 +0200
Subject: drm/i915: Use atomic plane state in the primary plane update.

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 110b8dc..0f5df11 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -13427,7 +13427,9 @@ intel_commit_primary_plane(struct drm_plane *plane,
 		/* FIXME: kill this fastboot hack */
 		intel_update_pipe_size(intel_crtc);
 
-	dev_priv->display.update_primary_plane(crtc, fb, crtc->x, crtc->y);
+	dev_priv->display.update_primary_plane(crtc, fb,
+					       state->src.x1 >> 16,
+					       state->src.y1 >> 16);
 }
 
 static void
-- 
cgit v0.10.2


From 23a48d5323bc488f0e350467996e7226f902e0a7 Mon Sep 17 00:00:00 2001
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Date: Thu, 10 Sep 2015 16:07:57 +0200
Subject: drm/i915: Use the plane state in intel_crtc_info.

Legacy state might not be updated any more.

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index 41629fa..72ae347 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -2817,11 +2817,13 @@ static void intel_crtc_info(struct seq_file *m, struct intel_crtc *intel_crtc)
 	struct drm_device *dev = node->minor->dev;
 	struct drm_crtc *crtc = &intel_crtc->base;
 	struct intel_encoder *intel_encoder;
+	struct drm_plane_state *plane_state = crtc->primary->state;
+	struct drm_framebuffer *fb = plane_state->fb;
 
-	if (crtc->primary->fb)
+	if (fb)
 		seq_printf(m, "\tfb: %d, pos: %dx%d, size: %dx%d\n",
-			   crtc->primary->fb->base.id, crtc->x, crtc->y,
-			   crtc->primary->fb->width, crtc->primary->fb->height);
+			   fb->base.id, plane_state->src_x >> 16,
+			   plane_state->src_y >> 16, fb->width, fb->height);
 	else
 		seq_puts(m, "\tprimary plane disabled\n");
 	for_each_encoder_on_crtc(dev, crtc, intel_encoder)
-- 
cgit v0.10.2


From 11c22da619101a5a4332df718268e6214e3919f3 Mon Sep 17 00:00:00 2001
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Date: Thu, 10 Sep 2015 16:07:58 +0200
Subject: drm/i915: Use the atomic state in intel_update_primary_planes.

This function was still using the legacy state, convert it to atomic.
While we're at it, fix the FIXME too and disable the primary plane.

v2 (Daniel):
- Add FIXME explaining that update_primary_planes should soon get
  removed anyway.
- Don't call ->disable_plane since we can't disable the primary plane
  with a CS flip (noticed by Ville).

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> (v1)
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 0f5df11..b7e6158 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -3135,24 +3135,20 @@ static void intel_complete_page_flips(struct drm_device *dev)
 
 static void intel_update_primary_planes(struct drm_device *dev)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_crtc *crtc;
 
 	for_each_crtc(dev, crtc) {
-		struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+		struct intel_plane *plane = to_intel_plane(crtc->primary);
+		struct intel_plane_state *plane_state;
 
-		drm_modeset_lock(&crtc->mutex, NULL);
-		/*
-		 * FIXME: Once we have proper support for primary planes (and
-		 * disabling them without disabling the entire crtc) allow again
-		 * a NULL crtc->primary->fb.
-		 */
-		if (intel_crtc->active && crtc->primary->fb)
-			dev_priv->display.update_primary_plane(crtc,
-							       crtc->primary->fb,
-							       crtc->x,
-							       crtc->y);
-		drm_modeset_unlock(&crtc->mutex);
+		drm_modeset_lock_crtc(crtc, &plane->base);
+
+		plane_state = to_intel_plane_state(plane->base.state);
+
+		if (plane_state->base.fb)
+			plane->commit_plane(&plane->base, plane_state);
+
+		drm_modeset_unlock_crtc(crtc);
 	}
 }
 
@@ -3196,6 +3192,9 @@ void intel_finish_reset(struct drm_device *dev)
 		 * so update the base address of all primary
 		 * planes to the the last fb to make sure we're
 		 * showing the correct fb after a reset.
+		 *
+		 * FIXME: Atomic will make this obsolete since we won't schedule
+		 * CS-based flips (which might get lost in gpu resets) any more.
 		 */
 		intel_update_primary_planes(dev);
 		return;
-- 
cgit v0.10.2


From 9b4101be5bb165484d143629d17d084d76fdd77c Mon Sep 17 00:00:00 2001
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Date: Thu, 10 Sep 2015 16:07:59 +0200
Subject: drm/i915: Use atomic state when changing cursor visibility.

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index b7e6158..ea3a4a5 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -9916,8 +9916,9 @@ static void intel_crtc_update_cursor(struct drm_crtc *crtc,
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	int pipe = intel_crtc->pipe;
-	int x = crtc->cursor_x;
-	int y = crtc->cursor_y;
+	struct drm_plane_state *cursor_state = crtc->cursor->state;
+	int x = cursor_state->crtc_x;
+	int y = cursor_state->crtc_y;
 	u32 base = 0, pos = 0;
 
 	if (on)
@@ -9930,7 +9931,7 @@ static void intel_crtc_update_cursor(struct drm_crtc *crtc,
 		base = 0;
 
 	if (x < 0) {
-		if (x + intel_crtc->base.cursor->state->crtc_w <= 0)
+		if (x + cursor_state->crtc_w <= 0)
 			base = 0;
 
 		pos |= CURSOR_POS_SIGN << CURSOR_X_SHIFT;
@@ -9939,7 +9940,7 @@ static void intel_crtc_update_cursor(struct drm_crtc *crtc,
 	pos |= x << CURSOR_X_SHIFT;
 
 	if (y < 0) {
-		if (y + intel_crtc->base.cursor->state->crtc_h <= 0)
+		if (y + cursor_state->crtc_h <= 0)
 			base = 0;
 
 		pos |= CURSOR_POS_SIGN << CURSOR_Y_SHIFT;
@@ -9955,8 +9956,8 @@ static void intel_crtc_update_cursor(struct drm_crtc *crtc,
 	/* ILK+ do this automagically */
 	if (HAS_GMCH_DISPLAY(dev) &&
 	    crtc->cursor->state->rotation == BIT(DRM_ROTATE_180)) {
-		base += (intel_crtc->base.cursor->state->crtc_h *
-			intel_crtc->base.cursor->state->crtc_w - 1) * 4;
+		base += (cursor_state->crtc_h *
+			 cursor_state->crtc_w - 1) * 4;
 	}
 
 	if (IS_845G(dev) || IS_I865G(dev))
-- 
cgit v0.10.2


From 8e0e9ceaa05317c6644635f8816705b609f2c274 Mon Sep 17 00:00:00 2001
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Date: Thu, 10 Sep 2015 16:08:00 +0200
Subject: drm/i915: Remove legacy plane updates for cursor and sprite planes.

Unfortunately fbc still depends on legacy primary state, so
it can't be killed off completely yet.

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index ea3a4a5..62573d5 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -13634,10 +13634,6 @@ intel_commit_cursor_plane(struct drm_plane *plane,
 	crtc = crtc ? crtc : plane->crtc;
 	intel_crtc = to_intel_crtc(crtc);
 
-	plane->fb = state->base.fb;
-	crtc->cursor_x = state->base.crtc_x;
-	crtc->cursor_y = state->base.crtc_y;
-
 	if (intel_crtc->cursor_bo == obj)
 		goto update;
 
diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c
index ca7e264..9553859 100644
--- a/drivers/gpu/drm/i915/intel_sprite.c
+++ b/drivers/gpu/drm/i915/intel_sprite.c
@@ -927,8 +927,6 @@ intel_commit_sprite_plane(struct drm_plane *plane,
 
 	crtc = crtc ? crtc : plane->crtc;
 
-	plane->fb = fb;
-
 	if (!crtc->state->active)
 		return;
 
-- 
cgit v0.10.2


From bca8013ec24210087bfed49678ec821c811982b1 Mon Sep 17 00:00:00 2001
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Date: Thu, 10 Sep 2015 16:08:02 +0200
Subject: drm/i915: Do not handle a null plane state.

With the conversion to atomic this cannot happen any more.

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_atomic_plane.c b/drivers/gpu/drm/i915/intel_atomic_plane.c
index f1ab8e4..a119806 100644
--- a/drivers/gpu/drm/i915/intel_atomic_plane.c
+++ b/drivers/gpu/drm/i915/intel_atomic_plane.c
@@ -76,11 +76,7 @@ intel_plane_duplicate_state(struct drm_plane *plane)
 	struct drm_plane_state *state;
 	struct intel_plane_state *intel_state;
 
-	if (WARN_ON(!plane->state))
-		intel_state = intel_create_plane_state(plane);
-	else
-		intel_state = kmemdup(plane->state, sizeof(*intel_state),
-				      GFP_KERNEL);
+	intel_state = kmemdup(plane->state, sizeof(*intel_state), GFP_KERNEL);
 
 	if (!intel_state)
 		return NULL;
-- 
cgit v0.10.2


From f2a066f3dee8a0e37015d15298e14e9ad6c0a53d Mon Sep 17 00:00:00 2001
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Date: Thu, 10 Sep 2015 16:08:03 +0200
Subject: drm/i915: Use crtc->state for duplication.

In async mode crtc->config can be updated after the locks are released,
resulting in the wrong state being duplicated.

Note that this also removes a spurious assignment of crtc_state->crtc
introduced in

commit f0c60574eb1216b0904c0d696c64d5096d6e4913
Author: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Date:   Tue Apr 21 17:12:58 2015 +0300

    drm/i915: Call drm helpers when duplicating crtc and plane states

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_atomic.c b/drivers/gpu/drm/i915/intel_atomic.c
index 9336e80..2c4b1c4 100644
--- a/drivers/gpu/drm/i915/intel_atomic.c
+++ b/drivers/gpu/drm/i915/intel_atomic.c
@@ -85,22 +85,14 @@ intel_connector_atomic_get_property(struct drm_connector *connector,
 struct drm_crtc_state *
 intel_crtc_duplicate_state(struct drm_crtc *crtc)
 {
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct intel_crtc_state *crtc_state;
 
-	if (WARN_ON(!intel_crtc->config))
-		crtc_state = kzalloc(sizeof(*crtc_state), GFP_KERNEL);
-	else
-		crtc_state = kmemdup(intel_crtc->config,
-				     sizeof(*intel_crtc->config), GFP_KERNEL);
-
+	crtc_state = kmemdup(crtc->state, sizeof(*crtc_state), GFP_KERNEL);
 	if (!crtc_state)
 		return NULL;
 
 	__drm_atomic_helper_crtc_duplicate_state(crtc, &crtc_state->base);
 
-	crtc_state->base.crtc = crtc;
-
 	return &crtc_state->base;
 }
 
-- 
cgit v0.10.2


From f0fea8dd362c77a06e7b8744d63f429fe065c6b4 Mon Sep 17 00:00:00 2001
From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Date: Tue, 8 Sep 2015 12:52:48 +0100
Subject: drm/i915: Remove one very outdated comment

Comment disagrees with the code which has changed a lot since
it was documented.

Note that the logic to remove -EIO handling was dropped in

commit 1488fc08c1706288616c602416654fd38c773deb
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Tue Apr 24 15:47:31 2012 +0100

    drm/i915: Remove the deferred-free list

Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 41263cd..760cfba 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -3230,10 +3230,6 @@ int i915_vma_unbind(struct i915_vma *vma)
 	ret = i915_gem_object_wait_rendering(obj, false);
 	if (ret)
 		return ret;
-	/* Continue on if we fail due to EIO, the GPU is hung so we
-	 * should be safe and we need to cleanup or else we might
-	 * cause memory corruption through use-after-free.
-	 */
 
 	if (i915_is_ggtt(vma->vm) &&
 	    vma->ggtt_view.type == I915_GGTT_VIEW_NORMAL) {
-- 
cgit v0.10.2


From c86fb9d997428e0d55ab5a47488b6ba3bc4b5c5b Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Thu, 10 Sep 2015 22:39:22 +0200
Subject: drm: Nuke drm_framebuffer->helper_private

It's completely unused and there's really no reason for this:
- drm_framebuffer structures are invariant after creation, no need for
  helpers to manipulate them.
- drm_framebuffer structures should just be embedded (and that's what
  all the drivers do).

Stumbled over this since some folks are apparently concerned with the
overhead of struct drm_framebuffer and this is an easy 8 byte saving.

More could be gained by ditching the legacy fields and recomputing
stuff from the fourcc value. But that would require some drm-wide
cocci and real justification.

Cc: gary.k.smith@intel.com
Reviewed-by: David Herrmann <dh.herrmann@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

diff --git a/include/drm/drm_crtc.h b/include/drm/drm_crtc.h
index 75f49c1..c0366e9 100644
--- a/include/drm/drm_crtc.h
+++ b/include/drm/drm_crtc.h
@@ -210,8 +210,6 @@ struct drm_framebuffer {
 	int flags;
 	uint32_t pixel_format; /* fourcc format */
 	struct list_head filp_head;
-	/* if you are using the helper */
-	void *helper_private;
 };
 
 struct drm_property_blob {
-- 
cgit v0.10.2


From fd1ee4cc9326c97b52154ee2ef8cdd23ac6aae1c Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Fri, 11 Sep 2015 21:57:24 +0200
Subject: drm/i915: Update DRIVER_DATE to 20150911

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 3e46187..da0f264 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -57,7 +57,7 @@
 
 #define DRIVER_NAME		"i915"
 #define DRIVER_DESC		"Intel Graphics"
-#define DRIVER_DATE		"20150828"
+#define DRIVER_DATE		"20150911"
 
 #undef WARN_ON
 /* Many gcc seem to no see through this and fall over :( */
-- 
cgit v0.10.2


From 216c59d65f99aa1ef1a92e1ae64f1f1c2590dddc Mon Sep 17 00:00:00 2001
From: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Date: Fri, 11 Sep 2015 00:07:19 +0300
Subject: drm/atomic-helper: Don't skip plane disabling on active CRTC

Since commit "drm/atomic-helper: Add option to update planes only on
active crtc" the drm_atomic_helper_commit_planes() function accepts an
active_only argument to skip updating planes when the associated CRTC is
inactive. Planes being disabled on an active CRTC are incorrectly
considered as associated with an inactive CRTC and are thus skipped,
preventing any plane disabling update from reaching drivers.

Fix it by checking the state of the CRTC stored in the old plane state
for planes being disabled.

Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c
index 9941167..94d6c8e 100644
--- a/drivers/gpu/drm/drm_atomic_helper.c
+++ b/drivers/gpu/drm/drm_atomic_helper.c
@@ -1227,23 +1227,35 @@ void drm_atomic_helper_commit_planes(struct drm_device *dev,
 
 	for_each_plane_in_state(old_state, plane, old_plane_state, i) {
 		const struct drm_plane_helper_funcs *funcs;
+		bool disabling;
 
 		funcs = plane->helper_private;
 
 		if (!funcs)
 			continue;
 
-		if (active_only && !plane_crtc_active(plane->state))
-			continue;
+		disabling = drm_atomic_plane_disabling(plane, old_plane_state);
+
+		if (active_only) {
+			/*
+			 * Skip planes related to inactive CRTCs. If the plane
+			 * is enabled use the state of the current CRTC. If the
+			 * plane is being disabled use the state of the old
+			 * CRTC to avoid skipping planes being disabled on an
+			 * active CRTC.
+			 */
+			if (!disabling && !plane_crtc_active(plane->state))
+				continue;
+			if (disabling && !plane_crtc_active(old_plane_state))
+				continue;
+		}
 
 		/*
 		 * Special-case disabling the plane if drivers support it.
 		 */
-		if (drm_atomic_plane_disabling(plane, old_plane_state) &&
-		    funcs->atomic_disable)
+		if (disabling && funcs->atomic_disable)
 			funcs->atomic_disable(plane, old_plane_state);
-		else if (plane->state->crtc ||
-			 drm_atomic_plane_disabling(plane, old_plane_state))
+		else if (plane->state->crtc || disabling)
 			funcs->atomic_update(plane, old_plane_state);
 	}
 
-- 
cgit v0.10.2


From 44522d852f1979ec0e6807eb4dc58bd7ecc8af08 Mon Sep 17 00:00:00 2001
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Date: Thu, 27 Aug 2015 15:44:02 +0200
Subject: drm/i915: Set csc coefficients in update_pipe_size.

This might not have been set during boot, and when we preserve
the initial mode this can result in a black screen.

Cc: Daniel Stone <daniels@collabora.com>
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 62573d5..1904919 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -3274,6 +3274,9 @@ static void intel_update_pipe_size(struct intel_crtc *crtc)
 	if (!i915.fastboot)
 		return;
 
+	if (HAS_DDI(dev))
+		intel_set_pipe_csc(&crtc->base);
+
 	/*
 	 * Update pipe size and adjust fitter if needed: the reason for this is
 	 * that in compute_mode_changes we check the native mode (not the pfit
-- 
cgit v0.10.2


From d551599181769571f4f68dd93e5d8b15868889af Mon Sep 17 00:00:00 2001
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Date: Thu, 27 Aug 2015 15:44:03 +0200
Subject: drm/i915: Remove references to crtc->active from intel_fbdev.c

It should really use the atomic state.

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_fbdev.c b/drivers/gpu/drm/i915/intel_fbdev.c
index 96476d7..0c9291d 100644
--- a/drivers/gpu/drm/i915/intel_fbdev.c
+++ b/drivers/gpu/drm/i915/intel_fbdev.c
@@ -550,7 +550,7 @@ static bool intel_fbdev_init_bios(struct drm_device *dev,
 			intel_fb_obj(crtc->primary->state->fb);
 		intel_crtc = to_intel_crtc(crtc);
 
-		if (!intel_crtc->active || !obj) {
+		if (!crtc->state->active || !obj) {
 			DRM_DEBUG_KMS("pipe %c not active or no fb, skipping\n",
 				      pipe_name(intel_crtc->pipe));
 			continue;
@@ -575,7 +575,7 @@ static bool intel_fbdev_init_bios(struct drm_device *dev,
 
 		intel_crtc = to_intel_crtc(crtc);
 
-		if (!intel_crtc->active) {
+		if (!crtc->state->active) {
 			DRM_DEBUG_KMS("pipe %c not active, skipping\n",
 				      pipe_name(intel_crtc->pipe));
 			continue;
@@ -638,7 +638,7 @@ static bool intel_fbdev_init_bios(struct drm_device *dev,
 	for_each_crtc(dev, crtc) {
 		intel_crtc = to_intel_crtc(crtc);
 
-		if (!intel_crtc->active)
+		if (!crtc->state->active)
 			continue;
 
 		WARN(!crtc->primary->fb,
-- 
cgit v0.10.2


From f4502c25ebd04691f284fdafff4a5613299c36dc Mon Sep 17 00:00:00 2001
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Date: Thu, 27 Aug 2015 15:44:04 +0200
Subject: drm/i915: Always try to inherit the initial fb.

The initial state is read out correctly and the state is atomic,
so it's safe to preserve the fb without any hacks if it's suitable.

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_fbdev.c b/drivers/gpu/drm/i915/intel_fbdev.c
index 0c9291d..73a14e8 100644
--- a/drivers/gpu/drm/i915/intel_fbdev.c
+++ b/drivers/gpu/drm/i915/intel_fbdev.c
@@ -541,9 +541,6 @@ static bool intel_fbdev_init_bios(struct drm_device *dev,
 	struct intel_crtc *intel_crtc;
 	unsigned int max_size = 0;
 
-	if (!i915.fastboot)
-		return false;
-
 	/* Find the largest fb */
 	for_each_crtc(dev, crtc) {
 		struct drm_i915_gem_object *obj =
-- 
cgit v0.10.2


From bfd16b2a23dc99f2af3f5c4f54aa153347f2dc3c Mon Sep 17 00:00:00 2001
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Date: Thu, 27 Aug 2015 15:44:05 +0200
Subject: drm/i915: Make updating pipe without modeset atomic.

Instead of doing a hack during primary plane commit the state
is updated during atomic evasion. It handles differences in
pipe size and the panel fitter.

This is continuing on top of Daniel's work to make faster
modesets atomic, and not yet enabled by default.

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
[danvet:
- simplify/future-proof if ladder that Jesse spotted
- resolve conflict in pipe_config_check and don't spuriously move the
  code.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_atomic.c b/drivers/gpu/drm/i915/intel_atomic.c
index 2c4b1c4..f1975f2 100644
--- a/drivers/gpu/drm/i915/intel_atomic.c
+++ b/drivers/gpu/drm/i915/intel_atomic.c
@@ -93,6 +93,8 @@ intel_crtc_duplicate_state(struct drm_crtc *crtc)
 
 	__drm_atomic_helper_crtc_duplicate_state(crtc, &crtc_state->base);
 
+	crtc_state->update_pipe = false;
+
 	return &crtc_state->base;
 }
 
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 1904919..1691ec2 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -112,6 +112,9 @@ static void skl_init_scalers(struct drm_device *dev, struct intel_crtc *intel_cr
 	struct intel_crtc_state *crtc_state);
 static int i9xx_get_refclk(const struct intel_crtc_state *crtc_state,
 			   int num_connectors);
+static void skylake_pfit_enable(struct intel_crtc *crtc);
+static void ironlake_pfit_disable(struct intel_crtc *crtc, bool force);
+static void ironlake_pfit_enable(struct intel_crtc *crtc);
 static void intel_modeset_setup_hw_state(struct drm_device *dev);
 
 typedef struct {
@@ -3265,14 +3268,20 @@ static bool intel_crtc_has_pending_flip(struct drm_crtc *crtc)
 	return pending;
 }
 
-static void intel_update_pipe_size(struct intel_crtc *crtc)
+static void intel_update_pipe_config(struct intel_crtc *crtc,
+				     struct intel_crtc_state *old_crtc_state)
 {
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	const struct drm_display_mode *adjusted_mode;
+	struct intel_crtc_state *pipe_config =
+		to_intel_crtc_state(crtc->base.state);
 
-	if (!i915.fastboot)
-		return;
+	/* drm_atomic_helper_update_legacy_modeset_state might not be called. */
+	crtc->base.mode = crtc->base.state->mode;
+
+	DRM_DEBUG_KMS("Updating pipe size %ix%i -> %ix%i\n",
+		      old_crtc_state->pipe_src_w, old_crtc_state->pipe_src_h,
+		      pipe_config->pipe_src_w, pipe_config->pipe_src_h);
 
 	if (HAS_DDI(dev))
 		intel_set_pipe_csc(&crtc->base);
@@ -3284,27 +3293,24 @@ static void intel_update_pipe_size(struct intel_crtc *crtc)
 	 * fastboot case, we'll flip, but if we don't update the pipesrc and
 	 * pfit state, we'll end up with a big fb scanned out into the wrong
 	 * sized surface.
-	 *
-	 * To fix this properly, we need to hoist the checks up into
-	 * compute_mode_changes (or above), check the actual pfit state and
-	 * whether the platform allows pfit disable with pipe active, and only
-	 * then update the pipesrc and pfit state, even on the flip path.
 	 */
 
-	adjusted_mode = &crtc->config->base.adjusted_mode;
-
 	I915_WRITE(PIPESRC(crtc->pipe),
-		   ((adjusted_mode->crtc_hdisplay - 1) << 16) |
-		   (adjusted_mode->crtc_vdisplay - 1));
-	if (!crtc->config->pch_pfit.enabled &&
-	    (intel_pipe_has_type(crtc, INTEL_OUTPUT_LVDS) ||
-	     intel_pipe_has_type(crtc, INTEL_OUTPUT_EDP))) {
-		I915_WRITE(PF_CTL(crtc->pipe), 0);
-		I915_WRITE(PF_WIN_POS(crtc->pipe), 0);
-		I915_WRITE(PF_WIN_SZ(crtc->pipe), 0);
+		   ((pipe_config->pipe_src_w - 1) << 16) |
+		   (pipe_config->pipe_src_h - 1));
+
+	/* on skylake this is done by detaching scalers */
+	if (INTEL_INFO(dev)->gen >= 9) {
+		skl_detach_scalers(crtc);
+
+		if (pipe_config->pch_pfit.enabled)
+			skylake_pfit_enable(crtc);
+	} else if (HAS_PCH_SPLIT(dev)) {
+		if (pipe_config->pch_pfit.enabled)
+			ironlake_pfit_enable(crtc);
+		else if (old_crtc_state->pch_pfit.enabled)
+			ironlake_pfit_disable(crtc, true);
 	}
-	crtc->config->pipe_src_w = adjusted_mode->crtc_hdisplay;
-	crtc->config->pipe_src_h = adjusted_mode->crtc_vdisplay;
 }
 
 static void intel_fdi_normal_train(struct drm_crtc *crtc)
@@ -4961,7 +4967,7 @@ static void haswell_crtc_enable(struct drm_crtc *crtc)
 	}
 }
 
-static void ironlake_pfit_disable(struct intel_crtc *crtc)
+static void ironlake_pfit_disable(struct intel_crtc *crtc, bool force)
 {
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -4969,7 +4975,7 @@ static void ironlake_pfit_disable(struct intel_crtc *crtc)
 
 	/* To avoid upsetting the power well on haswell only disable the pfit if
 	 * it's in use. The hw state code will make sure we get this right. */
-	if (crtc->config->pch_pfit.enabled) {
+	if (force || crtc->config->pch_pfit.enabled) {
 		I915_WRITE(PF_CTL(pipe), 0);
 		I915_WRITE(PF_WIN_POS(pipe), 0);
 		I915_WRITE(PF_WIN_SZ(pipe), 0);
@@ -4996,7 +5002,7 @@ static void ironlake_crtc_disable(struct drm_crtc *crtc)
 
 	intel_disable_pipe(intel_crtc);
 
-	ironlake_pfit_disable(intel_crtc);
+	ironlake_pfit_disable(intel_crtc, false);
 
 	if (intel_crtc->config->has_pch_encoder)
 		ironlake_fdi_disable(crtc);
@@ -5059,7 +5065,7 @@ static void haswell_crtc_disable(struct drm_crtc *crtc)
 	if (INTEL_INFO(dev)->gen >= 9)
 		skylake_scaler_disable(intel_crtc);
 	else
-		ironlake_pfit_disable(intel_crtc);
+		ironlake_pfit_disable(intel_crtc, false);
 
 	intel_ddi_disable_pipe_clock(intel_crtc);
 
@@ -12218,7 +12224,6 @@ static bool intel_fuzzy_clock_check(int clock1, int clock2)
 			    base.head) \
 		if (mask & (1 <<(intel_crtc)->pipe))
 
-
 static bool
 intel_compare_m_n(unsigned int m, unsigned int n,
 		  unsigned int m2, unsigned int n2,
@@ -12439,19 +12444,21 @@ intel_pipe_config_compare(struct drm_device *dev,
 				      DRM_MODE_FLAG_NVSYNC);
 	}
 
-	PIPE_CONF_CHECK_I(pipe_src_w);
-	PIPE_CONF_CHECK_I(pipe_src_h);
-
 	PIPE_CONF_CHECK_X(gmch_pfit.control);
 	/* pfit ratios are autocomputed by the hw on gen4+ */
 	if (INTEL_INFO(dev)->gen < 4)
 		PIPE_CONF_CHECK_I(gmch_pfit.pgm_ratios);
 	PIPE_CONF_CHECK_X(gmch_pfit.lvds_border_bits);
 
-	PIPE_CONF_CHECK_I(pch_pfit.enabled);
-	if (current_config->pch_pfit.enabled) {
-		PIPE_CONF_CHECK_X(pch_pfit.pos);
-		PIPE_CONF_CHECK_X(pch_pfit.size);
+	if (!adjust) {
+		PIPE_CONF_CHECK_I(pipe_src_w);
+		PIPE_CONF_CHECK_I(pipe_src_h);
+
+		PIPE_CONF_CHECK_I(pch_pfit.enabled);
+		if (current_config->pch_pfit.enabled) {
+			PIPE_CONF_CHECK_X(pch_pfit.pos);
+			PIPE_CONF_CHECK_X(pch_pfit.size);
+		}
 	}
 
 	PIPE_CONF_CHECK_I(scaler_state.scaler_id);
@@ -12616,7 +12623,8 @@ check_crtc_state(struct drm_device *dev, struct drm_atomic_state *old_state)
 		struct intel_crtc_state *pipe_config, *sw_config;
 		bool active;
 
-		if (!needs_modeset(crtc->state))
+		if (!needs_modeset(crtc->state) &&
+		    !to_intel_crtc_state(crtc->state)->update_pipe)
 			continue;
 
 		__drm_atomic_helper_crtc_destroy_state(crtc, old_crtc_state);
@@ -12912,7 +12920,6 @@ static int intel_modeset_all_pipes(struct drm_atomic_state *state)
 	return ret;
 }
 
-
 static int intel_modeset_checks(struct drm_atomic_state *state)
 {
 	struct drm_device *dev = state->dev;
@@ -13003,6 +13010,7 @@ static int intel_atomic_check(struct drm_device *dev,
 					to_intel_crtc_state(crtc->state),
 					pipe_config, true)) {
 			crtc_state->mode_changed = false;
+			to_intel_crtc_state(crtc_state)->update_pipe = true;
 		}
 
 		if (needs_modeset(crtc_state)) {
@@ -13100,16 +13108,30 @@ static int intel_atomic_commit(struct drm_device *dev,
 	for_each_crtc_in_state(state, crtc, crtc_state, i) {
 		struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 		bool modeset = needs_modeset(crtc->state);
+		bool update_pipe = !modeset &&
+			to_intel_crtc_state(crtc->state)->update_pipe;
+		unsigned long put_domains = 0;
 
 		if (modeset && crtc->state->active) {
 			update_scanline_offset(to_intel_crtc(crtc));
 			dev_priv->display.crtc_enable(crtc);
 		}
 
+		if (update_pipe) {
+			put_domains = modeset_get_crtc_power_domains(crtc);
+
+			/* make sure intel_modeset_check_state runs */
+			any_ms = true;
+		}
+
 		if (!modeset)
 			intel_pre_plane_update(intel_crtc);
 
 		drm_atomic_helper_commit_planes_on_crtc(crtc_state);
+
+		if (put_domains)
+			modeset_put_power_domains(dev_priv, put_domains);
+
 		intel_post_plane_update(intel_crtc);
 	}
 
@@ -13426,10 +13448,6 @@ intel_commit_primary_plane(struct drm_plane *plane,
 	if (!crtc->state->active)
 		return;
 
-	if (state->visible)
-		/* FIXME: kill this fastboot hack */
-		intel_update_pipe_size(intel_crtc);
-
 	dev_priv->display.update_primary_plane(crtc, fb,
 					       state->src.x1 >> 16,
 					       state->src.y1 >> 16);
@@ -13450,6 +13468,9 @@ static void intel_begin_crtc_commit(struct drm_crtc *crtc,
 {
 	struct drm_device *dev = crtc->dev;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	struct intel_crtc_state *old_intel_state =
+		to_intel_crtc_state(old_crtc_state);
+	bool modeset = needs_modeset(crtc->state);
 
 	if (intel_crtc->atomic.update_wm_pre)
 		intel_update_watermarks(crtc);
@@ -13458,7 +13479,12 @@ static void intel_begin_crtc_commit(struct drm_crtc *crtc,
 	if (crtc->state->active)
 		intel_pipe_update_start(intel_crtc);
 
-	if (!needs_modeset(crtc->state) && INTEL_INFO(dev)->gen >= 9)
+	if (modeset)
+		return;
+
+	if (to_intel_crtc_state(crtc->state)->update_pipe)
+		intel_update_pipe_config(intel_crtc, old_intel_state);
+	else if (INTEL_INFO(dev)->gen >= 9)
 		skl_detach_scalers(intel_crtc);
 }
 
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 129d99e..e9e5cfd 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -338,6 +338,8 @@ struct intel_crtc_state {
 #define PIPE_CONFIG_QUIRK_MODE_SYNC_FLAGS	(1<<0) /* unreliable sync mode.flags */
 	unsigned long quirks;
 
+	bool update_pipe;
+
 	/* Pipe source size (ie. panel fitter input size)
 	 * All planes will be positioned inside this space,
 	 * and get clipped at the edges. */
-- 
cgit v0.10.2


From 6764e9f8724f1231b4deac53b9a82286ac0830e7 Mon Sep 17 00:00:00 2001
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Date: Thu, 27 Aug 2015 15:44:06 +0200
Subject: drm/i915: skip modeset if compatible for everyone.

This is done as a separate commit, to make it easier to revert
when things break.

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index da0f264..2994ce5 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -2644,7 +2644,6 @@ struct i915_params {
 	int enable_cmd_parser;
 	/* leave bools at the end to not create holes */
 	bool enable_hangcheck;
-	bool fastboot;
 	bool prefault_disable;
 	bool load_detect_test;
 	bool reset;
diff --git a/drivers/gpu/drm/i915/i915_params.c b/drivers/gpu/drm/i915/i915_params.c
index 05053e2..ca060d5 100644
--- a/drivers/gpu/drm/i915/i915_params.c
+++ b/drivers/gpu/drm/i915/i915_params.c
@@ -40,7 +40,6 @@ struct i915_params i915 __read_mostly = {
 	.preliminary_hw_support = IS_ENABLED(CONFIG_DRM_I915_PRELIMINARY_HW_SUPPORT),
 	.disable_power_well = 1,
 	.enable_ips = 1,
-	.fastboot = 0,
 	.prefault_disable = 0,
 	.load_detect_test = 0,
 	.reset = true,
@@ -133,10 +132,6 @@ MODULE_PARM_DESC(disable_power_well,
 module_param_named(enable_ips, i915.enable_ips, int, 0600);
 MODULE_PARM_DESC(enable_ips, "Enable IPS (default: true)");
 
-module_param_named(fastboot, i915.fastboot, bool, 0600);
-MODULE_PARM_DESC(fastboot,
-	"Try to skip unnecessary mode sets at boot time (default: false)");
-
 module_param_named_unsafe(prefault_disable, i915.prefault_disable, bool, 0600);
 MODULE_PARM_DESC(prefault_disable,
 	"Disable page prefaulting for pread/pwrite/reloc (default:false). "
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 1691ec2..a001cbb 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -13005,8 +13005,7 @@ static int intel_atomic_check(struct drm_device *dev,
 		if (ret)
 			return ret;
 
-		if (i915.fastboot &&
-		    intel_pipe_config_compare(state->dev,
+		if (intel_pipe_config_compare(state->dev,
 					to_intel_crtc_state(crtc->state),
 					pipe_config, true)) {
 			crtc_state->mode_changed = false;
-- 
cgit v0.10.2


From ec72d5884c6a3cf4f27b52da51654a472b919545 Mon Sep 17 00:00:00 2001
From: Michel Thierry <michel.thierry@intel.com>
Date: Fri, 4 Sep 2015 12:59:14 +0100
Subject: drm/i915: WaEnableForceRestoreInCtxtDescForVCS is for video engines
 only

Also check for correct revision id in each Gen9 platform (SKL until B0
and BXT until A0).

Cc: Nick Hoath <nicholas.hoath@intel.com>
Signed-off-by: Michel Thierry <michel.thierry@intel.com>
Reviewed-by: Arun Siluvery <arun.siluvery@linux.intel.com>
Tested-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
index 28a712e..d8b605f 100644
--- a/drivers/gpu/drm/i915/intel_lrc.c
+++ b/drivers/gpu/drm/i915/intel_lrc.c
@@ -301,10 +301,10 @@ uint64_t intel_lr_context_descriptor(struct intel_context *ctx,
 	/* desc |= GEN8_CTX_FORCE_RESTORE; */
 
 	/* WaEnableForceRestoreInCtxtDescForVCS:skl */
-	if (IS_GEN9(dev) &&
-	    INTEL_REVID(dev) <= SKL_REVID_B0 &&
-	    (ring->id == BCS || ring->id == VCS ||
-	    ring->id == VECS || ring->id == VCS2))
+	/* WaEnableForceRestoreInCtxtDescForVCS:bxt */
+	if (((IS_SKYLAKE(dev) && INTEL_REVID(dev) <= SKL_REVID_B0) ||
+	     (IS_BROXTON(dev) && INTEL_REVID(dev) == BXT_REVID_A0)) &&
+	    (ring->id == VCS || ring->id == VCS2))
 		desc |= GEN8_CTX_FORCE_RESTORE;
 
 	return desc;
-- 
cgit v0.10.2


From 5af05fefb5be51d42effc4f802b6d806fbda34eb Mon Sep 17 00:00:00 2001
From: Michel Thierry <michel.thierry@intel.com>
Date: Fri, 4 Sep 2015 12:59:15 +0100
Subject: drm/i915/lrc: Prevent preemption when lite-restore is disabled

When WaEnableForceRestoreInCtxtDescForVCS is required, it is only
safe to send new contexts if the last reported event is "active to
idle". Otherwise the same context can fully preempt itself because
lite-restore is disabled.

Testcase: igt/gem_concurrent_blit
Reported-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Michel Thierry <michel.thierry@intel.com>
Reviewed-by: Arun Siluvery <arun.siluvery@linux.intel.com>
Tested-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
index d8b605f..c3fca4b 100644
--- a/drivers/gpu/drm/i915/intel_lrc.c
+++ b/drivers/gpu/drm/i915/intel_lrc.c
@@ -277,10 +277,18 @@ u32 intel_execlists_ctx_id(struct drm_i915_gem_object *ctx_obj)
 	return lrca >> 12;
 }
 
+static bool disable_lite_restore_wa(struct intel_engine_cs *ring)
+{
+	struct drm_device *dev = ring->dev;
+
+	return ((IS_SKYLAKE(dev) && INTEL_REVID(dev) <= SKL_REVID_B0) ||
+		(IS_BROXTON(dev) && INTEL_REVID(dev) == BXT_REVID_A0)) &&
+	       (ring->id == VCS || ring->id == VCS2);
+}
+
 uint64_t intel_lr_context_descriptor(struct intel_context *ctx,
 				     struct intel_engine_cs *ring)
 {
-	struct drm_device *dev = ring->dev;
 	struct drm_i915_gem_object *ctx_obj = ctx->engine[ring->id].state;
 	uint64_t desc;
 	uint64_t lrca = i915_gem_obj_ggtt_offset(ctx_obj) +
@@ -302,9 +310,7 @@ uint64_t intel_lr_context_descriptor(struct intel_context *ctx,
 
 	/* WaEnableForceRestoreInCtxtDescForVCS:skl */
 	/* WaEnableForceRestoreInCtxtDescForVCS:bxt */
-	if (((IS_SKYLAKE(dev) && INTEL_REVID(dev) <= SKL_REVID_B0) ||
-	     (IS_BROXTON(dev) && INTEL_REVID(dev) == BXT_REVID_A0)) &&
-	    (ring->id == VCS || ring->id == VCS2))
+	if (disable_lite_restore_wa(ring))
 		desc |= GEN8_CTX_FORCE_RESTORE;
 
 	return desc;
@@ -495,7 +501,7 @@ void intel_lrc_irq_handler(struct intel_engine_cs *ring)
 	u32 status_pointer;
 	u8 read_pointer;
 	u8 write_pointer;
-	u32 status;
+	u32 status = 0;
 	u32 status_id;
 	u32 submit_contexts = 0;
 
@@ -533,8 +539,14 @@ void intel_lrc_irq_handler(struct intel_engine_cs *ring)
 		}
 	}
 
-	if (submit_contexts != 0)
+	if (disable_lite_restore_wa(ring)) {
+		/* Prevent a ctx to preempt itself */
+		if ((status & GEN8_CTX_STATUS_ACTIVE_IDLE) &&
+		    (submit_contexts != 0))
+			execlists_context_unqueue(ring);
+	} else if (submit_contexts != 0) {
 		execlists_context_unqueue(ring);
+	}
 
 	spin_unlock(&ring->execlist_lock);
 
-- 
cgit v0.10.2


From e2ec35a58b3c9cdadbb7d913e8262ad9ddcb00f7 Mon Sep 17 00:00:00 2001
From: Sonika Jindal <sonika.jindal@intel.com>
Date: Fri, 11 Sep 2015 16:58:32 +0530
Subject: drm/i915/bxt: Use intel_encoder->hpd_pin to check live status

Using intel_encoder's hpd_pin to check the live status
because of BXT A0/A1 WA for HPD pins and hpd_pin contains the
updated pin for the corresponding port.

Signed-off-by: Sonika Jindal <sonika.jindal@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 45ab25e..b8c4fc5 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -4637,11 +4637,14 @@ static bool vlv_digital_port_connected(struct drm_i915_private *dev_priv,
 }
 
 static bool bxt_digital_port_connected(struct drm_i915_private *dev_priv,
-				       struct intel_digital_port *port)
+				       struct intel_digital_port *intel_dig_port)
 {
+	struct intel_encoder *intel_encoder = &intel_dig_port->base;
+	enum port port;
 	u32 bit;
 
-	switch (port->port) {
+	intel_hpd_pin_to_port(intel_encoder->hpd_pin, &port);
+	switch (port) {
 	case PORT_A:
 		bit = BXT_DE_PORT_HP_DDIA;
 		break;
@@ -4652,7 +4655,7 @@ static bool bxt_digital_port_connected(struct drm_i915_private *dev_priv,
 		bit = BXT_DE_PORT_HP_DDIC;
 		break;
 	default:
-		MISSING_CASE(port->port);
+		MISSING_CASE(port);
 		return false;
 	}
 
-- 
cgit v0.10.2


From 8c761609ec3beca88819e28d4fcf333b579d29b4 Mon Sep 17 00:00:00 2001
From: Arun Siluvery <arun.siluvery@linux.intel.com>
Date: Tue, 8 Sep 2015 10:31:48 +0100
Subject: drm/i915/gen9: Add WaDisableSamplerPowerBypassForSOPingPong

Signed-off-by: Arun Siluvery <arun.siluvery@linux.intel.com>
Reviewed-by: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index 20a75bb..684e069 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -983,6 +983,13 @@ static int gen9_init_workarounds(struct intel_engine_cs *ring)
 		tmp |= HDC_FORCE_CSR_NON_COHERENT_OVR_DISABLE;
 	WA_SET_BIT_MASKED(HDC_CHICKEN0, tmp);
 
+	/* WaDisableSamplerPowerBypassForSOPingPong:skl,bxt */
+	if (IS_SKYLAKE(dev) ||
+	    (IS_BROXTON(dev) && INTEL_REVID(dev) <= BXT_REVID_B0)) {
+		WA_SET_BIT_MASKED(HALF_SLICE_CHICKEN3,
+				  GEN8_SAMPLER_POWER_BYPASS_DIS);
+	}
+
 	return 0;
 }
 
-- 
cgit v0.10.2


From 5b88abacd4820c66bc347a6beeb3d6ad1144544b Mon Sep 17 00:00:00 2001
From: Arun Siluvery <arun.siluvery@linux.intel.com>
Date: Tue, 8 Sep 2015 10:31:49 +0100
Subject: drm/i915/bxt: Add WaSetClckGatingDisableMedia

Signed-off-by: Arun Siluvery <arun.siluvery@linux.intel.com>
Reviewed-by: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 65b5682..495ac17 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -6897,6 +6897,7 @@ enum skl_disp_power_wells {
 #define   GEN7_DOP_CLOCK_GATE_ENABLE		(1<<0)
 #define   GEN8_DOP_CLOCK_GATE_CFCLK_ENABLE	(1<<2)
 #define   GEN8_DOP_CLOCK_GATE_GUC_ENABLE	(1<<4)
+#define   GEN8_DOP_CLOCK_GATE_MEDIA_ENABLE     (1<<6)
 
 #define GEN8_GARBCNTL                   0xB004
 #define   GEN9_GAPS_TSV_CREDIT_DISABLE  (1<<7)
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index 64bc77e..920872a 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -134,6 +134,12 @@ static void bxt_init_clock_gating(struct drm_device *dev)
 		 */
 		I915_WRITE(TILECTL, I915_READ(TILECTL) | TILECTL_TLBPF);
 	}
+
+	/* WaSetClckGatingDisableMedia:bxt */
+	if (INTEL_REVID(dev) == BXT_REVID_A0) {
+		I915_WRITE(GEN7_MISCCPCTL, (I915_READ(GEN7_MISCCPCTL) &
+					    ~GEN8_DOP_CLOCK_GATE_MEDIA_ENABLE));
+	}
 }
 
 static void i915_pineview_get_mem_freq(struct drm_device *dev)
-- 
cgit v0.10.2


From b970b48685a7ae2953e659f7e657cde2949b12ae Mon Sep 17 00:00:00 2001
From: Nick Hoath <nicholas.hoath@intel.com>
Date: Tue, 8 Sep 2015 10:31:53 +0100
Subject: drm/i915/gen9: Add WaDisableMinuteIaClockGating

Signed-off-by: Nick Hoath <nicholas.hoath@intel.com>
Signed-off-by: Arun Siluvery <arun.siluvery@linux.intel.com>
Reviewed-by: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_guc_loader.c b/drivers/gpu/drm/i915/intel_guc_loader.c
index 5eafd31..e0601cc 100644
--- a/drivers/gpu/drm/i915/intel_guc_loader.c
+++ b/drivers/gpu/drm/i915/intel_guc_loader.c
@@ -330,6 +330,13 @@ static int guc_ucode_xfer(struct drm_i915_private *dev_priv)
 	/* Enable MIA caching. GuC clock gating is disabled. */
 	I915_WRITE(GUC_SHIM_CONTROL, GUC_SHIM_CONTROL_VALUE);
 
+	/* WaDisableMinuteIaClockGating:skl,bxt */
+	if ((IS_SKYLAKE(dev) && INTEL_REVID(dev) <= SKL_REVID_B0) ||
+	    (IS_BROXTON(dev) && INTEL_REVID(dev) == BXT_REVID_A0)) {
+		I915_WRITE(GUC_SHIM_CONTROL, (I915_READ(GUC_SHIM_CONTROL) &
+					      ~GUC_ENABLE_MIA_CLOCK_GATING));
+	}
+
 	/* WaC6DisallowByGfxPause*/
 	I915_WRITE(GEN6_GFXPAUSE, 0x30FFF);
 
-- 
cgit v0.10.2


From 6ff8ab0d0fab914755cb79b1e0469dd7123cc471 Mon Sep 17 00:00:00 2001
From: Jesse Barnes <jbarnes@virtuousgeek.org>
Date: Thu, 10 Sep 2015 08:20:28 -0700
Subject: drm/i915: make CSR firmware messages less verbose

Use WARN_ONCE in a bunch of places and demote a message that would
continually spam us.

Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Acked-by: Damien Lespiau <damien.lespiau@intel.com>
Acked-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_csr.c b/drivers/gpu/drm/i915/intel_csr.c
index ba1ae03..765dfcd 100644
--- a/drivers/gpu/drm/i915/intel_csr.c
+++ b/drivers/gpu/drm/i915/intel_csr.c
@@ -454,10 +454,10 @@ void intel_csr_ucode_fini(struct drm_device *dev)
 
 void assert_csr_loaded(struct drm_i915_private *dev_priv)
 {
-	WARN(intel_csr_load_status_get(dev_priv) != FW_LOADED,
-	     "CSR is not loaded.\n");
-	WARN(!I915_READ(CSR_PROGRAM_BASE),
-				"CSR program storage start is NULL\n");
-	WARN(!I915_READ(CSR_SSP_BASE), "CSR SSP Base Not fine\n");
-	WARN(!I915_READ(CSR_HTP_SKL), "CSR HTP Not fine\n");
+	WARN_ONCE(intel_csr_load_status_get(dev_priv) != FW_LOADED,
+		  "CSR is not loaded.\n");
+	WARN_ONCE(!I915_READ(CSR_PROGRAM_BASE),
+		  "CSR program storage start is NULL\n");
+	WARN_ONCE(!I915_READ(CSR_SSP_BASE), "CSR SSP Base Not fine\n");
+	WARN_ONCE(!I915_READ(CSR_HTP_SKL), "CSR HTP Not fine\n");
 }
diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c
index 3f682a1..85c35fd 100644
--- a/drivers/gpu/drm/i915/intel_runtime_pm.c
+++ b/drivers/gpu/drm/i915/intel_runtime_pm.c
@@ -463,14 +463,14 @@ static void assert_can_enable_dc5(struct drm_i915_private *dev_priv)
 	bool pg2_enabled = intel_display_power_well_is_enabled(dev_priv,
 					SKL_DISP_PW_2);
 
-	WARN(!IS_SKYLAKE(dev), "Platform doesn't support DC5.\n");
-	WARN(!HAS_RUNTIME_PM(dev), "Runtime PM not enabled.\n");
-	WARN(pg2_enabled, "PG2 not disabled to enable DC5.\n");
+	WARN_ONCE(!IS_SKYLAKE(dev), "Platform doesn't support DC5.\n");
+	WARN_ONCE(!HAS_RUNTIME_PM(dev), "Runtime PM not enabled.\n");
+	WARN_ONCE(pg2_enabled, "PG2 not disabled to enable DC5.\n");
 
-	WARN((I915_READ(DC_STATE_EN) & DC_STATE_EN_UPTO_DC5),
-				"DC5 already programmed to be enabled.\n");
-	WARN(dev_priv->pm.suspended,
-		"DC5 cannot be enabled, if platform is runtime-suspended.\n");
+	WARN_ONCE((I915_READ(DC_STATE_EN) & DC_STATE_EN_UPTO_DC5),
+		  "DC5 already programmed to be enabled.\n");
+	WARN_ONCE(dev_priv->pm.suspended,
+		  "DC5 cannot be enabled, if platform is runtime-suspended.\n");
 
 	assert_csr_loaded(dev_priv);
 }
@@ -486,8 +486,8 @@ static void assert_can_disable_dc5(struct drm_i915_private *dev_priv)
 	if (dev_priv->power_domains.initializing)
 		return;
 
-	WARN(!pg2_enabled, "PG2 not enabled to disable DC5.\n");
-	WARN(dev_priv->pm.suspended,
+	WARN_ONCE(!pg2_enabled, "PG2 not enabled to disable DC5.\n");
+	WARN_ONCE(dev_priv->pm.suspended,
 		"Disabling of DC5 while platform is runtime-suspended should never happen.\n");
 }
 
@@ -526,12 +526,12 @@ static void assert_can_enable_dc6(struct drm_i915_private *dev_priv)
 {
 	struct drm_device *dev = dev_priv->dev;
 
-	WARN(!IS_SKYLAKE(dev), "Platform doesn't support DC6.\n");
-	WARN(!HAS_RUNTIME_PM(dev), "Runtime PM not enabled.\n");
-	WARN(I915_READ(UTIL_PIN_CTL) & UTIL_PIN_ENABLE,
-		"Backlight is not disabled.\n");
-	WARN((I915_READ(DC_STATE_EN) & DC_STATE_EN_UPTO_DC6),
-		"DC6 already programmed to be enabled.\n");
+	WARN_ONCE(!IS_SKYLAKE(dev), "Platform doesn't support DC6.\n");
+	WARN_ONCE(!HAS_RUNTIME_PM(dev), "Runtime PM not enabled.\n");
+	WARN_ONCE(I915_READ(UTIL_PIN_CTL) & UTIL_PIN_ENABLE,
+		  "Backlight is not disabled.\n");
+	WARN_ONCE((I915_READ(DC_STATE_EN) & DC_STATE_EN_UPTO_DC6),
+		  "DC6 already programmed to be enabled.\n");
 
 	assert_csr_loaded(dev_priv);
 }
@@ -546,8 +546,8 @@ static void assert_can_disable_dc6(struct drm_i915_private *dev_priv)
 		return;
 
 	assert_csr_loaded(dev_priv);
-	WARN(!(I915_READ(DC_STATE_EN) & DC_STATE_EN_UPTO_DC6),
-		"DC6 already programmed to be disabled.\n");
+	WARN_ONCE(!(I915_READ(DC_STATE_EN) & DC_STATE_EN_UPTO_DC6),
+		  "DC6 already programmed to be disabled.\n");
 }
 
 static void skl_enable_dc6(struct drm_i915_private *dev_priv)
@@ -670,7 +670,7 @@ static void skl_set_power_well(struct drm_i915_private *dev_priv,
 				wait_for((state = intel_csr_load_status_get(dev_priv)) !=
 						FW_UNINITIALIZED, 1000);
 				if (state != FW_LOADED)
-					DRM_ERROR("CSR firmware not ready (%d)\n",
+					DRM_DEBUG("CSR firmware not ready (%d)\n",
 							state);
 				else
 					if (SKL_ENABLE_DC6(dev))
-- 
cgit v0.10.2


From 87bcdd2e275d75e374434a4ad7da66bbb6da1d17 Mon Sep 17 00:00:00 2001
From: Jesse Barnes <jbarnes@virtuousgeek.org>
Date: Thu, 10 Sep 2015 14:55:00 -0700
Subject: drm/i915: don't try to load GuC fw on pre-gen9

This avoids some bad register writes and generally feels more correct
than unconditionally trying to redirect interrupts and such.

References: https://bugs.freedesktop.org/show_bug.cgi?id=91777
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 760cfba..acb1f23 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -4687,19 +4687,22 @@ i915_gem_init_hw(struct drm_device *dev)
 	}
 
 	/* We can't enable contexts until all firmware is loaded */
-	ret = intel_guc_ucode_load(dev);
-	if (ret) {
-		/*
-		 * If we got an error and GuC submission is enabled, map
-		 * the error to -EIO so the GPU will be declared wedged.
-		 * OTOH, if we didn't intend to use the GuC anyway, just
-		 * discard the error and carry on.
-		 */
-		DRM_ERROR("Failed to initialize GuC, error %d%s\n", ret,
-			i915.enable_guc_submission ? "" : " (ignored)");
-		ret = i915.enable_guc_submission ? -EIO : 0;
-		if (ret)
-			goto out;
+	if (HAS_GUC_UCODE(dev)) {
+		ret = intel_guc_ucode_load(dev);
+		if (ret) {
+			/*
+			 * If we got an error and GuC submission is enabled, map
+			 * the error to -EIO so the GPU will be declared wedged.
+			 * OTOH, if we didn't intend to use the GuC anyway, just
+			 * discard the error and carry on.
+			 */
+			DRM_ERROR("Failed to initialize GuC, error %d%s\n", ret,
+				  i915.enable_guc_submission ? "" :
+				  " (ignored)");
+			ret = i915.enable_guc_submission ? -EIO : 0;
+			if (ret)
+				goto out;
+		}
 	}
 
 	/* Now it is safe to go back round and do everything else: */
-- 
cgit v0.10.2


From e84fe80337dc85cca07d0417ea97edbec4789d8b Mon Sep 17 00:00:00 2001
From: Nick Hoath <nicholas.hoath@intel.com>
Date: Fri, 11 Sep 2015 12:53:46 +0100
Subject: drm/i915: Split alloc from init for lrc

Extend init/init_hw split to context init.
   - Move context initialisation in to i915_gem_init_hw
   - Move one off initialisation for render ring to
        i915_gem_validate_context
   - Move default context initialisation to logical_ring_init

Rename intel_lr_context_deferred_create to
intel_lr_context_deferred_alloc, to reflect reduced functionality &
alloc/init split.

This patch is intended to split out the allocation of resources &
initialisation to allow easier reuse of code for resume/gpu reset.

v2: Removed function ptr wrapping of do_switch_context (Daniel Vetter)
    Left ->init_context int intel_lr_context_deferred_alloc
    (Daniel Vetter)
    Remove unnecessary init flag & ring type test. (Daniel Vetter)
    Improve commit message (Daniel Vetter)
v3: On init/reinit, set the hw next sequence number to the sw next
    sequence number. This is set to 1 at driver load time. This prevents
    the seqno being reset on reinit (Chris Wilson)
v4: Set seqno back to ~0 - 0x1000 at start-of-day, and increment by 0x100
    on reset.
    This makes it obvious which bbs are which after a reset. (David Gordon
    & John Harrison)
    Rebase.
v5: Rebase. Fixed rebase breakage. Put context pinning in separate
    function. Removed code churn. (Thomas Daniel)
v6: Cleanup up issues introduced in v2 & v5 (Thomas Daniel)

Issue: VIZ-4798
Signed-off-by: Nick Hoath <nicholas.hoath@intel.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: John Harrison <john.c.harrison@intel.com>
Cc: David Gordon <david.s.gordon@intel.com>
Cc: Thomas Daniel <thomas.daniel@intel.com>
Reviewed-by: Thomas Daniel <thomas.daniel@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 2994ce5..7ddf17b 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -890,7 +890,6 @@ struct intel_context {
 	} legacy_hw_ctx;
 
 	/* Execlists */
-	bool rcs_initialized;
 	struct {
 		struct drm_i915_gem_object *state;
 		struct intel_ringbuffer *ringbuf;
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index acb1f23..cb0df7e 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -4609,14 +4609,8 @@ int i915_gem_init_rings(struct drm_device *dev)
 			goto cleanup_vebox_ring;
 	}
 
-	ret = i915_gem_set_seqno(dev, ((u32)~0 - 0x1000));
-	if (ret)
-		goto cleanup_bsd2_ring;
-
 	return 0;
 
-cleanup_bsd2_ring:
-	intel_cleanup_ring_buffer(&dev_priv->ring[VCS2]);
 cleanup_vebox_ring:
 	intel_cleanup_ring_buffer(&dev_priv->ring[VECS]);
 cleanup_blt_ring:
@@ -4705,6 +4699,14 @@ i915_gem_init_hw(struct drm_device *dev)
 		}
 	}
 
+	/*
+	 * Increment the next seqno by 0x100 so we have a visible break
+	 * on re-initialisation
+	 */
+	ret = i915_gem_set_seqno(dev, dev_priv->next_seqno+0x100);
+	if (ret)
+		goto out;
+
 	/* Now it is safe to go back round and do everything else: */
 	for_each_ring(ring, dev_priv, i) {
 		struct drm_i915_gem_request *req;
@@ -4906,6 +4908,14 @@ i915_gem_load(struct drm_device *dev)
 		dev_priv->num_fence_regs =
 				I915_READ(vgtif_reg(avail_rs.fence_num));
 
+	/*
+	 * Set initial sequence number for requests.
+	 * Using this number allows the wraparound to happen early,
+	 * catching any obvious problems.
+	 */
+	dev_priv->next_seqno = ((u32)~0 - 0x1100);
+	dev_priv->last_seqno = ((u32)~0 - 0x1101);
+
 	/* Initialize fence registers to zero */
 	INIT_LIST_HEAD(&dev_priv->mm.fence_list);
 	i915_gem_restore_fences(dev);
diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index 923a3c4..98952b4 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -1009,7 +1009,7 @@ i915_gem_validate_context(struct drm_device *dev, struct drm_file *file,
 	}
 
 	if (i915.enable_execlists && !ctx->engine[ring->id].state) {
-		int ret = intel_lr_context_deferred_create(ctx, ring);
+		int ret = intel_lr_context_deferred_alloc(ctx, ring);
 		if (ret) {
 			DRM_DEBUG("Could not create LRC %u: %d\n", ctx_id, ret);
 			return ERR_PTR(ret);
diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
index c3fca4b..0a53c96 100644
--- a/drivers/gpu/drm/i915/intel_lrc.c
+++ b/drivers/gpu/drm/i915/intel_lrc.c
@@ -221,6 +221,9 @@ enum {
 #define CTX_RCS_INDIRECT_CTX_OFFSET_DEFAULT  0x17
 
 static int intel_lr_context_pin(struct drm_i915_gem_request *rq);
+static void lrc_setup_hardware_status_page(struct intel_engine_cs *ring,
+		struct drm_i915_gem_object *default_ctx_obj);
+
 
 /**
  * intel_sanitize_enable_execlists() - sanitize i915.enable_execlists
@@ -1020,39 +1023,54 @@ int logical_ring_flush_all_caches(struct drm_i915_gem_request *req)
 	return 0;
 }
 
-static int intel_lr_context_pin(struct drm_i915_gem_request *rq)
+static int intel_lr_context_do_pin(struct intel_engine_cs *ring,
+		struct drm_i915_gem_object *ctx_obj,
+		struct intel_ringbuffer *ringbuf)
 {
-	struct drm_i915_private *dev_priv = rq->i915;
-	struct intel_engine_cs *ring = rq->ring;
-	struct drm_i915_gem_object *ctx_obj = rq->ctx->engine[ring->id].state;
-	struct intel_ringbuffer *ringbuf = rq->ringbuf;
+	struct drm_device *dev = ring->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
 	int ret = 0;
 
 	WARN_ON(!mutex_is_locked(&ring->dev->struct_mutex));
-	if (rq->ctx->engine[ring->id].pin_count++ == 0) {
-		ret = i915_gem_obj_ggtt_pin(ctx_obj, GEN8_LR_CONTEXT_ALIGN,
-				PIN_OFFSET_BIAS | GUC_WOPCM_TOP);
-		if (ret)
-			goto reset_pin_count;
+	ret = i915_gem_obj_ggtt_pin(ctx_obj, GEN8_LR_CONTEXT_ALIGN,
+			PIN_OFFSET_BIAS | GUC_WOPCM_TOP);
+	if (ret)
+		return ret;
 
-		ret = intel_pin_and_map_ringbuffer_obj(ring->dev, ringbuf);
-		if (ret)
-			goto unpin_ctx_obj;
+	ret = intel_pin_and_map_ringbuffer_obj(ring->dev, ringbuf);
+	if (ret)
+		goto unpin_ctx_obj;
 
-		ctx_obj->dirty = true;
+	ctx_obj->dirty = true;
 
-		/* Invalidate GuC TLB. */
-		if (i915.enable_guc_submission)
-			I915_WRITE(GEN8_GTCR, GEN8_GTCR_INVALIDATE);
-	}
+	/* Invalidate GuC TLB. */
+	if (i915.enable_guc_submission)
+		I915_WRITE(GEN8_GTCR, GEN8_GTCR_INVALIDATE);
 
 	return ret;
 
 unpin_ctx_obj:
 	i915_gem_object_ggtt_unpin(ctx_obj);
+
+	return ret;
+}
+
+static int intel_lr_context_pin(struct drm_i915_gem_request *rq)
+{
+	int ret = 0;
+	struct intel_engine_cs *ring = rq->ring;
+	struct drm_i915_gem_object *ctx_obj = rq->ctx->engine[ring->id].state;
+	struct intel_ringbuffer *ringbuf = rq->ringbuf;
+
+	if (rq->ctx->engine[ring->id].pin_count++ == 0) {
+		ret = intel_lr_context_do_pin(ring, ctx_obj, ringbuf);
+		if (ret)
+			goto reset_pin_count;
+	}
+	return ret;
+
 reset_pin_count:
 	rq->ctx->engine[ring->id].pin_count = 0;
-
 	return ret;
 }
 
@@ -1462,6 +1480,9 @@ static int gen8_init_common_ring(struct intel_engine_cs *ring)
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
+	lrc_setup_hardware_status_page(ring,
+				ring->default_context->engine[ring->id].state);
+
 	I915_WRITE_IMR(ring, ~(ring->irq_enable_mask | ring->irq_keep_mask));
 	I915_WRITE(RING_HWSTAM(ring->mmio_base), 0xffffffff);
 
@@ -1901,7 +1922,21 @@ static int logical_ring_init(struct drm_device *dev, struct intel_engine_cs *rin
 	if (ret)
 		return ret;
 
-	ret = intel_lr_context_deferred_create(ring->default_context, ring);
+	ret = intel_lr_context_deferred_alloc(ring->default_context, ring);
+	if (ret)
+		return ret;
+
+	/* As this is the default context, always pin it */
+	ret = intel_lr_context_do_pin(
+			ring,
+			ring->default_context->engine[ring->id].state,
+			ring->default_context->engine[ring->id].ringbuf);
+	if (ret) {
+		DRM_ERROR(
+			"Failed to pin and map ringbuffer %s: %d\n",
+			ring->name, ret);
+		return ret;
+	}
 
 	return ret;
 }
@@ -2124,14 +2159,8 @@ int intel_logical_rings_init(struct drm_device *dev)
 			goto cleanup_vebox_ring;
 	}
 
-	ret = i915_gem_set_seqno(dev, ((u32)~0 - 0x1000));
-	if (ret)
-		goto cleanup_bsd2_ring;
-
 	return 0;
 
-cleanup_bsd2_ring:
-	intel_logical_ring_cleanup(&dev_priv->ring[VCS2]);
 cleanup_vebox_ring:
 	intel_logical_ring_cleanup(&dev_priv->ring[VECS]);
 cleanup_blt_ring:
@@ -2401,7 +2430,7 @@ static void lrc_setup_hardware_status_page(struct intel_engine_cs *ring,
 }
 
 /**
- * intel_lr_context_deferred_create() - create the LRC specific bits of a context
+ * intel_lr_context_deferred_alloc() - create the LRC specific bits of a context
  * @ctx: LR context to create.
  * @ring: engine to be used with the context.
  *
@@ -2413,12 +2442,11 @@ static void lrc_setup_hardware_status_page(struct intel_engine_cs *ring,
  *
  * Return: non-zero on error.
  */
-int intel_lr_context_deferred_create(struct intel_context *ctx,
+
+int intel_lr_context_deferred_alloc(struct intel_context *ctx,
 				     struct intel_engine_cs *ring)
 {
-	const bool is_global_default_ctx = (ctx == ring->default_context);
 	struct drm_device *dev = ring->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_gem_object *ctx_obj;
 	uint32_t context_size;
 	struct intel_ringbuffer *ringbuf;
@@ -2438,82 +2466,50 @@ int intel_lr_context_deferred_create(struct intel_context *ctx,
 		return -ENOMEM;
 	}
 
-	if (is_global_default_ctx) {
-		ret = i915_gem_obj_ggtt_pin(ctx_obj, GEN8_LR_CONTEXT_ALIGN,
-				PIN_OFFSET_BIAS | GUC_WOPCM_TOP);
-		if (ret) {
-			DRM_DEBUG_DRIVER("Pin LRC backing obj failed: %d\n",
-					ret);
-			drm_gem_object_unreference(&ctx_obj->base);
-			return ret;
-		}
-
-		/* Invalidate GuC TLB. */
-		if (i915.enable_guc_submission)
-			I915_WRITE(GEN8_GTCR, GEN8_GTCR_INVALIDATE);
-	}
-
 	ringbuf = intel_engine_create_ringbuffer(ring, 4 * PAGE_SIZE);
 	if (IS_ERR(ringbuf)) {
 		ret = PTR_ERR(ringbuf);
-		goto error_unpin_ctx;
-	}
-
-	if (is_global_default_ctx) {
-		ret = intel_pin_and_map_ringbuffer_obj(dev, ringbuf);
-		if (ret) {
-			DRM_ERROR(
-				  "Failed to pin and map ringbuffer %s: %d\n",
-				  ring->name, ret);
-			goto error_ringbuf;
-		}
+		goto error_deref_obj;
 	}
 
 	ret = populate_lr_context(ctx, ctx_obj, ring, ringbuf);
 	if (ret) {
 		DRM_DEBUG_DRIVER("Failed to populate LRC: %d\n", ret);
-		goto error;
+		goto error_ringbuf;
 	}
 
 	ctx->engine[ring->id].ringbuf = ringbuf;
 	ctx->engine[ring->id].state = ctx_obj;
 
-	if (ctx == ring->default_context)
-		lrc_setup_hardware_status_page(ring, ctx_obj);
-	else if (ring->id == RCS && !ctx->rcs_initialized) {
-		if (ring->init_context) {
-			struct drm_i915_gem_request *req;
-
-			ret = i915_gem_request_alloc(ring, ctx, &req);
-			if (ret)
-				return ret;
+	if (ctx != ring->default_context && ring->init_context) {
+		struct drm_i915_gem_request *req;
 
-			ret = ring->init_context(req);
-			if (ret) {
-				DRM_ERROR("ring init context: %d\n", ret);
-				i915_gem_request_cancel(req);
-				ctx->engine[ring->id].ringbuf = NULL;
-				ctx->engine[ring->id].state = NULL;
-				goto error;
-			}
-
-			i915_add_request_no_flush(req);
+		ret = i915_gem_request_alloc(ring,
+			ctx, &req);
+		if (ret) {
+			DRM_ERROR("ring create req: %d\n",
+				ret);
+			i915_gem_request_cancel(req);
+			goto error_ringbuf;
 		}
 
-		ctx->rcs_initialized = true;
+		ret = ring->init_context(req);
+		if (ret) {
+			DRM_ERROR("ring init context: %d\n",
+				ret);
+			i915_gem_request_cancel(req);
+			goto error_ringbuf;
+		}
+		i915_add_request_no_flush(req);
 	}
-
 	return 0;
 
-error:
-	if (is_global_default_ctx)
-		intel_unpin_ringbuffer_obj(ringbuf);
 error_ringbuf:
 	intel_ringbuffer_free(ringbuf);
-error_unpin_ctx:
-	if (is_global_default_ctx)
-		i915_gem_object_ggtt_unpin(ctx_obj);
+error_deref_obj:
 	drm_gem_object_unreference(&ctx_obj->base);
+	ctx->engine[ring->id].ringbuf = NULL;
+	ctx->engine[ring->id].state = NULL;
 	return ret;
 }
 
diff --git a/drivers/gpu/drm/i915/intel_lrc.h b/drivers/gpu/drm/i915/intel_lrc.h
index 4cc54b3..69d99f0 100644
--- a/drivers/gpu/drm/i915/intel_lrc.h
+++ b/drivers/gpu/drm/i915/intel_lrc.h
@@ -75,8 +75,8 @@ static inline void intel_logical_ring_emit(struct intel_ringbuffer *ringbuf,
 #define LRC_STATE_PN	(LRC_PPHWSP_PN + 1)
 
 void intel_lr_context_free(struct intel_context *ctx);
-int intel_lr_context_deferred_create(struct intel_context *ctx,
-				     struct intel_engine_cs *ring);
+int intel_lr_context_deferred_alloc(struct intel_context *ctx,
+				    struct intel_engine_cs *ring);
 void intel_lr_context_unpin(struct drm_i915_gem_request *req);
 void intel_lr_context_reset(struct drm_device *dev,
 			struct intel_context *ctx);
-- 
cgit v0.10.2


From 374887bae861be2e691381f704b3d6068a54457c Mon Sep 17 00:00:00 2001
From: Masanari Iida <standby24x7@gmail.com>
Date: Sun, 13 Sep 2015 21:08:31 +0900
Subject: drm/i915: Fix warnings while make xmldocs caused by intel_lrc.c

This patch fix following warnings while "make xmldocs".
.//drivers/gpu/drm/i915/intel_lrc.c:780: warning: No description
found for parameter 'req'
.//drivers/gpu/drm/i915/intel_lrc.c:780: warning: Excess function
 parameter 'request' description in 'intel_logical_ring_begin'
.//drivers/gpu/drm/i915/intel_lrc.c:780: warning: Excess function
 parameter 'ctx' description in 'intel_logical_ring_begin'

Signed-off-by: Masanari Iida <standby24x7@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
index 0a53c96..fe06accb0 100644
--- a/drivers/gpu/drm/i915/intel_lrc.c
+++ b/drivers/gpu/drm/i915/intel_lrc.c
@@ -802,8 +802,7 @@ static int logical_ring_prepare(struct drm_i915_gem_request *req, int bytes)
 /**
  * intel_logical_ring_begin() - prepare the logical ringbuffer to accept some commands
  *
- * @request: The request to start some new work for
- * @ctx: Logical ring context whose ringbuffer is being prepared.
+ * @req: The request to start some new work for
  * @num_dwords: number of DWORDs that we plan to write to the ringbuffer.
  *
  * The ringbuffer might not be ready to accept the commands right away (maybe it needs to
-- 
cgit v0.10.2


From d7884d69a524b92d9770bcdc03df3c5a6120c2d0 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 11 Sep 2015 21:14:29 +0300
Subject: drm/i915: Set stolen reserved to 0 for pre-g4x platforms
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This stolen reserved stuff was introduced on g4x, so no need to waste
stolen on older platforms. Unfortunately configdb is no more so I can't
look up the right way to detect this stuff. I do have one hint as to
where the register might be on ctg, but I don't have a ctg to test it,
and on the elk I have here it doesn't contain sensible looking data.
For ilk grits suggegsts it might be in the same place as on snb (the
original PCI reg, not the mirror) but I can't be entirely sure about it
The register shows a round zero on my ilk.

So when there's no really good data for any of these platforms leave the
current "assume 1MiB" approach in place.

Cc: Paulo Zanoni <paulo.r.zanoni@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Acked-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_stolen.c b/drivers/gpu/drm/i915/i915_gem_stolen.c
index f361c4a..bf26ecc 100644
--- a/drivers/gpu/drm/i915/i915_gem_stolen.c
+++ b/drivers/gpu/drm/i915/i915_gem_stolen.c
@@ -281,7 +281,7 @@ static void bdw_get_stolen_reserved(struct drm_i915_private *dev_priv,
 int i915_gem_init_stolen(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	unsigned long reserved_total, reserved_base, reserved_size;
+	unsigned long reserved_total, reserved_base = 0, reserved_size;
 	unsigned long stolen_top;
 
 	mutex_init(&dev_priv->mm.stolen_lock);
@@ -306,6 +306,9 @@ int i915_gem_init_stolen(struct drm_device *dev)
 	case 2:
 	case 3:
 	case 4:
+		if (!IS_G4X(dev))
+			break;
+		/* fall through */
 	case 5:
 		/* Assume the gen6 maximum for the older platforms. */
 		reserved_size = 1024 * 1024;
-- 
cgit v0.10.2


From 6b6d5626750d72a22180a6e094cf95acd1d85c9b Mon Sep 17 00:00:00 2001
From: Robert Beckett <robert.beckett@intel.com>
Date: Tue, 8 Sep 2015 10:31:52 +0100
Subject: drm/i915/gen9: WA ST Unit Power Optimization Disable

WaDisableSTUnitPowerOptimization:skl,bxt

Signed-off-by: Robert Beckett <robert.beckett@intel.com>
Signed-off-by: Arun Siluvery <arun.siluvery@linux.intel.com>
Reviewed-by: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 495ac17..dd3d235 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -6943,6 +6943,9 @@ enum skl_disp_power_wells {
 #define HSW_ROW_CHICKEN3		0xe49c
 #define  HSW_ROW_CHICKEN3_L3_GLOBAL_ATOMICS_DISABLE    (1 << 6)
 
+#define HALF_SLICE_CHICKEN2		0xe180
+#define   GEN8_ST_PO_DISABLE		(1<<13)
+
 #define HALF_SLICE_CHICKEN3		0xe184
 #define   HSW_SAMPLE_C_PERFORMANCE	(1<<9)
 #define   GEN8_CENTROID_PIXEL_OPT_DIS	(1<<8)
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index 684e069..16a4ead 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -990,6 +990,9 @@ static int gen9_init_workarounds(struct intel_engine_cs *ring)
 				  GEN8_SAMPLER_POWER_BYPASS_DIS);
 	}
 
+	/* WaDisableSTUnitPowerOptimization:skl,bxt */
+	WA_SET_BIT_MASKED(HALF_SLICE_CHICKEN2, GEN8_ST_PO_DISABLE);
+
 	return 0;
 }
 
-- 
cgit v0.10.2


From d1b9d039de79a1cdd789d1547251d5ba0fad573b Mon Sep 17 00:00:00 2001
From: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Date: Mon, 14 Sep 2015 21:35:42 +0530
Subject: drm/i915: Fix fb object's frontbuffer-bits

Shared frontbuffer bits are causing warnings when same FB is displayed
in another plane without clearing the bits from previous plane.

v2: Removing coversion of fb bits to 64 bit as it is not needed for now. (Daniel)

Change-Id: Ic2df80747f314b82afd22f8326297c57d1e652c6
Signed-off-by: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Signed-off-by: Kumar, Mahesh <mahesh1.kumar@intel.com>
[danvet: Drop INTEL_FRONTBUFFER_SPRITE_MASK since unused.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 7ddf17b..677faa8 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -2011,25 +2011,26 @@ struct drm_i915_gem_object_ops {
 
 /*
  * Frontbuffer tracking bits. Set in obj->frontbuffer_bits while a gem bo is
- * considered to be the frontbuffer for the given plane interface-vise. This
+ * considered to be the frontbuffer for the given plane interface-wise. This
  * doesn't mean that the hw necessarily already scans it out, but that any
  * rendering (by the cpu or gpu) will land in the frontbuffer eventually.
  *
  * We have one bit per pipe and per scanout plane type.
  */
-#define INTEL_FRONTBUFFER_BITS_PER_PIPE 4
+#define INTEL_MAX_SPRITE_BITS_PER_PIPE 5
+#define INTEL_FRONTBUFFER_BITS_PER_PIPE 8
 #define INTEL_FRONTBUFFER_BITS \
 	(INTEL_FRONTBUFFER_BITS_PER_PIPE * I915_MAX_PIPES)
 #define INTEL_FRONTBUFFER_PRIMARY(pipe) \
 	(1 << (INTEL_FRONTBUFFER_BITS_PER_PIPE * (pipe)))
 #define INTEL_FRONTBUFFER_CURSOR(pipe) \
-	(1 << (1 +(INTEL_FRONTBUFFER_BITS_PER_PIPE * (pipe))))
-#define INTEL_FRONTBUFFER_SPRITE(pipe) \
-	(1 << (2 +(INTEL_FRONTBUFFER_BITS_PER_PIPE * (pipe))))
+	(1 << (1 + (INTEL_FRONTBUFFER_BITS_PER_PIPE * (pipe))))
+#define INTEL_FRONTBUFFER_SPRITE(pipe, plane) \
+	(1 << (2 + plane + (INTEL_FRONTBUFFER_BITS_PER_PIPE * (pipe))))
 #define INTEL_FRONTBUFFER_OVERLAY(pipe) \
-	(1 << (3 +(INTEL_FRONTBUFFER_BITS_PER_PIPE * (pipe))))
+	(1 << (2 + INTEL_MAX_SPRITE_BITS_PER_PIPE + (INTEL_FRONTBUFFER_BITS_PER_PIPE * (pipe))))
 #define INTEL_FRONTBUFFER_ALL_MASK(pipe) \
-	(0xf << (INTEL_FRONTBUFFER_BITS_PER_PIPE * (pipe)))
+	(0xff << (INTEL_FRONTBUFFER_BITS_PER_PIPE * (pipe)))
 
 struct drm_i915_gem_object {
 	struct drm_gem_object base;
diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c
index 9553859..4d27243 100644
--- a/drivers/gpu/drm/i915/intel_sprite.c
+++ b/drivers/gpu/drm/i915/intel_sprite.c
@@ -1123,7 +1123,7 @@ intel_plane_init(struct drm_device *dev, enum pipe pipe, int plane)
 
 	intel_plane->pipe = pipe;
 	intel_plane->plane = plane;
-	intel_plane->frontbuffer_bit = INTEL_FRONTBUFFER_SPRITE(pipe);
+	intel_plane->frontbuffer_bit = INTEL_FRONTBUFFER_SPRITE(pipe, plane);
 	intel_plane->check_plane = intel_check_sprite_plane;
 	intel_plane->commit_plane = intel_commit_sprite_plane;
 	possible_crtcs = (1 << pipe);
-- 
cgit v0.10.2


From 3f3d0d00734de56a0c5996f4e4433046c745592e Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Sat, 6 Jun 2015 21:41:09 +0100
Subject: drm/i2c: tda998x: remove useless NULL checks

There is no way 'priv' can be NULL in tda998x_irq_thread() - this can
only happen if request_threaded_irq() was passed a NULL priv pointer,
and we would have crashed long before then if that was the case.

We also always ensure that priv->encoder is correctly setup, which
must have been initialised prior to the interrupt being claimed, so we
can remove this check as well.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/i2c/tda998x_drv.c b/drivers/gpu/drm/i2c/tda998x_drv.c
index 424228b..d8e9708 100644
--- a/drivers/gpu/drm/i2c/tda998x_drv.c
+++ b/drivers/gpu/drm/i2c/tda998x_drv.c
@@ -558,7 +558,7 @@ static void tda998x_hpd(struct work_struct *work)
 	struct tda998x_priv *priv =
 			container_of(dwork, struct tda998x_priv, dwork);
 
-	if (priv->encoder && priv->encoder->dev)
+	if (priv->encoder->dev)
 		drm_kms_helper_hotplug_event(priv->encoder->dev);
 }
 
@@ -570,8 +570,6 @@ static irqreturn_t tda998x_irq_thread(int irq, void *data)
 	struct tda998x_priv *priv = data;
 	u8 sta, cec, lvl, flag0, flag1, flag2;
 
-	if (!priv)
-		return IRQ_HANDLED;
 	sta = cec_read(priv, REG_CEC_INTSTATUS);
 	cec = cec_read(priv, REG_CEC_RXSHPDINT);
 	lvl = cec_read(priv, REG_CEC_RXSHPDLEV);
-- 
cgit v0.10.2


From f84a97d4804a09240372dc7b195f9d6162152228 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Sat, 6 Jun 2015 21:41:09 +0100
Subject: drm/i2c: tda998x: report whether we actually handled the IRQ

Rather than always reporting that the interrupt was handled, we should
report whether we did handle the interrupt.  Arrange to report IRQ_NONE
for cases where we found nothing to do.

This allows us to (eventually) recover from stuck-IRQ problems, rather
than causing the kernel to solidly lock up.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/i2c/tda998x_drv.c b/drivers/gpu/drm/i2c/tda998x_drv.c
index d8e9708..ad3ce34 100644
--- a/drivers/gpu/drm/i2c/tda998x_drv.c
+++ b/drivers/gpu/drm/i2c/tda998x_drv.c
@@ -569,6 +569,7 @@ static irqreturn_t tda998x_irq_thread(int irq, void *data)
 {
 	struct tda998x_priv *priv = data;
 	u8 sta, cec, lvl, flag0, flag1, flag2;
+	bool handled = false;
 
 	sta = cec_read(priv, REG_CEC_INTSTATUS);
 	cec = cec_read(priv, REG_CEC_RXSHPDINT);
@@ -582,10 +583,12 @@ static irqreturn_t tda998x_irq_thread(int irq, void *data)
 	if ((flag2 & INT_FLAGS_2_EDID_BLK_RD) && priv->wq_edid_wait) {
 		priv->wq_edid_wait = 0;
 		wake_up(&priv->wq_edid);
+		handled = true;
 	} else if (cec != 0) {			/* HPD change */
 		schedule_delayed_work(&priv->dwork, HZ/10);
+		handled = true;
 	}
-	return IRQ_HANDLED;
+	return IRQ_RETVAL(handled);
 }
 
 static uint8_t tda998x_cksum(uint8_t *buf, size_t bytes)
-- 
cgit v0.10.2


From 0fc6f44d9683c61678da4b0eebc89e8fa624de39 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Sat, 6 Jun 2015 21:41:09 +0100
Subject: drm/i2c: tda998x: re-implement "Fix EDID read timeout on HDMI
 connect"

Commit 6833d26ef823 ("drm: tda998x: Fix EDID read timeout on HDMI
connect") used a weak scheme to try and delay reading EDID on a HDMI
connect event.  It is weak because delaying the notification of a
hotplug event does not stop userspace from trying to read the EDID
within the 100ms delay.

The solution provided here solves this issue:
* When a HDMI connection event is detected, mark a blocking flag for
  EDID reads, and start a timer for the delay.
* If an EDID read is attempted, and the blocking flag is set, wait
  for the blocking flag to clear.
* When the timer expires, clear the blocking flag and wake any thread
  waiting for the EDID read.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/i2c/tda998x_drv.c b/drivers/gpu/drm/i2c/tda998x_drv.c
index ad3ce34..a53696f 100644
--- a/drivers/gpu/drm/i2c/tda998x_drv.c
+++ b/drivers/gpu/drm/i2c/tda998x_drv.c
@@ -34,7 +34,6 @@ struct tda998x_priv {
 	struct i2c_client *cec;
 	struct i2c_client *hdmi;
 	struct mutex mutex;
-	struct delayed_work dwork;
 	uint16_t rev;
 	uint8_t current_page;
 	int dpms;
@@ -47,6 +46,11 @@ struct tda998x_priv {
 	wait_queue_head_t wq_edid;
 	volatile int wq_edid_wait;
 	struct drm_encoder *encoder;
+
+	struct work_struct detect_work;
+	struct timer_list edid_delay_timer;
+	wait_queue_head_t edid_delay_waitq;
+	bool edid_delay_active;
 };
 
 #define to_tda998x_priv(x)  ((struct tda998x_priv *)to_encoder_slave(x)->slave_priv)
@@ -551,15 +555,50 @@ tda998x_reset(struct tda998x_priv *priv)
 	reg_write(priv, REG_MUX_VP_VIP_OUT, 0x24);
 }
 
-/* handle HDMI connect/disconnect */
-static void tda998x_hpd(struct work_struct *work)
+/*
+ * The TDA998x has a problem when trying to read the EDID close to a
+ * HPD assertion: it needs a delay of 100ms to avoid timing out while
+ * trying to read EDID data.
+ *
+ * However, tda998x_encoder_get_modes() may be called at any moment
+ * after tda998x_encoder_detect() indicates that we are connected, so
+ * we need to delay probing modes in tda998x_encoder_get_modes() after
+ * we have seen a HPD inactive->active transition.  This code implements
+ * that delay.
+ */
+static void tda998x_edid_delay_done(unsigned long data)
+{
+	struct tda998x_priv *priv = (struct tda998x_priv *)data;
+
+	priv->edid_delay_active = false;
+	wake_up(&priv->edid_delay_waitq);
+	schedule_work(&priv->detect_work);
+}
+
+static void tda998x_edid_delay_start(struct tda998x_priv *priv)
+{
+	priv->edid_delay_active = true;
+	mod_timer(&priv->edid_delay_timer, jiffies + HZ/10);
+}
+
+static int tda998x_edid_delay_wait(struct tda998x_priv *priv)
+{
+	return wait_event_killable(priv->edid_delay_waitq, !priv->edid_delay_active);
+}
+
+/*
+ * We need to run the KMS hotplug event helper outside of our threaded
+ * interrupt routine as this can call back into our get_modes method,
+ * which will want to make use of interrupts.
+ */
+static void tda998x_detect_work(struct work_struct *work)
 {
-	struct delayed_work *dwork = to_delayed_work(work);
 	struct tda998x_priv *priv =
-			container_of(dwork, struct tda998x_priv, dwork);
+		container_of(work, struct tda998x_priv, detect_work);
+	struct drm_device *dev = priv->encoder->dev;
 
-	if (priv->encoder->dev)
-		drm_kms_helper_hotplug_event(priv->encoder->dev);
+	if (dev)
+		drm_kms_helper_hotplug_event(dev);
 }
 
 /*
@@ -585,7 +624,11 @@ static irqreturn_t tda998x_irq_thread(int irq, void *data)
 		wake_up(&priv->wq_edid);
 		handled = true;
 	} else if (cec != 0) {			/* HPD change */
-		schedule_delayed_work(&priv->dwork, HZ/10);
+		if (lvl & CEC_RXSHPDLEV_HPD)
+			tda998x_edid_delay_start(priv);
+		else
+			schedule_work(&priv->detect_work);
+
 		handled = true;
 	}
 	return IRQ_RETVAL(handled);
@@ -1103,6 +1146,14 @@ tda998x_encoder_get_modes(struct tda998x_priv *priv,
 	struct edid *edid;
 	int n;
 
+	/*
+	 * If we get killed while waiting for the HPD timeout, return
+	 * no modes found: we are not in a restartable path, so we
+	 * can't handle signals gracefully.
+	 */
+	if (tda998x_edid_delay_wait(priv))
+		return 0;
+
 	if (priv->rev == TDA19988)
 		reg_clear(priv, REG_TX4, TX4_PD_RAM);
 
@@ -1149,10 +1200,12 @@ static void tda998x_destroy(struct tda998x_priv *priv)
 	/* disable all IRQs and free the IRQ handler */
 	cec_write(priv, REG_CEC_RXSHPDINTENA, 0);
 	reg_clear(priv, REG_INT_FLAGS_2, INT_FLAGS_2_EDID_BLK_RD);
-	if (priv->hdmi->irq) {
+
+	if (priv->hdmi->irq)
 		free_irq(priv->hdmi->irq, priv);
-		cancel_delayed_work_sync(&priv->dwork);
-	}
+
+	del_timer_sync(&priv->edid_delay_timer);
+	cancel_work_sync(&priv->detect_work);
 
 	i2c_unregister_device(priv->cec);
 }
@@ -1253,6 +1306,10 @@ static int tda998x_create(struct i2c_client *client, struct tda998x_priv *priv)
 	priv->dpms = DRM_MODE_DPMS_OFF;
 
 	mutex_init(&priv->mutex);	/* protect the page access */
+	init_waitqueue_head(&priv->edid_delay_waitq);
+	setup_timer(&priv->edid_delay_timer, tda998x_edid_delay_done,
+		    (unsigned long)priv);
+	INIT_WORK(&priv->detect_work, tda998x_detect_work);
 
 	/* wake up the device: */
 	cec_write(priv, REG_CEC_ENAMODS,
@@ -1311,7 +1368,6 @@ static int tda998x_create(struct i2c_client *client, struct tda998x_priv *priv)
 
 		/* init read EDID waitqueue and HDP work */
 		init_waitqueue_head(&priv->wq_edid);
-		INIT_DELAYED_WORK(&priv->dwork, tda998x_hpd);
 
 		/* clear pending interrupts */
 		reg_read(priv, REG_INT_FLAGS_0);
-- 
cgit v0.10.2


From e66e03abf80f701da60ae085cbb913e67ce6741d Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Sat, 6 Jun 2015 21:41:10 +0100
Subject: drm/i2c: tda998x: convert to u8/u16/u32 types

C99 types are against the style of the Linux kernel.  Convert to using
Linus-friendly types.  See https://lwn.net/Articles/113367/ for more
information.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/i2c/tda998x_drv.c b/drivers/gpu/drm/i2c/tda998x_drv.c
index a53696f..6d6aaad 100644
--- a/drivers/gpu/drm/i2c/tda998x_drv.c
+++ b/drivers/gpu/drm/i2c/tda998x_drv.c
@@ -34,8 +34,8 @@ struct tda998x_priv {
 	struct i2c_client *cec;
 	struct i2c_client *hdmi;
 	struct mutex mutex;
-	uint16_t rev;
-	uint8_t current_page;
+	u16 rev;
+	u8 current_page;
 	int dpms;
 	bool is_hdmi_sink;
 	u8 vip_cntrl_0;
@@ -349,10 +349,10 @@ struct tda998x_priv {
 #define TDA19988                  0x0301
 
 static void
-cec_write(struct tda998x_priv *priv, uint16_t addr, uint8_t val)
+cec_write(struct tda998x_priv *priv, u16 addr, u8 val)
 {
 	struct i2c_client *client = priv->cec;
-	uint8_t buf[] = {addr, val};
+	u8 buf[] = {addr, val};
 	int ret;
 
 	ret = i2c_master_send(client, buf, sizeof(buf));
@@ -360,11 +360,11 @@ cec_write(struct tda998x_priv *priv, uint16_t addr, uint8_t val)
 		dev_err(&client->dev, "Error %d writing to cec:0x%x\n", ret, addr);
 }
 
-static uint8_t
-cec_read(struct tda998x_priv *priv, uint8_t addr)
+static u8
+cec_read(struct tda998x_priv *priv, u8 addr)
 {
 	struct i2c_client *client = priv->cec;
-	uint8_t val;
+	u8 val;
 	int ret;
 
 	ret = i2c_master_send(client, &addr, sizeof(addr));
@@ -383,11 +383,11 @@ fail:
 }
 
 static int
-set_page(struct tda998x_priv *priv, uint16_t reg)
+set_page(struct tda998x_priv *priv, u16 reg)
 {
 	if (REG2PAGE(reg) != priv->current_page) {
 		struct i2c_client *client = priv->hdmi;
-		uint8_t buf[] = {
+		u8 buf[] = {
 				REG_CURPAGE, REG2PAGE(reg)
 		};
 		int ret = i2c_master_send(client, buf, sizeof(buf));
@@ -403,10 +403,10 @@ set_page(struct tda998x_priv *priv, uint16_t reg)
 }
 
 static int
-reg_read_range(struct tda998x_priv *priv, uint16_t reg, char *buf, int cnt)
+reg_read_range(struct tda998x_priv *priv, u16 reg, char *buf, int cnt)
 {
 	struct i2c_client *client = priv->hdmi;
-	uint8_t addr = REG2ADDR(reg);
+	u8 addr = REG2ADDR(reg);
 	int ret;
 
 	mutex_lock(&priv->mutex);
@@ -432,10 +432,10 @@ out:
 }
 
 static void
-reg_write_range(struct tda998x_priv *priv, uint16_t reg, uint8_t *p, int cnt)
+reg_write_range(struct tda998x_priv *priv, u16 reg, u8 *p, int cnt)
 {
 	struct i2c_client *client = priv->hdmi;
-	uint8_t buf[cnt+1];
+	u8 buf[cnt+1];
 	int ret;
 
 	buf[0] = REG2ADDR(reg);
@@ -454,9 +454,9 @@ out:
 }
 
 static int
-reg_read(struct tda998x_priv *priv, uint16_t reg)
+reg_read(struct tda998x_priv *priv, u16 reg)
 {
-	uint8_t val = 0;
+	u8 val = 0;
 	int ret;
 
 	ret = reg_read_range(priv, reg, &val, sizeof(val));
@@ -466,10 +466,10 @@ reg_read(struct tda998x_priv *priv, uint16_t reg)
 }
 
 static void
-reg_write(struct tda998x_priv *priv, uint16_t reg, uint8_t val)
+reg_write(struct tda998x_priv *priv, u16 reg, u8 val)
 {
 	struct i2c_client *client = priv->hdmi;
-	uint8_t buf[] = {REG2ADDR(reg), val};
+	u8 buf[] = {REG2ADDR(reg), val};
 	int ret;
 
 	mutex_lock(&priv->mutex);
@@ -485,10 +485,10 @@ out:
 }
 
 static void
-reg_write16(struct tda998x_priv *priv, uint16_t reg, uint16_t val)
+reg_write16(struct tda998x_priv *priv, u16 reg, u16 val)
 {
 	struct i2c_client *client = priv->hdmi;
-	uint8_t buf[] = {REG2ADDR(reg), val >> 8, val};
+	u8 buf[] = {REG2ADDR(reg), val >> 8, val};
 	int ret;
 
 	mutex_lock(&priv->mutex);
@@ -504,7 +504,7 @@ out:
 }
 
 static void
-reg_set(struct tda998x_priv *priv, uint16_t reg, uint8_t val)
+reg_set(struct tda998x_priv *priv, u16 reg, u8 val)
 {
 	int old_val;
 
@@ -514,7 +514,7 @@ reg_set(struct tda998x_priv *priv, uint16_t reg, uint8_t val)
 }
 
 static void
-reg_clear(struct tda998x_priv *priv, uint16_t reg, uint8_t val)
+reg_clear(struct tda998x_priv *priv, u16 reg, u8 val)
 {
 	int old_val;
 
@@ -634,7 +634,7 @@ static irqreturn_t tda998x_irq_thread(int irq, void *data)
 	return IRQ_RETVAL(handled);
 }
 
-static uint8_t tda998x_cksum(uint8_t *buf, size_t bytes)
+static u8 tda998x_cksum(u8 *buf, size_t bytes)
 {
 	int sum = 0;
 
@@ -647,8 +647,8 @@ static uint8_t tda998x_cksum(uint8_t *buf, size_t bytes)
 #define PB(x) (HB(2) + 1 + (x))
 
 static void
-tda998x_write_if(struct tda998x_priv *priv, uint8_t bit, uint16_t addr,
-		 uint8_t *buf, size_t size)
+tda998x_write_if(struct tda998x_priv *priv, u8 bit, u16 addr,
+		 u8 *buf, size_t size)
 {
 	reg_clear(priv, REG_DIP_IF_FLAGS, bit);
 	reg_write_range(priv, addr, buf, size);
@@ -711,8 +711,8 @@ static void
 tda998x_configure_audio(struct tda998x_priv *priv,
 		struct drm_display_mode *mode, struct tda998x_encoder_params *p)
 {
-	uint8_t buf[6], clksel_aip, clksel_fs, cts_n, adiv;
-	uint32_t n;
+	u8 buf[6], clksel_aip, clksel_fs, cts_n, adiv;
+	u32 n;
 
 	/* Enable audio ports */
 	reg_write(priv, REG_ENA_AP, p->audio_cfg);
@@ -888,14 +888,14 @@ tda998x_encoder_mode_set(struct tda998x_priv *priv,
 			 struct drm_display_mode *mode,
 			 struct drm_display_mode *adjusted_mode)
 {
-	uint16_t ref_pix, ref_line, n_pix, n_line;
-	uint16_t hs_pix_s, hs_pix_e;
-	uint16_t vs1_pix_s, vs1_pix_e, vs1_line_s, vs1_line_e;
-	uint16_t vs2_pix_s, vs2_pix_e, vs2_line_s, vs2_line_e;
-	uint16_t vwin1_line_s, vwin1_line_e;
-	uint16_t vwin2_line_s, vwin2_line_e;
-	uint16_t de_pix_s, de_pix_e;
-	uint8_t reg, div, rep;
+	u16 ref_pix, ref_line, n_pix, n_line;
+	u16 hs_pix_s, hs_pix_e;
+	u16 vs1_pix_s, vs1_pix_e, vs1_line_s, vs1_line_e;
+	u16 vs2_pix_s, vs2_pix_e, vs2_line_s, vs2_line_e;
+	u16 vwin1_line_s, vwin1_line_e;
+	u16 vwin2_line_s, vwin2_line_e;
+	u16 de_pix_s, de_pix_e;
+	u8 reg, div, rep;
 
 	/*
 	 * Internally TDA998x is using ITU-R BT.656 style sync but
@@ -1077,7 +1077,7 @@ tda998x_encoder_mode_set(struct tda998x_priv *priv,
 static enum drm_connector_status
 tda998x_encoder_detect(struct tda998x_priv *priv)
 {
-	uint8_t val = cec_read(priv, REG_CEC_RXSHPDLEV);
+	u8 val = cec_read(priv, REG_CEC_RXSHPDLEV);
 
 	return (val & CEC_RXSHPDLEV_HPD) ? connector_status_connected :
 			connector_status_disconnected;
@@ -1086,7 +1086,7 @@ tda998x_encoder_detect(struct tda998x_priv *priv)
 static int read_edid_block(void *data, u8 *buf, unsigned int blk, size_t length)
 {
 	struct tda998x_priv *priv = data;
-	uint8_t offset, segptr;
+	u8 offset, segptr;
 	int ret, i;
 
 	offset = (blk & 1) ? 128 : 0;
@@ -1558,7 +1558,7 @@ static int tda998x_bind(struct device *dev, struct device *master, void *data)
 	struct i2c_client *client = to_i2c_client(dev);
 	struct drm_device *drm = data;
 	struct tda998x_priv2 *priv;
-	uint32_t crtcs = 0;
+	u32 crtcs = 0;
 	int ret;
 
 	priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL);
-- 
cgit v0.10.2


From ec5d3e83d332ac5dfde8e0a1f57393fc91d55030 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Sat, 6 Jun 2015 21:41:10 +0100
Subject: drm/i2c: tda998x: handle all outstanding interrupts

As reading the interrupt registers clears the outstanding interrupts, we
must process all received interrupts to avoid dropping any.  Rearrange
the code to achieve this, and properly check for a HPD interrupt from
the CEC_RXSHPDINT register.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/i2c/tda998x_drv.c b/drivers/gpu/drm/i2c/tda998x_drv.c
index 6d6aaad..1285fb3 100644
--- a/drivers/gpu/drm/i2c/tda998x_drv.c
+++ b/drivers/gpu/drm/i2c/tda998x_drv.c
@@ -330,6 +330,8 @@ struct tda998x_priv {
 # define CEC_FRO_IM_CLK_CTRL_FRO_DIV   (1 << 0)
 #define REG_CEC_RXSHPDINTENA	  0xfc		      /* read/write */
 #define REG_CEC_RXSHPDINT	  0xfd		      /* read */
+# define CEC_RXSHPDINT_RXSENS     BIT(0)
+# define CEC_RXSHPDINT_HPD        BIT(1)
 #define REG_CEC_RXSHPDLEV         0xfe                /* read */
 # define CEC_RXSHPDLEV_RXSENS     (1 << 0)
 # define CEC_RXSHPDLEV_HPD        (1 << 1)
@@ -619,11 +621,8 @@ static irqreturn_t tda998x_irq_thread(int irq, void *data)
 	DRM_DEBUG_DRIVER(
 		"tda irq sta %02x cec %02x lvl %02x f0 %02x f1 %02x f2 %02x\n",
 		sta, cec, lvl, flag0, flag1, flag2);
-	if ((flag2 & INT_FLAGS_2_EDID_BLK_RD) && priv->wq_edid_wait) {
-		priv->wq_edid_wait = 0;
-		wake_up(&priv->wq_edid);
-		handled = true;
-	} else if (cec != 0) {			/* HPD change */
+
+	if (cec & CEC_RXSHPDINT_HPD) {
 		if (lvl & CEC_RXSHPDLEV_HPD)
 			tda998x_edid_delay_start(priv);
 		else
@@ -631,6 +630,13 @@ static irqreturn_t tda998x_irq_thread(int irq, void *data)
 
 		handled = true;
 	}
+
+	if ((flag2 & INT_FLAGS_2_EDID_BLK_RD) && priv->wq_edid_wait) {
+		priv->wq_edid_wait = 0;
+		wake_up(&priv->wq_edid);
+		handled = true;
+	}
+
 	return IRQ_RETVAL(handled);
 }
 
-- 
cgit v0.10.2


From 96795df15c89ce30261a31289740b4621bcec0fe Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Thu, 6 Aug 2015 10:52:05 +0100
Subject: drm/i2c: tda998x: use more HDMI helpers

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/i2c/tda998x_drv.c b/drivers/gpu/drm/i2c/tda998x_drv.c
index 1285fb3..1c6fc24 100644
--- a/drivers/gpu/drm/i2c/tda998x_drv.c
+++ b/drivers/gpu/drm/i2c/tda998x_drv.c
@@ -640,66 +640,57 @@ static irqreturn_t tda998x_irq_thread(int irq, void *data)
 	return IRQ_RETVAL(handled);
 }
 
-static u8 tda998x_cksum(u8 *buf, size_t bytes)
-{
-	int sum = 0;
-
-	while (bytes--)
-		sum -= *buf++;
-	return sum;
-}
-
-#define HB(x) (x)
-#define PB(x) (HB(2) + 1 + (x))
-
 static void
 tda998x_write_if(struct tda998x_priv *priv, u8 bit, u16 addr,
-		 u8 *buf, size_t size)
+		 union hdmi_infoframe *frame)
 {
+	u8 buf[32];
+	ssize_t len;
+
+	len = hdmi_infoframe_pack(frame, buf, sizeof(buf));
+	if (len < 0) {
+		dev_err(&priv->hdmi->dev,
+			"hdmi_infoframe_pack() type=0x%02x failed: %zd\n",
+			frame->any.type, len);
+		return;
+	}
+
 	reg_clear(priv, REG_DIP_IF_FLAGS, bit);
-	reg_write_range(priv, addr, buf, size);
+	reg_write_range(priv, addr, buf, len);
 	reg_set(priv, REG_DIP_IF_FLAGS, bit);
 }
 
 static void
 tda998x_write_aif(struct tda998x_priv *priv, struct tda998x_encoder_params *p)
 {
-	u8 buf[PB(HDMI_AUDIO_INFOFRAME_SIZE) + 1];
+	union hdmi_infoframe frame;
+
+	hdmi_audio_infoframe_init(&frame.audio);
 
-	memset(buf, 0, sizeof(buf));
-	buf[HB(0)] = HDMI_INFOFRAME_TYPE_AUDIO;
-	buf[HB(1)] = 0x01;
-	buf[HB(2)] = HDMI_AUDIO_INFOFRAME_SIZE;
-	buf[PB(1)] = p->audio_frame[1] & 0x07; /* CC */
-	buf[PB(2)] = p->audio_frame[2] & 0x1c; /* SF */
-	buf[PB(4)] = p->audio_frame[4];
-	buf[PB(5)] = p->audio_frame[5] & 0xf8; /* DM_INH + LSV */
+	frame.audio.channels = p->audio_frame[1] & 0x07;
+	frame.audio.channel_allocation = p->audio_frame[4];
+	frame.audio.level_shift_value = (p->audio_frame[5] & 0x78) >> 3;
+	frame.audio.downmix_inhibit = (p->audio_frame[5] & 0x80) >> 7;
 
-	buf[PB(0)] = tda998x_cksum(buf, sizeof(buf));
+	/*
+	 * L-PCM and IEC61937 compressed audio shall always set sample
+	 * frequency to "refer to stream".  For others, see the HDMI
+	 * specification.
+	 */
+	frame.audio.sample_frequency = (p->audio_frame[2] & 0x1c) >> 2;
 
-	tda998x_write_if(priv, DIP_IF_FLAGS_IF4, REG_IF4_HB0, buf,
-			 sizeof(buf));
+	tda998x_write_if(priv, DIP_IF_FLAGS_IF4, REG_IF4_HB0, &frame);
 }
 
 static void
 tda998x_write_avi(struct tda998x_priv *priv, struct drm_display_mode *mode)
 {
-	struct hdmi_avi_infoframe frame;
-	u8 buf[HDMI_INFOFRAME_HEADER_SIZE + HDMI_AVI_INFOFRAME_SIZE];
-	ssize_t len;
-
-	drm_hdmi_avi_infoframe_from_display_mode(&frame, mode);
+	union hdmi_infoframe frame;
 
-	frame.quantization_range = HDMI_QUANTIZATION_RANGE_FULL;
-
-	len = hdmi_avi_infoframe_pack(&frame, buf, sizeof(buf));
-	if (len < 0) {
-		dev_err(&priv->hdmi->dev,
-			"hdmi_avi_infoframe_pack() failed: %zd\n", len);
-		return;
-	}
+	drm_hdmi_avi_infoframe_from_display_mode(&frame.avi, mode);
+	frame.avi.quantization_range = HDMI_QUANTIZATION_RANGE_FULL;
 
-	tda998x_write_if(priv, DIP_IF_FLAGS_IF2, REG_IF2_HB0, buf, len);
+	tda998x_write_if(priv, DIP_IF_FLAGS_IF2, REG_IF2_HB0, &frame);
 }
 
 static void tda998x_audio_mute(struct tda998x_priv *priv, bool on)
-- 
cgit v0.10.2


From 0fb2970b4b6bfc26817a731b8bc29a9bf9177c20 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Sat, 6 Jun 2015 21:46:53 +0100
Subject: drm/armada: remove non-component support

Now that the transition of TDA998x to the component helpers is complete,
remove the non-componentised support from the Armada DRM driver.  All
outputs are expected to use the component helpers from now on.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/armada/Kconfig b/drivers/gpu/drm/armada/Kconfig
index 50ae88a..eb773e9 100644
--- a/drivers/gpu/drm/armada/Kconfig
+++ b/drivers/gpu/drm/armada/Kconfig
@@ -14,12 +14,3 @@ config DRM_ARMADA
 	  This driver provides no built-in acceleration; acceleration is
 	  performed by other IP found on the SoC.  This driver provides
 	  kernel mode setting and buffer management to userspace.
-
-config DRM_ARMADA_TDA1998X
-	bool "Support TDA1998X HDMI output"
-	depends on DRM_ARMADA != n
-	depends on I2C && DRM_I2C_NXP_TDA998X = y
-	default y
-	help
-	  Support the TDA1998x HDMI output device found on the Solid-Run
-	  CuBox.
diff --git a/drivers/gpu/drm/armada/Makefile b/drivers/gpu/drm/armada/Makefile
index d6f43e0..ffd6736 100644
--- a/drivers/gpu/drm/armada/Makefile
+++ b/drivers/gpu/drm/armada/Makefile
@@ -1,6 +1,5 @@
 armada-y	:= armada_crtc.o armada_drv.o armada_fb.o armada_fbdev.o \
-		   armada_gem.o armada_output.o armada_overlay.o \
-		   armada_slave.o
+		   armada_gem.o armada_overlay.o
 armada-y	+= armada_510.o
 armada-$(CONFIG_DEBUG_FS) += armada_debugfs.o
 
diff --git a/drivers/gpu/drm/armada/armada_crtc.c b/drivers/gpu/drm/armada/armada_crtc.c
index 01ffe9b..c7374a3 100644
--- a/drivers/gpu/drm/armada/armada_crtc.c
+++ b/drivers/gpu/drm/armada/armada_crtc.c
@@ -1044,7 +1044,7 @@ static int armada_drm_crtc_create_properties(struct drm_device *dev)
 	return 0;
 }
 
-int armada_drm_crtc_create(struct drm_device *drm, struct device *dev,
+static int armada_drm_crtc_create(struct drm_device *drm, struct device *dev,
 	struct resource *res, int irq, const struct armada_variant *variant,
 	struct device_node *port)
 {
diff --git a/drivers/gpu/drm/armada/armada_crtc.h b/drivers/gpu/drm/armada/armada_crtc.h
index 98102a5..a13469f 100644
--- a/drivers/gpu/drm/armada/armada_crtc.h
+++ b/drivers/gpu/drm/armada/armada_crtc.h
@@ -75,10 +75,6 @@ struct armada_crtc {
 };
 #define drm_to_armada_crtc(c) container_of(c, struct armada_crtc, crtc)
 
-struct device_node;
-int armada_drm_crtc_create(struct drm_device *, struct device *,
-	struct resource *, int, const struct armada_variant *,
-	struct device_node *);
 void armada_drm_crtc_gamma_set(struct drm_crtc *, u16, u16, u16, int);
 void armada_drm_crtc_gamma_get(struct drm_crtc *, u16 *, u16 *, u16 *, int);
 void armada_drm_crtc_disable_irq(struct armada_crtc *, u32);
diff --git a/drivers/gpu/drm/armada/armada_drv.c b/drivers/gpu/drm/armada/armada_drv.c
index 225034b..b373cf9 100644
--- a/drivers/gpu/drm/armada/armada_drv.c
+++ b/drivers/gpu/drm/armada/armada_drv.c
@@ -18,47 +18,6 @@
 #include <drm/armada_drm.h>
 #include "armada_ioctlP.h"
 
-#ifdef CONFIG_DRM_ARMADA_TDA1998X
-#include <drm/i2c/tda998x.h>
-#include "armada_slave.h"
-
-static struct tda998x_encoder_params params = {
-	/* With 0x24, there is no translation between vp_out and int_vp
-	FB	LCD out	Pins	VIP	Int Vp
-	R:23:16	R:7:0	VPC7:0	7:0	7:0[R]
-	G:15:8	G:15:8	VPB7:0	23:16	23:16[G]
-	B:7:0	B:23:16	VPA7:0	15:8	15:8[B]
-	*/
-	.swap_a = 2,
-	.swap_b = 3,
-	.swap_c = 4,
-	.swap_d = 5,
-	.swap_e = 0,
-	.swap_f = 1,
-	.audio_cfg = BIT(2),
-	.audio_frame[1] = 1,
-	.audio_format = AFMT_SPDIF,
-	.audio_sample_rate = 44100,
-};
-
-static const struct armada_drm_slave_config tda19988_config = {
-	.i2c_adapter_id = 0,
-	.crtcs = 1 << 0, /* Only LCD0 at the moment */
-	.polled = DRM_CONNECTOR_POLL_CONNECT | DRM_CONNECTOR_POLL_DISCONNECT,
-	.interlace_allowed = true,
-	.info = {
-		.type = "tda998x",
-		.addr = 0x70,
-		.platform_data = &params,
-	},
-};
-#endif
-
-static bool is_componentized(struct device *dev)
-{
-	return dev->of_node || dev->platform_data;
-}
-
 static void armada_drm_unref_work(struct work_struct *work)
 {
 	struct armada_private *priv =
@@ -91,16 +50,11 @@ void armada_drm_queue_unref_work(struct drm_device *dev,
 
 static int armada_drm_load(struct drm_device *dev, unsigned long flags)
 {
-	const struct platform_device_id *id;
-	const struct armada_variant *variant;
 	struct armada_private *priv;
-	struct resource *res[ARRAY_SIZE(priv->dcrtc)];
 	struct resource *mem = NULL;
-	int ret, n, i;
+	int ret, n;
 
-	memset(res, 0, sizeof(res));
-
-	for (n = i = 0; ; n++) {
+	for (n = 0; ; n++) {
 		struct resource *r = platform_get_resource(dev->platformdev,
 							   IORESOURCE_MEM, n);
 		if (!r)
@@ -109,8 +63,6 @@ static int armada_drm_load(struct drm_device *dev, unsigned long flags)
 		/* Resources above 64K are graphics memory */
 		if (resource_size(r) > SZ_64K)
 			mem = r;
-		else if (i < ARRAY_SIZE(priv->dcrtc))
-			res[i++] = r;
 		else
 			return -EINVAL;
 	}
@@ -131,13 +83,6 @@ static int armada_drm_load(struct drm_device *dev, unsigned long flags)
 	platform_set_drvdata(dev->platformdev, dev);
 	dev->dev_private = priv;
 
-	/* Get the implementation specific driver data. */
-	id = platform_get_device_id(dev->platformdev);
-	if (!id)
-		return -ENXIO;
-
-	variant = (const struct armada_variant *)id->driver_data;
-
 	INIT_WORK(&priv->fb_unref_work, armada_drm_unref_work);
 	INIT_KFIFO(priv->fb_unref);
 
@@ -157,34 +102,9 @@ static int armada_drm_load(struct drm_device *dev, unsigned long flags)
 	dev->mode_config.funcs = &armada_drm_mode_config_funcs;
 	drm_mm_init(&priv->linear, mem->start, resource_size(mem));
 
-	/* Create all LCD controllers */
-	for (n = 0; n < ARRAY_SIZE(priv->dcrtc); n++) {
-		int irq;
-
-		if (!res[n])
-			break;
-
-		irq = platform_get_irq(dev->platformdev, n);
-		if (irq < 0)
-			goto err_kms;
-
-		ret = armada_drm_crtc_create(dev, dev->dev, res[n], irq,
-					     variant, NULL);
-		if (ret)
-			goto err_kms;
-	}
-
-	if (is_componentized(dev->dev)) {
-		ret = component_bind_all(dev->dev, dev);
-		if (ret)
-			goto err_kms;
-	} else {
-#ifdef CONFIG_DRM_ARMADA_TDA1998X
-		ret = armada_drm_connector_slave_create(dev, &tda19988_config);
-		if (ret)
-			goto err_kms;
-#endif
-	}
+	ret = component_bind_all(dev->dev, dev);
+	if (ret)
+		goto err_kms;
 
 	ret = drm_vblank_init(dev, dev->mode_config.num_crtc);
 	if (ret)
@@ -202,8 +122,7 @@ static int armada_drm_load(struct drm_device *dev, unsigned long flags)
 	return 0;
 
  err_comp:
-	if (is_componentized(dev->dev))
-		component_unbind_all(dev->dev, dev);
+	component_unbind_all(dev->dev, dev);
  err_kms:
 	drm_mode_config_cleanup(dev);
 	drm_mm_takedown(&priv->linear);
@@ -219,8 +138,7 @@ static int armada_drm_unload(struct drm_device *dev)
 	drm_kms_helper_poll_fini(dev);
 	armada_fbdev_fini(dev);
 
-	if (is_componentized(dev->dev))
-		component_unbind_all(dev->dev, dev);
+	component_unbind_all(dev->dev, dev);
 
 	drm_mode_config_cleanup(dev);
 	drm_mm_takedown(&priv->linear);
@@ -435,37 +353,28 @@ static const struct component_master_ops armada_master_ops = {
 
 static int armada_drm_probe(struct platform_device *pdev)
 {
-	if (is_componentized(&pdev->dev)) {
-		struct component_match *match = NULL;
-		int ret;
-
-		ret = armada_drm_find_components(&pdev->dev, &match);
-		if (ret < 0)
-			return ret;
-
-		return component_master_add_with_match(&pdev->dev,
-				&armada_master_ops, match);
-	} else {
-		return drm_platform_init(&armada_drm_driver, pdev);
-	}
+	struct component_match *match = NULL;
+	int ret;
+
+	ret = armada_drm_find_components(&pdev->dev, &match);
+	if (ret < 0)
+		return ret;
+
+	return component_master_add_with_match(&pdev->dev, &armada_master_ops,
+					       match);
 }
 
 static int armada_drm_remove(struct platform_device *pdev)
 {
-	if (is_componentized(&pdev->dev))
-		component_master_del(&pdev->dev, &armada_master_ops);
-	else
-		drm_put_dev(platform_get_drvdata(pdev));
+	component_master_del(&pdev->dev, &armada_master_ops);
 	return 0;
 }
 
 static const struct platform_device_id armada_drm_platform_ids[] = {
 	{
 		.name		= "armada-drm",
-		.driver_data	= (unsigned long)&armada510_ops,
 	}, {
 		.name		= "armada-510-drm",
-		.driver_data	= (unsigned long)&armada510_ops,
 	},
 	{ },
 };
diff --git a/drivers/gpu/drm/armada/armada_output.c b/drivers/gpu/drm/armada/armada_output.c
deleted file mode 100644
index 5a98231..0000000
--- a/drivers/gpu/drm/armada/armada_output.c
+++ /dev/null
@@ -1,142 +0,0 @@
-/*
- * Copyright (C) 2012 Russell King
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-#include <drm/drmP.h>
-#include <drm/drm_crtc_helper.h>
-#include <drm/drm_edid.h>
-#include <drm/drm_encoder_slave.h>
-#include "armada_output.h"
-#include "armada_drm.h"
-
-struct armada_connector {
-	struct drm_connector conn;
-	const struct armada_output_type *type;
-};
-
-#define drm_to_armada_conn(c) container_of(c, struct armada_connector, conn)
-
-struct drm_encoder *armada_drm_connector_encoder(struct drm_connector *conn)
-{
-	struct drm_encoder *enc = conn->encoder;
-
-	return enc ? enc : drm_encoder_find(conn->dev, conn->encoder_ids[0]);
-}
-
-static enum drm_connector_status armada_drm_connector_detect(
-	struct drm_connector *conn, bool force)
-{
-	struct armada_connector *dconn = drm_to_armada_conn(conn);
-	enum drm_connector_status status = connector_status_disconnected;
-
-	if (dconn->type->detect) {
-		status = dconn->type->detect(conn, force);
-	} else {
-		struct drm_encoder *enc = armada_drm_connector_encoder(conn);
-
-		if (enc)
-			status = encoder_helper_funcs(enc)->detect(enc, conn);
-	}
-
-	return status;
-}
-
-static void armada_drm_connector_destroy(struct drm_connector *conn)
-{
-	struct armada_connector *dconn = drm_to_armada_conn(conn);
-
-	drm_connector_unregister(conn);
-	drm_connector_cleanup(conn);
-	kfree(dconn);
-}
-
-static int armada_drm_connector_set_property(struct drm_connector *conn,
-	struct drm_property *property, uint64_t value)
-{
-	struct armada_connector *dconn = drm_to_armada_conn(conn);
-
-	if (!dconn->type->set_property)
-		return -EINVAL;
-
-	return dconn->type->set_property(conn, property, value);
-}
-
-static const struct drm_connector_funcs armada_drm_conn_funcs = {
-	.dpms		= drm_helper_connector_dpms,
-	.fill_modes	= drm_helper_probe_single_connector_modes,
-	.detect		= armada_drm_connector_detect,
-	.destroy	= armada_drm_connector_destroy,
-	.set_property	= armada_drm_connector_set_property,
-};
-
-/* Shouldn't this be a generic helper function? */
-int armada_drm_slave_encoder_mode_valid(struct drm_connector *conn,
-	struct drm_display_mode *mode)
-{
-	struct drm_encoder *encoder = armada_drm_connector_encoder(conn);
-	int valid = MODE_BAD;
-
-	if (encoder) {
-		struct drm_encoder_slave *slave = to_encoder_slave(encoder);
-
-		valid = slave->slave_funcs->mode_valid(encoder, mode);
-	}
-	return valid;
-}
-
-int armada_drm_slave_encoder_set_property(struct drm_connector *conn,
-	struct drm_property *property, uint64_t value)
-{
-	struct drm_encoder *encoder = armada_drm_connector_encoder(conn);
-	int rc = -EINVAL;
-
-	if (encoder) {
-		struct drm_encoder_slave *slave = to_encoder_slave(encoder);
-
-		rc = slave->slave_funcs->set_property(encoder, conn, property,
-						      value);
-	}
-	return rc;
-}
-
-int armada_output_create(struct drm_device *dev,
-	const struct armada_output_type *type, const void *data)
-{
-	struct armada_connector *dconn;
-	int ret;
-
-	dconn = kzalloc(sizeof(*dconn), GFP_KERNEL);
-	if (!dconn)
-		return -ENOMEM;
-
-	dconn->type = type;
-
-	ret = drm_connector_init(dev, &dconn->conn, &armada_drm_conn_funcs,
-				 type->connector_type);
-	if (ret) {
-		DRM_ERROR("unable to init connector\n");
-		goto err_destroy_dconn;
-	}
-
-	ret = type->create(&dconn->conn, data);
-	if (ret)
-		goto err_conn;
-
-	ret = drm_connector_register(&dconn->conn);
-	if (ret)
-		goto err_sysfs;
-
-	return 0;
-
- err_sysfs:
-	if (dconn->conn.encoder)
-		dconn->conn.encoder->funcs->destroy(dconn->conn.encoder);
- err_conn:
-	drm_connector_cleanup(&dconn->conn);
- err_destroy_dconn:
-	kfree(dconn);
-	return ret;
-}
diff --git a/drivers/gpu/drm/armada/armada_output.h b/drivers/gpu/drm/armada/armada_output.h
deleted file mode 100644
index f448785..0000000
--- a/drivers/gpu/drm/armada/armada_output.h
+++ /dev/null
@@ -1,33 +0,0 @@
-/*
- * Copyright (C) 2012 Russell King
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-#ifndef ARMADA_CONNETOR_H
-#define ARMADA_CONNETOR_H
-
-#define encoder_helper_funcs(encoder) \
-	((const struct drm_encoder_helper_funcs *)encoder->helper_private)
-
-struct armada_output_type {
-	int connector_type;
-	enum drm_connector_status (*detect)(struct drm_connector *, bool);
-	int (*create)(struct drm_connector *, const void *);
-	int (*set_property)(struct drm_connector *, struct drm_property *,
-			    uint64_t);
-};
-
-struct drm_encoder *armada_drm_connector_encoder(struct drm_connector *conn);
-
-int armada_drm_slave_encoder_mode_valid(struct drm_connector *conn,
-	struct drm_display_mode *mode);
-
-int armada_drm_slave_encoder_set_property(struct drm_connector *conn,
-	struct drm_property *property, uint64_t value);
-
-int armada_output_create(struct drm_device *dev,
-	const struct armada_output_type *type, const void *data);
-
-#endif
diff --git a/drivers/gpu/drm/armada/armada_slave.c b/drivers/gpu/drm/armada/armada_slave.c
deleted file mode 100644
index 00d0fac..0000000
--- a/drivers/gpu/drm/armada/armada_slave.c
+++ /dev/null
@@ -1,139 +0,0 @@
-/*
- * Copyright (C) 2012 Russell King
- *  Rewritten from the dovefb driver, and Armada510 manuals.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-#include <drm/drmP.h>
-#include <drm/drm_crtc_helper.h>
-#include <drm/drm_edid.h>
-#include <drm/drm_encoder_slave.h>
-#include "armada_drm.h"
-#include "armada_output.h"
-#include "armada_slave.h"
-
-static int armada_drm_slave_get_modes(struct drm_connector *conn)
-{
-	struct drm_encoder *enc = armada_drm_connector_encoder(conn);
-	int count = 0;
-
-	if (enc) {
-		struct drm_encoder_slave *slave = to_encoder_slave(enc);
-
-		count = slave->slave_funcs->get_modes(enc, conn);
-	}
-
-	return count;
-}
-
-static void armada_drm_slave_destroy(struct drm_encoder *enc)
-{
-	struct drm_encoder_slave *slave = to_encoder_slave(enc);
-	struct i2c_client *client = drm_i2c_encoder_get_client(enc);
-
-	if (slave->slave_funcs)
-		slave->slave_funcs->destroy(enc);
-	if (client)
-		i2c_put_adapter(client->adapter);
-
-	drm_encoder_cleanup(&slave->base);
-	kfree(slave);
-}
-
-static const struct drm_encoder_funcs armada_drm_slave_encoder_funcs = {
-	.destroy	= armada_drm_slave_destroy,
-};
-
-static const struct drm_connector_helper_funcs armada_drm_slave_helper_funcs = {
-	.get_modes	= armada_drm_slave_get_modes,
-	.mode_valid	= armada_drm_slave_encoder_mode_valid,
-	.best_encoder	= armada_drm_connector_encoder,
-};
-
-static const struct drm_encoder_helper_funcs drm_slave_encoder_helpers = {
-	.dpms = drm_i2c_encoder_dpms,
-	.save = drm_i2c_encoder_save,
-	.restore = drm_i2c_encoder_restore,
-	.mode_fixup = drm_i2c_encoder_mode_fixup,
-	.prepare = drm_i2c_encoder_prepare,
-	.commit = drm_i2c_encoder_commit,
-	.mode_set = drm_i2c_encoder_mode_set,
-	.detect = drm_i2c_encoder_detect,
-};
-
-static int
-armada_drm_conn_slave_create(struct drm_connector *conn, const void *data)
-{
-	const struct armada_drm_slave_config *config = data;
-	struct drm_encoder_slave *slave;
-	struct i2c_adapter *adap;
-	int ret;
-
-	conn->interlace_allowed = config->interlace_allowed;
-	conn->doublescan_allowed = config->doublescan_allowed;
-	conn->polled = config->polled;
-
-	drm_connector_helper_add(conn, &armada_drm_slave_helper_funcs);
-
-	slave = kzalloc(sizeof(*slave), GFP_KERNEL);
-	if (!slave)
-		return -ENOMEM;
-
-	slave->base.possible_crtcs = config->crtcs;
-
-	adap = i2c_get_adapter(config->i2c_adapter_id);
-	if (!adap) {
-		kfree(slave);
-		return -EPROBE_DEFER;
-	}
-
-	ret = drm_encoder_init(conn->dev, &slave->base,
-			       &armada_drm_slave_encoder_funcs,
-			       DRM_MODE_ENCODER_TMDS);
-	if (ret) {
-		DRM_ERROR("unable to init encoder\n");
-		i2c_put_adapter(adap);
-		kfree(slave);
-		return ret;
-	}
-
-	ret = drm_i2c_encoder_init(conn->dev, slave, adap, &config->info);
-	i2c_put_adapter(adap);
-	if (ret) {
-		DRM_ERROR("unable to init encoder slave\n");
-		armada_drm_slave_destroy(&slave->base);
-		return ret;
-	}
-
-	drm_encoder_helper_add(&slave->base, &drm_slave_encoder_helpers);
-
-	ret = slave->slave_funcs->create_resources(&slave->base, conn);
-	if (ret) {
-		armada_drm_slave_destroy(&slave->base);
-		return ret;
-	}
-
-	ret = drm_mode_connector_attach_encoder(conn, &slave->base);
-	if (ret) {
-		armada_drm_slave_destroy(&slave->base);
-		return ret;
-	}
-
-	conn->encoder = &slave->base;
-
-	return ret;
-}
-
-static const struct armada_output_type armada_drm_conn_slave = {
-	.connector_type	= DRM_MODE_CONNECTOR_HDMIA,
-	.create		= armada_drm_conn_slave_create,
-	.set_property	= armada_drm_slave_encoder_set_property,
-};
-
-int armada_drm_connector_slave_create(struct drm_device *dev,
-	const struct armada_drm_slave_config *config)
-{
-	return armada_output_create(dev, &armada_drm_conn_slave, config);
-}
diff --git a/drivers/gpu/drm/armada/armada_slave.h b/drivers/gpu/drm/armada/armada_slave.h
deleted file mode 100644
index bf2374c..0000000
--- a/drivers/gpu/drm/armada/armada_slave.h
+++ /dev/null
@@ -1,26 +0,0 @@
-/*
- * Copyright (C) 2012 Russell King
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-#ifndef ARMADA_SLAVE_H
-#define ARMADA_SLAVE_H
-
-#include <linux/i2c.h>
-#include <drm/drmP.h>
-
-struct armada_drm_slave_config {
-	int i2c_adapter_id;
-	uint32_t crtcs;
-	uint8_t polled;
-	bool interlace_allowed;
-	bool doublescan_allowed;
-	struct i2c_board_info info;
-};
-
-int armada_drm_connector_slave_create(struct drm_device *dev,
-	const struct armada_drm_slave_config *);
-
-#endif
-- 
cgit v0.10.2


From 7c8f7e1abc75b853adf60d8ee0a589e058bcdb6b Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Mon, 29 Jun 2015 17:52:16 +0100
Subject: drm/armada: move vbl code into armada_crtc

Our vblank event code belongs in armada_crtc.c rather than the core of
the driver.  Move it there.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/armada/armada_crtc.c b/drivers/gpu/drm/armada/armada_crtc.c
index c7374a3..418594b 100644
--- a/drivers/gpu/drm/armada/armada_crtc.c
+++ b/drivers/gpu/drm/armada/armada_crtc.c
@@ -173,6 +173,44 @@ static unsigned armada_drm_crtc_calc_fb(struct drm_framebuffer *fb,
 	return i;
 }
 
+void armada_drm_vbl_event_add(struct armada_crtc *dcrtc,
+	struct armada_vbl_event *evt)
+{
+	unsigned long flags;
+	bool not_on_list;
+
+	WARN_ON(drm_vblank_get(dcrtc->crtc.dev, dcrtc->num));
+
+	spin_lock_irqsave(&dcrtc->irq_lock, flags);
+	not_on_list = list_empty(&evt->node);
+	if (not_on_list)
+		list_add_tail(&evt->node, &dcrtc->vbl_list);
+	spin_unlock_irqrestore(&dcrtc->irq_lock, flags);
+
+	if (!not_on_list)
+		drm_vblank_put(dcrtc->crtc.dev, dcrtc->num);
+}
+
+void armada_drm_vbl_event_remove(struct armada_crtc *dcrtc,
+	struct armada_vbl_event *evt)
+{
+	if (!list_empty(&evt->node)) {
+		list_del_init(&evt->node);
+		drm_vblank_put(dcrtc->crtc.dev, dcrtc->num);
+	}
+}
+
+static void armada_drm_vbl_event_run(struct armada_crtc *dcrtc)
+{
+	struct armada_vbl_event *e, *n;
+
+	list_for_each_entry_safe(e, n, &dcrtc->vbl_list, node) {
+		list_del_init(&e->node);
+		drm_vblank_put(dcrtc->crtc.dev, dcrtc->num);
+		e->fn(dcrtc, e->data);
+	}
+}
+
 static int armada_drm_crtc_queue_frame_work(struct armada_crtc *dcrtc,
 	struct armada_frame_work *work)
 {
@@ -356,7 +394,6 @@ static bool armada_drm_crtc_mode_fixup(struct drm_crtc *crtc,
 
 static void armada_drm_crtc_irq(struct armada_crtc *dcrtc, u32 stat)
 {
-	struct armada_vbl_event *e, *n;
 	void __iomem *base = dcrtc->base;
 
 	if (stat & DMA_FF_UNDERFLOW)
@@ -368,12 +405,7 @@ static void armada_drm_crtc_irq(struct armada_crtc *dcrtc, u32 stat)
 		drm_handle_vblank(dcrtc->crtc.dev, dcrtc->num);
 
 	spin_lock(&dcrtc->irq_lock);
-
-	list_for_each_entry_safe(e, n, &dcrtc->vbl_list, node) {
-		list_del_init(&e->node);
-		drm_vblank_put(dcrtc->crtc.dev, dcrtc->num);
-		e->fn(dcrtc, e->data);
-	}
+	armada_drm_vbl_event_run(dcrtc);
 
 	if (stat & GRA_FRAME_IRQ && dcrtc->interlaced) {
 		int i = stat & GRA_FRAME_IRQ0 ? 0 : 1;
diff --git a/drivers/gpu/drm/armada/armada_crtc.h b/drivers/gpu/drm/armada/armada_crtc.h
index a13469f..a86243e 100644
--- a/drivers/gpu/drm/armada/armada_crtc.h
+++ b/drivers/gpu/drm/armada/armada_crtc.h
@@ -75,6 +75,23 @@ struct armada_crtc {
 };
 #define drm_to_armada_crtc(c) container_of(c, struct armada_crtc, crtc)
 
+struct armada_vbl_event {
+	struct list_head	node;
+	void			*data;
+	void			(*fn)(struct armada_crtc *, void *);
+};
+
+void armada_drm_vbl_event_add(struct armada_crtc *,
+	struct armada_vbl_event *);
+void armada_drm_vbl_event_remove(struct armada_crtc *,
+	struct armada_vbl_event *);
+#define armada_drm_vbl_event_init(_e, _f, _d) do {	\
+	struct armada_vbl_event *__e = _e;		\
+	INIT_LIST_HEAD(&__e->node);			\
+	__e->data = _d;					\
+	__e->fn = _f;					\
+} while (0)
+
 void armada_drm_crtc_gamma_set(struct drm_crtc *, u16, u16, u16, int);
 void armada_drm_crtc_gamma_get(struct drm_crtc *, u16 *, u16 *, u16 *, int);
 void armada_drm_crtc_disable_irq(struct armada_crtc *, u32);
diff --git a/drivers/gpu/drm/armada/armada_drm.h b/drivers/gpu/drm/armada/armada_drm.h
index 5f6aef0..4df6f2a 100644
--- a/drivers/gpu/drm/armada/armada_drm.h
+++ b/drivers/gpu/drm/armada/armada_drm.h
@@ -37,22 +37,6 @@ static inline uint32_t armada_pitch(uint32_t width, uint32_t bpp)
 	return ALIGN(pitch, 128);
 }
 
-struct armada_vbl_event {
-	struct list_head	node;
-	void			*data;
-	void			(*fn)(struct armada_crtc *, void *);
-};
-void armada_drm_vbl_event_add(struct armada_crtc *,
-	struct armada_vbl_event *);
-void armada_drm_vbl_event_remove(struct armada_crtc *,
-	struct armada_vbl_event *);
-#define armada_drm_vbl_event_init(_e, _f, _d) do {	\
-	struct armada_vbl_event *__e = _e;		\
-	INIT_LIST_HEAD(&__e->node);			\
-	__e->data = _d;					\
-	__e->fn = _f;					\
-} while (0)
-
 
 struct armada_private;
 
diff --git a/drivers/gpu/drm/armada/armada_drv.c b/drivers/gpu/drm/armada/armada_drv.c
index b373cf9..3f1396e 100644
--- a/drivers/gpu/drm/armada/armada_drv.c
+++ b/drivers/gpu/drm/armada/armada_drv.c
@@ -148,29 +148,6 @@ static int armada_drm_unload(struct drm_device *dev)
 	return 0;
 }
 
-void armada_drm_vbl_event_add(struct armada_crtc *dcrtc,
-	struct armada_vbl_event *evt)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&dcrtc->irq_lock, flags);
-	if (list_empty(&evt->node)) {
-		list_add_tail(&evt->node, &dcrtc->vbl_list);
-
-		drm_vblank_get(dcrtc->crtc.dev, dcrtc->num);
-	}
-	spin_unlock_irqrestore(&dcrtc->irq_lock, flags);
-}
-
-void armada_drm_vbl_event_remove(struct armada_crtc *dcrtc,
-	struct armada_vbl_event *evt)
-{
-	if (!list_empty(&evt->node)) {
-		list_del_init(&evt->node);
-		drm_vblank_put(dcrtc->crtc.dev, dcrtc->num);
-	}
-}
-
 /* These are called under the vbl_lock. */
 static int armada_drm_enable_vblank(struct drm_device *dev, int crtc)
 {
-- 
cgit v0.10.2


From f8e140698234dae3a4ea7b971e7bf63a3e0c987a Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Mon, 29 Jun 2015 17:52:42 +0100
Subject: drm/armada: use drm_plane_force_disable() to disable the overlay
 plane

Use drm_plane_force_disable() to disable the overlay plane on a mode_set
rather than coding this ourselves.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/armada/armada_crtc.c b/drivers/gpu/drm/armada/armada_crtc.c
index 418594b..bbf5ff7 100644
--- a/drivers/gpu/drm/armada/armada_crtc.c
+++ b/drivers/gpu/drm/armada/armada_crtc.c
@@ -348,17 +348,11 @@ static void armada_drm_crtc_prepare(struct drm_crtc *crtc)
 	/*
 	 * If we have an overlay plane associated with this CRTC, disable
 	 * it before the modeset to avoid its coordinates being outside
-	 * the new mode parameters.  DRM doesn't provide help with this.
+	 * the new mode parameters.
 	 */
 	plane = dcrtc->plane;
-	if (plane) {
-		struct drm_framebuffer *fb = plane->fb;
-
-		plane->funcs->disable_plane(plane);
-		plane->fb = NULL;
-		plane->crtc = NULL;
-		drm_framebuffer_unreference(fb);
-	}
+	if (plane)
+		drm_plane_force_disable(plane);
 }
 
 /* The mode_config.mutex will be held for this call */
-- 
cgit v0.10.2


From e0ac5e9b4b14ab4be7fbba48d666fc619342fd88 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Mon, 29 Jun 2015 18:01:38 +0100
Subject: drm/armada: disable CRTC clock during DPMS

When the CRTC is in low power mode, it isn't running, and so there's
no point keeping the CRTC clock enabled.  Disable the CRTC clock during
DPMS.

We need to re-enable it in the mode_set callback to ensure that the
variant's compute_clock() continues to see its clock in the expected
state (enabled).

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/armada/armada_crtc.c b/drivers/gpu/drm/armada/armada_crtc.c
index bbf5ff7..8c43ecc 100644
--- a/drivers/gpu/drm/armada/armada_crtc.c
+++ b/drivers/gpu/drm/armada/armada_crtc.c
@@ -325,7 +325,11 @@ static void armada_drm_crtc_dpms(struct drm_crtc *crtc, int dpms)
 
 	if (dcrtc->dpms != dpms) {
 		dcrtc->dpms = dpms;
+		if (!IS_ERR(dcrtc->clk) && !dpms_blanked(dpms))
+			WARN_ON(clk_prepare_enable(dcrtc->clk));
 		armada_drm_crtc_update(dcrtc);
+		if (!IS_ERR(dcrtc->clk) && dpms_blanked(dpms))
+			clk_disable_unprepare(dcrtc->clk);
 		if (dpms_blanked(dpms))
 			armada_drm_vblank_off(dcrtc);
 		else
@@ -563,6 +567,13 @@ static int armada_drm_crtc_mode_set(struct drm_crtc *crtc,
 		writel_relaxed(val, dcrtc->base + LCD_SPU_DUMB_CTRL);
 	}
 
+	/*
+	 * If we are blanked, we would have disabled the clock.  Re-enable
+	 * it so that compute_clock() does the right thing.
+	 */
+	if (!IS_ERR(dcrtc->clk) && dpms_blanked(dcrtc->dpms))
+		WARN_ON(clk_prepare_enable(dcrtc->clk));
+
 	/* Now compute the divider for real */
 	dcrtc->variant->compute_clock(dcrtc, adj, &sclk);
 
-- 
cgit v0.10.2


From 709ffd82fc6ff760dc3a7f71bdf26d78a8e3caf0 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Wed, 15 Jul 2015 18:09:38 +0100
Subject: drm/armada: redo locking and atomics for
 armada_drm_crtc_complete_frame_work()

We can do better with armada_drm_crtc_complete_frame_work() - we can
avoid taking the event lock unless a call to drm_send_vblank_event()
is required, and using cmpxchg() and xchg(), we can eliminate the
locking around dcrtc->frame_work entirely.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/armada/armada_crtc.c b/drivers/gpu/drm/armada/armada_crtc.c
index 8c43ecc..5d62764 100644
--- a/drivers/gpu/drm/armada/armada_crtc.c
+++ b/drivers/gpu/drm/armada/armada_crtc.c
@@ -215,7 +215,6 @@ static int armada_drm_crtc_queue_frame_work(struct armada_crtc *dcrtc,
 	struct armada_frame_work *work)
 {
 	struct drm_device *dev = dcrtc->crtc.dev;
-	unsigned long flags;
 	int ret;
 
 	ret = drm_vblank_get(dev, dcrtc->num);
@@ -224,30 +223,29 @@ static int armada_drm_crtc_queue_frame_work(struct armada_crtc *dcrtc,
 		return ret;
 	}
 
-	spin_lock_irqsave(&dev->event_lock, flags);
-	if (!dcrtc->frame_work)
-		dcrtc->frame_work = work;
-	else
-		ret = -EBUSY;
-	spin_unlock_irqrestore(&dev->event_lock, flags);
-
-	if (ret)
+	if (cmpxchg(&dcrtc->frame_work, NULL, work)) {
 		drm_vblank_put(dev, dcrtc->num);
+		ret = -EBUSY;
+	}
 
 	return ret;
 }
 
-static void armada_drm_crtc_complete_frame_work(struct armada_crtc *dcrtc)
+static void armada_drm_crtc_complete_frame_work(struct armada_crtc *dcrtc,
+	struct armada_frame_work *work)
 {
 	struct drm_device *dev = dcrtc->crtc.dev;
-	struct armada_frame_work *work = dcrtc->frame_work;
-
-	dcrtc->frame_work = NULL;
+	unsigned long flags;
 
+	spin_lock_irqsave(&dcrtc->irq_lock, flags);
 	armada_drm_crtc_update_regs(dcrtc, work->regs);
+	spin_unlock_irqrestore(&dcrtc->irq_lock, flags);
 
-	if (work->event)
+	if (work->event) {
+		spin_lock_irqsave(&dev->event_lock, flags);
 		drm_send_vblank_event(dev, dcrtc->num, work->event);
+		spin_unlock_irqrestore(&dev->event_lock, flags);
+	}
 
 	drm_vblank_put(dev, dcrtc->num);
 
@@ -293,7 +291,7 @@ static void armada_drm_crtc_finish_fb(struct armada_crtc *dcrtc,
 
 static void armada_drm_vblank_off(struct armada_crtc *dcrtc)
 {
-	struct drm_device *dev = dcrtc->crtc.dev;
+	struct armada_frame_work *work;
 
 	/*
 	 * Tell the DRM core that vblank IRQs aren't going to happen for
@@ -302,10 +300,9 @@ static void armada_drm_vblank_off(struct armada_crtc *dcrtc)
 	drm_crtc_vblank_off(&dcrtc->crtc);
 
 	/* Handle any pending flip event. */
-	spin_lock_irq(&dev->event_lock);
-	if (dcrtc->frame_work)
-		armada_drm_crtc_complete_frame_work(dcrtc);
-	spin_unlock_irq(&dev->event_lock);
+	work = xchg(&dcrtc->frame_work, NULL);
+	if (work)
+		armada_drm_crtc_complete_frame_work(dcrtc, work);
 }
 
 void armada_drm_crtc_gamma_set(struct drm_crtc *crtc, u16 r, u16 g, u16 b,
@@ -434,12 +431,10 @@ static void armada_drm_crtc_irq(struct armada_crtc *dcrtc, u32 stat)
 	spin_unlock(&dcrtc->irq_lock);
 
 	if (stat & GRA_FRAME_IRQ) {
-		struct drm_device *dev = dcrtc->crtc.dev;
+		struct armada_frame_work *work = xchg(&dcrtc->frame_work, NULL);
 
-		spin_lock(&dev->event_lock);
-		if (dcrtc->frame_work)
-			armada_drm_crtc_complete_frame_work(dcrtc);
-		spin_unlock(&dev->event_lock);
+		if (work)
+			armada_drm_crtc_complete_frame_work(dcrtc, work);
 
 		wake_up(&dcrtc->frame_wait);
 	}
@@ -957,8 +952,6 @@ static int armada_drm_crtc_page_flip(struct drm_crtc *crtc,
 {
 	struct armada_crtc *dcrtc = drm_to_armada_crtc(crtc);
 	struct armada_frame_work *work;
-	struct drm_device *dev = crtc->dev;
-	unsigned long flags;
 	unsigned i;
 	int ret;
 
@@ -1004,10 +997,10 @@ static int armada_drm_crtc_page_flip(struct drm_crtc *crtc,
 	 * interrupt, so complete it now.
 	 */
 	if (dpms_blanked(dcrtc->dpms)) {
-		spin_lock_irqsave(&dev->event_lock, flags);
-		if (dcrtc->frame_work)
-			armada_drm_crtc_complete_frame_work(dcrtc);
-		spin_unlock_irqrestore(&dev->event_lock, flags);
+		struct armada_frame_work *work = xchg(&dcrtc->frame_work, NULL);
+
+		if (work)
+			armada_drm_crtc_complete_frame_work(dcrtc, work);
 	}
 
 	return 0;
-- 
cgit v0.10.2


From 28a2aebed6374c4af5224114e4b4273a3aae649a Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Wed, 15 Jul 2015 18:11:23 +0100
Subject: drm/armada: rename overlay identifiers

Include an _ovl infix into the overlay identifiers to separate them from
the primary plane.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/armada/armada_overlay.c b/drivers/gpu/drm/armada/armada_overlay.c
index e939fab..9393a15 100644
--- a/drivers/gpu/drm/armada/armada_overlay.c
+++ b/drivers/gpu/drm/armada/armada_overlay.c
@@ -16,7 +16,7 @@
 #include <drm/armada_drm.h>
 #include "armada_ioctlP.h"
 
-struct armada_plane_properties {
+struct armada_ovl_plane_properties {
 	uint32_t colorkey_yr;
 	uint32_t colorkey_ug;
 	uint32_t colorkey_vb;
@@ -29,7 +29,7 @@ struct armada_plane_properties {
 	uint32_t colorkey_mode;
 };
 
-struct armada_plane {
+struct armada_ovl_plane {
 	struct drm_plane base;
 	spinlock_t lock;
 	struct drm_framebuffer *old_fb;
@@ -42,13 +42,13 @@ struct armada_plane {
 		struct armada_regs regs[13];
 		wait_queue_head_t wait;
 	} vbl;
-	struct armada_plane_properties prop;
+	struct armada_ovl_plane_properties prop;
 };
-#define drm_to_armada_plane(p) container_of(p, struct armada_plane, base)
+#define drm_to_armada_ovl_plane(p) container_of(p, struct armada_ovl_plane, base)
 
 
 static void
-armada_ovl_update_attr(struct armada_plane_properties *prop,
+armada_ovl_update_attr(struct armada_ovl_plane_properties *prop,
 	struct armada_crtc *dcrtc)
 {
 	writel_relaxed(prop->colorkey_yr, dcrtc->base + LCD_SPU_COLORKEY_Y);
@@ -72,9 +72,9 @@ armada_ovl_update_attr(struct armada_plane_properties *prop,
 }
 
 /* === Plane support === */
-static void armada_plane_vbl(struct armada_crtc *dcrtc, void *data)
+static void armada_ovl_plane_vbl(struct armada_crtc *dcrtc, void *data)
 {
-	struct armada_plane *dplane = data;
+	struct armada_ovl_plane *dplane = data;
 	struct drm_framebuffer *fb;
 
 	armada_drm_crtc_update_regs(dcrtc, dplane->vbl.regs);
@@ -91,12 +91,12 @@ static void armada_plane_vbl(struct armada_crtc *dcrtc, void *data)
 }
 
 static int
-armada_plane_update(struct drm_plane *plane, struct drm_crtc *crtc,
+armada_ovl_plane_update(struct drm_plane *plane, struct drm_crtc *crtc,
 	struct drm_framebuffer *fb,
 	int crtc_x, int crtc_y, unsigned crtc_w, unsigned crtc_h,
 	uint32_t src_x, uint32_t src_y, uint32_t src_w, uint32_t src_h)
 {
-	struct armada_plane *dplane = drm_to_armada_plane(plane);
+	struct armada_ovl_plane *dplane = drm_to_armada_ovl_plane(plane);
 	struct armada_crtc *dcrtc = drm_to_armada_crtc(crtc);
 	struct drm_rect src = {
 		.x1 = src_x,
@@ -267,9 +267,9 @@ armada_plane_update(struct drm_plane *plane, struct drm_crtc *crtc,
 	return 0;
 }
 
-static int armada_plane_disable(struct drm_plane *plane)
+static int armada_ovl_plane_disable(struct drm_plane *plane)
 {
-	struct armada_plane *dplane = drm_to_armada_plane(plane);
+	struct armada_ovl_plane *dplane = drm_to_armada_ovl_plane(plane);
 	struct drm_framebuffer *fb;
 	struct armada_crtc *dcrtc;
 
@@ -302,20 +302,20 @@ static int armada_plane_disable(struct drm_plane *plane)
 	return 0;
 }
 
-static void armada_plane_destroy(struct drm_plane *plane)
+static void armada_ovl_plane_destroy(struct drm_plane *plane)
 {
-	struct armada_plane *dplane = drm_to_armada_plane(plane);
+	struct armada_ovl_plane *dplane = drm_to_armada_ovl_plane(plane);
 
 	drm_plane_cleanup(plane);
 
 	kfree(dplane);
 }
 
-static int armada_plane_set_property(struct drm_plane *plane,
+static int armada_ovl_plane_set_property(struct drm_plane *plane,
 	struct drm_property *property, uint64_t val)
 {
 	struct armada_private *priv = plane->dev->dev_private;
-	struct armada_plane *dplane = drm_to_armada_plane(plane);
+	struct armada_ovl_plane *dplane = drm_to_armada_ovl_plane(plane);
 	bool update_attr = false;
 
 	if (property == priv->colorkey_prop) {
@@ -379,14 +379,14 @@ static int armada_plane_set_property(struct drm_plane *plane,
 	return 0;
 }
 
-static const struct drm_plane_funcs armada_plane_funcs = {
-	.update_plane	= armada_plane_update,
-	.disable_plane	= armada_plane_disable,
-	.destroy	= armada_plane_destroy,
-	.set_property	= armada_plane_set_property,
+static const struct drm_plane_funcs armada_ovl_plane_funcs = {
+	.update_plane	= armada_ovl_plane_update,
+	.disable_plane	= armada_ovl_plane_disable,
+	.destroy	= armada_ovl_plane_destroy,
+	.set_property	= armada_ovl_plane_set_property,
 };
 
-static const uint32_t armada_formats[] = {
+static const uint32_t armada_ovl_formats[] = {
 	DRM_FORMAT_UYVY,
 	DRM_FORMAT_YUYV,
 	DRM_FORMAT_YUV420,
@@ -456,7 +456,7 @@ int armada_overlay_plane_create(struct drm_device *dev, unsigned long crtcs)
 {
 	struct armada_private *priv = dev->dev_private;
 	struct drm_mode_object *mobj;
-	struct armada_plane *dplane;
+	struct armada_ovl_plane *dplane;
 	int ret;
 
 	ret = armada_overlay_create_properties(dev);
@@ -469,11 +469,16 @@ int armada_overlay_plane_create(struct drm_device *dev, unsigned long crtcs)
 
 	spin_lock_init(&dplane->lock);
 	init_waitqueue_head(&dplane->vbl.wait);
-	armada_drm_vbl_event_init(&dplane->vbl.update, armada_plane_vbl,
+	armada_drm_vbl_event_init(&dplane->vbl.update, armada_ovl_plane_vbl,
 				  dplane);
 
-	drm_plane_init(dev, &dplane->base, crtcs, &armada_plane_funcs,
-		       armada_formats, ARRAY_SIZE(armada_formats), false);
+	drm_plane_init(dev, &dplane->base, crtcs, &armada_ovl_plane_funcs,
+		       armada_ovl_formats, ARRAY_SIZE(armada_ovl_formats),
+		       false);
+	if (ret) {
+		kfree(dplane);
+		return ret;
+	}
 
 	dplane->prop.colorkey_yr = 0xfefefe00;
 	dplane->prop.colorkey_ug = 0x01010100;
-- 
cgit v0.10.2


From fecfdb2db8b5fc2e6cb731a714889de5e43c2380 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Wed, 15 Jul 2015 18:11:24 +0100
Subject: drm/armada: factor out retirement of old fb

We have two identical places in the overlay code which retire the drm
framebuffer.  Factor these out into a common function.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/armada/armada_overlay.c b/drivers/gpu/drm/armada/armada_overlay.c
index 9393a15..093c2d4 100644
--- a/drivers/gpu/drm/armada/armada_overlay.c
+++ b/drivers/gpu/drm/armada/armada_overlay.c
@@ -71,21 +71,27 @@ armada_ovl_update_attr(struct armada_ovl_plane_properties *prop,
 	spin_unlock_irq(&dcrtc->irq_lock);
 }
 
+static void armada_ovl_retire_fb(struct armada_ovl_plane *dplane,
+	struct drm_framebuffer *fb)
+{
+	struct drm_framebuffer *old_fb;
+
+	spin_lock(&dplane->lock);
+	old_fb = dplane->old_fb;
+	dplane->old_fb = fb;
+	spin_unlock(&dplane->lock);
+
+	if (old_fb)
+		armada_drm_queue_unref_work(dplane->base.dev, old_fb);
+}
+
 /* === Plane support === */
 static void armada_ovl_plane_vbl(struct armada_crtc *dcrtc, void *data)
 {
 	struct armada_ovl_plane *dplane = data;
-	struct drm_framebuffer *fb;
 
 	armada_drm_crtc_update_regs(dcrtc, dplane->vbl.regs);
-
-	spin_lock(&dplane->lock);
-	fb = dplane->old_fb;
-	dplane->old_fb = NULL;
-	spin_unlock(&dplane->lock);
-
-	if (fb)
-		armada_drm_queue_unref_work(dcrtc->crtc.dev, fb);
+	armada_ovl_retire_fb(dplane, NULL);
 
 	wake_up(&dplane->vbl.wait);
 }
@@ -175,17 +181,8 @@ armada_ovl_plane_update(struct drm_plane *plane, struct drm_crtc *crtc,
 		 */
 		drm_framebuffer_reference(fb);
 
-		if (plane->fb) {
-			struct drm_framebuffer *older_fb;
-
-			spin_lock_irq(&dplane->lock);
-			older_fb = dplane->old_fb;
-			dplane->old_fb = plane->fb;
-			spin_unlock_irq(&dplane->lock);
-			if (older_fb)
-				armada_drm_queue_unref_work(dcrtc->crtc.dev,
-							    older_fb);
-		}
+		if (plane->fb)
+			armada_ovl_retire_fb(dplane, plane->fb);
 
 		src_y = src.y1 >> 16;
 		src_x = src.x1 >> 16;
-- 
cgit v0.10.2


From 9685cd9df75c1f5686308601c870f5e4ebc809be Mon Sep 17 00:00:00 2001
From: Rob Clark <robdclark@gmail.com>
Date: Tue, 25 Aug 2015 15:35:57 -0400
Subject: drm/fb-helper: add headerdoc for drm_fb_helper

Signed-off-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/include/drm/drm_fb_helper.h b/include/drm/drm_fb_helper.h
index 67de1f1..6254136 100644
--- a/include/drm/drm_fb_helper.h
+++ b/include/drm/drm_fb_helper.h
@@ -104,6 +104,20 @@ struct drm_fb_helper_connector {
 	struct drm_connector *connector;
 };
 
+/**
+ * struct drm_fb_helper - helper to emulate fbdev on top of kms
+ * @fb:  Scanout framebuffer object
+ * @dev:  DRM device
+ * @crtc_count: number of possible CRTCs
+ * @crtc_info: per-CRTC helper state (mode, x/y offset, etc)
+ * @connector_count: number of connected connectors
+ * @connector_info_alloc_count: size of connector_info
+ * @funcs: driver callbacks for fb helper
+ * @fbdev: emulated fbdev device info struct
+ * @pseudo_palette: fake palette of 16 colors
+ * @kernel_fb_list: list_head in kernel_fb_helper_list
+ * @delayed_hotplug: was there a hotplug while kms master active?
+ */
 struct drm_fb_helper {
 	struct drm_framebuffer *fb;
 	struct drm_device *dev;
-- 
cgit v0.10.2


From bbb1e52402b2a288b09ae37e8182599931c7e9df Mon Sep 17 00:00:00 2001
From: Rob Clark <robdclark@gmail.com>
Date: Tue, 25 Aug 2015 15:35:58 -0400
Subject: drm/fb-helper: atomic restore_fbdev_mode()..

Add support for using atomic code-paths for restore_fbdev_mode().

Signed-off-by: Rob Clark <robdclark@gmail.com>
[danvet: Bikeshed comments slightly.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c
index 94d6c8e..ee57ecf 100644
--- a/drivers/gpu/drm/drm_atomic_helper.c
+++ b/drivers/gpu/drm/drm_atomic_helper.c
@@ -1553,21 +1553,9 @@ retry:
 		goto fail;
 	}
 
-	ret = drm_atomic_set_crtc_for_plane(plane_state, NULL);
+	ret = __drm_atomic_helper_disable_plane(plane, plane_state);
 	if (ret != 0)
 		goto fail;
-	drm_atomic_set_fb_for_plane(plane_state, NULL);
-	plane_state->crtc_x = 0;
-	plane_state->crtc_y = 0;
-	plane_state->crtc_h = 0;
-	plane_state->crtc_w = 0;
-	plane_state->src_x = 0;
-	plane_state->src_y = 0;
-	plane_state->src_h = 0;
-	plane_state->src_w = 0;
-
-	if (plane == plane->crtc->cursor)
-		state->legacy_cursor_update = true;
 
 	ret = drm_atomic_commit(state);
 	if (ret != 0)
@@ -1597,6 +1585,32 @@ backoff:
 }
 EXPORT_SYMBOL(drm_atomic_helper_disable_plane);
 
+/* just used from fb-helper and atomic-helper: */
+int __drm_atomic_helper_disable_plane(struct drm_plane *plane,
+		struct drm_plane_state *plane_state)
+{
+	int ret;
+
+	ret = drm_atomic_set_crtc_for_plane(plane_state, NULL);
+	if (ret != 0)
+		return ret;
+
+	drm_atomic_set_fb_for_plane(plane_state, NULL);
+	plane_state->crtc_x = 0;
+	plane_state->crtc_y = 0;
+	plane_state->crtc_h = 0;
+	plane_state->crtc_w = 0;
+	plane_state->src_x = 0;
+	plane_state->src_y = 0;
+	plane_state->src_h = 0;
+	plane_state->src_w = 0;
+
+	if (plane->crtc && (plane == plane->crtc->cursor))
+		plane_state->state->legacy_cursor_update = true;
+
+	return 0;
+}
+
 static int update_output_state(struct drm_atomic_state *state,
 			       struct drm_mode_set *set)
 {
@@ -1680,8 +1694,6 @@ int drm_atomic_helper_set_config(struct drm_mode_set *set)
 {
 	struct drm_atomic_state *state;
 	struct drm_crtc *crtc = set->crtc;
-	struct drm_crtc_state *crtc_state;
-	struct drm_plane_state *primary_state;
 	int ret = 0;
 
 	state = drm_atomic_state_alloc(crtc->dev);
@@ -1690,17 +1702,54 @@ int drm_atomic_helper_set_config(struct drm_mode_set *set)
 
 	state->acquire_ctx = drm_modeset_legacy_acquire_ctx(crtc);
 retry:
-	crtc_state = drm_atomic_get_crtc_state(state, crtc);
-	if (IS_ERR(crtc_state)) {
-		ret = PTR_ERR(crtc_state);
+	ret = __drm_atomic_helper_set_config(set, state);
+	if (ret != 0)
 		goto fail;
-	}
 
-	primary_state = drm_atomic_get_plane_state(state, crtc->primary);
-	if (IS_ERR(primary_state)) {
-		ret = PTR_ERR(primary_state);
+	ret = drm_atomic_commit(state);
+	if (ret != 0)
 		goto fail;
-	}
+
+	/* Driver takes ownership of state on successful commit. */
+	return 0;
+fail:
+	if (ret == -EDEADLK)
+		goto backoff;
+
+	drm_atomic_state_free(state);
+
+	return ret;
+backoff:
+	drm_atomic_state_clear(state);
+	drm_atomic_legacy_backoff(state);
+
+	/*
+	 * Someone might have exchanged the framebuffer while we dropped locks
+	 * in the backoff code. We need to fix up the fb refcount tracking the
+	 * core does for us.
+	 */
+	crtc->primary->old_fb = crtc->primary->fb;
+
+	goto retry;
+}
+EXPORT_SYMBOL(drm_atomic_helper_set_config);
+
+/* just used from fb-helper and atomic-helper: */
+int __drm_atomic_helper_set_config(struct drm_mode_set *set,
+		struct drm_atomic_state *state)
+{
+	struct drm_crtc_state *crtc_state;
+	struct drm_plane_state *primary_state;
+	struct drm_crtc *crtc = set->crtc;
+	int ret;
+
+	crtc_state = drm_atomic_get_crtc_state(state, crtc);
+	if (IS_ERR(crtc_state))
+		return PTR_ERR(crtc_state);
+
+	primary_state = drm_atomic_get_plane_state(state, crtc->primary);
+	if (IS_ERR(primary_state))
+		return PTR_ERR(primary_state);
 
 	if (!set->mode) {
 		WARN_ON(set->fb);
@@ -1708,13 +1757,13 @@ retry:
 
 		ret = drm_atomic_set_mode_for_crtc(crtc_state, NULL);
 		if (ret != 0)
-			goto fail;
+			return ret;
 
 		crtc_state->active = false;
 
 		ret = drm_atomic_set_crtc_for_plane(primary_state, NULL);
 		if (ret != 0)
-			goto fail;
+			return ret;
 
 		drm_atomic_set_fb_for_plane(primary_state, NULL);
 
@@ -1726,13 +1775,14 @@ retry:
 
 	ret = drm_atomic_set_mode_for_crtc(crtc_state, set->mode);
 	if (ret != 0)
-		goto fail;
+		return ret;
 
 	crtc_state->active = true;
 
 	ret = drm_atomic_set_crtc_for_plane(primary_state, crtc);
 	if (ret != 0)
-		goto fail;
+		return ret;
+
 	drm_atomic_set_fb_for_plane(primary_state, set->fb);
 	primary_state->crtc_x = 0;
 	primary_state->crtc_y = 0;
@@ -1746,35 +1796,10 @@ retry:
 commit:
 	ret = update_output_state(state, set);
 	if (ret)
-		goto fail;
-
-	ret = drm_atomic_commit(state);
-	if (ret != 0)
-		goto fail;
+		return ret;
 
-	/* Driver takes ownership of state on successful commit. */
 	return 0;
-fail:
-	if (ret == -EDEADLK)
-		goto backoff;
-
-	drm_atomic_state_free(state);
-
-	return ret;
-backoff:
-	drm_atomic_state_clear(state);
-	drm_atomic_legacy_backoff(state);
-
-	/*
-	 * Someone might have exchanged the framebuffer while we dropped locks
-	 * in the backoff code. We need to fix up the fb refcount tracking the
-	 * core does for us.
-	 */
-	crtc->primary->old_fb = crtc->primary->fb;
-
-	goto retry;
 }
-EXPORT_SYMBOL(drm_atomic_helper_set_config);
 
 /**
  * drm_atomic_helper_crtc_set_property - helper for crtc properties
diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c
index ba12f51..0180fdd 100644
--- a/drivers/gpu/drm/drm_fb_helper.c
+++ b/drivers/gpu/drm/drm_fb_helper.c
@@ -38,6 +38,8 @@
 #include <drm/drm_crtc.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_crtc_helper.h>
+#include <drm/drm_atomic.h>
+#include <drm/drm_atomic_helper.h>
 
 static bool drm_fbdev_emulation = true;
 module_param_named(fbdev_emulation, drm_fbdev_emulation, bool, 0600);
@@ -334,6 +336,72 @@ int drm_fb_helper_debug_leave(struct fb_info *info)
 }
 EXPORT_SYMBOL(drm_fb_helper_debug_leave);
 
+static int restore_fbdev_mode_atomic(struct drm_fb_helper *fb_helper)
+{
+	struct drm_device *dev = fb_helper->dev;
+	struct drm_plane *plane;
+	struct drm_atomic_state *state;
+	int i, ret;
+
+	state = drm_atomic_state_alloc(dev);
+	if (!state)
+		return -ENOMEM;
+
+	state->acquire_ctx = dev->mode_config.acquire_ctx;
+retry:
+	drm_for_each_plane(plane, dev) {
+		struct drm_plane_state *plane_state;
+
+		plane_state = drm_atomic_get_plane_state(state, plane);
+		if (IS_ERR(plane_state)) {
+			ret = PTR_ERR(plane_state);
+			goto fail;
+		}
+
+		ret = drm_atomic_plane_set_property(plane, plane_state,
+				dev->mode_config.rotation_property,
+				BIT(DRM_ROTATE_0));
+		if (ret != 0)
+			goto fail;
+
+		/* disable non-primary: */
+		if (plane->type == DRM_PLANE_TYPE_PRIMARY)
+			continue;
+
+		ret = __drm_atomic_helper_disable_plane(plane, plane_state);
+		if (ret != 0)
+			goto fail;
+	}
+
+	for(i = 0; i < fb_helper->crtc_count; i++) {
+		struct drm_mode_set *mode_set = &fb_helper->crtc_info[i].mode_set;
+
+		ret = __drm_atomic_helper_set_config(mode_set, state);
+		if (ret != 0)
+			goto fail;
+	}
+
+	ret = drm_atomic_commit(state);
+	if (ret != 0)
+		goto fail;
+
+	return 0;
+
+fail:
+	if (ret == -EDEADLK)
+		goto backoff;
+
+	drm_atomic_state_free(state);
+
+	return ret;
+
+backoff:
+	drm_atomic_state_clear(state);
+	drm_atomic_legacy_backoff(state);
+
+	goto retry;
+}
+
 static int restore_fbdev_mode(struct drm_fb_helper *fb_helper)
 {
 	struct drm_device *dev = fb_helper->dev;
@@ -342,6 +410,9 @@ static int restore_fbdev_mode(struct drm_fb_helper *fb_helper)
 
 	drm_warn_on_modeset_not_all_locked(dev);
 
+	if (fb_helper->atomic)
+		return restore_fbdev_mode_atomic(fb_helper);
+
 	drm_for_each_plane(plane, dev) {
 		if (plane->type != DRM_PLANE_TYPE_PRIMARY)
 			drm_plane_force_disable(plane);
@@ -644,6 +715,8 @@ int drm_fb_helper_init(struct drm_device *dev,
 		i++;
 	}
 
+	fb_helper->atomic = !!drm_core_check_feature(dev, DRIVER_ATOMIC);
+
 	return 0;
 out_free:
 	drm_fb_helper_crtc_free(fb_helper);
diff --git a/include/drm/drm_atomic_helper.h b/include/drm/drm_atomic_helper.h
index 1547eb4..8cba54a 100644
--- a/include/drm/drm_atomic_helper.h
+++ b/include/drm/drm_atomic_helper.h
@@ -30,6 +30,8 @@
 
 #include <drm/drm_crtc.h>
 
+struct drm_atomic_state;
+
 int drm_atomic_helper_check_modeset(struct drm_device *dev,
 				struct drm_atomic_state *state);
 int drm_atomic_helper_check_planes(struct drm_device *dev,
@@ -73,7 +75,11 @@ int drm_atomic_helper_update_plane(struct drm_plane *plane,
 				   uint32_t src_x, uint32_t src_y,
 				   uint32_t src_w, uint32_t src_h);
 int drm_atomic_helper_disable_plane(struct drm_plane *plane);
+int __drm_atomic_helper_disable_plane(struct drm_plane *plane,
+		struct drm_plane_state *plane_state);
 int drm_atomic_helper_set_config(struct drm_mode_set *set);
+int __drm_atomic_helper_set_config(struct drm_mode_set *set,
+		struct drm_atomic_state *state);
 
 int drm_atomic_helper_crtc_set_property(struct drm_crtc *crtc,
 					struct drm_property *property,
diff --git a/include/drm/drm_fb_helper.h b/include/drm/drm_fb_helper.h
index 6254136..87b090c 100644
--- a/include/drm/drm_fb_helper.h
+++ b/include/drm/drm_fb_helper.h
@@ -134,6 +134,17 @@ struct drm_fb_helper {
 	/* we got a hotplug but fbdev wasn't running the console
 	   delay until next set_par */
 	bool delayed_hotplug;
+
+	/**
+	 * @atomic:
+	 *
+	 * Use atomic updates for restore_fbdev_mode(), etc.  This defaults to
+	 * true if driver has DRIVER_ATOMIC feature flag, but drivers can
+	 * override it to true after drm_fb_helper_init() if they support atomic
+	 * modeset but do not yet advertise DRIVER_ATOMIC (note that fb-helper
+	 * does not require ASYNC commits).
+	 */
+	bool atomic;
 };
 
 #ifdef CONFIG_DRM_FBDEV_EMULATION
-- 
cgit v0.10.2


From 1edf0269f02dc7e295f3ca7bdd698e3dcf7350bf Mon Sep 17 00:00:00 2001
From: Rob Clark <robdclark@gmail.com>
Date: Tue, 25 Aug 2015 15:35:59 -0400
Subject: drm/fb-helper: atomic pan_display()..

Signed-off-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c
index 0180fdd..64fc5ca 100644
--- a/drivers/gpu/drm/drm_fb_helper.c
+++ b/drivers/gpu/drm/drm_fb_helper.c
@@ -1217,6 +1217,57 @@ int drm_fb_helper_set_par(struct fb_info *info)
 }
 EXPORT_SYMBOL(drm_fb_helper_set_par);
 
+static int pan_display_atomic(struct fb_var_screeninfo *var,
+		struct fb_info *info)
+{
+	struct drm_fb_helper *fb_helper = info->par;
+	struct drm_device *dev = fb_helper->dev;
+	struct drm_atomic_state *state;
+	int i, ret;
+
+	state = drm_atomic_state_alloc(dev);
+	if (!state)
+		return -ENOMEM;
+
+	state->acquire_ctx = dev->mode_config.acquire_ctx;
+retry:
+	for(i = 0; i < fb_helper->crtc_count; i++) {
+		struct drm_mode_set *mode_set;
+
+		mode_set = &fb_helper->crtc_info[i].mode_set;
+
+		mode_set->x = var->xoffset;
+		mode_set->y = var->yoffset;
+
+		ret = __drm_atomic_helper_set_config(mode_set, state);
+		if (ret != 0)
+			goto fail;
+	}
+
+	ret = drm_atomic_commit(state);
+	if (ret != 0)
+		goto fail;
+
+	info->var.xoffset = var->xoffset;
+	info->var.yoffset = var->yoffset;
+
+	return 0;
+
+fail:
+	if (ret == -EDEADLK)
+		goto backoff;
+
+	drm_atomic_state_free(state);
+
+	return ret;
+
+backoff:
+	drm_atomic_state_clear(state);
+	drm_atomic_legacy_backoff(state);
+
+	goto retry;
+}
+
 /**
  * drm_fb_helper_pan_display - implementation for ->fb_pan_display
  * @var: updated screen information
@@ -1240,6 +1291,11 @@ int drm_fb_helper_pan_display(struct fb_var_screeninfo *var,
 		return -EBUSY;
 	}
 
+	if (fb_helper->atomic) {
+		ret = pan_display_atomic(var, info);
+		goto unlock;
+	}
+
 	for (i = 0; i < fb_helper->crtc_count; i++) {
 		modeset = &fb_helper->crtc_info[i].mode_set;
 
@@ -1254,6 +1310,7 @@ int drm_fb_helper_pan_display(struct fb_var_screeninfo *var,
 			}
 		}
 	}
+unlock:
 	drm_modeset_unlock_all(dev);
 	return ret;
 }
-- 
cgit v0.10.2


From 28cc504e8d52248962f5b485bdc65f539e3fe21d Mon Sep 17 00:00:00 2001
From: Rob Clark <robdclark@gmail.com>
Date: Tue, 25 Aug 2015 15:36:00 -0400
Subject: drm/i915: enable atomic fb-helper

i915 supports enough atomic to have atomic fb-helper paths, even though
it does not yet advertise DRIVER_ATOMIC.

Signed-off-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_fbdev.c b/drivers/gpu/drm/i915/intel_fbdev.c
index 8c6a6fa..ab2b856 100644
--- a/drivers/gpu/drm/i915/intel_fbdev.c
+++ b/drivers/gpu/drm/i915/intel_fbdev.c
@@ -689,6 +689,8 @@ int intel_fbdev_init(struct drm_device *dev)
 		return ret;
 	}
 
+	ifbdev->helper.atomic = true;
+
 	dev_priv->fbdev = ifbdev;
 	INIT_WORK(&dev_priv->fbdev_suspend_work, intel_fbdev_suspend_worker);
 
-- 
cgit v0.10.2


From d9072a3e80bd0f2e95d6a452987899039ba808ed Mon Sep 17 00:00:00 2001
From: Geliang Tang <geliangtang@163.com>
Date: Tue, 15 Sep 2015 05:58:44 -0700
Subject: drm/i915: fix kernel-doc warnings in i915_gem.c

Fix the following 'make htmldocs' warnings:

  .//drivers/gpu/drm/i915/i915_gem.c:1729: warning: No description found for parameter 'vma'
  .//drivers/gpu/drm/i915/i915_gem.c:1729: warning: No description found for parameter 'vmf'

  .//drivers/gpu/drm/i915/i915_gem.c:4962: warning: No description found for parameter 'old'
  .//drivers/gpu/drm/i915/i915_gem.c:4962: warning: No description found for parameter 'new'
  .//drivers/gpu/drm/i915/i915_gem.c:4962: warning: No description found for parameter 'frontbuffer_bits'

Signed-off-by: Geliang Tang <geliangtang@163.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>

diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index cb0df7e..e1b6457 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -1713,8 +1713,8 @@ i915_gem_mmap_ioctl(struct drm_device *dev, void *data,
 
 /**
  * i915_gem_fault - fault a page into the GTT
- * vma: VMA in question
- * vmf: fault info
+ * @vma: VMA in question
+ * @vmf: fault info
  *
  * The fault handler is set up by drm_gem_mmap() when a object is GTT mapped
  * from userspace.  The fault handler takes care of binding the object to
@@ -4985,9 +4985,9 @@ int i915_gem_open(struct drm_device *dev, struct drm_file *file)
 
 /**
  * i915_gem_track_fb - update frontbuffer tracking
- * old: current GEM buffer for the frontbuffer slots
- * new: new GEM buffer for the frontbuffer slots
- * frontbuffer_bits: bitmask of frontbuffer slots
+ * @old: current GEM buffer for the frontbuffer slots
+ * @new: new GEM buffer for the frontbuffer slots
+ * @frontbuffer_bits: bitmask of frontbuffer slots
  *
  * This updates the frontbuffer tracking bits @frontbuffer_bits by clearing them
  * from @old and setting them in @new. Both @old and @new can be NULL.
-- 
cgit v0.10.2


From eb120ef6a3e18780e7464e7de70221d06dfe5845 Mon Sep 17 00:00:00 2001
From: Jesse Barnes <jbarnes@virtuousgeek.org>
Date: Tue, 15 Sep 2015 14:19:32 -0700
Subject: drm/i915: add more debug info for when atomic updates fail v3
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

I used these additional fields to track down the issue I saw on HSW.

v2: move debug fields into a substruct (Ville)
v3: clean up debug code more (Ville)

References: https://bugs.freedesktop.org/show_bug.cgi?id=91579
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>

diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index e9e5cfd..8760814 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -565,8 +565,12 @@ struct intel_crtc {
 
 	int scanline_offset;
 
-	unsigned start_vbl_count;
-	ktime_t start_vbl_time;
+	struct {
+		unsigned start_vbl_count;
+		ktime_t start_vbl_time;
+		int min_vbl, max_vbl;
+		int scanline_start;
+	} debug;
 
 	struct intel_crtc_atomic_commit atomic;
 
diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c
index 4d27243..7de121d 100644
--- a/drivers/gpu/drm/i915/intel_sprite.c
+++ b/drivers/gpu/drm/i915/intel_sprite.c
@@ -95,7 +95,6 @@ void intel_pipe_update_start(struct intel_crtc *crtc)
 	max = vblank_start - 1;
 
 	local_irq_disable();
-	crtc->start_vbl_count = 0;
 
 	if (min <= 0 || max <= 0)
 		return;
@@ -134,11 +133,15 @@ void intel_pipe_update_start(struct intel_crtc *crtc)
 
 	drm_crtc_vblank_put(&crtc->base);
 
-	crtc->start_vbl_time = ktime_get();
-	crtc->start_vbl_count = dev->driver->get_vblank_counter(dev, pipe);
+	crtc->debug.min_vbl = min;
+	crtc->debug.max_vbl = max;
+	crtc->debug.scanline_start = scanline;
+	crtc->debug.start_vbl_time = ktime_get();
+	crtc->debug.start_vbl_count =
+		dev->driver->get_vblank_counter(dev, pipe);
 
 	trace_i915_pipe_update_vblank_evaded(crtc, min, max,
-					     crtc->start_vbl_count);
+					     crtc->debug.start_vbl_count);
 }
 
 /**
@@ -154,6 +157,7 @@ void intel_pipe_update_end(struct intel_crtc *crtc)
 {
 	struct drm_device *dev = crtc->base.dev;
 	enum pipe pipe = crtc->pipe;
+	int scanline_end = intel_get_crtc_scanline(crtc);
 	u32 end_vbl_count = dev->driver->get_vblank_counter(dev, pipe);
 	ktime_t end_vbl_time = ktime_get();
 
@@ -161,10 +165,15 @@ void intel_pipe_update_end(struct intel_crtc *crtc)
 
 	local_irq_enable();
 
-	if (crtc->start_vbl_count && crtc->start_vbl_count != end_vbl_count)
-		DRM_ERROR("Atomic update failure on pipe %c (start=%u end=%u) time %lld us\n",
-			  pipe_name(pipe), crtc->start_vbl_count, end_vbl_count,
-			  ktime_us_delta(end_vbl_time, crtc->start_vbl_time));
+	if (crtc->debug.start_vbl_count &&
+	    crtc->debug.start_vbl_count != end_vbl_count) {
+		DRM_ERROR("Atomic update failure on pipe %c (start=%u end=%u) time %lld us, min %d, max %d, scanline start %d, end %d\n",
+			  pipe_name(pipe), crtc->debug.start_vbl_count,
+			  end_vbl_count,
+			  ktime_us_delta(end_vbl_time, crtc->debug.start_vbl_time),
+			  crtc->debug.min_vbl, crtc->debug.max_vbl,
+			  crtc->debug.scanline_start, scanline_end);
+	}
 }
 
 static void
-- 
cgit v0.10.2


From d637ce3f6dcc0bec950ed4e03e53162500d6b28b Mon Sep 17 00:00:00 2001
From: Jesse Barnes <jbarnes@virtuousgeek.org>
Date: Thu, 17 Sep 2015 08:08:32 -0700
Subject: drm/i915: cleanup pipe_update trace functions with new crtc debug
 info v3
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Use the new debug info in the intel_crtc struct in these functions
rather than passing them as args.

v2: move min/max assignment back above first trace call (Ville)
    use scanline from crtc->debug rather than fetching a new one (Ville)
v3: fix up trace_i915_pipe_update_end, needs end scanline (Ville)

Requested-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>

diff --git a/drivers/gpu/drm/i915/i915_trace.h b/drivers/gpu/drm/i915/i915_trace.h
index e6b5c74..d0993bc 100644
--- a/drivers/gpu/drm/i915/i915_trace.h
+++ b/drivers/gpu/drm/i915/i915_trace.h
@@ -17,8 +17,8 @@
 /* pipe updates */
 
 TRACE_EVENT(i915_pipe_update_start,
-	    TP_PROTO(struct intel_crtc *crtc, u32 min, u32 max),
-	    TP_ARGS(crtc, min, max),
+	    TP_PROTO(struct intel_crtc *crtc),
+	    TP_ARGS(crtc),
 
 	    TP_STRUCT__entry(
 			     __field(enum pipe, pipe)
@@ -33,8 +33,8 @@ TRACE_EVENT(i915_pipe_update_start,
 			   __entry->frame = crtc->base.dev->driver->get_vblank_counter(crtc->base.dev,
 										       crtc->pipe);
 			   __entry->scanline = intel_get_crtc_scanline(crtc);
-			   __entry->min = min;
-			   __entry->max = max;
+			   __entry->min = crtc->debug.min_vbl;
+			   __entry->max = crtc->debug.max_vbl;
 			   ),
 
 	    TP_printk("pipe %c, frame=%u, scanline=%u, min=%u, max=%u",
@@ -43,8 +43,8 @@ TRACE_EVENT(i915_pipe_update_start,
 );
 
 TRACE_EVENT(i915_pipe_update_vblank_evaded,
-	    TP_PROTO(struct intel_crtc *crtc, u32 min, u32 max, u32 frame),
-	    TP_ARGS(crtc, min, max, frame),
+	    TP_PROTO(struct intel_crtc *crtc),
+	    TP_ARGS(crtc),
 
 	    TP_STRUCT__entry(
 			     __field(enum pipe, pipe)
@@ -56,10 +56,10 @@ TRACE_EVENT(i915_pipe_update_vblank_evaded,
 
 	    TP_fast_assign(
 			   __entry->pipe = crtc->pipe;
-			   __entry->frame = frame;
-			   __entry->scanline = intel_get_crtc_scanline(crtc);
-			   __entry->min = min;
-			   __entry->max = max;
+			   __entry->frame = crtc->debug.start_vbl_count;
+			   __entry->scanline = crtc->debug.scanline_start;
+			   __entry->min = crtc->debug.min_vbl;
+			   __entry->max = crtc->debug.max_vbl;
 			   ),
 
 	    TP_printk("pipe %c, frame=%u, scanline=%u, min=%u, max=%u",
@@ -68,8 +68,8 @@ TRACE_EVENT(i915_pipe_update_vblank_evaded,
 );
 
 TRACE_EVENT(i915_pipe_update_end,
-	    TP_PROTO(struct intel_crtc *crtc, u32 frame),
-	    TP_ARGS(crtc, frame),
+	    TP_PROTO(struct intel_crtc *crtc, u32 frame, int scanline_end),
+	    TP_ARGS(crtc, frame, scanline_end),
 
 	    TP_STRUCT__entry(
 			     __field(enum pipe, pipe)
@@ -80,7 +80,7 @@ TRACE_EVENT(i915_pipe_update_end,
 	    TP_fast_assign(
 			   __entry->pipe = crtc->pipe;
 			   __entry->frame = frame;
-			   __entry->scanline = intel_get_crtc_scanline(crtc);
+			   __entry->scanline = scanline_end;
 			   ),
 
 	    TP_printk("pipe %c, frame=%u, scanline=%u",
diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c
index 7de121d..79f7cc2 100644
--- a/drivers/gpu/drm/i915/intel_sprite.c
+++ b/drivers/gpu/drm/i915/intel_sprite.c
@@ -102,7 +102,9 @@ void intel_pipe_update_start(struct intel_crtc *crtc)
 	if (WARN_ON(drm_crtc_vblank_get(&crtc->base)))
 		return;
 
-	trace_i915_pipe_update_start(crtc, min, max);
+	crtc->debug.min_vbl = min;
+	crtc->debug.max_vbl = max;
+	trace_i915_pipe_update_start(crtc);
 
 	for (;;) {
 		/*
@@ -133,15 +135,12 @@ void intel_pipe_update_start(struct intel_crtc *crtc)
 
 	drm_crtc_vblank_put(&crtc->base);
 
-	crtc->debug.min_vbl = min;
-	crtc->debug.max_vbl = max;
 	crtc->debug.scanline_start = scanline;
 	crtc->debug.start_vbl_time = ktime_get();
 	crtc->debug.start_vbl_count =
 		dev->driver->get_vblank_counter(dev, pipe);
 
-	trace_i915_pipe_update_vblank_evaded(crtc, min, max,
-					     crtc->debug.start_vbl_count);
+	trace_i915_pipe_update_vblank_evaded(crtc);
 }
 
 /**
@@ -161,7 +160,7 @@ void intel_pipe_update_end(struct intel_crtc *crtc)
 	u32 end_vbl_count = dev->driver->get_vblank_counter(dev, pipe);
 	ktime_t end_vbl_time = ktime_get();
 
-	trace_i915_pipe_update_end(crtc, end_vbl_count);
+	trace_i915_pipe_update_end(crtc, end_vbl_count, scanline_end);
 
 	local_irq_enable();
 
-- 
cgit v0.10.2


From 9aa6114253a9d01513796f6b44eda9838dc7bf3b Mon Sep 17 00:00:00 2001
From: Matt Roper <matthew.d.roper@intel.com>
Date: Mon, 14 Sep 2015 19:24:18 -0700
Subject: drm/i915: Don't leak VBT mode data

We allocate memory for LVDS modes while parsing the VBT at startup, but
never free this memory when the driver is unloaded, causing a small
leak.

Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>

diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index 066a0ef..c069550 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -1137,6 +1137,10 @@ int i915_driver_unload(struct drm_device *dev)
 		dev_priv->vbt.child_dev = NULL;
 		dev_priv->vbt.child_dev_num = 0;
 	}
+	kfree(dev_priv->vbt.sdvo_lvds_vbt_mode);
+	dev_priv->vbt.sdvo_lvds_vbt_mode = NULL;
+	kfree(dev_priv->vbt.lfp_lvds_vbt_mode);
+	dev_priv->vbt.lfp_lvds_vbt_mode = NULL;
 
 	vga_switcheroo_unregister_client(dev->pdev);
 	vga_client_register(dev->pdev, NULL, NULL, NULL);
-- 
cgit v0.10.2


From b5dada82ad624b69a2229c819d6a30b664a5606f Mon Sep 17 00:00:00 2001
From: Damien Lespiau <damien.lespiau@intel.com>
Date: Thu, 17 Sep 2015 14:20:32 +0100
Subject: drm/i915/bxt: Fix wrongly placed ')' in I915_READ()

Not the first time! not the last time?

There is a possibility to use gcc 5's -Wbool-compare to try and compare
(reg) in those macros to a constant and gcc will warn that the
comparison between a boolean expression and a constant is always either
true or false. Maybe.

Cc: Imre Deak <imre.deak@intel.com>
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>

diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
index 4823184..5b600bf 100644
--- a/drivers/gpu/drm/i915/intel_ddi.c
+++ b/drivers/gpu/drm/i915/intel_ddi.c
@@ -2882,7 +2882,7 @@ static bool bxt_ddi_pll_get_hw_state(struct drm_i915_private *dev_priv,
 	 * here just read out lanes 0/1 and output a note if lanes 2/3 differ.
 	 */
 	hw_state->pcsdw12 = I915_READ(BXT_PORT_PCS_DW12_LN01(port));
-	if (I915_READ(BXT_PORT_PCS_DW12_LN23(port) != hw_state->pcsdw12))
+	if (I915_READ(BXT_PORT_PCS_DW12_LN23(port)) != hw_state->pcsdw12)
 		DRM_DEBUG_DRIVER("lane stagger config different for lane 01 (%08x) and 23 (%08x)\n",
 				 hw_state->pcsdw12,
 				 I915_READ(BXT_PORT_PCS_DW12_LN23(port)));
-- 
cgit v0.10.2


From a78695d3ea045cd9977a8397c89171d762708f86 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 18 Sep 2015 15:54:50 +0300
Subject: drm/i915/skl: handle port E in cpt_digital_port_connected

SKL port E handling was added in

commit 26951caf55d73ceb1967b0bf12f6d0b96853508e
Author: Xiong Zhang <xiong.y.zhang@intel.com>
Date:   Mon Aug 17 15:55:50 2015 +0800

    drm/i915/skl: enable DDI-E hotplug

but the whole function was moved in a another branch in

commit b93433ccf64846820b9448f5ff5dd4348b58a8ed
Author: Jani Nikula <jani.nikula@intel.com>
Date:   Thu Aug 20 10:47:36 2015 +0300

    drm/i915: move ibx_digital_port_connected to intel_dp.c

and the addition was lost at some backmerge that I was unable to
identify. Put it back in.

Tested-by: Tomi Sarvela <tomix.p.sarvela@intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index b8c4fc5..9700f77 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -4582,6 +4582,9 @@ static bool cpt_digital_port_connected(struct drm_i915_private *dev_priv,
 	case PORT_D:
 		bit = SDE_PORTD_HOTPLUG_CPT;
 		break;
+	case PORT_E:
+		bit = SDE_PORTE_HOTPLUG_SPT;
+		break;
 	default:
 		MISSING_CASE(port->port);
 		return false;
-- 
cgit v0.10.2


From a645654b817feba05e5156345325d19fc85ebc9f Mon Sep 17 00:00:00 2001
From: Lukas Wunner <lukas@wunner.de>
Date: Sun, 23 Aug 2015 15:18:55 +0200
Subject: vga_switcheroo: Document _ALL_ the things!

This adds an "Overview" DOC section plus two DOC sections for the modes
of use ("Manual switching and manual power control" and "Driver power
control").

Also included is kernel-doc for all public functions, structs and enums.

Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/vga/vga_switcheroo.c b/drivers/gpu/vga/vga_switcheroo.c
index 2106066..b19a72f 100644
--- a/drivers/gpu/vga/vga_switcheroo.c
+++ b/drivers/gpu/vga/vga_switcheroo.c
@@ -1,20 +1,31 @@
 /*
+ * vga_switcheroo.c - Support for laptop with dual GPU using one set of outputs
+ *
  * Copyright (c) 2010 Red Hat Inc.
  * Author : Dave Airlie <airlied@redhat.com>
  *
+ * Copyright (c) 2015 Lukas Wunner <lukas@wunner.de>
  *
- * Licensed under GPLv2
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
  *
- * vga_switcheroo.c - Support for laptop with dual GPU using one set of outputs
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
  *
- * Switcher interface - methods require for ATPX and DCM
- * - switchto - this throws the output MUX switch
- * - discrete_set_power - sets the power state for the discrete card
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS
+ * IN THE SOFTWARE.
  *
- * GPU driver interface
- * - set_gpu_state - this should do the equiv of s/r for the card
- *                 - this should *not* set the discrete power state
- * - switch_check  - check if the device is in a position to switch now
  */
 
 #define pr_fmt(fmt) "vga_switcheroo: " fmt
@@ -33,6 +44,61 @@
 
 #include <linux/vgaarb.h>
 
+/**
+ * DOC: Overview
+ *
+ * vga_switcheroo is the Linux subsystem for laptop hybrid graphics.
+ * These come in two flavors:
+ *
+ * * muxed: Dual GPUs with a multiplexer chip to switch outputs between GPUs.
+ * * muxless: Dual GPUs but only one of them is connected to outputs.
+ * 	The other one is merely used to offload rendering, its results
+ * 	are copied over PCIe into the framebuffer. On Linux this is
+ * 	supported with DRI PRIME.
+ *
+ * Hybrid graphics started to appear in the late Naughties and were initially
+ * all muxed. Newer laptops moved to a muxless architecture for cost reasons.
+ * A notable exception is the MacBook Pro which continues to use a mux.
+ * Muxes come with varying capabilities: Some switch only the panel, others
+ * can also switch external displays. Some switch all display pins at once
+ * while others can switch just the DDC lines. (To allow EDID probing
+ * for the inactive GPU.) Also, muxes are often used to cut power to the
+ * discrete GPU while it is not used.
+ *
+ * DRM drivers register GPUs with vga_switcheroo, these are heretoforth called
+ * clients. The mux is called the handler. Muxless machines also register a
+ * handler to control the power state of the discrete GPU, its ->switchto
+ * callback is a no-op for obvious reasons. The discrete GPU is often equipped
+ * with an HDA controller for the HDMI/DP audio signal, this will also
+ * register as a client so that vga_switcheroo can take care of the correct
+ * suspend/resume order when changing the discrete GPU's power state. In total
+ * there can thus be up to three clients: Two vga clients (GPUs) and one audio
+ * client (on the discrete GPU). The code is mostly prepared to support
+ * machines with more than two GPUs should they become available.
+ * The GPU to which the outputs are currently switched is called the
+ * active client in vga_switcheroo parlance. The GPU not in use is the
+ * inactive client.
+ */
+
+/**
+ * struct vga_switcheroo_client - registered client
+ * @pdev: client pci device
+ * @fb_info: framebuffer to which console is remapped on switching
+ * @pwr_state: current power state
+ * @ops: client callbacks
+ * @id: client identifier, see enum vga_switcheroo_client_id.
+ * 	Determining the id requires the handler, so GPUs are initially
+ * 	assigned -1 and later given their true id in vga_switcheroo_enable()
+ * @active: whether the outputs are currently switched to this client
+ * @driver_power_control: whether power state is controlled by the driver's
+ * 	runtime pm. If true, writing ON and OFF to the vga_switcheroo debugfs
+ * 	interface is a no-op so as not to interfere with runtime pm
+ * @list: client list
+ *
+ * Registered client. A client can be either a GPU or an audio device on a GPU.
+ * For audio clients, the @fb_info, @active and @driver_power_control members
+ * are bogus.
+ */
 struct vga_switcheroo_client {
 	struct pci_dev *pdev;
 	struct fb_info *fb_info;
@@ -44,10 +110,28 @@ struct vga_switcheroo_client {
 	struct list_head list;
 };
 
+/*
+ * protects access to struct vgasr_priv
+ */
 static DEFINE_MUTEX(vgasr_mutex);
 
+/**
+ * struct vgasr_priv - vga_switcheroo private data
+ * @active: whether vga_switcheroo is enabled.
+ * 	Prerequisite is the registration of two GPUs and a handler
+ * @delayed_switch_active: whether a delayed switch is pending
+ * @delayed_client_id: client to which a delayed switch is pending
+ * @debugfs_root: directory for vga_switcheroo debugfs interface
+ * @switch_file: file for vga_switcheroo debugfs interface
+ * @registered_clients: number of registered GPUs
+ * 	(counting only vga clients, not audio clients)
+ * @clients: list of registered clients
+ * @handler: registered handler
+ *
+ * vga_switcheroo private data. Currently only one vga_switcheroo instance
+ * per system is supported.
+ */
 struct vgasr_priv {
-
 	bool active;
 	bool delayed_switch_active;
 	enum vga_switcheroo_client_id delayed_client_id;
@@ -103,6 +187,15 @@ static void vga_switcheroo_enable(void)
 	vgasr_priv.active = true;
 }
 
+/**
+ * vga_switcheroo_register_handler() - register handler
+ * @handler: handler callbacks
+ *
+ * Register handler. Enable vga_switcheroo if two vga clients have already
+ * registered.
+ *
+ * Return: 0 on success, -EINVAL if a handler was already registered.
+ */
 int vga_switcheroo_register_handler(struct vga_switcheroo_handler *handler)
 {
 	mutex_lock(&vgasr_mutex);
@@ -121,6 +214,11 @@ int vga_switcheroo_register_handler(struct vga_switcheroo_handler *handler)
 }
 EXPORT_SYMBOL(vga_switcheroo_register_handler);
 
+/**
+ * vga_switcheroo_unregister_handler() - unregister handler
+ *
+ * Unregister handler. Disable vga_switcheroo.
+ */
 void vga_switcheroo_unregister_handler(void)
 {
 	mutex_lock(&vgasr_mutex);
@@ -164,6 +262,19 @@ static int register_client(struct pci_dev *pdev,
 	return 0;
 }
 
+/**
+ * vga_switcheroo_register_client - register vga client
+ * @pdev: client pci device
+ * @ops: client callbacks
+ * @driver_power_control: whether power state is controlled by the driver's
+ * 	runtime pm
+ *
+ * Register vga client (GPU). Enable vga_switcheroo if another GPU and a
+ * handler have already registered. The power state of the client is assumed
+ * to be ON.
+ *
+ * Return: 0 on success, -ENOMEM on memory allocation error.
+ */
 int vga_switcheroo_register_client(struct pci_dev *pdev,
 				   const struct vga_switcheroo_client_ops *ops,
 				   bool driver_power_control)
@@ -174,6 +285,18 @@ int vga_switcheroo_register_client(struct pci_dev *pdev,
 }
 EXPORT_SYMBOL(vga_switcheroo_register_client);
 
+/**
+ * vga_switcheroo_register_audio_client - register audio client
+ * @pdev: client pci device
+ * @ops: client callbacks
+ * @id: client identifier, see enum vga_switcheroo_client_id
+ * @active: whether the audio device is fully initialized
+ *
+ * Register audio client (audio device on a GPU). The power state of the
+ * client is assumed to be ON.
+ *
+ * Return: 0 on success, -ENOMEM on memory allocation error.
+ */
 int vga_switcheroo_register_audio_client(struct pci_dev *pdev,
 					 const struct vga_switcheroo_client_ops *ops,
 					 int id, bool active)
@@ -215,6 +338,15 @@ find_active_client(struct list_head *head)
 	return NULL;
 }
 
+/**
+ * vga_switcheroo_get_client_state() - obtain power state of a given client
+ * @pdev: client pci device
+ *
+ * Obtain power state of a given client as seen from vga_switcheroo.
+ * The function is only called from hda_intel.c.
+ *
+ * Return: Power state.
+ */
 int vga_switcheroo_get_client_state(struct pci_dev *pdev)
 {
 	struct vga_switcheroo_client *client;
@@ -228,6 +360,12 @@ int vga_switcheroo_get_client_state(struct pci_dev *pdev)
 }
 EXPORT_SYMBOL(vga_switcheroo_get_client_state);
 
+/**
+ * vga_switcheroo_unregister_client() - unregister client
+ * @pdev: client pci device
+ *
+ * Unregister client. Disable vga_switcheroo if this is a vga client (GPU).
+ */
 void vga_switcheroo_unregister_client(struct pci_dev *pdev)
 {
 	struct vga_switcheroo_client *client;
@@ -249,6 +387,14 @@ void vga_switcheroo_unregister_client(struct pci_dev *pdev)
 }
 EXPORT_SYMBOL(vga_switcheroo_unregister_client);
 
+/**
+ * vga_switcheroo_client_fb_set() - set framebuffer of a given client
+ * @pdev: client pci device
+ * @info: framebuffer
+ *
+ * Set framebuffer of a given client. The console will be remapped to this
+ * on switching.
+ */
 void vga_switcheroo_client_fb_set(struct pci_dev *pdev,
 				 struct fb_info *info)
 {
@@ -262,6 +408,42 @@ void vga_switcheroo_client_fb_set(struct pci_dev *pdev,
 }
 EXPORT_SYMBOL(vga_switcheroo_client_fb_set);
 
+/**
+ * DOC: Manual switching and manual power control
+ *
+ * In this mode of use, the file /sys/kernel/debug/vgaswitcheroo/switch
+ * can be read to retrieve the current vga_switcheroo state and commands
+ * can be written to it to change the state. The file appears as soon as
+ * two GPU drivers and one handler have registered with vga_switcheroo.
+ * The following commands are understood:
+ *
+ * * OFF: Power off the device not in use.
+ * * ON: Power on the device not in use.
+ * * IGD: Switch to the integrated graphics device.
+ * 	Power on the integrated GPU if necessary, power off the discrete GPU.
+ * 	Prerequisite is that no user space processes (e.g. Xorg, alsactl)
+ * 	have opened device files of the GPUs or the audio client. If the
+ * 	switch fails, the user may invoke lsof(8) or fuser(1) on /dev/dri/
+ * 	and /dev/snd/controlC1 to identify processes blocking the switch.
+ * * DIS: Switch to the discrete graphics device.
+ * * DIGD: Delayed switch to the integrated graphics device.
+ * 	This will perform the switch once the last user space process has
+ * 	closed the device files of the GPUs and the audio client.
+ * * DDIS: Delayed switch to the discrete graphics device.
+ * * MIGD: Mux-only switch to the integrated graphics device.
+ * 	Does not remap console or change the power state of either gpu.
+ * 	If the integrated GPU is currently off, the screen will turn black.
+ * 	If it is on, the screen will show whatever happens to be in VRAM.
+ * 	Either way, the user has to blindly enter the command to switch back.
+ * * MDIS: Mux-only switch to the discrete graphics device.
+ *
+ * For GPUs whose power state is controlled by the driver's runtime pm,
+ * the ON and OFF commands are a no-op (see next section).
+ *
+ * For muxless machines, the IGD/DIS, DIGD/DDIS and MIGD/MDIS commands
+ * should not be used.
+ */
+
 static int vga_switcheroo_show(struct seq_file *m, void *v)
 {
 	struct vga_switcheroo_client *client;
@@ -559,6 +741,16 @@ fail:
 	return -1;
 }
 
+/**
+ * vga_switcheroo_process_delayed_switch() - helper for delayed switching
+ *
+ * Process a delayed switch if one is pending. DRM drivers should call this
+ * from their ->lastclose callback.
+ *
+ * Return: 0 on success. -EINVAL if no delayed switch is pending, if the client
+ * has unregistered in the meantime or if there are other clients blocking the
+ * switch. If the actual switch fails, an error is reported and 0 is returned.
+ */
 int vga_switcheroo_process_delayed_switch(void)
 {
 	struct vga_switcheroo_client *client;
@@ -589,6 +781,39 @@ err:
 }
 EXPORT_SYMBOL(vga_switcheroo_process_delayed_switch);
 
+/**
+ * DOC: Driver power control
+ *
+ * In this mode of use, the discrete GPU automatically powers up and down at
+ * the discretion of the driver's runtime pm. On muxed machines, the user may
+ * still influence the muxer state by way of the debugfs interface, however
+ * the ON and OFF commands become a no-op for the discrete GPU.
+ *
+ * This mode is the default on Nvidia HybridPower/Optimus and ATI PowerXpress.
+ * Specifying nouveau.runpm=0, radeon.runpm=0 or amdgpu.runpm=0 on the kernel
+ * command line disables it.
+ *
+ * When the driver decides to power up or down, it notifies vga_switcheroo
+ * thereof so that it can (a) power the audio device on the GPU up or down,
+ * and (b) update its internal power state representation for the device.
+ * This is achieved by vga_switcheroo_set_dynamic_switch().
+ *
+ * After the GPU has been suspended, the handler needs to be called to cut
+ * power to the GPU. Likewise it needs to reinstate power before the GPU
+ * can resume. This is achieved by vga_switcheroo_init_domain_pm_ops(),
+ * which augments the GPU's suspend/resume functions by the requisite
+ * calls to the handler.
+ *
+ * When the audio device resumes, the GPU needs to be woken. This is achieved
+ * by vga_switcheroo_init_domain_pm_optimus_hdmi_audio(), which augments the
+ * audio device's resume function.
+ *
+ * On muxed machines, if the mux is initially switched to the discrete GPU,
+ * the user ends up with a black screen when the GPU powers down after boot.
+ * As a workaround, the mux is forced to the integrated GPU on runtime suspend,
+ * cf. https://bugs.freedesktop.org/show_bug.cgi?id=75917
+ */
+
 static void vga_switcheroo_power_switch(struct pci_dev *pdev,
 					enum vga_switcheroo_state state)
 {
@@ -607,8 +832,17 @@ static void vga_switcheroo_power_switch(struct pci_dev *pdev,
 	vgasr_priv.handler->power_state(client->id, state);
 }
 
-/* force a PCI device to a certain state - mainly to turn off audio clients */
-
+/**
+ * vga_switcheroo_set_dynamic_switch() - helper for driver power control
+ * @pdev: client pci device
+ * @dynamic: new power state
+ *
+ * Helper for GPUs whose power state is controlled by the driver's runtime pm.
+ * When the driver decides to power up or down, it notifies vga_switcheroo
+ * thereof using this helper so that it can (a) power the audio device on
+ * the GPU up or down, and (b) update its internal power state representation
+ * for the device.
+ */
 void vga_switcheroo_set_dynamic_switch(struct pci_dev *pdev,
 				       enum vga_switcheroo_state dynamic)
 {
@@ -654,8 +888,18 @@ static int vga_switcheroo_runtime_resume(struct device *dev)
 	return 0;
 }
 
-/* this version is for the case where the power switch is separate
-   to the device being powered down. */
+/**
+ * vga_switcheroo_init_domain_pm_ops() - helper for driver power control
+ * @dev: vga client device
+ * @domain: power domain
+ *
+ * Helper for GPUs whose power state is controlled by the driver's runtime pm.
+ * After the GPU has been suspended, the handler needs to be called to cut
+ * power to the GPU. Likewise it needs to reinstate power before the GPU
+ * can resume. To this end, this helper augments the suspend/resume functions
+ * by the requisite calls to the handler. It needs only be called on platforms
+ * where the power switch is separate to the device being powered down.
+ */
 int vga_switcheroo_init_domain_pm_ops(struct device *dev,
 				      struct dev_pm_domain *domain)
 {
@@ -709,6 +953,19 @@ static int vga_switcheroo_runtime_resume_hdmi_audio(struct device *dev)
 	return ret;
 }
 
+/**
+ * vga_switcheroo_init_domain_pm_optimus_hdmi_audio() - helper for driver
+ * 	power control
+ * @dev: audio client device
+ * @domain: power domain
+ *
+ * Helper for GPUs whose power state is controlled by the driver's runtime pm.
+ * When the audio device resumes, the GPU needs to be woken. This helper
+ * augments the audio device's resume function to do that.
+ *
+ * Return: 0 on success, -EINVAL if no power management operations are
+ * defined for this device.
+ */
 int
 vga_switcheroo_init_domain_pm_optimus_hdmi_audio(struct device *dev,
 						 struct dev_pm_domain *domain)
diff --git a/include/linux/vga_switcheroo.h b/include/linux/vga_switcheroo.h
index b483abd..fe90bfc 100644
--- a/include/linux/vga_switcheroo.h
+++ b/include/linux/vga_switcheroo.h
@@ -1,10 +1,31 @@
 /*
+ * vga_switcheroo.h - Support for laptop with dual GPU using one set of outputs
+ *
  * Copyright (c) 2010 Red Hat Inc.
  * Author : Dave Airlie <airlied@redhat.com>
  *
- * Licensed under GPLv2
+ * Copyright (c) 2015 Lukas Wunner <lukas@wunner.de>
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS
+ * IN THE SOFTWARE.
  *
- * vga_switcheroo.h - Support for laptop with dual GPU using one set of outputs
  */
 
 #ifndef _LINUX_VGA_SWITCHEROO_H_
@@ -14,6 +35,20 @@
 
 struct pci_dev;
 
+/**
+ * enum vga_switcheroo_state - client power state
+ * @VGA_SWITCHEROO_OFF: off
+ * @VGA_SWITCHEROO_ON: on
+ * @VGA_SWITCHEROO_INIT: client has registered with vga_switcheroo but
+ * 	vga_switcheroo is not enabled, i.e. no second client or no handler
+ * 	has registered. Only used in vga_switcheroo_get_client_state() which
+ * 	in turn is only called from hda_intel.c
+ * @VGA_SWITCHEROO_NOT_FOUND: client has not registered with vga_switcheroo.
+ * 	Only used in vga_switcheroo_get_client_state() which in turn is only
+ * 	called from hda_intel.c
+ *
+ * Client power state.
+ */
 enum vga_switcheroo_state {
 	VGA_SWITCHEROO_OFF,
 	VGA_SWITCHEROO_ON,
@@ -22,20 +57,64 @@ enum vga_switcheroo_state {
 	VGA_SWITCHEROO_NOT_FOUND,
 };
 
+/**
+ * enum vga_switcheroo_client_id - client identifier
+ * @VGA_SWITCHEROO_IGD: integrated graphics device
+ * @VGA_SWITCHEROO_DIS: discrete graphics device
+ * @VGA_SWITCHEROO_MAX_CLIENTS: currently no more than two GPUs are supported
+ *
+ * Client identifier. Audio clients use the same identifier & 0x100.
+ */
 enum vga_switcheroo_client_id {
 	VGA_SWITCHEROO_IGD,
 	VGA_SWITCHEROO_DIS,
 	VGA_SWITCHEROO_MAX_CLIENTS,
 };
 
+/**
+ * struct vga_switcheroo_handler - handler callbacks
+ * @init: initialize handler.
+ * 	Optional. This gets called when vga_switcheroo is enabled, i.e. when
+ * 	two vga clients have registered. It allows the handler to perform
+ * 	some delayed initialization that depends on the existence of the
+ * 	vga clients. Currently only the radeon and amdgpu drivers use this.
+ * 	The return value is ignored
+ * @switchto: switch outputs to given client.
+ * 	Mandatory. For muxless machines this should be a no-op. Returning 0
+ * 	denotes success, anything else failure (in which case the switch is
+ * 	aborted)
+ * @power_state: cut or reinstate power of given client.
+ * 	Optional. The return value is ignored
+ * @get_client_id: determine if given pci device is integrated or discrete GPU.
+ * 	Mandatory
+ *
+ * Handler callbacks. The multiplexer itself. The @switchto and @get_client_id
+ * methods are mandatory, all others may be set to NULL.
+ */
 struct vga_switcheroo_handler {
+	int (*init)(void);
 	int (*switchto)(enum vga_switcheroo_client_id id);
 	int (*power_state)(enum vga_switcheroo_client_id id,
 			   enum vga_switcheroo_state state);
-	int (*init)(void);
 	int (*get_client_id)(struct pci_dev *pdev);
 };
 
+/**
+ * struct vga_switcheroo_client_ops - client callbacks
+ * @set_gpu_state: do the equivalent of suspend/resume for the card.
+ * 	Mandatory. This should not cut power to the discrete GPU,
+ * 	which is the job of the handler
+ * @reprobe: poll outputs.
+ * 	Optional. This gets called after waking the GPU and switching
+ * 	the outputs to it
+ * @can_switch: check if the device is in a position to switch now.
+ * 	Mandatory. The client should return false if a user space process
+ * 	has one of its device files open
+ *
+ * Client callbacks. A client can be either a GPU or an audio device on a GPU.
+ * The @set_gpu_state and @can_switch methods are mandatory, @reprobe may be
+ * set to NULL. For audio clients, the @reprobe member is bogus.
+ */
 struct vga_switcheroo_client_ops {
 	void (*set_gpu_state)(struct pci_dev *dev, enum vga_switcheroo_state);
 	void (*reprobe)(struct pci_dev *dev);
-- 
cgit v0.10.2


From 5389e916c3b64c3dd249018dfaf983e2e3a06c6f Mon Sep 17 00:00:00 2001
From: Lukas Wunner <lukas@wunner.de>
Date: Fri, 4 Sep 2015 21:06:15 +0200
Subject: drm/i915: Spell vga_switcheroo consistently

Currently everyone and their dog has their own favourite spelling
for vga_switcheroo. This makes it hard to grep dmesg for log entries
relating to vga_switcheroo. It also makes it hard to find related
source files in the tree.

vga_switcheroo.c uses pr_fmt "vga_switcheroo". Use that everywhere.

Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_acpi.c b/drivers/gpu/drm/i915/intel_acpi.c
index d96eee1..8b13b9d 100644
--- a/drivers/gpu/drm/i915/intel_acpi.c
+++ b/drivers/gpu/drm/i915/intel_acpi.c
@@ -146,7 +146,7 @@ static bool intel_dsm_detect(void)
 
 	if (vga_count == 2 && has_dsm) {
 		acpi_get_name(intel_dsm_priv.dhandle, ACPI_FULL_PATHNAME, &buffer);
-		DRM_DEBUG_DRIVER("VGA switcheroo: detected DSM switching method %s handle\n",
+		DRM_DEBUG_DRIVER("vga_switcheroo: detected DSM switching method %s handle\n",
 				 acpi_method_name);
 		return true;
 	}
diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c
index 2034438a..2c11b4e 100644
--- a/drivers/gpu/drm/i915/intel_panel.c
+++ b/drivers/gpu/drm/i915/intel_panel.c
@@ -821,7 +821,7 @@ void intel_panel_disable_backlight(struct intel_connector *connector)
 		return;
 
 	/*
-	 * Do not disable backlight on the vgaswitcheroo path. When switching
+	 * Do not disable backlight on the vga_switcheroo path. When switching
 	 * away from i915, the other client may depend on i915 to handle the
 	 * backlight. This will leave the backlight on unnecessarily when
 	 * another client is not activated.
-- 
cgit v0.10.2


From 604ef7346862bb761a091932e99c14204ff4050a Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <a.hajda@samsung.com>
Date: Mon, 21 Sep 2015 15:33:35 +0200
Subject: drm/i915: fix handling gen8_emit_flush_coherentl3_wa result

The function can return negative value.

The problem has been detected using proposed semantic patch
scripts/coccinelle/tests/unsigned_lesser_than_zero.cocci [1].

[1]: http://permalink.gmane.org/gmane.linux.kernel/2038576

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
index fe06accb0..ff9a481 100644
--- a/drivers/gpu/drm/i915/intel_lrc.c
+++ b/drivers/gpu/drm/i915/intel_lrc.c
@@ -1254,9 +1254,10 @@ static int gen8_init_indirectctx_bb(struct intel_engine_cs *ring,
 
 	/* WaFlushCoherentL3CacheLinesAtContextSwitch:bdw */
 	if (IS_BROADWELL(ring->dev)) {
-		index = gen8_emit_flush_coherentl3_wa(ring, batch, index);
-		if (index < 0)
-			return index;
+		int rc = gen8_emit_flush_coherentl3_wa(ring, batch, index);
+		if (rc < 0)
+			return rc;
+		index = rc;
 	}
 
 	/* WaClearSlmSpaceAtContextSwitch:bdw,chv */
-- 
cgit v0.10.2


From 25e1793f6c642604b13e33bc81a9c8d48d652c86 Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Tue, 8 Sep 2015 13:56:23 +0200
Subject: drm/i915: Mark debug mod options as _unsafe

We don't want random people to touch these.

Especially true since we've just screwed up SKL by holding it way too
long under the preliminary flag because of some ABI issues. And now
there's howtos all over the internets about how to set this. Same
pretty much for anything else.

Cc: Jani Nikula <jani.nikula@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
[danvet: drop fastboot changes, that option is gone.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_params.c b/drivers/gpu/drm/i915/i915_params.c
index ca060d5..ca9b8f6 100644
--- a/drivers/gpu/drm/i915/i915_params.c
+++ b/drivers/gpu/drm/i915/i915_params.c
@@ -61,7 +61,7 @@ MODULE_PARM_DESC(modeset,
 	"Use kernel modesetting [KMS] (0=disable, "
 	"1=on, -1=force vga console preference [default])");
 
-module_param_named(panel_ignore_lid, i915.panel_ignore_lid, int, 0600);
+module_param_named_unsafe(panel_ignore_lid, i915.panel_ignore_lid, int, 0600);
 MODULE_PARM_DESC(panel_ignore_lid,
 	"Override lid status (0=autodetect, 1=autodetect disabled [default], "
 	"-1=force lid closed, -2=force lid open)");
@@ -84,17 +84,17 @@ MODULE_PARM_DESC(enable_fbc,
 	"Enable frame buffer compression for power savings "
 	"(default: -1 (use per-chip default))");
 
-module_param_named(lvds_channel_mode, i915.lvds_channel_mode, int, 0600);
+module_param_named_unsafe(lvds_channel_mode, i915.lvds_channel_mode, int, 0600);
 MODULE_PARM_DESC(lvds_channel_mode,
 	 "Specify LVDS channel mode "
 	 "(0=probe BIOS [default], 1=single-channel, 2=dual-channel)");
 
-module_param_named(lvds_use_ssc, i915.panel_use_ssc, int, 0600);
+module_param_named_unsafe(lvds_use_ssc, i915.panel_use_ssc, int, 0600);
 MODULE_PARM_DESC(lvds_use_ssc,
 	"Use Spread Spectrum Clock with panels [LVDS/eDP] "
 	"(default: auto from VBT)");
 
-module_param_named(vbt_sdvo_panel_type, i915.vbt_sdvo_panel_type, int, 0600);
+module_param_named_unsafe(vbt_sdvo_panel_type, i915.vbt_sdvo_panel_type, int, 0600);
 MODULE_PARM_DESC(vbt_sdvo_panel_type,
 	"Override/Ignore selection of SDVO panel mode in the VBT "
 	"(-2=ignore, -1=auto [default], index in VBT BIOS table)");
@@ -102,7 +102,7 @@ MODULE_PARM_DESC(vbt_sdvo_panel_type,
 module_param_named_unsafe(reset, i915.reset, bool, 0600);
 MODULE_PARM_DESC(reset, "Attempt GPU resets (default: true)");
 
-module_param_named(enable_hangcheck, i915.enable_hangcheck, bool, 0644);
+module_param_named_unsafe(enable_hangcheck, i915.enable_hangcheck, bool, 0644);
 MODULE_PARM_DESC(enable_hangcheck,
 	"Periodically check GPU activity for detecting hangs. "
 	"WARNING: Disabling this can cause system wide hangs. "
@@ -113,23 +113,23 @@ MODULE_PARM_DESC(enable_ppgtt,
 	"Override PPGTT usage. "
 	"(-1=auto [default], 0=disabled, 1=aliasing, 2=full)");
 
-module_param_named(enable_execlists, i915.enable_execlists, int, 0400);
+module_param_named_unsafe(enable_execlists, i915.enable_execlists, int, 0400);
 MODULE_PARM_DESC(enable_execlists,
 	"Override execlists usage. "
 	"(-1=auto [default], 0=disabled, 1=enabled)");
 
-module_param_named(enable_psr, i915.enable_psr, int, 0600);
+module_param_named_unsafe(enable_psr, i915.enable_psr, int, 0600);
 MODULE_PARM_DESC(enable_psr, "Enable PSR (default: false)");
 
-module_param_named(preliminary_hw_support, i915.preliminary_hw_support, int, 0600);
+module_param_named_unsafe(preliminary_hw_support, i915.preliminary_hw_support, int, 0600);
 MODULE_PARM_DESC(preliminary_hw_support,
 	"Enable preliminary hardware support.");
 
-module_param_named(disable_power_well, i915.disable_power_well, int, 0600);
+module_param_named_unsafe(disable_power_well, i915.disable_power_well, int, 0600);
 MODULE_PARM_DESC(disable_power_well,
 	"Disable the power well when possible (default: true)");
 
-module_param_named(enable_ips, i915.enable_ips, int, 0600);
+module_param_named_unsafe(enable_ips, i915.enable_ips, int, 0600);
 MODULE_PARM_DESC(enable_ips, "Enable IPS (default: true)");
 
 module_param_named_unsafe(prefault_disable, i915.prefault_disable, bool, 0600);
@@ -142,7 +142,7 @@ MODULE_PARM_DESC(load_detect_test,
 	"Force-enable the VGA load detect code for testing (default:false). "
 	"For developers only.");
 
-module_param_named(invert_brightness, i915.invert_brightness, int, 0600);
+module_param_named_unsafe(invert_brightness, i915.invert_brightness, int, 0600);
 MODULE_PARM_DESC(invert_brightness,
 	"Invert backlight brightness "
 	"(-1 force normal, 0 machine defaults, 1 force inversion), please "
@@ -153,14 +153,14 @@ MODULE_PARM_DESC(invert_brightness,
 module_param_named(disable_display, i915.disable_display, bool, 0600);
 MODULE_PARM_DESC(disable_display, "Disable display (default: false)");
 
-module_param_named(disable_vtd_wa, i915.disable_vtd_wa, bool, 0600);
+module_param_named_unsafe(disable_vtd_wa, i915.disable_vtd_wa, bool, 0600);
 MODULE_PARM_DESC(disable_vtd_wa, "Disable all VT-d workarounds (default: false)");
 
-module_param_named(enable_cmd_parser, i915.enable_cmd_parser, int, 0600);
+module_param_named_unsafe(enable_cmd_parser, i915.enable_cmd_parser, int, 0600);
 MODULE_PARM_DESC(enable_cmd_parser,
 		 "Enable command parsing (1=enabled [default], 0=disabled)");
 
-module_param_named(use_mmio_flip, i915.use_mmio_flip, int, 0600);
+module_param_named_unsafe(use_mmio_flip, i915.use_mmio_flip, int, 0600);
 MODULE_PARM_DESC(use_mmio_flip,
 		 "use MMIO flips (-1=never, 0=driver discretion [default], 1=always)");
 
-- 
cgit v0.10.2


From adda50b8b32bd559906d82fa32caeaf6c9306f59 Mon Sep 17 00:00:00 2001
From: Bob Paauwe <bob.j.paauwe@intel.com>
Date: Tue, 21 Jul 2015 10:42:53 -0700
Subject: drm/i915/skl: Don't clear all watermarks when updating. (v2)

Clearing the watermarks for all pipes/planes when updating the
watermarks for a single CRTC change seems like the wrong thing to
do here. As is, this code will ony update any pipe/plane watermarks
that need updating and leave the remaining set to zero.  Later, the
watermark checks in check_wm_state() will flag these zero'd out pipe/plane
watermarks and throw errors.

By clearing only the watermark values associated with the specific crtc
the other watermark values may remain unchanged.

v2: Make sure all the dirty flags are cleared. Damien
    Clear all values assoicated with crtc/pipe being updated.  Damien

Signed-off-by: Bob Paauwe <bob.j.paauwe@intel.com>
Reviewed-by: Damien Lespiau <damien.lespiau@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index 920872a..bec4e0e 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -3689,6 +3689,26 @@ static void skl_update_other_pipe_wm(struct drm_device *dev,
 	}
 }
 
+static void skl_clear_wm(struct skl_wm_values *watermarks, enum pipe pipe)
+{
+	watermarks->wm_linetime[pipe] = 0;
+	memset(watermarks->plane[pipe], 0,
+	       sizeof(uint32_t) * 8 * I915_MAX_PLANES);
+	memset(watermarks->cursor[pipe], 0, sizeof(uint32_t) * 8);
+	memset(watermarks->plane_trans[pipe],
+	       0, sizeof(uint32_t) * I915_MAX_PLANES);
+	watermarks->cursor_trans[pipe] = 0;
+
+	/* Clear ddb entries for pipe */
+	memset(&watermarks->ddb.pipe[pipe], 0, sizeof(struct skl_ddb_entry));
+	memset(&watermarks->ddb.plane[pipe], 0,
+	       sizeof(struct skl_ddb_entry) * I915_MAX_PLANES);
+	memset(&watermarks->ddb.y_plane[pipe], 0,
+	       sizeof(struct skl_ddb_entry) * I915_MAX_PLANES);
+	memset(&watermarks->ddb.cursor[pipe], 0, sizeof(struct skl_ddb_entry));
+
+}
+
 static void skl_update_wm(struct drm_crtc *crtc)
 {
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
@@ -3699,7 +3719,11 @@ static void skl_update_wm(struct drm_crtc *crtc)
 	struct skl_pipe_wm pipe_wm = {};
 	struct intel_wm_config config = {};
 
-	memset(results, 0, sizeof(*results));
+
+	/* Clear all dirty flags */
+	memset(results->dirty, 0, sizeof(bool) * I915_MAX_PIPES);
+
+	skl_clear_wm(results, intel_crtc->pipe);
 
 	skl_compute_wm_global_parameters(dev, &config);
 
-- 
cgit v0.10.2


From 18c237c0f9fa2e1c66e1eb4c305aeff4f28f89bf Mon Sep 17 00:00:00 2001
From: Animesh Manna <animesh.manna@intel.com>
Date: Tue, 4 Aug 2015 22:02:41 +0530
Subject: drm/i915/bxt: Path added of dmc firmware ver1 for BXT.

Broxton also has dmc to manage low-power display engine state.
Path of the firmware added in intel_csr.c.

Naming convention followed as <platform>_dmc_<api-version>.bin

v1: Initial version.

v2: Commit description added based on review comment from Sunil.

Cc: Damien Lespiau <damien.lespiau@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Sunil Kamath <sunil.kamath@intel.com>
Signed-off-by: Animesh Manna <animesh.manna@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_csr.c b/drivers/gpu/drm/i915/intel_csr.c
index 765dfcd..6d1aaff 100644
--- a/drivers/gpu/drm/i915/intel_csr.c
+++ b/drivers/gpu/drm/i915/intel_csr.c
@@ -42,8 +42,10 @@
  */
 
 #define I915_CSR_SKL "i915/skl_dmc_ver1.bin"
+#define I915_CSR_BXT "i915/bxt_dmc_ver1.bin"
 
 MODULE_FIRMWARE(I915_CSR_SKL);
+MODULE_FIRMWARE(I915_CSR_BXT);
 
 /*
 * SKL CSR registers for DC5 and DC6
@@ -409,6 +411,8 @@ void intel_csr_ucode_init(struct drm_device *dev)
 
 	if (IS_SKYLAKE(dev))
 		csr->fw_path = I915_CSR_SKL;
+	else if (IS_BROXTON(dev_priv))
+		csr->fw_path = I915_CSR_BXT;
 	else {
 		DRM_ERROR("Unexpected: no known CSR firmware for platform\n");
 		intel_csr_load_status_set(dev_priv, FW_FAILED);
-- 
cgit v0.10.2


From cff765fb1383dff6787fa735fcc88aa122d3c00d Mon Sep 17 00:00:00 2001
From: Animesh Manna <animesh.manna@intel.com>
Date: Tue, 4 Aug 2015 22:02:43 +0530
Subject: drm/i915/bxt: Stepping info added for bxt.

Added stepping info in intel_csr.c which is required to extract
specific firmware from packaged dmc firmware.

Stepping info is aligned with current bspec info.

Cc: Vetter, Daniel <daniel.vetter@intel.com>
Cc: Damien Lespiau <damien.lespiau@intel.com>
Cc: Imre Deak <imre.deak@intel.com>
Cc: Sunil Kamath <sunil.kamath@intel.com>
Signed-off-by: Animesh Manna <animesh.manna@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_csr.c b/drivers/gpu/drm/i915/intel_csr.c
index 6d1aaff..780d5d5 100644
--- a/drivers/gpu/drm/i915/intel_csr.c
+++ b/drivers/gpu/drm/i915/intel_csr.c
@@ -183,11 +183,19 @@ static const struct stepping_info skl_stepping_info[] = {
 		{'G', '0'}, {'H', '0'}, {'I', '0'}
 };
 
+static struct stepping_info bxt_stepping_info[] = {
+	{'A', '0'}, {'A', '1'}, {'A', '2'},
+	{'B', '0'}, {'B', '1'}, {'B', '2'}
+};
+
 static char intel_get_stepping(struct drm_device *dev)
 {
 	if (IS_SKYLAKE(dev) && (dev->pdev->revision <
 			ARRAY_SIZE(skl_stepping_info)))
 		return skl_stepping_info[dev->pdev->revision].stepping;
+	else if (IS_BROXTON(dev) && (dev->pdev->revision <
+				ARRAY_SIZE(bxt_stepping_info)))
+		return bxt_stepping_info[dev->pdev->revision].stepping;
 	else
 		return -ENODATA;
 }
@@ -197,6 +205,9 @@ static char intel_get_substepping(struct drm_device *dev)
 	if (IS_SKYLAKE(dev) && (dev->pdev->revision <
 			ARRAY_SIZE(skl_stepping_info)))
 		return skl_stepping_info[dev->pdev->revision].substepping;
+	else if (IS_BROXTON(dev) && (dev->pdev->revision <
+			ARRAY_SIZE(bxt_stepping_info)))
+		return bxt_stepping_info[dev->pdev->revision].substepping;
 	else
 		return -ENODATA;
 }
-- 
cgit v0.10.2


From 7b403ffba8ed8636f3467a9008bd2e060792b2e4 Mon Sep 17 00:00:00 2001
From: Animesh Manna <animesh.manna@intel.com>
Date: Tue, 4 Aug 2015 22:02:42 +0530
Subject: drm/i915/bxt: Modified HAS_CSR, added support for BXT

Modified HAS_CSR macro defination which earlier only supported
for skl, now added support for BXT.

v1: Initial version.

v2: Instaed of skylake/broxton check added gen9 check alone based
on review comment from Sunil.

Cc: Vetter, Daniel <daniel.vetter@intel.com>
Cc: Damien Lespiau <damien.lespiau@intel.com>
Cc: Imre Deak <imre.deak@intel.com>
Cc: Sunil Kamath <sunil.kamath@intel.com>
Signed-off-by: Animesh Manna <animesh.manna@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 677faa8..3795fdf 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -2578,7 +2578,7 @@ struct drm_i915_cmd_table {
 #define HAS_RC6(dev)		(INTEL_INFO(dev)->gen >= 6)
 #define HAS_RC6p(dev)		(INTEL_INFO(dev)->gen == 6 || IS_IVYBRIDGE(dev))
 
-#define HAS_CSR(dev)	(IS_SKYLAKE(dev))
+#define HAS_CSR(dev)	(IS_GEN9(dev))
 
 #define HAS_GUC_UCODE(dev)	(IS_GEN9(dev))
 #define HAS_GUC_SCHED(dev)	(IS_GEN9(dev))
-- 
cgit v0.10.2


From 23eafea6a9d1faac0588a5275d0c755cb261346e Mon Sep 17 00:00:00 2001
From: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Date: Sun, 23 Aug 2015 17:52:48 +0530
Subject: drm/i915/bxt: WaGsvDisableTurbo

Disable Turbo on steppings prior to B0 on BXT due to hangs seen during GT CPD exit.

Change-Id: I50c5c03f59f5ba092db19e17234951d89db42c6c
Signed-off-by: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Reviewed by: Alex Dai <yu.dai@intel.com>.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index bec4e0e..f52431b 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -4487,6 +4487,10 @@ static void gen6_set_rps(struct drm_device *dev, u8 val)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
+	/* WaGsvDisableTurbo: Workaround to disable turbo on BXT A* */
+	if (IS_BROXTON(dev) && (INTEL_REVID(dev) < BXT_REVID_B0))
+		return;
+
 	WARN_ON(!mutex_is_locked(&dev_priv->rps.hw_lock));
 	WARN_ON(val > dev_priv->rps.max_freq);
 	WARN_ON(val < dev_priv->rps.min_freq);
@@ -4807,6 +4811,12 @@ static void gen9_enable_rps(struct drm_device *dev)
 
 	gen6_init_rps_frequencies(dev);
 
+	/* WaGsvDisableTurbo: Workaround to disable turbo on BXT A* */
+	if (IS_BROXTON(dev) && (INTEL_REVID(dev) < BXT_REVID_B0)) {
+		intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+		return;
+	}
+
 	/* Program defaults and thresholds for RPS*/
 	I915_WRITE(GEN6_RC_VIDEO_FREQ,
 		GEN9_FREQUENCY(dev_priv->rps.rp1_freq));
-- 
cgit v0.10.2


From 83e3337204b2385d20e149c8fe91bb2719978df2 Mon Sep 17 00:00:00 2001
From: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Date: Sun, 23 Aug 2015 17:52:47 +0530
Subject: drm/i915: Increase maximum polling time to 50ms for forcewake
 request/clear ack

On BXT, We Observe timeout for forcewake request completion with 2ms polling period as given here:
[drm:fw_domains_get] ERROR render: timed out waiting for forcewake ack request.
Polling for 50ms is recommended to avoid these timeouts.

Change-Id: Ie715b0069a3049606e9602bc5e97a6511890864d
Signed-off-by: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Reviewed by: Alex Dai <yu.dai@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c
index 440e2a5..14d0831 100644
--- a/drivers/gpu/drm/i915/intel_uncore.c
+++ b/drivers/gpu/drm/i915/intel_uncore.c
@@ -27,7 +27,7 @@
 
 #include <linux/pm_runtime.h>
 
-#define FORCEWAKE_ACK_TIMEOUT_MS 2
+#define FORCEWAKE_ACK_TIMEOUT_MS 50
 
 #define __raw_i915_read8(dev_priv__, reg__) readb((dev_priv__)->regs + (reg__))
 #define __raw_i915_write8(dev_priv__, reg__, val__) writeb(val__, (dev_priv__)->regs + (reg__))
-- 
cgit v0.10.2


From cfe01a5eba1ff5723f3a47895fb7e9d890edf157 Mon Sep 17 00:00:00 2001
From: Shashank Sharma <shashank.sharma@intel.com>
Date: Tue, 1 Sep 2015 19:41:38 +0530
Subject: drm/i915/bxt: Enable BXT DSI PLL

This patch adds new functions for BXT clock and PLL programming.
They are:
1. configure_dsi_pll for BXT.
   This function does the basic math and generates the divider ratio
   based on requested pixclock, and program clock registers.
2. enable_dsi_pll function.
   This function programs the calculated clock values on the PLL.
3. intel_enable_dsi_pll
   Wrapper function to use same code for multiple platforms. It checks the
   platform and calls appropriate core pll enable function.

v2: Fixed Jani's review comments. Macros are adjusted as per convention.

v3: Removed a redundant change wrt code comment.

Signed-off-by: Shashank Sharma <shashank.sharma@intel.com>
Signed-off-by: Uma Shankar <uma.shankar@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index dd3d235..4ef16aa 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -7507,6 +7507,28 @@ enum skl_disp_power_wells {
 
 #define _MIPI_PORT(port, a, c)	_PORT3(port, a, 0, c)	/* ports A and C only */
 
+#define BXT_DSI_PLL_CTL			0x161000
+#define  BXT_DSI_PLL_PVD_RATIO_SHIFT	16
+#define  BXT_DSI_PLL_PVD_RATIO_MASK	(3 << BXT_DSI_PLL_PVD_RATIO_SHIFT)
+#define  BXT_DSI_PLL_PVD_RATIO_1	(1 << BXT_DSI_PLL_PVD_RATIO_SHIFT)
+#define  BXT_DSIC_16X_BY2		(1 << 10)
+#define  BXT_DSIC_16X_BY3		(2 << 10)
+#define  BXT_DSIC_16X_BY4		(3 << 10)
+#define  BXT_DSIA_16X_BY2		(1 << 8)
+#define  BXT_DSIA_16X_BY3		(2 << 8)
+#define  BXT_DSIA_16X_BY4		(3 << 8)
+#define  BXT_DSI_FREQ_SEL_SHIFT		8
+#define  BXT_DSI_FREQ_SEL_MASK		(0xF << BXT_DSI_FREQ_SEL_SHIFT)
+
+#define BXT_DSI_PLL_RATIO_MAX		0x7D
+#define BXT_DSI_PLL_RATIO_MIN		0x22
+#define BXT_DSI_PLL_RATIO_MASK		0xFF
+#define BXT_REF_CLOCK_KHZ		19500
+
+#define BXT_DSI_PLL_ENABLE		0x46080
+#define  BXT_DSI_PLL_DO_ENABLE		(1 << 31)
+#define  BXT_DSI_PLL_LOCKED		(1 << 30)
+
 #define _MIPIA_PORT_CTRL			(VLV_DISPLAY_BASE + 0x61190)
 #define _MIPIC_PORT_CTRL			(VLV_DISPLAY_BASE + 0x61700)
 #define MIPI_PORT_CTRL(port)	_MIPI_PORT(port, _MIPIA_PORT_CTRL, _MIPIC_PORT_CTRL)
diff --git a/drivers/gpu/drm/i915/intel_dsi.c b/drivers/gpu/drm/i915/intel_dsi.c
index 781c267..f2f8196 100644
--- a/drivers/gpu/drm/i915/intel_dsi.c
+++ b/drivers/gpu/drm/i915/intel_dsi.c
@@ -914,8 +914,8 @@ static void intel_dsi_pre_pll_enable(struct intel_encoder *encoder)
 	DRM_DEBUG_KMS("\n");
 
 	intel_dsi_prepare(encoder);
+	intel_enable_dsi_pll(encoder);
 
-	vlv_enable_dsi_pll(encoder);
 }
 
 static enum drm_connector_status
diff --git a/drivers/gpu/drm/i915/intel_dsi.h b/drivers/gpu/drm/i915/intel_dsi.h
index 42a6859..7f16c68 100644
--- a/drivers/gpu/drm/i915/intel_dsi.h
+++ b/drivers/gpu/drm/i915/intel_dsi.h
@@ -124,7 +124,7 @@ static inline struct intel_dsi *enc_to_intel_dsi(struct drm_encoder *encoder)
 	return container_of(encoder, struct intel_dsi, base.base);
 }
 
-extern void vlv_enable_dsi_pll(struct intel_encoder *encoder);
+extern void intel_enable_dsi_pll(struct intel_encoder *encoder);
 extern void vlv_disable_dsi_pll(struct intel_encoder *encoder);
 extern u32 vlv_get_dsi_pclk(struct intel_encoder *encoder, int pipe_bpp);
 
diff --git a/drivers/gpu/drm/i915/intel_dsi_pll.c b/drivers/gpu/drm/i915/intel_dsi_pll.c
index c6a8975..b07483b 100644
--- a/drivers/gpu/drm/i915/intel_dsi_pll.c
+++ b/drivers/gpu/drm/i915/intel_dsi_pll.c
@@ -246,7 +246,7 @@ static void vlv_configure_dsi_pll(struct intel_encoder *encoder)
 	vlv_cck_write(dev_priv, CCK_REG_DSI_PLL_CONTROL, dsi_mnp.dsi_pll_ctrl);
 }
 
-void vlv_enable_dsi_pll(struct intel_encoder *encoder)
+static void vlv_enable_dsi_pll(struct intel_encoder *encoder)
 {
 	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
 	u32 tmp;
@@ -363,3 +363,96 @@ u32 vlv_get_dsi_pclk(struct intel_encoder *encoder, int pipe_bpp)
 
 	return pclk;
 }
+
+static bool bxt_configure_dsi_pll(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
+	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
+	u8 dsi_ratio;
+	u32 dsi_clk;
+	u32 val;
+
+	dsi_clk = dsi_clk_from_pclk(intel_dsi->pclk, intel_dsi->pixel_format,
+			intel_dsi->lane_count);
+
+	/*
+	 * From clock diagram, to get PLL ratio divider, divide double of DSI
+	 * link rate (i.e., 2*8x=16x frequency value) by ref clock. Make sure to
+	 * round 'up' the result
+	 */
+	dsi_ratio = DIV_ROUND_UP(dsi_clk * 2, BXT_REF_CLOCK_KHZ);
+	if (dsi_ratio < BXT_DSI_PLL_RATIO_MIN ||
+			dsi_ratio > BXT_DSI_PLL_RATIO_MAX) {
+		DRM_ERROR("Cant get a suitable ratio from DSI PLL ratios\n");
+		return false;
+	}
+
+	/*
+	 * Program DSI ratio and Select MIPIC and MIPIA PLL output as 8x
+	 * Spec says both have to be programmed, even if one is not getting
+	 * used. Configure MIPI_CLOCK_CTL dividers in modeset
+	 */
+	val = I915_READ(BXT_DSI_PLL_CTL);
+	val &= ~BXT_DSI_PLL_PVD_RATIO_MASK;
+	val &= ~BXT_DSI_FREQ_SEL_MASK;
+	val &= ~BXT_DSI_PLL_RATIO_MASK;
+	val |= (dsi_ratio | BXT_DSIA_16X_BY2 | BXT_DSIC_16X_BY2);
+
+	/* As per recommendation from hardware team,
+	 * Prog PVD ratio =1 if dsi ratio <= 50
+	 */
+	if (dsi_ratio <= 50) {
+		val &= ~BXT_DSI_PLL_PVD_RATIO_MASK;
+		val |= BXT_DSI_PLL_PVD_RATIO_1;
+	}
+
+	I915_WRITE(BXT_DSI_PLL_CTL, val);
+	POSTING_READ(BXT_DSI_PLL_CTL);
+
+	return true;
+}
+
+static void bxt_enable_dsi_pll(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
+	u32 val;
+
+	DRM_DEBUG_KMS("\n");
+
+	val = I915_READ(BXT_DSI_PLL_ENABLE);
+
+	if (val & BXT_DSI_PLL_DO_ENABLE) {
+		WARN(1, "DSI PLL already enabled. Disabling it.\n");
+		val &= ~BXT_DSI_PLL_DO_ENABLE;
+		I915_WRITE(BXT_DSI_PLL_ENABLE, val);
+	}
+
+	/* Configure PLL vales */
+	if (!bxt_configure_dsi_pll(encoder)) {
+		DRM_ERROR("Configure DSI PLL failed, abort PLL enable\n");
+		return;
+	}
+
+	/* Enable DSI PLL */
+	val = I915_READ(BXT_DSI_PLL_ENABLE);
+	val |= BXT_DSI_PLL_DO_ENABLE;
+	I915_WRITE(BXT_DSI_PLL_ENABLE, val);
+
+	/* Timeout and fail if PLL not locked */
+	if (wait_for(I915_READ(BXT_DSI_PLL_ENABLE) & BXT_DSI_PLL_LOCKED, 1)) {
+		DRM_ERROR("Timed out waiting for DSI PLL to lock\n");
+		return;
+	}
+
+	DRM_DEBUG_KMS("DSI PLL locked\n");
+}
+
+void intel_enable_dsi_pll(struct intel_encoder *encoder)
+{
+	struct drm_device *dev = encoder->base.dev;
+
+	if (IS_VALLEYVIEW(dev))
+		vlv_enable_dsi_pll(encoder);
+	else if (IS_BROXTON(dev))
+		bxt_enable_dsi_pll(encoder);
+}
-- 
cgit v0.10.2


From fe88fc6828f6471c9eebc511aa15984d055d11c1 Mon Sep 17 00:00:00 2001
From: Shashank Sharma <shashank.sharma@intel.com>
Date: Tue, 1 Sep 2015 19:41:39 +0530
Subject: drm/i915/bxt: Disable DSI PLL for BXT

This patch adds two new functions:
- disable_dsi_pll.
  BXT DSI disable sequence and registers are
  different from previous platforms.
- intel_disable_dsi_pll
  wrapper function to re-use the same code for
  multiple platforms. It checks platform type and
  calls appropriate core pll disable function.

v2: Fixed Jani's review comments.

v3: Rebased on latest drm-nightly branch.

Signed-off-by: Shashank Sharma <shashank.sharma@intel.com>
Signed-off-by: Uma Shankar <uma.shankar@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dsi.c b/drivers/gpu/drm/i915/intel_dsi.c
index f2f8196..50159a9 100644
--- a/drivers/gpu/drm/i915/intel_dsi.c
+++ b/drivers/gpu/drm/i915/intel_dsi.c
@@ -557,7 +557,7 @@ static void intel_dsi_clear_device_ready(struct intel_encoder *encoder)
 		usleep_range(2000, 2500);
 	}
 
-	vlv_disable_dsi_pll(encoder);
+	intel_disable_dsi_pll(encoder);
 }
 
 static void intel_dsi_post_disable(struct intel_encoder *encoder)
diff --git a/drivers/gpu/drm/i915/intel_dsi.h b/drivers/gpu/drm/i915/intel_dsi.h
index 7f16c68..5cc46b4 100644
--- a/drivers/gpu/drm/i915/intel_dsi.h
+++ b/drivers/gpu/drm/i915/intel_dsi.h
@@ -125,7 +125,7 @@ static inline struct intel_dsi *enc_to_intel_dsi(struct drm_encoder *encoder)
 }
 
 extern void intel_enable_dsi_pll(struct intel_encoder *encoder);
-extern void vlv_disable_dsi_pll(struct intel_encoder *encoder);
+extern void intel_disable_dsi_pll(struct intel_encoder *encoder);
 extern u32 vlv_get_dsi_pclk(struct intel_encoder *encoder, int pipe_bpp);
 
 struct drm_panel *vbt_panel_init(struct intel_dsi *intel_dsi, u16 panel_id);
diff --git a/drivers/gpu/drm/i915/intel_dsi_pll.c b/drivers/gpu/drm/i915/intel_dsi_pll.c
index b07483b..f335e6c 100644
--- a/drivers/gpu/drm/i915/intel_dsi_pll.c
+++ b/drivers/gpu/drm/i915/intel_dsi_pll.c
@@ -276,7 +276,7 @@ static void vlv_enable_dsi_pll(struct intel_encoder *encoder)
 	DRM_DEBUG_KMS("DSI PLL locked\n");
 }
 
-void vlv_disable_dsi_pll(struct intel_encoder *encoder)
+static void vlv_disable_dsi_pll(struct intel_encoder *encoder)
 {
 	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
 	u32 tmp;
@@ -293,6 +293,26 @@ void vlv_disable_dsi_pll(struct intel_encoder *encoder)
 	mutex_unlock(&dev_priv->sb_lock);
 }
 
+static void bxt_disable_dsi_pll(struct intel_encoder *encoder)
+{
+	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
+	u32 val;
+
+	DRM_DEBUG_KMS("\n");
+
+	val = I915_READ(BXT_DSI_PLL_ENABLE);
+	val &= ~BXT_DSI_PLL_DO_ENABLE;
+	I915_WRITE(BXT_DSI_PLL_ENABLE, val);
+
+	/*
+	 * PLL lock should deassert within 200us.
+	 * Wait up to 1ms before timing out.
+	 */
+	if (wait_for((I915_READ(BXT_DSI_PLL_ENABLE)
+					& BXT_DSI_PLL_LOCKED) == 0, 1))
+		DRM_ERROR("Timeout waiting for PLL lock deassertion\n");
+}
+
 static void assert_bpp_mismatch(int pixel_format, int pipe_bpp)
 {
 	int bpp = dsi_pixel_format_bpp(pixel_format);
@@ -456,3 +476,13 @@ void intel_enable_dsi_pll(struct intel_encoder *encoder)
 	else if (IS_BROXTON(dev))
 		bxt_enable_dsi_pll(encoder);
 }
+
+void intel_disable_dsi_pll(struct intel_encoder *encoder)
+{
+	struct drm_device *dev = encoder->base.dev;
+
+	if (IS_VALLEYVIEW(dev))
+		vlv_disable_dsi_pll(encoder);
+	else if (IS_BROXTON(dev))
+		bxt_disable_dsi_pll(encoder);
+}
-- 
cgit v0.10.2


From d2e08c0f34438af791482de8abf2c8e4e573b1d3 Mon Sep 17 00:00:00 2001
From: Shashank Sharma <shashank.sharma@intel.com>
Date: Tue, 1 Sep 2015 19:41:40 +0530
Subject: drm/i915/bxt: DSI prepare changes for BXT

This patch modifies dsi_prepare() function to support the same
modeset prepare sequence for BXT also. Main changes are:
1. BXT port control register is different than VLV.
2. BXT modeset sequence needs vdisplay and hdisplay programmed
   for transcoder.
3. BXT can select PIPE for MIPI transcoders.
4. BXT needs to program register MIPI_INIT_COUNT for both the ports,
   even if only one is being used.

v2: Fixed Jani's review comments. Rectified the DSI Macros to get
    proper register offsets using _MIPI_PORT instead of _TRANSCODER

v3: Rebased on latest drm-nightly branch. Fixed Jani's review comments.

Signed-off-by: Shashank Sharma <shashank.sharma@intel.com>
Signed-off-by: Uma Shankar <uma.shankar@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 4ef16aa..c5287a1 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -7507,6 +7507,22 @@ enum skl_disp_power_wells {
 
 #define _MIPI_PORT(port, a, c)	_PORT3(port, a, 0, c)	/* ports A and C only */
 
+/* BXT MIPI mode configure */
+#define  _BXT_MIPIA_TRANS_HACTIVE			0x6B0F8
+#define  _BXT_MIPIC_TRANS_HACTIVE			0x6B8F8
+#define  BXT_MIPI_TRANS_HACTIVE(tc)	_MIPI_PORT(tc, \
+		_BXT_MIPIA_TRANS_HACTIVE, _BXT_MIPIC_TRANS_HACTIVE)
+
+#define  _BXT_MIPIA_TRANS_VACTIVE			0x6B0FC
+#define  _BXT_MIPIC_TRANS_VACTIVE			0x6B8FC
+#define  BXT_MIPI_TRANS_VACTIVE(tc)	_MIPI_PORT(tc, \
+		_BXT_MIPIA_TRANS_VACTIVE, _BXT_MIPIC_TRANS_VACTIVE)
+
+#define  _BXT_MIPIA_TRANS_VTOTAL			0x6B100
+#define  _BXT_MIPIC_TRANS_VTOTAL			0x6B900
+#define  BXT_MIPI_TRANS_VTOTAL(tc)	_MIPI_PORT(tc, \
+		_BXT_MIPIA_TRANS_VTOTAL, _BXT_MIPIC_TRANS_VTOTAL)
+
 #define BXT_DSI_PLL_CTL			0x161000
 #define  BXT_DSI_PLL_PVD_RATIO_SHIFT	16
 #define  BXT_DSI_PLL_PVD_RATIO_MASK	(3 << BXT_DSI_PLL_PVD_RATIO_SHIFT)
@@ -7942,6 +7958,11 @@ enum skl_disp_power_wells {
 #define  READ_REQUEST_PRIORITY_HIGH			(3 << 3)
 #define  RGB_FLIP_TO_BGR				(1 << 2)
 
+#define  BXT_PIPE_SELECT_MASK				(7 << 7)
+#define  BXT_PIPE_SELECT_C				(2 << 7)
+#define  BXT_PIPE_SELECT_B				(1 << 7)
+#define  BXT_PIPE_SELECT_A				(0 << 7)
+
 #define _MIPIA_DATA_ADDRESS		(dev_priv->mipi_mmio_base + 0xb108)
 #define _MIPIC_DATA_ADDRESS		(dev_priv->mipi_mmio_base + 0xb908)
 #define MIPI_DATA_ADDRESS(port)		_MIPI_PORT(port, _MIPIA_DATA_ADDRESS, \
diff --git a/drivers/gpu/drm/i915/intel_dsi.c b/drivers/gpu/drm/i915/intel_dsi.c
index 50159a9..5b69b41 100644
--- a/drivers/gpu/drm/i915/intel_dsi.c
+++ b/drivers/gpu/drm/i915/intel_dsi.c
@@ -737,6 +737,21 @@ static void set_dsi_timings(struct drm_encoder *encoder,
 	hbp = txbyteclkhs(hbp, bpp, lane_count, intel_dsi->burst_mode_ratio);
 
 	for_each_dsi_port(port, intel_dsi->ports) {
+		if (IS_BROXTON(dev)) {
+			/*
+			 * Program hdisplay and vdisplay on MIPI transcoder.
+			 * This is different from calculated hactive and
+			 * vactive, as they are calculated per channel basis,
+			 * whereas these values should be based on resolution.
+			 */
+			I915_WRITE(BXT_MIPI_TRANS_HACTIVE(port),
+					mode->hdisplay);
+			I915_WRITE(BXT_MIPI_TRANS_VACTIVE(port),
+					mode->vdisplay);
+			I915_WRITE(BXT_MIPI_TRANS_VTOTAL(port),
+					mode->vtotal);
+		}
+
 		I915_WRITE(MIPI_HACTIVE_AREA_COUNT(port), hactive);
 		I915_WRITE(MIPI_HFP_COUNT(port), hfp);
 
@@ -777,16 +792,39 @@ static void intel_dsi_prepare(struct intel_encoder *intel_encoder)
 	}
 
 	for_each_dsi_port(port, intel_dsi->ports) {
-		/* escape clock divider, 20MHz, shared for A and C.
-		 * device ready must be off when doing this! txclkesc? */
-		tmp = I915_READ(MIPI_CTRL(PORT_A));
-		tmp &= ~ESCAPE_CLOCK_DIVIDER_MASK;
-		I915_WRITE(MIPI_CTRL(PORT_A), tmp | ESCAPE_CLOCK_DIVIDER_1);
-
-		/* read request priority is per pipe */
-		tmp = I915_READ(MIPI_CTRL(port));
-		tmp &= ~READ_REQUEST_PRIORITY_MASK;
-		I915_WRITE(MIPI_CTRL(port), tmp | READ_REQUEST_PRIORITY_HIGH);
+		if (IS_VALLEYVIEW(dev)) {
+			/*
+			 * escape clock divider, 20MHz, shared for A and C.
+			 * device ready must be off when doing this! txclkesc?
+			 */
+			tmp = I915_READ(MIPI_CTRL(PORT_A));
+			tmp &= ~ESCAPE_CLOCK_DIVIDER_MASK;
+			I915_WRITE(MIPI_CTRL(PORT_A), tmp |
+					ESCAPE_CLOCK_DIVIDER_1);
+
+			/* read request priority is per pipe */
+			tmp = I915_READ(MIPI_CTRL(port));
+			tmp &= ~READ_REQUEST_PRIORITY_MASK;
+			I915_WRITE(MIPI_CTRL(port), tmp |
+					READ_REQUEST_PRIORITY_HIGH);
+		} else if (IS_BROXTON(dev)) {
+			/*
+			 * FIXME:
+			 * BXT can connect any PIPE to any MIPI port.
+			 * Select the pipe based on the MIPI port read from
+			 * VBT for now. Pick PIPE A for MIPI port A and C
+			 * for port C.
+			 */
+			tmp = I915_READ(MIPI_CTRL(port));
+			tmp &= ~BXT_PIPE_SELECT_MASK;
+
+			if (port == PORT_A)
+				tmp |= BXT_PIPE_SELECT_A;
+			else if (port == PORT_C)
+				tmp |= BXT_PIPE_SELECT_C;
+
+			I915_WRITE(MIPI_CTRL(port), tmp);
+		}
 
 		/* XXX: why here, why like this? handling in irq handler?! */
 		I915_WRITE(MIPI_INTR_STAT(port), 0xffffffff);
@@ -863,6 +901,17 @@ static void intel_dsi_prepare(struct intel_encoder *intel_encoder)
 		I915_WRITE(MIPI_INIT_COUNT(port),
 				txclkesc(intel_dsi->escape_clk_div, 100));
 
+		if (IS_BROXTON(dev) && (!intel_dsi->dual_link)) {
+			/*
+			 * BXT spec says write MIPI_INIT_COUNT for
+			 * both the ports, even if only one is
+			 * getting used. So write the other port
+			 * if not in dual link mode.
+			 */
+			I915_WRITE(MIPI_INIT_COUNT(port ==
+						PORT_A ? PORT_C : PORT_A),
+					intel_dsi->init_count);
+		}
 
 		/* recovery disables */
 		I915_WRITE(MIPI_EOT_DISABLE(port), tmp);
-- 
cgit v0.10.2


From 237ed86c693d8a8e4db476976aeb30df4deac74b Mon Sep 17 00:00:00 2001
From: Sonika Jindal <sonika.jindal@intel.com>
Date: Tue, 15 Sep 2015 09:44:20 +0530
Subject: drm/i915: Check live status before reading edid

The Bspec is very clear that Live status must be checked about before
trying to read EDID over DDC channel. This patch makes sure that HDMI
EDID is read only when live status is up.

The live status doesn't seem to perform very consistent across various
platforms when tested with different monitors. The reason behind that is
some monitors are late to provide right voltage to set live_status up.
So, after getting the interrupt, for a small duration, live status reg
fluctuates, and then settles down showing the correct staus.

This is explained here in, in a rough way:
HPD line  ________________
			 |\ T1 = Monitor Hotplug causing IRQ
			 | \______________________________________
			 | |
                         | |
			 | |   T2 = Live status is stable
			 | |  _____________________________________
			 | | /|
Live status _____________|_|/ |
			 | |  |
			 | |  |
			 | |  |
			T0 T1  T2

(Between T1 and T2 Live status fluctuates or can be even low, depending on
 the monitor)

After several experiments, we have concluded that a max delay
of 30ms is enough to allow the live status to settle down with
most of the monitors. This total delay of 30ms has been split into
a resolution of 3 retries of 10ms each, for the better cases.

This delay is kept at 30ms, keeping in consideration that, HDCP compliance
expect the HPD handler to respond a plug out in 100ms, by disabling port.

v2: Adding checks for VLV/CHV as well. Reusing old ibx and g4x functions
to check digital port status. Adding a separate function to get bxt live
status (Daniel)
v3: Using intel_encoder->hpd_pin to check the live status (Siva)
Moving the live status read to intel_hdmi_probe and passing parameter
to read/not to read the edid. (me)
v4:
* Added live status check for all platforms using
intel_digital_port_connected.
* Rebased on top of Jani's DP cleanup series
* Some monitors take time in setting the live status. So retry for few
times if this is a connect HPD
v5: Removed extra "drm/i915" from commit message. Adding Shashank's sob
    which was missed.
v6: Drop the (!detect_edid && !live_status check) check because for DDI
ports which are enumerated as hdmi as well as DP, we don't have a
mechanism to differentiate between DP and hdmi inside the encoder's
hot_plug. This leads to call to the hdmi's hot_plug hook for DP as well
as hdmi which leads to issues during unplug because of the above check.
v7: Make intel_digital_port_connected global in this patch, some
reformatting of while loop, adding a print when live status is not
up. (Rodrigo)
v8: Rebase it on nightly which involved skipping the hot_plug hook for now
and letting the live_status check happen in detect until the hpd handling
part is finalized (Daniel)

Signed-off-by: Shashank Sharma <shashank.sharma@intel.com>
Signed-off-by: Sonika Jindal <sonika.jindal@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 9700f77..02401a3 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -4672,7 +4672,7 @@ static bool bxt_digital_port_connected(struct drm_i915_private *dev_priv,
  *
  * Return %true if @port is connected, %false otherwise.
  */
-static bool intel_digital_port_connected(struct drm_i915_private *dev_priv,
+bool intel_digital_port_connected(struct drm_i915_private *dev_priv,
 					 struct intel_digital_port *port)
 {
 	if (HAS_PCH_IBX(dev_priv))
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 8760814..394018e 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -1218,6 +1218,8 @@ void intel_edp_drrs_disable(struct intel_dp *intel_dp);
 void intel_edp_drrs_invalidate(struct drm_device *dev,
 		unsigned frontbuffer_bits);
 void intel_edp_drrs_flush(struct drm_device *dev, unsigned frontbuffer_bits);
+bool intel_digital_port_connected(struct drm_i915_private *dev_priv,
+					 struct intel_digital_port *port);
 void hsw_dp_set_ddi_pll_sel(struct intel_crtc_state *pipe_config);
 
 /* intel_dp_mst.c */
diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index e978c59..bb33c66 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -1329,22 +1329,23 @@ intel_hdmi_unset_edid(struct drm_connector *connector)
 }
 
 static bool
-intel_hdmi_set_edid(struct drm_connector *connector)
+intel_hdmi_set_edid(struct drm_connector *connector, bool force)
 {
 	struct drm_i915_private *dev_priv = to_i915(connector->dev);
 	struct intel_hdmi *intel_hdmi = intel_attached_hdmi(connector);
 	struct intel_encoder *intel_encoder =
 		&hdmi_to_dig_port(intel_hdmi)->base;
 	enum intel_display_power_domain power_domain;
-	struct edid *edid;
+	struct edid *edid = NULL;
 	bool connected = false;
 
 	power_domain = intel_display_port_power_domain(intel_encoder);
 	intel_display_power_get(dev_priv, power_domain);
 
-	edid = drm_get_edid(connector,
-			    intel_gmbus_get_adapter(dev_priv,
-						    intel_hdmi->ddc_bus));
+	if (force)
+		edid = drm_get_edid(connector,
+				    intel_gmbus_get_adapter(dev_priv,
+				    intel_hdmi->ddc_bus));
 
 	intel_display_power_put(dev_priv, power_domain);
 
@@ -1372,13 +1373,26 @@ static enum drm_connector_status
 intel_hdmi_detect(struct drm_connector *connector, bool force)
 {
 	enum drm_connector_status status;
+	struct intel_hdmi *intel_hdmi = intel_attached_hdmi(connector);
+	struct drm_i915_private *dev_priv = to_i915(connector->dev);
+	bool live_status = false;
+	unsigned int retry = 3;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
 		      connector->base.id, connector->name);
 
+	while (!live_status && --retry) {
+		live_status = intel_digital_port_connected(dev_priv,
+				hdmi_to_dig_port(intel_hdmi));
+		mdelay(10);
+	}
+
+	if (!live_status)
+		DRM_DEBUG_KMS("Live status not up!");
+
 	intel_hdmi_unset_edid(connector);
 
-	if (intel_hdmi_set_edid(connector)) {
+	if (intel_hdmi_set_edid(connector, live_status)) {
 		struct intel_hdmi *intel_hdmi = intel_attached_hdmi(connector);
 
 		hdmi_to_dig_port(intel_hdmi)->base.type = INTEL_OUTPUT_HDMI;
@@ -1402,7 +1416,7 @@ intel_hdmi_force(struct drm_connector *connector)
 	if (connector->status != connector_status_connected)
 		return;
 
-	intel_hdmi_set_edid(connector);
+	intel_hdmi_set_edid(connector, true);
 	hdmi_to_dig_port(intel_hdmi)->base.type = INTEL_OUTPUT_HDMI;
 }
 
-- 
cgit v0.10.2


From 7f4c62840cc416986f983a8c68f7010c97e06daf Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 10 Sep 2015 18:59:07 +0300
Subject: drm/i915: Assign hwmode after encoder state readout
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The dotclock is often calculated in encoder .get_config(), so we
shouldn't copy the adjusted_mode to hwmode until we have read out the
dotclock.

Gets rid of some warnings like these:
[drm:drm_calc_timestamping_constants [drm]] *ERROR* crtc 21: Can't calculate constants, dotclock = 0!
[drm:i915_get_vblank_timestamp] crtc 0 is disabled

v2: Steal Maarten's idea to move crtc->mode etc. assignment too

Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Patrik Jakobsson <patrik.jakobsson@linux.intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=91428
Reviewed-by: Patrik Jakobsson <patrik.jakobsson@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index a001cbb..ce9bb47 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -15121,33 +15121,6 @@ static void intel_modeset_readout_hw_state(struct drm_device *dev)
 		crtc->base.state->active = crtc->active;
 		crtc->base.enabled = crtc->active;
 
-		memset(&crtc->base.mode, 0, sizeof(crtc->base.mode));
-		if (crtc->base.state->active) {
-			intel_mode_from_pipe_config(&crtc->base.mode, crtc->config);
-			intel_mode_from_pipe_config(&crtc->base.state->adjusted_mode, crtc->config);
-			WARN_ON(drm_atomic_set_mode_for_crtc(crtc->base.state, &crtc->base.mode));
-
-			/*
-			 * The initial mode needs to be set in order to keep
-			 * the atomic core happy. It wants a valid mode if the
-			 * crtc's enabled, so we do the above call.
-			 *
-			 * At this point some state updated by the connectors
-			 * in their ->detect() callback has not run yet, so
-			 * no recalculation can be done yet.
-			 *
-			 * Even if we could do a recalculation and modeset
-			 * right now it would cause a double modeset if
-			 * fbdev or userspace chooses a different initial mode.
-			 *
-			 * If that happens, someone indicated they wanted a
-			 * mode change, which means it's safe to do a full
-			 * recalculation.
-			 */
-			crtc->base.state->mode.private_flags = I915_MODE_FLAG_INHERITED;
-		}
-
-		crtc->base.hwmode = crtc->config->base.adjusted_mode;
 		readout_plane_state(crtc, to_intel_crtc_state(crtc->base.state));
 
 		DRM_DEBUG_KMS("[CRTC:%d] hw state readout: %s\n",
@@ -15207,6 +15180,36 @@ static void intel_modeset_readout_hw_state(struct drm_device *dev)
 			      connector->base.name,
 			      connector->base.encoder ? "enabled" : "disabled");
 	}
+
+	for_each_intel_crtc(dev, crtc) {
+		crtc->base.hwmode = crtc->config->base.adjusted_mode;
+
+		memset(&crtc->base.mode, 0, sizeof(crtc->base.mode));
+		if (crtc->base.state->active) {
+			intel_mode_from_pipe_config(&crtc->base.mode, crtc->config);
+			intel_mode_from_pipe_config(&crtc->base.state->adjusted_mode, crtc->config);
+			WARN_ON(drm_atomic_set_mode_for_crtc(crtc->base.state, &crtc->base.mode));
+
+			/*
+			 * The initial mode needs to be set in order to keep
+			 * the atomic core happy. It wants a valid mode if the
+			 * crtc's enabled, so we do the above call.
+			 *
+			 * At this point some state updated by the connectors
+			 * in their ->detect() callback has not run yet, so
+			 * no recalculation can be done yet.
+			 *
+			 * Even if we could do a recalculation and modeset
+			 * right now it would cause a double modeset if
+			 * fbdev or userspace chooses a different initial mode.
+			 *
+			 * If that happens, someone indicated they wanted a
+			 * mode change, which means it's safe to do a full
+			 * recalculation.
+			 */
+			crtc->base.state->mode.private_flags = I915_MODE_FLAG_INHERITED;
+		}
+	}
 }
 
 /* Scan out the current hw modeset state,
-- 
cgit v0.10.2


From f9cd7b881a042dcc8f1b1236ecea6deb3eb4d5bd Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 10 Sep 2015 18:59:08 +0300
Subject: drm/i915: Move sprite/cursor plane disable to intel_sanitize_crtc()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Move the sprite/cursor plane disabling to occur in intel_sanitize_crtc()
where it belongs instead of doing it in intel_modeset_readout_hw_state().

The plane disabling was first added in
4cf0ebbd4fafbdf8e6431dbb315e5511c3efdc3b drm/i915: Rework plane readout.

I got the idea from some patches from Partik and/or Maarten but those
moved also the plane state readout to intel_sanitize_crtc() which isn't
quite right in my opinion.

Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Patrik Jakobsson <patrik.jakobsson@linux.intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
References: https://bugs.freedesktop.org/show_bug.cgi?id=91910
Reviewed-by: Patrik Jakobsson <patrik.jakobsson@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index ce9bb47..5064d8f 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -14905,9 +14905,19 @@ static void intel_sanitize_crtc(struct intel_crtc *crtc)
 	/* restore vblank interrupts to correct state */
 	drm_crtc_vblank_reset(&crtc->base);
 	if (crtc->active) {
+		struct intel_plane *plane;
+
 		drm_calc_timestamping_constants(&crtc->base, &crtc->base.hwmode);
 		update_scanline_offset(crtc);
 		drm_crtc_vblank_on(&crtc->base);
+
+		/* Disable everything but the primary plane */
+		for_each_intel_plane_on_crtc(dev, crtc, plane) {
+			if (plane->base.type == DRM_PLANE_TYPE_PRIMARY)
+				continue;
+
+			plane->disable_plane(&plane->base, &crtc->base);
+		}
 	}
 
 	/* We need to sanitize the plane -> pipe mapping first because this will
@@ -15070,35 +15080,21 @@ void i915_redisable_vga(struct drm_device *dev)
 	i915_redisable_vga_power_on(dev);
 }
 
-static bool primary_get_hw_state(struct intel_crtc *crtc)
+static bool primary_get_hw_state(struct intel_plane *plane)
 {
-	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
+	struct drm_i915_private *dev_priv = to_i915(plane->base.dev);
 
-	return !!(I915_READ(DSPCNTR(crtc->plane)) & DISPLAY_PLANE_ENABLE);
+	return I915_READ(DSPCNTR(plane->plane)) & DISPLAY_PLANE_ENABLE;
 }
 
-static void readout_plane_state(struct intel_crtc *crtc,
-				struct intel_crtc_state *crtc_state)
+/* FIXME read out full plane state for all planes */
+static void readout_plane_state(struct intel_crtc *crtc)
 {
-	struct intel_plane *p;
-	struct intel_plane_state *plane_state;
-	bool active = crtc_state->base.active;
+	struct intel_plane_state *plane_state =
+		to_intel_plane_state(crtc->base.primary->state);
 
-	for_each_intel_plane(crtc->base.dev, p) {
-		if (crtc->pipe != p->pipe)
-			continue;
-
-		plane_state = to_intel_plane_state(p->base.state);
-
-		if (p->base.type == DRM_PLANE_TYPE_PRIMARY)
-			plane_state->visible = primary_get_hw_state(crtc);
-		else {
-			if (active)
-				p->disable_plane(&p->base, &crtc->base);
-
-			plane_state->visible = false;
-		}
-	}
+	plane_state->visible =
+		primary_get_hw_state(to_intel_plane(crtc->base.primary));
 }
 
 static void intel_modeset_readout_hw_state(struct drm_device *dev)
@@ -15121,7 +15117,7 @@ static void intel_modeset_readout_hw_state(struct drm_device *dev)
 		crtc->base.state->active = crtc->active;
 		crtc->base.enabled = crtc->active;
 
-		readout_plane_state(crtc, to_intel_crtc_state(crtc->base.state));
+		readout_plane_state(crtc);
 
 		DRM_DEBUG_KMS("[CRTC:%d] hw state readout: %s\n",
 			      crtc->base.base.id,
-- 
cgit v0.10.2


From 9eca6832f7254d49d25494da7d47c0f8a24f7862 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 10 Sep 2015 18:59:10 +0300
Subject: drm/i915: Move scanline_offset and timestamping constant setup to
 intel_modeset_readout_hw_state()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

intel_modeset_readout_hw_state() seems like the more appropriate place
for populating the scanline_offset and timestamping constants than
intel_sanitize_crtc() since they are basically part of the state we
read out.

Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Patrik Jakobsson <patrik.jakobsson@linux.intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Patrik Jakobsson <patrik.jakobsson@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 5064d8f..8db2b13 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -14907,8 +14907,6 @@ static void intel_sanitize_crtc(struct intel_crtc *crtc)
 	if (crtc->active) {
 		struct intel_plane *plane;
 
-		drm_calc_timestamping_constants(&crtc->base, &crtc->base.hwmode);
-		update_scanline_offset(crtc);
 		drm_crtc_vblank_on(&crtc->base);
 
 		/* Disable everything but the primary plane */
@@ -15204,6 +15202,9 @@ static void intel_modeset_readout_hw_state(struct drm_device *dev)
 			 * recalculation.
 			 */
 			crtc->base.state->mode.private_flags = I915_MODE_FLAG_INHERITED;
+
+			drm_calc_timestamping_constants(&crtc->base, &crtc->base.hwmode);
+			update_scanline_offset(crtc);
 		}
 	}
 }
-- 
cgit v0.10.2


From 3a5a0393ab5de28f8dad100bf604e4ab55550836 Mon Sep 17 00:00:00 2001
From: Jesse Barnes <jbarnes@virtuousgeek.org>
Date: Tue, 15 Sep 2015 10:03:01 -0700
Subject: drm/i915: fix crash in error state readout on non-execlist platforms
 v2

Looks like this was introduced in:
commit d1675198ed1f21aec6e036336e4340c40b726497
Author: Alex Dai <yu.dai@intel.com>
Date:   Wed Aug 12 15:43:43 2015 +0100

    drm/i915: Integrate GuC-based command submission

This patch assumed LRC contexts and HWS layout, which is incorrect on
platforms without execlists.  This can lead to a crash in GPU error
state readout on those platforms.

I don't see a bug filed for this, but there may be one that I haven't
found.

v2: fixup offset handling for error capture fix (Dave)

Cc: Alex Dai <yu.dai@intel.com>
Reviewed-by: Dave Gordon <david.s.gordon@intel.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
index 3379f9c..f95de05 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.c
+++ b/drivers/gpu/drm/i915/i915_gpu_error.c
@@ -457,17 +457,23 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m,
 		}
 
 		if ((obj = error->ring[i].hws_page)) {
+			u64 hws_offset = obj->gtt_offset;
+			u32 *hws_page = &obj->pages[0][0];
+
+			if (i915.enable_execlists) {
+				hws_offset += LRC_PPHWSP_PN * PAGE_SIZE;
+				hws_page = &obj->pages[LRC_PPHWSP_PN][0];
+			}
 			err_printf(m, "%s --- HW Status = 0x%08llx\n",
-				dev_priv->ring[i].name,
-				obj->gtt_offset + LRC_PPHWSP_PN * PAGE_SIZE);
+				   dev_priv->ring[i].name, hws_offset);
 			offset = 0;
 			for (elt = 0; elt < PAGE_SIZE/16; elt += 4) {
 				err_printf(m, "[%04x] %08x %08x %08x %08x\n",
 					   offset,
-					   obj->pages[LRC_PPHWSP_PN][elt],
-					   obj->pages[LRC_PPHWSP_PN][elt+1],
-					   obj->pages[LRC_PPHWSP_PN][elt+2],
-					   obj->pages[LRC_PPHWSP_PN][elt+3]);
+					   hws_page[elt],
+					   hws_page[elt+1],
+					   hws_page[elt+2],
+					   hws_page[elt+3]);
 					offset += 16;
 			}
 		}
-- 
cgit v0.10.2


From 7a58bad0e63295dfa803973efcebc80cb730c7bd Mon Sep 17 00:00:00 2001
From: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Date: Sat, 12 Sep 2015 10:17:50 +0530
Subject: drm/i915: Add IS_SKL_GT3 and IS_SKL_GT4 macro.

It will be usefull to specify w/a that affects only SKL GT3 and GT4.

Signed-off-by: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Reviewed-by: Alex Dai <yu.dai@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 3795fdf..7cbc81f 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -2488,6 +2488,11 @@ struct drm_i915_cmd_table {
 #define IS_SKL_ULX(dev)		(INTEL_DEVID(dev) == 0x190E || \
 				 INTEL_DEVID(dev) == 0x1915 || \
 				 INTEL_DEVID(dev) == 0x191E)
+#define IS_SKL_GT3(dev)		(IS_SKYLAKE(dev) && \
+				 (INTEL_DEVID(dev) & 0x00F0) == 0x0020)
+#define IS_SKL_GT4(dev)		(IS_SKYLAKE(dev) && \
+				 (INTEL_DEVID(dev) & 0x00F0) == 0x0030)
+
 #define IS_PRELIMINARY_HW(intel_info) ((intel_info)->is_preliminary)
 
 #define SKL_REVID_A0		(0x0)
-- 
cgit v0.10.2


From f2d2fe95072acd5404f8051b8bf1195c61a47fb5 Mon Sep 17 00:00:00 2001
From: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Date: Sat, 12 Sep 2015 10:17:51 +0530
Subject: drm/i915: WaRsDisableCoarsePowerGating

WaRsDisableCoarsePowerGating: Coarse Power Gating (CPG) needs to be
disabled for platforms prior to BXT B0 and SKL GT3/GT4 till E0.

v2: Added GT3/GT4 Check.

Change-Id: Ia3c4c16e050c88d3e259f601054875c812d69c3a
Signed-off-by: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Reviewed-by: Alex Dai <yu.dai@intel.com>
[danvet: Align continuation properly.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index f52431b..09be8fc 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -4877,11 +4877,14 @@ static void gen9_enable_rc6(struct drm_device *dev)
 
 	/*
 	 * 3b: Enable Coarse Power Gating only when RC6 is enabled.
-	 * WaDisableRenderPowerGating:skl,bxt - Render PG need to be disabled with RC6.
+	 * WaRsDisableCoarsePowerGating:skl,bxt - Render/Media PG need to be disabled with RC6.
 	 */
-	I915_WRITE(GEN9_PG_ENABLE, (rc6_mask & GEN6_RC_CTL_RC6_ENABLE) ?
-			GEN9_MEDIA_PG_ENABLE : 0);
-
+	if ((IS_BROXTON(dev) && (INTEL_REVID(dev) < BXT_REVID_B0)) ||
+	    ((IS_SKL_GT3(dev) || IS_SKL_GT4(dev)) && (INTEL_REVID(dev) <= SKL_REVID_E0)))
+		I915_WRITE(GEN9_PG_ENABLE, 0);
+	else
+		I915_WRITE(GEN9_PG_ENABLE, (rc6_mask & GEN6_RC_CTL_RC6_ENABLE) ?
+				(GEN9_RENDER_PG_ENABLE | GEN9_MEDIA_PG_ENABLE) : 0);
 
 	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
 
-- 
cgit v0.10.2


From e3429cd240b06c79df3ea90f28065a7e011744cd Mon Sep 17 00:00:00 2001
From: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Date: Sat, 12 Sep 2015 10:17:52 +0530
Subject: drm/i915: WaRsUseTimeoutMode

Enable TO mode for RC6 for SKL till D0 and BXT till A0.

Cc: Tom O'Rourke <Tom.O'Rourke@intel.com>
Cc: Akash Goel <akash.goel@intel.com>
Signed-off-by: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Reviewed-by: Alex Dai <yu.dai@intel.com>
[danvet: Fixup line continuation alignment.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index 09be8fc..b4e0d15 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -4871,9 +4871,16 @@ static void gen9_enable_rc6(struct drm_device *dev)
 		rc6_mask = GEN6_RC_CTL_RC6_ENABLE;
 	DRM_INFO("RC6 %s\n", (rc6_mask & GEN6_RC_CTL_RC6_ENABLE) ?
 			"on" : "off");
-	I915_WRITE(GEN6_RC_CONTROL, GEN6_RC_CTL_HW_ENABLE |
-				   GEN6_RC_CTL_EI_MODE(1) |
-				   rc6_mask);
+
+	if ((IS_SKYLAKE(dev) && INTEL_REVID(dev) <= SKL_REVID_D0) ||
+	    (IS_BROXTON(dev) && INTEL_REVID(dev) <= BXT_REVID_A0))
+		I915_WRITE(GEN6_RC_CONTROL, GEN6_RC_CTL_HW_ENABLE |
+			   GEN7_RC_CTL_TO_MODE |
+			   rc6_mask);
+	else
+		I915_WRITE(GEN6_RC_CONTROL, GEN6_RC_CTL_HW_ENABLE |
+			   GEN6_RC_CTL_EI_MODE(1) |
+			   rc6_mask);
 
 	/*
 	 * 3b: Enable Coarse Power Gating only when RC6 is enabled.
-- 
cgit v0.10.2


From 63a4dec2c168b74a39df1eac494501f0f6bf3708 Mon Sep 17 00:00:00 2001
From: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Date: Sat, 12 Sep 2015 10:17:53 +0530
Subject: drm/i915: WaRsDoubleRc6WrlWithCoarsePowerGating

Cc: Tom O'Rourke <Tom.O'Rourke@intel.com>
Cc: Akash Goel <akash.goel@intel.com>
Signed-off-by: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Reviewed-by: Alex Dai <yu.dai@intel.com>
[danvet: Fix continuation alignment.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index b4e0d15..61162b2 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -4854,7 +4854,13 @@ static void gen9_enable_rc6(struct drm_device *dev)
 	I915_WRITE(GEN6_RC_CONTROL, 0);
 
 	/* 2b: Program RC6 thresholds.*/
-	I915_WRITE(GEN6_RC6_WAKE_RATE_LIMIT, 54 << 16);
+
+	/* WaRsDoubleRc6WrlWithCoarsePowerGating: Doubling WRL only when CPG is enabled */
+	if (IS_SKYLAKE(dev) && !((IS_SKL_GT3(dev) || IS_SKL_GT4(dev)) &&
+				 (INTEL_REVID(dev) <= SKL_REVID_E0)))
+		I915_WRITE(GEN6_RC6_WAKE_RATE_LIMIT, 108 << 16);
+	else
+		I915_WRITE(GEN6_RC6_WAKE_RATE_LIMIT, 54 << 16);
 	I915_WRITE(GEN6_RC_EVALUATION_INTERVAL, 125000); /* 12500 * 1280ns */
 	I915_WRITE(GEN6_RC_IDLE_HYSTERSIS, 25); /* 25 * 1280ns */
 	for_each_ring(ring, dev_priv, unused)
-- 
cgit v0.10.2


From 97c322e713b92b5071ad276f65cabf8a503a7408 Mon Sep 17 00:00:00 2001
From: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Date: Sat, 12 Sep 2015 10:17:54 +0530
Subject: drm/i915: Program GuC MAX IDLE Count

Cc: Alex Dai <yu.dai@intel.com>
Cc: Tom O'Rourke <Tom.O'Rourke@intel.com>
Cc: Akash Goel <akash.goel@intel.com>
Signed-off-by: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Reviewed-by: Alex Dai <yu.dai@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_guc_reg.h b/drivers/gpu/drm/i915/i915_guc_reg.h
index 8c8e574..9d79a6b 100644
--- a/drivers/gpu/drm/i915/i915_guc_reg.h
+++ b/drivers/gpu/drm/i915/i915_guc_reg.h
@@ -53,6 +53,7 @@
 #define   START_DMA			  (1<<0)
 #define DMA_GUC_WOPCM_OFFSET		0xc340
 #define   GUC_WOPCM_OFFSET_VALUE	  0x80000	/* 512KB */
+#define GUC_MAX_IDLE_COUNT		0xC3E4
 
 #define GUC_WOPCM_SIZE			0xc050
 #define   GUC_WOPCM_SIZE_VALUE  	  (0x80 << 12)	/* 512KB */
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index 61162b2..566f5cd 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -4865,6 +4865,10 @@ static void gen9_enable_rc6(struct drm_device *dev)
 	I915_WRITE(GEN6_RC_IDLE_HYSTERSIS, 25); /* 25 * 1280ns */
 	for_each_ring(ring, dev_priv, unused)
 		I915_WRITE(RING_MAX_IDLE(ring->mmio_base), 10);
+
+	if (HAS_GUC_UCODE(dev))
+		I915_WRITE(GUC_MAX_IDLE_COUNT, 0xA);
+
 	I915_WRITE(GEN6_RC_SLEEP, 0);
 	I915_WRITE(GEN6_RC6_THRESHOLD, 37500); /* 37.5/125ms per EI */
 
-- 
cgit v0.10.2


From 7aefe2b50a0be5c85e7702bff110e5e187fbe4b5 Mon Sep 17 00:00:00 2001
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Date: Mon, 14 Sep 2015 11:30:10 +0200
Subject: drm/i915: Fix fastboot scalers for skylake.

The scaler_id in intel_pipe_config_compare should not be checked
when adjusting in intel_pipe_config_compare. The hw scaler id may
be changed in intel_update_pipe_config.

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 8db2b13..6d718c0 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -12459,9 +12459,9 @@ intel_pipe_config_compare(struct drm_device *dev,
 			PIPE_CONF_CHECK_X(pch_pfit.pos);
 			PIPE_CONF_CHECK_X(pch_pfit.size);
 		}
-	}
 
-	PIPE_CONF_CHECK_I(scaler_state.scaler_id);
+		PIPE_CONF_CHECK_I(scaler_state.scaler_id);
+	}
 
 	/* BDW+ don't expose a synchronous way to read the state */
 	if (IS_HASWELL(dev))
-- 
cgit v0.10.2


From e8cb8d69d125f56fa3ba5239b215a56718e2ca44 Mon Sep 17 00:00:00 2001
From: Paulo Zanoni <paulo.r.zanoni@intel.com>
Date: Mon, 14 Sep 2015 15:19:55 -0300
Subject: drm/i915: fix the FBC work allocation failure path
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Always update the currrent crtc, fb and vertical offset after calling
enable_fbc. We were forgetting to do so along the failure paths when
enabling fbc synchronously. Fix this with a new helper to enable_fbc()
and update the state simultaneously.

v2: Improve commit message (Chris).
v3: Constify struct drm_framebuffer (Ville).

Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_fbc.c b/drivers/gpu/drm/i915/intel_fbc.c
index 1f97fb5..9e42079 100644
--- a/drivers/gpu/drm/i915/intel_fbc.c
+++ b/drivers/gpu/drm/i915/intel_fbc.c
@@ -308,6 +308,18 @@ bool intel_fbc_enabled(struct drm_i915_private *dev_priv)
 	return dev_priv->fbc.enabled;
 }
 
+static void intel_fbc_enable(struct intel_crtc *crtc,
+			     const struct drm_framebuffer *fb)
+{
+	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
+
+	dev_priv->fbc.enable_fbc(crtc);
+
+	dev_priv->fbc.crtc = crtc;
+	dev_priv->fbc.fb_id = fb->base.id;
+	dev_priv->fbc.y = crtc->base.y;
+}
+
 static void intel_fbc_work_fn(struct work_struct *__work)
 {
 	struct intel_fbc_work *work =
@@ -321,13 +333,8 @@ static void intel_fbc_work_fn(struct work_struct *__work)
 		/* Double check that we haven't switched fb without cancelling
 		 * the prior work.
 		 */
-		if (crtc_fb == work->fb) {
-			dev_priv->fbc.enable_fbc(work->crtc);
-
-			dev_priv->fbc.crtc = work->crtc;
-			dev_priv->fbc.fb_id = crtc_fb->base.id;
-			dev_priv->fbc.y = work->crtc->base.y;
-		}
+		if (crtc_fb == work->fb)
+			intel_fbc_enable(work->crtc, work->fb);
 
 		dev_priv->fbc.fbc_work = NULL;
 	}
@@ -361,7 +368,7 @@ static void intel_fbc_cancel_work(struct drm_i915_private *dev_priv)
 	dev_priv->fbc.fbc_work = NULL;
 }
 
-static void intel_fbc_enable(struct intel_crtc *crtc)
+static void intel_fbc_schedule_enable(struct intel_crtc *crtc)
 {
 	struct intel_fbc_work *work;
 	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
@@ -373,7 +380,7 @@ static void intel_fbc_enable(struct intel_crtc *crtc)
 	work = kzalloc(sizeof(*work), GFP_KERNEL);
 	if (work == NULL) {
 		DRM_ERROR("Failed to allocate FBC work structure\n");
-		dev_priv->fbc.enable_fbc(crtc);
+		intel_fbc_enable(crtc, crtc->base.primary->fb);
 		return;
 	}
 
@@ -824,7 +831,7 @@ static void __intel_fbc_update(struct drm_i915_private *dev_priv)
 		__intel_fbc_disable(dev_priv);
 	}
 
-	intel_fbc_enable(intel_crtc);
+	intel_fbc_schedule_enable(intel_crtc);
 	dev_priv->fbc.no_fbc_reason = FBC_OK;
 	return;
 
-- 
cgit v0.10.2


From adf70c65cff66d843d66469dbe13ef1ed21e062b Mon Sep 17 00:00:00 2001
From: Paulo Zanoni <paulo.r.zanoni@intel.com>
Date: Mon, 14 Sep 2015 15:19:56 -0300
Subject: drm/i915: check for the supported strides on HSW+ FBC
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Don't allow FBC for cases where the spec says we can't FBC.

v2:
  - Just WARN_ON() the strides that should have been caught earlier
    (Daniel)
  - Make it a new function since I expect this to grow more.
v3:
  - Document which IGT test is exercised by this.
v4:
  - Implement the restrictions for gens 2-6 too (Ville).
  - Fix off-by-one mistake (Ville).

Testcase: igt/kms_frontbuffer_tracking/fbc-badstride
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 7cbc81f..d4240db 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -948,6 +948,7 @@ struct i915_fbc {
 		FBC_CHIP_DEFAULT, /* disabled by default on this chip */
 		FBC_ROTATION, /* rotation is not supported */
 		FBC_IN_DBG_MASTER, /* kernel debugger is active */
+		FBC_BAD_STRIDE, /* stride is not supported */
 	} no_fbc_reason;
 
 	bool (*fbc_enabled)(struct drm_i915_private *dev_priv);
diff --git a/drivers/gpu/drm/i915/intel_fbc.c b/drivers/gpu/drm/i915/intel_fbc.c
index 9e42079..db38091 100644
--- a/drivers/gpu/drm/i915/intel_fbc.c
+++ b/drivers/gpu/drm/i915/intel_fbc.c
@@ -480,6 +480,8 @@ const char *intel_no_fbc_reason_str(enum no_fbc_reason reason)
 		return "rotation unsupported";
 	case FBC_IN_DBG_MASTER:
 		return "Kernel debugger is active";
+	case FBC_BAD_STRIDE:
+		return "framebuffer stride not supported";
 	default:
 		MISSING_CASE(reason);
 		return "unknown reason";
@@ -671,6 +673,27 @@ static int intel_fbc_setup_cfb(struct drm_i915_private *dev_priv, int size,
 	return intel_fbc_alloc_cfb(dev_priv, size, fb_cpp);
 }
 
+static bool stride_is_valid(struct drm_i915_private *dev_priv,
+			    unsigned int stride)
+{
+	/* These should have been caught earlier. */
+	WARN_ON(stride < 512);
+	WARN_ON((stride & (64 - 1)) != 0);
+
+	/* Below are the additional FBC restrictions. */
+
+	if (IS_GEN2(dev_priv) || IS_GEN3(dev_priv))
+		return stride == 4096 || stride == 8192;
+
+	if (IS_GEN4(dev_priv) && !IS_G4X(dev_priv) && stride < 2048)
+		return false;
+
+	if (stride > 16384)
+		return false;
+
+	return true;
+}
+
 /**
  * __intel_fbc_update - enable/disable FBC as needed, unlocked
  * @dev_priv: i915 device instance
@@ -781,6 +804,11 @@ static void __intel_fbc_update(struct drm_i915_private *dev_priv)
 		goto out_disable;
 	}
 
+	if (!stride_is_valid(dev_priv, fb->pitches[0])) {
+		set_no_fbc_reason(dev_priv, FBC_BAD_STRIDE);
+		goto out_disable;
+	}
+
 	/* If the kernel debugger is active, always disable compression */
 	if (in_dbg_master()) {
 		set_no_fbc_reason(dev_priv, FBC_IN_DBG_MASTER);
-- 
cgit v0.10.2


From a9da512b3ed73045253afd778e40d4298f42905b Mon Sep 17 00:00:00 2001
From: Paulo Zanoni <paulo.r.zanoni@intel.com>
Date: Mon, 14 Sep 2015 15:19:57 -0300
Subject: drm/i915: avoid the last 8mb of stolen on BDW/SKL
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The FBC hardware for these platforms doesn't have access to the
bios_reserved range, so it always assumes the maximum (8mb) is used.
So avoid this range while allocating.

This solves a bunch of FIFO underruns that happen if you end up
putting the CFB in that memory range. On my machine, with 32mb of
stolen, I need a 2560x1440 mode for that.

Testcase: igt/kms_frontbuffer_tracking/fbc-* (given the right setup)
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index d4240db..ffd4ab1 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -3175,6 +3175,10 @@ static inline void i915_gem_chipset_flush(struct drm_device *dev)
 int i915_gem_stolen_insert_node(struct drm_i915_private *dev_priv,
 				struct drm_mm_node *node, u64 size,
 				unsigned alignment);
+int i915_gem_stolen_insert_node_in_range(struct drm_i915_private *dev_priv,
+					 struct drm_mm_node *node, u64 size,
+					 unsigned alignment, u64 start,
+					 u64 end);
 void i915_gem_stolen_remove_node(struct drm_i915_private *dev_priv,
 				 struct drm_mm_node *node);
 int i915_gem_init_stolen(struct drm_device *dev);
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.h b/drivers/gpu/drm/i915/i915_gem_gtt.h
index 8275007..96ebb98 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.h
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.h
@@ -341,6 +341,7 @@ struct i915_gtt {
 	struct i915_address_space base;
 
 	size_t stolen_size;		/* Total size of stolen memory */
+	size_t stolen_usable_size;	/* Total size minus BIOS reserved */
 	u64 mappable_end;		/* End offset that we can CPU map */
 	struct io_mapping *mappable;	/* Mapping to our CPU mappable region */
 	phys_addr_t mappable_base;	/* PA of our GMADR */
diff --git a/drivers/gpu/drm/i915/i915_gem_stolen.c b/drivers/gpu/drm/i915/i915_gem_stolen.c
index bf26ecc..081ef6d 100644
--- a/drivers/gpu/drm/i915/i915_gem_stolen.c
+++ b/drivers/gpu/drm/i915/i915_gem_stolen.c
@@ -42,9 +42,9 @@
  * for is a boon.
  */
 
-int i915_gem_stolen_insert_node(struct drm_i915_private *dev_priv,
-				struct drm_mm_node *node, u64 size,
-				unsigned alignment)
+int i915_gem_stolen_insert_node_in_range(struct drm_i915_private *dev_priv,
+					 struct drm_mm_node *node, u64 size,
+					 unsigned alignment, u64 start, u64 end)
 {
 	int ret;
 
@@ -52,13 +52,23 @@ int i915_gem_stolen_insert_node(struct drm_i915_private *dev_priv,
 		return -ENODEV;
 
 	mutex_lock(&dev_priv->mm.stolen_lock);
-	ret = drm_mm_insert_node(&dev_priv->mm.stolen, node, size, alignment,
-				 DRM_MM_SEARCH_DEFAULT);
+	ret = drm_mm_insert_node_in_range(&dev_priv->mm.stolen, node, size,
+					  alignment, start, end,
+					  DRM_MM_SEARCH_DEFAULT);
 	mutex_unlock(&dev_priv->mm.stolen_lock);
 
 	return ret;
 }
 
+int i915_gem_stolen_insert_node(struct drm_i915_private *dev_priv,
+				struct drm_mm_node *node, u64 size,
+				unsigned alignment)
+{
+	return i915_gem_stolen_insert_node_in_range(dev_priv, node, size,
+					alignment, 0,
+					dev_priv->gtt.stolen_usable_size);
+}
+
 void i915_gem_stolen_remove_node(struct drm_i915_private *dev_priv,
 				 struct drm_mm_node *node)
 {
@@ -355,9 +365,11 @@ int i915_gem_init_stolen(struct drm_device *dev)
 		      dev_priv->gtt.stolen_size >> 10,
 		      (dev_priv->gtt.stolen_size - reserved_total) >> 10);
 
+	dev_priv->gtt.stolen_usable_size = dev_priv->gtt.stolen_size -
+					   reserved_total;
+
 	/* Basic memrange allocator for stolen space */
-	drm_mm_init(&dev_priv->mm.stolen, 0, dev_priv->gtt.stolen_size -
-		    reserved_total);
+	drm_mm_init(&dev_priv->mm.stolen, 0, dev_priv->gtt.stolen_usable_size);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/i915/intel_fbc.c b/drivers/gpu/drm/i915/intel_fbc.c
index db38091..6f3c2ea 100644
--- a/drivers/gpu/drm/i915/intel_fbc.c
+++ b/drivers/gpu/drm/i915/intel_fbc.c
@@ -551,6 +551,16 @@ static int find_compression_threshold(struct drm_i915_private *dev_priv,
 {
 	int compression_threshold = 1;
 	int ret;
+	u64 end;
+
+	/* The FBC hardware for BDW/SKL doesn't have access to the stolen
+	 * reserved range size, so it always assumes the maximum (8mb) is used.
+	 * If we enable FBC using a CFB on that memory range we'll get FIFO
+	 * underruns, even if that range is not reserved by the BIOS. */
+	if (IS_BROADWELL(dev_priv) || IS_SKYLAKE(dev_priv))
+		end = dev_priv->gtt.stolen_size - 8 * 1024 * 1024;
+	else
+		end = dev_priv->gtt.stolen_usable_size;
 
 	/* HACK: This code depends on what we will do in *_enable_fbc. If that
 	 * code changes, this code needs to change as well.
@@ -560,7 +570,8 @@ static int find_compression_threshold(struct drm_i915_private *dev_priv,
 	 */
 
 	/* Try to over-allocate to reduce reallocations and fragmentation. */
-	ret = i915_gem_stolen_insert_node(dev_priv, node, size <<= 1, 4096);
+	ret = i915_gem_stolen_insert_node_in_range(dev_priv, node, size <<= 1,
+						   4096, 0, end);
 	if (ret == 0)
 		return compression_threshold;
 
@@ -570,7 +581,8 @@ again:
 	    (fb_cpp == 2 && compression_threshold == 2))
 		return 0;
 
-	ret = i915_gem_stolen_insert_node(dev_priv, node, size >>= 1, 4096);
+	ret = i915_gem_stolen_insert_node_in_range(dev_priv, node, size >>= 1,
+						   4096, 0, end);
 	if (ret && INTEL_INFO(dev_priv)->gen <= 4) {
 		return 0;
 	} else if (ret) {
-- 
cgit v0.10.2


From b8bf5d7fe0dee64ed0a73fa1e95ba8e464d605da Mon Sep 17 00:00:00 2001
From: Paulo Zanoni <paulo.r.zanoni@intel.com>
Date: Mon, 14 Sep 2015 15:19:58 -0300
Subject: drm/i915: print the correct amount of bytes allocated for the CFB
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

And also print the threshold. I was surprised to see a log message
claiming the CFB size was 32mb when there was less than 24mb available
for it.

Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_fbc.c b/drivers/gpu/drm/i915/intel_fbc.c
index 6f3c2ea..69726a7 100644
--- a/drivers/gpu/drm/i915/intel_fbc.c
+++ b/drivers/gpu/drm/i915/intel_fbc.c
@@ -634,8 +634,9 @@ static int intel_fbc_alloc_cfb(struct drm_i915_private *dev_priv, int size,
 
 	dev_priv->fbc.uncompressed_size = size;
 
-	DRM_DEBUG_KMS("reserved %d bytes of contiguous stolen space for FBC\n",
-		      size);
+	DRM_DEBUG_KMS("reserved %llu bytes of contiguous stolen space for FBC, threshold: %d\n",
+		      dev_priv->fbc.compressed_fb.size,
+		      dev_priv->fbc.threshold);
 
 	return 0;
 
-- 
cgit v0.10.2


From 7b24c9a696c1c68eaa471a27bf467e97a9986fa9 Mon Sep 17 00:00:00 2001
From: Paulo Zanoni <paulo.r.zanoni@intel.com>
Date: Mon, 14 Sep 2015 15:19:59 -0300
Subject: drm/i915: don't enable FBC when pixel rate exceeds 95% on HSW/BDW
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

BSpec says we shouldn't enable FBC on HSW/BDW when the pipe pixel rate
exceeds 95% of the core display clock.

v2:
  - HSW also needs the WA (Ville).
  - Add the WA name (Ville).
  - Use the current cdclk (Ville).

Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index ffd4ab1..e5242c0 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -949,6 +949,7 @@ struct i915_fbc {
 		FBC_ROTATION, /* rotation is not supported */
 		FBC_IN_DBG_MASTER, /* kernel debugger is active */
 		FBC_BAD_STRIDE, /* stride is not supported */
+		FBC_PIXEL_RATE, /* pixel rate is too big */
 	} no_fbc_reason;
 
 	bool (*fbc_enabled)(struct drm_i915_private *dev_priv);
diff --git a/drivers/gpu/drm/i915/intel_fbc.c b/drivers/gpu/drm/i915/intel_fbc.c
index 69726a7..1c4536a 100644
--- a/drivers/gpu/drm/i915/intel_fbc.c
+++ b/drivers/gpu/drm/i915/intel_fbc.c
@@ -482,6 +482,8 @@ const char *intel_no_fbc_reason_str(enum no_fbc_reason reason)
 		return "Kernel debugger is active";
 	case FBC_BAD_STRIDE:
 		return "framebuffer stride not supported";
+	case FBC_PIXEL_RATE:
+		return "pixel rate is too big";
 	default:
 		MISSING_CASE(reason);
 		return "unknown reason";
@@ -828,6 +830,14 @@ static void __intel_fbc_update(struct drm_i915_private *dev_priv)
 		goto out_disable;
 	}
 
+	/* WaFbcExceedCdClockThreshold:hsw,bdw */
+	if ((IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv)) &&
+	    ilk_pipe_pixel_rate(intel_crtc->config) >=
+	    dev_priv->cdclk_freq * 95 / 100) {
+		set_no_fbc_reason(dev_priv, FBC_PIXEL_RATE);
+		goto out_disable;
+	}
+
 	if (intel_fbc_setup_cfb(dev_priv, obj->base.size,
 				drm_format_plane_cpp(fb->pixel_format, 0))) {
 		set_no_fbc_reason(dev_priv, FBC_STOLEN_TOO_SMALL);
-- 
cgit v0.10.2


From 57012be9289b4f6d12b05878411d903982ec2a4e Mon Sep 17 00:00:00 2001
From: Paulo Zanoni <paulo.r.zanoni@intel.com>
Date: Mon, 14 Sep 2015 15:20:00 -0300
Subject: drm/i915: apply WaFbcAsynchFlipDisableFbcQueue earlier
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The spec says the register should have that value for the entire time
that FBC is enabled, so apply the WA before we enable FBC.

Notice that we also have this WA for ILK/SNB, but it is implemented at
init_clock_gating(). I could move the IVB/HSW/BDW WA code to
init_clock_gating() too, but since we recently had some complaints
about WAs not staying after being set, I'm going to play safe and keep
this here for now.

Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_fbc.c b/drivers/gpu/drm/i915/intel_fbc.c
index 1c4536a..2b75003 100644
--- a/drivers/gpu/drm/i915/intel_fbc.c
+++ b/drivers/gpu/drm/i915/intel_fbc.c
@@ -272,8 +272,6 @@ static void gen7_fbc_enable(struct intel_crtc *crtc)
 	if (dev_priv->fbc.false_color)
 		dpfc_ctl |= FBC_CTL_FALSE_COLOR;
 
-	I915_WRITE(ILK_DPFC_CONTROL, dpfc_ctl | DPFC_CTL_EN);
-
 	if (IS_IVYBRIDGE(dev_priv)) {
 		/* WaFbcAsynchFlipDisableFbcQueue:ivb */
 		I915_WRITE(ILK_DISPLAY_CHICKEN1,
@@ -286,6 +284,8 @@ static void gen7_fbc_enable(struct intel_crtc *crtc)
 			   HSW_FBCQ_DIS);
 	}
 
+	I915_WRITE(ILK_DPFC_CONTROL, dpfc_ctl | DPFC_CTL_EN);
+
 	I915_WRITE(SNB_DPFC_CTL_SA,
 		   SNB_CPU_FENCE_ENABLE | obj->fence_reg);
 	I915_WRITE(DPFC_CPU_FENCE_OFFSET, crtc->base.y);
-- 
cgit v0.10.2


From 40f4022ef67f30cc0b7ee7453926d1803277d1af Mon Sep 17 00:00:00 2001
From: Paulo Zanoni <paulo.r.zanoni@intel.com>
Date: Mon, 14 Sep 2015 15:20:01 -0300
Subject: drm/i915: don't apply WaFbcAsynchFlipDisableFbcQueue on SKL
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This WA is only for HSW/BDW.

Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_fbc.c b/drivers/gpu/drm/i915/intel_fbc.c
index 2b75003..56cf110 100644
--- a/drivers/gpu/drm/i915/intel_fbc.c
+++ b/drivers/gpu/drm/i915/intel_fbc.c
@@ -277,7 +277,7 @@ static void gen7_fbc_enable(struct intel_crtc *crtc)
 		I915_WRITE(ILK_DISPLAY_CHICKEN1,
 			   I915_READ(ILK_DISPLAY_CHICKEN1) |
 			   ILK_FBCQ_DIS);
-	} else {
+	} else if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv)) {
 		/* WaFbcAsynchFlipDisableFbcQueue:hsw,bdw */
 		I915_WRITE(CHICKEN_PIPESL_1(crtc->pipe),
 			   I915_READ(CHICKEN_PIPESL_1(crtc->pipe)) |
-- 
cgit v0.10.2


From b9e831dc3973bddfaa8e27629745c5948ed8b92d Mon Sep 17 00:00:00 2001
From: Paulo Zanoni <paulo.r.zanoni@intel.com>
Date: Mon, 21 Sep 2015 19:48:06 -0300
Subject: drm/i915: reject invalid formats for FBC
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This commit is essentially a rewrite of "drm/i915: Check pixel format
for fbc" from Ville Syrjälä. The idea is the same, but the code is
different due to all the changes that happened since his original
patch. So any bugs are due to my bad rewrite.

v2:
  - Drop the alpha formats (Ville).
v3:
  - Drop the stale comment (Ville).

Testcases: igt/kms_frontbuffer_tracking/*fbc*-${format_name}-draw-*
Credits-to: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index e5242c0..3044cbe 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -950,6 +950,7 @@ struct i915_fbc {
 		FBC_IN_DBG_MASTER, /* kernel debugger is active */
 		FBC_BAD_STRIDE, /* stride is not supported */
 		FBC_PIXEL_RATE, /* pixel rate is too big */
+		FBC_PIXEL_FORMAT /* pixel format is invalid */
 	} no_fbc_reason;
 
 	bool (*fbc_enabled)(struct drm_i915_private *dev_priv);
diff --git a/drivers/gpu/drm/i915/intel_fbc.c b/drivers/gpu/drm/i915/intel_fbc.c
index 56cf110..9d2f56e 100644
--- a/drivers/gpu/drm/i915/intel_fbc.c
+++ b/drivers/gpu/drm/i915/intel_fbc.c
@@ -484,6 +484,8 @@ const char *intel_no_fbc_reason_str(enum no_fbc_reason reason)
 		return "framebuffer stride not supported";
 	case FBC_PIXEL_RATE:
 		return "pixel rate is too big";
+	case FBC_PIXEL_FORMAT:
+		return "pixel format is invalid";
 	default:
 		MISSING_CASE(reason);
 		return "unknown reason";
@@ -709,6 +711,29 @@ static bool stride_is_valid(struct drm_i915_private *dev_priv,
 	return true;
 }
 
+static bool pixel_format_is_valid(struct drm_framebuffer *fb)
+{
+	struct drm_device *dev = fb->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	switch (fb->pixel_format) {
+	case DRM_FORMAT_XRGB8888:
+	case DRM_FORMAT_XBGR8888:
+		return true;
+	case DRM_FORMAT_XRGB1555:
+	case DRM_FORMAT_RGB565:
+		/* 16bpp not supported on gen2 */
+		if (IS_GEN2(dev))
+			return false;
+		/* WaFbcOnly1to1Ratio:ctg */
+		if (IS_G4X(dev_priv))
+			return false;
+		return true;
+	default:
+		return false;
+	}
+}
+
 /**
  * __intel_fbc_update - enable/disable FBC as needed, unlocked
  * @dev_priv: i915 device instance
@@ -824,6 +849,11 @@ static void __intel_fbc_update(struct drm_i915_private *dev_priv)
 		goto out_disable;
 	}
 
+	if (!pixel_format_is_valid(fb)) {
+		set_no_fbc_reason(dev_priv, FBC_PIXEL_FORMAT);
+		goto out_disable;
+	}
+
 	/* If the kernel debugger is active, always disable compression */
 	if (in_dbg_master()) {
 		set_no_fbc_reason(dev_priv, FBC_IN_DBG_MASTER);
-- 
cgit v0.10.2


From 2db3366b18e6ee5c6cb09b5f3902bcacfa3d534e Mon Sep 17 00:00:00 2001
From: Paulo Zanoni <paulo.r.zanoni@intel.com>
Date: Mon, 14 Sep 2015 15:20:03 -0300
Subject: drm/i915: fix FBC for cases where crtc->base.y is non-zero
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

I only tested this on BDW and SKL, but since the register description
is the same ever since gen4, let's assume that all gens take the same
register format. If that's not true, then hopefully someone will
bisect a bug to this patch and we'll fix it.

Notice that the wrong fence offset register just means that the
hardware tracking will be wrong.

Testcases:
 - igt/kms_frontbuffer_tracking/fbc-1p-primscrn-pri-shrfb-draw-mmap-gtt
 - igt/kms_frontbuffer_tracking/fbc-2p-primscrn-pri-shrfb-draw-mmap-gtt

v2:
  - Add intel_crtc->adjusted_{x,y} so this code can work independently
    of intel_gen4_compute_page_offset(). (Ville).
  - This version also works on SKL.

Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 6d718c0..bf8754f 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -2730,6 +2730,9 @@ static void i9xx_update_primary_plane(struct drm_crtc *crtc,
 			(intel_crtc->config->pipe_src_w - 1) * pixel_size;
 	}
 
+	intel_crtc->adjusted_x = x;
+	intel_crtc->adjusted_y = y;
+
 	I915_WRITE(reg, dspcntr);
 
 	I915_WRITE(DSPSTRIDE(plane), fb->pitches[0]);
@@ -2830,6 +2833,9 @@ static void ironlake_update_primary_plane(struct drm_crtc *crtc,
 		}
 	}
 
+	intel_crtc->adjusted_x = x;
+	intel_crtc->adjusted_y = y;
+
 	I915_WRITE(reg, dspcntr);
 
 	I915_WRITE(DSPSTRIDE(plane), fb->pitches[0]);
@@ -3082,6 +3088,9 @@ static void skylake_update_primary_plane(struct drm_crtc *crtc,
 	}
 	plane_offset = y_offset << 16 | x_offset;
 
+	intel_crtc->adjusted_x = x_offset;
+	intel_crtc->adjusted_y = y_offset;
+
 	I915_WRITE(PLANE_CTL(pipe, 0), plane_ctl);
 	I915_WRITE(PLANE_OFFSET(pipe, 0), plane_offset);
 	I915_WRITE(PLANE_SIZE(pipe, 0), plane_size);
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 394018e..ab3c296 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -537,6 +537,8 @@ struct intel_crtc {
 	 * gen4+ this only adjusts up to a tile, offsets within a tile are
 	 * handled in the hw itself (with the TILEOFF register). */
 	unsigned long dspaddr_offset;
+	int adjusted_x;
+	int adjusted_y;
 
 	struct drm_i915_gem_object *cursor_bo;
 	uint32_t cursor_addr;
diff --git a/drivers/gpu/drm/i915/intel_fbc.c b/drivers/gpu/drm/i915/intel_fbc.c
index 9d2f56e..d38f464 100644
--- a/drivers/gpu/drm/i915/intel_fbc.c
+++ b/drivers/gpu/drm/i915/intel_fbc.c
@@ -41,6 +41,19 @@
 #include "intel_drv.h"
 #include "i915_drv.h"
 
+/*
+ * In some platforms where the CRTC's x:0/y:0 coordinates doesn't match the
+ * frontbuffer's x:0/y:0 coordinates we lie to the hardware about the plane's
+ * origin so the x and y offsets can actually fit the registers. As a
+ * consequence, the fence doesn't really start exactly at the display plane
+ * address we program because it starts at the real start of the buffer, so we
+ * have to take this into consideration here.
+ */
+static unsigned int get_crtc_fence_y_offset(struct intel_crtc *crtc)
+{
+	return crtc->base.y - crtc->adjusted_y;
+}
+
 static void i8xx_fbc_disable(struct drm_i915_private *dev_priv)
 {
 	u32 fbc_ctl;
@@ -97,7 +110,7 @@ static void i8xx_fbc_enable(struct intel_crtc *crtc)
 		fbc_ctl2 = FBC_CTL_FENCE_DBL | FBC_CTL_IDLE_IMM | FBC_CTL_CPU_FENCE;
 		fbc_ctl2 |= FBC_CTL_PLANE(crtc->plane);
 		I915_WRITE(FBC_CONTROL2, fbc_ctl2);
-		I915_WRITE(FBC_FENCE_OFF, crtc->base.y);
+		I915_WRITE(FBC_FENCE_OFF, get_crtc_fence_y_offset(crtc));
 	}
 
 	/* enable it... */
@@ -135,7 +148,7 @@ static void g4x_fbc_enable(struct intel_crtc *crtc)
 		dpfc_ctl |= DPFC_CTL_LIMIT_1X;
 	dpfc_ctl |= DPFC_CTL_FENCE_EN | obj->fence_reg;
 
-	I915_WRITE(DPFC_FENCE_YOFF, crtc->base.y);
+	I915_WRITE(DPFC_FENCE_YOFF, get_crtc_fence_y_offset(crtc));
 
 	/* enable it... */
 	I915_WRITE(DPFC_CONTROL, dpfc_ctl | DPFC_CTL_EN);
@@ -177,6 +190,7 @@ static void ilk_fbc_enable(struct intel_crtc *crtc)
 	struct drm_i915_gem_object *obj = intel_fb_obj(fb);
 	u32 dpfc_ctl;
 	int threshold = dev_priv->fbc.threshold;
+	unsigned int y_offset;
 
 	dev_priv->fbc.enabled = true;
 
@@ -200,7 +214,8 @@ static void ilk_fbc_enable(struct intel_crtc *crtc)
 	if (IS_GEN5(dev_priv))
 		dpfc_ctl |= obj->fence_reg;
 
-	I915_WRITE(ILK_DPFC_FENCE_YOFF, crtc->base.y);
+	y_offset = get_crtc_fence_y_offset(crtc);
+	I915_WRITE(ILK_DPFC_FENCE_YOFF, y_offset);
 	I915_WRITE(ILK_FBC_RT_BASE, i915_gem_obj_ggtt_offset(obj) | ILK_FBC_RT_VALID);
 	/* enable it... */
 	I915_WRITE(ILK_DPFC_CONTROL, dpfc_ctl | DPFC_CTL_EN);
@@ -208,7 +223,7 @@ static void ilk_fbc_enable(struct intel_crtc *crtc)
 	if (IS_GEN6(dev_priv)) {
 		I915_WRITE(SNB_DPFC_CTL_SA,
 			   SNB_CPU_FENCE_ENABLE | obj->fence_reg);
-		I915_WRITE(DPFC_CPU_FENCE_OFFSET, crtc->base.y);
+		I915_WRITE(DPFC_CPU_FENCE_OFFSET, y_offset);
 	}
 
 	intel_fbc_nuke(dev_priv);
@@ -288,7 +303,7 @@ static void gen7_fbc_enable(struct intel_crtc *crtc)
 
 	I915_WRITE(SNB_DPFC_CTL_SA,
 		   SNB_CPU_FENCE_ENABLE | obj->fence_reg);
-	I915_WRITE(DPFC_CPU_FENCE_OFFSET, crtc->base.y);
+	I915_WRITE(DPFC_CPU_FENCE_OFFSET, get_crtc_fence_y_offset(crtc));
 
 	intel_fbc_nuke(dev_priv);
 
-- 
cgit v0.10.2


From 7d316aecf883a19c9883e4dcbc058806fd25b152 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 16 Sep 2015 21:28:50 +0300
Subject: drm/i915: Implement stolen reserved detection for ctg/elk
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Finally managed to dig up enough hints as to where the stolen
reserved stuff lives on ctg/elk. So add the code to decode it.
This was a combination of old chipset specs, diggin up an old
elk grits release with an ctg/elk AubLoad etc.

This was only tested on an elk as I don't have a ctg here
unfortunately.

This leaves ilk as the only platform that doesn't have a way
to detect this stuff. Looking at the register contents on my
ilk, it might be that the elk way works there too, but I
can't be sure since I can't affect the amount of reserved
memory on that machine, and if I am to trust the register
contents, by default it would reserve 0 bytes.

v2: s/WARN_ON_ONCE/WARN_ON/ since it's in one time init code
    anyway (Paulo)

Cc: Paulo Zanoni <paulo.r.zanoni@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Acked-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_stolen.c b/drivers/gpu/drm/i915/i915_gem_stolen.c
index 081ef6d..55df6ce 100644
--- a/drivers/gpu/drm/i915/i915_gem_stolen.c
+++ b/drivers/gpu/drm/i915/i915_gem_stolen.c
@@ -196,6 +196,29 @@ void i915_gem_cleanup_stolen(struct drm_device *dev)
 	drm_mm_takedown(&dev_priv->mm.stolen);
 }
 
+static void g4x_get_stolen_reserved(struct drm_i915_private *dev_priv,
+				    unsigned long *base, unsigned long *size)
+{
+	uint32_t reg_val = I915_READ(IS_GM45(dev_priv) ?
+				     CTG_STOLEN_RESERVED :
+				     ELK_STOLEN_RESERVED);
+	unsigned long stolen_top = dev_priv->mm.stolen_base +
+		dev_priv->gtt.stolen_size;
+
+	*base = (reg_val & G4X_STOLEN_RESERVED_ADDR2_MASK) << 16;
+
+	WARN_ON((reg_val & G4X_STOLEN_RESERVED_ADDR1_MASK) < *base);
+
+	/* On these platforms, the register doesn't have a size field, so the
+	 * size is the distance between the base and the top of the stolen
+	 * memory. We also have the genuine case where base is zero and there's
+	 * nothing reserved. */
+	if (*base == 0)
+		*size = 0;
+	else
+		*size = stolen_top - *base;
+}
+
 static void gen6_get_stolen_reserved(struct drm_i915_private *dev_priv,
 				     unsigned long *base, unsigned long *size)
 {
@@ -315,10 +338,12 @@ int i915_gem_init_stolen(struct drm_device *dev)
 	switch (INTEL_INFO(dev_priv)->gen) {
 	case 2:
 	case 3:
+		break;
 	case 4:
-		if (!IS_G4X(dev))
-			break;
-		/* fall through */
+		if (IS_G4X(dev))
+			g4x_get_stolen_reserved(dev_priv, &reserved_base,
+						&reserved_size);
+		break;
 	case 5:
 		/* Assume the gen6 maximum for the older platforms. */
 		reserved_size = 1024 * 1024;
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index c5287a1..881a47a 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -2494,6 +2494,11 @@ enum skl_disp_power_wells {
 
 #define MCHBAR_MIRROR_BASE_SNB	0x140000
 
+#define CTG_STOLEN_RESERVED		(MCHBAR_MIRROR_BASE + 0x34)
+#define ELK_STOLEN_RESERVED		(MCHBAR_MIRROR_BASE + 0x48)
+#define G4X_STOLEN_RESERVED_ADDR1_MASK	(0xFFFF << 16)
+#define G4X_STOLEN_RESERVED_ADDR2_MASK	(0xFFF << 4)
+
 /* Memory controller frequency in MCHBAR for Haswell (possible SNB+) */
 #define DCLK (MCHBAR_MIRROR_BASE_SNB + 0x5e04)
 
-- 
cgit v0.10.2


From a2cad9dff4dd44d0244b966d980de9d602d87593 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Micha=C5=82=20Winiarski?= <michal.winiarski@intel.com>
Date: Wed, 16 Sep 2015 11:49:00 +0200
Subject: drm/i915/gtt: Do not initialize drm_mm twice.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

It would be initialized just moments later by i915_init_vm.

Rearrange the code such that i915_init_vm() is next to its callers
inside i915_gem_gtt (and so we can make it static). After removing the
dance around the files, it is clear that we are repeating some work
inside the initializers (such as calling drm_mm_init() multiple times),
so take advantage of the refactor to also remove some redundant code and
clean up the interface.

v2: Commit msg update,
    s/i915_init_vm/i915_address_space_init, move to i915_gem_gtt.c,
    init address_space during i915_gem_setup_global_gtt for ggtt.
v3: Do not init global_link - we are adding it to vm_list moments later,
    make i915_address_space_init static, use OOP style parameter order.

Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Michel Thierry <michel.thierry@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
Reviewed-by Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 3044cbe..4e8178d 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -2809,8 +2809,6 @@ struct drm_i915_gem_object *i915_gem_alloc_object(struct drm_device *dev,
 						  size_t size);
 struct drm_i915_gem_object *i915_gem_object_create_from_data(
 		struct drm_device *dev, const void *data, size_t size);
-void i915_init_vm(struct drm_i915_private *dev_priv,
-		  struct i915_address_space *vm);
 void i915_gem_free_object(struct drm_gem_object *obj);
 void i915_gem_vma_destroy(struct i915_vma *vma);
 
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index e1b6457..58c7ad0 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -4844,18 +4844,6 @@ init_ring_lists(struct intel_engine_cs *ring)
 	INIT_LIST_HEAD(&ring->request_list);
 }
 
-void i915_init_vm(struct drm_i915_private *dev_priv,
-		  struct i915_address_space *vm)
-{
-	if (!i915_is_ggtt(vm))
-		drm_mm_init(&vm->mm, vm->start, vm->total);
-	vm->dev = dev_priv->dev;
-	INIT_LIST_HEAD(&vm->active_list);
-	INIT_LIST_HEAD(&vm->inactive_list);
-	INIT_LIST_HEAD(&vm->global_link);
-	list_add_tail(&vm->global_link, &dev_priv->vm_list);
-}
-
 void
 i915_gem_load(struct drm_device *dev)
 {
@@ -4879,8 +4867,6 @@ i915_gem_load(struct drm_device *dev)
 				  NULL);
 
 	INIT_LIST_HEAD(&dev_priv->vm_list);
-	i915_init_vm(dev_priv, &dev_priv->gtt.base);
-
 	INIT_LIST_HEAD(&dev_priv->context_list);
 	INIT_LIST_HEAD(&dev_priv->mm.unbound_list);
 	INIT_LIST_HEAD(&dev_priv->mm.bound_list);
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 8786281..01f3521 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -2121,6 +2121,16 @@ static int __hw_ppgtt_init(struct drm_device *dev, struct i915_hw_ppgtt *ppgtt)
 		return gen8_ppgtt_init(ppgtt);
 }
 
+static void i915_address_space_init(struct i915_address_space *vm,
+				    struct drm_i915_private *dev_priv)
+{
+	drm_mm_init(&vm->mm, vm->start, vm->total);
+	vm->dev = dev_priv->dev;
+	INIT_LIST_HEAD(&vm->active_list);
+	INIT_LIST_HEAD(&vm->inactive_list);
+	list_add_tail(&vm->global_link, &dev_priv->vm_list);
+}
+
 int i915_ppgtt_init(struct drm_device *dev, struct i915_hw_ppgtt *ppgtt)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -2129,9 +2139,7 @@ int i915_ppgtt_init(struct drm_device *dev, struct i915_hw_ppgtt *ppgtt)
 	ret = __hw_ppgtt_init(dev, ppgtt);
 	if (ret == 0) {
 		kref_init(&ppgtt->ref);
-		drm_mm_init(&ppgtt->base.mm, ppgtt->base.start,
-			    ppgtt->base.total);
-		i915_init_vm(dev_priv, &ppgtt->base);
+		i915_address_space_init(&ppgtt->base, dev_priv);
 	}
 
 	return ret;
@@ -2618,11 +2626,13 @@ static int i915_gem_setup_global_gtt(struct drm_device *dev,
 
 	BUG_ON(mappable_end > end);
 
-	/* Subtract the guard page ... */
-	drm_mm_init(&ggtt_vm->mm, start, end - start - PAGE_SIZE);
+	ggtt_vm->start = start;
 
-	dev_priv->gtt.base.start = start;
-	dev_priv->gtt.base.total = end - start;
+	/* Subtract the guard page before address space initialization to
+	 * shrink the range used by drm_mm */
+	ggtt_vm->total = end - start - PAGE_SIZE;
+	i915_address_space_init(ggtt_vm, dev_priv);
+	ggtt_vm->total += PAGE_SIZE;
 
 	if (intel_vgpu_active(dev)) {
 		ret = intel_vgt_balloon(dev);
@@ -2631,7 +2641,7 @@ static int i915_gem_setup_global_gtt(struct drm_device *dev,
 	}
 
 	if (!HAS_LLC(dev))
-		dev_priv->gtt.base.mm.color_adjust = i915_gtt_color_adjust;
+		ggtt_vm->mm.color_adjust = i915_gtt_color_adjust;
 
 	/* Mark any preallocated objects as occupied */
 	list_for_each_entry(obj, &dev_priv->mm.bound_list, global_list) {
-- 
cgit v0.10.2


From 8204502a6aa2eab8cd5779020c2881fbc6361a96 Mon Sep 17 00:00:00 2001
From: Dongwon Kim <dongwon.kim@intel.com>
Date: Thu, 17 Sep 2015 11:26:35 -0700
Subject: drm/i915: Do not hardcode s_max, ss_max and eu_mask for BXT

We can calculate BXT values correctly from GFX fuse values without
hardcoding special limits.

Cc: Imre Deak <imre.deak@intel.com>
Cc: Matthew D Roper <matthew.d.roper@intel.com>
Signed-off-by: Dongwon Kim <dongwon.kim@intel.com>
Reviewed-by: Arun Siluvery <arun.siluvery@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index c069550..817b05c 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -631,17 +631,6 @@ static void gen9_sseu_info_init(struct drm_device *dev)
 	u32 fuse2, s_enable, ss_disable, eu_disable;
 	u8 eu_mask = 0xff;
 
-	/*
-	 * BXT has a single slice. BXT also has at most 6 EU per subslice,
-	 * and therefore only the lowest 6 bits of the 8-bit EU disable
-	 * fields are valid.
-	*/
-	if (IS_BROXTON(dev)) {
-		s_max = 1;
-		eu_max = 6;
-		eu_mask = 0x3f;
-	}
-
 	info = (struct intel_device_info *)&dev_priv->info;
 	fuse2 = I915_READ(GEN8_FUSE2);
 	s_enable = (fuse2 & GEN8_F2_S_ENA_MASK) >>
-- 
cgit v0.10.2


From 0706f17c307b056ff6f1848320ba82d76945a6ff Mon Sep 17 00:00:00 2001
From: Egbert Eich <eich@suse.de>
Date: Wed, 23 Sep 2015 16:15:27 +0200
Subject: drm/i915: Avoid race of intel_crt_detect_hotplug() with HPD
 interrupt, v2

An HPD interrupt may fire while we are in a function that changes
the PORT_HOTPLUG_EN register - especially when an HPD interrupt
storm occurs.
Since the interrupt handler changes the enabled HPD lines when it
detects such a storm the read-modify-write cycles may interfere.
To avoid this, shiled the rmw cycles with IRQ save spinlocks.

Changes since v1:
- Implement a function which takes care of accessing PORT_HOTPLUG_EN.

Signed-off-by: Egbert Eich <eich@suse.de>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 4e8178d..ba4596e 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -2742,6 +2742,9 @@ i915_disable_pipestat(struct drm_i915_private *dev_priv, enum pipe pipe,
 
 void valleyview_enable_display_irqs(struct drm_i915_private *dev_priv);
 void valleyview_disable_display_irqs(struct drm_i915_private *dev_priv);
+void i915_hotplug_interrupt_update(struct drm_i915_private *dev_priv,
+				   uint32_t mask,
+				   uint32_t bits);
 void
 ironlake_enable_display_irq(struct drm_i915_private *dev_priv, u32 mask);
 void
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index ca6d806..6ca6c16 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -167,6 +167,44 @@ static const u32 hpd_bxt[HPD_NUM_PINS] = {
 
 static void gen6_rps_irq_handler(struct drm_i915_private *dev_priv, u32 pm_iir);
 
+/* For display hotplug interrupt */
+static inline void
+i915_hotplug_interrupt_update_locked(struct drm_i915_private *dev_priv,
+				     uint32_t mask,
+				     uint32_t bits)
+{
+	uint32_t val;
+
+	assert_spin_locked(&dev_priv->irq_lock);
+	WARN_ON(bits & ~mask);
+
+	val = I915_READ(PORT_HOTPLUG_EN);
+	val &= ~mask;
+	val |= bits;
+	I915_WRITE(PORT_HOTPLUG_EN, val);
+}
+
+/**
+ * i915_hotplug_interrupt_update - update hotplug interrupt enable
+ * @dev_priv: driver private
+ * @mask: bits to update
+ * @bits: bits to enable
+ * NOTE: the HPD enable bits are modified both inside and outside
+ * of an interrupt context. To avoid that read-modify-write cycles
+ * interfer, these bits are protected by a spinlock. Since this
+ * function is usually not called from a context where the lock is
+ * held already, this function acquires the lock itself. A non-locking
+ * version is also available.
+ */
+void i915_hotplug_interrupt_update(struct drm_i915_private *dev_priv,
+				   uint32_t mask,
+				   uint32_t bits)
+{
+	spin_lock_irq(&dev_priv->irq_lock);
+	i915_hotplug_interrupt_update_locked(dev_priv, mask, bits);
+	spin_unlock_irq(&dev_priv->irq_lock);
+}
+
 /**
  * ilk_update_display_irq - update DEIMR
  * @dev_priv: driver private
@@ -3050,7 +3088,7 @@ static void vlv_display_irq_reset(struct drm_i915_private *dev_priv)
 {
 	enum pipe pipe;
 
-	I915_WRITE(PORT_HOTPLUG_EN, 0);
+	i915_hotplug_interrupt_update(dev_priv, 0xFFFFFFFF, 0);
 	I915_WRITE(PORT_HOTPLUG_STAT, I915_READ(PORT_HOTPLUG_STAT));
 
 	for_each_pipe(dev_priv, pipe)
@@ -3466,7 +3504,7 @@ static void vlv_display_irq_postinstall(struct drm_i915_private *dev_priv)
 {
 	dev_priv->irq_mask = ~0;
 
-	I915_WRITE(PORT_HOTPLUG_EN, 0);
+	i915_hotplug_interrupt_update(dev_priv, 0xffffffff, 0);
 	POSTING_READ(PORT_HOTPLUG_EN);
 
 	I915_WRITE(VLV_IIR, 0xffffffff);
@@ -3840,7 +3878,7 @@ static void i915_irq_preinstall(struct drm_device * dev)
 	int pipe;
 
 	if (I915_HAS_HOTPLUG(dev)) {
-		I915_WRITE(PORT_HOTPLUG_EN, 0);
+		i915_hotplug_interrupt_update(dev_priv, 0xffffffff, 0);
 		I915_WRITE(PORT_HOTPLUG_STAT, I915_READ(PORT_HOTPLUG_STAT));
 	}
 
@@ -3874,7 +3912,7 @@ static int i915_irq_postinstall(struct drm_device *dev)
 		I915_USER_INTERRUPT;
 
 	if (I915_HAS_HOTPLUG(dev)) {
-		I915_WRITE(PORT_HOTPLUG_EN, 0);
+		i915_hotplug_interrupt_update(dev_priv, 0xffffffff, 0);
 		POSTING_READ(PORT_HOTPLUG_EN);
 
 		/* Enable in IER... */
@@ -4036,7 +4074,7 @@ static void i915_irq_uninstall(struct drm_device * dev)
 	int pipe;
 
 	if (I915_HAS_HOTPLUG(dev)) {
-		I915_WRITE(PORT_HOTPLUG_EN, 0);
+		i915_hotplug_interrupt_update(dev_priv, 0xffffffff, 0);
 		I915_WRITE(PORT_HOTPLUG_STAT, I915_READ(PORT_HOTPLUG_STAT));
 	}
 
@@ -4057,7 +4095,7 @@ static void i965_irq_preinstall(struct drm_device * dev)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int pipe;
 
-	I915_WRITE(PORT_HOTPLUG_EN, 0);
+	i915_hotplug_interrupt_update(dev_priv, 0xffffffff, 0);
 	I915_WRITE(PORT_HOTPLUG_STAT, I915_READ(PORT_HOTPLUG_STAT));
 
 	I915_WRITE(HWSTAM, 0xeffe);
@@ -4118,7 +4156,7 @@ static int i965_irq_postinstall(struct drm_device *dev)
 	I915_WRITE(IER, enable_mask);
 	POSTING_READ(IER);
 
-	I915_WRITE(PORT_HOTPLUG_EN, 0);
+	i915_hotplug_interrupt_update(dev_priv, 0xffffffff, 0);
 	POSTING_READ(PORT_HOTPLUG_EN);
 
 	i915_enable_asle_pipestat(dev);
@@ -4133,22 +4171,22 @@ static void i915_hpd_irq_setup(struct drm_device *dev)
 
 	assert_spin_locked(&dev_priv->irq_lock);
 
-	hotplug_en = I915_READ(PORT_HOTPLUG_EN);
-	hotplug_en &= ~HOTPLUG_INT_EN_MASK;
 	/* Note HDMI and DP share hotplug bits */
 	/* enable bits are the same for all generations */
-	hotplug_en |= intel_hpd_enabled_irqs(dev, hpd_mask_i915);
+	hotplug_en = intel_hpd_enabled_irqs(dev, hpd_mask_i915);
 	/* Programming the CRT detection parameters tends
 	   to generate a spurious hotplug event about three
 	   seconds later.  So just do it once.
 	*/
 	if (IS_G4X(dev))
 		hotplug_en |= CRT_HOTPLUG_ACTIVATION_PERIOD_64;
-	hotplug_en &= ~CRT_HOTPLUG_VOLTAGE_COMPARE_MASK;
 	hotplug_en |= CRT_HOTPLUG_VOLTAGE_COMPARE_50;
 
 	/* Ignore TV since it's buggy */
-	I915_WRITE(PORT_HOTPLUG_EN, hotplug_en);
+	i915_hotplug_interrupt_update_locked(dev_priv,
+				      (HOTPLUG_INT_EN_MASK
+				       | CRT_HOTPLUG_VOLTAGE_COMPARE_MASK),
+				      hotplug_en);
 }
 
 static irqreturn_t i965_irq_handler(int irq, void *arg)
@@ -4261,7 +4299,7 @@ static void i965_irq_uninstall(struct drm_device * dev)
 	if (!dev_priv)
 		return;
 
-	I915_WRITE(PORT_HOTPLUG_EN, 0);
+	i915_hotplug_interrupt_update(dev_priv, 0xffffffff, 0);
 	I915_WRITE(PORT_HOTPLUG_STAT, I915_READ(PORT_HOTPLUG_STAT));
 
 	I915_WRITE(HWSTAM, 0xffffffff);
diff --git a/drivers/gpu/drm/i915/intel_crt.c b/drivers/gpu/drm/i915/intel_crt.c
index af5e43b..6ce38e3 100644
--- a/drivers/gpu/drm/i915/intel_crt.c
+++ b/drivers/gpu/drm/i915/intel_crt.c
@@ -376,7 +376,7 @@ static bool intel_crt_detect_hotplug(struct drm_connector *connector)
 {
 	struct drm_device *dev = connector->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 hotplug_en, orig, stat;
+	u32 stat;
 	bool ret = false;
 	int i, tries = 0;
 
@@ -395,12 +395,12 @@ static bool intel_crt_detect_hotplug(struct drm_connector *connector)
 		tries = 2;
 	else
 		tries = 1;
-	hotplug_en = orig = I915_READ(PORT_HOTPLUG_EN);
-	hotplug_en |= CRT_HOTPLUG_FORCE_DETECT;
 
 	for (i = 0; i < tries ; i++) {
 		/* turn on the FORCE_DETECT */
-		I915_WRITE(PORT_HOTPLUG_EN, hotplug_en);
+		i915_hotplug_interrupt_update(dev_priv,
+					      CRT_HOTPLUG_FORCE_DETECT,
+					      CRT_HOTPLUG_FORCE_DETECT);
 		/* wait for FORCE_DETECT to go off */
 		if (wait_for((I915_READ(PORT_HOTPLUG_EN) &
 			      CRT_HOTPLUG_FORCE_DETECT) == 0,
@@ -415,8 +415,7 @@ static bool intel_crt_detect_hotplug(struct drm_connector *connector)
 	/* clear the interrupt we just generated, if any */
 	I915_WRITE(PORT_HOTPLUG_STAT, CRT_HOTPLUG_INT_STATUS);
 
-	/* and put the bits back */
-	I915_WRITE(PORT_HOTPLUG_EN, orig);
+	i915_hotplug_interrupt_update(dev_priv, CRT_HOTPLUG_FORCE_DETECT, 0);
 
 	return ret;
 }
-- 
cgit v0.10.2


From 53ce81a761fb130c5d7ad653239fc9f9adc32237 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 11 Sep 2015 21:04:38 +0300
Subject: drm/i915: Make sure we don't detect eDP on g4x
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We don't support eDP on g4x, so let's not even look at the VBT
to determine the port type, just in case the VBT is bonkers
on some g4x machines and indicates the precense of eDP.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 02401a3..5fe1648 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -5255,6 +5255,13 @@ bool intel_dp_is_edp(struct drm_device *dev, enum port port)
 		[PORT_E] = DVO_PORT_DPE,
 	};
 
+	/*
+	 * eDP not supported on g4x. so bail out early just
+	 * for a bit extra safety in case the VBT is bonkers.
+	 */
+	if (INTEL_INFO(dev)->gen < 5)
+		return false;
+
 	if (port == PORT_A)
 		return true;
 
-- 
cgit v0.10.2


From 972e7d71c82ea70100b808695d5cf735c1df5ef8 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 11 Sep 2015 21:04:39 +0300
Subject: drm/i915: Ignore "digital output" and "not HDMI output" bits for eDP
 detection
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Ignore DEVICE_TYPE_NOT_HDMI_OUTPUT and DEVICE_TYPE_DIGITAL_OUTPUT when
trying to determine the presence of eDP based on the VBT child device
type. Apparently a significant portion of VLV systems have these bits
set incorrectly, and so we currently fail to detect eDP on said systems.

This is based on an earlier patch [1] from Andreas Lampersperger.
Instead of ignoring the bits just on VLV as was done in the orignal
patch, we now ignore them for all platforms. We should still have
enough bits in there to avoid false positives (unless the VBT is totally
bonkers).

Quoting the orignal patch:
> When the i915.ko identify an eDP output on a valleyview
> board, it should be more slackly. The reason for that is,
> that BIOS DATA TABLES generated with intel BMP (Binary
> Modification Program) do not set bits for NOT_HDMI or
> DIGITAL_OUTPUT on the device type. Due to Adolfo
> Sanchez from Intel EMGD, this is not possible.
> To solve this problem and enable i915.ko on embedded
> vlv boards with eDP, we ignore this two bits.

[1] http://lists.freedesktop.org/archives/intel-gfx/2015-June/069416.html

Cc: Andreas Lampersperger <lampersperger.andreas@heidenhain.de>
Cc: "Sanchez, AdolfoX" <adolfox.sanchez@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Acked-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_bios.h b/drivers/gpu/drm/i915/intel_bios.h
index 1b7417e..7ec8c9a 100644
--- a/drivers/gpu/drm/i915/intel_bios.h
+++ b/drivers/gpu/drm/i915/intel_bios.h
@@ -741,7 +741,6 @@ int intel_parse_bios(struct drm_device *dev);
  */
 #define DEVICE_TYPE_eDP_BITS \
 	(DEVICE_TYPE_INTERNAL_CONNECTOR | \
-	 DEVICE_TYPE_NOT_HDMI_OUTPUT | \
 	 DEVICE_TYPE_MIPI_OUTPUT | \
 	 DEVICE_TYPE_COMPOSITE_OUTPUT | \
 	 DEVICE_TYPE_DUAL_CHANNEL | \
@@ -749,7 +748,6 @@ int intel_parse_bios(struct drm_device *dev);
 	 DEVICE_TYPE_TMDS_DVI_SIGNALING | \
 	 DEVICE_TYPE_VIDEO_SIGNALING | \
 	 DEVICE_TYPE_DISPLAYPORT_OUTPUT | \
-	 DEVICE_TYPE_DIGITAL_OUTPUT | \
 	 DEVICE_TYPE_ANALOG_OUTPUT)
 
 /* define the DVO port for HDMI output type */
-- 
cgit v0.10.2


From 8bd864b81a7472a19eb6ed536e028a10e7d71239 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 18 Sep 2015 20:03:14 +0300
Subject: drm/i915: Don't pass sdvo_reg to intel_sdvo_select_{ddc, i2c}_bus()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

intel_sdvo_select_ddc_bus() and intel_sdvo_select_i2c_bus() have no used
for the passed in 'reg', so just drop it.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c
index ca3dd7c..05521b5 100644
--- a/drivers/gpu/drm/i915/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/intel_sdvo.c
@@ -2222,7 +2222,7 @@ intel_sdvo_guess_ddc_bus(struct intel_sdvo *sdvo)
  */
 static void
 intel_sdvo_select_ddc_bus(struct drm_i915_private *dev_priv,
-			  struct intel_sdvo *sdvo, u32 reg)
+			  struct intel_sdvo *sdvo)
 {
 	struct sdvo_device_mapping *mapping;
 
@@ -2239,7 +2239,7 @@ intel_sdvo_select_ddc_bus(struct drm_i915_private *dev_priv,
 
 static void
 intel_sdvo_select_i2c_bus(struct drm_i915_private *dev_priv,
-			  struct intel_sdvo *sdvo, u32 reg)
+			  struct intel_sdvo *sdvo)
 {
 	struct sdvo_device_mapping *mapping;
 	u8 pin;
@@ -2925,7 +2925,7 @@ bool intel_sdvo_init(struct drm_device *dev, uint32_t sdvo_reg, bool is_sdvob)
 	intel_sdvo->sdvo_reg = sdvo_reg;
 	intel_sdvo->is_sdvob = is_sdvob;
 	intel_sdvo->slave_addr = intel_sdvo_get_slave_addr(dev, intel_sdvo) >> 1;
-	intel_sdvo_select_i2c_bus(dev_priv, intel_sdvo, sdvo_reg);
+	intel_sdvo_select_i2c_bus(dev_priv, intel_sdvo);
 	if (!intel_sdvo_init_ddc_proxy(intel_sdvo, dev))
 		goto err_i2c_bus;
 
@@ -2987,7 +2987,7 @@ bool intel_sdvo_init(struct drm_device *dev, uint32_t sdvo_reg, bool is_sdvob)
 	 */
 	intel_sdvo->base.cloneable = 0;
 
-	intel_sdvo_select_ddc_bus(dev_priv, intel_sdvo, sdvo_reg);
+	intel_sdvo_select_ddc_bus(dev_priv, intel_sdvo);
 
 	/* Set the input timing to the screen. Assume always input 0. */
 	if (!intel_sdvo_set_target_input(intel_sdvo))
-- 
cgit v0.10.2


From 83843d84fcd38a7e9b30ff0b9af04669c77ecf8c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 18 Sep 2015 20:03:15 +0300
Subject: drm/i915: Parametrize LRC registers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index 72ae347..5615d3d 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -2069,8 +2069,8 @@ static int i915_execlists(struct seq_file *m, void *data)
 
 		seq_printf(m, "%s\n", ring->name);
 
-		status = I915_READ(RING_EXECLIST_STATUS(ring));
-		ctx_id = I915_READ(RING_EXECLIST_STATUS(ring) + 4);
+		status = I915_READ(RING_EXECLIST_STATUS_LO(ring));
+		ctx_id = I915_READ(RING_EXECLIST_STATUS_HI(ring));
 		seq_printf(m, "\tExeclist status: 0x%08X, context: %u\n",
 			   status, ctx_id);
 
@@ -2085,8 +2085,8 @@ static int i915_execlists(struct seq_file *m, void *data)
 			   read_pointer, write_pointer);
 
 		for (i = 0; i < 6; i++) {
-			status = I915_READ(RING_CONTEXT_STATUS_BUF(ring) + 8*i);
-			ctx_id = I915_READ(RING_CONTEXT_STATUS_BUF(ring) + 8*i + 4);
+			status = I915_READ(RING_CONTEXT_STATUS_BUF_LO(ring, i));
+			ctx_id = I915_READ(RING_CONTEXT_STATUS_BUF_HI(ring, i));
 
 			seq_printf(m, "\tStatus buffer %d: 0x%08X, context: %u\n",
 				   i, status, ctx_id);
diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
index ff9a481..7417650 100644
--- a/drivers/gpu/drm/i915/intel_lrc.c
+++ b/drivers/gpu/drm/i915/intel_lrc.c
@@ -349,7 +349,7 @@ static void execlists_elsp_write(struct drm_i915_gem_request *rq0,
 	I915_WRITE_FW(RING_ELSP(ring), lower_32_bits(desc[0]));
 
 	/* ELSP is a wo register, use another nearby reg for posting */
-	POSTING_READ_FW(RING_EXECLIST_STATUS(ring));
+	POSTING_READ_FW(RING_EXECLIST_STATUS_LO(ring));
 	intel_uncore_forcewake_put__locked(dev_priv, FORCEWAKE_ALL);
 	spin_unlock(&dev_priv->uncore.lock);
 }
@@ -519,10 +519,8 @@ void intel_lrc_irq_handler(struct intel_engine_cs *ring)
 
 	while (read_pointer < write_pointer) {
 		read_pointer++;
-		status = I915_READ(RING_CONTEXT_STATUS_BUF(ring) +
-				(read_pointer % 6) * 8);
-		status_id = I915_READ(RING_CONTEXT_STATUS_BUF(ring) +
-				(read_pointer % 6) * 8 + 4);
+		status = I915_READ(RING_CONTEXT_STATUS_BUF_LO(ring, read_pointer % 6));
+		status_id = I915_READ(RING_CONTEXT_STATUS_BUF_HI(ring, read_pointer % 6));
 
 		if (status & GEN8_CTX_STATUS_IDLE_ACTIVE)
 			continue;
diff --git a/drivers/gpu/drm/i915/intel_lrc.h b/drivers/gpu/drm/i915/intel_lrc.h
index 69d99f0..8a08a27 100644
--- a/drivers/gpu/drm/i915/intel_lrc.h
+++ b/drivers/gpu/drm/i915/intel_lrc.h
@@ -28,12 +28,14 @@
 
 /* Execlists regs */
 #define RING_ELSP(ring)			((ring)->mmio_base+0x230)
-#define RING_EXECLIST_STATUS(ring)	((ring)->mmio_base+0x234)
+#define RING_EXECLIST_STATUS_LO(ring)	((ring)->mmio_base+0x234)
+#define RING_EXECLIST_STATUS_HI(ring)	((ring)->mmio_base+0x234 + 4)
 #define RING_CONTEXT_CONTROL(ring)	((ring)->mmio_base+0x244)
 #define	  CTX_CTRL_INHIBIT_SYN_CTX_SWITCH	(1 << 3)
 #define	  CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT	(1 << 0)
 #define   CTX_CTRL_RS_CTX_ENABLE                (1 << 1)
-#define RING_CONTEXT_STATUS_BUF(ring)	((ring)->mmio_base+0x370)
+#define RING_CONTEXT_STATUS_BUF_LO(ring, i)	((ring)->mmio_base+0x370 + (i) * 8)
+#define RING_CONTEXT_STATUS_BUF_HI(ring, i)	((ring)->mmio_base+0x370 + (i) * 8 + 4)
 #define RING_CONTEXT_STATUS_PTR(ring)	((ring)->mmio_base+0x3a0)
 
 /* Logical Rings */
-- 
cgit v0.10.2


From 22dfe79fd9c3951017bd68cad1d3257631307dde Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 18 Sep 2015 20:03:16 +0300
Subject: drm/i915: Parametrize GEN7_GT_SCRATCH and GEN7_LRA_LIMITS
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index bdec64c..947bd43 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -1120,7 +1120,7 @@ static void vlv_save_gunit_s0ix_state(struct drm_i915_private *dev_priv)
 	s->gfx_pend_tlb1	= I915_READ(GEN7_GFX_PEND_TLB1);
 
 	for (i = 0; i < ARRAY_SIZE(s->lra_limits); i++)
-		s->lra_limits[i] = I915_READ(GEN7_LRA_LIMITS_BASE + i * 4);
+		s->lra_limits[i] = I915_READ(GEN7_LRA_LIMITS(i));
 
 	s->media_max_req_count	= I915_READ(GEN7_MEDIA_MAX_REQ_COUNT);
 	s->gfx_max_req_count	= I915_READ(GEN7_GFX_MAX_REQ_COUNT);
@@ -1164,7 +1164,7 @@ static void vlv_save_gunit_s0ix_state(struct drm_i915_private *dev_priv)
 	s->pm_ier		= I915_READ(GEN6_PMIER);
 
 	for (i = 0; i < ARRAY_SIZE(s->gt_scratch); i++)
-		s->gt_scratch[i] = I915_READ(GEN7_GT_SCRATCH_BASE + i * 4);
+		s->gt_scratch[i] = I915_READ(GEN7_GT_SCRATCH(i));
 
 	/* GT SA CZ domain, 0x100000-0x138124 */
 	s->tilectl		= I915_READ(TILECTL);
@@ -1202,7 +1202,7 @@ static void vlv_restore_gunit_s0ix_state(struct drm_i915_private *dev_priv)
 	I915_WRITE(GEN7_GFX_PEND_TLB1,	s->gfx_pend_tlb1);
 
 	for (i = 0; i < ARRAY_SIZE(s->lra_limits); i++)
-		I915_WRITE(GEN7_LRA_LIMITS_BASE + i * 4, s->lra_limits[i]);
+		I915_WRITE(GEN7_LRA_LIMITS(i), s->lra_limits[i]);
 
 	I915_WRITE(GEN7_MEDIA_MAX_REQ_COUNT, s->media_max_req_count);
 	I915_WRITE(GEN7_GFX_MAX_REQ_COUNT, s->gfx_max_req_count);
@@ -1246,7 +1246,7 @@ static void vlv_restore_gunit_s0ix_state(struct drm_i915_private *dev_priv)
 	I915_WRITE(GEN6_PMIER,		s->pm_ier);
 
 	for (i = 0; i < ARRAY_SIZE(s->gt_scratch); i++)
-		I915_WRITE(GEN7_GT_SCRATCH_BASE + i * 4, s->gt_scratch[i]);
+		I915_WRITE(GEN7_GT_SCRATCH(i), s->gt_scratch[i]);
 
 	/* GT SA CZ domain, 0x100000-0x138124 */
 	I915_WRITE(TILECTL,			s->tilectl);
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 881a47a..90ce688 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -1527,7 +1527,7 @@ enum skl_disp_power_wells {
 #define GEN7_GFX_PEND_TLB0	0x4034
 #define GEN7_GFX_PEND_TLB1	0x4038
 /* L3, CVS, ZTLB, RCC, CASC LRA min, max values */
-#define GEN7_LRA_LIMITS_BASE	0x403C
+#define GEN7_LRA_LIMITS(i)	(0x403C + (i) * 4)
 #define GEN7_LRA_LIMITS_REG_NUM	13
 #define GEN7_MEDIA_MAX_REQ_COUNT	0x4070
 #define GEN7_GFX_MAX_REQ_COUNT		0x4074
@@ -6813,7 +6813,7 @@ enum skl_disp_power_wells {
 						 GEN6_PM_RP_DOWN_THRESHOLD | \
 						 GEN6_PM_RP_DOWN_TIMEOUT)
 
-#define GEN7_GT_SCRATCH_BASE			0x4F100
+#define GEN7_GT_SCRATCH(i)			(0x4F100 + (i) * 4)
 #define GEN7_GT_SCRATCH_REG_NUM			8
 
 #define VLV_GTLC_SURVIVABILITY_REG              0x130098
-- 
cgit v0.10.2


From 4d110c71fb87c5601832e918c36af8cd4ef739f9 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 18 Sep 2015 20:03:18 +0300
Subject: drm/i915: Parametrize FBC_TAG registers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 90ce688..ffd0f6b 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -2011,7 +2011,7 @@ enum skl_disp_power_wells {
 #define   FBC_CTL_CPU_FENCE	(1<<1)
 #define   FBC_CTL_PLANE(plane)	((plane)<<0)
 #define FBC_FENCE_OFF		0x03218 /* BSpec typo has 321Bh */
-#define FBC_TAG			0x03300
+#define FBC_TAG(i)		(0x03300 + (i) * 4)
 
 #define FBC_STATUS2		0x43214
 #define  FBC_COMPRESSION_MASK	0x7ff
diff --git a/drivers/gpu/drm/i915/intel_fbc.c b/drivers/gpu/drm/i915/intel_fbc.c
index d38f464..6777fbb 100644
--- a/drivers/gpu/drm/i915/intel_fbc.c
+++ b/drivers/gpu/drm/i915/intel_fbc.c
@@ -101,7 +101,7 @@ static void i8xx_fbc_enable(struct intel_crtc *crtc)
 
 	/* Clear old tags */
 	for (i = 0; i < (FBC_LL_SIZE / 32) + 1; i++)
-		I915_WRITE(FBC_TAG + (i * 4), 0);
+		I915_WRITE(FBC_TAG(i), 0);
 
 	if (IS_GEN4(dev_priv)) {
 		u32 fbc_ctl2;
-- 
cgit v0.10.2


From 616847e7e8569a28e46a56d4cc1ac28f061e29e4 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 18 Sep 2015 20:03:19 +0300
Subject: drm/i915: Parametrize ILK turbo registers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index ffd0f6b..07fcd06 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -2579,7 +2579,7 @@ enum skl_disp_power_wells {
 #define   TSFS_INTR_MASK	0x000000ff
 
 #define CRSTANDVID		0x11100
-#define PXVFREQ_BASE		0x11110 /* P[0-15]VIDFREQ (0x1114c) (Ironlake) */
+#define PXVFREQ(i)		(0x11110 + (i) * 4) /* P[0-15]VIDFREQ (0x1114c) (Ironlake) */
 #define   PXVFREQ_PX_MASK	0x7f000000
 #define   PXVFREQ_PX_SHIFT	24
 #define VIDFREQ_BASE		0x11110
@@ -2763,8 +2763,8 @@ enum skl_disp_power_wells {
 #define CSIEW0			0x11250
 #define CSIEW1			0x11254
 #define CSIEW2			0x11258
-#define PEW			0x1125c
-#define DEW			0x11270
+#define PEW(i)			(0x1125c + (i) * 4) /* 5 registers */
+#define DEW(i)			(0x11270 + (i) * 4) /* 3 registers */
 #define MCHAFE			0x112c0
 #define CSIEC			0x112e0
 #define DMIEC			0x112e4
@@ -2788,8 +2788,8 @@ enum skl_disp_power_wells {
 #define EG5			0x11624
 #define EG6			0x11628
 #define EG7			0x1162c
-#define PXW			0x11664
-#define PXWL			0x11680
+#define PXW(i)			(0x11664 + (i) * 4) /* 4 registers */
+#define PXWL(i)			(0x11680 + (i) * 4) /* 8 registers */
 #define LCFUSE02		0x116c0
 #define   LCFUSE_HIV_MASK	0x000000ff
 #define CSIPLL0			0x12c10
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index 566f5cd..cb747f1 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -4282,7 +4282,7 @@ static void ironlake_enable_drps(struct drm_device *dev)
 	fstart = (rgvmodectl & MEMMODE_FSTART_MASK) >>
 		MEMMODE_FSTART_SHIFT;
 
-	vstart = (I915_READ(PXVFREQ_BASE + (fstart * 4)) & PXVFREQ_PX_MASK) >>
+	vstart = (I915_READ(PXVFREQ(fstart)) & PXVFREQ_PX_MASK) >>
 		PXVFREQ_PX_SHIFT;
 
 	dev_priv->ips.fmax = fmax; /* IPS callback will increase this */
@@ -5915,7 +5915,7 @@ static unsigned long __i915_gfx_val(struct drm_i915_private *dev_priv)
 
 	assert_spin_locked(&mchdev_lock);
 
-	pxvid = I915_READ(PXVFREQ_BASE + (dev_priv->rps.cur_freq * 4));
+	pxvid = I915_READ(PXVFREQ(dev_priv->rps.cur_freq));
 	pxvid = (pxvid >> 24) & 0x7f;
 	ext_v = pvid_to_extvid(dev_priv, pxvid);
 
@@ -6158,13 +6158,13 @@ static void intel_init_emon(struct drm_device *dev)
 	I915_WRITE(CSIEW2, 0x04000004);
 
 	for (i = 0; i < 5; i++)
-		I915_WRITE(PEW + (i * 4), 0);
+		I915_WRITE(PEW(i), 0);
 	for (i = 0; i < 3; i++)
-		I915_WRITE(DEW + (i * 4), 0);
+		I915_WRITE(DEW(i), 0);
 
 	/* Program P-state weights to account for frequency power adjustment */
 	for (i = 0; i < 16; i++) {
-		u32 pxvidfreq = I915_READ(PXVFREQ_BASE + (i * 4));
+		u32 pxvidfreq = I915_READ(PXVFREQ(i));
 		unsigned long freq = intel_pxfreq(pxvidfreq);
 		unsigned long vid = (pxvidfreq & PXVFREQ_PX_MASK) >>
 			PXVFREQ_PX_SHIFT;
@@ -6185,7 +6185,7 @@ static void intel_init_emon(struct drm_device *dev)
 	for (i = 0; i < 4; i++) {
 		u32 val = (pxw[i*4] << 24) | (pxw[(i*4)+1] << 16) |
 			(pxw[(i*4)+2] << 8) | (pxw[(i*4)+3]);
-		I915_WRITE(PXW + (i * 4), val);
+		I915_WRITE(PXW(i), val);
 	}
 
 	/* Adjust magic regs to magic values (more experimental results) */
@@ -6201,7 +6201,7 @@ static void intel_init_emon(struct drm_device *dev)
 	I915_WRITE(EG7, 0);
 
 	for (i = 0; i < 8; i++)
-		I915_WRITE(PXWL + (i * 4), 0);
+		I915_WRITE(PXWL(i), 0);
 
 	/* Enable PMON + select events */
 	I915_WRITE(ECR, 0x80000019);
-- 
cgit v0.10.2


From 7d81c3e02b685074e437cdb48ea3a9efd82873c5 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 18 Sep 2015 20:03:20 +0300
Subject: drm/i915: Replace raw numbers with the approproate register name in
 ILK turbo code
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index cb747f1..6870069 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -4313,10 +4313,10 @@ static void ironlake_enable_drps(struct drm_device *dev)
 
 	ironlake_set_drps(dev, fstart);
 
-	dev_priv->ips.last_count1 = I915_READ(0x112e4) + I915_READ(0x112e8) +
-		I915_READ(0x112e0);
+	dev_priv->ips.last_count1 = I915_READ(DMIEC) +
+		I915_READ(DDREC) + I915_READ(CSIEC);
 	dev_priv->ips.last_time1 = jiffies_to_msecs(jiffies);
-	dev_priv->ips.last_count2 = I915_READ(0x112f4);
+	dev_priv->ips.last_count2 = I915_READ(GFXEC);
 	dev_priv->ips.last_time2 = ktime_get_raw_ns();
 
 	spin_unlock_irq(&mchdev_lock);
-- 
cgit v0.10.2


From 184d7c06227349aad07921e8da63f61ab67e6727 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 18 Sep 2015 20:03:21 +0300
Subject: drm/i915: Parametrize TV luma/chroma filter registers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 07fcd06..dad7307 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -4082,14 +4082,10 @@ enum skl_disp_power_wells {
 # define TV_CC_DATA_1_MASK		0x0000007f
 # define TV_CC_DATA_1_SHIFT		0
 
-#define TV_H_LUMA_0		0x68100
-#define TV_H_LUMA_59		0x681ec
-#define TV_H_CHROMA_0		0x68200
-#define TV_H_CHROMA_59		0x682ec
-#define TV_V_LUMA_0		0x68300
-#define TV_V_LUMA_42		0x683a8
-#define TV_V_CHROMA_0		0x68400
-#define TV_V_CHROMA_42		0x684a8
+#define TV_H_LUMA(i)		(0x68100 + (i) * 4) /* 60 registers */
+#define TV_H_CHROMA(i)		(0x68200 + (i) * 4) /* 60 registers */
+#define TV_V_LUMA(i)		(0x68300 + (i) * 4) /* 43 registers */
+#define TV_V_CHROMA(i)		(0x68400 + (i) * 4) /* 43 registers */
 
 /* Display Port */
 #define DP_A				0x64000 /* eDP */
diff --git a/drivers/gpu/drm/i915/intel_tv.c b/drivers/gpu/drm/i915/intel_tv.c
index cbe39dc..c69e1af 100644
--- a/drivers/gpu/drm/i915/intel_tv.c
+++ b/drivers/gpu/drm/i915/intel_tv.c
@@ -1138,13 +1138,13 @@ static void intel_tv_pre_enable(struct intel_encoder *encoder)
 
 	j = 0;
 	for (i = 0; i < 60; i++)
-		I915_WRITE(TV_H_LUMA_0 + (i<<2), tv_mode->filter_table[j++]);
+		I915_WRITE(TV_H_LUMA(i), tv_mode->filter_table[j++]);
 	for (i = 0; i < 60; i++)
-		I915_WRITE(TV_H_CHROMA_0 + (i<<2), tv_mode->filter_table[j++]);
+		I915_WRITE(TV_H_CHROMA(i), tv_mode->filter_table[j++]);
 	for (i = 0; i < 43; i++)
-		I915_WRITE(TV_V_LUMA_0 + (i<<2), tv_mode->filter_table[j++]);
+		I915_WRITE(TV_V_LUMA(i), tv_mode->filter_table[j++]);
 	for (i = 0; i < 43; i++)
-		I915_WRITE(TV_V_CHROMA_0 + (i<<2), tv_mode->filter_table[j++]);
+		I915_WRITE(TV_V_CHROMA(i), tv_mode->filter_table[j++]);
 	I915_WRITE(TV_DAC, I915_READ(TV_DAC) & TV_DAC_SAVE);
 	I915_WRITE(TV_CTL, tv_ctl);
 }
-- 
cgit v0.10.2


From 9712e68840cf09d138be6226f6e6f6dcd9b009f1 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 18 Sep 2015 20:03:22 +0300
Subject: drm/i915: Parametrize DDI_BUF_TRANS registers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

FIXME: Should there be a WARN(i != 9) or something, or what does the
entry 9 comment mean?

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index dad7307..56157eb 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -7197,7 +7197,8 @@ enum skl_disp_power_wells {
 /* DDI Buffer Translations */
 #define DDI_BUF_TRANS_A				0x64E00
 #define DDI_BUF_TRANS_B				0x64E60
-#define DDI_BUF_TRANS(port) _PORT(port, DDI_BUF_TRANS_A, DDI_BUF_TRANS_B)
+#define DDI_BUF_TRANS_LO(port, i) (_PORT(port, DDI_BUF_TRANS_A, DDI_BUF_TRANS_B) + (i) * 8)
+#define DDI_BUF_TRANS_HI(port, i) (_PORT(port, DDI_BUF_TRANS_A, DDI_BUF_TRANS_B) + (i) * 8 + 4)
 
 /* Sideband Interface (SBI) is programmed indirectly, via
  * SBI_ADDR, which contains the register offset; and SBI_DATA,
diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
index 5b600bf..9e640ea 100644
--- a/drivers/gpu/drm/i915/intel_ddi.c
+++ b/drivers/gpu/drm/i915/intel_ddi.c
@@ -414,7 +414,6 @@ static void intel_prepare_ddi_buffers(struct drm_device *dev, enum port port,
 				      bool supports_hdmi)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 reg;
 	u32 iboost_bit = 0;
 	int i, n_hdmi_entries, n_dp_entries, n_edp_entries, hdmi_default_entry,
 	    size;
@@ -505,11 +504,11 @@ static void intel_prepare_ddi_buffers(struct drm_device *dev, enum port port,
 		BUG();
 	}
 
-	for (i = 0, reg = DDI_BUF_TRANS(port); i < size; i++) {
-		I915_WRITE(reg, ddi_translations[i].trans1 | iboost_bit);
-		reg += 4;
-		I915_WRITE(reg, ddi_translations[i].trans2);
-		reg += 4;
+	for (i = 0; i < size; i++) {
+		I915_WRITE(DDI_BUF_TRANS_LO(port, i),
+			   ddi_translations[i].trans1 | iboost_bit);
+		I915_WRITE(DDI_BUF_TRANS_HI(port, i),
+			   ddi_translations[i].trans2);
 	}
 
 	if (!supports_hdmi)
@@ -521,10 +520,10 @@ static void intel_prepare_ddi_buffers(struct drm_device *dev, enum port port,
 		hdmi_level = hdmi_default_entry;
 
 	/* Entry 9 is for HDMI: */
-	I915_WRITE(reg, ddi_translations_hdmi[hdmi_level].trans1 | iboost_bit);
-	reg += 4;
-	I915_WRITE(reg, ddi_translations_hdmi[hdmi_level].trans2);
-	reg += 4;
+	I915_WRITE(DDI_BUF_TRANS_LO(port, i),
+		   ddi_translations_hdmi[hdmi_level].trans1 | iboost_bit);
+	I915_WRITE(DDI_BUF_TRANS_HI(port, i),
+		   ddi_translations_hdmi[hdmi_level].trans2);
 }
 
 /* Program DDI buffers translations for DP. By default, program ports A-D in DP
-- 
cgit v0.10.2


From d2aa5ae8e721bfe319e68ef5c08380385d452787 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 18 Sep 2015 20:03:23 +0300
Subject: drm/i915: Parametrize CSR_PROGRAM registers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_csr.c b/drivers/gpu/drm/i915/intel_csr.c
index 780d5d5..61b4a47 100644
--- a/drivers/gpu/drm/i915/intel_csr.c
+++ b/drivers/gpu/drm/i915/intel_csr.c
@@ -50,7 +50,7 @@ MODULE_FIRMWARE(I915_CSR_BXT);
 /*
 * SKL CSR registers for DC5 and DC6
 */
-#define CSR_PROGRAM_BASE		0x80000
+#define CSR_PROGRAM(i)			(0x80000 + (i) * 4)
 #define CSR_SSP_BASE_ADDR_GEN9		0x00002FC0
 #define CSR_HTP_ADDR_SKL		0x00500034
 #define CSR_SSP_BASE			0x8F074
@@ -268,8 +268,7 @@ void intel_csr_load_program(struct drm_device *dev)
 	mutex_lock(&dev_priv->csr_lock);
 	fw_size = dev_priv->csr.dmc_fw_size;
 	for (i = 0; i < fw_size; i++)
-		I915_WRITE(CSR_PROGRAM_BASE + i * 4,
-			payload[i]);
+		I915_WRITE(CSR_PROGRAM(i), payload[i]);
 
 	for (i = 0; i < dev_priv->csr.mmio_count; i++) {
 		I915_WRITE(dev_priv->csr.mmioaddr[i],
@@ -471,7 +470,7 @@ void assert_csr_loaded(struct drm_i915_private *dev_priv)
 {
 	WARN_ONCE(intel_csr_load_status_get(dev_priv) != FW_LOADED,
 		  "CSR is not loaded.\n");
-	WARN_ONCE(!I915_READ(CSR_PROGRAM_BASE),
+	WARN_ONCE(!I915_READ(CSR_PROGRAM(0)),
 		  "CSR program storage start is NULL\n");
 	WARN_ONCE(!I915_READ(CSR_SSP_BASE), "CSR SSP Base Not fine\n");
 	WARN_ONCE(!I915_READ(CSR_HTP_SKL), "CSR HTP Not fine\n");
-- 
cgit v0.10.2


From 5ca4163a612068d8f942c454218d3d631f22af1b Mon Sep 17 00:00:00 2001
From: Arun Siluvery <arun.siluvery@linux.intel.com>
Date: Fri, 18 Sep 2015 17:52:47 +0100
Subject: drm/i915/bxt: Update revision id for BXT C0

Cc: Nick Hoath <nicholas.hoath@intel.com>
Cc: Imre Deak <imre.deak@intel.com>
Signed-off-by: Arun Siluvery <arun.siluvery@linux.intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index ba4596e..6c3ef9c 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -2507,7 +2507,7 @@ struct drm_i915_cmd_table {
 
 #define BXT_REVID_A0		(0x0)
 #define BXT_REVID_B0		(0x3)
-#define BXT_REVID_C0		(0x6)
+#define BXT_REVID_C0		(0x9)
 
 /*
  * The genX designation typically refers to the render engine, so render
-- 
cgit v0.10.2


From fe47ea0cfa4adffe2a20edb22376d7af76b6a74c Mon Sep 17 00:00:00 2001
From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Date: Mon, 21 Sep 2015 10:45:32 +0100
Subject: drm/i915: Support planar formats in tile height calculations

This will be needed for NV12 support.

v2: Rebase.

Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index bf8754f..1f3af24 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -2190,7 +2190,7 @@ static bool need_vtd_wa(struct drm_device *dev)
 
 unsigned int
 intel_tile_height(struct drm_device *dev, uint32_t pixel_format,
-		  uint64_t fb_format_modifier)
+		  uint64_t fb_format_modifier, unsigned int plane)
 {
 	unsigned int tile_height;
 	uint32_t pixel_bytes;
@@ -2206,7 +2206,7 @@ intel_tile_height(struct drm_device *dev, uint32_t pixel_format,
 		tile_height = 32;
 		break;
 	case I915_FORMAT_MOD_Yf_TILED:
-		pixel_bytes = drm_format_plane_cpp(pixel_format, 0);
+		pixel_bytes = drm_format_plane_cpp(pixel_format, plane);
 		switch (pixel_bytes) {
 		default:
 		case 1:
@@ -2240,7 +2240,7 @@ intel_fb_align_height(struct drm_device *dev, unsigned int height,
 		      uint32_t pixel_format, uint64_t fb_format_modifier)
 {
 	return ALIGN(height, intel_tile_height(dev, pixel_format,
-					       fb_format_modifier));
+					       fb_format_modifier, 0));
 }
 
 static int
@@ -2266,7 +2266,7 @@ intel_fill_fb_ggtt_view(struct i915_ggtt_view *view, struct drm_framebuffer *fb,
 	info->fb_modifier = fb->modifier[0];
 
 	tile_height = intel_tile_height(fb->dev, fb->pixel_format,
-					fb->modifier[0]);
+					fb->modifier[0], 0);
 	tile_pitch = PAGE_SIZE / tile_height;
 	info->width_pages = DIV_ROUND_UP(fb->pitches[0], tile_pitch);
 	info->height_pages = DIV_ROUND_UP(fb->height, tile_height);
@@ -3075,7 +3075,7 @@ static void skylake_update_primary_plane(struct drm_crtc *crtc,
 	if (intel_rotation_90_or_270(rotation)) {
 		/* stride = Surface height in tiles */
 		tile_height = intel_tile_height(dev, fb->pixel_format,
-						fb->modifier[0]);
+						fb->modifier[0], 0);
 		stride = DIV_ROUND_UP(fb->height, tile_height);
 		x_offset = stride * tile_height - y - src_h;
 		y_offset = x;
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index ab3c296..858c382 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -1089,7 +1089,7 @@ int intel_plane_atomic_calc_changes(struct drm_crtc_state *crtc_state,
 
 unsigned int
 intel_tile_height(struct drm_device *dev, uint32_t pixel_format,
-		  uint64_t fb_format_modifier);
+		  uint64_t fb_format_modifier, unsigned int plane);
 
 static inline bool
 intel_rotation_90_or_270(unsigned int rotation)
diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c
index 79f7cc2..4372fa0 100644
--- a/drivers/gpu/drm/i915/intel_sprite.c
+++ b/drivers/gpu/drm/i915/intel_sprite.c
@@ -240,7 +240,7 @@ skl_update_plane(struct drm_plane *drm_plane, struct drm_crtc *crtc,
 	if (intel_rotation_90_or_270(rotation)) {
 		/* stride: Surface height in tiles */
 		tile_height = intel_tile_height(dev, fb->pixel_format,
-						fb->modifier[0]);
+						fb->modifier[0], 0);
 		stride = DIV_ROUND_UP(fb->height, tile_height);
 		plane_size = (src_w << 16) | src_h;
 		x_offset = stride * tile_height - y - (src_h + 1);
-- 
cgit v0.10.2


From 804beb4b5b76e41352b41027ab2804b8462c1686 Mon Sep 17 00:00:00 2001
From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Date: Mon, 21 Sep 2015 10:45:33 +0100
Subject: drm/i915: Support appending to the rotated pages mapping

By providing a start offset into the source array of pages, and returning the
end position in the scatter-gather table, we will be able to append the UV
plane to the rotated mapping in later patches.

v2: Rebase.

Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 01f3521..2a8f64d 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -3244,15 +3244,18 @@ i915_gem_obj_lookup_or_create_ggtt_vma(struct drm_i915_gem_object *obj,
 
 }
 
-static void
-rotate_pages(dma_addr_t *in, unsigned int width, unsigned int height,
-	     struct sg_table *st)
+static struct scatterlist *
+rotate_pages(dma_addr_t *in, unsigned int offset,
+	     unsigned int width, unsigned int height,
+	     struct sg_table *st, struct scatterlist *sg)
 {
 	unsigned int column, row;
 	unsigned int src_idx;
-	struct scatterlist *sg = st->sgl;
 
-	st->nents = 0;
+	if (!sg) {
+		st->nents = 0;
+		sg = st->sgl;
+	}
 
 	for (column = 0; column < width; column++) {
 		src_idx = width * (height - 1) + column;
@@ -3263,12 +3266,14 @@ rotate_pages(dma_addr_t *in, unsigned int width, unsigned int height,
 			 * The only thing we need are DMA addresses.
 			 */
 			sg_set_page(sg, NULL, PAGE_SIZE, 0);
-			sg_dma_address(sg) = in[src_idx];
+			sg_dma_address(sg) = in[offset + src_idx];
 			sg_dma_len(sg) = PAGE_SIZE;
 			sg = sg_next(sg);
 			src_idx -= width;
 		}
 	}
+
+	return sg;
 }
 
 static struct sg_table *
@@ -3306,9 +3311,9 @@ intel_rotate_fb_obj_pages(struct i915_ggtt_view *ggtt_view,
 	}
 
 	/* Rotate the pages. */
-	rotate_pages(page_addr_list,
+	rotate_pages(page_addr_list, 0,
 		     rot_info->width_pages, rot_info->height_pages,
-		     st);
+		     st, NULL);
 
 	DRM_DEBUG_KMS(
 		      "Created rotated page mapping for object size %zu (pitch=%u, height=%u, pixel_format=0x%x, %ux%u tiles, %u pages).\n",
-- 
cgit v0.10.2


From 89e3e1427629027dc33e576fc002880a02a7e50c Mon Sep 17 00:00:00 2001
From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Date: Mon, 21 Sep 2015 10:45:34 +0100
Subject: drm/i915: Support NV12 in rotated GGTT mapping

Just adding the rotated UV plane at the end of the rotated Y plane.

v2: Rebase.

Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 2a8f64d..ad0e7e0 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -3282,10 +3282,13 @@ intel_rotate_fb_obj_pages(struct i915_ggtt_view *ggtt_view,
 {
 	struct intel_rotation_info *rot_info = &ggtt_view->rotation_info;
 	unsigned int size_pages = rot_info->size >> PAGE_SHIFT;
+	unsigned int size_pages_uv;
 	struct sg_page_iter sg_iter;
 	unsigned long i;
 	dma_addr_t *page_addr_list;
 	struct sg_table *st;
+	unsigned int uv_start_page;
+	struct scatterlist *sg;
 	int ret = -ENOMEM;
 
 	/* Allocate a temporary list of source pages for random access. */
@@ -3294,12 +3297,18 @@ intel_rotate_fb_obj_pages(struct i915_ggtt_view *ggtt_view,
 	if (!page_addr_list)
 		return ERR_PTR(ret);
 
+	/* Account for UV plane with NV12. */
+	if (rot_info->pixel_format == DRM_FORMAT_NV12)
+		size_pages_uv = rot_info->size_uv >> PAGE_SHIFT;
+	else
+		size_pages_uv = 0;
+
 	/* Allocate target SG list. */
 	st = kmalloc(sizeof(*st), GFP_KERNEL);
 	if (!st)
 		goto err_st_alloc;
 
-	ret = sg_alloc_table(st, size_pages, GFP_KERNEL);
+	ret = sg_alloc_table(st, size_pages + size_pages_uv, GFP_KERNEL);
 	if (ret)
 		goto err_sg_alloc;
 
@@ -3311,15 +3320,30 @@ intel_rotate_fb_obj_pages(struct i915_ggtt_view *ggtt_view,
 	}
 
 	/* Rotate the pages. */
-	rotate_pages(page_addr_list, 0,
+	sg = rotate_pages(page_addr_list, 0,
 		     rot_info->width_pages, rot_info->height_pages,
 		     st, NULL);
 
+	/* Append the UV plane if NV12. */
+	if (rot_info->pixel_format == DRM_FORMAT_NV12) {
+		uv_start_page = size_pages;
+
+		/* Check for tile-row un-alignment. */
+		if (offset_in_page(rot_info->uv_offset))
+			uv_start_page--;
+
+		rotate_pages(page_addr_list, uv_start_page,
+			     rot_info->width_pages_uv,
+			     rot_info->height_pages_uv,
+			     st, sg);
+	}
+
 	DRM_DEBUG_KMS(
-		      "Created rotated page mapping for object size %zu (pitch=%u, height=%u, pixel_format=0x%x, %ux%u tiles, %u pages).\n",
+		      "Created rotated page mapping for object size %zu (pitch=%u, height=%u, pixel_format=0x%x, %ux%u tiles, %u pages (%u plane 0)).\n",
 		      obj->base.size, rot_info->pitch, rot_info->height,
 		      rot_info->pixel_format, rot_info->width_pages,
-		      rot_info->height_pages, size_pages);
+		      rot_info->height_pages, size_pages + size_pages_uv,
+		      size_pages);
 
 	drm_free_large(page_addr_list);
 
@@ -3331,10 +3355,11 @@ err_st_alloc:
 	drm_free_large(page_addr_list);
 
 	DRM_DEBUG_KMS(
-		      "Failed to create rotated mapping for object size %zu! (%d) (pitch=%u, height=%u, pixel_format=0x%x, %ux%u tiles, %u pages)\n",
+		      "Failed to create rotated mapping for object size %zu! (%d) (pitch=%u, height=%u, pixel_format=0x%x, %ux%u tiles, %u pages (%u plane 0))\n",
 		      obj->base.size, ret, rot_info->pitch, rot_info->height,
 		      rot_info->pixel_format, rot_info->width_pages,
-		      rot_info->height_pages, size_pages);
+		      rot_info->height_pages, size_pages + size_pages_uv,
+		      size_pages);
 	return ERR_PTR(ret);
 }
 
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.h b/drivers/gpu/drm/i915/i915_gem_gtt.h
index 96ebb98..c807586 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.h
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.h
@@ -138,10 +138,13 @@ enum i915_ggtt_view_type {
 struct intel_rotation_info {
 	unsigned int height;
 	unsigned int pitch;
+	unsigned int uv_offset;
 	uint32_t pixel_format;
 	uint64_t fb_modifier;
 	unsigned int width_pages, height_pages;
 	uint64_t size;
+	unsigned int width_pages_uv, height_pages_uv;
+	uint64_t size_uv;
 };
 
 struct i915_ggtt_view {
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 1f3af24..bdce458 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -2263,6 +2263,7 @@ intel_fill_fb_ggtt_view(struct i915_ggtt_view *view, struct drm_framebuffer *fb,
 	info->height = fb->height;
 	info->pixel_format = fb->pixel_format;
 	info->pitch = fb->pitches[0];
+	info->uv_offset = fb->offsets[1];
 	info->fb_modifier = fb->modifier[0];
 
 	tile_height = intel_tile_height(fb->dev, fb->pixel_format,
@@ -2272,6 +2273,17 @@ intel_fill_fb_ggtt_view(struct i915_ggtt_view *view, struct drm_framebuffer *fb,
 	info->height_pages = DIV_ROUND_UP(fb->height, tile_height);
 	info->size = info->width_pages * info->height_pages * PAGE_SIZE;
 
+	if (info->pixel_format == DRM_FORMAT_NV12) {
+		tile_height = intel_tile_height(fb->dev, fb->pixel_format,
+						fb->modifier[0], 1);
+		tile_pitch = PAGE_SIZE / tile_height;
+		info->width_pages_uv = DIV_ROUND_UP(fb->pitches[0], tile_pitch);
+		info->height_pages_uv = DIV_ROUND_UP(fb->height / 2,
+						     tile_height);
+		info->size_uv = info->width_pages_uv * info->height_pages_uv *
+				PAGE_SIZE;
+	}
+
 	return 0;
 }
 
-- 
cgit v0.10.2


From dedf278ce69cac6b29c324d280183da31093e0b0 Mon Sep 17 00:00:00 2001
From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Date: Mon, 21 Sep 2015 10:45:35 +0100
Subject: drm/i915: Enable querying offset of UV plane with
 intel_plane_obj_offset

v2: Rebase.

Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index ad0e7e0..bbb96a4 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -3332,6 +3332,8 @@ intel_rotate_fb_obj_pages(struct i915_ggtt_view *ggtt_view,
 		if (offset_in_page(rot_info->uv_offset))
 			uv_start_page--;
 
+		rot_info->uv_start_page = uv_start_page;
+
 		rotate_pages(page_addr_list, uv_start_page,
 			     rot_info->width_pages_uv,
 			     rot_info->height_pages_uv,
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.h b/drivers/gpu/drm/i915/i915_gem_gtt.h
index c807586..9fbb07d 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.h
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.h
@@ -145,6 +145,7 @@ struct intel_rotation_info {
 	uint64_t size;
 	unsigned int width_pages_uv, height_pages_uv;
 	uint64_t size_uv;
+	unsigned int uv_start_page;
 };
 
 struct i915_ggtt_view {
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index bdce458..9b19891 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -2897,14 +2897,29 @@ u32 intel_fb_stride_alignment(struct drm_device *dev, uint64_t fb_modifier,
 }
 
 unsigned long intel_plane_obj_offset(struct intel_plane *intel_plane,
-				     struct drm_i915_gem_object *obj)
+				     struct drm_i915_gem_object *obj,
+				     unsigned int plane)
 {
 	const struct i915_ggtt_view *view = &i915_ggtt_view_normal;
+	struct i915_vma *vma;
+	unsigned char *offset;
 
 	if (intel_rotation_90_or_270(intel_plane->base.state->rotation))
 		view = &i915_ggtt_view_rotated;
 
-	return i915_gem_obj_ggtt_offset_view(obj, view);
+	vma = i915_gem_obj_to_ggtt_view(obj, view);
+	if (WARN(!vma, "ggtt vma for display object not found! (view=%u)\n",
+		view->type))
+		return -1;
+
+	offset = (unsigned char *)vma->node.start;
+
+	if (plane == 1) {
+		offset += vma->ggtt_view.rotation_info.uv_start_page *
+			  PAGE_SIZE;
+	}
+
+	return (unsigned long)offset;
 }
 
 static void skl_detach_scaler(struct intel_crtc *intel_crtc, int id)
@@ -3060,7 +3075,7 @@ static void skylake_update_primary_plane(struct drm_crtc *crtc,
 	obj = intel_fb_obj(fb);
 	stride_div = intel_fb_stride_alignment(dev, fb->modifier[0],
 					       fb->pixel_format);
-	surf_addr = intel_plane_obj_offset(to_intel_plane(plane), obj);
+	surf_addr = intel_plane_obj_offset(to_intel_plane(plane), obj, 0);
 
 	/*
 	 * FIXME: intel_plane_state->src, dst aren't set when transitional
@@ -11423,8 +11438,9 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	if (ret)
 		goto cleanup_pending;
 
-	work->gtt_offset = intel_plane_obj_offset(to_intel_plane(primary), obj)
-						  + intel_crtc->dspaddr_offset;
+	work->gtt_offset = intel_plane_obj_offset(to_intel_plane(primary),
+						  obj, 0);
+	work->gtt_offset += intel_crtc->dspaddr_offset;
 
 	if (mmio_flip) {
 		ret = intel_queue_mmio_flip(dev, crtc, fb, obj, ring,
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 858c382..bc674aa 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -1170,7 +1170,9 @@ int skl_update_scaler_crtc(struct intel_crtc_state *crtc_state);
 int skl_max_scale(struct intel_crtc *crtc, struct intel_crtc_state *crtc_state);
 
 unsigned long intel_plane_obj_offset(struct intel_plane *intel_plane,
-				     struct drm_i915_gem_object *obj);
+				     struct drm_i915_gem_object *obj,
+				     unsigned int plane);
+
 u32 skl_plane_ctl_format(uint32_t pixel_format);
 u32 skl_plane_ctl_tiling(uint64_t fb_modifier);
 u32 skl_plane_ctl_rotation(unsigned int rotation);
diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c
index 4372fa0..4349fde 100644
--- a/drivers/gpu/drm/i915/intel_sprite.c
+++ b/drivers/gpu/drm/i915/intel_sprite.c
@@ -235,7 +235,7 @@ skl_update_plane(struct drm_plane *drm_plane, struct drm_crtc *crtc,
 	else if (key->flags & I915_SET_COLORKEY_SOURCE)
 		plane_ctl |= PLANE_CTL_KEY_ENABLE_SOURCE;
 
-	surf_addr = intel_plane_obj_offset(intel_plane, obj);
+	surf_addr = intel_plane_obj_offset(intel_plane, obj, 0);
 
 	if (intel_rotation_90_or_270(rotation)) {
 		/* stride: Surface height in tiles */
-- 
cgit v0.10.2


From cd1736ad0a6f0f7a31773ddf90040bd15969bc21 Mon Sep 17 00:00:00 2001
From: Nick Hoath <nicholas.hoath@intel.com>
Date: Mon, 21 Sep 2015 14:02:29 +0100
Subject: drm/i915: Remove extraneous request cancel.

Remove extraneous request cancel in request allocation failure path
in intel_lr_context_deferred_alloc (Tvrtko Ursulin)

Regression from:

  commit e84fe80337dc85cca07d0417ea97edbec4789d8b
  Author: Nick Hoath <nicholas.hoath@intel.com>
  Date:   Fri Sep 11 12:53:46 2015 +0100

      drm/i915: Split alloc from init for lrc

Signed-off-by: Nick Hoath <nicholas.hoath@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
index 7417650..256167b 100644
--- a/drivers/gpu/drm/i915/intel_lrc.c
+++ b/drivers/gpu/drm/i915/intel_lrc.c
@@ -2487,7 +2487,6 @@ int intel_lr_context_deferred_alloc(struct intel_context *ctx,
 		if (ret) {
 			DRM_ERROR("ring create req: %d\n",
 				ret);
-			i915_gem_request_cancel(req);
 			goto error_ringbuf;
 		}
 
-- 
cgit v0.10.2


From 7c4a7d60bcda9449bca1ed2ee47392cd7acb8535 Mon Sep 17 00:00:00 2001
From: Chris Wilson <chris@chris-wilson.co.uk>
Date: Thu, 24 Sep 2015 11:57:45 +0100
Subject: drm/i915: Defer adding preallocated stolen objects to the VM list
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

When preallocating a stolen object during early initialisation, we may
be running before we have setup the the global GTT VM state, in
particular before we have initialised the range manager and associated
lists. As this is the case, we defer binding the stolen object until we
call i915_gem_setup_global_gtt(). Not only should we defer the binding,
but we should also defer the VM list manipulation.

Fixes regression uncovered by commit a2cad9dff4dd44d0244b966d980de9d602d87593
Author: Michał Winiarski <michal.winiarski@intel.com>
Date:   Wed Sep 16 11:49:00 2015 +0200

    drm/i915/gtt: Do not initialize drm_mm twice.

Whilst I am here remove the duplicate work leaving dangling pointers
from the error path...

v2: Typos galore before coffee.

Reported-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=92099
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Jesse Barnes <jbarnes@virtuousgeek.org>
Cc: Michel Thierry <michel.thierry@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Cc: Michał Winiarski <michal.winiarski@intel.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Reviewed-by: Michał Winiarski <michal.winiarski@intel.com>
Tested-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index bbb96a4..47344d0 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -2533,7 +2533,6 @@ static int ggtt_bind_vma(struct i915_vma *vma,
 		 * the bound flag ourselves.
 		 */
 		vma->bound |= GLOBAL_BIND;
-
 	}
 
 	if (dev_priv->mm.aliasing_ppgtt && flags & LOCAL_BIND) {
@@ -2657,6 +2656,7 @@ static int i915_gem_setup_global_gtt(struct drm_device *dev,
 			return ret;
 		}
 		vma->bound |= GLOBAL_BIND;
+		list_add_tail(&vma->mm_list, &ggtt_vm->inactive_list);
 	}
 
 	/* Clear any non-preallocated blocks */
diff --git a/drivers/gpu/drm/i915/i915_gem_stolen.c b/drivers/gpu/drm/i915/i915_gem_stolen.c
index 55df6ce..1520779 100644
--- a/drivers/gpu/drm/i915/i915_gem_stolen.c
+++ b/drivers/gpu/drm/i915/i915_gem_stolen.c
@@ -584,7 +584,7 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,
 	vma = i915_gem_obj_lookup_or_create_vma(obj, ggtt);
 	if (IS_ERR(vma)) {
 		ret = PTR_ERR(vma);
-		goto err_out;
+		goto err;
 	}
 
 	/* To simplify the initialisation sequence between KMS and GTT,
@@ -598,23 +598,19 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,
 		ret = drm_mm_reserve_node(&ggtt->mm, &vma->node);
 		if (ret) {
 			DRM_DEBUG_KMS("failed to allocate stolen GTT space\n");
-			goto err_vma;
+			goto err;
 		}
-	}
 
-	vma->bound |= GLOBAL_BIND;
+		vma->bound |= GLOBAL_BIND;
+		list_add_tail(&vma->mm_list, &ggtt->inactive_list);
+	}
 
 	list_add_tail(&obj->global_list, &dev_priv->mm.bound_list);
-	list_add_tail(&vma->mm_list, &ggtt->inactive_list);
 	i915_gem_object_pin_pages(obj);
 
 	return obj;
 
-err_vma:
-	i915_gem_vma_destroy(vma);
-err_out:
-	i915_gem_stolen_remove_node(dev_priv, stolen);
-	kfree(stolen);
+err:
 	drm_gem_object_unreference(&obj->base);
 	return NULL;
 }
-- 
cgit v0.10.2


From 1da248a583de4e63a006dd84d9d53e6784b958a3 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 14 Sep 2015 22:43:42 +0300
Subject: drm: s/int crtc/unsigned int pipe/ straggles
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Finish the recent replacement of 'int pipe' with 'unsigned int pipe'

Cc: Thierry Reding <treding@nvidia.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Thierry Reding <treding@nvidia.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c
index 22d207e..8df4133 100644
--- a/drivers/gpu/drm/drm_irq.c
+++ b/drivers/gpu/drm/drm_irq.c
@@ -74,11 +74,11 @@ module_param_named(vblankoffdelay, drm_vblank_offdelay, int, 0600);
 module_param_named(timestamp_precision_usec, drm_timestamp_precision, int, 0600);
 module_param_named(timestamp_monotonic, drm_timestamp_monotonic, int, 0600);
 
-static void store_vblank(struct drm_device *dev, int crtc,
+static void store_vblank(struct drm_device *dev, unsigned int pipe,
 			 u32 vblank_count_inc,
 			 struct timeval *t_vblank)
 {
-	struct drm_vblank_crtc *vblank = &dev->vblank[crtc];
+	struct drm_vblank_crtc *vblank = &dev->vblank[pipe];
 	u32 tslot;
 
 	assert_spin_locked(&dev->vblank_time_lock);
@@ -88,7 +88,7 @@ static void store_vblank(struct drm_device *dev, int crtc,
 		 * the latching of vblank->count below.
 		 */
 		tslot = vblank->count + vblank_count_inc;
-		vblanktimestamp(dev, crtc, tslot) = *t_vblank;
+		vblanktimestamp(dev, pipe, tslot) = *t_vblank;
 	}
 
 	/*
@@ -110,7 +110,7 @@ static void store_vblank(struct drm_device *dev, int crtc,
  * @pipe: counter to update
  *
  * Call back into the driver to update the appropriate vblank counter
- * (specified by @crtc).  Deal with wraparound, if it occurred, and
+ * (specified by @pipe).  Deal with wraparound, if it occurred, and
  * update the last read value so we can deal with wraparound on the next
  * call if necessary.
  *
@@ -1154,8 +1154,8 @@ EXPORT_SYMBOL(drm_crtc_vblank_put);
  * @dev: DRM device
  * @pipe: CRTC index
  *
- * This waits for one vblank to pass on @crtc, using the irq driver interfaces.
- * It is a failure to call this when the vblank irq for @crtc is disabled, e.g.
+ * This waits for one vblank to pass on @pipe, using the irq driver interfaces.
+ * It is a failure to call this when the vblank irq for @pipe is disabled, e.g.
  * due to lack of driver support or because the crtc is off.
  */
 void drm_wait_one_vblank(struct drm_device *dev, unsigned int pipe)
@@ -1288,8 +1288,8 @@ void drm_crtc_vblank_reset(struct drm_crtc *drm_crtc)
 {
 	struct drm_device *dev = drm_crtc->dev;
 	unsigned long irqflags;
-	int crtc = drm_crtc_index(drm_crtc);
-	struct drm_vblank_crtc *vblank = &dev->vblank[crtc];
+	unsigned int pipe = drm_crtc_index(drm_crtc);
+	struct drm_vblank_crtc *vblank = &dev->vblank[pipe];
 
 	spin_lock_irqsave(&dev->vbl_lock, irqflags);
 	/*
-- 
cgit v0.10.2


From f15a66e68422ca6bb783142780ad440067f6cc89 Mon Sep 17 00:00:00 2001
From: Lukas Wunner <lukas@wunner.de>
Date: Sat, 5 Sep 2015 11:22:39 +0200
Subject: drm: Spell vga_switcheroo consistently

Currently everyone and their dog has their own favourite spelling
for vga_switcheroo. This makes it hard to grep dmesg for log entries
relating to vga_switcheroo. It also makes it hard to find related
source files in the tree.

vga_switcheroo.c uses pr_fmt "vga_switcheroo". Use that everywhere.

Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/Documentation/DocBook/drm.tmpl b/Documentation/DocBook/drm.tmpl
index 9ddf8c6..30401f9 100644
--- a/Documentation/DocBook/drm.tmpl
+++ b/Documentation/DocBook/drm.tmpl
@@ -3646,7 +3646,7 @@ void (*postclose) (struct drm_device *, struct drm_file *);</synopsis>
 	plane properties to default value, so that a subsequent open of the
 	device will not inherit state from the previous user. It can also be
 	used to execute delayed power switching state changes, e.g. in
-	conjunction with the vga-switcheroo infrastructure. Beyond that KMS
+	conjunction with the vga_switcheroo infrastructure. Beyond that KMS
 	drivers should not do any further cleanup. Only legacy UMS drivers might
 	need to clean up device state so that the vga console or an independent
 	fbdev driver could take over.
diff --git a/drivers/gpu/drm/omapdrm/omap_drv.c b/drivers/gpu/drm/omapdrm/omap_drv.c
index a5f9d8b..d685e23 100644
--- a/drivers/gpu/drm/omapdrm/omap_drv.c
+++ b/drivers/gpu/drm/omapdrm/omap_drv.c
@@ -753,7 +753,7 @@ static void dev_lastclose(struct drm_device *dev)
 {
 	int i;
 
-	/* we don't support vga-switcheroo.. so just make sure the fbdev
+	/* we don't support vga_switcheroo.. so just make sure the fbdev
 	 * mode is active
 	 */
 	struct omap_drm_private *priv = dev->dev_private;
diff --git a/include/linux/fb.h b/include/linux/fb.h
index bc9afa7..be40dba 100644
--- a/include/linux/fb.h
+++ b/include/linux/fb.h
@@ -156,7 +156,7 @@ struct fb_cursor_user {
 #define FB_EVENT_GET_REQ                0x0D
 /*      Unbind from the console if possible */
 #define FB_EVENT_FB_UNBIND              0x0E
-/*      CONSOLE-SPECIFIC: remap all consoles to new fb - for vga switcheroo */
+/*      CONSOLE-SPECIFIC: remap all consoles to new fb - for vga_switcheroo */
 #define FB_EVENT_REMAP_ALL_CONSOLE      0x0F
 /*      A hardware display blank early change occured */
 #define FB_EARLY_EVENT_BLANK		0x10
-- 
cgit v0.10.2


From 4127838c460ab66f60ea8cdb069654972f8c277f Mon Sep 17 00:00:00 2001
From: Lukas Wunner <lukas@wunner.de>
Date: Sat, 5 Sep 2015 13:40:23 +0200
Subject: vga_switcheroo: Sort headers alphabetically

Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/vga/vga_switcheroo.c b/drivers/gpu/vga/vga_switcheroo.c
index b19a72f..67a5709 100644
--- a/drivers/gpu/vga/vga_switcheroo.c
+++ b/drivers/gpu/vga/vga_switcheroo.c
@@ -30,19 +30,17 @@
 
 #define pr_fmt(fmt) "vga_switcheroo: " fmt
 
-#include <linux/module.h>
-#include <linux/seq_file.h>
-#include <linux/uaccess.h>
-#include <linux/fs.h>
+#include <linux/console.h>
 #include <linux/debugfs.h>
 #include <linux/fb.h>
-
+#include <linux/fs.h>
+#include <linux/module.h>
 #include <linux/pci.h>
-#include <linux/console.h>
-#include <linux/vga_switcheroo.h>
 #include <linux/pm_runtime.h>
-
+#include <linux/seq_file.h>
+#include <linux/uaccess.h>
 #include <linux/vgaarb.h>
+#include <linux/vga_switcheroo.h>
 
 /**
  * DOC: Overview
-- 
cgit v0.10.2


From 5e7d49446b5964d2866ea1912cc9f65ab33ed76f Mon Sep 17 00:00:00 2001
From: Geliang Tang <geliangtang@163.com>
Date: Wed, 16 Sep 2015 23:25:22 -0700
Subject: drm: fix kernel-doc warnings in drm_crtc.h

Fix the following 'make htmldocs' warning:

  .//include/drm/drm_crtc.h:929: warning: Excess struct/union/enum/typedef member 'base' description in 'drm_bridge'

Signed-off-by: Geliang Tang <geliangtang@163.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/include/drm/drm_crtc.h b/include/drm/drm_crtc.h
index c0366e9..6566f72 100644
--- a/include/drm/drm_crtc.h
+++ b/include/drm/drm_crtc.h
@@ -911,7 +911,6 @@ struct drm_bridge_funcs {
  * @next: the next bridge in the encoder chain
  * @of_node: device node pointer to the bridge
  * @list: to keep track of all added bridges
- * @base: base mode object
  * @funcs: control functions
  * @driver_private: pointer to the bridge driver's internal context
  */
-- 
cgit v0.10.2


From 942840371cde152fe57c15e0e8483b760e7763e3 Mon Sep 17 00:00:00 2001
From: Matt Roper <matthew.d.roper@intel.com>
Date: Mon, 21 Sep 2015 17:21:48 -0700
Subject: drm/fbdev: Update legacy plane->fb refcounting for atomic restore

Starting with commit

        commit 28cc504e8d52248962f5b485bdc65f539e3fe21d
        Author: Rob Clark <robdclark@gmail.com>
        Date:   Tue Aug 25 15:36:00 2015 -0400

            drm/i915: enable atomic fb-helper

I've been seeing some panics on i915 when the DRM master shuts down that appear
to be caused by using an already-freed framebuffer (i.e., we're unexpectedly
dropping our initial FB's reference count to 0 and freeing it, which causes a
crash when we try to restore it later).  Digging deeper, the state FB
refcounting is working as expected, but we seem to be missing proper
refcounting on the legacy plane->fb pointers in the new atomic fbdev code.

Tracking plane->old_fb and then doing a ref/unref at the end of the
fbdev restore like we do in the legacy ioctl's ensures we don't miscount
references on plane->fb and avoids the panics.

v2 from Daniel:

Really do what the atomic ioctl does:
- Also update plane->fb and plane->crtc.
- Clear out plane->old_fb on failures too.

v3: git add everything. Oops.

v4: Also clear old_fb in all other failure paths, spotted by David.

Cc: Rob Clark <robdclark@gmail.com>
Cc: intel-gfx@lists.freedesktop.org
Cc: David Herrmann <dh.herrmann@gmail.com>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com> (v1)
Reviewd-by: David Herrmann <dh.herrmann@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c
index 64fc5ca..abe9793 100644
--- a/drivers/gpu/drm/drm_fb_helper.c
+++ b/drivers/gpu/drm/drm_fb_helper.c
@@ -352,6 +352,8 @@ retry:
 	drm_for_each_plane(plane, dev) {
 		struct drm_plane_state *plane_state;
 
+		plane->old_fb = plane->fb;
+
 		plane_state = drm_atomic_get_plane_state(state, plane);
 		if (IS_ERR(plane_state)) {
 			ret = PTR_ERR(plane_state);
@@ -382,16 +384,27 @@ retry:
 	}
 
 	ret = drm_atomic_commit(state);
-	if (ret != 0)
-		goto fail;
-
-	return 0;
 
 fail:
+	drm_for_each_plane(plane, dev) {
+		if (ret == 0) {
+			struct drm_framebuffer *new_fb = plane->state->fb;
+			if (new_fb)
+				drm_framebuffer_reference(new_fb);
+			plane->fb = new_fb;
+			plane->crtc = plane->state->crtc;
+
+			if (plane->old_fb)
+				drm_framebuffer_unreference(plane->old_fb);
+		}
+		plane->old_fb = NULL;
+	}
+
 	if (ret == -EDEADLK)
 		goto backoff;
 
-	drm_atomic_state_free(state);
+	if (ret != 0)
+		drm_atomic_state_free(state);
 
 	return ret;
 
-- 
cgit v0.10.2


From eba1f35dfe145247c7eb690c7c32740fde8ec699 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 14 Sep 2015 22:43:43 +0300
Subject: drm: Move timestamping constants into drm_vblank_crtc
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Collect the timestamping constants alongside the rest of the relevant
stuff under drm_vblank_crtc.

We can now get rid of the 'refcrtc' parameter to
drm_calc_vbltimestamp_from_scanoutpos().

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
index 2236793..ecfa703 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
@@ -681,7 +681,7 @@ int amdgpu_get_vblank_timestamp_kms(struct drm_device *dev, int crtc,
 	/* Helper routine in DRM core does all the work: */
 	return drm_calc_vbltimestamp_from_scanoutpos(dev, crtc, max_error,
 						     vblank_time, flags,
-						     drmcrtc, &drmcrtc->hwmode);
+						     &drmcrtc->hwmode);
 }
 
 const struct drm_ioctl_desc amdgpu_ioctls_kms[] = {
diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c
index 8df4133..6b2fefd 100644
--- a/drivers/gpu/drm/drm_irq.c
+++ b/drivers/gpu/drm/drm_irq.c
@@ -603,6 +603,7 @@ int drm_control(struct drm_device *dev, void *data,
 void drm_calc_timestamping_constants(struct drm_crtc *crtc,
 				     const struct drm_display_mode *mode)
 {
+	struct drm_vblank_crtc *vblank = &crtc->dev->vblank[drm_crtc_index(crtc)];
 	int linedur_ns = 0, pixeldur_ns = 0, framedur_ns = 0;
 	int dotclock = mode->crtc_clock;
 
@@ -628,9 +629,9 @@ void drm_calc_timestamping_constants(struct drm_crtc *crtc,
 		DRM_ERROR("crtc %u: Can't calculate constants, dotclock = 0!\n",
 			  crtc->base.id);
 
-	crtc->pixeldur_ns = pixeldur_ns;
-	crtc->linedur_ns  = linedur_ns;
-	crtc->framedur_ns = framedur_ns;
+	vblank->pixeldur_ns = pixeldur_ns;
+	vblank->linedur_ns  = linedur_ns;
+	vblank->framedur_ns = framedur_ns;
 
 	DRM_DEBUG("crtc %u: hwmode: htotal %d, vtotal %d, vdisplay %d\n",
 		  crtc->base.id, mode->crtc_htotal,
@@ -651,7 +652,6 @@ EXPORT_SYMBOL(drm_calc_timestamping_constants);
  * @flags: Flags to pass to driver:
  *         0 = Default,
  *         DRM_CALLED_FROM_VBLIRQ = If function is called from vbl IRQ handler
- * @refcrtc: CRTC which defines scanout timing
  * @mode: mode which defines the scanout timings
  *
  * Implements calculation of exact vblank timestamps from given drm_display_mode
@@ -692,9 +692,9 @@ int drm_calc_vbltimestamp_from_scanoutpos(struct drm_device *dev,
 					  int *max_error,
 					  struct timeval *vblank_time,
 					  unsigned flags,
-					  const struct drm_crtc *refcrtc,
 					  const struct drm_display_mode *mode)
 {
+	struct drm_vblank_crtc *vblank = &dev->vblank[pipe];
 	struct timeval tv_etime;
 	ktime_t stime, etime;
 	int vbl_status;
@@ -714,9 +714,9 @@ int drm_calc_vbltimestamp_from_scanoutpos(struct drm_device *dev,
 	}
 
 	/* Durations of frames, lines, pixels in nanoseconds. */
-	framedur_ns = refcrtc->framedur_ns;
-	linedur_ns  = refcrtc->linedur_ns;
-	pixeldur_ns = refcrtc->pixeldur_ns;
+	framedur_ns = vblank->framedur_ns;
+	linedur_ns  = vblank->linedur_ns;
+	pixeldur_ns = vblank->pixeldur_ns;
 
 	/* If mode timing undefined, just return as no-op:
 	 * Happens during initial modesetting of a crtc.
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 5a244ab..4cbc722 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -810,7 +810,6 @@ static int i915_get_vblank_timestamp(struct drm_device *dev, int pipe,
 	/* Helper routine in DRM core does all the work: */
 	return drm_calc_vbltimestamp_from_scanoutpos(dev, pipe, max_error,
 						     vblank_time, flags,
-						     crtc,
 						     &crtc->hwmode);
 }
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_display.c b/drivers/gpu/drm/nouveau/nouveau_display.c
index cc6c228..425515f 100644
--- a/drivers/gpu/drm/nouveau/nouveau_display.c
+++ b/drivers/gpu/drm/nouveau/nouveau_display.c
@@ -103,6 +103,7 @@ nouveau_display_scanoutpos_head(struct drm_crtc *crtc, int *vpos, int *hpos,
 		.base.head = nouveau_crtc(crtc)->index,
 	};
 	struct nouveau_display *disp = nouveau_display(crtc->dev);
+	struct drm_vblank_crtc *vblank = &crtc->dev->vblank[drm_crtc_index(crtc)];
 	int ret, retry = 1;
 
 	do {
@@ -116,7 +117,7 @@ nouveau_display_scanoutpos_head(struct drm_crtc *crtc, int *vpos, int *hpos,
 			break;
 		}
 
-		if (retry) ndelay(crtc->linedur_ns);
+		if (retry) ndelay(vblank->linedur_ns);
 	} while (retry--);
 
 	*hpos = args.scan.hline;
@@ -155,7 +156,7 @@ nouveau_display_vblstamp(struct drm_device *dev, int head, int *max_error,
 	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
 		if (nouveau_crtc(crtc)->index == head) {
 			return drm_calc_vbltimestamp_from_scanoutpos(dev,
-					head, max_error, time, flags, crtc,
+					head, max_error, time, flags,
 					&crtc->hwmode);
 		}
 	}
diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c
index 4a119c2..fd9da28 100644
--- a/drivers/gpu/drm/radeon/radeon_kms.c
+++ b/drivers/gpu/drm/radeon/radeon_kms.c
@@ -841,7 +841,7 @@ int radeon_get_vblank_timestamp_kms(struct drm_device *dev, int crtc,
 	/* Helper routine in DRM core does all the work: */
 	return drm_calc_vbltimestamp_from_scanoutpos(dev, crtc, max_error,
 						     vblank_time, flags,
-						     drmcrtc, &drmcrtc->hwmode);
+						     &drmcrtc->hwmode);
 }
 
 #define KMS_INVALID_IOCTL(name)						\
diff --git a/include/drm/drmP.h b/include/drm/drmP.h
index 8b5ce7c..2998867 100644
--- a/include/drm/drmP.h
+++ b/include/drm/drmP.h
@@ -701,6 +701,9 @@ struct drm_vblank_crtc {
 	u32 last_wait;			/* Last vblank seqno waited per CRTC */
 	unsigned int inmodeset;		/* Display driver is setting mode */
 	unsigned int pipe;		/* crtc index */
+	int framedur_ns;		/* frame/field duration in ns */
+	int linedur_ns;			/* line duration in ns */
+	int pixeldur_ns;		/* pixel duration in ns */
 	bool enabled;			/* so we don't call enable more than
 					   once per disable */
 };
@@ -951,7 +954,6 @@ extern int drm_calc_vbltimestamp_from_scanoutpos(struct drm_device *dev,
 						 unsigned int pipe, int *max_error,
 						 struct timeval *vblank_time,
 						 unsigned flags,
-						 const struct drm_crtc *refcrtc,
 						 const struct drm_display_mode *mode);
 extern void drm_calc_timestamping_constants(struct drm_crtc *crtc,
 					    const struct drm_display_mode *mode);
diff --git a/include/drm/drm_crtc.h b/include/drm/drm_crtc.h
index 6566f72..683f142 100644
--- a/include/drm/drm_crtc.h
+++ b/include/drm/drm_crtc.h
@@ -413,9 +413,6 @@ struct drm_crtc_funcs {
  * @funcs: CRTC control functions
  * @gamma_size: size of gamma ramp
  * @gamma_store: gamma ramp values
- * @framedur_ns: precise frame timing
- * @linedur_ns: precise line timing
- * @pixeldur_ns: precise pixel timing
  * @helper_private: mid-layer private data
  * @properties: property tracking for this CRTC
  * @state: current atomic state for this CRTC
@@ -468,9 +465,6 @@ struct drm_crtc {
 	uint32_t gamma_size;
 	uint16_t *gamma_store;
 
-	/* Constants needed for precise vblank and swap timestamping. */
-	int framedur_ns, linedur_ns, pixeldur_ns;
-
 	/* if you are using the helper */
 	const void *helper_private;
 
-- 
cgit v0.10.2


From 3bb403bf421b5b00366a9041a7edc0a1f6494f5e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 14 Sep 2015 22:43:44 +0300
Subject: drm: Stop using linedur_ns and pixeldur_ns for vblank timestamps
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

linedur_ns, and especially pixeldur_ns are becoming rather inaccurate
to be used for the vblank timestamp correction. With 4k@60 the pixel
duration is already below 2ns, so the amount of error due to the
truncation to nanoseconds is introducing quite a bit of error.

We can avoid such problems if we instead calculate the timestamp
delta_ns directly from the dislay timings, avoiding the use of
these intermediate truncated values.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
[danvet: Squash in fixup from Thierry Reding for amdgpu.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
index e3d7077..9b34a34 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
@@ -745,7 +745,8 @@ bool amdgpu_crtc_scaling_mode_fixup(struct drm_crtc *crtc,
  *
  */
 int amdgpu_get_crtc_scanoutpos(struct drm_device *dev, int crtc, unsigned int flags,
-			       int *vpos, int *hpos, ktime_t *stime, ktime_t *etime)
+			       int *vpos, int *hpos, ktime_t *stime, ktime_t *etime,
+			       const struct drm_display_mode *mode)
 {
 	u32 vbl = 0, position = 0;
 	int vbl_start, vbl_end, vtotal, ret = 0;
@@ -781,7 +782,7 @@ int amdgpu_get_crtc_scanoutpos(struct drm_device *dev, int crtc, unsigned int fl
 	}
 	else {
 		/* No: Fake something reasonable which gives at least ok results. */
-		vbl_start = adev->mode_info.crtcs[crtc]->base.hwmode.crtc_vdisplay;
+		vbl_start = mode->crtc_vdisplay;
 		vbl_end = 0;
 	}
 
@@ -797,7 +798,7 @@ int amdgpu_get_crtc_scanoutpos(struct drm_device *dev, int crtc, unsigned int fl
 
 	/* Inside "upper part" of vblank area? Apply corrective offset if so: */
 	if (in_vbl && (*vpos >= vbl_start)) {
-		vtotal = adev->mode_info.crtcs[crtc]->base.hwmode.crtc_vtotal;
+		vtotal = mode->crtc_vtotal;
 		*vpos = *vpos - vtotal;
 	}
 
@@ -819,8 +820,8 @@ int amdgpu_get_crtc_scanoutpos(struct drm_device *dev, int crtc, unsigned int fl
 	 * We only do this if DRM_CALLED_FROM_VBLIRQ.
 	 */
 	if ((flags & DRM_CALLED_FROM_VBLIRQ) && !in_vbl) {
-		vbl_start = adev->mode_info.crtcs[crtc]->base.hwmode.crtc_vdisplay;
-		vtotal = adev->mode_info.crtcs[crtc]->base.hwmode.crtc_vtotal;
+		vbl_start = mode->crtc_vdisplay;
+		vtotal = mode->crtc_vtotal;
 
 		if (vbl_start - *vpos < vtotal / 100) {
 			*vpos -= vtotal;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h
index 64efe5b..2b03425 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h
@@ -543,7 +543,8 @@ void amdgpu_encoder_set_active_device(struct drm_encoder *encoder);
 int amdgpu_get_crtc_scanoutpos(struct drm_device *dev, int crtc,
 				      unsigned int flags,
 				      int *vpos, int *hpos, ktime_t *stime,
-				      ktime_t *etime);
+				      ktime_t *etime,
+				      const struct drm_display_mode *mode);
 
 int amdgpu_framebuffer_init(struct drm_device *dev,
 			     struct amdgpu_framebuffer *rfb,
diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c
index 6b2fefd..9fab333 100644
--- a/drivers/gpu/drm/drm_irq.c
+++ b/drivers/gpu/drm/drm_irq.c
@@ -694,12 +694,11 @@ int drm_calc_vbltimestamp_from_scanoutpos(struct drm_device *dev,
 					  unsigned flags,
 					  const struct drm_display_mode *mode)
 {
-	struct drm_vblank_crtc *vblank = &dev->vblank[pipe];
 	struct timeval tv_etime;
 	ktime_t stime, etime;
 	int vbl_status;
 	int vpos, hpos, i;
-	int framedur_ns, linedur_ns, pixeldur_ns, delta_ns, duration_ns;
+	int delta_ns, duration_ns;
 	bool invbl;
 
 	if (pipe >= dev->num_crtcs) {
@@ -713,15 +712,10 @@ int drm_calc_vbltimestamp_from_scanoutpos(struct drm_device *dev,
 		return -EIO;
 	}
 
-	/* Durations of frames, lines, pixels in nanoseconds. */
-	framedur_ns = vblank->framedur_ns;
-	linedur_ns  = vblank->linedur_ns;
-	pixeldur_ns = vblank->pixeldur_ns;
-
 	/* If mode timing undefined, just return as no-op:
 	 * Happens during initial modesetting of a crtc.
 	 */
-	if (framedur_ns == 0) {
+	if (mode->crtc_clock == 0) {
 		DRM_DEBUG("crtc %u: Noop due to uninitialized mode.\n", pipe);
 		return -EAGAIN;
 	}
@@ -738,8 +732,10 @@ int drm_calc_vbltimestamp_from_scanoutpos(struct drm_device *dev,
 		 * Get vertical and horizontal scanout position vpos, hpos,
 		 * and bounding timestamps stime, etime, pre/post query.
 		 */
-		vbl_status = dev->driver->get_scanout_position(dev, pipe, flags, &vpos,
-							       &hpos, &stime, &etime);
+		vbl_status = dev->driver->get_scanout_position(dev, pipe, flags,
+							       &vpos, &hpos,
+							       &stime, &etime,
+							       mode);
 
 		/* Return as no-op if scanout query unsupported or failed. */
 		if (!(vbl_status & DRM_SCANOUTPOS_VALID)) {
@@ -776,7 +772,8 @@ int drm_calc_vbltimestamp_from_scanoutpos(struct drm_device *dev,
 	 * since start of scanout at first display scanline. delta_ns
 	 * can be negative if start of scanout hasn't happened yet.
 	 */
-	delta_ns = vpos * linedur_ns + hpos * pixeldur_ns;
+	delta_ns = div_s64(1000000LL * (vpos * mode->crtc_htotal + hpos),
+			   mode->crtc_clock);
 
 	if (!drm_timestamp_monotonic)
 		etime = ktime_mono_to_real(etime);
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 4cbc722..ce1e0f5 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -648,12 +648,12 @@ static int __intel_get_crtc_scanline(struct intel_crtc *crtc)
 
 static int i915_get_crtc_scanoutpos(struct drm_device *dev, int pipe,
 				    unsigned int flags, int *vpos, int *hpos,
-				    ktime_t *stime, ktime_t *etime)
+				    ktime_t *stime, ktime_t *etime,
+				    const struct drm_display_mode *mode)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_crtc *crtc = dev_priv->pipe_to_crtc_mapping[pipe];
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	const struct drm_display_mode *mode = &intel_crtc->base.hwmode;
 	int position;
 	int vbl_start, vbl_end, hsync_start, htotal, vtotal;
 	bool in_vbl = true;
diff --git a/drivers/gpu/drm/nouveau/nouveau_display.c b/drivers/gpu/drm/nouveau/nouveau_display.c
index 425515f..a82c3cb 100644
--- a/drivers/gpu/drm/nouveau/nouveau_display.c
+++ b/drivers/gpu/drm/nouveau/nouveau_display.c
@@ -133,7 +133,8 @@ nouveau_display_scanoutpos_head(struct drm_crtc *crtc, int *vpos, int *hpos,
 
 int
 nouveau_display_scanoutpos(struct drm_device *dev, int head, unsigned int flags,
-			   int *vpos, int *hpos, ktime_t *stime, ktime_t *etime)
+			   int *vpos, int *hpos, ktime_t *stime, ktime_t *etime,
+			   const struct drm_display_mode *mode)
 {
 	struct drm_crtc *crtc;
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_display.h b/drivers/gpu/drm/nouveau/nouveau_display.h
index a6213e2..4182d21 100644
--- a/drivers/gpu/drm/nouveau/nouveau_display.h
+++ b/drivers/gpu/drm/nouveau/nouveau_display.h
@@ -68,7 +68,8 @@ void nouveau_display_resume(struct drm_device *dev, bool runtime);
 int  nouveau_display_vblank_enable(struct drm_device *, int);
 void nouveau_display_vblank_disable(struct drm_device *, int);
 int  nouveau_display_scanoutpos(struct drm_device *, int, unsigned int,
-				int *, int *, ktime_t *, ktime_t *);
+				int *, int *, ktime_t *, ktime_t *,
+				const struct drm_display_mode *);
 int  nouveau_display_vblstamp(struct drm_device *, int, int *,
 			      struct timeval *, unsigned);
 
diff --git a/drivers/gpu/drm/radeon/radeon_display.c b/drivers/gpu/drm/radeon/radeon_display.c
index d2e9e9e..0503af7 100644
--- a/drivers/gpu/drm/radeon/radeon_display.c
+++ b/drivers/gpu/drm/radeon/radeon_display.c
@@ -323,7 +323,8 @@ void radeon_crtc_handle_vblank(struct radeon_device *rdev, int crtc_id)
 	 */
 	if (update_pending &&
 	    (DRM_SCANOUTPOS_VALID & radeon_get_crtc_scanoutpos(rdev->ddev, crtc_id, 0,
-							       &vpos, &hpos, NULL, NULL)) &&
+							       &vpos, &hpos, NULL, NULL,
+							       &rdev->mode_info.crtcs[crtc_id]->base.hwmode)) &&
 	    ((vpos >= (99 * rdev->mode_info.crtcs[crtc_id]->base.hwmode.crtc_vdisplay)/100) ||
 	     (vpos < 0 && !ASIC_IS_AVIVO(rdev)))) {
 		/* crtc didn't flip in this target vblank interval,
@@ -1799,7 +1800,8 @@ bool radeon_crtc_scaling_mode_fixup(struct drm_crtc *crtc,
  *
  */
 int radeon_get_crtc_scanoutpos(struct drm_device *dev, int crtc, unsigned int flags,
-			       int *vpos, int *hpos, ktime_t *stime, ktime_t *etime)
+			       int *vpos, int *hpos, ktime_t *stime, ktime_t *etime,
+			       const struct drm_display_mode *mode)
 {
 	u32 stat_crtc = 0, vbl = 0, position = 0;
 	int vbl_start, vbl_end, vtotal, ret = 0;
@@ -1914,7 +1916,7 @@ int radeon_get_crtc_scanoutpos(struct drm_device *dev, int crtc, unsigned int fl
 	}
 	else {
 		/* No: Fake something reasonable which gives at least ok results. */
-		vbl_start = rdev->mode_info.crtcs[crtc]->base.hwmode.crtc_vdisplay;
+		vbl_start = mode->crtc_vdisplay;
 		vbl_end = 0;
 	}
 
@@ -1930,7 +1932,7 @@ int radeon_get_crtc_scanoutpos(struct drm_device *dev, int crtc, unsigned int fl
 
 	/* Inside "upper part" of vblank area? Apply corrective offset if so: */
 	if (in_vbl && (*vpos >= vbl_start)) {
-		vtotal = rdev->mode_info.crtcs[crtc]->base.hwmode.crtc_vtotal;
+		vtotal = mode->crtc_vtotal;
 		*vpos = *vpos - vtotal;
 	}
 
@@ -1952,8 +1954,8 @@ int radeon_get_crtc_scanoutpos(struct drm_device *dev, int crtc, unsigned int fl
 	 * We only do this if DRM_CALLED_FROM_VBLIRQ.
 	 */
 	if ((flags & DRM_CALLED_FROM_VBLIRQ) && !in_vbl) {
-		vbl_start = rdev->mode_info.crtcs[crtc]->base.hwmode.crtc_vdisplay;
-		vtotal = rdev->mode_info.crtcs[crtc]->base.hwmode.crtc_vtotal;
+		vbl_start = mode->crtc_vdisplay;
+		vtotal = mode->crtc_vtotal;
 
 		if (vbl_start - *vpos < vtotal / 100) {
 			*vpos -= vtotal;
diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c
index 5751446..e30c1d7 100644
--- a/drivers/gpu/drm/radeon/radeon_drv.c
+++ b/drivers/gpu/drm/radeon/radeon_drv.c
@@ -126,8 +126,9 @@ struct dma_buf *radeon_gem_prime_export(struct drm_device *dev,
 					int flags);
 extern int radeon_get_crtc_scanoutpos(struct drm_device *dev, int crtc,
 				      unsigned int flags,
-				      int *vpos, int *hpos, ktime_t *stime,
-				      ktime_t *etime);
+				      int *vpos, int *hpos,
+				      ktime_t *stime, ktime_t *etime,
+				      const struct drm_display_mode *mode);
 extern bool radeon_is_px(struct drm_device *dev);
 extern const struct drm_ioctl_desc radeon_ioctls_kms[];
 extern int radeon_max_kms_ioctl;
diff --git a/drivers/gpu/drm/radeon/radeon_mode.h b/drivers/gpu/drm/radeon/radeon_mode.h
index aecc3e3..2317d04 100644
--- a/drivers/gpu/drm/radeon/radeon_mode.h
+++ b/drivers/gpu/drm/radeon/radeon_mode.h
@@ -876,8 +876,9 @@ extern void radeon_cursor_reset(struct drm_crtc *crtc);
 
 extern int radeon_get_crtc_scanoutpos(struct drm_device *dev, int crtc,
 				      unsigned int flags,
-				      int *vpos, int *hpos, ktime_t *stime,
-				      ktime_t *etime);
+				      int *vpos, int *hpos,
+				      ktime_t *stime, ktime_t *etime,
+				      const struct drm_display_mode *mode);
 
 extern bool radeon_combios_check_hardcoded_edid(struct radeon_device *rdev);
 extern struct edid *
diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c
index 05751f3..10f4c12 100644
--- a/drivers/gpu/drm/radeon/radeon_pm.c
+++ b/drivers/gpu/drm/radeon/radeon_pm.c
@@ -1733,7 +1733,9 @@ static bool radeon_pm_in_vbl(struct radeon_device *rdev)
 	 */
 	for (crtc = 0; (crtc < rdev->num_crtc) && in_vbl; crtc++) {
 		if (rdev->pm.active_crtcs & (1 << crtc)) {
-			vbl_status = radeon_get_crtc_scanoutpos(rdev->ddev, crtc, 0, &vpos, &hpos, NULL, NULL);
+			vbl_status = radeon_get_crtc_scanoutpos(rdev->ddev, crtc, 0,
+								&vpos, &hpos, NULL, NULL,
+								&rdev->mode_info.crtcs[crtc]->base.hwmode);
 			if ((vbl_status & DRM_SCANOUTPOS_VALID) &&
 			    !(vbl_status & DRM_SCANOUTPOS_IN_VBLANK))
 				in_vbl = false;
diff --git a/include/drm/drmP.h b/include/drm/drmP.h
index 2998867..b2a95e7 100644
--- a/include/drm/drmP.h
+++ b/include/drm/drmP.h
@@ -482,6 +482,7 @@ struct drm_driver {
 	 *               scanout position query. Can be NULL to skip timestamp.
 	 * \param *etime Target location for timestamp taken immediately after
 	 *               scanout position query. Can be NULL to skip timestamp.
+	 * \param mode Current display timings.
 	 *
 	 * Returns vpos as a positive number while in active scanout area.
 	 * Returns vpos as a negative number inside vblank, counting the number
@@ -499,8 +500,9 @@ struct drm_driver {
 	 */
 	int (*get_scanout_position) (struct drm_device *dev, int crtc,
 				     unsigned int flags,
-				     int *vpos, int *hpos, ktime_t *stime,
-				     ktime_t *etime);
+				     int *vpos, int *hpos,
+				     ktime_t *stime, ktime_t *etime,
+				     const struct drm_display_mode *mode);
 
 	/**
 	 * Called by \c drm_get_last_vbltimestamp. Should return a precise
-- 
cgit v0.10.2


From 20b2020334110f9afb8316ba158b9549f2f07ff9 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 14 Sep 2015 22:43:45 +0300
Subject: drm: Kill pixeldur_ns
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

pixeldur_ns is now unsued, so kill it from drm_vblank_crtc. framedur_ns
is also currently unused but we will have use for it in the near future
so leave it be. linedur_ns is still used by nouveau for some internal
delays.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c
index 9fab333..ac17602 100644
--- a/drivers/gpu/drm/drm_irq.c
+++ b/drivers/gpu/drm/drm_irq.c
@@ -604,7 +604,7 @@ void drm_calc_timestamping_constants(struct drm_crtc *crtc,
 				     const struct drm_display_mode *mode)
 {
 	struct drm_vblank_crtc *vblank = &crtc->dev->vblank[drm_crtc_index(crtc)];
-	int linedur_ns = 0, pixeldur_ns = 0, framedur_ns = 0;
+	int linedur_ns = 0, framedur_ns = 0;
 	int dotclock = mode->crtc_clock;
 
 	/* Valid dotclock? */
@@ -613,10 +613,9 @@ void drm_calc_timestamping_constants(struct drm_crtc *crtc,
 
 		/*
 		 * Convert scanline length in pixels and video
-		 * dot clock to line duration, frame duration
-		 * and pixel duration in nanoseconds:
+		 * dot clock to line duration and frame duration
+		 * in nanoseconds:
 		 */
-		pixeldur_ns = 1000000 / dotclock;
 		linedur_ns  = div_u64((u64) mode->crtc_htotal * 1000000, dotclock);
 		framedur_ns = div_u64((u64) frame_size * 1000000, dotclock);
 
@@ -629,16 +628,14 @@ void drm_calc_timestamping_constants(struct drm_crtc *crtc,
 		DRM_ERROR("crtc %u: Can't calculate constants, dotclock = 0!\n",
 			  crtc->base.id);
 
-	vblank->pixeldur_ns = pixeldur_ns;
 	vblank->linedur_ns  = linedur_ns;
 	vblank->framedur_ns = framedur_ns;
 
 	DRM_DEBUG("crtc %u: hwmode: htotal %d, vtotal %d, vdisplay %d\n",
 		  crtc->base.id, mode->crtc_htotal,
 		  mode->crtc_vtotal, mode->crtc_vdisplay);
-	DRM_DEBUG("crtc %u: clock %d kHz framedur %d linedur %d, pixeldur %d\n",
-		  crtc->base.id, dotclock, framedur_ns,
-		  linedur_ns, pixeldur_ns);
+	DRM_DEBUG("crtc %u: clock %d kHz framedur %d linedur %d\n",
+		  crtc->base.id, dotclock, framedur_ns, linedur_ns);
 }
 EXPORT_SYMBOL(drm_calc_timestamping_constants);
 
diff --git a/include/drm/drmP.h b/include/drm/drmP.h
index b2a95e7..6717a7d 100644
--- a/include/drm/drmP.h
+++ b/include/drm/drmP.h
@@ -705,7 +705,6 @@ struct drm_vblank_crtc {
 	unsigned int pipe;		/* crtc index */
 	int framedur_ns;		/* frame/field duration in ns */
 	int linedur_ns;			/* line duration in ns */
-	int pixeldur_ns;		/* pixel duration in ns */
 	bool enabled;			/* so we don't call enable more than
 					   once per disable */
 };
-- 
cgit v0.10.2


From 66f59c5ccf50eec42bbe4f5029ad9434f73eaf75 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 14 Sep 2015 22:43:46 +0300
Subject: drm/i915: Fix vblank count variable types
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The vblank counts are u32 so make flip_queued_vblank and
flip_ready_vblank u32 as well.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index bfd1204..354432f 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -834,8 +834,8 @@ struct intel_unpin_work {
 	u32 flip_count;
 	u32 gtt_offset;
 	struct drm_i915_gem_request *flip_queued_req;
-	int flip_queued_vblank;
-	int flip_ready_vblank;
+	u32 flip_queued_vblank;
+	u32 flip_ready_vblank;
 	bool enable_stall_check;
 };
 
-- 
cgit v0.10.2


From a6e610dcce3c8cbd5217527a089c8811d4412184 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 14 Sep 2015 22:43:47 +0300
Subject: drm: Pass flags to drm_update_vblank_count()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We'll soon have use for the 'flags' in drm_update_vblank_count() so pass
it in.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c
index ac17602..0353224 100644
--- a/drivers/gpu/drm/drm_irq.c
+++ b/drivers/gpu/drm/drm_irq.c
@@ -120,7 +120,8 @@ static void store_vblank(struct drm_device *dev, unsigned int pipe,
  * Note: caller must hold dev->vbl_lock since this reads & writes
  * device vblank fields.
  */
-static void drm_update_vblank_count(struct drm_device *dev, unsigned int pipe)
+static void drm_update_vblank_count(struct drm_device *dev, unsigned int pipe,
+				    unsigned long flags)
 {
 	struct drm_vblank_crtc *vblank = &dev->vblank[pipe];
 	u32 cur_vblank, diff;
@@ -141,7 +142,7 @@ static void drm_update_vblank_count(struct drm_device *dev, unsigned int pipe)
 	 */
 	do {
 		cur_vblank = dev->driver->get_vblank_counter(dev, pipe);
-		rc = drm_get_last_vbltimestamp(dev, pipe, &t_vblank, 0);
+		rc = drm_get_last_vbltimestamp(dev, pipe, &t_vblank, flags);
 	} while (cur_vblank != dev->driver->get_vblank_counter(dev, pipe));
 
 	/* Deal with counter wrap */
@@ -207,7 +208,7 @@ static void vblank_disable_and_save(struct drm_device *dev, unsigned int pipe)
 	 */
 	if (!vblank->enabled &&
 	    drm_get_last_vbltimestamp(dev, pipe, &tvblank, 0)) {
-		drm_update_vblank_count(dev, pipe);
+		drm_update_vblank_count(dev, pipe, 0);
 		spin_unlock_irqrestore(&dev->vblank_time_lock, irqflags);
 		return;
 	}
@@ -1027,7 +1028,7 @@ static int drm_vblank_enable(struct drm_device *dev, unsigned int pipe)
 			atomic_dec(&vblank->refcount);
 		else {
 			vblank->enabled = true;
-			drm_update_vblank_count(dev, pipe);
+			drm_update_vblank_count(dev, pipe, 0);
 		}
 	}
 
-- 
cgit v0.10.2


From facfb062e8c958f321c366c6ea3d98db899a4c6e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 14 Sep 2015 22:43:48 +0300
Subject: drm: Limit the number of .get_vblank_counter() retries
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Pontential infinite loops in the vblank code are a bad idea. Add some
limits.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c
index 0353224..93fe582 100644
--- a/drivers/gpu/drm/drm_irq.c
+++ b/drivers/gpu/drm/drm_irq.c
@@ -127,6 +127,7 @@ static void drm_update_vblank_count(struct drm_device *dev, unsigned int pipe,
 	u32 cur_vblank, diff;
 	bool rc;
 	struct timeval t_vblank;
+	int count = DRM_TIMESTAMP_MAXRETRIES;
 
 	/*
 	 * Interrupts were disabled prior to this call, so deal with counter
@@ -143,7 +144,7 @@ static void drm_update_vblank_count(struct drm_device *dev, unsigned int pipe,
 	do {
 		cur_vblank = dev->driver->get_vblank_counter(dev, pipe);
 		rc = drm_get_last_vbltimestamp(dev, pipe, &t_vblank, flags);
-	} while (cur_vblank != dev->driver->get_vblank_counter(dev, pipe));
+	} while (cur_vblank != dev->driver->get_vblank_counter(dev, pipe) && --count > 0);
 
 	/* Deal with counter wrap */
 	diff = cur_vblank - vblank->last;
@@ -914,6 +915,7 @@ u32 drm_vblank_count_and_time(struct drm_device *dev, unsigned int pipe,
 			      struct timeval *vblanktime)
 {
 	struct drm_vblank_crtc *vblank = &dev->vblank[pipe];
+	int count = DRM_TIMESTAMP_MAXRETRIES;
 	u32 cur_vblank;
 
 	if (WARN_ON(pipe >= dev->num_crtcs))
@@ -929,7 +931,7 @@ u32 drm_vblank_count_and_time(struct drm_device *dev, unsigned int pipe,
 		smp_rmb();
 		*vblanktime = vblanktimestamp(dev, pipe, cur_vblank);
 		smp_rmb();
-	} while (cur_vblank != vblank->count);
+	} while (cur_vblank != vblank->count && --count > 0);
 
 	return cur_vblank;
 }
-- 
cgit v0.10.2


From ad1716ecca9d9099199cfc9278672c17d1484c75 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 14 Sep 2015 22:43:49 +0300
Subject: drm: Clean up drm_calc_vbltimestamp_from_scanoutpos() vbl_status
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Avoid confusion and don't use 'vbl_status' as both the
.get_scanout_position() return value and the return value from
drm_calc_vbltimestamp_from_scanoutpos().

While at it make 'vbl_status' unsigned and print it as hex in the
debug prints since it's a bitmask.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c
index 93fe582..aad4f1d 100644
--- a/drivers/gpu/drm/drm_irq.c
+++ b/drivers/gpu/drm/drm_irq.c
@@ -695,10 +695,10 @@ int drm_calc_vbltimestamp_from_scanoutpos(struct drm_device *dev,
 {
 	struct timeval tv_etime;
 	ktime_t stime, etime;
-	int vbl_status;
+	unsigned int vbl_status;
+	int ret = DRM_VBLANKTIME_SCANOUTPOS_METHOD;
 	int vpos, hpos, i;
 	int delta_ns, duration_ns;
-	bool invbl;
 
 	if (pipe >= dev->num_crtcs) {
 		DRM_ERROR("Invalid crtc %u\n", pipe);
@@ -738,7 +738,7 @@ int drm_calc_vbltimestamp_from_scanoutpos(struct drm_device *dev,
 
 		/* Return as no-op if scanout query unsupported or failed. */
 		if (!(vbl_status & DRM_SCANOUTPOS_VALID)) {
-			DRM_DEBUG("crtc %u : scanoutpos query failed [%d].\n",
+			DRM_DEBUG("crtc %u : scanoutpos query failed [0x%x].\n",
 				  pipe, vbl_status);
 			return -EIO;
 		}
@@ -765,7 +765,8 @@ int drm_calc_vbltimestamp_from_scanoutpos(struct drm_device *dev,
 	 * within vblank area, counting down the number of lines until
 	 * start of scanout.
 	 */
-	invbl = vbl_status & DRM_SCANOUTPOS_IN_VBLANK;
+	if (vbl_status & DRM_SCANOUTPOS_IN_VBLANK)
+		ret |= DRM_VBLANKTIME_IN_VBLANK;
 
 	/* Convert scanout position into elapsed time at raw_time query
 	 * since start of scanout at first display scanline. delta_ns
@@ -788,17 +789,13 @@ int drm_calc_vbltimestamp_from_scanoutpos(struct drm_device *dev,
 		etime = ktime_sub_ns(etime, delta_ns);
 	*vblank_time = ktime_to_timeval(etime);
 
-	DRM_DEBUG("crtc %u : v %d p(%d,%d)@ %ld.%ld -> %ld.%ld [e %d us, %d rep]\n",
-		  pipe, (int)vbl_status, hpos, vpos,
+	DRM_DEBUG("crtc %u : v 0x%x p(%d,%d)@ %ld.%ld -> %ld.%ld [e %d us, %d rep]\n",
+		  pipe, vbl_status, hpos, vpos,
 		  (long)tv_etime.tv_sec, (long)tv_etime.tv_usec,
 		  (long)vblank_time->tv_sec, (long)vblank_time->tv_usec,
 		  duration_ns/1000, i);
 
-	vbl_status = DRM_VBLANKTIME_SCANOUTPOS_METHOD;
-	if (invbl)
-		vbl_status |= DRM_VBLANKTIME_IN_VBLANK;
-
-	return vbl_status;
+	return ret;
 }
 EXPORT_SYMBOL(drm_calc_vbltimestamp_from_scanoutpos);
 
-- 
cgit v0.10.2


From 1b2eb71050915d88e62a0ca0c448e0e93484ae9e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 14 Sep 2015 22:43:50 +0300
Subject: drm: store_vblank() is never called with NULL timestamp
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Remove the NULL 't_vblank' checks from store_vblank() since that will
never happen.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c
index aad4f1d..07b0cb1 100644
--- a/drivers/gpu/drm/drm_irq.c
+++ b/drivers/gpu/drm/drm_irq.c
@@ -83,13 +83,11 @@ static void store_vblank(struct drm_device *dev, unsigned int pipe,
 
 	assert_spin_locked(&dev->vblank_time_lock);
 
-	if (t_vblank) {
-		/* All writers hold the spinlock, but readers are serialized by
-		 * the latching of vblank->count below.
-		 */
-		tslot = vblank->count + vblank_count_inc;
-		vblanktimestamp(dev, pipe, tslot) = *t_vblank;
-	}
+	/* All writers hold the spinlock, but readers are serialized by
+	 * the latching of vblank->count below.
+	 */
+	tslot = vblank->count + vblank_count_inc;
+	vblanktimestamp(dev, pipe, tslot) = *t_vblank;
 
 	/*
 	 * vblank timestamp updates are protected on the write side with
-- 
cgit v0.10.2


From 4dfd64862ff852df7b1198d667dda778715ee88f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 14 Sep 2015 22:43:51 +0300
Subject: drm: Use vblank timestamps to guesstimate how many vblanks were
 missed
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

When lacking am accurate hardware frame counter, we can fall back to
using the vblank timestamps to guesstimagte how many vblanks have
elapsed since the last time the vblank counter was updated.

Take the oppostunity to unify the vblank_disable_and_save() and
drm_handle_vblank_events() to call the same function
(drm_update_vblank_count()) to perform the vblank updates.

If the hardware/driver has an accurate frame counter use it instead of
the timestamp based guesstimate. If the hardware/driver has neither
a frame counter nor acurate vblank timestamps, we fall back to assuming
that each drm_handle_vblank_events() should increment the vblank count
by one.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c
index 07b0cb1..88fbee4 100644
--- a/drivers/gpu/drm/drm_irq.c
+++ b/drivers/gpu/drm/drm_irq.c
@@ -76,13 +76,15 @@ module_param_named(timestamp_monotonic, drm_timestamp_monotonic, int, 0600);
 
 static void store_vblank(struct drm_device *dev, unsigned int pipe,
 			 u32 vblank_count_inc,
-			 struct timeval *t_vblank)
+			 struct timeval *t_vblank, u32 last)
 {
 	struct drm_vblank_crtc *vblank = &dev->vblank[pipe];
 	u32 tslot;
 
 	assert_spin_locked(&dev->vblank_time_lock);
 
+	vblank->last = last;
+
 	/* All writers hold the spinlock, but readers are serialized by
 	 * the latching of vblank->count below.
 	 */
@@ -103,6 +105,54 @@ static void store_vblank(struct drm_device *dev, unsigned int pipe,
 }
 
 /**
+ * drm_reset_vblank_timestamp - reset the last timestamp to the last vblank
+ * @dev: DRM device
+ * @pipe: index of CRTC for which to reset the timestamp
+ *
+ * Reset the stored timestamp for the current vblank count to correspond
+ * to the last vblank occurred.
+ *
+ * Only to be called from drm_vblank_on().
+ *
+ * Note: caller must hold dev->vbl_lock since this reads & writes
+ * device vblank fields.
+ */
+static void drm_reset_vblank_timestamp(struct drm_device *dev, unsigned int pipe)
+{
+	u32 cur_vblank;
+	bool rc;
+	struct timeval t_vblank;
+	int count = DRM_TIMESTAMP_MAXRETRIES;
+
+	spin_lock(&dev->vblank_time_lock);
+
+	/*
+	 * sample the current counter to avoid random jumps
+	 * when drm_vblank_enable() applies the diff
+	 */
+	do {
+		cur_vblank = dev->driver->get_vblank_counter(dev, pipe);
+		rc = drm_get_last_vbltimestamp(dev, pipe, &t_vblank, 0);
+	} while (cur_vblank != dev->driver->get_vblank_counter(dev, pipe) && --count > 0);
+
+	/*
+	 * Only reinitialize corresponding vblank timestamp if high-precision query
+	 * available and didn't fail. Otherwise reinitialize delayed at next vblank
+	 * interrupt and assign 0 for now, to mark the vblanktimestamp as invalid.
+	 */
+	if (!rc)
+		t_vblank = (struct timeval) {0, 0};
+
+	/*
+	 * +1 to make sure user will never see the same
+	 * vblank counter value before and after a modeset
+	 */
+	store_vblank(dev, pipe, 1, &t_vblank, cur_vblank);
+
+	spin_unlock(&dev->vblank_time_lock);
+}
+
+/**
  * drm_update_vblank_count - update the master vblank counter
  * @dev: DRM device
  * @pipe: counter to update
@@ -126,6 +176,7 @@ static void drm_update_vblank_count(struct drm_device *dev, unsigned int pipe,
 	bool rc;
 	struct timeval t_vblank;
 	int count = DRM_TIMESTAMP_MAXRETRIES;
+	int framedur_ns = vblank->framedur_ns;
 
 	/*
 	 * Interrupts were disabled prior to this call, so deal with counter
@@ -144,20 +195,40 @@ static void drm_update_vblank_count(struct drm_device *dev, unsigned int pipe,
 		rc = drm_get_last_vbltimestamp(dev, pipe, &t_vblank, flags);
 	} while (cur_vblank != dev->driver->get_vblank_counter(dev, pipe) && --count > 0);
 
-	/* Deal with counter wrap */
-	diff = cur_vblank - vblank->last;
-	if (cur_vblank < vblank->last) {
-		diff += dev->max_vblank_count + 1;
+	if (dev->max_vblank_count != 0) {
+		/* trust the hw counter when it's around */
+		diff = (cur_vblank - vblank->last) & dev->max_vblank_count;
+	} else if (rc && framedur_ns) {
+		const struct timeval *t_old;
+		u64 diff_ns;
+
+		t_old = &vblanktimestamp(dev, pipe, vblank->count);
+		diff_ns = timeval_to_ns(&t_vblank) - timeval_to_ns(t_old);
 
-		DRM_DEBUG("last_vblank[%u]=0x%x, cur_vblank=0x%x => diff=0x%x\n",
-			  pipe, vblank->last, cur_vblank, diff);
+		/*
+		 * Figure out how many vblanks we've missed based
+		 * on the difference in the timestamps and the
+		 * frame/field duration.
+		 */
+		diff = DIV_ROUND_CLOSEST_ULL(diff_ns, framedur_ns);
+
+		if (diff == 0 && flags & DRM_CALLED_FROM_VBLIRQ)
+			DRM_DEBUG("crtc %u: Redundant vblirq ignored."
+				  " diff_ns = %lld, framedur_ns = %d)\n",
+				  pipe, (long long) diff_ns, framedur_ns);
+	} else {
+		/* some kind of default for drivers w/o accurate vbl timestamping */
+		diff = (flags & DRM_CALLED_FROM_VBLIRQ) != 0;
 	}
 
-	DRM_DEBUG("updating vblank count on crtc %u, missed %d\n",
-		  pipe, diff);
+	DRM_DEBUG("updating vblank count on crtc %u:"
+		  " current=%u, diff=%u, hw=%u hw_last=%u\n",
+		  pipe, vblank->count, diff, cur_vblank, vblank->last);
 
-	if (diff == 0)
+	if (diff == 0) {
+		WARN_ON_ONCE(cur_vblank != vblank->last);
 		return;
+	}
 
 	/*
 	 * Only reinitialize corresponding vblank timestamp if high-precision query
@@ -167,7 +238,7 @@ static void drm_update_vblank_count(struct drm_device *dev, unsigned int pipe,
 	if (!rc)
 		t_vblank = (struct timeval) {0, 0};
 
-	store_vblank(dev, pipe, diff, &t_vblank);
+	store_vblank(dev, pipe, diff, &t_vblank, cur_vblank);
 }
 
 /*
@@ -180,11 +251,6 @@ static void vblank_disable_and_save(struct drm_device *dev, unsigned int pipe)
 {
 	struct drm_vblank_crtc *vblank = &dev->vblank[pipe];
 	unsigned long irqflags;
-	u32 vblcount;
-	s64 diff_ns;
-	bool vblrc;
-	struct timeval tvblank;
-	int count = DRM_TIMESTAMP_MAXRETRIES;
 
 	/* Prevent vblank irq processing while disabling vblank irqs,
 	 * so no updates of timestamps or count can happen after we've
@@ -193,26 +259,6 @@ static void vblank_disable_and_save(struct drm_device *dev, unsigned int pipe)
 	spin_lock_irqsave(&dev->vblank_time_lock, irqflags);
 
 	/*
-	 * If the vblank interrupt was already disabled update the count
-	 * and timestamp to maintain the appearance that the counter
-	 * has been ticking all along until this time. This makes the
-	 * count account for the entire time between drm_vblank_on() and
-	 * drm_vblank_off().
-	 *
-	 * But only do this if precise vblank timestamps are available.
-	 * Otherwise we might read a totally bogus timestamp since drivers
-	 * lacking precise timestamp support rely upon sampling the system clock
-	 * at vblank interrupt time. Which obviously won't work out well if the
-	 * vblank interrupt is disabled.
-	 */
-	if (!vblank->enabled &&
-	    drm_get_last_vbltimestamp(dev, pipe, &tvblank, 0)) {
-		drm_update_vblank_count(dev, pipe, 0);
-		spin_unlock_irqrestore(&dev->vblank_time_lock, irqflags);
-		return;
-	}
-
-	/*
 	 * Only disable vblank interrupts if they're enabled. This avoids
 	 * calling the ->disable_vblank() operation in atomic context with the
 	 * hardware potentially runtime suspended.
@@ -222,47 +268,13 @@ static void vblank_disable_and_save(struct drm_device *dev, unsigned int pipe)
 		vblank->enabled = false;
 	}
 
-	/* No further vblank irq's will be processed after
-	 * this point. Get current hardware vblank count and
-	 * vblank timestamp, repeat until they are consistent.
-	 *
-	 * FIXME: There is still a race condition here and in
-	 * drm_update_vblank_count() which can cause off-by-one
-	 * reinitialization of software vblank counter. If gpu
-	 * vblank counter doesn't increment exactly at the leading
-	 * edge of a vblank interval, then we can lose 1 count if
-	 * we happen to execute between start of vblank and the
-	 * delayed gpu counter increment.
-	 */
-	do {
-		vblank->last = dev->driver->get_vblank_counter(dev, pipe);
-		vblrc = drm_get_last_vbltimestamp(dev, pipe, &tvblank, 0);
-	} while (vblank->last != dev->driver->get_vblank_counter(dev, pipe) && (--count) && vblrc);
-
-	if (!count)
-		vblrc = 0;
-
-	/* Compute time difference to stored timestamp of last vblank
-	 * as updated by last invocation of drm_handle_vblank() in vblank irq.
-	 */
-	vblcount = vblank->count;
-	diff_ns = timeval_to_ns(&tvblank) -
-		  timeval_to_ns(&vblanktimestamp(dev, pipe, vblcount));
-
-	/* If there is at least 1 msec difference between the last stored
-	 * timestamp and tvblank, then we are currently executing our
-	 * disable inside a new vblank interval, the tvblank timestamp
-	 * corresponds to this new vblank interval and the irq handler
-	 * for this vblank didn't run yet and won't run due to our disable.
-	 * Therefore we need to do the job of drm_handle_vblank() and
-	 * increment the vblank counter by one to account for this vblank.
-	 *
-	 * Skip this step if there isn't any high precision timestamp
-	 * available. In that case we can't account for this and just
-	 * hope for the best.
+	/*
+	 * Always update the count and timestamp to maintain the
+	 * appearance that the counter has been ticking all along until
+	 * this time. This makes the count account for the entire time
+	 * between drm_vblank_on() and drm_vblank_off().
 	 */
-	if (vblrc && (abs64(diff_ns) > 1000000))
-		store_vblank(dev, pipe, 1, &tvblank);
+	drm_update_vblank_count(dev, pipe, 0);
 
 	spin_unlock_irqrestore(&dev->vblank_time_lock, irqflags);
 }
@@ -1325,16 +1337,8 @@ void drm_vblank_on(struct drm_device *dev, unsigned int pipe)
 		vblank->inmodeset = 0;
 	}
 
-	/*
-	 * sample the current counter to avoid random jumps
-	 * when drm_vblank_enable() applies the diff
-	 *
-	 * -1 to make sure user will never see the same
-	 * vblank counter value before and after a modeset
-	 */
-	vblank->last =
-		(dev->driver->get_vblank_counter(dev, pipe) - 1) &
-		dev->max_vblank_count;
+	drm_reset_vblank_timestamp(dev, pipe);
+
 	/*
 	 * re-enable interrupts if there are users left, or the
 	 * user wishes vblank interrupts to be enabled all the time.
@@ -1717,9 +1721,6 @@ static void drm_handle_vblank_events(struct drm_device *dev, unsigned int pipe)
 bool drm_handle_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	struct drm_vblank_crtc *vblank = &dev->vblank[pipe];
-	u32 vblcount;
-	s64 diff_ns;
-	struct timeval tvblank;
 	unsigned long irqflags;
 
 	if (WARN_ON_ONCE(!dev->num_crtcs))
@@ -1743,32 +1744,7 @@ bool drm_handle_vblank(struct drm_device *dev, unsigned int pipe)
 		return false;
 	}
 
-	/* Fetch corresponding timestamp for this vblank interval from
-	 * driver and store it in proper slot of timestamp ringbuffer.
-	 */
-
-	/* Get current timestamp and count. */
-	vblcount = vblank->count;
-	drm_get_last_vbltimestamp(dev, pipe, &tvblank, DRM_CALLED_FROM_VBLIRQ);
-
-	/* Compute time difference to timestamp of last vblank */
-	diff_ns = timeval_to_ns(&tvblank) -
-		  timeval_to_ns(&vblanktimestamp(dev, pipe, vblcount));
-
-	/* Update vblank timestamp and count if at least
-	 * DRM_REDUNDANT_VBLIRQ_THRESH_NS nanoseconds
-	 * difference between last stored timestamp and current
-	 * timestamp. A smaller difference means basically
-	 * identical timestamps. Happens if this vblank has
-	 * been already processed and this is a redundant call,
-	 * e.g., due to spurious vblank interrupts. We need to
-	 * ignore those for accounting.
-	 */
-	if (abs64(diff_ns) > DRM_REDUNDANT_VBLIRQ_THRESH_NS)
-		store_vblank(dev, pipe, 1, &tvblank);
-	else
-		DRM_DEBUG("crtc %u: Redundant vblirq ignored. diff_ns = %d\n",
-			  pipe, (int) diff_ns);
+	drm_update_vblank_count(dev, pipe, DRM_CALLED_FROM_VBLIRQ);
 
 	spin_unlock(&dev->vblank_time_lock);
 
-- 
cgit v0.10.2


From 13803132818cf8084d169617be060fd8e3411a98 Mon Sep 17 00:00:00 2001
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Date: Wed, 9 Sep 2015 16:40:56 +0200
Subject: drm/core: Preserve the framebuffer after removing it.

Previously RMFB and fd close chose to disable any plane that had
an active framebuffer from this file. If it was a primary plane the
crtc was disabled. However the fbdev code or any system compositor
should restore the planes anyway so there's no need to do it twice.

The old fb_id is zero'd, so there's no danger of being able to
restore the fb from fb_id.

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: David Herrmann <dh.herrmann@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c
index 9b9c4b4..626b0a5 100644
--- a/drivers/gpu/drm/drm_crtc.c
+++ b/drivers/gpu/drm/drm_crtc.c
@@ -3327,7 +3327,7 @@ int drm_mode_rmfb(struct drm_device *dev,
 	mutex_unlock(&dev->mode_config.fb_lock);
 	mutex_unlock(&file_priv->fbs_lock);
 
-	drm_framebuffer_remove(fb);
+	drm_framebuffer_unreference(fb);
 
 	return 0;
 
@@ -3517,7 +3517,7 @@ void drm_fb_release(struct drm_file *priv)
 		list_del_init(&fb->filp_head);
 
 		/* This will also drop the fpriv->fbs reference. */
-		drm_framebuffer_remove(fb);
+		drm_framebuffer_unreference(fb);
 	}
 }
 
-- 
cgit v0.10.2


From 73f7570bc6c853ca1fad24f9d31815b20e405354 Mon Sep 17 00:00:00 2001
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Date: Wed, 9 Sep 2015 16:40:57 +0200
Subject: drm/core: Preserve the fb id on close.

Keep the fb_id, which means that any application exiting without
unsetting the framebuffer from all planes will preserve its contents.

This is similar to preserving the initial framebuffer, except all
planes are preserved.

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: David Herrmann <dh.herrmann@gmail.com>
[danvet: Remove unused variable, reported by Stephen Rothwell.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c
index 626b0a5..e600a5f 100644
--- a/drivers/gpu/drm/drm_crtc.c
+++ b/drivers/gpu/drm/drm_crtc.c
@@ -3320,9 +3320,6 @@ int drm_mode_rmfb(struct drm_device *dev,
 	if (!found)
 		goto fail_lookup;
 
-	/* Mark fb as reaped, we still have a ref from fpriv->fbs. */
-	__drm_framebuffer_unregister(dev, fb);
-
 	list_del_init(&fb->filp_head);
 	mutex_unlock(&dev->mode_config.fb_lock);
 	mutex_unlock(&file_priv->fbs_lock);
@@ -3494,7 +3491,6 @@ out_err1:
  */
 void drm_fb_release(struct drm_file *priv)
 {
-	struct drm_device *dev = priv->minor->dev;
 	struct drm_framebuffer *fb, *tfb;
 
 	/*
@@ -3508,15 +3504,9 @@ void drm_fb_release(struct drm_file *priv)
 	 * at it any more.
 	 */
 	list_for_each_entry_safe(fb, tfb, &priv->fbs, filp_head) {
-
-		mutex_lock(&dev->mode_config.fb_lock);
-		/* Mark fb as reaped, we still have a ref from fpriv->fbs. */
-		__drm_framebuffer_unregister(dev, fb);
-		mutex_unlock(&dev->mode_config.fb_lock);
-
 		list_del_init(&fb->filp_head);
 
-		/* This will also drop the fpriv->fbs reference. */
+		/* This drops the fpriv->fbs reference. */
 		drm_framebuffer_unreference(fb);
 	}
 }
-- 
cgit v0.10.2


From 21b45676b7c4b79334d8fe3c5a112af0517b66e9 Mon Sep 17 00:00:00 2001
From: Lukas Wunner <lukas@wunner.de>
Date: Thu, 27 Aug 2015 16:43:43 +0200
Subject: vga_switcheroo: Set active attribute to false for audio clients

The active attribute in struct vga_switcheroo_client denotes whether
the outputs are currently switched to this client. The attribute is
only meaningful for vga clients. It is never used for audio clients.

The function vga_switcheroo_register_audio_client() misuses this
attribute to store whether the audio device is fully initialized.
Most likely there was a misunderstanding about the meaning of
"active" when this was added.

Comment from Takashi's review:

"Not really.  The full initialization of audio was meant that the audio
is active indeed.  Admittedly, though, the active flag for each audio
client doesn't play any role because the audio always follows the gfx
state changes, and the value passed there doesn't reflect the actual
state due to the later change.  So, I agree with the removal of the
flag itself -- or let the audio active flag following the
corresponding gfx flag.  The latter will make the proc output more
consistent while the former is certainly more reduction of code."

Set the active attribute to false for audio clients. Remove the
active parameter from vga_switcheroo_register_audio_client() and
its sole caller, hda_intel.c:register_vga_switcheroo().

vga_switcheroo_register_audio_client() was introduced by 3e9e63dbd374
("vga_switcheroo: Add the support for audio clients"). Its use in
hda_intel.c was introduced by a82d51ed24bb ("ALSA: hda - Support
VGA-switcheroo").

v1.1: The changes above imply that in find_active_client() the call
to client_is_vga() is now superfluous. Drop it.

Cc: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Lukas Wunner <lukas@wunner.de>
[danvet: Add Takashi's clarification to the commit message.]
Reviewed-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/vga/vga_switcheroo.c b/drivers/gpu/vga/vga_switcheroo.c
index 67a5709..86c03b5 100644
--- a/drivers/gpu/vga/vga_switcheroo.c
+++ b/drivers/gpu/vga/vga_switcheroo.c
@@ -288,7 +288,6 @@ EXPORT_SYMBOL(vga_switcheroo_register_client);
  * @pdev: client pci device
  * @ops: client callbacks
  * @id: client identifier, see enum vga_switcheroo_client_id
- * @active: whether the audio device is fully initialized
  *
  * Register audio client (audio device on a GPU). The power state of the
  * client is assumed to be ON.
@@ -297,9 +296,9 @@ EXPORT_SYMBOL(vga_switcheroo_register_client);
  */
 int vga_switcheroo_register_audio_client(struct pci_dev *pdev,
 					 const struct vga_switcheroo_client_ops *ops,
-					 int id, bool active)
+					 int id)
 {
-	return register_client(pdev, ops, id | ID_BIT_AUDIO, active, false);
+	return register_client(pdev, ops, id | ID_BIT_AUDIO, false, false);
 }
 EXPORT_SYMBOL(vga_switcheroo_register_audio_client);
 
@@ -331,7 +330,7 @@ find_active_client(struct list_head *head)
 	struct vga_switcheroo_client *client;
 
 	list_for_each_entry(client, head, list)
-		if (client->active && client_is_vga(client))
+		if (client->active)
 			return client;
 	return NULL;
 }
diff --git a/include/linux/vga_switcheroo.h b/include/linux/vga_switcheroo.h
index fe90bfc..3764991 100644
--- a/include/linux/vga_switcheroo.h
+++ b/include/linux/vga_switcheroo.h
@@ -128,7 +128,7 @@ int vga_switcheroo_register_client(struct pci_dev *dev,
 				   bool driver_power_control);
 int vga_switcheroo_register_audio_client(struct pci_dev *pdev,
 					 const struct vga_switcheroo_client_ops *ops,
-					 int id, bool active);
+					 int id);
 
 void vga_switcheroo_client_fb_set(struct pci_dev *dev,
 				  struct fb_info *info);
@@ -154,7 +154,7 @@ static inline void vga_switcheroo_client_fb_set(struct pci_dev *dev, struct fb_i
 static inline int vga_switcheroo_register_handler(struct vga_switcheroo_handler *handler) { return 0; }
 static inline int vga_switcheroo_register_audio_client(struct pci_dev *pdev,
 	const struct vga_switcheroo_client_ops *ops,
-	int id, bool active) { return 0; }
+	int id) { return 0; }
 static inline void vga_switcheroo_unregister_handler(void) {}
 static inline int vga_switcheroo_process_delayed_switch(void) { return 0; }
 static inline int vga_switcheroo_get_client_state(struct pci_dev *dev) { return VGA_SWITCHEROO_ON; }
diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c
index c38c68f..e819013 100644
--- a/sound/pci/hda/hda_intel.c
+++ b/sound/pci/hda/hda_intel.c
@@ -1143,8 +1143,7 @@ static int register_vga_switcheroo(struct azx *chip)
 	 * is there any machine with two switchable HDMI audio controllers?
 	 */
 	err = vga_switcheroo_register_audio_client(chip->pci, &azx_vs_ops,
-						    VGA_SWITCHEROO_DIS,
-						    hda->probe_continued);
+						   VGA_SWITCHEROO_DIS);
 	if (err < 0)
 		return err;
 	hda->vga_switcheroo_registered = 1;
-- 
cgit v0.10.2


From 5fbec9014b5ab9deecb3338cd14aae82ed14d0ec Mon Sep 17 00:00:00 2001
From: Thierry Reding <treding@nvidia.com>
Date: Thu, 24 Sep 2015 10:07:32 +0200
Subject: drm/gma500: Remove DP_LINK_STATUS_SIZE redefinition

The DRM/DP helpers already contain a definition for this macro. Remove
the duplicate in the GMA500 driver to avoid having to keep both updated
synchronously.

Cc: Patrik Jakobsson <patrik.r.jakobsson@gmail.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
Reviewed-by: Patrik Jakobsson <patrik.r.jakobsson@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/gma500/cdv_intel_dp.c b/drivers/gpu/drm/gma500/cdv_intel_dp.c
index 0fafb8e..17cea40 100644
--- a/drivers/gpu/drm/gma500/cdv_intel_dp.c
+++ b/drivers/gpu/drm/gma500/cdv_intel_dp.c
@@ -247,7 +247,6 @@ i2c_dp_aux_add_bus(struct i2c_adapter *adapter)
 
 #define wait_for(COND, MS) _wait_for(COND, MS, 1)
 
-#define DP_LINK_STATUS_SIZE	6
 #define DP_LINK_CHECK_TIMEOUT	(10 * 1000)
 
 #define DP_LINK_CONFIGURATION_SIZE	9
-- 
cgit v0.10.2


From bf22f3be15630234b76d7ece0b1f97627ea833a8 Mon Sep 17 00:00:00 2001
From: Geliang Tang <geliangtang@163.com>
Date: Thu, 24 Sep 2015 03:01:03 -0700
Subject: drm: drm_atomic_crtc_get_property should be static

Fixes the following sparse warning:
 drivers/gpu/drm/drm_atomic.c:442:5: warning: symbol
 'drm_atomic_crtc_get_property' was not declared. Should it be static?

Signed-off-by: Geliang Tang <geliangtang@163.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_atomic.c b/drivers/gpu/drm/drm_atomic.c
index 940f80b..7bb3845 100644
--- a/drivers/gpu/drm/drm_atomic.c
+++ b/drivers/gpu/drm/drm_atomic.c
@@ -438,7 +438,8 @@ EXPORT_SYMBOL(drm_atomic_crtc_set_property);
  * consistent behavior you must call this function rather than the
  * driver hook directly.
  */
-int drm_atomic_crtc_get_property(struct drm_crtc *crtc,
+static int
+drm_atomic_crtc_get_property(struct drm_crtc *crtc,
 		const struct drm_crtc_state *state,
 		struct drm_property *property, uint64_t *val)
 {
-- 
cgit v0.10.2


From 2b193f023dc8f43b005bc7844b07e437a4e418db Mon Sep 17 00:00:00 2001
From: Thierry Reding <treding@nvidia.com>
Date: Thu, 24 Sep 2015 18:35:30 +0200
Subject: drm/irq: Rename drm_crtc -> crtc

Since the original crtc parameter was renamed to pipe, there is no
longer a need to artificially prefix the CRTC parameter.

Signed-off-by: Thierry Reding <treding@nvidia.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c
index 88fbee4..29a6dcd 100644
--- a/drivers/gpu/drm/drm_irq.c
+++ b/drivers/gpu/drm/drm_irq.c
@@ -1280,7 +1280,7 @@ EXPORT_SYMBOL(drm_crtc_vblank_off);
 
 /**
  * drm_crtc_vblank_reset - reset vblank state to off on a CRTC
- * @drm_crtc: CRTC in question
+ * @crtc: CRTC in question
  *
  * Drivers can use this function to reset the vblank state to off at load time.
  * Drivers should use this together with the drm_crtc_vblank_off() and
@@ -1288,11 +1288,11 @@ EXPORT_SYMBOL(drm_crtc_vblank_off);
  * drm_crtc_vblank_off() is that this function doesn't save the vblank counter
  * and hence doesn't need to call any driver hooks.
  */
-void drm_crtc_vblank_reset(struct drm_crtc *drm_crtc)
+void drm_crtc_vblank_reset(struct drm_crtc *crtc)
 {
-	struct drm_device *dev = drm_crtc->dev;
+	struct drm_device *dev = crtc->dev;
 	unsigned long irqflags;
-	unsigned int pipe = drm_crtc_index(drm_crtc);
+	unsigned int pipe = drm_crtc_index(crtc);
 	struct drm_vblank_crtc *vblank = &dev->vblank[pipe];
 
 	spin_lock_irqsave(&dev->vbl_lock, irqflags);
-- 
cgit v0.10.2


From cf6483050e9bf13979415d9fd388554d8c8f3477 Mon Sep 17 00:00:00 2001
From: Thierry Reding <treding@nvidia.com>
Date: Thu, 24 Sep 2015 18:35:36 +0200
Subject: drm/irq: Add drm_crtc_vblank_count_and_time()

This function is the KMS native variant of drm_vblank_count_and_time().
It takes a struct drm_crtc * instead of a struct drm_device * and an
index of the CRTC.

Eventually the goal is to access vblank data through the CRTC only so
that the per-CRTC data can be moved to struct drm_crtc.

Signed-off-by: Thierry Reding <treding@nvidia.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c
index 29a6dcd..ed2394e 100644
--- a/drivers/gpu/drm/drm_irq.c
+++ b/drivers/gpu/drm/drm_irq.c
@@ -917,6 +917,8 @@ EXPORT_SYMBOL(drm_crtc_vblank_count);
  * vblank events since the system was booted, including lost events due to
  * modesetting activity. Returns corresponding system timestamp of the time
  * of the vblank interval that corresponds to the current vblank counter value.
+ *
+ * This is the legacy version of drm_crtc_vblank_count_and_time().
  */
 u32 drm_vblank_count_and_time(struct drm_device *dev, unsigned int pipe,
 			      struct timeval *vblanktime)
@@ -944,6 +946,27 @@ u32 drm_vblank_count_and_time(struct drm_device *dev, unsigned int pipe,
 }
 EXPORT_SYMBOL(drm_vblank_count_and_time);
 
+/**
+ * drm_crtc_vblank_count_and_time - retrieve "cooked" vblank counter value
+ *     and the system timestamp corresponding to that vblank counter value
+ * @crtc: which counter to retrieve
+ * @vblanktime: Pointer to struct timeval to receive the vblank timestamp.
+ *
+ * Fetches the "cooked" vblank count value that represents the number of
+ * vblank events since the system was booted, including lost events due to
+ * modesetting activity. Returns corresponding system timestamp of the time
+ * of the vblank interval that corresponds to the current vblank counter value.
+ *
+ * This is the native KMS version of drm_vblank_count_and_time().
+ */
+u32 drm_crtc_vblank_count_and_time(struct drm_crtc *crtc,
+				   struct timeval *vblanktime)
+{
+	return drm_vblank_count_and_time(crtc->dev, drm_crtc_index(crtc),
+					 vblanktime);
+}
+EXPORT_SYMBOL(drm_crtc_vblank_count_and_time);
+
 static void send_vblank_event(struct drm_device *dev,
 		struct drm_pending_vblank_event *e,
 		unsigned long seq, struct timeval *now)
diff --git a/include/drm/drmP.h b/include/drm/drmP.h
index 6717a7d..d0251ac 100644
--- a/include/drm/drmP.h
+++ b/include/drm/drmP.h
@@ -932,6 +932,8 @@ extern u32 drm_vblank_count(struct drm_device *dev, int pipe);
 extern u32 drm_crtc_vblank_count(struct drm_crtc *crtc);
 extern u32 drm_vblank_count_and_time(struct drm_device *dev, unsigned int pipe,
 				     struct timeval *vblanktime);
+extern u32 drm_crtc_vblank_count_and_time(struct drm_crtc *crtc,
+					  struct timeval *vblanktime);
 extern void drm_send_vblank_event(struct drm_device *dev, unsigned int pipe,
 				  struct drm_pending_vblank_event *e);
 extern void drm_crtc_send_vblank_event(struct drm_crtc *crtc,
-- 
cgit v0.10.2


From 7cb5dff8d59da8008c6ec30d4161e8e96aa4afe8 Mon Sep 17 00:00:00 2001
From: Geliang Tang <geliangtang@163.com>
Date: Fri, 25 Sep 2015 03:58:11 -0700
Subject: drm/i915: fix task reference leak in i915_debugfs.c

Leak a task reference in i915_ppgtt_info(), add put_task_struct()
to fix it.

Introduced by

commit 1c60fef535d143860d5bf6593e24ab6417f5227c
Author: Ben Widawsky <ben@bwidawsk.net>
Date:   Fri Dec 6 14:11:30 2013 -0800

    drm/i915: Dump all ppgtt

Signed-off-by: Geliang Tang <geliangtang@163.com>
[danvet: Add note provided by Jani about which commit introduced this
issue.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index 5615d3d..8797717 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -2288,9 +2288,13 @@ static int i915_ppgtt_info(struct seq_file *m, void *data)
 
 	list_for_each_entry_reverse(file, &dev->filelist, lhead) {
 		struct drm_i915_file_private *file_priv = file->driver_priv;
+		struct task_struct *task;
 
-		seq_printf(m, "\nproc: %s\n",
-			   get_pid_task(file->pid, PIDTYPE_PID)->comm);
+		task = get_pid_task(file->pid, PIDTYPE_PID);
+		if (!task)
+			return -ESRCH;
+		seq_printf(m, "\nproc: %s\n", task->comm);
+		put_task_struct(task);
 		idr_for_each(&file_priv->context_idr, per_file_ctx,
 			     (void *)(unsigned long)m);
 	}
-- 
cgit v0.10.2


From 40a4a5727f21a0e439d317aa99953e24467605eb Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Mon, 28 Sep 2015 08:35:34 +0200
Subject: drm/i915: Update DRIVER_DATE to 20150928

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 6c3ef9c..c20f2c7 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -57,7 +57,7 @@
 
 #define DRIVER_NAME		"i915"
 #define DRIVER_DESC		"Intel Graphics"
-#define DRIVER_DATE		"20150911"
+#define DRIVER_DATE		"20150928"
 
 #undef WARN_ON
 /* Many gcc seem to no see through this and fall over :( */
-- 
cgit v0.10.2


From 3d58e31888318ebd157ba0680fb24cdcd1bc3d6f Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Fri, 14 Aug 2015 11:13:50 +0100
Subject: drm/i2c: tda998x: remove DRM slave encoder support

Remove the DRM slave encoder compatibility from the TDA998x driver.  We
now use the component helpers to manage the binding of DRM sub-drivers.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/i2c/tda998x_drv.c b/drivers/gpu/drm/i2c/tda998x_drv.c
index 1c6fc24..883025d 100644
--- a/drivers/gpu/drm/i2c/tda998x_drv.c
+++ b/drivers/gpu/drm/i2c/tda998x_drv.c
@@ -23,7 +23,6 @@
 
 #include <drm/drmP.h>
 #include <drm/drm_crtc_helper.h>
-#include <drm/drm_encoder_slave.h>
 #include <drm/drm_edid.h>
 #include <drm/drm_of.h>
 #include <drm/i2c/tda998x.h>
@@ -53,8 +52,6 @@ struct tda998x_priv {
 	bool edid_delay_active;
 };
 
-#define to_tda998x_priv(x)  ((struct tda998x_priv *)to_encoder_slave(x)->slave_priv)
-
 /* The TDA9988 series of devices use a paged register scheme.. to simplify
  * things we encode the page # in upper bits of the register #.  To read/
  * write a given register, we need to make sure CURPAGE register is set
@@ -1182,16 +1179,6 @@ static void tda998x_encoder_set_polling(struct tda998x_priv *priv,
 			DRM_CONNECTOR_POLL_DISCONNECT;
 }
 
-static int
-tda998x_encoder_set_property(struct drm_encoder *encoder,
-			    struct drm_connector *connector,
-			    struct drm_property *property,
-			    uint64_t val)
-{
-	DBG("");
-	return 0;
-}
-
 static void tda998x_destroy(struct tda998x_priv *priv)
 {
 	/* disable all IRQs and free the IRQ handler */
@@ -1207,78 +1194,6 @@ static void tda998x_destroy(struct tda998x_priv *priv)
 	i2c_unregister_device(priv->cec);
 }
 
-/* Slave encoder support */
-
-static void
-tda998x_encoder_slave_set_config(struct drm_encoder *encoder, void *params)
-{
-	tda998x_encoder_set_config(to_tda998x_priv(encoder), params);
-}
-
-static void tda998x_encoder_slave_destroy(struct drm_encoder *encoder)
-{
-	struct tda998x_priv *priv = to_tda998x_priv(encoder);
-
-	tda998x_destroy(priv);
-	drm_i2c_encoder_destroy(encoder);
-	kfree(priv);
-}
-
-static void tda998x_encoder_slave_dpms(struct drm_encoder *encoder, int mode)
-{
-	tda998x_encoder_dpms(to_tda998x_priv(encoder), mode);
-}
-
-static int tda998x_encoder_slave_mode_valid(struct drm_encoder *encoder,
-					    struct drm_display_mode *mode)
-{
-	return tda998x_encoder_mode_valid(to_tda998x_priv(encoder), mode);
-}
-
-static void
-tda998x_encoder_slave_mode_set(struct drm_encoder *encoder,
-			       struct drm_display_mode *mode,
-			       struct drm_display_mode *adjusted_mode)
-{
-	tda998x_encoder_mode_set(to_tda998x_priv(encoder), mode, adjusted_mode);
-}
-
-static enum drm_connector_status
-tda998x_encoder_slave_detect(struct drm_encoder *encoder,
-			     struct drm_connector *connector)
-{
-	return tda998x_encoder_detect(to_tda998x_priv(encoder));
-}
-
-static int tda998x_encoder_slave_get_modes(struct drm_encoder *encoder,
-					   struct drm_connector *connector)
-{
-	return tda998x_encoder_get_modes(to_tda998x_priv(encoder), connector);
-}
-
-static int
-tda998x_encoder_slave_create_resources(struct drm_encoder *encoder,
-				       struct drm_connector *connector)
-{
-	tda998x_encoder_set_polling(to_tda998x_priv(encoder), connector);
-	return 0;
-}
-
-static struct drm_encoder_slave_funcs tda998x_encoder_slave_funcs = {
-	.set_config = tda998x_encoder_slave_set_config,
-	.destroy = tda998x_encoder_slave_destroy,
-	.dpms = tda998x_encoder_slave_dpms,
-	.save = tda998x_encoder_save,
-	.restore = tda998x_encoder_restore,
-	.mode_fixup = tda998x_encoder_mode_fixup,
-	.mode_valid = tda998x_encoder_slave_mode_valid,
-	.mode_set = tda998x_encoder_slave_mode_set,
-	.detect = tda998x_encoder_slave_detect,
-	.get_modes = tda998x_encoder_slave_get_modes,
-	.create_resources = tda998x_encoder_slave_create_resources,
-	.set_property = tda998x_encoder_set_property,
-};
-
 /* I2C driver functions */
 
 static int tda998x_create(struct i2c_client *client, struct tda998x_priv *priv)
@@ -1413,31 +1328,6 @@ fail:
 	return -ENXIO;
 }
 
-static int tda998x_encoder_init(struct i2c_client *client,
-				struct drm_device *dev,
-				struct drm_encoder_slave *encoder_slave)
-{
-	struct tda998x_priv *priv;
-	int ret;
-
-	priv = kzalloc(sizeof(*priv), GFP_KERNEL);
-	if (!priv)
-		return -ENOMEM;
-
-	priv->encoder = &encoder_slave->base;
-
-	ret = tda998x_create(client, priv);
-	if (ret) {
-		kfree(priv);
-		return ret;
-	}
-
-	encoder_slave->slave_priv = priv;
-	encoder_slave->slave_funcs = &tda998x_encoder_slave_funcs;
-
-	return 0;
-}
-
 struct tda998x_priv2 {
 	struct tda998x_priv base;
 	struct drm_encoder encoder;
@@ -1659,38 +1549,18 @@ static struct i2c_device_id tda998x_ids[] = {
 };
 MODULE_DEVICE_TABLE(i2c, tda998x_ids);
 
-static struct drm_i2c_encoder_driver tda998x_driver = {
-	.i2c_driver = {
-		.probe = tda998x_probe,
-		.remove = tda998x_remove,
-		.driver = {
-			.name = "tda998x",
-			.of_match_table = of_match_ptr(tda998x_dt_ids),
-		},
-		.id_table = tda998x_ids,
+static struct i2c_driver tda998x_driver = {
+	.probe = tda998x_probe,
+	.remove = tda998x_remove,
+	.driver = {
+		.name = "tda998x",
+		.of_match_table = of_match_ptr(tda998x_dt_ids),
 	},
-	.encoder_init = tda998x_encoder_init,
+	.id_table = tda998x_ids,
 };
 
-/* Module initialization */
-
-static int __init
-tda998x_init(void)
-{
-	DBG("");
-	return drm_i2c_encoder_register(THIS_MODULE, &tda998x_driver);
-}
-
-static void __exit
-tda998x_exit(void)
-{
-	DBG("");
-	drm_i2c_encoder_unregister(&tda998x_driver);
-}
+module_i2c_driver(tda998x_driver);
 
 MODULE_AUTHOR("Rob Clark <robdclark@gmail.com");
 MODULE_DESCRIPTION("NXP Semiconductors TDA998X HDMI Encoder");
 MODULE_LICENSE("GPL");
-
-module_init(tda998x_init);
-module_exit(tda998x_exit);
-- 
cgit v0.10.2


From 78e401f9891a8ecdb075c0ca46f9f43ce0ed0c4e Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Fri, 14 Aug 2015 11:17:12 +0100
Subject: drm/i2c: tda998x: remove encoder pointer

Remove the encoder pointer from struct tda998x_priv, moving the encoder
itself from struct tda998x_priv2 here.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/i2c/tda998x_drv.c b/drivers/gpu/drm/i2c/tda998x_drv.c
index 883025d..e30a2a8 100644
--- a/drivers/gpu/drm/i2c/tda998x_drv.c
+++ b/drivers/gpu/drm/i2c/tda998x_drv.c
@@ -44,12 +44,13 @@ struct tda998x_priv {
 
 	wait_queue_head_t wq_edid;
 	volatile int wq_edid_wait;
-	struct drm_encoder *encoder;
 
 	struct work_struct detect_work;
 	struct timer_list edid_delay_timer;
 	wait_queue_head_t edid_delay_waitq;
 	bool edid_delay_active;
+
+	struct drm_encoder encoder;
 };
 
 /* The TDA9988 series of devices use a paged register scheme.. to simplify
@@ -594,7 +595,7 @@ static void tda998x_detect_work(struct work_struct *work)
 {
 	struct tda998x_priv *priv =
 		container_of(work, struct tda998x_priv, detect_work);
-	struct drm_device *dev = priv->encoder->dev;
+	struct drm_device *dev = priv->encoder.dev;
 
 	if (dev)
 		drm_kms_helper_hotplug_event(dev);
@@ -1330,7 +1331,6 @@ fail:
 
 struct tda998x_priv2 {
 	struct tda998x_priv base;
-	struct drm_encoder encoder;
 	struct drm_connector connector;
 };
 
@@ -1338,7 +1338,7 @@ struct tda998x_priv2 {
 	container_of(x, struct tda998x_priv2, connector);
 
 #define enc_to_tda998x_priv2(x) \
-	container_of(x, struct tda998x_priv2, encoder);
+	container_of(x, struct tda998x_priv2, base.encoder);
 
 static void tda998x_encoder2_dpms(struct drm_encoder *encoder, int mode)
 {
@@ -1408,7 +1408,7 @@ tda998x_connector_best_encoder(struct drm_connector *connector)
 {
 	struct tda998x_priv2 *priv = conn_to_tda998x_priv2(connector);
 
-	return &priv->encoder;
+	return &priv->base.encoder;
 }
 
 static
@@ -1463,9 +1463,8 @@ static int tda998x_bind(struct device *dev, struct device *master, void *data)
 		crtcs = 1 << 0;
 	}
 
-	priv->base.encoder = &priv->encoder;
 	priv->connector.interlace_allowed = 1;
-	priv->encoder.possible_crtcs = crtcs;
+	priv->base.encoder.possible_crtcs = crtcs;
 
 	ret = tda998x_create(client, &priv->base);
 	if (ret)
@@ -1476,8 +1475,8 @@ static int tda998x_bind(struct device *dev, struct device *master, void *data)
 
 	tda998x_encoder_set_polling(&priv->base, &priv->connector);
 
-	drm_encoder_helper_add(&priv->encoder, &tda998x_encoder_helper_funcs);
-	ret = drm_encoder_init(drm, &priv->encoder, &tda998x_encoder_funcs,
+	drm_encoder_helper_add(&priv->base.encoder, &tda998x_encoder_helper_funcs);
+	ret = drm_encoder_init(drm, &priv->base.encoder, &tda998x_encoder_funcs,
 			       DRM_MODE_ENCODER_TMDS);
 	if (ret)
 		goto err_encoder;
@@ -1494,15 +1493,15 @@ static int tda998x_bind(struct device *dev, struct device *master, void *data)
 	if (ret)
 		goto err_sysfs;
 
-	priv->connector.encoder = &priv->encoder;
-	drm_mode_connector_attach_encoder(&priv->connector, &priv->encoder);
+	priv->connector.encoder = &priv->base.encoder;
+	drm_mode_connector_attach_encoder(&priv->connector, &priv->base.encoder);
 
 	return 0;
 
 err_sysfs:
 	drm_connector_cleanup(&priv->connector);
 err_connector:
-	drm_encoder_cleanup(&priv->encoder);
+	drm_encoder_cleanup(&priv->base.encoder);
 err_encoder:
 	tda998x_destroy(&priv->base);
 	return ret;
@@ -1514,7 +1513,7 @@ static void tda998x_unbind(struct device *dev, struct device *master,
 	struct tda998x_priv2 *priv = dev_get_drvdata(dev);
 
 	drm_connector_cleanup(&priv->connector);
-	drm_encoder_cleanup(&priv->encoder);
+	drm_encoder_cleanup(&priv->base.encoder);
 	tda998x_destroy(&priv->base);
 }
 
-- 
cgit v0.10.2


From eed64b5963f1496be62ba41f536346b2186727a0 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Fri, 14 Aug 2015 11:18:28 +0100
Subject: drm/i2c: tda998x: move connector into struct tda998x_priv

Move the DRM connector structure into struct tda998x_priv from the old
struct tda998x_priv2.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/i2c/tda998x_drv.c b/drivers/gpu/drm/i2c/tda998x_drv.c
index e30a2a8..a2a463c 100644
--- a/drivers/gpu/drm/i2c/tda998x_drv.c
+++ b/drivers/gpu/drm/i2c/tda998x_drv.c
@@ -51,6 +51,7 @@ struct tda998x_priv {
 	bool edid_delay_active;
 
 	struct drm_encoder encoder;
+	struct drm_connector connector;
 };
 
 /* The TDA9988 series of devices use a paged register scheme.. to simplify
@@ -1331,11 +1332,10 @@ fail:
 
 struct tda998x_priv2 {
 	struct tda998x_priv base;
-	struct drm_connector connector;
 };
 
 #define conn_to_tda998x_priv2(x) \
-	container_of(x, struct tda998x_priv2, connector);
+	container_of(x, struct tda998x_priv2, base.connector);
 
 #define enc_to_tda998x_priv2(x) \
 	container_of(x, struct tda998x_priv2, base.encoder);
@@ -1463,7 +1463,7 @@ static int tda998x_bind(struct device *dev, struct device *master, void *data)
 		crtcs = 1 << 0;
 	}
 
-	priv->connector.interlace_allowed = 1;
+	priv->base.connector.interlace_allowed = 1;
 	priv->base.encoder.possible_crtcs = crtcs;
 
 	ret = tda998x_create(client, &priv->base);
@@ -1473,7 +1473,7 @@ static int tda998x_bind(struct device *dev, struct device *master, void *data)
 	if (!dev->of_node && params)
 		tda998x_encoder_set_config(&priv->base, params);
 
-	tda998x_encoder_set_polling(&priv->base, &priv->connector);
+	tda998x_encoder_set_polling(&priv->base, &priv->base.connector);
 
 	drm_encoder_helper_add(&priv->base.encoder, &tda998x_encoder_helper_funcs);
 	ret = drm_encoder_init(drm, &priv->base.encoder, &tda998x_encoder_funcs,
@@ -1481,25 +1481,25 @@ static int tda998x_bind(struct device *dev, struct device *master, void *data)
 	if (ret)
 		goto err_encoder;
 
-	drm_connector_helper_add(&priv->connector,
+	drm_connector_helper_add(&priv->base.connector,
 				 &tda998x_connector_helper_funcs);
-	ret = drm_connector_init(drm, &priv->connector,
+	ret = drm_connector_init(drm, &priv->base.connector,
 				 &tda998x_connector_funcs,
 				 DRM_MODE_CONNECTOR_HDMIA);
 	if (ret)
 		goto err_connector;
 
-	ret = drm_connector_register(&priv->connector);
+	ret = drm_connector_register(&priv->base.connector);
 	if (ret)
 		goto err_sysfs;
 
-	priv->connector.encoder = &priv->base.encoder;
-	drm_mode_connector_attach_encoder(&priv->connector, &priv->base.encoder);
+	priv->base.connector.encoder = &priv->base.encoder;
+	drm_mode_connector_attach_encoder(&priv->base.connector, &priv->base.encoder);
 
 	return 0;
 
 err_sysfs:
-	drm_connector_cleanup(&priv->connector);
+	drm_connector_cleanup(&priv->base.connector);
 err_connector:
 	drm_encoder_cleanup(&priv->base.encoder);
 err_encoder:
@@ -1512,7 +1512,7 @@ static void tda998x_unbind(struct device *dev, struct device *master,
 {
 	struct tda998x_priv2 *priv = dev_get_drvdata(dev);
 
-	drm_connector_cleanup(&priv->connector);
+	drm_connector_cleanup(&priv->base.connector);
 	drm_encoder_cleanup(&priv->base.encoder);
 	tda998x_destroy(&priv->base);
 }
-- 
cgit v0.10.2


From a3584f60f4898c9479931cdca1dc19f758af45fb Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Fri, 14 Aug 2015 11:22:50 +0100
Subject: drm/i2c: tda998x: kill struct tda998x_priv2

Kill the redundant tda998x_priv2 structure now that its only member is
the struct tda998x_priv.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/i2c/tda998x_drv.c b/drivers/gpu/drm/i2c/tda998x_drv.c
index a2a463c..8bca9155 100644
--- a/drivers/gpu/drm/i2c/tda998x_drv.c
+++ b/drivers/gpu/drm/i2c/tda998x_drv.c
@@ -1330,21 +1330,17 @@ fail:
 	return -ENXIO;
 }
 
-struct tda998x_priv2 {
-	struct tda998x_priv base;
-};
-
-#define conn_to_tda998x_priv2(x) \
-	container_of(x, struct tda998x_priv2, base.connector);
+#define conn_to_tda998x_priv(x) \
+	container_of(x, struct tda998x_priv, connector);
 
-#define enc_to_tda998x_priv2(x) \
-	container_of(x, struct tda998x_priv2, base.encoder);
+#define enc_to_tda998x_priv(x) \
+	container_of(x, struct tda998x_priv, encoder);
 
 static void tda998x_encoder2_dpms(struct drm_encoder *encoder, int mode)
 {
-	struct tda998x_priv2 *priv = enc_to_tda998x_priv2(encoder);
+	struct tda998x_priv *priv = enc_to_tda998x_priv(encoder);
 
-	tda998x_encoder_dpms(&priv->base, mode);
+	tda998x_encoder_dpms(priv, mode);
 }
 
 static void tda998x_encoder_prepare(struct drm_encoder *encoder)
@@ -1361,9 +1357,9 @@ static void tda998x_encoder2_mode_set(struct drm_encoder *encoder,
 				      struct drm_display_mode *mode,
 				      struct drm_display_mode *adjusted_mode)
 {
-	struct tda998x_priv2 *priv = enc_to_tda998x_priv2(encoder);
+	struct tda998x_priv *priv = enc_to_tda998x_priv(encoder);
 
-	tda998x_encoder_mode_set(&priv->base, mode, adjusted_mode);
+	tda998x_encoder_mode_set(priv, mode, adjusted_mode);
 }
 
 static const struct drm_encoder_helper_funcs tda998x_encoder_helper_funcs = {
@@ -1378,9 +1374,9 @@ static const struct drm_encoder_helper_funcs tda998x_encoder_helper_funcs = {
 
 static void tda998x_encoder_destroy(struct drm_encoder *encoder)
 {
-	struct tda998x_priv2 *priv = enc_to_tda998x_priv2(encoder);
+	struct tda998x_priv *priv = enc_to_tda998x_priv(encoder);
 
-	tda998x_destroy(&priv->base);
+	tda998x_destroy(priv);
 	drm_encoder_cleanup(encoder);
 }
 
@@ -1390,25 +1386,25 @@ static const struct drm_encoder_funcs tda998x_encoder_funcs = {
 
 static int tda998x_connector_get_modes(struct drm_connector *connector)
 {
-	struct tda998x_priv2 *priv = conn_to_tda998x_priv2(connector);
+	struct tda998x_priv *priv = conn_to_tda998x_priv(connector);
 
-	return tda998x_encoder_get_modes(&priv->base, connector);
+	return tda998x_encoder_get_modes(priv, connector);
 }
 
 static int tda998x_connector_mode_valid(struct drm_connector *connector,
 					struct drm_display_mode *mode)
 {
-	struct tda998x_priv2 *priv = conn_to_tda998x_priv2(connector);
+	struct tda998x_priv *priv = conn_to_tda998x_priv(connector);
 
-	return tda998x_encoder_mode_valid(&priv->base, mode);
+	return tda998x_encoder_mode_valid(priv, mode);
 }
 
 static struct drm_encoder *
 tda998x_connector_best_encoder(struct drm_connector *connector)
 {
-	struct tda998x_priv2 *priv = conn_to_tda998x_priv2(connector);
+	struct tda998x_priv *priv = conn_to_tda998x_priv(connector);
 
-	return &priv->base.encoder;
+	return &priv->encoder;
 }
 
 static
@@ -1421,9 +1417,9 @@ const struct drm_connector_helper_funcs tda998x_connector_helper_funcs = {
 static enum drm_connector_status
 tda998x_connector_detect(struct drm_connector *connector, bool force)
 {
-	struct tda998x_priv2 *priv = conn_to_tda998x_priv2(connector);
+	struct tda998x_priv *priv = conn_to_tda998x_priv(connector);
 
-	return tda998x_encoder_detect(&priv->base);
+	return tda998x_encoder_detect(priv);
 }
 
 static void tda998x_connector_destroy(struct drm_connector *connector)
@@ -1444,7 +1440,7 @@ static int tda998x_bind(struct device *dev, struct device *master, void *data)
 	struct tda998x_encoder_params *params = dev->platform_data;
 	struct i2c_client *client = to_i2c_client(dev);
 	struct drm_device *drm = data;
-	struct tda998x_priv2 *priv;
+	struct tda998x_priv *priv;
 	u32 crtcs = 0;
 	int ret;
 
@@ -1463,58 +1459,58 @@ static int tda998x_bind(struct device *dev, struct device *master, void *data)
 		crtcs = 1 << 0;
 	}
 
-	priv->base.connector.interlace_allowed = 1;
-	priv->base.encoder.possible_crtcs = crtcs;
+	priv->connector.interlace_allowed = 1;
+	priv->encoder.possible_crtcs = crtcs;
 
-	ret = tda998x_create(client, &priv->base);
+	ret = tda998x_create(client, priv);
 	if (ret)
 		return ret;
 
 	if (!dev->of_node && params)
-		tda998x_encoder_set_config(&priv->base, params);
+		tda998x_encoder_set_config(priv, params);
 
-	tda998x_encoder_set_polling(&priv->base, &priv->base.connector);
+	tda998x_encoder_set_polling(priv, &priv->connector);
 
-	drm_encoder_helper_add(&priv->base.encoder, &tda998x_encoder_helper_funcs);
-	ret = drm_encoder_init(drm, &priv->base.encoder, &tda998x_encoder_funcs,
+	drm_encoder_helper_add(&priv->encoder, &tda998x_encoder_helper_funcs);
+	ret = drm_encoder_init(drm, &priv->encoder, &tda998x_encoder_funcs,
 			       DRM_MODE_ENCODER_TMDS);
 	if (ret)
 		goto err_encoder;
 
-	drm_connector_helper_add(&priv->base.connector,
+	drm_connector_helper_add(&priv->connector,
 				 &tda998x_connector_helper_funcs);
-	ret = drm_connector_init(drm, &priv->base.connector,
+	ret = drm_connector_init(drm, &priv->connector,
 				 &tda998x_connector_funcs,
 				 DRM_MODE_CONNECTOR_HDMIA);
 	if (ret)
 		goto err_connector;
 
-	ret = drm_connector_register(&priv->base.connector);
+	ret = drm_connector_register(&priv->connector);
 	if (ret)
 		goto err_sysfs;
 
-	priv->base.connector.encoder = &priv->base.encoder;
-	drm_mode_connector_attach_encoder(&priv->base.connector, &priv->base.encoder);
+	priv->connector.encoder = &priv->encoder;
+	drm_mode_connector_attach_encoder(&priv->connector, &priv->encoder);
 
 	return 0;
 
 err_sysfs:
-	drm_connector_cleanup(&priv->base.connector);
+	drm_connector_cleanup(&priv->connector);
 err_connector:
-	drm_encoder_cleanup(&priv->base.encoder);
+	drm_encoder_cleanup(&priv->encoder);
 err_encoder:
-	tda998x_destroy(&priv->base);
+	tda998x_destroy(priv);
 	return ret;
 }
 
 static void tda998x_unbind(struct device *dev, struct device *master,
 			   void *data)
 {
-	struct tda998x_priv2 *priv = dev_get_drvdata(dev);
+	struct tda998x_priv *priv = dev_get_drvdata(dev);
 
-	drm_connector_cleanup(&priv->base.connector);
-	drm_encoder_cleanup(&priv->base.encoder);
-	tda998x_destroy(&priv->base);
+	drm_connector_cleanup(&priv->connector);
+	drm_encoder_cleanup(&priv->encoder);
+	tda998x_destroy(priv);
 }
 
 static const struct component_ops tda998x_ops = {
-- 
cgit v0.10.2


From 9525c4dd923f8ffe38818f86cee523a5f7b19617 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Fri, 14 Aug 2015 11:28:53 +0100
Subject: drm/i2c: tda998x: clean up after struct tda998x_priv2 removal

We can now kill a number of glue functions which were sitting between
the common tda998x code and the drm encoder/connector methods.  This
results in slightly cleaner code.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/i2c/tda998x_drv.c b/drivers/gpu/drm/i2c/tda998x_drv.c
index 8bca9155..896b6aa 100644
--- a/drivers/gpu/drm/i2c/tda998x_drv.c
+++ b/drivers/gpu/drm/i2c/tda998x_drv.c
@@ -54,6 +54,12 @@ struct tda998x_priv {
 	struct drm_connector connector;
 };
 
+#define conn_to_tda998x_priv(x) \
+	container_of(x, struct tda998x_priv, connector)
+
+#define enc_to_tda998x_priv(x) \
+	container_of(x, struct tda998x_priv, encoder)
+
 /* The TDA9988 series of devices use a paged register scheme.. to simplify
  * things we encode the page # in upper bits of the register #.  To read/
  * write a given register, we need to make sure CURPAGE register is set
@@ -562,7 +568,7 @@ tda998x_reset(struct tda998x_priv *priv)
  * trying to read EDID data.
  *
  * However, tda998x_encoder_get_modes() may be called at any moment
- * after tda998x_encoder_detect() indicates that we are connected, so
+ * after tda998x_connector_detect() indicates that we are connected, so
  * we need to delay probing modes in tda998x_encoder_get_modes() after
  * we have seen a HPD inactive->active transition.  This code implements
  * that delay.
@@ -816,8 +822,10 @@ static void tda998x_encoder_set_config(struct tda998x_priv *priv,
 	priv->params = *p;
 }
 
-static void tda998x_encoder_dpms(struct tda998x_priv *priv, int mode)
+static void tda998x_encoder_dpms(struct drm_encoder *encoder, int mode)
 {
+	struct tda998x_priv *priv = enc_to_tda998x_priv(encoder);
+
 	/* we only care about on or off: */
 	if (mode != DRM_MODE_DPMS_ON)
 		mode = DRM_MODE_DPMS_OFF;
@@ -867,8 +875,8 @@ tda998x_encoder_mode_fixup(struct drm_encoder *encoder,
 	return true;
 }
 
-static int tda998x_encoder_mode_valid(struct tda998x_priv *priv,
-				      struct drm_display_mode *mode)
+static int tda998x_connector_mode_valid(struct drm_connector *connector,
+					struct drm_display_mode *mode)
 {
 	if (mode->clock > 150000)
 		return MODE_CLOCK_HIGH;
@@ -880,10 +888,11 @@ static int tda998x_encoder_mode_valid(struct tda998x_priv *priv,
 }
 
 static void
-tda998x_encoder_mode_set(struct tda998x_priv *priv,
+tda998x_encoder_mode_set(struct drm_encoder *encoder,
 			 struct drm_display_mode *mode,
 			 struct drm_display_mode *adjusted_mode)
 {
+	struct tda998x_priv *priv = enc_to_tda998x_priv(encoder);
 	u16 ref_pix, ref_line, n_pix, n_line;
 	u16 hs_pix_s, hs_pix_e;
 	u16 vs1_pix_s, vs1_pix_e, vs1_line_s, vs1_line_e;
@@ -1071,8 +1080,9 @@ tda998x_encoder_mode_set(struct tda998x_priv *priv,
 }
 
 static enum drm_connector_status
-tda998x_encoder_detect(struct tda998x_priv *priv)
+tda998x_connector_detect(struct drm_connector *connector, bool force)
 {
+	struct tda998x_priv *priv = conn_to_tda998x_priv(connector);
 	u8 val = cec_read(priv, REG_CEC_RXSHPDLEV);
 
 	return (val & CEC_RXSHPDLEV_HPD) ? connector_status_connected :
@@ -1135,10 +1145,9 @@ static int read_edid_block(void *data, u8 *buf, unsigned int blk, size_t length)
 	return 0;
 }
 
-static int
-tda998x_encoder_get_modes(struct tda998x_priv *priv,
-			  struct drm_connector *connector)
+static int tda998x_connector_get_modes(struct drm_connector *connector)
 {
+	struct tda998x_priv *priv = conn_to_tda998x_priv(connector);
 	struct edid *edid;
 	int n;
 
@@ -1330,46 +1339,24 @@ fail:
 	return -ENXIO;
 }
 
-#define conn_to_tda998x_priv(x) \
-	container_of(x, struct tda998x_priv, connector);
-
-#define enc_to_tda998x_priv(x) \
-	container_of(x, struct tda998x_priv, encoder);
-
-static void tda998x_encoder2_dpms(struct drm_encoder *encoder, int mode)
-{
-	struct tda998x_priv *priv = enc_to_tda998x_priv(encoder);
-
-	tda998x_encoder_dpms(priv, mode);
-}
-
 static void tda998x_encoder_prepare(struct drm_encoder *encoder)
 {
-	tda998x_encoder2_dpms(encoder, DRM_MODE_DPMS_OFF);
+	tda998x_encoder_dpms(encoder, DRM_MODE_DPMS_OFF);
 }
 
 static void tda998x_encoder_commit(struct drm_encoder *encoder)
 {
-	tda998x_encoder2_dpms(encoder, DRM_MODE_DPMS_ON);
-}
-
-static void tda998x_encoder2_mode_set(struct drm_encoder *encoder,
-				      struct drm_display_mode *mode,
-				      struct drm_display_mode *adjusted_mode)
-{
-	struct tda998x_priv *priv = enc_to_tda998x_priv(encoder);
-
-	tda998x_encoder_mode_set(priv, mode, adjusted_mode);
+	tda998x_encoder_dpms(encoder, DRM_MODE_DPMS_ON);
 }
 
 static const struct drm_encoder_helper_funcs tda998x_encoder_helper_funcs = {
-	.dpms = tda998x_encoder2_dpms,
+	.dpms = tda998x_encoder_dpms,
 	.save = tda998x_encoder_save,
 	.restore = tda998x_encoder_restore,
 	.mode_fixup = tda998x_encoder_mode_fixup,
 	.prepare = tda998x_encoder_prepare,
 	.commit = tda998x_encoder_commit,
-	.mode_set = tda998x_encoder2_mode_set,
+	.mode_set = tda998x_encoder_mode_set,
 };
 
 static void tda998x_encoder_destroy(struct drm_encoder *encoder)
@@ -1384,21 +1371,6 @@ static const struct drm_encoder_funcs tda998x_encoder_funcs = {
 	.destroy = tda998x_encoder_destroy,
 };
 
-static int tda998x_connector_get_modes(struct drm_connector *connector)
-{
-	struct tda998x_priv *priv = conn_to_tda998x_priv(connector);
-
-	return tda998x_encoder_get_modes(priv, connector);
-}
-
-static int tda998x_connector_mode_valid(struct drm_connector *connector,
-					struct drm_display_mode *mode)
-{
-	struct tda998x_priv *priv = conn_to_tda998x_priv(connector);
-
-	return tda998x_encoder_mode_valid(priv, mode);
-}
-
 static struct drm_encoder *
 tda998x_connector_best_encoder(struct drm_connector *connector)
 {
@@ -1414,14 +1386,6 @@ const struct drm_connector_helper_funcs tda998x_connector_helper_funcs = {
 	.best_encoder = tda998x_connector_best_encoder,
 };
 
-static enum drm_connector_status
-tda998x_connector_detect(struct drm_connector *connector, bool force)
-{
-	struct tda998x_priv *priv = conn_to_tda998x_priv(connector);
-
-	return tda998x_encoder_detect(priv);
-}
-
 static void tda998x_connector_destroy(struct drm_connector *connector)
 {
 	drm_connector_unregister(connector);
-- 
cgit v0.10.2


From 6e3f797c9ae9bd381247e90907838da43f695306 Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Mon, 28 Sep 2015 21:46:35 +0200
Subject: drm/doc: Update docs about device instance setup

->load is deprecated, bus functions are deprecated and everyone
should use drm_dev_alloc&register.

So update the .tmpl (and pull a bunch of the overview docs into the
sourcecode to increase chances that it'll stay in sync in the future)
and add notes to functions which are deprecated. I didn't bother to
clean up and document the unload sequence similarly since that one is
still a bit a mess: drm_dev_unregister does way too much,
drm_unplug_dev does what _unregister should be doing but then has the
complication of promising something it doesn't actually do (it doesn't
unplug existing open fds for instance, only prevents new ones).

Motivated since I don't want to hunt every new driver for usage of
drm_platform_init any more ;-)

v2: Reword the deprecation note for ->load a bit, using Laurent's
suggestion as an example (but making the wording a bit stronger even).
Fix spelling in commit message.

v3: More spelling fixes from Laurent.

Cc: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Cc: David Herrmann <dh.herrmann@gmail.com>
Acked-by: David Herrmann <dh.herrmann@gmail.com> (v2)
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/Documentation/DocBook/drm.tmpl b/Documentation/DocBook/drm.tmpl
index 30401f9..11f5bfb 100644
--- a/Documentation/DocBook/drm.tmpl
+++ b/Documentation/DocBook/drm.tmpl
@@ -138,14 +138,10 @@
     <para>
       At the core of every DRM driver is a <structname>drm_driver</structname>
       structure. Drivers typically statically initialize a drm_driver structure,
-      and then pass it to one of the <function>drm_*_init()</function> functions
-      to register it with the DRM subsystem.
-    </para>
-    <para>
-      Newer drivers that no longer require a <structname>drm_bus</structname>
-      structure can alternatively use the low-level device initialization and
-      registration functions such as <function>drm_dev_alloc()</function> and
-      <function>drm_dev_register()</function> directly.
+      and then pass it to <function>drm_dev_alloc()</function> to allocate a
+      device instance. After the device instance is fully initialized it can be
+      registered (which makes it accessible from userspace) using
+      <function>drm_dev_register()</function>.
     </para>
     <para>
       The <structname>drm_driver</structname> structure contains static
@@ -296,83 +292,12 @@ char *date;</synopsis>
       </sect3>
     </sect2>
     <sect2>
-      <title>Device Registration</title>
-      <para>
-        A number of functions are provided to help with device registration.
-        The functions deal with PCI and platform devices, respectively.
-      </para>
-!Edrivers/gpu/drm/drm_pci.c
-!Edrivers/gpu/drm/drm_platform.c
-      <para>
-        New drivers that no longer rely on the services provided by the
-        <structname>drm_bus</structname> structure can call the low-level
-        device registration functions directly. The
-        <function>drm_dev_alloc()</function> function can be used to allocate
-        and initialize a new <structname>drm_device</structname> structure.
-        Drivers will typically want to perform some additional setup on this
-        structure, such as allocating driver-specific data and storing a
-        pointer to it in the DRM device's <structfield>dev_private</structfield>
-        field. Drivers should also set the device's unique name using the
-        <function>drm_dev_set_unique()</function> function. After it has been
-        set up a device can be registered with the DRM subsystem by calling
-        <function>drm_dev_register()</function>. This will cause the device to
-        be exposed to userspace and will call the driver's
-        <structfield>.load()</structfield> implementation. When a device is
-        removed, the DRM device can safely be unregistered and freed by calling
-        <function>drm_dev_unregister()</function> followed by a call to
-        <function>drm_dev_unref()</function>.
-      </para>
+      <title>Device Instance and Driver Handling</title>
+!Pdrivers/gpu/drm/drm_drv.c driver instance overview
 !Edrivers/gpu/drm/drm_drv.c
     </sect2>
     <sect2>
       <title>Driver Load</title>
-      <para>
-        The <methodname>load</methodname> method is the driver and device
-        initialization entry point. The method is responsible for allocating and
-	initializing driver private data, performing resource allocation and
-	mapping (e.g. acquiring
-        clocks, mapping registers or allocating command buffers), initializing
-        the memory manager (<xref linkend="drm-memory-management"/>), installing
-        the IRQ handler (<xref linkend="drm-irq-registration"/>), setting up
-        vertical blanking handling (<xref linkend="drm-vertical-blank"/>), mode
-	setting (<xref linkend="drm-mode-setting"/>) and initial output
-	configuration (<xref linkend="drm-kms-init"/>).
-      </para>
-      <note><para>
-        If compatibility is a concern (e.g. with drivers converted over from
-        User Mode Setting to Kernel Mode Setting), care must be taken to prevent
-        device initialization and control that is incompatible with currently
-        active userspace drivers. For instance, if user level mode setting
-        drivers are in use, it would be problematic to perform output discovery
-        &amp; configuration at load time. Likewise, if user-level drivers
-        unaware of memory management are in use, memory management and command
-        buffer setup may need to be omitted. These requirements are
-        driver-specific, and care needs to be taken to keep both old and new
-        applications and libraries working.
-      </para></note>
-      <synopsis>int (*load) (struct drm_device *, unsigned long flags);</synopsis>
-      <para>
-        The method takes two arguments, a pointer to the newly created
-	<structname>drm_device</structname> and flags. The flags are used to
-	pass the <structfield>driver_data</structfield> field of the device id
-	corresponding to the device passed to <function>drm_*_init()</function>.
-	Only PCI devices currently use this, USB and platform DRM drivers have
-	their <methodname>load</methodname> method called with flags to 0.
-      </para>
-      <sect3>
-        <title>Driver Private Data</title>
-        <para>
-          The driver private hangs off the main
-          <structname>drm_device</structname> structure and can be used for
-          tracking various device-specific bits of information, like register
-          offsets, command buffer status, register state for suspend/resume, etc.
-          At load time, a driver may simply allocate one and set
-          <structname>drm_device</structname>.<structfield>dev_priv</structfield>
-          appropriately; it should be freed and
-          <structname>drm_device</structname>.<structfield>dev_priv</structfield>
-          set to NULL when the driver is unloaded.
-        </para>
-      </sect3>
       <sect3 id="drm-irq-registration">
         <title>IRQ Registration</title>
         <para>
@@ -465,6 +390,18 @@ char *date;</synopsis>
         </para>
       </sect3>
     </sect2>
+    <sect2>
+      <title>Bus-specific Device Registration and PCI Support</title>
+      <para>
+        A number of functions are provided to help with device registration.
+	The functions deal with PCI and platform devices respectively and are
+	only provided for historical reasons. These are all deprecated and
+	shouldn't be used in new drivers. Besides that there's a few
+	helpers for pci drivers.
+      </para>
+!Edrivers/gpu/drm/drm_pci.c
+!Edrivers/gpu/drm/drm_platform.c
+    </sect2>
   </sect1>
 
   <!-- Internals: memory management -->
diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c
index 9ad823f..d01f8d6 100644
--- a/drivers/gpu/drm/drm_drv.c
+++ b/drivers/gpu/drm/drm_drv.c
@@ -397,15 +397,51 @@ void drm_minor_release(struct drm_minor *minor)
 }
 
 /**
+ * DOC: driver instance overview
+ *
+ * A device instance for a drm driver is represented by struct &drm_device. This
+ * is allocated with drm_dev_alloc(), usually from bus-specific ->probe()
+ * callbacks implemented by the driver. The driver then needs to initialize all
+ * the various subsystems for the drm device like memory management, vblank
+ * handling, modesetting support and intial output configuration plus obviously
+ * initialize all the corresponding hardware bits. An important part of this is
+ * also calling drm_dev_set_unique() to set the userspace-visible unique name of
+ * this device instance. Finally when everything is up and running and ready for
+ * userspace the device instance can be published using drm_dev_register().
+ *
+ * There is also deprecated support for initalizing device instances using
+ * bus-specific helpers and the ->load() callback. But due to
+ * backwards-compatibility needs the device instance have to be published too
+ * early, which requires unpretty global locking to make safe and is therefore
+ * only support for existing drivers not yet converted to the new scheme.
+ *
+ * When cleaning up a device instance everything needs to be done in reverse:
+ * First unpublish the device instance with drm_dev_unregister(). Then clean up
+ * any other resources allocated at device initialization and drop the driver's
+ * reference to &drm_device using drm_dev_unref().
+ *
+ * Note that the lifetime rules for &drm_device instance has still a lot of
+ * historical baggage. Hence use the reference counting provided by
+ * drm_dev_ref() and drm_dev_unref() only carefully.
+ *
+ * Also note that embedding of &drm_device is currently not (yet) supported (but
+ * it would be easy to add). Drivers can store driver-private data in the
+ * dev_priv field of &drm_device.
+ */
+
+/**
  * drm_put_dev - Unregister and release a DRM device
  * @dev: DRM device
  *
  * Called at module unload time or when a PCI device is unplugged.
  *
- * Use of this function is discouraged. It will eventually go away completely.
- * Please use drm_dev_unregister() and drm_dev_unref() explicitly instead.
- *
  * Cleans up all DRM device, calling drm_lastclose().
+ *
+ * Note: Use of this function is deprecated. It will eventually go away
+ * completely.  Please use drm_dev_unregister() and drm_dev_unref() explicitly
+ * instead to make sure that the device isn't userspace accessible any more
+ * while teardown is in progress, ensuring that userspace can't access an
+ * inconsistent state.
  */
 void drm_put_dev(struct drm_device *dev)
 {
@@ -518,7 +554,9 @@ static void drm_fs_inode_free(struct inode *inode)
  *
  * Allocate and initialize a new DRM device. No device registration is done.
  * Call drm_dev_register() to advertice the device to user space and register it
- * with other core subsystems.
+ * with other core subsystems. This should be done last in the device
+ * initialization sequence to make sure userspace can't access an inconsistent
+ * state.
  *
  * The initial ref-count of the object is 1. Use drm_dev_ref() and
  * drm_dev_unref() to take and drop further ref-counts.
@@ -673,6 +711,12 @@ EXPORT_SYMBOL(drm_dev_unref);
  *
  * Never call this twice on any device!
  *
+ * NOTE: To ensure backward compatibility with existing drivers method this
+ * function calls the ->load() method after registering the device nodes,
+ * creating race conditions. Usage of the ->load() methods is therefore
+ * deprecated, drivers must perform all initialization before calling
+ * drm_dev_register().
+ *
  * RETURNS:
  * 0 on success, negative error code on failure.
  */
@@ -720,6 +764,9 @@ EXPORT_SYMBOL(drm_dev_register);
  * Unregister the DRM device from the system. This does the reverse of
  * drm_dev_register() but does not deallocate the device. The caller must call
  * drm_dev_unref() to drop their final reference.
+ *
+ * This should be called first in the device teardown code to make sure
+ * userspace can't access the device instance any more.
  */
 void drm_dev_unregister(struct drm_device *dev)
 {
diff --git a/drivers/gpu/drm/drm_pci.c b/drivers/gpu/drm/drm_pci.c
index 1b1bd42..fcd2a86 100644
--- a/drivers/gpu/drm/drm_pci.c
+++ b/drivers/gpu/drm/drm_pci.c
@@ -266,6 +266,9 @@ void drm_pci_agp_destroy(struct drm_device *dev)
  * then register the character device and inter module information.
  * Try and register, if we fail to register, backout previous work.
  *
+ * NOTE: This function is deprecated, please use drm_dev_alloc() and
+ * drm_dev_register() instead and remove your ->load() callback.
+ *
  * Return: 0 on success or a negative error code on failure.
  */
 int drm_get_pci_dev(struct pci_dev *pdev, const struct pci_device_id *ent,
@@ -326,6 +329,10 @@ EXPORT_SYMBOL(drm_get_pci_dev);
  * Initializes a drm_device structures, registering the stubs and initializing
  * the AGP device.
  *
+ * NOTE: This function is deprecated. Modern modesetting drm drivers should use
+ * pci_register_driver() directly, this function only provides shadow-binding
+ * support for old legacy drivers on top of that core pci function.
+ *
  * Return: 0 on success or a negative error code on failure.
  */
 int drm_pci_init(struct drm_driver *driver, struct pci_driver *pdriver)
@@ -435,6 +442,10 @@ EXPORT_SYMBOL(drm_pci_init);
  *
  * Unregisters one or more devices matched by a PCI driver from the DRM
  * subsystem.
+ *
+ * NOTE: This function is deprecated. Modern modesetting drm drivers should use
+ * pci_unregister_driver() directly, this function only provides shadow-binding
+ * support for old legacy drivers on top of that core pci function.
  */
 void drm_pci_exit(struct drm_driver *driver, struct pci_driver *pdriver)
 {
diff --git a/drivers/gpu/drm/drm_platform.c b/drivers/gpu/drm/drm_platform.c
index 5314c9d..644169e 100644
--- a/drivers/gpu/drm/drm_platform.c
+++ b/drivers/gpu/drm/drm_platform.c
@@ -95,6 +95,9 @@ EXPORT_SYMBOL(drm_platform_set_busid);
  * subsystem, initializing a drm_device structure and calling the driver's
  * .load() function.
  *
+ * NOTE: This function is deprecated, please use drm_dev_alloc() and
+ * drm_dev_register() instead and remove your ->load() callback.
+ *
  * Return: 0 on success or a negative error code on failure.
  */
 int drm_platform_init(struct drm_driver *driver, struct platform_device *platform_device)
-- 
cgit v0.10.2


From 30c964a6cb7bbade28eabcbc3fce4b01be8f1a39 Mon Sep 17 00:00:00 2001
From: Robert Beckett <robert.beckett@intel.com>
Date: Fri, 28 Aug 2015 13:10:22 +0100
Subject: drm/i915: Detect virtual south bridge

Virtualized systems often use a virtual P2X4 south bridge.
Detect this in intel_detect_pch and make a best guess as to which PCH
we should be using.

This was seen on vmware esxi hypervisor. When passing the graphics device
through to a guest, it can not pass through the PCH. Instead it simulates
a P2X4 southbridge.

Signed-off-by: Robert Beckett <robert.beckett@intel.com>
Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index e6d7a69..b31485a 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -443,6 +443,34 @@ static const struct pci_device_id pciidlist[] = {		/* aka */
 
 MODULE_DEVICE_TABLE(pci, pciidlist);
 
+static enum intel_pch intel_virt_detect_pch(struct drm_device *dev)
+{
+	enum intel_pch ret = PCH_NOP;
+
+	/*
+	 * In a virtualized passthrough environment we can be in a
+	 * setup where the ISA bridge is not able to be passed through.
+	 * In this case, a south bridge can be emulated and we have to
+	 * make an educated guess as to which PCH is really there.
+	 */
+
+	if (IS_GEN5(dev)) {
+		ret = PCH_IBX;
+		DRM_DEBUG_KMS("Assuming Ibex Peak PCH\n");
+	} else if (IS_GEN6(dev) || IS_IVYBRIDGE(dev)) {
+		ret = PCH_CPT;
+		DRM_DEBUG_KMS("Assuming CouarPoint PCH\n");
+	} else if (IS_HASWELL(dev) || IS_BROADWELL(dev)) {
+		ret = PCH_LPT;
+		DRM_DEBUG_KMS("Assuming LynxPoint PCH\n");
+	} else if (IS_SKYLAKE(dev)) {
+		ret = PCH_SPT;
+		DRM_DEBUG_KMS("Assuming SunrisePoint PCH\n");
+	}
+
+	return ret;
+}
+
 void intel_detect_pch(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -503,6 +531,8 @@ void intel_detect_pch(struct drm_device *dev)
 				dev_priv->pch_type = PCH_SPT;
 				DRM_DEBUG_KMS("Found SunrisePoint LP PCH\n");
 				WARN_ON(!IS_SKYLAKE(dev));
+			} else if (id == INTEL_PCH_P2X_DEVICE_ID_TYPE) {
+				dev_priv->pch_type = intel_virt_detect_pch(dev);
 			} else
 				continue;
 
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 7d83f67..04d710f 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -2608,6 +2608,7 @@ struct drm_i915_cmd_table {
 #define INTEL_PCH_LPT_LP_DEVICE_ID_TYPE		0x9c00
 #define INTEL_PCH_SPT_DEVICE_ID_TYPE		0xA100
 #define INTEL_PCH_SPT_LP_DEVICE_ID_TYPE		0x9D00
+#define INTEL_PCH_P2X_DEVICE_ID_TYPE		0x7100
 
 #define INTEL_PCH_TYPE(dev) (__I915__(dev)->pch_type)
 #define HAS_PCH_SPT(dev) (INTEL_PCH_TYPE(dev) == PCH_SPT)
-- 
cgit v0.10.2


From 4b7ab5fca1976a53d59ef1558ddb06ba69fab356 Mon Sep 17 00:00:00 2001
From: Animesh Manna <animesh.manna@intel.com>
Date: Wed, 26 Aug 2015 01:36:05 +0530
Subject: drm/i915/skl: Added a check for the hardware status of csr fw before
 loading.

Dmc will restore the csr program except DC9, cold boot,
warm reset, PCI function level reset, and hibernate/suspend.

intel_csr_load_program() function is used to load the firmware
data from kernel memory to csr address space.

All values of csr address space will be zero if it got reset and
the first byte of csr program is always a non-zero if firmware
is loaded successfuly. Based on hardware status will load the
firmware.

Without this condition check if we overwrite the firmware data the
counters exposed for dc5/dc6 (help for debugging) will be nullified.

Note: Above commit message seems to be confused and the real problem
apparently going on is that for suspend-to-idle and system standby the
firmware survives (it's like runtime pm), but it doesn't for
suspend-to-mem and hibernate-to-disk. All the other talking about DC9
and pci reset are irrelevant for the path touched here (only driver
load and system resume functions) and might not be true statements.

v1: Initial version.

v2: Based on review comments from Daniel,
- Added a check to know hardware status and load the firmware if not loaded.

Cc: Daniel Vetter <daniel.vetter@intel.com>
Cc: Damien Lespiau <damien.lespiau@intel.com>
Cc: Imre Deak <imre.deak@intel.com>
Cc: Sunil Kamath <sunil.kamath@intel.com>
Signed-off-by: Animesh Manna <animesh.manna@intel.com>
Signed-off-by: Vathsala Nagaraju <vathsala.nagaraju@intel.com>
Reviewed-by: A.Sunil Kamath <sunil.kamath@intel.com>
[danvet: Explain clearly in the code comment when we need to reload
and when not and make it a FIXME. Also deconfuse the commit message
with a note. Also: Make. it. compile.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_csr.c b/drivers/gpu/drm/i915/intel_csr.c
index 3427dd4..9e530a7 100644
--- a/drivers/gpu/drm/i915/intel_csr.c
+++ b/drivers/gpu/drm/i915/intel_csr.c
@@ -265,6 +265,15 @@ void intel_csr_load_program(struct drm_device *dev)
 		return;
 	}
 
+	/*
+	 * FIXME: Firmware gets lost on S3/S4, but not when entering system
+	 * standby or suspend-to-idle (which is just like forced runtime pm).
+	 * Unfortunately the ACPI subsystem doesn't yet give us a way to
+	 * differentiate this, hence figure it out with this hack.
+	 */
+	if (I915_READ(CSR_PROGRAM(0)))
+		return;
+
 	mutex_lock(&dev_priv->csr_lock);
 	fw_size = dev_priv->csr.dmc_fw_size;
 	for (i = 0; i < fw_size; i++)
-- 
cgit v0.10.2


From c268444a2cecabc0ab567ca275662d80fa0ac813 Mon Sep 17 00:00:00 2001
From: Animesh Manna <animesh.manna@intel.com>
Date: Wed, 26 Aug 2015 01:36:06 +0530
Subject: drm/i915/skl Remove the call for csr uninitialization from suspend
 path

This patch remove the function call to set the firmware
loading status as uninitialized during suspend.

Dmc firmware will restore the firmware in normal suspend. In previous
patch added a check to directly read the hardware status and load
the firmware if got reset during resume from suspend-hibernation.

Cc: Daniel Vetter <daniel.vetter@intel.com>
Signed-off-by: Animesh Manna <animesh.manna@intel.com>
Signed-off-by: Vathsala Nagaraju <vathsala.nagaraju@intel.com>
Reviewed-by: A.Sunil Kamath <sunil.kamath@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index b31485a..1cb6b82 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -1051,12 +1051,6 @@ static int skl_suspend_complete(struct drm_i915_private *dev_priv)
 {
 	/* Enabling DC6 is not a hard requirement to enter runtime D3 */
 
-	/*
-	 * This is to ensure that CSR isn't identified as loaded before
-	 * CSR-loading program is called during runtime-resume.
-	 */
-	intel_csr_load_status_set(dev_priv, FW_UNINITIALIZED);
-
 	skl_uninit_cdclk(dev_priv);
 
 	return 0;
-- 
cgit v0.10.2


From 4e961e426cdeb5c9f27d65fb0afb0010fcecfeae Mon Sep 17 00:00:00 2001
From: Animesh Manna <animesh.manna@intel.com>
Date: Wed, 26 Aug 2015 01:36:08 +0530
Subject: drm/i915/skl: Do not disable cdclk PLL if csr firmware is present

While display engine entering into low power state no need to disable
cdclk pll as CSR firmware of dmc will take care. If pll is already
enabled firmware execution sequence will be blocked. This is one
of the criteria for dmc to work properly.

v1: Initial version.

v2: Based on review comment from Daniel added code commnent.

Cc: Daniel Vetter <daniel.vetter@intel.com>
Cc: Damien Lespiau <damien.lespiau@intel.com>
Cc: Imre Deak <imre.deak@intel.com>
Cc: Sunil Kamath <sunil.kamath@intel.com>
Signed-off-by: Animesh Manna <animesh.manna@intel.com>
Signed-off-bt: Vathsala Nagaraju <vathsala.nagaraju@intel.com>
Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@intel.com>
Reviewed-by: A.Sunil Kamath <sunil.kamath@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 1847257..5d8645e 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -5709,10 +5709,16 @@ void skl_uninit_cdclk(struct drm_i915_private *dev_priv)
 	if (I915_READ(DBUF_CTL) & DBUF_POWER_STATE)
 		DRM_ERROR("DBuf power disable timeout\n");
 
-	/* disable DPLL0 */
-	I915_WRITE(LCPLL1_CTL, I915_READ(LCPLL1_CTL) & ~LCPLL_PLL_ENABLE);
-	if (wait_for(!(I915_READ(LCPLL1_CTL) & LCPLL_PLL_LOCK), 1))
-		DRM_ERROR("Couldn't disable DPLL0\n");
+	/*
+	 * DMC assumes ownership of LCPLL and will get confused if we touch it.
+	 */
+	if (dev_priv->csr.dmc_payload) {
+		/* disable DPLL0 */
+		I915_WRITE(LCPLL1_CTL, I915_READ(LCPLL1_CTL) &
+					~LCPLL_PLL_ENABLE);
+		if (wait_for(!(I915_READ(LCPLL1_CTL) & LCPLL_PLL_LOCK), 1))
+			DRM_ERROR("Couldn't disable DPLL0\n");
+	}
 
 	intel_display_power_put(dev_priv, POWER_DOMAIN_PLLS);
 }
-- 
cgit v0.10.2


From 08aef7caa14f1c0b7c5d79d61d279bcedc188ab9 Mon Sep 17 00:00:00 2001
From: Animesh Manna <animesh.manna@intel.com>
Date: Wed, 26 Aug 2015 01:36:09 +0530
Subject: drm/i915/skl: Block disable call for pw1 if dmc firmware is present.

Another interesting criteria to work dmc as expected is pw1 to be
enabled by driver and dmc will shut it off in its execution
sequence. If already disabled by driver dmc will get confuse and
behave differently than expected found during pc10 entry issue
for skl.

So berfore we disable power-well 1, added check if dmc firmware is
present and driver will not disable power well 1, but for any reason
if firmware is not present of failed to load we can shut off the
power well 1 which will save some power.

As skl is currently fully dependent on dmc to go in lowest possible
power state (dc6) but the same is not applicable for bxt. Display
engine can enter into dc9 without dmc, hence unblocking disable call.

v1: Initial version.

v2: Rebased as per current patch series.

Cc: Daniel Vetter <daniel.vetter@intel.com>
Cc: Damien Lespiau <damien.lespiau@intel.com>
Cc: Imre Deak <imre.deak@intel.com>
Cc: Sunil Kamath <sunil.kamath@intel.com>
Signed-off-by: Animesh Manna <animesh.manna@intel.com>
Signed-off-by: Vathsala Nagaraju <vathsala.nagaraju@intel.com>
Reviewed-by: A.Sunil Kamath <sunil.kamath@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c
index 85c35fd..4a815bb 100644
--- a/drivers/gpu/drm/i915/intel_runtime_pm.c
+++ b/drivers/gpu/drm/i915/intel_runtime_pm.c
@@ -656,9 +656,15 @@ static void skl_set_power_well(struct drm_i915_private *dev_priv,
 		}
 	} else {
 		if (enable_requested) {
-			I915_WRITE(HSW_PWR_WELL_DRIVER,	tmp & ~req_mask);
-			POSTING_READ(HSW_PWR_WELL_DRIVER);
-			DRM_DEBUG_KMS("Disabling %s\n", power_well->name);
+			if (IS_SKYLAKE(dev) &&
+				(power_well->data == SKL_DISP_PW_1) &&
+				(intel_csr_load_status_get(dev_priv) == FW_LOADED))
+				DRM_DEBUG_KMS("Not Disabling PW1, dmc will handle\n");
+			else {
+				I915_WRITE(HSW_PWR_WELL_DRIVER,	tmp & ~req_mask);
+				POSTING_READ(HSW_PWR_WELL_DRIVER);
+				DRM_DEBUG_KMS("Disabling %s\n", power_well->name);
+			}
 
 			if ((GEN9_ENABLE_DC5(dev) || SKL_ENABLE_DC6(dev)) &&
 				power_well->data == SKL_DISP_PW_2) {
-- 
cgit v0.10.2


From 28694070d849c98290288734ef67a3ef484425ec Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 8 Sep 2015 13:40:44 +0300
Subject: drm/i915: Use intel_panel for DVO fixed mode handling
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Replace intel_dvo->panel_fixed_mode with the appropriate intel_panel
stuff. Now all connectors that have a fixed mode use intel_panel.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dvo.c b/drivers/gpu/drm/i915/intel_dvo.c
index c80fe1f..0bc8aa8 100644
--- a/drivers/gpu/drm/i915/intel_dvo.c
+++ b/drivers/gpu/drm/i915/intel_dvo.c
@@ -97,7 +97,8 @@ struct intel_dvo {
 
 	struct intel_dvo_device dev;
 
-	struct drm_display_mode *panel_fixed_mode;
+	struct intel_connector *attached_connector;
+
 	bool panel_wants_dither;
 };
 
@@ -201,6 +202,8 @@ intel_dvo_mode_valid(struct drm_connector *connector,
 		     struct drm_display_mode *mode)
 {
 	struct intel_dvo *intel_dvo = intel_attached_dvo(connector);
+	const struct drm_display_mode *fixed_mode =
+		to_intel_connector(connector)->panel.fixed_mode;
 	int max_dotclk = to_i915(connector->dev)->max_dotclk_freq;
 	int target_clock = mode->clock;
 
@@ -209,13 +212,13 @@ intel_dvo_mode_valid(struct drm_connector *connector,
 
 	/* XXX: Validate clock range */
 
-	if (intel_dvo->panel_fixed_mode) {
-		if (mode->hdisplay > intel_dvo->panel_fixed_mode->hdisplay)
+	if (fixed_mode) {
+		if (mode->hdisplay > fixed_mode->hdisplay)
 			return MODE_PANEL;
-		if (mode->vdisplay > intel_dvo->panel_fixed_mode->vdisplay)
+		if (mode->vdisplay > fixed_mode->vdisplay)
 			return MODE_PANEL;
 
-		target_clock = intel_dvo->panel_fixed_mode->clock;
+		target_clock = fixed_mode->clock;
 	}
 
 	if (target_clock > max_dotclk)
@@ -228,6 +231,8 @@ static bool intel_dvo_compute_config(struct intel_encoder *encoder,
 				     struct intel_crtc_state *pipe_config)
 {
 	struct intel_dvo *intel_dvo = enc_to_dvo(encoder);
+	const struct drm_display_mode *fixed_mode =
+		intel_dvo->attached_connector->panel.fixed_mode;
 	struct drm_display_mode *adjusted_mode = &pipe_config->base.adjusted_mode;
 
 	/* If we have timings from the BIOS for the panel, put them in
@@ -235,21 +240,8 @@ static bool intel_dvo_compute_config(struct intel_encoder *encoder,
 	 * with the panel scaling set up to source from the H/VDisplay
 	 * of the original mode.
 	 */
-	if (intel_dvo->panel_fixed_mode != NULL) {
-#define C(x) adjusted_mode->x = intel_dvo->panel_fixed_mode->x
-		C(hdisplay);
-		C(hsync_start);
-		C(hsync_end);
-		C(htotal);
-		C(vdisplay);
-		C(vsync_start);
-		C(vsync_end);
-		C(vtotal);
-		C(clock);
-#undef C
-
-		drm_mode_set_crtcinfo(adjusted_mode, 0);
-	}
+	if (fixed_mode)
+		intel_fixed_panel_mode(fixed_mode, adjusted_mode);
 
 	return true;
 }
@@ -318,8 +310,9 @@ intel_dvo_detect(struct drm_connector *connector, bool force)
 
 static int intel_dvo_get_modes(struct drm_connector *connector)
 {
-	struct intel_dvo *intel_dvo = intel_attached_dvo(connector);
 	struct drm_i915_private *dev_priv = connector->dev->dev_private;
+	const struct drm_display_mode *fixed_mode =
+		to_intel_connector(connector)->panel.fixed_mode;
 
 	/* We should probably have an i2c driver get_modes function for those
 	 * devices which will have a fixed set of modes determined by the chip
@@ -331,9 +324,9 @@ static int intel_dvo_get_modes(struct drm_connector *connector)
 	if (!list_empty(&connector->probed_modes))
 		return 1;
 
-	if (intel_dvo->panel_fixed_mode != NULL) {
+	if (fixed_mode) {
 		struct drm_display_mode *mode;
-		mode = drm_mode_duplicate(connector->dev, intel_dvo->panel_fixed_mode);
+		mode = drm_mode_duplicate(connector->dev, fixed_mode);
 		if (mode) {
 			drm_mode_probed_add(connector, mode);
 			return 1;
@@ -346,6 +339,7 @@ static int intel_dvo_get_modes(struct drm_connector *connector)
 static void intel_dvo_destroy(struct drm_connector *connector)
 {
 	drm_connector_cleanup(connector);
+	intel_panel_fini(&to_intel_connector(connector)->panel);
 	kfree(connector);
 }
 
@@ -372,8 +366,6 @@ static void intel_dvo_enc_destroy(struct drm_encoder *encoder)
 	if (intel_dvo->dev.dev_ops->destroy)
 		intel_dvo->dev.dev_ops->destroy(&intel_dvo->dev);
 
-	kfree(intel_dvo->panel_fixed_mode);
-
 	intel_encoder_destroy(encoder);
 }
 
@@ -438,6 +430,8 @@ void intel_dvo_init(struct drm_device *dev)
 		return;
 	}
 
+	intel_dvo->attached_connector = intel_connector;
+
 	intel_encoder = &intel_dvo->base;
 	drm_encoder_init(dev, &intel_encoder->base,
 			 &intel_dvo_enc_funcs, encoder_type);
@@ -542,8 +536,9 @@ void intel_dvo_init(struct drm_device *dev)
 			 * headers, likely), so for now, just get the current
 			 * mode being output through DVO.
 			 */
-			intel_dvo->panel_fixed_mode =
-				intel_dvo_get_current_mode(connector);
+			intel_panel_init(&intel_connector->panel,
+					 intel_dvo_get_current_mode(connector),
+					 NULL);
 			intel_dvo->panel_wants_dither = true;
 		}
 
-- 
cgit v0.10.2


From 124abe076fd8b360dd7c651046cb8b204268efeb Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 8 Sep 2015 13:40:45 +0300
Subject: drm/i915: Always call the adjusted mode 'adjusted_mode'
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Always name any variable pointing at the adjusted mode as
'adjustead_mode'. This will make it much easier to identify
when we should use the crtc_ timings and when we shoudln't.

Conversion was performed with coccinelle:
@@
expression E;
identifier I;
@@
- struct drm_display_mode *I = &E.adjusted_mode;
+ struct drm_display_mode *adjusted_mode = &E.adjusted_mode;
<...
- I
+ adjusted_mode
...>

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Mika Kahola <mika.kahola@intel.com>
[danvet: Fixup conflicts.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_audio.c b/drivers/gpu/drm/i915/intel_audio.c
index e35997e..d9876c5 100644
--- a/drivers/gpu/drm/i915/intel_audio.c
+++ b/drivers/gpu/drm/i915/intel_audio.c
@@ -396,7 +396,7 @@ void intel_audio_codec_enable(struct intel_encoder *intel_encoder)
 {
 	struct drm_encoder *encoder = &intel_encoder->base;
 	struct intel_crtc *crtc = to_intel_crtc(encoder->crtc);
-	struct drm_display_mode *mode = &crtc->config->base.adjusted_mode;
+	struct drm_display_mode *adjusted_mode = &crtc->config->base.adjusted_mode;
 	struct drm_connector *connector;
 	struct drm_device *dev = encoder->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -419,10 +419,11 @@ void intel_audio_codec_enable(struct intel_encoder *intel_encoder)
 	if (intel_pipe_has_type(crtc, INTEL_OUTPUT_DISPLAYPORT))
 		connector->eld[5] |= (1 << 2);
 
-	connector->eld[6] = drm_av_sync_delay(connector, mode) / 2;
+	connector->eld[6] = drm_av_sync_delay(connector, adjusted_mode) / 2;
 
 	if (dev_priv->display.audio_codec_enable)
-		dev_priv->display.audio_codec_enable(connector, intel_encoder, mode);
+		dev_priv->display.audio_codec_enable(connector, intel_encoder,
+						     adjusted_mode);
 
 	if (acomp && acomp->audio_ops && acomp->audio_ops->pin_eld_notify)
 		acomp->audio_ops->pin_eld_notify(acomp->audio_ops->audio_ptr, (int) port);
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 5d8645e..45ec0b8 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -4392,8 +4392,7 @@ skl_update_scaler(struct intel_crtc_state *crtc_state, bool force_detach,
 int skl_update_scaler_crtc(struct intel_crtc_state *state)
 {
 	struct intel_crtc *intel_crtc = to_intel_crtc(state->base.crtc);
-	struct drm_display_mode *adjusted_mode =
-		&state->base.adjusted_mode;
+	struct drm_display_mode *adjusted_mode = &state->base.adjusted_mode;
 
 	DRM_DEBUG_KMS("Updating scaler for [CRTC:%i] scaler_user index %u.%u\n",
 		      intel_crtc->base.base.id, intel_crtc->pipe, SKL_CRTC_INDEX);
@@ -7625,8 +7624,7 @@ static void intel_set_pipe_timings(struct intel_crtc *intel_crtc)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	enum pipe pipe = intel_crtc->pipe;
 	enum transcoder cpu_transcoder = intel_crtc->config->cpu_transcoder;
-	struct drm_display_mode *adjusted_mode =
-		&intel_crtc->config->base.adjusted_mode;
+	struct drm_display_mode *adjusted_mode = &intel_crtc->config->base.adjusted_mode;
 	uint32_t crtc_vtotal, crtc_vblank_end;
 	int vsyncshift = 0;
 
@@ -12821,11 +12819,11 @@ static void update_scanline_offset(struct intel_crtc *crtc)
 	 * one to the value.
 	 */
 	if (IS_GEN2(dev)) {
-		const struct drm_display_mode *mode = &crtc->config->base.adjusted_mode;
+		const struct drm_display_mode *adjusted_mode = &crtc->config->base.adjusted_mode;
 		int vtotal;
 
-		vtotal = mode->crtc_vtotal;
-		if (mode->flags & DRM_MODE_FLAG_INTERLACE)
+		vtotal = adjusted_mode->crtc_vtotal;
+		if (adjusted_mode->flags & DRM_MODE_FLAG_INTERLACE)
 			vtotal /= 2;
 
 		crtc->scanline_offset = vtotal - 1;
diff --git a/drivers/gpu/drm/i915/intel_dsi.c b/drivers/gpu/drm/i915/intel_dsi.c
index 61d69c2..23b7fc5 100644
--- a/drivers/gpu/drm/i915/intel_dsi.c
+++ b/drivers/gpu/drm/i915/intel_dsi.c
@@ -774,8 +774,7 @@ static void intel_dsi_prepare(struct intel_encoder *intel_encoder)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc);
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder);
-	struct drm_display_mode *adjusted_mode =
-		&intel_crtc->config->base.adjusted_mode;
+	struct drm_display_mode *adjusted_mode = &intel_crtc->config->base.adjusted_mode;
 	enum port port;
 	unsigned int bpp = intel_crtc->config->pipe_bpp;
 	u32 val, tmp;
@@ -880,14 +879,14 @@ static void intel_dsi_prepare(struct intel_encoder *intel_encoder)
 			intel_dsi->video_mode_format == VIDEO_MODE_BURST) {
 			I915_WRITE(MIPI_HS_TX_TIMEOUT(port),
 				txbyteclkhs(adjusted_mode->htotal, bpp,
-					intel_dsi->lane_count,
-					intel_dsi->burst_mode_ratio) + 1);
+					    intel_dsi->lane_count,
+					    intel_dsi->burst_mode_ratio) + 1);
 		} else {
 			I915_WRITE(MIPI_HS_TX_TIMEOUT(port),
 				txbyteclkhs(adjusted_mode->vtotal *
-					adjusted_mode->htotal,
-					bpp, intel_dsi->lane_count,
-					intel_dsi->burst_mode_ratio) + 1);
+					    adjusted_mode->htotal,
+					    bpp, intel_dsi->lane_count,
+					    intel_dsi->burst_mode_ratio) + 1);
 		}
 		I915_WRITE(MIPI_LP_RX_TIMEOUT(port), intel_dsi->lp_rx_timeout);
 		I915_WRITE(MIPI_TURN_AROUND_TIMEOUT(port),
diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index bb33c66..c11703a 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -1537,8 +1537,7 @@ static void intel_hdmi_pre_enable(struct intel_encoder *encoder)
 {
 	struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(&encoder->base);
 	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->base.crtc);
-	struct drm_display_mode *adjusted_mode =
-		&intel_crtc->config->base.adjusted_mode;
+	struct drm_display_mode *adjusted_mode = &intel_crtc->config->base.adjusted_mode;
 
 	intel_hdmi_prepare(encoder);
 
@@ -1555,8 +1554,7 @@ static void vlv_hdmi_pre_enable(struct intel_encoder *encoder)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc =
 		to_intel_crtc(encoder->base.crtc);
-	struct drm_display_mode *adjusted_mode =
-		&intel_crtc->config->base.adjusted_mode;
+	struct drm_display_mode *adjusted_mode = &intel_crtc->config->base.adjusted_mode;
 	enum dpio_channel port = vlv_dport_to_channel(dport);
 	int pipe = intel_crtc->pipe;
 	u32 val;
@@ -1822,8 +1820,7 @@ static void chv_hdmi_pre_enable(struct intel_encoder *encoder)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc =
 		to_intel_crtc(encoder->base.crtc);
-	struct drm_display_mode *adjusted_mode =
-		&intel_crtc->config->base.adjusted_mode;
+	struct drm_display_mode *adjusted_mode = &intel_crtc->config->base.adjusted_mode;
 	enum dpio_channel ch = vlv_dport_to_channel(dport);
 	int pipe = intel_crtc->pipe;
 	int data, i, stagger;
diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c
index 2c2d1f0..5e70acf 100644
--- a/drivers/gpu/drm/i915/intel_lvds.c
+++ b/drivers/gpu/drm/i915/intel_lvds.c
@@ -139,8 +139,7 @@ static void intel_pre_enable_lvds(struct intel_encoder *encoder)
 	struct drm_device *dev = encoder->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *crtc = to_intel_crtc(encoder->base.crtc);
-	const struct drm_display_mode *adjusted_mode =
-		&crtc->config->base.adjusted_mode;
+	const struct drm_display_mode *adjusted_mode = &crtc->config->base.adjusted_mode;
 	int pipe = crtc->pipe;
 	u32 temp;
 
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index ab5ac5e..d5c60bb 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -1502,8 +1502,7 @@ static void i965_update_wm(struct drm_crtc *unused_crtc)
 	if (crtc) {
 		/* self-refresh has much higher latency */
 		static const int sr_latency_ns = 12000;
-		const struct drm_display_mode *adjusted_mode =
-			&to_intel_crtc(crtc)->config->base.adjusted_mode;
+		const struct drm_display_mode *adjusted_mode = &to_intel_crtc(crtc)->config->base.adjusted_mode;
 		int clock = adjusted_mode->crtc_clock;
 		int htotal = adjusted_mode->crtc_htotal;
 		int hdisplay = to_intel_crtc(crtc)->config->pipe_src_w;
@@ -1650,8 +1649,7 @@ static void i9xx_update_wm(struct drm_crtc *unused_crtc)
 	if (HAS_FW_BLC(dev) && enabled) {
 		/* self-refresh has much higher latency */
 		static const int sr_latency_ns = 6000;
-		const struct drm_display_mode *adjusted_mode =
-			&to_intel_crtc(enabled)->config->base.adjusted_mode;
+		const struct drm_display_mode *adjusted_mode = &to_intel_crtc(enabled)->config->base.adjusted_mode;
 		int clock = adjusted_mode->crtc_clock;
 		int htotal = adjusted_mode->crtc_htotal;
 		int hdisplay = to_intel_crtc(enabled)->config->pipe_src_w;
@@ -2088,7 +2086,7 @@ hsw_compute_linetime_wm(struct drm_device *dev, struct drm_crtc *crtc)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	struct drm_display_mode *mode = &intel_crtc->config->base.adjusted_mode;
+	struct drm_display_mode *adjusted_mode = &intel_crtc->config->base.adjusted_mode;
 	u32 linetime, ips_linetime;
 
 	if (!intel_crtc->active)
@@ -2097,9 +2095,9 @@ hsw_compute_linetime_wm(struct drm_device *dev, struct drm_crtc *crtc)
 	/* The WM are computed with base on how long it takes to fill a single
 	 * row at the given clock rate, multiplied by 8.
 	 * */
-	linetime = DIV_ROUND_CLOSEST(mode->crtc_htotal * 1000 * 8,
-				     mode->crtc_clock);
-	ips_linetime = DIV_ROUND_CLOSEST(mode->crtc_htotal * 1000 * 8,
+	linetime = DIV_ROUND_CLOSEST(adjusted_mode->crtc_htotal * 1000 * 8,
+				     adjusted_mode->crtc_clock);
+	ips_linetime = DIV_ROUND_CLOSEST(adjusted_mode->crtc_htotal * 1000 * 8,
 					 dev_priv->cdclk_freq);
 
 	return PIPE_WM_LINETIME_IPS_LINETIME(ips_linetime) |
diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c
index 05521b5..b0b96fd 100644
--- a/drivers/gpu/drm/i915/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/intel_sdvo.c
@@ -1189,8 +1189,7 @@ static void intel_sdvo_pre_enable(struct intel_encoder *intel_encoder)
 	struct drm_device *dev = intel_encoder->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *crtc = to_intel_crtc(intel_encoder->base.crtc);
-	struct drm_display_mode *adjusted_mode =
-		&crtc->config->base.adjusted_mode;
+	struct drm_display_mode *adjusted_mode = &crtc->config->base.adjusted_mode;
 	struct drm_display_mode *mode = &crtc->config->base.mode;
 	struct intel_sdvo *intel_sdvo = to_sdvo(intel_encoder);
 	u32 sdvox;
diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c
index 4349fde..8218040 100644
--- a/drivers/gpu/drm/i915/intel_sprite.c
+++ b/drivers/gpu/drm/i915/intel_sprite.c
@@ -79,19 +79,19 @@ static int usecs_to_scanlines(const struct drm_display_mode *mode, int usecs)
 void intel_pipe_update_start(struct intel_crtc *crtc)
 {
 	struct drm_device *dev = crtc->base.dev;
-	const struct drm_display_mode *mode = &crtc->config->base.adjusted_mode;
+	const struct drm_display_mode *adjusted_mode = &crtc->config->base.adjusted_mode;
 	enum pipe pipe = crtc->pipe;
 	long timeout = msecs_to_jiffies_timeout(1);
 	int scanline, min, max, vblank_start;
 	wait_queue_head_t *wq = drm_crtc_vblank_waitqueue(&crtc->base);
 	DEFINE_WAIT(wait);
 
-	vblank_start = mode->crtc_vblank_start;
-	if (mode->flags & DRM_MODE_FLAG_INTERLACE)
+	vblank_start = adjusted_mode->crtc_vblank_start;
+	if (adjusted_mode->flags & DRM_MODE_FLAG_INTERLACE)
 		vblank_start = DIV_ROUND_UP(vblank_start, 2);
 
 	/* FIXME needs to be calibrated sensibly */
-	min = vblank_start - usecs_to_scanlines(mode, 100);
+	min = vblank_start - usecs_to_scanlines(adjusted_mode, 100);
 	max = vblank_start - 1;
 
 	local_irq_disable();
-- 
cgit v0.10.2


From 0d44d3fa2b9089b0039118351773d465204ad581 Mon Sep 17 00:00:00 2001
From: Alex Dai <yu.dai@intel.com>
Date: Tue, 22 Sep 2015 13:48:40 -0700
Subject: drm/i915/guc: Fix a bug in GuC status check

Bit 16 of GuC status indicates resuming from RC6. The LAPIC_DONE
status is a reliable readiness flag only when resuming from RC6.
This fix a racing issue that allocation of doorbell fails whilst
GuC init is not finished.

Signed-off-by: Alex Dai <yu.dai@intel.com>
Reviewed-by: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_guc_reg.h b/drivers/gpu/drm/i915/i915_guc_reg.h
index 9d79a6b..b355661 100644
--- a/drivers/gpu/drm/i915/i915_guc_reg.h
+++ b/drivers/gpu/drm/i915/i915_guc_reg.h
@@ -37,6 +37,7 @@
 #define   GS_UKERNEL_READY		  (0xF0 << GS_UKERNEL_SHIFT)
 #define   GS_MIA_SHIFT			16
 #define   GS_MIA_MASK			  (0x07 << GS_MIA_SHIFT)
+#define   GS_MIA_CORE_STATE		  (1 << GS_MIA_SHIFT)
 
 #define SOFT_SCRATCH(n)			(0xc180 + ((n) * 4))
 
diff --git a/drivers/gpu/drm/i915/intel_guc_loader.c b/drivers/gpu/drm/i915/intel_guc_loader.c
index e0601cc..40241f3 100644
--- a/drivers/gpu/drm/i915/intel_guc_loader.c
+++ b/drivers/gpu/drm/i915/intel_guc_loader.c
@@ -209,9 +209,10 @@ static inline bool guc_ucode_response(struct drm_i915_private *dev_priv,
 				      u32 *status)
 {
 	u32 val = I915_READ(GUC_STATUS);
+	u32 uk_val = val & GS_UKERNEL_MASK;
 	*status = val;
-	return ((val & GS_UKERNEL_MASK) == GS_UKERNEL_READY ||
-		(val & GS_UKERNEL_MASK) == GS_UKERNEL_LAPIC_DONE);
+	return (uk_val == GS_UKERNEL_READY ||
+		((val & GS_MIA_CORE_STATE) && uk_val == GS_UKERNEL_LAPIC_DONE));
 }
 
 /*
-- 
cgit v0.10.2


From 165ed87c47ae7dd0deab53d552a29d7985569c28 Mon Sep 17 00:00:00 2001
From: Jesse Barnes <jbarnes@virtuousgeek.org>
Date: Wed, 23 Sep 2015 14:37:17 -0700
Subject: drm/i915: fixup runtime PM handling v2

According to the PCI docs and Rafael, we don't need to be doing explicit
enables and disables in our init and teardown routines, as they're taken
care of by the PCI core.  So drop the pm_runtime_disable() at teardown
and pm_runtime_set_active() at init.

This fixes one failure of the basic-pci-d3-state test on my BYT.

v2: drop extra get_noresume() and put_noidle() (Rafael)

Signed-off-by: Jesse Barnes <jbarnes at virtuousgeek.org>
Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net>
Acked-by: "Rafael J. Wysocki" <rjw@rjwysocki.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c
index 4a815bb..e1fdbab 100644
--- a/drivers/gpu/drm/i915/intel_runtime_pm.c
+++ b/drivers/gpu/drm/i915/intel_runtime_pm.c
@@ -1828,7 +1828,6 @@ static void intel_runtime_pm_disable(struct drm_i915_private *dev_priv)
 
 	/* Make sure we're not suspended first. */
 	pm_runtime_get_sync(device);
-	pm_runtime_disable(device);
 }
 
 /**
@@ -2120,8 +2119,6 @@ void intel_runtime_pm_enable(struct drm_i915_private *dev_priv)
 	if (!HAS_RUNTIME_PM(dev))
 		return;
 
-	pm_runtime_set_active(device);
-
 	/*
 	 * RPM depends on RC6 to save restore the GT HW context, so make RC6 a
 	 * requirement.
-- 
cgit v0.10.2


From d9d7000d5a4dfda6aa54907b96cfa490f4140aa9 Mon Sep 17 00:00:00 2001
From: Sonika Jindal <sonika.jindal@intel.com>
Date: Thu, 24 Sep 2015 10:24:56 +0530
Subject: drm/i915/bxt: eDP low vswing support

Adding voltage swing table for edp to support low vswings.

v2: Rebased.

Signed-off-by: Sonika Jindal <sonika.jindal@intel.com>
Reviewed-by: Sivakumar Thulasimani <sivakumar.thulasimani@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
index 9e640ea..205f618 100644
--- a/drivers/gpu/drm/i915/intel_ddi.c
+++ b/drivers/gpu/drm/i915/intel_ddi.c
@@ -256,9 +256,6 @@ struct bxt_ddi_buf_trans {
 	bool default_index; /* true if the entry represents default value */
 };
 
-/* BSpec does not define separate vswing/pre-emphasis values for eDP.
- * Using DP values for eDP as well.
- */
 static const struct bxt_ddi_buf_trans bxt_ddi_translations_dp[] = {
 					/* Idx	NT mV diff	db  */
 	{ 52,  0x9A, 0, 128, true  },	/* 0:	400		0   */
@@ -273,6 +270,20 @@ static const struct bxt_ddi_buf_trans bxt_ddi_translations_dp[] = {
 	{ 154, 0x9A, 1, 128, false },	/* 9:	1200		0   */
 };
 
+static const struct bxt_ddi_buf_trans bxt_ddi_translations_edp[] = {
+					/* Idx	NT mV diff	db  */
+	{ 26, 0, 0, 128, false },	/* 0:	200		0   */
+	{ 38, 0, 0, 112, false },	/* 1:	200		1.5 */
+	{ 48, 0, 0, 96,  false },	/* 2:	200		4   */
+	{ 54, 0, 0, 69,  false },	/* 3:	200		6   */
+	{ 32, 0, 0, 128, false },	/* 4:	250		0   */
+	{ 48, 0, 0, 104, false },	/* 5:	250		1.5 */
+	{ 54, 0, 0, 85,  false },	/* 6:	250		4   */
+	{ 43, 0, 0, 128, false },	/* 7:	300		0   */
+	{ 54, 0, 0, 101, false },	/* 8:	300		1.5 */
+	{ 48, 0, 0, 128, false },	/* 9:	300		0   */
+};
+
 /* BSpec has 2 recommended values - entries 0 and 8.
  * Using the entry with higher vswing.
  */
@@ -2112,7 +2123,11 @@ static void bxt_ddi_vswing_sequence(struct drm_device *dev, u32 level,
 	u32 n_entries, i;
 	uint32_t val;
 
-	if (type == INTEL_OUTPUT_DISPLAYPORT || type == INTEL_OUTPUT_EDP) {
+	if (type == INTEL_OUTPUT_EDP && dev_priv->edp_low_vswing) {
+		n_entries = ARRAY_SIZE(bxt_ddi_translations_edp);
+		ddi_translations = bxt_ddi_translations_edp;
+	} else if (type == INTEL_OUTPUT_DISPLAYPORT
+			|| type == INTEL_OUTPUT_EDP) {
 		n_entries = ARRAY_SIZE(bxt_ddi_translations_dp);
 		ddi_translations = bxt_ddi_translations_dp;
 	} else if (type == INTEL_OUTPUT_HDMI) {
-- 
cgit v0.10.2


From 9c58a049566566a8dc5aa1874ddfb3d32f954260 Mon Sep 17 00:00:00 2001
From: Sonika Jindal <sonika.jindal@intel.com>
Date: Thu, 24 Sep 2015 10:22:54 +0530
Subject: drm/i915/bxt: Set oscaledcompmethod to enable scale value

Bspec update tells that we have to enable oscaledcompmethod instead of
ouniqetrangenmethod for enabling scale value during swing programming.

v2: Adding back 'don't care' values to bxt_ddi_translations_dp and add
error message if ouniquetrangemethod was set (Imre)

Signed-off-by: Sonika Jindal <sonika.jindal@intel.com>
Reviewed-by: Sivakumar Thulasimani <sivakumar.thulasimani@intel.com>(v1)
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 56157eb..184b123 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -1395,7 +1395,8 @@ enum skl_disp_power_wells {
 #define BXT_PORT_TX_DW3_LN0(port)	_PORT3(port, _PORT_TX_DW3_LN0_A,  \
 						     _PORT_TX_DW3_LN0_B,  \
 						     _PORT_TX_DW3_LN0_C)
-#define   UNIQE_TRANGE_EN_METHOD	(1 << 27)
+#define   SCALE_DCOMP_METHOD		(1 << 26)
+#define   UNIQUE_TRANGE_EN_METHOD	(1 << 27)
 
 #define _PORT_TX_DW4_LN0_A		0x162510
 #define _PORT_TX_DW4_LN0_B		0x6C510
diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
index 205f618..0e46679 100644
--- a/drivers/gpu/drm/i915/intel_ddi.c
+++ b/drivers/gpu/drm/i915/intel_ddi.c
@@ -2165,9 +2165,13 @@ static void bxt_ddi_vswing_sequence(struct drm_device *dev, u32 level,
 	I915_WRITE(BXT_PORT_TX_DW2_GRP(port), val);
 
 	val = I915_READ(BXT_PORT_TX_DW3_LN0(port));
-	val &= ~UNIQE_TRANGE_EN_METHOD;
+	val &= ~SCALE_DCOMP_METHOD;
 	if (ddi_translations[level].enable)
-		val |= UNIQE_TRANGE_EN_METHOD;
+		val |= SCALE_DCOMP_METHOD;
+
+	if ((val & UNIQUE_TRANGE_EN_METHOD) && !(val & SCALE_DCOMP_METHOD))
+		DRM_ERROR("Disabled scaling while ouniqetrangenmethod was set");
+
 	I915_WRITE(BXT_PORT_TX_DW3_GRP(port), val);
 
 	val = I915_READ(BXT_PORT_TX_DW4_LN0(port));
-- 
cgit v0.10.2


From eecf613a432c663d206a11c2543e9cca2ddc3aba Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 21 Sep 2015 18:05:14 +0300
Subject: drm/i915: Parametrize fence registers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

v2: Hide the 945 vs. rest of gen2/3 difference in the macro

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_fence.c b/drivers/gpu/drm/i915/i915_gem_fence.c
index 6077dff..1cbfd5b 100644
--- a/drivers/gpu/drm/i915/i915_gem_fence.c
+++ b/drivers/gpu/drm/i915/i915_gem_fence.c
@@ -59,19 +59,19 @@ static void i965_write_fence_reg(struct drm_device *dev, int reg,
 				 struct drm_i915_gem_object *obj)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	int fence_reg;
+	int fence_reg_lo, fence_reg_hi;
 	int fence_pitch_shift;
 
 	if (INTEL_INFO(dev)->gen >= 6) {
-		fence_reg = FENCE_REG_SANDYBRIDGE_0;
-		fence_pitch_shift = SANDYBRIDGE_FENCE_PITCH_SHIFT;
+		fence_reg_lo = FENCE_REG_GEN6_LO(reg);
+		fence_reg_hi = FENCE_REG_GEN6_HI(reg);
+		fence_pitch_shift = GEN6_FENCE_PITCH_SHIFT;
 	} else {
-		fence_reg = FENCE_REG_965_0;
+		fence_reg_lo = FENCE_REG_965_LO(reg);
+		fence_reg_hi = FENCE_REG_965_HI(reg);
 		fence_pitch_shift = I965_FENCE_PITCH_SHIFT;
 	}
 
-	fence_reg += reg * 8;
-
 	/* To w/a incoherency with non-atomic 64-bit register updates,
 	 * we split the 64-bit update into two 32-bit writes. In order
 	 * for a partial fence not to be evaluated between writes, we
@@ -81,8 +81,8 @@ static void i965_write_fence_reg(struct drm_device *dev, int reg,
 	 * For extra levels of paranoia, we make sure each step lands
 	 * before applying the next step.
 	 */
-	I915_WRITE(fence_reg, 0);
-	POSTING_READ(fence_reg);
+	I915_WRITE(fence_reg_lo, 0);
+	POSTING_READ(fence_reg_lo);
 
 	if (obj) {
 		u32 size = i915_gem_obj_ggtt_size(obj);
@@ -103,14 +103,14 @@ static void i965_write_fence_reg(struct drm_device *dev, int reg,
 			val |= 1 << I965_FENCE_TILING_Y_SHIFT;
 		val |= I965_FENCE_REG_VALID;
 
-		I915_WRITE(fence_reg + 4, val >> 32);
-		POSTING_READ(fence_reg + 4);
+		I915_WRITE(fence_reg_hi, val >> 32);
+		POSTING_READ(fence_reg_hi);
 
-		I915_WRITE(fence_reg + 0, val);
-		POSTING_READ(fence_reg);
+		I915_WRITE(fence_reg_lo, val);
+		POSTING_READ(fence_reg_lo);
 	} else {
-		I915_WRITE(fence_reg + 4, 0);
-		POSTING_READ(fence_reg + 4);
+		I915_WRITE(fence_reg_hi, 0);
+		POSTING_READ(fence_reg_hi);
 	}
 }
 
@@ -149,13 +149,8 @@ static void i915_write_fence_reg(struct drm_device *dev, int reg,
 	} else
 		val = 0;
 
-	if (reg < 8)
-		reg = FENCE_REG_830_0 + reg * 4;
-	else
-		reg = FENCE_REG_945_8 + (reg - 8) * 4;
-
-	I915_WRITE(reg, val);
-	POSTING_READ(reg);
+	I915_WRITE(FENCE_REG(reg), val);
+	POSTING_READ(FENCE_REG(reg));
 }
 
 static void i830_write_fence_reg(struct drm_device *dev, int reg,
@@ -186,8 +181,8 @@ static void i830_write_fence_reg(struct drm_device *dev, int reg,
 	} else
 		val = 0;
 
-	I915_WRITE(FENCE_REG_830_0 + reg * 4, val);
-	POSTING_READ(FENCE_REG_830_0 + reg * 4);
+	I915_WRITE(FENCE_REG(reg), val);
+	POSTING_READ(FENCE_REG(reg));
 }
 
 inline static bool i915_gem_object_needs_mb(struct drm_i915_gem_object *obj)
diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
index f95de05..d979dca 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.c
+++ b/drivers/gpu/drm/i915/i915_gpu_error.c
@@ -792,20 +792,15 @@ static void i915_gem_record_fences(struct drm_device *dev,
 	int i;
 
 	if (IS_GEN3(dev) || IS_GEN2(dev)) {
-		for (i = 0; i < 8; i++)
-			error->fence[i] = I915_READ(FENCE_REG_830_0 + (i * 4));
-		if (IS_I945G(dev) || IS_I945GM(dev) || IS_G33(dev))
-			for (i = 0; i < 8; i++)
-				error->fence[i+8] = I915_READ(FENCE_REG_945_8 +
-							      (i * 4));
-	} else if (IS_GEN5(dev) || IS_GEN4(dev))
-		for (i = 0; i < 16; i++)
-			error->fence[i] = I915_READ64(FENCE_REG_965_0 +
-						      (i * 8));
-	else if (INTEL_INFO(dev)->gen >= 6)
 		for (i = 0; i < dev_priv->num_fence_regs; i++)
-			error->fence[i] = I915_READ64(FENCE_REG_SANDYBRIDGE_0 +
-						      (i * 8));
+			error->fence[i] = I915_READ(FENCE_REG(i));
+	} else if (IS_GEN5(dev) || IS_GEN4(dev)) {
+		for (i = 0; i < dev_priv->num_fence_regs; i++)
+			error->fence[i] = I915_READ64(FENCE_REG_965_LO(i));
+	} else if (INTEL_INFO(dev)->gen >= 6) {
+		for (i = 0; i < dev_priv->num_fence_regs; i++)
+			error->fence[i] = I915_READ64(FENCE_REG_GEN6_LO(i));
+	}
 }
 
 
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 184b123..d5a0790 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -1437,9 +1437,15 @@ enum skl_disp_power_wells {
 
 /*
  * Fence registers
+ * [0-7]  @ 0x2000 gen2,gen3
+ * [8-15] @ 0x3000 945,g33,pnv
+ *
+ * [0-15] @ 0x3000 gen4,gen5
+ *
+ * [0-15] @ 0x100000 gen6,vlv,chv
+ * [0-31] @ 0x100000 gen7+
  */
-#define FENCE_REG_830_0			0x2000
-#define FENCE_REG_945_8			0x3000
+#define FENCE_REG(i)			(0x2000 + (((i) & 8) << 9) + ((i) & 7) * 4)
 #define   I830_FENCE_START_MASK		0x07f80000
 #define   I830_FENCE_TILING_Y_SHIFT	12
 #define   I830_FENCE_SIZE_BITS(size)	((ffs((size) >> 19) - 1) << 8)
@@ -1452,14 +1458,16 @@ enum skl_disp_power_wells {
 #define   I915_FENCE_START_MASK		0x0ff00000
 #define   I915_FENCE_SIZE_BITS(size)	((ffs((size) >> 20) - 1) << 8)
 
-#define FENCE_REG_965_0			0x03000
+#define FENCE_REG_965_LO(i)		(0x03000 + (i) * 8)
+#define FENCE_REG_965_HI(i)		(0x03000 + (i) * 8 + 4)
 #define   I965_FENCE_PITCH_SHIFT	2
 #define   I965_FENCE_TILING_Y_SHIFT	1
 #define   I965_FENCE_REG_VALID		(1<<0)
 #define   I965_FENCE_MAX_PITCH_VAL	0x0400
 
-#define FENCE_REG_SANDYBRIDGE_0		0x100000
-#define   SANDYBRIDGE_FENCE_PITCH_SHIFT	32
+#define FENCE_REG_GEN6_LO(i)	(0x100000 + (i) * 8)
+#define FENCE_REG_GEN6_HI(i)	(0x100000 + (i) * 8 + 4)
+#define   GEN6_FENCE_PITCH_SHIFT	32
 #define   GEN7_FENCE_MAX_PITCH_VAL	0x0800
 
 
-- 
cgit v0.10.2


From b26d3ea323b41a56db74f46131bc89e3c485f365 Mon Sep 17 00:00:00 2001
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Date: Wed, 23 Sep 2015 16:11:41 +0200
Subject: drm/i915: Add primary plane to mask if it's visible

This fixes the warnings like

"plane A assertion failure, should be disabled but not"

that on the initial modeset during boot. This can happen if
the primary plane is enabled by the firmware, but inheriting
it fails because the DMAR is active or for other reasons.

Most likely caused by

commit 36750f284b3a4f19b304fda1bb7d6e9e1275ea8d
Author: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Date:   Mon Jun 1 12:49:54 2015 +0200

    drm/i915: update plane state during init

This is the 4.4 version of

commit 721a09f7393de6c28a07516dccd654c6e995944a
Author: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Date:   Tue Sep 15 14:28:54 2015 +0200

    drm/i915: Add primary plane to mask if it's visible

Reported-by: Andreas Reis <andreas.reis@gmail.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=91429
Reported-and-tested-by: Emil Renner Berthing <kernel@esmil.dk>
Tested-by: Andreas Reis <andreas.reis@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 45ec0b8..036c704 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -15128,11 +15128,15 @@ static bool primary_get_hw_state(struct intel_plane *plane)
 /* FIXME read out full plane state for all planes */
 static void readout_plane_state(struct intel_crtc *crtc)
 {
+	struct drm_plane *primary = crtc->base.primary;
 	struct intel_plane_state *plane_state =
-		to_intel_plane_state(crtc->base.primary->state);
+		to_intel_plane_state(primary->state);
 
 	plane_state->visible =
-		primary_get_hw_state(to_intel_plane(crtc->base.primary));
+		primary_get_hw_state(to_intel_plane(primary));
+
+	if (plane_state->visible)
+		crtc->base.state->plane_mask |= 1 << drm_plane_index(primary);
 }
 
 static void intel_modeset_readout_hw_state(struct drm_device *dev)
-- 
cgit v0.10.2


From 1ca36d4cb3436bd783f8688b563830341a8d9150 Mon Sep 17 00:00:00 2001
From: Paulo Zanoni <paulo.r.zanoni@intel.com>
Date: Wed, 23 Sep 2015 12:52:22 -0300
Subject: drm/i915: don't use the first stolen page on Broadwell

The spec says we just can't use it.

v2:
  - Add WA name (Ville).
  - Add a big comment explaining that we still didn't fix the problem
    where we inherit a framebuffer on the first page (Chris, Ville).

Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Acked-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_stolen.c b/drivers/gpu/drm/i915/i915_gem_stolen.c
index 1520779..69eebc6 100644
--- a/drivers/gpu/drm/i915/i915_gem_stolen.c
+++ b/drivers/gpu/drm/i915/i915_gem_stolen.c
@@ -51,6 +51,11 @@ int i915_gem_stolen_insert_node_in_range(struct drm_i915_private *dev_priv,
 	if (!drm_mm_initialized(&dev_priv->mm.stolen))
 		return -ENODEV;
 
+	/* See the comment at the drm_mm_init() call for more about this check.
+	 * WaSkipStolenMemoryFirstPage:bdw,chv (incomplete) */
+	if (INTEL_INFO(dev_priv)->gen == 8 && start < 4096)
+		start = 4096;
+
 	mutex_lock(&dev_priv->mm.stolen_lock);
 	ret = drm_mm_insert_node_in_range(&dev_priv->mm.stolen, node, size,
 					  alignment, start, end,
@@ -393,7 +398,17 @@ int i915_gem_init_stolen(struct drm_device *dev)
 	dev_priv->gtt.stolen_usable_size = dev_priv->gtt.stolen_size -
 					   reserved_total;
 
-	/* Basic memrange allocator for stolen space */
+	/*
+	 * Basic memrange allocator for stolen space.
+	 *
+	 * TODO: Notice that some platforms require us to not use the first page
+	 * of the stolen memory but their BIOSes may still put the framebuffer
+	 * on the first page. So we don't reserve this page for now because of
+	 * that. Our current solution is to just prevent new nodes from being
+	 * inserted on the first page - see the check we have at
+	 * i915_gem_stolen_insert_node_in_range(). We may want to fix the fbcon
+	 * problem later.
+	 */
 	drm_mm_init(&dev_priv->mm.stolen, 0, dev_priv->gtt.stolen_usable_size);
 
 	return 0;
-- 
cgit v0.10.2


From 9f218336bd44719c64a6e7df142b101304452692 Mon Sep 17 00:00:00 2001
From: Paulo Zanoni <paulo.r.zanoni@intel.com>
Date: Wed, 23 Sep 2015 12:52:27 -0300
Subject: drm/i915: extract fbc_supported()

Make it clear that we're checking whether FBC is supported or not. The
fact that the vfunc is not NULL is just a consequence.

Another name option would have been fbc_initialized().

Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_fbc.c b/drivers/gpu/drm/i915/intel_fbc.c
index 6777fbb..20ab1b4 100644
--- a/drivers/gpu/drm/i915/intel_fbc.c
+++ b/drivers/gpu/drm/i915/intel_fbc.c
@@ -41,6 +41,11 @@
 #include "intel_drv.h"
 #include "i915_drv.h"
 
+static inline bool fbc_supported(struct drm_i915_private *dev_priv)
+{
+	return dev_priv->fbc.enable_fbc != NULL;
+}
+
 /*
  * In some platforms where the CRTC's x:0/y:0 coordinates doesn't match the
  * frontbuffer's x:0/y:0 coordinates we lie to the hardware about the plane's
@@ -439,7 +444,7 @@ static void __intel_fbc_disable(struct drm_i915_private *dev_priv)
  */
 void intel_fbc_disable(struct drm_i915_private *dev_priv)
 {
-	if (!dev_priv->fbc.enable_fbc)
+	if (!fbc_supported(dev_priv))
 		return;
 
 	mutex_lock(&dev_priv->fbc.lock);
@@ -457,7 +462,7 @@ void intel_fbc_disable_crtc(struct intel_crtc *crtc)
 {
 	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
 
-	if (!dev_priv->fbc.enable_fbc)
+	if (!fbc_supported(dev_priv))
 		return;
 
 	mutex_lock(&dev_priv->fbc.lock);
@@ -685,7 +690,7 @@ static void __intel_fbc_cleanup_cfb(struct drm_i915_private *dev_priv)
 
 void intel_fbc_cleanup_cfb(struct drm_i915_private *dev_priv)
 {
-	if (!dev_priv->fbc.enable_fbc)
+	if (!fbc_supported(dev_priv))
 		return;
 
 	mutex_lock(&dev_priv->fbc.lock);
@@ -948,7 +953,7 @@ out_disable:
  */
 void intel_fbc_update(struct drm_i915_private *dev_priv)
 {
-	if (!dev_priv->fbc.enable_fbc)
+	if (!fbc_supported(dev_priv))
 		return;
 
 	mutex_lock(&dev_priv->fbc.lock);
@@ -962,7 +967,7 @@ void intel_fbc_invalidate(struct drm_i915_private *dev_priv,
 {
 	unsigned int fbc_bits;
 
-	if (!dev_priv->fbc.enable_fbc)
+	if (!fbc_supported(dev_priv))
 		return;
 
 	if (origin == ORIGIN_GTT)
@@ -989,7 +994,7 @@ void intel_fbc_invalidate(struct drm_i915_private *dev_priv,
 void intel_fbc_flush(struct drm_i915_private *dev_priv,
 		     unsigned int frontbuffer_bits, enum fb_op_origin origin)
 {
-	if (!dev_priv->fbc.enable_fbc)
+	if (!fbc_supported(dev_priv))
 		return;
 
 	if (origin == ORIGIN_GTT)
-- 
cgit v0.10.2


From 3c5f174e383d6eddd9190637cebd005174287996 Mon Sep 17 00:00:00 2001
From: Paulo Zanoni <paulo.r.zanoni@intel.com>
Date: Wed, 23 Sep 2015 12:52:24 -0300
Subject: drm/i915: export size_is_valid() from __intel_fbc_update()

Make the giant function a little less giant.

Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
[danvet: Add pipe_ prefix as suggested by Chris.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_fbc.c b/drivers/gpu/drm/i915/intel_fbc.c
index 20ab1b4..1b2ebb2 100644
--- a/drivers/gpu/drm/i915/intel_fbc.c
+++ b/drivers/gpu/drm/i915/intel_fbc.c
@@ -754,6 +754,26 @@ static bool pixel_format_is_valid(struct drm_framebuffer *fb)
 	}
 }
 
+static bool pipe_size_is_valid(struct intel_crtc *crtc)
+{
+	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
+	unsigned int max_w, max_h;
+
+	if (INTEL_INFO(dev_priv)->gen >= 8 || IS_HASWELL(dev_priv)) {
+		max_w = 4096;
+		max_h = 4096;
+	} else if (IS_G4X(dev_priv) || INTEL_INFO(dev_priv)->gen >= 5) {
+		max_w = 4096;
+		max_h = 2048;
+	} else {
+		max_w = 2048;
+		max_h = 1536;
+	}
+
+	return crtc->config->pipe_src_w <= max_w &&
+	       crtc->config->pipe_src_h <= max_h;
+}
+
 /**
  * __intel_fbc_update - enable/disable FBC as needed, unlocked
  * @dev_priv: i915 device instance
@@ -780,7 +800,6 @@ static void __intel_fbc_update(struct drm_i915_private *dev_priv)
 	struct drm_framebuffer *fb;
 	struct drm_i915_gem_object *obj;
 	const struct drm_display_mode *adjusted_mode;
-	unsigned int max_width, max_height;
 
 	WARN_ON(!mutex_is_locked(&dev_priv->fbc.lock));
 
@@ -829,21 +848,11 @@ static void __intel_fbc_update(struct drm_i915_private *dev_priv)
 		goto out_disable;
 	}
 
-	if (INTEL_INFO(dev_priv)->gen >= 8 || IS_HASWELL(dev_priv)) {
-		max_width = 4096;
-		max_height = 4096;
-	} else if (IS_G4X(dev_priv) || INTEL_INFO(dev_priv)->gen >= 5) {
-		max_width = 4096;
-		max_height = 2048;
-	} else {
-		max_width = 2048;
-		max_height = 1536;
-	}
-	if (intel_crtc->config->pipe_src_w > max_width ||
-	    intel_crtc->config->pipe_src_h > max_height) {
+	if (!pipe_size_is_valid(intel_crtc)) {
 		set_no_fbc_reason(dev_priv, FBC_MODE_TOO_LARGE);
 		goto out_disable;
 	}
+
 	if ((INTEL_INFO(dev_priv)->gen < 4 || HAS_DDI(dev_priv)) &&
 	    intel_crtc->plane != PLANE_A) {
 		set_no_fbc_reason(dev_priv, FBC_BAD_PLANE);
-- 
cgit v0.10.2


From 5e7234c9ccf88c427448fbe147839b4dca82efde Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 25 Sep 2015 16:37:43 +0300
Subject: drm/i915: s/mode/adjusted_mode/ in functions that really get passed
 the adjusted_mode
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Rename the function argument to 'adjusted_mode' whenever the function
only ever gets passed the adjusted_mode.

v2: Update due to intel_dsi.c changes

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 04d710f..ad37b25 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -646,7 +646,7 @@ struct drm_i915_display_funcs {
 	void (*crtc_disable)(struct drm_crtc *crtc);
 	void (*audio_codec_enable)(struct drm_connector *connector,
 				   struct intel_encoder *encoder,
-				   struct drm_display_mode *mode);
+				   const struct drm_display_mode *adjusted_mode);
 	void (*audio_codec_disable)(struct intel_encoder *encoder);
 	void (*fdi_link_train)(struct drm_crtc *crtc);
 	void (*init_clock_gating)(struct drm_device *dev);
diff --git a/drivers/gpu/drm/i915/intel_audio.c b/drivers/gpu/drm/i915/intel_audio.c
index d9876c5..779f90c 100644
--- a/drivers/gpu/drm/i915/intel_audio.c
+++ b/drivers/gpu/drm/i915/intel_audio.c
@@ -69,17 +69,18 @@ static const struct {
 };
 
 /* get AUD_CONFIG_PIXEL_CLOCK_HDMI_* value for mode */
-static u32 audio_config_hdmi_pixel_clock(struct drm_display_mode *mode)
+static u32 audio_config_hdmi_pixel_clock(const struct drm_display_mode *adjusted_mode)
 {
 	int i;
 
 	for (i = 0; i < ARRAY_SIZE(hdmi_audio_clock); i++) {
-		if (mode->clock == hdmi_audio_clock[i].clock)
+		if (adjusted_mode->clock == hdmi_audio_clock[i].clock)
 			break;
 	}
 
 	if (i == ARRAY_SIZE(hdmi_audio_clock)) {
-		DRM_DEBUG_KMS("HDMI audio pixel clock setting for %d not found, falling back to defaults\n", mode->clock);
+		DRM_DEBUG_KMS("HDMI audio pixel clock setting for %d not found, falling back to defaults\n",
+			      adjusted_mode->clock);
 		i = 1;
 	}
 
@@ -138,7 +139,7 @@ static void g4x_audio_codec_disable(struct intel_encoder *encoder)
 
 static void g4x_audio_codec_enable(struct drm_connector *connector,
 				   struct intel_encoder *encoder,
-				   struct drm_display_mode *mode)
+				   const struct drm_display_mode *adjusted_mode)
 {
 	struct drm_i915_private *dev_priv = connector->dev->dev_private;
 	uint8_t *eld = connector->eld;
@@ -203,7 +204,7 @@ static void hsw_audio_codec_disable(struct intel_encoder *encoder)
 
 static void hsw_audio_codec_enable(struct drm_connector *connector,
 				   struct intel_encoder *encoder,
-				   struct drm_display_mode *mode)
+				   const struct drm_display_mode *adjusted_mode)
 {
 	struct drm_i915_private *dev_priv = connector->dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->base.crtc);
@@ -251,7 +252,7 @@ static void hsw_audio_codec_enable(struct drm_connector *connector,
 	if (intel_pipe_has_type(intel_crtc, INTEL_OUTPUT_DISPLAYPORT))
 		tmp |= AUD_CONFIG_N_VALUE_INDEX;
 	else
-		tmp |= audio_config_hdmi_pixel_clock(mode);
+		tmp |= audio_config_hdmi_pixel_clock(adjusted_mode);
 	I915_WRITE(HSW_AUD_CFG(pipe), tmp);
 }
 
@@ -304,7 +305,7 @@ static void ilk_audio_codec_disable(struct intel_encoder *encoder)
 
 static void ilk_audio_codec_enable(struct drm_connector *connector,
 				   struct intel_encoder *encoder,
-				   struct drm_display_mode *mode)
+				   const struct drm_display_mode *adjusted_mode)
 {
 	struct drm_i915_private *dev_priv = connector->dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->base.crtc);
@@ -381,7 +382,7 @@ static void ilk_audio_codec_enable(struct drm_connector *connector,
 	if (intel_pipe_has_type(intel_crtc, INTEL_OUTPUT_DISPLAYPORT))
 		tmp |= AUD_CONFIG_N_VALUE_INDEX;
 	else
-		tmp |= audio_config_hdmi_pixel_clock(mode);
+		tmp |= audio_config_hdmi_pixel_clock(adjusted_mode);
 	I915_WRITE(aud_config, tmp);
 }
 
diff --git a/drivers/gpu/drm/i915/intel_dsi.c b/drivers/gpu/drm/i915/intel_dsi.c
index 23b7fc5..b4d5213 100644
--- a/drivers/gpu/drm/i915/intel_dsi.c
+++ b/drivers/gpu/drm/i915/intel_dsi.c
@@ -698,7 +698,7 @@ static u16 txbyteclkhs(u16 pixels, int bpp, int lane_count,
 }
 
 static void set_dsi_timings(struct drm_encoder *encoder,
-			    const struct drm_display_mode *mode)
+			    const struct drm_display_mode *adjusted_mode)
 {
 	struct drm_device *dev = encoder->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -710,10 +710,10 @@ static void set_dsi_timings(struct drm_encoder *encoder,
 
 	u16 hactive, hfp, hsync, hbp, vfp, vsync, vbp;
 
-	hactive = mode->hdisplay;
-	hfp = mode->hsync_start - mode->hdisplay;
-	hsync = mode->hsync_end - mode->hsync_start;
-	hbp = mode->htotal - mode->hsync_end;
+	hactive = adjusted_mode->hdisplay;
+	hfp = adjusted_mode->hsync_start - adjusted_mode->hdisplay;
+	hsync = adjusted_mode->hsync_end - adjusted_mode->hsync_start;
+	hbp = adjusted_mode->htotal - adjusted_mode->hsync_end;
 
 	if (intel_dsi->dual_link) {
 		hactive /= 2;
@@ -724,9 +724,9 @@ static void set_dsi_timings(struct drm_encoder *encoder,
 		hbp /= 2;
 	}
 
-	vfp = mode->vsync_start - mode->vdisplay;
-	vsync = mode->vsync_end - mode->vsync_start;
-	vbp = mode->vtotal - mode->vsync_end;
+	vfp = adjusted_mode->vsync_start - adjusted_mode->vdisplay;
+	vsync = adjusted_mode->vsync_end - adjusted_mode->vsync_start;
+	vbp = adjusted_mode->vtotal - adjusted_mode->vsync_end;
 
 	/* horizontal values are in terms of high speed byte clock */
 	hactive = txbyteclkhs(hactive, bpp, lane_count,
@@ -745,11 +745,11 @@ static void set_dsi_timings(struct drm_encoder *encoder,
 			 * whereas these values should be based on resolution.
 			 */
 			I915_WRITE(BXT_MIPI_TRANS_HACTIVE(port),
-					mode->hdisplay);
+				   adjusted_mode->hdisplay);
 			I915_WRITE(BXT_MIPI_TRANS_VACTIVE(port),
-					mode->vdisplay);
+				   adjusted_mode->vdisplay);
 			I915_WRITE(BXT_MIPI_TRANS_VTOTAL(port),
-					mode->vtotal);
+				   adjusted_mode->vtotal);
 		}
 
 		I915_WRITE(MIPI_HACTIVE_AREA_COUNT(port), hactive);
diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c
index 2c11b4e..4294f60 100644
--- a/drivers/gpu/drm/i915/intel_panel.c
+++ b/drivers/gpu/drm/i915/intel_panel.c
@@ -172,46 +172,46 @@ done:
 }
 
 static void
-centre_horizontally(struct drm_display_mode *mode,
+centre_horizontally(struct drm_display_mode *adjusted_mode,
 		    int width)
 {
 	u32 border, sync_pos, blank_width, sync_width;
 
 	/* keep the hsync and hblank widths constant */
-	sync_width = mode->crtc_hsync_end - mode->crtc_hsync_start;
-	blank_width = mode->crtc_hblank_end - mode->crtc_hblank_start;
+	sync_width = adjusted_mode->crtc_hsync_end - adjusted_mode->crtc_hsync_start;
+	blank_width = adjusted_mode->crtc_hblank_end - adjusted_mode->crtc_hblank_start;
 	sync_pos = (blank_width - sync_width + 1) / 2;
 
-	border = (mode->hdisplay - width + 1) / 2;
+	border = (adjusted_mode->hdisplay - width + 1) / 2;
 	border += border & 1; /* make the border even */
 
-	mode->crtc_hdisplay = width;
-	mode->crtc_hblank_start = width + border;
-	mode->crtc_hblank_end = mode->crtc_hblank_start + blank_width;
+	adjusted_mode->crtc_hdisplay = width;
+	adjusted_mode->crtc_hblank_start = width + border;
+	adjusted_mode->crtc_hblank_end = adjusted_mode->crtc_hblank_start + blank_width;
 
-	mode->crtc_hsync_start = mode->crtc_hblank_start + sync_pos;
-	mode->crtc_hsync_end = mode->crtc_hsync_start + sync_width;
+	adjusted_mode->crtc_hsync_start = adjusted_mode->crtc_hblank_start + sync_pos;
+	adjusted_mode->crtc_hsync_end = adjusted_mode->crtc_hsync_start + sync_width;
 }
 
 static void
-centre_vertically(struct drm_display_mode *mode,
+centre_vertically(struct drm_display_mode *adjusted_mode,
 		  int height)
 {
 	u32 border, sync_pos, blank_width, sync_width;
 
 	/* keep the vsync and vblank widths constant */
-	sync_width = mode->crtc_vsync_end - mode->crtc_vsync_start;
-	blank_width = mode->crtc_vblank_end - mode->crtc_vblank_start;
+	sync_width = adjusted_mode->crtc_vsync_end - adjusted_mode->crtc_vsync_start;
+	blank_width = adjusted_mode->crtc_vblank_end - adjusted_mode->crtc_vblank_start;
 	sync_pos = (blank_width - sync_width + 1) / 2;
 
-	border = (mode->vdisplay - height + 1) / 2;
+	border = (adjusted_mode->vdisplay - height + 1) / 2;
 
-	mode->crtc_vdisplay = height;
-	mode->crtc_vblank_start = height + border;
-	mode->crtc_vblank_end = mode->crtc_vblank_start + blank_width;
+	adjusted_mode->crtc_vdisplay = height;
+	adjusted_mode->crtc_vblank_start = height + border;
+	adjusted_mode->crtc_vblank_end = adjusted_mode->crtc_vblank_start + blank_width;
 
-	mode->crtc_vsync_start = mode->crtc_vblank_start + sync_pos;
-	mode->crtc_vsync_end = mode->crtc_vsync_start + sync_width;
+	adjusted_mode->crtc_vsync_start = adjusted_mode->crtc_vblank_start + sync_pos;
+	adjusted_mode->crtc_vsync_end = adjusted_mode->crtc_vsync_start + sync_width;
 }
 
 static inline u32 panel_fitter_scaling(u32 source, u32 target)
diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c
index b0b96fd..ab75b61 100644
--- a/drivers/gpu/drm/i915/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/intel_sdvo.c
@@ -603,11 +603,11 @@ log_fail:
 	return false;
 }
 
-static int intel_sdvo_get_pixel_multiplier(struct drm_display_mode *mode)
+static int intel_sdvo_get_pixel_multiplier(const struct drm_display_mode *adjusted_mode)
 {
-	if (mode->clock >= 100000)
+	if (adjusted_mode->clock >= 100000)
 		return 1;
-	else if (mode->clock >= 50000)
+	else if (adjusted_mode->clock >= 50000)
 		return 2;
 	else
 		return 4;
diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c
index 8218040..b229c67 100644
--- a/drivers/gpu/drm/i915/intel_sprite.c
+++ b/drivers/gpu/drm/i915/intel_sprite.c
@@ -53,13 +53,15 @@ format_is_yuv(uint32_t format)
 	}
 }
 
-static int usecs_to_scanlines(const struct drm_display_mode *mode, int usecs)
+static int usecs_to_scanlines(const struct drm_display_mode *adjusted_mode,
+			      int usecs)
 {
 	/* paranoia */
-	if (!mode->crtc_htotal)
+	if (!adjusted_mode->crtc_htotal)
 		return 1;
 
-	return DIV_ROUND_UP(usecs * mode->crtc_clock, 1000 * mode->crtc_htotal);
+	return DIV_ROUND_UP(usecs * adjusted_mode->crtc_clock,
+			    1000 * adjusted_mode->crtc_htotal);
 }
 
 /**
-- 
cgit v0.10.2


From aad941d53f7aa2b642a798e6b3de520c19ba2e46 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 25 Sep 2015 16:38:56 +0300
Subject: drm/i915: Always use crtc_ timings when dealing with adjustead_mode
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The adjustead_mode crtc_ timings are what we will program into the hardware,
so it's those timings we should be looking practically everywhere.

The normal and crtc_ timings should differ only when stere doubling is
used. In that case the normal timings are the orignal non-doubled
timigns, and crtc_ timings are the doubled timings used by the hardware.

The only case where we continue to look at the normal timings is when we
pass the adjusted_mode to drm_match_{cea,hdmi}_mode() to find the VIC.
drm_edid keeps the modes aronund in the non-double form only, so it
needs the non-double timings to match against.

Done with sed
's/adjusted_mode->\([vhVH]\)/adjusted_mode->crtc_\1/g'
's/adjusted_mode->clock/adjusted_mode->crtc_clock/g'
with a manual s/VDisplay/vdisplay/ within the comment in intel_dvo.c

v2: Update due to intel_dsi.c changes

Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/dvo_ivch.c b/drivers/gpu/drm/i915/dvo_ivch.c
index 732ce87..e082f75 100644
--- a/drivers/gpu/drm/i915/dvo_ivch.c
+++ b/drivers/gpu/drm/i915/dvo_ivch.c
@@ -414,16 +414,16 @@ static void ivch_mode_set(struct intel_dvo_device *dvo,
 	vr40 = (VR40_STALL_ENABLE | VR40_VERTICAL_INTERP_ENABLE |
 		VR40_HORIZONTAL_INTERP_ENABLE);
 
-	if (mode->hdisplay != adjusted_mode->hdisplay ||
-	    mode->vdisplay != adjusted_mode->vdisplay) {
+	if (mode->hdisplay != adjusted_mode->crtc_hdisplay ||
+	    mode->vdisplay != adjusted_mode->crtc_vdisplay) {
 		uint16_t x_ratio, y_ratio;
 
 		vr01 |= VR01_PANEL_FIT_ENABLE;
 		vr40 |= VR40_CLOCK_GATING_ENABLE;
 		x_ratio = (((mode->hdisplay - 1) << 16) /
-			   (adjusted_mode->hdisplay - 1)) >> 2;
+			   (adjusted_mode->crtc_hdisplay - 1)) >> 2;
 		y_ratio = (((mode->vdisplay - 1) << 16) /
-			   (adjusted_mode->vdisplay - 1)) >> 2;
+			   (adjusted_mode->crtc_vdisplay - 1)) >> 2;
 		ivch_write(dvo, VR42, x_ratio);
 		ivch_write(dvo, VR41, y_ratio);
 	} else {
diff --git a/drivers/gpu/drm/i915/intel_audio.c b/drivers/gpu/drm/i915/intel_audio.c
index 779f90c..e4c30e8 100644
--- a/drivers/gpu/drm/i915/intel_audio.c
+++ b/drivers/gpu/drm/i915/intel_audio.c
@@ -74,13 +74,13 @@ static u32 audio_config_hdmi_pixel_clock(const struct drm_display_mode *adjusted
 	int i;
 
 	for (i = 0; i < ARRAY_SIZE(hdmi_audio_clock); i++) {
-		if (adjusted_mode->clock == hdmi_audio_clock[i].clock)
+		if (adjusted_mode->crtc_clock == hdmi_audio_clock[i].clock)
 			break;
 	}
 
 	if (i == ARRAY_SIZE(hdmi_audio_clock)) {
 		DRM_DEBUG_KMS("HDMI audio pixel clock setting for %d not found, falling back to defaults\n",
-			      adjusted_mode->clock);
+			      adjusted_mode->crtc_clock);
 		i = 1;
 	}
 
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 036c704..5105e09 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -4400,7 +4400,7 @@ int skl_update_scaler_crtc(struct intel_crtc_state *state)
 	return skl_update_scaler(state, !state->base.active, SKL_CRTC_INDEX,
 		&state->scaler_state.scaler_id, DRM_ROTATE_0,
 		state->pipe_src_w, state->pipe_src_h,
-		adjusted_mode->hdisplay, adjusted_mode->vdisplay);
+		adjusted_mode->crtc_hdisplay, adjusted_mode->crtc_vdisplay);
 }
 
 /**
@@ -6593,7 +6593,7 @@ static int intel_crtc_compute_config(struct intel_crtc *crtc,
 	 * WaPruneModeWithIncorrectHsyncOffset:ctg,elk,ilk,snb,ivb,vlv,hsw.
 	 */
 	if ((INTEL_INFO(dev)->gen > 4 || IS_G4X(dev)) &&
-		adjusted_mode->hsync_start == adjusted_mode->hdisplay)
+		adjusted_mode->crtc_hsync_start == adjusted_mode->crtc_hdisplay)
 		return -EINVAL;
 
 	if (HAS_IPS(dev))
diff --git a/drivers/gpu/drm/i915/intel_dp_mst.c b/drivers/gpu/drm/i915/intel_dp_mst.c
index ff8ba55..cccf014 100644
--- a/drivers/gpu/drm/i915/intel_dp_mst.c
+++ b/drivers/gpu/drm/i915/intel_dp_mst.c
@@ -78,7 +78,7 @@ static bool intel_dp_mst_compute_config(struct intel_encoder *encoder,
 		return false;
 	}
 
-	mst_pbn = drm_dp_calc_pbn_mode(adjusted_mode->clock, bpp);
+	mst_pbn = drm_dp_calc_pbn_mode(adjusted_mode->crtc_clock, bpp);
 
 	pipe_config->pbn = mst_pbn;
 	slots = drm_dp_find_vcpi_slots(&intel_dp->mst_mgr, mst_pbn);
diff --git a/drivers/gpu/drm/i915/intel_dsi.c b/drivers/gpu/drm/i915/intel_dsi.c
index b4d5213..4fb97c0 100644
--- a/drivers/gpu/drm/i915/intel_dsi.c
+++ b/drivers/gpu/drm/i915/intel_dsi.c
@@ -710,10 +710,10 @@ static void set_dsi_timings(struct drm_encoder *encoder,
 
 	u16 hactive, hfp, hsync, hbp, vfp, vsync, vbp;
 
-	hactive = adjusted_mode->hdisplay;
-	hfp = adjusted_mode->hsync_start - adjusted_mode->hdisplay;
-	hsync = adjusted_mode->hsync_end - adjusted_mode->hsync_start;
-	hbp = adjusted_mode->htotal - adjusted_mode->hsync_end;
+	hactive = adjusted_mode->crtc_hdisplay;
+	hfp = adjusted_mode->crtc_hsync_start - adjusted_mode->crtc_hdisplay;
+	hsync = adjusted_mode->crtc_hsync_end - adjusted_mode->crtc_hsync_start;
+	hbp = adjusted_mode->crtc_htotal - adjusted_mode->crtc_hsync_end;
 
 	if (intel_dsi->dual_link) {
 		hactive /= 2;
@@ -724,9 +724,9 @@ static void set_dsi_timings(struct drm_encoder *encoder,
 		hbp /= 2;
 	}
 
-	vfp = adjusted_mode->vsync_start - adjusted_mode->vdisplay;
-	vsync = adjusted_mode->vsync_end - adjusted_mode->vsync_start;
-	vbp = adjusted_mode->vtotal - adjusted_mode->vsync_end;
+	vfp = adjusted_mode->crtc_vsync_start - adjusted_mode->crtc_vdisplay;
+	vsync = adjusted_mode->crtc_vsync_end - adjusted_mode->crtc_vsync_start;
+	vbp = adjusted_mode->crtc_vtotal - adjusted_mode->crtc_vsync_end;
 
 	/* horizontal values are in terms of high speed byte clock */
 	hactive = txbyteclkhs(hactive, bpp, lane_count,
@@ -745,11 +745,11 @@ static void set_dsi_timings(struct drm_encoder *encoder,
 			 * whereas these values should be based on resolution.
 			 */
 			I915_WRITE(BXT_MIPI_TRANS_HACTIVE(port),
-				   adjusted_mode->hdisplay);
+				   adjusted_mode->crtc_hdisplay);
 			I915_WRITE(BXT_MIPI_TRANS_VACTIVE(port),
-				   adjusted_mode->vdisplay);
+				   adjusted_mode->crtc_vdisplay);
 			I915_WRITE(BXT_MIPI_TRANS_VTOTAL(port),
-				   adjusted_mode->vtotal);
+				   adjusted_mode->crtc_vtotal);
 		}
 
 		I915_WRITE(MIPI_HACTIVE_AREA_COUNT(port), hactive);
@@ -782,7 +782,7 @@ static void intel_dsi_prepare(struct intel_encoder *intel_encoder)
 
 	DRM_DEBUG_KMS("pipe %c\n", pipe_name(intel_crtc->pipe));
 
-	mode_hdisplay = adjusted_mode->hdisplay;
+	mode_hdisplay = adjusted_mode->crtc_hdisplay;
 
 	if (intel_dsi->dual_link) {
 		mode_hdisplay /= 2;
@@ -832,7 +832,7 @@ static void intel_dsi_prepare(struct intel_encoder *intel_encoder)
 		I915_WRITE(MIPI_DPHY_PARAM(port), intel_dsi->dphy_reg);
 
 		I915_WRITE(MIPI_DPI_RESOLUTION(port),
-			adjusted_mode->vdisplay << VERTICAL_ADDRESS_SHIFT |
+			adjusted_mode->crtc_vdisplay << VERTICAL_ADDRESS_SHIFT |
 			mode_hdisplay << HORIZONTAL_ADDRESS_SHIFT);
 	}
 
@@ -878,13 +878,13 @@ static void intel_dsi_prepare(struct intel_encoder *intel_encoder)
 		if (is_vid_mode(intel_dsi) &&
 			intel_dsi->video_mode_format == VIDEO_MODE_BURST) {
 			I915_WRITE(MIPI_HS_TX_TIMEOUT(port),
-				txbyteclkhs(adjusted_mode->htotal, bpp,
+				txbyteclkhs(adjusted_mode->crtc_htotal, bpp,
 					    intel_dsi->lane_count,
 					    intel_dsi->burst_mode_ratio) + 1);
 		} else {
 			I915_WRITE(MIPI_HS_TX_TIMEOUT(port),
-				txbyteclkhs(adjusted_mode->vtotal *
-					    adjusted_mode->htotal,
+				txbyteclkhs(adjusted_mode->crtc_vtotal *
+					    adjusted_mode->crtc_htotal,
 					    bpp, intel_dsi->lane_count,
 					    intel_dsi->burst_mode_ratio) + 1);
 		}
diff --git a/drivers/gpu/drm/i915/intel_dvo.c b/drivers/gpu/drm/i915/intel_dvo.c
index 0bc8aa8..555afbc 100644
--- a/drivers/gpu/drm/i915/intel_dvo.c
+++ b/drivers/gpu/drm/i915/intel_dvo.c
@@ -285,11 +285,11 @@ static void intel_dvo_pre_enable(struct intel_encoder *encoder)
 		dvo_val |= DVO_VSYNC_ACTIVE_HIGH;
 
 	/*I915_WRITE(DVOB_SRCDIM,
-	  (adjusted_mode->hdisplay << DVO_SRCDIM_HORIZONTAL_SHIFT) |
-	  (adjusted_mode->VDisplay << DVO_SRCDIM_VERTICAL_SHIFT));*/
+	  (adjusted_mode->crtc_hdisplay << DVO_SRCDIM_HORIZONTAL_SHIFT) |
+	  (adjusted_mode->crtc_vdisplay << DVO_SRCDIM_VERTICAL_SHIFT));*/
 	I915_WRITE(dvo_srcdim_reg,
-		   (adjusted_mode->hdisplay << DVO_SRCDIM_HORIZONTAL_SHIFT) |
-		   (adjusted_mode->vdisplay << DVO_SRCDIM_VERTICAL_SHIFT));
+		   (adjusted_mode->crtc_hdisplay << DVO_SRCDIM_HORIZONTAL_SHIFT) |
+		   (adjusted_mode->crtc_vdisplay << DVO_SRCDIM_VERTICAL_SHIFT));
 	/*I915_WRITE(DVOB, dvo_val);*/
 	I915_WRITE(dvo_reg, dvo_val);
 }
diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c
index 4294f60..2806049 100644
--- a/drivers/gpu/drm/i915/intel_panel.c
+++ b/drivers/gpu/drm/i915/intel_panel.c
@@ -113,51 +113,51 @@ intel_pch_panel_fitting(struct intel_crtc *intel_crtc,
 	x = y = width = height = 0;
 
 	/* Native modes don't need fitting */
-	if (adjusted_mode->hdisplay == pipe_config->pipe_src_w &&
-	    adjusted_mode->vdisplay == pipe_config->pipe_src_h)
+	if (adjusted_mode->crtc_hdisplay == pipe_config->pipe_src_w &&
+	    adjusted_mode->crtc_vdisplay == pipe_config->pipe_src_h)
 		goto done;
 
 	switch (fitting_mode) {
 	case DRM_MODE_SCALE_CENTER:
 		width = pipe_config->pipe_src_w;
 		height = pipe_config->pipe_src_h;
-		x = (adjusted_mode->hdisplay - width + 1)/2;
-		y = (adjusted_mode->vdisplay - height + 1)/2;
+		x = (adjusted_mode->crtc_hdisplay - width + 1)/2;
+		y = (adjusted_mode->crtc_vdisplay - height + 1)/2;
 		break;
 
 	case DRM_MODE_SCALE_ASPECT:
 		/* Scale but preserve the aspect ratio */
 		{
-			u32 scaled_width = adjusted_mode->hdisplay
+			u32 scaled_width = adjusted_mode->crtc_hdisplay
 				* pipe_config->pipe_src_h;
 			u32 scaled_height = pipe_config->pipe_src_w
-				* adjusted_mode->vdisplay;
+				* adjusted_mode->crtc_vdisplay;
 			if (scaled_width > scaled_height) { /* pillar */
 				width = scaled_height / pipe_config->pipe_src_h;
 				if (width & 1)
 					width++;
-				x = (adjusted_mode->hdisplay - width + 1) / 2;
+				x = (adjusted_mode->crtc_hdisplay - width + 1) / 2;
 				y = 0;
-				height = adjusted_mode->vdisplay;
+				height = adjusted_mode->crtc_vdisplay;
 			} else if (scaled_width < scaled_height) { /* letter */
 				height = scaled_width / pipe_config->pipe_src_w;
 				if (height & 1)
 				    height++;
-				y = (adjusted_mode->vdisplay - height + 1) / 2;
+				y = (adjusted_mode->crtc_vdisplay - height + 1) / 2;
 				x = 0;
-				width = adjusted_mode->hdisplay;
+				width = adjusted_mode->crtc_hdisplay;
 			} else {
 				x = y = 0;
-				width = adjusted_mode->hdisplay;
-				height = adjusted_mode->vdisplay;
+				width = adjusted_mode->crtc_hdisplay;
+				height = adjusted_mode->crtc_vdisplay;
 			}
 		}
 		break;
 
 	case DRM_MODE_SCALE_FULLSCREEN:
 		x = y = 0;
-		width = adjusted_mode->hdisplay;
-		height = adjusted_mode->vdisplay;
+		width = adjusted_mode->crtc_hdisplay;
+		height = adjusted_mode->crtc_vdisplay;
 		break;
 
 	default:
@@ -182,7 +182,7 @@ centre_horizontally(struct drm_display_mode *adjusted_mode,
 	blank_width = adjusted_mode->crtc_hblank_end - adjusted_mode->crtc_hblank_start;
 	sync_pos = (blank_width - sync_width + 1) / 2;
 
-	border = (adjusted_mode->hdisplay - width + 1) / 2;
+	border = (adjusted_mode->crtc_hdisplay - width + 1) / 2;
 	border += border & 1; /* make the border even */
 
 	adjusted_mode->crtc_hdisplay = width;
@@ -204,7 +204,7 @@ centre_vertically(struct drm_display_mode *adjusted_mode,
 	blank_width = adjusted_mode->crtc_vblank_end - adjusted_mode->crtc_vblank_start;
 	sync_pos = (blank_width - sync_width + 1) / 2;
 
-	border = (adjusted_mode->vdisplay - height + 1) / 2;
+	border = (adjusted_mode->crtc_vdisplay - height + 1) / 2;
 
 	adjusted_mode->crtc_vdisplay = height;
 	adjusted_mode->crtc_vblank_start = height + border;
@@ -231,10 +231,10 @@ static void i965_scale_aspect(struct intel_crtc_state *pipe_config,
 			      u32 *pfit_control)
 {
 	struct drm_display_mode *adjusted_mode = &pipe_config->base.adjusted_mode;
-	u32 scaled_width = adjusted_mode->hdisplay *
+	u32 scaled_width = adjusted_mode->crtc_hdisplay *
 		pipe_config->pipe_src_h;
 	u32 scaled_height = pipe_config->pipe_src_w *
-		adjusted_mode->vdisplay;
+		adjusted_mode->crtc_vdisplay;
 
 	/* 965+ is easy, it does everything in hw */
 	if (scaled_width > scaled_height)
@@ -243,7 +243,7 @@ static void i965_scale_aspect(struct intel_crtc_state *pipe_config,
 	else if (scaled_width < scaled_height)
 		*pfit_control |= PFIT_ENABLE |
 			PFIT_SCALING_LETTER;
-	else if (adjusted_mode->hdisplay != pipe_config->pipe_src_w)
+	else if (adjusted_mode->crtc_hdisplay != pipe_config->pipe_src_w)
 		*pfit_control |= PFIT_ENABLE | PFIT_SCALING_AUTO;
 }
 
@@ -252,10 +252,10 @@ static void i9xx_scale_aspect(struct intel_crtc_state *pipe_config,
 			      u32 *border)
 {
 	struct drm_display_mode *adjusted_mode = &pipe_config->base.adjusted_mode;
-	u32 scaled_width = adjusted_mode->hdisplay *
+	u32 scaled_width = adjusted_mode->crtc_hdisplay *
 		pipe_config->pipe_src_h;
 	u32 scaled_height = pipe_config->pipe_src_w *
-		adjusted_mode->vdisplay;
+		adjusted_mode->crtc_vdisplay;
 	u32 bits;
 
 	/*
@@ -269,9 +269,9 @@ static void i9xx_scale_aspect(struct intel_crtc_state *pipe_config,
 				    pipe_config->pipe_src_h);
 
 		*border = LVDS_BORDER_ENABLE;
-		if (pipe_config->pipe_src_h != adjusted_mode->vdisplay) {
+		if (pipe_config->pipe_src_h != adjusted_mode->crtc_vdisplay) {
 			bits = panel_fitter_scaling(pipe_config->pipe_src_h,
-						    adjusted_mode->vdisplay);
+						    adjusted_mode->crtc_vdisplay);
 
 			*pfit_pgm_ratios |= (bits << PFIT_HORIZ_SCALE_SHIFT |
 					     bits << PFIT_VERT_SCALE_SHIFT);
@@ -285,9 +285,9 @@ static void i9xx_scale_aspect(struct intel_crtc_state *pipe_config,
 				  pipe_config->pipe_src_w);
 
 		*border = LVDS_BORDER_ENABLE;
-		if (pipe_config->pipe_src_w != adjusted_mode->hdisplay) {
+		if (pipe_config->pipe_src_w != adjusted_mode->crtc_hdisplay) {
 			bits = panel_fitter_scaling(pipe_config->pipe_src_w,
-						    adjusted_mode->hdisplay);
+						    adjusted_mode->crtc_hdisplay);
 
 			*pfit_pgm_ratios |= (bits << PFIT_HORIZ_SCALE_SHIFT |
 					     bits << PFIT_VERT_SCALE_SHIFT);
@@ -315,8 +315,8 @@ void intel_gmch_panel_fitting(struct intel_crtc *intel_crtc,
 	adjusted_mode = &pipe_config->base.adjusted_mode;
 
 	/* Native modes don't need fitting */
-	if (adjusted_mode->hdisplay == pipe_config->pipe_src_w &&
-	    adjusted_mode->vdisplay == pipe_config->pipe_src_h)
+	if (adjusted_mode->crtc_hdisplay == pipe_config->pipe_src_w &&
+	    adjusted_mode->crtc_vdisplay == pipe_config->pipe_src_h)
 		goto out;
 
 	switch (fitting_mode) {
@@ -342,8 +342,8 @@ void intel_gmch_panel_fitting(struct intel_crtc *intel_crtc,
 		 * Full scaling, even if it changes the aspect ratio.
 		 * Fortunately this is all done for us in hw.
 		 */
-		if (pipe_config->pipe_src_h != adjusted_mode->vdisplay ||
-		    pipe_config->pipe_src_w != adjusted_mode->hdisplay) {
+		if (pipe_config->pipe_src_h != adjusted_mode->crtc_vdisplay ||
+		    pipe_config->pipe_src_w != adjusted_mode->crtc_hdisplay) {
 			pfit_control |= PFIT_ENABLE;
 			if (INTEL_INFO(dev)->gen >= 4)
 				pfit_control |= PFIT_SCALING_AUTO;
diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c
index ab75b61..9df6bbd 100644
--- a/drivers/gpu/drm/i915/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/intel_sdvo.c
@@ -605,9 +605,9 @@ log_fail:
 
 static int intel_sdvo_get_pixel_multiplier(const struct drm_display_mode *adjusted_mode)
 {
-	if (adjusted_mode->clock >= 100000)
+	if (adjusted_mode->crtc_clock >= 100000)
 		return 1;
-	else if (adjusted_mode->clock >= 50000)
+	else if (adjusted_mode->crtc_clock >= 50000)
 		return 2;
 	else
 		return 4;
-- 
cgit v0.10.2


From 28b468a008888928ce59bc2528713b913be60baa Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 8 Sep 2015 13:40:48 +0300
Subject: drm/i915: Move HDMI aspect ratio setup to .compute_config()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We shouldn't frob adjusted_mode after .compute_config(), so move the
infoframe aspect ratio setup to .compute_config() from
intel_hdmi_set_avi_infoframe().

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index c11703a..08f8399 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -454,9 +454,6 @@ static void intel_hdmi_set_avi_infoframe(struct drm_encoder *encoder,
 	union hdmi_infoframe frame;
 	int ret;
 
-	/* Set user selected PAR to incoming mode's member */
-	adjusted_mode->picture_aspect_ratio = intel_hdmi->aspect_ratio;
-
 	ret = drm_hdmi_avi_infoframe_from_display_mode(&frame.avi,
 						       adjusted_mode);
 	if (ret < 0) {
@@ -1312,6 +1309,9 @@ bool intel_hdmi_compute_config(struct intel_encoder *encoder,
 		return false;
 	}
 
+	/* Set user selected PAR to incoming mode's member */
+	adjusted_mode->picture_aspect_ratio = intel_hdmi->aspect_ratio;
+
 	return true;
 }
 
-- 
cgit v0.10.2


From 7c5f93b05efc72feb206348da00a0a77e631a3c2 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 8 Sep 2015 13:40:49 +0300
Subject: drm/i915: Constify adjusted_mode
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Make adjusted_mode const whereever we don't have to modify it. This only
covers cases when we have a local adjusted_mode variable, and doesn't
make any difference for cases where we just dereference
pipe_config->adjusted_mode.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/dvo.h b/drivers/gpu/drm/i915/dvo.h
index 3121633..0e2c1b9 100644
--- a/drivers/gpu/drm/i915/dvo.h
+++ b/drivers/gpu/drm/i915/dvo.h
@@ -94,8 +94,8 @@ struct intel_dvo_dev_ops {
 	 * after this function is called.
 	 */
 	void (*mode_set)(struct intel_dvo_device *dvo,
-			 struct drm_display_mode *mode,
-			 struct drm_display_mode *adjusted_mode);
+			 const struct drm_display_mode *mode,
+			 const struct drm_display_mode *adjusted_mode);
 
 	/*
 	 * Probe for a connected output, and return detect_status.
diff --git a/drivers/gpu/drm/i915/dvo_ch7017.c b/drivers/gpu/drm/i915/dvo_ch7017.c
index 86b27d1..cbb2202 100644
--- a/drivers/gpu/drm/i915/dvo_ch7017.c
+++ b/drivers/gpu/drm/i915/dvo_ch7017.c
@@ -255,8 +255,8 @@ static enum drm_mode_status ch7017_mode_valid(struct intel_dvo_device *dvo,
 }
 
 static void ch7017_mode_set(struct intel_dvo_device *dvo,
-			    struct drm_display_mode *mode,
-			    struct drm_display_mode *adjusted_mode)
+			    const struct drm_display_mode *mode,
+			    const struct drm_display_mode *adjusted_mode)
 {
 	uint8_t lvds_pll_feedback_div, lvds_pll_vco_control;
 	uint8_t outputs_enable, lvds_control_2, lvds_power_down;
diff --git a/drivers/gpu/drm/i915/dvo_ch7xxx.c b/drivers/gpu/drm/i915/dvo_ch7xxx.c
index 80449f4..4b4acc1 100644
--- a/drivers/gpu/drm/i915/dvo_ch7xxx.c
+++ b/drivers/gpu/drm/i915/dvo_ch7xxx.c
@@ -275,8 +275,8 @@ static enum drm_mode_status ch7xxx_mode_valid(struct intel_dvo_device *dvo,
 }
 
 static void ch7xxx_mode_set(struct intel_dvo_device *dvo,
-			    struct drm_display_mode *mode,
-			    struct drm_display_mode *adjusted_mode)
+			    const struct drm_display_mode *mode,
+			    const struct drm_display_mode *adjusted_mode)
 {
 	uint8_t tvco, tpcp, tpd, tlpf, idf;
 
diff --git a/drivers/gpu/drm/i915/dvo_ivch.c b/drivers/gpu/drm/i915/dvo_ivch.c
index e082f75..ff9f1b0 100644
--- a/drivers/gpu/drm/i915/dvo_ivch.c
+++ b/drivers/gpu/drm/i915/dvo_ivch.c
@@ -394,8 +394,8 @@ static bool ivch_get_hw_state(struct intel_dvo_device *dvo)
 }
 
 static void ivch_mode_set(struct intel_dvo_device *dvo,
-			  struct drm_display_mode *mode,
-			  struct drm_display_mode *adjusted_mode)
+			  const struct drm_display_mode *mode,
+			  const struct drm_display_mode *adjusted_mode)
 {
 	struct ivch_priv *priv = dvo->dev_priv;
 	uint16_t vr40 = 0;
diff --git a/drivers/gpu/drm/i915/dvo_ns2501.c b/drivers/gpu/drm/i915/dvo_ns2501.c
index 97ae8aa..063859f 100644
--- a/drivers/gpu/drm/i915/dvo_ns2501.c
+++ b/drivers/gpu/drm/i915/dvo_ns2501.c
@@ -546,8 +546,8 @@ static enum drm_mode_status ns2501_mode_valid(struct intel_dvo_device *dvo,
 }
 
 static void ns2501_mode_set(struct intel_dvo_device *dvo,
-			    struct drm_display_mode *mode,
-			    struct drm_display_mode *adjusted_mode)
+			    const struct drm_display_mode *mode,
+			    const struct drm_display_mode *adjusted_mode)
 {
 	const struct ns2501_configuration *conf;
 	struct ns2501_priv *ns = (struct ns2501_priv *)(dvo->dev_priv);
diff --git a/drivers/gpu/drm/i915/dvo_sil164.c b/drivers/gpu/drm/i915/dvo_sil164.c
index fa01149..26f13eb 100644
--- a/drivers/gpu/drm/i915/dvo_sil164.c
+++ b/drivers/gpu/drm/i915/dvo_sil164.c
@@ -190,8 +190,8 @@ static enum drm_mode_status sil164_mode_valid(struct intel_dvo_device *dvo,
 }
 
 static void sil164_mode_set(struct intel_dvo_device *dvo,
-			    struct drm_display_mode *mode,
-			    struct drm_display_mode *adjusted_mode)
+			    const struct drm_display_mode *mode,
+			    const struct drm_display_mode *adjusted_mode)
 {
 	/* As long as the basics are set up, since we don't have clock
 	 * dependencies in the mode setup, we can just leave the
diff --git a/drivers/gpu/drm/i915/dvo_tfp410.c b/drivers/gpu/drm/i915/dvo_tfp410.c
index 7853719..6f1a0a6 100644
--- a/drivers/gpu/drm/i915/dvo_tfp410.c
+++ b/drivers/gpu/drm/i915/dvo_tfp410.c
@@ -222,8 +222,8 @@ static enum drm_mode_status tfp410_mode_valid(struct intel_dvo_device *dvo,
 }
 
 static void tfp410_mode_set(struct intel_dvo_device *dvo,
-			    struct drm_display_mode *mode,
-			    struct drm_display_mode *adjusted_mode)
+			    const struct drm_display_mode *mode,
+			    const struct drm_display_mode *adjusted_mode)
 {
 	/* As long as the basics are set up, since we don't have clock dependencies
 	* in the mode setup, we can just leave the registers alone and everything
diff --git a/drivers/gpu/drm/i915/intel_audio.c b/drivers/gpu/drm/i915/intel_audio.c
index e4c30e8..b08b4a7 100644
--- a/drivers/gpu/drm/i915/intel_audio.c
+++ b/drivers/gpu/drm/i915/intel_audio.c
@@ -397,7 +397,7 @@ void intel_audio_codec_enable(struct intel_encoder *intel_encoder)
 {
 	struct drm_encoder *encoder = &intel_encoder->base;
 	struct intel_crtc *crtc = to_intel_crtc(encoder->crtc);
-	struct drm_display_mode *adjusted_mode = &crtc->config->base.adjusted_mode;
+	const struct drm_display_mode *adjusted_mode = &crtc->config->base.adjusted_mode;
 	struct drm_connector *connector;
 	struct drm_device *dev = encoder->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
diff --git a/drivers/gpu/drm/i915/intel_crt.c b/drivers/gpu/drm/i915/intel_crt.c
index 6ce38e3..cac7928 100644
--- a/drivers/gpu/drm/i915/intel_crt.c
+++ b/drivers/gpu/drm/i915/intel_crt.c
@@ -158,7 +158,7 @@ static void intel_crt_set_dpms(struct intel_encoder *encoder, int mode)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crt *crt = intel_encoder_to_crt(encoder);
 	struct intel_crtc *crtc = to_intel_crtc(encoder->base.crtc);
-	struct drm_display_mode *adjusted_mode = &crtc->config->base.adjusted_mode;
+	const struct drm_display_mode *adjusted_mode = &crtc->config->base.adjusted_mode;
 	u32 adpa;
 
 	if (INTEL_INFO(dev)->gen >= 5)
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 5105e09..47bf13f 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -4392,7 +4392,7 @@ skl_update_scaler(struct intel_crtc_state *crtc_state, bool force_detach,
 int skl_update_scaler_crtc(struct intel_crtc_state *state)
 {
 	struct intel_crtc *intel_crtc = to_intel_crtc(state->base.crtc);
-	struct drm_display_mode *adjusted_mode = &state->base.adjusted_mode;
+	const struct drm_display_mode *adjusted_mode = &state->base.adjusted_mode;
 
 	DRM_DEBUG_KMS("Updating scaler for [CRTC:%i] scaler_user index %u.%u\n",
 		      intel_crtc->base.base.id, intel_crtc->pipe, SKL_CRTC_INDEX);
@@ -6477,7 +6477,7 @@ static int ironlake_fdi_compute_config(struct intel_crtc *intel_crtc,
 				       struct intel_crtc_state *pipe_config)
 {
 	struct drm_device *dev = intel_crtc->base.dev;
-	struct drm_display_mode *adjusted_mode = &pipe_config->base.adjusted_mode;
+	const struct drm_display_mode *adjusted_mode = &pipe_config->base.adjusted_mode;
 	int lane, link_bw, fdi_dotclock, ret;
 	bool needs_recompute = false;
 
@@ -6556,7 +6556,7 @@ static int intel_crtc_compute_config(struct intel_crtc *crtc,
 {
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct drm_display_mode *adjusted_mode = &pipe_config->base.adjusted_mode;
+	const struct drm_display_mode *adjusted_mode = &pipe_config->base.adjusted_mode;
 
 	/* FIXME should check pixel clock limits on all platforms */
 	if (INTEL_INFO(dev)->gen < 4) {
@@ -7624,7 +7624,7 @@ static void intel_set_pipe_timings(struct intel_crtc *intel_crtc)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	enum pipe pipe = intel_crtc->pipe;
 	enum transcoder cpu_transcoder = intel_crtc->config->cpu_transcoder;
-	struct drm_display_mode *adjusted_mode = &intel_crtc->config->base.adjusted_mode;
+	const struct drm_display_mode *adjusted_mode = &intel_crtc->config->base.adjusted_mode;
 	uint32_t crtc_vtotal, crtc_vblank_end;
 	int vsyncshift = 0;
 
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 06a2b10..77f7330 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -1587,7 +1587,7 @@ static void intel_dp_prepare(struct intel_encoder *encoder)
 	struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base);
 	enum port port = dp_to_dig_port(intel_dp)->port;
 	struct intel_crtc *crtc = to_intel_crtc(encoder->base.crtc);
-	struct drm_display_mode *adjusted_mode = &crtc->config->base.adjusted_mode;
+	const struct drm_display_mode *adjusted_mode = &crtc->config->base.adjusted_mode;
 
 	intel_dp_set_link_params(intel_dp, crtc->config);
 
diff --git a/drivers/gpu/drm/i915/intel_dp_mst.c b/drivers/gpu/drm/i915/intel_dp_mst.c
index cccf014..ca4d022 100644
--- a/drivers/gpu/drm/i915/intel_dp_mst.c
+++ b/drivers/gpu/drm/i915/intel_dp_mst.c
@@ -40,7 +40,7 @@ static bool intel_dp_mst_compute_config(struct intel_encoder *encoder,
 	struct drm_atomic_state *state;
 	int bpp, i;
 	int lane_count, slots;
-	struct drm_display_mode *adjusted_mode = &pipe_config->base.adjusted_mode;
+	const struct drm_display_mode *adjusted_mode = &pipe_config->base.adjusted_mode;
 	struct drm_connector *drm_connector;
 	struct intel_connector *connector, *found = NULL;
 	struct drm_connector_state *connector_state;
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index c96289d..c98ed81 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -683,7 +683,7 @@ struct intel_hdmi {
 				const void *frame, ssize_t len);
 	void (*set_infoframes)(struct drm_encoder *encoder,
 			       bool enable,
-			       struct drm_display_mode *adjusted_mode);
+			       const struct drm_display_mode *adjusted_mode);
 	bool (*infoframe_enabled)(struct drm_encoder *encoder);
 };
 
diff --git a/drivers/gpu/drm/i915/intel_dsi.c b/drivers/gpu/drm/i915/intel_dsi.c
index 4fb97c0..eb86ba7 100644
--- a/drivers/gpu/drm/i915/intel_dsi.c
+++ b/drivers/gpu/drm/i915/intel_dsi.c
@@ -774,7 +774,7 @@ static void intel_dsi_prepare(struct intel_encoder *intel_encoder)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc);
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder);
-	struct drm_display_mode *adjusted_mode = &intel_crtc->config->base.adjusted_mode;
+	const struct drm_display_mode *adjusted_mode = &intel_crtc->config->base.adjusted_mode;
 	enum port port;
 	unsigned int bpp = intel_crtc->config->pipe_bpp;
 	u32 val, tmp;
diff --git a/drivers/gpu/drm/i915/intel_dvo.c b/drivers/gpu/drm/i915/intel_dvo.c
index 555afbc..8492053 100644
--- a/drivers/gpu/drm/i915/intel_dvo.c
+++ b/drivers/gpu/drm/i915/intel_dvo.c
@@ -251,7 +251,7 @@ static void intel_dvo_pre_enable(struct intel_encoder *encoder)
 	struct drm_device *dev = encoder->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *crtc = to_intel_crtc(encoder->base.crtc);
-	struct drm_display_mode *adjusted_mode = &crtc->config->base.adjusted_mode;
+	const struct drm_display_mode *adjusted_mode = &crtc->config->base.adjusted_mode;
 	struct intel_dvo *intel_dvo = enc_to_dvo(encoder);
 	int pipe = crtc->pipe;
 	u32 dvo_val;
diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index 08f8399..af231ed 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -447,7 +447,7 @@ static void intel_write_infoframe(struct drm_encoder *encoder,
 }
 
 static void intel_hdmi_set_avi_infoframe(struct drm_encoder *encoder,
-					 struct drm_display_mode *adjusted_mode)
+					 const struct drm_display_mode *adjusted_mode)
 {
 	struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder);
 	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc);
@@ -491,7 +491,7 @@ static void intel_hdmi_set_spd_infoframe(struct drm_encoder *encoder)
 
 static void
 intel_hdmi_set_hdmi_infoframe(struct drm_encoder *encoder,
-			      struct drm_display_mode *adjusted_mode)
+			      const struct drm_display_mode *adjusted_mode)
 {
 	union hdmi_infoframe frame;
 	int ret;
@@ -506,7 +506,7 @@ intel_hdmi_set_hdmi_infoframe(struct drm_encoder *encoder,
 
 static void g4x_set_infoframes(struct drm_encoder *encoder,
 			       bool enable,
-			       struct drm_display_mode *adjusted_mode)
+			       const struct drm_display_mode *adjusted_mode)
 {
 	struct drm_i915_private *dev_priv = encoder->dev->dev_private;
 	struct intel_digital_port *intel_dig_port = enc_to_dig_port(encoder);
@@ -658,7 +658,7 @@ static bool intel_hdmi_set_gcp_infoframe(struct drm_encoder *encoder)
 
 static void ibx_set_infoframes(struct drm_encoder *encoder,
 			       bool enable,
-			       struct drm_display_mode *adjusted_mode)
+			       const struct drm_display_mode *adjusted_mode)
 {
 	struct drm_i915_private *dev_priv = encoder->dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc);
@@ -710,7 +710,7 @@ static void ibx_set_infoframes(struct drm_encoder *encoder,
 
 static void cpt_set_infoframes(struct drm_encoder *encoder,
 			       bool enable,
-			       struct drm_display_mode *adjusted_mode)
+			       const struct drm_display_mode *adjusted_mode)
 {
 	struct drm_i915_private *dev_priv = encoder->dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc);
@@ -752,7 +752,7 @@ static void cpt_set_infoframes(struct drm_encoder *encoder,
 
 static void vlv_set_infoframes(struct drm_encoder *encoder,
 			       bool enable,
-			       struct drm_display_mode *adjusted_mode)
+			       const struct drm_display_mode *adjusted_mode)
 {
 	struct drm_i915_private *dev_priv = encoder->dev->dev_private;
 	struct intel_digital_port *intel_dig_port = enc_to_dig_port(encoder);
@@ -804,7 +804,7 @@ static void vlv_set_infoframes(struct drm_encoder *encoder,
 
 static void hsw_set_infoframes(struct drm_encoder *encoder,
 			       bool enable,
-			       struct drm_display_mode *adjusted_mode)
+			       const struct drm_display_mode *adjusted_mode)
 {
 	struct drm_i915_private *dev_priv = encoder->dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc);
@@ -841,7 +841,7 @@ static void intel_hdmi_prepare(struct intel_encoder *encoder)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *crtc = to_intel_crtc(encoder->base.crtc);
 	struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(&encoder->base);
-	struct drm_display_mode *adjusted_mode = &crtc->config->base.adjusted_mode;
+	const struct drm_display_mode *adjusted_mode = &crtc->config->base.adjusted_mode;
 	u32 hdmi_val;
 
 	hdmi_val = SDVO_ENCODING_HDMI;
@@ -1537,7 +1537,7 @@ static void intel_hdmi_pre_enable(struct intel_encoder *encoder)
 {
 	struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(&encoder->base);
 	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->base.crtc);
-	struct drm_display_mode *adjusted_mode = &intel_crtc->config->base.adjusted_mode;
+	const struct drm_display_mode *adjusted_mode = &intel_crtc->config->base.adjusted_mode;
 
 	intel_hdmi_prepare(encoder);
 
@@ -1554,7 +1554,7 @@ static void vlv_hdmi_pre_enable(struct intel_encoder *encoder)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc =
 		to_intel_crtc(encoder->base.crtc);
-	struct drm_display_mode *adjusted_mode = &intel_crtc->config->base.adjusted_mode;
+	const struct drm_display_mode *adjusted_mode = &intel_crtc->config->base.adjusted_mode;
 	enum dpio_channel port = vlv_dport_to_channel(dport);
 	int pipe = intel_crtc->pipe;
 	u32 val;
@@ -1820,7 +1820,7 @@ static void chv_hdmi_pre_enable(struct intel_encoder *encoder)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc =
 		to_intel_crtc(encoder->base.crtc);
-	struct drm_display_mode *adjusted_mode = &intel_crtc->config->base.adjusted_mode;
+	const struct drm_display_mode *adjusted_mode = &intel_crtc->config->base.adjusted_mode;
 	enum dpio_channel ch = vlv_dport_to_channel(dport);
 	int pipe = intel_crtc->pipe;
 	int data, i, stagger;
diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c
index 2806049..dd71e7f 100644
--- a/drivers/gpu/drm/i915/intel_panel.c
+++ b/drivers/gpu/drm/i915/intel_panel.c
@@ -105,12 +105,8 @@ intel_pch_panel_fitting(struct intel_crtc *intel_crtc,
 			struct intel_crtc_state *pipe_config,
 			int fitting_mode)
 {
-	struct drm_display_mode *adjusted_mode;
-	int x, y, width, height;
-
-	adjusted_mode = &pipe_config->base.adjusted_mode;
-
-	x = y = width = height = 0;
+	const struct drm_display_mode *adjusted_mode = &pipe_config->base.adjusted_mode;
+	int x = 0, y = 0, width = 0, height = 0;
 
 	/* Native modes don't need fitting */
 	if (adjusted_mode->crtc_hdisplay == pipe_config->pipe_src_w &&
@@ -230,7 +226,7 @@ static inline u32 panel_fitter_scaling(u32 source, u32 target)
 static void i965_scale_aspect(struct intel_crtc_state *pipe_config,
 			      u32 *pfit_control)
 {
-	struct drm_display_mode *adjusted_mode = &pipe_config->base.adjusted_mode;
+	const struct drm_display_mode *adjusted_mode = &pipe_config->base.adjusted_mode;
 	u32 scaled_width = adjusted_mode->crtc_hdisplay *
 		pipe_config->pipe_src_h;
 	u32 scaled_height = pipe_config->pipe_src_w *
@@ -310,9 +306,7 @@ void intel_gmch_panel_fitting(struct intel_crtc *intel_crtc,
 {
 	struct drm_device *dev = intel_crtc->base.dev;
 	u32 pfit_control = 0, pfit_pgm_ratios = 0, border = 0;
-	struct drm_display_mode *adjusted_mode;
-
-	adjusted_mode = &pipe_config->base.adjusted_mode;
+	struct drm_display_mode *adjusted_mode = &pipe_config->base.adjusted_mode;
 
 	/* Native modes don't need fitting */
 	if (adjusted_mode->crtc_hdisplay == pipe_config->pipe_src_w &&
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index d5c60bb..7da3358 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -703,12 +703,9 @@ static void pineview_update_wm(struct drm_crtc *unused_crtc)
 
 	crtc = single_enabled_crtc(dev);
 	if (crtc) {
-		const struct drm_display_mode *adjusted_mode;
+		const struct drm_display_mode *adjusted_mode = &to_intel_crtc(crtc)->config->base.adjusted_mode;
 		int pixel_size = crtc->primary->state->fb->bits_per_pixel / 8;
-		int clock;
-
-		adjusted_mode = &to_intel_crtc(crtc)->config->base.adjusted_mode;
-		clock = adjusted_mode->crtc_clock;
+		int clock = adjusted_mode->crtc_clock;
 
 		/* Display SR */
 		wm = intel_calculate_wm(clock, &pineview_display_wm,
@@ -2086,7 +2083,7 @@ hsw_compute_linetime_wm(struct drm_device *dev, struct drm_crtc *crtc)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	struct drm_display_mode *adjusted_mode = &intel_crtc->config->base.adjusted_mode;
+	const struct drm_display_mode *adjusted_mode = &intel_crtc->config->base.adjusted_mode;
 	u32 linetime, ips_linetime;
 
 	if (!intel_crtc->active)
diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c
index 9df6bbd..857b1bd 100644
--- a/drivers/gpu/drm/i915/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/intel_sdvo.c
@@ -1189,7 +1189,7 @@ static void intel_sdvo_pre_enable(struct intel_encoder *intel_encoder)
 	struct drm_device *dev = intel_encoder->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *crtc = to_intel_crtc(intel_encoder->base.crtc);
-	struct drm_display_mode *adjusted_mode = &crtc->config->base.adjusted_mode;
+	const struct drm_display_mode *adjusted_mode = &crtc->config->base.adjusted_mode;
 	struct drm_display_mode *mode = &crtc->config->base.mode;
 	struct intel_sdvo *intel_sdvo = to_sdvo(intel_encoder);
 	u32 sdvox;
-- 
cgit v0.10.2


From 7949dd47ba58e0fb7effd4651afa4468ba240460 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 25 Sep 2015 16:39:30 +0300
Subject: drm/i915: Add HDMI aspect ratio property for SDVO
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Handle the HDMI aspect ratio property the same way in the SDVO code
as we handle it in the HDMI code.

v2: Remove stray whitespace change

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index c98ed81..ed66a4f 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -1300,6 +1300,7 @@ int intel_connector_update_modes(struct drm_connector *connector,
 int intel_ddc_get_modes(struct drm_connector *c, struct i2c_adapter *adapter);
 void intel_attach_force_audio_property(struct drm_connector *connector);
 void intel_attach_broadcast_rgb_property(struct drm_connector *connector);
+void intel_attach_aspect_ratio_property(struct drm_connector *connector);
 
 
 /* intel_overlay.c */
diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index af231ed..3c5f2c9 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -2004,15 +2004,6 @@ static const struct drm_encoder_funcs intel_hdmi_enc_funcs = {
 };
 
 static void
-intel_attach_aspect_ratio_property(struct drm_connector *connector)
-{
-	if (!drm_mode_create_aspect_ratio_property(connector->dev))
-		drm_object_attach_property(&connector->base,
-			connector->dev->mode_config.aspect_ratio_property,
-			DRM_MODE_PICTURE_ASPECT_NONE);
-}
-
-static void
 intel_hdmi_add_properties(struct intel_hdmi *intel_hdmi, struct drm_connector *connector)
 {
 	intel_attach_force_audio_property(connector);
diff --git a/drivers/gpu/drm/i915/intel_modes.c b/drivers/gpu/drm/i915/intel_modes.c
index 0e860f3..38a4c8c 100644
--- a/drivers/gpu/drm/i915/intel_modes.c
+++ b/drivers/gpu/drm/i915/intel_modes.c
@@ -126,3 +126,12 @@ intel_attach_broadcast_rgb_property(struct drm_connector *connector)
 
 	drm_object_attach_property(&connector->base, prop, 0);
 }
+
+void
+intel_attach_aspect_ratio_property(struct drm_connector *connector)
+{
+	if (!drm_mode_create_aspect_ratio_property(connector->dev))
+		drm_object_attach_property(&connector->base,
+			connector->dev->mode_config.aspect_ratio_property,
+			DRM_MODE_PICTURE_ASPECT_NONE);
+}
diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c
index 857b1bd..c42b636 100644
--- a/drivers/gpu/drm/i915/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/intel_sdvo.c
@@ -107,6 +107,11 @@ struct intel_sdvo {
 	bool color_range_auto;
 
 	/**
+	 * HDMI user specified aspect ratio
+	 */
+	enum hdmi_picture_aspect aspect_ratio;
+
+	/**
 	 * This is set if we're going to treat the device as TV-out.
 	 *
 	 * While we have these nice friendly flags for output types that ought
@@ -1181,6 +1186,10 @@ static bool intel_sdvo_compute_config(struct intel_encoder *encoder,
 	if (intel_sdvo->is_tv)
 		i9xx_adjust_sdvo_tv_clock(pipe_config);
 
+	/* Set user selected PAR to incoming mode's member */
+	if (intel_sdvo->is_hdmi)
+		adjusted_mode->picture_aspect_ratio = intel_sdvo->aspect_ratio;
+
 	return true;
 }
 
@@ -2043,6 +2052,23 @@ intel_sdvo_set_property(struct drm_connector *connector,
 		goto done;
 	}
 
+	if (property == connector->dev->mode_config.aspect_ratio_property) {
+		switch (val) {
+		case DRM_MODE_PICTURE_ASPECT_NONE:
+			intel_sdvo->aspect_ratio = HDMI_PICTURE_ASPECT_NONE;
+			break;
+		case DRM_MODE_PICTURE_ASPECT_4_3:
+			intel_sdvo->aspect_ratio = HDMI_PICTURE_ASPECT_4_3;
+			break;
+		case DRM_MODE_PICTURE_ASPECT_16_9:
+			intel_sdvo->aspect_ratio = HDMI_PICTURE_ASPECT_16_9;
+			break;
+		default:
+			return -EINVAL;
+		}
+		goto done;
+	}
+
 #define CHECK_PROPERTY(name, NAME) \
 	if (intel_sdvo_connector->name == property) { \
 		if (intel_sdvo_connector->cur_##name == temp_value) return 0; \
@@ -2382,6 +2408,8 @@ intel_sdvo_add_hdmi_properties(struct intel_sdvo *intel_sdvo,
 		intel_attach_broadcast_rgb_property(&connector->base.base);
 		intel_sdvo->color_range_auto = true;
 	}
+	intel_attach_aspect_ratio_property(&connector->base.base);
+	intel_sdvo->aspect_ratio = HDMI_PICTURE_ASPECT_NONE;
 }
 
 static struct intel_sdvo_connector *intel_sdvo_connector_alloc(void)
-- 
cgit v0.10.2


From ab9cc558a3d793b8643846940f16faa440f7d32b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 18 Sep 2015 20:03:24 +0300
Subject: drm/i915: Parametrize UOS_RSA_SCRATCH
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_guc_reg.h b/drivers/gpu/drm/i915/i915_guc_reg.h
index b355661..c4cb1c0 100644
--- a/drivers/gpu/drm/i915/i915_guc_reg.h
+++ b/drivers/gpu/drm/i915/i915_guc_reg.h
@@ -41,7 +41,7 @@
 
 #define SOFT_SCRATCH(n)			(0xc180 + ((n) * 4))
 
-#define UOS_RSA_SCRATCH_0		0xc200
+#define UOS_RSA_SCRATCH(i)		(0xc200 + (i) * 4)
 #define DMA_ADDR_0_LOW			0xc300
 #define DMA_ADDR_0_HIGH			0xc304
 #define DMA_ADDR_1_LOW			0xc308
diff --git a/drivers/gpu/drm/i915/intel_guc_loader.c b/drivers/gpu/drm/i915/intel_guc_loader.c
index 40241f3..5d17b63 100644
--- a/drivers/gpu/drm/i915/intel_guc_loader.c
+++ b/drivers/gpu/drm/i915/intel_guc_loader.c
@@ -258,7 +258,7 @@ static int guc_ucode_xfer_dma(struct drm_i915_private *dev_priv)
 	/* Copy RSA signature from the fw image to HW for verification */
 	sg_pcopy_to_buffer(sg->sgl, sg->nents, rsa, UOS_RSA_SIG_SIZE, offset);
 	for (i = 0; i < UOS_RSA_SIG_SIZE / sizeof(u32); i++)
-		I915_WRITE(UOS_RSA_SCRATCH_0 + i * sizeof(u32), rsa[i]);
+		I915_WRITE(UOS_RSA_SCRATCH(i), rsa[i]);
 
 	/* Set the source address for the new blob */
 	offset = i915_gem_obj_ggtt_offset(fw_obj);
-- 
cgit v0.10.2


From 7e435ad2c263cf3fd51343b0754856f2d88b0cdb Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 18 Sep 2015 20:03:25 +0300
Subject: drm/i915: Add LO/HI PRIVATE_PAT registers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 47344d0..620d57e 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -2889,8 +2889,8 @@ static void bdw_setup_private_ppat(struct drm_i915_private *dev_priv)
 
 	/* XXX: spec defines this as 2 distinct registers. It's unclear if a 64b
 	 * write would work. */
-	I915_WRITE(GEN8_PRIVATE_PAT, pat);
-	I915_WRITE(GEN8_PRIVATE_PAT + 4, pat >> 32);
+	I915_WRITE(GEN8_PRIVATE_PAT_LO, pat);
+	I915_WRITE(GEN8_PRIVATE_PAT_HI, pat >> 32);
 }
 
 static void chv_setup_private_ppat(struct drm_i915_private *dev_priv)
@@ -2924,8 +2924,8 @@ static void chv_setup_private_ppat(struct drm_i915_private *dev_priv)
 	      GEN8_PPAT(6, CHV_PPAT_SNOOP) |
 	      GEN8_PPAT(7, CHV_PPAT_SNOOP);
 
-	I915_WRITE(GEN8_PRIVATE_PAT, pat);
-	I915_WRITE(GEN8_PRIVATE_PAT + 4, pat >> 32);
+	I915_WRITE(GEN8_PRIVATE_PAT_LO, pat);
+	I915_WRITE(GEN8_PRIVATE_PAT_HI, pat >> 32);
 }
 
 static int gen8_gmch_probe(struct drm_device *dev,
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index d5a0790..39b9213 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -1551,7 +1551,8 @@ enum skl_disp_power_wells {
 #define   RING_FAULT_FAULT_TYPE(x) ((x >> 1) & 0x3)
 #define   RING_FAULT_VALID	(1<<0)
 #define DONE_REG		0x40b0
-#define GEN8_PRIVATE_PAT	0x40e0
+#define GEN8_PRIVATE_PAT_LO	0x40e0
+#define GEN8_PRIVATE_PAT_HI	(0x40e0 + 4)
 #define BSD_HWS_PGA_GEN7	(0x04180)
 #define BLT_HWS_PGA_GEN7	(0x04280)
 #define VEBOX_HWS_PGA_GEN7	(0x04380)
-- 
cgit v0.10.2


From d3a93cbe47bfb49de2d517d68c7082cef176dbb9 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 18 Sep 2015 20:03:26 +0300
Subject: drm/i915: Always use GEN8_RING_PDP_{LDW, UDW} instead of hand rolling
 the register offsets
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index 8797717..007d892 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -2231,10 +2231,9 @@ static void gen8_ppgtt_info(struct seq_file *m, struct drm_device *dev)
 	for_each_ring(ring, dev_priv, unused) {
 		seq_printf(m, "%s\n", ring->name);
 		for (i = 0; i < 4; i++) {
-			u32 offset = 0x270 + i * 8;
-			u64 pdp = I915_READ(ring->mmio_base + offset + 4);
+			u64 pdp = I915_READ(GEN8_RING_PDP_UDW(ring, i));
 			pdp <<= 32;
-			pdp |= I915_READ(ring->mmio_base + offset);
+			pdp |= I915_READ(GEN8_RING_PDP_LDW(ring, i));
 			seq_printf(m, "\tPDP%d 0x%016llx\n", i, pdp);
 		}
 	}
-- 
cgit v0.10.2


From c039b7f2bf2e872b7b85d089ecf2f3cef215952e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 18 Sep 2015 20:03:27 +0300
Subject: drm/i915: Include MCHBAR_MIRROR_BASE in ILK_GDSR
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 39b9213..08d78fc 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -105,7 +105,7 @@
 #define  GRDOM_RESET_STATUS (1<<1)
 #define  GRDOM_RESET_ENABLE (1<<0)
 
-#define ILK_GDSR 0x2ca4 /* MCHBAR offset */
+#define ILK_GDSR (MCHBAR_MIRROR_BASE + 0x2ca4)
 #define  ILK_GRDOM_FULL		(0<<1)
 #define  ILK_GRDOM_RENDER	(1<<1)
 #define  ILK_GRDOM_MEDIA	(3<<1)
diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c
index 14d0831..b43c6d0 100644
--- a/drivers/gpu/drm/i915/intel_uncore.c
+++ b/drivers/gpu/drm/i915/intel_uncore.c
@@ -1429,21 +1429,21 @@ static int ironlake_do_reset(struct drm_device *dev)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int ret;
 
-	I915_WRITE(MCHBAR_MIRROR_BASE + ILK_GDSR,
+	I915_WRITE(ILK_GDSR,
 		   ILK_GRDOM_RENDER | ILK_GRDOM_RESET_ENABLE);
-	ret = wait_for((I915_READ(MCHBAR_MIRROR_BASE + ILK_GDSR) &
+	ret = wait_for((I915_READ(ILK_GDSR) &
 			ILK_GRDOM_RESET_ENABLE) == 0, 500);
 	if (ret)
 		return ret;
 
-	I915_WRITE(MCHBAR_MIRROR_BASE + ILK_GDSR,
+	I915_WRITE(ILK_GDSR,
 		   ILK_GRDOM_MEDIA | ILK_GRDOM_RESET_ENABLE);
-	ret = wait_for((I915_READ(MCHBAR_MIRROR_BASE + ILK_GDSR) &
+	ret = wait_for((I915_READ(ILK_GDSR) &
 			ILK_GRDOM_RESET_ENABLE) == 0, 500);
 	if (ret)
 		return ret;
 
-	I915_WRITE(MCHBAR_MIRROR_BASE + ILK_GDSR, 0);
+	I915_WRITE(ILK_GDSR, 0);
 
 	return 0;
 }
-- 
cgit v0.10.2


From f65a9c5bd7287b4e545b07a3551a1b05c1468349 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 18 Sep 2015 20:03:28 +0300
Subject: drm/i915: Parametrize PALETTE and LGC_PALETTE
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 08d78fc..2c3f8b3 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -2485,8 +2485,8 @@ enum skl_disp_power_wells {
 #define PALETTE_A_OFFSET 0xa000
 #define PALETTE_B_OFFSET 0xa800
 #define CHV_PALETTE_C_OFFSET 0xc000
-#define PALETTE(pipe) (dev_priv->info.palette_offsets[pipe] + \
-		       dev_priv->info.display_mmio_offset)
+#define PALETTE(pipe, i) (dev_priv->info.palette_offsets[pipe] + \
+			  dev_priv->info.display_mmio_offset + (i) * 4)
 
 /* MCH MMIO space */
 
@@ -5641,7 +5641,7 @@ enum skl_disp_power_wells {
 /* legacy palette */
 #define _LGC_PALETTE_A           0x4a000
 #define _LGC_PALETTE_B           0x4a800
-#define LGC_PALETTE(pipe) _PIPE(pipe, _LGC_PALETTE_A, _LGC_PALETTE_B)
+#define LGC_PALETTE(pipe, i) (_PIPE(pipe, _LGC_PALETTE_A, _LGC_PALETTE_B) + (i) * 4)
 
 #define _GAMMA_MODE_A		0x4a480
 #define _GAMMA_MODE_B		0x4ac80
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 47bf13f..203406b 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -4593,7 +4593,6 @@ static void intel_crtc_load_lut(struct drm_crtc *crtc)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	enum pipe pipe = intel_crtc->pipe;
-	int palreg = PALETTE(pipe);
 	int i;
 	bool reenable_ips = false;
 
@@ -4608,10 +4607,6 @@ static void intel_crtc_load_lut(struct drm_crtc *crtc)
 			assert_pll_enabled(dev_priv, pipe);
 	}
 
-	/* use legacy palette for Ironlake */
-	if (!HAS_GMCH_DISPLAY(dev))
-		palreg = LGC_PALETTE(pipe);
-
 	/* Workaround : Do not read or write the pipe palette/gamma data while
 	 * GAMMA_MODE is configured for split gamma and IPS_CTL has IPS enabled.
 	 */
@@ -4623,7 +4618,14 @@ static void intel_crtc_load_lut(struct drm_crtc *crtc)
 	}
 
 	for (i = 0; i < 256; i++) {
-		I915_WRITE(palreg + 4 * i,
+		u32 palreg;
+
+		if (HAS_GMCH_DISPLAY(dev))
+			palreg = PALETTE(pipe, i);
+		else
+			palreg = LGC_PALETTE(pipe, i);
+
+		I915_WRITE(palreg,
 			   (intel_crtc->lut_r[i] << 16) |
 			   (intel_crtc->lut_g[i] << 8) |
 			   intel_crtc->lut_b[i]);
-- 
cgit v0.10.2


From 0b87c24ea5ec9e950aaa0c933fa739a95aa43555 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 22 Sep 2015 19:47:51 +0300
Subject: drm/i915: s/_CURACNTR/CURCNTR(PIPE_A)/
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

v2: Deal with _CURABASE too

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index 007d892..7d88ecf 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -2909,7 +2909,7 @@ static bool cursor_active(struct drm_device *dev, int pipe)
 	u32 state;
 
 	if (IS_845G(dev) || IS_I865G(dev))
-		state = I915_READ(_CURACNTR) & CURSOR_ENABLE;
+		state = I915_READ(CURCNTR(PIPE_A)) & CURSOR_ENABLE;
 	else
 		state = I915_READ(CURCNTR(pipe)) & CURSOR_MODE;
 
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 203406b..86f5a6e 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -1295,7 +1295,7 @@ static void assert_cursor(struct drm_i915_private *dev_priv,
 	bool cur_state;
 
 	if (IS_845G(dev) || IS_I865G(dev))
-		cur_state = I915_READ(_CURACNTR) & CURSOR_ENABLE;
+		cur_state = I915_READ(CURCNTR(PIPE_A)) & CURSOR_ENABLE;
 	else
 		cur_state = I915_READ(CURCNTR(pipe)) & CURSOR_MODE;
 
@@ -9890,13 +9890,13 @@ static void i845_update_cursor(struct drm_crtc *crtc, u32 base)
 		/* On these chipsets we can only modify the base/size/stride
 		 * whilst the cursor is disabled.
 		 */
-		I915_WRITE(_CURACNTR, 0);
-		POSTING_READ(_CURACNTR);
+		I915_WRITE(CURCNTR(PIPE_A), 0);
+		POSTING_READ(CURCNTR(PIPE_A));
 		intel_crtc->cursor_cntl = 0;
 	}
 
 	if (intel_crtc->cursor_base != base) {
-		I915_WRITE(_CURABASE, base);
+		I915_WRITE(CURBASE(PIPE_A), base);
 		intel_crtc->cursor_base = base;
 	}
 
@@ -9906,8 +9906,8 @@ static void i845_update_cursor(struct drm_crtc *crtc, u32 base)
 	}
 
 	if (intel_crtc->cursor_cntl != cntl) {
-		I915_WRITE(_CURACNTR, cntl);
-		POSTING_READ(_CURACNTR);
+		I915_WRITE(CURCNTR(PIPE_A), cntl);
+		POSTING_READ(CURCNTR(PIPE_A));
 		intel_crtc->cursor_cntl = cntl;
 	}
 }
-- 
cgit v0.10.2


From e66eb81de2ff8228cc888946f2c1e307d5b19373 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 18 Sep 2015 20:03:34 +0300
Subject: drm/i915: Add VLV_HDMIB etc. which already include VLV_DISPLAY_BASE
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 2c3f8b3..59124a5 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -3274,7 +3274,9 @@ enum skl_disp_power_wells {
 #define GEN3_SDVOC	0x61160
 #define GEN4_HDMIB	GEN3_SDVOB
 #define GEN4_HDMIC	GEN3_SDVOC
-#define CHV_HDMID	0x6116C
+#define VLV_HDMIB	(VLV_DISPLAY_BASE + GEN4_HDMIB)
+#define VLV_HDMIC	(VLV_DISPLAY_BASE + GEN4_HDMIC)
+#define CHV_HDMID	(VLV_DISPLAY_BASE + 0x6116C)
 #define PCH_SDVOB	0xe1140
 #define PCH_HDMIB	PCH_SDVOB
 #define PCH_HDMIC	0xe1150
@@ -4103,6 +4105,10 @@ enum skl_disp_power_wells {
 #define DP_C				0x64200
 #define DP_D				0x64300
 
+#define VLV_DP_B			(VLV_DISPLAY_BASE + DP_B)
+#define VLV_DP_C			(VLV_DISPLAY_BASE + DP_C)
+#define CHV_DP_D			(VLV_DISPLAY_BASE + DP_D)
+
 #define   DP_PORT_EN			(1 << 31)
 #define   DP_PIPEB_SELECT		(1 << 30)
 #define   DP_PIPE_MASK			(1 << 30)
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 86f5a6e..91dba4f 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -14050,29 +14050,26 @@ static void intel_setup_outputs(struct drm_device *dev)
 		 * eDP ports. Consult the VBT as well as DP_DETECTED to
 		 * detect eDP ports.
 		 */
-		if (I915_READ(VLV_DISPLAY_BASE + GEN4_HDMIB) & SDVO_DETECTED &&
+		if (I915_READ(VLV_HDMIB) & SDVO_DETECTED &&
 		    !intel_dp_is_edp(dev, PORT_B))
-			intel_hdmi_init(dev, VLV_DISPLAY_BASE + GEN4_HDMIB,
-					PORT_B);
-		if (I915_READ(VLV_DISPLAY_BASE + DP_B) & DP_DETECTED ||
+			intel_hdmi_init(dev, VLV_HDMIB, PORT_B);
+		if (I915_READ(VLV_DP_B) & DP_DETECTED ||
 		    intel_dp_is_edp(dev, PORT_B))
-			intel_dp_init(dev, VLV_DISPLAY_BASE + DP_B, PORT_B);
+			intel_dp_init(dev, VLV_DP_B, PORT_B);
 
-		if (I915_READ(VLV_DISPLAY_BASE + GEN4_HDMIC) & SDVO_DETECTED &&
+		if (I915_READ(VLV_HDMIC) & SDVO_DETECTED &&
 		    !intel_dp_is_edp(dev, PORT_C))
-			intel_hdmi_init(dev, VLV_DISPLAY_BASE + GEN4_HDMIC,
-					PORT_C);
-		if (I915_READ(VLV_DISPLAY_BASE + DP_C) & DP_DETECTED ||
+			intel_hdmi_init(dev, VLV_HDMIC, PORT_C);
+		if (I915_READ(VLV_DP_C) & DP_DETECTED ||
 		    intel_dp_is_edp(dev, PORT_C))
-			intel_dp_init(dev, VLV_DISPLAY_BASE + DP_C, PORT_C);
+			intel_dp_init(dev, VLV_DP_C, PORT_C);
 
 		if (IS_CHERRYVIEW(dev)) {
-			if (I915_READ(VLV_DISPLAY_BASE + CHV_HDMID) & SDVO_DETECTED)
-				intel_hdmi_init(dev, VLV_DISPLAY_BASE + CHV_HDMID,
-						PORT_D);
 			/* eDP not supported on port D, so don't check VBT */
-			if (I915_READ(VLV_DISPLAY_BASE + DP_D) & DP_DETECTED)
-				intel_dp_init(dev, VLV_DISPLAY_BASE + DP_D, PORT_D);
+			if (I915_READ(CHV_HDMID) & SDVO_DETECTED)
+				intel_hdmi_init(dev, CHV_HDMID, PORT_D);
+			if (I915_READ(CHV_DP_D) & DP_DETECTED)
+				intel_dp_init(dev, CHV_DP_D, PORT_D);
 		}
 
 		intel_dsi_init(dev);
-- 
cgit v0.10.2


From 77179400cdf61f189fecb379e23347027b9ea68d Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 18 Sep 2015 20:03:35 +0300
Subject: drm/i915: s/DDI_BUF_CTL_A/DDI_BUF_CTL(PORT_A)/
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 91dba4f..7f3c482 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -13989,7 +13989,7 @@ static void intel_setup_outputs(struct drm_device *dev)
 		 * On SKL pre-D0 the strap isn't connected, so we assume
 		 * it's there.
 		 */
-		found = I915_READ(DDI_BUF_CTL_A) & DDI_INIT_DISPLAY_DETECTED;
+		found = I915_READ(DDI_BUF_CTL(PORT_A)) & DDI_INIT_DISPLAY_DETECTED;
 		/* WaIgnoreDDIAStrap: skl */
 		if (found || IS_SKYLAKE(dev))
 			intel_ddi_init(dev, PORT_A);
-- 
cgit v0.10.2


From 1170f28c0f7a450e1cb7f31ed2968e8be751b4fc Mon Sep 17 00:00:00 2001
From: Mika Kahola <mika.kahola@intel.com>
Date: Fri, 25 Sep 2015 14:00:32 +0300
Subject: drm/i915: Add CD and pixel clock information
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This patch adds information of current and maximum CD clock
frequency and pixel clock frequency information on 'i915_debugfs.c'.

v2:
- combined seperate patches for current CD clock, maximum CD clock
  and maximum pixel clock
- space added between the frequency value and the unit

Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index 7d88ecf..afa7982 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -1309,6 +1309,10 @@ static int i915_frequency_info(struct seq_file *m, void *unused)
 		seq_puts(m, "no P-state info available\n");
 	}
 
+	seq_printf(m, "Current CD clock frequency: %d kHz\n", dev_priv->cdclk_freq);
+	seq_printf(m, "Max CD clock frequency: %d kHz\n", dev_priv->max_cdclk_freq);
+	seq_printf(m, "Max pixel clock frequency: %d kHz\n", dev_priv->max_dotclk_freq);
+
 out:
 	intel_runtime_pm_put(dev_priv);
 	return ret;
-- 
cgit v0.10.2


From aa0011a896bc1d9aa35a1ed12da63dca92e7b3f1 Mon Sep 17 00:00:00 2001
From: Arun Siluvery <arun.siluvery@linux.intel.com>
Date: Fri, 25 Sep 2015 14:33:35 +0100
Subject: drm/i915/gen9: Handle error returned by gen9_init_workarounds
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Arun Siluvery <arun.siluvery@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index 16a4ead..2042bc5 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -1041,10 +1041,13 @@ static int skl_tune_iz_hashing(struct intel_engine_cs *ring)
 
 static int skl_init_workarounds(struct intel_engine_cs *ring)
 {
+	int ret;
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	gen9_init_workarounds(ring);
+	ret = gen9_init_workarounds(ring);
+	if (ret)
+		return ret;
 
 	/* WaDisablePowerCompilerClockGating:skl */
 	if (INTEL_REVID(dev) == SKL_REVID_B0)
@@ -1081,10 +1084,13 @@ static int skl_init_workarounds(struct intel_engine_cs *ring)
 
 static int bxt_init_workarounds(struct intel_engine_cs *ring)
 {
+	int ret;
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	gen9_init_workarounds(ring);
+	ret = gen9_init_workarounds(ring);
+	if (ret)
+		return ret;
 
 	/* WaDisableThreadStallDopClockGating:bxt */
 	WA_SET_BIT_MASKED(GEN8_ROW_CHICKEN,
-- 
cgit v0.10.2


From 60294683d4becb011f5e181062fd5214cf3690bd Mon Sep 17 00:00:00 2001
From: Arun Siluvery <arun.siluvery@linux.intel.com>
Date: Fri, 25 Sep 2015 14:33:37 +0100
Subject: drm/i915/gen9: Merge two WA as they part of same register
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Merge Wa4x4STCOptimizationDisable and WaDisablePartialResolveInVc to save
an entry in WA array.

Signed-off-by: Arun Siluvery <arun.siluvery@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index 2042bc5..a0b2219 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -961,10 +961,9 @@ static int gen9_init_workarounds(struct intel_engine_cs *ring)
 	}
 
 	/* Wa4x4STCOptimizationDisable:skl,bxt */
-	WA_SET_BIT_MASKED(CACHE_MODE_1, GEN8_4x4_STC_OPTIMIZATION_DISABLE);
-
 	/* WaDisablePartialResolveInVc:skl,bxt */
-	WA_SET_BIT_MASKED(CACHE_MODE_1, GEN9_PARTIAL_RESOLVE_IN_VC_DISABLE);
+	WA_SET_BIT_MASKED(CACHE_MODE_1, (GEN8_4x4_STC_OPTIMIZATION_DISABLE |
+					 GEN9_PARTIAL_RESOLVE_IN_VC_DISABLE));
 
 	/* WaCcsTlbPrefetchDisable:skl,bxt */
 	WA_CLR_BIT_MASKED(GEN9_HALF_SLICE_CHICKEN5,
-- 
cgit v0.10.2


From aa66c506f1c40af4fbb383ac35069b7baca7642c Mon Sep 17 00:00:00 2001
From: Arun Siluvery <arun.siluvery@linux.intel.com>
Date: Fri, 25 Sep 2015 14:33:40 +0100
Subject: drm/i915/bxt: Add WaStoreMultiplePTEenable name
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Updated WA with the name.

Signed-off-by: Arun Siluvery <arun.siluvery@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index 7da3358..c960559 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -127,13 +127,10 @@ static void bxt_init_clock_gating(struct drm_device *dev)
 	I915_WRITE(GEN8_UCGCTL6, I915_READ(GEN8_UCGCTL6) |
 		   GEN8_HDCUNIT_CLOCK_GATE_DISABLE_HDCREQ);
 
-	if (INTEL_REVID(dev) == BXT_REVID_A0) {
-		/*
-		 * Hardware specification requires this bit to be
-		 * set to 1 for A0
-		 */
+	/* WaStoreMultiplePTEenable:bxt */
+	/* This is a requirement according to Hardware specification */
+	if (INTEL_REVID(dev) == BXT_REVID_A0)
 		I915_WRITE(TILECTL, I915_READ(TILECTL) | TILECTL_TLBPF);
-	}
 
 	/* WaSetClckGatingDisableMedia:bxt */
 	if (INTEL_REVID(dev) == BXT_REVID_A0) {
-- 
cgit v0.10.2


From 68fc2b76a16c4dd6c768c838f91d956475e1b3df Mon Sep 17 00:00:00 2001
From: Arun Siluvery <arun.siluvery@linux.intel.com>
Date: Fri, 25 Sep 2015 14:33:42 +0100
Subject: drm/i915/skl: Remove WaDisableSDEUnitClockGating
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Dropping it because it is for pre-production stepping.

Signed-off-by: Arun Siluvery <arun.siluvery@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index c960559..eb5498a 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -73,12 +73,10 @@ static void skl_init_clock_gating(struct drm_device *dev)
 
 	if (INTEL_REVID(dev) <= SKL_REVID_B0) {
 		/*
-		 * WaDisableSDEUnitClockGating:skl
 		 * WaSetGAPSunitClckGateDisable:skl
 		 */
 		I915_WRITE(GEN8_UCGCTL6, I915_READ(GEN8_UCGCTL6) |
-			   GEN8_GAPSUNIT_CLOCK_GATE_DISABLE |
-			   GEN8_SDEUNIT_CLOCK_GATE_DISABLE);
+			   GEN8_GAPSUNIT_CLOCK_GATE_DISABLE);
 
 		/* WaDisableVFUnitClockGating:skl */
 		I915_WRITE(GEN6_UCGCTL2, I915_READ(GEN6_UCGCTL2) |
-- 
cgit v0.10.2


From 7019f0a7a68258ba05e6e5b17ef49ec547610a33 Mon Sep 17 00:00:00 2001
From: Arun Siluvery <arun.siluvery@linux.intel.com>
Date: Fri, 25 Sep 2015 14:33:43 +0100
Subject: drm/i915/skl: Remove WaSetGAPSunitClckGateDisable
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Dropping it because it is for pre-production stepping, also removed
bit definition in i915_reg as it is not used anywhere else.

Signed-off-by: Arun Siluvery <arun.siluvery@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
[danvet: Keep define as Ville suggested.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index eb5498a..ebb8356 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -72,12 +72,6 @@ static void skl_init_clock_gating(struct drm_device *dev)
 	gen9_init_clock_gating(dev);
 
 	if (INTEL_REVID(dev) <= SKL_REVID_B0) {
-		/*
-		 * WaSetGAPSunitClckGateDisable:skl
-		 */
-		I915_WRITE(GEN8_UCGCTL6, I915_READ(GEN8_UCGCTL6) |
-			   GEN8_GAPSUNIT_CLOCK_GATE_DISABLE);
-
 		/* WaDisableVFUnitClockGating:skl */
 		I915_WRITE(GEN6_UCGCTL2, I915_READ(GEN6_UCGCTL2) |
 			   GEN6_VFUNIT_CLOCK_GATE_DISABLE);
-- 
cgit v0.10.2


From 010e9f5fad88bce83dc21a864540fcbf6d17e9b3 Mon Sep 17 00:00:00 2001
From: Arun Siluvery <arun.siluvery@linux.intel.com>
Date: Fri, 25 Sep 2015 14:33:44 +0100
Subject: drm/i915/skl: Remove WaDisableVFUnitClockGating
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Dropping it because it is for pre-production stepping, also removed
bit definition in i915_reg.h as it is not used anywhere else.

Signed-off-by: Arun Siluvery <arun.siluvery@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
[danvet: Keep #define as Ville suggested.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index ebb8356..780b1e0 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -71,12 +71,6 @@ static void skl_init_clock_gating(struct drm_device *dev)
 
 	gen9_init_clock_gating(dev);
 
-	if (INTEL_REVID(dev) <= SKL_REVID_B0) {
-		/* WaDisableVFUnitClockGating:skl */
-		I915_WRITE(GEN6_UCGCTL2, I915_READ(GEN6_UCGCTL2) |
-			   GEN6_VFUNIT_CLOCK_GATE_DISABLE);
-	}
-
 	if (INTEL_REVID(dev) <= SKL_REVID_D0) {
 		/* WaDisableHDCInvalidation:skl */
 		I915_WRITE(GAM_ECOCHK, I915_READ(GAM_ECOCHK) |
-- 
cgit v0.10.2


From e9a64adaec93ce5dfcdee44bbd89c90b4d997f6f Mon Sep 17 00:00:00 2001
From: Arun Siluvery <arun.siluvery@linux.intel.com>
Date: Fri, 25 Sep 2015 17:40:37 +0100
Subject: drm/i915/gen8: Add gen8_init_workarounds for common WA
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

WA in this function should be ordered based on register address.
The following order is suggested (Ville),

instpm
mi_mode
row chicken
half slice chicken
common slice chicken
hdc chicken
cache_mode_0
cache_mode_1
gt_mode

Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Arun Siluvery <arun.siluvery@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index a0b2219..2c4f097 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -800,11 +800,22 @@ static int wa_add(struct drm_i915_private *dev_priv,
 
 #define WA_WRITE(addr, val) WA_REG(addr, 0xffffffff, val)
 
+static int gen8_init_workarounds(struct intel_engine_cs *ring)
+{
+
+	return 0;
+}
+
 static int bdw_init_workarounds(struct intel_engine_cs *ring)
 {
+	int ret;
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
+	ret = gen8_init_workarounds(ring);
+	if (ret)
+		return ret;
+
 	WA_SET_BIT_MASKED(INSTPM, INSTPM_FORCE_ORDERING);
 
 	/* WaDisableAsyncFlipPerfMode:bdw */
@@ -868,9 +879,14 @@ static int bdw_init_workarounds(struct intel_engine_cs *ring)
 
 static int chv_init_workarounds(struct intel_engine_cs *ring)
 {
+	int ret;
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
+	ret = gen8_init_workarounds(ring);
+	if (ret)
+		return ret;
+
 	WA_SET_BIT_MASKED(INSTPM, INSTPM_FORCE_ORDERING);
 
 	/* WaDisableAsyncFlipPerfMode:chv */
-- 
cgit v0.10.2


From 68c6198b33d22af5ac1649554f6b8982ea587a72 Mon Sep 17 00:00:00 2001
From: Arun Siluvery <arun.siluvery@linux.intel.com>
Date: Fri, 25 Sep 2015 17:40:38 +0100
Subject: drm/i915/gen8: Move INSTPM WA to common function
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Arun Siluvery <arun.siluvery@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index 2c4f097..76e5488 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -802,6 +802,10 @@ static int wa_add(struct drm_i915_private *dev_priv,
 
 static int gen8_init_workarounds(struct intel_engine_cs *ring)
 {
+	struct drm_device *dev = ring->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	WA_SET_BIT_MASKED(INSTPM, INSTPM_FORCE_ORDERING);
 
 	return 0;
 }
@@ -816,8 +820,6 @@ static int bdw_init_workarounds(struct intel_engine_cs *ring)
 	if (ret)
 		return ret;
 
-	WA_SET_BIT_MASKED(INSTPM, INSTPM_FORCE_ORDERING);
-
 	/* WaDisableAsyncFlipPerfMode:bdw */
 	WA_SET_BIT_MASKED(MI_MODE, ASYNC_FLIP_PERF_DISABLE);
 
@@ -887,8 +889,6 @@ static int chv_init_workarounds(struct intel_engine_cs *ring)
 	if (ret)
 		return ret;
 
-	WA_SET_BIT_MASKED(INSTPM, INSTPM_FORCE_ORDERING);
-
 	/* WaDisableAsyncFlipPerfMode:chv */
 	WA_SET_BIT_MASKED(MI_MODE, ASYNC_FLIP_PERF_DISABLE);
 
-- 
cgit v0.10.2


From 717d84d67e3a95f440c37c7482681b3535fdc7e2 Mon Sep 17 00:00:00 2001
From: Arun Siluvery <arun.siluvery@linux.intel.com>
Date: Fri, 25 Sep 2015 17:40:39 +0100
Subject: drm/i915/gen8: Move WaDisableAsyncFlipPerfMode to common init fn
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Arun Siluvery <arun.siluvery@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index 76e5488..ccb8c18 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -807,6 +807,9 @@ static int gen8_init_workarounds(struct intel_engine_cs *ring)
 
 	WA_SET_BIT_MASKED(INSTPM, INSTPM_FORCE_ORDERING);
 
+	/* WaDisableAsyncFlipPerfMode:bdw,chv */
+	WA_SET_BIT_MASKED(MI_MODE, ASYNC_FLIP_PERF_DISABLE);
+
 	return 0;
 }
 
@@ -820,9 +823,6 @@ static int bdw_init_workarounds(struct intel_engine_cs *ring)
 	if (ret)
 		return ret;
 
-	/* WaDisableAsyncFlipPerfMode:bdw */
-	WA_SET_BIT_MASKED(MI_MODE, ASYNC_FLIP_PERF_DISABLE);
-
 	/* WaDisablePartialInstShootdown:bdw */
 	/* WaDisableThreadStallDopClockGating:bdw (pre-production) */
 	WA_SET_BIT_MASKED(GEN8_ROW_CHICKEN,
@@ -889,9 +889,6 @@ static int chv_init_workarounds(struct intel_engine_cs *ring)
 	if (ret)
 		return ret;
 
-	/* WaDisableAsyncFlipPerfMode:chv */
-	WA_SET_BIT_MASKED(MI_MODE, ASYNC_FLIP_PERF_DISABLE);
-
 	/* WaDisablePartialInstShootdown:chv */
 	/* WaDisableThreadStallDopClockGating:chv */
 	WA_SET_BIT_MASKED(GEN8_ROW_CHICKEN,
-- 
cgit v0.10.2


From d0581194566eb9744fa657812b2693556392451b Mon Sep 17 00:00:00 2001
From: Arun Siluvery <arun.siluvery@linux.intel.com>
Date: Fri, 25 Sep 2015 17:40:40 +0100
Subject: drm/i915/gen8: Move WaDisablePartialInstShootdown to common init fn
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Arun Siluvery <arun.siluvery@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index ccb8c18..8c1d17a 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -810,6 +810,10 @@ static int gen8_init_workarounds(struct intel_engine_cs *ring)
 	/* WaDisableAsyncFlipPerfMode:bdw,chv */
 	WA_SET_BIT_MASKED(MI_MODE, ASYNC_FLIP_PERF_DISABLE);
 
+	/* WaDisablePartialInstShootdown:bdw,chv */
+	WA_SET_BIT_MASKED(GEN8_ROW_CHICKEN,
+			  PARTIAL_INSTRUCTION_SHOOTDOWN_DISABLE);
+
 	return 0;
 }
 
@@ -823,11 +827,8 @@ static int bdw_init_workarounds(struct intel_engine_cs *ring)
 	if (ret)
 		return ret;
 
-	/* WaDisablePartialInstShootdown:bdw */
 	/* WaDisableThreadStallDopClockGating:bdw (pre-production) */
-	WA_SET_BIT_MASKED(GEN8_ROW_CHICKEN,
-			  PARTIAL_INSTRUCTION_SHOOTDOWN_DISABLE |
-			  STALL_DOP_GATING_DISABLE);
+	WA_SET_BIT_MASKED(GEN8_ROW_CHICKEN, STALL_DOP_GATING_DISABLE);
 
 	/* WaDisableDopClockGating:bdw */
 	WA_SET_BIT_MASKED(GEN7_ROW_CHICKEN2,
@@ -889,11 +890,8 @@ static int chv_init_workarounds(struct intel_engine_cs *ring)
 	if (ret)
 		return ret;
 
-	/* WaDisablePartialInstShootdown:chv */
 	/* WaDisableThreadStallDopClockGating:chv */
-	WA_SET_BIT_MASKED(GEN8_ROW_CHICKEN,
-			  PARTIAL_INSTRUCTION_SHOOTDOWN_DISABLE |
-			  STALL_DOP_GATING_DISABLE);
+	WA_SET_BIT_MASKED(GEN8_ROW_CHICKEN, STALL_DOP_GATING_DISABLE);
 
 	/* Use Force Non-Coherent whenever executing a 3D context. This is a
 	 * workaround for a possible hang in the unlikely event a TLB
-- 
cgit v0.10.2


From 6def8fdd5d9528db7fb6dfebd994491f6ba45785 Mon Sep 17 00:00:00 2001
From: Arun Siluvery <arun.siluvery@linux.intel.com>
Date: Fri, 25 Sep 2015 17:40:42 +0100
Subject: drm/i915/gen8: Move HiZ RAW stall optimization disable WA to common
 init fn
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Arun Siluvery <arun.siluvery@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index 8c1d17a..6adc7f1 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -814,6 +814,16 @@ static int gen8_init_workarounds(struct intel_engine_cs *ring)
 	WA_SET_BIT_MASKED(GEN8_ROW_CHICKEN,
 			  PARTIAL_INSTRUCTION_SHOOTDOWN_DISABLE);
 
+	/* From the Haswell PRM, Command Reference: Registers, CACHE_MODE_0:
+	 * "The Hierarchical Z RAW Stall Optimization allows non-overlapping
+	 *  polygons in the same 8x4 pixel/sample area to be processed without
+	 *  stalling waiting for the earlier ones to write to Hierarchical Z
+	 *  buffer."
+	 *
+	 * This optimization is off by default for BDW and CHV; turn it on.
+	 */
+	WA_CLR_BIT_MASKED(CACHE_MODE_0_GEN7, HIZ_RAW_STALL_OPT_DISABLE);
+
 	return 0;
 }
 
@@ -851,16 +861,6 @@ static int bdw_init_workarounds(struct intel_engine_cs *ring)
 			  /* WaDisableFenceDestinationToSLM:bdw (pre-prod) */
 			  (IS_BDW_GT3(dev) ? HDC_FENCE_DEST_SLM_DISABLE : 0));
 
-	/* From the Haswell PRM, Command Reference: Registers, CACHE_MODE_0:
-	 * "The Hierarchical Z RAW Stall Optimization allows non-overlapping
-	 *  polygons in the same 8x4 pixel/sample area to be processed without
-	 *  stalling waiting for the earlier ones to write to Hierarchical Z
-	 *  buffer."
-	 *
-	 * This optimization is off by default for Broadwell; turn it on.
-	 */
-	WA_CLR_BIT_MASKED(CACHE_MODE_0_GEN7, HIZ_RAW_STALL_OPT_DISABLE);
-
 	/* Wa4x4STCOptimizationDisable:bdw */
 	WA_SET_BIT_MASKED(CACHE_MODE_1,
 			  GEN8_4x4_STC_OPTIMIZATION_DISABLE);
@@ -903,11 +903,6 @@ static int chv_init_workarounds(struct intel_engine_cs *ring)
 			  HDC_FORCE_NON_COHERENT |
 			  HDC_DONOT_FETCH_MEM_WHEN_MASKED);
 
-	/* According to the CACHE_MODE_0 default value documentation, some
-	 * CHV platforms disable this optimization by default.  Turn it on.
-	 */
-	WA_CLR_BIT_MASKED(CACHE_MODE_0_GEN7, HIZ_RAW_STALL_OPT_DISABLE);
-
 	/* Wa4x4STCOptimizationDisable:chv */
 	WA_SET_BIT_MASKED(CACHE_MODE_1,
 			  GEN8_4x4_STC_OPTIMIZATION_DISABLE);
-- 
cgit v0.10.2


From 484046362fcee7ad0c67e1c10c4c346fb14cf420 Mon Sep 17 00:00:00 2001
From: Arun Siluvery <arun.siluvery@linux.intel.com>
Date: Fri, 25 Sep 2015 17:40:43 +0100
Subject: drm/i915/gen8: Move Wa4x4STCOptimizationDisable to common init fn
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Arun Siluvery <arun.siluvery@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index 6adc7f1..d72e072 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -824,6 +824,9 @@ static int gen8_init_workarounds(struct intel_engine_cs *ring)
 	 */
 	WA_CLR_BIT_MASKED(CACHE_MODE_0_GEN7, HIZ_RAW_STALL_OPT_DISABLE);
 
+	/* Wa4x4STCOptimizationDisable:bdw,chv */
+	WA_SET_BIT_MASKED(CACHE_MODE_1, GEN8_4x4_STC_OPTIMIZATION_DISABLE);
+
 	return 0;
 }
 
@@ -861,10 +864,6 @@ static int bdw_init_workarounds(struct intel_engine_cs *ring)
 			  /* WaDisableFenceDestinationToSLM:bdw (pre-prod) */
 			  (IS_BDW_GT3(dev) ? HDC_FENCE_DEST_SLM_DISABLE : 0));
 
-	/* Wa4x4STCOptimizationDisable:bdw */
-	WA_SET_BIT_MASKED(CACHE_MODE_1,
-			  GEN8_4x4_STC_OPTIMIZATION_DISABLE);
-
 	/*
 	 * BSpec recommends 8x4 when MSAA is used,
 	 * however in practice 16x4 seems fastest.
@@ -903,10 +902,6 @@ static int chv_init_workarounds(struct intel_engine_cs *ring)
 			  HDC_FORCE_NON_COHERENT |
 			  HDC_DONOT_FETCH_MEM_WHEN_MASKED);
 
-	/* Wa4x4STCOptimizationDisable:chv */
-	WA_SET_BIT_MASKED(CACHE_MODE_1,
-			  GEN8_4x4_STC_OPTIMIZATION_DISABLE);
-
 	/* Improve HiZ throughput on CHV. */
 	WA_SET_BIT_MASKED(HIZ_CHICKEN, CHV_HZ_8X8_MODE_IN_1X);
 
-- 
cgit v0.10.2


From 7eebcde6e526a53e06d238f6daf788df3c8e7620 Mon Sep 17 00:00:00 2001
From: Arun Siluvery <arun.siluvery@linux.intel.com>
Date: Fri, 25 Sep 2015 17:40:44 +0100
Subject: drm/i915/gen8: Move GEN7_GT_MODE WA to common init fn
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Arun Siluvery <arun.siluvery@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index d72e072..199021c 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -827,6 +827,18 @@ static int gen8_init_workarounds(struct intel_engine_cs *ring)
 	/* Wa4x4STCOptimizationDisable:bdw,chv */
 	WA_SET_BIT_MASKED(CACHE_MODE_1, GEN8_4x4_STC_OPTIMIZATION_DISABLE);
 
+	/*
+	 * BSpec recommends 8x4 when MSAA is used,
+	 * however in practice 16x4 seems fastest.
+	 *
+	 * Note that PS/WM thread counts depend on the WIZ hashing
+	 * disable bit, which we don't touch here, but it's good
+	 * to keep in mind (see 3DSTATE_PS and 3DSTATE_WM).
+	 */
+	WA_SET_FIELD_MASKED(GEN7_GT_MODE,
+			    GEN6_WIZ_HASHING_MASK,
+			    GEN6_WIZ_HASHING_16x4);
+
 	return 0;
 }
 
@@ -864,18 +876,6 @@ static int bdw_init_workarounds(struct intel_engine_cs *ring)
 			  /* WaDisableFenceDestinationToSLM:bdw (pre-prod) */
 			  (IS_BDW_GT3(dev) ? HDC_FENCE_DEST_SLM_DISABLE : 0));
 
-	/*
-	 * BSpec recommends 8x4 when MSAA is used,
-	 * however in practice 16x4 seems fastest.
-	 *
-	 * Note that PS/WM thread counts depend on the WIZ hashing
-	 * disable bit, which we don't touch here, but it's good
-	 * to keep in mind (see 3DSTATE_PS and 3DSTATE_WM).
-	 */
-	WA_SET_FIELD_MASKED(GEN7_GT_MODE,
-			    GEN6_WIZ_HASHING_MASK,
-			    GEN6_WIZ_HASHING_16x4);
-
 	return 0;
 }
 
@@ -905,18 +905,6 @@ static int chv_init_workarounds(struct intel_engine_cs *ring)
 	/* Improve HiZ throughput on CHV. */
 	WA_SET_BIT_MASKED(HIZ_CHICKEN, CHV_HZ_8X8_MODE_IN_1X);
 
-	/*
-	 * BSpec recommends 8x4 when MSAA is used,
-	 * however in practice 16x4 seems fastest.
-	 *
-	 * Note that PS/WM thread counts depend on the WIZ hashing
-	 * disable bit, which we don't touch here, but it's good
-	 * to keep in mind (see 3DSTATE_PS and 3DSTATE_WM).
-	 */
-	WA_SET_FIELD_MASKED(GEN7_GT_MODE,
-			    GEN6_WIZ_HASHING_MASK,
-			    GEN6_WIZ_HASHING_16x4);
-
 	return 0;
 }
 
-- 
cgit v0.10.2


From a340af58730e892ccb507f7a70054f7909b28d49 Mon Sep 17 00:00:00 2001
From: Arun Siluvery <arun.siluvery@linux.intel.com>
Date: Fri, 25 Sep 2015 17:40:45 +0100
Subject: drm/i915/gen8: Move WaForceEnableNonCoherent to common init fn
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Arun Siluvery <arun.siluvery@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index 199021c..9b42157 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -814,6 +814,14 @@ static int gen8_init_workarounds(struct intel_engine_cs *ring)
 	WA_SET_BIT_MASKED(GEN8_ROW_CHICKEN,
 			  PARTIAL_INSTRUCTION_SHOOTDOWN_DISABLE);
 
+	/* Use Force Non-Coherent whenever executing a 3D context. This is a
+	 * workaround for for a possible hang in the unlikely event a TLB
+	 * invalidation occurs during a PSD flush.
+	 */
+	/* WaForceEnableNonCoherent:bdw,chv */
+	WA_SET_BIT_MASKED(HDC_CHICKEN0,
+			  HDC_FORCE_NON_COHERENT);
+
 	/* From the Haswell PRM, Command Reference: Registers, CACHE_MODE_0:
 	 * "The Hierarchical Z RAW Stall Optimization allows non-overlapping
 	 *  polygons in the same 8x4 pixel/sample area to be processed without
@@ -862,13 +870,7 @@ static int bdw_init_workarounds(struct intel_engine_cs *ring)
 	WA_SET_BIT_MASKED(HALF_SLICE_CHICKEN3,
 			  GEN8_SAMPLER_POWER_BYPASS_DIS);
 
-	/* Use Force Non-Coherent whenever executing a 3D context. This is a
-	 * workaround for for a possible hang in the unlikely event a TLB
-	 * invalidation occurs during a PSD flush.
-	 */
 	WA_SET_BIT_MASKED(HDC_CHICKEN0,
-			  /* WaForceEnableNonCoherent:bdw */
-			  HDC_FORCE_NON_COHERENT |
 			  /* WaForceContextSaveRestoreNonCoherent:bdw */
 			  HDC_FORCE_CONTEXT_SAVE_RESTORE_NON_COHERENT |
 			  /* WaHdcDisableFetchWhenMasked:bdw */
@@ -892,14 +894,8 @@ static int chv_init_workarounds(struct intel_engine_cs *ring)
 	/* WaDisableThreadStallDopClockGating:chv */
 	WA_SET_BIT_MASKED(GEN8_ROW_CHICKEN, STALL_DOP_GATING_DISABLE);
 
-	/* Use Force Non-Coherent whenever executing a 3D context. This is a
-	 * workaround for a possible hang in the unlikely event a TLB
-	 * invalidation occurs during a PSD flush.
-	 */
-	/* WaForceEnableNonCoherent:chv */
 	/* WaHdcDisableFetchWhenMasked:chv */
 	WA_SET_BIT_MASKED(HDC_CHICKEN0,
-			  HDC_FORCE_NON_COHERENT |
 			  HDC_DONOT_FETCH_MEM_WHEN_MASKED);
 
 	/* Improve HiZ throughput on CHV. */
-- 
cgit v0.10.2


From 120f5d28714b5c401179a1fd6010c86b13d912ff Mon Sep 17 00:00:00 2001
From: Arun Siluvery <arun.siluvery@linux.intel.com>
Date: Fri, 25 Sep 2015 17:40:46 +0100
Subject: drm/i915/gen8: Move WaHdcDisableFetchWhenMasked to common init fn
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Arun Siluvery <arun.siluvery@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index 9b42157..c82c74c 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -819,7 +819,9 @@ static int gen8_init_workarounds(struct intel_engine_cs *ring)
 	 * invalidation occurs during a PSD flush.
 	 */
 	/* WaForceEnableNonCoherent:bdw,chv */
+	/* WaHdcDisableFetchWhenMasked:bdw,chv */
 	WA_SET_BIT_MASKED(HDC_CHICKEN0,
+			  HDC_DONOT_FETCH_MEM_WHEN_MASKED |
 			  HDC_FORCE_NON_COHERENT);
 
 	/* From the Haswell PRM, Command Reference: Registers, CACHE_MODE_0:
@@ -873,8 +875,6 @@ static int bdw_init_workarounds(struct intel_engine_cs *ring)
 	WA_SET_BIT_MASKED(HDC_CHICKEN0,
 			  /* WaForceContextSaveRestoreNonCoherent:bdw */
 			  HDC_FORCE_CONTEXT_SAVE_RESTORE_NON_COHERENT |
-			  /* WaHdcDisableFetchWhenMasked:bdw */
-			  HDC_DONOT_FETCH_MEM_WHEN_MASKED |
 			  /* WaDisableFenceDestinationToSLM:bdw (pre-prod) */
 			  (IS_BDW_GT3(dev) ? HDC_FENCE_DEST_SLM_DISABLE : 0));
 
@@ -894,10 +894,6 @@ static int chv_init_workarounds(struct intel_engine_cs *ring)
 	/* WaDisableThreadStallDopClockGating:chv */
 	WA_SET_BIT_MASKED(GEN8_ROW_CHICKEN, STALL_DOP_GATING_DISABLE);
 
-	/* WaHdcDisableFetchWhenMasked:chv */
-	WA_SET_BIT_MASKED(HDC_CHICKEN0,
-			  HDC_DONOT_FETCH_MEM_WHEN_MASKED);
-
 	/* Improve HiZ throughput on CHV. */
 	WA_SET_BIT_MASKED(HIZ_CHICKEN, CHV_HZ_8X8_MODE_IN_1X);
 
-- 
cgit v0.10.2


From 2d05fa16fefb9922e021e0b7db4a0c515558f103 Mon Sep 17 00:00:00 2001
From: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Date: Mon, 28 Sep 2015 23:08:50 +0200
Subject: drm/i915: Fix comparison bug

->stolen->start has type u64 aka unsigned long long; relying on the
difference (effectively cast to int) for sorting is wrong.

It wouldn't be a problem in practice if the values compared are always
within INT_MAX of each other (so that the difference is actually
representable in an int), but 440fd5283a87 ("drm/mm: Support 4 GiB and
larger ranges") strongly suggests that's not the case.

Note: atm we don't support more than about 1G of stolen, so this is
impossible currenlty.

Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
[danvet: Add note that this is impossible currently.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index afa7982..9839831 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -253,7 +253,11 @@ static int obj_rank_by_stolen(void *priv,
 	struct drm_i915_gem_object *b =
 		container_of(B, struct drm_i915_gem_object, obj_exec_link);
 
-	return a->stolen->start - b->stolen->start;
+	if (a->stolen->start < b->stolen->start)
+		return -1;
+	if (a->stolen->start > b->stolen->start)
+		return 1;
+	return 0;
 }
 
 static int i915_gem_stolen_list_info(struct seq_file *m, void *data)
-- 
cgit v0.10.2


From 87d5d2593299cb8814e5f26c676aef3557650eea Mon Sep 17 00:00:00 2001
From: Vandana Kannan <vandana.kannan@intel.com>
Date: Thu, 24 Sep 2015 23:29:17 +0300
Subject: drm/i915: Renaming CCK related reg definitions
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Rename the DISPLAY_TRUNK_* and DISPLAY_FREQUENCY_* bits to CCK_... instead
of DISPLAY_... to make it clear they apply to all CCK clock control registers.
Suggested by Ville.

Signed-off-by: Vandana Kannan <vandana.kannan@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 59124a5..aca25bf 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -729,11 +729,11 @@ enum skl_disp_power_wells {
 #define  DSI_PLL_M1_DIV_SHIFT			0
 #define  DSI_PLL_M1_DIV_MASK			(0x1ff << 0)
 #define CCK_DISPLAY_CLOCK_CONTROL		0x6b
-#define  DISPLAY_TRUNK_FORCE_ON			(1 << 17)
-#define  DISPLAY_TRUNK_FORCE_OFF		(1 << 16)
-#define  DISPLAY_FREQUENCY_STATUS		(0x1f << 8)
-#define  DISPLAY_FREQUENCY_STATUS_SHIFT		8
-#define  DISPLAY_FREQUENCY_VALUES		(0x1f << 0)
+#define  CCK_TRUNK_FORCE_ON			(1 << 17)
+#define  CCK_TRUNK_FORCE_OFF			(1 << 16)
+#define  CCK_FREQUENCY_STATUS			(0x1f << 8)
+#define  CCK_FREQUENCY_STATUS_SHIFT		8
+#define  CCK_FREQUENCY_VALUES			(0x1f << 0)
 
 /**
  * DOC: DPIO
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 7f3c482..36591e3 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -5807,12 +5807,12 @@ static void valleyview_set_cdclk(struct drm_device *dev, int cdclk)
 
 		/* adjust cdclk divider */
 		val = vlv_cck_read(dev_priv, CCK_DISPLAY_CLOCK_CONTROL);
-		val &= ~DISPLAY_FREQUENCY_VALUES;
+		val &= ~CCK_FREQUENCY_VALUES;
 		val |= divider;
 		vlv_cck_write(dev_priv, CCK_DISPLAY_CLOCK_CONTROL, val);
 
 		if (wait_for((vlv_cck_read(dev_priv, CCK_DISPLAY_CLOCK_CONTROL) &
-			      DISPLAY_FREQUENCY_STATUS) == (divider << DISPLAY_FREQUENCY_STATUS_SHIFT),
+			      CCK_FREQUENCY_STATUS) == (divider << CCK_FREQUENCY_STATUS_SHIFT),
 			     50))
 			DRM_ERROR("timed out waiting for CDclk change\n");
 	}
@@ -6733,10 +6733,10 @@ static int valleyview_get_display_clock_speed(struct drm_device *dev)
 	val = vlv_cck_read(dev_priv, CCK_DISPLAY_CLOCK_CONTROL);
 	mutex_unlock(&dev_priv->sb_lock);
 
-	divider = val & DISPLAY_FREQUENCY_VALUES;
+	divider = val & CCK_FREQUENCY_VALUES;
 
-	WARN((val & DISPLAY_FREQUENCY_STATUS) !=
-	     (divider << DISPLAY_FREQUENCY_STATUS_SHIFT),
+	WARN((val & CCK_FREQUENCY_STATUS) !=
+	     (divider << CCK_FREQUENCY_STATUS_SHIFT),
 	     "cdclk change in progress\n");
 
 	return DIV_ROUND_CLOSEST(dev_priv->hpll_freq << 1, divider + 1);
-- 
cgit v0.10.2


From bfa7df01a092e92ae764d39fee07fc3f486ed6a9 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 24 Sep 2015 23:29:18 +0300
Subject: drm/i915: Read czclk from CCK on vlv/chv
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

As with the cdclk, read out czclk from CCK as well. This gives us the
real current value and avoids having to decode fuses and whatnot.

Also store it in kHz under dev_priv like we do for cdlck since it's not
just an rps related clock, and having it in kHz is more
standard/convenient for some things.

Imre also pointed out that we currently fail to read czclk on VLV, which
means the PFI credit programming isn't working as expected.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index ad37b25..4711a79 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -1146,7 +1146,6 @@ struct intel_gen6_power_mgmt {
 	u8 efficient_freq;	/* AKA RPe. Pre-determined balanced frequency */
 	u8 rp1_freq;		/* "less than" RP0 power/freqency */
 	u8 rp0_freq;		/* Non-overclocked max frequency. */
-	u32 cz_freq;
 
 	u8 up_threshold; /* Current %busy required to uplock */
 	u8 down_threshold; /* Current %busy required to downclock */
@@ -1810,6 +1809,7 @@ struct drm_i915_private {
 	unsigned int cdclk_freq, max_cdclk_freq;
 	unsigned int max_dotclk_freq;
 	unsigned int hpll_freq;
+	unsigned int czclk_freq;
 
 	/**
 	 * wq - Driver workqueue for GEM.
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index aca25bf..3709d6b 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -728,6 +728,7 @@ enum skl_disp_power_wells {
 #define  DSI_PLL_N1_DIV_MASK			(3 << 16)
 #define  DSI_PLL_M1_DIV_SHIFT			0
 #define  DSI_PLL_M1_DIV_MASK			(0x1ff << 0)
+#define CCK_CZ_CLOCK_CONTROL			0x62
 #define CCK_DISPLAY_CLOCK_CONTROL		0x6b
 #define  CCK_TRUNK_FORCE_ON			(1 << 17)
 #define  CCK_TRUNK_FORCE_OFF			(1 << 16)
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 36591e3..797b882 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -132,6 +132,42 @@ struct intel_limit {
 	intel_p2_t	    p2;
 };
 
+/* returns HPLL frequency in kHz */
+static int valleyview_get_vco(struct drm_i915_private *dev_priv)
+{
+	int hpll_freq, vco_freq[] = { 800, 1600, 2000, 2400 };
+
+	/* Obtain SKU information */
+	mutex_lock(&dev_priv->sb_lock);
+	hpll_freq = vlv_cck_read(dev_priv, CCK_FUSE_REG) &
+		CCK_FUSE_HPLL_FREQ_MASK;
+	mutex_unlock(&dev_priv->sb_lock);
+
+	return vco_freq[hpll_freq] * 1000;
+}
+
+static int vlv_get_cck_clock_hpll(struct drm_i915_private *dev_priv,
+				  const char *name, u32 reg)
+{
+	u32 val;
+	int divider;
+
+	if (dev_priv->hpll_freq == 0)
+		dev_priv->hpll_freq = valleyview_get_vco(dev_priv);
+
+	mutex_lock(&dev_priv->sb_lock);
+	val = vlv_cck_read(dev_priv, reg);
+	mutex_unlock(&dev_priv->sb_lock);
+
+	divider = val & CCK_FREQUENCY_VALUES;
+
+	WARN((val & CCK_FREQUENCY_STATUS) !=
+	     (divider << CCK_FREQUENCY_STATUS_SHIFT),
+	     "%s change in progress\n", name);
+
+	return DIV_ROUND_CLOSEST(dev_priv->hpll_freq << 1, divider + 1);
+}
+
 int
 intel_pch_rawclk(struct drm_device *dev)
 {
@@ -175,6 +211,17 @@ int intel_hrawclk(struct drm_device *dev)
 	}
 }
 
+static void intel_update_czclk(struct drm_i915_private *dev_priv)
+{
+	if (!IS_VALLEYVIEW(dev_priv))
+		return;
+
+	dev_priv->czclk_freq = vlv_get_cck_clock_hpll(dev_priv, "czclk",
+						      CCK_CZ_CLOCK_CONTROL);
+
+	DRM_DEBUG_DRIVER("CZ clock rate: %d kHz\n", dev_priv->czclk_freq);
+}
+
 static inline u32 /* units of 100MHz */
 intel_fdi_link_freq(struct drm_device *dev)
 {
@@ -5756,20 +5803,6 @@ void skl_init_cdclk(struct drm_i915_private *dev_priv)
 		DRM_ERROR("DBuf power enable timeout\n");
 }
 
-/* returns HPLL frequency in kHz */
-static int valleyview_get_vco(struct drm_i915_private *dev_priv)
-{
-	int hpll_freq, vco_freq[] = { 800, 1600, 2000, 2400 };
-
-	/* Obtain SKU information */
-	mutex_lock(&dev_priv->sb_lock);
-	hpll_freq = vlv_cck_read(dev_priv, CCK_FUSE_REG) &
-		CCK_FUSE_HPLL_FREQ_MASK;
-	mutex_unlock(&dev_priv->sb_lock);
-
-	return vco_freq[hpll_freq] * 1000;
-}
-
 /* Adjust CDclk dividers to allow high res or save power if possible */
 static void valleyview_set_cdclk(struct drm_device *dev, int cdclk)
 {
@@ -5990,7 +6023,7 @@ static void vlv_program_pfi_credits(struct drm_i915_private *dev_priv)
 	else
 		default_credits = PFI_CREDIT(8);
 
-	if (DIV_ROUND_CLOSEST(dev_priv->cdclk_freq, 1000) >= dev_priv->rps.cz_freq) {
+	if (dev_priv->cdclk_freq >= dev_priv->czclk_freq) {
 		/* CHV suggested value is 31 or 63 */
 		if (IS_CHERRYVIEW(dev_priv))
 			credits = PFI_CREDIT_63;
@@ -6722,24 +6755,8 @@ static int haswell_get_display_clock_speed(struct drm_device *dev)
 
 static int valleyview_get_display_clock_speed(struct drm_device *dev)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 val;
-	int divider;
-
-	if (dev_priv->hpll_freq == 0)
-		dev_priv->hpll_freq = valleyview_get_vco(dev_priv);
-
-	mutex_lock(&dev_priv->sb_lock);
-	val = vlv_cck_read(dev_priv, CCK_DISPLAY_CLOCK_CONTROL);
-	mutex_unlock(&dev_priv->sb_lock);
-
-	divider = val & CCK_FREQUENCY_VALUES;
-
-	WARN((val & CCK_FREQUENCY_STATUS) !=
-	     (divider << CCK_FREQUENCY_STATUS_SHIFT),
-	     "cdclk change in progress\n");
-
-	return DIV_ROUND_CLOSEST(dev_priv->hpll_freq << 1, divider + 1);
+	return vlv_get_cck_clock_hpll(to_i915(dev), "cdclk",
+				      CCK_DISPLAY_CLOCK_CONTROL);
 }
 
 static int ilk_get_display_clock_speed(struct drm_device *dev)
@@ -13329,8 +13346,6 @@ static void intel_shared_dpll_init(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	intel_update_cdclk(dev);
-
 	if (HAS_DDI(dev))
 		intel_ddi_pll_init(dev);
 	else if (HAS_PCH_IBX(dev) || HAS_PCH_CPT(dev))
@@ -14839,6 +14854,9 @@ void intel_modeset_init(struct drm_device *dev)
 		}
 	}
 
+	intel_update_czclk(dev_priv);
+	intel_update_cdclk(dev);
+
 	intel_shared_dpll_init(dev);
 
 	/* Just disable it once at startup */
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index 780b1e0..9643a7c 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -5460,25 +5460,10 @@ static void cherryview_init_gt_powersave(struct drm_device *dev)
 	mutex_unlock(&dev_priv->sb_lock);
 
 	switch ((val >> 2) & 0x7) {
-	case 0:
-	case 1:
-		dev_priv->rps.cz_freq = 200;
-		dev_priv->mem_freq = 1600;
-		break;
-	case 2:
-		dev_priv->rps.cz_freq = 267;
-		dev_priv->mem_freq = 1600;
-		break;
 	case 3:
-		dev_priv->rps.cz_freq = 333;
 		dev_priv->mem_freq = 2000;
 		break;
-	case 4:
-		dev_priv->rps.cz_freq = 320;
-		dev_priv->mem_freq = 1600;
-		break;
-	case 5:
-		dev_priv->rps.cz_freq = 400;
+	default:
 		dev_priv->mem_freq = 1600;
 		break;
 	}
@@ -7305,7 +7290,7 @@ static int vlv_gpu_freq_div(unsigned int czclk_freq)
 
 static int byt_gpu_freq(struct drm_i915_private *dev_priv, int val)
 {
-	int div, czclk_freq = DIV_ROUND_CLOSEST(dev_priv->mem_freq, 4);
+	int div, czclk_freq = DIV_ROUND_CLOSEST(dev_priv->czclk_freq, 1000);
 
 	div = vlv_gpu_freq_div(czclk_freq);
 	if (div < 0)
@@ -7316,7 +7301,7 @@ static int byt_gpu_freq(struct drm_i915_private *dev_priv, int val)
 
 static int byt_freq_opcode(struct drm_i915_private *dev_priv, int val)
 {
-	int mul, czclk_freq = DIV_ROUND_CLOSEST(dev_priv->mem_freq, 4);
+	int mul, czclk_freq = DIV_ROUND_CLOSEST(dev_priv->czclk_freq, 1000);
 
 	mul = vlv_gpu_freq_div(czclk_freq);
 	if (mul < 0)
@@ -7327,7 +7312,7 @@ static int byt_freq_opcode(struct drm_i915_private *dev_priv, int val)
 
 static int chv_gpu_freq(struct drm_i915_private *dev_priv, int val)
 {
-	int div, czclk_freq = dev_priv->rps.cz_freq;
+	int div, czclk_freq = DIV_ROUND_CLOSEST(dev_priv->czclk_freq, 1000);
 
 	div = vlv_gpu_freq_div(czclk_freq) / 2;
 	if (div < 0)
@@ -7338,7 +7323,7 @@ static int chv_gpu_freq(struct drm_i915_private *dev_priv, int val)
 
 static int chv_freq_opcode(struct drm_i915_private *dev_priv, int val)
 {
-	int mul, czclk_freq = dev_priv->rps.cz_freq;
+	int mul, czclk_freq = DIV_ROUND_CLOSEST(dev_priv->czclk_freq, 1000);
 
 	mul = vlv_gpu_freq_div(czclk_freq) / 2;
 	if (mul < 0)
-- 
cgit v0.10.2


From 2cc9fab180e8add4159ed1fc874131f984d0bcb1 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 28 Sep 2015 23:43:43 +0300
Subject: drm/i915: Simplify vlv/chv rc6 residency calculation
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We have the czclk frequency in dev_priv now, so let's just use it
when converting the rc6 counters to milliseconds. This eliminates
a bunch of hairy code that essentially tries to extract the czclk
frequency using yet another method.

v2: Fix typos in commit message (Imre)

Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_sysfs.c b/drivers/gpu/drm/i915/i915_sysfs.c
index 55bd04c..74086eb 100644
--- a/drivers/gpu/drm/i915/i915_sysfs.c
+++ b/drivers/gpu/drm/i915/i915_sysfs.c
@@ -39,7 +39,7 @@ static u32 calc_residency(struct drm_device *dev, const u32 reg)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u64 raw_time; /* 32b value may overflow during fixed point math */
-	u64 units = 128ULL, div = 100000ULL, bias = 100ULL;
+	u64 units = 128ULL, div = 100000ULL;
 	u32 ret;
 
 	if (!intel_enable_rc6(dev))
@@ -49,41 +49,16 @@ static u32 calc_residency(struct drm_device *dev, const u32 reg)
 
 	/* On VLV and CHV, residency time is in CZ units rather than 1.28us */
 	if (IS_VALLEYVIEW(dev)) {
-		u32 clk_reg, czcount_30ns;
-
-		if (IS_CHERRYVIEW(dev))
-			clk_reg = CHV_CLK_CTL1;
-		else
-			clk_reg = VLV_CLK_CTL2;
-
-		czcount_30ns = I915_READ(clk_reg) >> CLK_CTL2_CZCOUNT_30NS_SHIFT;
-
-		if (!czcount_30ns) {
-			WARN(!czcount_30ns, "bogus CZ count value");
-			ret = 0;
-			goto out;
-		}
-
-		if (IS_CHERRYVIEW(dev) && czcount_30ns == 1) {
-			/* Special case for 320Mhz */
-			div = 10000000ULL;
-			units = 3125ULL;
-		} else {
-			czcount_30ns += 1;
-			div = 1000000ULL;
-			units = DIV_ROUND_UP_ULL(30ULL * bias, czcount_30ns);
-		}
+		units = 1;
+		div = dev_priv->czclk_freq;
 
 		if (I915_READ(VLV_COUNTER_CONTROL) & VLV_COUNT_RANGE_HIGH)
 			units <<= 8;
-
-		div = div * bias;
 	}
 
 	raw_time = I915_READ(reg) * units;
 	ret = DIV_ROUND_UP_ULL(raw_time, div);
 
-out:
 	intel_runtime_pm_put(dev_priv);
 	return ret;
 }
-- 
cgit v0.10.2


From 7bad74d57c81b08218cc564af3a78427e12bf3ac Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 24 Sep 2015 23:29:20 +0300
Subject: drm/i915: Use czclk_freq in vlv c0 residency calculations
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Replace the use of mem_freq/4 with czclk_freq in the vlv c0 residency
calculations.

Also deal with VLV_COUNT_RANGE_HIGH which affects all RCx residency
counters. We have just enough bits to do this without intermediate
divisions.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index ff85eae..76bd40e 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -997,12 +997,16 @@ static bool vlv_c0_above(struct drm_i915_private *dev_priv,
 			 int threshold)
 {
 	u64 time, c0;
+	unsigned int mul = 100;
 
 	if (old->cz_clock == 0)
 		return false;
 
+	if (I915_READ(VLV_COUNTER_CONTROL) & VLV_COUNT_RANGE_HIGH)
+		mul <<= 8;
+
 	time = now->cz_clock - old->cz_clock;
-	time *= threshold * dev_priv->mem_freq;
+	time *= threshold * dev_priv->czclk_freq;
 
 	/* Workload can be split between render + media, e.g. SwapBuffers
 	 * being blitted in X after being rendered in mesa. To account for
@@ -1010,7 +1014,7 @@ static bool vlv_c0_above(struct drm_i915_private *dev_priv,
 	 */
 	c0 = now->render_c0 - old->render_c0;
 	c0 += now->media_c0 - old->media_c0;
-	c0 *= 100 * VLV_CZ_CLOCK_TO_MILLI_SEC * 4 / 1000;
+	c0 *= mul * VLV_CZ_CLOCK_TO_MILLI_SEC;
 
 	return c0 >= time;
 }
-- 
cgit v0.10.2


From 85b98a4c30c60272b2830d9acc43121eb2988bd2 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 29 Sep 2015 10:24:25 +0300
Subject: drm/i915/ddi: warn instead of oops on invalid ddi encoder type

It's more useful to limp on than bring the kernel down. Hitting this is
a more likely event with BXT DSI, although care should be taken not to
call the function for DSI.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
index 0e46679..fdcb456 100644
--- a/drivers/gpu/drm/i915/intel_ddi.c
+++ b/drivers/gpu/drm/i915/intel_ddi.c
@@ -322,8 +322,7 @@ static void ddi_get_encoder_port(struct intel_encoder *intel_encoder,
 		*dig_port = NULL;
 		*port = PORT_E;
 	} else {
-		DRM_ERROR("Invalid DDI encoder type %d\n", type);
-		BUG();
+		WARN(1, "Invalid DDI encoder type %d\n", type);
 	}
 }
 
-- 
cgit v0.10.2


From 8cd21b7f28ed1f12059935512edaaadf1deaeb67 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 29 Sep 2015 10:24:26 +0300
Subject: drm/i915/ddi: use switch case instead of if ladder for
 ddi_get_encoder_port

Make the alternatives stand  out better.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
index fdcb456..2b5a298 100644
--- a/drivers/gpu/drm/i915/intel_ddi.c
+++ b/drivers/gpu/drm/i915/intel_ddi.c
@@ -309,20 +309,26 @@ static void ddi_get_encoder_port(struct intel_encoder *intel_encoder,
 				 enum port *port)
 {
 	struct drm_encoder *encoder = &intel_encoder->base;
-	int type = intel_encoder->type;
 
-	if (type == INTEL_OUTPUT_DP_MST) {
+	switch (intel_encoder->type) {
+	case INTEL_OUTPUT_DP_MST:
 		*dig_port = enc_to_mst(encoder)->primary;
 		*port = (*dig_port)->port;
-	} else if (type == INTEL_OUTPUT_DISPLAYPORT || type == INTEL_OUTPUT_EDP ||
-	    type == INTEL_OUTPUT_HDMI || type == INTEL_OUTPUT_UNKNOWN) {
+		break;
+	case INTEL_OUTPUT_DISPLAYPORT:
+	case INTEL_OUTPUT_EDP:
+	case INTEL_OUTPUT_HDMI:
+	case INTEL_OUTPUT_UNKNOWN:
 		*dig_port = enc_to_dig_port(encoder);
 		*port = (*dig_port)->port;
-	} else if (type == INTEL_OUTPUT_ANALOG) {
+		break;
+	case INTEL_OUTPUT_ANALOG:
 		*dig_port = NULL;
 		*port = PORT_E;
-	} else {
-		WARN(1, "Invalid DDI encoder type %d\n", type);
+		break;
+	default:
+		WARN(1, "Invalid DDI encoder type %d\n", intel_encoder->type);
+		break;
 	}
 }
 
-- 
cgit v0.10.2


From 5507faeb03f87f952373314b21741548421a6944 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Mon, 14 Sep 2015 14:03:48 +0300
Subject: drm/i915: make backlight hooks connector specific

Previously we've relied on having basically one backlight and one
backlight type per platform. This is already a bit quirky with PMIC PWM
support on VLV/CHV platforms with MIPI DSI. In the foreseeable future
we'll have at least DPCD based backlight control on eDP and DCS command
based backlight control on MIPI DSI. Backlight is becoming more and more
connector specific, so reflect this fact by making the backlight control
hooks connector specific.

This enables further work to reuse generic backlight code in
intel_panel.c while adding more specific backlight code accessed via the
hooks.

Cc: Deepak M <m.deepak@intel.com>
Cc: Yetunde Adebisi <yetundex.adebisi@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Deepak M <m.deepak@intel.com>
Reviewed-by: Yetunde Adebisi <yetundex.adebisi@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 4711a79..ee478ac 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -664,15 +664,6 @@ struct drm_i915_display_funcs {
 	/* render clock increase/decrease */
 	/* display clock increase/decrease */
 	/* pll clock increase/decrease */
-
-	int (*setup_backlight)(struct intel_connector *connector, enum pipe pipe);
-	uint32_t (*get_backlight)(struct intel_connector *connector);
-	void (*set_backlight)(struct intel_connector *connector,
-			      uint32_t level);
-	void (*disable_backlight)(struct intel_connector *connector);
-	void (*enable_backlight)(struct intel_connector *connector);
-	uint32_t (*backlight_hz_to_pwm)(struct intel_connector *connector,
-					uint32_t hz);
 };
 
 enum forcewake_domain_id {
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 797b882..8e4fb80 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -14575,8 +14575,6 @@ static void intel_init_display(struct drm_device *dev)
 		dev_priv->display.queue_flip = intel_default_queue_flip;
 	}
 
-	intel_panel_init_backlight_funcs(dev);
-
 	mutex_init(&dev_priv->pps_mutex);
 }
 
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 77f7330..97ed418 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -6000,7 +6000,7 @@ static bool intel_edp_init_connector(struct intel_dp *intel_dp,
 	}
 
 	intel_panel_init(&intel_connector->panel, fixed_mode, downclock_mode);
-	intel_connector->panel.backlight_power = intel_edp_backlight_power;
+	intel_connector->panel.backlight.power = intel_edp_backlight_power;
 	intel_panel_setup_backlight(connector, pipe);
 
 	return true;
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index ed66a4f..1eaa9f9 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -182,9 +182,17 @@ struct intel_panel {
 		struct pwm_device *pwm;
 
 		struct backlight_device *device;
-	} backlight;
 
-	void (*backlight_power)(struct intel_connector *, bool enable);
+		/* Connector and platform specific backlight functions */
+		int (*setup)(struct intel_connector *connector, enum pipe pipe);
+		uint32_t (*get)(struct intel_connector *connector);
+		void (*set)(struct intel_connector *connector, uint32_t level);
+		void (*disable)(struct intel_connector *connector);
+		void (*enable)(struct intel_connector *connector);
+		uint32_t (*hz_to_pwm)(struct intel_connector *connector,
+				      uint32_t hz);
+		void (*power)(struct intel_connector *, bool enable);
+	} backlight;
 };
 
 struct intel_connector {
@@ -1333,7 +1341,6 @@ int intel_panel_setup_backlight(struct drm_connector *connector, enum pipe pipe)
 void intel_panel_enable_backlight(struct intel_connector *connector);
 void intel_panel_disable_backlight(struct intel_connector *connector);
 void intel_panel_destroy_backlight(struct drm_connector *connector);
-void intel_panel_init_backlight_funcs(struct drm_device *dev);
 enum drm_connector_status intel_panel_detect(struct drm_device *dev);
 extern struct drm_display_mode *intel_find_panel_downclock(
 				struct drm_device *dev,
diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c
index dd71e7f..4d28c7b 100644
--- a/drivers/gpu/drm/i915/intel_panel.c
+++ b/drivers/gpu/drm/i915/intel_panel.c
@@ -560,7 +560,7 @@ static u32 intel_panel_get_backlight(struct intel_connector *connector)
 	mutex_lock(&dev_priv->backlight_lock);
 
 	if (panel->backlight.enabled) {
-		val = dev_priv->display.get_backlight(connector);
+		val = panel->backlight.get(connector);
 		val = intel_panel_compute_brightness(connector, val);
 	}
 
@@ -649,13 +649,12 @@ static void pwm_set_backlight(struct intel_connector *connector, u32 level)
 static void
 intel_panel_actually_set_backlight(struct intel_connector *connector, u32 level)
 {
-	struct drm_device *dev = connector->base.dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_panel *panel = &connector->panel;
 
 	DRM_DEBUG_DRIVER("set backlight PWM = %d\n", level);
 
 	level = intel_panel_compute_brightness(connector, level);
-	dev_priv->display.set_backlight(connector, level);
+	panel->backlight.set(connector, level);
 }
 
 /* set backlight brightness to level in range [0..max], scaling wrt hw min */
@@ -830,7 +829,7 @@ void intel_panel_disable_backlight(struct intel_connector *connector)
 	if (panel->backlight.device)
 		panel->backlight.device->props.power = FB_BLANK_POWERDOWN;
 	panel->backlight.enabled = false;
-	dev_priv->display.disable_backlight(connector);
+	panel->backlight.disable(connector);
 
 	mutex_unlock(&dev_priv->backlight_lock);
 }
@@ -1079,7 +1078,7 @@ void intel_panel_enable_backlight(struct intel_connector *connector)
 						 panel->backlight.device->props.max_brightness);
 	}
 
-	dev_priv->display.enable_backlight(connector);
+	panel->backlight.enable(connector);
 	panel->backlight.enabled = true;
 	if (panel->backlight.device)
 		panel->backlight.device->props.power = FB_BLANK_UNBLANK;
@@ -1107,10 +1106,10 @@ static int intel_backlight_device_update_status(struct backlight_device *bd)
 	 * callback needs to take this into account.
 	 */
 	if (panel->backlight.enabled) {
-		if (panel->backlight_power) {
+		if (panel->backlight.power) {
 			bool enable = bd->props.power == FB_BLANK_UNBLANK &&
 				bd->props.brightness != 0;
-			panel->backlight_power(connector, enable);
+			panel->backlight.power(connector, enable);
 		}
 	} else {
 		bd->props.power = FB_BLANK_POWERDOWN;
@@ -1335,6 +1334,7 @@ static u32 get_backlight_max_vbt(struct intel_connector *connector)
 {
 	struct drm_device *dev = connector->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_panel *panel = &connector->panel;
 	u16 pwm_freq_hz = dev_priv->vbt.backlight.pwm_freq_hz;
 	u32 pwm;
 
@@ -1343,12 +1343,12 @@ static u32 get_backlight_max_vbt(struct intel_connector *connector)
 		return 0;
 	}
 
-	if (!dev_priv->display.backlight_hz_to_pwm) {
+	if (!panel->backlight.hz_to_pwm) {
 		DRM_DEBUG_KMS("backlight frequency setting from VBT currently not supported on this platform\n");
 		return 0;
 	}
 
-	pwm = dev_priv->display.backlight_hz_to_pwm(connector, pwm_freq_hz);
+	pwm = panel->backlight.hz_to_pwm(connector, pwm_freq_hz);
 	if (!pwm) {
 		DRM_DEBUG_KMS("backlight frequency conversion failed\n");
 		return 0;
@@ -1633,9 +1633,13 @@ int intel_panel_setup_backlight(struct drm_connector *connector, enum pipe pipe)
 		}
 	}
 
+	/* ensure intel_panel has been initialized first */
+	if (WARN_ON(!panel->backlight.setup))
+		return -ENODEV;
+
 	/* set level and max in panel struct */
 	mutex_lock(&dev_priv->backlight_lock);
-	ret = dev_priv->display.setup_backlight(intel_connector, pipe);
+	ret = panel->backlight.setup(intel_connector, pipe);
 	mutex_unlock(&dev_priv->backlight_lock);
 
 	if (ret) {
@@ -1667,62 +1671,66 @@ void intel_panel_destroy_backlight(struct drm_connector *connector)
 }
 
 /* Set up chip specific backlight functions */
-void intel_panel_init_backlight_funcs(struct drm_device *dev)
+static void
+intel_panel_init_backlight_funcs(struct intel_panel *panel)
 {
+	struct intel_connector *intel_connector =
+		container_of(panel, struct intel_connector, panel);
+	struct drm_device *dev = intel_connector->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
 	if (IS_BROXTON(dev)) {
-		dev_priv->display.setup_backlight = bxt_setup_backlight;
-		dev_priv->display.enable_backlight = bxt_enable_backlight;
-		dev_priv->display.disable_backlight = bxt_disable_backlight;
-		dev_priv->display.set_backlight = bxt_set_backlight;
-		dev_priv->display.get_backlight = bxt_get_backlight;
+		panel->backlight.setup = bxt_setup_backlight;
+		panel->backlight.enable = bxt_enable_backlight;
+		panel->backlight.disable = bxt_disable_backlight;
+		panel->backlight.set = bxt_set_backlight;
+		panel->backlight.get = bxt_get_backlight;
 	} else if (HAS_PCH_LPT(dev) || HAS_PCH_SPT(dev)) {
-		dev_priv->display.setup_backlight = lpt_setup_backlight;
-		dev_priv->display.enable_backlight = lpt_enable_backlight;
-		dev_priv->display.disable_backlight = lpt_disable_backlight;
-		dev_priv->display.set_backlight = lpt_set_backlight;
-		dev_priv->display.get_backlight = lpt_get_backlight;
+		panel->backlight.setup = lpt_setup_backlight;
+		panel->backlight.enable = lpt_enable_backlight;
+		panel->backlight.disable = lpt_disable_backlight;
+		panel->backlight.set = lpt_set_backlight;
+		panel->backlight.get = lpt_get_backlight;
 		if (HAS_PCH_LPT(dev))
-			dev_priv->display.backlight_hz_to_pwm = lpt_hz_to_pwm;
+			panel->backlight.hz_to_pwm = lpt_hz_to_pwm;
 		else
-			dev_priv->display.backlight_hz_to_pwm = spt_hz_to_pwm;
+			panel->backlight.hz_to_pwm = spt_hz_to_pwm;
 	} else if (HAS_PCH_SPLIT(dev)) {
-		dev_priv->display.setup_backlight = pch_setup_backlight;
-		dev_priv->display.enable_backlight = pch_enable_backlight;
-		dev_priv->display.disable_backlight = pch_disable_backlight;
-		dev_priv->display.set_backlight = pch_set_backlight;
-		dev_priv->display.get_backlight = pch_get_backlight;
-		dev_priv->display.backlight_hz_to_pwm = pch_hz_to_pwm;
+		panel->backlight.setup = pch_setup_backlight;
+		panel->backlight.enable = pch_enable_backlight;
+		panel->backlight.disable = pch_disable_backlight;
+		panel->backlight.set = pch_set_backlight;
+		panel->backlight.get = pch_get_backlight;
+		panel->backlight.hz_to_pwm = pch_hz_to_pwm;
 	} else if (IS_VALLEYVIEW(dev)) {
 		if (dev_priv->vbt.has_mipi) {
-			dev_priv->display.setup_backlight = pwm_setup_backlight;
-			dev_priv->display.enable_backlight = pwm_enable_backlight;
-			dev_priv->display.disable_backlight = pwm_disable_backlight;
-			dev_priv->display.set_backlight = pwm_set_backlight;
-			dev_priv->display.get_backlight = pwm_get_backlight;
+			panel->backlight.setup = pwm_setup_backlight;
+			panel->backlight.enable = pwm_enable_backlight;
+			panel->backlight.disable = pwm_disable_backlight;
+			panel->backlight.set = pwm_set_backlight;
+			panel->backlight.get = pwm_get_backlight;
 		} else {
-			dev_priv->display.setup_backlight = vlv_setup_backlight;
-			dev_priv->display.enable_backlight = vlv_enable_backlight;
-			dev_priv->display.disable_backlight = vlv_disable_backlight;
-			dev_priv->display.set_backlight = vlv_set_backlight;
-			dev_priv->display.get_backlight = vlv_get_backlight;
-			dev_priv->display.backlight_hz_to_pwm = vlv_hz_to_pwm;
+			panel->backlight.setup = vlv_setup_backlight;
+			panel->backlight.enable = vlv_enable_backlight;
+			panel->backlight.disable = vlv_disable_backlight;
+			panel->backlight.set = vlv_set_backlight;
+			panel->backlight.get = vlv_get_backlight;
+			panel->backlight.hz_to_pwm = vlv_hz_to_pwm;
 		}
 	} else if (IS_GEN4(dev)) {
-		dev_priv->display.setup_backlight = i965_setup_backlight;
-		dev_priv->display.enable_backlight = i965_enable_backlight;
-		dev_priv->display.disable_backlight = i965_disable_backlight;
-		dev_priv->display.set_backlight = i9xx_set_backlight;
-		dev_priv->display.get_backlight = i9xx_get_backlight;
-		dev_priv->display.backlight_hz_to_pwm = i965_hz_to_pwm;
+		panel->backlight.setup = i965_setup_backlight;
+		panel->backlight.enable = i965_enable_backlight;
+		panel->backlight.disable = i965_disable_backlight;
+		panel->backlight.set = i9xx_set_backlight;
+		panel->backlight.get = i9xx_get_backlight;
+		panel->backlight.hz_to_pwm = i965_hz_to_pwm;
 	} else {
-		dev_priv->display.setup_backlight = i9xx_setup_backlight;
-		dev_priv->display.enable_backlight = i9xx_enable_backlight;
-		dev_priv->display.disable_backlight = i9xx_disable_backlight;
-		dev_priv->display.set_backlight = i9xx_set_backlight;
-		dev_priv->display.get_backlight = i9xx_get_backlight;
-		dev_priv->display.backlight_hz_to_pwm = i9xx_hz_to_pwm;
+		panel->backlight.setup = i9xx_setup_backlight;
+		panel->backlight.enable = i9xx_enable_backlight;
+		panel->backlight.disable = i9xx_disable_backlight;
+		panel->backlight.set = i9xx_set_backlight;
+		panel->backlight.get = i9xx_get_backlight;
+		panel->backlight.hz_to_pwm = i9xx_hz_to_pwm;
 	}
 }
 
@@ -1730,6 +1738,8 @@ int intel_panel_init(struct intel_panel *panel,
 		     struct drm_display_mode *fixed_mode,
 		     struct drm_display_mode *downclock_mode)
 {
+	intel_panel_init_backlight_funcs(panel);
+
 	panel->fixed_mode = fixed_mode;
 	panel->downclock_mode = downclock_mode;
 
-- 
cgit v0.10.2


From 76b1cf211cbe120a2d6c46403abc97d4376c619e Mon Sep 17 00:00:00 2001
From: Julia Lawall <Julia.Lawall@lip6.fr>
Date: Sun, 13 Sep 2015 14:15:25 +0200
Subject: drm: i915: drop null test before destroy functions

Remove unneeded NULL test.

The semantic patch that makes this change is as follows:
(http://coccinelle.lip6.fr/)

// <smpl>
@@ expression x; @@
-if (x != NULL)
  \(kmem_cache_destroy\|mempool_destroy\|dma_pool_destroy\)(x);
// </smpl>

Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index 817b05c..ae1ba47 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -1074,12 +1074,9 @@ out_freecsr:
 put_bridge:
 	pci_dev_put(dev_priv->bridge_dev);
 free_priv:
-	if (dev_priv->requests)
-		kmem_cache_destroy(dev_priv->requests);
-	if (dev_priv->vmas)
-		kmem_cache_destroy(dev_priv->vmas);
-	if (dev_priv->objects)
-		kmem_cache_destroy(dev_priv->objects);
+	kmem_cache_destroy(dev_priv->requests);
+	kmem_cache_destroy(dev_priv->vmas);
+	kmem_cache_destroy(dev_priv->objects);
 	kfree(dev_priv);
 	return ret;
 }
@@ -1170,13 +1167,9 @@ int i915_driver_unload(struct drm_device *dev)
 	if (dev_priv->regs != NULL)
 		pci_iounmap(dev->pdev, dev_priv->regs);
 
-	if (dev_priv->requests)
-		kmem_cache_destroy(dev_priv->requests);
-	if (dev_priv->vmas)
-		kmem_cache_destroy(dev_priv->vmas);
-	if (dev_priv->objects)
-		kmem_cache_destroy(dev_priv->objects);
-
+	kmem_cache_destroy(dev_priv->requests);
+	kmem_cache_destroy(dev_priv->vmas);
+	kmem_cache_destroy(dev_priv->objects);
 	pci_dev_put(dev_priv->bridge_dev);
 	kfree(dev_priv);
 
-- 
cgit v0.10.2


From a7fb8a23c1afa607ec8ce9f61df645f37c529434 Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Wed, 9 Sep 2015 16:45:52 +0200
Subject: drm: Remove __OS_HAS_AGP
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We already express the drm/agp depencies correctly in Kconfig, so we
can rip this remnant from the shared drm core days.

Aside: Pretty much all the #ifdefs in radeon/nouveau could be killed
if ttm would provide dummy functions. I'm not going to volunteer for
that though.

v2: Use IS_ENABLED(CONFIG_AGP) as suggested by Ville

v3: Polish from Ville's review.

Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Christian König <christian.koenig@amd.com> (v2)
Reviewed-by: David Herrmann <dh.herrmann@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/Makefile b/drivers/gpu/drm/Makefile
index 45e7719..f458d6e 100644
--- a/drivers/gpu/drm/Makefile
+++ b/drivers/gpu/drm/Makefile
@@ -6,7 +6,7 @@ drm-y       :=	drm_auth.o drm_bufs.o drm_cache.o \
 		drm_context.o drm_dma.o \
 		drm_fops.o drm_gem.o drm_ioctl.o drm_irq.o \
 		drm_lock.o drm_memory.o drm_drv.o drm_vm.o \
-		drm_agpsupport.o drm_scatter.o drm_pci.o \
+		drm_scatter.o drm_pci.o \
 		drm_platform.o drm_sysfs.o drm_hashtab.o drm_mm.o \
 		drm_crtc.o drm_modes.o drm_edid.o \
 		drm_info.o drm_debugfs.o drm_encoder_slave.o \
@@ -19,6 +19,7 @@ drm-$(CONFIG_DRM_GEM_CMA_HELPER) += drm_gem_cma_helper.o
 drm-$(CONFIG_PCI) += ati_pcigart.o
 drm-$(CONFIG_DRM_PANEL) += drm_panel.o
 drm-$(CONFIG_OF) += drm_of.o
+drm-$(CONFIG_AGP) += drm_agpsupport.o
 
 drm_kms_helper-y := drm_crtc_helper.o drm_dp_helper.o drm_probe_helper.o \
 		drm_plane_helper.o drm_dp_mst_topology.o drm_atomic_helper.o
diff --git a/drivers/gpu/drm/drm_agpsupport.c b/drivers/gpu/drm/drm_agpsupport.c
index 4b2b4aa..a10ea6a 100644
--- a/drivers/gpu/drm/drm_agpsupport.c
+++ b/drivers/gpu/drm/drm_agpsupport.c
@@ -36,8 +36,6 @@
 #include <linux/slab.h>
 #include "drm_legacy.h"
 
-#if __OS_HAS_AGP
-
 #include <asm/agp.h>
 
 /**
@@ -502,5 +500,3 @@ drm_agp_bind_pages(struct drm_device *dev,
 	return mem;
 }
 EXPORT_SYMBOL(drm_agp_bind_pages);
-
-#endif /* __OS_HAS_AGP */
diff --git a/drivers/gpu/drm/drm_bufs.c b/drivers/gpu/drm/drm_bufs.c
index 569064a..f1a204d 100644
--- a/drivers/gpu/drm/drm_bufs.c
+++ b/drivers/gpu/drm/drm_bufs.c
@@ -582,7 +582,7 @@ static void drm_cleanup_buf_error(struct drm_device * dev,
 	}
 }
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 /**
  * Add AGP buffers for DMA transfers.
  *
@@ -756,7 +756,7 @@ int drm_legacy_addbufs_agp(struct drm_device *dev,
 	return 0;
 }
 EXPORT_SYMBOL(drm_legacy_addbufs_agp);
-#endif				/* __OS_HAS_AGP */
+#endif /* CONFIG_AGP */
 
 int drm_legacy_addbufs_pci(struct drm_device *dev,
 			   struct drm_buf_desc *request)
@@ -1145,7 +1145,7 @@ int drm_legacy_addbufs(struct drm_device *dev, void *data,
 	if (!drm_core_check_feature(dev, DRIVER_HAVE_DMA))
 		return -EINVAL;
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (request->flags & _DRM_AGP_BUFFER)
 		ret = drm_legacy_addbufs_agp(dev, request);
 	else
diff --git a/drivers/gpu/drm/drm_ioc32.c b/drivers/gpu/drm/drm_ioc32.c
index ddfa601..57676f8 100644
--- a/drivers/gpu/drm/drm_ioc32.c
+++ b/drivers/gpu/drm/drm_ioc32.c
@@ -720,7 +720,7 @@ static int compat_drm_dma(struct file *file, unsigned int cmd,
 	return 0;
 }
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 typedef struct drm_agp_mode32 {
 	u32 mode;	/**< AGP mode */
 } drm_agp_mode32_t;
@@ -882,7 +882,7 @@ static int compat_drm_agp_unbind(struct file *file, unsigned int cmd,
 
 	return drm_ioctl(file, DRM_IOCTL_AGP_UNBIND, (unsigned long)request);
 }
-#endif				/* __OS_HAS_AGP */
+#endif /* CONFIG_AGP */
 
 typedef struct drm_scatter_gather32 {
 	u32 size;	/**< In bytes -- will round to page boundary */
@@ -1090,7 +1090,7 @@ static drm_ioctl_compat_t *drm_compat_ioctls[] = {
 	[DRM_IOCTL_NR(DRM_IOCTL_GET_SAREA_CTX32)] = compat_drm_getsareactx,
 	[DRM_IOCTL_NR(DRM_IOCTL_RES_CTX32)] = compat_drm_resctx,
 	[DRM_IOCTL_NR(DRM_IOCTL_DMA32)] = compat_drm_dma,
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	[DRM_IOCTL_NR(DRM_IOCTL_AGP_ENABLE32)] = compat_drm_agp_enable,
 	[DRM_IOCTL_NR(DRM_IOCTL_AGP_INFO32)] = compat_drm_agp_info,
 	[DRM_IOCTL_NR(DRM_IOCTL_AGP_ALLOC32)] = compat_drm_agp_alloc,
diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c
index d93e737..cc94ecc 100644
--- a/drivers/gpu/drm/drm_ioctl.c
+++ b/drivers/gpu/drm/drm_ioctl.c
@@ -571,7 +571,7 @@ static const struct drm_ioctl_desc drm_ioctls[] = {
 
 	DRM_IOCTL_DEF(DRM_IOCTL_CONTROL, drm_control, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	DRM_IOCTL_DEF(DRM_IOCTL_AGP_ACQUIRE, drm_agp_acquire_ioctl, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
 	DRM_IOCTL_DEF(DRM_IOCTL_AGP_RELEASE, drm_agp_release_ioctl, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
 	DRM_IOCTL_DEF(DRM_IOCTL_AGP_ENABLE, drm_agp_enable_ioctl, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
diff --git a/drivers/gpu/drm/drm_memory.c b/drivers/gpu/drm/drm_memory.c
index a521ef6..87a8cb7 100644
--- a/drivers/gpu/drm/drm_memory.c
+++ b/drivers/gpu/drm/drm_memory.c
@@ -38,7 +38,7 @@
 #include <drm/drmP.h>
 #include "drm_legacy.h"
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 
 #ifdef HAVE_PAGE_AGP
 # include <asm/agp.h>
@@ -111,14 +111,14 @@ int drm_unbind_agp(struct agp_memory * handle)
 	return agp_unbind_memory(handle);
 }
 
-#else  /*  __OS_HAS_AGP  */
+#else /*  CONFIG_AGP  */
 static inline void *agp_remap(unsigned long offset, unsigned long size,
 			      struct drm_device * dev)
 {
 	return NULL;
 }
 
-#endif				/* agp */
+#endif /* CONFIG_AGP */
 
 void drm_legacy_ioremap(struct drm_local_map *map, struct drm_device *dev)
 {
diff --git a/drivers/gpu/drm/drm_vm.c b/drivers/gpu/drm/drm_vm.c
index aab49ee..f90bd5f 100644
--- a/drivers/gpu/drm/drm_vm.c
+++ b/drivers/gpu/drm/drm_vm.c
@@ -95,7 +95,7 @@ static pgprot_t drm_dma_prot(uint32_t map_type, struct vm_area_struct *vma)
  * Find the right map and if it's AGP memory find the real physical page to
  * map, get the page, increment the use count and return it.
  */
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 static int drm_do_vm_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 {
 	struct drm_file *priv = vma->vm_file->private_data;
@@ -168,12 +168,12 @@ static int drm_do_vm_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 vm_fault_error:
 	return VM_FAULT_SIGBUS;	/* Disallow mremap */
 }
-#else				/* __OS_HAS_AGP */
+#else
 static int drm_do_vm_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 {
 	return VM_FAULT_SIGBUS;
 }
-#endif				/* __OS_HAS_AGP */
+#endif
 
 /**
  * \c nopage method for shared virtual memory.
@@ -556,7 +556,7 @@ static int drm_mmap_locked(struct file *filp, struct vm_area_struct *vma)
 	 * --BenH.
 	 */
 	if (!vma->vm_pgoff
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	    && (!dev->agp
 		|| dev->agp->agp_info.device->vendor != PCI_VENDOR_ID_APPLE)
 #endif
diff --git a/drivers/gpu/drm/mga/mga_dma.c b/drivers/gpu/drm/mga/mga_dma.c
index 8cfa9cb..1f2f9ca 100644
--- a/drivers/gpu/drm/mga/mga_dma.c
+++ b/drivers/gpu/drm/mga/mga_dma.c
@@ -416,7 +416,7 @@ int mga_driver_load(struct drm_device *dev, unsigned long flags)
 	return 0;
 }
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 /**
  * Bootstrap the driver for AGP DMA.
  *
@@ -947,7 +947,7 @@ static int mga_do_cleanup_dma(struct drm_device *dev, int full_cleanup)
 			drm_legacy_ioremapfree(dev->agp_buffer_map, dev);
 
 		if (dev_priv->used_new_dma_init) {
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 			if (dev_priv->agp_handle != 0) {
 				struct drm_agp_binding unbind_req;
 				struct drm_agp_buffer free_req;
diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c
index 15057b3..78f520d 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bo.c
+++ b/drivers/gpu/drm/nouveau/nouveau_bo.c
@@ -574,7 +574,7 @@ static struct ttm_tt *
 nouveau_ttm_tt_create(struct ttm_bo_device *bdev, unsigned long size,
 		      uint32_t page_flags, struct page *dummy_read)
 {
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	struct nouveau_drm *drm = nouveau_bdev(bdev);
 
 	if (drm->agp.bridge) {
@@ -1366,7 +1366,7 @@ nouveau_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_mem_reg *mem)
 		/* System memory */
 		return 0;
 	case TTM_PL_TT:
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 		if (drm->agp.bridge) {
 			mem->bus.offset = mem->start << PAGE_SHIFT;
 			mem->bus.base = drm->agp.base;
@@ -1496,7 +1496,7 @@ nouveau_ttm_tt_populate(struct ttm_tt *ttm)
 	    ttm->caching_state == tt_uncached)
 		return ttm_dma_populate(ttm_dma, dev->dev);
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (drm->agp.bridge) {
 		return ttm_agp_tt_populate(ttm);
 	}
@@ -1563,7 +1563,7 @@ nouveau_ttm_tt_unpopulate(struct ttm_tt *ttm)
 		return;
 	}
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (drm->agp.bridge) {
 		ttm_agp_tt_unpopulate(ttm);
 		return;
diff --git a/drivers/gpu/drm/r128/r128_cce.c b/drivers/gpu/drm/r128/r128_cce.c
index 2c45ac9..14fd83b5f 100644
--- a/drivers/gpu/drm/r128/r128_cce.c
+++ b/drivers/gpu/drm/r128/r128_cce.c
@@ -311,7 +311,7 @@ static void r128_cce_init_ring_buffer(struct drm_device *dev,
 	/* The manual (p. 2) says this address is in "VM space".  This
 	 * means it's an offset from the start of AGP space.
 	 */
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (!dev_priv->is_pci)
 		ring_start = dev_priv->cce_ring->offset - dev->agp->base;
 	else
@@ -505,7 +505,7 @@ static int r128_do_init_cce(struct drm_device *dev, drm_r128_init_t *init)
 	    (drm_r128_sarea_t *) ((u8 *) dev_priv->sarea->handle +
 				  init->sarea_priv_offset);
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (!dev_priv->is_pci) {
 		drm_legacy_ioremap_wc(dev_priv->cce_ring, dev);
 		drm_legacy_ioremap_wc(dev_priv->ring_rptr, dev);
@@ -529,7 +529,7 @@ static int r128_do_init_cce(struct drm_device *dev, drm_r128_init_t *init)
 			(void *)(unsigned long)dev->agp_buffer_map->offset;
 	}
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (!dev_priv->is_pci)
 		dev_priv->cce_buffers_offset = dev->agp->base;
 	else
@@ -552,7 +552,7 @@ static int r128_do_init_cce(struct drm_device *dev, drm_r128_init_t *init)
 	dev_priv->sarea_priv->last_dispatch = 0;
 	R128_WRITE(R128_LAST_DISPATCH_REG, dev_priv->sarea_priv->last_dispatch);
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (dev_priv->is_pci) {
 #endif
 		dev_priv->gart_info.table_mask = DMA_BIT_MASK(32);
@@ -568,7 +568,7 @@ static int r128_do_init_cce(struct drm_device *dev, drm_r128_init_t *init)
 			return -ENOMEM;
 		}
 		R128_WRITE(R128_PCI_GART_PAGE, dev_priv->gart_info.bus_addr);
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	}
 #endif
 
@@ -600,7 +600,7 @@ int r128_do_cleanup_cce(struct drm_device *dev)
 	if (dev->dev_private) {
 		drm_r128_private_t *dev_priv = dev->dev_private;
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 		if (!dev_priv->is_pci) {
 			if (dev_priv->cce_ring != NULL)
 				drm_legacy_ioremapfree(dev_priv->cce_ring, dev);
diff --git a/drivers/gpu/drm/radeon/r600_cp.c b/drivers/gpu/drm/radeon/r600_cp.c
index 98f9ada..e231eea 100644
--- a/drivers/gpu/drm/radeon/r600_cp.c
+++ b/drivers/gpu/drm/radeon/r600_cp.c
@@ -1837,7 +1837,7 @@ static void r600_cp_init_ring_buffer(struct drm_device *dev,
 	SET_RING_HEAD(dev_priv, 0);
 	dev_priv->ring.tail = 0;
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (dev_priv->flags & RADEON_IS_AGP) {
 		rptr_addr = dev_priv->ring_rptr->offset
 			- dev->agp->base +
@@ -1863,7 +1863,7 @@ static void r600_cp_init_ring_buffer(struct drm_device *dev,
 		     dev_priv->ring.size_l2qw);
 #endif
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (dev_priv->flags & RADEON_IS_AGP) {
 		/* XXX */
 		radeon_write_agp_base(dev_priv, dev->agp->base);
@@ -1946,7 +1946,7 @@ int r600_do_cleanup_cp(struct drm_device *dev)
 	if (dev->irq_enabled)
 		drm_irq_uninstall(dev);
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (dev_priv->flags & RADEON_IS_AGP) {
 		if (dev_priv->cp_ring != NULL) {
 			drm_legacy_ioremapfree(dev_priv->cp_ring, dev);
@@ -2089,7 +2089,7 @@ int r600_do_init_cp(struct drm_device *dev, drm_radeon_init_t *init,
 		}
 	}
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	/* XXX */
 	if (dev_priv->flags & RADEON_IS_AGP) {
 		drm_legacy_ioremap_wc(dev_priv->cp_ring, dev);
@@ -2148,7 +2148,7 @@ int r600_do_init_cp(struct drm_device *dev, drm_radeon_init_t *init,
 		 * location in the card and on the bus, though we have to
 		 * align it down.
 		 */
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 		/* XXX */
 		if (dev_priv->flags & RADEON_IS_AGP) {
 			base = dev->agp->base;
@@ -2175,7 +2175,7 @@ int r600_do_init_cp(struct drm_device *dev, drm_radeon_init_t *init,
 				 base, dev_priv->gart_vm_start);
 	}
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	/* XXX */
 	if (dev_priv->flags & RADEON_IS_AGP)
 		dev_priv->gart_buffers_offset = (dev->agp_buffer_map->offset
@@ -2212,7 +2212,7 @@ int r600_do_init_cp(struct drm_device *dev, drm_radeon_init_t *init,
 
 	dev_priv->ring.high_mark = RADEON_RING_HIGH_MARK;
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (dev_priv->flags & RADEON_IS_AGP) {
 		/* XXX turn off pcie gart */
 	} else
diff --git a/drivers/gpu/drm/radeon/radeon_agp.c b/drivers/gpu/drm/radeon/radeon_agp.c
index a9297b2..fe994aa 100644
--- a/drivers/gpu/drm/radeon/radeon_agp.c
+++ b/drivers/gpu/drm/radeon/radeon_agp.c
@@ -28,7 +28,7 @@
 #include "radeon.h"
 #include <drm/radeon_drm.h>
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 
 struct radeon_agpmode_quirk {
 	u32 hostbridge_vendor;
@@ -123,7 +123,7 @@ static struct radeon_agpmode_quirk radeon_agpmode_quirk_list[] = {
 
 int radeon_agp_init(struct radeon_device *rdev)
 {
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	struct radeon_agpmode_quirk *p = radeon_agpmode_quirk_list;
 	struct drm_agp_mode mode;
 	struct drm_agp_info info;
@@ -257,7 +257,7 @@ int radeon_agp_init(struct radeon_device *rdev)
 
 void radeon_agp_resume(struct radeon_device *rdev)
 {
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	int r;
 	if (rdev->flags & RADEON_IS_AGP) {
 		r = radeon_agp_init(rdev);
@@ -269,7 +269,7 @@ void radeon_agp_resume(struct radeon_device *rdev)
 
 void radeon_agp_fini(struct radeon_device *rdev)
 {
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (rdev->ddev->agp && rdev->ddev->agp->acquired) {
 		drm_agp_release(rdev->ddev);
 	}
diff --git a/drivers/gpu/drm/radeon/radeon_cp.c b/drivers/gpu/drm/radeon/radeon_cp.c
index ea134a7..500287e 100644
--- a/drivers/gpu/drm/radeon/radeon_cp.c
+++ b/drivers/gpu/drm/radeon/radeon_cp.c
@@ -762,7 +762,7 @@ static void radeon_cp_init_ring_buffer(struct drm_device * dev,
 			     ((dev_priv->gart_vm_start - 1) & 0xffff0000)
 			     | (dev_priv->fb_location >> 16));
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (dev_priv->flags & RADEON_IS_AGP) {
 		radeon_write_agp_base(dev_priv, dev->agp->base);
 
@@ -791,7 +791,7 @@ static void radeon_cp_init_ring_buffer(struct drm_device * dev,
 	SET_RING_HEAD(dev_priv, cur_read_ptr);
 	dev_priv->ring.tail = cur_read_ptr;
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (dev_priv->flags & RADEON_IS_AGP) {
 		RADEON_WRITE(RADEON_CP_RB_RPTR_ADDR,
 			     dev_priv->ring_rptr->offset
@@ -1335,7 +1335,7 @@ static int radeon_do_init_cp(struct drm_device *dev, drm_radeon_init_t *init,
 		}
 	}
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (dev_priv->flags & RADEON_IS_AGP) {
 		drm_legacy_ioremap_wc(dev_priv->cp_ring, dev);
 		drm_legacy_ioremap_wc(dev_priv->ring_rptr, dev);
@@ -1394,7 +1394,7 @@ static int radeon_do_init_cp(struct drm_device *dev, drm_radeon_init_t *init,
 		 * location in the card and on the bus, though we have to
 		 * align it down.
 		 */
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 		if (dev_priv->flags & RADEON_IS_AGP) {
 			base = dev->agp->base;
 			/* Check if valid */
@@ -1424,7 +1424,7 @@ static int radeon_do_init_cp(struct drm_device *dev, drm_radeon_init_t *init,
 			RADEON_READ(RADEON_CONFIG_APER_SIZE);
 	}
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (dev_priv->flags & RADEON_IS_AGP)
 		dev_priv->gart_buffers_offset = (dev->agp_buffer_map->offset
 						 - dev->agp->base
@@ -1455,7 +1455,7 @@ static int radeon_do_init_cp(struct drm_device *dev, drm_radeon_init_t *init,
 
 	dev_priv->ring.high_mark = RADEON_RING_HIGH_MARK;
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (dev_priv->flags & RADEON_IS_AGP) {
 		/* Turn off PCI GART */
 		radeon_set_pcigart(dev_priv, 0);
@@ -1566,7 +1566,7 @@ static int radeon_do_cleanup_cp(struct drm_device * dev)
 	if (dev->irq_enabled)
 		drm_irq_uninstall(dev);
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (dev_priv->flags & RADEON_IS_AGP) {
 		if (dev_priv->cp_ring != NULL) {
 			drm_legacy_ioremapfree(dev_priv->cp_ring, dev);
@@ -1625,7 +1625,7 @@ static int radeon_do_resume_cp(struct drm_device *dev, struct drm_file *file_pri
 
 	DRM_DEBUG("Starting radeon_do_resume_cp()\n");
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (dev_priv->flags & RADEON_IS_AGP) {
 		/* Turn off PCI GART */
 		radeon_set_pcigart(dev_priv, 0);
diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c
index 06ac59f..e343074 100644
--- a/drivers/gpu/drm/radeon/radeon_ttm.c
+++ b/drivers/gpu/drm/radeon/radeon_ttm.c
@@ -144,7 +144,7 @@ static int radeon_init_mem_type(struct ttm_bo_device *bdev, uint32_t type,
 		man->available_caching = TTM_PL_MASK_CACHING;
 		man->default_caching = TTM_PL_FLAG_CACHED;
 		man->flags = TTM_MEMTYPE_FLAG_MAPPABLE | TTM_MEMTYPE_FLAG_CMA;
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 		if (rdev->flags & RADEON_IS_AGP) {
 			if (!rdev->ddev->agp) {
 				DRM_ERROR("AGP is not enabled for memory type %u\n",
@@ -461,7 +461,7 @@ static int radeon_ttm_io_mem_reserve(struct ttm_bo_device *bdev, struct ttm_mem_
 		/* system memory */
 		return 0;
 	case TTM_PL_TT:
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 		if (rdev->flags & RADEON_IS_AGP) {
 			/* RADEON_IS_AGP is set only if AGP is active */
 			mem->bus.offset = mem->start << PAGE_SHIFT;
@@ -680,7 +680,7 @@ static struct ttm_tt *radeon_ttm_tt_create(struct ttm_bo_device *bdev,
 	struct radeon_ttm_tt *gtt;
 
 	rdev = radeon_get_rdev(bdev);
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (rdev->flags & RADEON_IS_AGP) {
 		return ttm_agp_tt_create(bdev, rdev->ddev->agp->bridge,
 					 size, page_flags, dummy_read_page);
@@ -736,7 +736,7 @@ static int radeon_ttm_tt_populate(struct ttm_tt *ttm)
 	}
 
 	rdev = radeon_get_rdev(ttm->bdev);
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (rdev->flags & RADEON_IS_AGP) {
 		return ttm_agp_tt_populate(ttm);
 	}
@@ -787,7 +787,7 @@ static void radeon_ttm_tt_unpopulate(struct ttm_tt *ttm)
 		return;
 
 	rdev = radeon_get_rdev(ttm->bdev);
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 	if (rdev->flags & RADEON_IS_AGP) {
 		ttm_agp_tt_unpopulate(ttm);
 		return;
diff --git a/include/drm/drm_agpsupport.h b/include/drm/drm_agpsupport.h
index 055dc05..b0ec72f 100644
--- a/include/drm/drm_agpsupport.h
+++ b/include/drm/drm_agpsupport.h
@@ -12,9 +12,6 @@
 struct drm_device;
 struct drm_file;
 
-#define __OS_HAS_AGP (defined(CONFIG_AGP) || (defined(CONFIG_AGP_MODULE) && \
-					      defined(MODULE)))
-
 struct drm_agp_head {
 	struct agp_kern_info agp_info;
 	struct list_head memory;
@@ -28,7 +25,7 @@ struct drm_agp_head {
 	unsigned long page_mask;
 };
 
-#if __OS_HAS_AGP
+#if IS_ENABLED(CONFIG_AGP)
 
 void drm_free_agp(struct agp_memory * handle, int pages);
 int drm_bind_agp(struct agp_memory * handle, unsigned int start);
@@ -66,7 +63,7 @@ int drm_agp_bind(struct drm_device *dev, struct drm_agp_binding *request);
 int drm_agp_bind_ioctl(struct drm_device *dev, void *data,
 		       struct drm_file *file_priv);
 
-#else /* __OS_HAS_AGP */
+#else /* CONFIG_AGP */
 
 static inline void drm_free_agp(struct agp_memory * handle, int pages)
 {
@@ -194,6 +191,6 @@ static inline int drm_agp_bind_ioctl(struct drm_device *dev, void *data,
 	return -ENODEV;
 }
 
-#endif /* __OS_HAS_AGP */
+#endif /* CONFIG_AGP */
 
 #endif /* _DRM_AGPSUPPORT_H_ */
-- 
cgit v0.10.2


From 4b63539bb2f604b26ef4951c5c14828d24a7ce6c Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Tue, 8 Sep 2015 13:56:26 +0200
Subject: drm: Define a drm_invalid_op ioctl implementation
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

And use it in radeon to replace all the ioctls no longer valid in kms
mode. I plan to also use this later on when nuking the ums support for
i915.

Note that setting the function pointer in the ioctl table to NULL
would amount to the same, but that results in some debug output from
the drm_ioctl() function. I've figured it's cleaner to have a
special-purpose function.

Cc: Alex Deucher <alexdeucher@gmail.com>
Reviewed-by: David Herrmann <dh.herrmann@gmail.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c
index cc94ecc..f1c8a15 100644
--- a/drivers/gpu/drm/drm_ioctl.c
+++ b/drivers/gpu/drm/drm_ioctl.c
@@ -415,6 +415,13 @@ int drm_noop(struct drm_device *dev, void *data,
 }
 EXPORT_SYMBOL(drm_noop);
 
+int drm_invalid_op(struct drm_device *dev, void *data,
+		   struct drm_file *file_priv)
+{
+	return -EINVAL;
+}
+EXPORT_SYMBOL(drm_invalid_op);
+
 /**
  * Copy and IOCTL return string to user space
  */
diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c
index fd9da28..2773403 100644
--- a/drivers/gpu/drm/radeon/radeon_kms.c
+++ b/drivers/gpu/drm/radeon/radeon_kms.c
@@ -844,74 +844,34 @@ int radeon_get_vblank_timestamp_kms(struct drm_device *dev, int crtc,
 						     &drmcrtc->hwmode);
 }
 
-#define KMS_INVALID_IOCTL(name)						\
-static int name(struct drm_device *dev, void *data, struct drm_file	\
-		*file_priv)						\
-{									\
-	DRM_ERROR("invalid ioctl with kms %s\n", __func__);		\
-	return -EINVAL;							\
-}
-
-/*
- * All these ioctls are invalid in kms world.
- */
-KMS_INVALID_IOCTL(radeon_cp_init_kms)
-KMS_INVALID_IOCTL(radeon_cp_start_kms)
-KMS_INVALID_IOCTL(radeon_cp_stop_kms)
-KMS_INVALID_IOCTL(radeon_cp_reset_kms)
-KMS_INVALID_IOCTL(radeon_cp_idle_kms)
-KMS_INVALID_IOCTL(radeon_cp_resume_kms)
-KMS_INVALID_IOCTL(radeon_engine_reset_kms)
-KMS_INVALID_IOCTL(radeon_fullscreen_kms)
-KMS_INVALID_IOCTL(radeon_cp_swap_kms)
-KMS_INVALID_IOCTL(radeon_cp_clear_kms)
-KMS_INVALID_IOCTL(radeon_cp_vertex_kms)
-KMS_INVALID_IOCTL(radeon_cp_indices_kms)
-KMS_INVALID_IOCTL(radeon_cp_texture_kms)
-KMS_INVALID_IOCTL(radeon_cp_stipple_kms)
-KMS_INVALID_IOCTL(radeon_cp_indirect_kms)
-KMS_INVALID_IOCTL(radeon_cp_vertex2_kms)
-KMS_INVALID_IOCTL(radeon_cp_cmdbuf_kms)
-KMS_INVALID_IOCTL(radeon_cp_getparam_kms)
-KMS_INVALID_IOCTL(radeon_cp_flip_kms)
-KMS_INVALID_IOCTL(radeon_mem_alloc_kms)
-KMS_INVALID_IOCTL(radeon_mem_free_kms)
-KMS_INVALID_IOCTL(radeon_mem_init_heap_kms)
-KMS_INVALID_IOCTL(radeon_irq_emit_kms)
-KMS_INVALID_IOCTL(radeon_irq_wait_kms)
-KMS_INVALID_IOCTL(radeon_cp_setparam_kms)
-KMS_INVALID_IOCTL(radeon_surface_alloc_kms)
-KMS_INVALID_IOCTL(radeon_surface_free_kms)
-
-
 const struct drm_ioctl_desc radeon_ioctls_kms[] = {
-	DRM_IOCTL_DEF_DRV(RADEON_CP_INIT, radeon_cp_init_kms, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
-	DRM_IOCTL_DEF_DRV(RADEON_CP_START, radeon_cp_start_kms, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
-	DRM_IOCTL_DEF_DRV(RADEON_CP_STOP, radeon_cp_stop_kms, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
-	DRM_IOCTL_DEF_DRV(RADEON_CP_RESET, radeon_cp_reset_kms, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
-	DRM_IOCTL_DEF_DRV(RADEON_CP_IDLE, radeon_cp_idle_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_CP_RESUME, radeon_cp_resume_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_RESET, radeon_engine_reset_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_FULLSCREEN, radeon_fullscreen_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_SWAP, radeon_cp_swap_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_CLEAR, radeon_cp_clear_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_VERTEX, radeon_cp_vertex_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_INDICES, radeon_cp_indices_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_TEXTURE, radeon_cp_texture_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_STIPPLE, radeon_cp_stipple_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_INDIRECT, radeon_cp_indirect_kms, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
-	DRM_IOCTL_DEF_DRV(RADEON_VERTEX2, radeon_cp_vertex2_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_CMDBUF, radeon_cp_cmdbuf_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_GETPARAM, radeon_cp_getparam_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_FLIP, radeon_cp_flip_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_ALLOC, radeon_mem_alloc_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_FREE, radeon_mem_free_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_INIT_HEAP, radeon_mem_init_heap_kms, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
-	DRM_IOCTL_DEF_DRV(RADEON_IRQ_EMIT, radeon_irq_emit_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_IRQ_WAIT, radeon_irq_wait_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_SETPARAM, radeon_cp_setparam_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_SURF_ALLOC, radeon_surface_alloc_kms, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(RADEON_SURF_FREE, radeon_surface_free_kms, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_CP_INIT, drm_invalid_op, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
+	DRM_IOCTL_DEF_DRV(RADEON_CP_START, drm_invalid_op, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
+	DRM_IOCTL_DEF_DRV(RADEON_CP_STOP, drm_invalid_op, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
+	DRM_IOCTL_DEF_DRV(RADEON_CP_RESET, drm_invalid_op, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
+	DRM_IOCTL_DEF_DRV(RADEON_CP_IDLE, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_CP_RESUME, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_RESET, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_FULLSCREEN, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_SWAP, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_CLEAR, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_VERTEX, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_INDICES, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_TEXTURE, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_STIPPLE, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_INDIRECT, drm_invalid_op, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
+	DRM_IOCTL_DEF_DRV(RADEON_VERTEX2, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_CMDBUF, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_GETPARAM, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_FLIP, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_ALLOC, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_FREE, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_INIT_HEAP, drm_invalid_op, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
+	DRM_IOCTL_DEF_DRV(RADEON_IRQ_EMIT, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_IRQ_WAIT, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_SETPARAM, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_SURF_ALLOC, drm_invalid_op, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_SURF_FREE, drm_invalid_op, DRM_AUTH),
 	/* KMS */
 	DRM_IOCTL_DEF_DRV(RADEON_GEM_INFO, radeon_gem_info_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(RADEON_GEM_CREATE, radeon_gem_create_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
diff --git a/include/drm/drmP.h b/include/drm/drmP.h
index d0251ac..967d8a0 100644
--- a/include/drm/drmP.h
+++ b/include/drm/drmP.h
@@ -910,6 +910,8 @@ extern unsigned int drm_poll(struct file *filp, struct poll_table_struct *wait);
 /* Misc. IOCTL support (drm_ioctl.c) */
 int drm_noop(struct drm_device *dev, void *data,
 	     struct drm_file *file_priv);
+int drm_invalid_op(struct drm_device *dev, void *data,
+		   struct drm_file *file_priv);
 
 /* Cache management (drm_cache.c) */
 void drm_clflush_pages(struct page *pages[], unsigned long num_pages);
-- 
cgit v0.10.2


From 0aaf20cff6dc1b89ec9d63cb004803cf4c0efae7 Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Tue, 8 Sep 2015 13:56:27 +0200
Subject: drm/drm_ioctl.c: kerneldoc
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

As usual pull it into the drm docbook template, too. And again as
usual I've decided to only document stuff exported to drivers, so all
the old leftover markup from the shared drm repo days lost the magic
** signature.

Reviewed-by: David Herrmann <dh.herrmann@gmail.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

diff --git a/Documentation/DocBook/drm.tmpl b/Documentation/DocBook/drm.tmpl
index 11f5bfb..8d0cbf1 100644
--- a/Documentation/DocBook/drm.tmpl
+++ b/Documentation/DocBook/drm.tmpl
@@ -3689,6 +3689,7 @@ int num_ioctls;</synopsis>
 	  </itemizedlist>
 	</para>
       </para>
+!Edrivers/gpu/drm/drm_ioctl.c
     </sect2>
   </sect1>
   <sect1>
diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c
index f1c8a15..530c501 100644
--- a/drivers/gpu/drm/drm_ioctl.c
+++ b/drivers/gpu/drm/drm_ioctl.c
@@ -40,7 +40,7 @@
 static int drm_version(struct drm_device *dev, void *data,
 		       struct drm_file *file_priv);
 
-/**
+/*
  * Get the bus id.
  *
  * \param inode device inode.
@@ -75,7 +75,7 @@ drm_unset_busid(struct drm_device *dev,
 	master->unique_len = 0;
 }
 
-/**
+/*
  * Set the bus id.
  *
  * \param inode device inode.
@@ -149,7 +149,7 @@ static int drm_set_busid(struct drm_device *dev, struct drm_file *file_priv)
 	return 0;
 }
 
-/**
+/*
  * Get a mapping information.
  *
  * \param inode device inode.
@@ -201,7 +201,7 @@ static int drm_getmap(struct drm_device *dev, void *data,
 	return 0;
 }
 
-/**
+/*
  * Get client information.
  *
  * \param inode device inode.
@@ -244,7 +244,7 @@ static int drm_getclient(struct drm_device *dev, void *data,
 	}
 }
 
-/**
+/*
  * Get statistics information.
  *
  * \param inode device inode.
@@ -265,7 +265,7 @@ static int drm_getstats(struct drm_device *dev, void *data,
 	return 0;
 }
 
-/**
+/*
  * Get device/driver capabilities
  */
 static int drm_getcap(struct drm_device *dev, void *data, struct drm_file *file_priv)
@@ -318,7 +318,7 @@ static int drm_getcap(struct drm_device *dev, void *data, struct drm_file *file_
 	return 0;
 }
 
-/**
+/*
  * Set device/driver capabilities
  */
 static int
@@ -352,7 +352,7 @@ drm_setclientcap(struct drm_device *dev, void *data, struct drm_file *file_priv)
 	return 0;
 }
 
-/**
+/*
  * Setversion ioctl.
  *
  * \param inode device inode.
@@ -406,7 +406,18 @@ done:
 	return retcode;
 }
 
-/** No-op ioctl. */
+/**
+ * drm_noop - DRM no-op ioctl implemntation
+ * @dev: DRM device for the ioctl
+ * @data: data pointer for the ioctl
+ * @file_priv: DRM file for the ioctl call
+ *
+ * This no-op implementation for drm ioctls is useful for deprecated
+ * functionality where we can't return a failure code because existing userspace
+ * checks the result of the ioctl, but doesn't care about the action.
+ *
+ * Always returns successfully with 0.
+ */
 int drm_noop(struct drm_device *dev, void *data,
 	     struct drm_file *file_priv)
 {
@@ -415,6 +426,21 @@ int drm_noop(struct drm_device *dev, void *data,
 }
 EXPORT_SYMBOL(drm_noop);
 
+/**
+ * drm_invalid_op - DRM invalid ioctl implemntation
+ * @dev: DRM device for the ioctl
+ * @data: data pointer for the ioctl
+ * @file_priv: DRM file for the ioctl call
+ *
+ * This no-op implementation for drm ioctls is useful for deprecated
+ * functionality where we really don't want to allow userspace to call the ioctl
+ * any more. This is the case for old ums interfaces for drivers that
+ * transitioned to kms gradually and so kept the old legacy tables around. This
+ * only applies to radeon and i915 kms drivers, other drivers shouldn't need to
+ * use this function.
+ *
+ * Always fails with a return value of -EINVAL.
+ */
 int drm_invalid_op(struct drm_device *dev, void *data,
 		   struct drm_file *file_priv)
 {
@@ -422,7 +448,7 @@ int drm_invalid_op(struct drm_device *dev, void *data,
 }
 EXPORT_SYMBOL(drm_invalid_op);
 
-/**
+/*
  * Copy and IOCTL return string to user space
  */
 static int drm_copy_field(char __user *buf, size_t *buf_len, const char *value)
@@ -445,7 +471,7 @@ static int drm_copy_field(char __user *buf, size_t *buf_len, const char *value)
 	return 0;
 }
 
-/**
+/*
  * Get version information
  *
  * \param inode device inode.
@@ -477,7 +503,7 @@ static int drm_version(struct drm_device *dev, void *data,
 	return err;
 }
 
-/**
+/*
  * drm_ioctl_permit - Check ioctl permissions against caller
  *
  * @flags: ioctl permission flags.
@@ -525,7 +551,7 @@ EXPORT_SYMBOL(drm_ioctl_permit);
 		.name = #ioctl			\
 	}
 
-/** Ioctl table */
+/* Ioctl table */
 static const struct drm_ioctl_desc drm_ioctls[] = {
 	DRM_IOCTL_DEF(DRM_IOCTL_VERSION, drm_version,
 		      DRM_UNLOCKED|DRM_RENDER_ALLOW|DRM_CONTROL_ALLOW),
@@ -642,16 +668,16 @@ static const struct drm_ioctl_desc drm_ioctls[] = {
 #define DRM_CORE_IOCTL_COUNT	ARRAY_SIZE( drm_ioctls )
 
 /**
- * Called whenever a process performs an ioctl on /dev/drm.
- *
- * \param inode device inode.
- * \param file_priv DRM file private.
- * \param cmd command.
- * \param arg user argument.
- * \return zero on success or negative number on failure.
+ * drm_ioctl - ioctl callback implementation for DRM drivers
+ * @filp: file this ioctl is called on
+ * @cmd: ioctl cmd number
+ * @arg: user argument
  *
  * Looks up the ioctl function in the ::ioctls table, checking for root
  * previleges if so required, and dispatches to the respective function.
+ *
+ * Returns:
+ * Zero on success, negative error code on failure.
  */
 long drm_ioctl(struct file *filp,
 	      unsigned int cmd, unsigned long arg)
@@ -761,9 +787,15 @@ EXPORT_SYMBOL(drm_ioctl);
 
 /**
  * drm_ioctl_flags - Check for core ioctl and return ioctl permission flags
+ * @nr: ioctl number
+ * @flags: where to return the ioctl permission flags
+ *
+ * This ioctl is only used by the vmwgfx driver to augment the access checks
+ * done by the drm core and insofar a pretty decent layering violation. This
+ * shouldn't be used by any drivers.
  *
- * @nr: Ioctl number.
- * @flags: Where to return the ioctl permission flags
+ * Returns:
+ * True if the @nr corresponds to a DRM core ioctl numer, false otherwise.
  */
 bool drm_ioctl_flags(unsigned int nr, unsigned int *flags)
 {
-- 
cgit v0.10.2


From 87698c8aa77ce58c41a09b72a0c35fdbc9949025 Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Tue, 8 Sep 2015 13:56:29 +0200
Subject: drm/vmwgfx: Stop checking for DRM_UNLOCKED

drm core enforces now for DRIVER_MODESET that all ioctls are unlocked.
And all the old nasty ones from drm core aren't allowed for modern
drivers any more. Hence this is no longer needed.

Reviewed-by: David Herrmann <dh.herrmann@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
index 2c7a25c..8e7493d 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
@@ -1062,14 +1062,6 @@ static struct vmw_master *vmw_master_check(struct drm_device *dev,
 	mutex_unlock(&dev->master_mutex);
 
 	/*
-	 * Taking the drm_global_mutex after the TTM lock might deadlock
-	 */
-	if (!(flags & DRM_UNLOCKED)) {
-		DRM_ERROR("Refusing locked ioctl access.\n");
-		return ERR_PTR(-EDEADLK);
-	}
-
-	/*
 	 * Take the TTM lock. Possibly sleep waiting for the authenticating
 	 * master to become master again, or for a SIGTERM if the
 	 * authenticating master exits.
-- 
cgit v0.10.2


From 0731c65aceb2ac9b1d3bc64a057d6b0f0daae80d Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Tue, 8 Sep 2015 13:56:31 +0200
Subject: drm: Remove dummy agp ioctl wrappers

They're only used in the drm ioctl table, and there they're excluded
when AGP support is disabled. So this is just dead code ripe for
removal.

Reviewed-by: David Herrmann <dh.herrmann@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

diff --git a/include/drm/drm_agpsupport.h b/include/drm/drm_agpsupport.h
index b0ec72f..193ef19 100644
--- a/include/drm/drm_agpsupport.h
+++ b/include/drm/drm_agpsupport.h
@@ -102,95 +102,47 @@ static inline int drm_agp_acquire(struct drm_device *dev)
 	return -ENODEV;
 }
 
-static inline int drm_agp_acquire_ioctl(struct drm_device *dev, void *data,
-					struct drm_file *file_priv)
-{
-	return -ENODEV;
-}
-
 static inline int drm_agp_release(struct drm_device *dev)
 {
 	return -ENODEV;
 }
 
-static inline int drm_agp_release_ioctl(struct drm_device *dev, void *data,
-					struct drm_file *file_priv)
-{
-	return -ENODEV;
-}
-
 static inline int drm_agp_enable(struct drm_device *dev,
 				 struct drm_agp_mode mode)
 {
 	return -ENODEV;
 }
 
-static inline int drm_agp_enable_ioctl(struct drm_device *dev, void *data,
-				       struct drm_file *file_priv)
-{
-	return -ENODEV;
-}
-
 static inline int drm_agp_info(struct drm_device *dev,
 			       struct drm_agp_info *info)
 {
 	return -ENODEV;
 }
 
-static inline int drm_agp_info_ioctl(struct drm_device *dev, void *data,
-				     struct drm_file *file_priv)
-{
-	return -ENODEV;
-}
-
 static inline int drm_agp_alloc(struct drm_device *dev,
 				struct drm_agp_buffer *request)
 {
 	return -ENODEV;
 }
 
-static inline int drm_agp_alloc_ioctl(struct drm_device *dev, void *data,
-				      struct drm_file *file_priv)
-{
-	return -ENODEV;
-}
-
 static inline int drm_agp_free(struct drm_device *dev,
 			       struct drm_agp_buffer *request)
 {
 	return -ENODEV;
 }
 
-static inline int drm_agp_free_ioctl(struct drm_device *dev, void *data,
-				     struct drm_file *file_priv)
-{
-	return -ENODEV;
-}
-
 static inline int drm_agp_unbind(struct drm_device *dev,
 				 struct drm_agp_binding *request)
 {
 	return -ENODEV;
 }
 
-static inline int drm_agp_unbind_ioctl(struct drm_device *dev, void *data,
-				       struct drm_file *file_priv)
-{
-	return -ENODEV;
-}
-
 static inline int drm_agp_bind(struct drm_device *dev,
 			       struct drm_agp_binding *request)
 {
 	return -ENODEV;
 }
 
-static inline int drm_agp_bind_ioctl(struct drm_device *dev, void *data,
-				     struct drm_file *file_priv)
-{
-	return -ENODEV;
-}
-
 #endif /* CONFIG_AGP */
 
 #endif /* _DRM_AGPSUPPORT_H_ */
-- 
cgit v0.10.2


From 91bedd34abf0cd3f6276de642a145f75731acca5 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?=C5=81ukasz=20Daniluk?= <lukasz.daniluk@intel.com>
Date: Fri, 25 Sep 2015 11:54:58 +0200
Subject: drm/i915/bdw: Check for slice, subslice and EU count for BDW
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Added checks for available slices, subslices and EUs for Broadwell. This
information is filled in intel_device_info and is available to user with
GET_PARAM.
Added checks for enabled slices, subslices and EU for Broadwell. This
information is based on available counts but takes power gated slices
into account. It can be read in debugfs.
Introduce new register defines that contain information on slices on
Broadwell.

v2:
- Introduce GT_SLICE_INFO register
- Change Broadwell sseu_device_status function to use GT_SLICE_INFO
  register instead of RPCS register
- Undo removal of dev_priv variables in Cherryview and Gen9
  sseu_device_satus functions

v3:
- Fix style issues

v4:
- Corrected comment
- Reverted reordering of defines

Cc: Jeff Mcgee <jeff.mcgee@intel.com>
Cc: Arun Siluvery <arun.siluvery@linux.intel.com>
Signed-off-by: Łukasz Daniluk <lukasz.daniluk@intel.com>
Reviewed-by: Jeff McGee <jeff.mcgee@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index 9839831..2ac1ba8 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -5048,13 +5048,38 @@ static void gen9_sseu_device_status(struct drm_device *dev,
 	}
 }
 
+static void broadwell_sseu_device_status(struct drm_device *dev,
+					 struct sseu_dev_status *stat)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	int s;
+	u32 slice_info = I915_READ(GEN8_GT_SLICE_INFO);
+
+	stat->slice_total = hweight32(slice_info & GEN8_LSLICESTAT_MASK);
+
+	if (stat->slice_total) {
+		stat->subslice_per_slice = INTEL_INFO(dev)->subslice_per_slice;
+		stat->subslice_total = stat->slice_total *
+				       stat->subslice_per_slice;
+		stat->eu_per_subslice = INTEL_INFO(dev)->eu_per_subslice;
+		stat->eu_total = stat->eu_per_subslice * stat->subslice_total;
+
+		/* subtract fused off EU(s) from enabled slice(s) */
+		for (s = 0; s < stat->slice_total; s++) {
+			u8 subslice_7eu = INTEL_INFO(dev)->subslice_7eu[s];
+
+			stat->eu_total -= hweight8(subslice_7eu);
+		}
+	}
+}
+
 static int i915_sseu_status(struct seq_file *m, void *unused)
 {
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct sseu_dev_status stat;
 
-	if ((INTEL_INFO(dev)->gen < 8) || IS_BROADWELL(dev))
+	if (INTEL_INFO(dev)->gen < 8)
 		return -ENODEV;
 
 	seq_puts(m, "SSEU Device Info\n");
@@ -5079,6 +5104,8 @@ static int i915_sseu_status(struct seq_file *m, void *unused)
 	memset(&stat, 0, sizeof(stat));
 	if (IS_CHERRYVIEW(dev)) {
 		cherryview_sseu_device_status(dev, &stat);
+	} else if (IS_BROADWELL(dev)) {
+		broadwell_sseu_device_status(dev, &stat);
 	} else if (INTEL_INFO(dev)->gen >= 9) {
 		gen9_sseu_device_status(dev, &stat);
 	}
diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index ae1ba47..20b4b79 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -702,6 +702,82 @@ static void gen9_sseu_info_init(struct drm_device *dev)
 	info->has_eu_pg = (info->eu_per_subslice > 2);
 }
 
+static void broadwell_sseu_info_init(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_device_info *info;
+	const int s_max = 3, ss_max = 3, eu_max = 8;
+	int s, ss;
+	u32 fuse2, eu_disable[s_max], s_enable, ss_disable;
+
+	fuse2 = I915_READ(GEN8_FUSE2);
+	s_enable = (fuse2 & GEN8_F2_S_ENA_MASK) >> GEN8_F2_S_ENA_SHIFT;
+	ss_disable = (fuse2 & GEN8_F2_SS_DIS_MASK) >> GEN8_F2_SS_DIS_SHIFT;
+
+	eu_disable[0] = I915_READ(GEN8_EU_DISABLE0) & GEN8_EU_DIS0_S0_MASK;
+	eu_disable[1] = (I915_READ(GEN8_EU_DISABLE0) >> GEN8_EU_DIS0_S1_SHIFT) |
+			((I915_READ(GEN8_EU_DISABLE1) & GEN8_EU_DIS1_S1_MASK) <<
+			 (32 - GEN8_EU_DIS0_S1_SHIFT));
+	eu_disable[2] = (I915_READ(GEN8_EU_DISABLE1) >> GEN8_EU_DIS1_S2_SHIFT) |
+			((I915_READ(GEN8_EU_DISABLE2) & GEN8_EU_DIS2_S2_MASK) <<
+			 (32 - GEN8_EU_DIS1_S2_SHIFT));
+
+
+	info = (struct intel_device_info *)&dev_priv->info;
+	info->slice_total = hweight32(s_enable);
+
+	/*
+	 * The subslice disable field is global, i.e. it applies
+	 * to each of the enabled slices.
+	 */
+	info->subslice_per_slice = ss_max - hweight32(ss_disable);
+	info->subslice_total = info->slice_total * info->subslice_per_slice;
+
+	/*
+	 * Iterate through enabled slices and subslices to
+	 * count the total enabled EU.
+	 */
+	for (s = 0; s < s_max; s++) {
+		if (!(s_enable & (0x1 << s)))
+			/* skip disabled slice */
+			continue;
+
+		for (ss = 0; ss < ss_max; ss++) {
+			u32 n_disabled;
+
+			if (ss_disable & (0x1 << ss))
+				/* skip disabled subslice */
+				continue;
+
+			n_disabled = hweight8(eu_disable[s] >> (ss * eu_max));
+
+			/*
+			 * Record which subslices have 7 EUs.
+			 */
+			if (eu_max - n_disabled == 7)
+				info->subslice_7eu[s] |= 1 << ss;
+
+			info->eu_total += eu_max - n_disabled;
+		}
+	}
+
+	/*
+	 * BDW is expected to always have a uniform distribution of EU across
+	 * subslices with the exception that any one EU in any one subslice may
+	 * be fused off for die recovery.
+	 */
+	info->eu_per_subslice = info->subslice_total ?
+		DIV_ROUND_UP(info->eu_total, info->subslice_total) : 0;
+
+	/*
+	 * BDW supports slice power gating on devices with more than
+	 * one slice.
+	 */
+	info->has_slice_pg = (info->slice_total > 1);
+	info->has_subslice_pg = 0;
+	info->has_eu_pg = 0;
+}
+
 /*
  * Determine various intel_device_info fields at runtime.
  *
@@ -772,6 +848,8 @@ static void intel_device_info_runtime_init(struct drm_device *dev)
 	/* Initialize slice/subslice/EU info */
 	if (IS_CHERRYVIEW(dev))
 		cherryview_sseu_info_init(dev);
+	else if (IS_BROADWELL(dev))
+		broadwell_sseu_info_init(dev);
 	else if (INTEL_INFO(dev)->gen >= 9)
 		gen9_sseu_info_init(dev);
 
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 3709d6b..0b2e314 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -1887,12 +1887,27 @@ enum skl_disp_power_wells {
 #define   CHV_FGT_EU_DIS_SS1_R1_MASK	(0xf << CHV_FGT_EU_DIS_SS1_R1_SHIFT)
 
 #define GEN8_FUSE2			0x9120
+#define   GEN8_F2_SS_DIS_SHIFT		21
+#define   GEN8_F2_SS_DIS_MASK		(0x7 << GEN8_F2_SS_DIS_SHIFT)
 #define   GEN8_F2_S_ENA_SHIFT		25
 #define   GEN8_F2_S_ENA_MASK		(0x7 << GEN8_F2_S_ENA_SHIFT)
 
 #define   GEN9_F2_SS_DIS_SHIFT		20
 #define   GEN9_F2_SS_DIS_MASK		(0xf << GEN9_F2_SS_DIS_SHIFT)
 
+#define GEN8_EU_DISABLE0		0x9134
+#define   GEN8_EU_DIS0_S0_MASK		0xffffff
+#define   GEN8_EU_DIS0_S1_SHIFT		24
+#define   GEN8_EU_DIS0_S1_MASK		(0xff << GEN8_EU_DIS0_S1_SHIFT)
+
+#define GEN8_EU_DISABLE1		0x9138
+#define   GEN8_EU_DIS1_S1_MASK		0xffff
+#define   GEN8_EU_DIS1_S2_SHIFT		16
+#define   GEN8_EU_DIS1_S2_MASK		(0xffff << GEN8_EU_DIS1_S2_SHIFT)
+
+#define GEN8_EU_DISABLE2		0x913c
+#define   GEN8_EU_DIS2_S2_MASK		0xff
+
 #define GEN9_EU_DISABLE(slice)		(0x9134 + (slice)*0x4)
 
 #define GEN6_BSD_SLEEP_PSMI_CONTROL	0x12050
@@ -6885,6 +6900,9 @@ enum skl_disp_power_wells {
 #define   GEN6_RC6			3
 #define   GEN6_RC7			4
 
+#define GEN8_GT_SLICE_INFO		0x138064
+#define   GEN8_LSLICESTAT_MASK		0x7
+
 #define CHV_POWER_SS0_SIG1		0xa720
 #define CHV_POWER_SS1_SIG1		0xa728
 #define   CHV_SS_PG_ENABLE		(1<<1)
-- 
cgit v0.10.2


From c668cde5a3b5a1326923f341885ce9660d15091e Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Wed, 30 Sep 2015 10:46:59 +0200
Subject: drm/i915: Remove setparam ioctl
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This was only used for the ums+gem combo, so ripe for removal now that
we only have kms code left.

v2: Drop fence_reg_start since it's now unused, noticed by Ville.

Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Acked-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index e3ec904..40e77fd 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -957,7 +957,6 @@ static int i915_gem_fence_regs_info(struct seq_file *m, void *data)
 	if (ret)
 		return ret;
 
-	seq_printf(m, "Reserved fences = %d\n", dev_priv->fence_reg_start);
 	seq_printf(m, "Total fences = %d\n", dev_priv->num_fence_regs);
 	for (i = 0; i < dev_priv->num_fence_regs; i++) {
 		struct drm_i915_gem_object *obj = dev_priv->fence_regs[i].obj;
diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index ab37d11..d1454b5 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -75,7 +75,7 @@ static int i915_getparam(struct drm_device *dev, void *data,
 		value = 1;
 		break;
 	case I915_PARAM_NUM_FENCES_AVAIL:
-		value = dev_priv->num_fence_regs - dev_priv->fence_reg_start;
+		value = dev_priv->num_fence_regs;
 		break;
 	case I915_PARAM_HAS_OVERLAY:
 		value = dev_priv->overlay ? 1 : 0;
@@ -183,35 +183,6 @@ static int i915_getparam(struct drm_device *dev, void *data,
 	return 0;
 }
 
-static int i915_setparam(struct drm_device *dev, void *data,
-			 struct drm_file *file_priv)
-{
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	drm_i915_setparam_t *param = data;
-
-	switch (param->param) {
-	case I915_SETPARAM_USE_MI_BATCHBUFFER_START:
-	case I915_SETPARAM_TEX_LRU_LOG_GRANULARITY:
-	case I915_SETPARAM_ALLOW_BATCHBUFFER:
-		/* Reject all old ums/dri params. */
-		return -ENODEV;
-
-	case I915_SETPARAM_NUM_USED_FENCES:
-		if (param->value > dev_priv->num_fence_regs ||
-		    param->value < 0)
-			return -EINVAL;
-		/* Userspace can use first N regs */
-		dev_priv->fence_reg_start = param->value;
-		break;
-	default:
-		DRM_DEBUG_DRIVER("unknown parameter %d\n",
-					param->param);
-		return -EINVAL;
-	}
-
-	return 0;
-}
-
 static int i915_get_bridge_dev(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -1226,7 +1197,7 @@ const struct drm_ioctl_desc i915_ioctls[] = {
 	DRM_IOCTL_DEF_DRV(I915_IRQ_EMIT, drm_noop, DRM_AUTH),
 	DRM_IOCTL_DEF_DRV(I915_IRQ_WAIT, drm_noop, DRM_AUTH),
 	DRM_IOCTL_DEF_DRV(I915_GETPARAM, i915_getparam, DRM_AUTH|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(I915_SETPARAM, i915_setparam, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
+	DRM_IOCTL_DEF_DRV(I915_SETPARAM, drm_noop, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
 	DRM_IOCTL_DEF_DRV(I915_ALLOC, drm_noop, DRM_AUTH),
 	DRM_IOCTL_DEF_DRV(I915_FREE, drm_noop, DRM_AUTH),
 	DRM_IOCTL_DEF_DRV(I915_INIT_HEAP, drm_noop, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index e1db8de..9e13a77 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -1790,7 +1790,6 @@ struct drm_i915_private {
 	struct mutex pps_mutex;
 
 	struct drm_i915_fence_reg fence_regs[I915_MAX_NUM_FENCES]; /* assume 965 */
-	int fence_reg_start; /* 4 if userland hasn't ioctl'd us yet */
 	int num_fence_regs; /* 8 on pre-965, 16 otherwise */
 
 	unsigned int fsb_freq, mem_freq, is_ddr3;
diff --git a/drivers/gpu/drm/i915/i915_gem_fence.c b/drivers/gpu/drm/i915/i915_gem_fence.c
index af1f8c4..caa271b 100644
--- a/drivers/gpu/drm/i915/i915_gem_fence.c
+++ b/drivers/gpu/drm/i915/i915_gem_fence.c
@@ -322,7 +322,7 @@ i915_find_fence_reg(struct drm_device *dev)
 
 	/* First try to find a free reg */
 	avail = NULL;
-	for (i = dev_priv->fence_reg_start; i < dev_priv->num_fence_regs; i++) {
+	for (i = 0; i < dev_priv->num_fence_regs; i++) {
 		reg = &dev_priv->fence_regs[i];
 		if (!reg->obj)
 			return reg;
-- 
cgit v0.10.2


From eede3b53e9b5f7587e876ae9575392f92932951a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 18 Sep 2015 20:03:30 +0300
Subject: drm/i915: s/_FDI_RXA_.../FDI_RX_...(PIPE_A)/
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_crt.c b/drivers/gpu/drm/i915/intel_crt.c
index cac7928..b84aaa0 100644
--- a/drivers/gpu/drm/i915/intel_crt.c
+++ b/drivers/gpu/drm/i915/intel_crt.c
@@ -890,7 +890,7 @@ void intel_crt_init(struct drm_device *dev)
 		u32 fdi_config = FDI_RX_POLARITY_REVERSED_LPT |
 				 FDI_RX_LINK_REVERSAL_OVERRIDE;
 
-		dev_priv->fdi_rx_config = I915_READ(_FDI_RXA_CTL) & fdi_config;
+		dev_priv->fdi_rx_config = I915_READ(FDI_RX_CTL(PIPE_A)) & fdi_config;
 	}
 
 	intel_crt_reset(connector);
diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
index 2b5a298..9c11577 100644
--- a/drivers/gpu/drm/i915/intel_ddi.c
+++ b/drivers/gpu/drm/i915/intel_ddi.c
@@ -608,7 +608,7 @@ void hsw_fdi_link_train(struct drm_crtc *crtc)
 	 *
 	 * WaFDIAutoLinkSetTimingOverrride:hsw
 	 */
-	I915_WRITE(_FDI_RXA_MISC, FDI_RX_PWRDN_LANE1_VAL(2) |
+	I915_WRITE(FDI_RX_MISC(PIPE_A), FDI_RX_PWRDN_LANE1_VAL(2) |
 				  FDI_RX_PWRDN_LANE0_VAL(2) |
 				  FDI_RX_TP1_TO_TP2_48 | FDI_RX_FDI_DELAY_90);
 
@@ -616,13 +616,13 @@ void hsw_fdi_link_train(struct drm_crtc *crtc)
 	rx_ctl_val = dev_priv->fdi_rx_config | FDI_RX_ENHANCE_FRAME_ENABLE |
 		     FDI_RX_PLL_ENABLE |
 		     FDI_DP_PORT_WIDTH(intel_crtc->config->fdi_lanes);
-	I915_WRITE(_FDI_RXA_CTL, rx_ctl_val);
-	POSTING_READ(_FDI_RXA_CTL);
+	I915_WRITE(FDI_RX_CTL(PIPE_A), rx_ctl_val);
+	POSTING_READ(FDI_RX_CTL(PIPE_A));
 	udelay(220);
 
 	/* Switch from Rawclk to PCDclk */
 	rx_ctl_val |= FDI_PCDCLK;
-	I915_WRITE(_FDI_RXA_CTL, rx_ctl_val);
+	I915_WRITE(FDI_RX_CTL(PIPE_A), rx_ctl_val);
 
 	/* Configure Port Clock Select */
 	I915_WRITE(PORT_CLK_SEL(PORT_E), intel_crtc->config->ddi_pll_sel);
@@ -651,21 +651,21 @@ void hsw_fdi_link_train(struct drm_crtc *crtc)
 		udelay(600);
 
 		/* Program PCH FDI Receiver TU */
-		I915_WRITE(_FDI_RXA_TUSIZE1, TU_SIZE(64));
+		I915_WRITE(FDI_RX_TUSIZE1(PIPE_A), TU_SIZE(64));
 
 		/* Enable PCH FDI Receiver with auto-training */
 		rx_ctl_val |= FDI_RX_ENABLE | FDI_LINK_TRAIN_AUTO;
-		I915_WRITE(_FDI_RXA_CTL, rx_ctl_val);
-		POSTING_READ(_FDI_RXA_CTL);
+		I915_WRITE(FDI_RX_CTL(PIPE_A), rx_ctl_val);
+		POSTING_READ(FDI_RX_CTL(PIPE_A));
 
 		/* Wait for FDI receiver lane calibration */
 		udelay(30);
 
 		/* Unset FDI_RX_MISC pwrdn lanes */
-		temp = I915_READ(_FDI_RXA_MISC);
+		temp = I915_READ(FDI_RX_MISC(PIPE_A));
 		temp &= ~(FDI_RX_PWRDN_LANE1_MASK | FDI_RX_PWRDN_LANE0_MASK);
-		I915_WRITE(_FDI_RXA_MISC, temp);
-		POSTING_READ(_FDI_RXA_MISC);
+		I915_WRITE(FDI_RX_MISC(PIPE_A), temp);
+		POSTING_READ(FDI_RX_MISC(PIPE_A));
 
 		/* Wait for FDI auto training time */
 		udelay(5);
@@ -699,15 +699,15 @@ void hsw_fdi_link_train(struct drm_crtc *crtc)
 		intel_wait_ddi_buf_idle(dev_priv, PORT_E);
 
 		rx_ctl_val &= ~FDI_RX_ENABLE;
-		I915_WRITE(_FDI_RXA_CTL, rx_ctl_val);
-		POSTING_READ(_FDI_RXA_CTL);
+		I915_WRITE(FDI_RX_CTL(PIPE_A), rx_ctl_val);
+		POSTING_READ(FDI_RX_CTL(PIPE_A));
 
 		/* Reset FDI_RX_MISC pwrdn lanes */
-		temp = I915_READ(_FDI_RXA_MISC);
+		temp = I915_READ(FDI_RX_MISC(PIPE_A));
 		temp &= ~(FDI_RX_PWRDN_LANE1_MASK | FDI_RX_PWRDN_LANE0_MASK);
 		temp |= FDI_RX_PWRDN_LANE1_VAL(2) | FDI_RX_PWRDN_LANE0_VAL(2);
-		I915_WRITE(_FDI_RXA_MISC, temp);
-		POSTING_READ(_FDI_RXA_MISC);
+		I915_WRITE(FDI_RX_MISC(PIPE_A), temp);
+		POSTING_READ(FDI_RX_MISC(PIPE_A));
 	}
 
 	DRM_ERROR("FDI link training failed!\n");
@@ -3023,22 +3023,22 @@ void intel_ddi_fdi_disable(struct drm_crtc *crtc)
 
 	intel_ddi_post_disable(intel_encoder);
 
-	val = I915_READ(_FDI_RXA_CTL);
+	val = I915_READ(FDI_RX_CTL(PIPE_A));
 	val &= ~FDI_RX_ENABLE;
-	I915_WRITE(_FDI_RXA_CTL, val);
+	I915_WRITE(FDI_RX_CTL(PIPE_A), val);
 
-	val = I915_READ(_FDI_RXA_MISC);
+	val = I915_READ(FDI_RX_MISC(PIPE_A));
 	val &= ~(FDI_RX_PWRDN_LANE1_MASK | FDI_RX_PWRDN_LANE0_MASK);
 	val |= FDI_RX_PWRDN_LANE1_VAL(2) | FDI_RX_PWRDN_LANE0_VAL(2);
-	I915_WRITE(_FDI_RXA_MISC, val);
+	I915_WRITE(FDI_RX_MISC(PIPE_A), val);
 
-	val = I915_READ(_FDI_RXA_CTL);
+	val = I915_READ(FDI_RX_CTL(PIPE_A));
 	val &= ~FDI_PCDCLK;
-	I915_WRITE(_FDI_RXA_CTL, val);
+	I915_WRITE(FDI_RX_CTL(PIPE_A), val);
 
-	val = I915_READ(_FDI_RXA_CTL);
+	val = I915_READ(FDI_RX_CTL(PIPE_A));
 	val &= ~FDI_RX_PLL_ENABLE;
-	I915_WRITE(_FDI_RXA_CTL, val);
+	I915_WRITE(FDI_RX_CTL(PIPE_A), val);
 }
 
 void intel_ddi_get_config(struct intel_encoder *encoder,
-- 
cgit v0.10.2


From 36c0d0cf33ed31fada15caac34d50555b33208bb Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 18 Sep 2015 20:03:31 +0300
Subject: drm/i915: s/_TRANSA_CHICKEN/TRANS_CHICKEN(PIPE_A)/
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 8e4fb80..e7a3c1b 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -2050,9 +2050,9 @@ static void lpt_enable_pch_transcoder(struct drm_i915_private *dev_priv,
 	assert_fdi_rx_enabled(dev_priv, TRANSCODER_A);
 
 	/* Workaround: set timing override bit. */
-	val = I915_READ(_TRANSA_CHICKEN2);
+	val = I915_READ(TRANS_CHICKEN2(PIPE_A));
 	val |= TRANS_CHICKEN2_TIMING_OVERRIDE;
-	I915_WRITE(_TRANSA_CHICKEN2, val);
+	I915_WRITE(TRANS_CHICKEN2(PIPE_A), val);
 
 	val = TRANS_ENABLE;
 	pipeconf_val = I915_READ(PIPECONF(cpu_transcoder));
@@ -2110,9 +2110,9 @@ static void lpt_disable_pch_transcoder(struct drm_i915_private *dev_priv)
 		DRM_ERROR("Failed to disable PCH transcoder\n");
 
 	/* Workaround: clear timing override bit. */
-	val = I915_READ(_TRANSA_CHICKEN2);
+	val = I915_READ(TRANS_CHICKEN2(PIPE_A));
 	val &= ~TRANS_CHICKEN2_TIMING_OVERRIDE;
-	I915_WRITE(_TRANSA_CHICKEN2, val);
+	I915_WRITE(TRANS_CHICKEN2(PIPE_A), val);
 }
 
 /**
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index 9643a7c..cb409c2 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -6640,8 +6640,8 @@ static void lpt_init_clock_gating(struct drm_device *dev)
 			   PCH_LP_PARTITION_LEVEL_DISABLE);
 
 	/* WADPOClockGatingDisable:hsw */
-	I915_WRITE(_TRANSA_CHICKEN1,
-		   I915_READ(_TRANSA_CHICKEN1) |
+	I915_WRITE(TRANS_CHICKEN1(PIPE_A),
+		   I915_READ(TRANS_CHICKEN1(PIPE_A)) |
 		   TRANS_CHICKEN1_DP0UNIT_GC_DISABLE);
 }
 
-- 
cgit v0.10.2


From 93f253187c2f565678bd7e5ca5f64c1043774f1b Mon Sep 17 00:00:00 2001
From: Alex Dai <yu.dai@intel.com>
Date: Fri, 25 Sep 2015 11:46:56 -0700
Subject: drm/i915/guc: Media domain bit needed when notify GuC rc6 state

GuC expects two bits for Render and Media domain separately when
driver sends data via host2guc SAMPLE_FORCEWAKE. Bit 0 is for
Render and bit 1 is for Media domain.

v2: Keep sync with code for WaRsDoubleRc6WrlWithCoarsePowerGating

v1: Add parameters definition to avoid magic value

Signed-off-by: Alex Dai <yu.dai@intel.com>
Reviewed-by: Tom O'Rourke <Tom.O'Rourke@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_guc_submission.c b/drivers/gpu/drm/i915/i915_guc_submission.c
index 792d0b9..0b1797f 100644
--- a/drivers/gpu/drm/i915/i915_guc_submission.c
+++ b/drivers/gpu/drm/i915/i915_guc_submission.c
@@ -155,12 +155,21 @@ static int host2guc_sample_forcewake(struct intel_guc *guc,
 				     struct i915_guc_client *client)
 {
 	struct drm_i915_private *dev_priv = guc_to_i915(guc);
+	struct drm_device *dev = dev_priv->dev;
 	u32 data[2];
 
 	data[0] = HOST2GUC_ACTION_SAMPLE_FORCEWAKE;
-	data[1] = (intel_enable_rc6(dev_priv->dev)) ? 1 : 0;
+	/* WaRsDisableCoarsePowerGating:skl,bxt */
+	if (!intel_enable_rc6(dev_priv->dev) ||
+	    (IS_BROXTON(dev) && (INTEL_REVID(dev) < BXT_REVID_B0)) ||
+	    (IS_SKL_GT3(dev) && (INTEL_REVID(dev) <= SKL_REVID_E0)) ||
+	    (IS_SKL_GT4(dev) && (INTEL_REVID(dev) <= SKL_REVID_E0)))
+		data[1] = 0;
+	else
+		/* bit 0 and 1 are for Render and Media domain separately */
+		data[1] = GUC_FORCEWAKE_RENDER | GUC_FORCEWAKE_MEDIA;
 
-	return host2guc_action(guc, data, 2);
+	return host2guc_action(guc, data, ARRAY_SIZE(data));
 }
 
 /*
diff --git a/drivers/gpu/drm/i915/intel_guc_fwif.h b/drivers/gpu/drm/i915/intel_guc_fwif.h
index e1f47ba..6c78fdf 100644
--- a/drivers/gpu/drm/i915/intel_guc_fwif.h
+++ b/drivers/gpu/drm/i915/intel_guc_fwif.h
@@ -218,6 +218,9 @@ struct guc_context_desc {
 	u64 desc_private;
 } __packed;
 
+#define GUC_FORCEWAKE_RENDER	(1 << 0)
+#define GUC_FORCEWAKE_MEDIA	(1 << 1)
+
 /* This Action will be programmed in C180 - SOFT_SCRATCH_O_REG */
 enum host2guc_action {
 	HOST2GUC_ACTION_DEFAULT = 0x0,
-- 
cgit v0.10.2


From d8135109e45d4cf352387b7de66c216a176e37f0 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Tue, 29 Sep 2015 16:28:46 +0300
Subject: drm/i915/bxt: fix RC6 residency time calculation
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The RC6 residency time unit is 833.33ns on BXT according to the
specification, so update the calculation accordingly. Use the same way
as CHV/VLV to divide by the corresponding frequency, as I think this is
the more natural unit for what the HW does internally.

v2:
- add missing IS_BROXTON check (Ville)

Testcase: igt/pm_rc6_residency
Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_sysfs.c b/drivers/gpu/drm/i915/i915_sysfs.c
index 74086eb..50ce9ce 100644
--- a/drivers/gpu/drm/i915/i915_sysfs.c
+++ b/drivers/gpu/drm/i915/i915_sysfs.c
@@ -54,6 +54,9 @@ static u32 calc_residency(struct drm_device *dev, const u32 reg)
 
 		if (I915_READ(VLV_COUNTER_CONTROL) & VLV_COUNT_RANGE_HIGH)
 			units <<= 8;
+	} else if (IS_BROXTON(dev)) {
+		units = 1;
+		div = 1200;		/* 833.33ns */
 	}
 
 	raw_time = I915_READ(reg) * units;
-- 
cgit v0.10.2


From 37d9078b9b395185947784ba8d71c0980e5aacad Mon Sep 17 00:00:00 2001
From: Matt Roper <matthew.d.roper@intel.com>
Date: Thu, 24 Sep 2015 15:53:06 -0700
Subject: drm/i915: Drop redundant watermark programming

In commit

        commit e4ca061275ec6a48b66c6edebe08644e666994c0
        Author: Patrik Jakobsson <patrik.jakobsson@linux.intel.com>
        Date:   Wed Jul 8 15:31:52 2015 +0200

            drm/i915: Don't forget to mark crtc as inactive after disable

we added extra watermark updates to all of the .crtc_disable()
entrypoints to avoid problems problems with system resume on SKL.  Those
disable entrypoints are currently called in just two places in the
driver: intel_atomic_commit (i.e., during a modeset) and
intel_crtc_disable_noatomic (which is called during hardware readout).
It seems that this extra watermark recalculation should only be
important in the latter case (which happens during a resume operation);
the former case should always have appropriate watermark programming
happening at other points in the modeset sequence.

Let's move the watermark update out of the .crtc_disable() entrypoints
and place it directly in intel_crtc_disable_noatomic() so that it only
happens on S3 resume and not during a regular modeset (since the
existing watermark handling should properly update watermarks during
normal atomic commits).

Cc: Patrik Jakobsson <patrik.jakobsson@linux.intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index e7a3c1b..f2df9b1 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -5115,9 +5115,6 @@ static void ironlake_crtc_disable(struct drm_crtc *crtc)
 
 		ironlake_fdi_pll_disable(intel_crtc);
 	}
-
-	intel_crtc->active = false;
-	intel_update_watermarks(crtc);
 }
 
 static void haswell_crtc_disable(struct drm_crtc *crtc)
@@ -5161,9 +5158,6 @@ static void haswell_crtc_disable(struct drm_crtc *crtc)
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		if (encoder->post_disable)
 			encoder->post_disable(encoder);
-
-	intel_crtc->active = false;
-	intel_update_watermarks(crtc);
 }
 
 static void i9xx_pfit_enable(struct intel_crtc *crtc)
@@ -6254,9 +6248,6 @@ static void i9xx_crtc_disable(struct drm_crtc *crtc)
 
 	if (!IS_GEN2(dev))
 		intel_set_cpu_fifo_underrun_reporting(dev_priv, pipe, false);
-
-	intel_crtc->active = false;
-	intel_update_watermarks(crtc);
 }
 
 static void intel_crtc_disable_noatomic(struct drm_crtc *crtc)
@@ -6276,6 +6267,8 @@ static void intel_crtc_disable_noatomic(struct drm_crtc *crtc)
 
 	intel_crtc_disable_planes(crtc, crtc->state->plane_mask);
 	dev_priv->display.crtc_disable(crtc);
+	intel_crtc->active = false;
+	intel_update_watermarks(crtc);
 	intel_disable_shared_dpll(intel_crtc);
 
 	domains = intel_crtc->enabled_power_domains;
-- 
cgit v0.10.2


From 43d59eda1f69631c267e06ab6b94ed3c14f1f6d1 Mon Sep 17 00:00:00 2001
From: Matt Roper <matthew.d.roper@intel.com>
Date: Thu, 24 Sep 2015 15:53:07 -0700
Subject: drm/i915: Eliminate usage of plane_wm_parameters from ILK-style WM
 code (v2)

Just pull the info out of the plane state structure rather than staging
it in an additional structure.

v2: Add 'visible' condition to sprites_scaled so that we don't limit the
    WM level when the sprite isn't enabled.  (Ville)

Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by(v1): Ander Conselvan de Oliveira <conselvan2@gmail.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index cb409c2..bffb6eb 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -1777,9 +1777,6 @@ struct ilk_pipe_wm_parameters {
 	bool active;
 	uint32_t pipe_htotal;
 	uint32_t pixel_rate;
-	struct intel_plane_wm_parameters pri;
-	struct intel_plane_wm_parameters spr;
-	struct intel_plane_wm_parameters cur;
 };
 
 struct ilk_wm_maximums {
@@ -1801,25 +1798,25 @@ struct intel_wm_config {
  * mem_value must be in 0.1us units.
  */
 static uint32_t ilk_compute_pri_wm(const struct ilk_pipe_wm_parameters *params,
+				   const struct intel_plane_state *pstate,
 				   uint32_t mem_value,
 				   bool is_lp)
 {
+	int bpp = pstate->base.fb ? pstate->base.fb->bits_per_pixel / 8 : 0;
 	uint32_t method1, method2;
 
-	if (!params->active || !params->pri.enabled)
+	if (!params->active || !pstate->visible)
 		return 0;
 
-	method1 = ilk_wm_method1(params->pixel_rate,
-				 params->pri.bytes_per_pixel,
-				 mem_value);
+	method1 = ilk_wm_method1(params->pixel_rate, bpp, mem_value);
 
 	if (!is_lp)
 		return method1;
 
 	method2 = ilk_wm_method2(params->pixel_rate,
 				 params->pipe_htotal,
-				 params->pri.horiz_pixels,
-				 params->pri.bytes_per_pixel,
+				 drm_rect_width(&pstate->dst),
+				 bpp,
 				 mem_value);
 
 	return min(method1, method2);
@@ -1830,20 +1827,20 @@ static uint32_t ilk_compute_pri_wm(const struct ilk_pipe_wm_parameters *params,
  * mem_value must be in 0.1us units.
  */
 static uint32_t ilk_compute_spr_wm(const struct ilk_pipe_wm_parameters *params,
+				   const struct intel_plane_state *pstate,
 				   uint32_t mem_value)
 {
+	int bpp = pstate->base.fb ? pstate->base.fb->bits_per_pixel / 8 : 0;
 	uint32_t method1, method2;
 
-	if (!params->active || !params->spr.enabled)
+	if (!params->active || !pstate->visible)
 		return 0;
 
-	method1 = ilk_wm_method1(params->pixel_rate,
-				 params->spr.bytes_per_pixel,
-				 mem_value);
+	method1 = ilk_wm_method1(params->pixel_rate, bpp, mem_value);
 	method2 = ilk_wm_method2(params->pixel_rate,
 				 params->pipe_htotal,
-				 params->spr.horiz_pixels,
-				 params->spr.bytes_per_pixel,
+				 drm_rect_width(&pstate->dst),
+				 bpp,
 				 mem_value);
 	return min(method1, method2);
 }
@@ -1853,28 +1850,32 @@ static uint32_t ilk_compute_spr_wm(const struct ilk_pipe_wm_parameters *params,
  * mem_value must be in 0.1us units.
  */
 static uint32_t ilk_compute_cur_wm(const struct ilk_pipe_wm_parameters *params,
+				   const struct intel_plane_state *pstate,
 				   uint32_t mem_value)
 {
-	if (!params->active || !params->cur.enabled)
+	int bpp = pstate->base.fb ? pstate->base.fb->bits_per_pixel / 8 : 0;
+
+	if (!params->active || !pstate->visible)
 		return 0;
 
 	return ilk_wm_method2(params->pixel_rate,
 			      params->pipe_htotal,
-			      params->cur.horiz_pixels,
-			      params->cur.bytes_per_pixel,
+			      drm_rect_width(&pstate->dst),
+			      bpp,
 			      mem_value);
 }
 
 /* Only for WM_LP. */
 static uint32_t ilk_compute_fbc_wm(const struct ilk_pipe_wm_parameters *params,
+				   const struct intel_plane_state *pstate,
 				   uint32_t pri_val)
 {
-	if (!params->active || !params->pri.enabled)
+	int bpp = pstate->base.fb ? pstate->base.fb->bits_per_pixel / 8 : 0;
+
+	if (!params->active || !pstate->visible)
 		return 0;
 
-	return ilk_wm_fbc(pri_val,
-			  params->pri.horiz_pixels,
-			  params->pri.bytes_per_pixel);
+	return ilk_wm_fbc(pri_val, drm_rect_width(&pstate->dst), bpp);
 }
 
 static unsigned int ilk_display_fifo_size(const struct drm_device *dev)
@@ -2039,10 +2040,12 @@ static bool ilk_validate_wm_level(int level,
 }
 
 static void ilk_compute_wm_level(const struct drm_i915_private *dev_priv,
+				 const struct intel_crtc *intel_crtc,
 				 int level,
 				 const struct ilk_pipe_wm_parameters *p,
 				 struct intel_wm_level *result)
 {
+	struct intel_plane *intel_plane;
 	uint16_t pri_latency = dev_priv->wm.pri_latency[level];
 	uint16_t spr_latency = dev_priv->wm.spr_latency[level];
 	uint16_t cur_latency = dev_priv->wm.cur_latency[level];
@@ -2054,10 +2057,29 @@ static void ilk_compute_wm_level(const struct drm_i915_private *dev_priv,
 		cur_latency *= 5;
 	}
 
-	result->pri_val = ilk_compute_pri_wm(p, pri_latency, level);
-	result->spr_val = ilk_compute_spr_wm(p, spr_latency);
-	result->cur_val = ilk_compute_cur_wm(p, cur_latency);
-	result->fbc_val = ilk_compute_fbc_wm(p, result->pri_val);
+	for_each_intel_plane_on_crtc(dev_priv->dev, intel_crtc, intel_plane) {
+		struct intel_plane_state *pstate =
+			to_intel_plane_state(intel_plane->base.state);
+
+		switch (intel_plane->base.type) {
+		case DRM_PLANE_TYPE_PRIMARY:
+			result->pri_val = ilk_compute_pri_wm(p, pstate,
+							     pri_latency,
+							     level);
+			result->fbc_val = ilk_compute_fbc_wm(p, pstate,
+							     result->pri_val);
+			break;
+		case DRM_PLANE_TYPE_OVERLAY:
+			result->spr_val = ilk_compute_spr_wm(p, pstate,
+							     spr_latency);
+			break;
+		case DRM_PLANE_TYPE_CURSOR:
+			result->cur_val = ilk_compute_cur_wm(p, pstate,
+							     cur_latency);
+			break;
+		}
+	}
+
 	result->enable = true;
 }
 
@@ -2319,10 +2341,7 @@ static void skl_setup_wm_latency(struct drm_device *dev)
 static void ilk_compute_wm_parameters(struct drm_crtc *crtc,
 				      struct ilk_pipe_wm_parameters *p)
 {
-	struct drm_device *dev = crtc->dev;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	enum pipe pipe = intel_crtc->pipe;
-	struct drm_plane *plane;
 
 	if (!intel_crtc->active)
 		return;
@@ -2330,32 +2349,6 @@ static void ilk_compute_wm_parameters(struct drm_crtc *crtc,
 	p->active = true;
 	p->pipe_htotal = intel_crtc->config->base.adjusted_mode.crtc_htotal;
 	p->pixel_rate = ilk_pipe_pixel_rate(intel_crtc->config);
-
-	if (crtc->primary->state->fb)
-		p->pri.bytes_per_pixel =
-			crtc->primary->state->fb->bits_per_pixel / 8;
-	else
-		p->pri.bytes_per_pixel = 4;
-
-	p->cur.bytes_per_pixel = 4;
-	/*
-	 * TODO: for now, assume primary and cursor planes are always enabled.
-	 * Setting them to false makes the screen flicker.
-	 */
-	p->pri.enabled = true;
-	p->cur.enabled = true;
-
-	p->pri.horiz_pixels = intel_crtc->config->pipe_src_w;
-	p->cur.horiz_pixels = intel_crtc->base.cursor->state->crtc_w;
-
-	drm_for_each_legacy_plane(plane, dev) {
-		struct intel_plane *intel_plane = to_intel_plane(plane);
-
-		if (intel_plane->pipe == pipe) {
-			p->spr = intel_plane->wm;
-			break;
-		}
-	}
 }
 
 static void ilk_compute_wm_config(struct drm_device *dev,
@@ -2383,28 +2376,42 @@ static bool intel_compute_pipe_wm(struct drm_crtc *crtc,
 {
 	struct drm_device *dev = crtc->dev;
 	const struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	struct intel_plane *intel_plane;
+	struct intel_plane_state *sprstate = NULL;
 	int level, max_level = ilk_wm_max_level(dev);
 	/* LP0 watermark maximums depend on this pipe alone */
 	struct intel_wm_config config = {
 		.num_pipes_active = 1,
-		.sprites_enabled = params->spr.enabled,
-		.sprites_scaled = params->spr.scaled,
 	};
 	struct ilk_wm_maximums max;
 
+	for_each_intel_plane_on_crtc(dev, intel_crtc, intel_plane) {
+		if (intel_plane->base.type == DRM_PLANE_TYPE_OVERLAY) {
+			sprstate = to_intel_plane_state(intel_plane->base.state);
+			break;
+		}
+	}
+
+	config.sprites_enabled = sprstate->visible;
+	config.sprites_scaled = sprstate->visible &&
+		(drm_rect_width(&sprstate->dst) != drm_rect_width(&sprstate->src) >> 16 ||
+		drm_rect_height(&sprstate->dst) != drm_rect_height(&sprstate->src) >> 16);
+
+
 	pipe_wm->pipe_enabled = params->active;
-	pipe_wm->sprites_enabled = params->spr.enabled;
-	pipe_wm->sprites_scaled = params->spr.scaled;
+	pipe_wm->sprites_enabled = sprstate->visible;
+	pipe_wm->sprites_scaled = config.sprites_scaled;
 
 	/* ILK/SNB: LP2+ watermarks only w/o sprites */
-	if (INTEL_INFO(dev)->gen <= 6 && params->spr.enabled)
+	if (INTEL_INFO(dev)->gen <= 6 && sprstate->visible)
 		max_level = 1;
 
 	/* ILK/SNB/IVB: LP1+ watermarks only w/o scaling */
-	if (params->spr.scaled)
+	if (config.sprites_scaled)
 		max_level = 0;
 
-	ilk_compute_wm_level(dev_priv, 0, params, &pipe_wm->wm[0]);
+	ilk_compute_wm_level(dev_priv, intel_crtc, 0, params, &pipe_wm->wm[0]);
 
 	if (IS_HASWELL(dev) || IS_BROADWELL(dev))
 		pipe_wm->linetime = hsw_compute_linetime_wm(dev, crtc);
@@ -2421,7 +2428,7 @@ static bool intel_compute_pipe_wm(struct drm_crtc *crtc,
 	for (level = 1; level <= max_level; level++) {
 		struct intel_wm_level wm = {};
 
-		ilk_compute_wm_level(dev_priv, level, params, &wm);
+		ilk_compute_wm_level(dev_priv, intel_crtc, level, params, &wm);
 
 		/*
 		 * Disable any watermark level that exceeds the
-- 
cgit v0.10.2


From 7221fc333dbe1743a3dff155b03527fda90d4ec1 Mon Sep 17 00:00:00 2001
From: Matt Roper <matthew.d.roper@intel.com>
Date: Thu, 24 Sep 2015 15:53:08 -0700
Subject: drm/i915: Eliminate usage of pipe_wm_parameters from ILK-style WM
 (v2)

Just pull the info out of the CRTC state structure rather than staging
it in an additional structure.

Note that we use cstate->active rather than intel_crtc->active which may
appear to be a change in behavior.  However since we're no longer trying
to recalculate watermarks during the "pipe off" stage of a modeset,
intel_crtc->active and cstate->active should always be identical when
watermarks are calculated (at least for ILK-style platforms).

v2: Clarify reasoning for cstate->active and add a WARN_ON to the code
    to assert that it really is always identical to intel_crtc->active
    as expected.

Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index bffb6eb..a9615ba 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -1773,12 +1773,6 @@ struct skl_pipe_wm_parameters {
 	struct intel_plane_wm_parameters cursor;
 };
 
-struct ilk_pipe_wm_parameters {
-	bool active;
-	uint32_t pipe_htotal;
-	uint32_t pixel_rate;
-};
-
 struct ilk_wm_maximums {
 	uint16_t pri;
 	uint16_t spr;
@@ -1797,7 +1791,7 @@ struct intel_wm_config {
  * For both WM_PIPE and WM_LP.
  * mem_value must be in 0.1us units.
  */
-static uint32_t ilk_compute_pri_wm(const struct ilk_pipe_wm_parameters *params,
+static uint32_t ilk_compute_pri_wm(const struct intel_crtc_state *cstate,
 				   const struct intel_plane_state *pstate,
 				   uint32_t mem_value,
 				   bool is_lp)
@@ -1805,16 +1799,16 @@ static uint32_t ilk_compute_pri_wm(const struct ilk_pipe_wm_parameters *params,
 	int bpp = pstate->base.fb ? pstate->base.fb->bits_per_pixel / 8 : 0;
 	uint32_t method1, method2;
 
-	if (!params->active || !pstate->visible)
+	if (!cstate->base.active || !pstate->visible)
 		return 0;
 
-	method1 = ilk_wm_method1(params->pixel_rate, bpp, mem_value);
+	method1 = ilk_wm_method1(ilk_pipe_pixel_rate(cstate), bpp, mem_value);
 
 	if (!is_lp)
 		return method1;
 
-	method2 = ilk_wm_method2(params->pixel_rate,
-				 params->pipe_htotal,
+	method2 = ilk_wm_method2(ilk_pipe_pixel_rate(cstate),
+				 cstate->base.adjusted_mode.crtc_htotal,
 				 drm_rect_width(&pstate->dst),
 				 bpp,
 				 mem_value);
@@ -1826,19 +1820,19 @@ static uint32_t ilk_compute_pri_wm(const struct ilk_pipe_wm_parameters *params,
  * For both WM_PIPE and WM_LP.
  * mem_value must be in 0.1us units.
  */
-static uint32_t ilk_compute_spr_wm(const struct ilk_pipe_wm_parameters *params,
+static uint32_t ilk_compute_spr_wm(const struct intel_crtc_state *cstate,
 				   const struct intel_plane_state *pstate,
 				   uint32_t mem_value)
 {
 	int bpp = pstate->base.fb ? pstate->base.fb->bits_per_pixel / 8 : 0;
 	uint32_t method1, method2;
 
-	if (!params->active || !pstate->visible)
+	if (!cstate->base.active || !pstate->visible)
 		return 0;
 
-	method1 = ilk_wm_method1(params->pixel_rate, bpp, mem_value);
-	method2 = ilk_wm_method2(params->pixel_rate,
-				 params->pipe_htotal,
+	method1 = ilk_wm_method1(ilk_pipe_pixel_rate(cstate), bpp, mem_value);
+	method2 = ilk_wm_method2(ilk_pipe_pixel_rate(cstate),
+				 cstate->base.adjusted_mode.crtc_htotal,
 				 drm_rect_width(&pstate->dst),
 				 bpp,
 				 mem_value);
@@ -1849,30 +1843,30 @@ static uint32_t ilk_compute_spr_wm(const struct ilk_pipe_wm_parameters *params,
  * For both WM_PIPE and WM_LP.
  * mem_value must be in 0.1us units.
  */
-static uint32_t ilk_compute_cur_wm(const struct ilk_pipe_wm_parameters *params,
+static uint32_t ilk_compute_cur_wm(const struct intel_crtc_state *cstate,
 				   const struct intel_plane_state *pstate,
 				   uint32_t mem_value)
 {
 	int bpp = pstate->base.fb ? pstate->base.fb->bits_per_pixel / 8 : 0;
 
-	if (!params->active || !pstate->visible)
+	if (!cstate->base.active || !pstate->visible)
 		return 0;
 
-	return ilk_wm_method2(params->pixel_rate,
-			      params->pipe_htotal,
+	return ilk_wm_method2(ilk_pipe_pixel_rate(cstate),
+			      cstate->base.adjusted_mode.crtc_htotal,
 			      drm_rect_width(&pstate->dst),
 			      bpp,
 			      mem_value);
 }
 
 /* Only for WM_LP. */
-static uint32_t ilk_compute_fbc_wm(const struct ilk_pipe_wm_parameters *params,
+static uint32_t ilk_compute_fbc_wm(const struct intel_crtc_state *cstate,
 				   const struct intel_plane_state *pstate,
 				   uint32_t pri_val)
 {
 	int bpp = pstate->base.fb ? pstate->base.fb->bits_per_pixel / 8 : 0;
 
-	if (!params->active || !pstate->visible)
+	if (!cstate->base.active || !pstate->visible)
 		return 0;
 
 	return ilk_wm_fbc(pri_val, drm_rect_width(&pstate->dst), bpp);
@@ -2042,7 +2036,7 @@ static bool ilk_validate_wm_level(int level,
 static void ilk_compute_wm_level(const struct drm_i915_private *dev_priv,
 				 const struct intel_crtc *intel_crtc,
 				 int level,
-				 const struct ilk_pipe_wm_parameters *p,
+				 struct intel_crtc_state *cstate,
 				 struct intel_wm_level *result)
 {
 	struct intel_plane *intel_plane;
@@ -2063,18 +2057,18 @@ static void ilk_compute_wm_level(const struct drm_i915_private *dev_priv,
 
 		switch (intel_plane->base.type) {
 		case DRM_PLANE_TYPE_PRIMARY:
-			result->pri_val = ilk_compute_pri_wm(p, pstate,
+			result->pri_val = ilk_compute_pri_wm(cstate, pstate,
 							     pri_latency,
 							     level);
-			result->fbc_val = ilk_compute_fbc_wm(p, pstate,
+			result->fbc_val = ilk_compute_fbc_wm(cstate, pstate,
 							     result->pri_val);
 			break;
 		case DRM_PLANE_TYPE_OVERLAY:
-			result->spr_val = ilk_compute_spr_wm(p, pstate,
+			result->spr_val = ilk_compute_spr_wm(cstate, pstate,
 							     spr_latency);
 			break;
 		case DRM_PLANE_TYPE_CURSOR:
-			result->cur_val = ilk_compute_cur_wm(p, pstate,
+			result->cur_val = ilk_compute_cur_wm(cstate, pstate,
 							     cur_latency);
 			break;
 		}
@@ -2338,19 +2332,6 @@ static void skl_setup_wm_latency(struct drm_device *dev)
 	intel_print_wm_latency(dev, "Gen9 Plane", dev_priv->wm.skl_latency);
 }
 
-static void ilk_compute_wm_parameters(struct drm_crtc *crtc,
-				      struct ilk_pipe_wm_parameters *p)
-{
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-
-	if (!intel_crtc->active)
-		return;
-
-	p->active = true;
-	p->pipe_htotal = intel_crtc->config->base.adjusted_mode.crtc_htotal;
-	p->pixel_rate = ilk_pipe_pixel_rate(intel_crtc->config);
-}
-
 static void ilk_compute_wm_config(struct drm_device *dev,
 				  struct intel_wm_config *config)
 {
@@ -2370,10 +2351,10 @@ static void ilk_compute_wm_config(struct drm_device *dev,
 }
 
 /* Compute new watermarks for the pipe */
-static bool intel_compute_pipe_wm(struct drm_crtc *crtc,
-				  const struct ilk_pipe_wm_parameters *params,
+static bool intel_compute_pipe_wm(struct intel_crtc_state *cstate,
 				  struct intel_pipe_wm *pipe_wm)
 {
+	struct drm_crtc *crtc = cstate->base.crtc;
 	struct drm_device *dev = crtc->dev;
 	const struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
@@ -2398,8 +2379,7 @@ static bool intel_compute_pipe_wm(struct drm_crtc *crtc,
 		(drm_rect_width(&sprstate->dst) != drm_rect_width(&sprstate->src) >> 16 ||
 		drm_rect_height(&sprstate->dst) != drm_rect_height(&sprstate->src) >> 16);
 
-
-	pipe_wm->pipe_enabled = params->active;
+	pipe_wm->pipe_enabled = cstate->base.active;
 	pipe_wm->sprites_enabled = sprstate->visible;
 	pipe_wm->sprites_scaled = config.sprites_scaled;
 
@@ -2411,7 +2391,7 @@ static bool intel_compute_pipe_wm(struct drm_crtc *crtc,
 	if (config.sprites_scaled)
 		max_level = 0;
 
-	ilk_compute_wm_level(dev_priv, intel_crtc, 0, params, &pipe_wm->wm[0]);
+	ilk_compute_wm_level(dev_priv, intel_crtc, 0, cstate, &pipe_wm->wm[0]);
 
 	if (IS_HASWELL(dev) || IS_BROADWELL(dev))
 		pipe_wm->linetime = hsw_compute_linetime_wm(dev, crtc);
@@ -2428,7 +2408,7 @@ static bool intel_compute_pipe_wm(struct drm_crtc *crtc,
 	for (level = 1; level <= max_level; level++) {
 		struct intel_wm_level wm = {};
 
-		ilk_compute_wm_level(dev_priv, intel_crtc, level, params, &wm);
+		ilk_compute_wm_level(dev_priv, intel_crtc, level, cstate, &wm);
 
 		/*
 		 * Disable any watermark level that exceeds the
@@ -3759,19 +3739,19 @@ skl_update_sprite_wm(struct drm_plane *plane, struct drm_crtc *crtc,
 static void ilk_update_wm(struct drm_crtc *crtc)
 {
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	struct intel_crtc_state *cstate = to_intel_crtc_state(crtc->state);
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct ilk_wm_maximums max;
-	struct ilk_pipe_wm_parameters params = {};
 	struct ilk_wm_values results = {};
 	enum intel_ddb_partitioning partitioning;
 	struct intel_pipe_wm pipe_wm = {};
 	struct intel_pipe_wm lp_wm_1_2 = {}, lp_wm_5_6 = {}, *best_lp_wm;
 	struct intel_wm_config config = {};
 
-	ilk_compute_wm_parameters(crtc, &params);
+	WARN_ON(cstate->base.active != intel_crtc->active);
 
-	intel_compute_pipe_wm(crtc, &params, &pipe_wm);
+	intel_compute_pipe_wm(cstate, &pipe_wm);
 
 	if (!memcmp(&intel_crtc->wm.active, &pipe_wm, sizeof(pipe_wm)))
 		return;
@@ -3811,12 +3791,6 @@ ilk_update_sprite_wm(struct drm_plane *plane,
 	struct drm_device *dev = plane->dev;
 	struct intel_plane *intel_plane = to_intel_plane(plane);
 
-	intel_plane->wm.enabled = enabled;
-	intel_plane->wm.scaled = scaled;
-	intel_plane->wm.horiz_pixels = sprite_width;
-	intel_plane->wm.vert_pixels = sprite_width;
-	intel_plane->wm.bytes_per_pixel = pixel_size;
-
 	/*
 	 * IVB workaround: must disable low power watermarks for at least
 	 * one frame before enabling scaling.  LP watermarks can be re-enabled
-- 
cgit v0.10.2


From 31409e97ef708ba52ddcb9a7b65b8b878ecc08f3 Mon Sep 17 00:00:00 2001
From: Matt Roper <matthew.d.roper@intel.com>
Date: Thu, 24 Sep 2015 15:53:09 -0700
Subject: drm/i915: Determine I915_MAX_PLANES from plane enum

Let the compiler figure out what I915_MAX_PLANES is from 'enum plane' so
that we don't need a separate #define.

While we're at it, add the cursor plane to the enum.  This will cause
I915_MAX_PLANES to now include the cursor plane in its count (it didn't
previously).   This change is safe since we currently only use this
value in array declarations (never in the actual code logic); we just
wind up allocating slightly more memory than we need to.  A followup
patch will cause various parts of the code to start using the extra
array element where appropriate.

(This patch probably should have been squashed with the followup patch,
but I couldn't figure out how to get Coccinelle to modify enum
declarations...)

Suggested-by: Ander Conselvan De Oliveira <conselvan2@gmail.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index ee478ac..497ab4e 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -131,17 +131,17 @@ enum transcoder {
 #define transcoder_name(t) ((t) + 'A')
 
 /*
- * This is the maximum (across all platforms) number of planes (primary +
- * sprites) that can be active at the same time on one pipe.
- *
- * This value doesn't count the cursor plane.
+ * I915_MAX_PLANES in the enum below is the maximum (across all platforms)
+ * number of planes per CRTC.  Not all platforms really have this many planes,
+ * which means some arrays of size I915_MAX_PLANES may have unused entries
+ * between the topmost sprite plane and the cursor plane.
  */
-#define I915_MAX_PLANES	4
-
 enum plane {
 	PLANE_A = 0,
 	PLANE_B,
 	PLANE_C,
+	PLANE_CURSOR,
+	I915_MAX_PLANES,
 };
 #define plane_name(p) ((p) + 'A')
 
-- 
cgit v0.10.2


From 4969d33ed91d51262691e0479faad16f57688146 Mon Sep 17 00:00:00 2001
From: Matt Roper <matthew.d.roper@intel.com>
Date: Thu, 24 Sep 2015 15:53:10 -0700
Subject: drm/i915/skl: Simplify wm structures slightly (v2)

A bunch of SKL watermark-related structures have the cursor plane as a
separate entry from the rest of the planes.  Since a previous patch
updated I915_MAX_PLANES such that those plane arrays now have a slot for
the cursor, update the code to use the new slot in the existing plane
arrays and kill off the cursor-specific structures.

There shouldn't be any functional change here; this is just shuffling
around how the data is stored in some of the data structures.  The whole
patch is generated with Coccinelle via the following semantic patch:

        @@ struct skl_pipe_wm_parameters WMP; @@
        - WMP.cursor
        + WMP.plane[PLANE_CURSOR]

        @@ struct skl_pipe_wm_parameters *WMP; @@
        - WMP->cursor
        + WMP->plane[PLANE_CURSOR]

        @@ @@
        struct skl_pipe_wm_parameters {
        ...
        - struct intel_plane_wm_parameters cursor;
        ...
        };

        @@
        struct skl_ddb_allocation DDB;
        expression E;
        @@
        - DDB.cursor[E]
        + DDB.plane[E][PLANE_CURSOR]

        @@
        struct skl_ddb_allocation *DDB;
        expression E;
        @@
        - DDB->cursor[E]
        + DDB->plane[E][PLANE_CURSOR]

        @@ @@
        struct skl_ddb_allocation {
        ...
        - struct skl_ddb_entry cursor[I915_MAX_PIPES];
        ...
        };

        @@
        struct skl_wm_values WMV;
        expression E1, E2;
        @@
        (
        - WMV.cursor[E1][E2]
        + WMV.plane[E1][PLANE_CURSOR][E2]
        |
        - WMV.cursor_trans[E1]
        + WMV.plane_trans[E1][PLANE_CURSOR]
        )

        @@
        struct skl_wm_values *WMV;
        expression E1, E2;
        @@
        (
        - WMV->cursor[E1][E2]
        + WMV->plane[E1][PLANE_CURSOR][E2]
        |
        - WMV->cursor_trans[E1]
        + WMV->plane_trans[E1][PLANE_CURSOR]
        )

        @@ @@
        struct skl_wm_values {
        ...
        - uint32_t cursor[I915_MAX_PIPES][8];
        ...
        - uint32_t cursor_trans[I915_MAX_PIPES];
        ...
        };

        @@ struct skl_wm_level WML; @@
        (
        - WML.cursor_en
        + WML.plane_en[PLANE_CURSOR]
        |
        - WML.cursor_res_b
        + WML.plane_res_b[PLANE_CURSOR]
        |
        - WML.cursor_res_l
        + WML.plane_res_l[PLANE_CURSOR]
        )

        @@ struct skl_wm_level *WML; @@
        (
        - WML->cursor_en
        + WML->plane_en[PLANE_CURSOR]
        |
        - WML->cursor_res_b
        + WML->plane_res_b[PLANE_CURSOR]
        |
        - WML->cursor_res_l
        + WML->plane_res_l[PLANE_CURSOR]
        )

        @@ @@
        struct skl_wm_level {
        ...
        - bool cursor_en;
        ...
        - uint16_t cursor_res_b;
        - uint8_t cursor_res_l;
        ...
        };

v2: Use a PLANE_CURSOR enum entry rather than making the code reference
    I915_MAX_PLANES or I915_MAX_PLANES+1, which was confusing.  (Ander)

Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index 2ac1ba8..77ee8c5 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -3155,7 +3155,7 @@ static int i915_ddb_info(struct seq_file *m, void *unused)
 				   skl_ddb_entry_size(entry));
 		}
 
-		entry = &ddb->cursor[pipe];
+		entry = &ddb->plane[pipe][PLANE_CURSOR];
 		seq_printf(m, "  %-13s%8u%8u%8u\n", "Cursor", entry->start,
 			   entry->end, skl_ddb_entry_size(entry));
 	}
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 497ab4e..29ce11f 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -1578,8 +1578,7 @@ static inline bool skl_ddb_entry_equal(const struct skl_ddb_entry *e1,
 struct skl_ddb_allocation {
 	struct skl_ddb_entry pipe[I915_MAX_PIPES];
 	struct skl_ddb_entry plane[I915_MAX_PIPES][I915_MAX_PLANES]; /* packed/uv */
-	struct skl_ddb_entry y_plane[I915_MAX_PIPES][I915_MAX_PLANES]; /* y-plane */
-	struct skl_ddb_entry cursor[I915_MAX_PIPES];
+	struct skl_ddb_entry y_plane[I915_MAX_PIPES][I915_MAX_PLANES];
 };
 
 struct skl_wm_values {
@@ -1587,18 +1586,13 @@ struct skl_wm_values {
 	struct skl_ddb_allocation ddb;
 	uint32_t wm_linetime[I915_MAX_PIPES];
 	uint32_t plane[I915_MAX_PIPES][I915_MAX_PLANES][8];
-	uint32_t cursor[I915_MAX_PIPES][8];
 	uint32_t plane_trans[I915_MAX_PIPES][I915_MAX_PLANES];
-	uint32_t cursor_trans[I915_MAX_PIPES];
 };
 
 struct skl_wm_level {
 	bool plane_en[I915_MAX_PLANES];
-	bool cursor_en;
 	uint16_t plane_res_b[I915_MAX_PLANES];
 	uint8_t plane_res_l[I915_MAX_PLANES];
-	uint16_t cursor_res_b;
-	uint8_t cursor_res_l;
 };
 
 /*
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index f2df9b1..305abaa 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -12587,8 +12587,8 @@ static void check_wm_state(struct drm_device *dev)
 		}
 
 		/* cursor */
-		hw_entry = &hw_ddb.cursor[pipe];
-		sw_entry = &sw_ddb->cursor[pipe];
+		hw_entry = &hw_ddb.plane[pipe][PLANE_CURSOR];
+		sw_entry = &sw_ddb->plane[pipe][PLANE_CURSOR];
 
 		if (skl_ddb_entry_equal(hw_entry, sw_entry))
 			continue;
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index a9615ba..0004362 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -1770,7 +1770,6 @@ struct skl_pipe_wm_parameters {
 	uint32_t pipe_htotal;
 	uint32_t pixel_rate; /* in KHz */
 	struct intel_plane_wm_parameters plane[I915_MAX_PLANES];
-	struct intel_plane_wm_parameters cursor;
 };
 
 struct ilk_wm_maximums {
@@ -2884,7 +2883,8 @@ void skl_ddb_get_hw_state(struct drm_i915_private *dev_priv,
 		}
 
 		val = I915_READ(CUR_BUF_CFG(pipe));
-		skl_ddb_entry_init_from_hw(&ddb->cursor[pipe], val);
+		skl_ddb_entry_init_from_hw(&ddb->plane[pipe][PLANE_CURSOR],
+					   val);
 	}
 }
 
@@ -2953,13 +2953,14 @@ skl_allocate_pipe_ddb(struct drm_crtc *crtc,
 	alloc_size = skl_ddb_entry_size(alloc);
 	if (alloc_size == 0) {
 		memset(ddb->plane[pipe], 0, sizeof(ddb->plane[pipe]));
-		memset(&ddb->cursor[pipe], 0, sizeof(ddb->cursor[pipe]));
+		memset(&ddb->plane[pipe][PLANE_CURSOR], 0,
+		       sizeof(ddb->plane[pipe][PLANE_CURSOR]));
 		return;
 	}
 
 	cursor_blocks = skl_cursor_allocation(config);
-	ddb->cursor[pipe].start = alloc->end - cursor_blocks;
-	ddb->cursor[pipe].end = alloc->end;
+	ddb->plane[pipe][PLANE_CURSOR].start = alloc->end - cursor_blocks;
+	ddb->plane[pipe][PLANE_CURSOR].end = alloc->end;
 
 	alloc_size -= cursor_blocks;
 	alloc->end -= cursor_blocks;
@@ -3098,8 +3099,8 @@ static bool skl_ddb_allocation_changed(const struct skl_ddb_allocation *new_ddb,
 		   sizeof(new_ddb->plane[pipe])))
 		return true;
 
-	if (memcmp(&new_ddb->cursor[pipe], &cur_ddb->cursor[pipe],
-		    sizeof(new_ddb->cursor[pipe])))
+	if (memcmp(&new_ddb->plane[pipe][PLANE_CURSOR], &cur_ddb->plane[pipe][PLANE_CURSOR],
+		    sizeof(new_ddb->plane[pipe][PLANE_CURSOR])))
 		return true;
 
 	return false;
@@ -3159,17 +3160,17 @@ static void skl_compute_wm_pipe_parameters(struct drm_crtc *crtc,
 		p->plane[0].rotation = crtc->primary->state->rotation;
 
 		fb = crtc->cursor->state->fb;
-		p->cursor.y_bytes_per_pixel = 0;
+		p->plane[PLANE_CURSOR].y_bytes_per_pixel = 0;
 		if (fb) {
-			p->cursor.enabled = true;
-			p->cursor.bytes_per_pixel = fb->bits_per_pixel / 8;
-			p->cursor.horiz_pixels = crtc->cursor->state->crtc_w;
-			p->cursor.vert_pixels = crtc->cursor->state->crtc_h;
+			p->plane[PLANE_CURSOR].enabled = true;
+			p->plane[PLANE_CURSOR].bytes_per_pixel = fb->bits_per_pixel / 8;
+			p->plane[PLANE_CURSOR].horiz_pixels = crtc->cursor->state->crtc_w;
+			p->plane[PLANE_CURSOR].vert_pixels = crtc->cursor->state->crtc_h;
 		} else {
-			p->cursor.enabled = false;
-			p->cursor.bytes_per_pixel = 0;
-			p->cursor.horiz_pixels = 64;
-			p->cursor.vert_pixels = 64;
+			p->plane[PLANE_CURSOR].enabled = false;
+			p->plane[PLANE_CURSOR].bytes_per_pixel = 0;
+			p->plane[PLANE_CURSOR].horiz_pixels = 64;
+			p->plane[PLANE_CURSOR].vert_pixels = 64;
 		}
 	}
 
@@ -3283,11 +3284,12 @@ static void skl_compute_wm_level(const struct drm_i915_private *dev_priv,
 						&result->plane_res_l[i]);
 	}
 
-	ddb_blocks = skl_ddb_entry_size(&ddb->cursor[pipe]);
-	result->cursor_en = skl_compute_plane_wm(dev_priv, p, &p->cursor,
+	ddb_blocks = skl_ddb_entry_size(&ddb->plane[pipe][PLANE_CURSOR]);
+	result->plane_en[PLANE_CURSOR] = skl_compute_plane_wm(dev_priv, p,
+						 &p->plane[PLANE_CURSOR],
 						 ddb_blocks, level,
-						 &result->cursor_res_b,
-						 &result->cursor_res_l);
+						 &result->plane_res_b[PLANE_CURSOR],
+						 &result->plane_res_l[PLANE_CURSOR]);
 }
 
 static uint32_t
@@ -3315,7 +3317,7 @@ static void skl_compute_transition_wm(struct drm_crtc *crtc,
 	/* Until we know more, just disable transition WMs */
 	for (i = 0; i < intel_num_planes(intel_crtc); i++)
 		trans_wm->plane_en[i] = false;
-	trans_wm->cursor_en = false;
+	trans_wm->plane_en[PLANE_CURSOR] = false;
 }
 
 static void skl_compute_pipe_wm(struct drm_crtc *crtc,
@@ -3364,13 +3366,13 @@ static void skl_compute_wm_results(struct drm_device *dev,
 
 		temp = 0;
 
-		temp |= p_wm->wm[level].cursor_res_l << PLANE_WM_LINES_SHIFT;
-		temp |= p_wm->wm[level].cursor_res_b;
+		temp |= p_wm->wm[level].plane_res_l[PLANE_CURSOR] << PLANE_WM_LINES_SHIFT;
+		temp |= p_wm->wm[level].plane_res_b[PLANE_CURSOR];
 
-		if (p_wm->wm[level].cursor_en)
+		if (p_wm->wm[level].plane_en[PLANE_CURSOR])
 			temp |= PLANE_WM_EN;
 
-		r->cursor[pipe][level] = temp;
+		r->plane[pipe][PLANE_CURSOR][level] = temp;
 
 	}
 
@@ -3386,12 +3388,12 @@ static void skl_compute_wm_results(struct drm_device *dev,
 	}
 
 	temp = 0;
-	temp |= p_wm->trans_wm.cursor_res_l << PLANE_WM_LINES_SHIFT;
-	temp |= p_wm->trans_wm.cursor_res_b;
-	if (p_wm->trans_wm.cursor_en)
+	temp |= p_wm->trans_wm.plane_res_l[PLANE_CURSOR] << PLANE_WM_LINES_SHIFT;
+	temp |= p_wm->trans_wm.plane_res_b[PLANE_CURSOR];
+	if (p_wm->trans_wm.plane_en[PLANE_CURSOR])
 		temp |= PLANE_WM_EN;
 
-	r->cursor_trans[pipe] = temp;
+	r->plane_trans[pipe][PLANE_CURSOR] = temp;
 
 	r->wm_linetime[pipe] = p_wm->linetime;
 }
@@ -3425,12 +3427,13 @@ static void skl_write_wm_values(struct drm_i915_private *dev_priv,
 				I915_WRITE(PLANE_WM(pipe, i, level),
 					   new->plane[pipe][i][level]);
 			I915_WRITE(CUR_WM(pipe, level),
-				   new->cursor[pipe][level]);
+				   new->plane[pipe][PLANE_CURSOR][level]);
 		}
 		for (i = 0; i < intel_num_planes(crtc); i++)
 			I915_WRITE(PLANE_WM_TRANS(pipe, i),
 				   new->plane_trans[pipe][i]);
-		I915_WRITE(CUR_WM_TRANS(pipe), new->cursor_trans[pipe]);
+		I915_WRITE(CUR_WM_TRANS(pipe),
+			   new->plane_trans[pipe][PLANE_CURSOR]);
 
 		for (i = 0; i < intel_num_planes(crtc); i++) {
 			skl_ddb_entry_write(dev_priv,
@@ -3442,7 +3445,7 @@ static void skl_write_wm_values(struct drm_i915_private *dev_priv,
 		}
 
 		skl_ddb_entry_write(dev_priv, CUR_BUF_CFG(pipe),
-				    &new->ddb.cursor[pipe]);
+				    &new->ddb.plane[pipe][PLANE_CURSOR]);
 	}
 }
 
@@ -3655,10 +3658,9 @@ static void skl_clear_wm(struct skl_wm_values *watermarks, enum pipe pipe)
 	watermarks->wm_linetime[pipe] = 0;
 	memset(watermarks->plane[pipe], 0,
 	       sizeof(uint32_t) * 8 * I915_MAX_PLANES);
-	memset(watermarks->cursor[pipe], 0, sizeof(uint32_t) * 8);
 	memset(watermarks->plane_trans[pipe],
 	       0, sizeof(uint32_t) * I915_MAX_PLANES);
-	watermarks->cursor_trans[pipe] = 0;
+	watermarks->plane_trans[pipe][PLANE_CURSOR] = 0;
 
 	/* Clear ddb entries for pipe */
 	memset(&watermarks->ddb.pipe[pipe], 0, sizeof(struct skl_ddb_entry));
@@ -3666,7 +3668,8 @@ static void skl_clear_wm(struct skl_wm_values *watermarks, enum pipe pipe)
 	       sizeof(struct skl_ddb_entry) * I915_MAX_PLANES);
 	memset(&watermarks->ddb.y_plane[pipe], 0,
 	       sizeof(struct skl_ddb_entry) * I915_MAX_PLANES);
-	memset(&watermarks->ddb.cursor[pipe], 0, sizeof(struct skl_ddb_entry));
+	memset(&watermarks->ddb.plane[pipe][PLANE_CURSOR], 0,
+	       sizeof(struct skl_ddb_entry));
 
 }
 
@@ -3822,10 +3825,10 @@ static void skl_pipe_wm_active_state(uint32_t val,
 					(val >> PLANE_WM_LINES_SHIFT) &
 						PLANE_WM_LINES_MASK;
 		} else {
-			active->wm[level].cursor_en = is_enabled;
-			active->wm[level].cursor_res_b =
+			active->wm[level].plane_en[PLANE_CURSOR] = is_enabled;
+			active->wm[level].plane_res_b[PLANE_CURSOR] =
 					val & PLANE_WM_BLOCKS_MASK;
-			active->wm[level].cursor_res_l =
+			active->wm[level].plane_res_l[PLANE_CURSOR] =
 					(val >> PLANE_WM_LINES_SHIFT) &
 						PLANE_WM_LINES_MASK;
 		}
@@ -3838,10 +3841,10 @@ static void skl_pipe_wm_active_state(uint32_t val,
 					(val >> PLANE_WM_LINES_SHIFT) &
 						PLANE_WM_LINES_MASK;
 		} else {
-			active->trans_wm.cursor_en = is_enabled;
-			active->trans_wm.cursor_res_b =
+			active->trans_wm.plane_en[PLANE_CURSOR] = is_enabled;
+			active->trans_wm.plane_res_b[PLANE_CURSOR] =
 					val & PLANE_WM_BLOCKS_MASK;
-			active->trans_wm.cursor_res_l =
+			active->trans_wm.plane_res_l[PLANE_CURSOR] =
 					(val >> PLANE_WM_LINES_SHIFT) &
 						PLANE_WM_LINES_MASK;
 		}
@@ -3867,12 +3870,12 @@ static void skl_pipe_wm_get_hw_state(struct drm_crtc *crtc)
 		for (i = 0; i < intel_num_planes(intel_crtc); i++)
 			hw->plane[pipe][i][level] =
 					I915_READ(PLANE_WM(pipe, i, level));
-		hw->cursor[pipe][level] = I915_READ(CUR_WM(pipe, level));
+		hw->plane[pipe][PLANE_CURSOR][level] = I915_READ(CUR_WM(pipe, level));
 	}
 
 	for (i = 0; i < intel_num_planes(intel_crtc); i++)
 		hw->plane_trans[pipe][i] = I915_READ(PLANE_WM_TRANS(pipe, i));
-	hw->cursor_trans[pipe] = I915_READ(CUR_WM_TRANS(pipe));
+	hw->plane_trans[pipe][PLANE_CURSOR] = I915_READ(CUR_WM_TRANS(pipe));
 
 	if (!intel_crtc->active)
 		return;
@@ -3887,7 +3890,7 @@ static void skl_pipe_wm_get_hw_state(struct drm_crtc *crtc)
 			skl_pipe_wm_active_state(temp, active, false,
 						false, i, level);
 		}
-		temp = hw->cursor[pipe][level];
+		temp = hw->plane[pipe][PLANE_CURSOR][level];
 		skl_pipe_wm_active_state(temp, active, false, true, i, level);
 	}
 
@@ -3896,7 +3899,7 @@ static void skl_pipe_wm_get_hw_state(struct drm_crtc *crtc)
 		skl_pipe_wm_active_state(temp, active, true, false, i, 0);
 	}
 
-	temp = hw->cursor_trans[pipe];
+	temp = hw->plane_trans[pipe][PLANE_CURSOR];
 	skl_pipe_wm_active_state(temp, active, true, true, i, 0);
 }
 
-- 
cgit v0.10.2


From 3a05f5e2e78eab7ffe816abb59b6769e331a1957 Mon Sep 17 00:00:00 2001
From: Matt Roper <matthew.d.roper@intel.com>
Date: Thu, 24 Sep 2015 15:53:11 -0700
Subject: drm/i915/skl: Eliminate usage of pipe_wm_parameters from SKL-style WM
 (v3)

Just pull the info out of the state structures rather than staging
it in an additional set of structures.  To make this more
straightforward, we change the signature of several internal WM
functions to take the crtc state as a parameter.

v2:
 - Don't forget to skip cursor planes on a loop in the DDB allocation
   function to match original behavior.  (Ander)
 - Change a use of intel_crtc->active to cstate->active.  They should
   be identical, but it's better to be consistent.  (Ander)
 - Rework more function signatures to pass states rather than crtc for
   consistency. (Ander)

v3:
  - Add missing "+ 1" to skl_wm_plane_id()'s 'overlay' case. (Maarten)
  - Packed formats should pass '0' to drm_format_plane_cpp(), not 1.
    (Maarten)
  - Drop unwanted WARN_ON() for disabled planes when calculating data
    rate for SKL.  (Maarten)

Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index 0004362..3aca66d 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -1765,13 +1765,6 @@ static uint32_t ilk_wm_fbc(uint32_t pri_val, uint32_t horiz_pixels,
 	return DIV_ROUND_UP(pri_val * 64, horiz_pixels * bytes_per_pixel) + 2;
 }
 
-struct skl_pipe_wm_parameters {
-	bool active;
-	uint32_t pipe_htotal;
-	uint32_t pixel_rate; /* in KHz */
-	struct intel_plane_wm_parameters plane[I915_MAX_PLANES];
-};
-
 struct ilk_wm_maximums {
 	uint16_t pri;
 	uint16_t spr;
@@ -2812,18 +2805,40 @@ static bool ilk_disable_lp_wm(struct drm_device *dev)
 #define SKL_DDB_SIZE		896	/* in blocks */
 #define BXT_DDB_SIZE		512
 
+/*
+ * Return the index of a plane in the SKL DDB and wm result arrays.  Primary
+ * plane is always in slot 0, cursor is always in slot I915_MAX_PLANES-1, and
+ * other universal planes are in indices 1..n.  Note that this may leave unused
+ * indices between the top "sprite" plane and the cursor.
+ */
+static int
+skl_wm_plane_id(const struct intel_plane *plane)
+{
+	switch (plane->base.type) {
+	case DRM_PLANE_TYPE_PRIMARY:
+		return 0;
+	case DRM_PLANE_TYPE_CURSOR:
+		return PLANE_CURSOR;
+	case DRM_PLANE_TYPE_OVERLAY:
+		return plane->plane + 1;
+	default:
+		MISSING_CASE(plane->base.type);
+		return plane->plane;
+	}
+}
+
 static void
 skl_ddb_get_pipe_allocation_limits(struct drm_device *dev,
-				   struct drm_crtc *for_crtc,
+				   const struct intel_crtc_state *cstate,
 				   const struct intel_wm_config *config,
-				   const struct skl_pipe_wm_parameters *params,
 				   struct skl_ddb_entry *alloc /* out */)
 {
+	struct drm_crtc *for_crtc = cstate->base.crtc;
 	struct drm_crtc *crtc;
 	unsigned int pipe_size, ddb_size;
 	int nth_active_pipe;
 
-	if (!params->active) {
+	if (!cstate->base.active) {
 		alloc->start = 0;
 		alloc->end = 0;
 		return;
@@ -2889,19 +2904,29 @@ void skl_ddb_get_hw_state(struct drm_i915_private *dev_priv,
 }
 
 static unsigned int
-skl_plane_relative_data_rate(const struct intel_plane_wm_parameters *p, int y)
+skl_plane_relative_data_rate(const struct intel_crtc_state *cstate,
+			     const struct drm_plane_state *pstate,
+			     int y)
 {
+	struct intel_crtc *intel_crtc = to_intel_crtc(cstate->base.crtc);
+	struct drm_framebuffer *fb = pstate->fb;
 
 	/* for planar format */
-	if (p->y_bytes_per_pixel) {
+	if (fb->pixel_format == DRM_FORMAT_NV12) {
 		if (y)  /* y-plane data rate */
-			return p->horiz_pixels * p->vert_pixels * p->y_bytes_per_pixel;
+			return intel_crtc->config->pipe_src_w *
+				intel_crtc->config->pipe_src_h *
+				drm_format_plane_cpp(fb->pixel_format, 0);
 		else    /* uv-plane data rate */
-			return (p->horiz_pixels/2) * (p->vert_pixels/2) * p->bytes_per_pixel;
+			return (intel_crtc->config->pipe_src_w/2) *
+				(intel_crtc->config->pipe_src_h/2) *
+				drm_format_plane_cpp(fb->pixel_format, 1);
 	}
 
 	/* for packed formats */
-	return p->horiz_pixels * p->vert_pixels * p->bytes_per_pixel;
+	return intel_crtc->config->pipe_src_w *
+		intel_crtc->config->pipe_src_h *
+		drm_format_plane_cpp(fb->pixel_format, 0);
 }
 
 /*
@@ -2910,46 +2935,51 @@ skl_plane_relative_data_rate(const struct intel_plane_wm_parameters *p, int y)
  *   3 * 4096 * 8192  * 4 < 2^32
  */
 static unsigned int
-skl_get_total_relative_data_rate(struct intel_crtc *intel_crtc,
-				 const struct skl_pipe_wm_parameters *params)
+skl_get_total_relative_data_rate(const struct intel_crtc_state *cstate)
 {
+	struct intel_crtc *intel_crtc = to_intel_crtc(cstate->base.crtc);
+	struct drm_device *dev = intel_crtc->base.dev;
+	const struct intel_plane *intel_plane;
 	unsigned int total_data_rate = 0;
-	int plane;
 
-	for (plane = 0; plane < intel_num_planes(intel_crtc); plane++) {
-		const struct intel_plane_wm_parameters *p;
+	for_each_intel_plane_on_crtc(dev, intel_crtc, intel_plane) {
+		const struct drm_plane_state *pstate = intel_plane->base.state;
 
-		p = &params->plane[plane];
-		if (!p->enabled)
+		if (pstate->fb == NULL)
 			continue;
 
-		total_data_rate += skl_plane_relative_data_rate(p, 0); /* packed/uv */
-		if (p->y_bytes_per_pixel) {
-			total_data_rate += skl_plane_relative_data_rate(p, 1); /* y-plane */
-		}
+		/* packed/uv */
+		total_data_rate += skl_plane_relative_data_rate(cstate,
+								pstate,
+								0);
+
+		if (pstate->fb->pixel_format == DRM_FORMAT_NV12)
+			/* y-plane */
+			total_data_rate += skl_plane_relative_data_rate(cstate,
+									pstate,
+									1);
 	}
 
 	return total_data_rate;
 }
 
 static void
-skl_allocate_pipe_ddb(struct drm_crtc *crtc,
+skl_allocate_pipe_ddb(struct intel_crtc_state *cstate,
 		      const struct intel_wm_config *config,
-		      const struct skl_pipe_wm_parameters *params,
 		      struct skl_ddb_allocation *ddb /* out */)
 {
+	struct drm_crtc *crtc = cstate->base.crtc;
 	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	struct intel_plane *intel_plane;
 	enum pipe pipe = intel_crtc->pipe;
 	struct skl_ddb_entry *alloc = &ddb->pipe[pipe];
 	uint16_t alloc_size, start, cursor_blocks;
 	uint16_t minimum[I915_MAX_PLANES];
 	uint16_t y_minimum[I915_MAX_PLANES];
 	unsigned int total_data_rate;
-	int plane;
 
-	skl_ddb_get_pipe_allocation_limits(dev, crtc, config, params, alloc);
+	skl_ddb_get_pipe_allocation_limits(dev, cstate, config, alloc);
 	alloc_size = skl_ddb_entry_size(alloc);
 	if (alloc_size == 0) {
 		memset(ddb->plane[pipe], 0, sizeof(ddb->plane[pipe]));
@@ -2966,17 +2996,20 @@ skl_allocate_pipe_ddb(struct drm_crtc *crtc,
 	alloc->end -= cursor_blocks;
 
 	/* 1. Allocate the mininum required blocks for each active plane */
-	for_each_plane(dev_priv, pipe, plane) {
-		const struct intel_plane_wm_parameters *p;
+	for_each_intel_plane_on_crtc(dev, intel_crtc, intel_plane) {
+		struct drm_plane *plane = &intel_plane->base;
+		struct drm_framebuffer *fb = plane->fb;
+		int id = skl_wm_plane_id(intel_plane);
 
-		p = &params->plane[plane];
-		if (!p->enabled)
+		if (fb == NULL)
+			continue;
+		if (plane->type == DRM_PLANE_TYPE_CURSOR)
 			continue;
 
-		minimum[plane] = 8;
-		alloc_size -= minimum[plane];
-		y_minimum[plane] = p->y_bytes_per_pixel ? 8 : 0;
-		alloc_size -= y_minimum[plane];
+		minimum[id] = 8;
+		alloc_size -= minimum[id];
+		y_minimum[id] = (fb->pixel_format == DRM_FORMAT_NV12) ? 8 : 0;
+		alloc_size -= y_minimum[id];
 	}
 
 	/*
@@ -2985,45 +3018,50 @@ skl_allocate_pipe_ddb(struct drm_crtc *crtc,
 	 *
 	 * FIXME: we may not allocate every single block here.
 	 */
-	total_data_rate = skl_get_total_relative_data_rate(intel_crtc, params);
+	total_data_rate = skl_get_total_relative_data_rate(cstate);
 
 	start = alloc->start;
-	for (plane = 0; plane < intel_num_planes(intel_crtc); plane++) {
-		const struct intel_plane_wm_parameters *p;
+	for_each_intel_plane_on_crtc(dev, intel_crtc, intel_plane) {
+		struct drm_plane *plane = &intel_plane->base;
+		struct drm_plane_state *pstate = intel_plane->base.state;
 		unsigned int data_rate, y_data_rate;
 		uint16_t plane_blocks, y_plane_blocks = 0;
+		int id = skl_wm_plane_id(intel_plane);
 
-		p = &params->plane[plane];
-		if (!p->enabled)
+		if (pstate->fb == NULL)
+			continue;
+		if (plane->type == DRM_PLANE_TYPE_CURSOR)
 			continue;
 
-		data_rate = skl_plane_relative_data_rate(p, 0);
+		data_rate = skl_plane_relative_data_rate(cstate, pstate, 0);
 
 		/*
 		 * allocation for (packed formats) or (uv-plane part of planar format):
 		 * promote the expression to 64 bits to avoid overflowing, the
 		 * result is < available as data_rate / total_data_rate < 1
 		 */
-		plane_blocks = minimum[plane];
+		plane_blocks = minimum[id];
 		plane_blocks += div_u64((uint64_t)alloc_size * data_rate,
 					total_data_rate);
 
-		ddb->plane[pipe][plane].start = start;
-		ddb->plane[pipe][plane].end = start + plane_blocks;
+		ddb->plane[pipe][id].start = start;
+		ddb->plane[pipe][id].end = start + plane_blocks;
 
 		start += plane_blocks;
 
 		/*
 		 * allocation for y_plane part of planar format:
 		 */
-		if (p->y_bytes_per_pixel) {
-			y_data_rate = skl_plane_relative_data_rate(p, 1);
-			y_plane_blocks = y_minimum[plane];
+		if (pstate->fb->pixel_format == DRM_FORMAT_NV12) {
+			y_data_rate = skl_plane_relative_data_rate(cstate,
+								   pstate,
+								   1);
+			y_plane_blocks = y_minimum[id];
 			y_plane_blocks += div_u64((uint64_t)alloc_size * y_data_rate,
 						total_data_rate);
 
-			ddb->y_plane[pipe][plane].start = start;
-			ddb->y_plane[pipe][plane].end = start + y_plane_blocks;
+			ddb->y_plane[pipe][id].start = start;
+			ddb->y_plane[pipe][id].end = start + y_plane_blocks;
 
 			start += y_plane_blocks;
 		}
@@ -3124,73 +3162,16 @@ static void skl_compute_wm_global_parameters(struct drm_device *dev,
 	}
 }
 
-static void skl_compute_wm_pipe_parameters(struct drm_crtc *crtc,
-					   struct skl_pipe_wm_parameters *p)
-{
-	struct drm_device *dev = crtc->dev;
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	enum pipe pipe = intel_crtc->pipe;
-	struct drm_plane *plane;
-	struct drm_framebuffer *fb;
-	int i = 1; /* Index for sprite planes start */
-
-	p->active = intel_crtc->active;
-	if (p->active) {
-		p->pipe_htotal = intel_crtc->config->base.adjusted_mode.crtc_htotal;
-		p->pixel_rate = skl_pipe_pixel_rate(intel_crtc->config);
-
-		fb = crtc->primary->state->fb;
-		/* For planar: Bpp is for uv plane, y_Bpp is for y plane */
-		if (fb) {
-			p->plane[0].enabled = true;
-			p->plane[0].bytes_per_pixel = fb->pixel_format == DRM_FORMAT_NV12 ?
-				drm_format_plane_cpp(fb->pixel_format, 1) :
-				drm_format_plane_cpp(fb->pixel_format, 0);
-			p->plane[0].y_bytes_per_pixel = fb->pixel_format == DRM_FORMAT_NV12 ?
-				drm_format_plane_cpp(fb->pixel_format, 0) : 0;
-			p->plane[0].tiling = fb->modifier[0];
-		} else {
-			p->plane[0].enabled = false;
-			p->plane[0].bytes_per_pixel = 0;
-			p->plane[0].y_bytes_per_pixel = 0;
-			p->plane[0].tiling = DRM_FORMAT_MOD_NONE;
-		}
-		p->plane[0].horiz_pixels = intel_crtc->config->pipe_src_w;
-		p->plane[0].vert_pixels = intel_crtc->config->pipe_src_h;
-		p->plane[0].rotation = crtc->primary->state->rotation;
-
-		fb = crtc->cursor->state->fb;
-		p->plane[PLANE_CURSOR].y_bytes_per_pixel = 0;
-		if (fb) {
-			p->plane[PLANE_CURSOR].enabled = true;
-			p->plane[PLANE_CURSOR].bytes_per_pixel = fb->bits_per_pixel / 8;
-			p->plane[PLANE_CURSOR].horiz_pixels = crtc->cursor->state->crtc_w;
-			p->plane[PLANE_CURSOR].vert_pixels = crtc->cursor->state->crtc_h;
-		} else {
-			p->plane[PLANE_CURSOR].enabled = false;
-			p->plane[PLANE_CURSOR].bytes_per_pixel = 0;
-			p->plane[PLANE_CURSOR].horiz_pixels = 64;
-			p->plane[PLANE_CURSOR].vert_pixels = 64;
-		}
-	}
-
-	list_for_each_entry(plane, &dev->mode_config.plane_list, head) {
-		struct intel_plane *intel_plane = to_intel_plane(plane);
-
-		if (intel_plane->pipe == pipe &&
-			plane->type == DRM_PLANE_TYPE_OVERLAY)
-			p->plane[i++] = intel_plane->wm;
-	}
-}
-
 static bool skl_compute_plane_wm(const struct drm_i915_private *dev_priv,
-				 struct skl_pipe_wm_parameters *p,
-				 struct intel_plane_wm_parameters *p_params,
+				 struct intel_crtc_state *cstate,
+				 struct intel_plane *intel_plane,
 				 uint16_t ddb_allocation,
 				 int level,
 				 uint16_t *out_blocks, /* out */
 				 uint8_t *out_lines /* out */)
 {
+	struct drm_plane *plane = &intel_plane->base;
+	struct drm_framebuffer *fb = plane->state->fb;
 	uint32_t latency = dev_priv->wm.skl_latency[level];
 	uint32_t method1, method2;
 	uint32_t plane_bytes_per_line, plane_blocks_per_line;
@@ -3198,31 +3179,35 @@ static bool skl_compute_plane_wm(const struct drm_i915_private *dev_priv,
 	uint32_t selected_result;
 	uint8_t bytes_per_pixel;
 
-	if (latency == 0 || !p->active || !p_params->enabled)
+	if (latency == 0 || !cstate->base.active || !fb)
 		return false;
 
-	bytes_per_pixel = p_params->y_bytes_per_pixel ?
-		p_params->y_bytes_per_pixel :
-		p_params->bytes_per_pixel;
-	method1 = skl_wm_method1(p->pixel_rate,
+	bytes_per_pixel = (fb->pixel_format == DRM_FORMAT_NV12) ?
+		drm_format_plane_cpp(DRM_FORMAT_NV12, 0) :
+		drm_format_plane_cpp(DRM_FORMAT_NV12, 1);
+	method1 = skl_wm_method1(skl_pipe_pixel_rate(cstate),
 				 bytes_per_pixel,
 				 latency);
-	method2 = skl_wm_method2(p->pixel_rate,
-				 p->pipe_htotal,
-				 p_params->horiz_pixels,
+	method2 = skl_wm_method2(skl_pipe_pixel_rate(cstate),
+				 cstate->base.adjusted_mode.crtc_htotal,
+				 cstate->pipe_src_w,
 				 bytes_per_pixel,
-				 p_params->tiling,
+				 fb->modifier[0],
 				 latency);
 
-	plane_bytes_per_line = p_params->horiz_pixels * bytes_per_pixel;
+	plane_bytes_per_line = cstate->pipe_src_w * bytes_per_pixel;
 	plane_blocks_per_line = DIV_ROUND_UP(plane_bytes_per_line, 512);
 
-	if (p_params->tiling == I915_FORMAT_MOD_Y_TILED ||
-	    p_params->tiling == I915_FORMAT_MOD_Yf_TILED) {
+	if (fb->modifier[0] == I915_FORMAT_MOD_Y_TILED ||
+	    fb->modifier[0] == I915_FORMAT_MOD_Yf_TILED) {
 		uint32_t min_scanlines = 4;
 		uint32_t y_tile_minimum;
-		if (intel_rotation_90_or_270(p_params->rotation)) {
-			switch (p_params->bytes_per_pixel) {
+		if (intel_rotation_90_or_270(plane->state->rotation)) {
+			int bpp = (fb->pixel_format == DRM_FORMAT_NV12) ?
+				drm_format_plane_cpp(fb->pixel_format, 1) :
+				drm_format_plane_cpp(fb->pixel_format, 0);
+
+			switch (bpp) {
 			case 1:
 				min_scanlines = 16;
 				break;
@@ -3246,8 +3231,8 @@ static bool skl_compute_plane_wm(const struct drm_i915_private *dev_priv,
 	res_lines = DIV_ROUND_UP(selected_result, plane_blocks_per_line);
 
 	if (level >= 1 && level <= 7) {
-		if (p_params->tiling == I915_FORMAT_MOD_Y_TILED ||
-		    p_params->tiling == I915_FORMAT_MOD_Yf_TILED)
+		if (fb->modifier[0] == I915_FORMAT_MOD_Y_TILED ||
+		    fb->modifier[0] == I915_FORMAT_MOD_Yf_TILED)
 			res_lines += 4;
 		else
 			res_blocks++;
@@ -3264,84 +3249,80 @@ static bool skl_compute_plane_wm(const struct drm_i915_private *dev_priv,
 
 static void skl_compute_wm_level(const struct drm_i915_private *dev_priv,
 				 struct skl_ddb_allocation *ddb,
-				 struct skl_pipe_wm_parameters *p,
-				 enum pipe pipe,
+				 struct intel_crtc_state *cstate,
 				 int level,
-				 int num_planes,
 				 struct skl_wm_level *result)
 {
+	struct drm_device *dev = dev_priv->dev;
+	struct intel_crtc *intel_crtc = to_intel_crtc(cstate->base.crtc);
+	struct intel_plane *intel_plane;
 	uint16_t ddb_blocks;
-	int i;
+	enum pipe pipe = intel_crtc->pipe;
+
+	for_each_intel_plane_on_crtc(dev, intel_crtc, intel_plane) {
+		int i = skl_wm_plane_id(intel_plane);
 
-	for (i = 0; i < num_planes; i++) {
 		ddb_blocks = skl_ddb_entry_size(&ddb->plane[pipe][i]);
 
 		result->plane_en[i] = skl_compute_plane_wm(dev_priv,
-						p, &p->plane[i],
+						cstate,
+						intel_plane,
 						ddb_blocks,
 						level,
 						&result->plane_res_b[i],
 						&result->plane_res_l[i]);
 	}
-
-	ddb_blocks = skl_ddb_entry_size(&ddb->plane[pipe][PLANE_CURSOR]);
-	result->plane_en[PLANE_CURSOR] = skl_compute_plane_wm(dev_priv, p,
-						 &p->plane[PLANE_CURSOR],
-						 ddb_blocks, level,
-						 &result->plane_res_b[PLANE_CURSOR],
-						 &result->plane_res_l[PLANE_CURSOR]);
 }
 
 static uint32_t
-skl_compute_linetime_wm(struct drm_crtc *crtc, struct skl_pipe_wm_parameters *p)
+skl_compute_linetime_wm(struct intel_crtc_state *cstate)
 {
-	if (!to_intel_crtc(crtc)->active)
+	if (!cstate->base.active)
 		return 0;
 
-	if (WARN_ON(p->pixel_rate == 0))
+	if (WARN_ON(skl_pipe_pixel_rate(cstate) == 0))
 		return 0;
 
-	return DIV_ROUND_UP(8 * p->pipe_htotal * 1000, p->pixel_rate);
+	return DIV_ROUND_UP(8 * cstate->base.adjusted_mode.crtc_htotal * 1000,
+			    skl_pipe_pixel_rate(cstate));
 }
 
-static void skl_compute_transition_wm(struct drm_crtc *crtc,
-				      struct skl_pipe_wm_parameters *params,
+static void skl_compute_transition_wm(struct intel_crtc_state *cstate,
 				      struct skl_wm_level *trans_wm /* out */)
 {
+	struct drm_crtc *crtc = cstate->base.crtc;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	int i;
+	struct intel_plane *intel_plane;
 
-	if (!params->active)
+	if (!cstate->base.active)
 		return;
 
 	/* Until we know more, just disable transition WMs */
-	for (i = 0; i < intel_num_planes(intel_crtc); i++)
+	for_each_intel_plane_on_crtc(crtc->dev, intel_crtc, intel_plane) {
+		int i = skl_wm_plane_id(intel_plane);
+
 		trans_wm->plane_en[i] = false;
-	trans_wm->plane_en[PLANE_CURSOR] = false;
+	}
 }
 
-static void skl_compute_pipe_wm(struct drm_crtc *crtc,
+static void skl_compute_pipe_wm(struct intel_crtc_state *cstate,
 				struct skl_ddb_allocation *ddb,
-				struct skl_pipe_wm_parameters *params,
 				struct skl_pipe_wm *pipe_wm)
 {
-	struct drm_device *dev = crtc->dev;
+	struct drm_device *dev = cstate->base.crtc->dev;
 	const struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	int level, max_level = ilk_wm_max_level(dev);
 
 	for (level = 0; level <= max_level; level++) {
-		skl_compute_wm_level(dev_priv, ddb, params, intel_crtc->pipe,
-				     level, intel_num_planes(intel_crtc),
-				     &pipe_wm->wm[level]);
+		skl_compute_wm_level(dev_priv, ddb, cstate,
+				     level, &pipe_wm->wm[level]);
 	}
-	pipe_wm->linetime = skl_compute_linetime_wm(crtc, params);
+	pipe_wm->linetime = skl_compute_linetime_wm(cstate);
 
-	skl_compute_transition_wm(crtc, params, &pipe_wm->trans_wm);
+	skl_compute_transition_wm(cstate, &pipe_wm->trans_wm);
 }
 
 static void skl_compute_wm_results(struct drm_device *dev,
-				   struct skl_pipe_wm_parameters *p,
 				   struct skl_pipe_wm *p_wm,
 				   struct skl_wm_values *r,
 				   struct intel_crtc *intel_crtc)
@@ -3585,16 +3566,15 @@ static void skl_flush_wm_values(struct drm_i915_private *dev_priv,
 }
 
 static bool skl_update_pipe_wm(struct drm_crtc *crtc,
-			       struct skl_pipe_wm_parameters *params,
 			       struct intel_wm_config *config,
 			       struct skl_ddb_allocation *ddb, /* out */
 			       struct skl_pipe_wm *pipe_wm /* out */)
 {
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	struct intel_crtc_state *cstate = to_intel_crtc_state(crtc->state);
 
-	skl_compute_wm_pipe_parameters(crtc, params);
-	skl_allocate_pipe_ddb(crtc, config, params, ddb);
-	skl_compute_pipe_wm(crtc, ddb, params, pipe_wm);
+	skl_allocate_pipe_ddb(cstate, config, ddb);
+	skl_compute_pipe_wm(cstate, ddb, pipe_wm);
 
 	if (!memcmp(&intel_crtc->wm.skl_active, pipe_wm, sizeof(*pipe_wm)))
 		return false;
@@ -3627,7 +3607,6 @@ static void skl_update_other_pipe_wm(struct drm_device *dev,
 	 */
 	list_for_each_entry(intel_crtc, &dev->mode_config.crtc_list,
 				base.head) {
-		struct skl_pipe_wm_parameters params = {};
 		struct skl_pipe_wm pipe_wm = {};
 		bool wm_changed;
 
@@ -3637,8 +3616,7 @@ static void skl_update_other_pipe_wm(struct drm_device *dev,
 		if (!intel_crtc->active)
 			continue;
 
-		wm_changed = skl_update_pipe_wm(&intel_crtc->base,
-						&params, config,
+		wm_changed = skl_update_pipe_wm(&intel_crtc->base, config,
 						&r->ddb, &pipe_wm);
 
 		/*
@@ -3648,7 +3626,7 @@ static void skl_update_other_pipe_wm(struct drm_device *dev,
 		 */
 		WARN_ON(!wm_changed);
 
-		skl_compute_wm_results(dev, &params, &pipe_wm, r, intel_crtc);
+		skl_compute_wm_results(dev, &pipe_wm, r, intel_crtc);
 		r->dirty[intel_crtc->pipe] = true;
 	}
 }
@@ -3678,7 +3656,6 @@ static void skl_update_wm(struct drm_crtc *crtc)
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct skl_pipe_wm_parameters params = {};
 	struct skl_wm_values *results = &dev_priv->wm.skl_results;
 	struct skl_pipe_wm pipe_wm = {};
 	struct intel_wm_config config = {};
@@ -3691,11 +3668,10 @@ static void skl_update_wm(struct drm_crtc *crtc)
 
 	skl_compute_wm_global_parameters(dev, &config);
 
-	if (!skl_update_pipe_wm(crtc, &params, &config,
-				&results->ddb, &pipe_wm))
+	if (!skl_update_pipe_wm(crtc, &config, &results->ddb, &pipe_wm))
 		return;
 
-	skl_compute_wm_results(dev, &params, &pipe_wm, results, intel_crtc);
+	skl_compute_wm_results(dev, &pipe_wm, results, intel_crtc);
 	results->dirty[intel_crtc->pipe] = true;
 
 	skl_update_other_pipe_wm(dev, crtc, &config, results);
-- 
cgit v0.10.2


From 7809e5ae35b9d8d0710f0874b2e3f10be144e38b Mon Sep 17 00:00:00 2001
From: Matt Roper <matthew.d.roper@intel.com>
Date: Thu, 24 Sep 2015 15:53:12 -0700
Subject: drm/i915/ivb: Move WaCxSRDisabledForSpriteScaling w/a to atomic check

Determine whether we need to apply this workaround at atomic check time
and just set a flag that will be used by the main watermark update
routine.

Moving this workaround into the atomic framework reduces
ilk_update_sprite_wm() to just a standard watermark update, so drop it
completely and just ensure that ilk_update_wm() is called whenever a
sprite plane is updated in a way that would affect watermarks.

Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_atomic.c b/drivers/gpu/drm/i915/intel_atomic.c
index f1975f2..05b1203 100644
--- a/drivers/gpu/drm/i915/intel_atomic.c
+++ b/drivers/gpu/drm/i915/intel_atomic.c
@@ -94,6 +94,7 @@ intel_crtc_duplicate_state(struct drm_crtc *crtc)
 	__drm_atomic_helper_crtc_duplicate_state(crtc, &crtc_state->base);
 
 	crtc_state->update_pipe = false;
+	crtc_state->disable_lp_wm = false;
 
 	return &crtc_state->base;
 }
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 305abaa..d3dab99 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -11574,18 +11574,32 @@ retry:
 static bool intel_wm_need_update(struct drm_plane *plane,
 				 struct drm_plane_state *state)
 {
-	/* Update watermarks on tiling changes. */
+	struct intel_plane_state *new = to_intel_plane_state(state);
+	struct intel_plane_state *cur = to_intel_plane_state(plane->state);
+
+	/* Update watermarks on tiling or size changes. */
 	if (!plane->state->fb || !state->fb ||
 	    plane->state->fb->modifier[0] != state->fb->modifier[0] ||
-	    plane->state->rotation != state->rotation)
-		return true;
-
-	if (plane->state->crtc_w != state->crtc_w)
+	    plane->state->rotation != state->rotation ||
+	    drm_rect_width(&new->src) != drm_rect_width(&cur->src) ||
+	    drm_rect_height(&new->src) != drm_rect_height(&cur->src) ||
+	    drm_rect_width(&new->dst) != drm_rect_width(&cur->dst) ||
+	    drm_rect_height(&new->dst) != drm_rect_height(&cur->dst))
 		return true;
 
 	return false;
 }
 
+static bool needs_scaling(struct intel_plane_state *state)
+{
+	int src_w = drm_rect_width(&state->src) >> 16;
+	int src_h = drm_rect_height(&state->src) >> 16;
+	int dst_w = drm_rect_width(&state->dst);
+	int dst_h = drm_rect_height(&state->dst);
+
+	return (src_w != dst_w || src_h != dst_h);
+}
+
 int intel_plane_atomic_calc_changes(struct drm_crtc_state *crtc_state,
 				    struct drm_plane_state *plane_state)
 {
@@ -11601,7 +11615,6 @@ int intel_plane_atomic_calc_changes(struct drm_crtc_state *crtc_state,
 	bool mode_changed = needs_modeset(crtc_state);
 	bool was_crtc_enabled = crtc->state->active;
 	bool is_crtc_enabled = crtc_state->active;
-
 	bool turn_off, turn_on, visible, was_visible;
 	struct drm_framebuffer *fb = plane_state->fb;
 
@@ -11719,11 +11732,23 @@ int intel_plane_atomic_calc_changes(struct drm_crtc_state *crtc_state,
 	case DRM_PLANE_TYPE_CURSOR:
 		break;
 	case DRM_PLANE_TYPE_OVERLAY:
-		if (turn_off && !mode_changed) {
+		/*
+		 * WaCxSRDisabledForSpriteScaling:ivb
+		 *
+		 * cstate->update_wm was already set above, so this flag will
+		 * take effect when we commit and program watermarks.
+		 */
+		if (IS_IVYBRIDGE(dev) &&
+		    needs_scaling(to_intel_plane_state(plane_state)) &&
+		    !needs_scaling(old_plane_state)) {
+			to_intel_crtc_state(crtc_state)->disable_lp_wm = true;
+		} else if (turn_off && !mode_changed) {
 			intel_crtc->atomic.wait_vblank = true;
 			intel_crtc->atomic.update_sprite_watermarks |=
 				1 << i;
 		}
+
+		break;
 	}
 	return 0;
 }
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 1eaa9f9..d6d7ac7 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -466,6 +466,9 @@ struct intel_crtc_state {
 
 	/* w/a for waiting 2 vblanks during crtc enable */
 	enum pipe hsw_workaround_pipe;
+
+	/* IVB sprite scaling w/a (WaCxSRDisabledForSpriteScaling:ivb) */
+	bool disable_lp_wm;
 };
 
 struct vlv_wm_state {
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index 3aca66d..b8c3e22 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -3730,6 +3730,18 @@ static void ilk_update_wm(struct drm_crtc *crtc)
 
 	WARN_ON(cstate->base.active != intel_crtc->active);
 
+	/*
+	 * IVB workaround: must disable low power watermarks for at least
+	 * one frame before enabling scaling.  LP watermarks can be re-enabled
+	 * when scaling is disabled.
+	 *
+	 * WaCxSRDisabledForSpriteScaling:ivb
+	 */
+	if (cstate->disable_lp_wm) {
+		ilk_disable_lp_wm(dev);
+		intel_wait_for_vblank(dev, intel_crtc->pipe);
+	}
+
 	intel_compute_pipe_wm(cstate, &pipe_wm);
 
 	if (!memcmp(&intel_crtc->wm.active, &pipe_wm, sizeof(pipe_wm)))
@@ -3761,28 +3773,6 @@ static void ilk_update_wm(struct drm_crtc *crtc)
 	ilk_write_wm_values(dev_priv, &results);
 }
 
-static void
-ilk_update_sprite_wm(struct drm_plane *plane,
-		     struct drm_crtc *crtc,
-		     uint32_t sprite_width, uint32_t sprite_height,
-		     int pixel_size, bool enabled, bool scaled)
-{
-	struct drm_device *dev = plane->dev;
-	struct intel_plane *intel_plane = to_intel_plane(plane);
-
-	/*
-	 * IVB workaround: must disable low power watermarks for at least
-	 * one frame before enabling scaling.  LP watermarks can be re-enabled
-	 * when scaling is disabled.
-	 *
-	 * WaCxSRDisabledForSpriteScaling:ivb
-	 */
-	if (IS_IVYBRIDGE(dev) && scaled && ilk_disable_lp_wm(dev))
-		intel_wait_for_vblank(dev, intel_plane->pipe);
-
-	ilk_update_wm(crtc);
-}
-
 static void skl_pipe_wm_active_state(uint32_t val,
 				     struct skl_pipe_wm *active,
 				     bool is_transwm,
@@ -7108,7 +7098,6 @@ void intel_init_pm(struct drm_device *dev)
 		    (!IS_GEN5(dev) && dev_priv->wm.pri_latency[0] &&
 		     dev_priv->wm.spr_latency[0] && dev_priv->wm.cur_latency[0])) {
 			dev_priv->display.update_wm = ilk_update_wm;
-			dev_priv->display.update_sprite_wm = ilk_update_sprite_wm;
 		} else {
 			DRM_DEBUG_KMS("Failed to read display plane latency. "
 				      "Disable CxSR\n");
-- 
cgit v0.10.2


From 47c99438b52d12df50e182583634a4cfede3c920 Mon Sep 17 00:00:00 2001
From: Matt Roper <matthew.d.roper@intel.com>
Date: Thu, 24 Sep 2015 15:53:13 -0700
Subject: drm/i915: Drop intel_update_sprite_watermarks

The only platform that still has an update_sprite_wm entrypoint is SKL;
on SKL, intel_update_sprite_watermarks just updates intel_plane->wm and
then performs a regular watermark update.  However intel_plane->wm is
only used to update a couple fields in intel_wm_config, and those fields
are never used by the SKL code, so on SKL an update_sprite_wm is
effectively identical to an update_wm call.  Since we're already
ensuring that the regular intel_update_wm is called any time we'd try to
call intel_update_sprite_watermarks, the whole call is redundant and can
be dropped.

Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 29ce11f..268abbc 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -628,10 +628,6 @@ struct drm_i915_display_funcs {
 			  struct dpll *match_clock,
 			  struct dpll *best_clock);
 	void (*update_wm)(struct drm_crtc *crtc);
-	void (*update_sprite_wm)(struct drm_plane *plane,
-				 struct drm_crtc *crtc,
-				 uint32_t sprite_width, uint32_t sprite_height,
-				 int pixel_size, bool enable, bool scaled);
 	int (*modeset_calc_cdclk)(struct drm_atomic_state *state);
 	void (*modeset_commit_cdclk)(struct drm_atomic_state *state);
 	/* Returns the active state of the crtc, and if the crtc is active,
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index d3dab99..da894db 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -4805,7 +4805,6 @@ static void intel_post_plane_update(struct intel_crtc *crtc)
 	struct intel_crtc_atomic_commit *atomic = &crtc->atomic;
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct drm_plane *plane;
 
 	if (atomic->wait_vblank)
 		intel_wait_for_vblank(dev, crtc->pipe);
@@ -4824,10 +4823,6 @@ static void intel_post_plane_update(struct intel_crtc *crtc)
 	if (atomic->post_enable_primary)
 		intel_post_enable_primary(&crtc->base);
 
-	drm_for_each_plane_mask(plane, dev, atomic->update_sprite_watermarks)
-		intel_update_sprite_watermarks(plane, &crtc->base,
-					       0, 0, 0, false, false);
-
 	memset(atomic, 0, sizeof(*atomic));
 }
 
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index d6d7ac7..9431a12 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -1398,12 +1398,6 @@ void intel_init_clock_gating(struct drm_device *dev);
 void intel_suspend_hw(struct drm_device *dev);
 int ilk_wm_max_level(const struct drm_device *dev);
 void intel_update_watermarks(struct drm_crtc *crtc);
-void intel_update_sprite_watermarks(struct drm_plane *plane,
-				    struct drm_crtc *crtc,
-				    uint32_t sprite_width,
-				    uint32_t sprite_height,
-				    int pixel_size,
-				    bool enabled, bool scaled);
 void intel_init_pm(struct drm_device *dev);
 void intel_pm_setup(struct drm_device *dev);
 void intel_gpu_ips_init(struct drm_i915_private *dev_priv);
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index b8c3e22..6de3ef3 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -3148,18 +3148,9 @@ static void skl_compute_wm_global_parameters(struct drm_device *dev,
 					     struct intel_wm_config *config)
 {
 	struct drm_crtc *crtc;
-	struct drm_plane *plane;
 
 	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
 		config->num_pipes_active += to_intel_crtc(crtc)->active;
-
-	/* FIXME: I don't think we need those two global parameters on SKL */
-	list_for_each_entry(plane, &dev->mode_config.plane_list, head) {
-		struct intel_plane *intel_plane = to_intel_plane(plane);
-
-		config->sprites_enabled |= intel_plane->wm.enabled;
-		config->sprites_scaled |= intel_plane->wm.scaled;
-	}
 }
 
 static bool skl_compute_plane_wm(const struct drm_i915_private *dev_priv,
@@ -3682,39 +3673,6 @@ static void skl_update_wm(struct drm_crtc *crtc)
 	dev_priv->wm.skl_hw = *results;
 }
 
-static void
-skl_update_sprite_wm(struct drm_plane *plane, struct drm_crtc *crtc,
-		     uint32_t sprite_width, uint32_t sprite_height,
-		     int pixel_size, bool enabled, bool scaled)
-{
-	struct intel_plane *intel_plane = to_intel_plane(plane);
-	struct drm_framebuffer *fb = plane->state->fb;
-
-	intel_plane->wm.enabled = enabled;
-	intel_plane->wm.scaled = scaled;
-	intel_plane->wm.horiz_pixels = sprite_width;
-	intel_plane->wm.vert_pixels = sprite_height;
-	intel_plane->wm.tiling = DRM_FORMAT_MOD_NONE;
-
-	/* For planar: Bpp is for UV plane, y_Bpp is for Y plane */
-	intel_plane->wm.bytes_per_pixel =
-		(fb && fb->pixel_format == DRM_FORMAT_NV12) ?
-		drm_format_plane_cpp(plane->state->fb->pixel_format, 1) : pixel_size;
-	intel_plane->wm.y_bytes_per_pixel =
-		(fb && fb->pixel_format == DRM_FORMAT_NV12) ?
-		drm_format_plane_cpp(plane->state->fb->pixel_format, 0) : 0;
-
-	/*
-	 * Framebuffer can be NULL on plane disable, but it does not
-	 * matter for watermarks if we assume no tiling in that case.
-	 */
-	if (fb)
-		intel_plane->wm.tiling = fb->modifier[0];
-	intel_plane->wm.rotation = plane->state->rotation;
-
-	skl_update_wm(crtc);
-}
-
 static void ilk_update_wm(struct drm_crtc *crtc)
 {
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
@@ -4150,21 +4108,6 @@ void intel_update_watermarks(struct drm_crtc *crtc)
 		dev_priv->display.update_wm(crtc);
 }
 
-void intel_update_sprite_watermarks(struct drm_plane *plane,
-				    struct drm_crtc *crtc,
-				    uint32_t sprite_width,
-				    uint32_t sprite_height,
-				    int pixel_size,
-				    bool enabled, bool scaled)
-{
-	struct drm_i915_private *dev_priv = plane->dev->dev_private;
-
-	if (dev_priv->display.update_sprite_wm)
-		dev_priv->display.update_sprite_wm(plane, crtc,
-						   sprite_width, sprite_height,
-						   pixel_size, enabled, scaled);
-}
-
 /**
  * Lock protecting IPS related data structures
  */
@@ -7089,7 +7032,6 @@ void intel_init_pm(struct drm_device *dev)
 			dev_priv->display.init_clock_gating =
 				skl_init_clock_gating;
 		dev_priv->display.update_wm = skl_update_wm;
-		dev_priv->display.update_sprite_wm = skl_update_sprite_wm;
 	} else if (HAS_PCH_SPLIT(dev)) {
 		ilk_setup_wm_latency(dev);
 
diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c
index b229c67..dd2d568 100644
--- a/drivers/gpu/drm/i915/intel_sprite.c
+++ b/drivers/gpu/drm/i915/intel_sprite.c
@@ -192,7 +192,6 @@ skl_update_plane(struct drm_plane *drm_plane, struct drm_crtc *crtc,
 	const int pipe = intel_plane->pipe;
 	const int plane = intel_plane->plane + 1;
 	u32 plane_ctl, stride_div, stride;
-	int pixel_size = drm_format_plane_cpp(fb->pixel_format, 0);
 	const struct drm_intel_sprite_colorkey *key =
 		&to_intel_plane_state(drm_plane->state)->ckey;
 	unsigned long surf_addr;
@@ -211,10 +210,6 @@ skl_update_plane(struct drm_plane *drm_plane, struct drm_crtc *crtc,
 	rotation = drm_plane->state->rotation;
 	plane_ctl |= skl_plane_ctl_rotation(rotation);
 
-	intel_update_sprite_watermarks(drm_plane, crtc, src_w, src_h,
-				       pixel_size, true,
-				       src_w != crtc_w || src_h != crtc_h);
-
 	stride_div = intel_fb_stride_alignment(dev, fb->modifier[0],
 					       fb->pixel_format);
 
@@ -296,8 +291,6 @@ skl_disable_plane(struct drm_plane *dplane, struct drm_crtc *crtc)
 
 	I915_WRITE(PLANE_SURF(pipe, plane), 0);
 	POSTING_READ(PLANE_SURF(pipe, plane));
-
-	intel_update_sprite_watermarks(dplane, crtc, 0, 0, 0, false, false);
 }
 
 static void
@@ -540,10 +533,6 @@ ivb_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 	if (IS_HASWELL(dev) || IS_BROADWELL(dev))
 		sprctl |= SPRITE_PIPE_CSC_ENABLE;
 
-	intel_update_sprite_watermarks(plane, crtc, src_w, src_h, pixel_size,
-				       true,
-				       src_w != crtc_w || src_h != crtc_h);
-
 	/* Sizes are 0 based */
 	src_w--;
 	src_h--;
@@ -677,10 +666,6 @@ ilk_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 	if (IS_GEN6(dev))
 		dvscntr |= DVS_TRICKLE_FEED_DISABLE; /* must disable */
 
-	intel_update_sprite_watermarks(plane, crtc, src_w, src_h,
-				       pixel_size, true,
-				       src_w != crtc_w || src_h != crtc_h);
-
 	/* Sizes are 0 based */
 	src_w--;
 	src_h--;
-- 
cgit v0.10.2


From de165e0bccb4a8fe71debd5adb2f599609bcb8e0 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 24 Sep 2015 15:53:14 -0700
Subject: drm/i915: Refactor ilk_update_wm (v3)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Split ilk_update_wm() into two parts; one doing the programming
and the other the calculations.

v2: Fix typo in commit message

v3 (by Matt): Heavily rebased for current codebase.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index 6de3ef3..2f064de 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -3673,39 +3673,14 @@ static void skl_update_wm(struct drm_crtc *crtc)
 	dev_priv->wm.skl_hw = *results;
 }
 
-static void ilk_update_wm(struct drm_crtc *crtc)
+static void ilk_program_watermarks(struct drm_i915_private *dev_priv)
 {
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	struct intel_crtc_state *cstate = to_intel_crtc_state(crtc->state);
-	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_device *dev = dev_priv->dev;
+	struct intel_pipe_wm lp_wm_1_2 = {}, lp_wm_5_6 = {}, *best_lp_wm;
 	struct ilk_wm_maximums max;
+	struct intel_wm_config config = {};
 	struct ilk_wm_values results = {};
 	enum intel_ddb_partitioning partitioning;
-	struct intel_pipe_wm pipe_wm = {};
-	struct intel_pipe_wm lp_wm_1_2 = {}, lp_wm_5_6 = {}, *best_lp_wm;
-	struct intel_wm_config config = {};
-
-	WARN_ON(cstate->base.active != intel_crtc->active);
-
-	/*
-	 * IVB workaround: must disable low power watermarks for at least
-	 * one frame before enabling scaling.  LP watermarks can be re-enabled
-	 * when scaling is disabled.
-	 *
-	 * WaCxSRDisabledForSpriteScaling:ivb
-	 */
-	if (cstate->disable_lp_wm) {
-		ilk_disable_lp_wm(dev);
-		intel_wait_for_vblank(dev, intel_crtc->pipe);
-	}
-
-	intel_compute_pipe_wm(cstate, &pipe_wm);
-
-	if (!memcmp(&intel_crtc->wm.active, &pipe_wm, sizeof(pipe_wm)))
-		return;
-
-	intel_crtc->wm.active = pipe_wm;
 
 	ilk_compute_wm_config(dev, &config);
 
@@ -3731,6 +3706,37 @@ static void ilk_update_wm(struct drm_crtc *crtc)
 	ilk_write_wm_values(dev_priv, &results);
 }
 
+static void ilk_update_wm(struct drm_crtc *crtc)
+{
+	struct drm_i915_private *dev_priv = to_i915(crtc->dev);
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	struct intel_crtc_state *cstate = to_intel_crtc_state(crtc->state);
+	struct intel_pipe_wm pipe_wm = {};
+
+	WARN_ON(cstate->base.active != intel_crtc->active);
+
+	/*
+	 * IVB workaround: must disable low power watermarks for at least
+	 * one frame before enabling scaling.  LP watermarks can be re-enabled
+	 * when scaling is disabled.
+	 *
+	 * WaCxSRDisabledForSpriteScaling:ivb
+	 */
+	if (cstate->disable_lp_wm) {
+		ilk_disable_lp_wm(crtc->dev);
+		intel_wait_for_vblank(crtc->dev, intel_crtc->pipe);
+	}
+
+	intel_compute_pipe_wm(cstate, &pipe_wm);
+
+	if (!memcmp(&intel_crtc->wm.active, &pipe_wm, sizeof(pipe_wm)))
+		return;
+
+	intel_crtc->wm.active = pipe_wm;
+
+	ilk_program_watermarks(dev_priv);
+}
+
 static void skl_pipe_wm_active_state(uint32_t val,
 				     struct skl_pipe_wm *active,
 				     bool is_transwm,
-- 
cgit v0.10.2


From de4a9f83395e8a709ffe463dff6d1c51945da352 Mon Sep 17 00:00:00 2001
From: Matt Roper <matthew.d.roper@intel.com>
Date: Thu, 24 Sep 2015 15:53:15 -0700
Subject: drm/i915: Calculate pipe watermarks into CRTC state (v3)

A future patch will calculate these during the atomic 'check' phase
rather than at WM programming time, so let's store the watermark
values we're planning to use in the CRTC state; the values actually
active on the hardware remains in intel_crtc.

While we're at it, do some minor restructuring to keep ILK and SKL
values in a union.

v2: Don't move cxsr_allowed to state (Maarten)

v3: Only calculate watermarks in state.  Still keep active watermarks in
    intel_crtc itself.  (Ville)

Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 9431a12..4773292 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -332,6 +332,21 @@ struct intel_crtc_scaler_state {
 /* drm_mode->private_flags */
 #define I915_MODE_FLAG_INHERITED 1
 
+struct intel_pipe_wm {
+	struct intel_wm_level wm[5];
+	uint32_t linetime;
+	bool fbc_wm_enabled;
+	bool pipe_enabled;
+	bool sprites_enabled;
+	bool sprites_scaled;
+};
+
+struct skl_pipe_wm {
+	struct skl_wm_level wm[8];
+	struct skl_wm_level trans_wm;
+	uint32_t linetime;
+};
+
 struct intel_crtc_state {
 	struct drm_crtc_state base;
 
@@ -469,6 +484,17 @@ struct intel_crtc_state {
 
 	/* IVB sprite scaling w/a (WaCxSRDisabledForSpriteScaling:ivb) */
 	bool disable_lp_wm;
+
+	struct {
+		/*
+		 * optimal watermarks, programmed post-vblank when this state
+		 * is committed
+		 */
+		union {
+			struct intel_pipe_wm ilk;
+			struct skl_pipe_wm skl;
+		} optimal;
+	} wm;
 };
 
 struct vlv_wm_state {
@@ -480,15 +506,6 @@ struct vlv_wm_state {
 	bool cxsr;
 };
 
-struct intel_pipe_wm {
-	struct intel_wm_level wm[5];
-	uint32_t linetime;
-	bool fbc_wm_enabled;
-	bool pipe_enabled;
-	bool sprites_enabled;
-	bool sprites_scaled;
-};
-
 struct intel_mmio_flip {
 	struct work_struct work;
 	struct drm_i915_private *i915;
@@ -496,12 +513,6 @@ struct intel_mmio_flip {
 	struct intel_crtc *crtc;
 };
 
-struct skl_pipe_wm {
-	struct skl_wm_level wm[8];
-	struct skl_wm_level trans_wm;
-	uint32_t linetime;
-};
-
 /*
  * Tracking of operations that need to be performed at the beginning/end of an
  * atomic commit, outside the atomic section where interrupts are disabled.
@@ -569,9 +580,10 @@ struct intel_crtc {
 	/* per-pipe watermark state */
 	struct {
 		/* watermarks currently being used  */
-		struct intel_pipe_wm active;
-		/* SKL wm values currently in use */
-		struct skl_pipe_wm skl_active;
+		union {
+			struct intel_pipe_wm ilk;
+			struct skl_pipe_wm skl;
+		} active;
 		/* allow CxSR on this pipe */
 		bool cxsr_allowed;
 	} wm;
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index 2f064de..3857592 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -2331,7 +2331,7 @@ static void ilk_compute_wm_config(struct drm_device *dev,
 
 	/* Compute the currently _active_ config */
 	for_each_intel_crtc(dev, intel_crtc) {
-		const struct intel_pipe_wm *wm = &intel_crtc->wm.active;
+		const struct intel_pipe_wm *wm = &intel_crtc->wm.active.ilk;
 
 		if (!wm->pipe_enabled)
 			continue;
@@ -2428,7 +2428,9 @@ static void ilk_merge_wm_level(struct drm_device *dev,
 	ret_wm->enable = true;
 
 	for_each_intel_crtc(dev, intel_crtc) {
-		const struct intel_pipe_wm *active = &intel_crtc->wm.active;
+		const struct intel_crtc_state *cstate =
+			to_intel_crtc_state(intel_crtc->base.state);
+		const struct intel_pipe_wm *active = &cstate->wm.optimal.ilk;
 		const struct intel_wm_level *wm = &active->wm[level];
 
 		if (!active->pipe_enabled)
@@ -2576,14 +2578,15 @@ static void ilk_compute_wm_results(struct drm_device *dev,
 
 	/* LP0 register values */
 	for_each_intel_crtc(dev, intel_crtc) {
+		const struct intel_crtc_state *cstate =
+			to_intel_crtc_state(intel_crtc->base.state);
 		enum pipe pipe = intel_crtc->pipe;
-		const struct intel_wm_level *r =
-			&intel_crtc->wm.active.wm[0];
+		const struct intel_wm_level *r = &cstate->wm.optimal.ilk.wm[0];
 
 		if (WARN_ON(!r->enable))
 			continue;
 
-		results->wm_linetime[pipe] = intel_crtc->wm.active.linetime;
+		results->wm_linetime[pipe] = cstate->wm.optimal.ilk.linetime;
 
 		results->wm_pipe[pipe] =
 			(r->pri_val << WM0_PIPE_PLANE_SHIFT) |
@@ -3567,10 +3570,10 @@ static bool skl_update_pipe_wm(struct drm_crtc *crtc,
 	skl_allocate_pipe_ddb(cstate, config, ddb);
 	skl_compute_pipe_wm(cstate, ddb, pipe_wm);
 
-	if (!memcmp(&intel_crtc->wm.skl_active, pipe_wm, sizeof(*pipe_wm)))
+	if (!memcmp(&intel_crtc->wm.active.skl, pipe_wm, sizeof(*pipe_wm)))
 		return false;
 
-	intel_crtc->wm.skl_active = *pipe_wm;
+	intel_crtc->wm.active.skl = *pipe_wm;
 
 	return true;
 }
@@ -3648,7 +3651,8 @@ static void skl_update_wm(struct drm_crtc *crtc)
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct skl_wm_values *results = &dev_priv->wm.skl_results;
-	struct skl_pipe_wm pipe_wm = {};
+	struct intel_crtc_state *cstate = to_intel_crtc_state(crtc->state);
+	struct skl_pipe_wm *pipe_wm = &cstate->wm.optimal.skl;
 	struct intel_wm_config config = {};
 
 
@@ -3659,10 +3663,10 @@ static void skl_update_wm(struct drm_crtc *crtc)
 
 	skl_compute_wm_global_parameters(dev, &config);
 
-	if (!skl_update_pipe_wm(crtc, &config, &results->ddb, &pipe_wm))
+	if (!skl_update_pipe_wm(crtc, &config, &results->ddb, pipe_wm))
 		return;
 
-	skl_compute_wm_results(dev, &pipe_wm, results, intel_crtc);
+	skl_compute_wm_results(dev, pipe_wm, results, intel_crtc);
 	results->dirty[intel_crtc->pipe] = true;
 
 	skl_update_other_pipe_wm(dev, crtc, &config, results);
@@ -3711,7 +3715,6 @@ static void ilk_update_wm(struct drm_crtc *crtc)
 	struct drm_i915_private *dev_priv = to_i915(crtc->dev);
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct intel_crtc_state *cstate = to_intel_crtc_state(crtc->state);
-	struct intel_pipe_wm pipe_wm = {};
 
 	WARN_ON(cstate->base.active != intel_crtc->active);
 
@@ -3727,12 +3730,13 @@ static void ilk_update_wm(struct drm_crtc *crtc)
 		intel_wait_for_vblank(crtc->dev, intel_crtc->pipe);
 	}
 
-	intel_compute_pipe_wm(cstate, &pipe_wm);
+	intel_compute_pipe_wm(cstate, &cstate->wm.optimal.ilk);
 
-	if (!memcmp(&intel_crtc->wm.active, &pipe_wm, sizeof(pipe_wm)))
-		return;
+	if (!memcmp(&intel_crtc->wm.active.ilk,
+		    &cstate->wm.optimal.ilk,
+		    sizeof(cstate->wm.optimal.ilk)));
 
-	intel_crtc->wm.active = pipe_wm;
+	intel_crtc->wm.active.ilk = cstate->wm.optimal.ilk;
 
 	ilk_program_watermarks(dev_priv);
 }
@@ -3787,7 +3791,8 @@ static void skl_pipe_wm_get_hw_state(struct drm_crtc *crtc)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct skl_wm_values *hw = &dev_priv->wm.skl_hw;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	struct skl_pipe_wm *active = &intel_crtc->wm.skl_active;
+	struct intel_crtc_state *cstate = to_intel_crtc_state(crtc->state);
+	struct skl_pipe_wm *active = &cstate->wm.optimal.skl;
 	enum pipe pipe = intel_crtc->pipe;
 	int level, i, max_level;
 	uint32_t temp;
@@ -3831,6 +3836,8 @@ static void skl_pipe_wm_get_hw_state(struct drm_crtc *crtc)
 
 	temp = hw->plane_trans[pipe][PLANE_CURSOR];
 	skl_pipe_wm_active_state(temp, active, true, true, i, 0);
+
+	intel_crtc->wm.active.skl = *active;
 }
 
 void skl_wm_get_hw_state(struct drm_device *dev)
@@ -3850,7 +3857,8 @@ static void ilk_pipe_wm_get_hw_state(struct drm_crtc *crtc)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct ilk_wm_values *hw = &dev_priv->wm.hw;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	struct intel_pipe_wm *active = &intel_crtc->wm.active;
+	struct intel_crtc_state *cstate = to_intel_crtc_state(crtc->state);
+	struct intel_pipe_wm *active = &cstate->wm.optimal.ilk;
 	enum pipe pipe = intel_crtc->pipe;
 	static const unsigned int wm0_pipe_reg[] = {
 		[PIPE_A] = WM0_PIPEA_ILK,
@@ -3889,6 +3897,8 @@ static void ilk_pipe_wm_get_hw_state(struct drm_crtc *crtc)
 		for (level = 0; level <= max_level; level++)
 			active->wm[level].enable = true;
 	}
+
+	intel_crtc->wm.active.ilk = *active;
 }
 
 #define _FW_WM(value, plane) \
-- 
cgit v0.10.2


From a28170f3389f4e42db95e595b0d86384a79de696 Mon Sep 17 00:00:00 2001
From: Matt Roper <matthew.d.roper@intel.com>
Date: Thu, 24 Sep 2015 15:53:16 -0700
Subject: drm/i915: Calculate ILK-style watermarks during atomic check (v3)

Calculate pipe watermarks during atomic calculation phase, based on the
contents of the atomic transaction's state structure.  We still program
the watermarks at the same time we did before, but the computation now
happens much earlier.

While this patch isn't too exciting by itself, it paves the way for
future patches.  The eventual goal (which will be realized in future
patches in this series) is to calculate multiple sets up watermark
values up front, and then program them at different times (pre- vs
post-vblank) on the platforms that need a two-step watermark update.

While we're at it, s/intel_compute_pipe_wm/ilk_compute_pipe_wm/ since
this function only applies to ILK-style watermarks and we have a
completely different function for SKL-style watermarks.

Note that the original code had a memcmp() in ilk_update_wm() to avoid
calling ilk_program_watermarks() if the watermarks hadn't changed.  This
memcmp vanishes here, which means we may do some unnecessary result
generation and merging in cases where watermarks didn't change, but the
lower-level function ilk_write_wm_values already makes sure that we
don't actually try to program the watermark registers again.

v2: Squash a few commits from the original series together; no longer
    leave pre-calculated wm's in a separate temporary structure since
    it's easier to follow the logic if we just cut over to using the
    pre-calculated values directly.

v3:
 - Pass intel_crtc instead of drm_crtc to .compute_pipe_wm() entrypoint
   and use intel_atomic_get_crtc_state() to avoid need for extra
   casting.  (Ander)
 - Drop unused intel_check_crtc() function prototype.  (Ander)

Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 268abbc..00fef94 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -627,6 +627,8 @@ struct drm_i915_display_funcs {
 			  int target, int refclk,
 			  struct dpll *match_clock,
 			  struct dpll *best_clock);
+	int (*compute_pipe_wm)(struct intel_crtc *crtc,
+			       struct drm_atomic_state *state);
 	void (*update_wm)(struct drm_crtc *crtc);
 	int (*modeset_calc_cdclk)(struct drm_atomic_state *state);
 	void (*modeset_commit_cdclk)(struct drm_atomic_state *state);
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index da894db..f293de8 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -11828,6 +11828,12 @@ static int intel_crtc_atomic_check(struct drm_crtc *crtc,
 	}
 
 	ret = 0;
+	if (dev_priv->display.compute_pipe_wm) {
+		ret = dev_priv->display.compute_pipe_wm(intel_crtc, state);
+		if (ret)
+			return ret;
+	}
+
 	if (INTEL_INFO(dev)->gen >= 9) {
 		if (mode_changed)
 			ret = skl_update_scaler_crtc(pipe_config);
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index 3857592..52c8b6a 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -2029,9 +2029,11 @@ static void ilk_compute_wm_level(const struct drm_i915_private *dev_priv,
 				 const struct intel_crtc *intel_crtc,
 				 int level,
 				 struct intel_crtc_state *cstate,
+				 struct intel_plane_state *pristate,
+				 struct intel_plane_state *sprstate,
+				 struct intel_plane_state *curstate,
 				 struct intel_wm_level *result)
 {
-	struct intel_plane *intel_plane;
 	uint16_t pri_latency = dev_priv->wm.pri_latency[level];
 	uint16_t spr_latency = dev_priv->wm.spr_latency[level];
 	uint16_t cur_latency = dev_priv->wm.cur_latency[level];
@@ -2043,29 +2045,11 @@ static void ilk_compute_wm_level(const struct drm_i915_private *dev_priv,
 		cur_latency *= 5;
 	}
 
-	for_each_intel_plane_on_crtc(dev_priv->dev, intel_crtc, intel_plane) {
-		struct intel_plane_state *pstate =
-			to_intel_plane_state(intel_plane->base.state);
-
-		switch (intel_plane->base.type) {
-		case DRM_PLANE_TYPE_PRIMARY:
-			result->pri_val = ilk_compute_pri_wm(cstate, pstate,
-							     pri_latency,
-							     level);
-			result->fbc_val = ilk_compute_fbc_wm(cstate, pstate,
-							     result->pri_val);
-			break;
-		case DRM_PLANE_TYPE_OVERLAY:
-			result->spr_val = ilk_compute_spr_wm(cstate, pstate,
-							     spr_latency);
-			break;
-		case DRM_PLANE_TYPE_CURSOR:
-			result->cur_val = ilk_compute_cur_wm(cstate, pstate,
-							     cur_latency);
-			break;
-		}
-	}
-
+	result->pri_val = ilk_compute_pri_wm(cstate, pristate,
+					     pri_latency, level);
+	result->spr_val = ilk_compute_spr_wm(cstate, sprstate, spr_latency);
+	result->cur_val = ilk_compute_cur_wm(cstate, curstate, cur_latency);
+	result->fbc_val = ilk_compute_fbc_wm(cstate, pristate, result->pri_val);
 	result->enable = true;
 }
 
@@ -2343,15 +2327,18 @@ static void ilk_compute_wm_config(struct drm_device *dev,
 }
 
 /* Compute new watermarks for the pipe */
-static bool intel_compute_pipe_wm(struct intel_crtc_state *cstate,
-				  struct intel_pipe_wm *pipe_wm)
+static int ilk_compute_pipe_wm(struct intel_crtc *intel_crtc,
+			       struct drm_atomic_state *state)
 {
-	struct drm_crtc *crtc = cstate->base.crtc;
-	struct drm_device *dev = crtc->dev;
+	struct intel_pipe_wm *pipe_wm;
+	struct drm_device *dev = intel_crtc->base.dev;
 	const struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	struct intel_crtc_state *cstate = NULL;
 	struct intel_plane *intel_plane;
+	struct drm_plane_state *ps;
+	struct intel_plane_state *pristate = NULL;
 	struct intel_plane_state *sprstate = NULL;
+	struct intel_plane_state *curstate = NULL;
 	int level, max_level = ilk_wm_max_level(dev);
 	/* LP0 watermark maximums depend on this pipe alone */
 	struct intel_wm_config config = {
@@ -2359,11 +2346,24 @@ static bool intel_compute_pipe_wm(struct intel_crtc_state *cstate,
 	};
 	struct ilk_wm_maximums max;
 
+	cstate = intel_atomic_get_crtc_state(state, intel_crtc);
+	if (IS_ERR(cstate))
+		return PTR_ERR(cstate);
+
+	pipe_wm = &cstate->wm.optimal.ilk;
+
 	for_each_intel_plane_on_crtc(dev, intel_crtc, intel_plane) {
-		if (intel_plane->base.type == DRM_PLANE_TYPE_OVERLAY) {
-			sprstate = to_intel_plane_state(intel_plane->base.state);
-			break;
-		}
+		ps = drm_atomic_get_plane_state(state,
+						&intel_plane->base);
+		if (IS_ERR(ps))
+			return PTR_ERR(ps);
+
+		if (intel_plane->base.type == DRM_PLANE_TYPE_PRIMARY)
+			pristate = to_intel_plane_state(ps);
+		else if (intel_plane->base.type == DRM_PLANE_TYPE_OVERLAY)
+			sprstate = to_intel_plane_state(ps);
+		else if (intel_plane->base.type == DRM_PLANE_TYPE_CURSOR)
+			curstate = to_intel_plane_state(ps);
 	}
 
 	config.sprites_enabled = sprstate->visible;
@@ -2372,7 +2372,7 @@ static bool intel_compute_pipe_wm(struct intel_crtc_state *cstate,
 		drm_rect_height(&sprstate->dst) != drm_rect_height(&sprstate->src) >> 16);
 
 	pipe_wm->pipe_enabled = cstate->base.active;
-	pipe_wm->sprites_enabled = sprstate->visible;
+	pipe_wm->sprites_enabled = config.sprites_enabled;
 	pipe_wm->sprites_scaled = config.sprites_scaled;
 
 	/* ILK/SNB: LP2+ watermarks only w/o sprites */
@@ -2383,24 +2383,27 @@ static bool intel_compute_pipe_wm(struct intel_crtc_state *cstate,
 	if (config.sprites_scaled)
 		max_level = 0;
 
-	ilk_compute_wm_level(dev_priv, intel_crtc, 0, cstate, &pipe_wm->wm[0]);
+	ilk_compute_wm_level(dev_priv, intel_crtc, 0, cstate,
+			     pristate, sprstate, curstate, &pipe_wm->wm[0]);
 
 	if (IS_HASWELL(dev) || IS_BROADWELL(dev))
-		pipe_wm->linetime = hsw_compute_linetime_wm(dev, crtc);
+		pipe_wm->linetime = hsw_compute_linetime_wm(dev,
+							    &intel_crtc->base);
 
 	/* LP0 watermarks always use 1/2 DDB partitioning */
 	ilk_compute_wm_maximums(dev, 0, &config, INTEL_DDB_PART_1_2, &max);
 
 	/* At least LP0 must be valid */
 	if (!ilk_validate_wm_level(0, &max, &pipe_wm->wm[0]))
-		return false;
+		return -EINVAL;
 
 	ilk_compute_wm_reg_maximums(dev, 1, &max);
 
 	for (level = 1; level <= max_level; level++) {
 		struct intel_wm_level wm = {};
 
-		ilk_compute_wm_level(dev_priv, intel_crtc, level, cstate, &wm);
+		ilk_compute_wm_level(dev_priv, intel_crtc, level, cstate,
+				     pristate, sprstate, curstate, &wm);
 
 		/*
 		 * Disable any watermark level that exceeds the
@@ -2413,7 +2416,7 @@ static bool intel_compute_pipe_wm(struct intel_crtc_state *cstate,
 		pipe_wm->wm[level] = wm;
 	}
 
-	return true;
+	return 0;
 }
 
 /*
@@ -3730,12 +3733,6 @@ static void ilk_update_wm(struct drm_crtc *crtc)
 		intel_wait_for_vblank(crtc->dev, intel_crtc->pipe);
 	}
 
-	intel_compute_pipe_wm(cstate, &cstate->wm.optimal.ilk);
-
-	if (!memcmp(&intel_crtc->wm.active.ilk,
-		    &cstate->wm.optimal.ilk,
-		    sizeof(cstate->wm.optimal.ilk)));
-
 	intel_crtc->wm.active.ilk = cstate->wm.optimal.ilk;
 
 	ilk_program_watermarks(dev_priv);
@@ -7056,6 +7053,7 @@ void intel_init_pm(struct drm_device *dev)
 		    (!IS_GEN5(dev) && dev_priv->wm.pri_latency[0] &&
 		     dev_priv->wm.spr_latency[0] && dev_priv->wm.cur_latency[0])) {
 			dev_priv->display.update_wm = ilk_update_wm;
+			dev_priv->display.compute_pipe_wm = ilk_compute_pipe_wm;
 		} else {
 			DRM_DEBUG_KMS("Failed to read display plane latency. "
 				      "Disable CxSR\n");
-- 
cgit v0.10.2


From a4611e444694869a9ccec69e4e640aee9b7dc58e Mon Sep 17 00:00:00 2001
From: Matt Roper <matthew.d.roper@intel.com>
Date: Thu, 24 Sep 2015 15:53:17 -0700
Subject: drm/i915: Don't set plane visible during HW readout if CRTC is off

We already ensure that pstate->visible = false when crtc->active = false
during runtime programming; make sure we follow the same logic when
reading out initial hardware state.

Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index f293de8..b6a68bd 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -15166,7 +15166,7 @@ static void readout_plane_state(struct intel_crtc *crtc)
 	struct intel_plane_state *plane_state =
 		to_intel_plane_state(primary->state);
 
-	plane_state->visible =
+	plane_state->visible = crtc->active &&
 		primary_get_hw_state(to_intel_plane(primary));
 
 	if (plane_state->visible)
-- 
cgit v0.10.2


From 76305b1a68e196c4f76d15ec1897097bb7a7c641 Mon Sep 17 00:00:00 2001
From: Matt Roper <matthew.d.roper@intel.com>
Date: Thu, 24 Sep 2015 15:53:18 -0700
Subject: drm/i915: Calculate watermark configuration during atomic check (v2)

v2: Don't forget to actually check the cstate->active value when
    tallying up the number of active CRTC's.  (Ander)

Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 00fef94..b17dac6 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -1692,6 +1692,13 @@ struct i915_execbuffer_params {
 	struct drm_i915_gem_request     *request;
 };
 
+/* used in computing the new watermarks state */
+struct intel_wm_config {
+	unsigned int num_pipes_active;
+	bool sprites_enabled;
+	bool sprites_scaled;
+};
+
 struct drm_i915_private {
 	struct drm_device *dev;
 	struct kmem_cache *objects;
@@ -1912,6 +1919,9 @@ struct drm_i915_private {
 		 */
 		uint16_t skl_latency[8];
 
+		/* Committed wm config */
+		struct intel_wm_config config;
+
 		/*
 		 * The skl_wm_values structure is a bit too big for stack
 		 * allocation, so we keep the staging struct where we store
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index b6a68bd..bbeb6d3 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -13039,6 +13039,45 @@ static int intel_modeset_checks(struct drm_atomic_state *state)
 	return 0;
 }
 
+/*
+ * Handle calculation of various watermark data at the end of the atomic check
+ * phase.  The code here should be run after the per-crtc and per-plane 'check'
+ * handlers to ensure that all derived state has been updated.
+ */
+static void calc_watermark_data(struct drm_atomic_state *state)
+{
+	struct drm_device *dev = state->dev;
+	struct intel_atomic_state *intel_state = to_intel_atomic_state(state);
+	struct drm_crtc *crtc;
+	struct drm_crtc_state *cstate;
+	struct drm_plane *plane;
+	struct drm_plane_state *pstate;
+
+	/*
+	 * Calculate watermark configuration details now that derived
+	 * plane/crtc state is all properly updated.
+	 */
+	drm_for_each_crtc(crtc, dev) {
+		cstate = drm_atomic_get_existing_crtc_state(state, crtc) ?:
+			crtc->state;
+
+		if (cstate->active)
+			intel_state->wm_config.num_pipes_active++;
+	}
+	drm_for_each_legacy_plane(plane, dev) {
+		pstate = drm_atomic_get_existing_plane_state(state, plane) ?:
+			plane->state;
+
+		if (!to_intel_plane_state(pstate)->visible)
+			continue;
+
+		intel_state->wm_config.sprites_enabled = true;
+		if (pstate->crtc_w != pstate->src_w >> 16 ||
+		    pstate->crtc_h != pstate->src_h >> 16)
+			intel_state->wm_config.sprites_scaled = true;
+	}
+}
+
 /**
  * intel_atomic_check - validate state object
  * @dev: drm device
@@ -13047,6 +13086,7 @@ static int intel_modeset_checks(struct drm_atomic_state *state)
 static int intel_atomic_check(struct drm_device *dev,
 			      struct drm_atomic_state *state)
 {
+	struct intel_atomic_state *intel_state = to_intel_atomic_state(state);
 	struct drm_crtc *crtc;
 	struct drm_crtc_state *crtc_state;
 	int ret, i;
@@ -13110,10 +13150,15 @@ static int intel_atomic_check(struct drm_device *dev,
 		if (ret)
 			return ret;
 	} else
-		to_intel_atomic_state(state)->cdclk =
-			to_i915(state->dev)->cdclk_freq;
+		intel_state->cdclk = to_i915(state->dev)->cdclk_freq;
 
-	return drm_atomic_helper_check_planes(state->dev, state);
+	ret = drm_atomic_helper_check_planes(state->dev, state);
+	if (ret)
+		return ret;
+
+	calc_watermark_data(state);
+
+	return 0;
 }
 
 /**
@@ -13153,6 +13198,7 @@ static int intel_atomic_commit(struct drm_device *dev,
 		return ret;
 
 	drm_atomic_helper_swap_state(dev, state);
+	dev_priv->wm.config = to_intel_atomic_state(state)->wm_config;
 
 	for_each_crtc_in_state(state, crtc, crtc_state, i) {
 		struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 4773292..dfd2d10 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -248,6 +248,7 @@ struct intel_atomic_state {
 	unsigned int cdclk;
 	bool dpll_set;
 	struct intel_shared_dpll_config shared_dpll[I915_NUM_PLLS];
+	struct intel_wm_config wm_config;
 };
 
 struct intel_plane_state {
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index 52c8b6a..eb9a66d 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -1772,13 +1772,6 @@ struct ilk_wm_maximums {
 	uint16_t fbc;
 };
 
-/* used in computing the new watermarks state */
-struct intel_wm_config {
-	unsigned int num_pipes_active;
-	bool sprites_enabled;
-	bool sprites_scaled;
-};
-
 /*
  * For both WM_PIPE and WM_LP.
  * mem_value must be in 0.1us units.
@@ -2308,24 +2301,6 @@ static void skl_setup_wm_latency(struct drm_device *dev)
 	intel_print_wm_latency(dev, "Gen9 Plane", dev_priv->wm.skl_latency);
 }
 
-static void ilk_compute_wm_config(struct drm_device *dev,
-				  struct intel_wm_config *config)
-{
-	struct intel_crtc *intel_crtc;
-
-	/* Compute the currently _active_ config */
-	for_each_intel_crtc(dev, intel_crtc) {
-		const struct intel_pipe_wm *wm = &intel_crtc->wm.active.ilk;
-
-		if (!wm->pipe_enabled)
-			continue;
-
-		config->sprites_enabled |= wm->sprites_enabled;
-		config->sprites_scaled |= wm->sprites_scaled;
-		config->num_pipes_active++;
-	}
-}
-
 /* Compute new watermarks for the pipe */
 static int ilk_compute_pipe_wm(struct intel_crtc *intel_crtc,
 			       struct drm_atomic_state *state)
@@ -2971,11 +2946,12 @@ skl_get_total_relative_data_rate(const struct intel_crtc_state *cstate)
 
 static void
 skl_allocate_pipe_ddb(struct intel_crtc_state *cstate,
-		      const struct intel_wm_config *config,
 		      struct skl_ddb_allocation *ddb /* out */)
 {
 	struct drm_crtc *crtc = cstate->base.crtc;
 	struct drm_device *dev = crtc->dev;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct intel_wm_config *config = &dev_priv->wm.config;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct intel_plane *intel_plane;
 	enum pipe pipe = intel_crtc->pipe;
@@ -3150,15 +3126,6 @@ static bool skl_ddb_allocation_changed(const struct skl_ddb_allocation *new_ddb,
 	return false;
 }
 
-static void skl_compute_wm_global_parameters(struct drm_device *dev,
-					     struct intel_wm_config *config)
-{
-	struct drm_crtc *crtc;
-
-	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
-		config->num_pipes_active += to_intel_crtc(crtc)->active;
-}
-
 static bool skl_compute_plane_wm(const struct drm_i915_private *dev_priv,
 				 struct intel_crtc_state *cstate,
 				 struct intel_plane *intel_plane,
@@ -3563,14 +3530,13 @@ static void skl_flush_wm_values(struct drm_i915_private *dev_priv,
 }
 
 static bool skl_update_pipe_wm(struct drm_crtc *crtc,
-			       struct intel_wm_config *config,
 			       struct skl_ddb_allocation *ddb, /* out */
 			       struct skl_pipe_wm *pipe_wm /* out */)
 {
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct intel_crtc_state *cstate = to_intel_crtc_state(crtc->state);
 
-	skl_allocate_pipe_ddb(cstate, config, ddb);
+	skl_allocate_pipe_ddb(cstate, ddb);
 	skl_compute_pipe_wm(cstate, ddb, pipe_wm);
 
 	if (!memcmp(&intel_crtc->wm.active.skl, pipe_wm, sizeof(*pipe_wm)))
@@ -3583,7 +3549,6 @@ static bool skl_update_pipe_wm(struct drm_crtc *crtc,
 
 static void skl_update_other_pipe_wm(struct drm_device *dev,
 				     struct drm_crtc *crtc,
-				     struct intel_wm_config *config,
 				     struct skl_wm_values *r)
 {
 	struct intel_crtc *intel_crtc;
@@ -3613,7 +3578,7 @@ static void skl_update_other_pipe_wm(struct drm_device *dev,
 		if (!intel_crtc->active)
 			continue;
 
-		wm_changed = skl_update_pipe_wm(&intel_crtc->base, config,
+		wm_changed = skl_update_pipe_wm(&intel_crtc->base,
 						&r->ddb, &pipe_wm);
 
 		/*
@@ -3656,7 +3621,6 @@ static void skl_update_wm(struct drm_crtc *crtc)
 	struct skl_wm_values *results = &dev_priv->wm.skl_results;
 	struct intel_crtc_state *cstate = to_intel_crtc_state(crtc->state);
 	struct skl_pipe_wm *pipe_wm = &cstate->wm.optimal.skl;
-	struct intel_wm_config config = {};
 
 
 	/* Clear all dirty flags */
@@ -3664,15 +3628,13 @@ static void skl_update_wm(struct drm_crtc *crtc)
 
 	skl_clear_wm(results, intel_crtc->pipe);
 
-	skl_compute_wm_global_parameters(dev, &config);
-
-	if (!skl_update_pipe_wm(crtc, &config, &results->ddb, pipe_wm))
+	if (!skl_update_pipe_wm(crtc, &results->ddb, pipe_wm))
 		return;
 
 	skl_compute_wm_results(dev, pipe_wm, results, intel_crtc);
 	results->dirty[intel_crtc->pipe] = true;
 
-	skl_update_other_pipe_wm(dev, crtc, &config, results);
+	skl_update_other_pipe_wm(dev, crtc, results);
 	skl_write_wm_values(dev_priv, results);
 	skl_flush_wm_values(dev_priv, results);
 
@@ -3685,20 +3647,18 @@ static void ilk_program_watermarks(struct drm_i915_private *dev_priv)
 	struct drm_device *dev = dev_priv->dev;
 	struct intel_pipe_wm lp_wm_1_2 = {}, lp_wm_5_6 = {}, *best_lp_wm;
 	struct ilk_wm_maximums max;
-	struct intel_wm_config config = {};
+	struct intel_wm_config *config = &dev_priv->wm.config;
 	struct ilk_wm_values results = {};
 	enum intel_ddb_partitioning partitioning;
 
-	ilk_compute_wm_config(dev, &config);
-
-	ilk_compute_wm_maximums(dev, 1, &config, INTEL_DDB_PART_1_2, &max);
-	ilk_wm_merge(dev, &config, &max, &lp_wm_1_2);
+	ilk_compute_wm_maximums(dev, 1, config, INTEL_DDB_PART_1_2, &max);
+	ilk_wm_merge(dev, config, &max, &lp_wm_1_2);
 
 	/* 5/6 split only in single pipe config on IVB+ */
 	if (INTEL_INFO(dev)->gen >= 7 &&
-	    config.num_pipes_active == 1 && config.sprites_enabled) {
-		ilk_compute_wm_maximums(dev, 1, &config, INTEL_DDB_PART_5_6, &max);
-		ilk_wm_merge(dev, &config, &max, &lp_wm_5_6);
+	    config->num_pipes_active == 1 && config->sprites_enabled) {
+		ilk_compute_wm_maximums(dev, 1, config, INTEL_DDB_PART_5_6, &max);
+		ilk_wm_merge(dev, config, &max, &lp_wm_5_6);
 
 		best_lp_wm = ilk_find_best_result(dev, &lp_wm_1_2, &lp_wm_5_6);
 	} else {
-- 
cgit v0.10.2


From 8fa884dc355ffd0caa964a284a530bf747df5c77 Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Thu, 1 Oct 2015 08:44:24 +0200
Subject: drm: Hack around CONFIG_AGP=m build failures
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Surprisingly kbuild can't cope with tristates in the
<module>-$(CONFIG_FOO) pattern. This patch hacks up a solution.

Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Reported-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Michal Marek <mmarek@suse.com>
Cc: linux-kbuild@vger.kernel.org
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

diff --git a/drivers/gpu/drm/Makefile b/drivers/gpu/drm/Makefile
index f458d6e..e814517 100644
--- a/drivers/gpu/drm/Makefile
+++ b/drivers/gpu/drm/Makefile
@@ -21,6 +21,8 @@ drm-$(CONFIG_DRM_PANEL) += drm_panel.o
 drm-$(CONFIG_OF) += drm_of.o
 drm-$(CONFIG_AGP) += drm_agpsupport.o
 
+drm-y += $(drm-m)
+
 drm_kms_helper-y := drm_crtc_helper.o drm_dp_helper.o drm_probe_helper.o \
 		drm_plane_helper.o drm_dp_mst_topology.o drm_atomic_helper.o
 drm_kms_helper-$(CONFIG_DRM_LOAD_EDID_FIRMWARE) += drm_edid_load.o
-- 
cgit v0.10.2


From fa4270d8e0257b4b76f11baa2866f4313d29aaf5 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 30 Sep 2015 19:21:34 +0300
Subject: drm: Don't zero vblank timestamps from the irq handler
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

If we couldn't get a high precisions vblank timestamp, we currently
store a zeroed timestamp instead and assume the next vblank irq to
get us something better. This makes sense when trying to update the
timestamp from eg. vblank enable. But if we do this from the vblank
irq we will never get a vblank timestamp unless we high precision
timestamps are available and succeeded. This break weston for instance
on drivers lacking high precision timestamps.

To fix this, zero the timestamp only when not called from vbl irq.
When called from the irq, we still want the timestamp, even if not
perfect.

This fixes a regression from
4dfd64862ff852df drm: Use vblank timestamps to guesstimate how many vblanks were missed

Cc: Mario Kleiner <mario.kleiner.de@gmail.com>
Cc: Thierry Reding <thierry.reding@gmail.com>
Reported-by: Thierry Reding <thierry.reding@gmail.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Tested-by: Thierry Reding <treding@nvidia.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c
index ed2394e..6bff6d3 100644
--- a/drivers/gpu/drm/drm_irq.c
+++ b/drivers/gpu/drm/drm_irq.c
@@ -232,10 +232,11 @@ static void drm_update_vblank_count(struct drm_device *dev, unsigned int pipe,
 
 	/*
 	 * Only reinitialize corresponding vblank timestamp if high-precision query
-	 * available and didn't fail. Otherwise reinitialize delayed at next vblank
-	 * interrupt and assign 0 for now, to mark the vblanktimestamp as invalid.
+	 * available and didn't fail, or we were called from the vblank interrupt.
+	 * Otherwise reinitialize delayed at next vblank interrupt and assign 0
+	 * for now, to mark the vblanktimestamp as invalid.
 	 */
-	if (!rc)
+	if (!rc && (flags & DRM_CALLED_FROM_VBLIRQ) == 0)
 		t_vblank = (struct timeval) {0, 0};
 
 	store_vblank(dev, pipe, diff, &t_vblank, cur_vblank);
-- 
cgit v0.10.2


From 945ef5bb146e65c58edb866eb0708e98224e54ab Mon Sep 17 00:00:00 2001
From: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Date: Wed, 30 Sep 2015 21:57:06 +0200
Subject: vgaarb: use kzalloc in vga_arbiter_add_pci_device()

Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/vga/vgaarb.c b/drivers/gpu/vga/vgaarb.c
index a0b4334..3166e4b 100644
--- a/drivers/gpu/vga/vgaarb.c
+++ b/drivers/gpu/vga/vgaarb.c
@@ -531,7 +531,7 @@ static bool vga_arbiter_add_pci_device(struct pci_dev *pdev)
 		return false;
 
 	/* Allocate structure */
-	vgadev = kmalloc(sizeof(struct vga_device), GFP_KERNEL);
+	vgadev = kzalloc(sizeof(struct vga_device), GFP_KERNEL);
 	if (vgadev == NULL) {
 		pr_err("failed to allocate pci device\n");
 		/*
@@ -541,8 +541,6 @@ static bool vga_arbiter_add_pci_device(struct pci_dev *pdev)
 		return false;
 	}
 
-	memset(vgadev, 0, sizeof(*vgadev));
-
 	/* Take lock & check for duplicates */
 	spin_lock_irqsave(&vga_lock, flags);
 	if (vgadev_find(pdev) != NULL) {
-- 
cgit v0.10.2


From 66377efa3fd468283a092f17692e81c2344b03ed Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Wed, 15 Jul 2015 18:11:24 +0100
Subject: drm/armada: use xchg() to atomically update dplane->old_fb

Rather than using a spinlock, use xchg() to atomically update
dplane->old_fb.  This allows us to eliminate dplane->lock.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/armada/armada_overlay.c b/drivers/gpu/drm/armada/armada_overlay.c
index 093c2d4..5e2dcec 100644
--- a/drivers/gpu/drm/armada/armada_overlay.c
+++ b/drivers/gpu/drm/armada/armada_overlay.c
@@ -31,7 +31,6 @@ struct armada_ovl_plane_properties {
 
 struct armada_ovl_plane {
 	struct drm_plane base;
-	spinlock_t lock;
 	struct drm_framebuffer *old_fb;
 	uint32_t src_hw;
 	uint32_t dst_hw;
@@ -76,10 +75,7 @@ static void armada_ovl_retire_fb(struct armada_ovl_plane *dplane,
 {
 	struct drm_framebuffer *old_fb;
 
-	spin_lock(&dplane->lock);
-	old_fb = dplane->old_fb;
-	dplane->old_fb = fb;
-	spin_unlock(&dplane->lock);
+	old_fb = xchg(&dplane->old_fb, fb);
 
 	if (old_fb)
 		armada_drm_queue_unref_work(dplane->base.dev, old_fb);
@@ -289,10 +285,7 @@ static int armada_ovl_plane_disable(struct drm_plane *plane)
 	if (plane->fb)
 		drm_framebuffer_unreference(plane->fb);
 
-	spin_lock_irq(&dplane->lock);
-	fb = dplane->old_fb;
-	dplane->old_fb = NULL;
-	spin_unlock_irq(&dplane->lock);
+	fb = xchg(&dplane->old_fb, NULL);
 	if (fb)
 		drm_framebuffer_unreference(fb);
 
@@ -464,7 +457,6 @@ int armada_overlay_plane_create(struct drm_device *dev, unsigned long crtcs)
 	if (!dplane)
 		return -ENOMEM;
 
-	spin_lock_init(&dplane->lock);
 	init_waitqueue_head(&dplane->vbl.wait);
 	armada_drm_vbl_event_init(&dplane->vbl.update, armada_ovl_plane_vbl,
 				  dplane);
-- 
cgit v0.10.2


From d563c24514166d01b87cc96f92fe93b635d24c6e Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Wed, 15 Jul 2015 18:11:24 +0100
Subject: drm/armada: update armada overlay to use drm_universal_plane_init()

Use the new drm_universal_plane_init() rather than the legacy
drm_plane_init().

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/armada/armada_overlay.c b/drivers/gpu/drm/armada/armada_overlay.c
index 5e2dcec..8f2b216 100644
--- a/drivers/gpu/drm/armada/armada_overlay.c
+++ b/drivers/gpu/drm/armada/armada_overlay.c
@@ -461,9 +461,11 @@ int armada_overlay_plane_create(struct drm_device *dev, unsigned long crtcs)
 	armada_drm_vbl_event_init(&dplane->vbl.update, armada_ovl_plane_vbl,
 				  dplane);
 
-	drm_plane_init(dev, &dplane->base, crtcs, &armada_ovl_plane_funcs,
-		       armada_ovl_formats, ARRAY_SIZE(armada_ovl_formats),
-		       false);
+	ret = drm_universal_plane_init(dev, &dplane->base, crtcs,
+				       &armada_ovl_plane_funcs,
+				       armada_ovl_formats,
+				       ARRAY_SIZE(armada_ovl_formats),
+				       DRM_PLANE_TYPE_OVERLAY);
 	if (ret) {
 		kfree(dplane);
 		return ret;
-- 
cgit v0.10.2


From 561f60bc511f6ec054b566205b5c40ab9558a0ff Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Wed, 15 Jul 2015 18:11:24 +0100
Subject: drm/armada: introduce generic armada_plane struct

Introduce a generic armada_plane struct which will eventually be used
for both the primary and overlay planes.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/armada/armada_crtc.h b/drivers/gpu/drm/armada/armada_crtc.h
index a86243e..549b5f5 100644
--- a/drivers/gpu/drm/armada/armada_crtc.h
+++ b/drivers/gpu/drm/armada/armada_crtc.h
@@ -34,6 +34,11 @@ struct armada_regs {
 struct armada_frame_work;
 struct armada_variant;
 
+struct armada_plane {
+	struct drm_plane	base;
+};
+#define drm_to_armada_plane(p) container_of(p, struct armada_plane, base)
+
 struct armada_crtc {
 	struct drm_crtc		crtc;
 	const struct armada_variant *variant;
diff --git a/drivers/gpu/drm/armada/armada_overlay.c b/drivers/gpu/drm/armada/armada_overlay.c
index 8f2b216..e5a5b73 100644
--- a/drivers/gpu/drm/armada/armada_overlay.c
+++ b/drivers/gpu/drm/armada/armada_overlay.c
@@ -30,7 +30,7 @@ struct armada_ovl_plane_properties {
 };
 
 struct armada_ovl_plane {
-	struct drm_plane base;
+	struct armada_plane base;
 	struct drm_framebuffer *old_fb;
 	uint32_t src_hw;
 	uint32_t dst_hw;
@@ -43,7 +43,8 @@ struct armada_ovl_plane {
 	} vbl;
 	struct armada_ovl_plane_properties prop;
 };
-#define drm_to_armada_ovl_plane(p) container_of(p, struct armada_ovl_plane, base)
+#define drm_to_armada_ovl_plane(p) \
+	container_of(p, struct armada_ovl_plane, base.base)
 
 
 static void
@@ -78,7 +79,7 @@ static void armada_ovl_retire_fb(struct armada_ovl_plane *dplane,
 	old_fb = xchg(&dplane->old_fb, fb);
 
 	if (old_fb)
-		armada_drm_queue_unref_work(dplane->base.dev, old_fb);
+		armada_drm_queue_unref_work(dplane->base.base.dev, old_fb);
 }
 
 /* === Plane support === */
@@ -266,10 +267,10 @@ static int armada_ovl_plane_disable(struct drm_plane *plane)
 	struct drm_framebuffer *fb;
 	struct armada_crtc *dcrtc;
 
-	if (!dplane->base.crtc)
+	if (!dplane->base.base.crtc)
 		return 0;
 
-	dcrtc = drm_to_armada_crtc(dplane->base.crtc);
+	dcrtc = drm_to_armada_crtc(dplane->base.base.crtc);
 	dcrtc->plane = NULL;
 
 	spin_lock_irq(&dcrtc->irq_lock);
@@ -362,9 +363,9 @@ static int armada_ovl_plane_set_property(struct drm_plane *plane,
 		update_attr = true;
 	}
 
-	if (update_attr && dplane->base.crtc)
+	if (update_attr && dplane->base.base.crtc)
 		armada_ovl_update_attr(&dplane->prop,
-				       drm_to_armada_crtc(dplane->base.crtc));
+				       drm_to_armada_crtc(dplane->base.base.crtc));
 
 	return 0;
 }
@@ -461,7 +462,7 @@ int armada_overlay_plane_create(struct drm_device *dev, unsigned long crtcs)
 	armada_drm_vbl_event_init(&dplane->vbl.update, armada_ovl_plane_vbl,
 				  dplane);
 
-	ret = drm_universal_plane_init(dev, &dplane->base, crtcs,
+	ret = drm_universal_plane_init(dev, &dplane->base.base, crtcs,
 				       &armada_ovl_plane_funcs,
 				       armada_ovl_formats,
 				       ARRAY_SIZE(armada_ovl_formats),
@@ -479,7 +480,7 @@ int armada_overlay_plane_create(struct drm_device *dev, unsigned long crtcs)
 	dplane->prop.contrast = 0x4000;
 	dplane->prop.saturation = 0x4000;
 
-	mobj = &dplane->base.base;
+	mobj = &dplane->base.base.base;
 	drm_object_attach_property(mobj, priv->colorkey_prop,
 				   0x0101fe);
 	drm_object_attach_property(mobj, priv->colorkey_min_prop,
-- 
cgit v0.10.2


From 1c914cecb5bc4b097df07b799d39abac842ce193 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Wed, 15 Jul 2015 18:11:24 +0100
Subject: drm/armada: add primary plane creation

Use drm_primary_helper_create_plane() to create our primary plane, and
register the CRTC with drm_crtc_init_with_planes().  This enables the
primary plane to be initialised with the supported format information.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/armada/armada_crtc.c b/drivers/gpu/drm/armada/armada_crtc.c
index 5d62764..b96b77b 100644
--- a/drivers/gpu/drm/armada/armada_crtc.c
+++ b/drivers/gpu/drm/armada/armada_crtc.c
@@ -33,6 +33,23 @@ enum csc_mode {
 	CSC_RGB_STUDIO = 2,
 };
 
+static const uint32_t armada_primary_formats[] = {
+	DRM_FORMAT_UYVY,
+	DRM_FORMAT_YUYV,
+	DRM_FORMAT_VYUY,
+	DRM_FORMAT_YVYU,
+	DRM_FORMAT_ARGB8888,
+	DRM_FORMAT_ABGR8888,
+	DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_XBGR8888,
+	DRM_FORMAT_RGB888,
+	DRM_FORMAT_BGR888,
+	DRM_FORMAT_ARGB1555,
+	DRM_FORMAT_ABGR1555,
+	DRM_FORMAT_RGB565,
+	DRM_FORMAT_BGR565,
+};
+
 /*
  * A note about interlacing.  Let's consider HDMI 1920x1080i.
  * The timing parameters we have from X are:
@@ -1080,6 +1097,7 @@ static int armada_drm_crtc_create(struct drm_device *drm, struct device *dev,
 {
 	struct armada_private *priv = drm->dev_private;
 	struct armada_crtc *dcrtc;
+	struct drm_plane *primary;
 	void __iomem *base;
 	int ret;
 
@@ -1148,7 +1166,17 @@ static int armada_drm_crtc_create(struct drm_device *drm, struct device *dev,
 	priv->dcrtc[dcrtc->num] = dcrtc;
 
 	dcrtc->crtc.port = port;
-	drm_crtc_init(drm, &dcrtc->crtc, &armada_crtc_funcs);
+
+	primary = drm_primary_helper_create_plane(drm, armada_primary_formats,
+					ARRAY_SIZE(armada_primary_formats));
+	if (!primary)
+		return -ENOMEM;
+
+	ret = drm_crtc_init_with_planes(drm, &dcrtc->crtc, primary, NULL,
+					&armada_crtc_funcs);
+	if (ret)
+		goto err_crtc_init;
+
 	drm_crtc_helper_add(&dcrtc->crtc, &armada_crtc_helper_funcs);
 
 	drm_object_attach_property(&dcrtc->crtc.base, priv->csc_yuv_prop,
@@ -1157,6 +1185,10 @@ static int armada_drm_crtc_create(struct drm_device *drm, struct device *dev,
 				   dcrtc->csc_rgb_mode);
 
 	return armada_overlay_plane_create(drm, 1 << dcrtc->num);
+
+err_crtc_init:
+	primary->funcs->destroy(primary);
+	return ret;
 }
 
 static int
-- 
cgit v0.10.2


From de32301b86030b20a51151a12d81fff6429cad0c Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Wed, 15 Jul 2015 18:11:24 +0100
Subject: drm/armada: allocate primary plane ourselves

Allocate our own primary plane as an armada_plane.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/armada/armada_crtc.c b/drivers/gpu/drm/armada/armada_crtc.c
index b96b77b..f146fcf 100644
--- a/drivers/gpu/drm/armada/armada_crtc.c
+++ b/drivers/gpu/drm/armada/armada_crtc.c
@@ -1059,6 +1059,12 @@ static struct drm_crtc_funcs armada_crtc_funcs = {
 	.set_property	= armada_drm_crtc_set_property,
 };
 
+static const struct drm_plane_funcs armada_primary_plane_funcs = {
+	.update_plane	= drm_primary_helper_update,
+	.disable_plane	= drm_primary_helper_disable,
+	.destroy	= drm_primary_helper_destroy,
+};
+
 static struct drm_prop_enum_list armada_drm_csc_yuv_enum_list[] = {
 	{ CSC_AUTO,        "Auto" },
 	{ CSC_YUV_CCIR601, "CCIR601" },
@@ -1097,7 +1103,7 @@ static int armada_drm_crtc_create(struct drm_device *drm, struct device *dev,
 {
 	struct armada_private *priv = drm->dev_private;
 	struct armada_crtc *dcrtc;
-	struct drm_plane *primary;
+	struct armada_plane *primary;
 	void __iomem *base;
 	int ret;
 
@@ -1167,12 +1173,21 @@ static int armada_drm_crtc_create(struct drm_device *drm, struct device *dev,
 
 	dcrtc->crtc.port = port;
 
-	primary = drm_primary_helper_create_plane(drm, armada_primary_formats,
-					ARRAY_SIZE(armada_primary_formats));
+	primary = kzalloc(sizeof(*primary), GFP_KERNEL);
 	if (!primary)
 		return -ENOMEM;
 
-	ret = drm_crtc_init_with_planes(drm, &dcrtc->crtc, primary, NULL,
+	ret = drm_universal_plane_init(drm, &primary->base, 0,
+				       &armada_primary_plane_funcs,
+				       armada_primary_formats,
+				       ARRAY_SIZE(armada_primary_formats),
+				       DRM_PLANE_TYPE_PRIMARY);
+	if (ret) {
+		kfree(primary);
+		return ret;
+	}
+
+	ret = drm_crtc_init_with_planes(drm, &dcrtc->crtc, &primary->base, NULL,
 					&armada_crtc_funcs);
 	if (ret)
 		goto err_crtc_init;
@@ -1187,7 +1202,7 @@ static int armada_drm_crtc_create(struct drm_device *drm, struct device *dev,
 	return armada_overlay_plane_create(drm, 1 << dcrtc->num);
 
 err_crtc_init:
-	primary->funcs->destroy(primary);
+	primary->base.funcs->destroy(&primary->base);
 	return ret;
 }
 
-- 
cgit v0.10.2


From 583268035825fc5ef0bbc467631fea0358831cbe Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Wed, 15 Jul 2015 18:11:25 +0100
Subject: drm/armada: provide a common helper to disable a plane

Provide a common helper to disable either the overlay or the primary
plane.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/armada/armada_crtc.c b/drivers/gpu/drm/armada/armada_crtc.c
index f146fcf..007fc5d 100644
--- a/drivers/gpu/drm/armada/armada_crtc.c
+++ b/drivers/gpu/drm/armada/armada_crtc.c
@@ -700,18 +700,39 @@ static int armada_drm_crtc_mode_set_base(struct drm_crtc *crtc, int x, int y,
 	return 0;
 }
 
+void armada_drm_crtc_plane_disable(struct armada_crtc *dcrtc,
+	struct drm_plane *plane)
+{
+	u32 sram_para1;
+
+	/*
+	 * Drop our reference on any framebuffer attached to this plane.
+	 * We don't need to NULL this out as drm_plane_force_disable(),
+	 * and __setplane_internal() will do so for an overlay plane, and
+	 * __drm_helper_disable_unused_functions() will do so for the
+	 * primary plane.
+	 */
+	if (plane->fb)
+		drm_framebuffer_unreference(plane->fb);
+
+	/* Power down the Y/U/V FIFOs */
+	sram_para1 = CFG_PDWN16x66 | CFG_PDWN32x66;
+
+	/* Power down most RAMs and FIFOs if this is the primary plane */
+	if (plane->type == DRM_PLANE_TYPE_PRIMARY)
+		sram_para1 |= CFG_PDWN256x32 | CFG_PDWN256x24 | CFG_PDWN256x8 |
+			      CFG_PDWN32x32 | CFG_PDWN64x66;
+
+	armada_updatel(sram_para1, 0, dcrtc->base + LCD_SPU_SRAM_PARA1);
+}
+
 /* The mode_config.mutex will be held for this call */
 static void armada_drm_crtc_disable(struct drm_crtc *crtc)
 {
 	struct armada_crtc *dcrtc = drm_to_armada_crtc(crtc);
 
 	armada_drm_crtc_dpms(crtc, DRM_MODE_DPMS_OFF);
-	armada_drm_crtc_finish_fb(dcrtc, crtc->primary->fb, true);
-
-	/* Power down most RAMs and FIFOs */
-	writel_relaxed(CFG_PDWN256x32 | CFG_PDWN256x24 | CFG_PDWN256x8 |
-		       CFG_PDWN32x32 | CFG_PDWN16x66 | CFG_PDWN32x66 |
-		       CFG_PDWN64x66, dcrtc->base + LCD_SPU_SRAM_PARA1);
+	armada_drm_crtc_plane_disable(dcrtc, crtc->primary);
 }
 
 static const struct drm_crtc_helper_funcs armada_crtc_helper_funcs = {
diff --git a/drivers/gpu/drm/armada/armada_crtc.h b/drivers/gpu/drm/armada/armada_crtc.h
index 549b5f5..500ce0f 100644
--- a/drivers/gpu/drm/armada/armada_crtc.h
+++ b/drivers/gpu/drm/armada/armada_crtc.h
@@ -103,6 +103,9 @@ void armada_drm_crtc_disable_irq(struct armada_crtc *, u32);
 void armada_drm_crtc_enable_irq(struct armada_crtc *, u32);
 void armada_drm_crtc_update_regs(struct armada_crtc *, struct armada_regs *);
 
+void armada_drm_crtc_plane_disable(struct armada_crtc *dcrtc,
+	struct drm_plane *plane);
+
 extern struct platform_driver armada_lcd_platform_driver;
 
 #endif
diff --git a/drivers/gpu/drm/armada/armada_overlay.c b/drivers/gpu/drm/armada/armada_overlay.c
index e5a5b73..1032f9b 100644
--- a/drivers/gpu/drm/armada/armada_overlay.c
+++ b/drivers/gpu/drm/armada/armada_overlay.c
@@ -279,12 +279,7 @@ static int armada_ovl_plane_disable(struct drm_plane *plane)
 	dplane->ctrl0 = 0;
 	spin_unlock_irq(&dcrtc->irq_lock);
 
-	/* Power down the Y/U/V FIFOs */
-	armada_updatel(CFG_PDWN16x66 | CFG_PDWN32x66, 0,
-		       dcrtc->base + LCD_SPU_SRAM_PARA1);
-
-	if (plane->fb)
-		drm_framebuffer_unreference(plane->fb);
+	armada_drm_crtc_plane_disable(dcrtc, plane);
 
 	fb = xchg(&dplane->old_fb, NULL);
 	if (fb)
-- 
cgit v0.10.2


From 9099ea19ca8ad21ab7f2a7abc06e270adeb8b02f Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Wed, 15 Jul 2015 18:11:25 +0100
Subject: drm/armada: move write to dma_ctrl0 to
 armada_drm_crtc_plane_disable()

Move the write to clear the DMA enable bit, and augment it with clearing
the graphics enable bit for the primary plane.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/armada/armada_crtc.c b/drivers/gpu/drm/armada/armada_crtc.c
index 007fc5d..89decc5 100644
--- a/drivers/gpu/drm/armada/armada_crtc.c
+++ b/drivers/gpu/drm/armada/armada_crtc.c
@@ -703,7 +703,7 @@ static int armada_drm_crtc_mode_set_base(struct drm_crtc *crtc, int x, int y,
 void armada_drm_crtc_plane_disable(struct armada_crtc *dcrtc,
 	struct drm_plane *plane)
 {
-	u32 sram_para1;
+	u32 sram_para1, dma_ctrl0_mask;
 
 	/*
 	 * Drop our reference on any framebuffer attached to this plane.
@@ -719,9 +719,17 @@ void armada_drm_crtc_plane_disable(struct armada_crtc *dcrtc,
 	sram_para1 = CFG_PDWN16x66 | CFG_PDWN32x66;
 
 	/* Power down most RAMs and FIFOs if this is the primary plane */
-	if (plane->type == DRM_PLANE_TYPE_PRIMARY)
+	if (plane->type == DRM_PLANE_TYPE_PRIMARY) {
 		sram_para1 |= CFG_PDWN256x32 | CFG_PDWN256x24 | CFG_PDWN256x8 |
 			      CFG_PDWN32x32 | CFG_PDWN64x66;
+		dma_ctrl0_mask = CFG_GRA_ENA;
+	} else {
+		dma_ctrl0_mask = CFG_DMA_ENA;
+	}
+
+	spin_lock_irq(&dcrtc->irq_lock);
+	armada_updatel(0, dma_ctrl0_mask, dcrtc->base + LCD_SPU_DMA_CTRL0);
+	spin_unlock_irq(&dcrtc->irq_lock);
 
 	armada_updatel(sram_para1, 0, dcrtc->base + LCD_SPU_SRAM_PARA1);
 }
diff --git a/drivers/gpu/drm/armada/armada_overlay.c b/drivers/gpu/drm/armada/armada_overlay.c
index 1032f9b..9686d79 100644
--- a/drivers/gpu/drm/armada/armada_overlay.c
+++ b/drivers/gpu/drm/armada/armada_overlay.c
@@ -275,7 +275,6 @@ static int armada_ovl_plane_disable(struct drm_plane *plane)
 
 	spin_lock_irq(&dcrtc->irq_lock);
 	armada_drm_vbl_event_remove(dcrtc, &dplane->vbl.update);
-	armada_updatel(0, CFG_DMA_ENA, dcrtc->base + LCD_SPU_DMA_CTRL0);
 	dplane->ctrl0 = 0;
 	spin_unlock_irq(&dcrtc->irq_lock);
 
-- 
cgit v0.10.2


From 5c8752c6506abf29950d32366f826899dc87dde7 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Wed, 15 Jul 2015 18:11:25 +0100
Subject: drm/armada: move the update of dplane->ctrl0 out of spinlock

It is not necessary to write dplane->ctrl0 under the CRTC spinlock, as
this is only accessed under process context where the DRM locks will
protect us instead.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/armada/armada_overlay.c b/drivers/gpu/drm/armada/armada_overlay.c
index 9686d79..e7e020d 100644
--- a/drivers/gpu/drm/armada/armada_overlay.c
+++ b/drivers/gpu/drm/armada/armada_overlay.c
@@ -275,9 +275,10 @@ static int armada_ovl_plane_disable(struct drm_plane *plane)
 
 	spin_lock_irq(&dcrtc->irq_lock);
 	armada_drm_vbl_event_remove(dcrtc, &dplane->vbl.update);
-	dplane->ctrl0 = 0;
 	spin_unlock_irq(&dcrtc->irq_lock);
 
+	dplane->ctrl0 = 0;
+
 	armada_drm_crtc_plane_disable(dcrtc, plane);
 
 	fb = xchg(&dplane->old_fb, NULL);
-- 
cgit v0.10.2


From 6908cf755af74b38d67195ee6607976a55f53d95 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Wed, 15 Jul 2015 18:11:25 +0100
Subject: drm/armada: move the locking for armada_drm_vbl_event_remove()

Move the locking for armada_drm_vbl_event_remove() into itself, which
makes this function symmetrical with armada_drm_vbl_event_add().

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/armada/armada_crtc.c b/drivers/gpu/drm/armada/armada_crtc.c
index 89decc5..e3e6f81 100644
--- a/drivers/gpu/drm/armada/armada_crtc.c
+++ b/drivers/gpu/drm/armada/armada_crtc.c
@@ -211,10 +211,12 @@ void armada_drm_vbl_event_add(struct armada_crtc *dcrtc,
 void armada_drm_vbl_event_remove(struct armada_crtc *dcrtc,
 	struct armada_vbl_event *evt)
 {
+	spin_lock_irq(&dcrtc->irq_lock);
 	if (!list_empty(&evt->node)) {
 		list_del_init(&evt->node);
 		drm_vblank_put(dcrtc->crtc.dev, dcrtc->num);
 	}
+	spin_unlock_irq(&dcrtc->irq_lock);
 }
 
 static void armada_drm_vbl_event_run(struct armada_crtc *dcrtc)
diff --git a/drivers/gpu/drm/armada/armada_overlay.c b/drivers/gpu/drm/armada/armada_overlay.c
index e7e020d..6ec42eb 100644
--- a/drivers/gpu/drm/armada/armada_overlay.c
+++ b/drivers/gpu/drm/armada/armada_overlay.c
@@ -273,9 +273,7 @@ static int armada_ovl_plane_disable(struct drm_plane *plane)
 	dcrtc = drm_to_armada_crtc(dplane->base.base.crtc);
 	dcrtc->plane = NULL;
 
-	spin_lock_irq(&dcrtc->irq_lock);
 	armada_drm_vbl_event_remove(dcrtc, &dplane->vbl.update);
-	spin_unlock_irq(&dcrtc->irq_lock);
 
 	dplane->ctrl0 = 0;
 
-- 
cgit v0.10.2


From 5740d27fa5594344ed4d2c18d7ae7bea69002004 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Wed, 15 Jul 2015 18:11:25 +0100
Subject: drm/armada: move frame wait into armada_frame

Both the CRTC and overlay frames have their own wait queues.  It would
make more sense if these were part of the plane - the primary plane for
the CRTC and overlay plane for the overlay.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/armada/armada_crtc.c b/drivers/gpu/drm/armada/armada_crtc.c
index e3e6f81..46d932b 100644
--- a/drivers/gpu/drm/armada/armada_crtc.c
+++ b/drivers/gpu/drm/armada/armada_crtc.c
@@ -455,7 +455,7 @@ static void armada_drm_crtc_irq(struct armada_crtc *dcrtc, u32 stat)
 		if (work)
 			armada_drm_crtc_complete_frame_work(dcrtc, work);
 
-		wake_up(&dcrtc->frame_wait);
+		wake_up(&drm_to_armada_plane(dcrtc->crtc.primary)->frame_wait);
 	}
 }
 
@@ -571,7 +571,8 @@ static int armada_drm_crtc_mode_set(struct drm_crtc *crtc,
 		adj->crtc_vtotal, tm, bm);
 
 	/* Wait for pending flips to complete */
-	wait_event(dcrtc->frame_wait, !dcrtc->frame_work);
+	wait_event(drm_to_armada_plane(dcrtc->crtc.primary)->frame_wait,
+		   !dcrtc->frame_work);
 
 	drm_crtc_vblank_off(crtc);
 
@@ -688,7 +689,8 @@ static int armada_drm_crtc_mode_set_base(struct drm_crtc *crtc, int x, int y,
 	armada_reg_queue_end(regs, i);
 
 	/* Wait for pending flips to complete */
-	wait_event(dcrtc->frame_wait, !dcrtc->frame_work);
+	wait_event(drm_to_armada_plane(dcrtc->crtc.primary)->frame_wait,
+		   !dcrtc->frame_work);
 
 	/* Take a reference to the new fb as we're using it */
 	drm_framebuffer_reference(crtc->primary->fb);
@@ -1096,6 +1098,13 @@ static const struct drm_plane_funcs armada_primary_plane_funcs = {
 	.destroy	= drm_primary_helper_destroy,
 };
 
+int armada_drm_plane_init(struct armada_plane *plane)
+{
+	init_waitqueue_head(&plane->frame_wait);
+
+	return 0;
+}
+
 static struct drm_prop_enum_list armada_drm_csc_yuv_enum_list[] = {
 	{ CSC_AUTO,        "Auto" },
 	{ CSC_YUV_CCIR601, "CCIR601" },
@@ -1166,7 +1175,6 @@ static int armada_drm_crtc_create(struct drm_device *drm, struct device *dev,
 	spin_lock_init(&dcrtc->irq_lock);
 	dcrtc->irq_ena = CLEAN_SPU_IRQ_ISR;
 	INIT_LIST_HEAD(&dcrtc->vbl_list);
-	init_waitqueue_head(&dcrtc->frame_wait);
 
 	/* Initialize some registers which we don't otherwise set */
 	writel_relaxed(0x00000001, dcrtc->base + LCD_CFG_SCLK_DIV);
@@ -1208,6 +1216,12 @@ static int armada_drm_crtc_create(struct drm_device *drm, struct device *dev,
 	if (!primary)
 		return -ENOMEM;
 
+	ret = armada_drm_plane_init(primary);
+	if (ret) {
+		kfree(primary);
+		return ret;
+	}
+
 	ret = drm_universal_plane_init(drm, &primary->base, 0,
 				       &armada_primary_plane_funcs,
 				       armada_primary_formats,
diff --git a/drivers/gpu/drm/armada/armada_crtc.h b/drivers/gpu/drm/armada/armada_crtc.h
index 500ce0f..3ec5101 100644
--- a/drivers/gpu/drm/armada/armada_crtc.h
+++ b/drivers/gpu/drm/armada/armada_crtc.h
@@ -36,9 +36,12 @@ struct armada_variant;
 
 struct armada_plane {
 	struct drm_plane	base;
+	wait_queue_head_t	frame_wait;
 };
 #define drm_to_armada_plane(p) container_of(p, struct armada_plane, base)
 
+int armada_drm_plane_init(struct armada_plane *plane);
+
 struct armada_crtc {
 	struct drm_crtc		crtc;
 	const struct armada_variant *variant;
@@ -71,7 +74,6 @@ struct armada_crtc {
 	uint32_t		dumb_ctrl;
 	uint32_t		spu_iopad_ctrl;
 
-	wait_queue_head_t	frame_wait;
 	struct armada_frame_work *frame_work;
 
 	spinlock_t		irq_lock;
diff --git a/drivers/gpu/drm/armada/armada_overlay.c b/drivers/gpu/drm/armada/armada_overlay.c
index 6ec42eb..9a5bab7 100644
--- a/drivers/gpu/drm/armada/armada_overlay.c
+++ b/drivers/gpu/drm/armada/armada_overlay.c
@@ -39,7 +39,6 @@ struct armada_ovl_plane {
 	struct {
 		struct armada_vbl_event update;
 		struct armada_regs regs[13];
-		wait_queue_head_t wait;
 	} vbl;
 	struct armada_ovl_plane_properties prop;
 };
@@ -90,7 +89,7 @@ static void armada_ovl_plane_vbl(struct armada_crtc *dcrtc, void *data)
 	armada_drm_crtc_update_regs(dcrtc, dplane->vbl.regs);
 	armada_ovl_retire_fb(dplane, NULL);
 
-	wake_up(&dplane->vbl.wait);
+	wake_up(&dplane->base.frame_wait);
 }
 
 static int
@@ -163,7 +162,7 @@ armada_ovl_plane_update(struct drm_plane *plane, struct drm_crtc *crtc,
 			       dcrtc->base + LCD_SPU_SRAM_PARA1);
 	}
 
-	wait_event_timeout(dplane->vbl.wait,
+	wait_event_timeout(dplane->base.frame_wait,
 			   list_empty(&dplane->vbl.update.node),
 			   HZ/25);
 
@@ -451,7 +450,12 @@ int armada_overlay_plane_create(struct drm_device *dev, unsigned long crtcs)
 	if (!dplane)
 		return -ENOMEM;
 
-	init_waitqueue_head(&dplane->vbl.wait);
+	ret = armada_drm_plane_init(&dplane->base);
+	if (ret) {
+		kfree(dplane);
+		return ret;
+	}
+
 	armada_drm_vbl_event_init(&dplane->vbl.update, armada_ovl_plane_vbl,
 				  dplane);
 
-- 
cgit v0.10.2


From 4b5dda82c20c2eee500520010c0558789592d62f Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Thu, 6 Aug 2015 16:37:18 +0100
Subject: drm/armada: move CRTC flip work to primary plane work

Add a plane work implementation, and move the CRTC framebuffer flip
work to it for the primary plane.  The idea is to have a common
plane work implementation for both the primary and overlay planes.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/armada/armada_crtc.c b/drivers/gpu/drm/armada/armada_crtc.c
index 46d932b..0c1a152 100644
--- a/drivers/gpu/drm/armada/armada_crtc.c
+++ b/drivers/gpu/drm/armada/armada_crtc.c
@@ -20,6 +20,7 @@
 #include "armada_hw.h"
 
 struct armada_frame_work {
+	struct armada_plane_work work;
 	struct drm_pending_vblank_event *event;
 	struct armada_regs regs[4];
 	struct drm_framebuffer *old_fb;
@@ -190,6 +191,41 @@ static unsigned armada_drm_crtc_calc_fb(struct drm_framebuffer *fb,
 	return i;
 }
 
+static void armada_drm_plane_work_run(struct armada_crtc *dcrtc,
+	struct armada_plane *plane)
+{
+	struct armada_plane_work *work = xchg(&plane->work, NULL);
+
+	/* Handle any pending frame work. */
+	if (work) {
+		work->fn(dcrtc, plane, work);
+		drm_vblank_put(dcrtc->crtc.dev, dcrtc->num);
+	}
+}
+
+int armada_drm_plane_work_queue(struct armada_crtc *dcrtc,
+	struct armada_plane *plane, struct armada_plane_work *work)
+{
+	int ret;
+
+	ret = drm_vblank_get(dcrtc->crtc.dev, dcrtc->num);
+	if (ret) {
+		DRM_ERROR("failed to acquire vblank counter\n");
+		return ret;
+	}
+
+	ret = cmpxchg(&plane->work, NULL, work) ? -EBUSY : 0;
+	if (ret)
+		drm_vblank_put(dcrtc->crtc.dev, dcrtc->num);
+
+	return ret;
+}
+
+int armada_drm_plane_work_wait(struct armada_plane *plane, long timeout)
+{
+	return wait_event_timeout(plane->frame_wait, !plane->work, timeout);
+}
+
 void armada_drm_vbl_event_add(struct armada_crtc *dcrtc,
 	struct armada_vbl_event *evt)
 {
@@ -233,44 +269,31 @@ static void armada_drm_vbl_event_run(struct armada_crtc *dcrtc)
 static int armada_drm_crtc_queue_frame_work(struct armada_crtc *dcrtc,
 	struct armada_frame_work *work)
 {
-	struct drm_device *dev = dcrtc->crtc.dev;
-	int ret;
+	struct armada_plane *plane = drm_to_armada_plane(dcrtc->crtc.primary);
 
-	ret = drm_vblank_get(dev, dcrtc->num);
-	if (ret) {
-		DRM_ERROR("failed to acquire vblank counter\n");
-		return ret;
-	}
-
-	if (cmpxchg(&dcrtc->frame_work, NULL, work)) {
-		drm_vblank_put(dev, dcrtc->num);
-		ret = -EBUSY;
-	}
-
-	return ret;
+	return armada_drm_plane_work_queue(dcrtc, plane, &work->work);
 }
 
 static void armada_drm_crtc_complete_frame_work(struct armada_crtc *dcrtc,
-	struct armada_frame_work *work)
+	struct armada_plane *plane, struct armada_plane_work *work)
 {
+	struct armada_frame_work *fwork = container_of(work, struct armada_frame_work, work);
 	struct drm_device *dev = dcrtc->crtc.dev;
 	unsigned long flags;
 
 	spin_lock_irqsave(&dcrtc->irq_lock, flags);
-	armada_drm_crtc_update_regs(dcrtc, work->regs);
+	armada_drm_crtc_update_regs(dcrtc, fwork->regs);
 	spin_unlock_irqrestore(&dcrtc->irq_lock, flags);
 
-	if (work->event) {
+	if (fwork->event) {
 		spin_lock_irqsave(&dev->event_lock, flags);
-		drm_send_vblank_event(dev, dcrtc->num, work->event);
+		drm_send_vblank_event(dev, dcrtc->num, fwork->event);
 		spin_unlock_irqrestore(&dev->event_lock, flags);
 	}
 
-	drm_vblank_put(dev, dcrtc->num);
-
 	/* Finally, queue the process-half of the cleanup. */
-	__armada_drm_queue_unref_work(dcrtc->crtc.dev, work->old_fb);
-	kfree(work);
+	__armada_drm_queue_unref_work(dcrtc->crtc.dev, fwork->old_fb);
+	kfree(fwork);
 }
 
 static void armada_drm_crtc_finish_fb(struct armada_crtc *dcrtc,
@@ -290,6 +313,7 @@ static void armada_drm_crtc_finish_fb(struct armada_crtc *dcrtc,
 	work = kmalloc(sizeof(*work), GFP_KERNEL);
 	if (work) {
 		int i = 0;
+		work->work.fn = armada_drm_crtc_complete_frame_work;
 		work->event = NULL;
 		work->old_fb = fb;
 		armada_reg_queue_end(work->regs, i);
@@ -310,18 +334,14 @@ static void armada_drm_crtc_finish_fb(struct armada_crtc *dcrtc,
 
 static void armada_drm_vblank_off(struct armada_crtc *dcrtc)
 {
-	struct armada_frame_work *work;
+	struct armada_plane *plane = drm_to_armada_plane(dcrtc->crtc.primary);
 
 	/*
 	 * Tell the DRM core that vblank IRQs aren't going to happen for
 	 * a while.  This cleans up any pending vblank events for us.
 	 */
 	drm_crtc_vblank_off(&dcrtc->crtc);
-
-	/* Handle any pending flip event. */
-	work = xchg(&dcrtc->frame_work, NULL);
-	if (work)
-		armada_drm_crtc_complete_frame_work(dcrtc, work);
+	armada_drm_plane_work_run(dcrtc, plane);
 }
 
 void armada_drm_crtc_gamma_set(struct drm_crtc *crtc, u16 r, u16 g, u16 b,
@@ -450,12 +470,9 @@ static void armada_drm_crtc_irq(struct armada_crtc *dcrtc, u32 stat)
 	spin_unlock(&dcrtc->irq_lock);
 
 	if (stat & GRA_FRAME_IRQ) {
-		struct armada_frame_work *work = xchg(&dcrtc->frame_work, NULL);
-
-		if (work)
-			armada_drm_crtc_complete_frame_work(dcrtc, work);
-
-		wake_up(&drm_to_armada_plane(dcrtc->crtc.primary)->frame_wait);
+		struct armada_plane *plane = drm_to_armada_plane(dcrtc->crtc.primary);
+		armada_drm_plane_work_run(dcrtc, plane);
+		wake_up(&plane->frame_wait);
 	}
 }
 
@@ -571,8 +588,8 @@ static int armada_drm_crtc_mode_set(struct drm_crtc *crtc,
 		adj->crtc_vtotal, tm, bm);
 
 	/* Wait for pending flips to complete */
-	wait_event(drm_to_armada_plane(dcrtc->crtc.primary)->frame_wait,
-		   !dcrtc->frame_work);
+	armada_drm_plane_work_wait(drm_to_armada_plane(dcrtc->crtc.primary),
+				   MAX_SCHEDULE_TIMEOUT);
 
 	drm_crtc_vblank_off(crtc);
 
@@ -689,8 +706,8 @@ static int armada_drm_crtc_mode_set_base(struct drm_crtc *crtc, int x, int y,
 	armada_reg_queue_end(regs, i);
 
 	/* Wait for pending flips to complete */
-	wait_event(drm_to_armada_plane(dcrtc->crtc.primary)->frame_wait,
-		   !dcrtc->frame_work);
+	armada_drm_plane_work_wait(drm_to_armada_plane(dcrtc->crtc.primary),
+				   MAX_SCHEDULE_TIMEOUT);
 
 	/* Take a reference to the new fb as we're using it */
 	drm_framebuffer_reference(crtc->primary->fb);
@@ -1013,6 +1030,7 @@ static int armada_drm_crtc_page_flip(struct drm_crtc *crtc,
 	if (!work)
 		return -ENOMEM;
 
+	work->work.fn = armada_drm_crtc_complete_frame_work;
 	work->event = event;
 	work->old_fb = dcrtc->crtc.primary->fb;
 
@@ -1046,12 +1064,8 @@ static int armada_drm_crtc_page_flip(struct drm_crtc *crtc,
 	 * Finally, if the display is blanked, we won't receive an
 	 * interrupt, so complete it now.
 	 */
-	if (dpms_blanked(dcrtc->dpms)) {
-		struct armada_frame_work *work = xchg(&dcrtc->frame_work, NULL);
-
-		if (work)
-			armada_drm_crtc_complete_frame_work(dcrtc, work);
-	}
+	if (dpms_blanked(dcrtc->dpms))
+		armada_drm_plane_work_run(dcrtc, drm_to_armada_plane(dcrtc->crtc.primary));
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/armada/armada_crtc.h b/drivers/gpu/drm/armada/armada_crtc.h
index 3ec5101..aaad5ab 100644
--- a/drivers/gpu/drm/armada/armada_crtc.h
+++ b/drivers/gpu/drm/armada/armada_crtc.h
@@ -31,16 +31,27 @@ struct armada_regs {
 #define armada_reg_queue_end(_r, _i)		\
 	armada_reg_queue_mod(_r, _i, 0, 0, ~0)
 
-struct armada_frame_work;
+struct armada_crtc;
+struct armada_plane;
 struct armada_variant;
 
+struct armada_plane_work {
+	void			(*fn)(struct armada_crtc *,
+				      struct armada_plane *,
+				      struct armada_plane_work *);
+};
+
 struct armada_plane {
 	struct drm_plane	base;
 	wait_queue_head_t	frame_wait;
+	struct armada_plane_work *work;
 };
 #define drm_to_armada_plane(p) container_of(p, struct armada_plane, base)
 
 int armada_drm_plane_init(struct armada_plane *plane);
+int armada_drm_plane_work_queue(struct armada_crtc *dcrtc,
+	struct armada_plane *plane, struct armada_plane_work *work);
+int armada_drm_plane_work_wait(struct armada_plane *plane, long timeout);
 
 struct armada_crtc {
 	struct drm_crtc		crtc;
@@ -74,8 +85,6 @@ struct armada_crtc {
 	uint32_t		dumb_ctrl;
 	uint32_t		spu_iopad_ctrl;
 
-	struct armada_frame_work *frame_work;
-
 	spinlock_t		irq_lock;
 	uint32_t		irq_ena;
 	struct list_head	vbl_list;
-- 
cgit v0.10.2


From 4a8506d2d68724b6d326621118874c07095c6645 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Fri, 7 Aug 2015 09:33:05 +0100
Subject: drm/armada: convert overlay plane vbl worker to a armada plane worker

Convert the overlay plane to use the generic armada plane worker
infrastructure which is shared with the primary plane.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/armada/armada_crtc.c b/drivers/gpu/drm/armada/armada_crtc.c
index 0c1a152..418dbfa 100644
--- a/drivers/gpu/drm/armada/armada_crtc.c
+++ b/drivers/gpu/drm/armada/armada_crtc.c
@@ -226,44 +226,15 @@ int armada_drm_plane_work_wait(struct armada_plane *plane, long timeout)
 	return wait_event_timeout(plane->frame_wait, !plane->work, timeout);
 }
 
-void armada_drm_vbl_event_add(struct armada_crtc *dcrtc,
-	struct armada_vbl_event *evt)
+struct armada_plane_work *armada_drm_plane_work_cancel(
+	struct armada_crtc *dcrtc, struct armada_plane *plane)
 {
-	unsigned long flags;
-	bool not_on_list;
-
-	WARN_ON(drm_vblank_get(dcrtc->crtc.dev, dcrtc->num));
-
-	spin_lock_irqsave(&dcrtc->irq_lock, flags);
-	not_on_list = list_empty(&evt->node);
-	if (not_on_list)
-		list_add_tail(&evt->node, &dcrtc->vbl_list);
-	spin_unlock_irqrestore(&dcrtc->irq_lock, flags);
+	struct armada_plane_work *work = xchg(&plane->work, NULL);
 
-	if (!not_on_list)
+	if (work)
 		drm_vblank_put(dcrtc->crtc.dev, dcrtc->num);
-}
 
-void armada_drm_vbl_event_remove(struct armada_crtc *dcrtc,
-	struct armada_vbl_event *evt)
-{
-	spin_lock_irq(&dcrtc->irq_lock);
-	if (!list_empty(&evt->node)) {
-		list_del_init(&evt->node);
-		drm_vblank_put(dcrtc->crtc.dev, dcrtc->num);
-	}
-	spin_unlock_irq(&dcrtc->irq_lock);
-}
-
-static void armada_drm_vbl_event_run(struct armada_crtc *dcrtc)
-{
-	struct armada_vbl_event *e, *n;
-
-	list_for_each_entry_safe(e, n, &dcrtc->vbl_list, node) {
-		list_del_init(&e->node);
-		drm_vblank_put(dcrtc->crtc.dev, dcrtc->num);
-		e->fn(dcrtc, e->data);
-	}
+	return work;
 }
 
 static int armada_drm_crtc_queue_frame_work(struct armada_crtc *dcrtc,
@@ -429,6 +400,7 @@ static bool armada_drm_crtc_mode_fixup(struct drm_crtc *crtc,
 static void armada_drm_crtc_irq(struct armada_crtc *dcrtc, u32 stat)
 {
 	void __iomem *base = dcrtc->base;
+	struct drm_plane *ovl_plane;
 
 	if (stat & DMA_FF_UNDERFLOW)
 		DRM_ERROR("video underflow on crtc %u\n", dcrtc->num);
@@ -439,7 +411,12 @@ static void armada_drm_crtc_irq(struct armada_crtc *dcrtc, u32 stat)
 		drm_handle_vblank(dcrtc->crtc.dev, dcrtc->num);
 
 	spin_lock(&dcrtc->irq_lock);
-	armada_drm_vbl_event_run(dcrtc);
+	ovl_plane = dcrtc->plane;
+	if (ovl_plane) {
+		struct armada_plane *plane = drm_to_armada_plane(ovl_plane);
+		armada_drm_plane_work_run(dcrtc, plane);
+		wake_up(&plane->frame_wait);
+	}
 
 	if (stat & GRA_FRAME_IRQ && dcrtc->interlaced) {
 		int i = stat & GRA_FRAME_IRQ0 ? 0 : 1;
@@ -1188,7 +1165,6 @@ static int armada_drm_crtc_create(struct drm_device *drm, struct device *dev,
 	dcrtc->spu_iopad_ctrl = CFG_VSCALE_LN_EN | CFG_IOPAD_DUMB24;
 	spin_lock_init(&dcrtc->irq_lock);
 	dcrtc->irq_ena = CLEAN_SPU_IRQ_ISR;
-	INIT_LIST_HEAD(&dcrtc->vbl_list);
 
 	/* Initialize some registers which we don't otherwise set */
 	writel_relaxed(0x00000001, dcrtc->base + LCD_CFG_SCLK_DIV);
diff --git a/drivers/gpu/drm/armada/armada_crtc.h b/drivers/gpu/drm/armada/armada_crtc.h
index aaad5ab..04fdd22 100644
--- a/drivers/gpu/drm/armada/armada_crtc.h
+++ b/drivers/gpu/drm/armada/armada_crtc.h
@@ -52,6 +52,8 @@ int armada_drm_plane_init(struct armada_plane *plane);
 int armada_drm_plane_work_queue(struct armada_crtc *dcrtc,
 	struct armada_plane *plane, struct armada_plane_work *work);
 int armada_drm_plane_work_wait(struct armada_plane *plane, long timeout);
+struct armada_plane_work *armada_drm_plane_work_cancel(
+	struct armada_crtc *dcrtc, struct armada_plane *plane);
 
 struct armada_crtc {
 	struct drm_crtc		crtc;
@@ -87,27 +89,9 @@ struct armada_crtc {
 
 	spinlock_t		irq_lock;
 	uint32_t		irq_ena;
-	struct list_head	vbl_list;
 };
 #define drm_to_armada_crtc(c) container_of(c, struct armada_crtc, crtc)
 
-struct armada_vbl_event {
-	struct list_head	node;
-	void			*data;
-	void			(*fn)(struct armada_crtc *, void *);
-};
-
-void armada_drm_vbl_event_add(struct armada_crtc *,
-	struct armada_vbl_event *);
-void armada_drm_vbl_event_remove(struct armada_crtc *,
-	struct armada_vbl_event *);
-#define armada_drm_vbl_event_init(_e, _f, _d) do {	\
-	struct armada_vbl_event *__e = _e;		\
-	INIT_LIST_HEAD(&__e->node);			\
-	__e->data = _d;					\
-	__e->fn = _f;					\
-} while (0)
-
 void armada_drm_crtc_gamma_set(struct drm_crtc *, u16, u16, u16, int);
 void armada_drm_crtc_gamma_get(struct drm_crtc *, u16 *, u16 *, u16 *, int);
 void armada_drm_crtc_disable_irq(struct armada_crtc *, u32);
diff --git a/drivers/gpu/drm/armada/armada_overlay.c b/drivers/gpu/drm/armada/armada_overlay.c
index 9a5bab7..5c22b38 100644
--- a/drivers/gpu/drm/armada/armada_overlay.c
+++ b/drivers/gpu/drm/armada/armada_overlay.c
@@ -37,7 +37,7 @@ struct armada_ovl_plane {
 	uint32_t dst_yx;
 	uint32_t ctrl0;
 	struct {
-		struct armada_vbl_event update;
+		struct armada_plane_work work;
 		struct armada_regs regs[13];
 	} vbl;
 	struct armada_ovl_plane_properties prop;
@@ -82,14 +82,13 @@ static void armada_ovl_retire_fb(struct armada_ovl_plane *dplane,
 }
 
 /* === Plane support === */
-static void armada_ovl_plane_vbl(struct armada_crtc *dcrtc, void *data)
+static void armada_ovl_plane_work(struct armada_crtc *dcrtc,
+	struct armada_plane *plane, struct armada_plane_work *work)
 {
-	struct armada_ovl_plane *dplane = data;
+	struct armada_ovl_plane *dplane = container_of(plane, struct armada_ovl_plane, base);
 
 	armada_drm_crtc_update_regs(dcrtc, dplane->vbl.regs);
 	armada_ovl_retire_fb(dplane, NULL);
-
-	wake_up(&dplane->base.frame_wait);
 }
 
 static int
@@ -162,9 +161,8 @@ armada_ovl_plane_update(struct drm_plane *plane, struct drm_crtc *crtc,
 			       dcrtc->base + LCD_SPU_SRAM_PARA1);
 	}
 
-	wait_event_timeout(dplane->base.frame_wait,
-			   list_empty(&dplane->vbl.update.node),
-			   HZ/25);
+	if (armada_drm_plane_work_wait(&dplane->base, HZ / 25) == 0)
+		armada_drm_plane_work_cancel(dcrtc, &dplane->base);
 
 	if (plane->fb != fb) {
 		struct armada_gem_object *obj = drm_fb_obj(fb);
@@ -255,7 +253,8 @@ armada_ovl_plane_update(struct drm_plane *plane, struct drm_crtc *crtc,
 	}
 	if (idx) {
 		armada_reg_queue_end(dplane->vbl.regs, idx);
-		armada_drm_vbl_event_add(dcrtc, &dplane->vbl.update);
+		armada_drm_plane_work_queue(dcrtc, &dplane->base,
+					    &dplane->vbl.work);
 	}
 	return 0;
 }
@@ -270,14 +269,13 @@ static int armada_ovl_plane_disable(struct drm_plane *plane)
 		return 0;
 
 	dcrtc = drm_to_armada_crtc(dplane->base.base.crtc);
-	dcrtc->plane = NULL;
 
-	armada_drm_vbl_event_remove(dcrtc, &dplane->vbl.update);
+	armada_drm_plane_work_cancel(dcrtc, &dplane->base);
+	armada_drm_crtc_plane_disable(dcrtc, plane);
 
+	dcrtc->plane = NULL;
 	dplane->ctrl0 = 0;
 
-	armada_drm_crtc_plane_disable(dcrtc, plane);
-
 	fb = xchg(&dplane->old_fb, NULL);
 	if (fb)
 		drm_framebuffer_unreference(fb);
@@ -456,8 +454,7 @@ int armada_overlay_plane_create(struct drm_device *dev, unsigned long crtcs)
 		return ret;
 	}
 
-	armada_drm_vbl_event_init(&dplane->vbl.update, armada_ovl_plane_vbl,
-				  dplane);
+	dplane->vbl.work.fn = armada_ovl_plane_work;
 
 	ret = drm_universal_plane_init(dev, &dplane->base.base, crtcs,
 				       &armada_ovl_plane_funcs,
-- 
cgit v0.10.2


From 7cb410cdbdc9b71e2d4f356f2e4cea0b925abb21 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Fri, 7 Aug 2015 13:34:26 +0100
Subject: drm/armada: move frame wait wakeup into plane work

Move the wakeup for the frame wait into the armada plane work, to
ensure that it is woken up every time we run a work.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/armada/armada_crtc.c b/drivers/gpu/drm/armada/armada_crtc.c
index 418dbfa..cebcab5 100644
--- a/drivers/gpu/drm/armada/armada_crtc.c
+++ b/drivers/gpu/drm/armada/armada_crtc.c
@@ -201,6 +201,8 @@ static void armada_drm_plane_work_run(struct armada_crtc *dcrtc,
 		work->fn(dcrtc, plane, work);
 		drm_vblank_put(dcrtc->crtc.dev, dcrtc->num);
 	}
+
+	wake_up(&plane->frame_wait);
 }
 
 int armada_drm_plane_work_queue(struct armada_crtc *dcrtc,
@@ -415,7 +417,6 @@ static void armada_drm_crtc_irq(struct armada_crtc *dcrtc, u32 stat)
 	if (ovl_plane) {
 		struct armada_plane *plane = drm_to_armada_plane(ovl_plane);
 		armada_drm_plane_work_run(dcrtc, plane);
-		wake_up(&plane->frame_wait);
 	}
 
 	if (stat & GRA_FRAME_IRQ && dcrtc->interlaced) {
@@ -449,7 +450,6 @@ static void armada_drm_crtc_irq(struct armada_crtc *dcrtc, u32 stat)
 	if (stat & GRA_FRAME_IRQ) {
 		struct armada_plane *plane = drm_to_armada_plane(dcrtc->crtc.primary);
 		armada_drm_plane_work_run(dcrtc, plane);
-		wake_up(&plane->frame_wait);
 	}
 }
 
-- 
cgit v0.10.2


From 923c124107840d58c5002cc19734310743501811 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 30 Sep 2015 17:06:43 +0300
Subject: drm/i915: s/GET_CFG_CR1_REG/DPLL_CFGCR1/ etc.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

v2: Use SKL_DPLLx symbolic names instead of raw numbers

Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 0b2e314..9e3fdb3 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -7438,8 +7438,8 @@ enum skl_disp_power_wells {
 #define  DPLL_CFGCR2_PDIV_7 (4<<2)
 #define  DPLL_CFGCR2_CENTRAL_FREQ_MASK	(3)
 
-#define GET_CFG_CR1_REG(id) (DPLL1_CFGCR1 + (id - SKL_DPLL1) * 8)
-#define GET_CFG_CR2_REG(id) (DPLL1_CFGCR2 + (id - SKL_DPLL1) * 8)
+#define DPLL_CFGCR1(id) (DPLL1_CFGCR1 + ((id) - SKL_DPLL1) * 8)
+#define DPLL_CFGCR2(id) (DPLL1_CFGCR2 + ((id) - SKL_DPLL1) * 8)
 
 /* BXT display engine PLL */
 #define BXT_DE_PLL_CTL			0x6d000
diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
index 9c11577..2d3cc82 100644
--- a/drivers/gpu/drm/i915/intel_ddi.c
+++ b/drivers/gpu/drm/i915/intel_ddi.c
@@ -969,8 +969,8 @@ static int skl_calc_wrpll_link(struct drm_i915_private *dev_priv,
 	uint32_t cfgcr1_val, cfgcr2_val;
 	uint32_t p0, p1, p2, dco_freq;
 
-	cfgcr1_reg = GET_CFG_CR1_REG(dpll);
-	cfgcr2_reg = GET_CFG_CR2_REG(dpll);
+	cfgcr1_reg = DPLL_CFGCR1(dpll);
+	cfgcr2_reg = DPLL_CFGCR2(dpll);
 
 	cfgcr1_val = I915_READ(cfgcr1_reg);
 	cfgcr2_val = I915_READ(cfgcr2_reg);
@@ -2504,20 +2504,20 @@ static const struct skl_dpll_regs skl_dpll_regs[3] = {
 	{
 		/* DPLL 1 */
 		.ctl = LCPLL2_CTL,
-		.cfgcr1 = DPLL1_CFGCR1,
-		.cfgcr2 = DPLL1_CFGCR2,
+		.cfgcr1 = DPLL_CFGCR1(SKL_DPLL1),
+		.cfgcr2 = DPLL_CFGCR2(SKL_DPLL1),
 	},
 	{
 		/* DPLL 2 */
 		.ctl = WRPLL_CTL1,
-		.cfgcr1 = DPLL2_CFGCR1,
-		.cfgcr2 = DPLL2_CFGCR2,
+		.cfgcr1 = DPLL_CFGCR1(SKL_DPLL2),
+		.cfgcr2 = DPLL_CFGCR2(SKL_DPLL2),
 	},
 	{
 		/* DPLL 3 */
 		.ctl = WRPLL_CTL2,
-		.cfgcr1 = DPLL3_CFGCR1,
-		.cfgcr2 = DPLL3_CFGCR2,
+		.cfgcr1 = DPLL_CFGCR1(SKL_DPLL3),
+		.cfgcr2 = DPLL_CFGCR2(SKL_DPLL3),
 	},
 };
 
-- 
cgit v0.10.2


From 9d91a1bfae7d12eb2c91006dc977d9bac9000a30 Mon Sep 17 00:00:00 2001
From: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Date: Wed, 30 Sep 2015 09:57:37 -0700
Subject: drm/i915/guc: Don't forward flip interrupts to GuC

Due to flip interrupts GuC stays awake always and GT does not enter
RC6. Do not route those interrupts to GuC for now. Driver won't touch
DE_GUCRMR register and leave it as what default value.

Signed-off-by: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Signed-off-by: Alex Dai <yu.dai@intel.com>
Reviewed-by: Tom O'Rourke <Tom.O'Rourke@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_guc_loader.c b/drivers/gpu/drm/i915/intel_guc_loader.c
index 5d17b63..f87d833 100644
--- a/drivers/gpu/drm/i915/intel_guc_loader.c
+++ b/drivers/gpu/drm/i915/intel_guc_loader.c
@@ -90,9 +90,6 @@ static void direct_interrupts_to_host(struct drm_i915_private *dev_priv)
 	for_each_ring(ring, dev_priv, i)
 		I915_WRITE(RING_MODE_GEN7(ring), irqs);
 
-	/* tell DE to send nothing to GuC */
-	I915_WRITE(DE_GUCRMR, ~0);
-
 	/* route all GT interrupts to the host */
 	I915_WRITE(GUC_BCS_RCS_IER, 0);
 	I915_WRITE(GUC_VCS2_VCS1_IER, 0);
@@ -110,13 +107,6 @@ static void direct_interrupts_to_guc(struct drm_i915_private *dev_priv)
 	for_each_ring(ring, dev_priv, i)
 		I915_WRITE(RING_MODE_GEN7(ring), irqs);
 
-	/* tell DE to send (all) flip_done to GuC */
-	irqs = DERRMR_PIPEA_PRI_FLIP_DONE | DERRMR_PIPEA_SPR_FLIP_DONE |
-	       DERRMR_PIPEB_PRI_FLIP_DONE | DERRMR_PIPEB_SPR_FLIP_DONE |
-	       DERRMR_PIPEC_PRI_FLIP_DONE | DERRMR_PIPEC_SPR_FLIP_DONE;
-	/* Unmasked bits will cause GuC response message to be sent */
-	I915_WRITE(DE_GUCRMR, ~irqs);
-
 	/* route USER_INTERRUPT to Host, all others are sent to GuC. */
 	irqs = GT_RENDER_USER_INTERRUPT << GEN8_RCS_IRQ_SHIFT |
 	       GT_RENDER_USER_INTERRUPT << GEN8_BCS_IRQ_SHIFT;
-- 
cgit v0.10.2


From 101b506a7fc7be3f0d0a337ade270eb5eb5a2857 Mon Sep 17 00:00:00 2001
From: Michel Thierry <michel.thierry@intel.com>
Date: Thu, 1 Oct 2015 13:33:57 +0100
Subject: drm/i915: Wa32bitGeneralStateOffset & Wa32bitInstructionBaseOffset

There are some allocations that must be only referenced by 32-bit
offsets. To limit the chances of having the first 4GB already full,
objects not requiring this workaround use DRM_MM_SEARCH_BELOW/
DRM_MM_CREATE_TOP flags

In specific, any resource used with flat/heapless (0x00000000-0xfffff000)
General State Heap (GSH) or Instruction State Heap (ISH) must be in a
32-bit range, because the General State Offset and Instruction State
Offset are limited to 32-bits.

Objects must have EXEC_OBJECT_SUPPORTS_48B_ADDRESS flag to indicate if
they can be allocated above the 32-bit address range. To limit the
chances of having the first 4GB already full, objects will use
DRM_MM_SEARCH_BELOW + DRM_MM_CREATE_TOP flags when possible.

The libdrm user of the EXEC_OBJECT_SUPPORTS_48B_ADDRESS flag is here:
http://lists.freedesktop.org/archives/intel-gfx/2015-September/075836.html

v2: Changed flag logic from neeeds_32b, to supports_48b.
v3: Moved 48-bit support flag back to exec_object. (Chris, Daniel)
v4: Split pin flags into PIN_ZONE_4G and PIN_HIGH; update PIN_OFFSET_MASK
to use last PIN_ defined instead of hard-coded value; use correct limit
check in eb_vma_misplaced. (Chris)
v5: Don't touch PIN_OFFSET_MASK and update workaround comment (Chris)
v6: Apply pin-high for ggtt too (Chris)
v7: Handle simultaneous pin-high and pin-mappable end correctly (Akash)
    Fix check for entries currently using +4GB addresses, use min_t and
    other polish in object_bind_to_vm (Chris)
v8: Commit message updated to point to libdrm patch.
v9: vmas are allocated in the correct ozone, so only check flag when the
    vma has not been allocated. (Chris)

Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> (v4)
Signed-off-by: Michel Thierry <michel.thierry@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index b17dac6..1eab9ba 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -2819,6 +2819,8 @@ void i915_gem_vma_destroy(struct i915_vma *vma);
 #define PIN_OFFSET_BIAS	(1<<3)
 #define PIN_USER	(1<<4)
 #define PIN_UPDATE	(1<<5)
+#define PIN_ZONE_4G	(1<<6)
+#define PIN_HIGH	(1<<7)
 #define PIN_OFFSET_MASK (~4095)
 int __must_check
 i915_gem_object_pin(struct drm_i915_gem_object *obj,
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index bf5ef7a..f0cfbb9 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -3354,11 +3354,9 @@ i915_gem_object_bind_to_vm(struct drm_i915_gem_object *obj,
 	struct drm_device *dev = obj->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 fence_alignment, unfenced_alignment;
+	u32 search_flag, alloc_flag;
+	u64 start, end;
 	u64 size, fence_size;
-	u64 start =
-		flags & PIN_OFFSET_BIAS ? flags & PIN_OFFSET_MASK : 0;
-	u64 end =
-		flags & PIN_MAPPABLE ? dev_priv->gtt.mappable_end : vm->total;
 	struct i915_vma *vma;
 	int ret;
 
@@ -3398,6 +3396,13 @@ i915_gem_object_bind_to_vm(struct drm_i915_gem_object *obj,
 		size = flags & PIN_MAPPABLE ? fence_size : obj->base.size;
 	}
 
+	start = flags & PIN_OFFSET_BIAS ? flags & PIN_OFFSET_MASK : 0;
+	end = vm->total;
+	if (flags & PIN_MAPPABLE)
+		end = min_t(u64, end, dev_priv->gtt.mappable_end);
+	if (flags & PIN_ZONE_4G)
+		end = min_t(u64, end, (1ULL << 32));
+
 	if (alignment == 0)
 		alignment = flags & PIN_MAPPABLE ? fence_alignment :
 						unfenced_alignment;
@@ -3433,13 +3438,21 @@ i915_gem_object_bind_to_vm(struct drm_i915_gem_object *obj,
 	if (IS_ERR(vma))
 		goto err_unpin;
 
+	if (flags & PIN_HIGH) {
+		search_flag = DRM_MM_SEARCH_BELOW;
+		alloc_flag = DRM_MM_CREATE_TOP;
+	} else {
+		search_flag = DRM_MM_SEARCH_DEFAULT;
+		alloc_flag = DRM_MM_CREATE_DEFAULT;
+	}
+
 search_free:
 	ret = drm_mm_insert_node_in_range_generic(&vm->mm, &vma->node,
 						  size, alignment,
 						  obj->cache_level,
 						  start, end,
-						  DRM_MM_SEARCH_DEFAULT,
-						  DRM_MM_CREATE_DEFAULT);
+						  search_flag,
+						  alloc_flag);
 	if (ret) {
 		ret = i915_gem_evict_something(dev, vm, size, alignment,
 					       obj->cache_level,
diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index 67ef118..edc17be 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -590,10 +590,17 @@ i915_gem_execbuffer_reserve_vma(struct i915_vma *vma,
 		flags |= PIN_GLOBAL;
 
 	if (!drm_mm_node_allocated(&vma->node)) {
+		/* Wa32bitGeneralStateOffset & Wa32bitInstructionBaseOffset,
+		 * limit address to the first 4GBs for unflagged objects.
+		 */
+		if ((entry->flags & EXEC_OBJECT_SUPPORTS_48B_ADDRESS) == 0)
+			flags |= PIN_ZONE_4G;
 		if (entry->flags & __EXEC_OBJECT_NEEDS_MAP)
 			flags |= PIN_GLOBAL | PIN_MAPPABLE;
 		if (entry->flags & __EXEC_OBJECT_NEEDS_BIAS)
 			flags |= BATCH_OFFSET_BIAS | PIN_OFFSET_BIAS;
+		if ((flags & PIN_MAPPABLE) == 0)
+			flags |= PIN_HIGH;
 	}
 
 	ret = i915_gem_object_pin(obj, vma->vm, entry->alignment, flags);
@@ -671,6 +678,10 @@ eb_vma_misplaced(struct i915_vma *vma)
 	if (entry->flags & __EXEC_OBJECT_NEEDS_MAP && !obj->map_and_fenceable)
 		return !only_mappable_for_reloc(entry->flags);
 
+	if ((entry->flags & EXEC_OBJECT_SUPPORTS_48B_ADDRESS) == 0 &&
+	    (vma->node.start + vma->node.size - 1) >> 32)
+		return true;
+
 	return false;
 }
 
diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h
index fd5aa47..484a9fb 100644
--- a/include/uapi/drm/i915_drm.h
+++ b/include/uapi/drm/i915_drm.h
@@ -690,7 +690,8 @@ struct drm_i915_gem_exec_object2 {
 #define EXEC_OBJECT_NEEDS_FENCE (1<<0)
 #define EXEC_OBJECT_NEEDS_GTT	(1<<1)
 #define EXEC_OBJECT_WRITE	(1<<2)
-#define __EXEC_OBJECT_UNKNOWN_FLAGS -(EXEC_OBJECT_WRITE<<1)
+#define EXEC_OBJECT_SUPPORTS_48B_ADDRESS (1<<3)
+#define __EXEC_OBJECT_UNKNOWN_FLAGS -(EXEC_OBJECT_SUPPORTS_48B_ADDRESS<<1)
 	__u64 flags;
 
 	__u64 rsvd1;
-- 
cgit v0.10.2


From f1d543485344f11f90e5fac637cc3430841ddabf Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Wed, 30 Sep 2015 23:00:42 +0300
Subject: drm/i915: remove duplicate names for the render ring INSTDONE
 register

We use 3 different names to refer to the same render ring INSTDONE
register. This can be confusing when comparing two parts of the code
accessing the register via different names. Although the GEN4 version's
layout is different, we treat it the same way as the GEN7+ version, in
that we simply read it out during error capture. So remove the
duplicates and leave a comment about the GEN4 difference.

Note that there is also a GEN2 version of this register, but that's on a
different address so not handled in this patch.

Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
index d979dca..27423ed 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.c
+++ b/drivers/gpu/drm/i915/i915_gpu_error.c
@@ -1390,10 +1390,10 @@ void i915_get_extra_instdone(struct drm_device *dev, uint32_t *instdone)
 	if (IS_GEN2(dev) || IS_GEN3(dev))
 		instdone[0] = I915_READ(INSTDONE);
 	else if (IS_GEN4(dev) || IS_GEN5(dev) || IS_GEN6(dev)) {
-		instdone[0] = I915_READ(INSTDONE_I965);
+		instdone[0] = I915_READ(RING_INSTDONE(RENDER_RING_BASE));
 		instdone[1] = I915_READ(INSTDONE1);
 	} else if (INTEL_INFO(dev)->gen >= 7) {
-		instdone[0] = I915_READ(GEN7_INSTDONE_1);
+		instdone[0] = I915_READ(RING_INSTDONE(RENDER_RING_BASE));
 		instdone[1] = I915_READ(GEN7_SC_INSTDONE);
 		instdone[2] = I915_READ(GEN7_SAMPLER_INSTDONE);
 		instdone[3] = I915_READ(GEN7_ROW_INSTDONE);
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 9e3fdb3..d1e3c3c 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -1593,14 +1593,16 @@ enum skl_disp_power_wells {
 #endif
 #define IPEIR_I965	0x02064
 #define IPEHR_I965	0x02068
-#define INSTDONE_I965	0x0206c
-#define GEN7_INSTDONE_1		0x0206c
 #define GEN7_SC_INSTDONE	0x07100
 #define GEN7_SAMPLER_INSTDONE	0x0e160
 #define GEN7_ROW_INSTDONE	0x0e164
 #define I915_NUM_INSTDONE_REG	4
 #define RING_IPEIR(base)	((base)+0x64)
 #define RING_IPEHR(base)	((base)+0x68)
+/*
+ * On GEN4, only the render ring INSTDONE exists and has a different
+ * layout than the GEN7+ version.
+ */
 #define RING_INSTDONE(base)	((base)+0x6c)
 #define RING_INSTPS(base)	((base)+0x70)
 #define RING_DMA_FADD(base)	((base)+0x78)
-- 
cgit v0.10.2


From bd93a50e4dbae108a55a228bba1a69a2314096fb Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Wed, 30 Sep 2015 23:00:43 +0300
Subject: drm/i915: rename INSTDONE to GEN2_INSTDONE

We have a bunch of INSTDONE registers for different platforms and
purposes and it's not immediately clear which instance they are just by
looking at the register name. This one was added on GEN2, where it was
the only INSTDONE register, so mark it as such.

Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
index 27423ed..85d9a39 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.c
+++ b/drivers/gpu/drm/i915/i915_gpu_error.c
@@ -886,7 +886,7 @@ static void i915_record_ring_state(struct drm_device *dev,
 		ering->faddr = I915_READ(DMA_FADD_I8XX);
 		ering->ipeir = I915_READ(IPEIR);
 		ering->ipehr = I915_READ(IPEHR);
-		ering->instdone = I915_READ(INSTDONE);
+		ering->instdone = I915_READ(GEN2_INSTDONE);
 	}
 
 	ering->waiting = waitqueue_active(&ring->irq_queue);
@@ -1388,7 +1388,7 @@ void i915_get_extra_instdone(struct drm_device *dev, uint32_t *instdone)
 	memset(instdone, 0, sizeof(*instdone) * I915_NUM_INSTDONE_REG);
 
 	if (IS_GEN2(dev) || IS_GEN3(dev))
-		instdone[0] = I915_READ(INSTDONE);
+		instdone[0] = I915_READ(GEN2_INSTDONE);
 	else if (IS_GEN4(dev) || IS_GEN5(dev) || IS_GEN6(dev)) {
 		instdone[0] = I915_READ(RING_INSTDONE(RENDER_RING_BASE));
 		instdone[1] = I915_READ(INSTDONE1);
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index d1e3c3c..5d2da1e 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -1602,6 +1602,7 @@ enum skl_disp_power_wells {
 /*
  * On GEN4, only the render ring INSTDONE exists and has a different
  * layout than the GEN7+ version.
+ * The GEN2 counterpart of this register is GEN2_INSTDONE.
  */
 #define RING_INSTDONE(base)	((base)+0x6c)
 #define RING_INSTPS(base)	((base)+0x70)
@@ -1619,7 +1620,7 @@ enum skl_disp_power_wells {
 #define   PWRCTX_EN	(1<<0)
 #define IPEIR		0x02088
 #define IPEHR		0x0208c
-#define INSTDONE	0x02090
+#define GEN2_INSTDONE	0x02090
 #define NOPID		0x02094
 #define HWSTAM		0x02098
 #define DMA_FADD_I8XX	0x020d0
-- 
cgit v0.10.2


From 8f12a3118f554d2a38cc2c0ea17443d8b041b17f Mon Sep 17 00:00:00 2001
From: Lukas Wunner <lukas@wunner.de>
Date: Sun, 23 Aug 2015 23:23:02 +0200
Subject: vga_switcheroo: Add missing locking

The following functions iterate over the client list, invoke client
callbacks or invoke handler callbacks without locking anything at all:

- Introduced by c8e9cf7bb240 ("vga_switcheroo: Add a helper function to
  get the client state"):
  vga_switcheroo_get_client_state()

- Introduced by 0d69704ae348 ("gpu/vga_switcheroo: add driver control
  power feature. (v3)"):
  vga_switcheroo_set_dynamic_switch()
  vga_switcheroo_runtime_suspend()
  vga_switcheroo_runtime_resume()
  vga_switcheroo_runtime_resume_hdmi_audio()

Refactor vga_switcheroo_runtime_resume_hdmi_audio() a bit to be able to
release vgasr_mutex immediately after iterating over the client list.

Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/vga/vga_switcheroo.c b/drivers/gpu/vga/vga_switcheroo.c
index 86c03b5..1acbe20 100644
--- a/drivers/gpu/vga/vga_switcheroo.c
+++ b/drivers/gpu/vga/vga_switcheroo.c
@@ -347,13 +347,18 @@ find_active_client(struct list_head *head)
 int vga_switcheroo_get_client_state(struct pci_dev *pdev)
 {
 	struct vga_switcheroo_client *client;
+	enum vga_switcheroo_state ret;
 
+	mutex_lock(&vgasr_mutex);
 	client = find_client_from_pci(&vgasr_priv.clients, pdev);
 	if (!client)
-		return VGA_SWITCHEROO_NOT_FOUND;
-	if (!vgasr_priv.active)
-		return VGA_SWITCHEROO_INIT;
-	return client->pwr_state;
+		ret = VGA_SWITCHEROO_NOT_FOUND;
+	else if (!vgasr_priv.active)
+		ret = VGA_SWITCHEROO_INIT;
+	else
+		ret = client->pwr_state;
+	mutex_unlock(&vgasr_mutex);
+	return ret;
 }
 EXPORT_SYMBOL(vga_switcheroo_get_client_state);
 
@@ -845,15 +850,16 @@ void vga_switcheroo_set_dynamic_switch(struct pci_dev *pdev,
 {
 	struct vga_switcheroo_client *client;
 
+	mutex_lock(&vgasr_mutex);
 	client = find_client_from_pci(&vgasr_priv.clients, pdev);
-	if (!client)
-		return;
-
-	if (!client->driver_power_control)
+	if (!client || !client->driver_power_control) {
+		mutex_unlock(&vgasr_mutex);
 		return;
+	}
 
 	client->pwr_state = dynamic;
 	set_audio_state(client->id, dynamic);
+	mutex_unlock(&vgasr_mutex);
 }
 EXPORT_SYMBOL(vga_switcheroo_set_dynamic_switch);
 
@@ -866,9 +872,11 @@ static int vga_switcheroo_runtime_suspend(struct device *dev)
 	ret = dev->bus->pm->runtime_suspend(dev);
 	if (ret)
 		return ret;
+	mutex_lock(&vgasr_mutex);
 	if (vgasr_priv.handler->switchto)
 		vgasr_priv.handler->switchto(VGA_SWITCHEROO_IGD);
 	vga_switcheroo_power_switch(pdev, VGA_SWITCHEROO_OFF);
+	mutex_unlock(&vgasr_mutex);
 	return 0;
 }
 
@@ -877,7 +885,9 @@ static int vga_switcheroo_runtime_resume(struct device *dev)
 	struct pci_dev *pdev = to_pci_dev(dev);
 	int ret;
 
+	mutex_lock(&vgasr_mutex);
 	vga_switcheroo_power_switch(pdev, VGA_SWITCHEROO_ON);
+	mutex_unlock(&vgasr_mutex);
 	ret = dev->bus->pm->runtime_resume(dev);
 	if (ret)
 		return ret;
@@ -923,29 +933,33 @@ EXPORT_SYMBOL(vga_switcheroo_fini_domain_pm_ops);
 static int vga_switcheroo_runtime_resume_hdmi_audio(struct device *dev)
 {
 	struct pci_dev *pdev = to_pci_dev(dev);
+	struct vga_switcheroo_client *client;
+	struct device *video_dev = NULL;
 	int ret;
-	struct vga_switcheroo_client *client, *found = NULL;
 
 	/* we need to check if we have to switch back on the video
 	   device so the audio device can come back */
+	mutex_lock(&vgasr_mutex);
 	list_for_each_entry(client, &vgasr_priv.clients, list) {
 		if (PCI_SLOT(client->pdev->devfn) == PCI_SLOT(pdev->devfn) &&
 		    client_is_vga(client)) {
-			found = client;
-			ret = pm_runtime_get_sync(&client->pdev->dev);
-			if (ret) {
-				if (ret != 1)
-					return ret;
-			}
+			video_dev = &client->pdev->dev;
 			break;
 		}
 	}
+	mutex_unlock(&vgasr_mutex);
+
+	if (video_dev) {
+		ret = pm_runtime_get_sync(video_dev);
+		if (ret && ret != 1)
+			return ret;
+	}
 	ret = dev->bus->pm->runtime_resume(dev);
 
 	/* put the reference for the gpu */
-	if (found) {
-		pm_runtime_mark_last_busy(&found->pdev->dev);
-		pm_runtime_put_autosuspend(&found->pdev->dev);
+	if (video_dev) {
+		pm_runtime_mark_last_busy(video_dev);
+		pm_runtime_put_autosuspend(video_dev);
 	}
 	return ret;
 }
-- 
cgit v0.10.2


From 13d70b8135e92274c9044d5ba4a543e1d6f0ef59 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Wed, 30 Sep 2015 23:00:44 +0300
Subject: drm/i915: rename INSTDONE1 to GEN4_INSTDONE1

This register was added on GEN4, by the name INSTDONE_1 whereas the GEN6
specification calls it INSTDONE_2. Keep the original name with a
platform prefix to make it clearer which INSTDONE register instance this
is. Also add a comment about the SNB alternative name.

Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
index 85d9a39..2f04e4f 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.c
+++ b/drivers/gpu/drm/i915/i915_gpu_error.c
@@ -1391,7 +1391,7 @@ void i915_get_extra_instdone(struct drm_device *dev, uint32_t *instdone)
 		instdone[0] = I915_READ(GEN2_INSTDONE);
 	else if (IS_GEN4(dev) || IS_GEN5(dev) || IS_GEN6(dev)) {
 		instdone[0] = I915_READ(RING_INSTDONE(RENDER_RING_BASE));
-		instdone[1] = I915_READ(INSTDONE1);
+		instdone[1] = I915_READ(GEN4_INSTDONE1);
 	} else if (INTEL_INFO(dev)->gen >= 7) {
 		instdone[0] = I915_READ(RING_INSTDONE(RENDER_RING_BASE));
 		instdone[1] = I915_READ(GEN7_SC_INSTDONE);
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 5d2da1e..a1c313f 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -1611,7 +1611,7 @@ enum skl_disp_power_wells {
 #define RING_INSTPM(base)	((base)+0xc0)
 #define RING_MI_MODE(base)	((base)+0x9c)
 #define INSTPS		0x02070 /* 965+ only */
-#define INSTDONE1	0x0207c /* 965+ only */
+#define GEN4_INSTDONE1	0x0207c /* 965+ only, aka INSTDONE_2 on SNB */
 #define ACTHD_I965	0x02074
 #define HWS_PGA		0x02080
 #define HWS_ADDRESS_MASK	0xfffff000
-- 
cgit v0.10.2


From 7d4aefd0a90dcced6ec24fd1908e4b407a8d4793 Mon Sep 17 00:00:00 2001
From: Shashank Sharma <shashank.sharma@intel.com>
Date: Thu, 1 Oct 2015 22:23:49 +0530
Subject: drm/i915/bxt: DSI encoder support in CRTC modeset

SKL and BXT qualifies the HAS_DDI() check, and hence haswell
modeset functions are re-used for modeset sequence. But DDI
interface doesn't include support for DSI.
This patch adds:
1. cases for DSI encoder, in those modeset functions and allows
   a CRTC modeset
2. Adds call to pre_pll enabled from CRTC modeset function. Nothing
   needs to be done as such in CRTC for DSI encoder, as PLL, clock
   and and transcoder programming will be taken care in encoder's
   pre_enable and pre_pll_enable function.

v2: Fixed Jani's review comments. Added INVALID_PORT for non DDI
    encoder like DSI for platforms having HAS_DDI as true.

v3: Rebased on latest drm-nightly branch. Added a WARN_ON for invalid
    encoder.

v4: WARN_ON for invalid encoder is refactored as per Jani's suggestion.
    Fixed the sequence for pre_pll_enable.

v5: Protected DDI code paths in case of DSI encoder calls.

Signed-off-by: Shashank Sharma <shashank.sharma@intel.com>
Signed-off-by: Uma Shankar <uma.shankar@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
index 2d3cc82..b3f1748 100644
--- a/drivers/gpu/drm/i915/intel_ddi.c
+++ b/drivers/gpu/drm/i915/intel_ddi.c
@@ -558,8 +558,10 @@ void intel_prepare_ddi(struct drm_device *dev)
 		enum port port;
 		bool supports_hdmi;
 
-		ddi_get_encoder_port(intel_encoder, &intel_dig_port, &port);
+		if (intel_encoder->type == INTEL_OUTPUT_DSI)
+			continue;
 
+		ddi_get_encoder_port(intel_encoder, &intel_dig_port, &port);
 		if (visited[port])
 			continue;
 
@@ -2043,7 +2045,8 @@ bool intel_ddi_get_hw_state(struct intel_encoder *encoder,
 void intel_ddi_enable_pipe_clock(struct intel_crtc *intel_crtc)
 {
 	struct drm_crtc *crtc = &intel_crtc->base;
-	struct drm_i915_private *dev_priv = crtc->dev->dev_private;
+	struct drm_device *dev = crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_encoder *intel_encoder = intel_ddi_get_crtc_encoder(crtc);
 	enum port port = intel_ddi_get_encoder_port(intel_encoder);
 	enum transcoder cpu_transcoder = intel_crtc->config->cpu_transcoder;
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index bbeb6d3..539c373 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -4965,6 +4965,7 @@ static void haswell_crtc_enable(struct drm_crtc *crtc)
 	int pipe = intel_crtc->pipe, hsw_workaround_pipe;
 	struct intel_crtc_state *pipe_config =
 		to_intel_crtc_state(crtc->state);
+	bool is_dsi = intel_pipe_has_type(intel_crtc, INTEL_OUTPUT_DSI);
 
 	if (WARN_ON(intel_crtc->active))
 		return;
@@ -4994,9 +4995,12 @@ static void haswell_crtc_enable(struct drm_crtc *crtc)
 	intel_crtc->active = true;
 
 	intel_set_cpu_fifo_underrun_reporting(dev_priv, pipe, true);
-	for_each_encoder_on_crtc(dev, crtc, encoder)
+	for_each_encoder_on_crtc(dev, crtc, encoder) {
+		if (encoder->pre_pll_enable)
+			encoder->pre_pll_enable(encoder);
 		if (encoder->pre_enable)
 			encoder->pre_enable(encoder);
+	}
 
 	if (intel_crtc->config->has_pch_encoder) {
 		intel_set_pch_fifo_underrun_reporting(dev_priv, TRANSCODER_A,
@@ -5004,7 +5008,8 @@ static void haswell_crtc_enable(struct drm_crtc *crtc)
 		dev_priv->display.fdi_link_train(crtc);
 	}
 
-	intel_ddi_enable_pipe_clock(intel_crtc);
+	if (!is_dsi)
+		intel_ddi_enable_pipe_clock(intel_crtc);
 
 	if (INTEL_INFO(dev)->gen >= 9)
 		skylake_pfit_enable(intel_crtc);
@@ -5018,7 +5023,8 @@ static void haswell_crtc_enable(struct drm_crtc *crtc)
 	intel_crtc_load_lut(crtc);
 
 	intel_ddi_set_pipe_settings(crtc);
-	intel_ddi_enable_transcoder_func(crtc);
+	if (!is_dsi)
+		intel_ddi_enable_transcoder_func(crtc);
 
 	intel_update_watermarks(crtc);
 	intel_enable_pipe(intel_crtc);
@@ -5026,7 +5032,7 @@ static void haswell_crtc_enable(struct drm_crtc *crtc)
 	if (intel_crtc->config->has_pch_encoder)
 		lpt_pch_enable(crtc);
 
-	if (intel_crtc->config->dp_encoder_is_mst)
+	if (intel_crtc->config->dp_encoder_is_mst && !is_dsi)
 		intel_ddi_set_vc_payload_alloc(crtc, true);
 
 	assert_vblank_disabled(crtc);
@@ -5119,6 +5125,7 @@ static void haswell_crtc_disable(struct drm_crtc *crtc)
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct intel_encoder *encoder;
 	enum transcoder cpu_transcoder = intel_crtc->config->cpu_transcoder;
+	bool is_dsi = intel_pipe_has_type(intel_crtc, INTEL_OUTPUT_DSI);
 
 	for_each_encoder_on_crtc(dev, crtc, encoder) {
 		intel_opregion_notify_encoder(encoder, false);
@@ -5136,14 +5143,16 @@ static void haswell_crtc_disable(struct drm_crtc *crtc)
 	if (intel_crtc->config->dp_encoder_is_mst)
 		intel_ddi_set_vc_payload_alloc(crtc, false);
 
-	intel_ddi_disable_transcoder_func(dev_priv, cpu_transcoder);
+	if (!is_dsi)
+		intel_ddi_disable_transcoder_func(dev_priv, cpu_transcoder);
 
 	if (INTEL_INFO(dev)->gen >= 9)
 		skylake_scaler_disable(intel_crtc);
 	else
 		ironlake_pfit_disable(intel_crtc, false);
 
-	intel_ddi_disable_pipe_clock(intel_crtc);
+	if (!is_dsi)
+		intel_ddi_disable_pipe_clock(intel_crtc);
 
 	if (intel_crtc->config->has_pch_encoder) {
 		lpt_disable_pch_transcoder(dev_priv);
diff --git a/drivers/gpu/drm/i915/intel_opregion.c b/drivers/gpu/drm/i915/intel_opregion.c
index cb1c657..b706b4e 100644
--- a/drivers/gpu/drm/i915/intel_opregion.c
+++ b/drivers/gpu/drm/i915/intel_opregion.c
@@ -341,8 +341,12 @@ int intel_opregion_notify_encoder(struct intel_encoder *intel_encoder,
 	if (!HAS_DDI(dev))
 		return 0;
 
-	port = intel_ddi_get_encoder_port(intel_encoder);
-	if (port == PORT_E) {
+	if (intel_encoder->type == INTEL_OUTPUT_DSI)
+		port = 0;
+	else
+		port = intel_ddi_get_encoder_port(intel_encoder);
+
+	if (port == PORT_E)  {
 		port = 0;
 	} else {
 		parm |= 1 << port;
@@ -363,6 +367,7 @@ int intel_opregion_notify_encoder(struct intel_encoder *intel_encoder,
 		type = DISPLAY_TYPE_EXTERNAL_FLAT_PANEL;
 		break;
 	case INTEL_OUTPUT_EDP:
+	case INTEL_OUTPUT_DSI:
 		type = DISPLAY_TYPE_INTERNAL_FLAT_PANEL;
 		break;
 	default:
-- 
cgit v0.10.2


From 37ab0810c9b7e06ec3904c186c46e9c540b3793b Mon Sep 17 00:00:00 2001
From: Shashank Sharma <shashank.sharma@intel.com>
Date: Tue, 1 Sep 2015 19:41:42 +0530
Subject: drm/i915/bxt: DSI enable for BXT

This patch contains following changes:
1. MIPI device ready changes to support dsi_pre_enable. Changes
   are specific to BXT device ready sequence. Added check for
   ULPS mode(No effects on VLV).
2. Changes in dsi_enable to pick BXT port control register.
3. Changes in dsi_pre_enable to restrict DPIO programming for VLV

v2: Fixed Jani's review comments. Removed the changes in VLV/CHV
    code. Fixed the macros to get proper port offsets.

v3: Rebased on latest drm-nightly branch. Fixed Jani's review comments.

Signed-off-by: Shashank Sharma <shashank.sharma@intel.com>
Signed-off-by: Uma Shankar <uma.shankar@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index a1c313f..87de3a0 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -7588,6 +7588,13 @@ enum skl_disp_power_wells {
 #define _MIPIA_PORT_CTRL			(VLV_DISPLAY_BASE + 0x61190)
 #define _MIPIC_PORT_CTRL			(VLV_DISPLAY_BASE + 0x61700)
 #define MIPI_PORT_CTRL(port)	_MIPI_PORT(port, _MIPIA_PORT_CTRL, _MIPIC_PORT_CTRL)
+
+ /* BXT port control */
+#define _BXT_MIPIA_PORT_CTRL				0x6B0C0
+#define _BXT_MIPIC_PORT_CTRL				0x6B8C0
+#define BXT_MIPI_PORT_CTRL(tc)	_MIPI_PORT(tc, _BXT_MIPIA_PORT_CTRL, \
+						_BXT_MIPIC_PORT_CTRL)
+
 #define  DPI_ENABLE					(1 << 31) /* A + C */
 #define  MIPIA_MIPI4DPHY_DELAY_COUNT_SHIFT		27
 #define  MIPIA_MIPI4DPHY_DELAY_COUNT_MASK		(0xf << 27)
diff --git a/drivers/gpu/drm/i915/intel_dsi.c b/drivers/gpu/drm/i915/intel_dsi.c
index eb86ba7..b7eae52 100644
--- a/drivers/gpu/drm/i915/intel_dsi.c
+++ b/drivers/gpu/drm/i915/intel_dsi.c
@@ -282,58 +282,46 @@ static bool intel_dsi_compute_config(struct intel_encoder *encoder,
 	return true;
 }
 
-static void intel_dsi_port_enable(struct intel_encoder *encoder)
+static void bxt_dsi_device_ready(struct intel_encoder *encoder)
 {
-	struct drm_device *dev = encoder->base.dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->base.crtc);
+	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
 	enum port port;
-	u32 temp;
+	u32 val;
 
-	if (intel_dsi->dual_link == DSI_DUAL_LINK_FRONT_BACK) {
-		temp = I915_READ(VLV_CHICKEN_3);
-		temp &= ~PIXEL_OVERLAP_CNT_MASK |
-					intel_dsi->pixel_overlap <<
-					PIXEL_OVERLAP_CNT_SHIFT;
-		I915_WRITE(VLV_CHICKEN_3, temp);
-	}
+	DRM_DEBUG_KMS("\n");
 
+	/* Exit Low power state in 4 steps*/
 	for_each_dsi_port(port, intel_dsi->ports) {
-		temp = I915_READ(MIPI_PORT_CTRL(port));
-		temp &= ~LANE_CONFIGURATION_MASK;
-		temp &= ~DUAL_LINK_MODE_MASK;
 
-		if (intel_dsi->ports == ((1 << PORT_A) | (1 << PORT_C))) {
-			temp |= (intel_dsi->dual_link - 1)
-						<< DUAL_LINK_MODE_SHIFT;
-			temp |= intel_crtc->pipe ?
-					LANE_CONFIGURATION_DUAL_LINK_B :
-					LANE_CONFIGURATION_DUAL_LINK_A;
-		}
-		/* assert ip_tg_enable signal */
-		I915_WRITE(MIPI_PORT_CTRL(port), temp | DPI_ENABLE);
-		POSTING_READ(MIPI_PORT_CTRL(port));
-	}
-}
+		/* 1. Enable MIPI PHY transparent latch */
+		val = I915_READ(BXT_MIPI_PORT_CTRL(port));
+		I915_WRITE(BXT_MIPI_PORT_CTRL(port), val | LP_OUTPUT_HOLD);
+		usleep_range(2000, 2500);
 
-static void intel_dsi_port_disable(struct intel_encoder *encoder)
-{
-	struct drm_device *dev = encoder->base.dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
-	enum port port;
-	u32 temp;
+		/* 2. Enter ULPS */
+		val = I915_READ(MIPI_DEVICE_READY(port));
+		val &= ~ULPS_STATE_MASK;
+		val |= (ULPS_STATE_ENTER | DEVICE_READY);
+		I915_WRITE(MIPI_DEVICE_READY(port), val);
+		usleep_range(2, 3);
+
+		/* 3. Exit ULPS */
+		val = I915_READ(MIPI_DEVICE_READY(port));
+		val &= ~ULPS_STATE_MASK;
+		val |= (ULPS_STATE_EXIT | DEVICE_READY);
+		I915_WRITE(MIPI_DEVICE_READY(port), val);
+		usleep_range(1000, 1500);
 
-	for_each_dsi_port(port, intel_dsi->ports) {
-		/* de-assert ip_tg_enable signal */
-		temp = I915_READ(MIPI_PORT_CTRL(port));
-		I915_WRITE(MIPI_PORT_CTRL(port), temp & ~DPI_ENABLE);
-		POSTING_READ(MIPI_PORT_CTRL(port));
+		/* Clear ULPS and set device ready */
+		val = I915_READ(MIPI_DEVICE_READY(port));
+		val &= ~ULPS_STATE_MASK;
+		val |= DEVICE_READY;
+		I915_WRITE(MIPI_DEVICE_READY(port), val);
 	}
 }
 
-static void intel_dsi_device_ready(struct intel_encoder *encoder)
+static void vlv_dsi_device_ready(struct intel_encoder *encoder)
 {
 	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
@@ -372,6 +360,72 @@ static void intel_dsi_device_ready(struct intel_encoder *encoder)
 	}
 }
 
+static void intel_dsi_device_ready(struct intel_encoder *encoder)
+{
+	struct drm_device *dev = encoder->base.dev;
+
+	if (IS_VALLEYVIEW(dev))
+		vlv_dsi_device_ready(encoder);
+	else if (IS_BROXTON(dev))
+		bxt_dsi_device_ready(encoder);
+}
+
+static void intel_dsi_port_enable(struct intel_encoder *encoder)
+{
+	struct drm_device *dev = encoder->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->base.crtc);
+	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
+	enum port port;
+	u32 temp;
+	u32 port_ctrl;
+
+	if (intel_dsi->dual_link == DSI_DUAL_LINK_FRONT_BACK) {
+		temp = I915_READ(VLV_CHICKEN_3);
+		temp &= ~PIXEL_OVERLAP_CNT_MASK |
+					intel_dsi->pixel_overlap <<
+					PIXEL_OVERLAP_CNT_SHIFT;
+		I915_WRITE(VLV_CHICKEN_3, temp);
+	}
+
+	for_each_dsi_port(port, intel_dsi->ports) {
+		port_ctrl = IS_BROXTON(dev) ? BXT_MIPI_PORT_CTRL(port) :
+						MIPI_PORT_CTRL(port);
+
+		temp = I915_READ(port_ctrl);
+
+		temp &= ~LANE_CONFIGURATION_MASK;
+		temp &= ~DUAL_LINK_MODE_MASK;
+
+		if (intel_dsi->ports == ((1 << PORT_A) | (1 << PORT_C))) {
+			temp |= (intel_dsi->dual_link - 1)
+						<< DUAL_LINK_MODE_SHIFT;
+			temp |= intel_crtc->pipe ?
+					LANE_CONFIGURATION_DUAL_LINK_B :
+					LANE_CONFIGURATION_DUAL_LINK_A;
+		}
+		/* assert ip_tg_enable signal */
+		I915_WRITE(port_ctrl, temp | DPI_ENABLE);
+		POSTING_READ(port_ctrl);
+	}
+}
+
+static void intel_dsi_port_disable(struct intel_encoder *encoder)
+{
+	struct drm_device *dev = encoder->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
+	enum port port;
+	u32 temp;
+
+	for_each_dsi_port(port, intel_dsi->ports) {
+		/* de-assert ip_tg_enable signal */
+		temp = I915_READ(MIPI_PORT_CTRL(port));
+		I915_WRITE(MIPI_PORT_CTRL(port), temp & ~DPI_ENABLE);
+		POSTING_READ(MIPI_PORT_CTRL(port));
+	}
+}
+
 static void intel_dsi_enable(struct intel_encoder *encoder)
 {
 	struct drm_device *dev = encoder->base.dev;
@@ -419,19 +473,24 @@ static void intel_dsi_pre_enable(struct intel_encoder *encoder)
 
 	msleep(intel_dsi->panel_on_delay);
 
-	/* Disable DPOunit clock gating, can stall pipe
-	 * and we need DPLL REFA always enabled */
-	tmp = I915_READ(DPLL(pipe));
-	tmp |= DPLL_REF_CLK_ENABLE_VLV;
-	I915_WRITE(DPLL(pipe), tmp);
-
-	/* update the hw state for DPLL */
-	intel_crtc->config->dpll_hw_state.dpll = DPLL_INTEGRATED_REF_CLK_VLV |
-		DPLL_REF_CLK_ENABLE_VLV | DPLL_VGA_MODE_DIS;
-
-	tmp = I915_READ(DSPCLK_GATE_D);
-	tmp |= DPOUNIT_CLOCK_GATE_DISABLE;
-	I915_WRITE(DSPCLK_GATE_D, tmp);
+	if (IS_VALLEYVIEW(dev)) {
+		/*
+		 * Disable DPOunit clock gating, can stall pipe
+		 * and we need DPLL REFA always enabled
+		 */
+		tmp = I915_READ(DPLL(pipe));
+		tmp |= DPLL_REF_CLK_ENABLE_VLV;
+		I915_WRITE(DPLL(pipe), tmp);
+
+		/* update the hw state for DPLL */
+		intel_crtc->config->dpll_hw_state.dpll =
+				DPLL_INTEGRATED_REF_CLK_VLV |
+					DPLL_REF_CLK_ENABLE_VLV | DPLL_VGA_MODE_DIS;
+
+		tmp = I915_READ(DSPCLK_GATE_D);
+		tmp |= DPOUNIT_CLOCK_GATE_DISABLE;
+		I915_WRITE(DSPCLK_GATE_D, tmp);
+	}
 
 	/* put device in ready state */
 	intel_dsi_device_ready(encoder);
-- 
cgit v0.10.2


From 3fe2c7d4c85b4557c9772ff16f31ddc18cfd2521 Mon Sep 17 00:00:00 2001
From: Dmitry Osipenko <digetx@gmail.com>
Date: Sun, 28 Jun 2015 22:27:02 +0300
Subject: gpu: host1x: Fix MLOCK's debug info

MLOCK's debug info, spewed on CDMA timeout, contains meaningless MLOCK
owner channel ID because HOST1X_SYNC_MLOCK_OWNER_CHID_F() returns shifted
value, while unshifted should be used. Fix it by changing '_F' to '_V'.

Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Reviewed-By: Terje Bergstrom <tbergstrom@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>

diff --git a/drivers/gpu/host1x/hw/debug_hw.c b/drivers/gpu/host1x/hw/debug_hw.c
index 791de93..cc3f182 100644
--- a/drivers/gpu/host1x/hw/debug_hw.c
+++ b/drivers/gpu/host1x/hw/debug_hw.c
@@ -298,7 +298,7 @@ static void host1x_debug_show_mlocks(struct host1x *host, struct output *o)
 			host1x_sync_readl(host, HOST1X_SYNC_MLOCK_OWNER(i));
 		if (HOST1X_SYNC_MLOCK_OWNER_CH_OWNS_V(owner))
 			host1x_debug_output(o, "%d: locked by channel %d\n",
-				i, HOST1X_SYNC_MLOCK_OWNER_CHID_F(owner));
+				i, HOST1X_SYNC_MLOCK_OWNER_CHID_V(owner));
 		else if (HOST1X_SYNC_MLOCK_OWNER_CPU_OWNS_V(owner))
 			host1x_debug_output(o, "%d: locked by cpu\n", i);
 		else
diff --git a/drivers/gpu/host1x/hw/hw_host1x01_sync.h b/drivers/gpu/host1x/hw/hw_host1x01_sync.h
index ac704e5..31238c2 100644
--- a/drivers/gpu/host1x/hw/hw_host1x01_sync.h
+++ b/drivers/gpu/host1x/hw/hw_host1x01_sync.h
@@ -131,12 +131,12 @@ static inline u32 host1x_sync_mlock_owner_r(unsigned int id)
 }
 #define HOST1X_SYNC_MLOCK_OWNER(id) \
 	host1x_sync_mlock_owner_r(id)
-static inline u32 host1x_sync_mlock_owner_chid_f(u32 v)
+static inline u32 host1x_sync_mlock_owner_chid_v(u32 v)
 {
-	return (v & 0xf) << 8;
+	return (v >> 8) & 0xf;
 }
-#define HOST1X_SYNC_MLOCK_OWNER_CHID_F(v) \
-	host1x_sync_mlock_owner_chid_f(v)
+#define HOST1X_SYNC_MLOCK_OWNER_CHID_V(v) \
+	host1x_sync_mlock_owner_chid_v(v)
 static inline u32 host1x_sync_mlock_owner_cpu_owns_v(u32 r)
 {
 	return (r >> 1) & 0x1;
diff --git a/drivers/gpu/host1x/hw/hw_host1x02_sync.h b/drivers/gpu/host1x/hw/hw_host1x02_sync.h
index 4495401..540c7b6 100644
--- a/drivers/gpu/host1x/hw/hw_host1x02_sync.h
+++ b/drivers/gpu/host1x/hw/hw_host1x02_sync.h
@@ -131,12 +131,12 @@ static inline u32 host1x_sync_mlock_owner_r(unsigned int id)
 }
 #define HOST1X_SYNC_MLOCK_OWNER(id) \
 	host1x_sync_mlock_owner_r(id)
-static inline u32 host1x_sync_mlock_owner_chid_f(u32 v)
+static inline u32 host1x_sync_mlock_owner_chid_v(u32 v)
 {
-	return (v & 0xf) << 8;
+	return (v >> 8) & 0xf;
 }
-#define HOST1X_SYNC_MLOCK_OWNER_CHID_F(v) \
-	host1x_sync_mlock_owner_chid_f(v)
+#define HOST1X_SYNC_MLOCK_OWNER_CHID_V(v) \
+	host1x_sync_mlock_owner_chid_v(v)
 static inline u32 host1x_sync_mlock_owner_cpu_owns_v(u32 r)
 {
 	return (r >> 1) & 0x1;
diff --git a/drivers/gpu/host1x/hw/hw_host1x04_sync.h b/drivers/gpu/host1x/hw/hw_host1x04_sync.h
index ef2275b..3d6c8ec 100644
--- a/drivers/gpu/host1x/hw/hw_host1x04_sync.h
+++ b/drivers/gpu/host1x/hw/hw_host1x04_sync.h
@@ -131,12 +131,12 @@ static inline u32 host1x_sync_mlock_owner_r(unsigned int id)
 }
 #define HOST1X_SYNC_MLOCK_OWNER(id) \
 	host1x_sync_mlock_owner_r(id)
-static inline u32 host1x_sync_mlock_owner_chid_f(u32 v)
+static inline u32 host1x_sync_mlock_owner_chid_v(u32 v)
 {
-	return (v & 0xf) << 8;
+	return (v >> 8) & 0xf;
 }
-#define HOST1X_SYNC_MLOCK_OWNER_CHID_F(v) \
-	host1x_sync_mlock_owner_chid_f(v)
+#define HOST1X_SYNC_MLOCK_OWNER_CHID_V(v) \
+	host1x_sync_mlock_owner_chid_v(v)
 static inline u32 host1x_sync_mlock_owner_cpu_owns_v(u32 r)
 {
 	return (r >> 1) & 0x1;
-- 
cgit v0.10.2


From 11b8e4f58e1baa94e44400a076b1a3757612ea55 Mon Sep 17 00:00:00 2001
From: Shashank Sharma <shashank.sharma@intel.com>
Date: Wed, 23 Sep 2015 23:27:17 +0530
Subject: drm/i915/bxt: Program Tx Rx and Dphy clocks

BXT DSI clocks are different than previous platforms. So adding a
new function to program following clocks and dividers:
1. Program variable divider to generate input to Tx clock divider
   (Output value must be < 39.5Mhz)
2. Select divide by 2 option to get < 20Mhz for Tx clock
3. Program 8by3 divider to generate Rx clock

v2: Fixed Jani's review comments. Adjusted the Macro definition as
    per convention. Simplified the logic for bit definitions for
    MIPI PORT A and PORT C in same registers.

v3: Refactored the macros for TX, RX Escape and DPHY clocks as per
    Jani's suggestion.

v4: Addressed Jani's review comments.

Signed-off-by: Shashank Sharma <shashank.sharma@intel.com>
Signed-off-by: Uma Shankar <uma.shankar@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 87de3a0..8c24851 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -7547,6 +7547,68 @@ enum skl_disp_power_wells {
 
 #define _MIPI_PORT(port, a, c)	_PORT3(port, a, 0, c)	/* ports A and C only */
 
+/* BXT MIPI clock controls */
+#define BXT_MAX_VAR_OUTPUT_KHZ			39500
+
+#define BXT_MIPI_CLOCK_CTL			0x46090
+#define  BXT_MIPI1_DIV_SHIFT			26
+#define  BXT_MIPI2_DIV_SHIFT			10
+#define  BXT_MIPI_DIV_SHIFT(port)		\
+			_MIPI_PORT(port, BXT_MIPI1_DIV_SHIFT, \
+					BXT_MIPI2_DIV_SHIFT)
+/* Var clock divider to generate TX source. Result must be < 39.5 M */
+#define  BXT_MIPI1_ESCLK_VAR_DIV_MASK		(0x3F << 26)
+#define  BXT_MIPI2_ESCLK_VAR_DIV_MASK		(0x3F << 10)
+#define  BXT_MIPI_ESCLK_VAR_DIV_MASK(port)	\
+			_MIPI_PORT(port, BXT_MIPI1_ESCLK_VAR_DIV_MASK, \
+						BXT_MIPI2_ESCLK_VAR_DIV_MASK)
+
+#define  BXT_MIPI_ESCLK_VAR_DIV(port, val)	\
+			(val << BXT_MIPI_DIV_SHIFT(port))
+/* TX control divider to select actual TX clock output from (8x/var) */
+#define  BXT_MIPI1_TX_ESCLK_SHIFT		21
+#define  BXT_MIPI2_TX_ESCLK_SHIFT		5
+#define  BXT_MIPI_TX_ESCLK_SHIFT(port)		\
+			_MIPI_PORT(port, BXT_MIPI1_TX_ESCLK_SHIFT, \
+					BXT_MIPI2_TX_ESCLK_SHIFT)
+#define  BXT_MIPI1_TX_ESCLK_FIXDIV_MASK		(3 << 21)
+#define  BXT_MIPI2_TX_ESCLK_FIXDIV_MASK		(3 << 5)
+#define  BXT_MIPI_TX_ESCLK_FIXDIV_MASK(port)	\
+			_MIPI_PORT(port, BXT_MIPI1_TX_ESCLK_FIXDIV_MASK, \
+						BXT_MIPI2_TX_ESCLK_FIXDIV_MASK)
+#define  BXT_MIPI_TX_ESCLK_8XDIV_BY2(port)	\
+		(0x0 << BXT_MIPI_TX_ESCLK_SHIFT(port))
+#define  BXT_MIPI_TX_ESCLK_8XDIV_BY4(port)	\
+		(0x1 << BXT_MIPI_TX_ESCLK_SHIFT(port))
+#define  BXT_MIPI_TX_ESCLK_8XDIV_BY8(port)	\
+		(0x2 << BXT_MIPI_TX_ESCLK_SHIFT(port))
+/* RX control divider to select actual RX clock output from 8x*/
+#define  BXT_MIPI1_RX_ESCLK_SHIFT		19
+#define  BXT_MIPI2_RX_ESCLK_SHIFT		3
+#define  BXT_MIPI_RX_ESCLK_SHIFT(port)		\
+			_MIPI_PORT(port, BXT_MIPI1_RX_ESCLK_SHIFT, \
+					BXT_MIPI2_RX_ESCLK_SHIFT)
+#define  BXT_MIPI1_RX_ESCLK_FIXDIV_MASK		(3 << 19)
+#define  BXT_MIPI2_RX_ESCLK_FIXDIV_MASK		(3 << 3)
+#define  BXT_MIPI_RX_ESCLK_FIXDIV_MASK(port)	\
+		(3 << BXT_MIPI_RX_ESCLK_SHIFT(port))
+#define  BXT_MIPI_RX_ESCLK_8X_BY2(port)	\
+		(1 << BXT_MIPI_RX_ESCLK_SHIFT(port))
+#define  BXT_MIPI_RX_ESCLK_8X_BY3(port)	\
+		(2 << BXT_MIPI_RX_ESCLK_SHIFT(port))
+#define  BXT_MIPI_RX_ESCLK_8X_BY4(port)	\
+		(3 << BXT_MIPI_RX_ESCLK_SHIFT(port))
+/* BXT-A WA: Always prog DPHY dividers to 00 */
+#define  BXT_MIPI1_DPHY_DIV_SHIFT		16
+#define  BXT_MIPI2_DPHY_DIV_SHIFT		0
+#define  BXT_MIPI_DPHY_DIV_SHIFT(port)		\
+			_MIPI_PORT(port, BXT_MIPI1_DPHY_DIV_SHIFT, \
+					BXT_MIPI2_DPHY_DIV_SHIFT)
+#define  BXT_MIPI_1_DPHY_DIVIDER_MASK		(3 << 16)
+#define  BXT_MIPI_2_DPHY_DIVIDER_MASK		(3 << 0)
+#define  BXT_MIPI_DPHY_DIVIDER_MASK(port)	\
+		(3 << BXT_MIPI_DPHY_DIV_SHIFT(port))
+
 /* BXT MIPI mode configure */
 #define  _BXT_MIPIA_TRANS_HACTIVE			0x6B0F8
 #define  _BXT_MIPIC_TRANS_HACTIVE			0x6B8F8
diff --git a/drivers/gpu/drm/i915/intel_dsi_pll.c b/drivers/gpu/drm/i915/intel_dsi_pll.c
index f335e6c..3863362 100644
--- a/drivers/gpu/drm/i915/intel_dsi_pll.c
+++ b/drivers/gpu/drm/i915/intel_dsi_pll.c
@@ -384,6 +384,42 @@ u32 vlv_get_dsi_pclk(struct intel_encoder *encoder, int pipe_bpp)
 	return pclk;
 }
 
+/* Program BXT Mipi clocks and dividers */
+static void bxt_dsi_program_clocks(struct drm_device *dev, enum port port)
+{
+	u32 tmp;
+	u32 divider;
+	u32 dsi_rate;
+	u32 pll_ratio;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	/* Clear old configurations */
+	tmp = I915_READ(BXT_MIPI_CLOCK_CTL);
+	tmp &= ~(BXT_MIPI_TX_ESCLK_FIXDIV_MASK(port));
+	tmp &= ~(BXT_MIPI_RX_ESCLK_FIXDIV_MASK(port));
+	tmp &= ~(BXT_MIPI_ESCLK_VAR_DIV_MASK(port));
+	tmp &= ~(BXT_MIPI_DPHY_DIVIDER_MASK(port));
+
+	/* Get the current DSI rate(actual) */
+	pll_ratio = I915_READ(BXT_DSI_PLL_CTL) &
+				BXT_DSI_PLL_RATIO_MASK;
+	dsi_rate = (BXT_REF_CLOCK_KHZ * pll_ratio) / 2;
+
+	/* Max possible output of clock is 39.5 MHz, program value -1 */
+	divider = (dsi_rate / BXT_MAX_VAR_OUTPUT_KHZ) - 1;
+	tmp |= BXT_MIPI_ESCLK_VAR_DIV(port, divider);
+
+	/*
+	 * Tx escape clock must be as close to 20MHz possible, but should
+	 * not exceed it. Hence select divide by 2
+	 */
+	tmp |= BXT_MIPI_TX_ESCLK_8XDIV_BY2(port);
+
+	tmp |= BXT_MIPI_RX_ESCLK_8X_BY3(port);
+
+	I915_WRITE(BXT_MIPI_CLOCK_CTL, tmp);
+}
+
 static bool bxt_configure_dsi_pll(struct intel_encoder *encoder)
 {
 	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
@@ -435,6 +471,8 @@ static bool bxt_configure_dsi_pll(struct intel_encoder *encoder)
 static void bxt_enable_dsi_pll(struct intel_encoder *encoder)
 {
 	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
+	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
+	enum port port;
 	u32 val;
 
 	DRM_DEBUG_KMS("\n");
@@ -453,6 +491,10 @@ static void bxt_enable_dsi_pll(struct intel_encoder *encoder)
 		return;
 	}
 
+	/* Program TX, RX, Dphy clocks */
+	for_each_dsi_port(port, intel_dsi->ports)
+		bxt_dsi_program_clocks(encoder->base.dev, port);
+
 	/* Enable DSI PLL */
 	val = I915_READ(BXT_DSI_PLL_ENABLE);
 	val |= BXT_DSI_PLL_DO_ENABLE;
-- 
cgit v0.10.2


From b389a45c6b2f9b310ddf631d23b9e3e045cf747b Mon Sep 17 00:00:00 2001
From: Shashank Sharma <shashank.sharma@intel.com>
Date: Tue, 1 Sep 2015 19:41:44 +0530
Subject: drm/i915/bxt: DSI disable and post-disable

This patch contains changes to support DSI disble sequence in BXT.
The changes are:
1. BXT specific changes in clear_device_ready function.
2. BXT specific changes in DSI disable and post-disable functions.
3. Add a new function to reset BXT Dphy clock and dividers
   (bxt_dsi_reset_clocks).
4. Moved some part of the vlv clock reset code, in a new function
   (vlv_dsi_reset_clocks) maintaining the exact same sequence.
5. Wrapper function to call corresponding reset clock function.

v2: Fixed Jani's review comments.

v3: Removed the GET_DSI_PORT_CTRL Macro for consistency with earlier
    implementations as per Jani's suggestion.

Signed-off-by: Uma Shankar <uma.shankar@intel.com>
Signed-off-by: Shashank Sharma <shashank.sharma@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dsi.c b/drivers/gpu/drm/i915/intel_dsi.c
index b7eae52..505501e 100644
--- a/drivers/gpu/drm/i915/intel_dsi.c
+++ b/drivers/gpu/drm/i915/intel_dsi.c
@@ -417,12 +417,15 @@ static void intel_dsi_port_disable(struct intel_encoder *encoder)
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
 	enum port port;
 	u32 temp;
+	u32 port_ctrl;
 
 	for_each_dsi_port(port, intel_dsi->ports) {
 		/* de-assert ip_tg_enable signal */
-		temp = I915_READ(MIPI_PORT_CTRL(port));
-		I915_WRITE(MIPI_PORT_CTRL(port), temp & ~DPI_ENABLE);
-		POSTING_READ(MIPI_PORT_CTRL(port));
+		port_ctrl = IS_BROXTON(dev) ? BXT_MIPI_PORT_CTRL(port) :
+						MIPI_PORT_CTRL(port);
+		temp = I915_READ(port_ctrl);
+		I915_WRITE(port_ctrl, temp & ~DPI_ENABLE);
+		POSTING_READ(port_ctrl);
 	}
 }
 
@@ -554,12 +557,7 @@ static void intel_dsi_disable(struct intel_encoder *encoder)
 		/* Panel commands can be sent when clock is in LP11 */
 		I915_WRITE(MIPI_DEVICE_READY(port), 0x0);
 
-		temp = I915_READ(MIPI_CTRL(port));
-		temp &= ~ESCAPE_CLOCK_DIVIDER_MASK;
-		I915_WRITE(MIPI_CTRL(port), temp |
-			   intel_dsi->escape_clk_div <<
-			   ESCAPE_CLOCK_DIVIDER_SHIFT);
-
+		intel_dsi_reset_clocks(encoder, port);
 		I915_WRITE(MIPI_EOT_DISABLE(port), CLOCKSTOP);
 
 		temp = I915_READ(MIPI_DSI_FUNC_PRG(port));
@@ -578,10 +576,12 @@ static void intel_dsi_disable(struct intel_encoder *encoder)
 
 static void intel_dsi_clear_device_ready(struct intel_encoder *encoder)
 {
+	struct drm_device *dev = encoder->base.dev;
 	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
 	enum port port;
 	u32 val;
+	u32 port_ctrl = 0;
 
 	DRM_DEBUG_KMS("\n");
 	for_each_dsi_port(port, intel_dsi->ports) {
@@ -598,18 +598,22 @@ static void intel_dsi_clear_device_ready(struct intel_encoder *encoder)
 							ULPS_STATE_ENTER);
 		usleep_range(2000, 2500);
 
+		if (IS_BROXTON(dev))
+			port_ctrl = BXT_MIPI_PORT_CTRL(port);
+		else if (IS_VALLEYVIEW(dev))
+			/* Common bit for both MIPI Port A & MIPI Port C */
+			port_ctrl = MIPI_PORT_CTRL(PORT_A);
+
 		/* Wait till Clock lanes are in LP-00 state for MIPI Port A
 		 * only. MIPI Port C has no similar bit for checking
 		 */
-		if (wait_for(((I915_READ(MIPI_PORT_CTRL(PORT_A)) & AFE_LATCHOUT)
-							== 0x00000), 30))
+		if (wait_for(((I915_READ(port_ctrl) & AFE_LATCHOUT)
+						== 0x00000), 30))
 			DRM_ERROR("DSI LP not going Low\n");
 
-		/* Disable MIPI PHY transparent latch
-		 * Common bit for both MIPI Port A & MIPI Port C
-		 */
-		val = I915_READ(MIPI_PORT_CTRL(PORT_A));
-		I915_WRITE(MIPI_PORT_CTRL(PORT_A), val & ~LP_OUTPUT_HOLD);
+		/* Disable MIPI PHY transparent latch */
+		val = I915_READ(port_ctrl);
+		I915_WRITE(port_ctrl, val & ~LP_OUTPUT_HOLD);
 		usleep_range(1000, 1500);
 
 		I915_WRITE(MIPI_DEVICE_READY(port), 0x00);
diff --git a/drivers/gpu/drm/i915/intel_dsi.h b/drivers/gpu/drm/i915/intel_dsi.h
index 5cc46b4..797a612 100644
--- a/drivers/gpu/drm/i915/intel_dsi.h
+++ b/drivers/gpu/drm/i915/intel_dsi.h
@@ -127,6 +127,8 @@ static inline struct intel_dsi *enc_to_intel_dsi(struct drm_encoder *encoder)
 extern void intel_enable_dsi_pll(struct intel_encoder *encoder);
 extern void intel_disable_dsi_pll(struct intel_encoder *encoder);
 extern u32 vlv_get_dsi_pclk(struct intel_encoder *encoder, int pipe_bpp);
+extern void intel_dsi_reset_clocks(struct intel_encoder *encoder,
+							enum port port);
 
 struct drm_panel *vbt_panel_init(struct intel_dsi *intel_dsi, u16 panel_id);
 
diff --git a/drivers/gpu/drm/i915/intel_dsi_pll.c b/drivers/gpu/drm/i915/intel_dsi_pll.c
index 3863362..c9a38f4 100644
--- a/drivers/gpu/drm/i915/intel_dsi_pll.c
+++ b/drivers/gpu/drm/i915/intel_dsi_pll.c
@@ -384,6 +384,19 @@ u32 vlv_get_dsi_pclk(struct intel_encoder *encoder, int pipe_bpp)
 	return pclk;
 }
 
+void vlv_dsi_reset_clocks(struct intel_encoder *encoder, enum port port)
+{
+	u32 temp;
+	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
+	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
+
+	temp = I915_READ(MIPI_CTRL(port));
+	temp &= ~ESCAPE_CLOCK_DIVIDER_MASK;
+	I915_WRITE(MIPI_CTRL(port), temp |
+			intel_dsi->escape_clk_div <<
+			ESCAPE_CLOCK_DIVIDER_SHIFT);
+}
+
 /* Program BXT Mipi clocks and dividers */
 static void bxt_dsi_program_clocks(struct drm_device *dev, enum port port)
 {
@@ -528,3 +541,29 @@ void intel_disable_dsi_pll(struct intel_encoder *encoder)
 	else if (IS_BROXTON(dev))
 		bxt_disable_dsi_pll(encoder);
 }
+
+void bxt_dsi_reset_clocks(struct intel_encoder *encoder, enum port port)
+{
+	u32 tmp;
+	struct drm_device *dev = encoder->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	/* Clear old configurations */
+	tmp = I915_READ(BXT_MIPI_CLOCK_CTL);
+	tmp &= ~(BXT_MIPI_TX_ESCLK_FIXDIV_MASK(port));
+	tmp &= ~(BXT_MIPI_RX_ESCLK_FIXDIV_MASK(port));
+	tmp &= ~(BXT_MIPI_ESCLK_VAR_DIV_MASK(port));
+	tmp &= ~(BXT_MIPI_DPHY_DIVIDER_MASK(port));
+	I915_WRITE(BXT_MIPI_CLOCK_CTL, tmp);
+	I915_WRITE(MIPI_EOT_DISABLE(port), CLOCKSTOP);
+}
+
+void intel_dsi_reset_clocks(struct intel_encoder *encoder, enum port port)
+{
+	struct drm_device *dev = encoder->base.dev;
+
+	if (IS_BROXTON(dev))
+		bxt_dsi_reset_clocks(encoder, port);
+	else if (IS_VALLEYVIEW(dev))
+		vlv_dsi_reset_clocks(encoder, port);
+}
-- 
cgit v0.10.2


From baeac68a82f07f7c37bfc3d9624127b813a8d8b4 Mon Sep 17 00:00:00 2001
From: Shashank Sharma <shashank.sharma@intel.com>
Date: Tue, 1 Sep 2015 19:41:45 +0530
Subject: drm/i915/bxt: get_hw_state for BXT

Pick appropriate port control register (BXT or VLV), based on device.
Get the current hw state wrt Mipi port.

v2: Rebased on latest drm nightly branch.

v3: Removed the GET_DSI_PORT_CTRL Macro for consistency with earlier
    implementations as per Jani's suggestion.

Signed-off-by: Shashank Sharma <shashank.sharma@intel.com>
Signed-off-by: Uma Shankar <uma.shankar@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dsi.c b/drivers/gpu/drm/i915/intel_dsi.c
index 505501e..6591106 100644
--- a/drivers/gpu/drm/i915/intel_dsi.c
+++ b/drivers/gpu/drm/i915/intel_dsi.c
@@ -656,7 +656,7 @@ static bool intel_dsi_get_hw_state(struct intel_encoder *encoder,
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
 	struct drm_device *dev = encoder->base.dev;
 	enum intel_display_power_domain power_domain;
-	u32 dpi_enabled, func;
+	u32 dpi_enabled, func, ctrl_reg;
 	enum port port;
 
 	DRM_DEBUG_KMS("\n");
@@ -668,8 +668,9 @@ static bool intel_dsi_get_hw_state(struct intel_encoder *encoder,
 	/* XXX: this only works for one DSI output */
 	for_each_dsi_port(port, intel_dsi->ports) {
 		func = I915_READ(MIPI_DSI_FUNC_PRG(port));
-		dpi_enabled = I915_READ(MIPI_PORT_CTRL(port)) &
-							DPI_ENABLE;
+		ctrl_reg = IS_BROXTON(dev) ? BXT_MIPI_PORT_CTRL(port) :
+						MIPI_PORT_CTRL(port);
+		dpi_enabled = I915_READ(ctrl_reg) & DPI_ENABLE;
 
 		/* Due to some hardware limitations on BYT, MIPI Port C DPI
 		 * Enable bit does not get set. To check whether DSI Port C
-- 
cgit v0.10.2


From ce0c982152137789e6f09e9d8712034088adf3aa Mon Sep 17 00:00:00 2001
From: Shashank Sharma <shashank.sharma@intel.com>
Date: Tue, 1 Sep 2015 19:41:46 +0530
Subject: drm/i915/bxt: get DSI pixelclock

BXT's DSI PLL is different from that of VLV. So this patch
adds a new function to get the current DSI pixel clock based
on the PLL divider ratio and lane count.

This function is required for intel_dsi_get_config() function.

v2: Fixed Jani's review comments.

Signed-off-by: Shashank Sharma <shashank.sharma@intel.com>
Signed-off-by: Uma Shankar <uma.shankar@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dsi.c b/drivers/gpu/drm/i915/intel_dsi.c
index 6591106..170ae6f 100644
--- a/drivers/gpu/drm/i915/intel_dsi.c
+++ b/drivers/gpu/drm/i915/intel_dsi.c
@@ -695,7 +695,7 @@ static bool intel_dsi_get_hw_state(struct intel_encoder *encoder,
 static void intel_dsi_get_config(struct intel_encoder *encoder,
 				 struct intel_crtc_state *pipe_config)
 {
-	u32 pclk;
+	u32 pclk = 0;
 	DRM_DEBUG_KMS("\n");
 
 	/*
@@ -704,7 +704,11 @@ static void intel_dsi_get_config(struct intel_encoder *encoder,
 	 */
 	pipe_config->dpll_hw_state.dpll_md = 0;
 
-	pclk = vlv_get_dsi_pclk(encoder, pipe_config->pipe_bpp);
+	if (IS_BROXTON(encoder->base.dev))
+		pclk = bxt_get_dsi_pclk(encoder, pipe_config->pipe_bpp);
+	else if (IS_VALLEYVIEW(encoder->base.dev))
+		pclk = vlv_get_dsi_pclk(encoder, pipe_config->pipe_bpp);
+
 	if (!pclk)
 		return;
 
diff --git a/drivers/gpu/drm/i915/intel_dsi.h b/drivers/gpu/drm/i915/intel_dsi.h
index 797a612..e6cb252 100644
--- a/drivers/gpu/drm/i915/intel_dsi.h
+++ b/drivers/gpu/drm/i915/intel_dsi.h
@@ -127,6 +127,7 @@ static inline struct intel_dsi *enc_to_intel_dsi(struct drm_encoder *encoder)
 extern void intel_enable_dsi_pll(struct intel_encoder *encoder);
 extern void intel_disable_dsi_pll(struct intel_encoder *encoder);
 extern u32 vlv_get_dsi_pclk(struct intel_encoder *encoder, int pipe_bpp);
+extern u32 bxt_get_dsi_pclk(struct intel_encoder *encoder, int pipe_bpp);
 extern void intel_dsi_reset_clocks(struct intel_encoder *encoder,
 							enum port port);
 
diff --git a/drivers/gpu/drm/i915/intel_dsi_pll.c b/drivers/gpu/drm/i915/intel_dsi_pll.c
index c9a38f4..22c1ea9 100644
--- a/drivers/gpu/drm/i915/intel_dsi_pll.c
+++ b/drivers/gpu/drm/i915/intel_dsi_pll.c
@@ -384,6 +384,41 @@ u32 vlv_get_dsi_pclk(struct intel_encoder *encoder, int pipe_bpp)
 	return pclk;
 }
 
+u32 bxt_get_dsi_pclk(struct intel_encoder *encoder, int pipe_bpp)
+{
+	u32 pclk;
+	u32 dsi_clk;
+	u32 dsi_ratio;
+	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
+	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
+
+	/* Divide by zero */
+	if (!pipe_bpp) {
+		DRM_ERROR("Invalid BPP(0)\n");
+		return 0;
+	}
+
+	dsi_ratio = I915_READ(BXT_DSI_PLL_CTL) &
+				BXT_DSI_PLL_RATIO_MASK;
+
+	/* Invalid DSI ratio ? */
+	if (dsi_ratio < BXT_DSI_PLL_RATIO_MIN ||
+			dsi_ratio > BXT_DSI_PLL_RATIO_MAX) {
+		DRM_ERROR("Invalid DSI pll ratio(%u) programmed\n", dsi_ratio);
+		return 0;
+	}
+
+	dsi_clk = (dsi_ratio * BXT_REF_CLOCK_KHZ) / 2;
+
+	/* pixel_format and pipe_bpp should agree */
+	assert_bpp_mismatch(intel_dsi->pixel_format, pipe_bpp);
+
+	pclk = DIV_ROUND_CLOSEST(dsi_clk * intel_dsi->lane_count, pipe_bpp);
+
+	DRM_DEBUG_DRIVER("Calculated pclk=%u\n", pclk);
+	return pclk;
+}
+
 void vlv_dsi_reset_clocks(struct intel_encoder *encoder, enum port port)
 {
 	u32 temp;
-- 
cgit v0.10.2


From 022e4e52a750066047b22031733df70e136ae299 Mon Sep 17 00:00:00 2001
From: Sunil Kamath <sunil.kamath@intel.com>
Date: Wed, 30 Sep 2015 22:34:57 +0530
Subject: drm/i915/bxt: Modify BXT BLC according to VBT changes

Latest VBT mentions which set of registers will be used for BLC,
as controller number field. Making use of this field in BXT
BLC implementation. Also, the registers are used in case control
pin indicates display DDI. Adding a check for this.
According to Bspec, BLC_PWM_*_2 uses the display utility pin for output.
To use backlight 2, enable the utility pin with mode = PWM
   v2: Jani's review comments
   addressed
       - Add a prefix _ to BXT BLC registers definitions.
       - Add "bxt only" comment for u8 controller
       - Remove control_pin check for DDI controller
       - Check for valid controller values
       - Set pipe bits in UTIL_PIN_CTL
       - Enable/Disable UTIL_PIN_CTL in enable/disable_backlight()
       - If BLC 2 is used, read active_low_pwm from UTIL_PIN polarity
   Satheesh's review comment addressed
       - If UTIL PIN is already enabled, BIOS would have programmed it. No
       need to disable and enable again.
   v3: Jani's review comments
       - add UTIL_PIN_PIPE_MASK and UTIL_PIN_MODE_MASK
       - Disable UTIL_PIN if controller 1 is used
       - Mask out UTIL_PIN_PIPE_MASK and UTIL_PIN_MODE_MASK before enabling
       UTIL_PIN
       - check valid controller value in intel_bios.c
       - add backlight.util_pin_active_low
       - disable util pin before enabling
   v4: Change for BXT-PO branch:
   Stubbed unwanted definition which was existing before
   because of DC6 patch.
   UTIL_PIN_MODE_PWM     (0x1b << 24)

v2: Fixed Jani's review comment.

v3: Split the backight PWM frequency programming into separate patch,
    in cases BIOS doesn't initializes it.

v4: Starting afresh and not modifying existing state for backlight, as
    per Jani's recommendation.

v5: Fixed Jani's review comment wrt util pin enable

Signed-off-by: Vandana Kannan <vandana.kannan@intel.com>
Signed-off-by: Sunil Kamath <sunil.kamath@intel.com>
Signed-off-by: Uma Shankar <uma.shankar@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 8c24851..8d7be66 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -3627,17 +3627,29 @@ enum skl_disp_power_wells {
 #define UTIL_PIN_CTL		0x48400
 #define   UTIL_PIN_ENABLE	(1 << 31)
 
+#define   UTIL_PIN_PIPE(x)     ((x) << 29)
+#define   UTIL_PIN_PIPE_MASK   (3 << 29)
+#define   UTIL_PIN_MODE_PWM    (1 << 24)
+#define   UTIL_PIN_MODE_MASK   (0xf << 24)
+#define   UTIL_PIN_POLARITY    (1 << 22)
+
 /* BXT backlight register definition. */
-#define BXT_BLC_PWM_CTL1			0xC8250
+#define _BXT_BLC_PWM_CTL1			0xC8250
 #define   BXT_BLC_PWM_ENABLE			(1 << 31)
 #define   BXT_BLC_PWM_POLARITY			(1 << 29)
-#define BXT_BLC_PWM_FREQ1			0xC8254
-#define BXT_BLC_PWM_DUTY1			0xC8258
-
-#define BXT_BLC_PWM_CTL2			0xC8350
-#define BXT_BLC_PWM_FREQ2			0xC8354
-#define BXT_BLC_PWM_DUTY2			0xC8358
-
+#define _BXT_BLC_PWM_FREQ1			0xC8254
+#define _BXT_BLC_PWM_DUTY1			0xC8258
+
+#define _BXT_BLC_PWM_CTL2			0xC8350
+#define _BXT_BLC_PWM_FREQ2			0xC8354
+#define _BXT_BLC_PWM_DUTY2			0xC8358
+
+#define BXT_BLC_PWM_CTL(controller)    _PIPE(controller, \
+					_BXT_BLC_PWM_CTL1, _BXT_BLC_PWM_CTL2)
+#define BXT_BLC_PWM_FREQ(controller)   _PIPE(controller, \
+					_BXT_BLC_PWM_FREQ1, _BXT_BLC_PWM_FREQ2)
+#define BXT_BLC_PWM_DUTY(controller)   _PIPE(controller, \
+					_BXT_BLC_PWM_DUTY1, _BXT_BLC_PWM_DUTY2)
 
 #define PCH_GTC_CTL		0xe7000
 #define   PCH_GTC_ENABLE	(1 << 31)
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index dfd2d10..563c4f1 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -179,6 +179,8 @@ struct intel_panel {
 		bool active_low_pwm;
 
 		/* PWM chip */
+		bool util_pin_active_low;	/* bxt+ */
+		u8 controller;		/* bxt+ only */
 		struct pwm_device *pwm;
 
 		struct backlight_device *device;
diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c
index 4d28c7b..f30c996 100644
--- a/drivers/gpu/drm/i915/intel_panel.c
+++ b/drivers/gpu/drm/i915/intel_panel.c
@@ -536,9 +536,10 @@ static u32 vlv_get_backlight(struct intel_connector *connector)
 static u32 bxt_get_backlight(struct intel_connector *connector)
 {
 	struct drm_device *dev = connector->base.dev;
+	struct intel_panel *panel = &connector->panel;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	return I915_READ(BXT_BLC_PWM_DUTY1);
+	return I915_READ(BXT_BLC_PWM_DUTY(panel->backlight.controller));
 }
 
 static u32 pwm_get_backlight(struct intel_connector *connector)
@@ -634,8 +635,9 @@ static void bxt_set_backlight(struct intel_connector *connector, u32 level)
 {
 	struct drm_device *dev = connector->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_panel *panel = &connector->panel;
 
-	I915_WRITE(BXT_BLC_PWM_DUTY1, level);
+	I915_WRITE(BXT_BLC_PWM_DUTY(panel->backlight.controller), level);
 }
 
 static void pwm_set_backlight(struct intel_connector *connector, u32 level)
@@ -786,12 +788,20 @@ static void bxt_disable_backlight(struct intel_connector *connector)
 {
 	struct drm_device *dev = connector->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 tmp;
+	struct intel_panel *panel = &connector->panel;
+	u32 tmp, val;
 
 	intel_panel_actually_set_backlight(connector, 0);
 
-	tmp = I915_READ(BXT_BLC_PWM_CTL1);
-	I915_WRITE(BXT_BLC_PWM_CTL1, tmp & ~BXT_BLC_PWM_ENABLE);
+	tmp = I915_READ(BXT_BLC_PWM_CTL(panel->backlight.controller));
+	I915_WRITE(BXT_BLC_PWM_CTL(panel->backlight.controller),
+			tmp & ~BXT_BLC_PWM_ENABLE);
+
+	if (panel->backlight.controller == 1) {
+		val = I915_READ(UTIL_PIN_CTL);
+		val &= ~UTIL_PIN_ENABLE;
+		I915_WRITE(UTIL_PIN_CTL, val);
+	}
 }
 
 static void pwm_disable_backlight(struct intel_connector *connector)
@@ -1023,16 +1033,38 @@ static void bxt_enable_backlight(struct intel_connector *connector)
 	struct drm_device *dev = connector->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_panel *panel = &connector->panel;
-	u32 pwm_ctl;
+	enum pipe pipe = intel_get_pipe_from_connector(connector);
+	u32 pwm_ctl, val;
+
+	/* To use 2nd set of backlight registers, utility pin has to be
+	 * enabled with PWM mode.
+	 * The field should only be changed when the utility pin is disabled
+	 */
+	if (panel->backlight.controller == 1) {
+		val = I915_READ(UTIL_PIN_CTL);
+		if (val & UTIL_PIN_ENABLE) {
+			DRM_DEBUG_KMS("util pin already enabled\n");
+			val &= ~UTIL_PIN_ENABLE;
+			I915_WRITE(UTIL_PIN_CTL, val);
+		}
 
-	pwm_ctl = I915_READ(BXT_BLC_PWM_CTL1);
+		val = 0;
+		if (panel->backlight.util_pin_active_low)
+			val |= UTIL_PIN_POLARITY;
+		I915_WRITE(UTIL_PIN_CTL, val | UTIL_PIN_PIPE(pipe) |
+				UTIL_PIN_MODE_PWM | UTIL_PIN_ENABLE);
+	}
+
+	pwm_ctl = I915_READ(BXT_BLC_PWM_CTL(panel->backlight.controller));
 	if (pwm_ctl & BXT_BLC_PWM_ENABLE) {
 		DRM_DEBUG_KMS("backlight already enabled\n");
 		pwm_ctl &= ~BXT_BLC_PWM_ENABLE;
-		I915_WRITE(BXT_BLC_PWM_CTL1, pwm_ctl);
+		I915_WRITE(BXT_BLC_PWM_CTL(panel->backlight.controller),
+				pwm_ctl);
 	}
 
-	I915_WRITE(BXT_BLC_PWM_FREQ1, panel->backlight.max);
+	I915_WRITE(BXT_BLC_PWM_FREQ(panel->backlight.controller),
+			panel->backlight.max);
 
 	intel_panel_actually_set_backlight(connector, panel->backlight.level);
 
@@ -1040,9 +1072,10 @@ static void bxt_enable_backlight(struct intel_connector *connector)
 	if (panel->backlight.active_low_pwm)
 		pwm_ctl |= BXT_BLC_PWM_POLARITY;
 
-	I915_WRITE(BXT_BLC_PWM_CTL1, pwm_ctl);
-	POSTING_READ(BXT_BLC_PWM_CTL1);
-	I915_WRITE(BXT_BLC_PWM_CTL1, pwm_ctl | BXT_BLC_PWM_ENABLE);
+	I915_WRITE(BXT_BLC_PWM_CTL(panel->backlight.controller), pwm_ctl);
+	POSTING_READ(BXT_BLC_PWM_CTL(panel->backlight.controller));
+	I915_WRITE(BXT_BLC_PWM_CTL(panel->backlight.controller),
+			pwm_ctl | BXT_BLC_PWM_ENABLE);
 }
 
 static void pwm_enable_backlight(struct intel_connector *connector)
@@ -1562,10 +1595,28 @@ bxt_setup_backlight(struct intel_connector *connector, enum pipe unused)
 	struct intel_panel *panel = &connector->panel;
 	u32 pwm_ctl, val;
 
-	pwm_ctl = I915_READ(BXT_BLC_PWM_CTL1);
-	panel->backlight.active_low_pwm = pwm_ctl & BXT_BLC_PWM_POLARITY;
+	/*
+	 * For BXT hard coding the Backlight controller to 0.
+	 * TODO : Read the controller value from VBT and generalize
+	 */
+	panel->backlight.controller = 0;
 
-	panel->backlight.max = I915_READ(BXT_BLC_PWM_FREQ1);
+	pwm_ctl = I915_READ(BXT_BLC_PWM_CTL(panel->backlight.controller));
+
+	/* Keeping the check if controller 1 is to be programmed.
+	 * This will come into affect once the VBT parsing
+	 * is fixed for controller selection, and controller 1 is used
+	 * for a prticular display configuration.
+	 */
+	if (panel->backlight.controller == 1) {
+		val = I915_READ(UTIL_PIN_CTL);
+		panel->backlight.util_pin_active_low =
+					val & UTIL_PIN_POLARITY;
+	}
+
+	panel->backlight.active_low_pwm = pwm_ctl & BXT_BLC_PWM_POLARITY;
+	panel->backlight.max =
+		I915_READ(BXT_BLC_PWM_FREQ(panel->backlight.controller));
 
 	if (!panel->backlight.max)
 		panel->backlight.max = get_backlight_max_vbt(connector);
-- 
cgit v0.10.2


From 8e5de1d8c34c63331f12cf7bf7bbf115bba0317a Mon Sep 17 00:00:00 2001
From: Lukas Wunner <lukas@wunner.de>
Date: Sat, 5 Sep 2015 11:14:43 +0200
Subject: drm/radeon: Spell vga_switcheroo consistently

Currently everyone and their dog has their own favourite spelling
for vga_switcheroo. This makes it hard to grep dmesg for log entries
relating to vga_switcheroo. It also makes it hard to find related
source files in the tree.

vga_switcheroo.c uses pr_fmt "vga_switcheroo". Use that everywhere.

Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/radeon/radeon_atpx_handler.c b/drivers/gpu/drm/radeon/radeon_atpx_handler.c
index 8bc7d0b..a771b9f 100644
--- a/drivers/gpu/drm/radeon/radeon_atpx_handler.c
+++ b/drivers/gpu/drm/radeon/radeon_atpx_handler.c
@@ -535,7 +535,7 @@ static bool radeon_atpx_detect(void)
 
 	if (has_atpx && vga_count == 2) {
 		acpi_get_name(radeon_atpx_priv.atpx.handle, ACPI_FULL_PATHNAME, &buffer);
-		printk(KERN_INFO "VGA switcheroo: detected switching method %s handle\n",
+		printk(KERN_INFO "vga_switcheroo: detected switching method %s handle\n",
 		       acpi_method_name);
 		radeon_atpx_priv.atpx_detected = true;
 		return true;
diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c
index f3f562f..c566993 100644
--- a/drivers/gpu/drm/radeon/radeon_device.c
+++ b/drivers/gpu/drm/radeon/radeon_device.c
@@ -1197,7 +1197,7 @@ static void radeon_check_arguments(struct radeon_device *rdev)
  * radeon_switcheroo_set_state - set switcheroo state
  *
  * @pdev: pci dev pointer
- * @state: vga switcheroo state
+ * @state: vga_switcheroo state
  *
  * Callback for the switcheroo driver.  Suspends or resumes the
  * the asics before or after it is powered up using ACPI methods.
diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c
index fd9da28..977f1a5 100644
--- a/drivers/gpu/drm/radeon/radeon_kms.c
+++ b/drivers/gpu/drm/radeon/radeon_kms.c
@@ -602,7 +602,7 @@ static int radeon_info_ioctl(struct drm_device *dev, void *data, struct drm_file
  *
  * @dev: drm dev pointer
  *
- * Switch vga switcheroo state after last close (all asics).
+ * Switch vga_switcheroo state after last close (all asics).
  */
 void radeon_driver_lastclose_kms(struct drm_device *dev)
 {
-- 
cgit v0.10.2


From 1694467b96cadd1b166b3bd0365b2431a7685143 Mon Sep 17 00:00:00 2001
From: Lukas Wunner <lukas@wunner.de>
Date: Sat, 5 Sep 2015 11:17:35 +0200
Subject: drm/amdgpu: Spell vga_switcheroo consistently

Currently everyone and their dog has their own favourite spelling
for vga_switcheroo. This makes it hard to grep dmesg for log entries
relating to vga_switcheroo. It also makes it hard to find related
source files in the tree.

vga_switcheroo.c uses pr_fmt "vga_switcheroo". Use that everywhere.

Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c
index 3f7aaa4..1a6b239 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c
@@ -536,7 +536,7 @@ static bool amdgpu_atpx_detect(void)
 
 	if (has_atpx && vga_count == 2) {
 		acpi_get_name(amdgpu_atpx_priv.atpx.handle, ACPI_FULL_PATHNAME, &buffer);
-		printk(KERN_INFO "VGA switcheroo: detected switching method %s handle\n",
+		printk(KERN_INFO "vga_switcheroo: detected switching method %s handle\n",
 		       acpi_method_name);
 		amdgpu_atpx_priv.atpx_detected = true;
 		return true;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 6068d82..4256b00 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -1022,7 +1022,7 @@ static void amdgpu_check_arguments(struct amdgpu_device *adev)
  * amdgpu_switcheroo_set_state - set switcheroo state
  *
  * @pdev: pci dev pointer
- * @state: vga switcheroo state
+ * @state: vga_switcheroo state
  *
  * Callback for the switcheroo driver.  Suspends or resumes the
  * the asics before or after it is powered up using ACPI methods.
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
index 275f1c3..7823322 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
@@ -489,7 +489,7 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file
  *
  * @dev: drm dev pointer
  *
- * Switch vga switcheroo state after last close (all asics).
+ * Switch vga_switcheroo state after last close (all asics).
  */
 void amdgpu_driver_lastclose_kms(struct drm_device *dev)
 {
-- 
cgit v0.10.2


From 4a985353d461951e4ed51abc281c09a780d30058 Mon Sep 17 00:00:00 2001
From: Grazvydas Ignotas <notasas@gmail.com>
Date: Sun, 23 Aug 2015 03:57:35 +0300
Subject: drm/radeon: simplify register checker

To avoid having to distinguish between CAYMAN or older on every register
check, place a pointer in evergreen_cs_track and use it unconditionally.
Also make use of the fact that both reg_safe_bm[] arrays are of the same
length to remove another CAYMAN check.

Reviewed-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Grazvydas Ignotas <notasas@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/radeon/evergreen_cs.c b/drivers/gpu/drm/radeon/evergreen_cs.c
index c9e0fbb..5c840da 100644
--- a/drivers/gpu/drm/radeon/evergreen_cs.c
+++ b/drivers/gpu/drm/radeon/evergreen_cs.c
@@ -34,6 +34,8 @@
 #define MAX(a,b)                   (((a)>(b))?(a):(b))
 #define MIN(a,b)                   (((a)<(b))?(a):(b))
 
+#define REG_SAFE_BM_SIZE ARRAY_SIZE(evergreen_reg_safe_bm)
+
 int r600_dma_cs_next_reloc(struct radeon_cs_parser *p,
 			   struct radeon_bo_list **cs_reloc);
 struct evergreen_cs_track {
@@ -84,6 +86,7 @@ struct evergreen_cs_track {
 	u32			htile_surface;
 	struct radeon_bo	*htile_bo;
 	unsigned long		indirect_draw_buffer_size;
+	const unsigned		*reg_safe_bm;
 };
 
 static u32 evergreen_cs_get_aray_mode(u32 tiling_flags)
@@ -1096,28 +1099,17 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 {
 	struct evergreen_cs_track *track = (struct evergreen_cs_track *)p->track;
 	struct radeon_bo_list *reloc;
-	u32 last_reg;
 	u32 m, i, tmp, *ib;
 	int r;
 
-	if (p->rdev->family >= CHIP_CAYMAN)
-		last_reg = ARRAY_SIZE(cayman_reg_safe_bm);
-	else
-		last_reg = ARRAY_SIZE(evergreen_reg_safe_bm);
-
 	i = (reg >> 7);
-	if (i >= last_reg) {
+	if (unlikely(i >= REG_SAFE_BM_SIZE)) {
 		dev_warn(p->dev, "forbidden register 0x%08x at %d\n", reg, idx);
 		return -EINVAL;
 	}
 	m = 1 << ((reg >> 2) & 31);
-	if (p->rdev->family >= CHIP_CAYMAN) {
-		if (!(cayman_reg_safe_bm[i] & m))
-			return 0;
-	} else {
-		if (!(evergreen_reg_safe_bm[i] & m))
-			return 0;
-	}
+	if (!(track->reg_safe_bm[i] & m))
+		return 0;
 	ib = p->ib.ptr;
 	switch (reg) {
 	/* force following reg to 0 in an attempt to disable out buffer
@@ -1766,26 +1758,17 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 
 static bool evergreen_is_safe_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 {
-	u32 last_reg, m, i;
-
-	if (p->rdev->family >= CHIP_CAYMAN)
-		last_reg = ARRAY_SIZE(cayman_reg_safe_bm);
-	else
-		last_reg = ARRAY_SIZE(evergreen_reg_safe_bm);
+	struct evergreen_cs_track *track = p->track;
+	u32 m, i;
 
 	i = (reg >> 7);
-	if (i >= last_reg) {
+	if (unlikely(i >= REG_SAFE_BM_SIZE)) {
 		dev_warn(p->dev, "forbidden register 0x%08x at %d\n", reg, idx);
 		return false;
 	}
 	m = 1 << ((reg >> 2) & 31);
-	if (p->rdev->family >= CHIP_CAYMAN) {
-		if (!(cayman_reg_safe_bm[i] & m))
-			return true;
-	} else {
-		if (!(evergreen_reg_safe_bm[i] & m))
-			return true;
-	}
+	if (!(track->reg_safe_bm[i] & m))
+		return true;
 	dev_warn(p->dev, "forbidden register 0x%08x at %d\n", reg, idx);
 	return false;
 }
@@ -2644,11 +2627,15 @@ int evergreen_cs_parse(struct radeon_cs_parser *p)
 		if (track == NULL)
 			return -ENOMEM;
 		evergreen_cs_track_init(track);
-		if (p->rdev->family >= CHIP_CAYMAN)
+		if (p->rdev->family >= CHIP_CAYMAN) {
 			tmp = p->rdev->config.cayman.tile_config;
-		else
+			track->reg_safe_bm = cayman_reg_safe_bm;
+		} else {
 			tmp = p->rdev->config.evergreen.tile_config;
-
+			track->reg_safe_bm = evergreen_reg_safe_bm;
+		}
+		BUILD_BUG_ON(ARRAY_SIZE(cayman_reg_safe_bm) != REG_SAFE_BM_SIZE);
+		BUILD_BUG_ON(ARRAY_SIZE(evergreen_reg_safe_bm) != REG_SAFE_BM_SIZE);
 		switch (tmp & 0xf) {
 		case 0:
 			track->npipes = 1;
-- 
cgit v0.10.2


From e5b69da6badca8286a67c1d0d1c6634e65c7e669 Mon Sep 17 00:00:00 2001
From: Grazvydas Ignotas <notasas@gmail.com>
Date: Sun, 23 Aug 2015 03:57:36 +0300
Subject: drm/radeon: split evergreen_cs_check_reg

evergreen_cs_check_reg() is a large function and gcc doesn't want to
inline it. It has a quick check for reg_safe_bm[] to see if register
needs special handling, which often results in early exit. However
because the function is large, it has a long prologue/epilogue to
save/restore all the callee-save registers which according to perf is
taking significant amount of time. To avoid this, we can reuse
evergreen_is_safe_reg() to do the early check directly in register loop.

Reviewed-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Grazvydas Ignotas <notasas@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/radeon/evergreen_cs.c b/drivers/gpu/drm/radeon/evergreen_cs.c
index 5c840da..4453799 100644
--- a/drivers/gpu/drm/radeon/evergreen_cs.c
+++ b/drivers/gpu/drm/radeon/evergreen_cs.c
@@ -1086,30 +1086,18 @@ static int evergreen_cs_parse_packet0(struct radeon_cs_parser *p,
 }
 
 /**
- * evergreen_cs_check_reg() - check if register is authorized or not
+ * evergreen_cs_handle_reg() - process registers that need special handling.
  * @parser: parser structure holding parsing context
  * @reg: register we are testing
  * @idx: index into the cs buffer
- *
- * This function will test against evergreen_reg_safe_bm and return 0
- * if register is safe. If register is not flag as safe this function
- * will test it against a list of register needind special handling.
  */
-static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
+static int evergreen_cs_handle_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 {
 	struct evergreen_cs_track *track = (struct evergreen_cs_track *)p->track;
 	struct radeon_bo_list *reloc;
-	u32 m, i, tmp, *ib;
+	u32 tmp, *ib;
 	int r;
 
-	i = (reg >> 7);
-	if (unlikely(i >= REG_SAFE_BM_SIZE)) {
-		dev_warn(p->dev, "forbidden register 0x%08x at %d\n", reg, idx);
-		return -EINVAL;
-	}
-	m = 1 << ((reg >> 2) & 31);
-	if (!(track->reg_safe_bm[i] & m))
-		return 0;
 	ib = p->ib.ptr;
 	switch (reg) {
 	/* force following reg to 0 in an attempt to disable out buffer
@@ -1756,20 +1744,27 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	return 0;
 }
 
-static bool evergreen_is_safe_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
+/**
+ * evergreen_is_safe_reg() - check if register is authorized or not
+ * @parser: parser structure holding parsing context
+ * @reg: register we are testing
+ *
+ * This function will test against reg_safe_bm and return true
+ * if register is safe or false otherwise.
+ */
+static inline bool evergreen_is_safe_reg(struct radeon_cs_parser *p, u32 reg)
 {
 	struct evergreen_cs_track *track = p->track;
 	u32 m, i;
 
 	i = (reg >> 7);
 	if (unlikely(i >= REG_SAFE_BM_SIZE)) {
-		dev_warn(p->dev, "forbidden register 0x%08x at %d\n", reg, idx);
 		return false;
 	}
 	m = 1 << ((reg >> 2) & 31);
 	if (!(track->reg_safe_bm[i] & m))
 		return true;
-	dev_warn(p->dev, "forbidden register 0x%08x at %d\n", reg, idx);
+
 	return false;
 }
 
@@ -2306,7 +2301,9 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		}
 		for (i = 0; i < pkt->count; i++) {
 			reg = start_reg + (4 * i);
-			r = evergreen_cs_check_reg(p, reg, idx+1+i);
+			if (evergreen_is_safe_reg(p, reg))
+				continue;
+			r = evergreen_cs_handle_reg(p, reg, idx + 1 + i);
 			if (r)
 				return r;
 		}
@@ -2322,7 +2319,9 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		}
 		for (i = 0; i < pkt->count; i++) {
 			reg = start_reg + (4 * i);
-			r = evergreen_cs_check_reg(p, reg, idx+1+i);
+			if (evergreen_is_safe_reg(p, reg))
+				continue;
+			r = evergreen_cs_handle_reg(p, reg, idx + 1 + i);
 			if (r)
 				return r;
 		}
@@ -2577,8 +2576,11 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		} else {
 			/* SRC is a reg. */
 			reg = radeon_get_ib_value(p, idx+1) << 2;
-			if (!evergreen_is_safe_reg(p, reg, idx+1))
+			if (!evergreen_is_safe_reg(p, reg)) {
+				dev_warn(p->dev, "forbidden register 0x%08x at %d\n",
+					 reg, idx + 1);
 				return -EINVAL;
+			}
 		}
 		if (idx_value & 0x2) {
 			u64 offset;
@@ -2601,8 +2603,11 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		} else {
 			/* DST is a reg. */
 			reg = radeon_get_ib_value(p, idx+3) << 2;
-			if (!evergreen_is_safe_reg(p, reg, idx+3))
+			if (!evergreen_is_safe_reg(p, reg)) {
+				dev_warn(p->dev, "forbidden register 0x%08x at %d\n",
+					 reg, idx + 3);
 				return -EINVAL;
+			}
 		}
 		break;
 	case PACKET3_NOP:
-- 
cgit v0.10.2


From 7874d3901d752dc9ea31bbb32f519bc9a2c1b6aa Mon Sep 17 00:00:00 2001
From: Grazvydas Ignotas <notasas@gmail.com>
Date: Sun, 23 Aug 2015 03:57:37 +0300
Subject: drm/radeon: refactor register check loop

After this patch the register check loop does the same thing as before,
except that now gcc does better job optimizing it: it now sees that
end_reg was already checked against PACKET3_SET_CONTEXT_REG_END and can
optimize REG_SAFE_BM_SIZE comparison out of evergreen_is_safe_reg()
as (PACKET3_SET_CONTEXT_REG_END >> 7) < REG_SAFE_BM_SIZE.

Reviewed-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Grazvydas Ignotas <notasas@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/radeon/evergreen_cs.c b/drivers/gpu/drm/radeon/evergreen_cs.c
index 4453799..e31076e 100644
--- a/drivers/gpu/drm/radeon/evergreen_cs.c
+++ b/drivers/gpu/drm/radeon/evergreen_cs.c
@@ -2299,11 +2299,10 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			DRM_ERROR("bad PACKET3_SET_CONFIG_REG\n");
 			return -EINVAL;
 		}
-		for (i = 0; i < pkt->count; i++) {
-			reg = start_reg + (4 * i);
+		for (reg = start_reg, idx++; reg <= end_reg; reg += 4, idx++) {
 			if (evergreen_is_safe_reg(p, reg))
 				continue;
-			r = evergreen_cs_handle_reg(p, reg, idx + 1 + i);
+			r = evergreen_cs_handle_reg(p, reg, idx);
 			if (r)
 				return r;
 		}
@@ -2317,11 +2316,10 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			DRM_ERROR("bad PACKET3_SET_CONTEXT_REG\n");
 			return -EINVAL;
 		}
-		for (i = 0; i < pkt->count; i++) {
-			reg = start_reg + (4 * i);
+		for (reg = start_reg, idx++; reg <= end_reg; reg += 4, idx++) {
 			if (evergreen_is_safe_reg(p, reg))
 				continue;
-			r = evergreen_cs_handle_reg(p, reg, idx + 1 + i);
+			r = evergreen_cs_handle_reg(p, reg, idx);
 			if (r)
 				return r;
 		}
-- 
cgit v0.10.2


From 1c5dc33514002395cf07fa818b95ac4ac21364db Mon Sep 17 00:00:00 2001
From: Grazvydas Ignotas <notasas@gmail.com>
Date: Sun, 23 Aug 2015 03:57:38 +0300
Subject: drm/radeon: remove volatile qualifier

There doesn't seem to be any need to have 'ib' volatile, the code is
not even consistent with it and some places already miss it. As it is
now it's just making gcc produce worse code. If there are special
requirements for that memory, then proper primitives like memory
barriers or accessor functions should be used, but it doesn't look
like that is needed here.
While at it, change the type to match the one in radeon_ib structure.

Reviewed-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Grazvydas Ignotas <notasas@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/radeon/evergreen_cs.c b/drivers/gpu/drm/radeon/evergreen_cs.c
index e31076e..46f87d4 100644
--- a/drivers/gpu/drm/radeon/evergreen_cs.c
+++ b/drivers/gpu/drm/radeon/evergreen_cs.c
@@ -447,7 +447,7 @@ static int evergreen_cs_track_validate_cb(struct radeon_cs_parser *p, unsigned i
 		 * command stream.
 		 */
 		if (!surf.mode) {
-			volatile u32 *ib = p->ib.ptr;
+			uint32_t *ib = p->ib.ptr;
 			unsigned long tmp, nby, bsize, size, min = 0;
 
 			/* find the height the ddx wants */
@@ -1773,7 +1773,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 {
 	struct radeon_bo_list *reloc;
 	struct evergreen_cs_track *track;
-	volatile u32 *ib;
+	uint32_t *ib;
 	unsigned idx;
 	unsigned i;
 	unsigned start_reg, end_reg, reg;
@@ -2747,7 +2747,7 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
 	struct radeon_cs_chunk *ib_chunk = p->chunk_ib;
 	struct radeon_bo_list *src_reloc, *dst_reloc, *dst2_reloc;
 	u32 header, cmd, count, sub_cmd;
-	volatile u32 *ib = p->ib.ptr;
+	uint32_t *ib = p->ib.ptr;
 	u32 idx;
 	u64 src_offset, dst_offset, dst2_offset;
 	int r;
-- 
cgit v0.10.2


From 068127601eefa0565f4b2f4d82750cad9e37000c Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@oracle.com>
Date: Fri, 2 Oct 2015 18:14:22 +0300
Subject: drm/i915: unlock on error in i915_ppgtt_info()

We need to call intel_runtime_pm_put() and mutex_unlock() before
returning.

Fixes: 7cb5dff8d59d ('drm/i915: fix task reference leak in i915_debugfs.c')
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Acked-by: Geliang Tang <geliangtang@163.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index 77ee8c5..7e65015 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -2298,18 +2298,21 @@ static int i915_ppgtt_info(struct seq_file *m, void *data)
 		struct task_struct *task;
 
 		task = get_pid_task(file->pid, PIDTYPE_PID);
-		if (!task)
-			return -ESRCH;
+		if (!task) {
+			ret = -ESRCH;
+			goto out_put;
+		}
 		seq_printf(m, "\nproc: %s\n", task->comm);
 		put_task_struct(task);
 		idr_for_each(&file_priv->context_idr, per_file_ctx,
 			     (void *)(unsigned long)m);
 	}
 
+out_put:
 	intel_runtime_pm_put(dev_priv);
 	mutex_unlock(&dev->struct_mutex);
 
-	return 0;
+	return ret;
 }
 
 static int count_irq_waiters(struct drm_i915_private *i915)
-- 
cgit v0.10.2


From 6220907089cc3eb4ab2fa7073bbf617b019666c5 Mon Sep 17 00:00:00 2001
From: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Date: Thu, 1 Oct 2015 10:00:57 +0300
Subject: drm: Add DRM_ROTATE_MASK and DRM_REFLECT_MASK
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Makes it cleaner to separate the two from rotation variable.

Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/include/drm/drm_crtc.h b/include/drm/drm_crtc.h
index 683f142..33ddedd 100644
--- a/include/drm/drm_crtc.h
+++ b/include/drm/drm_crtc.h
@@ -86,10 +86,12 @@ static inline uint64_t I642U64(int64_t val)
 }
 
 /* rotation property bits */
+#define DRM_ROTATE_MASK 0x0f
 #define DRM_ROTATE_0	0
 #define DRM_ROTATE_90	1
 #define DRM_ROTATE_180	2
 #define DRM_ROTATE_270	3
+#define DRM_REFLECT_MASK (~DRM_ROTATE_MASK)
 #define DRM_REFLECT_X	4
 #define DRM_REFLECT_Y	5
 
-- 
cgit v0.10.2


From 14152c8d30bbe5155d0438feb0a1931aee0e5bce Mon Sep 17 00:00:00 2001
From: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Date: Thu, 1 Oct 2015 10:00:58 +0300
Subject: drm: Use DRM_ROTATE_MASK and DRM_REFLECT_MASK
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Avoid magic numbers and use the introduced defines.

Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_plane.c b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_plane.c
index 36fda86..d0299ae 100644
--- a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_plane.c
+++ b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_plane.c
@@ -633,7 +633,7 @@ static int atmel_hlcdc_plane_atomic_check(struct drm_plane *p,
 		if (!state->bpp[i])
 			return -EINVAL;
 
-		switch (state->base.rotation & 0xf) {
+		switch (state->base.rotation & DRM_ROTATE_MASK) {
 		case BIT(DRM_ROTATE_90):
 			offset = ((y_offset + state->src_y + patched_src_w - 1) /
 				  ydiv) * fb->pitches[i];
diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c
index e600a5f..e7c8422 100644
--- a/drivers/gpu/drm/drm_crtc.c
+++ b/drivers/gpu/drm/drm_crtc.c
@@ -5629,7 +5629,8 @@ unsigned int drm_rotation_simplify(unsigned int rotation,
 {
 	if (rotation & ~supported_rotations) {
 		rotation ^= BIT(DRM_REFLECT_X) | BIT(DRM_REFLECT_Y);
-		rotation = (rotation & ~0xf) | BIT((ffs(rotation & 0xf) + 1) % 4);
+		rotation = (rotation & DRM_REFLECT_MASK) |
+		           BIT((ffs(rotation & DRM_ROTATE_MASK) + 1) % 4);
 	}
 
 	return rotation;
diff --git a/drivers/gpu/drm/drm_rect.c b/drivers/gpu/drm/drm_rect.c
index 631f5af..531ac4c 100644
--- a/drivers/gpu/drm/drm_rect.c
+++ b/drivers/gpu/drm/drm_rect.c
@@ -330,7 +330,7 @@ void drm_rect_rotate(struct drm_rect *r,
 		}
 	}
 
-	switch (rotation & 0xf) {
+	switch (rotation & DRM_ROTATE_MASK) {
 	case BIT(DRM_ROTATE_0):
 		break;
 	case BIT(DRM_ROTATE_90):
@@ -390,7 +390,7 @@ void drm_rect_rotate_inv(struct drm_rect *r,
 {
 	struct drm_rect tmp;
 
-	switch (rotation & 0xf) {
+	switch (rotation & DRM_ROTATE_MASK) {
 	case BIT(DRM_ROTATE_0):
 		break;
 	case BIT(DRM_ROTATE_90):
diff --git a/drivers/gpu/drm/omapdrm/omap_fb.c b/drivers/gpu/drm/omapdrm/omap_fb.c
index 51b1219..636a1f9 100644
--- a/drivers/gpu/drm/omapdrm/omap_fb.c
+++ b/drivers/gpu/drm/omapdrm/omap_fb.c
@@ -171,7 +171,7 @@ void omap_framebuffer_update_scanout(struct drm_framebuffer *fb,
 		uint32_t w = win->src_w;
 		uint32_t h = win->src_h;
 
-		switch (win->rotation & 0xf) {
+		switch (win->rotation & DRM_ROTATE_MASK) {
 		default:
 			dev_err(fb->dev->dev, "invalid rotation: %02x",
 					(uint32_t)win->rotation);
@@ -209,7 +209,7 @@ void omap_framebuffer_update_scanout(struct drm_framebuffer *fb,
 		info->rotation_type = OMAP_DSS_ROT_TILER;
 		info->screen_width  = omap_gem_tiled_stride(plane->bo, orient);
 	} else {
-		switch (win->rotation & 0xf) {
+		switch (win->rotation & DRM_ROTATE_MASK) {
 		case 0:
 		case BIT(DRM_ROTATE_0):
 			/* OK */
diff --git a/drivers/gpu/drm/omapdrm/omap_plane.c b/drivers/gpu/drm/omapdrm/omap_plane.c
index 09e363b..3054bda 100644
--- a/drivers/gpu/drm/omapdrm/omap_plane.c
+++ b/drivers/gpu/drm/omapdrm/omap_plane.c
@@ -108,7 +108,7 @@ static void omap_plane_atomic_update(struct drm_plane *plane,
 	win.src_x = state->src_x >> 16;
 	win.src_y = state->src_y >> 16;
 
-	switch (state->rotation & 0xf) {
+	switch (state->rotation & DRM_ROTATE_MASK) {
 	case BIT(DRM_ROTATE_90):
 	case BIT(DRM_ROTATE_270):
 		win.src_w = state->src_h >> 16;
-- 
cgit v0.10.2


From 5d250b05918c002b63632c7db91c3c5f924c6a3b Mon Sep 17 00:00:00 2001
From: Sonika Jindal <sonika.jindal@intel.com>
Date: Mon, 5 Oct 2015 16:43:14 +0530
Subject: drm/i915: Call encoder hotplug for init and resume cases

For all the encoders, call the hot_plug if it is registered.
This is required for connected boot and resume cases to generate
fake hpd resulting in reading of edid.
Removing the initial sdvo hot_plug call too so that it will be called
just once from this loop.

Signed-off-by: Sonika Jindal <sonika.jindal@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_hotplug.c b/drivers/gpu/drm/i915/intel_hotplug.c
index 53c0173..eac4757 100644
--- a/drivers/gpu/drm/i915/intel_hotplug.c
+++ b/drivers/gpu/drm/i915/intel_hotplug.c
@@ -458,6 +458,7 @@ void intel_hpd_init(struct drm_i915_private *dev_priv)
 {
 	struct drm_device *dev = dev_priv->dev;
 	struct drm_mode_config *mode_config = &dev->mode_config;
+	struct intel_encoder *encoder;
 	struct drm_connector *connector;
 	int i;
 
@@ -482,6 +483,16 @@ void intel_hpd_init(struct drm_i915_private *dev_priv)
 	if (dev_priv->display.hpd_irq_setup)
 		dev_priv->display.hpd_irq_setup(dev);
 	spin_unlock_irq(&dev_priv->irq_lock);
+
+	/*
+	 * Connected boot / resume scenarios can't generate new hot plug.
+	 * So, probe it manually.
+	 */
+	list_for_each_entry(encoder, &dev->mode_config.encoder_list,
+			    base.head) {
+		if (encoder->hot_plug)
+			encoder->hot_plug(encoder);
+	}
 }
 
 void intel_hpd_init_work(struct drm_i915_private *dev_priv)
diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c
index c42b636..853f4b2 100644
--- a/drivers/gpu/drm/i915/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/intel_sdvo.c
@@ -2460,7 +2460,6 @@ intel_sdvo_dvi_init(struct intel_sdvo *intel_sdvo, int device)
 		 * Ensure that they get re-enabled when an interrupt happens.
 		 */
 		intel_encoder->hot_plug = intel_sdvo_enable_hotplug;
-		intel_sdvo_enable_hotplug(intel_encoder);
 	} else {
 		intel_connector->polled = DRM_CONNECTOR_POLL_CONNECT | DRM_CONNECTOR_POLL_DISCONNECT;
 	}
-- 
cgit v0.10.2


From 9571b190cf474ae02a7e018a21d14052d750070f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 8 Sep 2015 18:05:44 +0300
Subject: drm/i915: Don't bypass LRC on CHV
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The docs are unclear as usual, so it's not clear whether LRC should be
bypassed, performed normally or GRC code should be used as the LRC code.
Some old docs stated that LRC bypass ought to be used, more recent ones
no longer say that. Some docs indicated that we could use GRC as the LRC
code on CHV, but the BIOS doesn't do that, so let's not do it either.

Besides to enable LRC bypass properly, I believe we should set the bit
already before deasserting cmnreset.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Deepak S<deepak.s@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 97ed418..19cbc2e 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -3417,11 +3417,6 @@ static uint32_t chv_signal_levels(struct intel_dp *intel_dp)
 		vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW10(ch), val);
 	}
 
-	/* LRC Bypass */
-	val = vlv_dpio_read(dev_priv, pipe, CHV_CMN_DW30);
-	val |= DPIO_LRC_BYPASS;
-	vlv_dpio_write(dev_priv, pipe, CHV_CMN_DW30, val);
-
 	mutex_unlock(&dev_priv->sb_lock);
 
 	return 0;
diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index 3c5f2c9..3b28ed3 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -1952,11 +1952,6 @@ static void chv_hdmi_pre_enable(struct intel_encoder *encoder)
 	val |= DPIO_PCS_SWING_CALC_TX0_TX2 | DPIO_PCS_SWING_CALC_TX1_TX3;
 	vlv_dpio_write(dev_priv, pipe, VLV_PCS23_DW10(ch), val);
 
-	/* LRC Bypass */
-	val = vlv_dpio_read(dev_priv, pipe, CHV_CMN_DW30);
-	val |= DPIO_LRC_BYPASS;
-	vlv_dpio_write(dev_priv, pipe, CHV_CMN_DW30, val);
-
 	mutex_unlock(&dev_priv->sb_lock);
 
 	intel_hdmi->set_infoframes(&encoder->base,
-- 
cgit v0.10.2


From 3be60de9e9dc92c852c196cf4b62e287fa53963c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 8 Sep 2015 18:05:45 +0300
Subject: drm/i915: Skip CHV PHY asserts until PHY has been fully reset
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The BIOS can leave the CHV display PHY in some odd state where
some of the LDOs/lanes won't power down fully when unused. This
will trigger a host of asserts that were added in:
30142273a3e83936fd7b45aa5339311a9295ca51 drm/i915: Add CHV PHY LDO power sanity checks
6669e39f95b5530ca8cb9137703ceb5e83e5d648 drm/i915: Add some CHV DPIO lane power state asserts

To avoid that, skip the asserts until the PHY power well has been
disabled at least once. That will fully reset the PHY, and once
brought back up, the dynamic power down features will work correctly.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Deepak S<deepak.s@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 1eab9ba..35bf5cb 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -1953,6 +1953,9 @@ struct drm_i915_private {
 
 	bool edp_low_vswing;
 
+	/* perform PHY state sanity checks? */
+	bool chv_phy_assert[2];
+
 	/*
 	 * NOTE: This is the dri1/ums dungeon, don't add stuff here. Your patch
 	 * will be rejected. Instead look for a better place.
diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c
index e1fdbab..0cfe4c1 100644
--- a/drivers/gpu/drm/i915/intel_runtime_pm.c
+++ b/drivers/gpu/drm/i915/intel_runtime_pm.c
@@ -993,8 +993,29 @@ static void assert_chv_phy_status(struct drm_i915_private *dev_priv)
 		lookup_power_well(dev_priv, PUNIT_POWER_WELL_DPIO_CMN_D);
 	u32 phy_control = dev_priv->chv_phy_control;
 	u32 phy_status = 0;
+	u32 phy_status_mask = 0xffffffff;
 	u32 tmp;
 
+	/*
+	 * The BIOS can leave the PHY is some weird state
+	 * where it doesn't fully power down some parts.
+	 * Disable the asserts until the PHY has been fully
+	 * reset (ie. the power well has been disabled at
+	 * least once).
+	 */
+	if (!dev_priv->chv_phy_assert[DPIO_PHY0])
+		phy_status_mask &= ~(PHY_STATUS_CMN_LDO(DPIO_PHY0, DPIO_CH0) |
+				     PHY_STATUS_SPLINE_LDO(DPIO_PHY0, DPIO_CH0, 0) |
+				     PHY_STATUS_SPLINE_LDO(DPIO_PHY0, DPIO_CH0, 1) |
+				     PHY_STATUS_CMN_LDO(DPIO_PHY0, DPIO_CH1) |
+				     PHY_STATUS_SPLINE_LDO(DPIO_PHY0, DPIO_CH1, 0) |
+				     PHY_STATUS_SPLINE_LDO(DPIO_PHY0, DPIO_CH1, 1));
+
+	if (!dev_priv->chv_phy_assert[DPIO_PHY1])
+		phy_status_mask &= ~(PHY_STATUS_CMN_LDO(DPIO_PHY1, DPIO_CH0) |
+				     PHY_STATUS_SPLINE_LDO(DPIO_PHY1, DPIO_CH0, 0) |
+				     PHY_STATUS_SPLINE_LDO(DPIO_PHY1, DPIO_CH0, 1));
+
 	if (cmn_bc->ops->is_enabled(dev_priv, cmn_bc)) {
 		phy_status |= PHY_POWERGOOD(DPIO_PHY0);
 
@@ -1055,11 +1076,13 @@ static void assert_chv_phy_status(struct drm_i915_private *dev_priv)
 			phy_status |= PHY_STATUS_SPLINE_LDO(DPIO_PHY1, DPIO_CH0, 1);
 	}
 
+	phy_status &= phy_status_mask;
+
 	/*
 	 * The PHY may be busy with some initial calibration and whatnot,
 	 * so the power state can take a while to actually change.
 	 */
-	if (wait_for((tmp = I915_READ(DISPLAY_PHY_STATUS)) == phy_status, 10))
+	if (wait_for((tmp = I915_READ(DISPLAY_PHY_STATUS) & phy_status_mask) == phy_status, 10))
 		WARN(phy_status != tmp,
 		     "Unexpected PHY_STATUS 0x%08x, expected 0x%08x (PHY_CONTROL=0x%08x)\n",
 		     tmp, phy_status, dev_priv->chv_phy_control);
@@ -1152,6 +1175,9 @@ static void chv_dpio_cmn_power_well_disable(struct drm_i915_private *dev_priv,
 	DRM_DEBUG_KMS("Disabled DPIO PHY%d (PHY_CONTROL=0x%08x)\n",
 		      phy, dev_priv->chv_phy_control);
 
+	/* PHY is fully reset now, so we can enable the PHY state asserts */
+	dev_priv->chv_phy_assert[phy] = true;
+
 	assert_chv_phy_status(dev_priv);
 }
 
@@ -1161,6 +1187,16 @@ static void assert_chv_phy_powergate(struct drm_i915_private *dev_priv, enum dpi
 	enum pipe pipe = phy == DPIO_PHY0 ? PIPE_A : PIPE_C;
 	u32 reg, val, expected, actual;
 
+	/*
+	 * The BIOS can leave the PHY is some weird state
+	 * where it doesn't fully power down some parts.
+	 * Disable the asserts until the PHY has been fully
+	 * reset (ie. the power well has been disabled at
+	 * least once).
+	 */
+	if (!dev_priv->chv_phy_assert[phy])
+		return;
+
 	if (ch == DPIO_CH0)
 		reg = _CHV_CMN_DW0_CH0;
 	else
@@ -1916,6 +1952,10 @@ static void chv_phy_control_init(struct drm_i915_private *dev_priv)
 			PHY_CH_POWER_DOWN_OVRD(mask, DPIO_PHY0, DPIO_CH1);
 
 		dev_priv->chv_phy_control |= PHY_COM_LANE_RESET_DEASSERT(DPIO_PHY0);
+
+		dev_priv->chv_phy_assert[DPIO_PHY0] = false;
+	} else {
+		dev_priv->chv_phy_assert[DPIO_PHY0] = true;
 	}
 
 	if (cmn_d->ops->is_enabled(dev_priv, cmn_d)) {
@@ -1934,6 +1974,10 @@ static void chv_phy_control_init(struct drm_i915_private *dev_priv)
 			PHY_CH_POWER_DOWN_OVRD(mask, DPIO_PHY1, DPIO_CH0);
 
 		dev_priv->chv_phy_control |= PHY_COM_LANE_RESET_DEASSERT(DPIO_PHY1);
+
+		dev_priv->chv_phy_assert[DPIO_PHY1] = false;
+	} else {
+		dev_priv->chv_phy_assert[DPIO_PHY1] = true;
 	}
 
 	I915_WRITE(DISPLAY_PHY_CONTROL, dev_priv->chv_phy_control);
-- 
cgit v0.10.2


From a1c4199414539d7eea34e416fb64fb5178221381 Mon Sep 17 00:00:00 2001
From: Alex Dai <yu.dai@intel.com>
Date: Wed, 30 Sep 2015 09:46:37 -0700
Subject: drm/i915/guc: Add host2guc notification for suspend and resume

Add host2guc interface to notify GuC power state changes when
enter or resume from power saving state.

v3: Move intel_guc_suspend to i915_drm_suspend for consistency.

v2: Add GuC suspend/resume to runtime suspend/resume too

v1: Change to a more flexible way when fill host to GuC scratch
data in order to remove hard coding.

Signed-off-by: Alex Dai <yu.dai@intel.com>
Reviewed-by: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 1cb6b82..760e0ce 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -638,6 +638,8 @@ static int i915_drm_suspend(struct drm_device *dev)
 		return error;
 	}
 
+	intel_guc_suspend(dev);
+
 	intel_suspend_gt_powersave(dev);
 
 	/*
@@ -767,6 +769,8 @@ static int i915_drm_resume(struct drm_device *dev)
 	}
 	mutex_unlock(&dev->struct_mutex);
 
+	intel_guc_resume(dev);
+
 	intel_modeset_init_hw(dev);
 
 	spin_lock_irq(&dev_priv->irq_lock);
@@ -1500,6 +1504,8 @@ static int intel_runtime_suspend(struct device *device)
 	i915_gem_release_all_mmaps(dev_priv);
 	mutex_unlock(&dev->struct_mutex);
 
+	intel_guc_suspend(dev);
+
 	intel_suspend_gt_powersave(dev);
 	intel_runtime_pm_disable_interrupts(dev_priv);
 
@@ -1559,6 +1565,8 @@ static int intel_runtime_resume(struct device *device)
 	intel_opregion_notify_adapter(dev, PCI_D0);
 	dev_priv->pm.suspended = false;
 
+	intel_guc_resume(dev);
+
 	if (IS_GEN6(dev_priv))
 		intel_init_pch_refclk(dev);
 
diff --git a/drivers/gpu/drm/i915/i915_guc_submission.c b/drivers/gpu/drm/i915/i915_guc_submission.c
index 0b1797f..036b42b 100644
--- a/drivers/gpu/drm/i915/i915_guc_submission.c
+++ b/drivers/gpu/drm/i915/i915_guc_submission.c
@@ -923,3 +923,53 @@ void i915_guc_submission_fini(struct drm_device *dev)
 	gem_release_guc_obj(guc->ctx_pool_obj);
 	guc->ctx_pool_obj = NULL;
 }
+
+/**
+ * intel_guc_suspend() - notify GuC entering suspend state
+ * @dev:	drm device
+ */
+int intel_guc_suspend(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_guc *guc = &dev_priv->guc;
+	struct intel_context *ctx;
+	u32 data[3];
+
+	if (!i915.enable_guc_submission)
+		return 0;
+
+	ctx = dev_priv->ring[RCS].default_context;
+
+	data[0] = HOST2GUC_ACTION_ENTER_S_STATE;
+	/* any value greater than GUC_POWER_D0 */
+	data[1] = GUC_POWER_D1;
+	/* first page is shared data with GuC */
+	data[2] = i915_gem_obj_ggtt_offset(ctx->engine[RCS].state);
+
+	return host2guc_action(guc, data, ARRAY_SIZE(data));
+}
+
+
+/**
+ * intel_guc_resume() - notify GuC resuming from suspend state
+ * @dev:	drm device
+ */
+int intel_guc_resume(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_guc *guc = &dev_priv->guc;
+	struct intel_context *ctx;
+	u32 data[3];
+
+	if (!i915.enable_guc_submission)
+		return 0;
+
+	ctx = dev_priv->ring[RCS].default_context;
+
+	data[0] = HOST2GUC_ACTION_EXIT_S_STATE;
+	data[1] = GUC_POWER_D0;
+	/* first page is shared data with GuC */
+	data[2] = i915_gem_obj_ggtt_offset(ctx->engine[RCS].state);
+
+	return host2guc_action(guc, data, ARRAY_SIZE(data));
+}
diff --git a/drivers/gpu/drm/i915/intel_guc.h b/drivers/gpu/drm/i915/intel_guc.h
index 4ec2d27..081d5f6 100644
--- a/drivers/gpu/drm/i915/intel_guc.h
+++ b/drivers/gpu/drm/i915/intel_guc.h
@@ -110,6 +110,8 @@ extern void intel_guc_ucode_init(struct drm_device *dev);
 extern int intel_guc_ucode_load(struct drm_device *dev);
 extern void intel_guc_ucode_fini(struct drm_device *dev);
 extern const char *intel_guc_fw_status_repr(enum intel_guc_fw_status status);
+extern int intel_guc_suspend(struct drm_device *dev);
+extern int intel_guc_resume(struct drm_device *dev);
 
 /* i915_guc_submission.c */
 int i915_guc_submission_init(struct drm_device *dev);
diff --git a/drivers/gpu/drm/i915/intel_guc_fwif.h b/drivers/gpu/drm/i915/intel_guc_fwif.h
index 6c78fdf..593d2f5 100644
--- a/drivers/gpu/drm/i915/intel_guc_fwif.h
+++ b/drivers/gpu/drm/i915/intel_guc_fwif.h
@@ -221,12 +221,20 @@ struct guc_context_desc {
 #define GUC_FORCEWAKE_RENDER	(1 << 0)
 #define GUC_FORCEWAKE_MEDIA	(1 << 1)
 
+#define GUC_POWER_UNSPECIFIED	0
+#define GUC_POWER_D0		1
+#define GUC_POWER_D1		2
+#define GUC_POWER_D2		3
+#define GUC_POWER_D3		4
+
 /* This Action will be programmed in C180 - SOFT_SCRATCH_O_REG */
 enum host2guc_action {
 	HOST2GUC_ACTION_DEFAULT = 0x0,
 	HOST2GUC_ACTION_SAMPLE_FORCEWAKE = 0x6,
 	HOST2GUC_ACTION_ALLOCATE_DOORBELL = 0x10,
 	HOST2GUC_ACTION_DEALLOCATE_DOORBELL = 0x20,
+	HOST2GUC_ACTION_ENTER_S_STATE = 0x501,
+	HOST2GUC_ACTION_EXIT_S_STATE = 0x502,
 	HOST2GUC_ACTION_SLPC_REQUEST = 0x3003,
 	HOST2GUC_ACTION_LIMIT
 };
diff --git a/drivers/gpu/drm/i915/intel_guc_loader.c b/drivers/gpu/drm/i915/intel_guc_loader.c
index f87d833..a17b6a5 100644
--- a/drivers/gpu/drm/i915/intel_guc_loader.c
+++ b/drivers/gpu/drm/i915/intel_guc_loader.c
@@ -383,7 +383,6 @@ int intel_guc_ucode_load(struct drm_device *dev)
 		intel_guc_fw_status_repr(guc_fw->guc_fw_load_status));
 
 	direct_interrupts_to_host(dev_priv);
-	i915_guc_submission_disable(dev);
 
 	if (guc_fw->guc_fw_fetch_status == GUC_FIRMWARE_NONE)
 		return 0;
@@ -433,6 +432,9 @@ int intel_guc_ucode_load(struct drm_device *dev)
 		intel_guc_fw_status_repr(guc_fw->guc_fw_load_status));
 
 	if (i915.enable_guc_submission) {
+		/* The execbuf_client will be recreated. Release it first. */
+		i915_guc_submission_disable(dev);
+
 		err = i915_guc_submission_enable(dev);
 		if (err)
 			goto fail;
-- 
cgit v0.10.2


From 3e7732a05d3b14e2ae33013622ae9c691b419312 Mon Sep 17 00:00:00 2001
From: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Date: Thu, 1 Oct 2015 20:29:27 +0530
Subject: drm/i915: Update Promotion timer for RC6 TO Mode

When using RC6 timeout mode, the timeout value
should be written to GEN6_RC6_THRESHOLD.

v2: Updated commit message. (Tom)

v3: Rebase over whitespace differences. (Daniel)

Cc: Tom O'Rourke <Tom.O'Rourke@intel.com>
Signed-off-by: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Reviewed-by: Tom O'Rourke <Tom.O'Rourke@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index eb9a66d..60d120c 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -4730,7 +4730,6 @@ static void gen9_enable_rc6(struct drm_device *dev)
 		I915_WRITE(GUC_MAX_IDLE_COUNT, 0xA);
 
 	I915_WRITE(GEN6_RC_SLEEP, 0);
-	I915_WRITE(GEN6_RC6_THRESHOLD, 37500); /* 37.5/125ms per EI */
 
 	/* 2c: Program Coarse Power Gating Policies. */
 	I915_WRITE(GEN9_MEDIA_PG_IDLE_HYSTERESIS, 25);
@@ -4741,16 +4740,19 @@ static void gen9_enable_rc6(struct drm_device *dev)
 		rc6_mask = GEN6_RC_CTL_RC6_ENABLE;
 	DRM_INFO("RC6 %s\n", (rc6_mask & GEN6_RC_CTL_RC6_ENABLE) ?
 			"on" : "off");
-
+	/* WaRsUseTimeoutMode */
 	if ((IS_SKYLAKE(dev) && INTEL_REVID(dev) <= SKL_REVID_D0) ||
-	    (IS_BROXTON(dev) && INTEL_REVID(dev) <= BXT_REVID_A0))
+	    (IS_BROXTON(dev) && INTEL_REVID(dev) <= BXT_REVID_A0)) {
+		I915_WRITE(GEN6_RC6_THRESHOLD, 625); /* 800us */
 		I915_WRITE(GEN6_RC_CONTROL, GEN6_RC_CTL_HW_ENABLE |
 			   GEN7_RC_CTL_TO_MODE |
 			   rc6_mask);
-	else
+	} else {
+		I915_WRITE(GEN6_RC6_THRESHOLD, 37500); /* 37.5/125ms per EI */
 		I915_WRITE(GEN6_RC_CONTROL, GEN6_RC_CTL_HW_ENABLE |
 			   GEN6_RC_CTL_EI_MODE(1) |
 			   rc6_mask);
+	}
 
 	/*
 	 * 3b: Enable Coarse Power Gating only when RC6 is enabled.
-- 
cgit v0.10.2


From 7b9748cb513a6bef4af87b79f0da3ff7e8b56cd8 Mon Sep 17 00:00:00 2001
From: Jordan Justen <jordan.l.justen@intel.com>
Date: Thu, 1 Oct 2015 23:09:58 -0700
Subject: drm/i915: Add GEN7_GPGPU_DISPATCHDIMX/Y/Z to the register whitelist
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This is required to support glDispatchComputeIndirect for gen7.

Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_cmd_parser.c b/drivers/gpu/drm/i915/i915_cmd_parser.c
index 09932ca..db58c8d 100644
--- a/drivers/gpu/drm/i915/i915_cmd_parser.c
+++ b/drivers/gpu/drm/i915/i915_cmd_parser.c
@@ -448,6 +448,9 @@ static const struct drm_i915_reg_descriptor gen7_render_regs[] = {
 	REG32(GEN7_3DPRIM_INSTANCE_COUNT),
 	REG32(GEN7_3DPRIM_START_INSTANCE),
 	REG32(GEN7_3DPRIM_BASE_VERTEX),
+	REG32(GEN7_GPGPU_DISPATCHDIMX),
+	REG32(GEN7_GPGPU_DISPATCHDIMY),
+	REG32(GEN7_GPGPU_DISPATCHDIMZ),
 	REG64(GEN7_SO_NUM_PRIMS_WRITTEN(0)),
 	REG64(GEN7_SO_NUM_PRIMS_WRITTEN(1)),
 	REG64(GEN7_SO_NUM_PRIMS_WRITTEN(2)),
@@ -1214,6 +1217,7 @@ int i915_cmd_parser_get_version(void)
 	 *    MI_PREDICATE_SRC1 registers.
 	 * 3. Allow access to the GPGPU_THREADS_DISPATCHED register.
 	 * 4. L3 atomic chicken bits of HSW_SCRATCH1 and HSW_ROW_CHICKEN3.
+	 * 5. GPGPU dispatch compute indirect registers.
 	 */
-	return 4;
+	return 5;
 }
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 8d7be66..07588b6 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -536,6 +536,10 @@
 #define GEN7_3DPRIM_START_INSTANCE      0x243C
 #define GEN7_3DPRIM_BASE_VERTEX         0x2440
 
+#define GEN7_GPGPU_DISPATCHDIMX         0x2500
+#define GEN7_GPGPU_DISPATCHDIMY         0x2504
+#define GEN7_GPGPU_DISPATCHDIMZ         0x2508
+
 #define OACONTROL 0x2360
 
 #define _GEN7_PIPEA_DE_LOAD_SL	0x70068
-- 
cgit v0.10.2


From 0b5e88dc25ee6c9040c0355e6e025dcbc9c8de92 Mon Sep 17 00:00:00 2001
From: Sonika Jindal <sonika.jindal@intel.com>
Date: Mon, 5 Oct 2015 16:43:15 +0530
Subject: drm/i915: Add hot_plug hook for hdmi encoder

This patch adds a separate probe function for HDMI
EDID read over DDC channel. This function has been
registered as a .hot_plug handler for HDMI encoder.

The current implementation of hdmi_detect()
function re-sets the cached HDMI edid (in connector->detect_edid) in
every detect call.This function gets called many times, sometimes
directly from userspace probes, forcing drivers to read EDID every
detect function call.This causes several problems like:

1. Race conditions in multiple hot_plug / unplug cases, between
   interrupts bottom halves and userspace detections.
2. Many Un-necessary EDID reads for single hotplug/unplug
3. HDMI complaince failures which expects only one EDID read per hotplug

This function will be serving the purpose of really reading the EDID
by really probing the DDC channel, and updating the cached EDID.

The plan is to:
1. i915 IRQ handler bottom half function already calls
   intel_encoder->hotplug() function. Adding This probe function which
   will read the EDID only in case of a hotplug / unplug.
2. During init_connector this probe will be called to read the edid
3. Reuse the cached EDID in hdmi_detect() function.

The "< gen7" check is there because this was tested only for >=gen7
platforms. For older platforms the hotplug/reading edid path remains same.

v2: Calling set_edid instead of hdmi_probe during init.
Also, for platforms having DDI, intel_encoder for DP and HDMI is same
(taken from intel_dig_port), so for DP also, hot_plug function gets called
which is not intended here. So, check for HDMI in intel_hdmi_probe
Rely on HPD for updating edid only for platforms gen > 8 and also for VLV.

v3: Dropping the gen < 8 || !VLV  check. Now all platforms should rely on
hotplug or init for updating the edid.(Daniel)
Also, calling hdmi_probe in init instead of set_edid

v4: Renaming intel_hdmi_probe to intel_hdmi_hot_plug.
Also calling this hotplug handler from intel_hpd_init to take care of init
resume scenarios.

v5: Moved the call to encoder hotplug during init to separate patch(Daniel)

Signed-off-by: Shashank Sharma <shashank.sharma@intel.com>
Signed-off-by: Sonika Jindal <sonika.jindal@intel.com>
[danvet: Mark intel_hdmi_hot_plug as static.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index 3b28ed3..03d8590 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -1369,18 +1369,16 @@ intel_hdmi_set_edid(struct drm_connector *connector, bool force)
 	return connected;
 }
 
-static enum drm_connector_status
-intel_hdmi_detect(struct drm_connector *connector, bool force)
+static void intel_hdmi_hot_plug(struct intel_encoder *intel_encoder)
 {
-	enum drm_connector_status status;
-	struct intel_hdmi *intel_hdmi = intel_attached_hdmi(connector);
-	struct drm_i915_private *dev_priv = to_i915(connector->dev);
+	struct intel_hdmi *intel_hdmi =
+			enc_to_intel_hdmi(&intel_encoder->base);
+	struct intel_connector *intel_connector =
+				intel_hdmi->attached_connector;
+	struct drm_i915_private *dev_priv = to_i915(intel_encoder->base.dev);
 	bool live_status = false;
 	unsigned int retry = 3;
 
-	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
-		      connector->base.id, connector->name);
-
 	while (!live_status && --retry) {
 		live_status = intel_digital_port_connected(dev_priv,
 				hdmi_to_dig_port(intel_hdmi));
@@ -1390,15 +1388,48 @@ intel_hdmi_detect(struct drm_connector *connector, bool force)
 	if (!live_status)
 		DRM_DEBUG_KMS("Live status not up!");
 
-	intel_hdmi_unset_edid(connector);
+	/*
+	 * We are here, means there is a hotplug or a force
+	 * detection. Clear the cached EDID and probe the
+	 * DDC bus to check the current status of HDMI.
+	 */
+	intel_hdmi_unset_edid(&intel_connector->base);
+	if (intel_hdmi_set_edid(&intel_connector->base, live_status))
+		DRM_DEBUG_DRIVER("DDC probe: got EDID\n");
+	else
+		DRM_DEBUG_DRIVER("DDC probe: no EDID\n");
+}
 
-	if (intel_hdmi_set_edid(connector, live_status)) {
+static enum drm_connector_status
+intel_hdmi_detect(struct drm_connector *connector, bool force)
+{
+	enum drm_connector_status status;
+	struct intel_connector *intel_connector =
+				to_intel_connector(connector);
+
+	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
+		      connector->base.id, connector->name);
+
+	/*
+	 * There are many userspace calls which probe EDID from
+	 * detect path. In case of multiple hotplug/unplug, these
+	 * can cause race conditions while probing EDID. Also its
+	 * waste of CPU cycles to read the EDID again and again
+	 * unless there is a real hotplug.
+	 * So, rely on hotplugs and init to read edid.
+	 * Check connector status based on availability of cached EDID.
+	 */
+
+	if (intel_connector->detect_edid) {
 		struct intel_hdmi *intel_hdmi = intel_attached_hdmi(connector);
 
 		hdmi_to_dig_port(intel_hdmi)->base.type = INTEL_OUTPUT_HDMI;
 		status = connector_status_connected;
-	} else
+		DRM_DEBUG_DRIVER("hdmi status = connected\n");
+	} else {
 		status = connector_status_disconnected;
+		DRM_DEBUG_DRIVER("hdmi status = disconnected\n");
+	}
 
 	return status;
 }
@@ -2114,6 +2145,7 @@ void intel_hdmi_init_connector(struct intel_digital_port *intel_dig_port,
 	intel_connector->unregister = intel_connector_unregister;
 
 	intel_hdmi_add_properties(intel_hdmi, connector);
+	intel_encoder->hot_plug = intel_hdmi_hot_plug;
 
 	intel_connector_attach_encoder(intel_connector, intel_encoder);
 	drm_connector_register(connector);
-- 
cgit v0.10.2


From 2493f21f56316ad55508f173463ebf0b3f0a15bb Mon Sep 17 00:00:00 2001
From: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Date: Mon, 5 Oct 2015 10:01:13 +0300
Subject: drm/i915: Rename DP link training functions

The link training functions had confusing names. The start function
actually does the clock recovery phase of the link training, and the
complete function does the channel equalization. So call them that
instead. Also, every call to intel_dp_start_link_train() was followed
by a call to intel_dp_complete_link_train(), so add a new start
function that calls clock_recory and channel_equalization.

Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
index b3f1748..b25e99a 100644
--- a/drivers/gpu/drm/i915/intel_ddi.c
+++ b/drivers/gpu/drm/i915/intel_ddi.c
@@ -2320,7 +2320,6 @@ static void intel_ddi_pre_enable(struct intel_encoder *intel_encoder)
 
 		intel_dp_sink_dpms(intel_dp, DRM_MODE_DPMS_ON);
 		intel_dp_start_link_train(intel_dp);
-		intel_dp_complete_link_train(intel_dp);
 		if (port != PORT_A || INTEL_INFO(dev)->gen >= 9)
 			intel_dp_stop_link_train(intel_dp);
 	} else if (type == INTEL_OUTPUT_HDMI) {
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 19cbc2e..8d34ca7 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -2604,7 +2604,6 @@ static void intel_enable_dp(struct intel_encoder *encoder)
 
 	intel_dp_sink_dpms(intel_dp, DRM_MODE_DPMS_ON);
 	intel_dp_start_link_train(intel_dp);
-	intel_dp_complete_link_train(intel_dp);
 	intel_dp_stop_link_train(intel_dp);
 
 	if (crtc->config->has_audio) {
@@ -3691,8 +3690,8 @@ static void intel_dp_set_idle_link_train(struct intel_dp *intel_dp)
 }
 
 /* Enable corresponding port and start training pattern 1 */
-void
-intel_dp_start_link_train(struct intel_dp *intel_dp)
+static void
+intel_dp_link_training_clock_recovery(struct intel_dp *intel_dp)
 {
 	struct drm_encoder *encoder = &dp_to_dig_port(intel_dp)->base.base;
 	struct drm_device *dev = encoder->dev;
@@ -3805,8 +3804,8 @@ intel_dp_start_link_train(struct intel_dp *intel_dp)
 	intel_dp->DP = DP;
 }
 
-void
-intel_dp_complete_link_train(struct intel_dp *intel_dp)
+static void
+intel_dp_link_training_channel_equalization(struct intel_dp *intel_dp)
 {
 	struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp);
 	struct drm_device *dev = dig_port->base.base.dev;
@@ -3859,7 +3858,7 @@ intel_dp_complete_link_train(struct intel_dp *intel_dp)
 		if (!drm_dp_clock_recovery_ok(link_status,
 					      intel_dp->lane_count)) {
 			intel_dp->train_set_valid = false;
-			intel_dp_start_link_train(intel_dp);
+			intel_dp_link_training_clock_recovery(intel_dp);
 			intel_dp_set_link_train(intel_dp, &DP,
 						training_pattern |
 						DP_LINK_SCRAMBLING_DISABLE);
@@ -3876,7 +3875,7 @@ intel_dp_complete_link_train(struct intel_dp *intel_dp)
 		/* Try 5 times, then try clock recovery if that fails */
 		if (tries > 5) {
 			intel_dp->train_set_valid = false;
-			intel_dp_start_link_train(intel_dp);
+			intel_dp_link_training_clock_recovery(intel_dp);
 			intel_dp_set_link_train(intel_dp, &DP,
 						training_pattern |
 						DP_LINK_SCRAMBLING_DISABLE);
@@ -3909,6 +3908,13 @@ void intel_dp_stop_link_train(struct intel_dp *intel_dp)
 				DP_TRAINING_PATTERN_DISABLE);
 }
 
+void
+intel_dp_start_link_train(struct intel_dp *intel_dp)
+{
+	intel_dp_link_training_clock_recovery(intel_dp);
+	intel_dp_link_training_channel_equalization(intel_dp);
+}
+
 static void
 intel_dp_link_down(struct intel_dp *intel_dp)
 {
@@ -4377,7 +4383,6 @@ go_again:
 			    !drm_dp_channel_eq_ok(&esi[10], intel_dp->lane_count)) {
 				DRM_DEBUG_KMS("channel EQ not ok, retraining\n");
 				intel_dp_start_link_train(intel_dp);
-				intel_dp_complete_link_train(intel_dp);
 				intel_dp_stop_link_train(intel_dp);
 			}
 
@@ -4468,7 +4473,6 @@ intel_dp_check_link_status(struct intel_dp *intel_dp)
 		DRM_DEBUG_KMS("%s: channel EQ not ok, retraining\n",
 			      intel_encoder->base.name);
 		intel_dp_start_link_train(intel_dp);
-		intel_dp_complete_link_train(intel_dp);
 		intel_dp_stop_link_train(intel_dp);
 	}
 }
diff --git a/drivers/gpu/drm/i915/intel_dp_mst.c b/drivers/gpu/drm/i915/intel_dp_mst.c
index ca4d022..1537259 100644
--- a/drivers/gpu/drm/i915/intel_dp_mst.c
+++ b/drivers/gpu/drm/i915/intel_dp_mst.c
@@ -188,7 +188,6 @@ static void intel_mst_pre_enable_dp(struct intel_encoder *encoder)
 
 
 		intel_dp_start_link_train(intel_dp);
-		intel_dp_complete_link_train(intel_dp);
 		intel_dp_stop_link_train(intel_dp);
 	}
 
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 563c4f1..e320825 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -1217,7 +1217,6 @@ bool intel_dp_init_connector(struct intel_digital_port *intel_dig_port,
 void intel_dp_set_link_params(struct intel_dp *intel_dp,
 			      const struct intel_crtc_state *pipe_config);
 void intel_dp_start_link_train(struct intel_dp *intel_dp);
-void intel_dp_complete_link_train(struct intel_dp *intel_dp);
 void intel_dp_stop_link_train(struct intel_dp *intel_dp);
 void intel_dp_sink_dpms(struct intel_dp *intel_dp, int mode);
 void intel_dp_encoder_destroy(struct drm_encoder *encoder);
-- 
cgit v0.10.2


From b248e6548e344a26fc92c57772fd4224a490bca9 Mon Sep 17 00:00:00 2001
From: kbuild test robot <fengguang.wu@intel.com>
Date: Mon, 5 Oct 2015 16:21:11 +0800
Subject: drm/i915/bxt: vlv_dsi_reset_clocks() can be static

Signed-off-by: Fengguang Wu <fengguang.wu@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dsi_pll.c b/drivers/gpu/drm/i915/intel_dsi_pll.c
index 22c1ea9..cb3cf39 100644
--- a/drivers/gpu/drm/i915/intel_dsi_pll.c
+++ b/drivers/gpu/drm/i915/intel_dsi_pll.c
@@ -419,7 +419,7 @@ u32 bxt_get_dsi_pclk(struct intel_encoder *encoder, int pipe_bpp)
 	return pclk;
 }
 
-void vlv_dsi_reset_clocks(struct intel_encoder *encoder, enum port port)
+static void vlv_dsi_reset_clocks(struct intel_encoder *encoder, enum port port)
 {
 	u32 temp;
 	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
@@ -577,7 +577,7 @@ void intel_disable_dsi_pll(struct intel_encoder *encoder)
 		bxt_disable_dsi_pll(encoder);
 }
 
-void bxt_dsi_reset_clocks(struct intel_encoder *encoder, enum port port)
+static void bxt_dsi_reset_clocks(struct intel_encoder *encoder, enum port port)
 {
 	u32 tmp;
 	struct drm_device *dev = encoder->base.dev;
-- 
cgit v0.10.2


From 88e72717c2de4181d8a6de1b04315953ad2bebdf Mon Sep 17 00:00:00 2001
From: Thierry Reding <treding@nvidia.com>
Date: Thu, 24 Sep 2015 18:35:31 +0200
Subject: drm/irq: Use unsigned int pipe in public API
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This continues the pattern started in commit cc1ef118fc09 ("drm/irq:
Make pipe unsigned and name consistent"). This is applied to the public
APIs and driver callbacks, so pretty much all drivers need to be updated
to match the new prototypes.

Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Russell King <rmk+kernel@arm.linux.org.uk>
Cc: Inki Dae <inki.dae@samsung.com>
Cc: Jianwei Wang <jianwei.wang.chn@gmail.com>
Cc: Alison Wang <alison.wang@freescale.com>
Cc: Patrik Jakobsson <patrik.r.jakobsson@gmail.com>
Cc: Daniel Vetter <daniel.vetter@intel.com>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Cc: Philipp Zabel <p.zabel@pengutronix.de>
Cc: David Airlie <airlied@linux.ie>
Cc: Rob Clark <robdclark@gmail.com>
Cc: Ben Skeggs <bskeggs@redhat.com>
Cc: Tomi Valkeinen <tomi.valkeinen@ti.com>
Cc: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Cc: Mark Yao <mark.yao@rock-chips.com>
Cc: Benjamin Gaignard <benjamin.gaignard@linaro.org>
Cc: Vincent Abriou <vincent.abriou@st.com>
Cc: Thomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 6647fb2..5e43178d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -2349,10 +2349,10 @@ void amdgpu_driver_preclose_kms(struct drm_device *dev,
 				struct drm_file *file_priv);
 int amdgpu_suspend_kms(struct drm_device *dev, bool suspend, bool fbcon);
 int amdgpu_resume_kms(struct drm_device *dev, bool resume, bool fbcon);
-u32 amdgpu_get_vblank_counter_kms(struct drm_device *dev, int crtc);
-int amdgpu_enable_vblank_kms(struct drm_device *dev, int crtc);
-void amdgpu_disable_vblank_kms(struct drm_device *dev, int crtc);
-int amdgpu_get_vblank_timestamp_kms(struct drm_device *dev, int crtc,
+u32 amdgpu_get_vblank_counter_kms(struct drm_device *dev, unsigned int pipe);
+int amdgpu_enable_vblank_kms(struct drm_device *dev, unsigned int pipe);
+void amdgpu_disable_vblank_kms(struct drm_device *dev, unsigned int pipe);
+int amdgpu_get_vblank_timestamp_kms(struct drm_device *dev, unsigned int pipe,
 				    int *max_error,
 				    struct timeval *vblank_time,
 				    unsigned flags);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
index 9b34a34..de11639 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
@@ -721,7 +721,7 @@ bool amdgpu_crtc_scaling_mode_fixup(struct drm_crtc *crtc,
  * an optional accurate timestamp of when query happened.
  *
  * \param dev Device to query.
- * \param crtc Crtc to query.
+ * \param pipe Crtc to query.
  * \param flags Flags from caller (DRM_CALLED_FROM_VBLIRQ or 0).
  * \param *vpos Location where vertical scanout position should be stored.
  * \param *hpos Location where horizontal scanout position should go.
@@ -744,8 +744,9 @@ bool amdgpu_crtc_scaling_mode_fixup(struct drm_crtc *crtc,
  * unknown small number of scanlines wrt. real scanout position.
  *
  */
-int amdgpu_get_crtc_scanoutpos(struct drm_device *dev, int crtc, unsigned int flags,
-			       int *vpos, int *hpos, ktime_t *stime, ktime_t *etime,
+int amdgpu_get_crtc_scanoutpos(struct drm_device *dev, unsigned int pipe,
+			       unsigned int flags, int *vpos, int *hpos,
+			       ktime_t *stime, ktime_t *etime,
 			       const struct drm_display_mode *mode)
 {
 	u32 vbl = 0, position = 0;
@@ -760,7 +761,7 @@ int amdgpu_get_crtc_scanoutpos(struct drm_device *dev, int crtc, unsigned int fl
 	if (stime)
 		*stime = ktime_get();
 
-	if (amdgpu_display_page_flip_get_scanoutpos(adev, crtc, &vbl, &position) == 0)
+	if (amdgpu_display_page_flip_get_scanoutpos(adev, pipe, &vbl, &position) == 0)
 		ret |= DRM_SCANOUTPOS_VALID;
 
 	/* Get optional system timestamp after query. */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
index 275f1c3..b9faaf8 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
@@ -600,36 +600,36 @@ void amdgpu_driver_preclose_kms(struct drm_device *dev,
  * amdgpu_get_vblank_counter_kms - get frame count
  *
  * @dev: drm dev pointer
- * @crtc: crtc to get the frame count from
+ * @pipe: crtc to get the frame count from
  *
  * Gets the frame count on the requested crtc (all asics).
  * Returns frame count on success, -EINVAL on failure.
  */
-u32 amdgpu_get_vblank_counter_kms(struct drm_device *dev, int crtc)
+u32 amdgpu_get_vblank_counter_kms(struct drm_device *dev, unsigned int pipe)
 {
 	struct amdgpu_device *adev = dev->dev_private;
 
-	if (crtc < 0 || crtc >= adev->mode_info.num_crtc) {
-		DRM_ERROR("Invalid crtc %d\n", crtc);
+	if (pipe >= adev->mode_info.num_crtc) {
+		DRM_ERROR("Invalid crtc %u\n", pipe);
 		return -EINVAL;
 	}
 
-	return amdgpu_display_vblank_get_counter(adev, crtc);
+	return amdgpu_display_vblank_get_counter(adev, pipe);
 }
 
 /**
  * amdgpu_enable_vblank_kms - enable vblank interrupt
  *
  * @dev: drm dev pointer
- * @crtc: crtc to enable vblank interrupt for
+ * @pipe: crtc to enable vblank interrupt for
  *
  * Enable the interrupt on the requested crtc (all asics).
  * Returns 0 on success, -EINVAL on failure.
  */
-int amdgpu_enable_vblank_kms(struct drm_device *dev, int crtc)
+int amdgpu_enable_vblank_kms(struct drm_device *dev, unsigned int pipe)
 {
 	struct amdgpu_device *adev = dev->dev_private;
-	int idx = amdgpu_crtc_idx_to_irq_type(adev, crtc);
+	int idx = amdgpu_crtc_idx_to_irq_type(adev, pipe);
 
 	return amdgpu_irq_get(adev, &adev->crtc_irq, idx);
 }
@@ -638,14 +638,14 @@ int amdgpu_enable_vblank_kms(struct drm_device *dev, int crtc)
  * amdgpu_disable_vblank_kms - disable vblank interrupt
  *
  * @dev: drm dev pointer
- * @crtc: crtc to disable vblank interrupt for
+ * @pipe: crtc to disable vblank interrupt for
  *
  * Disable the interrupt on the requested crtc (all asics).
  */
-void amdgpu_disable_vblank_kms(struct drm_device *dev, int crtc)
+void amdgpu_disable_vblank_kms(struct drm_device *dev, unsigned int pipe)
 {
 	struct amdgpu_device *adev = dev->dev_private;
-	int idx = amdgpu_crtc_idx_to_irq_type(adev, crtc);
+	int idx = amdgpu_crtc_idx_to_irq_type(adev, pipe);
 
 	amdgpu_irq_put(adev, &adev->crtc_irq, idx);
 }
@@ -663,26 +663,26 @@ void amdgpu_disable_vblank_kms(struct drm_device *dev, int crtc)
  * scanout position.  (all asics).
  * Returns postive status flags on success, negative error on failure.
  */
-int amdgpu_get_vblank_timestamp_kms(struct drm_device *dev, int crtc,
+int amdgpu_get_vblank_timestamp_kms(struct drm_device *dev, unsigned int pipe,
 				    int *max_error,
 				    struct timeval *vblank_time,
 				    unsigned flags)
 {
-	struct drm_crtc *drmcrtc;
+	struct drm_crtc *crtc;
 	struct amdgpu_device *adev = dev->dev_private;
 
-	if (crtc < 0 || crtc >= dev->num_crtcs) {
-		DRM_ERROR("Invalid crtc %d\n", crtc);
+	if (pipe >= dev->num_crtcs) {
+		DRM_ERROR("Invalid crtc %u\n", pipe);
 		return -EINVAL;
 	}
 
 	/* Get associated drm_crtc: */
-	drmcrtc = &adev->mode_info.crtcs[crtc]->base;
+	crtc = &adev->mode_info.crtcs[pipe]->base;
 
 	/* Helper routine in DRM core does all the work: */
-	return drm_calc_vbltimestamp_from_scanoutpos(dev, crtc, max_error,
+	return drm_calc_vbltimestamp_from_scanoutpos(dev, pipe, max_error,
 						     vblank_time, flags,
-						     &drmcrtc->hwmode);
+						     &crtc->hwmode);
 }
 
 const struct drm_ioctl_desc amdgpu_ioctls_kms[] = {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h
index 2b03425..f6b0299 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h
@@ -540,11 +540,10 @@ bool amdgpu_ddc_probe(struct amdgpu_connector *amdgpu_connector, bool use_aux);
 
 void amdgpu_encoder_set_active_device(struct drm_encoder *encoder);
 
-int amdgpu_get_crtc_scanoutpos(struct drm_device *dev, int crtc,
-				      unsigned int flags,
-				      int *vpos, int *hpos, ktime_t *stime,
-				      ktime_t *etime,
-				      const struct drm_display_mode *mode);
+int amdgpu_get_crtc_scanoutpos(struct drm_device *dev, unsigned int pipe,
+			       unsigned int flags, int *vpos, int *hpos,
+			       ktime_t *stime, ktime_t *etime,
+			       const struct drm_display_mode *mode);
 
 int amdgpu_framebuffer_init(struct drm_device *dev,
 			     struct amdgpu_framebuffer *rfb,
diff --git a/drivers/gpu/drm/armada/armada_drv.c b/drivers/gpu/drm/armada/armada_drv.c
index 225034b..a438886 100644
--- a/drivers/gpu/drm/armada/armada_drv.c
+++ b/drivers/gpu/drm/armada/armada_drv.c
@@ -254,17 +254,17 @@ void armada_drm_vbl_event_remove(struct armada_crtc *dcrtc,
 }
 
 /* These are called under the vbl_lock. */
-static int armada_drm_enable_vblank(struct drm_device *dev, int crtc)
+static int armada_drm_enable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	struct armada_private *priv = dev->dev_private;
-	armada_drm_crtc_enable_irq(priv->dcrtc[crtc], VSYNC_IRQ_ENA);
+	armada_drm_crtc_enable_irq(priv->dcrtc[pipe], VSYNC_IRQ_ENA);
 	return 0;
 }
 
-static void armada_drm_disable_vblank(struct drm_device *dev, int crtc)
+static void armada_drm_disable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	struct armada_private *priv = dev->dev_private;
-	armada_drm_crtc_disable_irq(priv->dcrtc[crtc], VSYNC_IRQ_ENA);
+	armada_drm_crtc_disable_irq(priv->dcrtc[pipe], VSYNC_IRQ_ENA);
 }
 
 static struct drm_ioctl_desc armada_ioctls[] = {
diff --git a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c
index 8bc62ec..6dfb63a 100644
--- a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c
+++ b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c
@@ -656,7 +656,8 @@ static void atmel_hlcdc_dc_irq_uninstall(struct drm_device *dev)
 	regmap_read(dc->hlcdc->regmap, ATMEL_HLCDC_ISR, &isr);
 }
 
-static int atmel_hlcdc_dc_enable_vblank(struct drm_device *dev, int crtc)
+static int atmel_hlcdc_dc_enable_vblank(struct drm_device *dev,
+					unsigned int pipe)
 {
 	struct atmel_hlcdc_dc *dc = dev->dev_private;
 
@@ -666,7 +667,8 @@ static int atmel_hlcdc_dc_enable_vblank(struct drm_device *dev, int crtc)
 	return 0;
 }
 
-static void atmel_hlcdc_dc_disable_vblank(struct drm_device *dev, int crtc)
+static void atmel_hlcdc_dc_disable_vblank(struct drm_device *dev,
+					  unsigned int pipe)
 {
 	struct atmel_hlcdc_dc *dc = dev->dev_private;
 
diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c
index 6bff6d3..0659d99 100644
--- a/drivers/gpu/drm/drm_irq.c
+++ b/drivers/gpu/drm/drm_irq.c
@@ -877,7 +877,7 @@ drm_get_last_vbltimestamp(struct drm_device *dev, unsigned int pipe,
  * Returns:
  * The software vblank counter.
  */
-u32 drm_vblank_count(struct drm_device *dev, int pipe)
+u32 drm_vblank_count(struct drm_device *dev, unsigned int pipe)
 {
 	struct drm_vblank_crtc *vblank = &dev->vblank[pipe];
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_crtc.c b/drivers/gpu/drm/exynos/exynos_drm_crtc.c
index 0872aa2f..f364d69 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_crtc.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_crtc.c
@@ -167,7 +167,7 @@ err_crtc:
 	return ERR_PTR(ret);
 }
 
-int exynos_drm_crtc_enable_vblank(struct drm_device *dev, int pipe)
+int exynos_drm_crtc_enable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	struct exynos_drm_private *private = dev->dev_private;
 	struct exynos_drm_crtc *exynos_crtc =
@@ -179,7 +179,7 @@ int exynos_drm_crtc_enable_vblank(struct drm_device *dev, int pipe)
 	return 0;
 }
 
-void exynos_drm_crtc_disable_vblank(struct drm_device *dev, int pipe)
+void exynos_drm_crtc_disable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	struct exynos_drm_private *private = dev->dev_private;
 	struct exynos_drm_crtc *exynos_crtc =
diff --git a/drivers/gpu/drm/exynos/exynos_drm_crtc.h b/drivers/gpu/drm/exynos/exynos_drm_crtc.h
index f87d4ab..f9f365b 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_crtc.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_crtc.h
@@ -23,8 +23,8 @@ struct exynos_drm_crtc *exynos_drm_crtc_create(struct drm_device *drm_dev,
 					enum exynos_drm_output_type type,
 					const struct exynos_drm_crtc_ops *ops,
 					void *context);
-int exynos_drm_crtc_enable_vblank(struct drm_device *dev, int pipe);
-void exynos_drm_crtc_disable_vblank(struct drm_device *dev, int pipe);
+int exynos_drm_crtc_enable_vblank(struct drm_device *dev, unsigned int pipe);
+void exynos_drm_crtc_disable_vblank(struct drm_device *dev, unsigned int pipe);
 void exynos_drm_crtc_wait_pending_update(struct exynos_drm_crtc *exynos_crtc);
 void exynos_drm_crtc_finish_update(struct exynos_drm_crtc *exynos_crtc,
 				   struct exynos_drm_plane *exynos_plane);
diff --git a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_drv.c b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_drv.c
index 9a8e2da..f1fd986 100644
--- a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_drv.c
+++ b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_drv.c
@@ -140,7 +140,7 @@ static irqreturn_t fsl_dcu_drm_irq(int irq, void *arg)
 	return IRQ_HANDLED;
 }
 
-static int fsl_dcu_drm_enable_vblank(struct drm_device *dev, int crtc)
+static int fsl_dcu_drm_enable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	struct fsl_dcu_drm_device *fsl_dev = dev->dev_private;
 	unsigned int value;
@@ -156,7 +156,8 @@ static int fsl_dcu_drm_enable_vblank(struct drm_device *dev, int crtc)
 	return 0;
 }
 
-static void fsl_dcu_drm_disable_vblank(struct drm_device *dev, int crtc)
+static void fsl_dcu_drm_disable_vblank(struct drm_device *dev,
+				       unsigned int pipe)
 {
 	struct fsl_dcu_drm_device *fsl_dev = dev->dev_private;
 	unsigned int value;
diff --git a/drivers/gpu/drm/gma500/psb_drv.h b/drivers/gpu/drm/gma500/psb_drv.h
index e38057b..e21726e 100644
--- a/drivers/gpu/drm/gma500/psb_drv.h
+++ b/drivers/gpu/drm/gma500/psb_drv.h
@@ -687,15 +687,15 @@ extern void psb_irq_turn_off_dpst(struct drm_device *dev);
 extern void psb_irq_uninstall_islands(struct drm_device *dev, int hw_islands);
 extern int psb_vblank_wait2(struct drm_device *dev, unsigned int *sequence);
 extern int psb_vblank_wait(struct drm_device *dev, unsigned int *sequence);
-extern int psb_enable_vblank(struct drm_device *dev, int crtc);
-extern void psb_disable_vblank(struct drm_device *dev, int crtc);
+extern int psb_enable_vblank(struct drm_device *dev, unsigned int pipe);
+extern void psb_disable_vblank(struct drm_device *dev, unsigned int pipe);
 void
 psb_enable_pipestat(struct drm_psb_private *dev_priv, int pipe, u32 mask);
 
 void
 psb_disable_pipestat(struct drm_psb_private *dev_priv, int pipe, u32 mask);
 
-extern u32 psb_get_vblank_counter(struct drm_device *dev, int crtc);
+extern u32 psb_get_vblank_counter(struct drm_device *dev, unsigned int pipe);
 
 /* framebuffer.c */
 extern int psbfb_probed(struct drm_device *dev);
diff --git a/drivers/gpu/drm/gma500/psb_irq.c b/drivers/gpu/drm/gma500/psb_irq.c
index 624eb36..78eb109 100644
--- a/drivers/gpu/drm/gma500/psb_irq.c
+++ b/drivers/gpu/drm/gma500/psb_irq.c
@@ -510,7 +510,7 @@ int psb_irq_disable_dpst(struct drm_device *dev)
 /*
  * It is used to enable VBLANK interrupt
  */
-int psb_enable_vblank(struct drm_device *dev, int pipe)
+int psb_enable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	struct drm_psb_private *dev_priv = dev->dev_private;
 	unsigned long irqflags;
@@ -549,7 +549,7 @@ int psb_enable_vblank(struct drm_device *dev, int pipe)
 /*
  * It is used to disable VBLANK interrupt
  */
-void psb_disable_vblank(struct drm_device *dev, int pipe)
+void psb_disable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	struct drm_psb_private *dev_priv = dev->dev_private;
 	unsigned long irqflags;
@@ -622,7 +622,7 @@ void mdfld_disable_te(struct drm_device *dev, int pipe)
 /* Called from drm generic code, passed a 'crtc', which
  * we use as a pipe index
  */
-u32 psb_get_vblank_counter(struct drm_device *dev, int pipe)
+u32 psb_get_vblank_counter(struct drm_device *dev, unsigned int pipe)
 {
 	uint32_t high_frame = PIPEAFRAMEHIGH;
 	uint32_t low_frame = PIPEAFRAMEPIXEL;
@@ -654,7 +654,7 @@ u32 psb_get_vblank_counter(struct drm_device *dev, int pipe)
 	reg_val = REG_READ(pipeconf_reg);
 
 	if (!(reg_val & PIPEACONF_ENABLE)) {
-		dev_err(dev->dev, "trying to get vblank count for disabled pipe %d\n",
+		dev_err(dev->dev, "trying to get vblank count for disabled pipe %u\n",
 								pipe);
 		goto psb_get_vblank_counter_exit;
 	}
diff --git a/drivers/gpu/drm/gma500/psb_irq.h b/drivers/gpu/drm/gma500/psb_irq.h
index d0b45ff..e6a81a8 100644
--- a/drivers/gpu/drm/gma500/psb_irq.h
+++ b/drivers/gpu/drm/gma500/psb_irq.h
@@ -38,9 +38,9 @@ int psb_irq_enable_dpst(struct drm_device *dev);
 int psb_irq_disable_dpst(struct drm_device *dev);
 void psb_irq_turn_on_dpst(struct drm_device *dev);
 void psb_irq_turn_off_dpst(struct drm_device *dev);
-int  psb_enable_vblank(struct drm_device *dev, int pipe);
-void psb_disable_vblank(struct drm_device *dev, int pipe);
-u32  psb_get_vblank_counter(struct drm_device *dev, int pipe);
+int  psb_enable_vblank(struct drm_device *dev, unsigned int pipe);
+void psb_disable_vblank(struct drm_device *dev, unsigned int pipe);
+u32  psb_get_vblank_counter(struct drm_device *dev, unsigned int pipe);
 
 int mdfld_enable_te(struct drm_device *dev, int pipe);
 void mdfld_disable_te(struct drm_device *dev, int pipe);
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 88d064e..bc732eb 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -554,7 +554,7 @@ static void i915_enable_asle_pipestat(struct drm_device *dev)
  *   of horizontal active on the first line of vertical active
  */
 
-static u32 i8xx_get_vblank_counter(struct drm_device *dev, int pipe)
+static u32 i8xx_get_vblank_counter(struct drm_device *dev, unsigned int pipe)
 {
 	/* Gen2 doesn't have a hardware frame counter */
 	return 0;
@@ -563,7 +563,7 @@ static u32 i8xx_get_vblank_counter(struct drm_device *dev, int pipe)
 /* Called from drm generic code, passed a 'crtc', which
  * we use as a pipe index
  */
-static u32 i915_get_vblank_counter(struct drm_device *dev, int pipe)
+static u32 i915_get_vblank_counter(struct drm_device *dev, unsigned int pipe)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long high_frame;
@@ -611,7 +611,7 @@ static u32 i915_get_vblank_counter(struct drm_device *dev, int pipe)
 	return (((high1 << 8) | low) + (pixel >= vbl_start)) & 0xffffff;
 }
 
-static u32 gm45_get_vblank_counter(struct drm_device *dev, int pipe)
+static u32 gm45_get_vblank_counter(struct drm_device *dev, unsigned int pipe)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int reg = PIPE_FRMCOUNT_GM45(pipe);
@@ -672,7 +672,7 @@ static int __intel_get_crtc_scanline(struct intel_crtc *crtc)
 	return (position + crtc->scanline_offset) % vtotal;
 }
 
-static int i915_get_crtc_scanoutpos(struct drm_device *dev, int pipe,
+static int i915_get_crtc_scanoutpos(struct drm_device *dev, unsigned int pipe,
 				    unsigned int flags, int *vpos, int *hpos,
 				    ktime_t *stime, ktime_t *etime,
 				    const struct drm_display_mode *mode)
@@ -809,27 +809,27 @@ int intel_get_crtc_scanline(struct intel_crtc *crtc)
 	return position;
 }
 
-static int i915_get_vblank_timestamp(struct drm_device *dev, int pipe,
+static int i915_get_vblank_timestamp(struct drm_device *dev, unsigned int pipe,
 			      int *max_error,
 			      struct timeval *vblank_time,
 			      unsigned flags)
 {
 	struct drm_crtc *crtc;
 
-	if (pipe < 0 || pipe >= INTEL_INFO(dev)->num_pipes) {
-		DRM_ERROR("Invalid crtc %d\n", pipe);
+	if (pipe >= INTEL_INFO(dev)->num_pipes) {
+		DRM_ERROR("Invalid crtc %u\n", pipe);
 		return -EINVAL;
 	}
 
 	/* Get drm_crtc to timestamp: */
 	crtc = intel_get_crtc_for_pipe(dev, pipe);
 	if (crtc == NULL) {
-		DRM_ERROR("Invalid crtc %d\n", pipe);
+		DRM_ERROR("Invalid crtc %u\n", pipe);
 		return -EINVAL;
 	}
 
 	if (!crtc->hwmode.crtc_clock) {
-		DRM_DEBUG_KMS("crtc %d is disabled\n", pipe);
+		DRM_DEBUG_KMS("crtc %u is disabled\n", pipe);
 		return -EBUSY;
 	}
 
@@ -2431,7 +2431,7 @@ void i915_handle_error(struct drm_device *dev, bool wedged,
 /* Called from drm generic code, passed 'crtc' which
  * we use as a pipe index
  */
-static int i915_enable_vblank(struct drm_device *dev, int pipe)
+static int i915_enable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long irqflags;
@@ -2448,7 +2448,7 @@ static int i915_enable_vblank(struct drm_device *dev, int pipe)
 	return 0;
 }
 
-static int ironlake_enable_vblank(struct drm_device *dev, int pipe)
+static int ironlake_enable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long irqflags;
@@ -2462,7 +2462,7 @@ static int ironlake_enable_vblank(struct drm_device *dev, int pipe)
 	return 0;
 }
 
-static int valleyview_enable_vblank(struct drm_device *dev, int pipe)
+static int valleyview_enable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long irqflags;
@@ -2475,7 +2475,7 @@ static int valleyview_enable_vblank(struct drm_device *dev, int pipe)
 	return 0;
 }
 
-static int gen8_enable_vblank(struct drm_device *dev, int pipe)
+static int gen8_enable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long irqflags;
@@ -2491,7 +2491,7 @@ static int gen8_enable_vblank(struct drm_device *dev, int pipe)
 /* Called from drm generic code, passed 'crtc' which
  * we use as a pipe index
  */
-static void i915_disable_vblank(struct drm_device *dev, int pipe)
+static void i915_disable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long irqflags;
@@ -2503,7 +2503,7 @@ static void i915_disable_vblank(struct drm_device *dev, int pipe)
 	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
 }
 
-static void ironlake_disable_vblank(struct drm_device *dev, int pipe)
+static void ironlake_disable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long irqflags;
@@ -2515,7 +2515,7 @@ static void ironlake_disable_vblank(struct drm_device *dev, int pipe)
 	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
 }
 
-static void valleyview_disable_vblank(struct drm_device *dev, int pipe)
+static void valleyview_disable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long irqflags;
@@ -2526,7 +2526,7 @@ static void valleyview_disable_vblank(struct drm_device *dev, int pipe)
 	spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
 }
 
-static void gen8_disable_vblank(struct drm_device *dev, int pipe)
+static void gen8_disable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	unsigned long irqflags;
diff --git a/drivers/gpu/drm/imx/imx-drm-core.c b/drivers/gpu/drm/imx/imx-drm-core.c
index 74f505b..b880c12 100644
--- a/drivers/gpu/drm/imx/imx-drm-core.c
+++ b/drivers/gpu/drm/imx/imx-drm-core.c
@@ -145,10 +145,10 @@ void imx_drm_handle_vblank(struct imx_drm_crtc *imx_drm_crtc)
 }
 EXPORT_SYMBOL_GPL(imx_drm_handle_vblank);
 
-static int imx_drm_enable_vblank(struct drm_device *drm, int crtc)
+static int imx_drm_enable_vblank(struct drm_device *drm, unsigned int pipe)
 {
 	struct imx_drm_device *imxdrm = drm->dev_private;
-	struct imx_drm_crtc *imx_drm_crtc = imxdrm->crtc[crtc];
+	struct imx_drm_crtc *imx_drm_crtc = imxdrm->crtc[pipe];
 	int ret;
 
 	if (!imx_drm_crtc)
@@ -163,10 +163,10 @@ static int imx_drm_enable_vblank(struct drm_device *drm, int crtc)
 	return ret;
 }
 
-static void imx_drm_disable_vblank(struct drm_device *drm, int crtc)
+static void imx_drm_disable_vblank(struct drm_device *drm, unsigned int pipe)
 {
 	struct imx_drm_device *imxdrm = drm->dev_private;
-	struct imx_drm_crtc *imx_drm_crtc = imxdrm->crtc[crtc];
+	struct imx_drm_crtc *imx_drm_crtc = imxdrm->crtc[pipe];
 
 	if (!imx_drm_crtc)
 		return;
diff --git a/drivers/gpu/drm/mga/mga_drv.h b/drivers/gpu/drm/mga/mga_drv.h
index b4a20149..bb31233 100644
--- a/drivers/gpu/drm/mga/mga_drv.h
+++ b/drivers/gpu/drm/mga/mga_drv.h
@@ -183,9 +183,9 @@ extern int mga_warp_install_microcode(drm_mga_private_t *dev_priv);
 extern int mga_warp_init(drm_mga_private_t *dev_priv);
 
 				/* mga_irq.c */
-extern int mga_enable_vblank(struct drm_device *dev, int crtc);
-extern void mga_disable_vblank(struct drm_device *dev, int crtc);
-extern u32 mga_get_vblank_counter(struct drm_device *dev, int crtc);
+extern int mga_enable_vblank(struct drm_device *dev, unsigned int pipe);
+extern void mga_disable_vblank(struct drm_device *dev, unsigned int pipe);
+extern u32 mga_get_vblank_counter(struct drm_device *dev, unsigned int pipe);
 extern int mga_driver_fence_wait(struct drm_device *dev, unsigned int *sequence);
 extern int mga_driver_vblank_wait(struct drm_device *dev, unsigned int *sequence);
 extern irqreturn_t mga_driver_irq_handler(int irq, void *arg);
diff --git a/drivers/gpu/drm/mga/mga_irq.c b/drivers/gpu/drm/mga/mga_irq.c
index 1b071b8..693ba70 100644
--- a/drivers/gpu/drm/mga/mga_irq.c
+++ b/drivers/gpu/drm/mga/mga_irq.c
@@ -35,12 +35,12 @@
 #include <drm/mga_drm.h>
 #include "mga_drv.h"
 
-u32 mga_get_vblank_counter(struct drm_device *dev, int crtc)
+u32 mga_get_vblank_counter(struct drm_device *dev, unsigned int pipe)
 {
 	const drm_mga_private_t *const dev_priv =
 		(drm_mga_private_t *) dev->dev_private;
 
-	if (crtc != 0)
+	if (pipe != 0)
 		return 0;
 
 	return atomic_read(&dev_priv->vbl_received);
@@ -88,13 +88,13 @@ irqreturn_t mga_driver_irq_handler(int irq, void *arg)
 	return IRQ_NONE;
 }
 
-int mga_enable_vblank(struct drm_device *dev, int crtc)
+int mga_enable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	drm_mga_private_t *dev_priv = (drm_mga_private_t *) dev->dev_private;
 
-	if (crtc != 0) {
-		DRM_ERROR("tried to enable vblank on non-existent crtc %d\n",
-			  crtc);
+	if (pipe != 0) {
+		DRM_ERROR("tried to enable vblank on non-existent crtc %u\n",
+			  pipe);
 		return 0;
 	}
 
@@ -103,11 +103,11 @@ int mga_enable_vblank(struct drm_device *dev, int crtc)
 }
 
 
-void mga_disable_vblank(struct drm_device *dev, int crtc)
+void mga_disable_vblank(struct drm_device *dev, unsigned int pipe)
 {
-	if (crtc != 0) {
-		DRM_ERROR("tried to disable vblank on non-existent crtc %d\n",
-			  crtc);
+	if (pipe != 0) {
+		DRM_ERROR("tried to disable vblank on non-existent crtc %u\n",
+			  pipe);
 	}
 
 	/* Do *NOT* disable the vertical refresh interrupt.  MGA doesn't have
diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c
index 0339c5d..7e44511 100644
--- a/drivers/gpu/drm/msm/msm_drv.c
+++ b/drivers/gpu/drm/msm/msm_drv.c
@@ -531,24 +531,24 @@ static void msm_irq_uninstall(struct drm_device *dev)
 	kms->funcs->irq_uninstall(kms);
 }
 
-static int msm_enable_vblank(struct drm_device *dev, int crtc_id)
+static int msm_enable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	struct msm_drm_private *priv = dev->dev_private;
 	struct msm_kms *kms = priv->kms;
 	if (!kms)
 		return -ENXIO;
-	DBG("dev=%p, crtc=%d", dev, crtc_id);
-	return vblank_ctrl_queue_work(priv, crtc_id, true);
+	DBG("dev=%p, crtc=%u", dev, pipe);
+	return vblank_ctrl_queue_work(priv, pipe, true);
 }
 
-static void msm_disable_vblank(struct drm_device *dev, int crtc_id)
+static void msm_disable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	struct msm_drm_private *priv = dev->dev_private;
 	struct msm_kms *kms = priv->kms;
 	if (!kms)
 		return;
-	DBG("dev=%p, crtc=%d", dev, crtc_id);
-	vblank_ctrl_queue_work(priv, crtc_id, false);
+	DBG("dev=%p, crtc=%u", dev, pipe);
+	vblank_ctrl_queue_work(priv, pipe, false);
 }
 
 /*
diff --git a/drivers/gpu/drm/nouveau/nouveau_display.c b/drivers/gpu/drm/nouveau/nouveau_display.c
index a82c3cb..886079d 100644
--- a/drivers/gpu/drm/nouveau/nouveau_display.c
+++ b/drivers/gpu/drm/nouveau/nouveau_display.c
@@ -51,12 +51,12 @@ nouveau_display_vblank_handler(struct nvif_notify *notify)
 }
 
 int
-nouveau_display_vblank_enable(struct drm_device *dev, int head)
+nouveau_display_vblank_enable(struct drm_device *dev, unsigned int pipe)
 {
 	struct drm_crtc *crtc;
 	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
 		struct nouveau_crtc *nv_crtc = nouveau_crtc(crtc);
-		if (nv_crtc->index == head) {
+		if (nv_crtc->index == pipe) {
 			nvif_notify_get(&nv_crtc->vblank);
 			return 0;
 		}
@@ -65,12 +65,12 @@ nouveau_display_vblank_enable(struct drm_device *dev, int head)
 }
 
 void
-nouveau_display_vblank_disable(struct drm_device *dev, int head)
+nouveau_display_vblank_disable(struct drm_device *dev, unsigned int pipe)
 {
 	struct drm_crtc *crtc;
 	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
 		struct nouveau_crtc *nv_crtc = nouveau_crtc(crtc);
-		if (nv_crtc->index == head) {
+		if (nv_crtc->index == pipe) {
 			nvif_notify_put(&nv_crtc->vblank);
 			return;
 		}
@@ -132,14 +132,15 @@ nouveau_display_scanoutpos_head(struct drm_crtc *crtc, int *vpos, int *hpos,
 }
 
 int
-nouveau_display_scanoutpos(struct drm_device *dev, int head, unsigned int flags,
-			   int *vpos, int *hpos, ktime_t *stime, ktime_t *etime,
+nouveau_display_scanoutpos(struct drm_device *dev, unsigned int pipe,
+			   unsigned int flags, int *vpos, int *hpos,
+			   ktime_t *stime, ktime_t *etime,
 			   const struct drm_display_mode *mode)
 {
 	struct drm_crtc *crtc;
 
 	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
-		if (nouveau_crtc(crtc)->index == head) {
+		if (nouveau_crtc(crtc)->index == pipe) {
 			return nouveau_display_scanoutpos_head(crtc, vpos, hpos,
 							       stime, etime);
 		}
@@ -149,15 +150,15 @@ nouveau_display_scanoutpos(struct drm_device *dev, int head, unsigned int flags,
 }
 
 int
-nouveau_display_vblstamp(struct drm_device *dev, int head, int *max_error,
-			 struct timeval *time, unsigned flags)
+nouveau_display_vblstamp(struct drm_device *dev, unsigned int pipe,
+			 int *max_error, struct timeval *time, unsigned flags)
 {
 	struct drm_crtc *crtc;
 
 	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
-		if (nouveau_crtc(crtc)->index == head) {
+		if (nouveau_crtc(crtc)->index == pipe) {
 			return drm_calc_vbltimestamp_from_scanoutpos(dev,
-					head, max_error, time, flags,
+					pipe, max_error, time, flags,
 					&crtc->hwmode);
 		}
 	}
diff --git a/drivers/gpu/drm/nouveau/nouveau_display.h b/drivers/gpu/drm/nouveau/nouveau_display.h
index 4182d21..856abe0 100644
--- a/drivers/gpu/drm/nouveau/nouveau_display.h
+++ b/drivers/gpu/drm/nouveau/nouveau_display.h
@@ -65,12 +65,12 @@ int  nouveau_display_init(struct drm_device *dev);
 void nouveau_display_fini(struct drm_device *dev);
 int  nouveau_display_suspend(struct drm_device *dev, bool runtime);
 void nouveau_display_resume(struct drm_device *dev, bool runtime);
-int  nouveau_display_vblank_enable(struct drm_device *, int);
-void nouveau_display_vblank_disable(struct drm_device *, int);
-int  nouveau_display_scanoutpos(struct drm_device *, int, unsigned int,
-				int *, int *, ktime_t *, ktime_t *,
-				const struct drm_display_mode *);
-int  nouveau_display_vblstamp(struct drm_device *, int, int *,
+int  nouveau_display_vblank_enable(struct drm_device *, unsigned int);
+void nouveau_display_vblank_disable(struct drm_device *, unsigned int);
+int  nouveau_display_scanoutpos(struct drm_device *, unsigned int,
+				unsigned int, int *, int *, ktime_t *,
+				ktime_t *, const struct drm_display_mode *);
+int  nouveau_display_vblstamp(struct drm_device *, unsigned int, int *,
 			      struct timeval *, unsigned);
 
 int  nouveau_crtc_page_flip(struct drm_crtc *crtc, struct drm_framebuffer *fb,
diff --git a/drivers/gpu/drm/omapdrm/omap_drv.h b/drivers/gpu/drm/omapdrm/omap_drv.h
index 12081e6..5c367aa 100644
--- a/drivers/gpu/drm/omapdrm/omap_drv.h
+++ b/drivers/gpu/drm/omapdrm/omap_drv.h
@@ -129,8 +129,8 @@ void omap_gem_describe_objects(struct list_head *list, struct seq_file *m);
 int omap_gem_resume(struct device *dev);
 #endif
 
-int omap_irq_enable_vblank(struct drm_device *dev, int crtc_id);
-void omap_irq_disable_vblank(struct drm_device *dev, int crtc_id);
+int omap_irq_enable_vblank(struct drm_device *dev, unsigned int pipe);
+void omap_irq_disable_vblank(struct drm_device *dev, unsigned int pipe);
 void __omap_irq_register(struct drm_device *dev, struct omap_drm_irq *irq);
 void __omap_irq_unregister(struct drm_device *dev, struct omap_drm_irq *irq);
 void omap_irq_register(struct drm_device *dev, struct omap_drm_irq *irq);
diff --git a/drivers/gpu/drm/omapdrm/omap_irq.c b/drivers/gpu/drm/omapdrm/omap_irq.c
index 249c0330..60e1e80 100644
--- a/drivers/gpu/drm/omapdrm/omap_irq.c
+++ b/drivers/gpu/drm/omapdrm/omap_irq.c
@@ -134,7 +134,7 @@ int omap_irq_wait(struct drm_device *dev, struct omap_irq_wait *wait,
 /**
  * enable_vblank - enable vblank interrupt events
  * @dev: DRM device
- * @crtc: which irq to enable
+ * @pipe: which irq to enable
  *
  * Enable vblank interrupts for @crtc.  If the device doesn't have
  * a hardware vblank counter, this routine should be a no-op, since
@@ -144,13 +144,13 @@ int omap_irq_wait(struct drm_device *dev, struct omap_irq_wait *wait,
  * Zero on success, appropriate errno if the given @crtc's vblank
  * interrupt cannot be enabled.
  */
-int omap_irq_enable_vblank(struct drm_device *dev, int crtc_id)
+int omap_irq_enable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	struct omap_drm_private *priv = dev->dev_private;
-	struct drm_crtc *crtc = priv->crtcs[crtc_id];
+	struct drm_crtc *crtc = priv->crtcs[pipe];
 	unsigned long flags;
 
-	DBG("dev=%p, crtc=%d", dev, crtc_id);
+	DBG("dev=%p, crtc=%u", dev, pipe);
 
 	spin_lock_irqsave(&list_lock, flags);
 	priv->vblank_mask |= pipe2vbl(crtc);
@@ -163,19 +163,19 @@ int omap_irq_enable_vblank(struct drm_device *dev, int crtc_id)
 /**
  * disable_vblank - disable vblank interrupt events
  * @dev: DRM device
- * @crtc: which irq to enable
+ * @pipe: which irq to enable
  *
  * Disable vblank interrupts for @crtc.  If the device doesn't have
  * a hardware vblank counter, this routine should be a no-op, since
  * interrupts will have to stay on to keep the count accurate.
  */
-void omap_irq_disable_vblank(struct drm_device *dev, int crtc_id)
+void omap_irq_disable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	struct omap_drm_private *priv = dev->dev_private;
-	struct drm_crtc *crtc = priv->crtcs[crtc_id];
+	struct drm_crtc *crtc = priv->crtcs[pipe];
 	unsigned long flags;
 
-	DBG("dev=%p, crtc=%d", dev, crtc_id);
+	DBG("dev=%p, crtc=%u", dev, pipe);
 
 	spin_lock_irqsave(&list_lock, flags);
 	priv->vblank_mask &= ~pipe2vbl(crtc);
diff --git a/drivers/gpu/drm/qxl/qxl_drv.c b/drivers/gpu/drm/qxl/qxl_drv.c
index 83f6f0b..7307b07 100644
--- a/drivers/gpu/drm/qxl/qxl_drv.c
+++ b/drivers/gpu/drm/qxl/qxl_drv.c
@@ -196,17 +196,18 @@ static int qxl_pm_restore(struct device *dev)
 	return qxl_drm_resume(drm_dev, false);
 }
 
-static u32 qxl_noop_get_vblank_counter(struct drm_device *dev, int crtc)
+static u32 qxl_noop_get_vblank_counter(struct drm_device *dev,
+				       unsigned int pipe)
 {
 	return 0;
 }
 
-static int qxl_noop_enable_vblank(struct drm_device *dev, int crtc)
+static int qxl_noop_enable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	return 0;
 }
 
-static void qxl_noop_disable_vblank(struct drm_device *dev, int crtc)
+static void qxl_noop_disable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 }
 
diff --git a/drivers/gpu/drm/r128/r128_drv.h b/drivers/gpu/drm/r128/r128_drv.h
index 723e5d6..09143b8 100644
--- a/drivers/gpu/drm/r128/r128_drv.h
+++ b/drivers/gpu/drm/r128/r128_drv.h
@@ -154,9 +154,9 @@ extern int r128_wait_ring(drm_r128_private_t *dev_priv, int n);
 extern int r128_do_cce_idle(drm_r128_private_t *dev_priv);
 extern int r128_do_cleanup_cce(struct drm_device *dev);
 
-extern int r128_enable_vblank(struct drm_device *dev, int crtc);
-extern void r128_disable_vblank(struct drm_device *dev, int crtc);
-extern u32 r128_get_vblank_counter(struct drm_device *dev, int crtc);
+extern int r128_enable_vblank(struct drm_device *dev, unsigned int pipe);
+extern void r128_disable_vblank(struct drm_device *dev, unsigned int pipe);
+extern u32 r128_get_vblank_counter(struct drm_device *dev, unsigned int pipe);
 extern irqreturn_t r128_driver_irq_handler(int irq, void *arg);
 extern void r128_driver_irq_preinstall(struct drm_device *dev);
 extern int r128_driver_irq_postinstall(struct drm_device *dev);
diff --git a/drivers/gpu/drm/r128/r128_irq.c b/drivers/gpu/drm/r128/r128_irq.c
index c2ae496..9730f49 100644
--- a/drivers/gpu/drm/r128/r128_irq.c
+++ b/drivers/gpu/drm/r128/r128_irq.c
@@ -34,11 +34,11 @@
 #include <drm/r128_drm.h>
 #include "r128_drv.h"
 
-u32 r128_get_vblank_counter(struct drm_device *dev, int crtc)
+u32 r128_get_vblank_counter(struct drm_device *dev, unsigned int pipe)
 {
 	const drm_r128_private_t *dev_priv = dev->dev_private;
 
-	if (crtc != 0)
+	if (pipe != 0)
 		return 0;
 
 	return atomic_read(&dev_priv->vbl_received);
@@ -62,12 +62,12 @@ irqreturn_t r128_driver_irq_handler(int irq, void *arg)
 	return IRQ_NONE;
 }
 
-int r128_enable_vblank(struct drm_device *dev, int crtc)
+int r128_enable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	drm_r128_private_t *dev_priv = dev->dev_private;
 
-	if (crtc != 0) {
-		DRM_ERROR("%s:  bad crtc %d\n", __func__, crtc);
+	if (pipe != 0) {
+		DRM_ERROR("%s:  bad crtc %u\n", __func__, pipe);
 		return -EINVAL;
 	}
 
@@ -75,10 +75,10 @@ int r128_enable_vblank(struct drm_device *dev, int crtc)
 	return 0;
 }
 
-void r128_disable_vblank(struct drm_device *dev, int crtc)
+void r128_disable_vblank(struct drm_device *dev, unsigned int pipe)
 {
-	if (crtc != 0)
-		DRM_ERROR("%s:  bad crtc %d\n", __func__, crtc);
+	if (pipe != 0)
+		DRM_ERROR("%s:  bad crtc %u\n", __func__, pipe);
 
 	/*
 	 * FIXME: implement proper interrupt disable by using the vblank
diff --git a/drivers/gpu/drm/radeon/radeon_display.c b/drivers/gpu/drm/radeon/radeon_display.c
index 0503af7..a58635c 100644
--- a/drivers/gpu/drm/radeon/radeon_display.c
+++ b/drivers/gpu/drm/radeon/radeon_display.c
@@ -1799,8 +1799,9 @@ bool radeon_crtc_scaling_mode_fixup(struct drm_crtc *crtc,
  * unknown small number of scanlines wrt. real scanout position.
  *
  */
-int radeon_get_crtc_scanoutpos(struct drm_device *dev, int crtc, unsigned int flags,
-			       int *vpos, int *hpos, ktime_t *stime, ktime_t *etime,
+int radeon_get_crtc_scanoutpos(struct drm_device *dev, unsigned int pipe,
+			       unsigned int flags, int *vpos, int *hpos,
+			       ktime_t *stime, ktime_t *etime,
 			       const struct drm_display_mode *mode)
 {
 	u32 stat_crtc = 0, vbl = 0, position = 0;
@@ -1816,42 +1817,42 @@ int radeon_get_crtc_scanoutpos(struct drm_device *dev, int crtc, unsigned int fl
 		*stime = ktime_get();
 
 	if (ASIC_IS_DCE4(rdev)) {
-		if (crtc == 0) {
+		if (pipe == 0) {
 			vbl = RREG32(EVERGREEN_CRTC_V_BLANK_START_END +
 				     EVERGREEN_CRTC0_REGISTER_OFFSET);
 			position = RREG32(EVERGREEN_CRTC_STATUS_POSITION +
 					  EVERGREEN_CRTC0_REGISTER_OFFSET);
 			ret |= DRM_SCANOUTPOS_VALID;
 		}
-		if (crtc == 1) {
+		if (pipe == 1) {
 			vbl = RREG32(EVERGREEN_CRTC_V_BLANK_START_END +
 				     EVERGREEN_CRTC1_REGISTER_OFFSET);
 			position = RREG32(EVERGREEN_CRTC_STATUS_POSITION +
 					  EVERGREEN_CRTC1_REGISTER_OFFSET);
 			ret |= DRM_SCANOUTPOS_VALID;
 		}
-		if (crtc == 2) {
+		if (pipe == 2) {
 			vbl = RREG32(EVERGREEN_CRTC_V_BLANK_START_END +
 				     EVERGREEN_CRTC2_REGISTER_OFFSET);
 			position = RREG32(EVERGREEN_CRTC_STATUS_POSITION +
 					  EVERGREEN_CRTC2_REGISTER_OFFSET);
 			ret |= DRM_SCANOUTPOS_VALID;
 		}
-		if (crtc == 3) {
+		if (pipe == 3) {
 			vbl = RREG32(EVERGREEN_CRTC_V_BLANK_START_END +
 				     EVERGREEN_CRTC3_REGISTER_OFFSET);
 			position = RREG32(EVERGREEN_CRTC_STATUS_POSITION +
 					  EVERGREEN_CRTC3_REGISTER_OFFSET);
 			ret |= DRM_SCANOUTPOS_VALID;
 		}
-		if (crtc == 4) {
+		if (pipe == 4) {
 			vbl = RREG32(EVERGREEN_CRTC_V_BLANK_START_END +
 				     EVERGREEN_CRTC4_REGISTER_OFFSET);
 			position = RREG32(EVERGREEN_CRTC_STATUS_POSITION +
 					  EVERGREEN_CRTC4_REGISTER_OFFSET);
 			ret |= DRM_SCANOUTPOS_VALID;
 		}
-		if (crtc == 5) {
+		if (pipe == 5) {
 			vbl = RREG32(EVERGREEN_CRTC_V_BLANK_START_END +
 				     EVERGREEN_CRTC5_REGISTER_OFFSET);
 			position = RREG32(EVERGREEN_CRTC_STATUS_POSITION +
@@ -1859,19 +1860,19 @@ int radeon_get_crtc_scanoutpos(struct drm_device *dev, int crtc, unsigned int fl
 			ret |= DRM_SCANOUTPOS_VALID;
 		}
 	} else if (ASIC_IS_AVIVO(rdev)) {
-		if (crtc == 0) {
+		if (pipe == 0) {
 			vbl = RREG32(AVIVO_D1CRTC_V_BLANK_START_END);
 			position = RREG32(AVIVO_D1CRTC_STATUS_POSITION);
 			ret |= DRM_SCANOUTPOS_VALID;
 		}
-		if (crtc == 1) {
+		if (pipe == 1) {
 			vbl = RREG32(AVIVO_D2CRTC_V_BLANK_START_END);
 			position = RREG32(AVIVO_D2CRTC_STATUS_POSITION);
 			ret |= DRM_SCANOUTPOS_VALID;
 		}
 	} else {
 		/* Pre-AVIVO: Different encoding of scanout pos and vblank interval. */
-		if (crtc == 0) {
+		if (pipe == 0) {
 			/* Assume vbl_end == 0, get vbl_start from
 			 * upper 16 bits.
 			 */
@@ -1885,7 +1886,7 @@ int radeon_get_crtc_scanoutpos(struct drm_device *dev, int crtc, unsigned int fl
 
 			ret |= DRM_SCANOUTPOS_VALID;
 		}
-		if (crtc == 1) {
+		if (pipe == 1) {
 			vbl = (RREG32(RADEON_CRTC2_V_TOTAL_DISP) &
 				RADEON_CRTC_V_DISP) >> RADEON_CRTC_V_DISP_SHIFT;
 			position = (RREG32(RADEON_CRTC2_VLINE_CRNT_VLINE) >> 16) & RADEON_CRTC_V_TOTAL;
diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c
index e30c1d7..5b6a6f5 100644
--- a/drivers/gpu/drm/radeon/radeon_drv.c
+++ b/drivers/gpu/drm/radeon/radeon_drv.c
@@ -105,10 +105,10 @@ void radeon_driver_preclose_kms(struct drm_device *dev,
 				struct drm_file *file_priv);
 int radeon_suspend_kms(struct drm_device *dev, bool suspend, bool fbcon);
 int radeon_resume_kms(struct drm_device *dev, bool resume, bool fbcon);
-u32 radeon_get_vblank_counter_kms(struct drm_device *dev, int crtc);
-int radeon_enable_vblank_kms(struct drm_device *dev, int crtc);
-void radeon_disable_vblank_kms(struct drm_device *dev, int crtc);
-int radeon_get_vblank_timestamp_kms(struct drm_device *dev, int crtc,
+u32 radeon_get_vblank_counter_kms(struct drm_device *dev, unsigned int pipe);
+int radeon_enable_vblank_kms(struct drm_device *dev, unsigned int pipe);
+void radeon_disable_vblank_kms(struct drm_device *dev, unsigned int pipe);
+int radeon_get_vblank_timestamp_kms(struct drm_device *dev, unsigned int pipe,
 				    int *max_error,
 				    struct timeval *vblank_time,
 				    unsigned flags);
@@ -124,9 +124,8 @@ void radeon_gem_object_close(struct drm_gem_object *obj,
 struct dma_buf *radeon_gem_prime_export(struct drm_device *dev,
 					struct drm_gem_object *gobj,
 					int flags);
-extern int radeon_get_crtc_scanoutpos(struct drm_device *dev, int crtc,
-				      unsigned int flags,
-				      int *vpos, int *hpos,
+extern int radeon_get_crtc_scanoutpos(struct drm_device *dev, unsigned int crtc,
+				      unsigned int flags, int *vpos, int *hpos,
 				      ktime_t *stime, ktime_t *etime,
 				      const struct drm_display_mode *mode);
 extern bool radeon_is_px(struct drm_device *dev);
diff --git a/drivers/gpu/drm/radeon/radeon_drv.h b/drivers/gpu/drm/radeon/radeon_drv.h
index 46bd393..0caafc7 100644
--- a/drivers/gpu/drm/radeon/radeon_drv.h
+++ b/drivers/gpu/drm/radeon/radeon_drv.h
@@ -404,9 +404,9 @@ extern int radeon_irq_emit(struct drm_device *dev, void *data, struct drm_file *
 extern int radeon_irq_wait(struct drm_device *dev, void *data, struct drm_file *file_priv);
 
 extern void radeon_do_release(struct drm_device * dev);
-extern u32 radeon_get_vblank_counter(struct drm_device *dev, int crtc);
-extern int radeon_enable_vblank(struct drm_device *dev, int crtc);
-extern void radeon_disable_vblank(struct drm_device *dev, int crtc);
+extern u32 radeon_get_vblank_counter(struct drm_device *dev, unsigned int pipe);
+extern int radeon_enable_vblank(struct drm_device *dev, unsigned int pipe);
+extern void radeon_disable_vblank(struct drm_device *dev, unsigned int pipe);
 extern irqreturn_t radeon_driver_irq_handler(int irq, void *arg);
 extern void radeon_driver_irq_preinstall(struct drm_device * dev);
 extern int radeon_driver_irq_postinstall(struct drm_device *dev);
diff --git a/drivers/gpu/drm/radeon/radeon_irq.c b/drivers/gpu/drm/radeon/radeon_irq.c
index 244b19b..688afb6 100644
--- a/drivers/gpu/drm/radeon/radeon_irq.c
+++ b/drivers/gpu/drm/radeon/radeon_irq.c
@@ -62,12 +62,12 @@ static void r500_vbl_irq_set_state(struct drm_device *dev, u32 mask, int state)
 		RADEON_WRITE(R500_DxMODE_INT_MASK, dev_priv->r500_disp_irq_reg);
 }
 
-int radeon_enable_vblank(struct drm_device *dev, int crtc)
+int radeon_enable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	drm_radeon_private_t *dev_priv = dev->dev_private;
 
 	if ((dev_priv->flags & RADEON_FAMILY_MASK) >= CHIP_RS600) {
-		switch (crtc) {
+		switch (pipe) {
 		case 0:
 			r500_vbl_irq_set_state(dev, R500_D1MODE_INT_MASK, 1);
 			break;
@@ -75,12 +75,12 @@ int radeon_enable_vblank(struct drm_device *dev, int crtc)
 			r500_vbl_irq_set_state(dev, R500_D2MODE_INT_MASK, 1);
 			break;
 		default:
-			DRM_ERROR("tried to enable vblank on non-existent crtc %d\n",
-				  crtc);
+			DRM_ERROR("tried to enable vblank on non-existent crtc %u\n",
+				  pipe);
 			return -EINVAL;
 		}
 	} else {
-		switch (crtc) {
+		switch (pipe) {
 		case 0:
 			radeon_irq_set_state(dev, RADEON_CRTC_VBLANK_MASK, 1);
 			break;
@@ -88,8 +88,8 @@ int radeon_enable_vblank(struct drm_device *dev, int crtc)
 			radeon_irq_set_state(dev, RADEON_CRTC2_VBLANK_MASK, 1);
 			break;
 		default:
-			DRM_ERROR("tried to enable vblank on non-existent crtc %d\n",
-				  crtc);
+			DRM_ERROR("tried to enable vblank on non-existent crtc %u\n",
+				  pipe);
 			return -EINVAL;
 		}
 	}
@@ -97,12 +97,12 @@ int radeon_enable_vblank(struct drm_device *dev, int crtc)
 	return 0;
 }
 
-void radeon_disable_vblank(struct drm_device *dev, int crtc)
+void radeon_disable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	drm_radeon_private_t *dev_priv = dev->dev_private;
 
 	if ((dev_priv->flags & RADEON_FAMILY_MASK) >= CHIP_RS600) {
-		switch (crtc) {
+		switch (pipe) {
 		case 0:
 			r500_vbl_irq_set_state(dev, R500_D1MODE_INT_MASK, 0);
 			break;
@@ -110,12 +110,12 @@ void radeon_disable_vblank(struct drm_device *dev, int crtc)
 			r500_vbl_irq_set_state(dev, R500_D2MODE_INT_MASK, 0);
 			break;
 		default:
-			DRM_ERROR("tried to enable vblank on non-existent crtc %d\n",
-				  crtc);
+			DRM_ERROR("tried to enable vblank on non-existent crtc %u\n",
+				  pipe);
 			break;
 		}
 	} else {
-		switch (crtc) {
+		switch (pipe) {
 		case 0:
 			radeon_irq_set_state(dev, RADEON_CRTC_VBLANK_MASK, 0);
 			break;
@@ -123,8 +123,8 @@ void radeon_disable_vblank(struct drm_device *dev, int crtc)
 			radeon_irq_set_state(dev, RADEON_CRTC2_VBLANK_MASK, 0);
 			break;
 		default:
-			DRM_ERROR("tried to enable vblank on non-existent crtc %d\n",
-				  crtc);
+			DRM_ERROR("tried to enable vblank on non-existent crtc %u\n",
+				  pipe);
 			break;
 		}
 	}
@@ -255,7 +255,7 @@ static int radeon_wait_irq(struct drm_device * dev, int swi_nr)
 	return ret;
 }
 
-u32 radeon_get_vblank_counter(struct drm_device *dev, int crtc)
+u32 radeon_get_vblank_counter(struct drm_device *dev, unsigned int pipe)
 {
 	drm_radeon_private_t *dev_priv = dev->dev_private;
 
@@ -264,18 +264,18 @@ u32 radeon_get_vblank_counter(struct drm_device *dev, int crtc)
 		return -EINVAL;
 	}
 
-	if (crtc < 0 || crtc > 1) {
-		DRM_ERROR("Invalid crtc %d\n", crtc);
+	if (pipe > 1) {
+		DRM_ERROR("Invalid crtc %u\n", pipe);
 		return -EINVAL;
 	}
 
 	if ((dev_priv->flags & RADEON_FAMILY_MASK) >= CHIP_RS600) {
-		if (crtc == 0)
+		if (pipe == 0)
 			return RADEON_READ(R500_D1CRTC_FRAME_COUNT);
 		else
 			return RADEON_READ(R500_D2CRTC_FRAME_COUNT);
 	} else {
-		if (crtc == 0)
+		if (pipe == 0)
 			return RADEON_READ(RADEON_CRTC_CRNT_FRAME);
 		else
 			return RADEON_READ(RADEON_CRTC2_CRNT_FRAME);
diff --git a/drivers/gpu/drm/radeon/radeon_mode.h b/drivers/gpu/drm/radeon/radeon_mode.h
index 2317d04..de18f06 100644
--- a/drivers/gpu/drm/radeon/radeon_mode.h
+++ b/drivers/gpu/drm/radeon/radeon_mode.h
@@ -874,9 +874,8 @@ extern int radeon_crtc_cursor_move(struct drm_crtc *crtc,
 				   int x, int y);
 extern void radeon_cursor_reset(struct drm_crtc *crtc);
 
-extern int radeon_get_crtc_scanoutpos(struct drm_device *dev, int crtc,
-				      unsigned int flags,
-				      int *vpos, int *hpos,
+extern int radeon_get_crtc_scanoutpos(struct drm_device *dev, unsigned int pipe,
+				      unsigned int flags, int *vpos, int *hpos,
 				      ktime_t *stime, ktime_t *etime,
 				      const struct drm_display_mode *mode);
 
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_drv.c b/drivers/gpu/drm/rcar-du/rcar_du_drv.c
index 780ca11..bb806c4 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_drv.c
+++ b/drivers/gpu/drm/rcar-du/rcar_du_drv.c
@@ -221,20 +221,20 @@ static void rcar_du_lastclose(struct drm_device *dev)
 	drm_fbdev_cma_restore_mode(rcdu->fbdev);
 }
 
-static int rcar_du_enable_vblank(struct drm_device *dev, int crtc)
+static int rcar_du_enable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	struct rcar_du_device *rcdu = dev->dev_private;
 
-	rcar_du_crtc_enable_vblank(&rcdu->crtcs[crtc], true);
+	rcar_du_crtc_enable_vblank(&rcdu->crtcs[pipe], true);
 
 	return 0;
 }
 
-static void rcar_du_disable_vblank(struct drm_device *dev, int crtc)
+static void rcar_du_disable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	struct rcar_du_device *rcdu = dev->dev_private;
 
-	rcar_du_crtc_enable_vblank(&rcdu->crtcs[crtc], false);
+	rcar_du_crtc_enable_vblank(&rcdu->crtcs[pipe], false);
 }
 
 static const struct file_operations rcar_du_fops = {
diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_drv.c b/drivers/gpu/drm/rockchip/rockchip_drm_drv.c
index 9a0c291..32c6098 100644
--- a/drivers/gpu/drm/rockchip/rockchip_drm_drv.c
+++ b/drivers/gpu/drm/rockchip/rockchip_drm_drv.c
@@ -103,7 +103,8 @@ static struct drm_crtc *rockchip_crtc_from_pipe(struct drm_device *drm,
 	return NULL;
 }
 
-static int rockchip_drm_crtc_enable_vblank(struct drm_device *dev, int pipe)
+static int rockchip_drm_crtc_enable_vblank(struct drm_device *dev,
+					   unsigned int pipe)
 {
 	struct rockchip_drm_private *priv = dev->dev_private;
 	struct drm_crtc *crtc = rockchip_crtc_from_pipe(dev, pipe);
@@ -115,7 +116,8 @@ static int rockchip_drm_crtc_enable_vblank(struct drm_device *dev, int pipe)
 	return 0;
 }
 
-static void rockchip_drm_crtc_disable_vblank(struct drm_device *dev, int pipe)
+static void rockchip_drm_crtc_disable_vblank(struct drm_device *dev,
+					     unsigned int pipe)
 {
 	struct rockchip_drm_private *priv = dev->dev_private;
 	struct drm_crtc *crtc = rockchip_crtc_from_pipe(dev, pipe);
diff --git a/drivers/gpu/drm/shmobile/shmob_drm_drv.c b/drivers/gpu/drm/shmobile/shmob_drm_drv.c
index 666321d..ca2f918 100644
--- a/drivers/gpu/drm/shmobile/shmob_drm_drv.c
+++ b/drivers/gpu/drm/shmobile/shmob_drm_drv.c
@@ -231,7 +231,7 @@ static irqreturn_t shmob_drm_irq(int irq, void *arg)
 	return IRQ_HANDLED;
 }
 
-static int shmob_drm_enable_vblank(struct drm_device *dev, int crtc)
+static int shmob_drm_enable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	struct shmob_drm_device *sdev = dev->dev_private;
 
@@ -240,7 +240,7 @@ static int shmob_drm_enable_vblank(struct drm_device *dev, int crtc)
 	return 0;
 }
 
-static void shmob_drm_disable_vblank(struct drm_device *dev, int crtc)
+static void shmob_drm_disable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	struct shmob_drm_device *sdev = dev->dev_private;
 
diff --git a/drivers/gpu/drm/sti/sti_crtc.c b/drivers/gpu/drm/sti/sti_crtc.c
index 018ffc9..493c4a3 100644
--- a/drivers/gpu/drm/sti/sti_crtc.c
+++ b/drivers/gpu/drm/sti/sti_crtc.c
@@ -299,7 +299,7 @@ int sti_crtc_vblank_cb(struct notifier_block *nb,
 	return 0;
 }
 
-int sti_crtc_enable_vblank(struct drm_device *dev, int crtc)
+int sti_crtc_enable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	struct sti_private *dev_priv = dev->dev_private;
 	struct sti_compositor *compo = dev_priv->compo;
@@ -307,9 +307,9 @@ int sti_crtc_enable_vblank(struct drm_device *dev, int crtc)
 
 	DRM_DEBUG_DRIVER("\n");
 
-	if (sti_vtg_register_client(crtc == STI_MIXER_MAIN ?
+	if (sti_vtg_register_client(pipe == STI_MIXER_MAIN ?
 			compo->vtg_main : compo->vtg_aux,
-			vtg_vblank_nb, crtc)) {
+			vtg_vblank_nb, pipe)) {
 		DRM_ERROR("Cannot register VTG notifier\n");
 		return -EINVAL;
 	}
@@ -318,7 +318,7 @@ int sti_crtc_enable_vblank(struct drm_device *dev, int crtc)
 }
 EXPORT_SYMBOL(sti_crtc_enable_vblank);
 
-void sti_crtc_disable_vblank(struct drm_device *drm_dev, int crtc)
+void sti_crtc_disable_vblank(struct drm_device *drm_dev, unsigned int pipe)
 {
 	struct sti_private *priv = drm_dev->dev_private;
 	struct sti_compositor *compo = priv->compo;
@@ -326,14 +326,14 @@ void sti_crtc_disable_vblank(struct drm_device *drm_dev, int crtc)
 
 	DRM_DEBUG_DRIVER("\n");
 
-	if (sti_vtg_unregister_client(crtc == STI_MIXER_MAIN ?
+	if (sti_vtg_unregister_client(pipe == STI_MIXER_MAIN ?
 			compo->vtg_main : compo->vtg_aux, vtg_vblank_nb))
 		DRM_DEBUG_DRIVER("Warning: cannot unregister VTG notifier\n");
 
 	/* free the resources of the pending requests */
-	if (compo->mixer[crtc]->pending_event) {
-		drm_vblank_put(drm_dev, crtc);
-		compo->mixer[crtc]->pending_event = NULL;
+	if (compo->mixer[pipe]->pending_event) {
+		drm_vblank_put(drm_dev, pipe);
+		compo->mixer[pipe]->pending_event = NULL;
 	}
 }
 EXPORT_SYMBOL(sti_crtc_disable_vblank);
diff --git a/drivers/gpu/drm/sti/sti_crtc.h b/drivers/gpu/drm/sti/sti_crtc.h
index 51963e6..3f2d89a 100644
--- a/drivers/gpu/drm/sti/sti_crtc.h
+++ b/drivers/gpu/drm/sti/sti_crtc.h
@@ -13,8 +13,8 @@ struct sti_mixer;
 
 int sti_crtc_init(struct drm_device *drm_dev, struct sti_mixer *mixer,
 		  struct drm_plane *primary, struct drm_plane *cursor);
-int sti_crtc_enable_vblank(struct drm_device *dev, int crtc);
-void sti_crtc_disable_vblank(struct drm_device *dev, int crtc);
+int sti_crtc_enable_vblank(struct drm_device *dev, unsigned int pipe);
+void sti_crtc_disable_vblank(struct drm_device *dev, unsigned int pipe);
 int sti_crtc_vblank_cb(struct notifier_block *nb,
 		       unsigned long event, void *data);
 bool sti_crtc_is_main(struct drm_crtc *drm_crtc);
diff --git a/drivers/gpu/drm/tegra/drm.c b/drivers/gpu/drm/tegra/drm.c
index 2486bc2..759e6af 100644
--- a/drivers/gpu/drm/tegra/drm.c
+++ b/drivers/gpu/drm/tegra/drm.c
@@ -822,7 +822,8 @@ static struct drm_crtc *tegra_crtc_from_pipe(struct drm_device *drm,
 	return NULL;
 }
 
-static u32 tegra_drm_get_vblank_counter(struct drm_device *drm, int pipe)
+static u32 tegra_drm_get_vblank_counter(struct drm_device *drm,
+					unsigned int pipe)
 {
 	struct drm_crtc *crtc = tegra_crtc_from_pipe(drm, pipe);
 	struct tegra_dc *dc = to_tegra_dc(crtc);
@@ -833,7 +834,7 @@ static u32 tegra_drm_get_vblank_counter(struct drm_device *drm, int pipe)
 	return tegra_dc_get_vblank_counter(dc);
 }
 
-static int tegra_drm_enable_vblank(struct drm_device *drm, int pipe)
+static int tegra_drm_enable_vblank(struct drm_device *drm, unsigned int pipe)
 {
 	struct drm_crtc *crtc = tegra_crtc_from_pipe(drm, pipe);
 	struct tegra_dc *dc = to_tegra_dc(crtc);
@@ -846,7 +847,7 @@ static int tegra_drm_enable_vblank(struct drm_device *drm, int pipe)
 	return 0;
 }
 
-static void tegra_drm_disable_vblank(struct drm_device *drm, int pipe)
+static void tegra_drm_disable_vblank(struct drm_device *drm, unsigned int pipe)
 {
 	struct drm_crtc *crtc = tegra_crtc_from_pipe(drm, pipe);
 	struct tegra_dc *dc = to_tegra_dc(crtc);
diff --git a/drivers/gpu/drm/tilcdc/tilcdc_drv.c b/drivers/gpu/drm/tilcdc/tilcdc_drv.c
index 0f283a3..a5b8f5d 100644
--- a/drivers/gpu/drm/tilcdc/tilcdc_drv.c
+++ b/drivers/gpu/drm/tilcdc/tilcdc_drv.c
@@ -425,13 +425,13 @@ static void enable_vblank(struct drm_device *dev, bool enable)
 		tilcdc_clear(dev, reg, mask);
 }
 
-static int tilcdc_enable_vblank(struct drm_device *dev, int crtc)
+static int tilcdc_enable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	enable_vblank(dev, true);
 	return 0;
 }
 
-static void tilcdc_disable_vblank(struct drm_device *dev, int crtc)
+static void tilcdc_disable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	enable_vblank(dev, false);
 }
diff --git a/drivers/gpu/drm/via/via_drv.h b/drivers/gpu/drm/via/via_drv.h
index ef8c500..644093f 100644
--- a/drivers/gpu/drm/via/via_drv.h
+++ b/drivers/gpu/drm/via/via_drv.h
@@ -136,9 +136,9 @@ extern int via_init_context(struct drm_device *dev, int context);
 extern int via_final_context(struct drm_device *dev, int context);
 
 extern int via_do_cleanup_map(struct drm_device *dev);
-extern u32 via_get_vblank_counter(struct drm_device *dev, int crtc);
-extern int via_enable_vblank(struct drm_device *dev, int crtc);
-extern void via_disable_vblank(struct drm_device *dev, int crtc);
+extern u32 via_get_vblank_counter(struct drm_device *dev, unsigned int pipe);
+extern int via_enable_vblank(struct drm_device *dev, unsigned int pipe);
+extern void via_disable_vblank(struct drm_device *dev, unsigned int pipe);
 
 extern irqreturn_t via_driver_irq_handler(int irq, void *arg);
 extern void via_driver_irq_preinstall(struct drm_device *dev);
diff --git a/drivers/gpu/drm/via/via_irq.c b/drivers/gpu/drm/via/via_irq.c
index 1319433..ea8172c 100644
--- a/drivers/gpu/drm/via/via_irq.c
+++ b/drivers/gpu/drm/via/via_irq.c
@@ -95,10 +95,11 @@ static unsigned time_diff(struct timeval *now, struct timeval *then)
 		1000000 - (then->tv_usec - now->tv_usec);
 }
 
-u32 via_get_vblank_counter(struct drm_device *dev, int crtc)
+u32 via_get_vblank_counter(struct drm_device *dev, unsigned int pipe)
 {
 	drm_via_private_t *dev_priv = dev->dev_private;
-	if (crtc != 0)
+
+	if (pipe != 0)
 		return 0;
 
 	return atomic_read(&dev_priv->vbl_received);
@@ -170,13 +171,13 @@ static __inline__ void viadrv_acknowledge_irqs(drm_via_private_t *dev_priv)
 	}
 }
 
-int via_enable_vblank(struct drm_device *dev, int crtc)
+int via_enable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	drm_via_private_t *dev_priv = dev->dev_private;
 	u32 status;
 
-	if (crtc != 0) {
-		DRM_ERROR("%s:  bad crtc %d\n", __func__, crtc);
+	if (pipe != 0) {
+		DRM_ERROR("%s:  bad crtc %u\n", __func__, pipe);
 		return -EINVAL;
 	}
 
@@ -189,7 +190,7 @@ int via_enable_vblank(struct drm_device *dev, int crtc)
 	return 0;
 }
 
-void via_disable_vblank(struct drm_device *dev, int crtc)
+void via_disable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	drm_via_private_t *dev_priv = dev->dev_private;
 	u32 status;
@@ -200,8 +201,8 @@ void via_disable_vblank(struct drm_device *dev, int crtc)
 	VIA_WRITE8(0x83d4, 0x11);
 	VIA_WRITE8(0x83d5, VIA_READ8(0x83d5) & ~0x30);
 
-	if (crtc != 0)
-		DRM_ERROR("%s:  bad crtc %d\n", __func__, crtc);
+	if (pipe != 0)
+		DRM_ERROR("%s:  bad crtc %u\n", __func__, pipe);
 }
 
 static int
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
index f19fd39..a613bd4 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
@@ -914,9 +914,9 @@ void vmw_kms_idle_workqueues(struct vmw_master *vmaster);
 bool vmw_kms_validate_mode_vram(struct vmw_private *dev_priv,
 				uint32_t pitch,
 				uint32_t height);
-u32 vmw_get_vblank_counter(struct drm_device *dev, int crtc);
-int vmw_enable_vblank(struct drm_device *dev, int crtc);
-void vmw_disable_vblank(struct drm_device *dev, int crtc);
+u32 vmw_get_vblank_counter(struct drm_device *dev, unsigned int pipe);
+int vmw_enable_vblank(struct drm_device *dev, unsigned int pipe);
+void vmw_disable_vblank(struct drm_device *dev, unsigned int pipe);
 int vmw_kms_present(struct vmw_private *dev_priv,
 		    struct drm_file *file_priv,
 		    struct vmw_framebuffer *vfb,
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
index 15a6c01..03ffab2 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
@@ -1263,7 +1263,7 @@ bool vmw_kms_validate_mode_vram(struct vmw_private *dev_priv,
 /**
  * Function called by DRM code called with vbl_lock held.
  */
-u32 vmw_get_vblank_counter(struct drm_device *dev, int crtc)
+u32 vmw_get_vblank_counter(struct drm_device *dev, unsigned int pipe)
 {
 	return 0;
 }
@@ -1271,7 +1271,7 @@ u32 vmw_get_vblank_counter(struct drm_device *dev, int crtc)
 /**
  * Function called by DRM code called with vbl_lock held.
  */
-int vmw_enable_vblank(struct drm_device *dev, int crtc)
+int vmw_enable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 	return -ENOSYS;
 }
@@ -1279,7 +1279,7 @@ int vmw_enable_vblank(struct drm_device *dev, int crtc)
 /**
  * Function called by DRM code called with vbl_lock held.
  */
-void vmw_disable_vblank(struct drm_device *dev, int crtc)
+void vmw_disable_vblank(struct drm_device *dev, unsigned int pipe)
 {
 }
 
diff --git a/include/drm/drmP.h b/include/drm/drmP.h
index 967d8a0..1cb1e84 100644
--- a/include/drm/drmP.h
+++ b/include/drm/drmP.h
@@ -412,7 +412,7 @@ struct drm_driver {
 	/**
 	 * get_vblank_counter - get raw hardware vblank counter
 	 * @dev: DRM device
-	 * @crtc: counter to fetch
+	 * @pipe: counter to fetch
 	 *
 	 * Driver callback for fetching a raw hardware vblank counter for @crtc.
 	 * If a device doesn't have a hardware counter, the driver can simply
@@ -426,12 +426,12 @@ struct drm_driver {
 	 * RETURNS
 	 * Raw vblank counter value.
 	 */
-	u32 (*get_vblank_counter) (struct drm_device *dev, int crtc);
+	u32 (*get_vblank_counter) (struct drm_device *dev, unsigned int pipe);
 
 	/**
 	 * enable_vblank - enable vblank interrupt events
 	 * @dev: DRM device
-	 * @crtc: which irq to enable
+	 * @pipe: which irq to enable
 	 *
 	 * Enable vblank interrupts for @crtc.  If the device doesn't have
 	 * a hardware vblank counter, this routine should be a no-op, since
@@ -441,18 +441,18 @@ struct drm_driver {
 	 * Zero on success, appropriate errno if the given @crtc's vblank
 	 * interrupt cannot be enabled.
 	 */
-	int (*enable_vblank) (struct drm_device *dev, int crtc);
+	int (*enable_vblank) (struct drm_device *dev, unsigned int pipe);
 
 	/**
 	 * disable_vblank - disable vblank interrupt events
 	 * @dev: DRM device
-	 * @crtc: which irq to enable
+	 * @pipe: which irq to enable
 	 *
 	 * Disable vblank interrupts for @crtc.  If the device doesn't have
 	 * a hardware vblank counter, this routine should be a no-op, since
 	 * interrupts will have to stay on to keep the count accurate.
 	 */
-	void (*disable_vblank) (struct drm_device *dev, int crtc);
+	void (*disable_vblank) (struct drm_device *dev, unsigned int pipe);
 
 	/**
 	 * Called by \c drm_device_is_agp.  Typically used to determine if a
@@ -474,7 +474,7 @@ struct drm_driver {
 	 * optional accurate ktime_get timestamp of when position was measured.
 	 *
 	 * \param dev  DRM device.
-	 * \param crtc Id of the crtc to query.
+	 * \param pipe Id of the crtc to query.
 	 * \param flags Flags from the caller (DRM_CALLED_FROM_VBLIRQ or 0).
 	 * \param *vpos Target location for current vertical scanout position.
 	 * \param *hpos Target location for current horizontal scanout position.
@@ -498,9 +498,8 @@ struct drm_driver {
 	 * but unknown small number of scanlines wrt. real scanout position.
 	 *
 	 */
-	int (*get_scanout_position) (struct drm_device *dev, int crtc,
-				     unsigned int flags,
-				     int *vpos, int *hpos,
+	int (*get_scanout_position) (struct drm_device *dev, unsigned int pipe,
+				     unsigned int flags, int *vpos, int *hpos,
 				     ktime_t *stime, ktime_t *etime,
 				     const struct drm_display_mode *mode);
 
@@ -518,7 +517,7 @@ struct drm_driver {
 	 * to the OpenML OML_sync_control extension specification.
 	 *
 	 * \param dev dev DRM device handle.
-	 * \param crtc crtc for which timestamp should be returned.
+	 * \param pipe crtc for which timestamp should be returned.
 	 * \param *max_error Maximum allowable timestamp error in nanoseconds.
 	 *                   Implementation should strive to provide timestamp
 	 *                   with an error of at most *max_error nanoseconds.
@@ -534,7 +533,7 @@ struct drm_driver {
 	 * negative number on failure. A positive status code on success,
 	 * which describes how the vblank_time timestamp was computed.
 	 */
-	int (*get_vblank_timestamp) (struct drm_device *dev, int crtc,
+	int (*get_vblank_timestamp) (struct drm_device *dev, unsigned int pipe,
 				     int *max_error,
 				     struct timeval *vblank_time,
 				     unsigned flags);
@@ -930,7 +929,7 @@ extern int drm_irq_uninstall(struct drm_device *dev);
 extern int drm_vblank_init(struct drm_device *dev, unsigned int num_crtcs);
 extern int drm_wait_vblank(struct drm_device *dev, void *data,
 			   struct drm_file *filp);
-extern u32 drm_vblank_count(struct drm_device *dev, int pipe);
+extern u32 drm_vblank_count(struct drm_device *dev, unsigned int pipe);
 extern u32 drm_crtc_vblank_count(struct drm_crtc *crtc);
 extern u32 drm_vblank_count_and_time(struct drm_device *dev, unsigned int pipe,
 				     struct timeval *vblanktime);
-- 
cgit v0.10.2


From e9f24d5fb7cf3628b195b18ff3ac4e37937ceeae Mon Sep 17 00:00:00 2001
From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Date: Mon, 5 Oct 2015 13:26:36 +0100
Subject: drm/i915: Clean up associated VMAs on context destruction

Prevent leaking VMAs and PPGTT VMs when objects are imported
via flink.

Scenario is that any VMAs created by the importer will be left
dangling after the importer exits, or destroys the PPGTT context
with which they are associated.

This is caused by object destruction not running when the
importer closes the buffer object handle due the reference held
by the exporter. This also leaks the VM since the VMA has a
reference on it.

In practice these leaks can be observed by stopping and starting
the X server on a kernel with fbcon compiled in. Every time
X server exits another VMA will be leaked against the fbcon's
frame buffer object.

Also on systems where flink buffer sharing is used extensively,
like Android, this leak has even more serious consequences.

This version is takes a general approach from the  earlier work
by Rafael Barbalho (drm/i915: Clean-up PPGTT on context
destruction) and tries to incorporate the subsequent discussion
between Chris Wilson and Daniel Vetter.

v2:

Removed immediate cleanup on object retire - it was causing a
recursive VMA unbind via i915_gem_object_wait_rendering. And
it is in fact not even needed since by definition context
cleanup worker runs only after the last context reference has
been dropped, hence all VMAs against the VM belonging to the
context are already on the inactive list.

v3:

Previous version could deadlock since VMA unbind waits on any
rendering on an object to complete. Objects can be busy in a
different VM which would mean that the cleanup loop would do
the wait with the struct mutex held.

This is an even simpler approach where we just unbind VMAs
without waiting since we know all VMAs belonging to this VM
are idle, and there is nothing in flight, at the point
context destructor runs.

v4:

Double underscore prefix for __915_vma_unbind_no_wait and a
commit message typo fix. (Michel Thierry)

Note that this is just a partial/interim fix since we have a bit a
fundamental issue with cleaning up, e.g.

https://bugs.freedesktop.org/show_bug.cgi?id=87729

Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Testcase: igt/gem_ppgtt.c/flink-and-exit-vma-leak
Reviewed-by: Michel Thierry <michel.thierry@intel.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Rafael Barbalho <rafael.barbalho@intel.com>
Cc: Michel Thierry <michel.thierry@intel.com>
[danvet: Add a note that this isn't everything.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 35bf5cb..c93f9a1 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -2839,6 +2839,11 @@ i915_gem_object_ggtt_pin(struct drm_i915_gem_object *obj,
 int i915_vma_bind(struct i915_vma *vma, enum i915_cache_level cache_level,
 		  u32 flags);
 int __must_check i915_vma_unbind(struct i915_vma *vma);
+/*
+ * BEWARE: Do not use the function below unless you can _absolutely_
+ * _guarantee_ VMA in question is _not in use_ anywhere.
+ */
+int __must_check __i915_vma_unbind_no_wait(struct i915_vma *vma);
 int i915_gem_object_put_pages(struct drm_i915_gem_object *obj);
 void i915_gem_release_all_mmaps(struct drm_i915_private *dev_priv);
 void i915_gem_release_mmap(struct drm_i915_gem_object *obj);
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index f0cfbb9..52642af 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -3208,7 +3208,7 @@ static void i915_gem_object_finish_gtt(struct drm_i915_gem_object *obj)
 					    old_write_domain);
 }
 
-int i915_vma_unbind(struct i915_vma *vma)
+static int __i915_vma_unbind(struct i915_vma *vma, bool wait)
 {
 	struct drm_i915_gem_object *obj = vma->obj;
 	struct drm_i915_private *dev_priv = obj->base.dev->dev_private;
@@ -3227,9 +3227,11 @@ int i915_vma_unbind(struct i915_vma *vma)
 
 	BUG_ON(obj->pages == NULL);
 
-	ret = i915_gem_object_wait_rendering(obj, false);
-	if (ret)
-		return ret;
+	if (wait) {
+		ret = i915_gem_object_wait_rendering(obj, false);
+		if (ret)
+			return ret;
+	}
 
 	if (i915_is_ggtt(vma->vm) &&
 	    vma->ggtt_view.type == I915_GGTT_VIEW_NORMAL) {
@@ -3274,6 +3276,16 @@ int i915_vma_unbind(struct i915_vma *vma)
 	return 0;
 }
 
+int i915_vma_unbind(struct i915_vma *vma)
+{
+	return __i915_vma_unbind(vma, true);
+}
+
+int __i915_vma_unbind_no_wait(struct i915_vma *vma)
+{
+	return __i915_vma_unbind(vma, false);
+}
+
 int i915_gpu_idle(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c
index 74aa0c9..680b4c9 100644
--- a/drivers/gpu/drm/i915/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/i915_gem_context.c
@@ -133,6 +133,23 @@ static int get_context_size(struct drm_device *dev)
 	return ret;
 }
 
+static void i915_gem_context_clean(struct intel_context *ctx)
+{
+	struct i915_hw_ppgtt *ppgtt = ctx->ppgtt;
+	struct i915_vma *vma, *next;
+
+	if (WARN_ON_ONCE(!ppgtt))
+		return;
+
+	WARN_ON(!list_empty(&ppgtt->base.active_list));
+
+	list_for_each_entry_safe(vma, next, &ppgtt->base.inactive_list,
+				 mm_list) {
+		if (WARN_ON(__i915_vma_unbind_no_wait(vma)))
+			break;
+	}
+}
+
 void i915_gem_context_free(struct kref *ctx_ref)
 {
 	struct intel_context *ctx = container_of(ctx_ref, typeof(*ctx), ref);
@@ -142,6 +159,13 @@ void i915_gem_context_free(struct kref *ctx_ref)
 	if (i915.enable_execlists)
 		intel_lr_context_free(ctx);
 
+	/*
+	 * This context is going away and we need to remove all VMAs still
+	 * around. This is to handle imported shared objects for which
+	 * destructor did not run when their handles were closed.
+	 */
+	i915_gem_context_clean(ctx);
+
 	i915_ppgtt_put(ctx->ppgtt);
 
 	if (ctx->legacy_hw_ctx.rcs_state)
-- 
cgit v0.10.2


From 24dfd0736c9fc01d096e5760c656032b5a07e962 Mon Sep 17 00:00:00 2001
From: Michel Thierry <michel.thierry@intel.com>
Date: Fri, 2 Oct 2015 14:16:53 +0100
Subject: drm/i915: prevent out of range pt in the PDE macros (take 3)

We tried to fix this in commit fdc454c1484a ("drm/i915: Prevent out of
range pt in gen6_for_each_pde").

But the static analyzer still complains that, just before we break due
to "iter < I915_PDES", we do "pt = (pd)->page_table[iter]" with an
iter value that is bigger than I915_PDES. Of course, this isn't really
a problem since no one uses pt outside the macro. Still, every single
new usage of the macro will create a new issue for us to mark as a
false positive.

Also, Paulo re-started the discussion a while ago [1], but didn't end up
implemented.

In order to "solve" this "problem", this patch takes the ideas from
Chris and Dave, but that check would change the desired behavior of the
code, because the object (for example pdp->page_directory[iter]) can be
null during init/alloc, and C would take this as false, breaking the for
loop immediately.

This has been already verified with "static analysis tools".

[1]http://lists.freedesktop.org/archives/intel-gfx/2015-June/068548.html

v2: Make it a single statement, while preventing the common subexpression
elimination (Chris)

Cc: Paulo Zanoni <paulo.r.zanoni@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Dave Gordon <david.s.gordon@intel.com>
Signed-off-by: Michel Thierry <michel.thierry@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.h b/drivers/gpu/drm/i915/i915_gem_gtt.h
index 9fbb07d..a216397 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.h
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.h
@@ -394,7 +394,8 @@ struct i915_hw_ppgtt {
  */
 #define gen6_for_each_pde(pt, pd, start, length, temp, iter) \
 	for (iter = gen6_pde_index(start); \
-	     pt = (pd)->page_table[iter], length > 0 && iter < I915_PDES; \
+	     length > 0 && iter < I915_PDES ? \
+			(pt = (pd)->page_table[iter]), 1 : 0; \
 	     iter++, \
 	     temp = ALIGN(start+1, 1 << GEN6_PDE_SHIFT) - start, \
 	     temp = min_t(unsigned, temp, length), \
@@ -459,7 +460,8 @@ static inline uint32_t gen6_pde_index(uint32_t addr)
  */
 #define gen8_for_each_pde(pt, pd, start, length, temp, iter)		\
 	for (iter = gen8_pde_index(start); \
-	     pt = (pd)->page_table[iter], length > 0 && iter < I915_PDES;	\
+	     length > 0 && iter < I915_PDES ? \
+			(pt = (pd)->page_table[iter]), 1 : 0; \
 	     iter++,				\
 	     temp = ALIGN(start+1, 1 << GEN8_PDE_SHIFT) - start,	\
 	     temp = min(temp, length),					\
@@ -467,8 +469,8 @@ static inline uint32_t gen6_pde_index(uint32_t addr)
 
 #define gen8_for_each_pdpe(pd, pdp, start, length, temp, iter)	\
 	for (iter = gen8_pdpe_index(start); \
-	     pd = (pdp)->page_directory[iter], \
-	     length > 0 && (iter < I915_PDPES_PER_PDP(dev)); \
+	     length > 0 && (iter < I915_PDPES_PER_PDP(dev)) ? \
+			(pd = (pdp)->page_directory[iter]), 1 : 0; \
 	     iter++,				\
 	     temp = ALIGN(start+1, 1 << GEN8_PDPE_SHIFT) - start,	\
 	     temp = min(temp, length),					\
@@ -476,8 +478,8 @@ static inline uint32_t gen6_pde_index(uint32_t addr)
 
 #define gen8_for_each_pml4e(pdp, pml4, start, length, temp, iter)	\
 	for (iter = gen8_pml4e_index(start);	\
-	     pdp = (pml4)->pdps[iter], \
-	     length > 0 && iter < GEN8_PML4ES_PER_PML4; \
+	     length > 0 && iter < GEN8_PML4ES_PER_PML4 ? \
+			(pdp = (pml4)->pdps[iter]), 1 : 0; \
 	     iter++,				\
 	     temp = ALIGN(start+1, 1ULL << GEN8_PML4E_SHIFT) - start,	\
 	     temp = min(temp, length),					\
-- 
cgit v0.10.2


From 68d6c840595849c0d29f6c52bc75b44ded66b41f Mon Sep 17 00:00:00 2001
From: Chris Wilson <chris@chris-wilson.co.uk>
Date: Thu, 1 Oct 2015 12:34:45 +0100
Subject: drm/i915: Only update the current userptr worker

The userptr worker allows for a slight race condition where upon there
may two or more threads calling get_user_pages for the same object. When
we have the array of pages, then we serialise the update of the object.
However, the worker should only overwrite the obj->userptr.work pointer
if and only if it is the active one. Currently we clear it for a
secondary worker with the effect that we may rarely force a second
lookup.

v2: Rebase and rename a variable to avoid 80cols
v3: Mention v2

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_userptr.c b/drivers/gpu/drm/i915/i915_gem_userptr.c
index d11901d..800a539 100644
--- a/drivers/gpu/drm/i915/i915_gem_userptr.c
+++ b/drivers/gpu/drm/i915/i915_gem_userptr.c
@@ -571,25 +571,25 @@ __i915_gem_userptr_get_pages_worker(struct work_struct *_work)
 	struct get_pages_work *work = container_of(_work, typeof(*work), work);
 	struct drm_i915_gem_object *obj = work->obj;
 	struct drm_device *dev = obj->base.dev;
-	const int num_pages = obj->base.size >> PAGE_SHIFT;
+	const int npages = obj->base.size >> PAGE_SHIFT;
 	struct page **pvec;
 	int pinned, ret;
 
 	ret = -ENOMEM;
 	pinned = 0;
 
-	pvec = kmalloc(num_pages*sizeof(struct page *),
+	pvec = kmalloc(npages*sizeof(struct page *),
 		       GFP_TEMPORARY | __GFP_NOWARN | __GFP_NORETRY);
 	if (pvec == NULL)
-		pvec = drm_malloc_ab(num_pages, sizeof(struct page *));
+		pvec = drm_malloc_ab(npages, sizeof(struct page *));
 	if (pvec != NULL) {
 		struct mm_struct *mm = obj->userptr.mm->mm;
 
 		down_read(&mm->mmap_sem);
-		while (pinned < num_pages) {
+		while (pinned < npages) {
 			ret = get_user_pages(work->task, mm,
 					     obj->userptr.ptr + pinned * PAGE_SIZE,
-					     num_pages - pinned,
+					     npages - pinned,
 					     !obj->userptr.read_only, 0,
 					     pvec + pinned, NULL);
 			if (ret < 0)
@@ -601,20 +601,20 @@ __i915_gem_userptr_get_pages_worker(struct work_struct *_work)
 	}
 
 	mutex_lock(&dev->struct_mutex);
-	if (obj->userptr.work != &work->work) {
-		ret = 0;
-	} else if (pinned == num_pages) {
-		ret = __i915_gem_userptr_set_pages(obj, pvec, num_pages);
-		if (ret == 0) {
-			list_add_tail(&obj->global_list, &to_i915(dev)->mm.unbound_list);
-			obj->get_page.sg = obj->pages->sgl;
-			obj->get_page.last = 0;
-
-			pinned = 0;
+	if (obj->userptr.work == &work->work) {
+		if (pinned == npages) {
+			ret = __i915_gem_userptr_set_pages(obj, pvec, npages);
+			if (ret == 0) {
+				list_add_tail(&obj->global_list,
+					      &to_i915(dev)->mm.unbound_list);
+				obj->get_page.sg = obj->pages->sgl;
+				obj->get_page.last = 0;
+				pinned = 0;
+			}
 		}
+		obj->userptr.work = ERR_PTR(ret);
 	}
 
-	obj->userptr.work = ERR_PTR(ret);
 	obj->userptr.workers--;
 	drm_gem_object_unreference(&obj->base);
 	mutex_unlock(&dev->struct_mutex);
-- 
cgit v0.10.2


From e4b946bfe1e36680e27a5f39163980979fa61a5d Mon Sep 17 00:00:00 2001
From: Chris Wilson <chris@chris-wilson.co.uk>
Date: Thu, 1 Oct 2015 12:34:46 +0100
Subject: drm/i915: Fix userptr deadlock with aliased GTT mmappings
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Michał Winiarski found a really evil way to trigger a struct_mutex
deadlock with userptr. He found that if he allocated a userptr bo and
then GTT mmaped another bo, or even itself, at the same address as the
userptr using MAP_FIXED, he could then cause a deadlock any time we then
had to invalidate the GTT mmappings (so at will). Tvrtko then found by
repeatedly allocating GTT mmappings he could alias with an old userptr
mmap and also trigger the deadlock.

To counter act the deadlock, we make the observation that we only need
to take the struct_mutex if the object has any pages to revoke, and that
before userspace can alias with the userptr address space, it must have
invalidated the userptr->pages. Thus if we can check for those pages
outside of the struct_mutex, we can avoid the deadlock. To do so we
introduce a separate flag for userptr objects that we can inspect from
the mmu-notifier underneath its spinlock.

The patch makes one eye-catching change. That is the removal serial=0
after detecting a to-be-freed object inside the invalidate walker. I
felt setting serial=0 was a questionable pessimisation: it denies us the
chance to reuse the current iterator for the next loop (before it is
freed) and being explicit makes the reader question the validity of the
locking (since the object-free race could occur elsewhere). The
serialisation of the iterator is through the spinlock, if the object is
freed before the next loop then the notifier.serial will be incremented
and we start the walk from the beginning as we detect the invalid cache.

To try and tame the error paths and interactions with the userptr->active
flag, we have to do a fair amount of rearranging of get_pages_userptr().

v2: Grammar fixes
v3: Reorder set-active so that it is only set when obj->pages is set
(and so needs cancellation). Only the order of setting obj->pages and
the active-flag is crucial. Calling gup after invalidate-range begin
means the userptr sees the new set of backing storage (and so will not
need to invalidate its new pages), but we have to be careful not to set
the active-flag prior to successfully establishing obj->pages.
v4: Take the active->flag early so we know in the mmu-notifier when we
have to cancel a pending gup-worker.
v5: Rearrange the error path so that is not so convoluted
v6: Set pinned to 0 when negative before calling release_pages()

Reported-by: Michał Winiarski <michal.winiarski@intel.com>
Testcase: igt/gem_userptr_blits/map-fixed*
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Michał Winiarski <michal.winiarski@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: stable@vger.kernel.org
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_userptr.c b/drivers/gpu/drm/i915/i915_gem_userptr.c
index 800a539..161f7fb 100644
--- a/drivers/gpu/drm/i915/i915_gem_userptr.c
+++ b/drivers/gpu/drm/i915/i915_gem_userptr.c
@@ -59,6 +59,7 @@ struct i915_mmu_object {
 	struct interval_tree_node it;
 	struct list_head link;
 	struct drm_i915_gem_object *obj;
+	bool active;
 	bool is_linear;
 };
 
@@ -114,7 +115,8 @@ restart:
 
 		obj = mo->obj;
 
-		if (!kref_get_unless_zero(&obj->base.refcount))
+		if (!mo->active ||
+		    !kref_get_unless_zero(&obj->base.refcount))
 			continue;
 
 		spin_unlock(&mn->lock);
@@ -151,7 +153,8 @@ static void i915_gem_userptr_mn_invalidate_range_start(struct mmu_notifier *_mn,
 		else
 			it = interval_tree_iter_first(&mn->objects, start, end);
 		if (it != NULL) {
-			obj = container_of(it, struct i915_mmu_object, it)->obj;
+			struct i915_mmu_object *mo =
+				container_of(it, struct i915_mmu_object, it);
 
 			/* The mmu_object is released late when destroying the
 			 * GEM object so it is entirely possible to gain a
@@ -160,11 +163,9 @@ static void i915_gem_userptr_mn_invalidate_range_start(struct mmu_notifier *_mn,
 			 * the struct_mutex - and consequently use it after it
 			 * is freed and then double free it.
 			 */
-			if (!kref_get_unless_zero(&obj->base.refcount)) {
-				spin_unlock(&mn->lock);
-				serial = 0;
-				continue;
-			}
+			if (mo->active &&
+			    kref_get_unless_zero(&mo->obj->base.refcount))
+				obj = mo->obj;
 
 			serial = mn->serial;
 		}
@@ -566,6 +567,30 @@ __i915_gem_userptr_set_pages(struct drm_i915_gem_object *obj,
 }
 
 static void
+__i915_gem_userptr_set_active(struct drm_i915_gem_object *obj,
+			      bool value)
+{
+	/* During mm_invalidate_range we need to cancel any userptr that
+	 * overlaps the range being invalidated. Doing so requires the
+	 * struct_mutex, and that risks recursion. In order to cause
+	 * recursion, the user must alias the userptr address space with
+	 * a GTT mmapping (possible with a MAP_FIXED) - then when we have
+	 * to invalidate that mmaping, mm_invalidate_range is called with
+	 * the userptr address *and* the struct_mutex held.  To prevent that
+	 * we set a flag under the i915_mmu_notifier spinlock to indicate
+	 * whether this object is valid.
+	 */
+#if defined(CONFIG_MMU_NOTIFIER)
+	if (obj->userptr.mmu_object == NULL)
+		return;
+
+	spin_lock(&obj->userptr.mmu_object->mn->lock);
+	obj->userptr.mmu_object->active = value;
+	spin_unlock(&obj->userptr.mmu_object->mn->lock);
+#endif
+}
+
+static void
 __i915_gem_userptr_get_pages_worker(struct work_struct *_work)
 {
 	struct get_pages_work *work = container_of(_work, typeof(*work), work);
@@ -613,6 +638,8 @@ __i915_gem_userptr_get_pages_worker(struct work_struct *_work)
 			}
 		}
 		obj->userptr.work = ERR_PTR(ret);
+		if (ret)
+			__i915_gem_userptr_set_active(obj, false);
 	}
 
 	obj->userptr.workers--;
@@ -627,11 +654,60 @@ __i915_gem_userptr_get_pages_worker(struct work_struct *_work)
 }
 
 static int
+__i915_gem_userptr_get_pages_schedule(struct drm_i915_gem_object *obj,
+				      bool *active)
+{
+	struct get_pages_work *work;
+
+	/* Spawn a worker so that we can acquire the
+	 * user pages without holding our mutex. Access
+	 * to the user pages requires mmap_sem, and we have
+	 * a strict lock ordering of mmap_sem, struct_mutex -
+	 * we already hold struct_mutex here and so cannot
+	 * call gup without encountering a lock inversion.
+	 *
+	 * Userspace will keep on repeating the operation
+	 * (thanks to EAGAIN) until either we hit the fast
+	 * path or the worker completes. If the worker is
+	 * cancelled or superseded, the task is still run
+	 * but the results ignored. (This leads to
+	 * complications that we may have a stray object
+	 * refcount that we need to be wary of when
+	 * checking for existing objects during creation.)
+	 * If the worker encounters an error, it reports
+	 * that error back to this function through
+	 * obj->userptr.work = ERR_PTR.
+	 */
+	if (obj->userptr.workers >= I915_GEM_USERPTR_MAX_WORKERS)
+		return -EAGAIN;
+
+	work = kmalloc(sizeof(*work), GFP_KERNEL);
+	if (work == NULL)
+		return -ENOMEM;
+
+	obj->userptr.work = &work->work;
+	obj->userptr.workers++;
+
+	work->obj = obj;
+	drm_gem_object_reference(&obj->base);
+
+	work->task = current;
+	get_task_struct(work->task);
+
+	INIT_WORK(&work->work, __i915_gem_userptr_get_pages_worker);
+	schedule_work(&work->work);
+
+	*active = true;
+	return -EAGAIN;
+}
+
+static int
 i915_gem_userptr_get_pages(struct drm_i915_gem_object *obj)
 {
 	const int num_pages = obj->base.size >> PAGE_SHIFT;
 	struct page **pvec;
 	int pinned, ret;
+	bool active;
 
 	/* If userspace should engineer that these pages are replaced in
 	 * the vma between us binding this page into the GTT and completion
@@ -649,6 +725,18 @@ i915_gem_userptr_get_pages(struct drm_i915_gem_object *obj)
 	 * to the vma (discard or cloning) which should prevent the more
 	 * egregious cases from causing harm.
 	 */
+	if (IS_ERR(obj->userptr.work)) {
+		/* active flag will have been dropped already by the worker */
+		ret = PTR_ERR(obj->userptr.work);
+		obj->userptr.work = NULL;
+		return ret;
+	}
+	if (obj->userptr.work)
+		/* active flag should still be held for the pending work */
+		return -EAGAIN;
+
+	/* Let the mmu-notifier know that we have begun and need cancellation */
+	__i915_gem_userptr_set_active(obj, true);
 
 	pvec = NULL;
 	pinned = 0;
@@ -657,73 +745,27 @@ i915_gem_userptr_get_pages(struct drm_i915_gem_object *obj)
 			       GFP_TEMPORARY | __GFP_NOWARN | __GFP_NORETRY);
 		if (pvec == NULL) {
 			pvec = drm_malloc_ab(num_pages, sizeof(struct page *));
-			if (pvec == NULL)
+			if (pvec == NULL) {
+				__i915_gem_userptr_set_active(obj, false);
 				return -ENOMEM;
+			}
 		}
 
 		pinned = __get_user_pages_fast(obj->userptr.ptr, num_pages,
 					       !obj->userptr.read_only, pvec);
 	}
-	if (pinned < num_pages) {
-		if (pinned < 0) {
-			ret = pinned;
-			pinned = 0;
-		} else {
-			/* Spawn a worker so that we can acquire the
-			 * user pages without holding our mutex. Access
-			 * to the user pages requires mmap_sem, and we have
-			 * a strict lock ordering of mmap_sem, struct_mutex -
-			 * we already hold struct_mutex here and so cannot
-			 * call gup without encountering a lock inversion.
-			 *
-			 * Userspace will keep on repeating the operation
-			 * (thanks to EAGAIN) until either we hit the fast
-			 * path or the worker completes. If the worker is
-			 * cancelled or superseded, the task is still run
-			 * but the results ignored. (This leads to
-			 * complications that we may have a stray object
-			 * refcount that we need to be wary of when
-			 * checking for existing objects during creation.)
-			 * If the worker encounters an error, it reports
-			 * that error back to this function through
-			 * obj->userptr.work = ERR_PTR.
-			 */
-			ret = -EAGAIN;
-			if (obj->userptr.work == NULL &&
-			    obj->userptr.workers < I915_GEM_USERPTR_MAX_WORKERS) {
-				struct get_pages_work *work;
-
-				work = kmalloc(sizeof(*work), GFP_KERNEL);
-				if (work != NULL) {
-					obj->userptr.work = &work->work;
-					obj->userptr.workers++;
-
-					work->obj = obj;
-					drm_gem_object_reference(&obj->base);
-
-					work->task = current;
-					get_task_struct(work->task);
-
-					INIT_WORK(&work->work, __i915_gem_userptr_get_pages_worker);
-					schedule_work(&work->work);
-				} else
-					ret = -ENOMEM;
-			} else {
-				if (IS_ERR(obj->userptr.work)) {
-					ret = PTR_ERR(obj->userptr.work);
-					obj->userptr.work = NULL;
-				}
-			}
-		}
-	} else {
+
+	active = false;
+	if (pinned < 0)
+		ret = pinned, pinned = 0;
+	else if (pinned < num_pages)
+		ret = __i915_gem_userptr_get_pages_schedule(obj, &active);
+	else
 		ret = __i915_gem_userptr_set_pages(obj, pvec, num_pages);
-		if (ret == 0) {
-			obj->userptr.work = NULL;
-			pinned = 0;
-		}
+	if (ret) {
+		__i915_gem_userptr_set_active(obj, active);
+		release_pages(pvec, pinned, 0);
 	}
-
-	release_pages(pvec, pinned, 0);
 	drm_free_large(pvec);
 	return ret;
 }
@@ -734,6 +776,7 @@ i915_gem_userptr_put_pages(struct drm_i915_gem_object *obj)
 	struct sg_page_iter sg_iter;
 
 	BUG_ON(obj->userptr.work != NULL);
+	__i915_gem_userptr_set_active(obj, false);
 
 	if (obj->madv != I915_MADV_WILLNEED)
 		obj->dirty = 0;
-- 
cgit v0.10.2


From 380996aab50eea1d66269e1633cd2f971f06da6d Mon Sep 17 00:00:00 2001
From: Chris Wilson <chris@chris-wilson.co.uk>
Date: Thu, 1 Oct 2015 12:34:47 +0100
Subject: drm/i915: Use a task to cancel the userptr on invalidate_range
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Whilst discussing possible ways to trigger an invalidate_range on a
userptr with an aliased GGTT mmapping (and so cause a struct_mutex
deadlock), the conclusion is that we can, and we must, prevent any
possible deadlock by avoiding taking the mutex at all during
invalidate_range. This has numerous advantages all of which stem from
avoid the sleeping function from inside the unknown context. In
particular, it simplifies the invalidate_range because we no longer
have to juggle the spinlock/mutex and can just hold the spinlock
for the entire walk. To compensate, we have to make get_pages a bit more
complicated in order to serialise with a pending cancel_userptr worker.
As we hold the struct_mutex, we have no choice but to return EAGAIN and
hope that the worker is then flushed before we retry after reacquiring
the struct_mutex.

The important caveat is that the invalidate_range itself is no longer
synchronous. There exists a small but definite period in time in which
the old PTE's page remain accessible via the GPU. Note however that the
physical pages themselves are not invalidated by the mmu_notifier, just
the CPU view of the address space. The impact should be limited to a
delay in pages being flushed, rather than a possibility of writing to
the wrong pages. The only race condition that this worsens is remapping
an userptr active on the GPU where fresh work may still reference the
old pages due to struct_mutex contention. Given that userspace is racing
with the GPU, it is fair to say that the results are undefined.

v2: Only queue (and importantly only take one refcnt) the worker once.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Michał Winiarski <michal.winiarski@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_userptr.c b/drivers/gpu/drm/i915/i915_gem_userptr.c
index 161f7fb..1b3b451 100644
--- a/drivers/gpu/drm/i915/i915_gem_userptr.c
+++ b/drivers/gpu/drm/i915/i915_gem_userptr.c
@@ -50,7 +50,6 @@ struct i915_mmu_notifier {
 	struct mmu_notifier mn;
 	struct rb_root objects;
 	struct list_head linear;
-	unsigned long serial;
 	bool has_linear;
 };
 
@@ -59,14 +58,16 @@ struct i915_mmu_object {
 	struct interval_tree_node it;
 	struct list_head link;
 	struct drm_i915_gem_object *obj;
+	struct work_struct work;
 	bool active;
 	bool is_linear;
 };
 
-static unsigned long cancel_userptr(struct drm_i915_gem_object *obj)
+static void __cancel_userptr__worker(struct work_struct *work)
 {
+	struct i915_mmu_object *mo = container_of(work, typeof(*mo), work);
+	struct drm_i915_gem_object *obj = mo->obj;
 	struct drm_device *dev = obj->base.dev;
-	unsigned long end;
 
 	mutex_lock(&dev->struct_mutex);
 	/* Cancel any active worker and force us to re-evaluate gup */
@@ -89,46 +90,28 @@ static unsigned long cancel_userptr(struct drm_i915_gem_object *obj)
 		dev_priv->mm.interruptible = was_interruptible;
 	}
 
-	end = obj->userptr.ptr + obj->base.size;
-
 	drm_gem_object_unreference(&obj->base);
 	mutex_unlock(&dev->struct_mutex);
-
-	return end;
 }
 
-static void *invalidate_range__linear(struct i915_mmu_notifier *mn,
-				      struct mm_struct *mm,
-				      unsigned long start,
-				      unsigned long end)
+static unsigned long cancel_userptr(struct i915_mmu_object *mo)
 {
-	struct i915_mmu_object *mo;
-	unsigned long serial;
-
-restart:
-	serial = mn->serial;
-	list_for_each_entry(mo, &mn->linear, link) {
-		struct drm_i915_gem_object *obj;
-
-		if (mo->it.last < start || mo->it.start > end)
-			continue;
-
-		obj = mo->obj;
-
-		if (!mo->active ||
-		    !kref_get_unless_zero(&obj->base.refcount))
-			continue;
-
-		spin_unlock(&mn->lock);
-
-		cancel_userptr(obj);
-
-		spin_lock(&mn->lock);
-		if (serial != mn->serial)
-			goto restart;
+	unsigned long end = mo->obj->userptr.ptr + mo->obj->base.size;
+
+	/* The mmu_object is released late when destroying the
+	 * GEM object so it is entirely possible to gain a
+	 * reference on an object in the process of being freed
+	 * since our serialisation is via the spinlock and not
+	 * the struct_mutex - and consequently use it after it
+	 * is freed and then double free it.
+	 */
+	if (mo->active && kref_get_unless_zero(&mo->obj->base.refcount)) {
+		schedule_work(&mo->work);
+		/* only schedule one work packet to avoid the refleak */
+		mo->active = false;
 	}
 
-	return NULL;
+	return end;
 }
 
 static void i915_gem_userptr_mn_invalidate_range_start(struct mmu_notifier *_mn,
@@ -136,45 +119,32 @@ static void i915_gem_userptr_mn_invalidate_range_start(struct mmu_notifier *_mn,
 						       unsigned long start,
 						       unsigned long end)
 {
-	struct i915_mmu_notifier *mn = container_of(_mn, struct i915_mmu_notifier, mn);
-	struct interval_tree_node *it = NULL;
-	unsigned long next = start;
-	unsigned long serial = 0;
-
-	end--; /* interval ranges are inclusive, but invalidate range is exclusive */
-	while (next < end) {
-		struct drm_i915_gem_object *obj = NULL;
-
-		spin_lock(&mn->lock);
-		if (mn->has_linear)
-			it = invalidate_range__linear(mn, mm, start, end);
-		else if (serial == mn->serial)
-			it = interval_tree_iter_next(it, next, end);
-		else
-			it = interval_tree_iter_first(&mn->objects, start, end);
-		if (it != NULL) {
-			struct i915_mmu_object *mo =
-				container_of(it, struct i915_mmu_object, it);
-
-			/* The mmu_object is released late when destroying the
-			 * GEM object so it is entirely possible to gain a
-			 * reference on an object in the process of being freed
-			 * since our serialisation is via the spinlock and not
-			 * the struct_mutex - and consequently use it after it
-			 * is freed and then double free it.
-			 */
-			if (mo->active &&
-			    kref_get_unless_zero(&mo->obj->base.refcount))
-				obj = mo->obj;
-
-			serial = mn->serial;
+	struct i915_mmu_notifier *mn =
+		container_of(_mn, struct i915_mmu_notifier, mn);
+	struct i915_mmu_object *mo;
+
+	/* interval ranges are inclusive, but invalidate range is exclusive */
+	end--;
+
+	spin_lock(&mn->lock);
+	if (mn->has_linear) {
+		list_for_each_entry(mo, &mn->linear, link) {
+			if (mo->it.last < start || mo->it.start > end)
+				continue;
+
+			cancel_userptr(mo);
 		}
-		spin_unlock(&mn->lock);
-		if (obj == NULL)
-			return;
+	} else {
+		struct interval_tree_node *it;
 
-		next = cancel_userptr(obj);
+		it = interval_tree_iter_first(&mn->objects, start, end);
+		while (it) {
+			mo = container_of(it, struct i915_mmu_object, it);
+			start = cancel_userptr(mo);
+			it = interval_tree_iter_next(it, start, end);
+		}
 	}
+	spin_unlock(&mn->lock);
 }
 
 static const struct mmu_notifier_ops i915_gem_userptr_notifier = {
@@ -194,7 +164,6 @@ i915_mmu_notifier_create(struct mm_struct *mm)
 	spin_lock_init(&mn->lock);
 	mn->mn.ops = &i915_gem_userptr_notifier;
 	mn->objects = RB_ROOT;
-	mn->serial = 1;
 	INIT_LIST_HEAD(&mn->linear);
 	mn->has_linear = false;
 
@@ -208,12 +177,6 @@ i915_mmu_notifier_create(struct mm_struct *mm)
 	return mn;
 }
 
-static void __i915_mmu_notifier_update_serial(struct i915_mmu_notifier *mn)
-{
-	if (++mn->serial == 0)
-		mn->serial = 1;
-}
-
 static int
 i915_mmu_notifier_add(struct drm_device *dev,
 		      struct i915_mmu_notifier *mn,
@@ -260,10 +223,9 @@ i915_mmu_notifier_add(struct drm_device *dev,
 	} else
 		interval_tree_insert(&mo->it, &mn->objects);
 
-	if (ret == 0) {
+	if (ret == 0)
 		list_add(&mo->link, &mn->linear);
-		__i915_mmu_notifier_update_serial(mn);
-	}
+
 	spin_unlock(&mn->lock);
 	mutex_unlock(&dev->struct_mutex);
 
@@ -291,7 +253,6 @@ i915_mmu_notifier_del(struct i915_mmu_notifier *mn,
 		mn->has_linear = i915_mmu_notifier_has_linear(mn);
 	else
 		interval_tree_remove(&mo->it, &mn->objects);
-	__i915_mmu_notifier_update_serial(mn);
 	spin_unlock(&mn->lock);
 }
 
@@ -358,6 +319,7 @@ i915_gem_userptr_init__mmu_notifier(struct drm_i915_gem_object *obj,
 	mo->it.start = obj->userptr.ptr;
 	mo->it.last = mo->it.start + obj->base.size - 1;
 	mo->obj = obj;
+	INIT_WORK(&mo->work, __cancel_userptr__worker);
 
 	ret = i915_mmu_notifier_add(obj->base.dev, mn, mo);
 	if (ret) {
@@ -566,10 +528,12 @@ __i915_gem_userptr_set_pages(struct drm_i915_gem_object *obj,
 	return ret;
 }
 
-static void
+static int
 __i915_gem_userptr_set_active(struct drm_i915_gem_object *obj,
 			      bool value)
 {
+	int ret = 0;
+
 	/* During mm_invalidate_range we need to cancel any userptr that
 	 * overlaps the range being invalidated. Doing so requires the
 	 * struct_mutex, and that risks recursion. In order to cause
@@ -582,12 +546,20 @@ __i915_gem_userptr_set_active(struct drm_i915_gem_object *obj,
 	 */
 #if defined(CONFIG_MMU_NOTIFIER)
 	if (obj->userptr.mmu_object == NULL)
-		return;
+		return 0;
 
 	spin_lock(&obj->userptr.mmu_object->mn->lock);
-	obj->userptr.mmu_object->active = value;
+	/* In order to serialise get_pages with an outstanding
+	 * cancel_userptr, we must drop the struct_mutex and try again.
+	 */
+	if (!value || !work_pending(&obj->userptr.mmu_object->work))
+		obj->userptr.mmu_object->active = value;
+	else
+		ret = -EAGAIN;
 	spin_unlock(&obj->userptr.mmu_object->mn->lock);
 #endif
+
+	return ret;
 }
 
 static void
@@ -736,7 +708,9 @@ i915_gem_userptr_get_pages(struct drm_i915_gem_object *obj)
 		return -EAGAIN;
 
 	/* Let the mmu-notifier know that we have begun and need cancellation */
-	__i915_gem_userptr_set_active(obj, true);
+	ret = __i915_gem_userptr_set_active(obj, true);
+	if (ret)
+		return ret;
 
 	pvec = NULL;
 	pinned = 0;
-- 
cgit v0.10.2


From f94ab604db9db1a8a0cf0827f5573bfd2d2b89aa Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Tue, 21 Jul 2015 10:00:15 +0100
Subject: gpu: imx: simplify sync polarity setting

Use a function to convert the sync pin to a bit mask for the DI_GENERAL
register, and move this out of the interlace/non-interlace path to the
common path.

Tested-by: Philipp Zabel <p.zabel@pengutronix.de>
Reviewed-by: Fabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/ipu-v3/ipu-di.c b/drivers/gpu/ipu-v3/ipu-di.c
index 2970c6b..a96991c 100644
--- a/drivers/gpu/ipu-v3/ipu-di.c
+++ b/drivers/gpu/ipu-v3/ipu-di.c
@@ -543,6 +543,29 @@ int ipu_di_adjust_videomode(struct ipu_di *di, struct videomode *mode)
 }
 EXPORT_SYMBOL_GPL(ipu_di_adjust_videomode);
 
+static u32 ipu_di_gen_polarity(int pin)
+{
+	switch (pin) {
+	case 1:
+		return DI_GEN_POLARITY_1;
+	case 2:
+		return DI_GEN_POLARITY_2;
+	case 3:
+		return DI_GEN_POLARITY_3;
+	case 4:
+		return DI_GEN_POLARITY_4;
+	case 5:
+		return DI_GEN_POLARITY_5;
+	case 6:
+		return DI_GEN_POLARITY_6;
+	case 7:
+		return DI_GEN_POLARITY_7;
+	case 8:
+		return DI_GEN_POLARITY_8;
+	}
+	return 0;
+}
+
 int ipu_di_init_sync_panel(struct ipu_di *di, struct ipu_di_signal_cfg *sig)
 {
 	u32 reg;
@@ -586,11 +609,6 @@ int ipu_di_init_sync_panel(struct ipu_di *di, struct ipu_di_signal_cfg *sig)
 		di_gen |= DI_GEN_POLARITY_8;
 
 		vsync_cnt = 7;
-
-		if (sig->mode.flags & DISPLAY_FLAGS_HSYNC_HIGH)
-			di_gen |= DI_GEN_POLARITY_3;
-		if (sig->mode.flags & DISPLAY_FLAGS_VSYNC_HIGH)
-			di_gen |= DI_GEN_POLARITY_2;
 	} else {
 		ipu_di_sync_config_noninterlaced(di, sig, div);
 
@@ -602,25 +620,13 @@ int ipu_di_init_sync_panel(struct ipu_di *di, struct ipu_di_signal_cfg *sig)
 			 */
 			if (!(sig->hsync_pin == 2 && sig->vsync_pin == 3))
 				vsync_cnt = 6;
-
-		if (sig->mode.flags & DISPLAY_FLAGS_HSYNC_HIGH) {
-			if (sig->hsync_pin == 2)
-				di_gen |= DI_GEN_POLARITY_2;
-			else if (sig->hsync_pin == 4)
-				di_gen |= DI_GEN_POLARITY_4;
-			else if (sig->hsync_pin == 7)
-				di_gen |= DI_GEN_POLARITY_7;
-		}
-		if (sig->mode.flags & DISPLAY_FLAGS_VSYNC_HIGH) {
-			if (sig->vsync_pin == 3)
-				di_gen |= DI_GEN_POLARITY_3;
-			else if (sig->vsync_pin == 6)
-				di_gen |= DI_GEN_POLARITY_6;
-			else if (sig->vsync_pin == 8)
-				di_gen |= DI_GEN_POLARITY_8;
-		}
 	}
 
+	if (sig->mode.flags & DISPLAY_FLAGS_HSYNC_HIGH)
+		di_gen |= ipu_di_gen_polarity(sig->hsync_pin);
+	if (sig->mode.flags & DISPLAY_FLAGS_VSYNC_HIGH)
+		di_gen |= ipu_di_gen_polarity(sig->vsync_pin);
+
 	if (sig->clk_pol)
 		di_gen |= DI_GEN_POLARITY_DISP_CLK;
 
-- 
cgit v0.10.2


From aefa627fa3f231c5d86aeb229f15e8478cb6e910 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Tue, 21 Jul 2015 10:22:29 +0100
Subject: gpu: imx: fix support for interlaced modes

The support for interlaced video modes seems to be broken; we don't use
anything other than the vtotal/htotal from the timing information to
define the various sync counters.

Freescale patches for interlaced video support contain an alternative
sync counter setup, which we include here.  This setup produces the
hsync and vsync via the normal counter 2 and 3, but moves the display
enable signal from counter 5 to counter 6.  Therefore, we need to
change the display controller setup as well.

The corresponding Freescale patches for this change are:
  iMX6-HDMI-support-interlaced-display-mode.patch
  IPU-fine-tuning-the-interlace-display-timing-for-CEA.patch

This produces a working interlace format output from the IPU.

Tested-by: Philipp Zabel <p.zabel@pengutronix.de>
Reviewed-by: Fabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/ipu-v3/ipu-dc.c b/drivers/gpu/ipu-v3/ipu-dc.c
index 9ef2e1f..d3ad534 100644
--- a/drivers/gpu/ipu-v3/ipu-dc.c
+++ b/drivers/gpu/ipu-v3/ipu-dc.c
@@ -183,12 +183,19 @@ int ipu_dc_init_sync(struct ipu_dc *dc, struct ipu_di *di, bool interlaced,
 	}
 
 	if (interlaced) {
-		dc_link_event(dc, DC_EVT_NL, 0, 3);
-		dc_link_event(dc, DC_EVT_EOL, 0, 2);
-		dc_link_event(dc, DC_EVT_NEW_DATA, 0, 1);
+		int addr;
+
+		if (dc->di)
+			addr = 1;
+		else
+			addr = 0;
+
+		dc_link_event(dc, DC_EVT_NL, addr, 3);
+		dc_link_event(dc, DC_EVT_EOL, addr, 2);
+		dc_link_event(dc, DC_EVT_NEW_DATA, addr, 1);
 
 		/* Init template microcode */
-		dc_write_tmpl(dc, 0, WROD(0), 0, map, SYNC_WAVE, 0, 8, 1);
+		dc_write_tmpl(dc, addr, WROD(0), 0, map, SYNC_WAVE, 0, 6, 1);
 	} else {
 		if (dc->di) {
 			dc_link_event(dc, DC_EVT_NL, 2, 3);
diff --git a/drivers/gpu/ipu-v3/ipu-di.c b/drivers/gpu/ipu-v3/ipu-di.c
index a96991c..359268e 100644
--- a/drivers/gpu/ipu-v3/ipu-di.c
+++ b/drivers/gpu/ipu-v3/ipu-di.c
@@ -71,6 +71,10 @@ enum di_sync_wave {
 	DI_SYNC_HSYNC = 3,
 	DI_SYNC_VSYNC = 4,
 	DI_SYNC_DE = 6,
+
+	DI_SYNC_CNT1 = 2,	/* counter >= 2 only */
+	DI_SYNC_CNT4 = 5,	/* counter >= 5 only */
+	DI_SYNC_CNT5 = 6,	/* counter >= 6 only */
 };
 
 #define SYNC_WAVE 0
@@ -211,66 +215,59 @@ static void ipu_di_sync_config_interlaced(struct ipu_di *di,
 		sig->mode.hback_porch + sig->mode.hfront_porch;
 	u32 v_total = sig->mode.vactive + sig->mode.vsync_len +
 		sig->mode.vback_porch + sig->mode.vfront_porch;
-	u32 reg;
 	struct di_sync_config cfg[] = {
 		{
-			.run_count = h_total / 2 - 1,
-			.run_src = DI_SYNC_CLK,
+			/* 1: internal VSYNC for each frame */
+			.run_count = v_total * 2 - 1,
+			.run_src = 3,			/* == counter 7 */
 		}, {
-			.run_count = h_total - 11,
+			/* PIN2: HSYNC waveform */
+			.run_count = h_total - 1,
 			.run_src = DI_SYNC_CLK,
-			.cnt_down = 4,
+			.cnt_polarity_gen_en = 1,
+			.cnt_polarity_trigger_src = DI_SYNC_CLK,
+			.cnt_down = sig->mode.hsync_len * 2,
 		}, {
-			.run_count = v_total * 2 - 1,
-			.run_src = DI_SYNC_INT_HSYNC,
-			.offset_count = 1,
-			.offset_src = DI_SYNC_INT_HSYNC,
-			.cnt_down = 4,
+			/* PIN3: VSYNC waveform */
+			.run_count = v_total - 1,
+			.run_src = 4,			/* == counter 7 */
+			.cnt_polarity_gen_en = 1,
+			.cnt_polarity_trigger_src = 4,	/* == counter 7 */
+			.cnt_down = sig->mode.vsync_len * 2,
+			.cnt_clr_src = DI_SYNC_CNT1,
 		}, {
-			.run_count = v_total / 2 - 1,
+			/* 4: Field */
+			.run_count = v_total / 2,
 			.run_src = DI_SYNC_HSYNC,
-			.offset_count = sig->mode.vback_porch,
-			.offset_src = DI_SYNC_HSYNC,
+			.offset_count = h_total / 2,
+			.offset_src = DI_SYNC_CLK,
 			.repeat_count = 2,
-			.cnt_clr_src = DI_SYNC_VSYNC,
+			.cnt_clr_src = DI_SYNC_CNT1,
 		}, {
+			/* 5: Active lines */
 			.run_src = DI_SYNC_HSYNC,
-			.repeat_count = sig->mode.vactive / 2,
-			.cnt_clr_src = 4,
-		}, {
-			.run_count = v_total - 1,
-			.run_src = DI_SYNC_HSYNC,
-		}, {
-			.run_count = v_total / 2 - 1,
-			.run_src = DI_SYNC_HSYNC,
-			.offset_count = 9,
+			.offset_count = (sig->mode.vsync_len +
+					 sig->mode.vback_porch) / 2,
 			.offset_src = DI_SYNC_HSYNC,
-			.repeat_count = 2,
-			.cnt_clr_src = DI_SYNC_VSYNC,
+			.repeat_count = sig->mode.vactive / 2,
+			.cnt_clr_src = DI_SYNC_CNT4,
 		}, {
+			/* 6: Active pixel, referenced by DC */
 			.run_src = DI_SYNC_CLK,
-			.offset_count = sig->mode.hback_porch,
+			.offset_count = sig->mode.hsync_len +
+					sig->mode.hback_porch,
 			.offset_src = DI_SYNC_CLK,
 			.repeat_count = sig->mode.hactive,
-			.cnt_clr_src = 5,
+			.cnt_clr_src = DI_SYNC_CNT5,
 		}, {
-			.run_count = v_total - 1,
-			.run_src = DI_SYNC_INT_HSYNC,
-			.offset_count = v_total / 2,
-			.offset_src = DI_SYNC_INT_HSYNC,
-			.cnt_clr_src = DI_SYNC_HSYNC,
-			.cnt_down = 4,
+			/* 7: Half line HSYNC */
+			.run_count = h_total / 2 - 1,
+			.run_src = DI_SYNC_CLK,
 		}
 	};
 
 	ipu_di_sync_config(di, cfg, 0, ARRAY_SIZE(cfg));
 
-	/* set gentime select and tag sel */
-	reg = ipu_di_read(di, DI_SW_GEN1(9));
-	reg &= 0x1FFFFFFF;
-	reg |= (3 - 1) << 29 | 0x00008000;
-	ipu_di_write(di, reg, DI_SW_GEN1(9));
-
 	ipu_di_write(di, v_total / 2 - 1, DI_SCR_CONF);
 }
 
@@ -605,10 +602,8 @@ int ipu_di_init_sync_panel(struct ipu_di *di, struct ipu_di_signal_cfg *sig)
 
 		/* set y_sel = 1 */
 		di_gen |= 0x10000000;
-		di_gen |= DI_GEN_POLARITY_5;
-		di_gen |= DI_GEN_POLARITY_8;
 
-		vsync_cnt = 7;
+		vsync_cnt = 3;
 	} else {
 		ipu_di_sync_config_noninterlaced(di, sig, div);
 
-- 
cgit v0.10.2


From e80b9f4eea2649f67edc05d4f2e7ddc250f91453 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Tue, 21 Jul 2015 11:08:25 +0100
Subject: drm: bridge/dw_hdmi: add support for interlaced video modes

Add support for interlaced video modes to the dw_hdmi bridge.  This
mainly involves halving the vertical parameters to be programmed into
the bridge registers, and setting the interlace_allowed connector flag.

This brings working 1080i support.  However, 480i and 576i fail to
work due to the lack of proper pixel repetition support, which is not
trivial to add due to the tabular PLL parameterisation.  Hence, we
filter out these modes in our mode_valid() method.

Tested-by: Philipp Zabel <p.zabel@pengutronix.de>
Reviewed-by: Fabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/bridge/dw_hdmi.c b/drivers/gpu/drm/bridge/dw_hdmi.c
index 0083d4e..dc1a382 100644
--- a/drivers/gpu/drm/bridge/dw_hdmi.c
+++ b/drivers/gpu/drm/bridge/dw_hdmi.c
@@ -1063,6 +1063,7 @@ static void hdmi_av_composer(struct dw_hdmi *hdmi,
 	u8 inv_val;
 	struct hdmi_vmode *vmode = &hdmi->hdmi_data.video_mode;
 	int hblank, vblank, h_de_hs, v_de_vs, hsync_len, vsync_len;
+	unsigned int vdisplay;
 
 	vmode->mpixelclock = mode->clock * 1000;
 
@@ -1102,13 +1103,29 @@ static void hdmi_av_composer(struct dw_hdmi *hdmi,
 
 	hdmi_writeb(hdmi, inv_val, HDMI_FC_INVIDCONF);
 
+	vdisplay = mode->vdisplay;
+	vblank = mode->vtotal - mode->vdisplay;
+	v_de_vs = mode->vsync_start - mode->vdisplay;
+	vsync_len = mode->vsync_end - mode->vsync_start;
+
+	/*
+	 * When we're setting an interlaced mode, we need
+	 * to adjust the vertical timing to suit.
+	 */
+	if (mode->flags & DRM_MODE_FLAG_INTERLACE) {
+		vdisplay /= 2;
+		vblank /= 2;
+		v_de_vs /= 2;
+		vsync_len /= 2;
+	}
+
 	/* Set up horizontal active pixel width */
 	hdmi_writeb(hdmi, mode->hdisplay >> 8, HDMI_FC_INHACTV1);
 	hdmi_writeb(hdmi, mode->hdisplay, HDMI_FC_INHACTV0);
 
 	/* Set up vertical active lines */
-	hdmi_writeb(hdmi, mode->vdisplay >> 8, HDMI_FC_INVACTV1);
-	hdmi_writeb(hdmi, mode->vdisplay, HDMI_FC_INVACTV0);
+	hdmi_writeb(hdmi, vdisplay >> 8, HDMI_FC_INVACTV1);
+	hdmi_writeb(hdmi, vdisplay, HDMI_FC_INVACTV0);
 
 	/* Set up horizontal blanking pixel region width */
 	hblank = mode->htotal - mode->hdisplay;
@@ -1116,7 +1133,6 @@ static void hdmi_av_composer(struct dw_hdmi *hdmi,
 	hdmi_writeb(hdmi, hblank, HDMI_FC_INHBLANK0);
 
 	/* Set up vertical blanking pixel region width */
-	vblank = mode->vtotal - mode->vdisplay;
 	hdmi_writeb(hdmi, vblank, HDMI_FC_INVBLANK);
 
 	/* Set up HSYNC active edge delay width (in pixel clks) */
@@ -1125,7 +1141,6 @@ static void hdmi_av_composer(struct dw_hdmi *hdmi,
 	hdmi_writeb(hdmi, h_de_hs, HDMI_FC_HSYNCINDELAY0);
 
 	/* Set up VSYNC active edge delay (in lines) */
-	v_de_vs = mode->vsync_start - mode->vdisplay;
 	hdmi_writeb(hdmi, v_de_vs, HDMI_FC_VSYNCINDELAY);
 
 	/* Set up HSYNC active pulse width (in pixel clks) */
@@ -1134,7 +1149,6 @@ static void hdmi_av_composer(struct dw_hdmi *hdmi,
 	hdmi_writeb(hdmi, hsync_len, HDMI_FC_HSYNCINWIDTH0);
 
 	/* Set up VSYNC active edge delay (in lines) */
-	vsync_len = mode->vsync_end - mode->vsync_start;
 	hdmi_writeb(hdmi, vsync_len, HDMI_FC_VSYNCINWIDTH);
 }
 
@@ -1608,6 +1622,8 @@ int dw_hdmi_bind(struct device *dev, struct device *master,
 	if (!hdmi)
 		return -ENOMEM;
 
+	hdmi->connector.interlace_allowed = 1;
+
 	hdmi->plat_data = plat_data;
 	hdmi->dev = dev;
 	hdmi->dev_type = plat_data->dev_type;
-- 
cgit v0.10.2


From 381f05a7a8428c84483d6bbd0aa3f4ced463c278 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Fri, 5 Jun 2015 15:25:08 +0100
Subject: drm: bridge/dw_hdmi: add connector mode forcing

When connected to HDMI sources, some DVI monitors de-assert their HPD
signal and TDMS loads for one seconds every four seconds when there is
no signal present on the connection.

Unfortunately, this behaviour is indistinguishable from a proper HDMI
setup with an AV receiver in the path to the display: the HDMI spec
requires us to detect HPD deassertions as short as 100ms, which indicate
that the EDID has changed.

Since it is possible to connect a DVI monitor to an AV receiver and then
to a HDMI source, merely working around this by detecting the lack of
HDMI vendor block in the EDID is insufficient - the AV receiver is at
liberty to modify the EDID as it sees fit, and it will place its own
parameters into the EDID including the HDMI vendor block.

DRM has support for forcing the state of a connector, which we should
implement to allow us to work around these broken DVI monitors - we can
tell DRM to force the connection state to indicate that there is always
a device connected to work around this problem.  Although this requires
manual configuration, it is better than nothing at all.

When a forced connection state has been set, there is no point handling
our RXSENSE interrupts, so disable them in this circumstance.

Tested-by: Philipp Zabel <p.zabel@pengutronix.de>
Reviewed-by: Fabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/bridge/dw_hdmi.c b/drivers/gpu/drm/bridge/dw_hdmi.c
index dc1a382..7dcae9d 100644
--- a/drivers/gpu/drm/bridge/dw_hdmi.c
+++ b/drivers/gpu/drm/bridge/dw_hdmi.c
@@ -126,7 +126,9 @@ struct dw_hdmi {
 	bool sink_has_audio;
 
 	struct mutex mutex;		/* for state below and previous_mode */
+	enum drm_connector_force force;	/* mutex-protected force state */
 	bool disabled;			/* DRM has disabled our bridge */
+	bool bridge_is_on;		/* indicates the bridge is on */
 
 	spinlock_t audio_lock;
 	struct mutex audio_mutex;
@@ -1378,12 +1380,36 @@ static void initialize_hdmi_ih_mutes(struct dw_hdmi *hdmi)
 
 static void dw_hdmi_poweron(struct dw_hdmi *hdmi)
 {
+	hdmi->bridge_is_on = true;
 	dw_hdmi_setup(hdmi, &hdmi->previous_mode);
 }
 
 static void dw_hdmi_poweroff(struct dw_hdmi *hdmi)
 {
 	dw_hdmi_phy_disable(hdmi);
+	hdmi->bridge_is_on = false;
+}
+
+static void dw_hdmi_update_power(struct dw_hdmi *hdmi)
+{
+	int force = hdmi->force;
+
+	if (hdmi->disabled) {
+		force = DRM_FORCE_OFF;
+	} else if (force == DRM_FORCE_UNSPECIFIED) {
+		if (hdmi_readb(hdmi, HDMI_PHY_STAT0) & HDMI_PHY_HPD)
+			force = DRM_FORCE_ON;
+		else
+			force = DRM_FORCE_OFF;
+	}
+
+	if (force == DRM_FORCE_OFF) {
+		if (hdmi->bridge_is_on)
+			dw_hdmi_poweroff(hdmi);
+	} else {
+		if (!hdmi->bridge_is_on)
+			dw_hdmi_poweron(hdmi);
+	}
 }
 
 static void dw_hdmi_bridge_mode_set(struct drm_bridge *bridge,
@@ -1413,7 +1439,7 @@ static void dw_hdmi_bridge_disable(struct drm_bridge *bridge)
 
 	mutex_lock(&hdmi->mutex);
 	hdmi->disabled = true;
-	dw_hdmi_poweroff(hdmi);
+	dw_hdmi_update_power(hdmi);
 	mutex_unlock(&hdmi->mutex);
 }
 
@@ -1422,8 +1448,8 @@ static void dw_hdmi_bridge_enable(struct drm_bridge *bridge)
 	struct dw_hdmi *hdmi = bridge->driver_private;
 
 	mutex_lock(&hdmi->mutex);
-	dw_hdmi_poweron(hdmi);
 	hdmi->disabled = false;
+	dw_hdmi_update_power(hdmi);
 	mutex_unlock(&hdmi->mutex);
 }
 
@@ -1438,6 +1464,11 @@ dw_hdmi_connector_detect(struct drm_connector *connector, bool force)
 	struct dw_hdmi *hdmi = container_of(connector, struct dw_hdmi,
 					     connector);
 
+	mutex_lock(&hdmi->mutex);
+	hdmi->force = DRM_FORCE_UNSPECIFIED;
+	dw_hdmi_update_power(hdmi);
+	mutex_unlock(&hdmi->mutex);
+
 	return hdmi_readb(hdmi, HDMI_PHY_STAT0) & HDMI_PHY_HPD ?
 		connector_status_connected : connector_status_disconnected;
 }
@@ -1502,11 +1533,23 @@ static void dw_hdmi_connector_destroy(struct drm_connector *connector)
 	drm_connector_cleanup(connector);
 }
 
+static void dw_hdmi_connector_force(struct drm_connector *connector)
+{
+	struct dw_hdmi *hdmi = container_of(connector, struct dw_hdmi,
+					     connector);
+
+	mutex_lock(&hdmi->mutex);
+	hdmi->force = connector->force;
+	dw_hdmi_update_power(hdmi);
+	mutex_unlock(&hdmi->mutex);
+}
+
 static struct drm_connector_funcs dw_hdmi_connector_funcs = {
 	.dpms = drm_helper_connector_dpms,
 	.fill_modes = drm_helper_probe_single_connector_modes,
 	.detect = dw_hdmi_connector_detect,
 	.destroy = dw_hdmi_connector_destroy,
+	.force = dw_hdmi_connector_force,
 };
 
 static struct drm_connector_helper_funcs dw_hdmi_connector_helper_funcs = {
@@ -1552,12 +1595,12 @@ static irqreturn_t dw_hdmi_irq(int irq, void *dev_id)
 		if (phy_int_pol & HDMI_PHY_HPD) {
 			dev_dbg(hdmi->dev, "EVENT=plugin\n");
 
-			if (!hdmi->disabled)
+			if (!hdmi->disabled && !hdmi->force)
 				dw_hdmi_poweron(hdmi);
 		} else {
 			dev_dbg(hdmi->dev, "EVENT=plugout\n");
 
-			if (!hdmi->disabled)
+			if (!hdmi->disabled && !hdmi->force)
 				dw_hdmi_poweroff(hdmi);
 		}
 		mutex_unlock(&hdmi->mutex);
-- 
cgit v0.10.2


From aeac23bda87fd98049352d3bc27a1cb80445e7a0 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Fri, 5 Jun 2015 13:46:22 +0100
Subject: drm: bridge/dw_hdmi: improve HDMI enable/disable handling

HDMI sinks are permitted to de-assert and re-assert the HPD signal to
indicate that their EDID has been updated, which may not involve a
change of video information.

An example of where such a situation can arise is when an AV receiver
is connected between the source and the display device.  Events which
can cause the HPD to be deasserted include:

 * turning on or switching to standby the AV receiver.
 * turning on or switching to standby the display device.

Each of these can change the entire EDID data, or just a part of the
EDID data - it's up to the connected HDMI sink to do what they desire
here.  For example

 - with the AV receiver and display device both in standby, a source
   connected to the AV receiver may provide its own EDID to the source.
 - turning on the display device causes the display device's EDID to be
   made available in an unmodified form to the source.
 - subsequently turning on the AV receiver then provides a modified
   version of the display device's EDID.

Moreover, HPD doesn't tell us whether something is actually listening
on the HDMI TDMS signals.  The phy gives us a set of RXSENSE indications
which tell us whether there is a sink connected to the TMDS signals.

Currently, we use the HPD signal to enable or disable the HDMI block,
which is questionable when HPD is used in this manner.  Using the
RXSENSE would be more appropriate, but there is some bad behaviour
which needs to be coped with.  The iMX6 implementation lets the TMDS
signals float when the phy is "powered down", which cause spurious
interrupts.  Rather than just using RXSENSE, use RXSENSE and HPD
becoming both active to signal the presence of a device, but loss
of RXSENSE to indicate that the device has been unplugged.

The side effect of this change is that a sink deasserting the HPD
signal to cause a re-read of the EDID data will not cause the bridge
to immediately disable the video signal.

Tested-by: Philipp Zabel <p.zabel@pengutronix.de>
Reviewed-by: Fabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/bridge/dw_hdmi.c b/drivers/gpu/drm/bridge/dw_hdmi.c
index 7dcae9d..7a130f6 100644
--- a/drivers/gpu/drm/bridge/dw_hdmi.c
+++ b/drivers/gpu/drm/bridge/dw_hdmi.c
@@ -129,6 +129,8 @@ struct dw_hdmi {
 	enum drm_connector_force force;	/* mutex-protected force state */
 	bool disabled;			/* DRM has disabled our bridge */
 	bool bridge_is_on;		/* indicates the bridge is on */
+	bool rxsense;			/* rxsense state */
+	u8 phy_mask;			/* desired phy int mask settings */
 
 	spinlock_t audio_lock;
 	struct mutex audio_mutex;
@@ -142,6 +144,14 @@ struct dw_hdmi {
 	u8 (*read)(struct dw_hdmi *hdmi, int offset);
 };
 
+#define HDMI_IH_PHY_STAT0_RX_SENSE \
+	(HDMI_IH_PHY_STAT0_RX_SENSE0 | HDMI_IH_PHY_STAT0_RX_SENSE1 | \
+	 HDMI_IH_PHY_STAT0_RX_SENSE2 | HDMI_IH_PHY_STAT0_RX_SENSE3)
+
+#define HDMI_PHY_RX_SENSE \
+	(HDMI_PHY_RX_SENSE0 | HDMI_PHY_RX_SENSE1 | \
+	 HDMI_PHY_RX_SENSE2 | HDMI_PHY_RX_SENSE3)
+
 static void dw_hdmi_writel(struct dw_hdmi *hdmi, u8 val, int offset)
 {
 	writel(val, hdmi->regs + (offset << 2));
@@ -1318,10 +1328,11 @@ static int dw_hdmi_fb_registered(struct dw_hdmi *hdmi)
 		    HDMI_PHY_I2CM_CTLINT_ADDR);
 
 	/* enable cable hot plug irq */
-	hdmi_writeb(hdmi, (u8)~HDMI_PHY_HPD, HDMI_PHY_MASK0);
+	hdmi_writeb(hdmi, hdmi->phy_mask, HDMI_PHY_MASK0);
 
 	/* Clear Hotplug interrupts */
-	hdmi_writeb(hdmi, HDMI_IH_PHY_STAT0_HPD, HDMI_IH_PHY_STAT0);
+	hdmi_writeb(hdmi, HDMI_IH_PHY_STAT0_HPD | HDMI_IH_PHY_STAT0_RX_SENSE,
+		    HDMI_IH_PHY_STAT0);
 
 	return 0;
 }
@@ -1397,7 +1408,7 @@ static void dw_hdmi_update_power(struct dw_hdmi *hdmi)
 	if (hdmi->disabled) {
 		force = DRM_FORCE_OFF;
 	} else if (force == DRM_FORCE_UNSPECIFIED) {
-		if (hdmi_readb(hdmi, HDMI_PHY_STAT0) & HDMI_PHY_HPD)
+		if (hdmi->rxsense)
 			force = DRM_FORCE_ON;
 		else
 			force = DRM_FORCE_OFF;
@@ -1412,6 +1423,31 @@ static void dw_hdmi_update_power(struct dw_hdmi *hdmi)
 	}
 }
 
+/*
+ * Adjust the detection of RXSENSE according to whether we have a forced
+ * connection mode enabled, or whether we have been disabled.  There is
+ * no point processing RXSENSE interrupts if we have a forced connection
+ * state, or DRM has us disabled.
+ *
+ * We also disable rxsense interrupts when we think we're disconnected
+ * to avoid floating TDMS signals giving false rxsense interrupts.
+ *
+ * Note: we still need to listen for HPD interrupts even when DRM has us
+ * disabled so that we can detect a connect event.
+ */
+static void dw_hdmi_update_phy_mask(struct dw_hdmi *hdmi)
+{
+	u8 old_mask = hdmi->phy_mask;
+
+	if (hdmi->force || hdmi->disabled || !hdmi->rxsense)
+		hdmi->phy_mask |= HDMI_PHY_RX_SENSE;
+	else
+		hdmi->phy_mask &= ~HDMI_PHY_RX_SENSE;
+
+	if (old_mask != hdmi->phy_mask)
+		hdmi_writeb(hdmi, hdmi->phy_mask, HDMI_PHY_MASK0);
+}
+
 static void dw_hdmi_bridge_mode_set(struct drm_bridge *bridge,
 				    struct drm_display_mode *orig_mode,
 				    struct drm_display_mode *mode)
@@ -1440,6 +1476,7 @@ static void dw_hdmi_bridge_disable(struct drm_bridge *bridge)
 	mutex_lock(&hdmi->mutex);
 	hdmi->disabled = true;
 	dw_hdmi_update_power(hdmi);
+	dw_hdmi_update_phy_mask(hdmi);
 	mutex_unlock(&hdmi->mutex);
 }
 
@@ -1450,6 +1487,7 @@ static void dw_hdmi_bridge_enable(struct drm_bridge *bridge)
 	mutex_lock(&hdmi->mutex);
 	hdmi->disabled = false;
 	dw_hdmi_update_power(hdmi);
+	dw_hdmi_update_phy_mask(hdmi);
 	mutex_unlock(&hdmi->mutex);
 }
 
@@ -1467,6 +1505,7 @@ dw_hdmi_connector_detect(struct drm_connector *connector, bool force)
 	mutex_lock(&hdmi->mutex);
 	hdmi->force = DRM_FORCE_UNSPECIFIED;
 	dw_hdmi_update_power(hdmi);
+	dw_hdmi_update_phy_mask(hdmi);
 	mutex_unlock(&hdmi->mutex);
 
 	return hdmi_readb(hdmi, HDMI_PHY_STAT0) & HDMI_PHY_HPD ?
@@ -1541,6 +1580,7 @@ static void dw_hdmi_connector_force(struct drm_connector *connector)
 	mutex_lock(&hdmi->mutex);
 	hdmi->force = connector->force;
 	dw_hdmi_update_power(hdmi);
+	dw_hdmi_update_phy_mask(hdmi);
 	mutex_unlock(&hdmi->mutex);
 }
 
@@ -1582,33 +1622,69 @@ static irqreturn_t dw_hdmi_hardirq(int irq, void *dev_id)
 static irqreturn_t dw_hdmi_irq(int irq, void *dev_id)
 {
 	struct dw_hdmi *hdmi = dev_id;
-	u8 intr_stat;
-	u8 phy_int_pol;
+	u8 intr_stat, phy_int_pol, phy_pol_mask, phy_stat;
 
 	intr_stat = hdmi_readb(hdmi, HDMI_IH_PHY_STAT0);
-
 	phy_int_pol = hdmi_readb(hdmi, HDMI_PHY_POL0);
+	phy_stat = hdmi_readb(hdmi, HDMI_PHY_STAT0);
+
+	phy_pol_mask = 0;
+	if (intr_stat & HDMI_IH_PHY_STAT0_HPD)
+		phy_pol_mask |= HDMI_PHY_HPD;
+	if (intr_stat & HDMI_IH_PHY_STAT0_RX_SENSE0)
+		phy_pol_mask |= HDMI_PHY_RX_SENSE0;
+	if (intr_stat & HDMI_IH_PHY_STAT0_RX_SENSE1)
+		phy_pol_mask |= HDMI_PHY_RX_SENSE1;
+	if (intr_stat & HDMI_IH_PHY_STAT0_RX_SENSE2)
+		phy_pol_mask |= HDMI_PHY_RX_SENSE2;
+	if (intr_stat & HDMI_IH_PHY_STAT0_RX_SENSE3)
+		phy_pol_mask |= HDMI_PHY_RX_SENSE3;
+
+	if (phy_pol_mask)
+		hdmi_modb(hdmi, ~phy_int_pol, phy_pol_mask, HDMI_PHY_POL0);
 
-	if (intr_stat & HDMI_IH_PHY_STAT0_HPD) {
-		hdmi_modb(hdmi, ~phy_int_pol, HDMI_PHY_HPD, HDMI_PHY_POL0);
+	/*
+	 * RX sense tells us whether the TDMS transmitters are detecting
+	 * load - in other words, there's something listening on the
+	 * other end of the link.  Use this to decide whether we should
+	 * power on the phy as HPD may be toggled by the sink to merely
+	 * ask the source to re-read the EDID.
+	 */
+	if (intr_stat &
+	    (HDMI_IH_PHY_STAT0_RX_SENSE | HDMI_IH_PHY_STAT0_HPD)) {
 		mutex_lock(&hdmi->mutex);
-		if (phy_int_pol & HDMI_PHY_HPD) {
-			dev_dbg(hdmi->dev, "EVENT=plugin\n");
-
-			if (!hdmi->disabled && !hdmi->force)
-				dw_hdmi_poweron(hdmi);
-		} else {
-			dev_dbg(hdmi->dev, "EVENT=plugout\n");
-
-			if (!hdmi->disabled && !hdmi->force)
-				dw_hdmi_poweroff(hdmi);
+		if (!hdmi->disabled && !hdmi->force) {
+			/*
+			 * If the RX sense status indicates we're disconnected,
+			 * clear the software rxsense status.
+			 */
+			if (!(phy_stat & HDMI_PHY_RX_SENSE))
+				hdmi->rxsense = false;
+
+			/*
+			 * Only set the software rxsense status when both
+			 * rxsense and hpd indicates we're connected.
+			 * This avoids what seems to be bad behaviour in
+			 * at least iMX6S versions of the phy.
+			 */
+			if (phy_stat & HDMI_PHY_HPD)
+				hdmi->rxsense = true;
+
+			dw_hdmi_update_power(hdmi);
+			dw_hdmi_update_phy_mask(hdmi);
 		}
 		mutex_unlock(&hdmi->mutex);
+	}
+
+	if (intr_stat & HDMI_IH_PHY_STAT0_HPD) {
+		dev_dbg(hdmi->dev, "EVENT=%s\n",
+			phy_int_pol & HDMI_PHY_HPD ? "plugin" : "plugout");
 		drm_helper_hpd_irq_event(hdmi->bridge->dev);
 	}
 
 	hdmi_writeb(hdmi, intr_stat, HDMI_IH_PHY_STAT0);
-	hdmi_writeb(hdmi, ~HDMI_IH_PHY_STAT0_HPD, HDMI_IH_MUTE_PHY_STAT0);
+	hdmi_writeb(hdmi, ~(HDMI_IH_PHY_STAT0_HPD | HDMI_IH_PHY_STAT0_RX_SENSE),
+		    HDMI_IH_MUTE_PHY_STAT0);
 
 	return IRQ_HANDLED;
 }
@@ -1674,6 +1750,8 @@ int dw_hdmi_bind(struct device *dev, struct device *master,
 	hdmi->ratio = 100;
 	hdmi->encoder = encoder;
 	hdmi->disabled = true;
+	hdmi->rxsense = true;
+	hdmi->phy_mask = (u8)~(HDMI_PHY_HPD | HDMI_PHY_RX_SENSE);
 
 	mutex_init(&hdmi->mutex);
 	mutex_init(&hdmi->audio_mutex);
@@ -1764,10 +1842,11 @@ int dw_hdmi_bind(struct device *dev, struct device *master,
 	 * Configure registers related to HDMI interrupt
 	 * generation before registering IRQ.
 	 */
-	hdmi_writeb(hdmi, HDMI_PHY_HPD, HDMI_PHY_POL0);
+	hdmi_writeb(hdmi, HDMI_PHY_HPD | HDMI_PHY_RX_SENSE, HDMI_PHY_POL0);
 
 	/* Clear Hotplug interrupts */
-	hdmi_writeb(hdmi, HDMI_IH_PHY_STAT0_HPD, HDMI_IH_PHY_STAT0);
+	hdmi_writeb(hdmi, HDMI_IH_PHY_STAT0_HPD | HDMI_IH_PHY_STAT0_RX_SENSE,
+		    HDMI_IH_PHY_STAT0);
 
 	ret = dw_hdmi_fb_registered(hdmi);
 	if (ret)
@@ -1778,7 +1857,8 @@ int dw_hdmi_bind(struct device *dev, struct device *master,
 		goto err_iahb;
 
 	/* Unmute interrupts */
-	hdmi_writeb(hdmi, ~HDMI_IH_PHY_STAT0_HPD, HDMI_IH_MUTE_PHY_STAT0);
+	hdmi_writeb(hdmi, ~(HDMI_IH_PHY_STAT0_HPD | HDMI_IH_PHY_STAT0_RX_SENSE),
+		    HDMI_IH_MUTE_PHY_STAT0);
 
 	dev_set_drvdata(dev, hdmi);
 
-- 
cgit v0.10.2


From 1f2449cdd689931f5f511235899b2a0e28010e50 Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Tue, 6 Oct 2015 14:47:55 +0200
Subject: drm/i915: Fix kerneldoc for i915_gem_shrink_all

I've botched this, so let's fix it.

Botched in

commit eb0b44adc08c0be01a027eb009e9cdadc31e65a2
Author: Daniel Vetter <daniel.vetter@ffwll.ch>
Date:   Wed Mar 18 14:47:59 2015 +0100

    drm/i915: kerneldoc for i915_gem_shrinker.c

v2: Be a good citizen^Wmaintainer and add the proper commit citation.
Noticed by Jani.

Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

diff --git a/drivers/gpu/drm/i915/i915_gem_shrinker.c b/drivers/gpu/drm/i915/i915_gem_shrinker.c
index f6ecbda..6743417 100644
--- a/drivers/gpu/drm/i915/i915_gem_shrinker.c
+++ b/drivers/gpu/drm/i915/i915_gem_shrinker.c
@@ -143,7 +143,7 @@ i915_gem_shrink(struct drm_i915_private *dev_priv,
 }
 
 /**
- * i915_gem_shrink - Shrink buffer object caches completely
+ * i915_gem_shrink_all - Shrink buffer object caches completely
  * @dev_priv: i915 device
  *
  * This is a simple wraper around i915_gem_shrink() to aggressively shrink all
-- 
cgit v0.10.2


From 143875400b33fc20e9dd060e90034f2e9e9ffc98 Mon Sep 17 00:00:00 2001
From: Chris Wilson <chris@chris-wilson.co.uk>
Date: Thu, 1 Oct 2015 12:18:25 +0100
Subject: drm/i915: shrinker_control->nr_to_scan is now unsigned long

As the shrinker_control now passes us unsigned long targets, update our
shrinker functions to match.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index c93f9a1..85c4894 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -3203,7 +3203,7 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,
 
 /* i915_gem_shrinker.c */
 unsigned long i915_gem_shrink(struct drm_i915_private *dev_priv,
-			      long target,
+			      unsigned long target,
 			      unsigned flags);
 #define I915_SHRINK_PURGEABLE 0x1
 #define I915_SHRINK_UNBOUND 0x2
diff --git a/drivers/gpu/drm/i915/i915_gem_shrinker.c b/drivers/gpu/drm/i915/i915_gem_shrinker.c
index 6743417..858df2b 100644
--- a/drivers/gpu/drm/i915/i915_gem_shrinker.c
+++ b/drivers/gpu/drm/i915/i915_gem_shrinker.c
@@ -73,7 +73,7 @@ static bool mutex_is_locked_by(struct mutex *mutex, struct task_struct *task)
  */
 unsigned long
 i915_gem_shrink(struct drm_i915_private *dev_priv,
-		long target, unsigned flags)
+		unsigned long target, unsigned flags)
 {
 	const struct {
 		struct list_head *list;
@@ -159,7 +159,7 @@ i915_gem_shrink(struct drm_i915_private *dev_priv,
 unsigned long i915_gem_shrink_all(struct drm_i915_private *dev_priv)
 {
 	i915_gem_evict_everything(dev_priv->dev);
-	return i915_gem_shrink(dev_priv, LONG_MAX,
+	return i915_gem_shrink(dev_priv, -1UL,
 			       I915_SHRINK_BOUND | I915_SHRINK_UNBOUND);
 }
 
-- 
cgit v0.10.2


From 3f4c90bd203125c807a96f18d3195cf3a1988279 Mon Sep 17 00:00:00 2001
From: Libin Yang <libin.yang@intel.com>
Date: Thu, 1 Oct 2015 17:01:08 +0800
Subject: drm/i915: add kerneldoc for i915_audio_component

Add the kerneldoc for i915_audio_component in i915_component.h

Signed-off-by: Libin Yang <libin.yang@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/include/drm/i915_component.h b/include/drm/i915_component.h
index 89dc7d6..30d89e0 100644
--- a/include/drm/i915_component.h
+++ b/include/drm/i915_component.h
@@ -30,38 +30,49 @@
  */
 #define MAX_PORTS 5
 
-struct i915_audio_component {
-	struct device *dev;
+/**
+ * struct i915_audio_component_ops - callbacks defined in gfx driver
+ * @owner: the module owner
+ * @get_power: get the POWER_DOMAIN_AUDIO power well
+ * @put_power: put the POWER_DOMAIN_AUDIO power well
+ * @codec_wake_override: Enable/Disable generating the codec wake signal
+ * @get_cdclk_freq: get the Core Display Clock in KHz
+ * @sync_audio_rate: set n/cts based on the sample rate
+ */
+struct i915_audio_component_ops {
+	struct module *owner;
+	void (*get_power)(struct device *);
+	void (*put_power)(struct device *);
+	void (*codec_wake_override)(struct device *, bool enable);
+	int (*get_cdclk_freq)(struct device *);
+	int (*sync_audio_rate)(struct device *, int port, int rate);
+};
+
+struct i915_audio_component_audio_ops {
+	void *audio_ptr;
 	/**
-	 * @aud_sample_rate: the array of audio sample rate per port
+	 * Call from i915 driver, notifying the HDA driver that
+	 * pin sense and/or ELD information has changed.
+	 * @audio_ptr:		HDA driver object
+	 * @port:	Which port has changed (PORTA / PORTB / PORTC etc)
 	 */
+	void (*pin_eld_notify)(void *audio_ptr, int port);
+};
+
+/**
+ * struct i915_audio_component - used for audio video interaction
+ * @dev: the device from gfx driver
+ * @aud_sample_rate: the array of audio sample rate per port
+ * @ops: callback for audio driver calling
+ * @audio_ops: Call from i915 driver
+ */
+struct i915_audio_component {
+	struct device *dev;
 	int aud_sample_rate[MAX_PORTS];
 
-	const struct i915_audio_component_ops {
-		struct module *owner;
-		void (*get_power)(struct device *);
-		void (*put_power)(struct device *);
-		void (*codec_wake_override)(struct device *, bool enable);
-		int (*get_cdclk_freq)(struct device *);
-		/**
-		 * @sync_audio_rate: set n/cts based on the sample rate
-		 *
-		 * Called from audio driver. After audio driver sets the
-		 * sample rate, it will call this function to set n/cts
-		 */
-		int (*sync_audio_rate)(struct device *, int port, int rate);
-	} *ops;
+	const struct i915_audio_component_ops *ops;
 
-	const struct i915_audio_component_audio_ops {
-		void *audio_ptr;
-		/**
-		 * Call from i915 driver, notifying the HDA driver that
-		 * pin sense and/or ELD information has changed.
-		 * @audio_ptr:		HDA driver object
-		 * @port:		Which port has changed (PORTA / PORTB / PORTC etc)
-		 */
-		void (*pin_eld_notify)(void *audio_ptr, int port);
-	} *audio_ops;
+	const struct i915_audio_component_audio_ops *audio_ops;
 };
 
 #endif /* _I915_COMPONENT_H_ */
-- 
cgit v0.10.2


From cb422619976f3f1b71f68f0c1b5a764e9f90fb0c Mon Sep 17 00:00:00 2001
From: Libin Yang <libin.yang@intel.com>
Date: Thu, 1 Oct 2015 17:01:09 +0800
Subject: drm/i915: DocBook add i915_component.h support

Add the item of i915_component.h in DocBook and add the DOC for
i915_component.h. Explain the struct i915_audio_component_ops and
struct i915_audio_component_audio_ops usage.

Signed-off-by: Libin Yang <libin.yang@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/Documentation/DocBook/drm.tmpl b/Documentation/DocBook/drm.tmpl
index f78ca7f..a249c73 100644
--- a/Documentation/DocBook/drm.tmpl
+++ b/Documentation/DocBook/drm.tmpl
@@ -4051,6 +4051,7 @@ int num_ioctls;</synopsis>
 	<title>High Definition Audio</title>
 !Pdrivers/gpu/drm/i915/intel_audio.c High Definition Audio over HDMI and Display Port
 !Idrivers/gpu/drm/i915/intel_audio.c
+!Iinclude/drm/i915_component.h
       </sect2>
       <sect2>
 	<title>Panel Self Refresh PSR (PSR/SRD)</title>
diff --git a/drivers/gpu/drm/i915/intel_audio.c b/drivers/gpu/drm/i915/intel_audio.c
index 72d696b..56c2f54 100644
--- a/drivers/gpu/drm/i915/intel_audio.c
+++ b/drivers/gpu/drm/i915/intel_audio.c
@@ -50,6 +50,11 @@
  * co-operation between the graphics and audio drivers is handled via audio
  * related registers. (The notable exception is the power management, not
  * covered here.)
+ *
+ * The struct i915_audio_component is used to interact between the graphics
+ * and audio drivers. The struct i915_audio_component_ops *ops in it is
+ * defined in graphics driver and called in audio driver. The
+ * struct i915_audio_component_audio_ops *audio_ops is called from i915 driver.
  */
 
 static const struct {
-- 
cgit v0.10.2


From 3abafa539d2c17d46df3f47b35eb784fdf3020a1 Mon Sep 17 00:00:00 2001
From: Chris Wilson <chris@chris-wilson.co.uk>
Date: Thu, 1 Oct 2015 12:18:26 +0100
Subject: drm/i915: Add a tracepoint for the shrinker

Often it is very useful to know why we suddenly purge vast tracts of
memory and surprisingly up until now we didn't even have a tracepoint
for when we shrink our memory.

Note that there are slab_start/end tracepoints already, but those
don't cover the internal recursion when we directly call into our
shrinker code. Hence a separate tracepoint seems justified. Also note
that we don't really need a separate tracepoint for the actual amount
of pages freed since we already have an unbind tracpoint for that.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
[danvet: Add a note that there's also slab_start/end and why they're
insufficient.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_shrinker.c b/drivers/gpu/drm/i915/i915_gem_shrinker.c
index 858df2b..1b66e1d 100644
--- a/drivers/gpu/drm/i915/i915_gem_shrinker.c
+++ b/drivers/gpu/drm/i915/i915_gem_shrinker.c
@@ -85,6 +85,8 @@ i915_gem_shrink(struct drm_i915_private *dev_priv,
 	}, *phase;
 	unsigned long count = 0;
 
+	trace_i915_gem_shrink(dev_priv, target, flags);
+
 	/*
 	 * As we may completely rewrite the (un)bound list whilst unbinding
 	 * (due to retiring requests) we have to strictly process only
diff --git a/drivers/gpu/drm/i915/i915_trace.h b/drivers/gpu/drm/i915/i915_trace.h
index d0993bc..04fe849 100644
--- a/drivers/gpu/drm/i915/i915_trace.h
+++ b/drivers/gpu/drm/i915/i915_trace.h
@@ -107,6 +107,26 @@ TRACE_EVENT(i915_gem_object_create,
 	    TP_printk("obj=%p, size=%u", __entry->obj, __entry->size)
 );
 
+TRACE_EVENT(i915_gem_shrink,
+	    TP_PROTO(struct drm_i915_private *i915, unsigned long target, unsigned flags),
+	    TP_ARGS(i915, target, flags),
+
+	    TP_STRUCT__entry(
+			     __field(int, dev)
+			     __field(unsigned long, target)
+			     __field(unsigned, flags)
+			     ),
+
+	    TP_fast_assign(
+			   __entry->dev = i915->dev->primary->index;
+			   __entry->target = target;
+			   __entry->flags = flags;
+			   ),
+
+	    TP_printk("dev=%d, target=%lu, flags=%x",
+		      __entry->dev, __entry->target, __entry->flags)
+);
+
 TRACE_EVENT(i915_vma_bind,
 	    TP_PROTO(struct i915_vma *vma, unsigned flags),
 	    TP_ARGS(vma, flags),
-- 
cgit v0.10.2


From c9c0f5ea1b60b1f759a928505205335775e20512 Mon Sep 17 00:00:00 2001
From: Chris Wilson <chris@chris-wilson.co.uk>
Date: Thu, 1 Oct 2015 12:18:27 +0100
Subject: drm/i915: During shrink_all we only need to idle the GPU

We can forgo an evict-everything here as the shrinker operation itself
will unbind any vma as required. If we explicitly idle the GPU through a
switch to the default context, we not only create a request in an
illegal context (e.g. whilst shrinking during execbuf with a request
already allocated), but switching to the default context will not free
up the memory backing the active contexts - unless in the unlikely
situation that context had already been closed (and just kept arrive by
being the current context). The saving is near zero and the danger real.

To compensate for the loss of the forced retire, add a couple of
retire-requests to i915_gem_shirnk() - this should help free up any
transitive cache from the requests.

Note that the second retire_requests is for the benefit of the
hand-rolled execlist ctx active tracking: We need to manually kick
requests to get those unpinned again. Once that's fixed we can try to
remove this again.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
[danvet: Add summary of why we need a pile of retire_requests.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_shrinker.c b/drivers/gpu/drm/i915/i915_gem_shrinker.c
index 1b66e1d..03cf3cc 100644
--- a/drivers/gpu/drm/i915/i915_gem_shrinker.c
+++ b/drivers/gpu/drm/i915/i915_gem_shrinker.c
@@ -86,6 +86,7 @@ i915_gem_shrink(struct drm_i915_private *dev_priv,
 	unsigned long count = 0;
 
 	trace_i915_gem_shrink(dev_priv, target, flags);
+	i915_gem_retire_requests(dev_priv->dev);
 
 	/*
 	 * As we may completely rewrite the (un)bound list whilst unbinding
@@ -141,6 +142,8 @@ i915_gem_shrink(struct drm_i915_private *dev_priv,
 		list_splice(&still_in_list, phase->list);
 	}
 
+	i915_gem_retire_requests(dev_priv->dev);
+
 	return count;
 }
 
@@ -160,7 +163,6 @@ i915_gem_shrink(struct drm_i915_private *dev_priv,
  */
 unsigned long i915_gem_shrink_all(struct drm_i915_private *dev_priv)
 {
-	i915_gem_evict_everything(dev_priv->dev);
 	return i915_gem_shrink(dev_priv, -1UL,
 			       I915_SHRINK_BOUND | I915_SHRINK_UNBOUND);
 }
-- 
cgit v0.10.2


From ce8daef3580ae38fea9599e2193e4c368357c4c6 Mon Sep 17 00:00:00 2001
From: Chris Wilson <chris@chris-wilson.co.uk>
Date: Thu, 1 Oct 2015 12:18:28 +0100
Subject: drm/i915: Remove dead i915_gem_evict_everything()

With UMS gone, we no longer use it during suspend. And with the last
user removed from the shrinker, we can remove the dead code.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 1626f3d..8031052 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -3177,7 +3177,6 @@ int __must_check i915_gem_evict_something(struct drm_device *dev,
 					  unsigned long end,
 					  unsigned flags);
 int i915_gem_evict_vm(struct i915_address_space *vm, bool do_idle);
-int i915_gem_evict_everything(struct drm_device *dev);
 
 /* belongs in i915_gem_gtt.h */
 static inline void i915_gem_chipset_flush(struct drm_device *dev)
diff --git a/drivers/gpu/drm/i915/i915_gem_evict.c b/drivers/gpu/drm/i915/i915_gem_evict.c
index d09e35e..d71a133 100644
--- a/drivers/gpu/drm/i915/i915_gem_evict.c
+++ b/drivers/gpu/drm/i915/i915_gem_evict.c
@@ -237,48 +237,3 @@ int i915_gem_evict_vm(struct i915_address_space *vm, bool do_idle)
 
 	return 0;
 }
-
-/**
- * i915_gem_evict_everything - Try to evict all objects
- * @dev: Device to evict objects for
- *
- * This functions tries to evict all gem objects from all address spaces. Used
- * by the shrinker as a last-ditch effort and for suspend, before releasing the
- * backing storage of all unbound objects.
- */
-int
-i915_gem_evict_everything(struct drm_device *dev)
-{
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct i915_address_space *vm, *v;
-	bool lists_empty = true;
-	int ret;
-
-	list_for_each_entry(vm, &dev_priv->vm_list, global_link) {
-		lists_empty = (list_empty(&vm->inactive_list) &&
-			       list_empty(&vm->active_list));
-		if (!lists_empty)
-			lists_empty = false;
-	}
-
-	if (lists_empty)
-		return -ENOSPC;
-
-	trace_i915_gem_evict_everything(dev);
-
-	/* The gpu_idle will flush everything in the write domain to the
-	 * active list. Then we must move everything off the active list
-	 * with retire requests.
-	 */
-	ret = i915_gpu_idle(dev);
-	if (ret)
-		return ret;
-
-	i915_gem_retire_requests(dev);
-
-	/* Having flushed everything, unbind() should never raise an error */
-	list_for_each_entry_safe(vm, v, &dev_priv->vm_list, global_link)
-		WARN_ON(i915_gem_evict_vm(vm, false));
-
-	return 0;
-}
-- 
cgit v0.10.2


From 5763ff04dc4ebdd13d069d44513b10805ebebd8c Mon Sep 17 00:00:00 2001
From: Chris Wilson <chris@chris-wilson.co.uk>
Date: Thu, 1 Oct 2015 12:18:29 +0100
Subject: drm/i915: Avoid GPU stalls from kswapd

Exclude active GPU pages from the purview of the background shrinker
(kswapd), as these cause uncontrollable GPU stalls. Given that the
shrinker is rerun until the freelists are satisfied, we should have
opportunity in subsequent passes to recover the pages once idle. If the
machine does run out of memory entirely, we have the forced idling in the
oom-notifier as a means of releasing all the pages we can before an oom
is prematurely executed.

Note that this relies upon an up-front retire_requests to keep the
inactive list in shape, which was added in a previous patch, mostly as
execlist ctx pinning band-aids.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Damien Lespiau <damien.lespiau@intel.com>
[danvet: Add note about retire_requests.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 8031052..b39b5cc 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -3212,6 +3212,7 @@ unsigned long i915_gem_shrink(struct drm_i915_private *dev_priv,
 #define I915_SHRINK_PURGEABLE 0x1
 #define I915_SHRINK_UNBOUND 0x2
 #define I915_SHRINK_BOUND 0x4
+#define I915_SHRINK_ACTIVE 0x8
 unsigned long i915_gem_shrink_all(struct drm_i915_private *dev_priv);
 void i915_gem_shrinker_init(struct drm_i915_private *dev_priv);
 
diff --git a/drivers/gpu/drm/i915/i915_gem_shrinker.c b/drivers/gpu/drm/i915/i915_gem_shrinker.c
index 03cf3cc..f7df54a 100644
--- a/drivers/gpu/drm/i915/i915_gem_shrinker.c
+++ b/drivers/gpu/drm/i915/i915_gem_shrinker.c
@@ -126,6 +126,9 @@ i915_gem_shrink(struct drm_i915_private *dev_priv,
 			    obj->madv != I915_MADV_DONTNEED)
 				continue;
 
+			if ((flags & I915_SHRINK_ACTIVE) == 0 && obj->active)
+				continue;
+
 			drm_gem_object_reference(&obj->base);
 
 			/* For the unbound phase, this should be a no-op! */
@@ -164,7 +167,9 @@ i915_gem_shrink(struct drm_i915_private *dev_priv,
 unsigned long i915_gem_shrink_all(struct drm_i915_private *dev_priv)
 {
 	return i915_gem_shrink(dev_priv, -1UL,
-			       I915_SHRINK_BOUND | I915_SHRINK_UNBOUND);
+			       I915_SHRINK_BOUND |
+			       I915_SHRINK_UNBOUND |
+			       I915_SHRINK_ACTIVE);
 }
 
 static bool i915_gem_shrinker_lock(struct drm_device *dev, bool *unlock)
@@ -217,7 +222,7 @@ i915_gem_shrinker_count(struct shrinker *shrinker, struct shrink_control *sc)
 			count += obj->base.size >> PAGE_SHIFT;
 
 	list_for_each_entry(obj, &dev_priv->mm.bound_list, global_list) {
-		if (obj->pages_pin_count == num_vma_bound(obj))
+		if (!obj->active && obj->pages_pin_count == num_vma_bound(obj))
 			count += obj->base.size >> PAGE_SHIFT;
 	}
 
-- 
cgit v0.10.2


From 2f5945bc9076bd7b1086c1b4ac47dea8ab336e57 Mon Sep 17 00:00:00 2001
From: Chris Wilson <chris@chris-wilson.co.uk>
Date: Tue, 6 Oct 2015 11:39:55 +0100
Subject: drm/i915: Kill DRI1 cliprects

Passing cliprects into the kernel for it to re-execute the batch buffer
with different CMD_DRAWRECT died out long ago. As DRI1 support has been
removed from the kernel, we can now simply reject any execbuf trying to
use this "feature".

To keep Daniel happy with the prospect of being able to reuse these
fields in the next decade, continue to ensure that current userspace is
not passing garbage in through the dead fields.

v2: Fix the cliprects_ptr check

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Dave Gordon <david.s.gordon@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index edc17be..6ed7d63a 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -945,7 +945,21 @@ i915_gem_check_execbuffer(struct drm_i915_gem_execbuffer2 *exec)
 	if (exec->flags & __I915_EXEC_UNKNOWN_FLAGS)
 		return false;
 
-	return ((exec->batch_start_offset | exec->batch_len) & 0x7) == 0;
+	/* Kernel clipping was a DRI1 misfeature */
+	if (exec->num_cliprects || exec->cliprects_ptr)
+		return false;
+
+	if (exec->DR4 == 0xffffffff) {
+		DRM_DEBUG("UXA submitting garbage DR4, fixing up\n");
+		exec->DR4 = 0;
+	}
+	if (exec->DR1 || exec->DR4)
+		return false;
+
+	if ((exec->batch_start_offset | exec->batch_len) & 0x7)
+		return false;
+
+	return true;
 }
 
 static int
@@ -1109,47 +1123,6 @@ i915_reset_gen7_sol_offsets(struct drm_device *dev,
 	return 0;
 }
 
-static int
-i915_emit_box(struct drm_i915_gem_request *req,
-	      struct drm_clip_rect *box,
-	      int DR1, int DR4)
-{
-	struct intel_engine_cs *ring = req->ring;
-	int ret;
-
-	if (box->y2 <= box->y1 || box->x2 <= box->x1 ||
-	    box->y2 <= 0 || box->x2 <= 0) {
-		DRM_ERROR("Bad box %d,%d..%d,%d\n",
-			  box->x1, box->y1, box->x2, box->y2);
-		return -EINVAL;
-	}
-
-	if (INTEL_INFO(ring->dev)->gen >= 4) {
-		ret = intel_ring_begin(req, 4);
-		if (ret)
-			return ret;
-
-		intel_ring_emit(ring, GFX_OP_DRAWRECT_INFO_I965);
-		intel_ring_emit(ring, (box->x1 & 0xffff) | box->y1 << 16);
-		intel_ring_emit(ring, ((box->x2 - 1) & 0xffff) | (box->y2 - 1) << 16);
-		intel_ring_emit(ring, DR4);
-	} else {
-		ret = intel_ring_begin(req, 6);
-		if (ret)
-			return ret;
-
-		intel_ring_emit(ring, GFX_OP_DRAWRECT_INFO);
-		intel_ring_emit(ring, DR1);
-		intel_ring_emit(ring, (box->x1 & 0xffff) | box->y1 << 16);
-		intel_ring_emit(ring, ((box->x2 - 1) & 0xffff) | (box->y2 - 1) << 16);
-		intel_ring_emit(ring, DR4);
-		intel_ring_emit(ring, 0);
-	}
-	intel_ring_advance(ring);
-
-	return 0;
-}
-
 static struct drm_i915_gem_object*
 i915_gem_execbuffer_parse(struct intel_engine_cs *ring,
 			  struct drm_i915_gem_exec_object2 *shadow_exec_entry,
@@ -1208,65 +1181,21 @@ i915_gem_ringbuffer_submission(struct i915_execbuffer_params *params,
 			       struct drm_i915_gem_execbuffer2 *args,
 			       struct list_head *vmas)
 {
-	struct drm_clip_rect *cliprects = NULL;
 	struct drm_device *dev = params->dev;
 	struct intel_engine_cs *ring = params->ring;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u64 exec_start, exec_len;
 	int instp_mode;
 	u32 instp_mask;
-	int i, ret = 0;
-
-	if (args->num_cliprects != 0) {
-		if (ring != &dev_priv->ring[RCS]) {
-			DRM_DEBUG("clip rectangles are only valid with the render ring\n");
-			return -EINVAL;
-		}
-
-		if (INTEL_INFO(dev)->gen >= 5) {
-			DRM_DEBUG("clip rectangles are only valid on pre-gen5\n");
-			return -EINVAL;
-		}
-
-		if (args->num_cliprects > UINT_MAX / sizeof(*cliprects)) {
-			DRM_DEBUG("execbuf with %u cliprects\n",
-				  args->num_cliprects);
-			return -EINVAL;
-		}
-
-		cliprects = kcalloc(args->num_cliprects,
-				    sizeof(*cliprects),
-				    GFP_KERNEL);
-		if (cliprects == NULL) {
-			ret = -ENOMEM;
-			goto error;
-		}
-
-		if (copy_from_user(cliprects,
-				   to_user_ptr(args->cliprects_ptr),
-				   sizeof(*cliprects)*args->num_cliprects)) {
-			ret = -EFAULT;
-			goto error;
-		}
-	} else {
-		if (args->DR4 == 0xffffffff) {
-			DRM_DEBUG("UXA submitting garbage DR4, fixing up\n");
-			args->DR4 = 0;
-		}
-
-		if (args->DR1 || args->DR4 || args->cliprects_ptr) {
-			DRM_DEBUG("0 cliprects but dirt in cliprects fields\n");
-			return -EINVAL;
-		}
-	}
+	int ret;
 
 	ret = i915_gem_execbuffer_move_to_gpu(params->request, vmas);
 	if (ret)
-		goto error;
+		return ret;
 
 	ret = i915_switch_context(params->request);
 	if (ret)
-		goto error;
+		return ret;
 
 	WARN(params->ctx->ppgtt && params->ctx->ppgtt->pd_dirty_rings & (1<<ring->id),
 	     "%s didn't clear reload\n", ring->name);
@@ -1279,22 +1208,19 @@ i915_gem_ringbuffer_submission(struct i915_execbuffer_params *params,
 	case I915_EXEC_CONSTANTS_REL_SURFACE:
 		if (instp_mode != 0 && ring != &dev_priv->ring[RCS]) {
 			DRM_DEBUG("non-0 rel constants mode on non-RCS\n");
-			ret = -EINVAL;
-			goto error;
+			return -EINVAL;
 		}
 
 		if (instp_mode != dev_priv->relative_constants_mode) {
 			if (INTEL_INFO(dev)->gen < 4) {
 				DRM_DEBUG("no rel constants on pre-gen4\n");
-				ret = -EINVAL;
-				goto error;
+				return -EINVAL;
 			}
 
 			if (INTEL_INFO(dev)->gen > 5 &&
 			    instp_mode == I915_EXEC_CONSTANTS_REL_SURFACE) {
 				DRM_DEBUG("rel surface constants mode invalid on gen5+\n");
-				ret = -EINVAL;
-				goto error;
+				return -EINVAL;
 			}
 
 			/* The HW changed the meaning on this bit on gen6 */
@@ -1304,15 +1230,14 @@ i915_gem_ringbuffer_submission(struct i915_execbuffer_params *params,
 		break;
 	default:
 		DRM_DEBUG("execbuf with unknown constants: %d\n", instp_mode);
-		ret = -EINVAL;
-		goto error;
+		return -EINVAL;
 	}
 
 	if (ring == &dev_priv->ring[RCS] &&
-			instp_mode != dev_priv->relative_constants_mode) {
+	    instp_mode != dev_priv->relative_constants_mode) {
 		ret = intel_ring_begin(params->request, 4);
 		if (ret)
-			goto error;
+			return ret;
 
 		intel_ring_emit(ring, MI_NOOP);
 		intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(1));
@@ -1326,42 +1251,25 @@ i915_gem_ringbuffer_submission(struct i915_execbuffer_params *params,
 	if (args->flags & I915_EXEC_GEN7_SOL_RESET) {
 		ret = i915_reset_gen7_sol_offsets(dev, params->request);
 		if (ret)
-			goto error;
+			return ret;
 	}
 
 	exec_len   = args->batch_len;
 	exec_start = params->batch_obj_vm_offset +
 		     params->args_batch_start_offset;
 
-	if (cliprects) {
-		for (i = 0; i < args->num_cliprects; i++) {
-			ret = i915_emit_box(params->request, &cliprects[i],
-					    args->DR1, args->DR4);
-			if (ret)
-				goto error;
-
-			ret = ring->dispatch_execbuffer(params->request,
-							exec_start, exec_len,
-							params->dispatch_flags);
-			if (ret)
-				goto error;
-		}
-	} else {
-		ret = ring->dispatch_execbuffer(params->request,
-						exec_start, exec_len,
-						params->dispatch_flags);
-		if (ret)
-			return ret;
-	}
+	ret = ring->dispatch_execbuffer(params->request,
+					exec_start, exec_len,
+					params->dispatch_flags);
+	if (ret)
+		return ret;
 
 	trace_i915_gem_ring_dispatch(params->request, params->dispatch_flags);
 
 	i915_gem_execbuffer_move_to_active(vmas, params->request);
 	i915_gem_execbuffer_retire_commands(params);
 
-error:
-	kfree(cliprects);
-	return ret;
+	return 0;
 }
 
 /**
diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
index 256167b..1bb1c9c 100644
--- a/drivers/gpu/drm/i915/intel_lrc.c
+++ b/drivers/gpu/drm/i915/intel_lrc.c
@@ -902,21 +902,6 @@ int intel_execlists_submission(struct i915_execbuffer_params *params,
 		return -EINVAL;
 	}
 
-	if (args->num_cliprects != 0) {
-		DRM_DEBUG("clip rectangles are only valid on pre-gen5\n");
-		return -EINVAL;
-	} else {
-		if (args->DR4 == 0xffffffff) {
-			DRM_DEBUG("UXA submitting garbage DR4, fixing up\n");
-			args->DR4 = 0;
-		}
-
-		if (args->DR1 || args->DR4 || args->cliprects_ptr) {
-			DRM_DEBUG("0 cliprects but dirt in cliprects fields\n");
-			return -EINVAL;
-		}
-	}
-
 	if (args->flags & I915_EXEC_GEN7_SOL_RESET) {
 		DRM_DEBUG("sol reset is gen7 only\n");
 		return -EINVAL;
-- 
cgit v0.10.2


From 26148bd3c0f1fbd8f2b0dae994f3195316f677db Mon Sep 17 00:00:00 2001
From: Akash Goel <akash.goel@intel.com>
Date: Fri, 18 Sep 2015 23:39:51 +0530
Subject: drm/i915/bxt: Set time interval unit to 0.833us

Note that in Bspec you have to dig around in a section called
"Timestamp bases" and Bspec update request is filed.

Signed-off-by: Ankitprasad Sharma <ankitprasad.r.sharma@intel.com>
Signed-off-by: Akash Goel <akash.goel@intel.com>
Signed-off-by: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
[danvet: Add note about state of Bspec.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 07588b6..6be853d 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -2840,8 +2840,11 @@ enum skl_disp_power_wells {
 
 #define INTERVAL_1_28_US(us)	(((us) * 100) >> 7)
 #define INTERVAL_1_33_US(us)	(((us) * 3)   >> 2)
+#define INTERVAL_0_833_US(us)	(((us) * 6) / 5)
 #define GT_INTERVAL_FROM_US(dev_priv, us) (IS_GEN9(dev_priv) ? \
-				INTERVAL_1_33_US(us) : \
+				(IS_BROXTON(dev_priv) ? \
+				INTERVAL_0_833_US(us) : \
+				INTERVAL_1_33_US(us)) : \
 				INTERVAL_1_28_US(us))
 
 /*
-- 
cgit v0.10.2


From b44f84081b8db1b5830cbd30280ba1109cc1a084 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 30 Sep 2015 16:46:48 +0300
Subject: drm: Stop using drm_vblank_count() as the hw frame counter
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

drm_vblank_count() returns the software counter. We should not pretend
it's the hw counter since we use the hw counter to figuere out what the
software counter value should be. So instead provide a new function
drm_vblank_no_hw_counter() for drivers that don't have a real hw
counter. The new function simply returns 0, which is about the only
thing it can do.

Cc: Vincent Abriou <vincent.abriou@st.com>
Cc: Thierry Reding <treding@nvidia.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Vincent Abriou <vincent.abriou@st.com>
[danvet: s/int pipe/unsigned int pipe/ to follow Thierry's interface
change.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/armada/armada_drv.c b/drivers/gpu/drm/armada/armada_drv.c
index a438886..f91a496 100644
--- a/drivers/gpu/drm/armada/armada_drv.c
+++ b/drivers/gpu/drm/armada/armada_drv.c
@@ -300,7 +300,7 @@ static struct drm_driver armada_drm_driver = {
 	.lastclose		= armada_drm_lastclose,
 	.unload			= armada_drm_unload,
 	.set_busid		= drm_platform_set_busid,
-	.get_vblank_counter	= drm_vblank_count,
+	.get_vblank_counter	= drm_vblank_no_hw_counter,
 	.enable_vblank		= armada_drm_enable_vblank,
 	.disable_vblank		= armada_drm_disable_vblank,
 #ifdef CONFIG_DEBUG_FS
diff --git a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c
index 6dfb63a..244df0a 100644
--- a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c
+++ b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c
@@ -699,7 +699,7 @@ static struct drm_driver atmel_hlcdc_dc_driver = {
 	.irq_preinstall = atmel_hlcdc_dc_irq_uninstall,
 	.irq_postinstall = atmel_hlcdc_dc_irq_postinstall,
 	.irq_uninstall = atmel_hlcdc_dc_irq_uninstall,
-	.get_vblank_counter = drm_vblank_count,
+	.get_vblank_counter = drm_vblank_no_hw_counter,
 	.enable_vblank = atmel_hlcdc_dc_enable_vblank,
 	.disable_vblank = atmel_hlcdc_dc_disable_vblank,
 	.gem_free_object = drm_gem_cma_free_object,
diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c
index 0659d99..7bdf247 100644
--- a/drivers/gpu/drm/drm_irq.c
+++ b/drivers/gpu/drm/drm_irq.c
@@ -1798,3 +1798,20 @@ bool drm_crtc_handle_vblank(struct drm_crtc *crtc)
 	return drm_handle_vblank(crtc->dev, drm_crtc_index(crtc));
 }
 EXPORT_SYMBOL(drm_crtc_handle_vblank);
+
+/**
+ * drm_vblank_no_hw_counter - "No hw counter" implementation of .get_vblank_counter()
+ * @dev: DRM device
+ * @pipe: CRTC for which to read the counter
+ *
+ * Drivers can plug this into the .get_vblank_counter() function if
+ * there is no useable hardware frame counter available.
+ *
+ * Returns:
+ * 0
+ */
+u32 drm_vblank_no_hw_counter(struct drm_device *dev, unsigned int pipe)
+{
+	return 0;
+}
+EXPORT_SYMBOL(drm_vblank_no_hw_counter);
diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.c b/drivers/gpu/drm/exynos/exynos_drm_drv.c
index f0a5839..fb9cfc5 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_drv.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_drv.c
@@ -447,7 +447,7 @@ static struct drm_driver exynos_drm_driver = {
 	.lastclose		= exynos_drm_lastclose,
 	.postclose		= exynos_drm_postclose,
 	.set_busid		= drm_platform_set_busid,
-	.get_vblank_counter	= drm_vblank_count,
+	.get_vblank_counter	= drm_vblank_no_hw_counter,
 	.enable_vblank		= exynos_drm_crtc_enable_vblank,
 	.disable_vblank		= exynos_drm_crtc_disable_vblank,
 	.gem_free_object	= exynos_drm_gem_free_object,
diff --git a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_drv.c b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_drv.c
index f1fd986..1930234 100644
--- a/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_drv.c
+++ b/drivers/gpu/drm/fsl-dcu/fsl_dcu_drm_drv.c
@@ -193,7 +193,7 @@ static struct drm_driver fsl_dcu_drm_driver = {
 	.unload			= fsl_dcu_unload,
 	.preclose		= fsl_dcu_drm_preclose,
 	.irq_handler		= fsl_dcu_drm_irq,
-	.get_vblank_counter	= drm_vblank_count,
+	.get_vblank_counter	= drm_vblank_no_hw_counter,
 	.enable_vblank		= fsl_dcu_drm_enable_vblank,
 	.disable_vblank		= fsl_dcu_drm_disable_vblank,
 	.gem_free_object	= drm_gem_cma_free_object,
diff --git a/drivers/gpu/drm/imx/imx-drm-core.c b/drivers/gpu/drm/imx/imx-drm-core.c
index b880c12..de00a6c 100644
--- a/drivers/gpu/drm/imx/imx-drm-core.c
+++ b/drivers/gpu/drm/imx/imx-drm-core.c
@@ -487,7 +487,7 @@ static struct drm_driver imx_drm_driver = {
 	.gem_prime_vmap		= drm_gem_cma_prime_vmap,
 	.gem_prime_vunmap	= drm_gem_cma_prime_vunmap,
 	.gem_prime_mmap		= drm_gem_cma_prime_mmap,
-	.get_vblank_counter	= drm_vblank_count,
+	.get_vblank_counter	= drm_vblank_no_hw_counter,
 	.enable_vblank		= imx_drm_enable_vblank,
 	.disable_vblank		= imx_drm_disable_vblank,
 	.ioctls			= imx_drm_ioctls,
diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c
index 7e44511..a06ec71 100644
--- a/drivers/gpu/drm/msm/msm_drv.c
+++ b/drivers/gpu/drm/msm/msm_drv.c
@@ -978,7 +978,7 @@ static struct drm_driver msm_driver = {
 	.irq_preinstall     = msm_irq_preinstall,
 	.irq_postinstall    = msm_irq_postinstall,
 	.irq_uninstall      = msm_irq_uninstall,
-	.get_vblank_counter = drm_vblank_count,
+	.get_vblank_counter = drm_vblank_no_hw_counter,
 	.enable_vblank      = msm_enable_vblank,
 	.disable_vblank     = msm_disable_vblank,
 	.gem_free_object    = msm_gem_free_object,
diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c
index ccefb64..2416c7d 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_drm.c
@@ -934,7 +934,7 @@ driver_stub = {
 	.debugfs_cleanup = nouveau_debugfs_takedown,
 #endif
 
-	.get_vblank_counter = drm_vblank_count,
+	.get_vblank_counter = drm_vblank_no_hw_counter,
 	.enable_vblank = nouveau_display_vblank_enable,
 	.disable_vblank = nouveau_display_vblank_disable,
 	.get_scanout_position = nouveau_display_scanoutpos,
diff --git a/drivers/gpu/drm/omapdrm/omap_drv.c b/drivers/gpu/drm/omapdrm/omap_drv.c
index d685e23..4d58934 100644
--- a/drivers/gpu/drm/omapdrm/omap_drv.c
+++ b/drivers/gpu/drm/omapdrm/omap_drv.c
@@ -839,7 +839,7 @@ static struct drm_driver omap_drm_driver = {
 	.preclose = dev_preclose,
 	.postclose = dev_postclose,
 	.set_busid = drm_platform_set_busid,
-	.get_vblank_counter = drm_vblank_count,
+	.get_vblank_counter = drm_vblank_no_hw_counter,
 	.enable_vblank = omap_irq_enable_vblank,
 	.disable_vblank = omap_irq_disable_vblank,
 #ifdef CONFIG_DEBUG_FS
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_drv.c b/drivers/gpu/drm/rcar-du/rcar_du_drv.c
index bb806c4..feddda0 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_drv.c
+++ b/drivers/gpu/drm/rcar-du/rcar_du_drv.c
@@ -259,7 +259,7 @@ static struct drm_driver rcar_du_driver = {
 	.preclose		= rcar_du_preclose,
 	.lastclose		= rcar_du_lastclose,
 	.set_busid		= drm_platform_set_busid,
-	.get_vblank_counter	= drm_vblank_count,
+	.get_vblank_counter	= drm_vblank_no_hw_counter,
 	.enable_vblank		= rcar_du_enable_vblank,
 	.disable_vblank		= rcar_du_disable_vblank,
 	.gem_free_object	= drm_gem_cma_free_object,
diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_drv.c b/drivers/gpu/drm/rockchip/rockchip_drm_drv.c
index 32c6098..f22e1e1 100644
--- a/drivers/gpu/drm/rockchip/rockchip_drm_drv.c
+++ b/drivers/gpu/drm/rockchip/rockchip_drm_drv.c
@@ -279,7 +279,7 @@ static struct drm_driver rockchip_drm_driver = {
 	.load			= rockchip_drm_load,
 	.unload			= rockchip_drm_unload,
 	.lastclose		= rockchip_drm_lastclose,
-	.get_vblank_counter	= drm_vblank_count,
+	.get_vblank_counter	= drm_vblank_no_hw_counter,
 	.enable_vblank		= rockchip_drm_crtc_enable_vblank,
 	.disable_vblank		= rockchip_drm_crtc_disable_vblank,
 	.gem_vm_ops		= &rockchip_drm_vm_ops,
diff --git a/drivers/gpu/drm/shmobile/shmob_drm_drv.c b/drivers/gpu/drm/shmobile/shmob_drm_drv.c
index ca2f918..04e66e3 100644
--- a/drivers/gpu/drm/shmobile/shmob_drm_drv.c
+++ b/drivers/gpu/drm/shmobile/shmob_drm_drv.c
@@ -269,7 +269,7 @@ static struct drm_driver shmob_drm_driver = {
 	.preclose		= shmob_drm_preclose,
 	.set_busid		= drm_platform_set_busid,
 	.irq_handler		= shmob_drm_irq,
-	.get_vblank_counter	= drm_vblank_count,
+	.get_vblank_counter	= drm_vblank_no_hw_counter,
 	.enable_vblank		= shmob_drm_enable_vblank,
 	.disable_vblank		= shmob_drm_disable_vblank,
 	.gem_free_object	= drm_gem_cma_free_object,
diff --git a/drivers/gpu/drm/sti/sti_drv.c b/drivers/gpu/drm/sti/sti_drv.c
index 9f85988..f846996 100644
--- a/drivers/gpu/drm/sti/sti_drv.c
+++ b/drivers/gpu/drm/sti/sti_drv.c
@@ -201,7 +201,7 @@ static struct drm_driver sti_driver = {
 	.dumb_destroy = drm_gem_dumb_destroy,
 	.fops = &sti_driver_fops,
 
-	.get_vblank_counter = drm_vblank_count,
+	.get_vblank_counter = drm_vblank_no_hw_counter,
 	.enable_vblank = sti_crtc_enable_vblank,
 	.disable_vblank = sti_crtc_disable_vblank,
 
diff --git a/drivers/gpu/drm/tilcdc/tilcdc_drv.c b/drivers/gpu/drm/tilcdc/tilcdc_drv.c
index a5b8f5d..876cad5 100644
--- a/drivers/gpu/drm/tilcdc/tilcdc_drv.c
+++ b/drivers/gpu/drm/tilcdc/tilcdc_drv.c
@@ -563,7 +563,7 @@ static struct drm_driver tilcdc_driver = {
 	.irq_preinstall     = tilcdc_irq_preinstall,
 	.irq_postinstall    = tilcdc_irq_postinstall,
 	.irq_uninstall      = tilcdc_irq_uninstall,
-	.get_vblank_counter = drm_vblank_count,
+	.get_vblank_counter = drm_vblank_no_hw_counter,
 	.enable_vblank      = tilcdc_enable_vblank,
 	.disable_vblank     = tilcdc_disable_vblank,
 	.gem_free_object    = drm_gem_cma_free_object,
diff --git a/include/drm/drmP.h b/include/drm/drmP.h
index 1cb1e84..3dc56d3 100644
--- a/include/drm/drmP.h
+++ b/include/drm/drmP.h
@@ -953,6 +953,7 @@ extern void drm_crtc_vblank_off(struct drm_crtc *crtc);
 extern void drm_crtc_vblank_reset(struct drm_crtc *crtc);
 extern void drm_crtc_vblank_on(struct drm_crtc *crtc);
 extern void drm_vblank_cleanup(struct drm_device *dev);
+extern u32 drm_vblank_no_hw_counter(struct drm_device *dev, unsigned int pipe);
 
 extern int drm_calc_vbltimestamp_from_scanoutpos(struct drm_device *dev,
 						 unsigned int pipe, int *max_error,
-- 
cgit v0.10.2


From d39398f5374e2ae35832ad8e01e60e8deb6d4e1c Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Wed, 7 Oct 2015 11:17:44 +0300
Subject: drm/i915/snb: remove pre-production hardware workaround
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 52642af..1e67484 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -4558,22 +4558,6 @@ void i915_gem_init_swizzling(struct drm_device *dev)
 		BUG();
 }
 
-static bool
-intel_enable_blt(struct drm_device *dev)
-{
-	if (!HAS_BLT(dev))
-		return false;
-
-	/* The blitter was dysfunctional on early prototypes */
-	if (IS_GEN6(dev) && dev->pdev->revision < 8) {
-		DRM_INFO("BLT not supported on this pre-production hardware;"
-			 " graphics performance will be degraded.\n");
-		return false;
-	}
-
-	return true;
-}
-
 static void init_unused_ring(struct drm_device *dev, u32 base)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -4616,7 +4600,7 @@ int i915_gem_init_rings(struct drm_device *dev)
 			goto cleanup_render_ring;
 	}
 
-	if (intel_enable_blt(dev)) {
+	if (HAS_BLT(dev)) {
 		ret = intel_init_blt_ring_buffer(dev);
 		if (ret)
 			goto cleanup_bsd_ring;
-- 
cgit v0.10.2


From 5b5929cbe3f77e1a01cd2709c8dd7fe0302af1b6 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Wed, 7 Oct 2015 11:17:46 +0300
Subject: drm/i915/chv: remove pre-production hardware workarounds
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
[danvet: Appease gcc and remove the unused variable.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index 60d120c..3f9b3c0 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -5061,32 +5061,27 @@ static int cherryview_rps_max_freq(struct drm_i915_private *dev_priv)
 	struct drm_device *dev = dev_priv->dev;
 	u32 val, rp0;
 
-	if (dev->pdev->revision >= 0x20) {
-		val = vlv_punit_read(dev_priv, FB_GFX_FMAX_AT_VMAX_FUSE);
+	val = vlv_punit_read(dev_priv, FB_GFX_FMAX_AT_VMAX_FUSE);
 
-		switch (INTEL_INFO(dev)->eu_total) {
-		case 8:
-				/* (2 * 4) config */
-				rp0 = (val >> FB_GFX_FMAX_AT_VMAX_2SS4EU_FUSE_SHIFT);
-				break;
-		case 12:
-				/* (2 * 6) config */
-				rp0 = (val >> FB_GFX_FMAX_AT_VMAX_2SS6EU_FUSE_SHIFT);
-				break;
-		case 16:
-				/* (2 * 8) config */
-		default:
-				/* Setting (2 * 8) Min RP0 for any other combination */
-				rp0 = (val >> FB_GFX_FMAX_AT_VMAX_2SS8EU_FUSE_SHIFT);
-				break;
-		}
-		rp0 = (rp0 & FB_GFX_FREQ_FUSE_MASK);
-	} else {
-		/* For pre-production hardware */
-		val = vlv_punit_read(dev_priv, PUNIT_GPU_STATUS_REG);
-		rp0 = (val >> PUNIT_GPU_STATUS_MAX_FREQ_SHIFT) &
-		       PUNIT_GPU_STATUS_MAX_FREQ_MASK;
+	switch (INTEL_INFO(dev)->eu_total) {
+	case 8:
+		/* (2 * 4) config */
+		rp0 = (val >> FB_GFX_FMAX_AT_VMAX_2SS4EU_FUSE_SHIFT);
+		break;
+	case 12:
+		/* (2 * 6) config */
+		rp0 = (val >> FB_GFX_FMAX_AT_VMAX_2SS6EU_FUSE_SHIFT);
+		break;
+	case 16:
+		/* (2 * 8) config */
+	default:
+		/* Setting (2 * 8) Min RP0 for any other combination */
+		rp0 = (val >> FB_GFX_FMAX_AT_VMAX_2SS8EU_FUSE_SHIFT);
+		break;
 	}
+
+	rp0 = (rp0 & FB_GFX_FREQ_FUSE_MASK);
+
 	return rp0;
 }
 
@@ -5102,18 +5097,11 @@ static int cherryview_rps_rpe_freq(struct drm_i915_private *dev_priv)
 
 static int cherryview_rps_guar_freq(struct drm_i915_private *dev_priv)
 {
-	struct drm_device *dev = dev_priv->dev;
 	u32 val, rp1;
 
-	if (dev->pdev->revision >= 0x20) {
-		val = vlv_punit_read(dev_priv, FB_GFX_FMAX_AT_VMAX_FUSE);
-		rp1 = (val & FB_GFX_FREQ_FUSE_MASK);
-	} else {
-		/* For pre-production hardware */
-		val = vlv_punit_read(dev_priv, PUNIT_REG_GPU_FREQ_STS);
-		rp1 = ((val >> PUNIT_GPU_STATUS_MAX_FREQ_SHIFT) &
-		       PUNIT_GPU_STATUS_MAX_FREQ_MASK);
-	}
+	val = vlv_punit_read(dev_priv, FB_GFX_FMAX_AT_VMAX_FUSE);
+	rp1 = (val & FB_GFX_FREQ_FUSE_MASK);
+
 	return rp1;
 }
 
-- 
cgit v0.10.2


From 6a8aadc43974729349cb6b823d774c091afd78cd Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Thu, 1 Oct 2015 18:06:50 +0200
Subject: drm/i915: Resurrect golden context on gen6/7

In

commit 8f0e2b9d95a88ca5d8349deef2375644faf184ae
Author: Daniel Vetter <daniel.vetter@ffwll.ch>
Date:   Tue Dec 2 16:19:07 2014 +0100

    drm/i915: Move golden context init into ->init_context

I've shuffled around per-ctx init code a bit for legacy contexts but
accidentally dropped the render state init call on gen6/7. Resurrect
it.

Reported-by: Francisco Jerez <currojerez@riseup.net>
Cc: Francisco Jerez <currojerez@riseup.net>
Cc: Dave Gordon <david.s.gordon@intel.com>
Cc: Thomas Daniel <thomas.daniel@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index c82c74c..9035f8c 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -2623,6 +2623,7 @@ int intel_init_render_ring_buffer(struct drm_device *dev)
 			GEN8_RING_SEMAPHORE_INIT;
 		}
 	} else if (INTEL_INFO(dev)->gen >= 6) {
+		ring->init_context = i915_gem_render_state_init;
 		ring->add_request = gen6_add_request;
 		ring->flush = gen7_render_ring_flush;
 		if (INTEL_INFO(dev)->gen == 6)
-- 
cgit v0.10.2


From 02235808b61cd9382d224b0df263193006dd9913 Mon Sep 17 00:00:00 2001
From: Francisco Jerez <currojerez@riseup.net>
Date: Wed, 7 Oct 2015 14:44:01 +0300
Subject: drm/i915: Don't warn if the workaround list is empty.

It's not an error for the workaround list to be empty if no
workarounds are needed.  This will avoid spamming the logs
unnecessarily on Gen6 after the workaround list is hooked up on
pre-Gen8 hardware by the following commits.

Signed-off-by: Francisco Jerez <currojerez@riseup.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index 9035f8c..702d3a2 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -717,7 +717,7 @@ static int intel_ring_workarounds_emit(struct drm_i915_gem_request *req)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct i915_workarounds *w = &dev_priv->workarounds;
 
-	if (WARN_ON_ONCE(w->count == 0))
+	if (w->count == 0)
 		return 0;
 
 	ring->gpu_caches_dirty = true;
-- 
cgit v0.10.2


From 4f91fc6d2c1d98b3da7a8341ef5ec66ccf2916b9 Mon Sep 17 00:00:00 2001
From: Francisco Jerez <currojerez@riseup.net>
Date: Wed, 7 Oct 2015 14:44:02 +0300
Subject: drm/i915: Hook up ring workaround writes at context creation time on
 Gen6-7.

intel_rcs_ctx_init() emits all workaround register writes on the list
to the ring, in addition to calling i915_gem_render_state_init().  The
workaround list is currently empty on Gen6-7 so this shouldn't cause
any functional changes.

Signed-off-by: Francisco Jerez <currojerez@riseup.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index 702d3a2..654ae99 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -2623,7 +2623,7 @@ int intel_init_render_ring_buffer(struct drm_device *dev)
 			GEN8_RING_SEMAPHORE_INIT;
 		}
 	} else if (INTEL_INFO(dev)->gen >= 6) {
-		ring->init_context = i915_gem_render_state_init;
+		ring->init_context = intel_rcs_ctx_init;
 		ring->add_request = gen6_add_request;
 		ring->flush = gen7_render_ring_flush;
 		if (INTEL_INFO(dev)->gen == 6)
-- 
cgit v0.10.2


From 0bde3a95eaa90f130dbb71a846243aa6f0e678cf Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Thu, 3 Sep 2015 17:58:06 -0400
Subject: drm/amdgpu: split gfx8 gpu init into sw and hw parts
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Calculate the driver state in sw_init and program the
registers in hw init.

Acked-by: Leo Liu <leo.liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index cb4f68f..718250a 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -903,6 +903,191 @@ static int gfx_v8_0_mec_init(struct amdgpu_device *adev)
 	return 0;
 }
 
+static void gfx_v8_0_gpu_early_init(struct amdgpu_device *adev)
+{
+	u32 gb_addr_config;
+	u32 mc_shared_chmap, mc_arb_ramcfg;
+	u32 dimm00_addr_map, dimm01_addr_map, dimm10_addr_map, dimm11_addr_map;
+	u32 tmp;
+
+	switch (adev->asic_type) {
+	case CHIP_TOPAZ:
+		adev->gfx.config.max_shader_engines = 1;
+		adev->gfx.config.max_tile_pipes = 2;
+		adev->gfx.config.max_cu_per_sh = 6;
+		adev->gfx.config.max_sh_per_se = 1;
+		adev->gfx.config.max_backends_per_se = 2;
+		adev->gfx.config.max_texture_channel_caches = 2;
+		adev->gfx.config.max_gprs = 256;
+		adev->gfx.config.max_gs_threads = 32;
+		adev->gfx.config.max_hw_contexts = 8;
+
+		adev->gfx.config.sc_prim_fifo_size_frontend = 0x20;
+		adev->gfx.config.sc_prim_fifo_size_backend = 0x100;
+		adev->gfx.config.sc_hiz_tile_fifo_size = 0x30;
+		adev->gfx.config.sc_earlyz_tile_fifo_size = 0x130;
+		gb_addr_config = TOPAZ_GB_ADDR_CONFIG_GOLDEN;
+		break;
+	case CHIP_FIJI:
+		adev->gfx.config.max_shader_engines = 4;
+		adev->gfx.config.max_tile_pipes = 16;
+		adev->gfx.config.max_cu_per_sh = 16;
+		adev->gfx.config.max_sh_per_se = 1;
+		adev->gfx.config.max_backends_per_se = 4;
+		adev->gfx.config.max_texture_channel_caches = 8;
+		adev->gfx.config.max_gprs = 256;
+		adev->gfx.config.max_gs_threads = 32;
+		adev->gfx.config.max_hw_contexts = 8;
+
+		adev->gfx.config.sc_prim_fifo_size_frontend = 0x20;
+		adev->gfx.config.sc_prim_fifo_size_backend = 0x100;
+		adev->gfx.config.sc_hiz_tile_fifo_size = 0x30;
+		adev->gfx.config.sc_earlyz_tile_fifo_size = 0x130;
+		gb_addr_config = TONGA_GB_ADDR_CONFIG_GOLDEN;
+		break;
+	case CHIP_TONGA:
+		adev->gfx.config.max_shader_engines = 4;
+		adev->gfx.config.max_tile_pipes = 8;
+		adev->gfx.config.max_cu_per_sh = 8;
+		adev->gfx.config.max_sh_per_se = 1;
+		adev->gfx.config.max_backends_per_se = 2;
+		adev->gfx.config.max_texture_channel_caches = 8;
+		adev->gfx.config.max_gprs = 256;
+		adev->gfx.config.max_gs_threads = 32;
+		adev->gfx.config.max_hw_contexts = 8;
+
+		adev->gfx.config.sc_prim_fifo_size_frontend = 0x20;
+		adev->gfx.config.sc_prim_fifo_size_backend = 0x100;
+		adev->gfx.config.sc_hiz_tile_fifo_size = 0x30;
+		adev->gfx.config.sc_earlyz_tile_fifo_size = 0x130;
+		gb_addr_config = TONGA_GB_ADDR_CONFIG_GOLDEN;
+		break;
+	case CHIP_CARRIZO:
+		adev->gfx.config.max_shader_engines = 1;
+		adev->gfx.config.max_tile_pipes = 2;
+		adev->gfx.config.max_sh_per_se = 1;
+		adev->gfx.config.max_backends_per_se = 2;
+
+		switch (adev->pdev->revision) {
+		case 0xc4:
+		case 0x84:
+		case 0xc8:
+		case 0xcc:
+			/* B10 */
+			adev->gfx.config.max_cu_per_sh = 8;
+			break;
+		case 0xc5:
+		case 0x81:
+		case 0x85:
+		case 0xc9:
+		case 0xcd:
+			/* B8 */
+			adev->gfx.config.max_cu_per_sh = 6;
+			break;
+		case 0xc6:
+		case 0xca:
+		case 0xce:
+			/* B6 */
+			adev->gfx.config.max_cu_per_sh = 6;
+			break;
+		case 0xc7:
+		case 0x87:
+		case 0xcb:
+		default:
+			/* B4 */
+			adev->gfx.config.max_cu_per_sh = 4;
+			break;
+		}
+
+		adev->gfx.config.max_texture_channel_caches = 2;
+		adev->gfx.config.max_gprs = 256;
+		adev->gfx.config.max_gs_threads = 32;
+		adev->gfx.config.max_hw_contexts = 8;
+
+		adev->gfx.config.sc_prim_fifo_size_frontend = 0x20;
+		adev->gfx.config.sc_prim_fifo_size_backend = 0x100;
+		adev->gfx.config.sc_hiz_tile_fifo_size = 0x30;
+		adev->gfx.config.sc_earlyz_tile_fifo_size = 0x130;
+		gb_addr_config = CARRIZO_GB_ADDR_CONFIG_GOLDEN;
+		break;
+	default:
+		adev->gfx.config.max_shader_engines = 2;
+		adev->gfx.config.max_tile_pipes = 4;
+		adev->gfx.config.max_cu_per_sh = 2;
+		adev->gfx.config.max_sh_per_se = 1;
+		adev->gfx.config.max_backends_per_se = 2;
+		adev->gfx.config.max_texture_channel_caches = 4;
+		adev->gfx.config.max_gprs = 256;
+		adev->gfx.config.max_gs_threads = 32;
+		adev->gfx.config.max_hw_contexts = 8;
+
+		adev->gfx.config.sc_prim_fifo_size_frontend = 0x20;
+		adev->gfx.config.sc_prim_fifo_size_backend = 0x100;
+		adev->gfx.config.sc_hiz_tile_fifo_size = 0x30;
+		adev->gfx.config.sc_earlyz_tile_fifo_size = 0x130;
+		gb_addr_config = TONGA_GB_ADDR_CONFIG_GOLDEN;
+		break;
+	}
+
+	mc_shared_chmap = RREG32(mmMC_SHARED_CHMAP);
+	adev->gfx.config.mc_arb_ramcfg = RREG32(mmMC_ARB_RAMCFG);
+	mc_arb_ramcfg = adev->gfx.config.mc_arb_ramcfg;
+
+	adev->gfx.config.num_tile_pipes = adev->gfx.config.max_tile_pipes;
+	adev->gfx.config.mem_max_burst_length_bytes = 256;
+	if (adev->flags & AMD_IS_APU) {
+		/* Get memory bank mapping mode. */
+		tmp = RREG32(mmMC_FUS_DRAM0_BANK_ADDR_MAPPING);
+		dimm00_addr_map = REG_GET_FIELD(tmp, MC_FUS_DRAM0_BANK_ADDR_MAPPING, DIMM0ADDRMAP);
+		dimm01_addr_map = REG_GET_FIELD(tmp, MC_FUS_DRAM0_BANK_ADDR_MAPPING, DIMM1ADDRMAP);
+
+		tmp = RREG32(mmMC_FUS_DRAM1_BANK_ADDR_MAPPING);
+		dimm10_addr_map = REG_GET_FIELD(tmp, MC_FUS_DRAM1_BANK_ADDR_MAPPING, DIMM0ADDRMAP);
+		dimm11_addr_map = REG_GET_FIELD(tmp, MC_FUS_DRAM1_BANK_ADDR_MAPPING, DIMM1ADDRMAP);
+
+		/* Validate settings in case only one DIMM installed. */
+		if ((dimm00_addr_map == 0) || (dimm00_addr_map == 3) || (dimm00_addr_map == 4) || (dimm00_addr_map > 12))
+			dimm00_addr_map = 0;
+		if ((dimm01_addr_map == 0) || (dimm01_addr_map == 3) || (dimm01_addr_map == 4) || (dimm01_addr_map > 12))
+			dimm01_addr_map = 0;
+		if ((dimm10_addr_map == 0) || (dimm10_addr_map == 3) || (dimm10_addr_map == 4) || (dimm10_addr_map > 12))
+			dimm10_addr_map = 0;
+		if ((dimm11_addr_map == 0) || (dimm11_addr_map == 3) || (dimm11_addr_map == 4) || (dimm11_addr_map > 12))
+			dimm11_addr_map = 0;
+
+		/* If DIMM Addr map is 8GB, ROW size should be 2KB. Otherwise 1KB. */
+		/* If ROW size(DIMM1) != ROW size(DMIMM0), ROW size should be larger one. */
+		if ((dimm00_addr_map == 11) || (dimm01_addr_map == 11) || (dimm10_addr_map == 11) || (dimm11_addr_map == 11))
+			adev->gfx.config.mem_row_size_in_kb = 2;
+		else
+			adev->gfx.config.mem_row_size_in_kb = 1;
+	} else {
+		tmp = REG_GET_FIELD(mc_arb_ramcfg, MC_ARB_RAMCFG, NOOFCOLS);
+		adev->gfx.config.mem_row_size_in_kb = (4 * (1 << (8 + tmp))) / 1024;
+		if (adev->gfx.config.mem_row_size_in_kb > 4)
+			adev->gfx.config.mem_row_size_in_kb = 4;
+	}
+
+	adev->gfx.config.shader_engine_tile_size = 32;
+	adev->gfx.config.num_gpus = 1;
+	adev->gfx.config.multi_gpu_tile_size = 64;
+
+	/* fix up row size */
+	switch (adev->gfx.config.mem_row_size_in_kb) {
+	case 1:
+	default:
+		gb_addr_config = REG_SET_FIELD(gb_addr_config, GB_ADDR_CONFIG, ROW_SIZE, 0);
+		break;
+	case 2:
+		gb_addr_config = REG_SET_FIELD(gb_addr_config, GB_ADDR_CONFIG, ROW_SIZE, 1);
+		break;
+	case 4:
+		gb_addr_config = REG_SET_FIELD(gb_addr_config, GB_ADDR_CONFIG, ROW_SIZE, 2);
+		break;
+	}
+	adev->gfx.config.gb_addr_config = gb_addr_config;
+}
+
 static int gfx_v8_0_sw_init(void *handle)
 {
 	int i, r;
@@ -1010,6 +1195,8 @@ static int gfx_v8_0_sw_init(void *handle)
 
 	adev->gfx.ce_ram_size = 0x8000;
 
+	gfx_v8_0_gpu_early_init(adev);
+
 	return 0;
 }
 
@@ -2043,203 +2230,23 @@ static void gfx_v8_0_init_compute_vmid(struct amdgpu_device *adev)
 
 static void gfx_v8_0_gpu_init(struct amdgpu_device *adev)
 {
-	u32 gb_addr_config;
-	u32 mc_shared_chmap, mc_arb_ramcfg;
-	u32 dimm00_addr_map, dimm01_addr_map, dimm10_addr_map, dimm11_addr_map;
 	u32 tmp;
 	int i;
 
-	switch (adev->asic_type) {
-	case CHIP_TOPAZ:
-		adev->gfx.config.max_shader_engines = 1;
-		adev->gfx.config.max_tile_pipes = 2;
-		adev->gfx.config.max_cu_per_sh = 6;
-		adev->gfx.config.max_sh_per_se = 1;
-		adev->gfx.config.max_backends_per_se = 2;
-		adev->gfx.config.max_texture_channel_caches = 2;
-		adev->gfx.config.max_gprs = 256;
-		adev->gfx.config.max_gs_threads = 32;
-		adev->gfx.config.max_hw_contexts = 8;
-
-		adev->gfx.config.sc_prim_fifo_size_frontend = 0x20;
-		adev->gfx.config.sc_prim_fifo_size_backend = 0x100;
-		adev->gfx.config.sc_hiz_tile_fifo_size = 0x30;
-		adev->gfx.config.sc_earlyz_tile_fifo_size = 0x130;
-		gb_addr_config = TOPAZ_GB_ADDR_CONFIG_GOLDEN;
-		break;
-	case CHIP_FIJI:
-		adev->gfx.config.max_shader_engines = 4;
-		adev->gfx.config.max_tile_pipes = 16;
-		adev->gfx.config.max_cu_per_sh = 16;
-		adev->gfx.config.max_sh_per_se = 1;
-		adev->gfx.config.max_backends_per_se = 4;
-		adev->gfx.config.max_texture_channel_caches = 8;
-		adev->gfx.config.max_gprs = 256;
-		adev->gfx.config.max_gs_threads = 32;
-		adev->gfx.config.max_hw_contexts = 8;
-
-		adev->gfx.config.sc_prim_fifo_size_frontend = 0x20;
-		adev->gfx.config.sc_prim_fifo_size_backend = 0x100;
-		adev->gfx.config.sc_hiz_tile_fifo_size = 0x30;
-		adev->gfx.config.sc_earlyz_tile_fifo_size = 0x130;
-		gb_addr_config = TONGA_GB_ADDR_CONFIG_GOLDEN;
-		break;
-	case CHIP_TONGA:
-		adev->gfx.config.max_shader_engines = 4;
-		adev->gfx.config.max_tile_pipes = 8;
-		adev->gfx.config.max_cu_per_sh = 8;
-		adev->gfx.config.max_sh_per_se = 1;
-		adev->gfx.config.max_backends_per_se = 2;
-		adev->gfx.config.max_texture_channel_caches = 8;
-		adev->gfx.config.max_gprs = 256;
-		adev->gfx.config.max_gs_threads = 32;
-		adev->gfx.config.max_hw_contexts = 8;
-
-		adev->gfx.config.sc_prim_fifo_size_frontend = 0x20;
-		adev->gfx.config.sc_prim_fifo_size_backend = 0x100;
-		adev->gfx.config.sc_hiz_tile_fifo_size = 0x30;
-		adev->gfx.config.sc_earlyz_tile_fifo_size = 0x130;
-		gb_addr_config = TONGA_GB_ADDR_CONFIG_GOLDEN;
-		break;
-	case CHIP_CARRIZO:
-		adev->gfx.config.max_shader_engines = 1;
-		adev->gfx.config.max_tile_pipes = 2;
-		adev->gfx.config.max_sh_per_se = 1;
-		adev->gfx.config.max_backends_per_se = 2;
-
-		switch (adev->pdev->revision) {
-		case 0xc4:
-		case 0x84:
-		case 0xc8:
-		case 0xcc:
-			/* B10 */
-			adev->gfx.config.max_cu_per_sh = 8;
-			break;
-		case 0xc5:
-		case 0x81:
-		case 0x85:
-		case 0xc9:
-		case 0xcd:
-			/* B8 */
-			adev->gfx.config.max_cu_per_sh = 6;
-			break;
-		case 0xc6:
-		case 0xca:
-		case 0xce:
-			/* B6 */
-			adev->gfx.config.max_cu_per_sh = 6;
-			break;
-		case 0xc7:
-		case 0x87:
-		case 0xcb:
-		default:
-			/* B4 */
-			adev->gfx.config.max_cu_per_sh = 4;
-			break;
-		}
-
-		adev->gfx.config.max_texture_channel_caches = 2;
-		adev->gfx.config.max_gprs = 256;
-		adev->gfx.config.max_gs_threads = 32;
-		adev->gfx.config.max_hw_contexts = 8;
-
-		adev->gfx.config.sc_prim_fifo_size_frontend = 0x20;
-		adev->gfx.config.sc_prim_fifo_size_backend = 0x100;
-		adev->gfx.config.sc_hiz_tile_fifo_size = 0x30;
-		adev->gfx.config.sc_earlyz_tile_fifo_size = 0x130;
-		gb_addr_config = CARRIZO_GB_ADDR_CONFIG_GOLDEN;
-		break;
-	default:
-		adev->gfx.config.max_shader_engines = 2;
-		adev->gfx.config.max_tile_pipes = 4;
-		adev->gfx.config.max_cu_per_sh = 2;
-		adev->gfx.config.max_sh_per_se = 1;
-		adev->gfx.config.max_backends_per_se = 2;
-		adev->gfx.config.max_texture_channel_caches = 4;
-		adev->gfx.config.max_gprs = 256;
-		adev->gfx.config.max_gs_threads = 32;
-		adev->gfx.config.max_hw_contexts = 8;
-
-		adev->gfx.config.sc_prim_fifo_size_frontend = 0x20;
-		adev->gfx.config.sc_prim_fifo_size_backend = 0x100;
-		adev->gfx.config.sc_hiz_tile_fifo_size = 0x30;
-		adev->gfx.config.sc_earlyz_tile_fifo_size = 0x130;
-		gb_addr_config = TONGA_GB_ADDR_CONFIG_GOLDEN;
-		break;
-	}
-
 	tmp = RREG32(mmGRBM_CNTL);
 	tmp = REG_SET_FIELD(tmp, GRBM_CNTL, READ_TIMEOUT, 0xff);
 	WREG32(mmGRBM_CNTL, tmp);
 
-	mc_shared_chmap = RREG32(mmMC_SHARED_CHMAP);
-	adev->gfx.config.mc_arb_ramcfg = RREG32(mmMC_ARB_RAMCFG);
-	mc_arb_ramcfg = adev->gfx.config.mc_arb_ramcfg;
-
-	adev->gfx.config.num_tile_pipes = adev->gfx.config.max_tile_pipes;
-	adev->gfx.config.mem_max_burst_length_bytes = 256;
-	if (adev->flags & AMD_IS_APU) {
-		/* Get memory bank mapping mode. */
-		tmp = RREG32(mmMC_FUS_DRAM0_BANK_ADDR_MAPPING);
-		dimm00_addr_map = REG_GET_FIELD(tmp, MC_FUS_DRAM0_BANK_ADDR_MAPPING, DIMM0ADDRMAP);
-		dimm01_addr_map = REG_GET_FIELD(tmp, MC_FUS_DRAM0_BANK_ADDR_MAPPING, DIMM1ADDRMAP);
-
-		tmp = RREG32(mmMC_FUS_DRAM1_BANK_ADDR_MAPPING);
-		dimm10_addr_map = REG_GET_FIELD(tmp, MC_FUS_DRAM1_BANK_ADDR_MAPPING, DIMM0ADDRMAP);
-		dimm11_addr_map = REG_GET_FIELD(tmp, MC_FUS_DRAM1_BANK_ADDR_MAPPING, DIMM1ADDRMAP);
-
-		/* Validate settings in case only one DIMM installed. */
-		if ((dimm00_addr_map == 0) || (dimm00_addr_map == 3) || (dimm00_addr_map == 4) || (dimm00_addr_map > 12))
-			dimm00_addr_map = 0;
-		if ((dimm01_addr_map == 0) || (dimm01_addr_map == 3) || (dimm01_addr_map == 4) || (dimm01_addr_map > 12))
-			dimm01_addr_map = 0;
-		if ((dimm10_addr_map == 0) || (dimm10_addr_map == 3) || (dimm10_addr_map == 4) || (dimm10_addr_map > 12))
-			dimm10_addr_map = 0;
-		if ((dimm11_addr_map == 0) || (dimm11_addr_map == 3) || (dimm11_addr_map == 4) || (dimm11_addr_map > 12))
-			dimm11_addr_map = 0;
-
-		/* If DIMM Addr map is 8GB, ROW size should be 2KB. Otherwise 1KB. */
-		/* If ROW size(DIMM1) != ROW size(DMIMM0), ROW size should be larger one. */
-		if ((dimm00_addr_map == 11) || (dimm01_addr_map == 11) || (dimm10_addr_map == 11) || (dimm11_addr_map == 11))
-			adev->gfx.config.mem_row_size_in_kb = 2;
-		else
-			adev->gfx.config.mem_row_size_in_kb = 1;
-	} else {
-		tmp = REG_GET_FIELD(mc_arb_ramcfg, MC_ARB_RAMCFG, NOOFCOLS);
-		adev->gfx.config.mem_row_size_in_kb = (4 * (1 << (8 + tmp))) / 1024;
-		if (adev->gfx.config.mem_row_size_in_kb > 4)
-			adev->gfx.config.mem_row_size_in_kb = 4;
-	}
-
-	adev->gfx.config.shader_engine_tile_size = 32;
-	adev->gfx.config.num_gpus = 1;
-	adev->gfx.config.multi_gpu_tile_size = 64;
-
-	/* fix up row size */
-	switch (adev->gfx.config.mem_row_size_in_kb) {
-	case 1:
-	default:
-		gb_addr_config = REG_SET_FIELD(gb_addr_config, GB_ADDR_CONFIG, ROW_SIZE, 0);
-		break;
-	case 2:
-		gb_addr_config = REG_SET_FIELD(gb_addr_config, GB_ADDR_CONFIG, ROW_SIZE, 1);
-		break;
-	case 4:
-		gb_addr_config = REG_SET_FIELD(gb_addr_config, GB_ADDR_CONFIG, ROW_SIZE, 2);
-		break;
-	}
-	adev->gfx.config.gb_addr_config = gb_addr_config;
-
-	WREG32(mmGB_ADDR_CONFIG, gb_addr_config);
-	WREG32(mmHDP_ADDR_CONFIG, gb_addr_config);
-	WREG32(mmDMIF_ADDR_CALC, gb_addr_config);
+	WREG32(mmGB_ADDR_CONFIG, adev->gfx.config.gb_addr_config);
+	WREG32(mmHDP_ADDR_CONFIG, adev->gfx.config.gb_addr_config);
+	WREG32(mmDMIF_ADDR_CALC, adev->gfx.config.gb_addr_config);
 	WREG32(mmSDMA0_TILING_CONFIG + SDMA0_REGISTER_OFFSET,
-	       gb_addr_config & 0x70);
+	       adev->gfx.config.gb_addr_config & 0x70);
 	WREG32(mmSDMA0_TILING_CONFIG + SDMA1_REGISTER_OFFSET,
-	       gb_addr_config & 0x70);
-	WREG32(mmUVD_UDEC_ADDR_CONFIG, gb_addr_config);
-	WREG32(mmUVD_UDEC_DB_ADDR_CONFIG, gb_addr_config);
-	WREG32(mmUVD_UDEC_DBW_ADDR_CONFIG, gb_addr_config);
+	       adev->gfx.config.gb_addr_config & 0x70);
+	WREG32(mmUVD_UDEC_ADDR_CONFIG, adev->gfx.config.gb_addr_config);
+	WREG32(mmUVD_UDEC_DB_ADDR_CONFIG, adev->gfx.config.gb_addr_config);
+	WREG32(mmUVD_UDEC_DBW_ADDR_CONFIG, adev->gfx.config.gb_addr_config);
 
 	gfx_v8_0_tiling_mode_table_init(adev);
 
@@ -2256,13 +2263,13 @@ static void gfx_v8_0_gpu_init(struct amdgpu_device *adev)
 		if (i == 0) {
 			tmp = REG_SET_FIELD(0, SH_MEM_CONFIG, DEFAULT_MTYPE, MTYPE_UC);
 			tmp = REG_SET_FIELD(tmp, SH_MEM_CONFIG, APE1_MTYPE, MTYPE_UC);
-			tmp = REG_SET_FIELD(tmp, SH_MEM_CONFIG, ALIGNMENT_MODE, 
+			tmp = REG_SET_FIELD(tmp, SH_MEM_CONFIG, ALIGNMENT_MODE,
 					    SH_MEM_ALIGNMENT_MODE_UNALIGNED);
 			WREG32(mmSH_MEM_CONFIG, tmp);
 		} else {
 			tmp = REG_SET_FIELD(0, SH_MEM_CONFIG, DEFAULT_MTYPE, MTYPE_NC);
 			tmp = REG_SET_FIELD(tmp, SH_MEM_CONFIG, APE1_MTYPE, MTYPE_NC);
-			tmp = REG_SET_FIELD(tmp, SH_MEM_CONFIG, ALIGNMENT_MODE, 
+			tmp = REG_SET_FIELD(tmp, SH_MEM_CONFIG, ALIGNMENT_MODE,
 					    SH_MEM_ALIGNMENT_MODE_UNALIGNED);
 			WREG32(mmSH_MEM_CONFIG, tmp);
 		}
-- 
cgit v0.10.2


From a2ef8a974931bd4a28e81a3de0a3813e97acfc30 Mon Sep 17 00:00:00 2001
From: Chunming Zhou <david1.zhou@amd.com>
Date: Tue, 22 Sep 2015 18:20:50 +0800
Subject: drm/amdgpu: add vram usage into debugfs
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index 364cbe9..a089e69 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -1072,6 +1072,11 @@ static int amdgpu_mm_dump_table(struct seq_file *m, void *data)
 	spin_lock(&glob->lru_lock);
 	ret = drm_mm_dump_table(m, mm);
 	spin_unlock(&glob->lru_lock);
+	if (ttm_pl == TTM_PL_VRAM)
+		seq_printf(m, "man size:%llu pages, ram usage:%luMB, vis usage:%luMB\n",
+			   adev->mman.bdev.man[ttm_pl].size,
+			   atomic64_read(&adev->vram_usage) >> 20,
+			   atomic64_read(&adev->vram_vis_usage) >> 20);
 	return ret;
 }
 
-- 
cgit v0.10.2


From 95d7918323be0090a1cc36f0ebdd1d3af88f2335 Mon Sep 17 00:00:00 2001
From: Chunming Zhou <david1.zhou@amd.com>
Date: Wed, 23 Sep 2015 17:22:43 +0800
Subject: drm/amdgpu: add TOPDOWN flag to the whole vram
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

need to decrease visible vram usage by default.

Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: monk.liu <monk.liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
index 1a7708f..0d52438 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
@@ -132,6 +132,8 @@ static void amdgpu_ttm_placement_init(struct amdgpu_device *adev,
 		placements[c].fpfn = 0;
 		placements[c++].flags = TTM_PL_FLAG_WC | TTM_PL_FLAG_UNCACHED |
 			TTM_PL_FLAG_VRAM;
+		if (!(flags & AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED))
+			placements[c - 1].flags |= TTM_PL_FLAG_TOPDOWN;
 	}
 
 	if (domain & AMDGPU_GEM_DOMAIN_GTT) {
-- 
cgit v0.10.2


From 36b4ba07d673032d71045bc149930b0f176e6292 Mon Sep 17 00:00:00 2001
From: Chunming Zhou <david1.zhou@amd.com>
Date: Tue, 15 Sep 2015 16:56:10 +0800
Subject: drm/amdgpu: enable scheduler by default
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Change-Id: Idce64f63e8422324996fc5d583d0bc9a5ac60d0c
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index adb4835..fcbff9f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -76,7 +76,7 @@ int amdgpu_deep_color = 0;
 int amdgpu_vm_size = 8;
 int amdgpu_vm_block_size = -1;
 int amdgpu_exp_hw_support = 0;
-int amdgpu_enable_scheduler = 0;
+int amdgpu_enable_scheduler = 1;
 int amdgpu_sched_jobs = 16;
 int amdgpu_sched_hw_submission = 2;
 int amdgpu_enable_semaphores = 1;
@@ -144,7 +144,7 @@ module_param_named(vm_block_size, amdgpu_vm_block_size, int, 0444);
 MODULE_PARM_DESC(exp_hw_support, "experimental hw support (1 = enable, 0 = disable (default))");
 module_param_named(exp_hw_support, amdgpu_exp_hw_support, int, 0444);
 
-MODULE_PARM_DESC(enable_scheduler, "enable SW GPU scheduler (1 = enable, 0 = disable ((default))");
+MODULE_PARM_DESC(enable_scheduler, "enable SW GPU scheduler (1 = enable (default), 0 = disable)");
 module_param_named(enable_scheduler, amdgpu_enable_scheduler, int, 0444);
 
 MODULE_PARM_DESC(sched_jobs, "the max number of jobs supported in the sw queue (default 16)");
-- 
cgit v0.10.2


From d4edda3f2b3c74379c78c656a06657b328c813aa Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Wed, 30 Sep 2015 12:24:19 -0400
Subject: drm/amdgpu: disable hw semaphores by default

These are buggy on some asics and not really used anymore
now that the GPU schedular is enabled.

Change-Id: I67182b409d64de308392a15d1a0a15018071dc0b
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index fcbff9f..c183772 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -79,7 +79,7 @@ int amdgpu_exp_hw_support = 0;
 int amdgpu_enable_scheduler = 1;
 int amdgpu_sched_jobs = 16;
 int amdgpu_sched_hw_submission = 2;
-int amdgpu_enable_semaphores = 1;
+int amdgpu_enable_semaphores = 0;
 
 MODULE_PARM_DESC(vramlimit, "Restrict VRAM for testing, in megabytes");
 module_param_named(vramlimit, amdgpu_vram_limit, int, 0600);
@@ -153,7 +153,7 @@ module_param_named(sched_jobs, amdgpu_sched_jobs, int, 0444);
 MODULE_PARM_DESC(sched_hw_submission, "the max number of HW submissions (default 2)");
 module_param_named(sched_hw_submission, amdgpu_sched_hw_submission, int, 0444);
 
-MODULE_PARM_DESC(enable_semaphores, "Enable semaphores (1 = enable (default), 0 = disable)");
+MODULE_PARM_DESC(enable_semaphores, "Enable semaphores (1 = enable, 0 = disable (default))");
 module_param_named(enable_semaphores, amdgpu_enable_semaphores, int, 0644);
 
 static struct pci_device_id pciidlist[] = {
-- 
cgit v0.10.2


From 468f9d2903bcf23e50663679eab5a18da5b886f6 Mon Sep 17 00:00:00 2001
From: Javier Martinez Canillas <javier@osg.samsung.com>
Date: Thu, 8 Oct 2015 09:54:44 +0200
Subject: drm/i915/irq: Fix kernel-doc warnings

Add the dev parameter for the functions i915_enable_asle_pipestat() and
i915_reset_and_wakeup() to the kernel-doc to fix the following warnings:

.//drivers/gpu/drm/i915/i915_irq.c:586: warning: No description found for parameter 'dev'
.//drivers/gpu/drm/i915/i915_irq.c:2400: warning: No description found for parameter 'dev'

Signed-off-by: Javier Martinez Canillas <javier@osg.samsung.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 76bd40e..18b4754 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -581,6 +581,7 @@ i915_disable_pipestat(struct drm_i915_private *dev_priv, enum pipe pipe,
 
 /**
  * i915_enable_asle_pipestat - enable ASLE pipestat for OpRegion
+ * @dev: drm device
  */
 static void i915_enable_asle_pipestat(struct drm_device *dev)
 {
@@ -2392,6 +2393,7 @@ static void i915_error_wake_up(struct drm_i915_private *dev_priv,
 
 /**
  * i915_reset_and_wakeup - do process context error handling work
+ * @dev: drm device
  *
  * Fire an error uevent so userspace can see that a hang or error
  * was detected.
-- 
cgit v0.10.2


From aafd858192bc98fab9230f6a634af85c4079f80f Mon Sep 17 00:00:00 2001
From: Javier Martinez Canillas <javier@osg.samsung.com>
Date: Thu, 8 Oct 2015 09:57:49 +0200
Subject: drm/i915/irq: Fix misspelled word register in kernel-doc

There is a typo in the function i915_handle_error()
kernel-doc and the word register is spelled wrongly.

Signed-off-by: Javier Martinez Canillas <javier@osg.samsung.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 18b4754..bd56ca6 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -2571,7 +2571,7 @@ static void i915_report_and_clear_eir(struct drm_device *dev)
  * i915_handle_error - handle a gpu error
  * @dev: drm device
  *
- * Do some basic checking of regsiter state at error time and
+ * Do some basic checking of register state at error time and
  * dump it to the syslog.  Also call i915_capture_error_state() to make
  * sure we get a record and make it available in debugfs.  Fire a uevent
  * so userspace knows something bad happened (should trigger collection
-- 
cgit v0.10.2


From 11aee0f6d1b6520620c5e7af04504ce9de588520 Mon Sep 17 00:00:00 2001
From: Sudip Mukherjee <sudipm.mukherjee@gmail.com>
Date: Thu, 8 Oct 2015 19:27:59 +0530
Subject: drm/i915: use error path

Use goto to handle the error path to avoid duplicating the same code. In
the error path intel_dig_port is the last one to be released as it was
the first one to be allocated and ideally the error path should be the
reverse of the execution path.

Cc: Daniel Vetter <daniel.vetter@intel.com>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 8d34ca7..18bcfbe 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -6168,10 +6168,8 @@ intel_dp_init(struct drm_device *dev, int output_reg, enum port port)
 		return;
 
 	intel_connector = intel_connector_alloc();
-	if (!intel_connector) {
-		kfree(intel_dig_port);
-		return;
-	}
+	if (!intel_connector)
+		goto err_connector_alloc;
 
 	intel_encoder = &intel_dig_port->base;
 	encoder = &intel_encoder->base;
@@ -6219,11 +6217,18 @@ intel_dp_init(struct drm_device *dev, int output_reg, enum port port)
 	intel_dig_port->hpd_pulse = intel_dp_hpd_pulse;
 	dev_priv->hotplug.irq_port[port] = intel_dig_port;
 
-	if (!intel_dp_init_connector(intel_dig_port, intel_connector)) {
-		drm_encoder_cleanup(encoder);
-		kfree(intel_dig_port);
-		kfree(intel_connector);
-	}
+	if (!intel_dp_init_connector(intel_dig_port, intel_connector))
+		goto err_init_connector;
+
+	return;
+
+err_init_connector:
+	drm_encoder_cleanup(encoder);
+	kfree(intel_connector);
+err_connector_alloc:
+	kfree(intel_dig_port);
+
+	return;
 }
 
 void intel_dp_mst_suspend(struct drm_device *dev)
-- 
cgit v0.10.2


From 8166fcead6b29589f264b9d900c0770f501964c3 Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Thu, 8 Oct 2015 21:50:57 +0200
Subject: Revert "drm/i915: Add hot_plug hook for hdmi encoder"

This reverts commit 0b5e88dc25ee6c9040c0355e6e025dcbc9c8de92.

It completely breaks booting on at least bsw (and maybe more).

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=88081
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index 03d8590..3b28ed3 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -1369,16 +1369,18 @@ intel_hdmi_set_edid(struct drm_connector *connector, bool force)
 	return connected;
 }
 
-static void intel_hdmi_hot_plug(struct intel_encoder *intel_encoder)
+static enum drm_connector_status
+intel_hdmi_detect(struct drm_connector *connector, bool force)
 {
-	struct intel_hdmi *intel_hdmi =
-			enc_to_intel_hdmi(&intel_encoder->base);
-	struct intel_connector *intel_connector =
-				intel_hdmi->attached_connector;
-	struct drm_i915_private *dev_priv = to_i915(intel_encoder->base.dev);
+	enum drm_connector_status status;
+	struct intel_hdmi *intel_hdmi = intel_attached_hdmi(connector);
+	struct drm_i915_private *dev_priv = to_i915(connector->dev);
 	bool live_status = false;
 	unsigned int retry = 3;
 
+	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
+		      connector->base.id, connector->name);
+
 	while (!live_status && --retry) {
 		live_status = intel_digital_port_connected(dev_priv,
 				hdmi_to_dig_port(intel_hdmi));
@@ -1388,48 +1390,15 @@ static void intel_hdmi_hot_plug(struct intel_encoder *intel_encoder)
 	if (!live_status)
 		DRM_DEBUG_KMS("Live status not up!");
 
-	/*
-	 * We are here, means there is a hotplug or a force
-	 * detection. Clear the cached EDID and probe the
-	 * DDC bus to check the current status of HDMI.
-	 */
-	intel_hdmi_unset_edid(&intel_connector->base);
-	if (intel_hdmi_set_edid(&intel_connector->base, live_status))
-		DRM_DEBUG_DRIVER("DDC probe: got EDID\n");
-	else
-		DRM_DEBUG_DRIVER("DDC probe: no EDID\n");
-}
-
-static enum drm_connector_status
-intel_hdmi_detect(struct drm_connector *connector, bool force)
-{
-	enum drm_connector_status status;
-	struct intel_connector *intel_connector =
-				to_intel_connector(connector);
-
-	DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n",
-		      connector->base.id, connector->name);
-
-	/*
-	 * There are many userspace calls which probe EDID from
-	 * detect path. In case of multiple hotplug/unplug, these
-	 * can cause race conditions while probing EDID. Also its
-	 * waste of CPU cycles to read the EDID again and again
-	 * unless there is a real hotplug.
-	 * So, rely on hotplugs and init to read edid.
-	 * Check connector status based on availability of cached EDID.
-	 */
+	intel_hdmi_unset_edid(connector);
 
-	if (intel_connector->detect_edid) {
+	if (intel_hdmi_set_edid(connector, live_status)) {
 		struct intel_hdmi *intel_hdmi = intel_attached_hdmi(connector);
 
 		hdmi_to_dig_port(intel_hdmi)->base.type = INTEL_OUTPUT_HDMI;
 		status = connector_status_connected;
-		DRM_DEBUG_DRIVER("hdmi status = connected\n");
-	} else {
+	} else
 		status = connector_status_disconnected;
-		DRM_DEBUG_DRIVER("hdmi status = disconnected\n");
-	}
 
 	return status;
 }
@@ -2145,7 +2114,6 @@ void intel_hdmi_init_connector(struct intel_digital_port *intel_dig_port,
 	intel_connector->unregister = intel_connector_unregister;
 
 	intel_hdmi_add_properties(intel_hdmi, connector);
-	intel_encoder->hot_plug = intel_hdmi_hot_plug;
 
 	intel_connector_attach_encoder(intel_connector, intel_encoder);
 	drm_connector_register(connector);
-- 
cgit v0.10.2


From 3a2fb2c394ff0f0ea90bacc557147d9ac733d13c Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Thu, 8 Oct 2015 21:51:57 +0200
Subject: Revert "drm/i915: Call encoder hotplug for init and resume cases"
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This reverts commit 5d250b05918c002b63632c7db91c3c5f924c6a3b.

It results on a deadlock on platforms where we need to (at least
partially) re-init hpd interrupts from power domain code, since
->hot_plug might again grab a power well reference (to do edid/dp_aux
transactions. At least chv is affected.

Reported-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
References: http://mid.gmane.org/20151008133548.GX26517@intel.com
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

diff --git a/drivers/gpu/drm/i915/intel_hotplug.c b/drivers/gpu/drm/i915/intel_hotplug.c
index eac4757..53c0173 100644
--- a/drivers/gpu/drm/i915/intel_hotplug.c
+++ b/drivers/gpu/drm/i915/intel_hotplug.c
@@ -458,7 +458,6 @@ void intel_hpd_init(struct drm_i915_private *dev_priv)
 {
 	struct drm_device *dev = dev_priv->dev;
 	struct drm_mode_config *mode_config = &dev->mode_config;
-	struct intel_encoder *encoder;
 	struct drm_connector *connector;
 	int i;
 
@@ -483,16 +482,6 @@ void intel_hpd_init(struct drm_i915_private *dev_priv)
 	if (dev_priv->display.hpd_irq_setup)
 		dev_priv->display.hpd_irq_setup(dev);
 	spin_unlock_irq(&dev_priv->irq_lock);
-
-	/*
-	 * Connected boot / resume scenarios can't generate new hot plug.
-	 * So, probe it manually.
-	 */
-	list_for_each_entry(encoder, &dev->mode_config.encoder_list,
-			    base.head) {
-		if (encoder->hot_plug)
-			encoder->hot_plug(encoder);
-	}
 }
 
 void intel_hpd_init_work(struct drm_i915_private *dev_priv)
diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c
index 853f4b2..c42b636 100644
--- a/drivers/gpu/drm/i915/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/intel_sdvo.c
@@ -2460,6 +2460,7 @@ intel_sdvo_dvi_init(struct intel_sdvo *intel_sdvo, int device)
 		 * Ensure that they get re-enabled when an interrupt happens.
 		 */
 		intel_encoder->hot_plug = intel_sdvo_enable_hotplug;
+		intel_sdvo_enable_hotplug(intel_encoder);
 	} else {
 		intel_connector->polled = DRM_CONNECTOR_POLL_CONNECT | DRM_CONNECTOR_POLL_DISCONNECT;
 	}
-- 
cgit v0.10.2


From 3badb49f08f66fbdbf9da8335c5db4b05a1677c8 Mon Sep 17 00:00:00 2001
From: Paulo Zanoni <paulo.r.zanoni@intel.com>
Date: Wed, 23 Sep 2015 12:52:23 -0300
Subject: drm/i915: don't allocate fbcon from stolen memory if it's too big

Technology has evolved and now we have eDP panels with 3200x1800
resolution. In the meantime, the BIOS guys didn't change the default
32mb for stolen memory. On top of that, we can't assume our users will
be able to increase the default stolen memory size to more than 32mb -
I'm not even sure all BIOSes allow that.

So just the fbcon buffer alone eats 22mb of my stolen memroy, and due
to the BDW/SKL restriction of not using the last 8mb of stolen memory,
all that's left for FBC is 2mb! Since fbcon is not the coolest feature
ever, I think it's better to save our precious stolen resource to FBC
and the other guys.

On the other hand, we really want to use as much stolen memory as
possible, since on some older systems the stolen memory may be a
considerable percentage of the total available memory.

This patch tries to achieve a little balance using a simple heuristic:
if the fbcon wants more than half of the available stolen memory,
don't use stolen memory in order to leave some for FBC and the other
features.

The long term plan should be to implement a way to set priorities for
stolen memory allocation and then evict low priority users when the
high priority ones need the memory. While we still don't have that,
let's try to make FBC usable with the simple solution.

Cc: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 539c373..184ba09 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -2545,6 +2545,7 @@ intel_alloc_initial_plane_obj(struct intel_crtc *crtc,
 			      struct intel_initial_plane_config *plane_config)
 {
 	struct drm_device *dev = crtc->base.dev;
+	struct drm_i915_private *dev_priv = to_i915(dev);
 	struct drm_i915_gem_object *obj = NULL;
 	struct drm_mode_fb_cmd2 mode_cmd = { 0 };
 	struct drm_framebuffer *fb = &plane_config->fb->base;
@@ -2557,6 +2558,12 @@ intel_alloc_initial_plane_obj(struct intel_crtc *crtc,
 	if (plane_config->size == 0)
 		return false;
 
+	/* If the FB is too big, just don't use it since fbdev is not very
+	 * important and we should probably use that space with FBC or other
+	 * features. */
+	if (size_aligned * 2 > dev_priv->gtt.stolen_usable_size)
+		return false;
+
 	obj = i915_gem_object_create_stolen_for_preallocated(dev,
 							     base_aligned,
 							     base_aligned,
diff --git a/drivers/gpu/drm/i915/intel_fbdev.c b/drivers/gpu/drm/i915/intel_fbdev.c
index 6532912..4fd5fdf 100644
--- a/drivers/gpu/drm/i915/intel_fbdev.c
+++ b/drivers/gpu/drm/i915/intel_fbdev.c
@@ -121,8 +121,9 @@ static int intelfb_alloc(struct drm_fb_helper *helper,
 		container_of(helper, struct intel_fbdev, helper);
 	struct drm_framebuffer *fb;
 	struct drm_device *dev = helper->dev;
+	struct drm_i915_private *dev_priv = to_i915(dev);
 	struct drm_mode_fb_cmd2 mode_cmd = {};
-	struct drm_i915_gem_object *obj;
+	struct drm_i915_gem_object *obj = NULL;
 	int size, ret;
 
 	/* we don't do packed 24bpp */
@@ -139,7 +140,12 @@ static int intelfb_alloc(struct drm_fb_helper *helper,
 
 	size = mode_cmd.pitches[0] * mode_cmd.height;
 	size = PAGE_ALIGN(size);
-	obj = i915_gem_object_create_stolen(dev, size);
+
+	/* If the FB is too big, just don't use it since fbdev is not very
+	 * important and we should probably use that space with FBC or other
+	 * features. */
+	if (size * 2 < dev_priv->gtt.stolen_usable_size)
+		obj = i915_gem_object_create_stolen(dev, size);
 	if (obj == NULL)
 		obj = i915_gem_alloc_object(dev, size);
 	if (!obj) {
-- 
cgit v0.10.2


From a42e5a23ed1939a8eca5a753c19bb8b0e5cee475 Mon Sep 17 00:00:00 2001
From: Paulo Zanoni <paulo.r.zanoni@intel.com>
Date: Wed, 30 Sep 2015 17:05:43 -0300
Subject: drm/i915: remove pre-atomic check from SKL update_primary_plane
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The comment suggests the check was there for some non-fully-atomic
case, and I couldn't find a case where we wouldn't correctly
initialize plane_state, so remove the check.

Let's leave a WARN there just in case.

Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Acked-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 184ba09..6b3b65e 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -3131,27 +3131,19 @@ static void skylake_update_primary_plane(struct drm_crtc *crtc,
 					       fb->pixel_format);
 	surf_addr = intel_plane_obj_offset(to_intel_plane(plane), obj, 0);
 
-	/*
-	 * FIXME: intel_plane_state->src, dst aren't set when transitional
-	 * update_plane helpers are called from legacy paths.
-	 * Once full atomic crtc is available, below check can be avoided.
-	 */
-	if (drm_rect_width(&plane_state->src)) {
-		scaler_id = plane_state->scaler_id;
-		src_x = plane_state->src.x1 >> 16;
-		src_y = plane_state->src.y1 >> 16;
-		src_w = drm_rect_width(&plane_state->src) >> 16;
-		src_h = drm_rect_height(&plane_state->src) >> 16;
-		dst_x = plane_state->dst.x1;
-		dst_y = plane_state->dst.y1;
-		dst_w = drm_rect_width(&plane_state->dst);
-		dst_h = drm_rect_height(&plane_state->dst);
-
-		WARN_ON(x != src_x || y != src_y);
-	} else {
-		src_w = intel_crtc->config->pipe_src_w;
-		src_h = intel_crtc->config->pipe_src_h;
-	}
+	WARN_ON(drm_rect_width(&plane_state->src) == 0);
+
+	scaler_id = plane_state->scaler_id;
+	src_x = plane_state->src.x1 >> 16;
+	src_y = plane_state->src.y1 >> 16;
+	src_w = drm_rect_width(&plane_state->src) >> 16;
+	src_h = drm_rect_height(&plane_state->src) >> 16;
+	dst_x = plane_state->dst.x1;
+	dst_y = plane_state->dst.y1;
+	dst_w = drm_rect_width(&plane_state->dst);
+	dst_h = drm_rect_height(&plane_state->dst);
+
+	WARN_ON(x != src_x || y != src_y);
 
 	if (intel_rotation_90_or_270(rotation)) {
 		/* stride = Surface height in tiles */
-- 
cgit v0.10.2


From c4ffd40908c30a33291227920e921f6b45b9e8f7 Mon Sep 17 00:00:00 2001
From: Paulo Zanoni <paulo.r.zanoni@intel.com>
Date: Thu, 1 Oct 2015 19:55:57 -0300
Subject: drm/i915: fix CFB size calculation
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We were considering the whole framebuffer height, but the spec says we
should only consider the active display height size. There were still
some unclear questions based on the spec, but the hardware guys
clarified them for us. According to them:

- CFB size = CFB stride * Number of lines FBC writes to CFB
- CFB stride = plane stride / compression limit
- Number of lines FBC writes to CFB = MIN(plane source height, maximum
  number of lines FBC writes to CFB)
- Plane source height =
  - pipe source height (PIPE_SRCSZ register) (before SKL)
  - plane size register height (PLANE_SIZE register) (SKL+)
- Maximum number of lines FBC writes to CFB =
  - plane source height (before HSW)
  - 2048 (HSW+)

For the plane source height, I could just have made our code do
I915_READ() in order to be more future proof, but since it's not cool
to do register reads I decided to just recalculate the values we use
when we actually write to those registers.

With this patch, depending on your machine configuration, a lot of the
kms_frontbuffer_tracking subtests that used to result in a SKIP due to
not enough stolen memory still start resulting in a PASS.

v2: Use the clipped src size instead of pipe_src_h (Ville).
v3: Use the appropriate information provided by the hardware guys.
v4: Bikesheds: s/sizes/size/, s/fb_cpp/cpp/ (Ville).
v5: - Don't use crtc->config->pipe_src_x for BDW- (Ville).
    - Fix the register name written in the comment.

Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_fbc.c b/drivers/gpu/drm/i915/intel_fbc.c
index 1b2ebb2..18e228b 100644
--- a/drivers/gpu/drm/i915/intel_fbc.c
+++ b/drivers/gpu/drm/i915/intel_fbc.c
@@ -698,16 +698,61 @@ void intel_fbc_cleanup_cfb(struct drm_i915_private *dev_priv)
 	mutex_unlock(&dev_priv->fbc.lock);
 }
 
-static int intel_fbc_setup_cfb(struct drm_i915_private *dev_priv, int size,
-			       int fb_cpp)
+/*
+ * For SKL+, the plane source size used by the hardware is based on the value we
+ * write to the PLANE_SIZE register. For BDW-, the hardware looks at the value
+ * we wrote to PIPESRC.
+ */
+static void intel_fbc_get_plane_source_size(struct intel_crtc *crtc,
+					    int *width, int *height)
 {
+	struct intel_plane_state *plane_state =
+			to_intel_plane_state(crtc->base.primary->state);
+	int w, h;
+
+	if (intel_rotation_90_or_270(plane_state->base.rotation)) {
+		w = drm_rect_height(&plane_state->src) >> 16;
+		h = drm_rect_width(&plane_state->src) >> 16;
+	} else {
+		w = drm_rect_width(&plane_state->src) >> 16;
+		h = drm_rect_height(&plane_state->src) >> 16;
+	}
+
+	if (width)
+		*width = w;
+	if (height)
+		*height = h;
+}
+
+static int intel_fbc_calculate_cfb_size(struct intel_crtc *crtc)
+{
+	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
+	struct drm_framebuffer *fb = crtc->base.primary->fb;
+	int lines;
+
+	intel_fbc_get_plane_source_size(crtc, NULL, &lines);
+	if (INTEL_INFO(dev_priv)->gen >= 7)
+		lines = min(lines, 2048);
+
+	return lines * fb->pitches[0];
+}
+
+static int intel_fbc_setup_cfb(struct intel_crtc *crtc)
+{
+	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
+	struct drm_framebuffer *fb = crtc->base.primary->fb;
+	int size, cpp;
+
+	size = intel_fbc_calculate_cfb_size(crtc);
+	cpp = drm_format_plane_cpp(fb->pixel_format, 0);
+
 	if (size <= dev_priv->fbc.uncompressed_size)
 		return 0;
 
 	/* Release any current block */
 	__intel_fbc_cleanup_cfb(dev_priv);
 
-	return intel_fbc_alloc_cfb(dev_priv, size, fb_cpp);
+	return intel_fbc_alloc_cfb(dev_priv, size, cpp);
 }
 
 static bool stride_is_valid(struct drm_i915_private *dev_priv,
@@ -897,8 +942,7 @@ static void __intel_fbc_update(struct drm_i915_private *dev_priv)
 		goto out_disable;
 	}
 
-	if (intel_fbc_setup_cfb(dev_priv, obj->base.size,
-				drm_format_plane_cpp(fb->pixel_format, 0))) {
+	if (intel_fbc_setup_cfb(intel_crtc)) {
 		set_no_fbc_reason(dev_priv, FBC_STOLEN_TOO_SMALL);
 		goto out_disable;
 	}
-- 
cgit v0.10.2


From 856312aeb124b0477d452aa43415711109b46cbe Mon Sep 17 00:00:00 2001
From: Paulo Zanoni <paulo.r.zanoni@intel.com>
Date: Thu, 1 Oct 2015 19:57:12 -0300
Subject: drm/i915: fix FBC buffer size checks
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

According to my experiments (and later confirmation from the hardware
developers), the maximum sizes mentioned in the specification delimit
how far in the buffer the hardware tracking can go. And the hardware
calculates the size based on the plane address we provide - and the
provided plane address might not be the real x:0,y:0 point due to the
compute_page_offset() function.

On platforms that do the x/y offset adjustment trick it will be really
hard to reproduce a bug, but on the current SKL we can reproduce the
bug with igt/kms_frontbuffer_tracking/fbc-farfromfence. With this
patch, we'll go from "CRC assertion failure" to "FBC unexpectedly
disabled", which is still a failure on the test suite but is not a
perceived user bug - you will just not save as much power as you could
if FBC is disabled.

v2, rewrite patch after clarification from the Hadware guys:
  - Rename function so it's clear what the check is for.
  - Use the new intel_fbc_get_plane_source_sizes() function in order
    to get the proper sizes as seen by FBC.
v3:
  - Rebase after the s/sizes/size/ on the previous patch.
  - Adjust comment wording (Ville).
  - s/used_/effective_/ (Ville).

Testcase: igt/kms_frontbuffer_tracking/fbc-farfromfence (SKL)
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_fbc.c b/drivers/gpu/drm/i915/intel_fbc.c
index 18e228b..cf47352 100644
--- a/drivers/gpu/drm/i915/intel_fbc.c
+++ b/drivers/gpu/drm/i915/intel_fbc.c
@@ -799,10 +799,16 @@ static bool pixel_format_is_valid(struct drm_framebuffer *fb)
 	}
 }
 
-static bool pipe_size_is_valid(struct intel_crtc *crtc)
+/*
+ * For some reason, the hardware tracking starts looking at whatever we
+ * programmed as the display plane base address register. It does not look at
+ * the X and Y offset registers. That's why we look at the crtc->adjusted{x,y}
+ * variables instead of just looking at the pipe/plane size.
+ */
+static bool intel_fbc_hw_tracking_covers_screen(struct intel_crtc *crtc)
 {
 	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
-	unsigned int max_w, max_h;
+	unsigned int effective_w, effective_h, max_w, max_h;
 
 	if (INTEL_INFO(dev_priv)->gen >= 8 || IS_HASWELL(dev_priv)) {
 		max_w = 4096;
@@ -815,8 +821,11 @@ static bool pipe_size_is_valid(struct intel_crtc *crtc)
 		max_h = 1536;
 	}
 
-	return crtc->config->pipe_src_w <= max_w &&
-	       crtc->config->pipe_src_h <= max_h;
+	intel_fbc_get_plane_source_size(crtc, &effective_w, &effective_h);
+	effective_w += crtc->adjusted_x;
+	effective_h += crtc->adjusted_y;
+
+	return effective_w <= max_w && effective_h <= max_h;
 }
 
 /**
@@ -893,7 +902,7 @@ static void __intel_fbc_update(struct drm_i915_private *dev_priv)
 		goto out_disable;
 	}
 
-	if (!pipe_size_is_valid(intel_crtc)) {
+	if (!intel_fbc_hw_tracking_covers_screen(intel_crtc)) {
 		set_no_fbc_reason(dev_priv, FBC_MODE_TOO_LARGE);
 		goto out_disable;
 	}
-- 
cgit v0.10.2


From 0ad98c74e093041eb163b01013eb46989f2124aa Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 8 Oct 2015 12:08:20 +0300
Subject: drm/i915: Determine the stolen memory base address on gen2
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

There isn't an explicit stolen memory base register on gen2.
Some old comment in the i915 code suggests we should get it via
max_low_pfn_mapped, but that's clearly a bad idea on my MGM.

The e820 map in said machine looks like this:
[    0.000000] BIOS-e820: [mem 0x0000000000000000-0x000000000009f7ff] usable
[    0.000000] BIOS-e820: [mem 0x000000000009f800-0x000000000009ffff] reserved
[    0.000000] BIOS-e820: [mem 0x00000000000ce000-0x00000000000cffff] reserved
[    0.000000] BIOS-e820: [mem 0x00000000000dc000-0x00000000000fffff] reserved
[    0.000000] BIOS-e820: [mem 0x0000000000100000-0x000000001f6effff] usable
[    0.000000] BIOS-e820: [mem 0x000000001f6f0000-0x000000001f6f7fff] ACPI data
[    0.000000] BIOS-e820: [mem 0x000000001f6f8000-0x000000001f6fffff] ACPI NVS
[    0.000000] BIOS-e820: [mem 0x000000001f700000-0x000000001fffffff] reserved
[    0.000000] BIOS-e820: [mem 0x00000000fec10000-0x00000000fec1ffff] reserved
[    0.000000] BIOS-e820: [mem 0x00000000ffb00000-0x00000000ffbfffff] reserved
[    0.000000] BIOS-e820: [mem 0x00000000fff00000-0x00000000ffffffff] reserved

That makes max_low_pfn_mapped = 1f6f0000, so assuming our stolen memory
would start there would place it on top of some ACPI memory regions.
So not a good idea as already stated.

The 9MB region after the ACPI regions at 0x1f700000 however looks
promising given that the macine reports the stolen memory size to be
8MB. Looking at the PGTBL_CTL register, the GTT entries are at offset
0x1fee00000, and given that the GTT entries occupy 128KB, it looks like
the stolen memory could start at 0x1f700000 and the GTT entries would
occupy the last 128KB of the stolen memory.

After some more digging through chipset documentation, I've determined
the BIOS first allocates space for something called TSEG (something to
do with SMM) from the top of memory, and then it allocates the graphics
stolen memory below that. Accordind to the chipset documentation TSEG
has a fixed size of 1MB on 855. So that explains the top 1MB in the
e820 region. And it also confirms that the GTT entries are in fact at
the end of the the stolen memory region.

Derive the stolen memory base address on gen2 the same as the BIOS does
(TOM-TSEG_SIZE-stolen_size). There are a few differences between the
registers on various gen2 chipsets, so a few different codepaths are
required.

865G is again bit more special since it seems to support enough memory
to hit 4GB address space issues. This means the PCI allocations will
also affect the location of the stolen memory. Fortunately there
appears to be the TOUD register which may give us the correct answer
directly. But the chipset docs are a bit unclear, so I'm not 100%
sure that the graphics stolen memory is always the last thing the
BIOS steals. Someone would need to verify it on a real system.

I tested this on the my 830 and 855 machines, and so far everything
looks peachy.

v2: Rewrite to use the TOM-TSEG_SIZE-stolen_size and TOUD methods
v3: Fix TSEG size for 830
v4: Add missing 'else' (Chris)

Tested-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_stolen.c b/drivers/gpu/drm/i915/i915_gem_stolen.c
index 69eebc6..cdacf3f 100644
--- a/drivers/gpu/drm/i915/i915_gem_stolen.c
+++ b/drivers/gpu/drm/i915/i915_gem_stolen.c
@@ -30,6 +30,9 @@
 #include <drm/i915_drm.h>
 #include "i915_drv.h"
 
+#define KB(x) ((x) * 1024)
+#define MB(x) (KB(x) * 1024)
+
 /*
  * The BIOS typically reserves some of the system's memory for the exclusive
  * use of the integrated graphics. This memory is no longer available for
@@ -91,24 +94,91 @@ static unsigned long i915_stolen_to_physical(struct drm_device *dev)
 	/* Almost universally we can find the Graphics Base of Stolen Memory
 	 * at offset 0x5c in the igfx configuration space. On a few (desktop)
 	 * machines this is also mirrored in the bridge device at different
-	 * locations, or in the MCHBAR. On gen2, the layout is again slightly
-	 * different with the Graphics Segment immediately following Top of
-	 * Memory (or Top of Usable DRAM). Note it appears that TOUD is only
-	 * reported by 865g, so we just use the top of memory as determined
-	 * by the e820 probe.
+	 * locations, or in the MCHBAR.
+	 *
+	 * On 865 we just check the TOUD register.
+	 *
+	 * On 830/845/85x the stolen memory base isn't available in any
+	 * register. We need to calculate it as TOM-TSEG_SIZE-stolen_size.
 	 *
-	 * XXX However gen2 requires an unavailable symbol.
 	 */
 	base = 0;
 	if (INTEL_INFO(dev)->gen >= 3) {
 		/* Read Graphics Base of Stolen Memory directly */
 		pci_read_config_dword(dev->pdev, 0x5c, &base);
 		base &= ~((1<<20) - 1);
-	} else { /* GEN2 */
-#if 0
-		/* Stolen is immediately above Top of Memory */
-		base = max_low_pfn_mapped << PAGE_SHIFT;
-#endif
+	} else if (IS_I865G(dev)) {
+		u16 toud = 0;
+
+		/*
+		 * FIXME is the graphics stolen memory region
+		 * always at TOUD? Ie. is it always the last
+		 * one to be allocated by the BIOS?
+		 */
+		pci_bus_read_config_word(dev->pdev->bus, PCI_DEVFN(0, 0),
+					 I865_TOUD, &toud);
+
+		base = toud << 16;
+	} else if (IS_I85X(dev)) {
+		u32 tseg_size = 0;
+		u32 tom;
+		u8 tmp;
+
+		pci_bus_read_config_byte(dev->pdev->bus, PCI_DEVFN(0, 0),
+					 I85X_ESMRAMC, &tmp);
+
+		if (tmp & TSEG_ENABLE)
+			tseg_size = MB(1);
+
+		pci_bus_read_config_byte(dev->pdev->bus, PCI_DEVFN(0, 1),
+					 I85X_DRB3, &tmp);
+		tom = tmp * MB(32);
+
+		base = tom - tseg_size - dev_priv->gtt.stolen_size;
+	} else if (IS_845G(dev)) {
+		u32 tseg_size = 0;
+		u32 tom;
+		u8 tmp;
+
+		pci_bus_read_config_byte(dev->pdev->bus, PCI_DEVFN(0, 0),
+					 I845_ESMRAMC, &tmp);
+
+		if (tmp & TSEG_ENABLE) {
+			switch (tmp & I845_TSEG_SIZE_MASK) {
+			case I845_TSEG_SIZE_512K:
+				tseg_size = KB(512);
+				break;
+			case I845_TSEG_SIZE_1M:
+				tseg_size = MB(1);
+				break;
+			}
+		}
+
+		pci_bus_read_config_byte(dev->pdev->bus, PCI_DEVFN(0, 0),
+					 I830_DRB3, &tmp);
+		tom = tmp * MB(32);
+
+		base = tom - tseg_size - dev_priv->gtt.stolen_size;
+	} else if (IS_I830(dev)) {
+		u32 tseg_size = 0;
+		u32 tom;
+		u8 tmp;
+
+		pci_bus_read_config_byte(dev->pdev->bus, PCI_DEVFN(0, 0),
+					 I830_ESMRAMC, &tmp);
+
+		if (tmp & TSEG_ENABLE) {
+			if (tmp & I830_TSEG_SIZE_1M)
+				tseg_size = MB(1);
+			else
+				tseg_size = KB(512);
+		}
+
+		pci_bus_read_config_byte(dev->pdev->bus, PCI_DEVFN(0, 0),
+					 I830_DRB3, &tmp);
+		tom = tmp * MB(32);
+
+		base = tom - tseg_size - dev_priv->gtt.stolen_size;
 	}
 
 	if (base == 0)
-- 
cgit v0.10.2


From 61fb58815192c558d74016721dae6235c18c0fbf Mon Sep 17 00:00:00 2001
From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Date: Thu, 8 Oct 2015 15:37:00 +0100
Subject: drm/i915: Remove wrong warning from i915_gem_context_clean

commit e9f24d5fb7cf3628b195b18ff3ac4e37937ceeae
Author: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Date:   Mon Oct 5 13:26:36 2015 +0100

    drm/i915: Clean up associated VMAs on context destruction

Introduced a wrong assumption that all contexts have a ppgtt
instance. This is not true when full PPGTT is not active so
remove the WARN_ON_ONCE from the context cleanup code.

Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Michel Thierry <michel.thierry@intel.com>
Reviewed-by: Michel Thierry <michel.thierry@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c
index 680b4c9..8c688a5 100644
--- a/drivers/gpu/drm/i915/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/i915_gem_context.c
@@ -138,7 +138,7 @@ static void i915_gem_context_clean(struct intel_context *ctx)
 	struct i915_hw_ppgtt *ppgtt = ctx->ppgtt;
 	struct i915_vma *vma, *next;
 
-	if (WARN_ON_ONCE(!ppgtt))
+	if (!ppgtt)
 		return;
 
 	WARN_ON(!list_empty(&ppgtt->base.active_list));
-- 
cgit v0.10.2


From 381e8ae377d9f0708a5073cb6ee2fa24ef303623 Mon Sep 17 00:00:00 2001
From: Tomas Elf <tomas.elf@intel.com>
Date: Thu, 8 Oct 2015 19:31:33 +0100
Subject: drm/i915: Early exit from semaphore_waits_for for execlist mode.

When submitting semaphores in execlist mode the hang checker crashes in this
function because it is only runnable in ring submission mode. The reason this
is of particular interest to the TDR patch series is because we use semaphores
as a mean to induce hangs during testing (which is the recommended way to
induce hangs for gen8+). It's not clear how this is supposed to work in
execlist mode since:

1. This function requires a ring buffer.

2. Retrieving a ring buffer in execlist mode requires us to retrieve the
corresponding context, which we get from a request.

3. Retieving a request from the hang checker is not straight-forward since that
requires us to grab the struct_mutex in order to synchronize against the
request retirement thread.

4. Grabbing the struct_mutex from the hang checker is nothing that we will do
since that puts us at risk of deadlock since a hung thread might be holding the
struct_mutex already.

Therefore it's not obvious how we're supposed to deal with this. For now, we're
doing an early exit from this function, which avoids any kernel panic situation
when running our own internal TDR ULT.

* v2: (Chris Wilson)
Turned the execlist mode check into a ringbuffer NULL check to make it more
submission mode agnostic and less of a layering violation.

Signed-off-by: Tomas Elf <tomas.elf@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index bd56ca6..637c132 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -2784,6 +2784,26 @@ semaphore_waits_for(struct intel_engine_cs *ring, u32 *seqno)
 	u64 offset = 0;
 	int i, backwards;
 
+	/*
+	 * This function does not support execlist mode - any attempt to
+	 * proceed further into this function will result in a kernel panic
+	 * when dereferencing ring->buffer, which is not set up in execlist
+	 * mode.
+	 *
+	 * The correct way of doing it would be to derive the currently
+	 * executing ring buffer from the current context, which is derived
+	 * from the currently running request. Unfortunately, to get the
+	 * current request we would have to grab the struct_mutex before doing
+	 * anything else, which would be ill-advised since some other thread
+	 * might have grabbed it already and managed to hang itself, causing
+	 * the hang checker to deadlock.
+	 *
+	 * Therefore, this function does not support execlist mode in its
+	 * current form. Just return NULL and move on.
+	 */
+	if (ring->buffer == NULL)
+		return NULL;
+
 	ipehr = I915_READ(RING_IPEHR(ring->mmio_base));
 	if (!ipehr_is_semaphore_wait(ring->dev, ipehr))
 		return NULL;
-- 
cgit v0.10.2


From 261a27d11fa1dec47c47ece6968eaaba55861eca Mon Sep 17 00:00:00 2001
From: Matt Roper <matthew.d.roper@intel.com>
Date: Thu, 8 Oct 2015 15:28:25 -0700
Subject: drm/i915: Partial revert of atomic watermark series

It's been reported that the atomic watermark series triggers some
regressions on SKL, which we haven't been able to track down yet.  Let's
temporarily revert these patches while we track down the root cause.

This commit squashes the reverts of:
  76305b1 drm/i915: Calculate watermark configuration during atomic check (v2)
  a4611e4 drm/i915: Don't set plane visible during HW readout if CRTC is off
  a28170f drm/i915: Calculate ILK-style watermarks during atomic check (v3)
  de4a9f8 drm/i915: Calculate pipe watermarks into CRTC state (v3)
  de165e0 drm/i915: Refactor ilk_update_wm (v3)

Reference: http://lists.freedesktop.org/archives/intel-gfx/2015-October/077190.html
Cc: "Zanoni, Paulo R" <paulo.r.zanoni@intel.com>
Cc: "Vetter, Daniel" <daniel.vetter@intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index b39b5cc..4feec19 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -627,8 +627,6 @@ struct drm_i915_display_funcs {
 			  int target, int refclk,
 			  struct dpll *match_clock,
 			  struct dpll *best_clock);
-	int (*compute_pipe_wm)(struct intel_crtc *crtc,
-			       struct drm_atomic_state *state);
 	void (*update_wm)(struct drm_crtc *crtc);
 	int (*modeset_calc_cdclk)(struct drm_atomic_state *state);
 	void (*modeset_commit_cdclk)(struct drm_atomic_state *state);
@@ -1692,13 +1690,6 @@ struct i915_execbuffer_params {
 	struct drm_i915_gem_request     *request;
 };
 
-/* used in computing the new watermarks state */
-struct intel_wm_config {
-	unsigned int num_pipes_active;
-	bool sprites_enabled;
-	bool sprites_scaled;
-};
-
 struct drm_i915_private {
 	struct drm_device *dev;
 	struct kmem_cache *objects;
@@ -1924,9 +1915,6 @@ struct drm_i915_private {
 		 */
 		uint16_t skl_latency[8];
 
-		/* Committed wm config */
-		struct intel_wm_config config;
-
 		/*
 		 * The skl_wm_values structure is a bit too big for stack
 		 * allocation, so we keep the staging struct where we store
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 6b3b65e..cddb0c6 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -11836,12 +11836,6 @@ static int intel_crtc_atomic_check(struct drm_crtc *crtc,
 	}
 
 	ret = 0;
-	if (dev_priv->display.compute_pipe_wm) {
-		ret = dev_priv->display.compute_pipe_wm(intel_crtc, state);
-		if (ret)
-			return ret;
-	}
-
 	if (INTEL_INFO(dev)->gen >= 9) {
 		if (mode_changed)
 			ret = skl_update_scaler_crtc(pipe_config);
@@ -13047,45 +13041,6 @@ static int intel_modeset_checks(struct drm_atomic_state *state)
 	return 0;
 }
 
-/*
- * Handle calculation of various watermark data at the end of the atomic check
- * phase.  The code here should be run after the per-crtc and per-plane 'check'
- * handlers to ensure that all derived state has been updated.
- */
-static void calc_watermark_data(struct drm_atomic_state *state)
-{
-	struct drm_device *dev = state->dev;
-	struct intel_atomic_state *intel_state = to_intel_atomic_state(state);
-	struct drm_crtc *crtc;
-	struct drm_crtc_state *cstate;
-	struct drm_plane *plane;
-	struct drm_plane_state *pstate;
-
-	/*
-	 * Calculate watermark configuration details now that derived
-	 * plane/crtc state is all properly updated.
-	 */
-	drm_for_each_crtc(crtc, dev) {
-		cstate = drm_atomic_get_existing_crtc_state(state, crtc) ?:
-			crtc->state;
-
-		if (cstate->active)
-			intel_state->wm_config.num_pipes_active++;
-	}
-	drm_for_each_legacy_plane(plane, dev) {
-		pstate = drm_atomic_get_existing_plane_state(state, plane) ?:
-			plane->state;
-
-		if (!to_intel_plane_state(pstate)->visible)
-			continue;
-
-		intel_state->wm_config.sprites_enabled = true;
-		if (pstate->crtc_w != pstate->src_w >> 16 ||
-		    pstate->crtc_h != pstate->src_h >> 16)
-			intel_state->wm_config.sprites_scaled = true;
-	}
-}
-
 /**
  * intel_atomic_check - validate state object
  * @dev: drm device
@@ -13094,7 +13049,6 @@ static void calc_watermark_data(struct drm_atomic_state *state)
 static int intel_atomic_check(struct drm_device *dev,
 			      struct drm_atomic_state *state)
 {
-	struct intel_atomic_state *intel_state = to_intel_atomic_state(state);
 	struct drm_crtc *crtc;
 	struct drm_crtc_state *crtc_state;
 	int ret, i;
@@ -13158,15 +13112,10 @@ static int intel_atomic_check(struct drm_device *dev,
 		if (ret)
 			return ret;
 	} else
-		intel_state->cdclk = to_i915(state->dev)->cdclk_freq;
-
-	ret = drm_atomic_helper_check_planes(state->dev, state);
-	if (ret)
-		return ret;
-
-	calc_watermark_data(state);
+		to_intel_atomic_state(state)->cdclk =
+			to_i915(state->dev)->cdclk_freq;
 
-	return 0;
+	return drm_atomic_helper_check_planes(state->dev, state);
 }
 
 /**
@@ -13206,7 +13155,6 @@ static int intel_atomic_commit(struct drm_device *dev,
 		return ret;
 
 	drm_atomic_helper_swap_state(dev, state);
-	dev_priv->wm.config = to_intel_atomic_state(state)->wm_config;
 
 	for_each_crtc_in_state(state, crtc, crtc_state, i) {
 		struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
@@ -15220,7 +15168,7 @@ static void readout_plane_state(struct intel_crtc *crtc)
 	struct intel_plane_state *plane_state =
 		to_intel_plane_state(primary->state);
 
-	plane_state->visible = crtc->active &&
+	plane_state->visible =
 		primary_get_hw_state(to_intel_plane(primary));
 
 	if (plane_state->visible)
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index e320825..91b6b40 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -250,7 +250,6 @@ struct intel_atomic_state {
 	unsigned int cdclk;
 	bool dpll_set;
 	struct intel_shared_dpll_config shared_dpll[I915_NUM_PLLS];
-	struct intel_wm_config wm_config;
 };
 
 struct intel_plane_state {
@@ -335,21 +334,6 @@ struct intel_crtc_scaler_state {
 /* drm_mode->private_flags */
 #define I915_MODE_FLAG_INHERITED 1
 
-struct intel_pipe_wm {
-	struct intel_wm_level wm[5];
-	uint32_t linetime;
-	bool fbc_wm_enabled;
-	bool pipe_enabled;
-	bool sprites_enabled;
-	bool sprites_scaled;
-};
-
-struct skl_pipe_wm {
-	struct skl_wm_level wm[8];
-	struct skl_wm_level trans_wm;
-	uint32_t linetime;
-};
-
 struct intel_crtc_state {
 	struct drm_crtc_state base;
 
@@ -487,17 +471,6 @@ struct intel_crtc_state {
 
 	/* IVB sprite scaling w/a (WaCxSRDisabledForSpriteScaling:ivb) */
 	bool disable_lp_wm;
-
-	struct {
-		/*
-		 * optimal watermarks, programmed post-vblank when this state
-		 * is committed
-		 */
-		union {
-			struct intel_pipe_wm ilk;
-			struct skl_pipe_wm skl;
-		} optimal;
-	} wm;
 };
 
 struct vlv_wm_state {
@@ -509,6 +482,15 @@ struct vlv_wm_state {
 	bool cxsr;
 };
 
+struct intel_pipe_wm {
+	struct intel_wm_level wm[5];
+	uint32_t linetime;
+	bool fbc_wm_enabled;
+	bool pipe_enabled;
+	bool sprites_enabled;
+	bool sprites_scaled;
+};
+
 struct intel_mmio_flip {
 	struct work_struct work;
 	struct drm_i915_private *i915;
@@ -516,6 +498,12 @@ struct intel_mmio_flip {
 	struct intel_crtc *crtc;
 };
 
+struct skl_pipe_wm {
+	struct skl_wm_level wm[8];
+	struct skl_wm_level trans_wm;
+	uint32_t linetime;
+};
+
 /*
  * Tracking of operations that need to be performed at the beginning/end of an
  * atomic commit, outside the atomic section where interrupts are disabled.
@@ -583,10 +571,9 @@ struct intel_crtc {
 	/* per-pipe watermark state */
 	struct {
 		/* watermarks currently being used  */
-		union {
-			struct intel_pipe_wm ilk;
-			struct skl_pipe_wm skl;
-		} active;
+		struct intel_pipe_wm active;
+		/* SKL wm values currently in use */
+		struct skl_pipe_wm skl_active;
 		/* allow CxSR on this pipe */
 		bool cxsr_allowed;
 	} wm;
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index 3f9b3c0..d031d74 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -1772,6 +1772,13 @@ struct ilk_wm_maximums {
 	uint16_t fbc;
 };
 
+/* used in computing the new watermarks state */
+struct intel_wm_config {
+	unsigned int num_pipes_active;
+	bool sprites_enabled;
+	bool sprites_scaled;
+};
+
 /*
  * For both WM_PIPE and WM_LP.
  * mem_value must be in 0.1us units.
@@ -2022,11 +2029,9 @@ static void ilk_compute_wm_level(const struct drm_i915_private *dev_priv,
 				 const struct intel_crtc *intel_crtc,
 				 int level,
 				 struct intel_crtc_state *cstate,
-				 struct intel_plane_state *pristate,
-				 struct intel_plane_state *sprstate,
-				 struct intel_plane_state *curstate,
 				 struct intel_wm_level *result)
 {
+	struct intel_plane *intel_plane;
 	uint16_t pri_latency = dev_priv->wm.pri_latency[level];
 	uint16_t spr_latency = dev_priv->wm.spr_latency[level];
 	uint16_t cur_latency = dev_priv->wm.cur_latency[level];
@@ -2038,11 +2043,29 @@ static void ilk_compute_wm_level(const struct drm_i915_private *dev_priv,
 		cur_latency *= 5;
 	}
 
-	result->pri_val = ilk_compute_pri_wm(cstate, pristate,
-					     pri_latency, level);
-	result->spr_val = ilk_compute_spr_wm(cstate, sprstate, spr_latency);
-	result->cur_val = ilk_compute_cur_wm(cstate, curstate, cur_latency);
-	result->fbc_val = ilk_compute_fbc_wm(cstate, pristate, result->pri_val);
+	for_each_intel_plane_on_crtc(dev_priv->dev, intel_crtc, intel_plane) {
+		struct intel_plane_state *pstate =
+			to_intel_plane_state(intel_plane->base.state);
+
+		switch (intel_plane->base.type) {
+		case DRM_PLANE_TYPE_PRIMARY:
+			result->pri_val = ilk_compute_pri_wm(cstate, pstate,
+							     pri_latency,
+							     level);
+			result->fbc_val = ilk_compute_fbc_wm(cstate, pstate,
+							     result->pri_val);
+			break;
+		case DRM_PLANE_TYPE_OVERLAY:
+			result->spr_val = ilk_compute_spr_wm(cstate, pstate,
+							     spr_latency);
+			break;
+		case DRM_PLANE_TYPE_CURSOR:
+			result->cur_val = ilk_compute_cur_wm(cstate, pstate,
+							     cur_latency);
+			break;
+		}
+	}
+
 	result->enable = true;
 }
 
@@ -2301,19 +2324,34 @@ static void skl_setup_wm_latency(struct drm_device *dev)
 	intel_print_wm_latency(dev, "Gen9 Plane", dev_priv->wm.skl_latency);
 }
 
+static void ilk_compute_wm_config(struct drm_device *dev,
+				  struct intel_wm_config *config)
+{
+	struct intel_crtc *intel_crtc;
+
+	/* Compute the currently _active_ config */
+	for_each_intel_crtc(dev, intel_crtc) {
+		const struct intel_pipe_wm *wm = &intel_crtc->wm.active;
+
+		if (!wm->pipe_enabled)
+			continue;
+
+		config->sprites_enabled |= wm->sprites_enabled;
+		config->sprites_scaled |= wm->sprites_scaled;
+		config->num_pipes_active++;
+	}
+}
+
 /* Compute new watermarks for the pipe */
-static int ilk_compute_pipe_wm(struct intel_crtc *intel_crtc,
-			       struct drm_atomic_state *state)
+static bool intel_compute_pipe_wm(struct intel_crtc_state *cstate,
+				  struct intel_pipe_wm *pipe_wm)
 {
-	struct intel_pipe_wm *pipe_wm;
-	struct drm_device *dev = intel_crtc->base.dev;
+	struct drm_crtc *crtc = cstate->base.crtc;
+	struct drm_device *dev = crtc->dev;
 	const struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_crtc_state *cstate = NULL;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct intel_plane *intel_plane;
-	struct drm_plane_state *ps;
-	struct intel_plane_state *pristate = NULL;
 	struct intel_plane_state *sprstate = NULL;
-	struct intel_plane_state *curstate = NULL;
 	int level, max_level = ilk_wm_max_level(dev);
 	/* LP0 watermark maximums depend on this pipe alone */
 	struct intel_wm_config config = {
@@ -2321,24 +2359,11 @@ static int ilk_compute_pipe_wm(struct intel_crtc *intel_crtc,
 	};
 	struct ilk_wm_maximums max;
 
-	cstate = intel_atomic_get_crtc_state(state, intel_crtc);
-	if (IS_ERR(cstate))
-		return PTR_ERR(cstate);
-
-	pipe_wm = &cstate->wm.optimal.ilk;
-
 	for_each_intel_plane_on_crtc(dev, intel_crtc, intel_plane) {
-		ps = drm_atomic_get_plane_state(state,
-						&intel_plane->base);
-		if (IS_ERR(ps))
-			return PTR_ERR(ps);
-
-		if (intel_plane->base.type == DRM_PLANE_TYPE_PRIMARY)
-			pristate = to_intel_plane_state(ps);
-		else if (intel_plane->base.type == DRM_PLANE_TYPE_OVERLAY)
-			sprstate = to_intel_plane_state(ps);
-		else if (intel_plane->base.type == DRM_PLANE_TYPE_CURSOR)
-			curstate = to_intel_plane_state(ps);
+		if (intel_plane->base.type == DRM_PLANE_TYPE_OVERLAY) {
+			sprstate = to_intel_plane_state(intel_plane->base.state);
+			break;
+		}
 	}
 
 	config.sprites_enabled = sprstate->visible;
@@ -2347,7 +2372,7 @@ static int ilk_compute_pipe_wm(struct intel_crtc *intel_crtc,
 		drm_rect_height(&sprstate->dst) != drm_rect_height(&sprstate->src) >> 16);
 
 	pipe_wm->pipe_enabled = cstate->base.active;
-	pipe_wm->sprites_enabled = config.sprites_enabled;
+	pipe_wm->sprites_enabled = sprstate->visible;
 	pipe_wm->sprites_scaled = config.sprites_scaled;
 
 	/* ILK/SNB: LP2+ watermarks only w/o sprites */
@@ -2358,27 +2383,24 @@ static int ilk_compute_pipe_wm(struct intel_crtc *intel_crtc,
 	if (config.sprites_scaled)
 		max_level = 0;
 
-	ilk_compute_wm_level(dev_priv, intel_crtc, 0, cstate,
-			     pristate, sprstate, curstate, &pipe_wm->wm[0]);
+	ilk_compute_wm_level(dev_priv, intel_crtc, 0, cstate, &pipe_wm->wm[0]);
 
 	if (IS_HASWELL(dev) || IS_BROADWELL(dev))
-		pipe_wm->linetime = hsw_compute_linetime_wm(dev,
-							    &intel_crtc->base);
+		pipe_wm->linetime = hsw_compute_linetime_wm(dev, crtc);
 
 	/* LP0 watermarks always use 1/2 DDB partitioning */
 	ilk_compute_wm_maximums(dev, 0, &config, INTEL_DDB_PART_1_2, &max);
 
 	/* At least LP0 must be valid */
 	if (!ilk_validate_wm_level(0, &max, &pipe_wm->wm[0]))
-		return -EINVAL;
+		return false;
 
 	ilk_compute_wm_reg_maximums(dev, 1, &max);
 
 	for (level = 1; level <= max_level; level++) {
 		struct intel_wm_level wm = {};
 
-		ilk_compute_wm_level(dev_priv, intel_crtc, level, cstate,
-				     pristate, sprstate, curstate, &wm);
+		ilk_compute_wm_level(dev_priv, intel_crtc, level, cstate, &wm);
 
 		/*
 		 * Disable any watermark level that exceeds the
@@ -2391,7 +2413,7 @@ static int ilk_compute_pipe_wm(struct intel_crtc *intel_crtc,
 		pipe_wm->wm[level] = wm;
 	}
 
-	return 0;
+	return true;
 }
 
 /*
@@ -2406,9 +2428,7 @@ static void ilk_merge_wm_level(struct drm_device *dev,
 	ret_wm->enable = true;
 
 	for_each_intel_crtc(dev, intel_crtc) {
-		const struct intel_crtc_state *cstate =
-			to_intel_crtc_state(intel_crtc->base.state);
-		const struct intel_pipe_wm *active = &cstate->wm.optimal.ilk;
+		const struct intel_pipe_wm *active = &intel_crtc->wm.active;
 		const struct intel_wm_level *wm = &active->wm[level];
 
 		if (!active->pipe_enabled)
@@ -2556,15 +2576,14 @@ static void ilk_compute_wm_results(struct drm_device *dev,
 
 	/* LP0 register values */
 	for_each_intel_crtc(dev, intel_crtc) {
-		const struct intel_crtc_state *cstate =
-			to_intel_crtc_state(intel_crtc->base.state);
 		enum pipe pipe = intel_crtc->pipe;
-		const struct intel_wm_level *r = &cstate->wm.optimal.ilk.wm[0];
+		const struct intel_wm_level *r =
+			&intel_crtc->wm.active.wm[0];
 
 		if (WARN_ON(!r->enable))
 			continue;
 
-		results->wm_linetime[pipe] = cstate->wm.optimal.ilk.linetime;
+		results->wm_linetime[pipe] = intel_crtc->wm.active.linetime;
 
 		results->wm_pipe[pipe] =
 			(r->pri_val << WM0_PIPE_PLANE_SHIFT) |
@@ -2946,12 +2965,11 @@ skl_get_total_relative_data_rate(const struct intel_crtc_state *cstate)
 
 static void
 skl_allocate_pipe_ddb(struct intel_crtc_state *cstate,
+		      const struct intel_wm_config *config,
 		      struct skl_ddb_allocation *ddb /* out */)
 {
 	struct drm_crtc *crtc = cstate->base.crtc;
 	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = to_i915(dev);
-	struct intel_wm_config *config = &dev_priv->wm.config;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct intel_plane *intel_plane;
 	enum pipe pipe = intel_crtc->pipe;
@@ -3126,6 +3144,15 @@ static bool skl_ddb_allocation_changed(const struct skl_ddb_allocation *new_ddb,
 	return false;
 }
 
+static void skl_compute_wm_global_parameters(struct drm_device *dev,
+					     struct intel_wm_config *config)
+{
+	struct drm_crtc *crtc;
+
+	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
+		config->num_pipes_active += to_intel_crtc(crtc)->active;
+}
+
 static bool skl_compute_plane_wm(const struct drm_i915_private *dev_priv,
 				 struct intel_crtc_state *cstate,
 				 struct intel_plane *intel_plane,
@@ -3530,25 +3557,27 @@ static void skl_flush_wm_values(struct drm_i915_private *dev_priv,
 }
 
 static bool skl_update_pipe_wm(struct drm_crtc *crtc,
+			       struct intel_wm_config *config,
 			       struct skl_ddb_allocation *ddb, /* out */
 			       struct skl_pipe_wm *pipe_wm /* out */)
 {
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct intel_crtc_state *cstate = to_intel_crtc_state(crtc->state);
 
-	skl_allocate_pipe_ddb(cstate, ddb);
+	skl_allocate_pipe_ddb(cstate, config, ddb);
 	skl_compute_pipe_wm(cstate, ddb, pipe_wm);
 
-	if (!memcmp(&intel_crtc->wm.active.skl, pipe_wm, sizeof(*pipe_wm)))
+	if (!memcmp(&intel_crtc->wm.skl_active, pipe_wm, sizeof(*pipe_wm)))
 		return false;
 
-	intel_crtc->wm.active.skl = *pipe_wm;
+	intel_crtc->wm.skl_active = *pipe_wm;
 
 	return true;
 }
 
 static void skl_update_other_pipe_wm(struct drm_device *dev,
 				     struct drm_crtc *crtc,
+				     struct intel_wm_config *config,
 				     struct skl_wm_values *r)
 {
 	struct intel_crtc *intel_crtc;
@@ -3578,7 +3607,7 @@ static void skl_update_other_pipe_wm(struct drm_device *dev,
 		if (!intel_crtc->active)
 			continue;
 
-		wm_changed = skl_update_pipe_wm(&intel_crtc->base,
+		wm_changed = skl_update_pipe_wm(&intel_crtc->base, config,
 						&r->ddb, &pipe_wm);
 
 		/*
@@ -3619,8 +3648,8 @@ static void skl_update_wm(struct drm_crtc *crtc)
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct skl_wm_values *results = &dev_priv->wm.skl_results;
-	struct intel_crtc_state *cstate = to_intel_crtc_state(crtc->state);
-	struct skl_pipe_wm *pipe_wm = &cstate->wm.optimal.skl;
+	struct skl_pipe_wm pipe_wm = {};
+	struct intel_wm_config config = {};
 
 
 	/* Clear all dirty flags */
@@ -3628,13 +3657,15 @@ static void skl_update_wm(struct drm_crtc *crtc)
 
 	skl_clear_wm(results, intel_crtc->pipe);
 
-	if (!skl_update_pipe_wm(crtc, &results->ddb, pipe_wm))
+	skl_compute_wm_global_parameters(dev, &config);
+
+	if (!skl_update_pipe_wm(crtc, &config, &results->ddb, &pipe_wm))
 		return;
 
-	skl_compute_wm_results(dev, pipe_wm, results, intel_crtc);
+	skl_compute_wm_results(dev, &pipe_wm, results, intel_crtc);
 	results->dirty[intel_crtc->pipe] = true;
 
-	skl_update_other_pipe_wm(dev, crtc, results);
+	skl_update_other_pipe_wm(dev, crtc, &config, results);
 	skl_write_wm_values(dev_priv, results);
 	skl_flush_wm_values(dev_priv, results);
 
@@ -3642,23 +3673,50 @@ static void skl_update_wm(struct drm_crtc *crtc)
 	dev_priv->wm.skl_hw = *results;
 }
 
-static void ilk_program_watermarks(struct drm_i915_private *dev_priv)
+static void ilk_update_wm(struct drm_crtc *crtc)
 {
-	struct drm_device *dev = dev_priv->dev;
-	struct intel_pipe_wm lp_wm_1_2 = {}, lp_wm_5_6 = {}, *best_lp_wm;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	struct intel_crtc_state *cstate = to_intel_crtc_state(crtc->state);
+	struct drm_device *dev = crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct ilk_wm_maximums max;
-	struct intel_wm_config *config = &dev_priv->wm.config;
 	struct ilk_wm_values results = {};
 	enum intel_ddb_partitioning partitioning;
+	struct intel_pipe_wm pipe_wm = {};
+	struct intel_pipe_wm lp_wm_1_2 = {}, lp_wm_5_6 = {}, *best_lp_wm;
+	struct intel_wm_config config = {};
 
-	ilk_compute_wm_maximums(dev, 1, config, INTEL_DDB_PART_1_2, &max);
-	ilk_wm_merge(dev, config, &max, &lp_wm_1_2);
+	WARN_ON(cstate->base.active != intel_crtc->active);
+
+	/*
+	 * IVB workaround: must disable low power watermarks for at least
+	 * one frame before enabling scaling.  LP watermarks can be re-enabled
+	 * when scaling is disabled.
+	 *
+	 * WaCxSRDisabledForSpriteScaling:ivb
+	 */
+	if (cstate->disable_lp_wm) {
+		ilk_disable_lp_wm(dev);
+		intel_wait_for_vblank(dev, intel_crtc->pipe);
+	}
+
+	intel_compute_pipe_wm(cstate, &pipe_wm);
+
+	if (!memcmp(&intel_crtc->wm.active, &pipe_wm, sizeof(pipe_wm)))
+		return;
+
+	intel_crtc->wm.active = pipe_wm;
+
+	ilk_compute_wm_config(dev, &config);
+
+	ilk_compute_wm_maximums(dev, 1, &config, INTEL_DDB_PART_1_2, &max);
+	ilk_wm_merge(dev, &config, &max, &lp_wm_1_2);
 
 	/* 5/6 split only in single pipe config on IVB+ */
 	if (INTEL_INFO(dev)->gen >= 7 &&
-	    config->num_pipes_active == 1 && config->sprites_enabled) {
-		ilk_compute_wm_maximums(dev, 1, config, INTEL_DDB_PART_5_6, &max);
-		ilk_wm_merge(dev, config, &max, &lp_wm_5_6);
+	    config.num_pipes_active == 1 && config.sprites_enabled) {
+		ilk_compute_wm_maximums(dev, 1, &config, INTEL_DDB_PART_5_6, &max);
+		ilk_wm_merge(dev, &config, &max, &lp_wm_5_6);
 
 		best_lp_wm = ilk_find_best_result(dev, &lp_wm_1_2, &lp_wm_5_6);
 	} else {
@@ -3673,31 +3731,6 @@ static void ilk_program_watermarks(struct drm_i915_private *dev_priv)
 	ilk_write_wm_values(dev_priv, &results);
 }
 
-static void ilk_update_wm(struct drm_crtc *crtc)
-{
-	struct drm_i915_private *dev_priv = to_i915(crtc->dev);
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	struct intel_crtc_state *cstate = to_intel_crtc_state(crtc->state);
-
-	WARN_ON(cstate->base.active != intel_crtc->active);
-
-	/*
-	 * IVB workaround: must disable low power watermarks for at least
-	 * one frame before enabling scaling.  LP watermarks can be re-enabled
-	 * when scaling is disabled.
-	 *
-	 * WaCxSRDisabledForSpriteScaling:ivb
-	 */
-	if (cstate->disable_lp_wm) {
-		ilk_disable_lp_wm(crtc->dev);
-		intel_wait_for_vblank(crtc->dev, intel_crtc->pipe);
-	}
-
-	intel_crtc->wm.active.ilk = cstate->wm.optimal.ilk;
-
-	ilk_program_watermarks(dev_priv);
-}
-
 static void skl_pipe_wm_active_state(uint32_t val,
 				     struct skl_pipe_wm *active,
 				     bool is_transwm,
@@ -3748,8 +3781,7 @@ static void skl_pipe_wm_get_hw_state(struct drm_crtc *crtc)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct skl_wm_values *hw = &dev_priv->wm.skl_hw;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	struct intel_crtc_state *cstate = to_intel_crtc_state(crtc->state);
-	struct skl_pipe_wm *active = &cstate->wm.optimal.skl;
+	struct skl_pipe_wm *active = &intel_crtc->wm.skl_active;
 	enum pipe pipe = intel_crtc->pipe;
 	int level, i, max_level;
 	uint32_t temp;
@@ -3793,8 +3825,6 @@ static void skl_pipe_wm_get_hw_state(struct drm_crtc *crtc)
 
 	temp = hw->plane_trans[pipe][PLANE_CURSOR];
 	skl_pipe_wm_active_state(temp, active, true, true, i, 0);
-
-	intel_crtc->wm.active.skl = *active;
 }
 
 void skl_wm_get_hw_state(struct drm_device *dev)
@@ -3814,8 +3844,7 @@ static void ilk_pipe_wm_get_hw_state(struct drm_crtc *crtc)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct ilk_wm_values *hw = &dev_priv->wm.hw;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	struct intel_crtc_state *cstate = to_intel_crtc_state(crtc->state);
-	struct intel_pipe_wm *active = &cstate->wm.optimal.ilk;
+	struct intel_pipe_wm *active = &intel_crtc->wm.active;
 	enum pipe pipe = intel_crtc->pipe;
 	static const unsigned int wm0_pipe_reg[] = {
 		[PIPE_A] = WM0_PIPEA_ILK,
@@ -3854,8 +3883,6 @@ static void ilk_pipe_wm_get_hw_state(struct drm_crtc *crtc)
 		for (level = 0; level <= max_level; level++)
 			active->wm[level].enable = true;
 	}
-
-	intel_crtc->wm.active.ilk = *active;
 }
 
 #define _FW_WM(value, plane) \
@@ -7003,7 +7030,6 @@ void intel_init_pm(struct drm_device *dev)
 		    (!IS_GEN5(dev) && dev_priv->wm.pri_latency[0] &&
 		     dev_priv->wm.spr_latency[0] && dev_priv->wm.cur_latency[0])) {
 			dev_priv->display.update_wm = ilk_update_wm;
-			dev_priv->display.compute_pipe_wm = ilk_compute_pipe_wm;
 		} else {
 			DRM_DEBUG_KMS("Failed to read display plane latency. "
 				      "Disable CxSR\n");
-- 
cgit v0.10.2


From 7ed6c665e19d4c0456d01c01147e82563c482c4d Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Thu, 7 Nov 2013 16:01:45 +0000
Subject: drm: bridge/dw_hdmi-ahb-audio: add audio driver

Add ALSA based HDMI AHB audio driver for dw_hdmi.  The only buffer
format supported by the hardware is its own special IEC958 based format,
which is not compatible with any ALSA format.  To avoid doing too much
data manipulation within the driver, we support only ALSAs IEC958 LE and
24-bit PCM formats for 2 to 6 channels, which we convert to its hardware
format.

A more desirable solution would be to have this conversion in userspace,
but ALSA does not appear to allow such transformations outside of
libasound itself.

Reviewed-by: Takashi Iwai <tiwai@suse.de>
Tested-by: Fabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/bridge/Kconfig b/drivers/gpu/drm/bridge/Kconfig
index 2de52a5..0d0a4f7 100644
--- a/drivers/gpu/drm/bridge/Kconfig
+++ b/drivers/gpu/drm/bridge/Kconfig
@@ -11,6 +11,17 @@ config DRM_DW_HDMI
 	tristate
 	select DRM_KMS_HELPER
 
+config DRM_DW_HDMI_AHB_AUDIO
+	tristate "Synopsis Designware AHB Audio interface"
+	depends on DRM_DW_HDMI && SND
+	select SND_PCM
+	select SND_PCM_IEC958
+	help
+	  Support the AHB Audio interface which is part of the Synopsis
+	  Designware HDMI block.  This is used in conjunction with
+	  the i.MX6 HDMI driver.
+
+
 config DRM_NXP_PTN3460
 	tristate "NXP PTN3460 DP/LVDS bridge"
 	depends on OF
diff --git a/drivers/gpu/drm/bridge/Makefile b/drivers/gpu/drm/bridge/Makefile
index e2eef1c..d4e28be 100644
--- a/drivers/gpu/drm/bridge/Makefile
+++ b/drivers/gpu/drm/bridge/Makefile
@@ -1,5 +1,6 @@
 ccflags-y := -Iinclude/drm
 
 obj-$(CONFIG_DRM_DW_HDMI) += dw_hdmi.o
+obj-$(CONFIG_DRM_DW_HDMI_AHB_AUDIO) += dw_hdmi-ahb-audio.o
 obj-$(CONFIG_DRM_NXP_PTN3460) += nxp-ptn3460.o
 obj-$(CONFIG_DRM_PARADE_PS8622) += parade-ps8622.o
diff --git a/drivers/gpu/drm/bridge/dw_hdmi-ahb-audio.c b/drivers/gpu/drm/bridge/dw_hdmi-ahb-audio.c
new file mode 100644
index 0000000..bf379310
--- /dev/null
+++ b/drivers/gpu/drm/bridge/dw_hdmi-ahb-audio.c
@@ -0,0 +1,579 @@
+/*
+ * DesignWare HDMI audio driver
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * Written and tested against the Designware HDMI Tx found in iMX6.
+ */
+#include <linux/io.h>
+#include <linux/interrupt.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <drm/bridge/dw_hdmi.h>
+
+#include <sound/asoundef.h>
+#include <sound/core.h>
+#include <sound/initval.h>
+#include <sound/pcm.h>
+#include <sound/pcm_iec958.h>
+
+#include "dw_hdmi-audio.h"
+
+#define DRIVER_NAME "dw-hdmi-ahb-audio"
+
+/* Provide some bits rather than bit offsets */
+enum {
+	HDMI_AHB_DMA_CONF0_SW_FIFO_RST = BIT(7),
+	HDMI_AHB_DMA_CONF0_EN_HLOCK = BIT(3),
+	HDMI_AHB_DMA_START_START = BIT(0),
+	HDMI_AHB_DMA_STOP_STOP = BIT(0),
+	HDMI_IH_MUTE_AHBDMAAUD_STAT0_ERROR = BIT(5),
+	HDMI_IH_MUTE_AHBDMAAUD_STAT0_LOST = BIT(4),
+	HDMI_IH_MUTE_AHBDMAAUD_STAT0_RETRY = BIT(3),
+	HDMI_IH_MUTE_AHBDMAAUD_STAT0_DONE = BIT(2),
+	HDMI_IH_MUTE_AHBDMAAUD_STAT0_BUFFFULL = BIT(1),
+	HDMI_IH_MUTE_AHBDMAAUD_STAT0_BUFFEMPTY = BIT(0),
+	HDMI_IH_MUTE_AHBDMAAUD_STAT0_ALL =
+		HDMI_IH_MUTE_AHBDMAAUD_STAT0_ERROR |
+		HDMI_IH_MUTE_AHBDMAAUD_STAT0_LOST |
+		HDMI_IH_MUTE_AHBDMAAUD_STAT0_RETRY |
+		HDMI_IH_MUTE_AHBDMAAUD_STAT0_DONE |
+		HDMI_IH_MUTE_AHBDMAAUD_STAT0_BUFFFULL |
+		HDMI_IH_MUTE_AHBDMAAUD_STAT0_BUFFEMPTY,
+	HDMI_IH_AHBDMAAUD_STAT0_ERROR = BIT(5),
+	HDMI_IH_AHBDMAAUD_STAT0_LOST = BIT(4),
+	HDMI_IH_AHBDMAAUD_STAT0_RETRY = BIT(3),
+	HDMI_IH_AHBDMAAUD_STAT0_DONE = BIT(2),
+	HDMI_IH_AHBDMAAUD_STAT0_BUFFFULL = BIT(1),
+	HDMI_IH_AHBDMAAUD_STAT0_BUFFEMPTY = BIT(0),
+	HDMI_IH_AHBDMAAUD_STAT0_ALL =
+		HDMI_IH_AHBDMAAUD_STAT0_ERROR |
+		HDMI_IH_AHBDMAAUD_STAT0_LOST |
+		HDMI_IH_AHBDMAAUD_STAT0_RETRY |
+		HDMI_IH_AHBDMAAUD_STAT0_DONE |
+		HDMI_IH_AHBDMAAUD_STAT0_BUFFFULL |
+		HDMI_IH_AHBDMAAUD_STAT0_BUFFEMPTY,
+	HDMI_AHB_DMA_CONF0_INCR16 = 2 << 1,
+	HDMI_AHB_DMA_CONF0_INCR8 = 1 << 1,
+	HDMI_AHB_DMA_CONF0_INCR4 = 0,
+	HDMI_AHB_DMA_CONF0_BURST_MODE = BIT(0),
+	HDMI_AHB_DMA_MASK_DONE = BIT(7),
+	HDMI_REVISION_ID = 0x0001,
+	HDMI_IH_AHBDMAAUD_STAT0 = 0x0109,
+	HDMI_IH_MUTE_AHBDMAAUD_STAT0 = 0x0189,
+	HDMI_AHB_DMA_CONF0 = 0x3600,
+	HDMI_AHB_DMA_START = 0x3601,
+	HDMI_AHB_DMA_STOP = 0x3602,
+	HDMI_AHB_DMA_THRSLD = 0x3603,
+	HDMI_AHB_DMA_STRADDR0 = 0x3604,
+	HDMI_AHB_DMA_STPADDR0 = 0x3608,
+	HDMI_AHB_DMA_MASK = 0x3614,
+	HDMI_AHB_DMA_POL = 0x3615,
+	HDMI_AHB_DMA_CONF1 = 0x3616,
+	HDMI_AHB_DMA_BUFFPOL = 0x361a,
+};
+
+struct snd_dw_hdmi {
+	struct snd_card *card;
+	struct snd_pcm *pcm;
+	spinlock_t lock;
+	struct dw_hdmi_audio_data data;
+	struct snd_pcm_substream *substream;
+	void (*reformat)(struct snd_dw_hdmi *, size_t, size_t);
+	void *buf_src;
+	void *buf_dst;
+	dma_addr_t buf_addr;
+	unsigned buf_offset;
+	unsigned buf_period;
+	unsigned buf_size;
+	unsigned channels;
+	u8 revision;
+	u8 iec_offset;
+	u8 cs[192][8];
+};
+
+static void dw_hdmi_writel(u32 val, void __iomem *ptr)
+{
+	writeb_relaxed(val, ptr);
+	writeb_relaxed(val >> 8, ptr + 1);
+	writeb_relaxed(val >> 16, ptr + 2);
+	writeb_relaxed(val >> 24, ptr + 3);
+}
+
+/*
+ * Convert to hardware format: The userspace buffer contains IEC958 samples,
+ * with the PCUV bits in bits 31..28 and audio samples in bits 27..4.  We
+ * need these to be in bits 27..24, with the IEC B bit in bit 28, and audio
+ * samples in 23..0.
+ *
+ * Default preamble in bits 3..0: 8 = block start, 4 = even 2 = odd
+ *
+ * Ideally, we could do with having the data properly formatted in userspace.
+ */
+static void dw_hdmi_reformat_iec958(struct snd_dw_hdmi *dw,
+	size_t offset, size_t bytes)
+{
+	u32 *src = dw->buf_src + offset;
+	u32 *dst = dw->buf_dst + offset;
+	u32 *end = dw->buf_src + offset + bytes;
+
+	do {
+		u32 b, sample = *src++;
+
+		b = (sample & 8) << (28 - 3);
+
+		sample >>= 4;
+
+		*dst++ = sample | b;
+	} while (src < end);
+}
+
+static u32 parity(u32 sample)
+{
+	sample ^= sample >> 16;
+	sample ^= sample >> 8;
+	sample ^= sample >> 4;
+	sample ^= sample >> 2;
+	sample ^= sample >> 1;
+	return (sample & 1) << 27;
+}
+
+static void dw_hdmi_reformat_s24(struct snd_dw_hdmi *dw,
+	size_t offset, size_t bytes)
+{
+	u32 *src = dw->buf_src + offset;
+	u32 *dst = dw->buf_dst + offset;
+	u32 *end = dw->buf_src + offset + bytes;
+
+	do {
+		unsigned i;
+		u8 *cs;
+
+		cs = dw->cs[dw->iec_offset++];
+		if (dw->iec_offset >= 192)
+			dw->iec_offset = 0;
+
+		i = dw->channels;
+		do {
+			u32 sample = *src++;
+
+			sample &= ~0xff000000;
+			sample |= *cs++ << 24;
+			sample |= parity(sample & ~0xf8000000);
+
+			*dst++ = sample;
+		} while (--i);
+	} while (src < end);
+}
+
+static void dw_hdmi_create_cs(struct snd_dw_hdmi *dw,
+	struct snd_pcm_runtime *runtime)
+{
+	u8 cs[4];
+	unsigned ch, i, j;
+
+	snd_pcm_create_iec958_consumer(runtime, cs, sizeof(cs));
+
+	memset(dw->cs, 0, sizeof(dw->cs));
+
+	for (ch = 0; ch < 8; ch++) {
+		cs[2] &= ~IEC958_AES2_CON_CHANNEL;
+		cs[2] |= (ch + 1) << 4;
+
+		for (i = 0; i < ARRAY_SIZE(cs); i++) {
+			unsigned c = cs[i];
+
+			for (j = 0; j < 8; j++, c >>= 1)
+				dw->cs[i * 8 + j][ch] = (c & 1) << 2;
+		}
+	}
+	dw->cs[0][0] |= BIT(4);
+}
+
+static void dw_hdmi_start_dma(struct snd_dw_hdmi *dw)
+{
+	void __iomem *base = dw->data.base;
+	unsigned offset = dw->buf_offset;
+	unsigned period = dw->buf_period;
+	u32 start, stop;
+
+	dw->reformat(dw, offset, period);
+
+	/* Clear all irqs before enabling irqs and starting DMA */
+	writeb_relaxed(HDMI_IH_AHBDMAAUD_STAT0_ALL,
+		       base + HDMI_IH_AHBDMAAUD_STAT0);
+
+	start = dw->buf_addr + offset;
+	stop = start + period - 1;
+
+	/* Setup the hardware start/stop addresses */
+	dw_hdmi_writel(start, base + HDMI_AHB_DMA_STRADDR0);
+	dw_hdmi_writel(stop, base + HDMI_AHB_DMA_STPADDR0);
+
+	writeb_relaxed((u8)~HDMI_AHB_DMA_MASK_DONE, base + HDMI_AHB_DMA_MASK);
+	writeb(HDMI_AHB_DMA_START_START, base + HDMI_AHB_DMA_START);
+
+	offset += period;
+	if (offset >= dw->buf_size)
+		offset = 0;
+	dw->buf_offset = offset;
+}
+
+static void dw_hdmi_stop_dma(struct snd_dw_hdmi *dw)
+{
+	/* Disable interrupts before disabling DMA */
+	writeb_relaxed(~0, dw->data.base + HDMI_AHB_DMA_MASK);
+	writeb_relaxed(HDMI_AHB_DMA_STOP_STOP, dw->data.base + HDMI_AHB_DMA_STOP);
+}
+
+static irqreturn_t snd_dw_hdmi_irq(int irq, void *data)
+{
+	struct snd_dw_hdmi *dw = data;
+	struct snd_pcm_substream *substream;
+	unsigned stat;
+
+	stat = readb_relaxed(dw->data.base + HDMI_IH_AHBDMAAUD_STAT0);
+	if (!stat)
+		return IRQ_NONE;
+
+	writeb_relaxed(stat, dw->data.base + HDMI_IH_AHBDMAAUD_STAT0);
+
+	substream = dw->substream;
+	if (stat & HDMI_IH_AHBDMAAUD_STAT0_DONE && substream) {
+		snd_pcm_period_elapsed(substream);
+
+		spin_lock(&dw->lock);
+		if (dw->substream)
+			dw_hdmi_start_dma(dw);
+		spin_unlock(&dw->lock);
+	}
+
+	return IRQ_HANDLED;
+}
+
+static struct snd_pcm_hardware dw_hdmi_hw = {
+	.info = SNDRV_PCM_INFO_INTERLEAVED |
+		SNDRV_PCM_INFO_BLOCK_TRANSFER |
+		SNDRV_PCM_INFO_MMAP |
+		SNDRV_PCM_INFO_MMAP_VALID,
+	.formats = SNDRV_PCM_FMTBIT_IEC958_SUBFRAME_LE |
+		   SNDRV_PCM_FMTBIT_S24_LE,
+	.rates = SNDRV_PCM_RATE_32000 |
+		 SNDRV_PCM_RATE_44100 |
+		 SNDRV_PCM_RATE_48000 |
+		 SNDRV_PCM_RATE_88200 |
+		 SNDRV_PCM_RATE_96000 |
+		 SNDRV_PCM_RATE_176400 |
+		 SNDRV_PCM_RATE_192000,
+	.channels_min = 2,
+	.channels_max = 8,
+	.buffer_bytes_max = 64 * 1024,
+	.period_bytes_min = 256,
+	.period_bytes_max = 8192,	/* ERR004323: must limit to 8k */
+	.periods_min = 2,
+	.periods_max = 16,
+	.fifo_size = 0,
+};
+
+static int dw_hdmi_open(struct snd_pcm_substream *substream)
+{
+	struct snd_pcm_runtime *runtime = substream->runtime;
+	struct snd_dw_hdmi *dw = substream->private_data;
+	void __iomem *base = dw->data.base;
+	int ret;
+
+	runtime->hw = dw_hdmi_hw;
+
+	ret = snd_pcm_limit_hw_rates(runtime);
+	if (ret < 0)
+		return ret;
+
+	ret = snd_pcm_hw_constraint_integer(runtime, SNDRV_PCM_HW_PARAM_PERIODS);
+	if (ret < 0)
+		return ret;
+
+	/* Clear FIFO */
+	writeb_relaxed(HDMI_AHB_DMA_CONF0_SW_FIFO_RST,
+		       base + HDMI_AHB_DMA_CONF0);
+
+	/* Configure interrupt polarities */
+	writeb_relaxed(~0, base + HDMI_AHB_DMA_POL);
+	writeb_relaxed(~0, base + HDMI_AHB_DMA_BUFFPOL);
+
+	/* Keep interrupts masked, and clear any pending */
+	writeb_relaxed(~0, base + HDMI_AHB_DMA_MASK);
+	writeb_relaxed(~0, base + HDMI_IH_AHBDMAAUD_STAT0);
+
+	ret = request_irq(dw->data.irq, snd_dw_hdmi_irq, IRQF_SHARED,
+			  "dw-hdmi-audio", dw);
+	if (ret)
+		return ret;
+
+	/* Un-mute done interrupt */
+	writeb_relaxed(HDMI_IH_MUTE_AHBDMAAUD_STAT0_ALL &
+		       ~HDMI_IH_MUTE_AHBDMAAUD_STAT0_DONE,
+		       base + HDMI_IH_MUTE_AHBDMAAUD_STAT0);
+
+	return 0;
+}
+
+static int dw_hdmi_close(struct snd_pcm_substream *substream)
+{
+	struct snd_dw_hdmi *dw = substream->private_data;
+
+	/* Mute all interrupts */
+	writeb_relaxed(HDMI_IH_MUTE_AHBDMAAUD_STAT0_ALL,
+		       dw->data.base + HDMI_IH_MUTE_AHBDMAAUD_STAT0);
+
+	free_irq(dw->data.irq, dw);
+
+	return 0;
+}
+
+static int dw_hdmi_hw_free(struct snd_pcm_substream *substream)
+{
+	return snd_pcm_lib_free_vmalloc_buffer(substream);
+}
+
+static int dw_hdmi_hw_params(struct snd_pcm_substream *substream,
+	struct snd_pcm_hw_params *params)
+{
+	return snd_pcm_lib_alloc_vmalloc_buffer(substream,
+						params_buffer_bytes(params));
+}
+
+static int dw_hdmi_prepare(struct snd_pcm_substream *substream)
+{
+	struct snd_pcm_runtime *runtime = substream->runtime;
+	struct snd_dw_hdmi *dw = substream->private_data;
+	u8 threshold, conf0, conf1;
+
+	/* Setup as per 3.0.5 FSL 4.1.0 BSP */
+	switch (dw->revision) {
+	case 0x0a:
+		conf0 = HDMI_AHB_DMA_CONF0_BURST_MODE |
+			HDMI_AHB_DMA_CONF0_INCR4;
+		if (runtime->channels == 2)
+			threshold = 126;
+		else
+			threshold = 124;
+		break;
+	case 0x1a:
+		conf0 = HDMI_AHB_DMA_CONF0_BURST_MODE |
+			HDMI_AHB_DMA_CONF0_INCR8;
+		threshold = 128;
+		break;
+	default:
+		/* NOTREACHED */
+		return -EINVAL;
+	}
+
+	dw_hdmi_set_sample_rate(dw->data.hdmi, runtime->rate);
+
+	/* Minimum number of bytes in the fifo. */
+	runtime->hw.fifo_size = threshold * 32;
+
+	conf0 |= HDMI_AHB_DMA_CONF0_EN_HLOCK;
+	conf1 = (1 << runtime->channels) - 1;
+
+	writeb_relaxed(threshold, dw->data.base + HDMI_AHB_DMA_THRSLD);
+	writeb_relaxed(conf0, dw->data.base + HDMI_AHB_DMA_CONF0);
+	writeb_relaxed(conf1, dw->data.base + HDMI_AHB_DMA_CONF1);
+
+	switch (runtime->format) {
+	case SNDRV_PCM_FORMAT_IEC958_SUBFRAME_LE:
+		dw->reformat = dw_hdmi_reformat_iec958;
+		break;
+	case SNDRV_PCM_FORMAT_S24_LE:
+		dw_hdmi_create_cs(dw, runtime);
+		dw->reformat = dw_hdmi_reformat_s24;
+		break;
+	}
+	dw->iec_offset = 0;
+	dw->channels = runtime->channels;
+	dw->buf_src  = runtime->dma_area;
+	dw->buf_dst  = substream->dma_buffer.area;
+	dw->buf_addr = substream->dma_buffer.addr;
+	dw->buf_period = snd_pcm_lib_period_bytes(substream);
+	dw->buf_size = snd_pcm_lib_buffer_bytes(substream);
+
+	return 0;
+}
+
+static int dw_hdmi_trigger(struct snd_pcm_substream *substream, int cmd)
+{
+	struct snd_dw_hdmi *dw = substream->private_data;
+	unsigned long flags;
+	int ret = 0;
+
+	switch (cmd) {
+	case SNDRV_PCM_TRIGGER_START:
+		spin_lock_irqsave(&dw->lock, flags);
+		dw->buf_offset = 0;
+		dw->substream = substream;
+		dw_hdmi_start_dma(dw);
+		dw_hdmi_audio_enable(dw->data.hdmi);
+		spin_unlock_irqrestore(&dw->lock, flags);
+		substream->runtime->delay = substream->runtime->period_size;
+		break;
+
+	case SNDRV_PCM_TRIGGER_STOP:
+		spin_lock_irqsave(&dw->lock, flags);
+		dw->substream = NULL;
+		dw_hdmi_stop_dma(dw);
+		dw_hdmi_audio_disable(dw->data.hdmi);
+		spin_unlock_irqrestore(&dw->lock, flags);
+		break;
+
+	default:
+		ret = -EINVAL;
+		break;
+	}
+
+	return ret;
+}
+
+static snd_pcm_uframes_t dw_hdmi_pointer(struct snd_pcm_substream *substream)
+{
+	struct snd_pcm_runtime *runtime = substream->runtime;
+	struct snd_dw_hdmi *dw = substream->private_data;
+
+	/*
+	 * We are unable to report the exact hardware position as
+	 * reading the 32-bit DMA position using 8-bit reads is racy.
+	 */
+	return bytes_to_frames(runtime, dw->buf_offset);
+}
+
+static struct snd_pcm_ops snd_dw_hdmi_ops = {
+	.open = dw_hdmi_open,
+	.close = dw_hdmi_close,
+	.ioctl = snd_pcm_lib_ioctl,
+	.hw_params = dw_hdmi_hw_params,
+	.hw_free = dw_hdmi_hw_free,
+	.prepare = dw_hdmi_prepare,
+	.trigger = dw_hdmi_trigger,
+	.pointer = dw_hdmi_pointer,
+	.page = snd_pcm_lib_get_vmalloc_page,
+};
+
+static int snd_dw_hdmi_probe(struct platform_device *pdev)
+{
+	const struct dw_hdmi_audio_data *data = pdev->dev.platform_data;
+	struct device *dev = pdev->dev.parent;
+	struct snd_dw_hdmi *dw;
+	struct snd_card *card;
+	struct snd_pcm *pcm;
+	unsigned revision;
+	int ret;
+
+	writeb_relaxed(HDMI_IH_MUTE_AHBDMAAUD_STAT0_ALL,
+		       data->base + HDMI_IH_MUTE_AHBDMAAUD_STAT0);
+	revision = readb_relaxed(data->base + HDMI_REVISION_ID);
+	if (revision != 0x0a && revision != 0x1a) {
+		dev_err(dev, "dw-hdmi-audio: unknown revision 0x%02x\n",
+			revision);
+		return -ENXIO;
+	}
+
+	ret = snd_card_new(dev, SNDRV_DEFAULT_IDX1, SNDRV_DEFAULT_STR1,
+			      THIS_MODULE, sizeof(struct snd_dw_hdmi), &card);
+	if (ret < 0)
+		return ret;
+
+	strlcpy(card->driver, DRIVER_NAME, sizeof(card->driver));
+	strlcpy(card->shortname, "DW-HDMI", sizeof(card->shortname));
+	snprintf(card->longname, sizeof(card->longname),
+		 "%s rev 0x%02x, irq %d", card->shortname, revision,
+		 data->irq);
+
+	dw = card->private_data;
+	dw->card = card;
+	dw->data = *data;
+	dw->revision = revision;
+
+	spin_lock_init(&dw->lock);
+
+	ret = snd_pcm_new(card, "DW HDMI", 0, 1, 0, &pcm);
+	if (ret < 0)
+		goto err;
+
+	dw->pcm = pcm;
+	pcm->private_data = dw;
+	strlcpy(pcm->name, DRIVER_NAME, sizeof(pcm->name));
+	snd_pcm_set_ops(pcm, SNDRV_PCM_STREAM_PLAYBACK, &snd_dw_hdmi_ops);
+
+	snd_pcm_lib_preallocate_pages_for_all(pcm, SNDRV_DMA_TYPE_DEV,
+			dev, 64 * 1024, 64 * 1024);
+
+	ret = snd_card_register(card);
+	if (ret < 0)
+		goto err;
+
+	platform_set_drvdata(pdev, dw);
+
+	return 0;
+
+err:
+	snd_card_free(card);
+	return ret;
+}
+
+static int snd_dw_hdmi_remove(struct platform_device *pdev)
+{
+	struct snd_dw_hdmi *dw = platform_get_drvdata(pdev);
+
+	snd_card_free(dw->card);
+
+	return 0;
+}
+
+#if defined(CONFIG_PM_SLEEP) && defined(IS_NOT_BROKEN)
+/*
+ * This code is fine, but requires implementation in the dw_hdmi_trigger()
+ * method which is currently missing as I have no way to test this.
+ */
+static int snd_dw_hdmi_suspend(struct device *dev)
+{
+	struct snd_dw_hdmi *dw = dev_get_drvdata(dev);
+
+	snd_power_change_state(dw->card, SNDRV_CTL_POWER_D3cold);
+	snd_pcm_suspend_all(dw->pcm);
+
+	return 0;
+}
+
+static int snd_dw_hdmi_resume(struct device *dev)
+{
+	struct snd_dw_hdmi *dw = dev_get_drvdata(dev);
+
+	snd_power_change_state(dw->card, SNDRV_CTL_POWER_D0);
+
+	return 0;
+}
+
+static SIMPLE_DEV_PM_OPS(snd_dw_hdmi_pm, snd_dw_hdmi_suspend,
+			 snd_dw_hdmi_resume);
+#define PM_OPS &snd_dw_hdmi_pm
+#else
+#define PM_OPS NULL
+#endif
+
+static struct platform_driver snd_dw_hdmi_driver = {
+	.probe	= snd_dw_hdmi_probe,
+	.remove	= snd_dw_hdmi_remove,
+	.driver	= {
+		.name = DRIVER_NAME,
+		.owner = THIS_MODULE,
+		.pm = PM_OPS,
+	},
+};
+
+module_platform_driver(snd_dw_hdmi_driver);
+
+MODULE_AUTHOR("Russell King <rmk+kernel@arm.linux.org.uk>");
+MODULE_DESCRIPTION("Synopsis Designware HDMI AHB ALSA interface");
+MODULE_LICENSE("GPL v2");
+MODULE_ALIAS("platform:" DRIVER_NAME);
diff --git a/drivers/gpu/drm/bridge/dw_hdmi-audio.h b/drivers/gpu/drm/bridge/dw_hdmi-audio.h
new file mode 100644
index 0000000..1e84011
--- /dev/null
+++ b/drivers/gpu/drm/bridge/dw_hdmi-audio.h
@@ -0,0 +1,13 @@
+#ifndef DW_HDMI_AUDIO_H
+#define DW_HDMI_AUDIO_H
+
+struct dw_hdmi;
+
+struct dw_hdmi_audio_data {
+	phys_addr_t phys;
+	void __iomem *base;
+	int irq;
+	struct dw_hdmi *hdmi;
+};
+
+#endif
diff --git a/drivers/gpu/drm/bridge/dw_hdmi.c b/drivers/gpu/drm/bridge/dw_hdmi.c
index 7a130f6..d61a9fb4 100644
--- a/drivers/gpu/drm/bridge/dw_hdmi.c
+++ b/drivers/gpu/drm/bridge/dw_hdmi.c
@@ -28,6 +28,7 @@
 #include <drm/bridge/dw_hdmi.h>
 
 #include "dw_hdmi.h"
+#include "dw_hdmi-audio.h"
 
 #define HDMI_EDID_LEN		512
 
@@ -104,6 +105,7 @@ struct dw_hdmi {
 	struct drm_encoder *encoder;
 	struct drm_bridge *bridge;
 
+	struct platform_device *audio;
 	enum dw_hdmi_devtype dev_type;
 	struct device *dev;
 	struct clk *isfr_clk;
@@ -1732,7 +1734,9 @@ int dw_hdmi_bind(struct device *dev, struct device *master,
 {
 	struct drm_device *drm = data;
 	struct device_node *np = dev->of_node;
+	struct platform_device_info pdevinfo;
 	struct device_node *ddc_node;
+	struct dw_hdmi_audio_data audio;
 	struct dw_hdmi *hdmi;
 	int ret;
 	u32 val = 1;
@@ -1860,6 +1864,23 @@ int dw_hdmi_bind(struct device *dev, struct device *master,
 	hdmi_writeb(hdmi, ~(HDMI_IH_PHY_STAT0_HPD | HDMI_IH_PHY_STAT0_RX_SENSE),
 		    HDMI_IH_MUTE_PHY_STAT0);
 
+	memset(&pdevinfo, 0, sizeof(pdevinfo));
+	pdevinfo.parent = dev;
+	pdevinfo.id = PLATFORM_DEVID_AUTO;
+
+	if (hdmi_readb(hdmi, HDMI_CONFIG1_ID) & HDMI_CONFIG1_AHB) {
+		audio.phys = iores->start;
+		audio.base = hdmi->regs;
+		audio.irq = irq;
+		audio.hdmi = hdmi;
+
+		pdevinfo.name = "dw-hdmi-ahb-audio";
+		pdevinfo.data = &audio;
+		pdevinfo.size_data = sizeof(audio);
+		pdevinfo.dma_mask = DMA_BIT_MASK(32);
+		hdmi->audio = platform_device_register_full(&pdevinfo);
+	}
+
 	dev_set_drvdata(dev, hdmi);
 
 	return 0;
@@ -1877,6 +1898,9 @@ void dw_hdmi_unbind(struct device *dev, struct device *master, void *data)
 {
 	struct dw_hdmi *hdmi = dev_get_drvdata(dev);
 
+	if (hdmi->audio && !IS_ERR(hdmi->audio))
+		platform_device_unregister(hdmi->audio);
+
 	/* Disable all interrupts */
 	hdmi_writeb(hdmi, ~0, HDMI_IH_MUTE_PHY_STAT0);
 
diff --git a/drivers/gpu/drm/bridge/dw_hdmi.h b/drivers/gpu/drm/bridge/dw_hdmi.h
index ee7f7ed..fc9a560 100644
--- a/drivers/gpu/drm/bridge/dw_hdmi.h
+++ b/drivers/gpu/drm/bridge/dw_hdmi.h
@@ -545,6 +545,9 @@
 #define HDMI_I2CM_FS_SCL_LCNT_0_ADDR            0x7E12
 
 enum {
+/* CONFIG1_ID field values */
+	HDMI_CONFIG1_AHB = 0x01,
+
 /* IH_FC_INT2 field values */
 	HDMI_IH_FC_INT2_OVERFLOW_MASK = 0x03,
 	HDMI_IH_FC_INT2_LOW_PRIORITY_OVERFLOW = 0x02,
-- 
cgit v0.10.2


From f5ce405722e410a69da4604285ae5e9210011ffe Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Thu, 7 Nov 2013 16:06:01 +0000
Subject: drm: bridge/dw_hdmi-ahb-audio: parse ELD from HDMI driver

Parse the ELD (EDID like data) stored from the HDMI driver to restrict
the sample rates and channels which are available to ALSA.  This causes
the ALSA device to reflect the capabilities of the overall audio path,
not just what is supported at the HDMI source interface level.

Tested-by: Fabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/bridge/Kconfig b/drivers/gpu/drm/bridge/Kconfig
index 0d0a4f7..6dddd39 100644
--- a/drivers/gpu/drm/bridge/Kconfig
+++ b/drivers/gpu/drm/bridge/Kconfig
@@ -15,6 +15,7 @@ config DRM_DW_HDMI_AHB_AUDIO
 	tristate "Synopsis Designware AHB Audio interface"
 	depends on DRM_DW_HDMI && SND
 	select SND_PCM
+	select SND_PCM_ELD
 	select SND_PCM_IEC958
 	help
 	  Support the AHB Audio interface which is part of the Synopsis
diff --git a/drivers/gpu/drm/bridge/dw_hdmi-ahb-audio.c b/drivers/gpu/drm/bridge/dw_hdmi-ahb-audio.c
index bf379310..fbcef26 100644
--- a/drivers/gpu/drm/bridge/dw_hdmi-ahb-audio.c
+++ b/drivers/gpu/drm/bridge/dw_hdmi-ahb-audio.c
@@ -12,11 +12,13 @@
 #include <linux/module.h>
 #include <linux/platform_device.h>
 #include <drm/bridge/dw_hdmi.h>
+#include <drm/drm_edid.h>
 
 #include <sound/asoundef.h>
 #include <sound/core.h>
 #include <sound/initval.h>
 #include <sound/pcm.h>
+#include <sound/pcm_drm_eld.h>
 #include <sound/pcm_iec958.h>
 
 #include "dw_hdmi-audio.h"
@@ -286,6 +288,10 @@ static int dw_hdmi_open(struct snd_pcm_substream *substream)
 
 	runtime->hw = dw_hdmi_hw;
 
+	ret = snd_pcm_hw_constraint_eld(runtime, dw->data.eld);
+	if (ret < 0)
+		return ret;
+
 	ret = snd_pcm_limit_hw_rates(runtime);
 	if (ret < 0)
 		return ret;
diff --git a/drivers/gpu/drm/bridge/dw_hdmi-audio.h b/drivers/gpu/drm/bridge/dw_hdmi-audio.h
index 1e84011..91f631b 100644
--- a/drivers/gpu/drm/bridge/dw_hdmi-audio.h
+++ b/drivers/gpu/drm/bridge/dw_hdmi-audio.h
@@ -8,6 +8,7 @@ struct dw_hdmi_audio_data {
 	void __iomem *base;
 	int irq;
 	struct dw_hdmi *hdmi;
+	u8 *eld;
 };
 
 #endif
diff --git a/drivers/gpu/drm/bridge/dw_hdmi.c b/drivers/gpu/drm/bridge/dw_hdmi.c
index d61a9fb4..70e7a93 100644
--- a/drivers/gpu/drm/bridge/dw_hdmi.c
+++ b/drivers/gpu/drm/bridge/dw_hdmi.c
@@ -1533,6 +1533,8 @@ static int dw_hdmi_connector_get_modes(struct drm_connector *connector)
 		hdmi->sink_has_audio = drm_detect_monitor_audio(edid);
 		drm_mode_connector_update_edid_property(connector, edid);
 		ret = drm_add_edid_modes(connector, edid);
+		/* Store the ELD */
+		drm_edid_to_eld(connector, edid);
 		kfree(edid);
 	} else {
 		dev_dbg(hdmi->dev, "failed to get edid\n");
@@ -1873,6 +1875,7 @@ int dw_hdmi_bind(struct device *dev, struct device *master,
 		audio.base = hdmi->regs;
 		audio.irq = irq;
 		audio.hdmi = hdmi;
+		audio.eld = hdmi->connector.eld;
 
 		pdevinfo.name = "dw-hdmi-ahb-audio";
 		pdevinfo.data = &audio;
-- 
cgit v0.10.2


From 9dc515f8dfc8c2b5d4e8355933e6ac71abeb7b1e Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Thu, 4 Jun 2015 10:13:28 +0100
Subject: drm: bridge/dw_hdmi-ahb-audio: basic support for multi-channel PCM
 audio

Add basic support for multi-channel PCM audio, with fixed speaker
mappings.  This has been tested with an AV receiver, and appears to
work for low sample rates up to 8 channels.

It should be noted that multi-channel mode using the IEC958 alsa-lib
conversion plugin requires correct AES channel status for the AV
receiver to recognise the stream, especially the sample rate bits.
"Not identified" does not work there.

Tested-by: Fabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/bridge/dw_hdmi-ahb-audio.c b/drivers/gpu/drm/bridge/dw_hdmi-ahb-audio.c
index fbcef26..e779c16 100644
--- a/drivers/gpu/drm/bridge/dw_hdmi-ahb-audio.c
+++ b/drivers/gpu/drm/bridge/dw_hdmi-ahb-audio.c
@@ -62,9 +62,14 @@ enum {
 	HDMI_AHB_DMA_CONF0_INCR4 = 0,
 	HDMI_AHB_DMA_CONF0_BURST_MODE = BIT(0),
 	HDMI_AHB_DMA_MASK_DONE = BIT(7),
+
 	HDMI_REVISION_ID = 0x0001,
 	HDMI_IH_AHBDMAAUD_STAT0 = 0x0109,
 	HDMI_IH_MUTE_AHBDMAAUD_STAT0 = 0x0189,
+	HDMI_FC_AUDICONF2 = 0x1027,
+	HDMI_FC_AUDSCONF = 0x1063,
+	HDMI_FC_AUDSCONF_LAYOUT1 = 1 << 0,
+	HDMI_FC_AUDSCONF_LAYOUT0 = 0 << 0,
 	HDMI_AHB_DMA_CONF0 = 0x3600,
 	HDMI_AHB_DMA_START = 0x3601,
 	HDMI_AHB_DMA_STOP = 0x3602,
@@ -77,6 +82,44 @@ enum {
 	HDMI_AHB_DMA_BUFFPOL = 0x361a,
 };
 
+struct dw_hdmi_channel_conf {
+	u8 conf1;
+	u8 ca;
+};
+
+/*
+ * The default mapping of ALSA channels to HDMI channels and speaker
+ * allocation bits.  Note that we can't do channel remapping here -
+ * channels must be in the same order.
+ *
+ * Mappings for alsa-lib pcm/surround*.conf files:
+ *
+ *		Front	Sur4.0	Sur4.1	Sur5.0	Sur5.1	Sur7.1
+ * Channels	2	4	6	6	6	8
+ *
+ * Our mapping from ALSA channel to CEA686D speaker name and HDMI channel:
+ *
+ *				Number of ALSA channels
+ * ALSA Channel	2	3	4	5	6	7	8
+ * 0		FL:0	=	=	=	=	=	=
+ * 1		FR:1	=	=	=	=	=	=
+ * 2			FC:3	RL:4	LFE:2	=	=	=
+ * 3				RR:5	RL:4	FC:3	=	=
+ * 4					RR:5	RL:4	=	=
+ * 5						RR:5	=	=
+ * 6							RC:6	=
+ * 7							RLC/FRC	RLC/FRC
+ */
+static struct dw_hdmi_channel_conf default_hdmi_channel_config[7] = {
+	{ 0x03, 0x00 },	/* FL,FR */
+	{ 0x0b, 0x02 },	/* FL,FR,FC */
+	{ 0x33, 0x08 },	/* FL,FR,RL,RR */
+	{ 0x37, 0x09 },	/* FL,FR,LFE,RL,RR */
+	{ 0x3f, 0x0b },	/* FL,FR,LFE,FC,RL,RR */
+	{ 0x7f, 0x0f },	/* FL,FR,LFE,FC,RL,RR,RC */
+	{ 0xff, 0x13 },	/* FL,FR,LFE,FC,RL,RR,[FR]RC,[FR]LC */
+};
+
 struct snd_dw_hdmi {
 	struct snd_card *card;
 	struct snd_pcm *pcm;
@@ -354,7 +397,7 @@ static int dw_hdmi_prepare(struct snd_pcm_substream *substream)
 {
 	struct snd_pcm_runtime *runtime = substream->runtime;
 	struct snd_dw_hdmi *dw = substream->private_data;
-	u8 threshold, conf0, conf1;
+	u8 threshold, conf0, conf1, layout, ca;
 
 	/* Setup as per 3.0.5 FSL 4.1.0 BSP */
 	switch (dw->revision) {
@@ -382,11 +425,23 @@ static int dw_hdmi_prepare(struct snd_pcm_substream *substream)
 	runtime->hw.fifo_size = threshold * 32;
 
 	conf0 |= HDMI_AHB_DMA_CONF0_EN_HLOCK;
-	conf1 = (1 << runtime->channels) - 1;
+	conf1 = default_hdmi_channel_config[runtime->channels - 2].conf1;
+	ca = default_hdmi_channel_config[runtime->channels - 2].ca;
+
+	/*
+	 * For >2 channel PCM audio, we need to select layout 1
+	 * and set an appropriate channel map.
+	 */
+	if (runtime->channels > 2)
+		layout = HDMI_FC_AUDSCONF_LAYOUT1;
+	else
+		layout = HDMI_FC_AUDSCONF_LAYOUT0;
 
 	writeb_relaxed(threshold, dw->data.base + HDMI_AHB_DMA_THRSLD);
 	writeb_relaxed(conf0, dw->data.base + HDMI_AHB_DMA_CONF0);
 	writeb_relaxed(conf1, dw->data.base + HDMI_AHB_DMA_CONF1);
+	writeb_relaxed(layout, dw->data.base + HDMI_FC_AUDSCONF);
+	writeb_relaxed(ca, dw->data.base + HDMI_FC_AUDICONF2);
 
 	switch (runtime->format) {
 	case SNDRV_PCM_FORMAT_IEC958_SUBFRAME_LE:
-- 
cgit v0.10.2


From 91cd69088ecc375d057eb3721bfbcf927e9f2dd2 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Thu, 4 Jun 2015 10:24:33 +0100
Subject: drm: bridge/dw_hdmi-ahb-audio: allow larger buffer sizes

With multichannel audio, we need to allow larger buffer sizes to avoid
XRUNs during playback.  Push the buffer size up to 1024K, but as we
maintain two buffers, ensure that the vmalloc buffer does not exceed
the userspace buffer size.

Tested-by: Fabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/bridge/dw_hdmi-ahb-audio.c b/drivers/gpu/drm/bridge/dw_hdmi-ahb-audio.c
index e779c16..59f630f 100644
--- a/drivers/gpu/drm/bridge/dw_hdmi-ahb-audio.c
+++ b/drivers/gpu/drm/bridge/dw_hdmi-ahb-audio.c
@@ -314,7 +314,7 @@ static struct snd_pcm_hardware dw_hdmi_hw = {
 		 SNDRV_PCM_RATE_192000,
 	.channels_min = 2,
 	.channels_max = 8,
-	.buffer_bytes_max = 64 * 1024,
+	.buffer_bytes_max = 1024 * 1024,
 	.period_bytes_min = 256,
 	.period_bytes_max = 8192,	/* ERR004323: must limit to 8k */
 	.periods_min = 2,
@@ -339,7 +339,15 @@ static int dw_hdmi_open(struct snd_pcm_substream *substream)
 	if (ret < 0)
 		return ret;
 
-	ret = snd_pcm_hw_constraint_integer(runtime, SNDRV_PCM_HW_PARAM_PERIODS);
+	ret = snd_pcm_hw_constraint_integer(runtime,
+					    SNDRV_PCM_HW_PARAM_PERIODS);
+	if (ret < 0)
+		return ret;
+
+	/* Limit the buffer size to the size of the preallocated buffer */
+	ret = snd_pcm_hw_constraint_minmax(runtime,
+					   SNDRV_PCM_HW_PARAM_BUFFER_SIZE,
+					   0, substream->dma_buffer.bytes);
 	if (ret < 0)
 		return ret;
 
@@ -389,6 +397,7 @@ static int dw_hdmi_hw_free(struct snd_pcm_substream *substream)
 static int dw_hdmi_hw_params(struct snd_pcm_substream *substream,
 	struct snd_pcm_hw_params *params)
 {
+	/* Allocate the PCM runtime buffer, which is exposed to userspace. */
 	return snd_pcm_lib_alloc_vmalloc_buffer(substream,
 						params_buffer_bytes(params));
 }
@@ -566,8 +575,12 @@ static int snd_dw_hdmi_probe(struct platform_device *pdev)
 	strlcpy(pcm->name, DRIVER_NAME, sizeof(pcm->name));
 	snd_pcm_set_ops(pcm, SNDRV_PCM_STREAM_PLAYBACK, &snd_dw_hdmi_ops);
 
+	/*
+	 * To support 8-channel 96kHz audio reliably, we need 512k
+	 * to satisfy alsa with our restricted period (ERR004323).
+	 */
 	snd_pcm_lib_preallocate_pages_for_all(pcm, SNDRV_DMA_TYPE_DEV,
-			dev, 64 * 1024, 64 * 1024);
+			dev, 128 * 1024, 1024 * 1024);
 
 	ret = snd_card_register(card);
 	if (ret < 0)
-- 
cgit v0.10.2


From d0c96d168021a60555c692d860334814cc694d8b Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Wed, 22 Jul 2015 10:35:41 +0100
Subject: drm: bridge/dw_hdmi: avoid being recursive in N calculation

There's no need to be recursive when computing the N value for the ACR
packet - we can instead calculate the multiplier prior to our switch()
based lookup, and multiply the N value appropriately afterwards.

Tested-by: Fabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/bridge/dw_hdmi.c b/drivers/gpu/drm/bridge/dw_hdmi.c
index 70e7a93..e5ee923 100644
--- a/drivers/gpu/drm/bridge/dw_hdmi.c
+++ b/drivers/gpu/drm/bridge/dw_hdmi.c
@@ -221,6 +221,12 @@ static unsigned int hdmi_compute_n(unsigned int freq, unsigned long pixel_clk,
 				   unsigned int ratio)
 {
 	unsigned int n = (128 * freq) / 1000;
+	unsigned int mult = 1;
+
+	while (freq > 48000) {
+		mult *= 2;
+		freq /= 2;
+	}
 
 	switch (freq) {
 	case 32000:
@@ -232,6 +238,7 @@ static unsigned int hdmi_compute_n(unsigned int freq, unsigned long pixel_clk,
 			n = 11648;
 		else
 			n = 4096;
+		n *= mult;
 		break;
 
 	case 44100:
@@ -243,6 +250,7 @@ static unsigned int hdmi_compute_n(unsigned int freq, unsigned long pixel_clk,
 			n = (ratio == 150) ? 17836 : 8918;
 		else
 			n = 6272;
+		n *= mult;
 		break;
 
 	case 48000:
@@ -256,22 +264,7 @@ static unsigned int hdmi_compute_n(unsigned int freq, unsigned long pixel_clk,
 			n = (ratio == 150) ? 11648 : 5824;
 		else
 			n = 6144;
-		break;
-
-	case 88200:
-		n = hdmi_compute_n(44100, pixel_clk, ratio) * 2;
-		break;
-
-	case 96000:
-		n = hdmi_compute_n(48000, pixel_clk, ratio) * 2;
-		break;
-
-	case 176400:
-		n = hdmi_compute_n(44100, pixel_clk, ratio) * 4;
-		break;
-
-	case 192000:
-		n = hdmi_compute_n(48000, pixel_clk, ratio) * 4;
+		n *= mult;
 		break;
 
 	default:
-- 
cgit v0.10.2


From 426701d0ccc13a2819fd99938ec50a888817e77d Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Wed, 22 Jul 2015 10:39:27 +0100
Subject: drm: bridge/dw_hdmi: adjust pixel clock values in N calculation

Adjust the pixel clock values in the N calculation to match the more
accurate clock values we're given by the DRM subsystem, which are the
kHz pixel rate, with any fractional kHz rounded down in the case of
the non-240, non-480 line modes, or rounded up for the others.  So,

	 25.20 / 1.001 =>  25175
	 27.00 * 1.001 =>  27027
	 74.25 / 1.001 =>  74176
	148.50 / 1.001 => 148352

DRM derives these rates from the EDID CEA mode identifiers, which are
looked up in the tables in drivers/gpu/drm/drm_edid.c.  The values on
the right are the clock values found in these tables, and are
currently expected to be passed to the HDMI driver unchanged.

Tested-by: Fabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/bridge/dw_hdmi.c b/drivers/gpu/drm/bridge/dw_hdmi.c
index e5ee923..2ef13d9 100644
--- a/drivers/gpu/drm/bridge/dw_hdmi.c
+++ b/drivers/gpu/drm/bridge/dw_hdmi.c
@@ -230,11 +230,11 @@ static unsigned int hdmi_compute_n(unsigned int freq, unsigned long pixel_clk,
 
 	switch (freq) {
 	case 32000:
-		if (pixel_clk == 25170000)
+		if (pixel_clk == 25175000)
 			n = (ratio == 150) ? 9152 : 4576;
-		else if (pixel_clk == 27020000)
+		else if (pixel_clk == 27027000)
 			n = (ratio == 150) ? 8192 : 4096;
-		else if (pixel_clk == 74170000 || pixel_clk == 148350000)
+		else if (pixel_clk == 74176000 || pixel_clk == 148352000)
 			n = 11648;
 		else
 			n = 4096;
@@ -242,11 +242,11 @@ static unsigned int hdmi_compute_n(unsigned int freq, unsigned long pixel_clk,
 		break;
 
 	case 44100:
-		if (pixel_clk == 25170000)
+		if (pixel_clk == 25175000)
 			n = 7007;
-		else if (pixel_clk == 74170000)
+		else if (pixel_clk == 74176000)
 			n = 17836;
-		else if (pixel_clk == 148350000)
+		else if (pixel_clk == 148352000)
 			n = (ratio == 150) ? 17836 : 8918;
 		else
 			n = 6272;
@@ -254,13 +254,13 @@ static unsigned int hdmi_compute_n(unsigned int freq, unsigned long pixel_clk,
 		break;
 
 	case 48000:
-		if (pixel_clk == 25170000)
+		if (pixel_clk == 25175000)
 			n = (ratio == 150) ? 9152 : 6864;
-		else if (pixel_clk == 27020000)
+		else if (pixel_clk == 27027000)
 			n = (ratio == 150) ? 8192 : 6144;
-		else if (pixel_clk == 74170000)
+		else if (pixel_clk == 74176000)
 			n = 11648;
-		else if (pixel_clk == 148350000)
+		else if (pixel_clk == 148352000)
 			n = (ratio == 150) ? 11648 : 5824;
 		else
 			n = 6144;
-- 
cgit v0.10.2


From b195fbdbdf5f6712beec09bb9009f99ff9ecd912 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Wed, 22 Jul 2015 11:28:16 +0100
Subject: drm: bridge/dw_hdmi: remove ratio support from ACR code

We never set the ratio for CTS/N calculation for the audio clock
regenerator (ACR) to anything but 100, so this adds pointless
complexity.  Should we support pixel repetition, we should update the
CTS/N calculation code to use those parameters or the actual TMDS clock
rate instead of a ratio.

Tested-by: Fabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/bridge/dw_hdmi.c b/drivers/gpu/drm/bridge/dw_hdmi.c
index 2ef13d9..62ee302 100644
--- a/drivers/gpu/drm/bridge/dw_hdmi.c
+++ b/drivers/gpu/drm/bridge/dw_hdmi.c
@@ -140,7 +140,6 @@ struct dw_hdmi {
 	unsigned int audio_cts;
 	unsigned int audio_n;
 	bool audio_enable;
-	int ratio;
 
 	void (*write)(struct dw_hdmi *hdmi, u8 val, int offset);
 	u8 (*read)(struct dw_hdmi *hdmi, int offset);
@@ -217,8 +216,7 @@ static void hdmi_set_cts_n(struct dw_hdmi *hdmi, unsigned int cts,
 	hdmi_writeb(hdmi, n & 0xff, HDMI_AUD_N1);
 }
 
-static unsigned int hdmi_compute_n(unsigned int freq, unsigned long pixel_clk,
-				   unsigned int ratio)
+static unsigned int hdmi_compute_n(unsigned int freq, unsigned long pixel_clk)
 {
 	unsigned int n = (128 * freq) / 1000;
 	unsigned int mult = 1;
@@ -231,9 +229,9 @@ static unsigned int hdmi_compute_n(unsigned int freq, unsigned long pixel_clk,
 	switch (freq) {
 	case 32000:
 		if (pixel_clk == 25175000)
-			n = (ratio == 150) ? 9152 : 4576;
+			n = 4576;
 		else if (pixel_clk == 27027000)
-			n = (ratio == 150) ? 8192 : 4096;
+			n = 4096;
 		else if (pixel_clk == 74176000 || pixel_clk == 148352000)
 			n = 11648;
 		else
@@ -247,7 +245,7 @@ static unsigned int hdmi_compute_n(unsigned int freq, unsigned long pixel_clk,
 		else if (pixel_clk == 74176000)
 			n = 17836;
 		else if (pixel_clk == 148352000)
-			n = (ratio == 150) ? 17836 : 8918;
+			n = 8918;
 		else
 			n = 6272;
 		n *= mult;
@@ -255,13 +253,13 @@ static unsigned int hdmi_compute_n(unsigned int freq, unsigned long pixel_clk,
 
 	case 48000:
 		if (pixel_clk == 25175000)
-			n = (ratio == 150) ? 9152 : 6864;
+			n = 6864;
 		else if (pixel_clk == 27027000)
-			n = (ratio == 150) ? 8192 : 6144;
+			n = 6144;
 		else if (pixel_clk == 74176000)
 			n = 11648;
 		else if (pixel_clk == 148352000)
-			n = (ratio == 150) ? 11648 : 5824;
+			n = 5824;
 		else
 			n = 6144;
 		n *= mult;
@@ -274,13 +272,11 @@ static unsigned int hdmi_compute_n(unsigned int freq, unsigned long pixel_clk,
 	return n;
 }
 
-static unsigned int hdmi_compute_cts(unsigned int freq, unsigned long pixel_clk,
-				     unsigned int ratio)
+static unsigned int hdmi_compute_cts(unsigned int freq, unsigned long pixel_clk)
 {
 	unsigned int cts = 0;
 
-	pr_debug("%s: freq: %d pixel_clk: %ld ratio: %d\n", __func__, freq,
-		 pixel_clk, ratio);
+	pr_debug("%s: freq: %d pixel_clk: %ld\n", __func__, freq, pixel_clk);
 
 	switch (freq) {
 	case 32000:
@@ -341,26 +337,24 @@ static unsigned int hdmi_compute_cts(unsigned int freq, unsigned long pixel_clk,
 	default:
 		break;
 	}
-	if (ratio == 100)
-		return cts;
-	return (cts * ratio) / 100;
+	return cts;
 }
 
 static void hdmi_set_clk_regenerator(struct dw_hdmi *hdmi,
-	unsigned long pixel_clk, unsigned int sample_rate, unsigned int ratio)
+	unsigned long pixel_clk, unsigned int sample_rate)
 {
 	unsigned int n, cts;
 
-	n = hdmi_compute_n(sample_rate, pixel_clk, ratio);
-	cts = hdmi_compute_cts(sample_rate, pixel_clk, ratio);
+	n = hdmi_compute_n(sample_rate, pixel_clk);
+	cts = hdmi_compute_cts(sample_rate, pixel_clk);
 	if (!cts) {
 		dev_err(hdmi->dev,
 			"%s: pixel clock/sample rate not supported: %luMHz / %ukHz\n",
 			__func__, pixel_clk, sample_rate);
 	}
 
-	dev_dbg(hdmi->dev, "%s: samplerate=%ukHz ratio=%d pixelclk=%luMHz N=%d cts=%d\n",
-		__func__, sample_rate, ratio, pixel_clk, n, cts);
+	dev_dbg(hdmi->dev, "%s: samplerate=%ukHz pixelclk=%luMHz N=%d cts=%d\n",
+		__func__, sample_rate, pixel_clk, n, cts);
 
 	spin_lock_irq(&hdmi->audio_lock);
 	hdmi->audio_n = n;
@@ -372,8 +366,7 @@ static void hdmi_set_clk_regenerator(struct dw_hdmi *hdmi,
 static void hdmi_init_clk_regenerator(struct dw_hdmi *hdmi)
 {
 	mutex_lock(&hdmi->audio_mutex);
-	hdmi_set_clk_regenerator(hdmi, 74250000, hdmi->sample_rate,
-				 hdmi->ratio);
+	hdmi_set_clk_regenerator(hdmi, 74250000, hdmi->sample_rate);
 	mutex_unlock(&hdmi->audio_mutex);
 }
 
@@ -381,7 +374,7 @@ static void hdmi_clk_regenerator_update_pixel_clock(struct dw_hdmi *hdmi)
 {
 	mutex_lock(&hdmi->audio_mutex);
 	hdmi_set_clk_regenerator(hdmi, hdmi->hdmi_data.video_mode.mpixelclock,
-				 hdmi->sample_rate, hdmi->ratio);
+				 hdmi->sample_rate);
 	mutex_unlock(&hdmi->audio_mutex);
 }
 
@@ -390,7 +383,7 @@ void dw_hdmi_set_sample_rate(struct dw_hdmi *hdmi, unsigned int rate)
 	mutex_lock(&hdmi->audio_mutex);
 	hdmi->sample_rate = rate;
 	hdmi_set_clk_regenerator(hdmi, hdmi->hdmi_data.video_mode.mpixelclock,
-				 hdmi->sample_rate, hdmi->ratio);
+				 hdmi->sample_rate);
 	mutex_unlock(&hdmi->audio_mutex);
 }
 EXPORT_SYMBOL_GPL(dw_hdmi_set_sample_rate);
@@ -1746,7 +1739,6 @@ int dw_hdmi_bind(struct device *dev, struct device *master,
 	hdmi->dev = dev;
 	hdmi->dev_type = plat_data->dev_type;
 	hdmi->sample_rate = 48000;
-	hdmi->ratio = 100;
 	hdmi->encoder = encoder;
 	hdmi->disabled = true;
 	hdmi->rxsense = true;
-- 
cgit v0.10.2


From dfbdaf50460479446a258ef781683e7d7d6349d7 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Wed, 22 Jul 2015 16:54:37 +0100
Subject: drm: bridge/dw_hdmi: replace CTS calculation for the ACR

Given the TDMS clock, audio sample rate, and the N parameter, we can
calculate the CTS value for the audio clock regenerator (ACR) using the
following calculation given in the HDMI specification:

	CTS = ftdms * N / (128 * fs)

The specification says that the CTS value is an average value, which is
true if the source hardware measures it.  Where source hardware needs it
to be programmed, it is particularly difficult to alternate between two
values correctly to ensure that we achieve a correct "average"
fractional value at the sink.

Also, there's the problem that our "ftdms" is not a fully accurate
value; it is rounded to a kHz value.  This introduces an unnecessary
(and harmless) fractional value into the above equation for combinations
like 148.5MHz/1.001 for 44100Hz - we still calculate the correct CTS
value.

Tested-by: Fabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

diff --git a/drivers/gpu/drm/bridge/dw_hdmi.c b/drivers/gpu/drm/bridge/dw_hdmi.c
index 62ee302..56de9f1 100644
--- a/drivers/gpu/drm/bridge/dw_hdmi.c
+++ b/drivers/gpu/drm/bridge/dw_hdmi.c
@@ -272,89 +272,29 @@ static unsigned int hdmi_compute_n(unsigned int freq, unsigned long pixel_clk)
 	return n;
 }
 
-static unsigned int hdmi_compute_cts(unsigned int freq, unsigned long pixel_clk)
-{
-	unsigned int cts = 0;
-
-	pr_debug("%s: freq: %d pixel_clk: %ld\n", __func__, freq, pixel_clk);
-
-	switch (freq) {
-	case 32000:
-		if (pixel_clk == 297000000) {
-			cts = 222750;
-			break;
-		}
-	case 48000:
-	case 96000:
-	case 192000:
-		switch (pixel_clk) {
-		case 25200000:
-		case 27000000:
-		case 54000000:
-		case 74250000:
-		case 148500000:
-			cts = pixel_clk / 1000;
-			break;
-		case 297000000:
-			cts = 247500;
-			break;
-		/*
-		 * All other TMDS clocks are not supported by
-		 * DWC_hdmi_tx. The TMDS clocks divided or
-		 * multiplied by 1,001 coefficients are not
-		 * supported.
-		 */
-		default:
-			break;
-		}
-		break;
-	case 44100:
-	case 88200:
-	case 176400:
-		switch (pixel_clk) {
-		case 25200000:
-			cts = 28000;
-			break;
-		case 27000000:
-			cts = 30000;
-			break;
-		case 54000000:
-			cts = 60000;
-			break;
-		case 74250000:
-			cts = 82500;
-			break;
-		case 148500000:
-			cts = 165000;
-			break;
-		case 297000000:
-			cts = 247500;
-			break;
-		default:
-			break;
-		}
-		break;
-	default:
-		break;
-	}
-	return cts;
-}
-
 static void hdmi_set_clk_regenerator(struct dw_hdmi *hdmi,
 	unsigned long pixel_clk, unsigned int sample_rate)
 {
+	unsigned long ftdms = pixel_clk;
 	unsigned int n, cts;
+	u64 tmp;
 
 	n = hdmi_compute_n(sample_rate, pixel_clk);
-	cts = hdmi_compute_cts(sample_rate, pixel_clk);
-	if (!cts) {
-		dev_err(hdmi->dev,
-			"%s: pixel clock/sample rate not supported: %luMHz / %ukHz\n",
-			__func__, pixel_clk, sample_rate);
-	}
 
-	dev_dbg(hdmi->dev, "%s: samplerate=%ukHz pixelclk=%luMHz N=%d cts=%d\n",
-		__func__, sample_rate, pixel_clk, n, cts);
+	/*
+	 * Compute the CTS value from the N value.  Note that CTS and N
+	 * can be up to 20 bits in total, so we need 64-bit math.  Also
+	 * note that our TDMS clock is not fully accurate; it is accurate
+	 * to kHz.  This can introduce an unnecessary remainder in the
+	 * calculation below, so we don't try to warn about that.
+	 */
+	tmp = (u64)ftdms * n;
+	do_div(tmp, 128 * sample_rate);
+	cts = tmp;
+
+	dev_dbg(hdmi->dev, "%s: fs=%uHz ftdms=%lu.%03luMHz N=%d cts=%d\n",
+		__func__, sample_rate, ftdms / 1000000, (ftdms / 1000) % 1000,
+		n, cts);
 
 	spin_lock_irq(&hdmi->audio_lock);
 	hdmi->audio_n = n;
-- 
cgit v0.10.2


From 80bea1897d7bc35e2b201847e12029a9d677cf12 Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Sat, 10 Oct 2015 13:35:42 +0200
Subject: drm/i915: Update DRIVER_DATE to 20151010

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 4feec19..3c52a77 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -57,7 +57,7 @@
 
 #define DRIVER_NAME		"i915"
 #define DRIVER_DESC		"Intel Graphics"
-#define DRIVER_DATE		"20150928"
+#define DRIVER_DATE		"20151010"
 
 #undef WARN_ON
 /* Many gcc seem to no see through this and fall over :( */
-- 
cgit v0.10.2


From adfb46721e4a4ef373dee8b773b424c40a176fc2 Mon Sep 17 00:00:00 2001
From: Lukas Wunner <lukas@wunner.de>
Date: Mon, 12 Oct 2015 10:57:43 +0200
Subject: drm/i915: Drop unnecessary #include <linux/vga_switcheroo.h>

Commit 599bbb9de0fe ("drm/i915: i915 cannot provide switcher services.")
removed all remaining vga_switcheroo symbols from intel_acpi.c but left
the include. Drop it.

Signed-off-by: Lukas Wunner <lukas@wunner.de>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_acpi.c b/drivers/gpu/drm/i915/intel_acpi.c
index 8b13b9d..eb638a1 100644
--- a/drivers/gpu/drm/i915/intel_acpi.c
+++ b/drivers/gpu/drm/i915/intel_acpi.c
@@ -5,7 +5,6 @@
  */
 #include <linux/pci.h>
 #include <linux/acpi.h>
-#include <linux/vga_switcheroo.h>
 #include <drm/drmP.h>
 #include "i915_drv.h"
 
-- 
cgit v0.10.2


From 115719fceaa733d646e39cdce83cc32ddb891a49 Mon Sep 17 00:00:00 2001
From: "Williams, Dan J" <dan.j.williams@intel.com>
Date: Mon, 12 Oct 2015 21:12:57 +0000
Subject: i915: switch from acpi_os_ioremap to memremap

i915 expects the OpRegion to be cached (i.e. not __iomem), so explicitly
map it with memremap rather than the implied cache setting of
acpi_os_ioremap().

Cc: Daniel Vetter <daniel.vetter@intel.com>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Cc: intel-gfx@lists.freedesktop.org
Cc: David Airlie <airlied@linux.ie>
Cc: dri-devel@lists.freedesktop.org
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index 7e65015..1676388 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -1851,7 +1851,7 @@ static int i915_opregion(struct seq_file *m, void *unused)
 		goto out;
 
 	if (opregion->header) {
-		memcpy_fromio(data, opregion->header, OPREGION_SIZE);
+		memcpy(data, opregion->header, OPREGION_SIZE);
 		seq_write(m, data, OPREGION_SIZE);
 	}
 
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 3c52a77..604f55d 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -450,14 +450,14 @@ struct opregion_swsci;
 struct opregion_asle;
 
 struct intel_opregion {
-	struct opregion_header __iomem *header;
-	struct opregion_acpi __iomem *acpi;
-	struct opregion_swsci __iomem *swsci;
+	struct opregion_header *header;
+	struct opregion_acpi *acpi;
+	struct opregion_swsci *swsci;
 	u32 swsci_gbda_sub_functions;
 	u32 swsci_sbcb_sub_functions;
-	struct opregion_asle __iomem *asle;
-	void __iomem *vbt;
-	u32 __iomem *lid_state;
+	struct opregion_asle *asle;
+	void *vbt;
+	u32 *lid_state;
 	struct work_struct asle_work;
 };
 #define OPREGION_SIZE            (8*1024)
diff --git a/drivers/gpu/drm/i915/intel_bios.c b/drivers/gpu/drm/i915/intel_bios.c
index 68421c2..ce82f9c 100644
--- a/drivers/gpu/drm/i915/intel_bios.c
+++ b/drivers/gpu/drm/i915/intel_bios.c
@@ -1231,20 +1231,13 @@ static const struct dmi_system_id intel_no_opregion_vbt[] = {
 	{ }
 };
 
-static const struct bdb_header *validate_vbt(const void __iomem *_base,
+static const struct bdb_header *validate_vbt(const void *base,
 					     size_t size,
-					     const void __iomem *_vbt,
+					     const void *_vbt,
 					     const char *source)
 {
-	/*
-	 * This is the one place where we explicitly discard the address space
-	 * (__iomem) of the BIOS/VBT. (And this will cause a sparse complaint.)
-	 * From now on everything is based on 'base', and treated as regular
-	 * memory.
-	 */
-	const void *base = (const void *) _base;
-	size_t offset = _vbt - _base;
-	const struct vbt_header *vbt = base + offset;
+	size_t offset = _vbt - base;
+	const struct vbt_header *vbt = _vbt;
 	const struct bdb_header *bdb;
 
 	if (offset + sizeof(struct vbt_header) > size) {
@@ -1282,7 +1275,15 @@ static const struct bdb_header *find_vbt(void __iomem *bios, size_t size)
 	/* Scour memory looking for the VBT signature. */
 	for (i = 0; i + 4 < size; i++) {
 		if (ioread32(bios + i) == *((const u32 *) "$VBT")) {
-			bdb = validate_vbt(bios, size, bios + i, "PCI ROM");
+			/*
+			 * This is the one place where we explicitly discard the
+			 * address space (__iomem) of the BIOS/VBT. From now on
+			 * everything is based on 'base', and treated as regular
+			 * memory.
+			 */
+			void *_bios = (void __force *) bios;
+
+			bdb = validate_vbt(_bios, size, _bios + i, "PCI ROM");
 			break;
 		}
 	}
diff --git a/drivers/gpu/drm/i915/intel_opregion.c b/drivers/gpu/drm/i915/intel_opregion.c
index b706b4e..6dc13c0 100644
--- a/drivers/gpu/drm/i915/intel_opregion.c
+++ b/drivers/gpu/drm/i915/intel_opregion.c
@@ -239,7 +239,7 @@ struct opregion_asle {
 static int swsci(struct drm_device *dev, u32 function, u32 parm, u32 *parm_out)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct opregion_swsci __iomem *swsci = dev_priv->opregion.swsci;
+	struct opregion_swsci *swsci = dev_priv->opregion.swsci;
 	u32 main_function, sub_function, scic;
 	u16 pci_swsci;
 	u32 dslp;
@@ -264,7 +264,7 @@ static int swsci(struct drm_device *dev, u32 function, u32 parm, u32 *parm_out)
 	}
 
 	/* Driver sleep timeout in ms. */
-	dslp = ioread32(&swsci->dslp);
+	dslp = swsci->dslp;
 	if (!dslp) {
 		/* The spec says 2ms should be the default, but it's too small
 		 * for some machines. */
@@ -277,7 +277,7 @@ static int swsci(struct drm_device *dev, u32 function, u32 parm, u32 *parm_out)
 	}
 
 	/* The spec tells us to do this, but we are the only user... */
-	scic = ioread32(&swsci->scic);
+	scic = swsci->scic;
 	if (scic & SWSCI_SCIC_INDICATOR) {
 		DRM_DEBUG_DRIVER("SWSCI request already in progress\n");
 		return -EBUSY;
@@ -285,8 +285,8 @@ static int swsci(struct drm_device *dev, u32 function, u32 parm, u32 *parm_out)
 
 	scic = function | SWSCI_SCIC_INDICATOR;
 
-	iowrite32(parm, &swsci->parm);
-	iowrite32(scic, &swsci->scic);
+	swsci->parm = parm;
+	swsci->scic = scic;
 
 	/* Ensure SCI event is selected and event trigger is cleared. */
 	pci_read_config_word(dev->pdev, PCI_SWSCI, &pci_swsci);
@@ -301,7 +301,7 @@ static int swsci(struct drm_device *dev, u32 function, u32 parm, u32 *parm_out)
 	pci_write_config_word(dev->pdev, PCI_SWSCI, pci_swsci);
 
 	/* Poll for the result. */
-#define C (((scic = ioread32(&swsci->scic)) & SWSCI_SCIC_INDICATOR) == 0)
+#define C (((scic = swsci->scic) & SWSCI_SCIC_INDICATOR) == 0)
 	if (wait_for(C, dslp)) {
 		DRM_DEBUG_DRIVER("SWSCI request timed out\n");
 		return -ETIMEDOUT;
@@ -317,7 +317,7 @@ static int swsci(struct drm_device *dev, u32 function, u32 parm, u32 *parm_out)
 	}
 
 	if (parm_out)
-		*parm_out = ioread32(&swsci->parm);
+		*parm_out = swsci->parm;
 
 	return 0;
 
@@ -412,7 +412,7 @@ static u32 asle_set_backlight(struct drm_device *dev, u32 bclp)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_connector *intel_connector;
-	struct opregion_asle __iomem *asle = dev_priv->opregion.asle;
+	struct opregion_asle *asle = dev_priv->opregion.asle;
 
 	DRM_DEBUG_DRIVER("bclp = 0x%08x\n", bclp);
 
@@ -437,7 +437,7 @@ static u32 asle_set_backlight(struct drm_device *dev, u32 bclp)
 	DRM_DEBUG_KMS("updating opregion backlight %d/255\n", bclp);
 	list_for_each_entry(intel_connector, &dev->mode_config.connector_list, base.head)
 		intel_panel_set_backlight_acpi(intel_connector, bclp, 255);
-	iowrite32(DIV_ROUND_UP(bclp * 100, 255) | ASLE_CBLV_VALID, &asle->cblv);
+	asle->cblv = DIV_ROUND_UP(bclp * 100, 255) | ASLE_CBLV_VALID;
 
 	drm_modeset_unlock(&dev->mode_config.connection_mutex);
 
@@ -524,14 +524,14 @@ static void asle_work(struct work_struct *work)
 	struct drm_i915_private *dev_priv =
 		container_of(opregion, struct drm_i915_private, opregion);
 	struct drm_device *dev = dev_priv->dev;
-	struct opregion_asle __iomem *asle = dev_priv->opregion.asle;
+	struct opregion_asle *asle = dev_priv->opregion.asle;
 	u32 aslc_stat = 0;
 	u32 aslc_req;
 
 	if (!asle)
 		return;
 
-	aslc_req = ioread32(&asle->aslc);
+	aslc_req = asle->aslc;
 
 	if (!(aslc_req & ASLC_REQ_MSK)) {
 		DRM_DEBUG_DRIVER("No request on ASLC interrupt 0x%08x\n",
@@ -540,34 +540,34 @@ static void asle_work(struct work_struct *work)
 	}
 
 	if (aslc_req & ASLC_SET_ALS_ILLUM)
-		aslc_stat |= asle_set_als_illum(dev, ioread32(&asle->alsi));
+		aslc_stat |= asle_set_als_illum(dev, asle->alsi);
 
 	if (aslc_req & ASLC_SET_BACKLIGHT)
-		aslc_stat |= asle_set_backlight(dev, ioread32(&asle->bclp));
+		aslc_stat |= asle_set_backlight(dev, asle->bclp);
 
 	if (aslc_req & ASLC_SET_PFIT)
-		aslc_stat |= asle_set_pfit(dev, ioread32(&asle->pfit));
+		aslc_stat |= asle_set_pfit(dev, asle->pfit);
 
 	if (aslc_req & ASLC_SET_PWM_FREQ)
-		aslc_stat |= asle_set_pwm_freq(dev, ioread32(&asle->pfmb));
+		aslc_stat |= asle_set_pwm_freq(dev, asle->pfmb);
 
 	if (aslc_req & ASLC_SUPPORTED_ROTATION_ANGLES)
 		aslc_stat |= asle_set_supported_rotation_angles(dev,
-							ioread32(&asle->srot));
+							asle->srot);
 
 	if (aslc_req & ASLC_BUTTON_ARRAY)
-		aslc_stat |= asle_set_button_array(dev, ioread32(&asle->iuer));
+		aslc_stat |= asle_set_button_array(dev, asle->iuer);
 
 	if (aslc_req & ASLC_CONVERTIBLE_INDICATOR)
-		aslc_stat |= asle_set_convertible(dev, ioread32(&asle->iuer));
+		aslc_stat |= asle_set_convertible(dev, asle->iuer);
 
 	if (aslc_req & ASLC_DOCKING_INDICATOR)
-		aslc_stat |= asle_set_docking(dev, ioread32(&asle->iuer));
+		aslc_stat |= asle_set_docking(dev, asle->iuer);
 
 	if (aslc_req & ASLC_ISCT_STATE_CHANGE)
 		aslc_stat |= asle_isct_state(dev);
 
-	iowrite32(aslc_stat, &asle->aslc);
+	asle->aslc = aslc_stat;
 }
 
 void intel_opregion_asle_intr(struct drm_device *dev)
@@ -592,8 +592,8 @@ static int intel_opregion_video_event(struct notifier_block *nb,
 	   Linux, these are handled by the dock, button and video drivers.
 	*/
 
-	struct opregion_acpi __iomem *acpi;
 	struct acpi_bus_event *event = data;
+	struct opregion_acpi *acpi;
 	int ret = NOTIFY_OK;
 
 	if (strcmp(event->device_class, ACPI_VIDEO_CLASS) != 0)
@@ -604,11 +604,10 @@ static int intel_opregion_video_event(struct notifier_block *nb,
 
 	acpi = system_opregion->acpi;
 
-	if (event->type == 0x80 &&
-	    (ioread32(&acpi->cevt) & 1) == 0)
+	if (event->type == 0x80 && ((acpi->cevt & 1) == 0))
 		ret = NOTIFY_BAD;
 
-	iowrite32(0, &acpi->csts);
+	acpi->csts = 0;
 
 	return ret;
 }
@@ -628,14 +627,14 @@ static u32 get_did(struct intel_opregion *opregion, int i)
 	u32 did;
 
 	if (i < ARRAY_SIZE(opregion->acpi->didl)) {
-		did = ioread32(&opregion->acpi->didl[i]);
+		did = opregion->acpi->didl[i];
 	} else {
 		i -= ARRAY_SIZE(opregion->acpi->didl);
 
 		if (WARN_ON(i >= ARRAY_SIZE(opregion->acpi->did2)))
 			return 0;
 
-		did = ioread32(&opregion->acpi->did2[i]);
+		did = opregion->acpi->did2[i];
 	}
 
 	return did;
@@ -644,14 +643,14 @@ static u32 get_did(struct intel_opregion *opregion, int i)
 static void set_did(struct intel_opregion *opregion, int i, u32 val)
 {
 	if (i < ARRAY_SIZE(opregion->acpi->didl)) {
-		iowrite32(val, &opregion->acpi->didl[i]);
+		opregion->acpi->didl[i] = val;
 	} else {
 		i -= ARRAY_SIZE(opregion->acpi->didl);
 
 		if (WARN_ON(i >= ARRAY_SIZE(opregion->acpi->did2)))
 			return;
 
-		iowrite32(val, &opregion->acpi->did2[i]);
+		opregion->acpi->did2[i] = val;
 	}
 }
 
@@ -773,7 +772,7 @@ static void intel_setup_cadls(struct drm_device *dev)
 	 * there are less than eight devices. */
 	do {
 		disp_id = get_did(opregion, i);
-		iowrite32(disp_id, &opregion->acpi->cadl[i]);
+		opregion->acpi->cadl[i] = disp_id;
 	} while (++i < 8 && disp_id != 0);
 }
 
@@ -792,16 +791,16 @@ void intel_opregion_init(struct drm_device *dev)
 		/* Notify BIOS we are ready to handle ACPI video ext notifs.
 		 * Right now, all the events are handled by the ACPI video module.
 		 * We don't actually need to do anything with them. */
-		iowrite32(0, &opregion->acpi->csts);
-		iowrite32(1, &opregion->acpi->drdy);
+		opregion->acpi->csts = 0;
+		opregion->acpi->drdy = 1;
 
 		system_opregion = opregion;
 		register_acpi_notifier(&intel_opregion_notifier);
 	}
 
 	if (opregion->asle) {
-		iowrite32(ASLE_TCHE_BLC_EN, &opregion->asle->tche);
-		iowrite32(ASLE_ARDY_READY, &opregion->asle->ardy);
+		opregion->asle->tche = ASLE_TCHE_BLC_EN;
+		opregion->asle->ardy = ASLE_ARDY_READY;
 	}
 }
 
@@ -814,19 +813,19 @@ void intel_opregion_fini(struct drm_device *dev)
 		return;
 
 	if (opregion->asle)
-		iowrite32(ASLE_ARDY_NOT_READY, &opregion->asle->ardy);
+		opregion->asle->ardy = ASLE_ARDY_NOT_READY;
 
 	cancel_work_sync(&dev_priv->opregion.asle_work);
 
 	if (opregion->acpi) {
-		iowrite32(0, &opregion->acpi->drdy);
+		opregion->acpi->drdy = 0;
 
 		system_opregion = NULL;
 		unregister_acpi_notifier(&intel_opregion_notifier);
 	}
 
 	/* just clear all opregion memory pointers now */
-	iounmap(opregion->header);
+	memunmap(opregion->header);
 	opregion->header = NULL;
 	opregion->acpi = NULL;
 	opregion->swsci = NULL;
@@ -899,10 +898,10 @@ int intel_opregion_setup(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_opregion *opregion = &dev_priv->opregion;
-	void __iomem *base;
 	u32 asls, mboxes;
 	char buf[sizeof(OPREGION_SIGNATURE)];
 	int err = 0;
+	void *base;
 
 	BUILD_BUG_ON(sizeof(struct opregion_header) != 0x100);
 	BUILD_BUG_ON(sizeof(struct opregion_acpi) != 0x100);
@@ -920,11 +919,11 @@ int intel_opregion_setup(struct drm_device *dev)
 	INIT_WORK(&opregion->asle_work, asle_work);
 #endif
 
-	base = acpi_os_ioremap(asls, OPREGION_SIZE);
+	base = memremap(asls, OPREGION_SIZE, MEMREMAP_WB);
 	if (!base)
 		return -ENOMEM;
 
-	memcpy_fromio(buf, base, sizeof(buf));
+	memcpy(buf, base, sizeof(buf));
 
 	if (memcmp(buf, OPREGION_SIGNATURE, 16)) {
 		DRM_DEBUG_DRIVER("opregion signature mismatch\n");
@@ -936,7 +935,7 @@ int intel_opregion_setup(struct drm_device *dev)
 
 	opregion->lid_state = base + ACPI_CLID;
 
-	mboxes = ioread32(&opregion->header->mboxes);
+	mboxes = opregion->header->mboxes;
 	if (mboxes & MBOX_ACPI) {
 		DRM_DEBUG_DRIVER("Public ACPI methods supported\n");
 		opregion->acpi = base + OPREGION_ACPI_OFFSET;
@@ -951,12 +950,12 @@ int intel_opregion_setup(struct drm_device *dev)
 		DRM_DEBUG_DRIVER("ASLE supported\n");
 		opregion->asle = base + OPREGION_ASLE_OFFSET;
 
-		iowrite32(ASLE_ARDY_NOT_READY, &opregion->asle->ardy);
+		opregion->asle->ardy = ASLE_ARDY_NOT_READY;
 	}
 
 	return 0;
 
 err_out:
-	iounmap(base);
+	memunmap(base);
 	return err;
 }
diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c
index f30c996..b05c6d9 100644
--- a/drivers/gpu/drm/i915/intel_panel.c
+++ b/drivers/gpu/drm/i915/intel_panel.c
@@ -381,7 +381,7 @@ intel_panel_detect(struct drm_device *dev)
 
 	/* Assume that the BIOS does not lie through the OpRegion... */
 	if (!i915.panel_ignore_lid && dev_priv->opregion.lid_state) {
-		return ioread32(dev_priv->opregion.lid_state) & 0x1 ?
+		return *dev_priv->opregion.lid_state & 0x1 ?
 			connector_status_connected :
 			connector_status_disconnected;
 	}
-- 
cgit v0.10.2


From 6042639c4c7f9e8a0c634245f034eb94827591a8 Mon Sep 17 00:00:00 2001
From: Chris Wilson <chris@chris-wilson.co.uk>
Date: Sat, 10 Oct 2015 10:44:32 +0100
Subject: drm/i915: Hold dev->event_lock whilst inspecting
 intel_crtc->unpin_work
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We should serialise access to the intel_crtc->unpin_work through the
dev->event_lock spinlock. It should not be possible for it to disappear
without severe error as the mmio_flip worker has not tagged the
unpin_work pending flip-completion. Similarly if the error exists, just
taking the unpin_work whilst holding the spinlock and then using it
unserialised just masks the race. (It is supposed to be valid as the
unpin_work exists until the flip completion interrupt which should not
fire until we flush the mmio writes to update the display base which is
the last time we access the unpin_work from the kthread.)

References: https://bugs.freedesktop.org/show_bug.cgi?id=92335
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index cddb0c6..71d7298 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -10848,11 +10848,11 @@ void intel_prepare_page_flip(struct drm_device *dev, int plane)
 	spin_unlock_irqrestore(&dev->event_lock, flags);
 }
 
-static inline void intel_mark_page_flip_active(struct intel_crtc *intel_crtc)
+static inline void intel_mark_page_flip_active(struct intel_unpin_work *work)
 {
 	/* Ensure that the work item is consistent when activating it ... */
 	smp_wmb();
-	atomic_set(&intel_crtc->unpin_work->pending, INTEL_FLIP_PENDING);
+	atomic_set(&work->pending, INTEL_FLIP_PENDING);
 	/* and that it is marked active as soon as the irq could fire. */
 	smp_wmb();
 }
@@ -10888,7 +10888,7 @@ static int intel_gen2_queue_flip(struct drm_device *dev,
 	intel_ring_emit(ring, intel_crtc->unpin_work->gtt_offset);
 	intel_ring_emit(ring, 0); /* aux display base address, unused */
 
-	intel_mark_page_flip_active(intel_crtc);
+	intel_mark_page_flip_active(intel_crtc->unpin_work);
 	return 0;
 }
 
@@ -10920,7 +10920,7 @@ static int intel_gen3_queue_flip(struct drm_device *dev,
 	intel_ring_emit(ring, intel_crtc->unpin_work->gtt_offset);
 	intel_ring_emit(ring, MI_NOOP);
 
-	intel_mark_page_flip_active(intel_crtc);
+	intel_mark_page_flip_active(intel_crtc->unpin_work);
 	return 0;
 }
 
@@ -10959,7 +10959,7 @@ static int intel_gen4_queue_flip(struct drm_device *dev,
 	pipesrc = I915_READ(PIPESRC(intel_crtc->pipe)) & 0x0fff0fff;
 	intel_ring_emit(ring, pf | pipesrc);
 
-	intel_mark_page_flip_active(intel_crtc);
+	intel_mark_page_flip_active(intel_crtc->unpin_work);
 	return 0;
 }
 
@@ -10995,7 +10995,7 @@ static int intel_gen6_queue_flip(struct drm_device *dev,
 	pipesrc = I915_READ(PIPESRC(intel_crtc->pipe)) & 0x0fff0fff;
 	intel_ring_emit(ring, pf | pipesrc);
 
-	intel_mark_page_flip_active(intel_crtc);
+	intel_mark_page_flip_active(intel_crtc->unpin_work);
 	return 0;
 }
 
@@ -11090,7 +11090,7 @@ static int intel_gen7_queue_flip(struct drm_device *dev,
 	intel_ring_emit(ring, intel_crtc->unpin_work->gtt_offset);
 	intel_ring_emit(ring, (MI_NOOP));
 
-	intel_mark_page_flip_active(intel_crtc);
+	intel_mark_page_flip_active(intel_crtc->unpin_work);
 	return 0;
 }
 
@@ -11121,7 +11121,8 @@ static bool use_mmio_flip(struct intel_engine_cs *ring,
 		return ring != i915_gem_request_get_ring(obj->last_write_req);
 }
 
-static void skl_do_mmio_flip(struct intel_crtc *intel_crtc)
+static void skl_do_mmio_flip(struct intel_crtc *intel_crtc,
+			     struct intel_unpin_work *work)
 {
 	struct drm_device *dev = intel_crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -11162,11 +11163,12 @@ static void skl_do_mmio_flip(struct intel_crtc *intel_crtc)
 	I915_WRITE(PLANE_CTL(pipe, 0), ctl);
 	I915_WRITE(PLANE_STRIDE(pipe, 0), stride);
 
-	I915_WRITE(PLANE_SURF(pipe, 0), intel_crtc->unpin_work->gtt_offset);
+	I915_WRITE(PLANE_SURF(pipe, 0), work->gtt_offset);
 	POSTING_READ(PLANE_SURF(pipe, 0));
 }
 
-static void ilk_do_mmio_flip(struct intel_crtc *intel_crtc)
+static void ilk_do_mmio_flip(struct intel_crtc *intel_crtc,
+			     struct intel_unpin_work *work)
 {
 	struct drm_device *dev = intel_crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -11186,31 +11188,36 @@ static void ilk_do_mmio_flip(struct intel_crtc *intel_crtc)
 
 	I915_WRITE(reg, dspcntr);
 
-	I915_WRITE(DSPSURF(intel_crtc->plane),
-		   intel_crtc->unpin_work->gtt_offset);
+	I915_WRITE(DSPSURF(intel_crtc->plane), work->gtt_offset);
 	POSTING_READ(DSPSURF(intel_crtc->plane));
-
 }
 
 /*
  * XXX: This is the temporary way to update the plane registers until we get
  * around to using the usual plane update functions for MMIO flips
  */
-static void intel_do_mmio_flip(struct intel_crtc *intel_crtc)
+static void intel_do_mmio_flip(struct intel_mmio_flip *mmio_flip)
 {
-	struct drm_device *dev = intel_crtc->base.dev;
+	struct intel_crtc *crtc = mmio_flip->crtc;
+	struct intel_unpin_work *work;
 
-	intel_mark_page_flip_active(intel_crtc);
+	spin_lock_irq(&crtc->base.dev->event_lock);
+	work = crtc->unpin_work;
+	spin_unlock_irq(&crtc->base.dev->event_lock);
+	if (work == NULL)
+		return;
 
-	intel_pipe_update_start(intel_crtc);
+	intel_mark_page_flip_active(work);
 
-	if (INTEL_INFO(dev)->gen >= 9)
-		skl_do_mmio_flip(intel_crtc);
+	intel_pipe_update_start(crtc);
+
+	if (INTEL_INFO(mmio_flip->i915)->gen >= 9)
+		skl_do_mmio_flip(crtc, work);
 	else
 		/* use_mmio_flip() retricts MMIO flips to ilk+ */
-		ilk_do_mmio_flip(intel_crtc);
+		ilk_do_mmio_flip(crtc, work);
 
-	intel_pipe_update_end(intel_crtc);
+	intel_pipe_update_end(crtc);
 }
 
 static void intel_mmio_flip_work_func(struct work_struct *work)
@@ -11218,15 +11225,15 @@ static void intel_mmio_flip_work_func(struct work_struct *work)
 	struct intel_mmio_flip *mmio_flip =
 		container_of(work, struct intel_mmio_flip, work);
 
-	if (mmio_flip->req)
+	if (mmio_flip->req) {
 		WARN_ON(__i915_wait_request(mmio_flip->req,
 					    mmio_flip->crtc->reset_counter,
 					    false, NULL,
 					    &mmio_flip->i915->rps.mmioflips));
+		i915_gem_request_unreference__unlocked(mmio_flip->req);
+	}
 
-	intel_do_mmio_flip(mmio_flip->crtc);
-
-	i915_gem_request_unreference__unlocked(mmio_flip->req);
+	intel_do_mmio_flip(mmio_flip);
 	kfree(mmio_flip);
 }
 
-- 
cgit v0.10.2


From 03999f0436325106ae4c6ca18646c68e5acc7f1b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Mon, 12 Oct 2015 19:41:08 +0300
Subject: drm/i915: Eliminate weird parameter inversion from BXT PPS registers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

v2: Keep using the same registers (PCH_*) instead of accidentally
    starting to use the other ones (BXT_*)2) (Jesse)

Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 6be853d..9d051ef 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -6587,10 +6587,10 @@ enum skl_disp_power_wells {
 #define _BXT_PP_ON_DELAYS2	0xc7308
 #define _BXT_PP_OFF_DELAYS2	0xc730c
 
-#define BXT_PP_STATUS(n)	((!n) ? PCH_PP_STATUS : _BXT_PP_STATUS2)
-#define BXT_PP_CONTROL(n)	((!n) ? PCH_PP_CONTROL : _BXT_PP_CONTROL2)
-#define BXT_PP_ON_DELAYS(n)	((!n) ? PCH_PP_ON_DELAYS : _BXT_PP_ON_DELAYS2)
-#define BXT_PP_OFF_DELAYS(n)	((!n) ? PCH_PP_OFF_DELAYS : _BXT_PP_OFF_DELAYS2)
+#define BXT_PP_STATUS(n)	_PIPE(n, PCH_PP_STATUS, _BXT_PP_STATUS2)
+#define BXT_PP_CONTROL(n)	_PIPE(n, PCH_PP_CONTROL, _BXT_PP_CONTROL2)
+#define BXT_PP_ON_DELAYS(n)	_PIPE(n, PCH_PP_ON_DELAYS, _BXT_PP_ON_DELAYS2)
+#define BXT_PP_OFF_DELAYS(n)	_PIPE(n, PCH_PP_OFF_DELAYS, _BXT_PP_OFF_DELAYS2)
 
 #define PCH_DP_B		0xe4100
 #define PCH_DPB_AUX_CH_CTL	0xe4110
-- 
cgit v0.10.2


From 436c6d4a145b63621cd40d81231b39d22ae84292 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 18 Sep 2015 20:03:37 +0300
Subject: drm/i915: Parametrize HSW video DIP data registers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 9d051ef..72741f4 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -6286,16 +6286,16 @@ enum skl_disp_power_wells {
 
 #define HSW_TVIDEO_DIP_CTL(trans) \
 	 _TRANSCODER2(trans, HSW_VIDEO_DIP_CTL_A)
-#define HSW_TVIDEO_DIP_AVI_DATA(trans) \
-	 _TRANSCODER2(trans, HSW_VIDEO_DIP_AVI_DATA_A)
-#define HSW_TVIDEO_DIP_VS_DATA(trans) \
-	 _TRANSCODER2(trans, HSW_VIDEO_DIP_VS_DATA_A)
-#define HSW_TVIDEO_DIP_SPD_DATA(trans) \
-	 _TRANSCODER2(trans, HSW_VIDEO_DIP_SPD_DATA_A)
+#define HSW_TVIDEO_DIP_AVI_DATA(trans, i) \
+	(_TRANSCODER2(trans, HSW_VIDEO_DIP_AVI_DATA_A) + (i) * 4)
+#define HSW_TVIDEO_DIP_VS_DATA(trans, i) \
+	(_TRANSCODER2(trans, HSW_VIDEO_DIP_VS_DATA_A) + (i) * 4)
+#define HSW_TVIDEO_DIP_SPD_DATA(trans, i) \
+	(_TRANSCODER2(trans, HSW_VIDEO_DIP_SPD_DATA_A) + (i) * 4)
 #define HSW_TVIDEO_DIP_GCP(trans) \
 	_TRANSCODER2(trans, HSW_VIDEO_DIP_GCP_A)
-#define HSW_TVIDEO_DIP_VSC_DATA(trans) \
-	 _TRANSCODER2(trans, HSW_VIDEO_DIP_VSC_DATA_A)
+#define HSW_TVIDEO_DIP_VSC_DATA(trans, i) \
+	(_TRANSCODER2(trans, HSW_VIDEO_DIP_VSC_DATA_A) + (i) * 4)
 
 #define HSW_STEREO_3D_CTL_A	0x70020
 #define   S3D_ENABLE		(1<<31)
diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index 3b28ed3..9eafa19 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -113,17 +113,18 @@ static u32 hsw_infoframe_enable(enum hdmi_infoframe_type type)
 	}
 }
 
-static u32 hsw_infoframe_data_reg(enum hdmi_infoframe_type type,
-				  enum transcoder cpu_transcoder,
-				  struct drm_i915_private *dev_priv)
+static u32 hsw_dip_data_reg(struct drm_i915_private *dev_priv,
+			    enum transcoder cpu_transcoder,
+			    enum hdmi_infoframe_type type,
+			    int i)
 {
 	switch (type) {
 	case HDMI_INFOFRAME_TYPE_AVI:
-		return HSW_TVIDEO_DIP_AVI_DATA(cpu_transcoder);
+		return HSW_TVIDEO_DIP_AVI_DATA(cpu_transcoder, i);
 	case HDMI_INFOFRAME_TYPE_SPD:
-		return HSW_TVIDEO_DIP_SPD_DATA(cpu_transcoder);
+		return HSW_TVIDEO_DIP_SPD_DATA(cpu_transcoder, i);
 	case HDMI_INFOFRAME_TYPE_VENDOR:
-		return HSW_TVIDEO_DIP_VS_DATA(cpu_transcoder);
+		return HSW_TVIDEO_DIP_VS_DATA(cpu_transcoder, i);
 	default:
 		DRM_DEBUG_DRIVER("unknown info frame type %d\n", type);
 		return 0;
@@ -365,14 +366,13 @@ static void hsw_write_infoframe(struct drm_encoder *encoder,
 	struct drm_device *dev = encoder->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc);
-	u32 ctl_reg = HSW_TVIDEO_DIP_CTL(intel_crtc->config->cpu_transcoder);
+	enum transcoder cpu_transcoder = intel_crtc->config->cpu_transcoder;
+	u32 ctl_reg = HSW_TVIDEO_DIP_CTL(cpu_transcoder);
 	u32 data_reg;
 	int i;
 	u32 val = I915_READ(ctl_reg);
 
-	data_reg = hsw_infoframe_data_reg(type,
-					  intel_crtc->config->cpu_transcoder,
-					  dev_priv);
+	data_reg = hsw_dip_data_reg(dev_priv, cpu_transcoder, type, 0);
 	if (data_reg == 0)
 		return;
 
@@ -381,12 +381,14 @@ static void hsw_write_infoframe(struct drm_encoder *encoder,
 
 	mmiowb();
 	for (i = 0; i < len; i += 4) {
-		I915_WRITE(data_reg + i, *data);
+		I915_WRITE(hsw_dip_data_reg(dev_priv, cpu_transcoder,
+					    type, i >> 2), *data);
 		data++;
 	}
 	/* Write every possible data byte to force correct ECC calculation. */
 	for (; i < VIDEO_DIP_DATA_SIZE; i += 4)
-		I915_WRITE(data_reg + i, 0);
+		I915_WRITE(hsw_dip_data_reg(dev_priv, cpu_transcoder,
+					    type, i >> 2), 0);
 	mmiowb();
 
 	val |= hsw_infoframe_enable(type);
diff --git a/drivers/gpu/drm/i915/intel_psr.c b/drivers/gpu/drm/i915/intel_psr.c
index a04b4dc..213581c 100644
--- a/drivers/gpu/drm/i915/intel_psr.c
+++ b/drivers/gpu/drm/i915/intel_psr.c
@@ -73,14 +73,14 @@ static bool vlv_is_psr_active_on_pipe(struct drm_device *dev, int pipe)
 }
 
 static void intel_psr_write_vsc(struct intel_dp *intel_dp,
-				    struct edp_vsc_psr *vsc_psr)
+				const struct edp_vsc_psr *vsc_psr)
 {
 	struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp);
 	struct drm_device *dev = dig_port->base.base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *crtc = to_intel_crtc(dig_port->base.base.crtc);
-	u32 ctl_reg = HSW_TVIDEO_DIP_CTL(crtc->config->cpu_transcoder);
-	u32 data_reg = HSW_TVIDEO_DIP_VSC_DATA(crtc->config->cpu_transcoder);
+	enum transcoder cpu_transcoder = crtc->config->cpu_transcoder;
+	u32 ctl_reg = HSW_TVIDEO_DIP_CTL(cpu_transcoder);
 	uint32_t *data = (uint32_t *) vsc_psr;
 	unsigned int i;
 
@@ -90,12 +90,14 @@ static void intel_psr_write_vsc(struct intel_dp *intel_dp,
 	I915_WRITE(ctl_reg, 0);
 	POSTING_READ(ctl_reg);
 
-	for (i = 0; i < VIDEO_DIP_VSC_DATA_SIZE; i += 4) {
-		if (i < sizeof(struct edp_vsc_psr))
-			I915_WRITE(data_reg + i, *data++);
-		else
-			I915_WRITE(data_reg + i, 0);
+	for (i = 0; i < sizeof(*vsc_psr); i += 4) {
+		I915_WRITE(HSW_TVIDEO_DIP_VSC_DATA(cpu_transcoder,
+						   i >> 2), *data);
+		data++;
 	}
+	for (; i < VIDEO_DIP_VSC_DATA_SIZE; i += 4)
+		I915_WRITE(HSW_TVIDEO_DIP_VSC_DATA(cpu_transcoder,
+						   i >> 2), 0);
 
 	I915_WRITE(ctl_reg, VIDEO_DIP_ENABLE_VSC_HSW);
 	POSTING_READ(ctl_reg);
-- 
cgit v0.10.2


From 699fc401da1c9cc8c6bda578ca3d6310924276a2 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 18 Sep 2015 20:03:38 +0300
Subject: drm/i915: Include gpio_mmio_base in GMBUS reg defines
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 72741f4..b02356f 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -2144,7 +2144,7 @@ enum skl_disp_power_wells {
 # define GPIO_DATA_VAL_IN		(1 << 12)
 # define GPIO_DATA_PULLUP_DISABLE	(1 << 13)
 
-#define GMBUS0			0x5100 /* clock/port select */
+#define GMBUS0			(dev_priv->gpio_mmio_base + 0x5100) /* clock/port select */
 #define   GMBUS_RATE_100KHZ	(0<<8)
 #define   GMBUS_RATE_50KHZ	(1<<8)
 #define   GMBUS_RATE_400KHZ	(2<<8) /* reserved on Pineview */
@@ -2163,7 +2163,7 @@ enum skl_disp_power_wells {
 #define   GMBUS_PIN_2_BXT	2
 #define   GMBUS_PIN_3_BXT	3
 #define   GMBUS_NUM_PINS	7 /* including 0 */
-#define GMBUS1			0x5104 /* command/status */
+#define GMBUS1			(dev_priv->gpio_mmio_base + 0x5104) /* command/status */
 #define   GMBUS_SW_CLR_INT	(1<<31)
 #define   GMBUS_SW_RDY		(1<<30)
 #define   GMBUS_ENT		(1<<29) /* enable timeout */
@@ -2177,7 +2177,7 @@ enum skl_disp_power_wells {
 #define   GMBUS_SLAVE_ADDR_SHIFT 1
 #define   GMBUS_SLAVE_READ	(1<<0)
 #define   GMBUS_SLAVE_WRITE	(0<<0)
-#define GMBUS2			0x5108 /* status */
+#define GMBUS2			(dev_priv->gpio_mmio_base + 0x5108) /* status */
 #define   GMBUS_INUSE		(1<<15)
 #define   GMBUS_HW_WAIT_PHASE	(1<<14)
 #define   GMBUS_STALL_TIMEOUT	(1<<13)
@@ -2185,14 +2185,14 @@ enum skl_disp_power_wells {
 #define   GMBUS_HW_RDY		(1<<11)
 #define   GMBUS_SATOER		(1<<10)
 #define   GMBUS_ACTIVE		(1<<9)
-#define GMBUS3			0x510c /* data buffer bytes 3-0 */
-#define GMBUS4			0x5110 /* interrupt mask (Pineview+) */
+#define GMBUS3			(dev_priv->gpio_mmio_base + 0x510c) /* data buffer bytes 3-0 */
+#define GMBUS4			(dev_priv->gpio_mmio_base + 0x5110) /* interrupt mask (Pineview+) */
 #define   GMBUS_SLAVE_TIMEOUT_EN (1<<4)
 #define   GMBUS_NAK_EN		(1<<3)
 #define   GMBUS_IDLE_EN		(1<<2)
 #define   GMBUS_HW_WAIT_EN	(1<<1)
 #define   GMBUS_HW_RDY_EN	(1<<0)
-#define GMBUS5			0x5120 /* byte index */
+#define GMBUS5			(dev_priv->gpio_mmio_base + 0x5120) /* byte index */
 #define   GMBUS_2BYTE_INDEX_EN	(1<<31)
 
 /*
diff --git a/drivers/gpu/drm/i915/intel_i2c.c b/drivers/gpu/drm/i915/intel_i2c.c
index a64f26c..1369fc4 100644
--- a/drivers/gpu/drm/i915/intel_i2c.c
+++ b/drivers/gpu/drm/i915/intel_i2c.c
@@ -114,8 +114,8 @@ intel_i2c_reset(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	I915_WRITE(dev_priv->gpio_mmio_base + GMBUS0, 0);
-	I915_WRITE(dev_priv->gpio_mmio_base + GMBUS4, 0);
+	I915_WRITE(GMBUS0, 0);
+	I915_WRITE(GMBUS4, 0);
 }
 
 static void intel_i2c_quirk_set(struct drm_i915_private *dev_priv, bool enable)
@@ -261,7 +261,6 @@ gmbus_wait_hw_status(struct drm_i915_private *dev_priv,
 		     u32 gmbus4_irq_en)
 {
 	int i;
-	int reg_offset = dev_priv->gpio_mmio_base;
 	u32 gmbus2 = 0;
 	DEFINE_WAIT(wait);
 
@@ -271,13 +270,13 @@ gmbus_wait_hw_status(struct drm_i915_private *dev_priv,
 	/* Important: The hw handles only the first bit, so set only one! Since
 	 * we also need to check for NAKs besides the hw ready/idle signal, we
 	 * need to wake up periodically and check that ourselves. */
-	I915_WRITE(GMBUS4 + reg_offset, gmbus4_irq_en);
+	I915_WRITE(GMBUS4, gmbus4_irq_en);
 
 	for (i = 0; i < msecs_to_jiffies_timeout(50); i++) {
 		prepare_to_wait(&dev_priv->gmbus_wait_queue, &wait,
 				TASK_UNINTERRUPTIBLE);
 
-		gmbus2 = I915_READ_NOTRACE(GMBUS2 + reg_offset);
+		gmbus2 = I915_READ_NOTRACE(GMBUS2);
 		if (gmbus2 & (GMBUS_SATOER | gmbus2_status))
 			break;
 
@@ -285,7 +284,7 @@ gmbus_wait_hw_status(struct drm_i915_private *dev_priv,
 	}
 	finish_wait(&dev_priv->gmbus_wait_queue, &wait);
 
-	I915_WRITE(GMBUS4 + reg_offset, 0);
+	I915_WRITE(GMBUS4, 0);
 
 	if (gmbus2 & GMBUS_SATOER)
 		return -ENXIO;
@@ -298,20 +297,19 @@ static int
 gmbus_wait_idle(struct drm_i915_private *dev_priv)
 {
 	int ret;
-	int reg_offset = dev_priv->gpio_mmio_base;
 
-#define C ((I915_READ_NOTRACE(GMBUS2 + reg_offset) & GMBUS_ACTIVE) == 0)
+#define C ((I915_READ_NOTRACE(GMBUS2) & GMBUS_ACTIVE) == 0)
 
 	if (!HAS_GMBUS_IRQ(dev_priv->dev))
 		return wait_for(C, 10);
 
 	/* Important: The hw handles only the first bit, so set only one! */
-	I915_WRITE(GMBUS4 + reg_offset, GMBUS_IDLE_EN);
+	I915_WRITE(GMBUS4, GMBUS_IDLE_EN);
 
 	ret = wait_event_timeout(dev_priv->gmbus_wait_queue, C,
 				 msecs_to_jiffies_timeout(10));
 
-	I915_WRITE(GMBUS4 + reg_offset, 0);
+	I915_WRITE(GMBUS4, 0);
 
 	if (ret)
 		return 0;
@@ -325,9 +323,7 @@ gmbus_xfer_read_chunk(struct drm_i915_private *dev_priv,
 		      unsigned short addr, u8 *buf, unsigned int len,
 		      u32 gmbus1_index)
 {
-	int reg_offset = dev_priv->gpio_mmio_base;
-
-	I915_WRITE(GMBUS1 + reg_offset,
+	I915_WRITE(GMBUS1,
 		   gmbus1_index |
 		   GMBUS_CYCLE_WAIT |
 		   (len << GMBUS_BYTE_COUNT_SHIFT) |
@@ -342,7 +338,7 @@ gmbus_xfer_read_chunk(struct drm_i915_private *dev_priv,
 		if (ret)
 			return ret;
 
-		val = I915_READ(GMBUS3 + reg_offset);
+		val = I915_READ(GMBUS3);
 		do {
 			*buf++ = val & 0xff;
 			val >>= 8;
@@ -380,7 +376,6 @@ static int
 gmbus_xfer_write_chunk(struct drm_i915_private *dev_priv,
 		       unsigned short addr, u8 *buf, unsigned int len)
 {
-	int reg_offset = dev_priv->gpio_mmio_base;
 	unsigned int chunk_size = len;
 	u32 val, loop;
 
@@ -390,8 +385,8 @@ gmbus_xfer_write_chunk(struct drm_i915_private *dev_priv,
 		len -= 1;
 	}
 
-	I915_WRITE(GMBUS3 + reg_offset, val);
-	I915_WRITE(GMBUS1 + reg_offset,
+	I915_WRITE(GMBUS3, val);
+	I915_WRITE(GMBUS1,
 		   GMBUS_CYCLE_WAIT |
 		   (chunk_size << GMBUS_BYTE_COUNT_SHIFT) |
 		   (addr << GMBUS_SLAVE_ADDR_SHIFT) |
@@ -404,7 +399,7 @@ gmbus_xfer_write_chunk(struct drm_i915_private *dev_priv,
 			val |= *buf++ << (8 * loop);
 		} while (--len && ++loop < 4);
 
-		I915_WRITE(GMBUS3 + reg_offset, val);
+		I915_WRITE(GMBUS3, val);
 
 		ret = gmbus_wait_hw_status(dev_priv, GMBUS_HW_RDY,
 					   GMBUS_HW_RDY_EN);
@@ -452,7 +447,6 @@ gmbus_is_index_read(struct i2c_msg *msgs, int i, int num)
 static int
 gmbus_xfer_index_read(struct drm_i915_private *dev_priv, struct i2c_msg *msgs)
 {
-	int reg_offset = dev_priv->gpio_mmio_base;
 	u32 gmbus1_index = 0;
 	u32 gmbus5 = 0;
 	int ret;
@@ -466,13 +460,13 @@ gmbus_xfer_index_read(struct drm_i915_private *dev_priv, struct i2c_msg *msgs)
 
 	/* GMBUS5 holds 16-bit index */
 	if (gmbus5)
-		I915_WRITE(GMBUS5 + reg_offset, gmbus5);
+		I915_WRITE(GMBUS5, gmbus5);
 
 	ret = gmbus_xfer_read(dev_priv, &msgs[1], gmbus1_index);
 
 	/* Clear GMBUS5 after each index transfer */
 	if (gmbus5)
-		I915_WRITE(GMBUS5 + reg_offset, 0);
+		I915_WRITE(GMBUS5, 0);
 
 	return ret;
 }
@@ -486,7 +480,7 @@ gmbus_xfer(struct i2c_adapter *adapter,
 					       struct intel_gmbus,
 					       adapter);
 	struct drm_i915_private *dev_priv = bus->dev_priv;
-	int i = 0, inc, try = 0, reg_offset;
+	int i = 0, inc, try = 0;
 	int ret = 0;
 
 	intel_aux_display_runtime_get(dev_priv);
@@ -497,10 +491,8 @@ gmbus_xfer(struct i2c_adapter *adapter,
 		goto out;
 	}
 
-	reg_offset = dev_priv->gpio_mmio_base;
-
 retry:
-	I915_WRITE(GMBUS0 + reg_offset, bus->reg0);
+	I915_WRITE(GMBUS0, bus->reg0);
 
 	for (; i < num; i += inc) {
 		inc = 1;
@@ -530,7 +522,7 @@ retry:
 	 * a STOP on the very first cycle. To simplify the code we
 	 * unconditionally generate the STOP condition with an additional gmbus
 	 * cycle. */
-	I915_WRITE(GMBUS1 + reg_offset, GMBUS_CYCLE_STOP | GMBUS_SW_RDY);
+	I915_WRITE(GMBUS1, GMBUS_CYCLE_STOP | GMBUS_SW_RDY);
 
 	/* Mark the GMBUS interface as disabled after waiting for idle.
 	 * We will re-enable it at the start of the next xfer,
@@ -541,7 +533,7 @@ retry:
 			 adapter->name);
 		ret = -ETIMEDOUT;
 	}
-	I915_WRITE(GMBUS0 + reg_offset, 0);
+	I915_WRITE(GMBUS0, 0);
 	ret = ret ?: i;
 	goto out;
 
@@ -570,9 +562,9 @@ clear_err:
 	 * of resetting the GMBUS controller and so clearing the
 	 * BUS_ERROR raised by the slave's NAK.
 	 */
-	I915_WRITE(GMBUS1 + reg_offset, GMBUS_SW_CLR_INT);
-	I915_WRITE(GMBUS1 + reg_offset, 0);
-	I915_WRITE(GMBUS0 + reg_offset, 0);
+	I915_WRITE(GMBUS1, GMBUS_SW_CLR_INT);
+	I915_WRITE(GMBUS1, 0);
+	I915_WRITE(GMBUS0, 0);
 
 	DRM_DEBUG_KMS("GMBUS [%s] NAK for addr: %04x %c(%d)\n",
 			 adapter->name, msgs[i].addr,
@@ -595,7 +587,7 @@ clear_err:
 timeout:
 	DRM_INFO("GMBUS [%s] timed out, falling back to bit banging on pin %d\n",
 		 bus->adapter.name, bus->reg0 & 0xff);
-	I915_WRITE(GMBUS0 + reg_offset, 0);
+	I915_WRITE(GMBUS0, 0);
 
 	/* Hardware may not support GMBUS over these pins? Try GPIO bitbanging instead. */
 	bus->force_bit = 1;
-- 
cgit v0.10.2


From 68d9753837db0e45dadd16a312d479adf3170b2c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 18 Sep 2015 20:03:39 +0300
Subject: drm/i915: Protect register macro arguments
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Always put parens around macro argument evaluations.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index b02356f..93c6ec1 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -429,7 +429,7 @@
 #define   ASYNC_FLIP                (1<<22)
 #define   DISPLAY_PLANE_A           (0<<20)
 #define   DISPLAY_PLANE_B           (1<<20)
-#define GFX_OP_PIPE_CONTROL(len)	((0x3<<29)|(0x3<<27)|(0x2<<24)|(len-2))
+#define GFX_OP_PIPE_CONTROL(len)	((0x3<<29)|(0x3<<27)|(0x2<<24)|((len)-2))
 #define   PIPE_CONTROL_FLUSH_L3				(1<<27)
 #define   PIPE_CONTROL_GLOBAL_GTT_IVB			(1<<24) /* gen7+ */
 #define   PIPE_CONTROL_MMIO_WRITE			(1<<23)
@@ -1255,7 +1255,7 @@ enum skl_disp_power_wells {
 #define  PORT_PLL_DCO_AMP_OVR_EN_H	(1<<27)
 #define  PORT_PLL_DCO_AMP_DEFAULT	15
 #define  PORT_PLL_DCO_AMP_MASK		0x3c00
-#define  PORT_PLL_DCO_AMP(x)		(x<<10)
+#define  PORT_PLL_DCO_AMP(x)		((x)<<10)
 #define _PORT_PLL_BASE(port)		_PORT3(port, _PORT_PLL_0_A,	\
 						_PORT_PLL_0_B,		\
 						_PORT_PLL_0_C)
@@ -1552,8 +1552,8 @@ enum skl_disp_power_wells {
 #define RENDER_HWS_PGA_GEN7	(0x04080)
 #define RING_FAULT_REG(ring)	(0x4094 + 0x100*(ring)->id)
 #define   RING_FAULT_GTTSEL_MASK (1<<11)
-#define   RING_FAULT_SRCID(x)	((x >> 3) & 0xff)
-#define   RING_FAULT_FAULT_TYPE(x) ((x >> 1) & 0x3)
+#define   RING_FAULT_SRCID(x)	(((x) >> 3) & 0xff)
+#define   RING_FAULT_FAULT_TYPE(x) (((x) >> 1) & 0x3)
 #define   RING_FAULT_VALID	(1<<0)
 #define DONE_REG		0x40b0
 #define GEN8_PRIVATE_PAT_LO	0x40e0
@@ -1641,9 +1641,9 @@ enum skl_disp_power_wells {
 #define   ERR_INT_PIPE_CRC_DONE_B	(1<<5)
 #define   ERR_INT_FIFO_UNDERRUN_B	(1<<3)
 #define   ERR_INT_PIPE_CRC_DONE_A	(1<<2)
-#define   ERR_INT_PIPE_CRC_DONE(pipe)	(1<<(2 + pipe*3))
+#define   ERR_INT_PIPE_CRC_DONE(pipe)	(1<<(2 + (pipe)*3))
 #define   ERR_INT_FIFO_UNDERRUN_A	(1<<0)
-#define   ERR_INT_FIFO_UNDERRUN(pipe)	(1<<(pipe*3))
+#define   ERR_INT_FIFO_UNDERRUN(pipe)	(1<<((pipe)*3))
 
 #define GEN8_FAULT_TLB_DATA0		0x04b10
 #define GEN8_FAULT_TLB_DATA1		0x04b14
@@ -1704,8 +1704,8 @@ enum skl_disp_power_wells {
 #define   GEN6_WIZ_HASHING_16x4				GEN6_WIZ_HASHING(1, 0)
 #define   GEN6_WIZ_HASHING_MASK				GEN6_WIZ_HASHING(1, 1)
 #define   GEN6_TD_FOUR_ROW_DISPATCH_DISABLE		(1 << 5)
-#define   GEN9_IZ_HASHING_MASK(slice)			(0x3 << (slice * 2))
-#define   GEN9_IZ_HASHING(slice, val)			((val) << (slice * 2))
+#define   GEN9_IZ_HASHING_MASK(slice)			(0x3 << ((slice) * 2))
+#define   GEN9_IZ_HASHING(slice, val)			((val) << ((slice) * 2))
 
 #define GFX_MODE	0x02520
 #define GFX_MODE_GEN7	0x0229c
@@ -2866,21 +2866,21 @@ enum skl_disp_power_wells {
  *   doesn't need saving on GT1
  */
 #define CXT_SIZE		0x21a0
-#define GEN6_CXT_POWER_SIZE(cxt_reg)	((cxt_reg >> 24) & 0x3f)
-#define GEN6_CXT_RING_SIZE(cxt_reg)	((cxt_reg >> 18) & 0x3f)
-#define GEN6_CXT_RENDER_SIZE(cxt_reg)	((cxt_reg >> 12) & 0x3f)
-#define GEN6_CXT_EXTENDED_SIZE(cxt_reg)	((cxt_reg >> 6) & 0x3f)
-#define GEN6_CXT_PIPELINE_SIZE(cxt_reg)	((cxt_reg >> 0) & 0x3f)
+#define GEN6_CXT_POWER_SIZE(cxt_reg)	(((cxt_reg) >> 24) & 0x3f)
+#define GEN6_CXT_RING_SIZE(cxt_reg)	(((cxt_reg) >> 18) & 0x3f)
+#define GEN6_CXT_RENDER_SIZE(cxt_reg)	(((cxt_reg) >> 12) & 0x3f)
+#define GEN6_CXT_EXTENDED_SIZE(cxt_reg)	(((cxt_reg) >> 6) & 0x3f)
+#define GEN6_CXT_PIPELINE_SIZE(cxt_reg)	(((cxt_reg) >> 0) & 0x3f)
 #define GEN6_CXT_TOTAL_SIZE(cxt_reg)	(GEN6_CXT_RING_SIZE(cxt_reg) + \
 					GEN6_CXT_EXTENDED_SIZE(cxt_reg) + \
 					GEN6_CXT_PIPELINE_SIZE(cxt_reg))
 #define GEN7_CXT_SIZE		0x21a8
-#define GEN7_CXT_POWER_SIZE(ctx_reg)	((ctx_reg >> 25) & 0x7f)
-#define GEN7_CXT_RING_SIZE(ctx_reg)	((ctx_reg >> 22) & 0x7)
-#define GEN7_CXT_RENDER_SIZE(ctx_reg)	((ctx_reg >> 16) & 0x3f)
-#define GEN7_CXT_EXTENDED_SIZE(ctx_reg)	((ctx_reg >> 9) & 0x7f)
-#define GEN7_CXT_GT1_SIZE(ctx_reg)	((ctx_reg >> 6) & 0x7)
-#define GEN7_CXT_VFSTATE_SIZE(ctx_reg)	((ctx_reg >> 0) & 0x3f)
+#define GEN7_CXT_POWER_SIZE(ctx_reg)	(((ctx_reg) >> 25) & 0x7f)
+#define GEN7_CXT_RING_SIZE(ctx_reg)	(((ctx_reg) >> 22) & 0x7)
+#define GEN7_CXT_RENDER_SIZE(ctx_reg)	(((ctx_reg) >> 16) & 0x3f)
+#define GEN7_CXT_EXTENDED_SIZE(ctx_reg)	(((ctx_reg) >> 9) & 0x7f)
+#define GEN7_CXT_GT1_SIZE(ctx_reg)	(((ctx_reg) >> 6) & 0x7)
+#define GEN7_CXT_VFSTATE_SIZE(ctx_reg)	(((ctx_reg) >> 0) & 0x3f)
 #define GEN7_CXT_TOTAL_SIZE(ctx_reg)	(GEN7_CXT_EXTENDED_SIZE(ctx_reg) + \
 					 GEN7_CXT_VFSTATE_SIZE(ctx_reg))
 /* Haswell does have the CXT_SIZE register however it does not appear to be
@@ -5223,18 +5223,18 @@ enum skl_disp_power_wells {
 #define _SPBCONSTALPHA		(VLV_DISPLAY_BASE + 0x722a8)
 #define _SPBGAMC		(VLV_DISPLAY_BASE + 0x722f4)
 
-#define SPCNTR(pipe, plane) _PIPE(pipe * 2 + plane, _SPACNTR, _SPBCNTR)
-#define SPLINOFF(pipe, plane) _PIPE(pipe * 2 + plane, _SPALINOFF, _SPBLINOFF)
-#define SPSTRIDE(pipe, plane) _PIPE(pipe * 2 + plane, _SPASTRIDE, _SPBSTRIDE)
-#define SPPOS(pipe, plane) _PIPE(pipe * 2 + plane, _SPAPOS, _SPBPOS)
-#define SPSIZE(pipe, plane) _PIPE(pipe * 2 + plane, _SPASIZE, _SPBSIZE)
-#define SPKEYMINVAL(pipe, plane) _PIPE(pipe * 2 + plane, _SPAKEYMINVAL, _SPBKEYMINVAL)
-#define SPKEYMSK(pipe, plane) _PIPE(pipe * 2 + plane, _SPAKEYMSK, _SPBKEYMSK)
-#define SPSURF(pipe, plane) _PIPE(pipe * 2 + plane, _SPASURF, _SPBSURF)
-#define SPKEYMAXVAL(pipe, plane) _PIPE(pipe * 2 + plane, _SPAKEYMAXVAL, _SPBKEYMAXVAL)
-#define SPTILEOFF(pipe, plane) _PIPE(pipe * 2 + plane, _SPATILEOFF, _SPBTILEOFF)
-#define SPCONSTALPHA(pipe, plane) _PIPE(pipe * 2 + plane, _SPACONSTALPHA, _SPBCONSTALPHA)
-#define SPGAMC(pipe, plane) _PIPE(pipe * 2 + plane, _SPAGAMC, _SPBGAMC)
+#define SPCNTR(pipe, plane) _PIPE((pipe) * 2 + (plane), _SPACNTR, _SPBCNTR)
+#define SPLINOFF(pipe, plane) _PIPE((pipe) * 2 + (plane), _SPALINOFF, _SPBLINOFF)
+#define SPSTRIDE(pipe, plane) _PIPE((pipe) * 2 + (plane), _SPASTRIDE, _SPBSTRIDE)
+#define SPPOS(pipe, plane) _PIPE((pipe) * 2 + (plane), _SPAPOS, _SPBPOS)
+#define SPSIZE(pipe, plane) _PIPE((pipe) * 2 + (plane), _SPASIZE, _SPBSIZE)
+#define SPKEYMINVAL(pipe, plane) _PIPE((pipe) * 2 + (plane), _SPAKEYMINVAL, _SPBKEYMINVAL)
+#define SPKEYMSK(pipe, plane) _PIPE((pipe) * 2 + (plane), _SPAKEYMSK, _SPBKEYMSK)
+#define SPSURF(pipe, plane) _PIPE((pipe) * 2 + (plane), _SPASURF, _SPBSURF)
+#define SPKEYMAXVAL(pipe, plane) _PIPE((pipe) * 2 + (plane), _SPAKEYMAXVAL, _SPBKEYMAXVAL)
+#define SPTILEOFF(pipe, plane) _PIPE((pipe) * 2 + (plane), _SPATILEOFF, _SPBTILEOFF)
+#define SPCONSTALPHA(pipe, plane) _PIPE((pipe) * 2 + (plane), _SPACONSTALPHA, _SPBCONSTALPHA)
+#define SPGAMC(pipe, plane) _PIPE((pipe) * 2 + (plane), _SPAGAMC, _SPBGAMC)
 
 /*
  * CHV pipe B sprite CSC
@@ -5580,7 +5580,7 @@ enum skl_disp_power_wells {
 #define PS_SCALER_MODE_DYN  (0 << 28)
 #define PS_SCALER_MODE_HQ  (1 << 28)
 #define PS_PLANE_SEL_MASK  (7 << 25)
-#define PS_PLANE_SEL(plane) ((plane + 1) << 25)
+#define PS_PLANE_SEL(plane) (((plane) + 1) << 25)
 #define PS_FILTER_MASK         (3 << 23)
 #define PS_FILTER_MEDIUM       (0 << 23)
 #define PS_FILTER_EDGE_ENHANCE (2 << 23)
@@ -5745,7 +5745,7 @@ enum skl_disp_power_wells {
 #define DE_PLANEA_FLIP_DONE_IVB		(1<<3)
 #define DE_PLANE_FLIP_DONE_IVB(plane)	(1<< (3 + 5*(plane)))
 #define DE_PIPEA_VBLANK_IVB		(1<<0)
-#define DE_PIPE_VBLANK_IVB(pipe)	(1 << (pipe * 5))
+#define DE_PIPE_VBLANK_IVB(pipe)	(1 << ((pipe) * 5))
 
 #define VLV_MASTER_IER			0x4400c /* Gunit master IER */
 #define   MASTER_INTERRUPT_ENABLE	(1<<31)
@@ -5769,7 +5769,7 @@ enum skl_disp_power_wells {
 #define  GEN8_DE_PIPE_C_IRQ		(1<<18)
 #define  GEN8_DE_PIPE_B_IRQ		(1<<17)
 #define  GEN8_DE_PIPE_A_IRQ		(1<<16)
-#define  GEN8_DE_PIPE_IRQ(pipe)		(1<<(16+pipe))
+#define  GEN8_DE_PIPE_IRQ(pipe)		(1<<(16+(pipe)))
 #define  GEN8_GT_VECS_IRQ		(1<<6)
 #define  GEN8_GT_PM_IRQ			(1<<4)
 #define  GEN8_GT_VCS2_IRQ		(1<<3)
@@ -5813,7 +5813,7 @@ enum skl_disp_power_wells {
 #define  GEN9_PIPE_PLANE3_FLIP_DONE	(1 << 5)
 #define  GEN9_PIPE_PLANE2_FLIP_DONE	(1 << 4)
 #define  GEN9_PIPE_PLANE1_FLIP_DONE	(1 << 3)
-#define  GEN9_PIPE_PLANE_FLIP_DONE(p)	(1 << (3 + p))
+#define  GEN9_PIPE_PLANE_FLIP_DONE(p)	(1 << (3 + (p)))
 #define GEN8_DE_PIPE_IRQ_FAULT_ERRORS \
 	(GEN8_PIPE_CURSOR_FAULT | \
 	 GEN8_PIPE_SPRITE_FAULT | \
@@ -6072,7 +6072,7 @@ enum skl_disp_power_wells {
 #define  SERR_INT_TRANS_C_FIFO_UNDERRUN	(1<<6)
 #define  SERR_INT_TRANS_B_FIFO_UNDERRUN	(1<<3)
 #define  SERR_INT_TRANS_A_FIFO_UNDERRUN	(1<<0)
-#define  SERR_INT_TRANS_FIFO_UNDERRUN(pipe)	(1<<(pipe*3))
+#define  SERR_INT_TRANS_FIFO_UNDERRUN(pipe)	(1<<((pipe)*3))
 
 /* digital port hotplug */
 #define PCH_PORT_HOTPLUG		0xc4030	/* SHOTPLUG_CTL */
@@ -6183,9 +6183,9 @@ enum skl_disp_power_wells {
 #define PCH_SSC4_AUX_PARMS      0xc6214
 
 #define PCH_DPLL_SEL		0xc7000
-#define	 TRANS_DPLLB_SEL(pipe)		(1 << (pipe * 4))
+#define	 TRANS_DPLLB_SEL(pipe)		(1 << ((pipe) * 4))
 #define	 TRANS_DPLLA_SEL(pipe)		0
-#define  TRANS_DPLL_ENABLE(pipe)	(1 << (pipe * 4 + 3))
+#define  TRANS_DPLL_ENABLE(pipe)	(1 << ((pipe) * 4 + 3))
 
 /* transcoder */
 
@@ -7348,7 +7348,7 @@ enum skl_disp_power_wells {
 #define TRANS_CLK_SEL(tran) _TRANSCODER(tran, TRANS_CLK_SEL_A, TRANS_CLK_SEL_B)
 /* For each transcoder, we need to select the corresponding port clock */
 #define  TRANS_CLK_SEL_DISABLED		(0x0<<29)
-#define  TRANS_CLK_SEL_PORT(x)		((x+1)<<29)
+#define  TRANS_CLK_SEL_PORT(x)		(((x)+1)<<29)
 
 #define TRANSA_MSA_MISC			0x60410
 #define TRANSB_MSA_MISC			0x61410
@@ -7421,10 +7421,10 @@ enum skl_disp_power_wells {
 
 /* DPLL control2 */
 #define DPLL_CTRL2				0x6C05C
-#define  DPLL_CTRL2_DDI_CLK_OFF(port)		(1<<(port+15))
+#define  DPLL_CTRL2_DDI_CLK_OFF(port)		(1<<((port)+15))
 #define  DPLL_CTRL2_DDI_CLK_SEL_MASK(port)	(3<<((port)*3+1))
 #define  DPLL_CTRL2_DDI_CLK_SEL_SHIFT(port)    ((port)*3+1)
-#define  DPLL_CTRL2_DDI_CLK_SEL(clk, port)	(clk<<((port)*3+1))
+#define  DPLL_CTRL2_DDI_CLK_SEL(clk, port)	((clk)<<((port)*3+1))
 #define  DPLL_CTRL2_DDI_SEL_OVERRIDE(port)     (1<<((port)*3))
 
 /* DPLL Status */
@@ -7437,23 +7437,23 @@ enum skl_disp_power_wells {
 #define DPLL3_CFGCR1	0x6C050
 #define  DPLL_CFGCR1_FREQ_ENABLE	(1<<31)
 #define  DPLL_CFGCR1_DCO_FRACTION_MASK	(0x7fff<<9)
-#define  DPLL_CFGCR1_DCO_FRACTION(x)	(x<<9)
+#define  DPLL_CFGCR1_DCO_FRACTION(x)	((x)<<9)
 #define  DPLL_CFGCR1_DCO_INTEGER_MASK	(0x1ff)
 
 #define DPLL1_CFGCR2	0x6C044
 #define DPLL2_CFGCR2	0x6C04C
 #define DPLL3_CFGCR2	0x6C054
 #define  DPLL_CFGCR2_QDIV_RATIO_MASK	(0xff<<8)
-#define  DPLL_CFGCR2_QDIV_RATIO(x)	(x<<8)
-#define  DPLL_CFGCR2_QDIV_MODE(x)	(x<<7)
+#define  DPLL_CFGCR2_QDIV_RATIO(x)	((x)<<8)
+#define  DPLL_CFGCR2_QDIV_MODE(x)	((x)<<7)
 #define  DPLL_CFGCR2_KDIV_MASK		(3<<5)
-#define  DPLL_CFGCR2_KDIV(x)		(x<<5)
+#define  DPLL_CFGCR2_KDIV(x)		((x)<<5)
 #define  DPLL_CFGCR2_KDIV_5 (0<<5)
 #define  DPLL_CFGCR2_KDIV_2 (1<<5)
 #define  DPLL_CFGCR2_KDIV_3 (2<<5)
 #define  DPLL_CFGCR2_KDIV_1 (3<<5)
 #define  DPLL_CFGCR2_PDIV_MASK		(7<<2)
-#define  DPLL_CFGCR2_PDIV(x)		(x<<2)
+#define  DPLL_CFGCR2_PDIV(x)		((x)<<2)
 #define  DPLL_CFGCR2_PDIV_1 (0<<2)
 #define  DPLL_CFGCR2_PDIV_2 (1<<2)
 #define  DPLL_CFGCR2_PDIV_3 (2<<2)
-- 
cgit v0.10.2


From 395b2913e36ffb6a09057ea0b069113960dd3a06 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 18 Sep 2015 20:03:40 +0300
Subject: drm/i915: Fix a few bad hex numbers in register defines
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

A few register mask defines were missing the '0x' from hex numbers. Or
at least I assume those were meant to be hex numbers. Put the '0x' in
place.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 93c6ec1..f24782b 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -4284,7 +4284,7 @@ enum skl_disp_power_wells {
 #define   DP_AUX_CH_CTL_PSR_DATA_AUX_REG_SKL	(1 << 14)
 #define   DP_AUX_CH_CTL_FS_DATA_AUX_REG_SKL	(1 << 13)
 #define   DP_AUX_CH_CTL_GTC_DATA_AUX_REG_SKL	(1 << 12)
-#define   DP_AUX_CH_CTL_FW_SYNC_PULSE_SKL_MASK (1f << 5)
+#define   DP_AUX_CH_CTL_FW_SYNC_PULSE_SKL_MASK (0x1f << 5)
 #define   DP_AUX_CH_CTL_FW_SYNC_PULSE_SKL(c) (((c) - 1) << 5)
 #define   DP_AUX_CH_CTL_SYNC_PULSE_SKL(c)   ((c) - 1)
 
@@ -7979,7 +7979,7 @@ enum skl_disp_power_wells {
 #define  VIRTUAL_CHANNEL_SHIFT				6
 #define  VIRTUAL_CHANNEL_MASK				(3 << 6)
 #define  DATA_TYPE_SHIFT				0
-#define  DATA_TYPE_MASK					(3f << 0)
+#define  DATA_TYPE_MASK					(0x3f << 0)
 /* data type values, see include/video/mipi_display.h */
 
 #define _MIPIA_GEN_FIFO_STAT		(dev_priv->mipi_mmio_base + 0xb074)
-- 
cgit v0.10.2


From b51a284278c634c675fa22e26e9e9b97dec4518a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 18 Sep 2015 20:03:41 +0300
Subject: drm/i915: Turn GEN5_ASSERT_IIR_IS_ZERO() into a function
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 637c132..e6b406e 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -139,27 +139,30 @@ static const u32 hpd_bxt[HPD_NUM_PINS] = {
 /*
  * We should clear IMR at preinstall/uninstall, and just check at postinstall.
  */
-#define GEN5_ASSERT_IIR_IS_ZERO(reg) do { \
-	u32 val = I915_READ(reg); \
-	if (val) { \
-		WARN(1, "Interrupt register 0x%x is not zero: 0x%08x\n", \
-		     (reg), val); \
-		I915_WRITE((reg), 0xffffffff); \
-		POSTING_READ(reg); \
-		I915_WRITE((reg), 0xffffffff); \
-		POSTING_READ(reg); \
-	} \
-} while (0)
+static void gen5_assert_iir_is_zero(struct drm_i915_private *dev_priv, u32 reg)
+{
+	u32 val = I915_READ(reg);
+
+	if (val == 0)
+		return;
+
+	WARN(1, "Interrupt register 0x%x is not zero: 0x%08x\n",
+	     reg, val);
+	I915_WRITE(reg, 0xffffffff);
+	POSTING_READ(reg);
+	I915_WRITE(reg, 0xffffffff);
+	POSTING_READ(reg);
+}
 
 #define GEN8_IRQ_INIT_NDX(type, which, imr_val, ier_val) do { \
-	GEN5_ASSERT_IIR_IS_ZERO(GEN8_##type##_IIR(which)); \
+	gen5_assert_iir_is_zero(dev_priv, GEN8_##type##_IIR(which)); \
 	I915_WRITE(GEN8_##type##_IER(which), (ier_val)); \
 	I915_WRITE(GEN8_##type##_IMR(which), (imr_val)); \
 	POSTING_READ(GEN8_##type##_IMR(which)); \
 } while (0)
 
 #define GEN5_IRQ_INIT(type, imr_val, ier_val) do { \
-	GEN5_ASSERT_IIR_IS_ZERO(type##IIR); \
+	gen5_assert_iir_is_zero(dev_priv, type##IIR); \
 	I915_WRITE(type##IER, (ier_val)); \
 	I915_WRITE(type##IMR, (imr_val)); \
 	POSTING_READ(type##IMR); \
@@ -3365,7 +3368,7 @@ static void ibx_irq_postinstall(struct drm_device *dev)
 	else
 		mask = SDE_GMBUS_CPT | SDE_AUX_MASK_CPT;
 
-	GEN5_ASSERT_IIR_IS_ZERO(SDEIIR);
+	gen5_assert_iir_is_zero(dev_priv, SDEIIR);
 	I915_WRITE(SDEIMR, ~mask);
 }
 
-- 
cgit v0.10.2


From fd8f507c0de97c3cff39c2e905ad92f15debbdcf Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 18 Sep 2015 20:03:42 +0300
Subject: drm/i915: s/PIPE_FRMCOUNT_GM45/PIPE_FRMCOUNT_G4X/ etc.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The PIPE_FRMCOUNT_GM45 and PIPE_FLIPCOUNT_GM45 names have bothered me
for a long time. The work equally well for ELK and onwards, so let's
s/GM45/G4X/.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index e6b406e..6dd24a1 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -710,10 +710,10 @@ static u32 i915_get_vblank_counter(struct drm_device *dev, int pipe)
 	return (((high1 << 8) | low) + (pixel >= vbl_start)) & 0xffffff;
 }
 
-static u32 gm45_get_vblank_counter(struct drm_device *dev, int pipe)
+static u32 g4x_get_vblank_counter(struct drm_device *dev, int pipe)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	int reg = PIPE_FRMCOUNT_GM45(pipe);
+	int reg = PIPE_FRMCOUNT_G4X(pipe);
 
 	return I915_READ(reg);
 }
@@ -4400,7 +4400,7 @@ void intel_irq_init(struct drm_i915_private *dev_priv)
 		dev->driver->get_vblank_counter = i8xx_get_vblank_counter;
 	} else if (IS_G4X(dev_priv) || INTEL_INFO(dev_priv)->gen >= 5) {
 		dev->max_vblank_count = 0xffffffff; /* full 32 bit counter */
-		dev->driver->get_vblank_counter = gm45_get_vblank_counter;
+		dev->driver->get_vblank_counter = g4x_get_vblank_counter;
 	} else {
 		dev->driver->get_vblank_counter = i915_get_vblank_counter;
 		dev->max_vblank_count = 0xffffff; /* only 24 bits of frame count */
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index f24782b..a838304 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -4846,10 +4846,10 @@ enum skl_disp_power_wells {
 #define   PIPE_PIXEL_MASK         0x00ffffff
 #define   PIPE_PIXEL_SHIFT        0
 /* GM45+ just has to be different */
-#define _PIPEA_FRMCOUNT_GM45	0x70040
-#define _PIPEA_FLIPCOUNT_GM45	0x70044
-#define PIPE_FRMCOUNT_GM45(pipe) _PIPE2(pipe, _PIPEA_FRMCOUNT_GM45)
-#define PIPE_FLIPCOUNT_GM45(pipe) _PIPE2(pipe, _PIPEA_FLIPCOUNT_GM45)
+#define _PIPEA_FRMCOUNT_G4X	0x70040
+#define _PIPEA_FLIPCOUNT_G4X	0x70044
+#define PIPE_FRMCOUNT_G4X(pipe) _PIPE2(pipe, _PIPEA_FRMCOUNT_G4X)
+#define PIPE_FLIPCOUNT_G4X(pipe) _PIPE2(pipe, _PIPEA_FLIPCOUNT_G4X)
 
 /* Cursor A & B regs */
 #define _CURACNTR		0x70080
@@ -5012,8 +5012,8 @@ enum skl_disp_power_wells {
 #define _PIPEBSTAT		(dev_priv->info.display_mmio_offset + 0x71024)
 #define _PIPEBFRAMEHIGH		0x71040
 #define _PIPEBFRAMEPIXEL	0x71044
-#define _PIPEB_FRMCOUNT_GM45	(dev_priv->info.display_mmio_offset + 0x71040)
-#define _PIPEB_FLIPCOUNT_GM45	(dev_priv->info.display_mmio_offset + 0x71044)
+#define _PIPEB_FRMCOUNT_G4X	(dev_priv->info.display_mmio_offset + 0x71040)
+#define _PIPEB_FLIPCOUNT_G4X	(dev_priv->info.display_mmio_offset + 0x71044)
 
 
 /* Display B control */
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 71d7298..b2232ed 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -10822,7 +10822,7 @@ static bool page_flip_finished(struct intel_crtc *crtc)
 	 */
 	return (I915_READ(DSPSURFLIVE(crtc->plane)) & ~0xfff) ==
 		crtc->unpin_work->gtt_offset &&
-		g4x_flip_count_after_eq(I915_READ(PIPE_FLIPCOUNT_GM45(crtc->pipe)),
+		g4x_flip_count_after_eq(I915_READ(PIPE_FLIPCOUNT_G4X(crtc->pipe)),
 				    crtc->unpin_work->flip_count);
 }
 
@@ -11434,7 +11434,7 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	intel_crtc->reset_counter = atomic_read(&dev_priv->gpu_error.reset_counter);
 
 	if (INTEL_INFO(dev)->gen >= 5 || IS_G4X(dev))
-		work->flip_count = I915_READ(PIPE_FLIPCOUNT_GM45(pipe)) + 1;
+		work->flip_count = I915_READ(PIPE_FLIPCOUNT_G4X(pipe)) + 1;
 
 	if (IS_VALLEYVIEW(dev)) {
 		ring = &dev_priv->ring[BCS];
-- 
cgit v0.10.2


From 85fa792beee342598f24cede8f0cc45eb741f469 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 18 Sep 2015 20:03:43 +0300
Subject: drm/i915: Parametrize and fix SWF registers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Parametrize the SWF registers. This also fixes the register offsets,
which were mostly garbage in the old defines.

Also save/restore only as many SWF registers that each platform has.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 604f55d..b6d9b2d 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -1031,7 +1031,7 @@ struct i915_suspend_saved_registers {
 	u32 saveMI_ARB_STATE;
 	u32 saveSWF0[16];
 	u32 saveSWF1[16];
-	u32 saveSWF2[3];
+	u32 saveSWF3[3];
 	uint64_t saveFENCE[I915_MAX_NUM_FENCES];
 	u32 savePCH_PORT_HOTPLUG;
 	u16 saveGCDGMBUS;
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index a838304..bc7b8fa 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -4991,20 +4991,20 @@ enum skl_disp_power_wells {
 #define I915_LO_DISPBASE(val)	(val & ~DISP_BASEADDR_MASK)
 #define I915_HI_DISPBASE(val)	(val & DISP_BASEADDR_MASK)
 
-/* VBIOS flags */
-#define SWF00			(dev_priv->info.display_mmio_offset + 0x71410)
-#define SWF01			(dev_priv->info.display_mmio_offset + 0x71414)
-#define SWF02			(dev_priv->info.display_mmio_offset + 0x71418)
-#define SWF03			(dev_priv->info.display_mmio_offset + 0x7141c)
-#define SWF04			(dev_priv->info.display_mmio_offset + 0x71420)
-#define SWF05			(dev_priv->info.display_mmio_offset + 0x71424)
-#define SWF06			(dev_priv->info.display_mmio_offset + 0x71428)
-#define SWF10			(dev_priv->info.display_mmio_offset + 0x70410)
-#define SWF11			(dev_priv->info.display_mmio_offset + 0x70414)
-#define SWF14			(dev_priv->info.display_mmio_offset + 0x71420)
-#define SWF30			(dev_priv->info.display_mmio_offset + 0x72414)
-#define SWF31			(dev_priv->info.display_mmio_offset + 0x72418)
-#define SWF32			(dev_priv->info.display_mmio_offset + 0x7241c)
+/*
+ * VBIOS flags
+ * gen2:
+ * [00:06] alm,mgm
+ * [10:16] all
+ * [30:32] alm,mgm
+ * gen3+:
+ * [00:0f] all
+ * [10:1f] all
+ * [30:32] all
+ */
+#define SWF0(i)	(dev_priv->info.display_mmio_offset + 0x70410 + (i) * 4)
+#define SWF1(i)	(dev_priv->info.display_mmio_offset + 0x71410 + (i) * 4)
+#define SWF3(i)	(dev_priv->info.display_mmio_offset + 0x72414 + (i) * 4)
 
 /* Pipe B */
 #define _PIPEBDSL		(dev_priv->info.display_mmio_offset + 0x71000)
diff --git a/drivers/gpu/drm/i915/i915_suspend.c b/drivers/gpu/drm/i915/i915_suspend.c
index 1ccac61..2d91821 100644
--- a/drivers/gpu/drm/i915/i915_suspend.c
+++ b/drivers/gpu/drm/i915/i915_suspend.c
@@ -122,12 +122,24 @@ int i915_save_state(struct drm_device *dev)
 	dev_priv->regfile.saveMI_ARB_STATE = I915_READ(MI_ARB_STATE);
 
 	/* Scratch space */
-	for (i = 0; i < 16; i++) {
-		dev_priv->regfile.saveSWF0[i] = I915_READ(SWF00 + (i << 2));
-		dev_priv->regfile.saveSWF1[i] = I915_READ(SWF10 + (i << 2));
+	if (IS_GEN2(dev_priv) && IS_MOBILE(dev_priv)) {
+		for (i = 0; i < 7; i++) {
+			dev_priv->regfile.saveSWF0[i] = I915_READ(SWF0(i));
+			dev_priv->regfile.saveSWF1[i] = I915_READ(SWF1(i));
+		}
+		for (i = 0; i < 3; i++)
+			dev_priv->regfile.saveSWF3[i] = I915_READ(SWF3(i));
+	} else if (IS_GEN2(dev_priv)) {
+		for (i = 0; i < 7; i++)
+			dev_priv->regfile.saveSWF1[i] = I915_READ(SWF1(i));
+	} else if (HAS_GMCH_DISPLAY(dev_priv)) {
+		for (i = 0; i < 16; i++) {
+			dev_priv->regfile.saveSWF0[i] = I915_READ(SWF0(i));
+			dev_priv->regfile.saveSWF1[i] = I915_READ(SWF1(i));
+		}
+		for (i = 0; i < 3; i++)
+			dev_priv->regfile.saveSWF3[i] = I915_READ(SWF3(i));
 	}
-	for (i = 0; i < 3; i++)
-		dev_priv->regfile.saveSWF2[i] = I915_READ(SWF30 + (i << 2));
 
 	mutex_unlock(&dev->struct_mutex);
 
@@ -156,12 +168,25 @@ int i915_restore_state(struct drm_device *dev)
 	/* Memory arbitration state */
 	I915_WRITE(MI_ARB_STATE, dev_priv->regfile.saveMI_ARB_STATE | 0xffff0000);
 
-	for (i = 0; i < 16; i++) {
-		I915_WRITE(SWF00 + (i << 2), dev_priv->regfile.saveSWF0[i]);
-		I915_WRITE(SWF10 + (i << 2), dev_priv->regfile.saveSWF1[i]);
+	/* Scratch space */
+	if (IS_GEN2(dev_priv) && IS_MOBILE(dev_priv)) {
+		for (i = 0; i < 7; i++) {
+			I915_WRITE(SWF0(i), dev_priv->regfile.saveSWF0[i]);
+			I915_WRITE(SWF1(i), dev_priv->regfile.saveSWF1[i]);
+		}
+		for (i = 0; i < 3; i++)
+			I915_WRITE(SWF3(i), dev_priv->regfile.saveSWF3[i]);
+	} else if (IS_GEN2(dev_priv)) {
+		for (i = 0; i < 7; i++)
+			I915_WRITE(SWF1(i), dev_priv->regfile.saveSWF1[i]);
+	} else if (HAS_GMCH_DISPLAY(dev_priv)) {
+		for (i = 0; i < 16; i++) {
+			I915_WRITE(SWF0(i), dev_priv->regfile.saveSWF0[i]);
+			I915_WRITE(SWF1(i), dev_priv->regfile.saveSWF1[i]);
+		}
+		for (i = 0; i < 3; i++)
+			I915_WRITE(SWF3(i), dev_priv->regfile.saveSWF3[i]);
 	}
-	for (i = 0; i < 3; i++)
-		I915_WRITE(SWF30 + (i << 2), dev_priv->regfile.saveSWF2[i]);
 
 	mutex_unlock(&dev->struct_mutex);
 
-- 
cgit v0.10.2


From 649636ef8274cb7bca10fceef2dca1e0813d8cef Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 22 Sep 2015 19:50:01 +0300
Subject: drm/i915: Throw out some useless variables
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Drop some useless 'reg' variables when we only use them once.

v2: A few more, including a few variable moves

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 6dd24a1..e24378ee 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -713,9 +713,8 @@ static u32 i915_get_vblank_counter(struct drm_device *dev, int pipe)
 static u32 g4x_get_vblank_counter(struct drm_device *dev, int pipe)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	int reg = PIPE_FRMCOUNT_G4X(pipe);
 
-	return I915_READ(reg);
+	return I915_READ(PIPE_FRMCOUNT_G4X(pipe));
 }
 
 /* raw reads, only for fast reads of display block, no need for forcewake etc. */
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index b2232ed..7a35492 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -1157,12 +1157,10 @@ static const char *state_string(bool enabled)
 void assert_pll(struct drm_i915_private *dev_priv,
 		enum pipe pipe, bool state)
 {
-	int reg;
 	u32 val;
 	bool cur_state;
 
-	reg = DPLL(pipe);
-	val = I915_READ(reg);
+	val = I915_READ(DPLL(pipe));
 	cur_state = !!(val & DPLL_VCO_ENABLE);
 	I915_STATE_WARN(cur_state != state,
 	     "PLL state assertion failure (expected %s, current %s)\n",
@@ -1219,20 +1217,16 @@ void assert_shared_dpll(struct drm_i915_private *dev_priv,
 static void assert_fdi_tx(struct drm_i915_private *dev_priv,
 			  enum pipe pipe, bool state)
 {
-	int reg;
-	u32 val;
 	bool cur_state;
 	enum transcoder cpu_transcoder = intel_pipe_to_cpu_transcoder(dev_priv,
 								      pipe);
 
 	if (HAS_DDI(dev_priv->dev)) {
 		/* DDI does not have a specific FDI_TX register */
-		reg = TRANS_DDI_FUNC_CTL(cpu_transcoder);
-		val = I915_READ(reg);
+		u32 val = I915_READ(TRANS_DDI_FUNC_CTL(cpu_transcoder));
 		cur_state = !!(val & TRANS_DDI_FUNC_ENABLE);
 	} else {
-		reg = FDI_TX_CTL(pipe);
-		val = I915_READ(reg);
+		u32 val = I915_READ(FDI_TX_CTL(pipe));
 		cur_state = !!(val & FDI_TX_ENABLE);
 	}
 	I915_STATE_WARN(cur_state != state,
@@ -1245,12 +1239,10 @@ static void assert_fdi_tx(struct drm_i915_private *dev_priv,
 static void assert_fdi_rx(struct drm_i915_private *dev_priv,
 			  enum pipe pipe, bool state)
 {
-	int reg;
 	u32 val;
 	bool cur_state;
 
-	reg = FDI_RX_CTL(pipe);
-	val = I915_READ(reg);
+	val = I915_READ(FDI_RX_CTL(pipe));
 	cur_state = !!(val & FDI_RX_ENABLE);
 	I915_STATE_WARN(cur_state != state,
 	     "FDI RX state assertion failure (expected %s, current %s)\n",
@@ -1262,7 +1254,6 @@ static void assert_fdi_rx(struct drm_i915_private *dev_priv,
 static void assert_fdi_tx_pll_enabled(struct drm_i915_private *dev_priv,
 				      enum pipe pipe)
 {
-	int reg;
 	u32 val;
 
 	/* ILK FDI PLL is always enabled */
@@ -1273,20 +1264,17 @@ static void assert_fdi_tx_pll_enabled(struct drm_i915_private *dev_priv,
 	if (HAS_DDI(dev_priv->dev))
 		return;
 
-	reg = FDI_TX_CTL(pipe);
-	val = I915_READ(reg);
+	val = I915_READ(FDI_TX_CTL(pipe));
 	I915_STATE_WARN(!(val & FDI_TX_PLL_ENABLE), "FDI TX PLL assertion failure, should be active but is disabled\n");
 }
 
 void assert_fdi_rx_pll(struct drm_i915_private *dev_priv,
 		       enum pipe pipe, bool state)
 {
-	int reg;
 	u32 val;
 	bool cur_state;
 
-	reg = FDI_RX_CTL(pipe);
-	val = I915_READ(reg);
+	val = I915_READ(FDI_RX_CTL(pipe));
 	cur_state = !!(val & FDI_RX_PLL_ENABLE);
 	I915_STATE_WARN(cur_state != state,
 	     "FDI RX PLL assertion failure (expected %s, current %s)\n",
@@ -1356,8 +1344,6 @@ static void assert_cursor(struct drm_i915_private *dev_priv,
 void assert_pipe(struct drm_i915_private *dev_priv,
 		 enum pipe pipe, bool state)
 {
-	int reg;
-	u32 val;
 	bool cur_state;
 	enum transcoder cpu_transcoder = intel_pipe_to_cpu_transcoder(dev_priv,
 								      pipe);
@@ -1371,8 +1357,7 @@ void assert_pipe(struct drm_i915_private *dev_priv,
 				POWER_DOMAIN_TRANSCODER(cpu_transcoder))) {
 		cur_state = false;
 	} else {
-		reg = PIPECONF(cpu_transcoder);
-		val = I915_READ(reg);
+		u32 val = I915_READ(PIPECONF(cpu_transcoder));
 		cur_state = !!(val & PIPECONF_ENABLE);
 	}
 
@@ -1384,12 +1369,10 @@ void assert_pipe(struct drm_i915_private *dev_priv,
 static void assert_plane(struct drm_i915_private *dev_priv,
 			 enum plane plane, bool state)
 {
-	int reg;
 	u32 val;
 	bool cur_state;
 
-	reg = DSPCNTR(plane);
-	val = I915_READ(reg);
+	val = I915_READ(DSPCNTR(plane));
 	cur_state = !!(val & DISPLAY_PLANE_ENABLE);
 	I915_STATE_WARN(cur_state != state,
 	     "plane %c assertion failure (expected %s, current %s)\n",
@@ -1403,14 +1386,11 @@ static void assert_planes_disabled(struct drm_i915_private *dev_priv,
 				   enum pipe pipe)
 {
 	struct drm_device *dev = dev_priv->dev;
-	int reg, i;
-	u32 val;
-	int cur_pipe;
+	int i;
 
 	/* Primary planes are fixed to pipes on gen4+ */
 	if (INTEL_INFO(dev)->gen >= 4) {
-		reg = DSPCNTR(pipe);
-		val = I915_READ(reg);
+		u32 val = I915_READ(DSPCNTR(pipe));
 		I915_STATE_WARN(val & DISPLAY_PLANE_ENABLE,
 		     "plane %c assertion failure, should be disabled but not\n",
 		     plane_name(pipe));
@@ -1419,9 +1399,8 @@ static void assert_planes_disabled(struct drm_i915_private *dev_priv,
 
 	/* Need to check both planes against the pipe */
 	for_each_pipe(dev_priv, i) {
-		reg = DSPCNTR(i);
-		val = I915_READ(reg);
-		cur_pipe = (val & DISPPLANE_SEL_PIPE_MASK) >>
+		u32 val = I915_READ(DSPCNTR(i));
+		enum pipe cur_pipe = (val & DISPPLANE_SEL_PIPE_MASK) >>
 			DISPPLANE_SEL_PIPE_SHIFT;
 		I915_STATE_WARN((val & DISPLAY_PLANE_ENABLE) && pipe == cur_pipe,
 		     "plane %c assertion failure, should be off on pipe %c but is still active\n",
@@ -1433,33 +1412,29 @@ static void assert_sprites_disabled(struct drm_i915_private *dev_priv,
 				    enum pipe pipe)
 {
 	struct drm_device *dev = dev_priv->dev;
-	int reg, sprite;
-	u32 val;
+	int sprite;
 
 	if (INTEL_INFO(dev)->gen >= 9) {
 		for_each_sprite(dev_priv, pipe, sprite) {
-			val = I915_READ(PLANE_CTL(pipe, sprite));
+			u32 val = I915_READ(PLANE_CTL(pipe, sprite));
 			I915_STATE_WARN(val & PLANE_CTL_ENABLE,
 			     "plane %d assertion failure, should be off on pipe %c but is still active\n",
 			     sprite, pipe_name(pipe));
 		}
 	} else if (IS_VALLEYVIEW(dev)) {
 		for_each_sprite(dev_priv, pipe, sprite) {
-			reg = SPCNTR(pipe, sprite);
-			val = I915_READ(reg);
+			u32 val = I915_READ(SPCNTR(pipe, sprite));
 			I915_STATE_WARN(val & SP_ENABLE,
 			     "sprite %c assertion failure, should be off on pipe %c but is still active\n",
 			     sprite_name(pipe, sprite), pipe_name(pipe));
 		}
 	} else if (INTEL_INFO(dev)->gen >= 7) {
-		reg = SPRCTL(pipe);
-		val = I915_READ(reg);
+		u32 val = I915_READ(SPRCTL(pipe));
 		I915_STATE_WARN(val & SPRITE_ENABLE,
 		     "sprite %c assertion failure, should be off on pipe %c but is still active\n",
 		     plane_name(pipe), pipe_name(pipe));
 	} else if (INTEL_INFO(dev)->gen >= 5) {
-		reg = DVSCNTR(pipe);
-		val = I915_READ(reg);
+		u32 val = I915_READ(DVSCNTR(pipe));
 		I915_STATE_WARN(val & DVS_ENABLE,
 		     "sprite %c assertion failure, should be off on pipe %c but is still active\n",
 		     plane_name(pipe), pipe_name(pipe));
@@ -1488,12 +1463,10 @@ static void ibx_assert_pch_refclk_enabled(struct drm_i915_private *dev_priv)
 static void assert_pch_transcoder_disabled(struct drm_i915_private *dev_priv,
 					   enum pipe pipe)
 {
-	int reg;
 	u32 val;
 	bool enabled;
 
-	reg = PCH_TRANSCONF(pipe);
-	val = I915_READ(reg);
+	val = I915_READ(PCH_TRANSCONF(pipe));
 	enabled = !!(val & TRANS_ENABLE);
 	I915_STATE_WARN(enabled,
 	     "transcoder assertion failed, should be off on pipe %c but is still active\n",
@@ -1600,21 +1573,18 @@ static void assert_pch_hdmi_disabled(struct drm_i915_private *dev_priv,
 static void assert_pch_ports_disabled(struct drm_i915_private *dev_priv,
 				      enum pipe pipe)
 {
-	int reg;
 	u32 val;
 
 	assert_pch_dp_disabled(dev_priv, pipe, PCH_DP_B, TRANS_DP_PORT_SEL_B);
 	assert_pch_dp_disabled(dev_priv, pipe, PCH_DP_C, TRANS_DP_PORT_SEL_C);
 	assert_pch_dp_disabled(dev_priv, pipe, PCH_DP_D, TRANS_DP_PORT_SEL_D);
 
-	reg = PCH_ADPA;
-	val = I915_READ(reg);
+	val = I915_READ(PCH_ADPA);
 	I915_STATE_WARN(adpa_pipe_enabled(dev_priv, pipe, val),
 	     "PCH VGA enabled on transcoder %c, should be disabled\n",
 	     pipe_name(pipe));
 
-	reg = PCH_LVDS;
-	val = I915_READ(reg);
+	val = I915_READ(PCH_LVDS);
 	I915_STATE_WARN(lvds_pipe_enabled(dev_priv, pipe, val),
 	     "PCH LVDS enabled on transcoder %c, should be disabled\n",
 	     pipe_name(pipe));
@@ -14949,13 +14919,12 @@ intel_check_plane_mapping(struct intel_crtc *crtc)
 {
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 reg, val;
+	u32 val;
 
 	if (INTEL_INFO(dev)->num_pipes == 1)
 		return true;
 
-	reg = DSPCNTR(!crtc->plane);
-	val = I915_READ(reg);
+	val = I915_READ(DSPCNTR(!crtc->plane));
 
 	if ((val & DISPLAY_PLANE_ENABLE) &&
 	    (!!(val & DISPPLANE_SEL_PIPE_MASK) == crtc->pipe))
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 18bcfbe..09bdd94 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -574,8 +574,6 @@ static int edp_notify_handler(struct notifier_block *this, unsigned long code,
 						 edp_notifier);
 	struct drm_device *dev = intel_dp_to_dev(intel_dp);
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 pp_div;
-	u32 pp_ctrl_reg, pp_div_reg;
 
 	if (!is_edp(intel_dp) || code != SYS_RESTART)
 		return 0;
@@ -584,6 +582,8 @@ static int edp_notify_handler(struct notifier_block *this, unsigned long code,
 
 	if (IS_VALLEYVIEW(dev)) {
 		enum pipe pipe = vlv_power_sequencer_pipe(intel_dp);
+		u32 pp_ctrl_reg, pp_div_reg;
+		u32 pp_div;
 
 		pp_ctrl_reg = VLV_PIPE_PP_CONTROL(pipe);
 		pp_div_reg  = VLV_PIPE_PP_DIVISOR(pipe);
@@ -5536,7 +5536,6 @@ static void intel_dp_set_drrs_state(struct drm_device *dev, int refresh_rate)
 	struct intel_dp *intel_dp = dev_priv->drrs.dp;
 	struct intel_crtc_state *config = NULL;
 	struct intel_crtc *intel_crtc = NULL;
-	u32 reg, val;
 	enum drrs_refresh_rate_type index = DRRS_HIGH_RR;
 
 	if (refresh_rate <= 0) {
@@ -5598,9 +5597,10 @@ static void intel_dp_set_drrs_state(struct drm_device *dev, int refresh_rate)
 			DRM_ERROR("Unsupported refreshrate type\n");
 		}
 	} else if (INTEL_INFO(dev)->gen > 6) {
-		reg = PIPECONF(intel_crtc->config->cpu_transcoder);
-		val = I915_READ(reg);
+		u32 reg = PIPECONF(intel_crtc->config->cpu_transcoder);
+		u32 val;
 
+		val = I915_READ(reg);
 		if (index > DRRS_HIGH_RR) {
 			if (IS_VALLEYVIEW(dev))
 				val |= PIPECONF_EDP_RR_MODE_SWITCH_VLV;
-- 
cgit v0.10.2


From d0669d007542d7f4829a6a3db50297a4b940fb54 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 18 Sep 2015 20:03:45 +0300
Subject: drm/i915: Clean up LVDS register handling
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Keep single 'lvds_reg' and 'lvds' variable around in
intel_lvds_init(), and read it just once at the start.

Also intel_lvds_get_config() doesn't need to figure out which reg to use
since it can just consult lvds_encoder->reg.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c
index 5e70acf..7f39b8a 100644
--- a/drivers/gpu/drm/i915/intel_lvds.c
+++ b/drivers/gpu/drm/i915/intel_lvds.c
@@ -98,15 +98,11 @@ static void intel_lvds_get_config(struct intel_encoder *encoder,
 {
 	struct drm_device *dev = encoder->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 lvds_reg, tmp, flags = 0;
+	struct intel_lvds_encoder *lvds_encoder = to_lvds_encoder(&encoder->base);
+	u32 tmp, flags = 0;
 	int dotclock;
 
-	if (HAS_PCH_SPLIT(dev))
-		lvds_reg = PCH_LVDS;
-	else
-		lvds_reg = LVDS;
-
-	tmp = I915_READ(lvds_reg);
+	tmp = I915_READ(lvds_encoder->reg);
 	if (tmp & LVDS_HSYNC_POLARITY)
 		flags |= DRM_MODE_FLAG_NHSYNC;
 	else
@@ -943,6 +939,7 @@ void intel_lvds_init(struct drm_device *dev)
 	struct drm_display_mode *downclock_mode = NULL;
 	struct edid *edid;
 	struct drm_crtc *crtc;
+	u32 lvds_reg;
 	u32 lvds;
 	int pipe;
 	u8 pin;
@@ -965,8 +962,15 @@ void intel_lvds_init(struct drm_device *dev)
 	if (dmi_check_system(intel_no_lvds))
 		return;
 
+	if (HAS_PCH_SPLIT(dev))
+		lvds_reg = PCH_LVDS;
+	else
+		lvds_reg = LVDS;
+
+	lvds = I915_READ(lvds_reg);
+
 	if (HAS_PCH_SPLIT(dev)) {
-		if ((I915_READ(PCH_LVDS) & LVDS_DETECTED) == 0)
+		if ((lvds & LVDS_DETECTED) == 0)
 			return;
 		if (dev_priv->vbt.edp_support) {
 			DRM_DEBUG_KMS("disable LVDS for eDP support\n");
@@ -976,8 +980,7 @@ void intel_lvds_init(struct drm_device *dev)
 
 	pin = GMBUS_PIN_PANEL;
 	if (!lvds_is_present_in_vbt(dev, &pin)) {
-		u32 reg = HAS_PCH_SPLIT(dev) ? PCH_LVDS : LVDS;
-		if ((I915_READ(reg) & LVDS_PORT_EN) == 0) {
+		if ((lvds & LVDS_PORT_EN) == 0) {
 			DRM_DEBUG_KMS("LVDS is not present in VBT\n");
 			return;
 		}
@@ -1054,11 +1057,7 @@ void intel_lvds_init(struct drm_device *dev)
 	connector->interlace_allowed = false;
 	connector->doublescan_allowed = false;
 
-	if (HAS_PCH_SPLIT(dev)) {
-		lvds_encoder->reg = PCH_LVDS;
-	} else {
-		lvds_encoder->reg = LVDS;
-	}
+	lvds_encoder->reg = lvds_reg;
 
 	/* create the scaling mode property */
 	drm_mode_create_scaling_mode_property(dev);
@@ -1139,7 +1138,6 @@ void intel_lvds_init(struct drm_device *dev)
 	if (HAS_PCH_SPLIT(dev))
 		goto failed;
 
-	lvds = I915_READ(LVDS);
 	pipe = (lvds & LVDS_PIPEB_SELECT) ? 1 : 0;
 	crtc = intel_get_crtc_for_pipe(dev, pipe);
 
-- 
cgit v0.10.2


From ded1749371db498233b7e8a73f51a1759494bcd1 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 18 Sep 2015 20:03:46 +0300
Subject: drm/i915: Remove dev_priv argument from NEEDS_FORCE_WAKE
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c
index b43c6d0..1663ea5 100644
--- a/drivers/gpu/drm/i915/intel_uncore.c
+++ b/drivers/gpu/drm/i915/intel_uncore.c
@@ -525,7 +525,7 @@ void assert_forcewakes_inactive(struct drm_i915_private *dev_priv)
 }
 
 /* We give fast paths for the really cool registers */
-#define NEEDS_FORCE_WAKE(dev_priv, reg) \
+#define NEEDS_FORCE_WAKE(reg) \
 	 ((reg) < 0x40000 && (reg) != FORCEWAKE)
 
 #define REG_RANGE(reg, start, end) ((reg) >= (start) && (reg) < (end))
@@ -727,7 +727,7 @@ static u##x \
 gen6_read##x(struct drm_i915_private *dev_priv, off_t reg, bool trace) { \
 	GEN6_READ_HEADER(x); \
 	hsw_unclaimed_reg_debug(dev_priv, reg, true, true); \
-	if (NEEDS_FORCE_WAKE((dev_priv), (reg))) \
+	if (NEEDS_FORCE_WAKE(reg)) \
 		__force_wake_get(dev_priv, FORCEWAKE_RENDER); \
 	val = __raw_i915_read##x(dev_priv, reg); \
 	hsw_unclaimed_reg_debug(dev_priv, reg, true, false); \
@@ -761,7 +761,7 @@ chv_read##x(struct drm_i915_private *dev_priv, off_t reg, bool trace) { \
 	GEN6_READ_FOOTER; \
 }
 
-#define SKL_NEEDS_FORCE_WAKE(dev_priv, reg)	\
+#define SKL_NEEDS_FORCE_WAKE(reg) \
 	 ((reg) < 0x40000 && !FORCEWAKE_GEN9_UNCORE_RANGE_OFFSET(reg))
 
 #define __gen9_read(x) \
@@ -770,9 +770,9 @@ gen9_read##x(struct drm_i915_private *dev_priv, off_t reg, bool trace) { \
 	enum forcewake_domains fw_engine; \
 	GEN6_READ_HEADER(x); \
 	hsw_unclaimed_reg_debug(dev_priv, reg, true, true); \
-	if (!SKL_NEEDS_FORCE_WAKE((dev_priv), (reg)))	\
+	if (!SKL_NEEDS_FORCE_WAKE(reg)) \
 		fw_engine = 0; \
-	else if (FORCEWAKE_GEN9_RENDER_RANGE_OFFSET(reg))	\
+	else if (FORCEWAKE_GEN9_RENDER_RANGE_OFFSET(reg)) \
 		fw_engine = FORCEWAKE_RENDER; \
 	else if (FORCEWAKE_GEN9_MEDIA_RANGE_OFFSET(reg)) \
 		fw_engine = FORCEWAKE_MEDIA; \
@@ -868,7 +868,7 @@ static void \
 gen6_write##x(struct drm_i915_private *dev_priv, off_t reg, u##x val, bool trace) { \
 	u32 __fifo_ret = 0; \
 	GEN6_WRITE_HEADER; \
-	if (NEEDS_FORCE_WAKE((dev_priv), (reg))) { \
+	if (NEEDS_FORCE_WAKE(reg)) { \
 		__fifo_ret = __gen6_gt_wait_for_fifo(dev_priv); \
 	} \
 	__raw_i915_write##x(dev_priv, reg, val); \
@@ -883,7 +883,7 @@ static void \
 hsw_write##x(struct drm_i915_private *dev_priv, off_t reg, u##x val, bool trace) { \
 	u32 __fifo_ret = 0; \
 	GEN6_WRITE_HEADER; \
-	if (NEEDS_FORCE_WAKE((dev_priv), (reg))) { \
+	if (NEEDS_FORCE_WAKE(reg)) { \
 		__fifo_ret = __gen6_gt_wait_for_fifo(dev_priv); \
 	} \
 	hsw_unclaimed_reg_debug(dev_priv, reg, false, true); \
@@ -985,7 +985,7 @@ gen9_write##x(struct drm_i915_private *dev_priv, off_t reg, u##x val, \
 	enum forcewake_domains fw_engine; \
 	GEN6_WRITE_HEADER; \
 	hsw_unclaimed_reg_debug(dev_priv, reg, false, true); \
-	if (!SKL_NEEDS_FORCE_WAKE((dev_priv), (reg)) ||	\
+	if (!SKL_NEEDS_FORCE_WAKE(reg) || \
 	    is_gen9_shadowed(dev_priv, reg)) \
 		fw_engine = 0; \
 	else if (FORCEWAKE_GEN9_RENDER_RANGE_OFFSET(reg)) \
-- 
cgit v0.10.2


From 2791a16ca43302d07ac74cbe7c048e367c4632c4 Mon Sep 17 00:00:00 2001
From: Paulo Zanoni <paulo.r.zanoni@intel.com>
Date: Fri, 9 Oct 2015 18:22:43 -0300
Subject: drm/i915: revert a few more watermark commits

This is a squash of the following commits:

Revert "drm/i915: Drop intel_update_sprite_watermarks"
This reverts commit 47c99438b52d12df50e182583634a4cfede3c920.

Revert "drm/i915/ivb: Move WaCxSRDisabledForSpriteScaling w/a to atomic check"
This reverts commit 7809e5ae35b9d8d0710f0874b2e3f10be144e38b.

Revert "drm/i915/skl: Eliminate usage of pipe_wm_parameters from SKL-style WM (v3)"
This reverts commit 3a05f5e2e78eab7ffe816abb59b6769e331a1957.

With these reverts, SKL finally stops failing every single FBC test
with FIFO underrun error messages. After some brief testing, it also
seems that this commit prevents the machine from completely freezing
when we run igt/kms_fbc_crc (see fd.o #92355).

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=92355
Cc: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index b6d9b2d..dd0429c 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -628,6 +628,10 @@ struct drm_i915_display_funcs {
 			  struct dpll *match_clock,
 			  struct dpll *best_clock);
 	void (*update_wm)(struct drm_crtc *crtc);
+	void (*update_sprite_wm)(struct drm_plane *plane,
+				 struct drm_crtc *crtc,
+				 uint32_t sprite_width, uint32_t sprite_height,
+				 int pixel_size, bool enable, bool scaled);
 	int (*modeset_calc_cdclk)(struct drm_atomic_state *state);
 	void (*modeset_commit_cdclk)(struct drm_atomic_state *state);
 	/* Returns the active state of the crtc, and if the crtc is active,
diff --git a/drivers/gpu/drm/i915/intel_atomic.c b/drivers/gpu/drm/i915/intel_atomic.c
index 05b1203..f1975f2 100644
--- a/drivers/gpu/drm/i915/intel_atomic.c
+++ b/drivers/gpu/drm/i915/intel_atomic.c
@@ -94,7 +94,6 @@ intel_crtc_duplicate_state(struct drm_crtc *crtc)
 	__drm_atomic_helper_crtc_duplicate_state(crtc, &crtc_state->base);
 
 	crtc_state->update_pipe = false;
-	crtc_state->disable_lp_wm = false;
 
 	return &crtc_state->base;
 }
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 7a35492..d37b7a1 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -4774,6 +4774,7 @@ static void intel_post_plane_update(struct intel_crtc *crtc)
 	struct intel_crtc_atomic_commit *atomic = &crtc->atomic;
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct drm_plane *plane;
 
 	if (atomic->wait_vblank)
 		intel_wait_for_vblank(dev, crtc->pipe);
@@ -4792,6 +4793,10 @@ static void intel_post_plane_update(struct intel_crtc *crtc)
 	if (atomic->post_enable_primary)
 		intel_post_enable_primary(&crtc->base);
 
+	drm_for_each_plane_mask(plane, dev, atomic->update_sprite_watermarks)
+		intel_update_sprite_watermarks(plane, &crtc->base,
+					       0, 0, 0, false, false);
+
 	memset(atomic, 0, sizeof(*atomic));
 }
 
@@ -11554,30 +11559,16 @@ retry:
 static bool intel_wm_need_update(struct drm_plane *plane,
 				 struct drm_plane_state *state)
 {
-	struct intel_plane_state *new = to_intel_plane_state(state);
-	struct intel_plane_state *cur = to_intel_plane_state(plane->state);
-
-	/* Update watermarks on tiling or size changes. */
+	/* Update watermarks on tiling changes. */
 	if (!plane->state->fb || !state->fb ||
 	    plane->state->fb->modifier[0] != state->fb->modifier[0] ||
-	    plane->state->rotation != state->rotation ||
-	    drm_rect_width(&new->src) != drm_rect_width(&cur->src) ||
-	    drm_rect_height(&new->src) != drm_rect_height(&cur->src) ||
-	    drm_rect_width(&new->dst) != drm_rect_width(&cur->dst) ||
-	    drm_rect_height(&new->dst) != drm_rect_height(&cur->dst))
+	    plane->state->rotation != state->rotation)
 		return true;
 
-	return false;
-}
-
-static bool needs_scaling(struct intel_plane_state *state)
-{
-	int src_w = drm_rect_width(&state->src) >> 16;
-	int src_h = drm_rect_height(&state->src) >> 16;
-	int dst_w = drm_rect_width(&state->dst);
-	int dst_h = drm_rect_height(&state->dst);
+	if (plane->state->crtc_w != state->crtc_w)
+		return true;
 
-	return (src_w != dst_w || src_h != dst_h);
+	return false;
 }
 
 int intel_plane_atomic_calc_changes(struct drm_crtc_state *crtc_state,
@@ -11595,6 +11586,7 @@ int intel_plane_atomic_calc_changes(struct drm_crtc_state *crtc_state,
 	bool mode_changed = needs_modeset(crtc_state);
 	bool was_crtc_enabled = crtc->state->active;
 	bool is_crtc_enabled = crtc_state->active;
+
 	bool turn_off, turn_on, visible, was_visible;
 	struct drm_framebuffer *fb = plane_state->fb;
 
@@ -11712,23 +11704,11 @@ int intel_plane_atomic_calc_changes(struct drm_crtc_state *crtc_state,
 	case DRM_PLANE_TYPE_CURSOR:
 		break;
 	case DRM_PLANE_TYPE_OVERLAY:
-		/*
-		 * WaCxSRDisabledForSpriteScaling:ivb
-		 *
-		 * cstate->update_wm was already set above, so this flag will
-		 * take effect when we commit and program watermarks.
-		 */
-		if (IS_IVYBRIDGE(dev) &&
-		    needs_scaling(to_intel_plane_state(plane_state)) &&
-		    !needs_scaling(old_plane_state)) {
-			to_intel_crtc_state(crtc_state)->disable_lp_wm = true;
-		} else if (turn_off && !mode_changed) {
+		if (turn_off && !mode_changed) {
 			intel_crtc->atomic.wait_vblank = true;
 			intel_crtc->atomic.update_sprite_watermarks |=
 				1 << i;
 		}
-
-		break;
 	}
 	return 0;
 }
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 91b6b40..0598932 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -468,9 +468,6 @@ struct intel_crtc_state {
 
 	/* w/a for waiting 2 vblanks during crtc enable */
 	enum pipe hsw_workaround_pipe;
-
-	/* IVB sprite scaling w/a (WaCxSRDisabledForSpriteScaling:ivb) */
-	bool disable_lp_wm;
 };
 
 struct vlv_wm_state {
@@ -1399,6 +1396,12 @@ void intel_init_clock_gating(struct drm_device *dev);
 void intel_suspend_hw(struct drm_device *dev);
 int ilk_wm_max_level(const struct drm_device *dev);
 void intel_update_watermarks(struct drm_crtc *crtc);
+void intel_update_sprite_watermarks(struct drm_plane *plane,
+				    struct drm_crtc *crtc,
+				    uint32_t sprite_width,
+				    uint32_t sprite_height,
+				    int pixel_size,
+				    bool enabled, bool scaled);
 void intel_init_pm(struct drm_device *dev);
 void intel_pm_setup(struct drm_device *dev);
 void intel_gpu_ips_init(struct drm_i915_private *dev_priv);
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index d031d74..d13551f 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -1765,6 +1765,13 @@ static uint32_t ilk_wm_fbc(uint32_t pri_val, uint32_t horiz_pixels,
 	return DIV_ROUND_UP(pri_val * 64, horiz_pixels * bytes_per_pixel) + 2;
 }
 
+struct skl_pipe_wm_parameters {
+	bool active;
+	uint32_t pipe_htotal;
+	uint32_t pixel_rate; /* in KHz */
+	struct intel_plane_wm_parameters plane[I915_MAX_PLANES];
+};
+
 struct ilk_wm_maximums {
 	uint16_t pri;
 	uint16_t spr;
@@ -2805,40 +2812,18 @@ static bool ilk_disable_lp_wm(struct drm_device *dev)
 #define SKL_DDB_SIZE		896	/* in blocks */
 #define BXT_DDB_SIZE		512
 
-/*
- * Return the index of a plane in the SKL DDB and wm result arrays.  Primary
- * plane is always in slot 0, cursor is always in slot I915_MAX_PLANES-1, and
- * other universal planes are in indices 1..n.  Note that this may leave unused
- * indices between the top "sprite" plane and the cursor.
- */
-static int
-skl_wm_plane_id(const struct intel_plane *plane)
-{
-	switch (plane->base.type) {
-	case DRM_PLANE_TYPE_PRIMARY:
-		return 0;
-	case DRM_PLANE_TYPE_CURSOR:
-		return PLANE_CURSOR;
-	case DRM_PLANE_TYPE_OVERLAY:
-		return plane->plane + 1;
-	default:
-		MISSING_CASE(plane->base.type);
-		return plane->plane;
-	}
-}
-
 static void
 skl_ddb_get_pipe_allocation_limits(struct drm_device *dev,
-				   const struct intel_crtc_state *cstate,
+				   struct drm_crtc *for_crtc,
 				   const struct intel_wm_config *config,
+				   const struct skl_pipe_wm_parameters *params,
 				   struct skl_ddb_entry *alloc /* out */)
 {
-	struct drm_crtc *for_crtc = cstate->base.crtc;
 	struct drm_crtc *crtc;
 	unsigned int pipe_size, ddb_size;
 	int nth_active_pipe;
 
-	if (!cstate->base.active) {
+	if (!params->active) {
 		alloc->start = 0;
 		alloc->end = 0;
 		return;
@@ -2904,29 +2889,19 @@ void skl_ddb_get_hw_state(struct drm_i915_private *dev_priv,
 }
 
 static unsigned int
-skl_plane_relative_data_rate(const struct intel_crtc_state *cstate,
-			     const struct drm_plane_state *pstate,
-			     int y)
+skl_plane_relative_data_rate(const struct intel_plane_wm_parameters *p, int y)
 {
-	struct intel_crtc *intel_crtc = to_intel_crtc(cstate->base.crtc);
-	struct drm_framebuffer *fb = pstate->fb;
 
 	/* for planar format */
-	if (fb->pixel_format == DRM_FORMAT_NV12) {
+	if (p->y_bytes_per_pixel) {
 		if (y)  /* y-plane data rate */
-			return intel_crtc->config->pipe_src_w *
-				intel_crtc->config->pipe_src_h *
-				drm_format_plane_cpp(fb->pixel_format, 0);
+			return p->horiz_pixels * p->vert_pixels * p->y_bytes_per_pixel;
 		else    /* uv-plane data rate */
-			return (intel_crtc->config->pipe_src_w/2) *
-				(intel_crtc->config->pipe_src_h/2) *
-				drm_format_plane_cpp(fb->pixel_format, 1);
+			return (p->horiz_pixels/2) * (p->vert_pixels/2) * p->bytes_per_pixel;
 	}
 
 	/* for packed formats */
-	return intel_crtc->config->pipe_src_w *
-		intel_crtc->config->pipe_src_h *
-		drm_format_plane_cpp(fb->pixel_format, 0);
+	return p->horiz_pixels * p->vert_pixels * p->bytes_per_pixel;
 }
 
 /*
@@ -2935,51 +2910,46 @@ skl_plane_relative_data_rate(const struct intel_crtc_state *cstate,
  *   3 * 4096 * 8192  * 4 < 2^32
  */
 static unsigned int
-skl_get_total_relative_data_rate(const struct intel_crtc_state *cstate)
+skl_get_total_relative_data_rate(struct intel_crtc *intel_crtc,
+				 const struct skl_pipe_wm_parameters *params)
 {
-	struct intel_crtc *intel_crtc = to_intel_crtc(cstate->base.crtc);
-	struct drm_device *dev = intel_crtc->base.dev;
-	const struct intel_plane *intel_plane;
 	unsigned int total_data_rate = 0;
+	int plane;
 
-	for_each_intel_plane_on_crtc(dev, intel_crtc, intel_plane) {
-		const struct drm_plane_state *pstate = intel_plane->base.state;
+	for (plane = 0; plane < intel_num_planes(intel_crtc); plane++) {
+		const struct intel_plane_wm_parameters *p;
 
-		if (pstate->fb == NULL)
+		p = &params->plane[plane];
+		if (!p->enabled)
 			continue;
 
-		/* packed/uv */
-		total_data_rate += skl_plane_relative_data_rate(cstate,
-								pstate,
-								0);
-
-		if (pstate->fb->pixel_format == DRM_FORMAT_NV12)
-			/* y-plane */
-			total_data_rate += skl_plane_relative_data_rate(cstate,
-									pstate,
-									1);
+		total_data_rate += skl_plane_relative_data_rate(p, 0); /* packed/uv */
+		if (p->y_bytes_per_pixel) {
+			total_data_rate += skl_plane_relative_data_rate(p, 1); /* y-plane */
+		}
 	}
 
 	return total_data_rate;
 }
 
 static void
-skl_allocate_pipe_ddb(struct intel_crtc_state *cstate,
+skl_allocate_pipe_ddb(struct drm_crtc *crtc,
 		      const struct intel_wm_config *config,
+		      const struct skl_pipe_wm_parameters *params,
 		      struct skl_ddb_allocation *ddb /* out */)
 {
-	struct drm_crtc *crtc = cstate->base.crtc;
 	struct drm_device *dev = crtc->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	struct intel_plane *intel_plane;
 	enum pipe pipe = intel_crtc->pipe;
 	struct skl_ddb_entry *alloc = &ddb->pipe[pipe];
 	uint16_t alloc_size, start, cursor_blocks;
 	uint16_t minimum[I915_MAX_PLANES];
 	uint16_t y_minimum[I915_MAX_PLANES];
 	unsigned int total_data_rate;
+	int plane;
 
-	skl_ddb_get_pipe_allocation_limits(dev, cstate, config, alloc);
+	skl_ddb_get_pipe_allocation_limits(dev, crtc, config, params, alloc);
 	alloc_size = skl_ddb_entry_size(alloc);
 	if (alloc_size == 0) {
 		memset(ddb->plane[pipe], 0, sizeof(ddb->plane[pipe]));
@@ -2996,20 +2966,17 @@ skl_allocate_pipe_ddb(struct intel_crtc_state *cstate,
 	alloc->end -= cursor_blocks;
 
 	/* 1. Allocate the mininum required blocks for each active plane */
-	for_each_intel_plane_on_crtc(dev, intel_crtc, intel_plane) {
-		struct drm_plane *plane = &intel_plane->base;
-		struct drm_framebuffer *fb = plane->fb;
-		int id = skl_wm_plane_id(intel_plane);
+	for_each_plane(dev_priv, pipe, plane) {
+		const struct intel_plane_wm_parameters *p;
 
-		if (fb == NULL)
-			continue;
-		if (plane->type == DRM_PLANE_TYPE_CURSOR)
+		p = &params->plane[plane];
+		if (!p->enabled)
 			continue;
 
-		minimum[id] = 8;
-		alloc_size -= minimum[id];
-		y_minimum[id] = (fb->pixel_format == DRM_FORMAT_NV12) ? 8 : 0;
-		alloc_size -= y_minimum[id];
+		minimum[plane] = 8;
+		alloc_size -= minimum[plane];
+		y_minimum[plane] = p->y_bytes_per_pixel ? 8 : 0;
+		alloc_size -= y_minimum[plane];
 	}
 
 	/*
@@ -3018,50 +2985,45 @@ skl_allocate_pipe_ddb(struct intel_crtc_state *cstate,
 	 *
 	 * FIXME: we may not allocate every single block here.
 	 */
-	total_data_rate = skl_get_total_relative_data_rate(cstate);
+	total_data_rate = skl_get_total_relative_data_rate(intel_crtc, params);
 
 	start = alloc->start;
-	for_each_intel_plane_on_crtc(dev, intel_crtc, intel_plane) {
-		struct drm_plane *plane = &intel_plane->base;
-		struct drm_plane_state *pstate = intel_plane->base.state;
+	for (plane = 0; plane < intel_num_planes(intel_crtc); plane++) {
+		const struct intel_plane_wm_parameters *p;
 		unsigned int data_rate, y_data_rate;
 		uint16_t plane_blocks, y_plane_blocks = 0;
-		int id = skl_wm_plane_id(intel_plane);
 
-		if (pstate->fb == NULL)
-			continue;
-		if (plane->type == DRM_PLANE_TYPE_CURSOR)
+		p = &params->plane[plane];
+		if (!p->enabled)
 			continue;
 
-		data_rate = skl_plane_relative_data_rate(cstate, pstate, 0);
+		data_rate = skl_plane_relative_data_rate(p, 0);
 
 		/*
 		 * allocation for (packed formats) or (uv-plane part of planar format):
 		 * promote the expression to 64 bits to avoid overflowing, the
 		 * result is < available as data_rate / total_data_rate < 1
 		 */
-		plane_blocks = minimum[id];
+		plane_blocks = minimum[plane];
 		plane_blocks += div_u64((uint64_t)alloc_size * data_rate,
 					total_data_rate);
 
-		ddb->plane[pipe][id].start = start;
-		ddb->plane[pipe][id].end = start + plane_blocks;
+		ddb->plane[pipe][plane].start = start;
+		ddb->plane[pipe][plane].end = start + plane_blocks;
 
 		start += plane_blocks;
 
 		/*
 		 * allocation for y_plane part of planar format:
 		 */
-		if (pstate->fb->pixel_format == DRM_FORMAT_NV12) {
-			y_data_rate = skl_plane_relative_data_rate(cstate,
-								   pstate,
-								   1);
-			y_plane_blocks = y_minimum[id];
+		if (p->y_bytes_per_pixel) {
+			y_data_rate = skl_plane_relative_data_rate(p, 1);
+			y_plane_blocks = y_minimum[plane];
 			y_plane_blocks += div_u64((uint64_t)alloc_size * y_data_rate,
 						total_data_rate);
 
-			ddb->y_plane[pipe][id].start = start;
-			ddb->y_plane[pipe][id].end = start + y_plane_blocks;
+			ddb->y_plane[pipe][plane].start = start;
+			ddb->y_plane[pipe][plane].end = start + y_plane_blocks;
 
 			start += y_plane_blocks;
 		}
@@ -3148,21 +3110,87 @@ static void skl_compute_wm_global_parameters(struct drm_device *dev,
 					     struct intel_wm_config *config)
 {
 	struct drm_crtc *crtc;
+	struct drm_plane *plane;
 
 	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
 		config->num_pipes_active += to_intel_crtc(crtc)->active;
+
+	/* FIXME: I don't think we need those two global parameters on SKL */
+	list_for_each_entry(plane, &dev->mode_config.plane_list, head) {
+		struct intel_plane *intel_plane = to_intel_plane(plane);
+
+		config->sprites_enabled |= intel_plane->wm.enabled;
+		config->sprites_scaled |= intel_plane->wm.scaled;
+	}
+}
+
+static void skl_compute_wm_pipe_parameters(struct drm_crtc *crtc,
+					   struct skl_pipe_wm_parameters *p)
+{
+	struct drm_device *dev = crtc->dev;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	enum pipe pipe = intel_crtc->pipe;
+	struct drm_plane *plane;
+	struct drm_framebuffer *fb;
+	int i = 1; /* Index for sprite planes start */
+
+	p->active = intel_crtc->active;
+	if (p->active) {
+		p->pipe_htotal = intel_crtc->config->base.adjusted_mode.crtc_htotal;
+		p->pixel_rate = skl_pipe_pixel_rate(intel_crtc->config);
+
+		fb = crtc->primary->state->fb;
+		/* For planar: Bpp is for uv plane, y_Bpp is for y plane */
+		if (fb) {
+			p->plane[0].enabled = true;
+			p->plane[0].bytes_per_pixel = fb->pixel_format == DRM_FORMAT_NV12 ?
+				drm_format_plane_cpp(fb->pixel_format, 1) :
+				drm_format_plane_cpp(fb->pixel_format, 0);
+			p->plane[0].y_bytes_per_pixel = fb->pixel_format == DRM_FORMAT_NV12 ?
+				drm_format_plane_cpp(fb->pixel_format, 0) : 0;
+			p->plane[0].tiling = fb->modifier[0];
+		} else {
+			p->plane[0].enabled = false;
+			p->plane[0].bytes_per_pixel = 0;
+			p->plane[0].y_bytes_per_pixel = 0;
+			p->plane[0].tiling = DRM_FORMAT_MOD_NONE;
+		}
+		p->plane[0].horiz_pixels = intel_crtc->config->pipe_src_w;
+		p->plane[0].vert_pixels = intel_crtc->config->pipe_src_h;
+		p->plane[0].rotation = crtc->primary->state->rotation;
+
+		fb = crtc->cursor->state->fb;
+		p->plane[PLANE_CURSOR].y_bytes_per_pixel = 0;
+		if (fb) {
+			p->plane[PLANE_CURSOR].enabled = true;
+			p->plane[PLANE_CURSOR].bytes_per_pixel = fb->bits_per_pixel / 8;
+			p->plane[PLANE_CURSOR].horiz_pixels = crtc->cursor->state->crtc_w;
+			p->plane[PLANE_CURSOR].vert_pixels = crtc->cursor->state->crtc_h;
+		} else {
+			p->plane[PLANE_CURSOR].enabled = false;
+			p->plane[PLANE_CURSOR].bytes_per_pixel = 0;
+			p->plane[PLANE_CURSOR].horiz_pixels = 64;
+			p->plane[PLANE_CURSOR].vert_pixels = 64;
+		}
+	}
+
+	list_for_each_entry(plane, &dev->mode_config.plane_list, head) {
+		struct intel_plane *intel_plane = to_intel_plane(plane);
+
+		if (intel_plane->pipe == pipe &&
+			plane->type == DRM_PLANE_TYPE_OVERLAY)
+			p->plane[i++] = intel_plane->wm;
+	}
 }
 
 static bool skl_compute_plane_wm(const struct drm_i915_private *dev_priv,
-				 struct intel_crtc_state *cstate,
-				 struct intel_plane *intel_plane,
+				 struct skl_pipe_wm_parameters *p,
+				 struct intel_plane_wm_parameters *p_params,
 				 uint16_t ddb_allocation,
 				 int level,
 				 uint16_t *out_blocks, /* out */
 				 uint8_t *out_lines /* out */)
 {
-	struct drm_plane *plane = &intel_plane->base;
-	struct drm_framebuffer *fb = plane->state->fb;
 	uint32_t latency = dev_priv->wm.skl_latency[level];
 	uint32_t method1, method2;
 	uint32_t plane_bytes_per_line, plane_blocks_per_line;
@@ -3170,35 +3198,31 @@ static bool skl_compute_plane_wm(const struct drm_i915_private *dev_priv,
 	uint32_t selected_result;
 	uint8_t bytes_per_pixel;
 
-	if (latency == 0 || !cstate->base.active || !fb)
+	if (latency == 0 || !p->active || !p_params->enabled)
 		return false;
 
-	bytes_per_pixel = (fb->pixel_format == DRM_FORMAT_NV12) ?
-		drm_format_plane_cpp(DRM_FORMAT_NV12, 0) :
-		drm_format_plane_cpp(DRM_FORMAT_NV12, 1);
-	method1 = skl_wm_method1(skl_pipe_pixel_rate(cstate),
+	bytes_per_pixel = p_params->y_bytes_per_pixel ?
+		p_params->y_bytes_per_pixel :
+		p_params->bytes_per_pixel;
+	method1 = skl_wm_method1(p->pixel_rate,
 				 bytes_per_pixel,
 				 latency);
-	method2 = skl_wm_method2(skl_pipe_pixel_rate(cstate),
-				 cstate->base.adjusted_mode.crtc_htotal,
-				 cstate->pipe_src_w,
+	method2 = skl_wm_method2(p->pixel_rate,
+				 p->pipe_htotal,
+				 p_params->horiz_pixels,
 				 bytes_per_pixel,
-				 fb->modifier[0],
+				 p_params->tiling,
 				 latency);
 
-	plane_bytes_per_line = cstate->pipe_src_w * bytes_per_pixel;
+	plane_bytes_per_line = p_params->horiz_pixels * bytes_per_pixel;
 	plane_blocks_per_line = DIV_ROUND_UP(plane_bytes_per_line, 512);
 
-	if (fb->modifier[0] == I915_FORMAT_MOD_Y_TILED ||
-	    fb->modifier[0] == I915_FORMAT_MOD_Yf_TILED) {
+	if (p_params->tiling == I915_FORMAT_MOD_Y_TILED ||
+	    p_params->tiling == I915_FORMAT_MOD_Yf_TILED) {
 		uint32_t min_scanlines = 4;
 		uint32_t y_tile_minimum;
-		if (intel_rotation_90_or_270(plane->state->rotation)) {
-			int bpp = (fb->pixel_format == DRM_FORMAT_NV12) ?
-				drm_format_plane_cpp(fb->pixel_format, 1) :
-				drm_format_plane_cpp(fb->pixel_format, 0);
-
-			switch (bpp) {
+		if (intel_rotation_90_or_270(p_params->rotation)) {
+			switch (p_params->bytes_per_pixel) {
 			case 1:
 				min_scanlines = 16;
 				break;
@@ -3222,8 +3246,8 @@ static bool skl_compute_plane_wm(const struct drm_i915_private *dev_priv,
 	res_lines = DIV_ROUND_UP(selected_result, plane_blocks_per_line);
 
 	if (level >= 1 && level <= 7) {
-		if (fb->modifier[0] == I915_FORMAT_MOD_Y_TILED ||
-		    fb->modifier[0] == I915_FORMAT_MOD_Yf_TILED)
+		if (p_params->tiling == I915_FORMAT_MOD_Y_TILED ||
+		    p_params->tiling == I915_FORMAT_MOD_Yf_TILED)
 			res_lines += 4;
 		else
 			res_blocks++;
@@ -3240,80 +3264,84 @@ static bool skl_compute_plane_wm(const struct drm_i915_private *dev_priv,
 
 static void skl_compute_wm_level(const struct drm_i915_private *dev_priv,
 				 struct skl_ddb_allocation *ddb,
-				 struct intel_crtc_state *cstate,
+				 struct skl_pipe_wm_parameters *p,
+				 enum pipe pipe,
 				 int level,
+				 int num_planes,
 				 struct skl_wm_level *result)
 {
-	struct drm_device *dev = dev_priv->dev;
-	struct intel_crtc *intel_crtc = to_intel_crtc(cstate->base.crtc);
-	struct intel_plane *intel_plane;
 	uint16_t ddb_blocks;
-	enum pipe pipe = intel_crtc->pipe;
-
-	for_each_intel_plane_on_crtc(dev, intel_crtc, intel_plane) {
-		int i = skl_wm_plane_id(intel_plane);
+	int i;
 
+	for (i = 0; i < num_planes; i++) {
 		ddb_blocks = skl_ddb_entry_size(&ddb->plane[pipe][i]);
 
 		result->plane_en[i] = skl_compute_plane_wm(dev_priv,
-						cstate,
-						intel_plane,
+						p, &p->plane[i],
 						ddb_blocks,
 						level,
 						&result->plane_res_b[i],
 						&result->plane_res_l[i]);
 	}
+
+	ddb_blocks = skl_ddb_entry_size(&ddb->plane[pipe][PLANE_CURSOR]);
+	result->plane_en[PLANE_CURSOR] = skl_compute_plane_wm(dev_priv, p,
+						 &p->plane[PLANE_CURSOR],
+						 ddb_blocks, level,
+						 &result->plane_res_b[PLANE_CURSOR],
+						 &result->plane_res_l[PLANE_CURSOR]);
 }
 
 static uint32_t
-skl_compute_linetime_wm(struct intel_crtc_state *cstate)
+skl_compute_linetime_wm(struct drm_crtc *crtc, struct skl_pipe_wm_parameters *p)
 {
-	if (!cstate->base.active)
+	if (!to_intel_crtc(crtc)->active)
 		return 0;
 
-	if (WARN_ON(skl_pipe_pixel_rate(cstate) == 0))
+	if (WARN_ON(p->pixel_rate == 0))
 		return 0;
 
-	return DIV_ROUND_UP(8 * cstate->base.adjusted_mode.crtc_htotal * 1000,
-			    skl_pipe_pixel_rate(cstate));
+	return DIV_ROUND_UP(8 * p->pipe_htotal * 1000, p->pixel_rate);
 }
 
-static void skl_compute_transition_wm(struct intel_crtc_state *cstate,
+static void skl_compute_transition_wm(struct drm_crtc *crtc,
+				      struct skl_pipe_wm_parameters *params,
 				      struct skl_wm_level *trans_wm /* out */)
 {
-	struct drm_crtc *crtc = cstate->base.crtc;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	struct intel_plane *intel_plane;
+	int i;
 
-	if (!cstate->base.active)
+	if (!params->active)
 		return;
 
 	/* Until we know more, just disable transition WMs */
-	for_each_intel_plane_on_crtc(crtc->dev, intel_crtc, intel_plane) {
-		int i = skl_wm_plane_id(intel_plane);
-
+	for (i = 0; i < intel_num_planes(intel_crtc); i++)
 		trans_wm->plane_en[i] = false;
-	}
+	trans_wm->plane_en[PLANE_CURSOR] = false;
 }
 
-static void skl_compute_pipe_wm(struct intel_crtc_state *cstate,
+static void skl_compute_pipe_wm(struct drm_crtc *crtc,
 				struct skl_ddb_allocation *ddb,
+				struct skl_pipe_wm_parameters *params,
 				struct skl_pipe_wm *pipe_wm)
 {
-	struct drm_device *dev = cstate->base.crtc->dev;
+	struct drm_device *dev = crtc->dev;
 	const struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	int level, max_level = ilk_wm_max_level(dev);
 
 	for (level = 0; level <= max_level; level++) {
-		skl_compute_wm_level(dev_priv, ddb, cstate,
-				     level, &pipe_wm->wm[level]);
+		skl_compute_wm_level(dev_priv, ddb, params, intel_crtc->pipe,
+				     level, intel_num_planes(intel_crtc),
+				     &pipe_wm->wm[level]);
 	}
-	pipe_wm->linetime = skl_compute_linetime_wm(cstate);
+	pipe_wm->linetime = skl_compute_linetime_wm(crtc, params);
 
-	skl_compute_transition_wm(cstate, &pipe_wm->trans_wm);
+	skl_compute_transition_wm(crtc, params, &pipe_wm->trans_wm);
 }
 
 static void skl_compute_wm_results(struct drm_device *dev,
+				   struct skl_pipe_wm_parameters *p,
 				   struct skl_pipe_wm *p_wm,
 				   struct skl_wm_values *r,
 				   struct intel_crtc *intel_crtc)
@@ -3557,15 +3585,16 @@ static void skl_flush_wm_values(struct drm_i915_private *dev_priv,
 }
 
 static bool skl_update_pipe_wm(struct drm_crtc *crtc,
+			       struct skl_pipe_wm_parameters *params,
 			       struct intel_wm_config *config,
 			       struct skl_ddb_allocation *ddb, /* out */
 			       struct skl_pipe_wm *pipe_wm /* out */)
 {
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	struct intel_crtc_state *cstate = to_intel_crtc_state(crtc->state);
 
-	skl_allocate_pipe_ddb(cstate, config, ddb);
-	skl_compute_pipe_wm(cstate, ddb, pipe_wm);
+	skl_compute_wm_pipe_parameters(crtc, params);
+	skl_allocate_pipe_ddb(crtc, config, params, ddb);
+	skl_compute_pipe_wm(crtc, ddb, params, pipe_wm);
 
 	if (!memcmp(&intel_crtc->wm.skl_active, pipe_wm, sizeof(*pipe_wm)))
 		return false;
@@ -3598,6 +3627,7 @@ static void skl_update_other_pipe_wm(struct drm_device *dev,
 	 */
 	list_for_each_entry(intel_crtc, &dev->mode_config.crtc_list,
 				base.head) {
+		struct skl_pipe_wm_parameters params = {};
 		struct skl_pipe_wm pipe_wm = {};
 		bool wm_changed;
 
@@ -3607,7 +3637,8 @@ static void skl_update_other_pipe_wm(struct drm_device *dev,
 		if (!intel_crtc->active)
 			continue;
 
-		wm_changed = skl_update_pipe_wm(&intel_crtc->base, config,
+		wm_changed = skl_update_pipe_wm(&intel_crtc->base,
+						&params, config,
 						&r->ddb, &pipe_wm);
 
 		/*
@@ -3617,7 +3648,7 @@ static void skl_update_other_pipe_wm(struct drm_device *dev,
 		 */
 		WARN_ON(!wm_changed);
 
-		skl_compute_wm_results(dev, &pipe_wm, r, intel_crtc);
+		skl_compute_wm_results(dev, &params, &pipe_wm, r, intel_crtc);
 		r->dirty[intel_crtc->pipe] = true;
 	}
 }
@@ -3647,6 +3678,7 @@ static void skl_update_wm(struct drm_crtc *crtc)
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct skl_pipe_wm_parameters params = {};
 	struct skl_wm_values *results = &dev_priv->wm.skl_results;
 	struct skl_pipe_wm pipe_wm = {};
 	struct intel_wm_config config = {};
@@ -3659,10 +3691,11 @@ static void skl_update_wm(struct drm_crtc *crtc)
 
 	skl_compute_wm_global_parameters(dev, &config);
 
-	if (!skl_update_pipe_wm(crtc, &config, &results->ddb, &pipe_wm))
+	if (!skl_update_pipe_wm(crtc, &params, &config,
+				&results->ddb, &pipe_wm))
 		return;
 
-	skl_compute_wm_results(dev, &pipe_wm, results, intel_crtc);
+	skl_compute_wm_results(dev, &params, &pipe_wm, results, intel_crtc);
 	results->dirty[intel_crtc->pipe] = true;
 
 	skl_update_other_pipe_wm(dev, crtc, &config, results);
@@ -3673,6 +3706,39 @@ static void skl_update_wm(struct drm_crtc *crtc)
 	dev_priv->wm.skl_hw = *results;
 }
 
+static void
+skl_update_sprite_wm(struct drm_plane *plane, struct drm_crtc *crtc,
+		     uint32_t sprite_width, uint32_t sprite_height,
+		     int pixel_size, bool enabled, bool scaled)
+{
+	struct intel_plane *intel_plane = to_intel_plane(plane);
+	struct drm_framebuffer *fb = plane->state->fb;
+
+	intel_plane->wm.enabled = enabled;
+	intel_plane->wm.scaled = scaled;
+	intel_plane->wm.horiz_pixels = sprite_width;
+	intel_plane->wm.vert_pixels = sprite_height;
+	intel_plane->wm.tiling = DRM_FORMAT_MOD_NONE;
+
+	/* For planar: Bpp is for UV plane, y_Bpp is for Y plane */
+	intel_plane->wm.bytes_per_pixel =
+		(fb && fb->pixel_format == DRM_FORMAT_NV12) ?
+		drm_format_plane_cpp(plane->state->fb->pixel_format, 1) : pixel_size;
+	intel_plane->wm.y_bytes_per_pixel =
+		(fb && fb->pixel_format == DRM_FORMAT_NV12) ?
+		drm_format_plane_cpp(plane->state->fb->pixel_format, 0) : 0;
+
+	/*
+	 * Framebuffer can be NULL on plane disable, but it does not
+	 * matter for watermarks if we assume no tiling in that case.
+	 */
+	if (fb)
+		intel_plane->wm.tiling = fb->modifier[0];
+	intel_plane->wm.rotation = plane->state->rotation;
+
+	skl_update_wm(crtc);
+}
+
 static void ilk_update_wm(struct drm_crtc *crtc)
 {
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
@@ -3688,18 +3754,6 @@ static void ilk_update_wm(struct drm_crtc *crtc)
 
 	WARN_ON(cstate->base.active != intel_crtc->active);
 
-	/*
-	 * IVB workaround: must disable low power watermarks for at least
-	 * one frame before enabling scaling.  LP watermarks can be re-enabled
-	 * when scaling is disabled.
-	 *
-	 * WaCxSRDisabledForSpriteScaling:ivb
-	 */
-	if (cstate->disable_lp_wm) {
-		ilk_disable_lp_wm(dev);
-		intel_wait_for_vblank(dev, intel_crtc->pipe);
-	}
-
 	intel_compute_pipe_wm(cstate, &pipe_wm);
 
 	if (!memcmp(&intel_crtc->wm.active, &pipe_wm, sizeof(pipe_wm)))
@@ -3731,6 +3785,28 @@ static void ilk_update_wm(struct drm_crtc *crtc)
 	ilk_write_wm_values(dev_priv, &results);
 }
 
+static void
+ilk_update_sprite_wm(struct drm_plane *plane,
+		     struct drm_crtc *crtc,
+		     uint32_t sprite_width, uint32_t sprite_height,
+		     int pixel_size, bool enabled, bool scaled)
+{
+	struct drm_device *dev = plane->dev;
+	struct intel_plane *intel_plane = to_intel_plane(plane);
+
+	/*
+	 * IVB workaround: must disable low power watermarks for at least
+	 * one frame before enabling scaling.  LP watermarks can be re-enabled
+	 * when scaling is disabled.
+	 *
+	 * WaCxSRDisabledForSpriteScaling:ivb
+	 */
+	if (IS_IVYBRIDGE(dev) && scaled && ilk_disable_lp_wm(dev))
+		intel_wait_for_vblank(dev, intel_plane->pipe);
+
+	ilk_update_wm(crtc);
+}
+
 static void skl_pipe_wm_active_state(uint32_t val,
 				     struct skl_pipe_wm *active,
 				     bool is_transwm,
@@ -4108,6 +4184,21 @@ void intel_update_watermarks(struct drm_crtc *crtc)
 		dev_priv->display.update_wm(crtc);
 }
 
+void intel_update_sprite_watermarks(struct drm_plane *plane,
+				    struct drm_crtc *crtc,
+				    uint32_t sprite_width,
+				    uint32_t sprite_height,
+				    int pixel_size,
+				    bool enabled, bool scaled)
+{
+	struct drm_i915_private *dev_priv = plane->dev->dev_private;
+
+	if (dev_priv->display.update_sprite_wm)
+		dev_priv->display.update_sprite_wm(plane, crtc,
+						   sprite_width, sprite_height,
+						   pixel_size, enabled, scaled);
+}
+
 /**
  * Lock protecting IPS related data structures
  */
@@ -7022,6 +7113,7 @@ void intel_init_pm(struct drm_device *dev)
 			dev_priv->display.init_clock_gating =
 				skl_init_clock_gating;
 		dev_priv->display.update_wm = skl_update_wm;
+		dev_priv->display.update_sprite_wm = skl_update_sprite_wm;
 	} else if (HAS_PCH_SPLIT(dev)) {
 		ilk_setup_wm_latency(dev);
 
@@ -7030,6 +7122,7 @@ void intel_init_pm(struct drm_device *dev)
 		    (!IS_GEN5(dev) && dev_priv->wm.pri_latency[0] &&
 		     dev_priv->wm.spr_latency[0] && dev_priv->wm.cur_latency[0])) {
 			dev_priv->display.update_wm = ilk_update_wm;
+			dev_priv->display.update_sprite_wm = ilk_update_sprite_wm;
 		} else {
 			DRM_DEBUG_KMS("Failed to read display plane latency. "
 				      "Disable CxSR\n");
diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c
index dd2d568..b229c67 100644
--- a/drivers/gpu/drm/i915/intel_sprite.c
+++ b/drivers/gpu/drm/i915/intel_sprite.c
@@ -192,6 +192,7 @@ skl_update_plane(struct drm_plane *drm_plane, struct drm_crtc *crtc,
 	const int pipe = intel_plane->pipe;
 	const int plane = intel_plane->plane + 1;
 	u32 plane_ctl, stride_div, stride;
+	int pixel_size = drm_format_plane_cpp(fb->pixel_format, 0);
 	const struct drm_intel_sprite_colorkey *key =
 		&to_intel_plane_state(drm_plane->state)->ckey;
 	unsigned long surf_addr;
@@ -210,6 +211,10 @@ skl_update_plane(struct drm_plane *drm_plane, struct drm_crtc *crtc,
 	rotation = drm_plane->state->rotation;
 	plane_ctl |= skl_plane_ctl_rotation(rotation);
 
+	intel_update_sprite_watermarks(drm_plane, crtc, src_w, src_h,
+				       pixel_size, true,
+				       src_w != crtc_w || src_h != crtc_h);
+
 	stride_div = intel_fb_stride_alignment(dev, fb->modifier[0],
 					       fb->pixel_format);
 
@@ -291,6 +296,8 @@ skl_disable_plane(struct drm_plane *dplane, struct drm_crtc *crtc)
 
 	I915_WRITE(PLANE_SURF(pipe, plane), 0);
 	POSTING_READ(PLANE_SURF(pipe, plane));
+
+	intel_update_sprite_watermarks(dplane, crtc, 0, 0, 0, false, false);
 }
 
 static void
@@ -533,6 +540,10 @@ ivb_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 	if (IS_HASWELL(dev) || IS_BROADWELL(dev))
 		sprctl |= SPRITE_PIPE_CSC_ENABLE;
 
+	intel_update_sprite_watermarks(plane, crtc, src_w, src_h, pixel_size,
+				       true,
+				       src_w != crtc_w || src_h != crtc_h);
+
 	/* Sizes are 0 based */
 	src_w--;
 	src_h--;
@@ -666,6 +677,10 @@ ilk_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 	if (IS_GEN6(dev))
 		dvscntr |= DVS_TRICKLE_FEED_DISABLE; /* must disable */
 
+	intel_update_sprite_watermarks(plane, crtc, src_w, src_h,
+				       pixel_size, true,
+				       src_w != crtc_w || src_h != crtc_h);
+
 	/* Sizes are 0 based */
 	src_w--;
 	src_h--;
-- 
cgit v0.10.2


From ef55f92a92eee54238e16269823a52cfcbb2330c Mon Sep 17 00:00:00 2001
From: Chris Wilson <chris@chris-wilson.co.uk>
Date: Fri, 9 Oct 2015 14:11:27 +0100
Subject: drm/i915: Drop i915_gem_obj_is_pinned() from set-cache-level

Since the remove of the pin-ioctl, we only care about not changing the
cache level on buffers pinned to the hardware as indicated by
obj->pin_display. By knowing that only objects pinned to the hardware
will have an elevated vma->pin_count, so we can coallesce many of the
linear walks over the obj->vma_list.

v2: Try and retrospectively add comments explaining the steps in
rebinding the active VMA.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 1e67484..e57061a 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -3657,53 +3657,106 @@ i915_gem_object_set_to_gtt_domain(struct drm_i915_gem_object *obj, bool write)
 	return 0;
 }
 
+/**
+ * Changes the cache-level of an object across all VMA.
+ *
+ * After this function returns, the object will be in the new cache-level
+ * across all GTT and the contents of the backing storage will be coherent,
+ * with respect to the new cache-level. In order to keep the backing storage
+ * coherent for all users, we only allow a single cache level to be set
+ * globally on the object and prevent it from being changed whilst the
+ * hardware is reading from the object. That is if the object is currently
+ * on the scanout it will be set to uncached (or equivalent display
+ * cache coherency) and all non-MOCS GPU access will also be uncached so
+ * that all direct access to the scanout remains coherent.
+ */
 int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,
 				    enum i915_cache_level cache_level)
 {
 	struct drm_device *dev = obj->base.dev;
 	struct i915_vma *vma, *next;
+	bool bound = false;
 	int ret = 0;
 
 	if (obj->cache_level == cache_level)
 		goto out;
 
-	if (i915_gem_obj_is_pinned(obj)) {
-		DRM_DEBUG("can not change the cache level of pinned objects\n");
-		return -EBUSY;
-	}
-
+	/* Inspect the list of currently bound VMA and unbind any that would
+	 * be invalid given the new cache-level. This is principally to
+	 * catch the issue of the CS prefetch crossing page boundaries and
+	 * reading an invalid PTE on older architectures.
+	 */
 	list_for_each_entry_safe(vma, next, &obj->vma_list, vma_link) {
+		if (!drm_mm_node_allocated(&vma->node))
+			continue;
+
+		if (vma->pin_count) {
+			DRM_DEBUG("can not change the cache level of pinned objects\n");
+			return -EBUSY;
+		}
+
 		if (!i915_gem_valid_gtt_space(vma, cache_level)) {
 			ret = i915_vma_unbind(vma);
 			if (ret)
 				return ret;
-		}
+		} else
+			bound = true;
 	}
 
-	if (i915_gem_obj_bound_any(obj)) {
+	/* We can reuse the existing drm_mm nodes but need to change the
+	 * cache-level on the PTE. We could simply unbind them all and
+	 * rebind with the correct cache-level on next use. However since
+	 * we already have a valid slot, dma mapping, pages etc, we may as
+	 * rewrite the PTE in the belief that doing so tramples upon less
+	 * state and so involves less work.
+	 */
+	if (bound) {
+		/* Before we change the PTE, the GPU must not be accessing it.
+		 * If we wait upon the object, we know that all the bound
+		 * VMA are no longer active.
+		 */
 		ret = i915_gem_object_wait_rendering(obj, false);
 		if (ret)
 			return ret;
 
-		i915_gem_object_finish_gtt(obj);
-
-		/* Before SandyBridge, you could not use tiling or fence
-		 * registers with snooped memory, so relinquish any fences
-		 * currently pointing to our region in the aperture.
-		 */
-		if (INTEL_INFO(dev)->gen < 6) {
+		if (!HAS_LLC(dev) && cache_level != I915_CACHE_NONE) {
+			/* Access to snoopable pages through the GTT is
+			 * incoherent and on some machines causes a hard
+			 * lockup. Relinquish the CPU mmaping to force
+			 * userspace to refault in the pages and we can
+			 * then double check if the GTT mapping is still
+			 * valid for that pointer access.
+			 */
+			i915_gem_release_mmap(obj);
+
+			/* As we no longer need a fence for GTT access,
+			 * we can relinquish it now (and so prevent having
+			 * to steal a fence from someone else on the next
+			 * fence request). Note GPU activity would have
+			 * dropped the fence as all snoopable access is
+			 * supposed to be linear.
+			 */
 			ret = i915_gem_object_put_fence(obj);
 			if (ret)
 				return ret;
+		} else {
+			/* We either have incoherent backing store and
+			 * so no GTT access or the architecture is fully
+			 * coherent. In such cases, existing GTT mmaps
+			 * ignore the cache bit in the PTE and we can
+			 * rewrite it without confusing the GPU or having
+			 * to force userspace to fault back in its mmaps.
+			 */
 		}
 
-		list_for_each_entry(vma, &obj->vma_list, vma_link)
-			if (drm_mm_node_allocated(&vma->node)) {
-				ret = i915_vma_bind(vma, cache_level,
-						    PIN_UPDATE);
-				if (ret)
-					return ret;
-			}
+		list_for_each_entry(vma, &obj->vma_list, vma_link) {
+			if (!drm_mm_node_allocated(&vma->node))
+				continue;
+
+			ret = i915_vma_bind(vma, cache_level, PIN_UPDATE);
+			if (ret)
+				return ret;
+		}
 	}
 
 	list_for_each_entry(vma, &obj->vma_list, vma_link)
@@ -3711,6 +3764,10 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,
 	obj->cache_level = cache_level;
 
 out:
+	/* Flush the dirty CPU caches to the backing storage so that the
+	 * object is now coherent at its new cache level (with respect
+	 * to the access domain).
+	 */
 	if (obj->cache_dirty &&
 	    obj->base.write_domain != I915_GEM_DOMAIN_CPU &&
 	    cpu_write_needs_clflush(obj)) {
-- 
cgit v0.10.2


From 9c4cbf8212e8d8be4bc6e16cc2c21af2bbaab2c1 Mon Sep 17 00:00:00 2001
From: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Date: Mon, 12 Oct 2015 13:20:59 +0300
Subject: drm/i915: Move skl/bxt gt specific workarounds to ring init
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Some registers are, naturally, lost in gpu reset/suspend cycle.
And some registers, for example in display domain, are not subject
to gpu reset so they retain their contents.

As hang recovery triggers a reset, recoverable gpu hang can currently
flush out essential workarounds and cause havoc later on.

When register GEN8_GARBNTL is missing the WaEnableGapsTsvCreditFix:skl,
it can cause random system hangs [1]. This workaround was added in:
commit 245d96670d26 ("drm/i915:skl: Add WaEnableGapsTsvCreditFix")
But another set of system hangs were observed and the failure pattern
indicated that there was random gpu hang preceding the system hang [2].
This lead to the realization that we lose this workaround and BDW_SCRATCH1
on reset.

Add these workarounds setup in display init to skl/bxt ring init
where LRI workarounds are also setup. This way their setup is not
dependent on display side init.

References: [1] https://bugs.freedesktop.org/show_bug.cgi?id=90854
References: [2] https://bugs.freedesktop.org/show_bug.cgi?id=92315
Reported-by: Tomi Sarvela <tomix.p.sarvela@intel.com>
Cc: Tomi Sarvela <tomix.p.sarvela@intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Tested-by: Tomi Sarvela <tomix.p.sarvela@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index d13551f..9dda3ea 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -52,56 +52,10 @@
 #define INTEL_RC6p_ENABLE			(1<<1)
 #define INTEL_RC6pp_ENABLE			(1<<2)
 
-static void gen9_init_clock_gating(struct drm_device *dev)
-{
-	struct drm_i915_private *dev_priv = dev->dev_private;
-
-	/* WaEnableLbsSlaRetryTimerDecrement:skl */
-	I915_WRITE(BDW_SCRATCH1, I915_READ(BDW_SCRATCH1) |
-		   GEN9_LBS_SLA_RETRY_TIMER_DECREMENT_ENABLE);
-
-	/* WaDisableKillLogic:bxt,skl */
-	I915_WRITE(GAM_ECOCHK, I915_READ(GAM_ECOCHK) |
-		   ECOCHK_DIS_TLB);
-}
-
-static void skl_init_clock_gating(struct drm_device *dev)
-{
-	struct drm_i915_private *dev_priv = dev->dev_private;
-
-	gen9_init_clock_gating(dev);
-
-	if (INTEL_REVID(dev) <= SKL_REVID_D0) {
-		/* WaDisableHDCInvalidation:skl */
-		I915_WRITE(GAM_ECOCHK, I915_READ(GAM_ECOCHK) |
-			   BDW_DISABLE_HDC_INVALIDATION);
-
-		/* WaDisableChickenBitTSGBarrierAckForFFSliceCS:skl */
-		I915_WRITE(FF_SLICE_CS_CHICKEN2,
-			   _MASKED_BIT_ENABLE(GEN9_TSG_BARRIER_ACK_DISABLE));
-	}
-
-	/* GEN8_L3SQCREG4 has a dependency with WA batch so any new changes
-	 * involving this register should also be added to WA batch as required.
-	 */
-	if (INTEL_REVID(dev) <= SKL_REVID_E0)
-		/* WaDisableLSQCROPERFforOCL:skl */
-		I915_WRITE(GEN8_L3SQCREG4, I915_READ(GEN8_L3SQCREG4) |
-			   GEN8_LQSC_RO_PERF_DIS);
-
-	/* WaEnableGapsTsvCreditFix:skl */
-	if (IS_SKYLAKE(dev) && (INTEL_REVID(dev) >= SKL_REVID_C0)) {
-		I915_WRITE(GEN8_GARBCNTL, (I915_READ(GEN8_GARBCNTL) |
-					   GEN9_GAPS_TSV_CREDIT_DISABLE));
-	}
-}
-
 static void bxt_init_clock_gating(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	gen9_init_clock_gating(dev);
-
 	/* WaDisableSDEUnitClockGating:bxt */
 	I915_WRITE(GEN8_UCGCTL6, I915_READ(GEN8_UCGCTL6) |
 		   GEN8_SDEUNIT_CLOCK_GATE_DISABLE);
@@ -112,17 +66,6 @@ static void bxt_init_clock_gating(struct drm_device *dev)
 	 */
 	I915_WRITE(GEN8_UCGCTL6, I915_READ(GEN8_UCGCTL6) |
 		   GEN8_HDCUNIT_CLOCK_GATE_DISABLE_HDCREQ);
-
-	/* WaStoreMultiplePTEenable:bxt */
-	/* This is a requirement according to Hardware specification */
-	if (INTEL_REVID(dev) == BXT_REVID_A0)
-		I915_WRITE(TILECTL, I915_READ(TILECTL) | TILECTL_TLBPF);
-
-	/* WaSetClckGatingDisableMedia:bxt */
-	if (INTEL_REVID(dev) == BXT_REVID_A0) {
-		I915_WRITE(GEN7_MISCCPCTL, (I915_READ(GEN7_MISCCPCTL) &
-					    ~GEN8_DOP_CLOCK_GATE_MEDIA_ENABLE));
-	}
 }
 
 static void i915_pineview_get_mem_freq(struct drm_device *dev)
@@ -7109,9 +7052,6 @@ void intel_init_pm(struct drm_device *dev)
 		if (IS_BROXTON(dev))
 			dev_priv->display.init_clock_gating =
 				bxt_init_clock_gating;
-		else if (IS_SKYLAKE(dev))
-			dev_priv->display.init_clock_gating =
-				skl_init_clock_gating;
 		dev_priv->display.update_wm = skl_update_wm;
 		dev_priv->display.update_sprite_wm = skl_update_sprite_wm;
 	} else if (HAS_PCH_SPLIT(dev)) {
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index 654ae99..0359736 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -906,6 +906,14 @@ static int gen9_init_workarounds(struct intel_engine_cs *ring)
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	uint32_t tmp;
 
+	/* WaEnableLbsSlaRetryTimerDecrement:skl */
+	I915_WRITE(BDW_SCRATCH1, I915_READ(BDW_SCRATCH1) |
+		   GEN9_LBS_SLA_RETRY_TIMER_DECREMENT_ENABLE);
+
+	/* WaDisableKillLogic:bxt,skl */
+	I915_WRITE(GAM_ECOCHK, I915_READ(GAM_ECOCHK) |
+		   ECOCHK_DIS_TLB);
+
 	/* WaDisablePartialInstShootdown:skl,bxt */
 	WA_SET_BIT_MASKED(GEN8_ROW_CHICKEN,
 			  PARTIAL_INSTRUCTION_SHOOTDOWN_DISABLE);
@@ -1018,7 +1026,6 @@ static int skl_tune_iz_hashing(struct intel_engine_cs *ring)
 	return 0;
 }
 
-
 static int skl_init_workarounds(struct intel_engine_cs *ring)
 {
 	int ret;
@@ -1029,6 +1036,30 @@ static int skl_init_workarounds(struct intel_engine_cs *ring)
 	if (ret)
 		return ret;
 
+	if (INTEL_REVID(dev) <= SKL_REVID_D0) {
+		/* WaDisableHDCInvalidation:skl */
+		I915_WRITE(GAM_ECOCHK, I915_READ(GAM_ECOCHK) |
+			   BDW_DISABLE_HDC_INVALIDATION);
+
+		/* WaDisableChickenBitTSGBarrierAckForFFSliceCS:skl */
+		I915_WRITE(FF_SLICE_CS_CHICKEN2,
+			   _MASKED_BIT_ENABLE(GEN9_TSG_BARRIER_ACK_DISABLE));
+	}
+
+	/* GEN8_L3SQCREG4 has a dependency with WA batch so any new changes
+	 * involving this register should also be added to WA batch as required.
+	 */
+	if (INTEL_REVID(dev) <= SKL_REVID_E0)
+		/* WaDisableLSQCROPERFforOCL:skl */
+		I915_WRITE(GEN8_L3SQCREG4, I915_READ(GEN8_L3SQCREG4) |
+			   GEN8_LQSC_RO_PERF_DIS);
+
+	/* WaEnableGapsTsvCreditFix:skl */
+	if (IS_SKYLAKE(dev) && (INTEL_REVID(dev) >= SKL_REVID_C0)) {
+		I915_WRITE(GEN8_GARBCNTL, (I915_READ(GEN8_GARBCNTL) |
+					   GEN9_GAPS_TSV_CREDIT_DISABLE));
+	}
+
 	/* WaDisablePowerCompilerClockGating:skl */
 	if (INTEL_REVID(dev) == SKL_REVID_B0)
 		WA_SET_BIT_MASKED(HIZ_CHICKEN,
@@ -1072,6 +1103,17 @@ static int bxt_init_workarounds(struct intel_engine_cs *ring)
 	if (ret)
 		return ret;
 
+	/* WaStoreMultiplePTEenable:bxt */
+	/* This is a requirement according to Hardware specification */
+	if (INTEL_REVID(dev) == BXT_REVID_A0)
+		I915_WRITE(TILECTL, I915_READ(TILECTL) | TILECTL_TLBPF);
+
+	/* WaSetClckGatingDisableMedia:bxt */
+	if (INTEL_REVID(dev) == BXT_REVID_A0) {
+		I915_WRITE(GEN7_MISCCPCTL, (I915_READ(GEN7_MISCCPCTL) &
+					    ~GEN8_DOP_CLOCK_GATE_MEDIA_ENABLE));
+	}
+
 	/* WaDisableThreadStallDopClockGating:bxt */
 	WA_SET_BIT_MASKED(GEN8_ROW_CHICKEN,
 			  STALL_DOP_GATING_DISABLE);
-- 
cgit v0.10.2


From 7564fde3a13e000867729a291e77cdefa0854b8e Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 13 Oct 2015 11:16:48 +0300
Subject: MAINTAINERS: add link to the Intel Graphics for Linux web site

There's plenty of drm/i915 related hardware and software documentation,
and firmware downloads for the latest platforms.

Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/MAINTAINERS b/MAINTAINERS
index 9f6685f..c8ca988 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -3584,6 +3584,7 @@ M:	Daniel Vetter <daniel.vetter@intel.com>
 M:	Jani Nikula <jani.nikula@linux.intel.com>
 L:	intel-gfx@lists.freedesktop.org
 L:	dri-devel@lists.freedesktop.org
+W:	https://01.org/linuxgraphics/
 Q:	http://patchwork.freedesktop.org/project/intel-gfx/
 T:	git git://anongit.freedesktop.org/drm-intel
 S:	Supported
-- 
cgit v0.10.2


From f76097c099ab84befce13c0cbc827ece6ea3bcdb Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Fri, 2 Oct 2015 13:53:33 -0400
Subject: drm/amdgpu/atom: implement debug opcode
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Basically a nop.

Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/atom.c b/drivers/gpu/drm/amd/amdgpu/atom.c
index a0346a9..331bd05 100644
--- a/drivers/gpu/drm/amd/amdgpu/atom.c
+++ b/drivers/gpu/drm/amd/amdgpu/atom.c
@@ -1022,7 +1022,8 @@ static void atom_op_xor(atom_exec_context *ctx, int *ptr, int arg)
 
 static void atom_op_debug(atom_exec_context *ctx, int *ptr, int arg)
 {
-	printk(KERN_INFO "unimplemented!\n");
+	uint8_t val = U8((*ptr)++);
+	SDEBUG("DEBUG output: 0x%02X\n", val);
 }
 
 static struct {
-- 
cgit v0.10.2


From 554384198c11717d9d9fdb2c9aa83ab78cd50fdf Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Fri, 2 Oct 2015 14:03:26 -0400
Subject: drm/amdgpu/atom: add support for process ds opcode
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Just skips a data section.

Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/atom.c b/drivers/gpu/drm/amd/amdgpu/atom.c
index 331bd05..885d3d3 100644
--- a/drivers/gpu/drm/amd/amdgpu/atom.c
+++ b/drivers/gpu/drm/amd/amdgpu/atom.c
@@ -1026,6 +1026,13 @@ static void atom_op_debug(atom_exec_context *ctx, int *ptr, int arg)
 	SDEBUG("DEBUG output: 0x%02X\n", val);
 }
 
+static void atom_op_processds(atom_exec_context *ctx, int *ptr, int arg)
+{
+	uint16_t val = U16(*ptr);
+	(*ptr) += val + 2;
+	SDEBUG("PROCESSDS output: 0x%02X\n", val);
+}
+
 static struct {
 	void (*func) (atom_exec_context *, int *, int);
 	int arg;
@@ -1152,7 +1159,9 @@ static struct {
 	atom_op_shr, ATOM_ARG_FB}, {
 	atom_op_shr, ATOM_ARG_PLL}, {
 	atom_op_shr, ATOM_ARG_MC}, {
-atom_op_debug, 0},};
+	atom_op_debug, 0}, {
+	atom_op_processds, 0},
+};
 
 static int amdgpu_atom_execute_table_locked(struct atom_context *ctx, int index, uint32_t * params)
 {
-- 
cgit v0.10.2


From c9c145021f8fc8445fb07d16073696330b6186c8 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Fri, 2 Oct 2015 14:16:11 -0400
Subject: drm/amdgpu/atom: add support for new mul32 opcodes (v2)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Better precision than the regular mul opcode.

v2: handle big endian properly.

Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/atom.c b/drivers/gpu/drm/amd/amdgpu/atom.c
index 885d3d3..474357d 100644
--- a/drivers/gpu/drm/amd/amdgpu/atom.c
+++ b/drivers/gpu/drm/amd/amdgpu/atom.c
@@ -788,6 +788,20 @@ static void atom_op_mul(atom_exec_context *ctx, int *ptr, int arg)
 	ctx->ctx->divmul[0] = dst * src;
 }
 
+static void atom_op_mul32(atom_exec_context *ctx, int *ptr, int arg)
+{
+	uint64_t val64;
+	uint8_t attr = U8((*ptr)++);
+	uint32_t dst, src;
+	SDEBUG("   src1: ");
+	dst = atom_get_dst(ctx, arg, attr, ptr, NULL, 1);
+	SDEBUG("   src2: ");
+	src = atom_get_src(ctx, attr, ptr);
+	val64 = (uint64_t)dst * (uint64_t)src;
+	ctx->ctx->divmul[0] = lower_32_bits(val64);
+	ctx->ctx->divmul[1] = upper_32_bits(val64);
+}
+
 static void atom_op_nop(atom_exec_context *ctx, int *ptr, int arg)
 {
 	/* nothing */
@@ -1160,7 +1174,9 @@ static struct {
 	atom_op_shr, ATOM_ARG_PLL}, {
 	atom_op_shr, ATOM_ARG_MC}, {
 	atom_op_debug, 0}, {
-	atom_op_processds, 0},
+	atom_op_processds, 0}, {
+	atom_op_mul32, ATOM_ARG_PS}, {
+	atom_op_mul32, ATOM_ARG_WS},
 };
 
 static int amdgpu_atom_execute_table_locked(struct atom_context *ctx, int index, uint32_t * params)
diff --git a/drivers/gpu/drm/amd/amdgpu/atom.h b/drivers/gpu/drm/amd/amdgpu/atom.h
index 09d0f82..a940eb3 100644
--- a/drivers/gpu/drm/amd/amdgpu/atom.h
+++ b/drivers/gpu/drm/amd/amdgpu/atom.h
@@ -60,7 +60,7 @@
 #define ATOM_CT_PS_MASK		0x7F
 #define ATOM_CT_CODE_PTR	6
 
-#define ATOM_OP_CNT		123
+#define ATOM_OP_CNT		125
 #define ATOM_OP_EOT		91
 
 #define ATOM_CASE_MAGIC		0x63
-- 
cgit v0.10.2


From c2fe16aa36d2bc976f7e79600d3a118fafdcc8dc Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Fri, 2 Oct 2015 14:26:41 -0400
Subject: drm/amdgpu/atom: add support for new div32 opcodes (v3)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Better precision than the regular div opcode.

v2: drop 64 bit divide
v3: fix op handling.  This actually is a 64 bit divide.

Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/atom.c b/drivers/gpu/drm/amd/amdgpu/atom.c
index 474357d..1b50e6c 100644
--- a/drivers/gpu/drm/amd/amdgpu/atom.c
+++ b/drivers/gpu/drm/amd/amdgpu/atom.c
@@ -685,6 +685,27 @@ static void atom_op_div(atom_exec_context *ctx, int *ptr, int arg)
 	}
 }
 
+static void atom_op_div32(atom_exec_context *ctx, int *ptr, int arg)
+{
+	uint64_t val64;
+	uint8_t attr = U8((*ptr)++);
+	uint32_t dst, src;
+	SDEBUG("   src1: ");
+	dst = atom_get_dst(ctx, arg, attr, ptr, NULL, 1);
+	SDEBUG("   src2: ");
+	src = atom_get_src(ctx, attr, ptr);
+	if (src != 0) {
+		val64 = dst;
+		val64 |= ((uint64_t)ctx->ctx->divmul[1]) << 32;
+		do_div(val64, src);
+		ctx->ctx->divmul[0] = lower_32_bits(val64);
+		ctx->ctx->divmul[1] = upper_32_bits(val64);
+	} else {
+		ctx->ctx->divmul[0] = 0;
+		ctx->ctx->divmul[1] = 0;
+	}
+}
+
 static void atom_op_eot(atom_exec_context *ctx, int *ptr, int arg)
 {
 	/* functionally, a nop */
@@ -1176,7 +1197,9 @@ static struct {
 	atom_op_debug, 0}, {
 	atom_op_processds, 0}, {
 	atom_op_mul32, ATOM_ARG_PS}, {
-	atom_op_mul32, ATOM_ARG_WS},
+	atom_op_mul32, ATOM_ARG_WS}, {
+	atom_op_div32, ATOM_ARG_PS}, {
+	atom_op_div32, ATOM_ARG_WS},
 };
 
 static int amdgpu_atom_execute_table_locked(struct atom_context *ctx, int index, uint32_t * params)
diff --git a/drivers/gpu/drm/amd/amdgpu/atom.h b/drivers/gpu/drm/amd/amdgpu/atom.h
index a940eb3..fece8f4 100644
--- a/drivers/gpu/drm/amd/amdgpu/atom.h
+++ b/drivers/gpu/drm/amd/amdgpu/atom.h
@@ -60,7 +60,7 @@
 #define ATOM_CT_PS_MASK		0x7F
 #define ATOM_CT_CODE_PTR	6
 
-#define ATOM_OP_CNT		125
+#define ATOM_OP_CNT		127
 #define ATOM_OP_EOT		91
 
 #define ATOM_CASE_MAGIC		0x63
-- 
cgit v0.10.2


From 29275a9b09ce2abd8fa01268b217c0b56ff22d83 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Thu, 24 Sep 2015 17:29:44 -0400
Subject: drm/amdgpu/dce10: Use cursor_set2 hook for enabling / disabling the
 HW cursor
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The cursor_set2 hook provides the cursor hotspot position within the
cursor image. When the hotspot position changes, we can adjust the cursor
position such that the hotspot doesn't move on the screen. This prevents
the cursor from appearing to intermittently jump around on the screen
when the position of the hotspot within the cursor image changes.

Port of radeon commits:
78b1a6010b46a69bcd47b723a80f92693f26d17b
3feba08d79c32777a845c3c8a4ab93092bdf6f19

Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h
index 2b03425..d8dea52 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h
@@ -373,6 +373,10 @@ struct amdgpu_crtc {
 	uint32_t crtc_offset;
 	struct drm_gem_object *cursor_bo;
 	uint64_t cursor_addr;
+	int cursor_x;
+	int cursor_y;
+	int cursor_hot_x;
+	int cursor_hot_y;
 	int cursor_width;
 	int cursor_height;
 	int max_cursor_width;
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
index e4d101b..3057996 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
@@ -2505,20 +2505,8 @@ static void dce_v10_0_show_cursor(struct drm_crtc *crtc)
 	WREG32_IDX(mmCUR_CONTROL + amdgpu_crtc->crtc_offset, tmp);
 }
 
-static void dce_v10_0_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *obj,
-			      uint64_t gpu_addr)
-{
-	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	struct amdgpu_device *adev = crtc->dev->dev_private;
-
-	WREG32(mmCUR_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset,
-	       upper_32_bits(gpu_addr));
-	WREG32(mmCUR_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset,
-	       lower_32_bits(gpu_addr));
-}
-
-static int dce_v10_0_crtc_cursor_move(struct drm_crtc *crtc,
-				     int x, int y)
+static int dce_v10_0_cursor_move_locked(struct drm_crtc *crtc,
+					int x, int y)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct amdgpu_device *adev = crtc->dev->dev_private;
@@ -2538,21 +2526,61 @@ static int dce_v10_0_crtc_cursor_move(struct drm_crtc *crtc,
 		y = 0;
 	}
 
-	dce_v10_0_lock_cursor(crtc, true);
 	WREG32(mmCUR_POSITION + amdgpu_crtc->crtc_offset, (x << 16) | y);
 	WREG32(mmCUR_HOT_SPOT + amdgpu_crtc->crtc_offset, (xorigin << 16) | yorigin);
 	WREG32(mmCUR_SIZE + amdgpu_crtc->crtc_offset,
 	       ((amdgpu_crtc->cursor_width - 1) << 16) | (amdgpu_crtc->cursor_height - 1));
-	dce_v10_0_lock_cursor(crtc, false);
+
+	amdgpu_crtc->cursor_x = x;
+	amdgpu_crtc->cursor_y = y;
 
 	return 0;
 }
 
-static int dce_v10_0_crtc_cursor_set(struct drm_crtc *crtc,
-				    struct drm_file *file_priv,
-				    uint32_t handle,
-				    uint32_t width,
-				    uint32_t height)
+static void dce_v10_0_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *obj,
+				 uint64_t gpu_addr, int hot_x, int hot_y)
+{
+	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
+	struct amdgpu_device *adev = crtc->dev->dev_private;
+
+	WREG32(mmCUR_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset,
+	       upper_32_bits(gpu_addr));
+	WREG32(mmCUR_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset,
+	       lower_32_bits(gpu_addr));
+
+	if (hot_x != amdgpu_crtc->cursor_hot_x ||
+	    hot_y != amdgpu_crtc->cursor_hot_y) {
+		int x, y;
+
+		x = amdgpu_crtc->cursor_x + amdgpu_crtc->cursor_hot_x - hot_x;
+		y = amdgpu_crtc->cursor_y + amdgpu_crtc->cursor_hot_y - hot_y;
+
+		dce_v10_0_cursor_move_locked(crtc, x, y);
+
+		amdgpu_crtc->cursor_hot_x = hot_x;
+		amdgpu_crtc->cursor_hot_y = hot_y;
+	}
+}
+
+static int dce_v10_0_crtc_cursor_move(struct drm_crtc *crtc,
+				      int x, int y)
+{
+	int ret;
+
+	dce_v10_0_lock_cursor(crtc, true);
+	ret = dce_v10_0_cursor_move_locked(crtc, x, y);
+	dce_v10_0_lock_cursor(crtc, false);
+
+	return ret;
+}
+
+static int dce_v10_0_crtc_cursor_set2(struct drm_crtc *crtc,
+				      struct drm_file *file_priv,
+				      uint32_t handle,
+				      uint32_t width,
+				      uint32_t height,
+				      int32_t hot_x,
+				      int32_t hot_y)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_gem_object *obj;
@@ -2593,7 +2621,7 @@ static int dce_v10_0_crtc_cursor_set(struct drm_crtc *crtc,
 	amdgpu_crtc->cursor_height = height;
 
 	dce_v10_0_lock_cursor(crtc, true);
-	dce_v10_0_set_cursor(crtc, obj, gpu_addr);
+	dce_v10_0_set_cursor(crtc, obj, gpu_addr, hot_x, hot_y);
 	dce_v10_0_show_cursor(crtc);
 	dce_v10_0_lock_cursor(crtc, false);
 
@@ -2641,7 +2669,7 @@ static void dce_v10_0_crtc_destroy(struct drm_crtc *crtc)
 }
 
 static const struct drm_crtc_funcs dce_v10_0_crtc_funcs = {
-	.cursor_set = dce_v10_0_crtc_cursor_set,
+	.cursor_set2 = dce_v10_0_crtc_cursor_set2,
 	.cursor_move = dce_v10_0_crtc_cursor_move,
 	.gamma_set = dce_v10_0_crtc_gamma_set,
 	.set_config = amdgpu_crtc_set_config,
-- 
cgit v0.10.2


From dd0b5d2f886daced723c374fdb0e013e6186493b Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Thu, 24 Sep 2015 17:43:09 -0400
Subject: drm/amdgpu/dce10: Re-show the cursor after a modeset (v2)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Setting a mode seems to clear the cursor registers, so we need to
re-program them to make sure the cursor is visible.

Port of radeon commit:
6d3759fac636028849f3bbec80c4b77e9bfdb1d2

v2: change radeon reference in error message

Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
index 3057996..75fcad6 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
@@ -2537,11 +2537,23 @@ static int dce_v10_0_cursor_move_locked(struct drm_crtc *crtc,
 	return 0;
 }
 
-static void dce_v10_0_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *obj,
-				 uint64_t gpu_addr, int hot_x, int hot_y)
+static int dce_v10_0_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *obj,
+				int hot_x, int hot_y)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct amdgpu_device *adev = crtc->dev->dev_private;
+	struct amdgpu_bo *aobj = gem_to_amdgpu_bo(obj);
+	uint64_t gpu_addr;
+	int ret;
+
+	ret = amdgpu_bo_reserve(aobj, false);
+	if (unlikely(ret != 0))
+		goto fail;
+
+	ret = amdgpu_bo_pin(aobj, AMDGPU_GEM_DOMAIN_VRAM, &gpu_addr);
+	amdgpu_bo_unreserve(aobj);
+	if (ret)
+		goto fail;
 
 	WREG32(mmCUR_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset,
 	       upper_32_bits(gpu_addr));
@@ -2560,6 +2572,13 @@ static void dce_v10_0_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *o
 		amdgpu_crtc->cursor_hot_x = hot_x;
 		amdgpu_crtc->cursor_hot_y = hot_y;
 	}
+
+	return 0;
+
+fail:
+	drm_gem_object_unreference_unlocked(obj);
+
+	return ret;
 }
 
 static int dce_v10_0_crtc_cursor_move(struct drm_crtc *crtc,
@@ -2584,8 +2603,6 @@ static int dce_v10_0_crtc_cursor_set2(struct drm_crtc *crtc,
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_gem_object *obj;
-	struct amdgpu_bo *robj;
-	uint64_t gpu_addr;
 	int ret;
 
 	if (!handle) {
@@ -2607,41 +2624,56 @@ static int dce_v10_0_crtc_cursor_set2(struct drm_crtc *crtc,
 		return -ENOENT;
 	}
 
-	robj = gem_to_amdgpu_bo(obj);
-	ret = amdgpu_bo_reserve(robj, false);
-	if (unlikely(ret != 0))
-		goto fail;
-	ret = amdgpu_bo_pin_restricted(robj, AMDGPU_GEM_DOMAIN_VRAM,
-				       0, 0, &gpu_addr);
-	amdgpu_bo_unreserve(robj);
-	if (ret)
-		goto fail;
-
 	amdgpu_crtc->cursor_width = width;
 	amdgpu_crtc->cursor_height = height;
 
 	dce_v10_0_lock_cursor(crtc, true);
-	dce_v10_0_set_cursor(crtc, obj, gpu_addr, hot_x, hot_y);
-	dce_v10_0_show_cursor(crtc);
+	ret = dce_v10_0_set_cursor(crtc, obj, hot_x, hot_y);
+	if (ret)
+		DRM_ERROR("dce_v10_0_set_cursor returned %d, not changing cursor\n",
+			  ret);
+	else
+		dce_v10_0_show_cursor(crtc);
 	dce_v10_0_lock_cursor(crtc, false);
 
 unpin:
 	if (amdgpu_crtc->cursor_bo) {
-		robj = gem_to_amdgpu_bo(amdgpu_crtc->cursor_bo);
-		ret = amdgpu_bo_reserve(robj, false);
+		struct amdgpu_bo *aobj = gem_to_amdgpu_bo(amdgpu_crtc->cursor_bo);
+		ret = amdgpu_bo_reserve(aobj, false);
 		if (likely(ret == 0)) {
-			amdgpu_bo_unpin(robj);
-			amdgpu_bo_unreserve(robj);
+			amdgpu_bo_unpin(aobj);
+			amdgpu_bo_unreserve(aobj);
 		}
-		drm_gem_object_unreference_unlocked(amdgpu_crtc->cursor_bo);
+		if (amdgpu_crtc->cursor_bo != obj)
+			drm_gem_object_unreference_unlocked(amdgpu_crtc->cursor_bo);
 	}
 
 	amdgpu_crtc->cursor_bo = obj;
 	return 0;
-fail:
-	drm_gem_object_unreference_unlocked(obj);
+}
 
-	return ret;
+static void dce_v10_0_cursor_reset(struct drm_crtc *crtc)
+{
+	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
+	int ret;
+
+	if (amdgpu_crtc->cursor_bo) {
+		dce_v10_0_lock_cursor(crtc, true);
+
+		dce_v10_0_cursor_move_locked(crtc, amdgpu_crtc->cursor_x,
+					     amdgpu_crtc->cursor_y);
+
+		ret = dce_v10_0_set_cursor(crtc, amdgpu_crtc->cursor_bo,
+					   amdgpu_crtc->cursor_hot_x,
+					   amdgpu_crtc->cursor_hot_y);
+		if (ret)
+			DRM_ERROR("dce_v10_0_set_cursor returned %d, not showing "
+				  "cursor\n", ret);
+		else
+			dce_v10_0_show_cursor(crtc);
+
+		dce_v10_0_lock_cursor(crtc, false);
+	}
 }
 
 static void dce_v10_0_crtc_gamma_set(struct drm_crtc *crtc, u16 *red, u16 *green,
@@ -2802,6 +2834,7 @@ static int dce_v10_0_crtc_mode_set(struct drm_crtc *crtc,
 	dce_v10_0_crtc_do_set_base(crtc, old_fb, x, y, 0);
 	amdgpu_atombios_crtc_overscan_setup(crtc, mode, adjusted_mode);
 	amdgpu_atombios_crtc_scaler_setup(crtc);
+	dce_v10_0_cursor_reset(crtc);
 	/* update the hw version fpr dpm */
 	amdgpu_crtc->hw_mode = *adjusted_mode;
 
-- 
cgit v0.10.2


From ef67e38cfe63c70b7901bd5fdb8a170a94b2ef69 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Thu, 24 Sep 2015 17:46:19 -0400
Subject: drm/amdgpu/dce10: Move hotspot handling out of set_cursor
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

It's only needed in dce10 crtc_cursor_set2.

Port of radeon commit:
2e007e611bc0a46dfed1547e0228a57f800e703d

Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
index 75fcad6..f2f93b3 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
@@ -2537,8 +2537,7 @@ static int dce_v10_0_cursor_move_locked(struct drm_crtc *crtc,
 	return 0;
 }
 
-static int dce_v10_0_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *obj,
-				int hot_x, int hot_y)
+static int dce_v10_0_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *obj)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct amdgpu_device *adev = crtc->dev->dev_private;
@@ -2560,19 +2559,6 @@ static int dce_v10_0_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *ob
 	WREG32(mmCUR_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset,
 	       lower_32_bits(gpu_addr));
 
-	if (hot_x != amdgpu_crtc->cursor_hot_x ||
-	    hot_y != amdgpu_crtc->cursor_hot_y) {
-		int x, y;
-
-		x = amdgpu_crtc->cursor_x + amdgpu_crtc->cursor_hot_x - hot_x;
-		y = amdgpu_crtc->cursor_y + amdgpu_crtc->cursor_hot_y - hot_y;
-
-		dce_v10_0_cursor_move_locked(crtc, x, y);
-
-		amdgpu_crtc->cursor_hot_x = hot_x;
-		amdgpu_crtc->cursor_hot_y = hot_y;
-	}
-
 	return 0;
 
 fail:
@@ -2628,7 +2614,21 @@ static int dce_v10_0_crtc_cursor_set2(struct drm_crtc *crtc,
 	amdgpu_crtc->cursor_height = height;
 
 	dce_v10_0_lock_cursor(crtc, true);
-	ret = dce_v10_0_set_cursor(crtc, obj, hot_x, hot_y);
+
+	if (hot_x != amdgpu_crtc->cursor_hot_x ||
+	    hot_y != amdgpu_crtc->cursor_hot_y) {
+		int x, y;
+
+		x = amdgpu_crtc->cursor_x + amdgpu_crtc->cursor_hot_x - hot_x;
+		y = amdgpu_crtc->cursor_y + amdgpu_crtc->cursor_hot_y - hot_y;
+
+		dce_v10_0_cursor_move_locked(crtc, x, y);
+
+		amdgpu_crtc->cursor_hot_x = hot_x;
+		amdgpu_crtc->cursor_hot_y = hot_y;
+	}
+
+	ret = dce_v10_0_set_cursor(crtc, obj);
 	if (ret)
 		DRM_ERROR("dce_v10_0_set_cursor returned %d, not changing cursor\n",
 			  ret);
@@ -2663,9 +2663,7 @@ static void dce_v10_0_cursor_reset(struct drm_crtc *crtc)
 		dce_v10_0_cursor_move_locked(crtc, amdgpu_crtc->cursor_x,
 					     amdgpu_crtc->cursor_y);
 
-		ret = dce_v10_0_set_cursor(crtc, amdgpu_crtc->cursor_bo,
-					   amdgpu_crtc->cursor_hot_x,
-					   amdgpu_crtc->cursor_hot_y);
+		ret = dce_v10_0_set_cursor(crtc, amdgpu_crtc->cursor_bo);
 		if (ret)
 			DRM_ERROR("dce_v10_0_set_cursor returned %d, not showing "
 				  "cursor\n", ret);
-- 
cgit v0.10.2


From f9242d1baff77aa148a6220f906870611dbc4746 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Thu, 8 Oct 2015 11:25:38 -0400
Subject: drm/amdgpu/dce10: Clean up reference counting and pinning of the
 cursor BOs
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Take a GEM reference for and pin the new cursor BO, unpin and drop the
GEM reference for the old cursor BO in dce10 crtc_cursor_set2, and use
amdgpu_crtc->cursor_addr in dce10 set_cursor.

This fixes dce10 cursor_reset accidentally incrementing the cursor BO
pin count, and cleans up the code a little.

Port of radeon commit:
cd404af0c930104462aa91344f07d002cf8248ed

Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
index f2f93b3..17692f3 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
@@ -2537,34 +2537,15 @@ static int dce_v10_0_cursor_move_locked(struct drm_crtc *crtc,
 	return 0;
 }
 
-static int dce_v10_0_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *obj)
+static void dce_v10_0_set_cursor(struct drm_crtc *crtc)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct amdgpu_device *adev = crtc->dev->dev_private;
-	struct amdgpu_bo *aobj = gem_to_amdgpu_bo(obj);
-	uint64_t gpu_addr;
-	int ret;
-
-	ret = amdgpu_bo_reserve(aobj, false);
-	if (unlikely(ret != 0))
-		goto fail;
-
-	ret = amdgpu_bo_pin(aobj, AMDGPU_GEM_DOMAIN_VRAM, &gpu_addr);
-	amdgpu_bo_unreserve(aobj);
-	if (ret)
-		goto fail;
 
 	WREG32(mmCUR_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset,
-	       upper_32_bits(gpu_addr));
+	       upper_32_bits(amdgpu_crtc->cursor_addr));
 	WREG32(mmCUR_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset,
-	       lower_32_bits(gpu_addr));
-
-	return 0;
-
-fail:
-	drm_gem_object_unreference_unlocked(obj);
-
-	return ret;
+	       lower_32_bits(amdgpu_crtc->cursor_addr));
 }
 
 static int dce_v10_0_crtc_cursor_move(struct drm_crtc *crtc,
@@ -2589,6 +2570,7 @@ static int dce_v10_0_crtc_cursor_set2(struct drm_crtc *crtc,
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_gem_object *obj;
+	struct amdgpu_bo *aobj;
 	int ret;
 
 	if (!handle) {
@@ -2610,6 +2592,21 @@ static int dce_v10_0_crtc_cursor_set2(struct drm_crtc *crtc,
 		return -ENOENT;
 	}
 
+	aobj = gem_to_amdgpu_bo(obj);
+	ret = amdgpu_bo_reserve(aobj, false);
+	if (ret != 0) {
+		drm_gem_object_unreference_unlocked(obj);
+		return ret;
+	}
+
+	ret = amdgpu_bo_pin(aobj, AMDGPU_GEM_DOMAIN_VRAM, &amdgpu_crtc->cursor_addr);
+	amdgpu_bo_unreserve(aobj);
+	if (ret) {
+		DRM_ERROR("Failed to pin new cursor BO (%d)\n", ret);
+		drm_gem_object_unreference_unlocked(obj);
+		return ret;
+	}
+
 	amdgpu_crtc->cursor_width = width;
 	amdgpu_crtc->cursor_height = height;
 
@@ -2628,12 +2625,8 @@ static int dce_v10_0_crtc_cursor_set2(struct drm_crtc *crtc,
 		amdgpu_crtc->cursor_hot_y = hot_y;
 	}
 
-	ret = dce_v10_0_set_cursor(crtc, obj);
-	if (ret)
-		DRM_ERROR("dce_v10_0_set_cursor returned %d, not changing cursor\n",
-			  ret);
-	else
-		dce_v10_0_show_cursor(crtc);
+	dce_v10_0_set_cursor(crtc);
+	dce_v10_0_show_cursor(crtc);
 	dce_v10_0_lock_cursor(crtc, false);
 
 unpin:
@@ -2644,8 +2637,7 @@ unpin:
 			amdgpu_bo_unpin(aobj);
 			amdgpu_bo_unreserve(aobj);
 		}
-		if (amdgpu_crtc->cursor_bo != obj)
-			drm_gem_object_unreference_unlocked(amdgpu_crtc->cursor_bo);
+		drm_gem_object_unreference_unlocked(amdgpu_crtc->cursor_bo);
 	}
 
 	amdgpu_crtc->cursor_bo = obj;
@@ -2655,7 +2647,6 @@ unpin:
 static void dce_v10_0_cursor_reset(struct drm_crtc *crtc)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	int ret;
 
 	if (amdgpu_crtc->cursor_bo) {
 		dce_v10_0_lock_cursor(crtc, true);
@@ -2663,12 +2654,8 @@ static void dce_v10_0_cursor_reset(struct drm_crtc *crtc)
 		dce_v10_0_cursor_move_locked(crtc, amdgpu_crtc->cursor_x,
 					     amdgpu_crtc->cursor_y);
 
-		ret = dce_v10_0_set_cursor(crtc, amdgpu_crtc->cursor_bo);
-		if (ret)
-			DRM_ERROR("dce_v10_0_set_cursor returned %d, not showing "
-				  "cursor\n", ret);
-		else
-			dce_v10_0_show_cursor(crtc);
+		dce_v10_0_set_cursor(crtc);
+		dce_v10_0_show_cursor(crtc);
 
 		dce_v10_0_lock_cursor(crtc, false);
 	}
-- 
cgit v0.10.2


From 3c681718b4bf73d80fd6a11033e12fd4402a7a1c Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Fri, 25 Sep 2015 09:51:59 -0400
Subject: drm/amdgpu/dce10: Fold set_cursor() into show_cursor()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Port of radeon commit:
8991668ab4e26f985a8485719bce5d6d0623a644

Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
index 17692f3..57e1232 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
@@ -2499,6 +2499,11 @@ static void dce_v10_0_show_cursor(struct drm_crtc *crtc)
 	struct amdgpu_device *adev = crtc->dev->dev_private;
 	u32 tmp;
 
+	WREG32(mmCUR_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset,
+	       upper_32_bits(amdgpu_crtc->cursor_addr));
+	WREG32(mmCUR_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset,
+	       lower_32_bits(amdgpu_crtc->cursor_addr));
+
 	tmp = RREG32_IDX(mmCUR_CONTROL + amdgpu_crtc->crtc_offset);
 	tmp = REG_SET_FIELD(tmp, CUR_CONTROL, CURSOR_EN, 1);
 	tmp = REG_SET_FIELD(tmp, CUR_CONTROL, CURSOR_MODE, 2);
@@ -2537,17 +2542,6 @@ static int dce_v10_0_cursor_move_locked(struct drm_crtc *crtc,
 	return 0;
 }
 
-static void dce_v10_0_set_cursor(struct drm_crtc *crtc)
-{
-	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	struct amdgpu_device *adev = crtc->dev->dev_private;
-
-	WREG32(mmCUR_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset,
-	       upper_32_bits(amdgpu_crtc->cursor_addr));
-	WREG32(mmCUR_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset,
-	       lower_32_bits(amdgpu_crtc->cursor_addr));
-}
-
 static int dce_v10_0_crtc_cursor_move(struct drm_crtc *crtc,
 				      int x, int y)
 {
@@ -2625,7 +2619,6 @@ static int dce_v10_0_crtc_cursor_set2(struct drm_crtc *crtc,
 		amdgpu_crtc->cursor_hot_y = hot_y;
 	}
 
-	dce_v10_0_set_cursor(crtc);
 	dce_v10_0_show_cursor(crtc);
 	dce_v10_0_lock_cursor(crtc, false);
 
@@ -2654,7 +2647,6 @@ static void dce_v10_0_cursor_reset(struct drm_crtc *crtc)
 		dce_v10_0_cursor_move_locked(crtc, amdgpu_crtc->cursor_x,
 					     amdgpu_crtc->cursor_y);
 
-		dce_v10_0_set_cursor(crtc);
 		dce_v10_0_show_cursor(crtc);
 
 		dce_v10_0_lock_cursor(crtc, false);
-- 
cgit v0.10.2


From d8ee89c6393d78b52b42fe0f188fe382fa8fc162 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Wed, 7 Oct 2015 17:07:01 -0400
Subject: drm/amdgpu/dce11: Use cursor_set2 hook for enabling / disabling the
 HW cursor
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The cursor_set2 hook provides the cursor hotspot position within the
cursor image. When the hotspot position changes, we can adjust the cursor
position such that the hotspot doesn't move on the screen. This prevents
the cursor from appearing to intermittently jump around on the screen
when the position of the hotspot within the cursor image changes.

Port of radeon commits:
78b1a6010b46a69bcd47b723a80f92693f26d17b
3feba08d79c32777a845c3c8a4ab93092bdf6f19

Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
index 6411e82..5be7783 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
@@ -2482,20 +2482,8 @@ static void dce_v11_0_show_cursor(struct drm_crtc *crtc)
 	WREG32_IDX(mmCUR_CONTROL + amdgpu_crtc->crtc_offset, tmp);
 }
 
-static void dce_v11_0_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *obj,
-			      uint64_t gpu_addr)
-{
-	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	struct amdgpu_device *adev = crtc->dev->dev_private;
-
-	WREG32(mmCUR_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset,
-	       upper_32_bits(gpu_addr));
-	WREG32(mmCUR_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset,
-	       lower_32_bits(gpu_addr));
-}
-
-static int dce_v11_0_crtc_cursor_move(struct drm_crtc *crtc,
-				     int x, int y)
+static int dce_v11_0_cursor_move_locked(struct drm_crtc *crtc,
+					int x, int y)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct amdgpu_device *adev = crtc->dev->dev_private;
@@ -2515,21 +2503,61 @@ static int dce_v11_0_crtc_cursor_move(struct drm_crtc *crtc,
 		y = 0;
 	}
 
-	dce_v11_0_lock_cursor(crtc, true);
 	WREG32(mmCUR_POSITION + amdgpu_crtc->crtc_offset, (x << 16) | y);
 	WREG32(mmCUR_HOT_SPOT + amdgpu_crtc->crtc_offset, (xorigin << 16) | yorigin);
 	WREG32(mmCUR_SIZE + amdgpu_crtc->crtc_offset,
 	       ((amdgpu_crtc->cursor_width - 1) << 16) | (amdgpu_crtc->cursor_height - 1));
-	dce_v11_0_lock_cursor(crtc, false);
+
+	amdgpu_crtc->cursor_x = x;
+	amdgpu_crtc->cursor_y = y;
 
 	return 0;
 }
 
-static int dce_v11_0_crtc_cursor_set(struct drm_crtc *crtc,
-				    struct drm_file *file_priv,
-				    uint32_t handle,
-				    uint32_t width,
-				    uint32_t height)
+static void dce_v11_0_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *obj,
+				 uint64_t gpu_addr, int hot_x, int hot_y)
+{
+	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
+	struct amdgpu_device *adev = crtc->dev->dev_private;
+
+	WREG32(mmCUR_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset,
+	       upper_32_bits(gpu_addr));
+	WREG32(mmCUR_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset,
+	       lower_32_bits(gpu_addr));
+
+	if (hot_x != amdgpu_crtc->cursor_hot_x ||
+	    hot_y != amdgpu_crtc->cursor_hot_y) {
+		int x, y;
+
+		x = amdgpu_crtc->cursor_x + amdgpu_crtc->cursor_hot_x - hot_x;
+		y = amdgpu_crtc->cursor_y + amdgpu_crtc->cursor_hot_y - hot_y;
+
+		dce_v11_0_cursor_move_locked(crtc, x, y);
+
+		amdgpu_crtc->cursor_hot_x = hot_x;
+		amdgpu_crtc->cursor_hot_y = hot_y;
+	}
+}
+
+static int dce_v11_0_crtc_cursor_move(struct drm_crtc *crtc,
+				      int x, int y)
+{
+	int ret;
+
+	dce_v11_0_lock_cursor(crtc, true);
+	ret = dce_v11_0_cursor_move_locked(crtc, x, y);
+	dce_v11_0_lock_cursor(crtc, false);
+
+	return ret;
+}
+
+static int dce_v11_0_crtc_cursor_set2(struct drm_crtc *crtc,
+				      struct drm_file *file_priv,
+				      uint32_t handle,
+				      uint32_t width,
+				      uint32_t height,
+				      int32_t hot_x,
+				      int32_t hot_y)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_gem_object *obj;
@@ -2570,7 +2598,7 @@ static int dce_v11_0_crtc_cursor_set(struct drm_crtc *crtc,
 	amdgpu_crtc->cursor_height = height;
 
 	dce_v11_0_lock_cursor(crtc, true);
-	dce_v11_0_set_cursor(crtc, obj, gpu_addr);
+	dce_v11_0_set_cursor(crtc, obj, gpu_addr, hot_x, hot_y);
 	dce_v11_0_show_cursor(crtc);
 	dce_v11_0_lock_cursor(crtc, false);
 
@@ -2618,7 +2646,7 @@ static void dce_v11_0_crtc_destroy(struct drm_crtc *crtc)
 }
 
 static const struct drm_crtc_funcs dce_v11_0_crtc_funcs = {
-	.cursor_set = dce_v11_0_crtc_cursor_set,
+	.cursor_set2 = dce_v11_0_crtc_cursor_set2,
 	.cursor_move = dce_v11_0_crtc_cursor_move,
 	.gamma_set = dce_v11_0_crtc_gamma_set,
 	.set_config = amdgpu_crtc_set_config,
-- 
cgit v0.10.2


From 8ddef5a5e51435075b9412698519f8ddb36ee53c Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Wed, 7 Oct 2015 17:22:23 -0400
Subject: drm/amdgpu/dce11: Re-show the cursor after a modeset (v2)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Setting a mode seems to clear the cursor registers, so we need to
re-program them to make sure the cursor is visible.

Port of radeon commit:
6d3759fac636028849f3bbec80c4b77e9bfdb1d2

v2: change radeon reference in error output

Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
index 5be7783..bebb879 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
@@ -2514,11 +2514,23 @@ static int dce_v11_0_cursor_move_locked(struct drm_crtc *crtc,
 	return 0;
 }
 
-static void dce_v11_0_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *obj,
-				 uint64_t gpu_addr, int hot_x, int hot_y)
+static int dce_v11_0_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *obj,
+				int hot_x, int hot_y)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct amdgpu_device *adev = crtc->dev->dev_private;
+	struct amdgpu_bo *aobj = gem_to_amdgpu_bo(obj);
+	uint64_t gpu_addr;
+	int ret;
+
+	ret = amdgpu_bo_reserve(aobj, false);
+	if (unlikely(ret != 0))
+		goto fail;
+
+	ret = amdgpu_bo_pin(aobj, AMDGPU_GEM_DOMAIN_VRAM, &gpu_addr);
+	amdgpu_bo_unreserve(aobj);
+	if (ret)
+		goto fail;
 
 	WREG32(mmCUR_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset,
 	       upper_32_bits(gpu_addr));
@@ -2537,6 +2549,13 @@ static void dce_v11_0_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *o
 		amdgpu_crtc->cursor_hot_x = hot_x;
 		amdgpu_crtc->cursor_hot_y = hot_y;
 	}
+
+	return 0;
+
+fail:
+	drm_gem_object_unreference_unlocked(obj);
+
+	return ret;
 }
 
 static int dce_v11_0_crtc_cursor_move(struct drm_crtc *crtc,
@@ -2561,8 +2580,6 @@ static int dce_v11_0_crtc_cursor_set2(struct drm_crtc *crtc,
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_gem_object *obj;
-	struct amdgpu_bo *robj;
-	uint64_t gpu_addr;
 	int ret;
 
 	if (!handle) {
@@ -2584,41 +2601,56 @@ static int dce_v11_0_crtc_cursor_set2(struct drm_crtc *crtc,
 		return -ENOENT;
 	}
 
-	robj = gem_to_amdgpu_bo(obj);
-	ret = amdgpu_bo_reserve(robj, false);
-	if (unlikely(ret != 0))
-		goto fail;
-	ret = amdgpu_bo_pin_restricted(robj, AMDGPU_GEM_DOMAIN_VRAM,
-				       0, 0, &gpu_addr);
-	amdgpu_bo_unreserve(robj);
-	if (ret)
-		goto fail;
-
 	amdgpu_crtc->cursor_width = width;
 	amdgpu_crtc->cursor_height = height;
 
 	dce_v11_0_lock_cursor(crtc, true);
-	dce_v11_0_set_cursor(crtc, obj, gpu_addr, hot_x, hot_y);
-	dce_v11_0_show_cursor(crtc);
+	ret = dce_v11_0_set_cursor(crtc, obj, hot_x, hot_y);
+	if (ret)
+		DRM_ERROR("dce_v11_0_set_cursor returned %d, not changing cursor\n",
+			  ret);
+	else
+		dce_v11_0_show_cursor(crtc);
 	dce_v11_0_lock_cursor(crtc, false);
 
 unpin:
 	if (amdgpu_crtc->cursor_bo) {
-		robj = gem_to_amdgpu_bo(amdgpu_crtc->cursor_bo);
-		ret = amdgpu_bo_reserve(robj, false);
+		struct amdgpu_bo *aobj = gem_to_amdgpu_bo(amdgpu_crtc->cursor_bo);
+		ret = amdgpu_bo_reserve(aobj, false);
 		if (likely(ret == 0)) {
-			amdgpu_bo_unpin(robj);
-			amdgpu_bo_unreserve(robj);
+			amdgpu_bo_unpin(aobj);
+			amdgpu_bo_unreserve(aobj);
 		}
-		drm_gem_object_unreference_unlocked(amdgpu_crtc->cursor_bo);
+		if (amdgpu_crtc->cursor_bo != obj)
+			drm_gem_object_unreference_unlocked(amdgpu_crtc->cursor_bo);
 	}
 
 	amdgpu_crtc->cursor_bo = obj;
 	return 0;
-fail:
-	drm_gem_object_unreference_unlocked(obj);
+}
 
-	return ret;
+static void dce_v11_0_cursor_reset(struct drm_crtc *crtc)
+{
+	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
+	int ret;
+
+	if (amdgpu_crtc->cursor_bo) {
+		dce_v11_0_lock_cursor(crtc, true);
+
+		dce_v11_0_cursor_move_locked(crtc, amdgpu_crtc->cursor_x,
+					     amdgpu_crtc->cursor_y);
+
+		ret = dce_v11_0_set_cursor(crtc, amdgpu_crtc->cursor_bo,
+					   amdgpu_crtc->cursor_hot_x,
+					   amdgpu_crtc->cursor_hot_y);
+		if (ret)
+			DRM_ERROR("dce_v11_0_set_cursor returned %d, not showing "
+				  "cursor\n", ret);
+		else
+			dce_v11_0_show_cursor(crtc);
+
+		dce_v11_0_lock_cursor(crtc, false);
+	}
 }
 
 static void dce_v11_0_crtc_gamma_set(struct drm_crtc *crtc, u16 *red, u16 *green,
@@ -2779,6 +2811,7 @@ static int dce_v11_0_crtc_mode_set(struct drm_crtc *crtc,
 	dce_v11_0_crtc_do_set_base(crtc, old_fb, x, y, 0);
 	amdgpu_atombios_crtc_overscan_setup(crtc, mode, adjusted_mode);
 	amdgpu_atombios_crtc_scaler_setup(crtc);
+	dce_v11_0_cursor_reset(crtc);
 	/* update the hw version fpr dpm */
 	amdgpu_crtc->hw_mode = *adjusted_mode;
 
-- 
cgit v0.10.2


From 1996ea09cd89a5bba8b1e193e08850e273046fd6 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Wed, 7 Oct 2015 17:25:54 -0400
Subject: drm/amdgpu/dce11: Move hotspot handling out of set_cursor
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

It's only needed in dce11 crtc_cursor_set2.

Port of radeon commit:
2e007e611bc0a46dfed1547e0228a57f800e703d

Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
index bebb879..9690c02 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
@@ -2514,8 +2514,7 @@ static int dce_v11_0_cursor_move_locked(struct drm_crtc *crtc,
 	return 0;
 }
 
-static int dce_v11_0_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *obj,
-				int hot_x, int hot_y)
+static int dce_v11_0_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *obj)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct amdgpu_device *adev = crtc->dev->dev_private;
@@ -2537,19 +2536,6 @@ static int dce_v11_0_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *ob
 	WREG32(mmCUR_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset,
 	       lower_32_bits(gpu_addr));
 
-	if (hot_x != amdgpu_crtc->cursor_hot_x ||
-	    hot_y != amdgpu_crtc->cursor_hot_y) {
-		int x, y;
-
-		x = amdgpu_crtc->cursor_x + amdgpu_crtc->cursor_hot_x - hot_x;
-		y = amdgpu_crtc->cursor_y + amdgpu_crtc->cursor_hot_y - hot_y;
-
-		dce_v11_0_cursor_move_locked(crtc, x, y);
-
-		amdgpu_crtc->cursor_hot_x = hot_x;
-		amdgpu_crtc->cursor_hot_y = hot_y;
-	}
-
 	return 0;
 
 fail:
@@ -2605,7 +2591,21 @@ static int dce_v11_0_crtc_cursor_set2(struct drm_crtc *crtc,
 	amdgpu_crtc->cursor_height = height;
 
 	dce_v11_0_lock_cursor(crtc, true);
-	ret = dce_v11_0_set_cursor(crtc, obj, hot_x, hot_y);
+
+	if (hot_x != amdgpu_crtc->cursor_hot_x ||
+	    hot_y != amdgpu_crtc->cursor_hot_y) {
+		int x, y;
+
+		x = amdgpu_crtc->cursor_x + amdgpu_crtc->cursor_hot_x - hot_x;
+		y = amdgpu_crtc->cursor_y + amdgpu_crtc->cursor_hot_y - hot_y;
+
+		dce_v11_0_cursor_move_locked(crtc, x, y);
+
+		amdgpu_crtc->cursor_hot_x = hot_x;
+		amdgpu_crtc->cursor_hot_y = hot_y;
+	}
+
+	ret = dce_v11_0_set_cursor(crtc, obj);
 	if (ret)
 		DRM_ERROR("dce_v11_0_set_cursor returned %d, not changing cursor\n",
 			  ret);
@@ -2640,9 +2640,7 @@ static void dce_v11_0_cursor_reset(struct drm_crtc *crtc)
 		dce_v11_0_cursor_move_locked(crtc, amdgpu_crtc->cursor_x,
 					     amdgpu_crtc->cursor_y);
 
-		ret = dce_v11_0_set_cursor(crtc, amdgpu_crtc->cursor_bo,
-					   amdgpu_crtc->cursor_hot_x,
-					   amdgpu_crtc->cursor_hot_y);
+		ret = dce_v11_0_set_cursor(crtc, amdgpu_crtc->cursor_bo);
 		if (ret)
 			DRM_ERROR("dce_v11_0_set_cursor returned %d, not showing "
 				  "cursor\n", ret);
-- 
cgit v0.10.2


From 232cc6520a6c84a3e5acd075605ddc1bca6bd157 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Thu, 8 Oct 2015 11:28:49 -0400
Subject: drm/amdgpu/dce11: Clean up reference counting and pinning of the
 cursor BOs
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Take a GEM reference for and pin the new cursor BO, unpin and drop the
GEM reference for the old cursor BO in dce11 crtc_cursor_set2, and use
amdgpu_crtc->cursor_addr in dce11 set_cursor.

This fixes dce11 cursor_reset accidentally incrementing the cursor BO
pin count, and cleans up the code a little.

Port of radeon commit:
cd404af0c930104462aa91344f07d002cf8248ed

Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
index 9690c02..5a1ee25 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
@@ -2514,34 +2514,15 @@ static int dce_v11_0_cursor_move_locked(struct drm_crtc *crtc,
 	return 0;
 }
 
-static int dce_v11_0_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *obj)
+static void dce_v11_0_set_cursor(struct drm_crtc *crtc)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct amdgpu_device *adev = crtc->dev->dev_private;
-	struct amdgpu_bo *aobj = gem_to_amdgpu_bo(obj);
-	uint64_t gpu_addr;
-	int ret;
-
-	ret = amdgpu_bo_reserve(aobj, false);
-	if (unlikely(ret != 0))
-		goto fail;
-
-	ret = amdgpu_bo_pin(aobj, AMDGPU_GEM_DOMAIN_VRAM, &gpu_addr);
-	amdgpu_bo_unreserve(aobj);
-	if (ret)
-		goto fail;
 
 	WREG32(mmCUR_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset,
-	       upper_32_bits(gpu_addr));
+	       upper_32_bits(amdgpu_crtc->cursor_addr));
 	WREG32(mmCUR_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset,
-	       lower_32_bits(gpu_addr));
-
-	return 0;
-
-fail:
-	drm_gem_object_unreference_unlocked(obj);
-
-	return ret;
+	       lower_32_bits(amdgpu_crtc->cursor_addr));
 }
 
 static int dce_v11_0_crtc_cursor_move(struct drm_crtc *crtc,
@@ -2566,6 +2547,7 @@ static int dce_v11_0_crtc_cursor_set2(struct drm_crtc *crtc,
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_gem_object *obj;
+	struct amdgpu_bo *aobj;
 	int ret;
 
 	if (!handle) {
@@ -2587,6 +2569,21 @@ static int dce_v11_0_crtc_cursor_set2(struct drm_crtc *crtc,
 		return -ENOENT;
 	}
 
+	aobj = gem_to_amdgpu_bo(obj);
+	ret = amdgpu_bo_reserve(aobj, false);
+	if (ret != 0) {
+		drm_gem_object_unreference_unlocked(obj);
+		return ret;
+	}
+
+	ret = amdgpu_bo_pin(aobj, AMDGPU_GEM_DOMAIN_VRAM, &amdgpu_crtc->cursor_addr);
+	amdgpu_bo_unreserve(aobj);
+	if (ret) {
+		DRM_ERROR("Failed to pin new cursor BO (%d)\n", ret);
+		drm_gem_object_unreference_unlocked(obj);
+		return ret;
+	}
+
 	amdgpu_crtc->cursor_width = width;
 	amdgpu_crtc->cursor_height = height;
 
@@ -2605,12 +2602,8 @@ static int dce_v11_0_crtc_cursor_set2(struct drm_crtc *crtc,
 		amdgpu_crtc->cursor_hot_y = hot_y;
 	}
 
-	ret = dce_v11_0_set_cursor(crtc, obj);
-	if (ret)
-		DRM_ERROR("dce_v11_0_set_cursor returned %d, not changing cursor\n",
-			  ret);
-	else
-		dce_v11_0_show_cursor(crtc);
+	dce_v11_0_set_cursor(crtc);
+	dce_v11_0_show_cursor(crtc);
 	dce_v11_0_lock_cursor(crtc, false);
 
 unpin:
@@ -2621,8 +2614,7 @@ unpin:
 			amdgpu_bo_unpin(aobj);
 			amdgpu_bo_unreserve(aobj);
 		}
-		if (amdgpu_crtc->cursor_bo != obj)
-			drm_gem_object_unreference_unlocked(amdgpu_crtc->cursor_bo);
+		drm_gem_object_unreference_unlocked(amdgpu_crtc->cursor_bo);
 	}
 
 	amdgpu_crtc->cursor_bo = obj;
@@ -2632,7 +2624,6 @@ unpin:
 static void dce_v11_0_cursor_reset(struct drm_crtc *crtc)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	int ret;
 
 	if (amdgpu_crtc->cursor_bo) {
 		dce_v11_0_lock_cursor(crtc, true);
@@ -2640,12 +2631,8 @@ static void dce_v11_0_cursor_reset(struct drm_crtc *crtc)
 		dce_v11_0_cursor_move_locked(crtc, amdgpu_crtc->cursor_x,
 					     amdgpu_crtc->cursor_y);
 
-		ret = dce_v11_0_set_cursor(crtc, amdgpu_crtc->cursor_bo);
-		if (ret)
-			DRM_ERROR("dce_v11_0_set_cursor returned %d, not showing "
-				  "cursor\n", ret);
-		else
-			dce_v11_0_show_cursor(crtc);
+		dce_v11_0_set_cursor(crtc);
+		dce_v11_0_show_cursor(crtc);
 
 		dce_v11_0_lock_cursor(crtc, false);
 	}
-- 
cgit v0.10.2


From ec9353dca9a02b429f31ac58ec7b55ee61757441 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Wed, 7 Oct 2015 23:20:45 -0400
Subject: drm/amdgpu/dce11: Fold set_cursor() into show_cursor()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Port of radeon commit:
8991668ab4e26f985a8485719bce5d6d0623a644

Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
index 5a1ee25..378932a 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
@@ -2476,6 +2476,11 @@ static void dce_v11_0_show_cursor(struct drm_crtc *crtc)
 	struct amdgpu_device *adev = crtc->dev->dev_private;
 	u32 tmp;
 
+	WREG32(mmCUR_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset,
+	       upper_32_bits(amdgpu_crtc->cursor_addr));
+	WREG32(mmCUR_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset,
+	       lower_32_bits(amdgpu_crtc->cursor_addr));
+
 	tmp = RREG32_IDX(mmCUR_CONTROL + amdgpu_crtc->crtc_offset);
 	tmp = REG_SET_FIELD(tmp, CUR_CONTROL, CURSOR_EN, 1);
 	tmp = REG_SET_FIELD(tmp, CUR_CONTROL, CURSOR_MODE, 2);
@@ -2514,17 +2519,6 @@ static int dce_v11_0_cursor_move_locked(struct drm_crtc *crtc,
 	return 0;
 }
 
-static void dce_v11_0_set_cursor(struct drm_crtc *crtc)
-{
-	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	struct amdgpu_device *adev = crtc->dev->dev_private;
-
-	WREG32(mmCUR_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset,
-	       upper_32_bits(amdgpu_crtc->cursor_addr));
-	WREG32(mmCUR_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset,
-	       lower_32_bits(amdgpu_crtc->cursor_addr));
-}
-
 static int dce_v11_0_crtc_cursor_move(struct drm_crtc *crtc,
 				      int x, int y)
 {
@@ -2602,7 +2596,6 @@ static int dce_v11_0_crtc_cursor_set2(struct drm_crtc *crtc,
 		amdgpu_crtc->cursor_hot_y = hot_y;
 	}
 
-	dce_v11_0_set_cursor(crtc);
 	dce_v11_0_show_cursor(crtc);
 	dce_v11_0_lock_cursor(crtc, false);
 
@@ -2631,7 +2624,6 @@ static void dce_v11_0_cursor_reset(struct drm_crtc *crtc)
 		dce_v11_0_cursor_move_locked(crtc, amdgpu_crtc->cursor_x,
 					     amdgpu_crtc->cursor_y);
 
-		dce_v11_0_set_cursor(crtc);
 		dce_v11_0_show_cursor(crtc);
 
 		dce_v11_0_lock_cursor(crtc, false);
-- 
cgit v0.10.2


From 77ed35b889e38775649ed985048cac8c134cd3a3 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Wed, 7 Oct 2015 23:26:15 -0400
Subject: drm/amdgpu/dce8: Use cursor_set2 hook for enabling / disabling the HW
 cursor
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The cursor_set2 hook provides the cursor hotspot position within the
cursor image. When the hotspot position changes, we can adjust the cursor
position such that the hotspot doesn't move on the screen. This prevents
the cursor from appearing to intermittently jump around on the screen
when the position of the hotspot within the cursor image changes.

Port of radeon commits:
78b1a6010b46a69bcd47b723a80f92693f26d17b
3feba08d79c32777a845c3c8a4ab93092bdf6f19

Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
index c86911c..aaa6561 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
@@ -2417,20 +2417,8 @@ static void dce_v8_0_show_cursor(struct drm_crtc *crtc)
 		   (CURSOR_URGENT_1_2 << CUR_CONTROL__CURSOR_URGENT_CONTROL__SHIFT));
 }
 
-static void dce_v8_0_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *obj,
-			      uint64_t gpu_addr)
-{
-	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	struct amdgpu_device *adev = crtc->dev->dev_private;
-
-	WREG32(mmCUR_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset,
-	       upper_32_bits(gpu_addr));
-	WREG32(mmCUR_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset,
-	       gpu_addr & 0xffffffff);
-}
-
-static int dce_v8_0_crtc_cursor_move(struct drm_crtc *crtc,
-				     int x, int y)
+static int dce_v8_0_cursor_move_locked(struct drm_crtc *crtc,
+				       int x, int y)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct amdgpu_device *adev = crtc->dev->dev_private;
@@ -2450,21 +2438,61 @@ static int dce_v8_0_crtc_cursor_move(struct drm_crtc *crtc,
 		y = 0;
 	}
 
-	dce_v8_0_lock_cursor(crtc, true);
 	WREG32(mmCUR_POSITION + amdgpu_crtc->crtc_offset, (x << 16) | y);
 	WREG32(mmCUR_HOT_SPOT + amdgpu_crtc->crtc_offset, (xorigin << 16) | yorigin);
 	WREG32(mmCUR_SIZE + amdgpu_crtc->crtc_offset,
 	       ((amdgpu_crtc->cursor_width - 1) << 16) | (amdgpu_crtc->cursor_height - 1));
-	dce_v8_0_lock_cursor(crtc, false);
+
+	amdgpu_crtc->cursor_x = x;
+	amdgpu_crtc->cursor_y = y;
 
 	return 0;
 }
 
-static int dce_v8_0_crtc_cursor_set(struct drm_crtc *crtc,
-				    struct drm_file *file_priv,
-				    uint32_t handle,
-				    uint32_t width,
-				    uint32_t height)
+static void dce_v8_0_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *obj,
+				 uint64_t gpu_addr, int hot_x, int hot_y)
+{
+	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
+	struct amdgpu_device *adev = crtc->dev->dev_private;
+
+	WREG32(mmCUR_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset,
+	       upper_32_bits(gpu_addr));
+	WREG32(mmCUR_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset,
+	       lower_32_bits(gpu_addr));
+
+	if (hot_x != amdgpu_crtc->cursor_hot_x ||
+	    hot_y != amdgpu_crtc->cursor_hot_y) {
+		int x, y;
+
+		x = amdgpu_crtc->cursor_x + amdgpu_crtc->cursor_hot_x - hot_x;
+		y = amdgpu_crtc->cursor_y + amdgpu_crtc->cursor_hot_y - hot_y;
+
+		dce_v8_0_cursor_move_locked(crtc, x, y);
+
+		amdgpu_crtc->cursor_hot_x = hot_x;
+		amdgpu_crtc->cursor_hot_y = hot_y;
+	}
+}
+
+static int dce_v8_0_crtc_cursor_move(struct drm_crtc *crtc,
+				     int x, int y)
+{
+	int ret;
+
+	dce_v8_0_lock_cursor(crtc, true);
+	ret = dce_v8_0_cursor_move_locked(crtc, x, y);
+	dce_v8_0_lock_cursor(crtc, false);
+
+	return ret;
+}
+
+static int dce_v8_0_crtc_cursor_set2(struct drm_crtc *crtc,
+				     struct drm_file *file_priv,
+				     uint32_t handle,
+				     uint32_t width,
+				     uint32_t height,
+				     int32_t hot_x,
+				     int32_t hot_y)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_gem_object *obj;
@@ -2505,7 +2533,7 @@ static int dce_v8_0_crtc_cursor_set(struct drm_crtc *crtc,
 	amdgpu_crtc->cursor_height = height;
 
 	dce_v8_0_lock_cursor(crtc, true);
-	dce_v8_0_set_cursor(crtc, obj, gpu_addr);
+	dce_v8_0_set_cursor(crtc, obj, gpu_addr, hot_x, hot_y);
 	dce_v8_0_show_cursor(crtc);
 	dce_v8_0_lock_cursor(crtc, false);
 
@@ -2553,7 +2581,7 @@ static void dce_v8_0_crtc_destroy(struct drm_crtc *crtc)
 }
 
 static const struct drm_crtc_funcs dce_v8_0_crtc_funcs = {
-	.cursor_set = dce_v8_0_crtc_cursor_set,
+	.cursor_set2 = dce_v8_0_crtc_cursor_set2,
 	.cursor_move = dce_v8_0_crtc_cursor_move,
 	.gamma_set = dce_v8_0_crtc_gamma_set,
 	.set_config = amdgpu_crtc_set_config,
-- 
cgit v0.10.2


From fd70cf63e24a0cf3105853e3b59eaa52d8b9df17 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Wed, 7 Oct 2015 23:34:57 -0400
Subject: drm/amdgpu/dce8: Re-show the cursor after a modeset (v2)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Setting a mode seems to clear the cursor registers, so we need to
re-program them to make sure the cursor is visible.

Port of radeon commit:
6d3759fac636028849f3bbec80c4b77e9bfdb1d2

v2: change radeon reference in error message

Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
index aaa6561..d7f88f3 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
@@ -2449,11 +2449,23 @@ static int dce_v8_0_cursor_move_locked(struct drm_crtc *crtc,
 	return 0;
 }
 
-static void dce_v8_0_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *obj,
-				 uint64_t gpu_addr, int hot_x, int hot_y)
+static int dce_v8_0_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *obj,
+			       int hot_x, int hot_y)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct amdgpu_device *adev = crtc->dev->dev_private;
+	struct amdgpu_bo *aobj = gem_to_amdgpu_bo(obj);
+	uint64_t gpu_addr;
+	int ret;
+
+	ret = amdgpu_bo_reserve(aobj, false);
+	if (unlikely(ret != 0))
+		goto fail;
+
+	ret = amdgpu_bo_pin(aobj, AMDGPU_GEM_DOMAIN_VRAM, &gpu_addr);
+	amdgpu_bo_unreserve(aobj);
+	if (ret)
+		goto fail;
 
 	WREG32(mmCUR_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset,
 	       upper_32_bits(gpu_addr));
@@ -2472,6 +2484,13 @@ static void dce_v8_0_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *ob
 		amdgpu_crtc->cursor_hot_x = hot_x;
 		amdgpu_crtc->cursor_hot_y = hot_y;
 	}
+
+	return 0;
+
+fail:
+	drm_gem_object_unreference_unlocked(obj);
+
+	return ret;
 }
 
 static int dce_v8_0_crtc_cursor_move(struct drm_crtc *crtc,
@@ -2496,8 +2515,6 @@ static int dce_v8_0_crtc_cursor_set2(struct drm_crtc *crtc,
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_gem_object *obj;
-	struct amdgpu_bo *robj;
-	uint64_t gpu_addr;
 	int ret;
 
 	if (!handle) {
@@ -2519,41 +2536,56 @@ static int dce_v8_0_crtc_cursor_set2(struct drm_crtc *crtc,
 		return -ENOENT;
 	}
 
-	robj = gem_to_amdgpu_bo(obj);
-	ret = amdgpu_bo_reserve(robj, false);
-	if (unlikely(ret != 0))
-		goto fail;
-	ret = amdgpu_bo_pin_restricted(robj, AMDGPU_GEM_DOMAIN_VRAM,
-				       0, 0, &gpu_addr);
-	amdgpu_bo_unreserve(robj);
-	if (ret)
-		goto fail;
-
 	amdgpu_crtc->cursor_width = width;
 	amdgpu_crtc->cursor_height = height;
 
 	dce_v8_0_lock_cursor(crtc, true);
-	dce_v8_0_set_cursor(crtc, obj, gpu_addr, hot_x, hot_y);
-	dce_v8_0_show_cursor(crtc);
+	ret = dce_v8_0_set_cursor(crtc, obj, hot_x, hot_y);
+	if (ret)
+		DRM_ERROR("dce_v8_0_set_cursor returned %d, not changing cursor\n",
+			  ret);
+	else
+		dce_v8_0_show_cursor(crtc);
 	dce_v8_0_lock_cursor(crtc, false);
 
 unpin:
 	if (amdgpu_crtc->cursor_bo) {
-		robj = gem_to_amdgpu_bo(amdgpu_crtc->cursor_bo);
-		ret = amdgpu_bo_reserve(robj, false);
+		struct amdgpu_bo *aobj = gem_to_amdgpu_bo(amdgpu_crtc->cursor_bo);
+		ret = amdgpu_bo_reserve(aobj, false);
 		if (likely(ret == 0)) {
-			amdgpu_bo_unpin(robj);
-			amdgpu_bo_unreserve(robj);
+			amdgpu_bo_unpin(aobj);
+			amdgpu_bo_unreserve(aobj);
 		}
-		drm_gem_object_unreference_unlocked(amdgpu_crtc->cursor_bo);
+		if (amdgpu_crtc->cursor_bo != obj)
+			drm_gem_object_unreference_unlocked(amdgpu_crtc->cursor_bo);
 	}
 
 	amdgpu_crtc->cursor_bo = obj;
 	return 0;
-fail:
-	drm_gem_object_unreference_unlocked(obj);
+}
 
-	return ret;
+static void dce_v8_0_cursor_reset(struct drm_crtc *crtc)
+{
+	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
+	int ret;
+
+	if (amdgpu_crtc->cursor_bo) {
+		dce_v8_0_lock_cursor(crtc, true);
+
+		dce_v8_0_cursor_move_locked(crtc, amdgpu_crtc->cursor_x,
+					    amdgpu_crtc->cursor_y);
+
+		ret = dce_v8_0_set_cursor(crtc, amdgpu_crtc->cursor_bo,
+					  amdgpu_crtc->cursor_hot_x,
+					  amdgpu_crtc->cursor_hot_y);
+		if (ret)
+			DRM_ERROR("dce_v8_0_set_cursor returned %d, not showing "
+				  "cursor\n", ret);
+		else
+			dce_v8_0_show_cursor(crtc);
+
+		dce_v8_0_lock_cursor(crtc, false);
+	}
 }
 
 static void dce_v8_0_crtc_gamma_set(struct drm_crtc *crtc, u16 *red, u16 *green,
@@ -2721,6 +2753,7 @@ static int dce_v8_0_crtc_mode_set(struct drm_crtc *crtc,
 	dce_v8_0_crtc_do_set_base(crtc, old_fb, x, y, 0);
 	amdgpu_atombios_crtc_overscan_setup(crtc, mode, adjusted_mode);
 	amdgpu_atombios_crtc_scaler_setup(crtc);
+	dce_v8_0_cursor_reset(crtc);
 	/* update the hw version fpr dpm */
 	amdgpu_crtc->hw_mode = *adjusted_mode;
 
-- 
cgit v0.10.2


From c4e0dfadb2d4c054ca06641da77294fd5c61725a Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Wed, 7 Oct 2015 23:38:03 -0400
Subject: drm/amdgpu/dce8: Move hotspot handling out of set_cursor
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

It's only needed in dce8 crtc_cursor_set2.

Port of radeon commit:
2e007e611bc0a46dfed1547e0228a57f800e703d

Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
index d7f88f3..aac9980 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
@@ -2449,8 +2449,7 @@ static int dce_v8_0_cursor_move_locked(struct drm_crtc *crtc,
 	return 0;
 }
 
-static int dce_v8_0_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *obj,
-			       int hot_x, int hot_y)
+static int dce_v8_0_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *obj)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct amdgpu_device *adev = crtc->dev->dev_private;
@@ -2472,19 +2471,6 @@ static int dce_v8_0_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *obj
 	WREG32(mmCUR_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset,
 	       lower_32_bits(gpu_addr));
 
-	if (hot_x != amdgpu_crtc->cursor_hot_x ||
-	    hot_y != amdgpu_crtc->cursor_hot_y) {
-		int x, y;
-
-		x = amdgpu_crtc->cursor_x + amdgpu_crtc->cursor_hot_x - hot_x;
-		y = amdgpu_crtc->cursor_y + amdgpu_crtc->cursor_hot_y - hot_y;
-
-		dce_v8_0_cursor_move_locked(crtc, x, y);
-
-		amdgpu_crtc->cursor_hot_x = hot_x;
-		amdgpu_crtc->cursor_hot_y = hot_y;
-	}
-
 	return 0;
 
 fail:
@@ -2540,7 +2526,21 @@ static int dce_v8_0_crtc_cursor_set2(struct drm_crtc *crtc,
 	amdgpu_crtc->cursor_height = height;
 
 	dce_v8_0_lock_cursor(crtc, true);
-	ret = dce_v8_0_set_cursor(crtc, obj, hot_x, hot_y);
+
+	if (hot_x != amdgpu_crtc->cursor_hot_x ||
+	    hot_y != amdgpu_crtc->cursor_hot_y) {
+		int x, y;
+
+		x = amdgpu_crtc->cursor_x + amdgpu_crtc->cursor_hot_x - hot_x;
+		y = amdgpu_crtc->cursor_y + amdgpu_crtc->cursor_hot_y - hot_y;
+
+		dce_v8_0_cursor_move_locked(crtc, x, y);
+
+		amdgpu_crtc->cursor_hot_x = hot_x;
+		amdgpu_crtc->cursor_hot_y = hot_y;
+	}
+
+	ret = dce_v8_0_set_cursor(crtc, obj);
 	if (ret)
 		DRM_ERROR("dce_v8_0_set_cursor returned %d, not changing cursor\n",
 			  ret);
@@ -2575,9 +2575,7 @@ static void dce_v8_0_cursor_reset(struct drm_crtc *crtc)
 		dce_v8_0_cursor_move_locked(crtc, amdgpu_crtc->cursor_x,
 					    amdgpu_crtc->cursor_y);
 
-		ret = dce_v8_0_set_cursor(crtc, amdgpu_crtc->cursor_bo,
-					  amdgpu_crtc->cursor_hot_x,
-					  amdgpu_crtc->cursor_hot_y);
+		ret = dce_v8_0_set_cursor(crtc, amdgpu_crtc->cursor_bo);
 		if (ret)
 			DRM_ERROR("dce_v8_0_set_cursor returned %d, not showing "
 				  "cursor\n", ret);
-- 
cgit v0.10.2


From 72b400675a900e80a75cad57db6aa726c68ffb3f Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Thu, 8 Oct 2015 11:31:58 -0400
Subject: drm/amdgpu/dce8: Clean up reference counting and pinning of the
 cursor BOs
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Take a GEM reference for and pin the new cursor BO, unpin and drop the
GEM reference for the old cursor BO in dce8 crtc_cursor_set2, and use
amdgpu_crtc->cursor_addr in dce8 set_cursor.

This fixes dce8 cursor_reset accidentally incrementing the cursor BO
pin count, and cleans up the code a little.

Port of radeon commit:
cd404af0c930104462aa91344f07d002cf8248ed

Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
index aac9980..aebfe3c 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
@@ -2449,34 +2449,15 @@ static int dce_v8_0_cursor_move_locked(struct drm_crtc *crtc,
 	return 0;
 }
 
-static int dce_v8_0_set_cursor(struct drm_crtc *crtc, struct drm_gem_object *obj)
+static void dce_v8_0_set_cursor(struct drm_crtc *crtc)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct amdgpu_device *adev = crtc->dev->dev_private;
-	struct amdgpu_bo *aobj = gem_to_amdgpu_bo(obj);
-	uint64_t gpu_addr;
-	int ret;
-
-	ret = amdgpu_bo_reserve(aobj, false);
-	if (unlikely(ret != 0))
-		goto fail;
-
-	ret = amdgpu_bo_pin(aobj, AMDGPU_GEM_DOMAIN_VRAM, &gpu_addr);
-	amdgpu_bo_unreserve(aobj);
-	if (ret)
-		goto fail;
 
 	WREG32(mmCUR_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset,
-	       upper_32_bits(gpu_addr));
+	       upper_32_bits(amdgpu_crtc->cursor_addr));
 	WREG32(mmCUR_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset,
-	       lower_32_bits(gpu_addr));
-
-	return 0;
-
-fail:
-	drm_gem_object_unreference_unlocked(obj);
-
-	return ret;
+	       lower_32_bits(amdgpu_crtc->cursor_addr));
 }
 
 static int dce_v8_0_crtc_cursor_move(struct drm_crtc *crtc,
@@ -2501,6 +2482,7 @@ static int dce_v8_0_crtc_cursor_set2(struct drm_crtc *crtc,
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct drm_gem_object *obj;
+	struct amdgpu_bo *aobj;
 	int ret;
 
 	if (!handle) {
@@ -2522,6 +2504,21 @@ static int dce_v8_0_crtc_cursor_set2(struct drm_crtc *crtc,
 		return -ENOENT;
 	}
 
+	aobj = gem_to_amdgpu_bo(obj);
+	ret = amdgpu_bo_reserve(aobj, false);
+	if (ret != 0) {
+		drm_gem_object_unreference_unlocked(obj);
+		return ret;
+	}
+
+	ret = amdgpu_bo_pin(aobj, AMDGPU_GEM_DOMAIN_VRAM, &amdgpu_crtc->cursor_addr);
+	amdgpu_bo_unreserve(aobj);
+	if (ret) {
+		DRM_ERROR("Failed to pin new cursor BO (%d)\n", ret);
+		drm_gem_object_unreference_unlocked(obj);
+		return ret;
+	}
+
 	amdgpu_crtc->cursor_width = width;
 	amdgpu_crtc->cursor_height = height;
 
@@ -2540,12 +2537,8 @@ static int dce_v8_0_crtc_cursor_set2(struct drm_crtc *crtc,
 		amdgpu_crtc->cursor_hot_y = hot_y;
 	}
 
-	ret = dce_v8_0_set_cursor(crtc, obj);
-	if (ret)
-		DRM_ERROR("dce_v8_0_set_cursor returned %d, not changing cursor\n",
-			  ret);
-	else
-		dce_v8_0_show_cursor(crtc);
+	dce_v8_0_set_cursor(crtc);
+	dce_v8_0_show_cursor(crtc);
 	dce_v8_0_lock_cursor(crtc, false);
 
 unpin:
@@ -2556,8 +2549,7 @@ unpin:
 			amdgpu_bo_unpin(aobj);
 			amdgpu_bo_unreserve(aobj);
 		}
-		if (amdgpu_crtc->cursor_bo != obj)
-			drm_gem_object_unreference_unlocked(amdgpu_crtc->cursor_bo);
+		drm_gem_object_unreference_unlocked(amdgpu_crtc->cursor_bo);
 	}
 
 	amdgpu_crtc->cursor_bo = obj;
@@ -2567,7 +2559,6 @@ unpin:
 static void dce_v8_0_cursor_reset(struct drm_crtc *crtc)
 {
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	int ret;
 
 	if (amdgpu_crtc->cursor_bo) {
 		dce_v8_0_lock_cursor(crtc, true);
@@ -2575,12 +2566,8 @@ static void dce_v8_0_cursor_reset(struct drm_crtc *crtc)
 		dce_v8_0_cursor_move_locked(crtc, amdgpu_crtc->cursor_x,
 					    amdgpu_crtc->cursor_y);
 
-		ret = dce_v8_0_set_cursor(crtc, amdgpu_crtc->cursor_bo);
-		if (ret)
-			DRM_ERROR("dce_v8_0_set_cursor returned %d, not showing "
-				  "cursor\n", ret);
-		else
-			dce_v8_0_show_cursor(crtc);
+		dce_v8_0_set_cursor(crtc);
+		dce_v8_0_show_cursor(crtc);
 
 		dce_v8_0_lock_cursor(crtc, false);
 	}
-- 
cgit v0.10.2


From a2df42da61109653782949789ffe6c358e5d2685 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Wed, 7 Oct 2015 23:46:49 -0400
Subject: drm/amdgpu/dce8: Fold set_cursor() into show_cursor()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Port of radeon commit:
8991668ab4e26f985a8485719bce5d6d0623a644

Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
index aebfe3c..0d2c50d 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
@@ -2411,6 +2411,11 @@ static void dce_v8_0_show_cursor(struct drm_crtc *crtc)
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 	struct amdgpu_device *adev = crtc->dev->dev_private;
 
+	WREG32(mmCUR_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset,
+	       upper_32_bits(amdgpu_crtc->cursor_addr));
+	WREG32(mmCUR_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset,
+	       lower_32_bits(amdgpu_crtc->cursor_addr));
+
 	WREG32_IDX(mmCUR_CONTROL + amdgpu_crtc->crtc_offset,
 		   CUR_CONTROL__CURSOR_EN_MASK |
 		   (CURSOR_24_8_PRE_MULT << CUR_CONTROL__CURSOR_MODE__SHIFT) |
@@ -2449,17 +2454,6 @@ static int dce_v8_0_cursor_move_locked(struct drm_crtc *crtc,
 	return 0;
 }
 
-static void dce_v8_0_set_cursor(struct drm_crtc *crtc)
-{
-	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	struct amdgpu_device *adev = crtc->dev->dev_private;
-
-	WREG32(mmCUR_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset,
-	       upper_32_bits(amdgpu_crtc->cursor_addr));
-	WREG32(mmCUR_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset,
-	       lower_32_bits(amdgpu_crtc->cursor_addr));
-}
-
 static int dce_v8_0_crtc_cursor_move(struct drm_crtc *crtc,
 				     int x, int y)
 {
@@ -2537,7 +2531,6 @@ static int dce_v8_0_crtc_cursor_set2(struct drm_crtc *crtc,
 		amdgpu_crtc->cursor_hot_y = hot_y;
 	}
 
-	dce_v8_0_set_cursor(crtc);
 	dce_v8_0_show_cursor(crtc);
 	dce_v8_0_lock_cursor(crtc, false);
 
@@ -2566,7 +2559,6 @@ static void dce_v8_0_cursor_reset(struct drm_crtc *crtc)
 		dce_v8_0_cursor_move_locked(crtc, amdgpu_crtc->cursor_x,
 					    amdgpu_crtc->cursor_y);
 
-		dce_v8_0_set_cursor(crtc);
 		dce_v8_0_show_cursor(crtc);
 
 		dce_v8_0_lock_cursor(crtc, false);
-- 
cgit v0.10.2


From 756e6880c51376d2e4d53050582d66d88fc281c7 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Thu, 8 Oct 2015 00:03:36 -0400
Subject: drm/amdgpu: unpin cursor BOs on suspend and pin them again on resume
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Everything is evicted from VRAM before suspend, so we need to make
sure all BOs are unpinned and re-pinned after resume. Fixes broken
mouse cursor after resume introduced by commit b9729b17.

Port of radeon commit:
f3cbb17bcf676a2fc6aedebe9fbebd59e550c51a

Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 4256b00..901a460 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -1657,11 +1657,21 @@ int amdgpu_suspend_kms(struct drm_device *dev, bool suspend, bool fbcon)
 	}
 	drm_modeset_unlock_all(dev);
 
-	/* unpin the front buffers */
+	/* unpin the front buffers and cursors */
 	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+		struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
 		struct amdgpu_framebuffer *rfb = to_amdgpu_framebuffer(crtc->primary->fb);
 		struct amdgpu_bo *robj;
 
+		if (amdgpu_crtc->cursor_bo) {
+			struct amdgpu_bo *aobj = gem_to_amdgpu_bo(amdgpu_crtc->cursor_bo);
+			r = amdgpu_bo_reserve(aobj, false);
+			if (r == 0) {
+				amdgpu_bo_unpin(aobj);
+				amdgpu_bo_unreserve(aobj);
+			}
+		}
+
 		if (rfb == NULL || rfb->obj == NULL) {
 			continue;
 		}
@@ -1713,6 +1723,7 @@ int amdgpu_resume_kms(struct drm_device *dev, bool resume, bool fbcon)
 {
 	struct drm_connector *connector;
 	struct amdgpu_device *adev = dev->dev_private;
+	struct drm_crtc *crtc;
 	int r;
 
 	if (dev->switch_power_state == DRM_SWITCH_POWER_OFF)
@@ -1746,6 +1757,24 @@ int amdgpu_resume_kms(struct drm_device *dev, bool resume, bool fbcon)
 	if (r)
 		return r;
 
+	/* pin cursors */
+	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+		struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
+
+		if (amdgpu_crtc->cursor_bo) {
+			struct amdgpu_bo *aobj = gem_to_amdgpu_bo(amdgpu_crtc->cursor_bo);
+			r = amdgpu_bo_reserve(aobj, false);
+			if (r == 0) {
+				r = amdgpu_bo_pin(aobj,
+						  AMDGPU_GEM_DOMAIN_VRAM,
+						  &amdgpu_crtc->cursor_addr);
+				if (r != 0)
+					DRM_ERROR("Failed to pin cursor BO (%d)\n", r);
+				amdgpu_bo_unreserve(aobj);
+			}
+		}
+	}
+
 	/* blat the mode back in */
 	if (fbcon) {
 		drm_helper_resume_force_mode(dev);
-- 
cgit v0.10.2


From c113ea1c4f4a7592f17e53d658873b83af0dbb4b Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Thu, 8 Oct 2015 16:30:37 -0400
Subject: drm/amdgpu: rework sdma structures
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Rework the sdma structures in the driver to
consolidate all of the sdma info into a single
structure and allow for asics that may have
different numbers of sdma instances.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 6647fb2..afc9848 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -1708,7 +1708,7 @@ struct amdgpu_vce {
 /*
  * SDMA
  */
-struct amdgpu_sdma {
+struct amdgpu_sdma_instance {
 	/* SDMA firmware */
 	const struct firmware	*fw;
 	uint32_t		fw_version;
@@ -1718,6 +1718,13 @@ struct amdgpu_sdma {
 	bool			burst_nop;
 };
 
+struct amdgpu_sdma {
+	struct amdgpu_sdma_instance instance[AMDGPU_MAX_SDMA_INSTANCES];
+	struct amdgpu_irq_src	trap_irq;
+	struct amdgpu_irq_src	illegal_inst_irq;
+	int 			num_instances;
+};
+
 /*
  * Firmware
  */
@@ -2064,9 +2071,7 @@ struct amdgpu_device {
 	struct amdgpu_gfx		gfx;
 
 	/* sdma */
-	struct amdgpu_sdma		sdma[AMDGPU_MAX_SDMA_INSTANCES];
-	struct amdgpu_irq_src		sdma_trap_irq;
-	struct amdgpu_irq_src		sdma_illegal_inst_irq;
+	struct amdgpu_sdma		sdma;
 
 	/* uvd */
 	bool				has_uvd;
@@ -2203,17 +2208,18 @@ static inline void amdgpu_ring_write(struct amdgpu_ring *ring, uint32_t v)
 	ring->ring_free_dw--;
 }
 
-static inline struct amdgpu_sdma * amdgpu_get_sdma_instance(struct amdgpu_ring *ring)
+static inline struct amdgpu_sdma_instance *
+amdgpu_get_sdma_instance(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
 	int i;
 
-	for (i = 0; i < AMDGPU_MAX_SDMA_INSTANCES; i++)
-		if (&adev->sdma[i].ring == ring)
+	for (i = 0; i < adev->sdma.num_instances; i++)
+		if (&adev->sdma.instance[i].ring == ring)
 			break;
 
 	if (i < AMDGPU_MAX_SDMA_INSTANCES)
-		return &adev->sdma[i];
+		return &adev->sdma.instance[i];
 	else
 		return NULL;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c
index dd2037b..0e13763 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c
@@ -649,12 +649,12 @@ static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type)
 
 	case KGD_ENGINE_SDMA1:
 		hdr = (const union amdgpu_firmware_header *)
-							adev->sdma[0].fw->data;
+							adev->sdma.instance[0].fw->data;
 		break;
 
 	case KGD_ENGINE_SDMA2:
 		hdr = (const union amdgpu_firmware_header *)
-							adev->sdma[1].fw->data;
+							adev->sdma.instance[1].fw->data;
 		break;
 
 	default:
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c
index dfd1d50..79fa5c7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c
@@ -523,12 +523,12 @@ static uint16_t get_fw_version(struct kgd_dev *kgd, enum kgd_engine_type type)
 
 	case KGD_ENGINE_SDMA1:
 		hdr = (const union amdgpu_firmware_header *)
-							adev->sdma[0].fw->data;
+							adev->sdma.instance[0].fw->data;
 		break;
 
 	case KGD_ENGINE_SDMA2:
 		hdr = (const union amdgpu_firmware_header *)
-							adev->sdma[1].fw->data;
+							adev->sdma.instance[1].fw->data;
 		break;
 
 	default:
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index 749420f..29fc45c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -104,10 +104,11 @@ int amdgpu_cs_get_ring(struct amdgpu_device *adev, u32 ip_type,
 		}
 		break;
 	case AMDGPU_HW_IP_DMA:
-		if (ring < 2) {
-			*out_ring = &adev->sdma[ring].ring;
+		if (ring < adev->sdma.num_instances) {
+			*out_ring = &adev->sdma.instance[ring].ring;
 		} else {
-			DRM_ERROR("only two SDMA rings are supported\n");
+			DRM_ERROR("only %d SDMA rings are supported\n",
+				  adev->sdma.num_instances);
 			return -EINVAL;
 		}
 		break;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
index 7823322..3f5f2d5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
@@ -218,8 +218,8 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file
 			break;
 		case AMDGPU_HW_IP_DMA:
 			type = AMD_IP_BLOCK_TYPE_SDMA;
-			ring_mask = adev->sdma[0].ring.ready ? 1 : 0;
-			ring_mask |= ((adev->sdma[1].ring.ready ? 1 : 0) << 1);
+			for (i = 0; i < adev->sdma.num_instances; i++)
+				ring_mask |= ((adev->sdma.instance[i].ring.ready ? 1 : 0) << i);
 			ib_start_alignment = AMDGPU_GPU_PAGE_SIZE;
 			ib_size_alignment = 1;
 			break;
@@ -341,10 +341,10 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file
 			fw_info.feature = 0;
 			break;
 		case AMDGPU_INFO_FW_SDMA:
-			if (info->query_fw.index >= 2)
+			if (info->query_fw.index >= adev->sdma.num_instances)
 				return -EINVAL;
-			fw_info.ver = adev->sdma[info->query_fw.index].fw_version;
-			fw_info.feature = adev->sdma[info->query_fw.index].feature_version;
+			fw_info.ver = adev->sdma.instance[info->query_fw.index].fw_version;
+			fw_info.feature = adev->sdma.instance[info->query_fw.index].feature_version;
 			break;
 		default:
 			return -EINVAL;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
index 30dce23..b13a74b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
@@ -540,8 +540,8 @@ static int amdgpu_debugfs_ring_info(struct seq_file *m, void *data)
 static int amdgpu_gfx_index = offsetof(struct amdgpu_device, gfx.gfx_ring[0]);
 static int cayman_cp1_index = offsetof(struct amdgpu_device, gfx.compute_ring[0]);
 static int cayman_cp2_index = offsetof(struct amdgpu_device, gfx.compute_ring[1]);
-static int amdgpu_dma1_index = offsetof(struct amdgpu_device, sdma[0].ring);
-static int amdgpu_dma2_index = offsetof(struct amdgpu_device, sdma[1].ring);
+static int amdgpu_dma1_index = offsetof(struct amdgpu_device, sdma.instance[0].ring);
+static int amdgpu_dma2_index = offsetof(struct amdgpu_device, sdma.instance[1].ring);
 static int r600_uvd_index = offsetof(struct amdgpu_device, uvd.ring);
 static int si_vce1_index = offsetof(struct amdgpu_device, vce.ring[0]);
 static int si_vce2_index = offsetof(struct amdgpu_device, vce.ring[1]);
diff --git a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
index 9ea9de4..814598e 100644
--- a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
+++ b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
@@ -96,7 +96,7 @@ static int cik_sdma_init_microcode(struct amdgpu_device *adev)
 {
 	const char *chip_name;
 	char fw_name[30];
-	int err, i;
+	int err = 0, i;
 
 	DRM_DEBUG("\n");
 
@@ -119,24 +119,24 @@ static int cik_sdma_init_microcode(struct amdgpu_device *adev)
 	default: BUG();
 	}
 
-	for (i = 0; i < SDMA_MAX_INSTANCE; i++) {
+	for (i = 0; i < adev->sdma.num_instances; i++) {
 		if (i == 0)
 			snprintf(fw_name, sizeof(fw_name), "radeon/%s_sdma.bin", chip_name);
 		else
 			snprintf(fw_name, sizeof(fw_name), "radeon/%s_sdma1.bin", chip_name);
-		err = request_firmware(&adev->sdma[i].fw, fw_name, adev->dev);
+		err = request_firmware(&adev->sdma.instance[i].fw, fw_name, adev->dev);
 		if (err)
 			goto out;
-		err = amdgpu_ucode_validate(adev->sdma[i].fw);
+		err = amdgpu_ucode_validate(adev->sdma.instance[i].fw);
 	}
 out:
 	if (err) {
 		printk(KERN_ERR
 		       "cik_sdma: Failed to load firmware \"%s\"\n",
 		       fw_name);
-		for (i = 0; i < SDMA_MAX_INSTANCE; i++) {
-			release_firmware(adev->sdma[i].fw);
-			adev->sdma[i].fw = NULL;
+		for (i = 0; i < adev->sdma.num_instances; i++) {
+			release_firmware(adev->sdma.instance[i].fw);
+			adev->sdma.instance[i].fw = NULL;
 		}
 	}
 	return err;
@@ -168,7 +168,7 @@ static uint32_t cik_sdma_ring_get_rptr(struct amdgpu_ring *ring)
 static uint32_t cik_sdma_ring_get_wptr(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
-	u32 me = (ring == &adev->sdma[0].ring) ? 0 : 1;
+	u32 me = (ring == &adev->sdma.instance[0].ring) ? 0 : 1;
 
 	return (RREG32(mmSDMA0_GFX_RB_WPTR + sdma_offsets[me]) & 0x3fffc) >> 2;
 }
@@ -183,14 +183,14 @@ static uint32_t cik_sdma_ring_get_wptr(struct amdgpu_ring *ring)
 static void cik_sdma_ring_set_wptr(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
-	u32 me = (ring == &adev->sdma[0].ring) ? 0 : 1;
+	u32 me = (ring == &adev->sdma.instance[0].ring) ? 0 : 1;
 
 	WREG32(mmSDMA0_GFX_RB_WPTR + sdma_offsets[me], (ring->wptr << 2) & 0x3fffc);
 }
 
 static void cik_sdma_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count)
 {
-	struct amdgpu_sdma *sdma = amdgpu_get_sdma_instance(ring);
+	struct amdgpu_sdma_instance *sdma = amdgpu_get_sdma_instance(ring);
 	int i;
 
 	for (i = 0; i < count; i++)
@@ -248,7 +248,7 @@ static void cik_sdma_ring_emit_hdp_flush(struct amdgpu_ring *ring)
 			  SDMA_POLL_REG_MEM_EXTRA_FUNC(3)); /* == */
 	u32 ref_and_mask;
 
-	if (ring == &ring->adev->sdma[0].ring)
+	if (ring == &ring->adev->sdma.instance[0].ring)
 		ref_and_mask = GPU_HDP_FLUSH_DONE__SDMA0_MASK;
 	else
 		ref_and_mask = GPU_HDP_FLUSH_DONE__SDMA1_MASK;
@@ -327,8 +327,8 @@ static bool cik_sdma_ring_emit_semaphore(struct amdgpu_ring *ring,
  */
 static void cik_sdma_gfx_stop(struct amdgpu_device *adev)
 {
-	struct amdgpu_ring *sdma0 = &adev->sdma[0].ring;
-	struct amdgpu_ring *sdma1 = &adev->sdma[1].ring;
+	struct amdgpu_ring *sdma0 = &adev->sdma.instance[0].ring;
+	struct amdgpu_ring *sdma1 = &adev->sdma.instance[1].ring;
 	u32 rb_cntl;
 	int i;
 
@@ -336,7 +336,7 @@ static void cik_sdma_gfx_stop(struct amdgpu_device *adev)
 	    (adev->mman.buffer_funcs_ring == sdma1))
 		amdgpu_ttm_set_active_vram_size(adev, adev->mc.visible_vram_size);
 
-	for (i = 0; i < SDMA_MAX_INSTANCE; i++) {
+	for (i = 0; i < adev->sdma.num_instances; i++) {
 		rb_cntl = RREG32(mmSDMA0_GFX_RB_CNTL + sdma_offsets[i]);
 		rb_cntl &= ~SDMA0_GFX_RB_CNTL__RB_ENABLE_MASK;
 		WREG32(mmSDMA0_GFX_RB_CNTL + sdma_offsets[i], rb_cntl);
@@ -376,7 +376,7 @@ static void cik_sdma_enable(struct amdgpu_device *adev, bool enable)
 		cik_sdma_rlc_stop(adev);
 	}
 
-	for (i = 0; i < SDMA_MAX_INSTANCE; i++) {
+	for (i = 0; i < adev->sdma.num_instances; i++) {
 		me_cntl = RREG32(mmSDMA0_F32_CNTL + sdma_offsets[i]);
 		if (enable)
 			me_cntl &= ~SDMA0_F32_CNTL__HALT_MASK;
@@ -402,8 +402,8 @@ static int cik_sdma_gfx_resume(struct amdgpu_device *adev)
 	u32 wb_offset;
 	int i, j, r;
 
-	for (i = 0; i < SDMA_MAX_INSTANCE; i++) {
-		ring = &adev->sdma[i].ring;
+	for (i = 0; i < adev->sdma.num_instances; i++) {
+		ring = &adev->sdma.instance[i].ring;
 		wb_offset = (ring->rptr_offs * 4);
 
 		mutex_lock(&adev->srbm_mutex);
@@ -502,26 +502,25 @@ static int cik_sdma_load_microcode(struct amdgpu_device *adev)
 	u32 fw_size;
 	int i, j;
 
-	if (!adev->sdma[0].fw || !adev->sdma[1].fw)
-		return -EINVAL;
-
 	/* halt the MEs */
 	cik_sdma_enable(adev, false);
 
-	for (i = 0; i < SDMA_MAX_INSTANCE; i++) {
-		hdr = (const struct sdma_firmware_header_v1_0 *)adev->sdma[i].fw->data;
+	for (i = 0; i < adev->sdma.num_instances; i++) {
+		if (!adev->sdma.instance[i].fw)
+			return -EINVAL;
+		hdr = (const struct sdma_firmware_header_v1_0 *)adev->sdma.instance[i].fw->data;
 		amdgpu_ucode_print_sdma_hdr(&hdr->header);
 		fw_size = le32_to_cpu(hdr->header.ucode_size_bytes) / 4;
-		adev->sdma[i].fw_version = le32_to_cpu(hdr->header.ucode_version);
-		adev->sdma[i].feature_version = le32_to_cpu(hdr->ucode_feature_version);
-		if (adev->sdma[i].feature_version >= 20)
-			adev->sdma[i].burst_nop = true;
+		adev->sdma.instance[i].fw_version = le32_to_cpu(hdr->header.ucode_version);
+		adev->sdma.instance[i].feature_version = le32_to_cpu(hdr->ucode_feature_version);
+		if (adev->sdma.instance[i].feature_version >= 20)
+			adev->sdma.instance[i].burst_nop = true;
 		fw_data = (const __le32 *)
-			(adev->sdma[i].fw->data + le32_to_cpu(hdr->header.ucode_array_offset_bytes));
+			(adev->sdma.instance[i].fw->data + le32_to_cpu(hdr->header.ucode_array_offset_bytes));
 		WREG32(mmSDMA0_UCODE_ADDR + sdma_offsets[i], 0);
 		for (j = 0; j < fw_size; j++)
 			WREG32(mmSDMA0_UCODE_DATA + sdma_offsets[i], le32_to_cpup(fw_data++));
-		WREG32(mmSDMA0_UCODE_ADDR + sdma_offsets[i], adev->sdma[i].fw_version);
+		WREG32(mmSDMA0_UCODE_ADDR + sdma_offsets[i], adev->sdma.instance[i].fw_version);
 	}
 
 	return 0;
@@ -830,7 +829,7 @@ static void cik_sdma_vm_set_pte_pde(struct amdgpu_ib *ib,
  */
 static void cik_sdma_vm_pad_ib(struct amdgpu_ib *ib)
 {
-	struct amdgpu_sdma *sdma = amdgpu_get_sdma_instance(ib->ring);
+	struct amdgpu_sdma_instance *sdma = amdgpu_get_sdma_instance(ib->ring);
 	u32 pad_count;
 	int i;
 
@@ -934,6 +933,8 @@ static int cik_sdma_early_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
+	adev->sdma.num_instances = SDMA_MAX_INSTANCE;
+
 	cik_sdma_set_ring_funcs(adev);
 	cik_sdma_set_irq_funcs(adev);
 	cik_sdma_set_buffer_funcs(adev);
@@ -946,7 +947,7 @@ static int cik_sdma_sw_init(void *handle)
 {
 	struct amdgpu_ring *ring;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	int r;
+	int r, i;
 
 	r = cik_sdma_init_microcode(adev);
 	if (r) {
@@ -955,43 +956,33 @@ static int cik_sdma_sw_init(void *handle)
 	}
 
 	/* SDMA trap event */
-	r = amdgpu_irq_add_id(adev, 224, &adev->sdma_trap_irq);
+	r = amdgpu_irq_add_id(adev, 224, &adev->sdma.trap_irq);
 	if (r)
 		return r;
 
 	/* SDMA Privileged inst */
-	r = amdgpu_irq_add_id(adev, 241, &adev->sdma_illegal_inst_irq);
+	r = amdgpu_irq_add_id(adev, 241, &adev->sdma.illegal_inst_irq);
 	if (r)
 		return r;
 
 	/* SDMA Privileged inst */
-	r = amdgpu_irq_add_id(adev, 247, &adev->sdma_illegal_inst_irq);
-	if (r)
-		return r;
-
-	ring = &adev->sdma[0].ring;
-	ring->ring_obj = NULL;
-
-	ring = &adev->sdma[1].ring;
-	ring->ring_obj = NULL;
-
-	ring = &adev->sdma[0].ring;
-	sprintf(ring->name, "sdma0");
-	r = amdgpu_ring_init(adev, ring, 256 * 1024,
-			     SDMA_PACKET(SDMA_OPCODE_NOP, 0, 0), 0xf,
-			     &adev->sdma_trap_irq, AMDGPU_SDMA_IRQ_TRAP0,
-			     AMDGPU_RING_TYPE_SDMA);
+	r = amdgpu_irq_add_id(adev, 247, &adev->sdma.illegal_inst_irq);
 	if (r)
 		return r;
 
-	ring = &adev->sdma[1].ring;
-	sprintf(ring->name, "sdma1");
-	r = amdgpu_ring_init(adev, ring, 256 * 1024,
-			     SDMA_PACKET(SDMA_OPCODE_NOP, 0, 0), 0xf,
-			     &adev->sdma_trap_irq, AMDGPU_SDMA_IRQ_TRAP1,
-			     AMDGPU_RING_TYPE_SDMA);
-	if (r)
-		return r;
+	for (i = 0; i < adev->sdma.num_instances; i++) {
+		ring = &adev->sdma.instance[i].ring;
+		ring->ring_obj = NULL;
+		sprintf(ring->name, "sdma%d", i);
+		r = amdgpu_ring_init(adev, ring, 256 * 1024,
+				     SDMA_PACKET(SDMA_OPCODE_NOP, 0, 0), 0xf,
+				     &adev->sdma.trap_irq,
+				     (i == 0) ?
+				     AMDGPU_SDMA_IRQ_TRAP0 : AMDGPU_SDMA_IRQ_TRAP1,
+				     AMDGPU_RING_TYPE_SDMA);
+		if (r)
+			return r;
+	}
 
 	return r;
 }
@@ -999,9 +990,10 @@ static int cik_sdma_sw_init(void *handle)
 static int cik_sdma_sw_fini(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	int i;
 
-	amdgpu_ring_fini(&adev->sdma[0].ring);
-	amdgpu_ring_fini(&adev->sdma[1].ring);
+	for (i = 0; i < adev->sdma.num_instances; i++)
+		amdgpu_ring_fini(&adev->sdma.instance[i].ring);
 
 	return 0;
 }
@@ -1078,7 +1070,7 @@ static void cik_sdma_print_status(void *handle)
 	dev_info(adev->dev, "CIK SDMA registers\n");
 	dev_info(adev->dev, "  SRBM_STATUS2=0x%08X\n",
 		 RREG32(mmSRBM_STATUS2));
-	for (i = 0; i < SDMA_MAX_INSTANCE; i++) {
+	for (i = 0; i < adev->sdma.num_instances; i++) {
 		dev_info(adev->dev, "  SDMA%d_STATUS_REG=0x%08X\n",
 			 i, RREG32(mmSDMA0_STATUS_REG + sdma_offsets[i]));
 		dev_info(adev->dev, "  SDMA%d_ME_CNTL=0x%08X\n",
@@ -1223,7 +1215,7 @@ static int cik_sdma_process_trap_irq(struct amdgpu_device *adev,
 	case 0:
 		switch (queue_id) {
 		case 0:
-			amdgpu_fence_process(&adev->sdma[0].ring);
+			amdgpu_fence_process(&adev->sdma.instance[0].ring);
 			break;
 		case 1:
 			/* XXX compute */
@@ -1236,7 +1228,7 @@ static int cik_sdma_process_trap_irq(struct amdgpu_device *adev,
 	case 1:
 		switch (queue_id) {
 		case 0:
-			amdgpu_fence_process(&adev->sdma[1].ring);
+			amdgpu_fence_process(&adev->sdma.instance[1].ring);
 			break;
 		case 1:
 			/* XXX compute */
@@ -1334,8 +1326,10 @@ static const struct amdgpu_ring_funcs cik_sdma_ring_funcs = {
 
 static void cik_sdma_set_ring_funcs(struct amdgpu_device *adev)
 {
-	adev->sdma[0].ring.funcs = &cik_sdma_ring_funcs;
-	adev->sdma[1].ring.funcs = &cik_sdma_ring_funcs;
+	int i;
+
+	for (i = 0; i < adev->sdma.num_instances; i++)
+		adev->sdma.instance[i].ring.funcs = &cik_sdma_ring_funcs;
 }
 
 static const struct amdgpu_irq_src_funcs cik_sdma_trap_irq_funcs = {
@@ -1349,9 +1343,9 @@ static const struct amdgpu_irq_src_funcs cik_sdma_illegal_inst_irq_funcs = {
 
 static void cik_sdma_set_irq_funcs(struct amdgpu_device *adev)
 {
-	adev->sdma_trap_irq.num_types = AMDGPU_SDMA_IRQ_LAST;
-	adev->sdma_trap_irq.funcs = &cik_sdma_trap_irq_funcs;
-	adev->sdma_illegal_inst_irq.funcs = &cik_sdma_illegal_inst_irq_funcs;
+	adev->sdma.trap_irq.num_types = AMDGPU_SDMA_IRQ_LAST;
+	adev->sdma.trap_irq.funcs = &cik_sdma_trap_irq_funcs;
+	adev->sdma.illegal_inst_irq.funcs = &cik_sdma_illegal_inst_irq_funcs;
 }
 
 /**
@@ -1416,7 +1410,7 @@ static void cik_sdma_set_buffer_funcs(struct amdgpu_device *adev)
 {
 	if (adev->mman.buffer_funcs == NULL) {
 		adev->mman.buffer_funcs = &cik_sdma_buffer_funcs;
-		adev->mman.buffer_funcs_ring = &adev->sdma[0].ring;
+		adev->mman.buffer_funcs_ring = &adev->sdma.instance[0].ring;
 	}
 }
 
@@ -1431,7 +1425,7 @@ static void cik_sdma_set_vm_pte_funcs(struct amdgpu_device *adev)
 {
 	if (adev->vm_manager.vm_pte_funcs == NULL) {
 		adev->vm_manager.vm_pte_funcs = &cik_sdma_vm_pte_funcs;
-		adev->vm_manager.vm_pte_funcs_ring = &adev->sdma[0].ring;
+		adev->vm_manager.vm_pte_funcs_ring = &adev->sdma.instance[0].ring;
 		adev->vm_manager.vm_pte_funcs_ring->is_pte_ring = true;
 	}
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c
index 14e8723..f8b868c 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c
@@ -118,7 +118,7 @@ static int sdma_v2_4_init_microcode(struct amdgpu_device *adev)
 {
 	const char *chip_name;
 	char fw_name[30];
-	int err, i;
+	int err = 0, i;
 	struct amdgpu_firmware_info *info = NULL;
 	const struct common_firmware_header *header = NULL;
 	const struct sdma_firmware_header_v1_0 *hdr;
@@ -132,27 +132,27 @@ static int sdma_v2_4_init_microcode(struct amdgpu_device *adev)
 	default: BUG();
 	}
 
-	for (i = 0; i < SDMA_MAX_INSTANCE; i++) {
+	for (i = 0; i < adev->sdma.num_instances; i++) {
 		if (i == 0)
 			snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_sdma.bin", chip_name);
 		else
 			snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_sdma1.bin", chip_name);
-		err = request_firmware(&adev->sdma[i].fw, fw_name, adev->dev);
+		err = request_firmware(&adev->sdma.instance[i].fw, fw_name, adev->dev);
 		if (err)
 			goto out;
-		err = amdgpu_ucode_validate(adev->sdma[i].fw);
+		err = amdgpu_ucode_validate(adev->sdma.instance[i].fw);
 		if (err)
 			goto out;
-		hdr = (const struct sdma_firmware_header_v1_0 *)adev->sdma[i].fw->data;
-		adev->sdma[i].fw_version = le32_to_cpu(hdr->header.ucode_version);
-		adev->sdma[i].feature_version = le32_to_cpu(hdr->ucode_feature_version);
-		if (adev->sdma[i].feature_version >= 20)
-			adev->sdma[i].burst_nop = true;
+		hdr = (const struct sdma_firmware_header_v1_0 *)adev->sdma.instance[i].fw->data;
+		adev->sdma.instance[i].fw_version = le32_to_cpu(hdr->header.ucode_version);
+		adev->sdma.instance[i].feature_version = le32_to_cpu(hdr->ucode_feature_version);
+		if (adev->sdma.instance[i].feature_version >= 20)
+			adev->sdma.instance[i].burst_nop = true;
 
 		if (adev->firmware.smu_load) {
 			info = &adev->firmware.ucode[AMDGPU_UCODE_ID_SDMA0 + i];
 			info->ucode_id = AMDGPU_UCODE_ID_SDMA0 + i;
-			info->fw = adev->sdma[i].fw;
+			info->fw = adev->sdma.instance[i].fw;
 			header = (const struct common_firmware_header *)info->fw->data;
 			adev->firmware.fw_size +=
 				ALIGN(le32_to_cpu(header->ucode_size_bytes), PAGE_SIZE);
@@ -164,9 +164,9 @@ out:
 		printk(KERN_ERR
 		       "sdma_v2_4: Failed to load firmware \"%s\"\n",
 		       fw_name);
-		for (i = 0; i < SDMA_MAX_INSTANCE; i++) {
-			release_firmware(adev->sdma[i].fw);
-			adev->sdma[i].fw = NULL;
+		for (i = 0; i < adev->sdma.num_instances; i++) {
+			release_firmware(adev->sdma.instance[i].fw);
+			adev->sdma.instance[i].fw = NULL;
 		}
 	}
 	return err;
@@ -199,7 +199,7 @@ static uint32_t sdma_v2_4_ring_get_rptr(struct amdgpu_ring *ring)
 static uint32_t sdma_v2_4_ring_get_wptr(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
-	int me = (ring == &ring->adev->sdma[0].ring) ? 0 : 1;
+	int me = (ring == &ring->adev->sdma.instance[0].ring) ? 0 : 1;
 	u32 wptr = RREG32(mmSDMA0_GFX_RB_WPTR + sdma_offsets[me]) >> 2;
 
 	return wptr;
@@ -215,14 +215,14 @@ static uint32_t sdma_v2_4_ring_get_wptr(struct amdgpu_ring *ring)
 static void sdma_v2_4_ring_set_wptr(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
-	int me = (ring == &ring->adev->sdma[0].ring) ? 0 : 1;
+	int me = (ring == &ring->adev->sdma.instance[0].ring) ? 0 : 1;
 
 	WREG32(mmSDMA0_GFX_RB_WPTR + sdma_offsets[me], ring->wptr << 2);
 }
 
 static void sdma_v2_4_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count)
 {
-	struct amdgpu_sdma *sdma = amdgpu_get_sdma_instance(ring);
+	struct amdgpu_sdma_instance *sdma = amdgpu_get_sdma_instance(ring);
 	int i;
 
 	for (i = 0; i < count; i++)
@@ -284,7 +284,7 @@ static void sdma_v2_4_ring_emit_hdp_flush(struct amdgpu_ring *ring)
 {
 	u32 ref_and_mask = 0;
 
-	if (ring == &ring->adev->sdma[0].ring)
+	if (ring == &ring->adev->sdma.instance[0].ring)
 		ref_and_mask = REG_SET_FIELD(ref_and_mask, GPU_HDP_FLUSH_DONE, SDMA0, 1);
 	else
 		ref_and_mask = REG_SET_FIELD(ref_and_mask, GPU_HDP_FLUSH_DONE, SDMA1, 1);
@@ -368,8 +368,8 @@ static bool sdma_v2_4_ring_emit_semaphore(struct amdgpu_ring *ring,
  */
 static void sdma_v2_4_gfx_stop(struct amdgpu_device *adev)
 {
-	struct amdgpu_ring *sdma0 = &adev->sdma[0].ring;
-	struct amdgpu_ring *sdma1 = &adev->sdma[1].ring;
+	struct amdgpu_ring *sdma0 = &adev->sdma.instance[0].ring;
+	struct amdgpu_ring *sdma1 = &adev->sdma.instance[1].ring;
 	u32 rb_cntl, ib_cntl;
 	int i;
 
@@ -377,7 +377,7 @@ static void sdma_v2_4_gfx_stop(struct amdgpu_device *adev)
 	    (adev->mman.buffer_funcs_ring == sdma1))
 		amdgpu_ttm_set_active_vram_size(adev, adev->mc.visible_vram_size);
 
-	for (i = 0; i < SDMA_MAX_INSTANCE; i++) {
+	for (i = 0; i < adev->sdma.num_instances; i++) {
 		rb_cntl = RREG32(mmSDMA0_GFX_RB_CNTL + sdma_offsets[i]);
 		rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_ENABLE, 0);
 		WREG32(mmSDMA0_GFX_RB_CNTL + sdma_offsets[i], rb_cntl);
@@ -419,7 +419,7 @@ static void sdma_v2_4_enable(struct amdgpu_device *adev, bool enable)
 		sdma_v2_4_rlc_stop(adev);
 	}
 
-	for (i = 0; i < SDMA_MAX_INSTANCE; i++) {
+	for (i = 0; i < adev->sdma.num_instances; i++) {
 		f32_cntl = RREG32(mmSDMA0_F32_CNTL + sdma_offsets[i]);
 		if (enable)
 			f32_cntl = REG_SET_FIELD(f32_cntl, SDMA0_F32_CNTL, HALT, 0);
@@ -445,8 +445,8 @@ static int sdma_v2_4_gfx_resume(struct amdgpu_device *adev)
 	u32 wb_offset;
 	int i, j, r;
 
-	for (i = 0; i < SDMA_MAX_INSTANCE; i++) {
-		ring = &adev->sdma[i].ring;
+	for (i = 0; i < adev->sdma.num_instances; i++) {
+		ring = &adev->sdma.instance[i].ring;
 		wb_offset = (ring->rptr_offs * 4);
 
 		mutex_lock(&adev->srbm_mutex);
@@ -545,29 +545,23 @@ static int sdma_v2_4_load_microcode(struct amdgpu_device *adev)
 	const __le32 *fw_data;
 	u32 fw_size;
 	int i, j;
-	bool smc_loads_fw = false; /* XXX fix me */
-
-	if (!adev->sdma[0].fw || !adev->sdma[1].fw)
-		return -EINVAL;
 
 	/* halt the MEs */
 	sdma_v2_4_enable(adev, false);
 
-	if (smc_loads_fw) {
-		/* XXX query SMC for fw load complete */
-	} else {
-		for (i = 0; i < SDMA_MAX_INSTANCE; i++) {
-			hdr = (const struct sdma_firmware_header_v1_0 *)adev->sdma[i].fw->data;
-			amdgpu_ucode_print_sdma_hdr(&hdr->header);
-			fw_size = le32_to_cpu(hdr->header.ucode_size_bytes) / 4;
-			fw_data = (const __le32 *)
-				(adev->sdma[i].fw->data +
-				 le32_to_cpu(hdr->header.ucode_array_offset_bytes));
-			WREG32(mmSDMA0_UCODE_ADDR + sdma_offsets[i], 0);
-			for (j = 0; j < fw_size; j++)
-				WREG32(mmSDMA0_UCODE_DATA + sdma_offsets[i], le32_to_cpup(fw_data++));
-			WREG32(mmSDMA0_UCODE_ADDR + sdma_offsets[i], adev->sdma[i].fw_version);
-		}
+	for (i = 0; i < adev->sdma.num_instances; i++) {
+		if (!adev->sdma.instance[i].fw)
+			return -EINVAL;
+		hdr = (const struct sdma_firmware_header_v1_0 *)adev->sdma.instance[i].fw->data;
+		amdgpu_ucode_print_sdma_hdr(&hdr->header);
+		fw_size = le32_to_cpu(hdr->header.ucode_size_bytes) / 4;
+		fw_data = (const __le32 *)
+			(adev->sdma.instance[i].fw->data +
+			 le32_to_cpu(hdr->header.ucode_array_offset_bytes));
+		WREG32(mmSDMA0_UCODE_ADDR + sdma_offsets[i], 0);
+		for (j = 0; j < fw_size; j++)
+			WREG32(mmSDMA0_UCODE_DATA + sdma_offsets[i], le32_to_cpup(fw_data++));
+		WREG32(mmSDMA0_UCODE_ADDR + sdma_offsets[i], adev->sdma.instance[i].fw_version);
 	}
 
 	return 0;
@@ -894,7 +888,7 @@ static void sdma_v2_4_vm_set_pte_pde(struct amdgpu_ib *ib,
  */
 static void sdma_v2_4_vm_pad_ib(struct amdgpu_ib *ib)
 {
-	struct amdgpu_sdma *sdma = amdgpu_get_sdma_instance(ib->ring);
+	struct amdgpu_sdma_instance *sdma = amdgpu_get_sdma_instance(ib->ring);
 	u32 pad_count;
 	int i;
 
@@ -952,6 +946,8 @@ static int sdma_v2_4_early_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
+	adev->sdma.num_instances = SDMA_MAX_INSTANCE;
+
 	sdma_v2_4_set_ring_funcs(adev);
 	sdma_v2_4_set_buffer_funcs(adev);
 	sdma_v2_4_set_vm_pte_funcs(adev);
@@ -963,21 +959,21 @@ static int sdma_v2_4_early_init(void *handle)
 static int sdma_v2_4_sw_init(void *handle)
 {
 	struct amdgpu_ring *ring;
-	int r;
+	int r, i;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
 	/* SDMA trap event */
-	r = amdgpu_irq_add_id(adev, 224, &adev->sdma_trap_irq);
+	r = amdgpu_irq_add_id(adev, 224, &adev->sdma.trap_irq);
 	if (r)
 		return r;
 
 	/* SDMA Privileged inst */
-	r = amdgpu_irq_add_id(adev, 241, &adev->sdma_illegal_inst_irq);
+	r = amdgpu_irq_add_id(adev, 241, &adev->sdma.illegal_inst_irq);
 	if (r)
 		return r;
 
 	/* SDMA Privileged inst */
-	r = amdgpu_irq_add_id(adev, 247, &adev->sdma_illegal_inst_irq);
+	r = amdgpu_irq_add_id(adev, 247, &adev->sdma.illegal_inst_irq);
 	if (r)
 		return r;
 
@@ -987,31 +983,20 @@ static int sdma_v2_4_sw_init(void *handle)
 		return r;
 	}
 
-	ring = &adev->sdma[0].ring;
-	ring->ring_obj = NULL;
-	ring->use_doorbell = false;
-
-	ring = &adev->sdma[1].ring;
-	ring->ring_obj = NULL;
-	ring->use_doorbell = false;
-
-	ring = &adev->sdma[0].ring;
-	sprintf(ring->name, "sdma0");
-	r = amdgpu_ring_init(adev, ring, 256 * 1024,
-			     SDMA_PKT_NOP_HEADER_OP(SDMA_OP_NOP), 0xf,
-			     &adev->sdma_trap_irq, AMDGPU_SDMA_IRQ_TRAP0,
-			     AMDGPU_RING_TYPE_SDMA);
-	if (r)
-		return r;
-
-	ring = &adev->sdma[1].ring;
-	sprintf(ring->name, "sdma1");
-	r = amdgpu_ring_init(adev, ring, 256 * 1024,
-			     SDMA_PKT_NOP_HEADER_OP(SDMA_OP_NOP), 0xf,
-			     &adev->sdma_trap_irq, AMDGPU_SDMA_IRQ_TRAP1,
-			     AMDGPU_RING_TYPE_SDMA);
-	if (r)
-		return r;
+	for (i = 0; i < adev->sdma.num_instances; i++) {
+		ring = &adev->sdma.instance[i].ring;
+		ring->ring_obj = NULL;
+		ring->use_doorbell = false;
+		sprintf(ring->name, "sdma%d", i);
+		r = amdgpu_ring_init(adev, ring, 256 * 1024,
+				     SDMA_PKT_NOP_HEADER_OP(SDMA_OP_NOP), 0xf,
+				     &adev->sdma.trap_irq,
+				     (i == 0) ?
+				     AMDGPU_SDMA_IRQ_TRAP0 : AMDGPU_SDMA_IRQ_TRAP1,
+				     AMDGPU_RING_TYPE_SDMA);
+		if (r)
+			return r;
+	}
 
 	return r;
 }
@@ -1019,9 +1004,10 @@ static int sdma_v2_4_sw_init(void *handle)
 static int sdma_v2_4_sw_fini(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	int i;
 
-	amdgpu_ring_fini(&adev->sdma[0].ring);
-	amdgpu_ring_fini(&adev->sdma[1].ring);
+	for (i = 0; i < adev->sdma.num_instances; i++)
+		amdgpu_ring_fini(&adev->sdma.instance[i].ring);
 
 	return 0;
 }
@@ -1100,7 +1086,7 @@ static void sdma_v2_4_print_status(void *handle)
 	dev_info(adev->dev, "VI SDMA registers\n");
 	dev_info(adev->dev, "  SRBM_STATUS2=0x%08X\n",
 		 RREG32(mmSRBM_STATUS2));
-	for (i = 0; i < SDMA_MAX_INSTANCE; i++) {
+	for (i = 0; i < adev->sdma.num_instances; i++) {
 		dev_info(adev->dev, "  SDMA%d_STATUS_REG=0x%08X\n",
 			 i, RREG32(mmSDMA0_STATUS_REG + sdma_offsets[i]));
 		dev_info(adev->dev, "  SDMA%d_F32_CNTL=0x%08X\n",
@@ -1243,7 +1229,7 @@ static int sdma_v2_4_process_trap_irq(struct amdgpu_device *adev,
 	case 0:
 		switch (queue_id) {
 		case 0:
-			amdgpu_fence_process(&adev->sdma[0].ring);
+			amdgpu_fence_process(&adev->sdma.instance[0].ring);
 			break;
 		case 1:
 			/* XXX compute */
@@ -1256,7 +1242,7 @@ static int sdma_v2_4_process_trap_irq(struct amdgpu_device *adev,
 	case 1:
 		switch (queue_id) {
 		case 0:
-			amdgpu_fence_process(&adev->sdma[1].ring);
+			amdgpu_fence_process(&adev->sdma.instance[1].ring);
 			break;
 		case 1:
 			/* XXX compute */
@@ -1345,8 +1331,10 @@ static const struct amdgpu_ring_funcs sdma_v2_4_ring_funcs = {
 
 static void sdma_v2_4_set_ring_funcs(struct amdgpu_device *adev)
 {
-	adev->sdma[0].ring.funcs = &sdma_v2_4_ring_funcs;
-	adev->sdma[1].ring.funcs = &sdma_v2_4_ring_funcs;
+	int i;
+
+	for (i = 0; i < adev->sdma.num_instances; i++)
+		adev->sdma.instance[i].ring.funcs = &sdma_v2_4_ring_funcs;
 }
 
 static const struct amdgpu_irq_src_funcs sdma_v2_4_trap_irq_funcs = {
@@ -1360,9 +1348,9 @@ static const struct amdgpu_irq_src_funcs sdma_v2_4_illegal_inst_irq_funcs = {
 
 static void sdma_v2_4_set_irq_funcs(struct amdgpu_device *adev)
 {
-	adev->sdma_trap_irq.num_types = AMDGPU_SDMA_IRQ_LAST;
-	adev->sdma_trap_irq.funcs = &sdma_v2_4_trap_irq_funcs;
-	adev->sdma_illegal_inst_irq.funcs = &sdma_v2_4_illegal_inst_irq_funcs;
+	adev->sdma.trap_irq.num_types = AMDGPU_SDMA_IRQ_LAST;
+	adev->sdma.trap_irq.funcs = &sdma_v2_4_trap_irq_funcs;
+	adev->sdma.illegal_inst_irq.funcs = &sdma_v2_4_illegal_inst_irq_funcs;
 }
 
 /**
@@ -1428,7 +1416,7 @@ static void sdma_v2_4_set_buffer_funcs(struct amdgpu_device *adev)
 {
 	if (adev->mman.buffer_funcs == NULL) {
 		adev->mman.buffer_funcs = &sdma_v2_4_buffer_funcs;
-		adev->mman.buffer_funcs_ring = &adev->sdma[0].ring;
+		adev->mman.buffer_funcs_ring = &adev->sdma.instance[0].ring;
 	}
 }
 
@@ -1443,7 +1431,7 @@ static void sdma_v2_4_set_vm_pte_funcs(struct amdgpu_device *adev)
 {
 	if (adev->vm_manager.vm_pte_funcs == NULL) {
 		adev->vm_manager.vm_pte_funcs = &sdma_v2_4_vm_pte_funcs;
-		adev->vm_manager.vm_pte_funcs_ring = &adev->sdma[0].ring;
+		adev->vm_manager.vm_pte_funcs_ring = &adev->sdma.instance[0].ring;
 		adev->vm_manager.vm_pte_funcs_ring->is_pte_ring = true;
 	}
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
index 9bfe92d..670555a 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
@@ -184,7 +184,7 @@ static int sdma_v3_0_init_microcode(struct amdgpu_device *adev)
 {
 	const char *chip_name;
 	char fw_name[30];
-	int err, i;
+	int err = 0, i;
 	struct amdgpu_firmware_info *info = NULL;
 	const struct common_firmware_header *header = NULL;
 	const struct sdma_firmware_header_v1_0 *hdr;
@@ -204,27 +204,27 @@ static int sdma_v3_0_init_microcode(struct amdgpu_device *adev)
 	default: BUG();
 	}
 
-	for (i = 0; i < SDMA_MAX_INSTANCE; i++) {
+	for (i = 0; i < adev->sdma.num_instances; i++) {
 		if (i == 0)
 			snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_sdma.bin", chip_name);
 		else
 			snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_sdma1.bin", chip_name);
-		err = request_firmware(&adev->sdma[i].fw, fw_name, adev->dev);
+		err = request_firmware(&adev->sdma.instance[i].fw, fw_name, adev->dev);
 		if (err)
 			goto out;
-		err = amdgpu_ucode_validate(adev->sdma[i].fw);
+		err = amdgpu_ucode_validate(adev->sdma.instance[i].fw);
 		if (err)
 			goto out;
-		hdr = (const struct sdma_firmware_header_v1_0 *)adev->sdma[i].fw->data;
-		adev->sdma[i].fw_version = le32_to_cpu(hdr->header.ucode_version);
-		adev->sdma[i].feature_version = le32_to_cpu(hdr->ucode_feature_version);
-		if (adev->sdma[i].feature_version >= 20)
-			adev->sdma[i].burst_nop = true;
+		hdr = (const struct sdma_firmware_header_v1_0 *)adev->sdma.instance[i].fw->data;
+		adev->sdma.instance[i].fw_version = le32_to_cpu(hdr->header.ucode_version);
+		adev->sdma.instance[i].feature_version = le32_to_cpu(hdr->ucode_feature_version);
+		if (adev->sdma.instance[i].feature_version >= 20)
+			adev->sdma.instance[i].burst_nop = true;
 
 		if (adev->firmware.smu_load) {
 			info = &adev->firmware.ucode[AMDGPU_UCODE_ID_SDMA0 + i];
 			info->ucode_id = AMDGPU_UCODE_ID_SDMA0 + i;
-			info->fw = adev->sdma[i].fw;
+			info->fw = adev->sdma.instance[i].fw;
 			header = (const struct common_firmware_header *)info->fw->data;
 			adev->firmware.fw_size +=
 				ALIGN(le32_to_cpu(header->ucode_size_bytes), PAGE_SIZE);
@@ -235,9 +235,9 @@ out:
 		printk(KERN_ERR
 		       "sdma_v3_0: Failed to load firmware \"%s\"\n",
 		       fw_name);
-		for (i = 0; i < SDMA_MAX_INSTANCE; i++) {
-			release_firmware(adev->sdma[i].fw);
-			adev->sdma[i].fw = NULL;
+		for (i = 0; i < adev->sdma.num_instances; i++) {
+			release_firmware(adev->sdma.instance[i].fw);
+			adev->sdma.instance[i].fw = NULL;
 		}
 	}
 	return err;
@@ -276,7 +276,7 @@ static uint32_t sdma_v3_0_ring_get_wptr(struct amdgpu_ring *ring)
 		/* XXX check if swapping is necessary on BE */
 		wptr = ring->adev->wb.wb[ring->wptr_offs] >> 2;
 	} else {
-		int me = (ring == &ring->adev->sdma[0].ring) ? 0 : 1;
+		int me = (ring == &ring->adev->sdma.instance[0].ring) ? 0 : 1;
 
 		wptr = RREG32(mmSDMA0_GFX_RB_WPTR + sdma_offsets[me]) >> 2;
 	}
@@ -300,7 +300,7 @@ static void sdma_v3_0_ring_set_wptr(struct amdgpu_ring *ring)
 		adev->wb.wb[ring->wptr_offs] = ring->wptr << 2;
 		WDOORBELL32(ring->doorbell_index, ring->wptr << 2);
 	} else {
-		int me = (ring == &ring->adev->sdma[0].ring) ? 0 : 1;
+		int me = (ring == &ring->adev->sdma.instance[0].ring) ? 0 : 1;
 
 		WREG32(mmSDMA0_GFX_RB_WPTR + sdma_offsets[me], ring->wptr << 2);
 	}
@@ -308,7 +308,7 @@ static void sdma_v3_0_ring_set_wptr(struct amdgpu_ring *ring)
 
 static void sdma_v3_0_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count)
 {
-	struct amdgpu_sdma *sdma = amdgpu_get_sdma_instance(ring);
+	struct amdgpu_sdma_instance *sdma = amdgpu_get_sdma_instance(ring);
 	int i;
 
 	for (i = 0; i < count; i++)
@@ -369,7 +369,7 @@ static void sdma_v3_0_ring_emit_hdp_flush(struct amdgpu_ring *ring)
 {
 	u32 ref_and_mask = 0;
 
-	if (ring == &ring->adev->sdma[0].ring)
+	if (ring == &ring->adev->sdma.instance[0].ring)
 		ref_and_mask = REG_SET_FIELD(ref_and_mask, GPU_HDP_FLUSH_DONE, SDMA0, 1);
 	else
 		ref_and_mask = REG_SET_FIELD(ref_and_mask, GPU_HDP_FLUSH_DONE, SDMA1, 1);
@@ -454,8 +454,8 @@ static bool sdma_v3_0_ring_emit_semaphore(struct amdgpu_ring *ring,
  */
 static void sdma_v3_0_gfx_stop(struct amdgpu_device *adev)
 {
-	struct amdgpu_ring *sdma0 = &adev->sdma[0].ring;
-	struct amdgpu_ring *sdma1 = &adev->sdma[1].ring;
+	struct amdgpu_ring *sdma0 = &adev->sdma.instance[0].ring;
+	struct amdgpu_ring *sdma1 = &adev->sdma.instance[1].ring;
 	u32 rb_cntl, ib_cntl;
 	int i;
 
@@ -463,7 +463,7 @@ static void sdma_v3_0_gfx_stop(struct amdgpu_device *adev)
 	    (adev->mman.buffer_funcs_ring == sdma1))
 		amdgpu_ttm_set_active_vram_size(adev, adev->mc.visible_vram_size);
 
-	for (i = 0; i < SDMA_MAX_INSTANCE; i++) {
+	for (i = 0; i < adev->sdma.num_instances; i++) {
 		rb_cntl = RREG32(mmSDMA0_GFX_RB_CNTL + sdma_offsets[i]);
 		rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_ENABLE, 0);
 		WREG32(mmSDMA0_GFX_RB_CNTL + sdma_offsets[i], rb_cntl);
@@ -500,7 +500,7 @@ static void sdma_v3_0_ctx_switch_enable(struct amdgpu_device *adev, bool enable)
 	u32 f32_cntl;
 	int i;
 
-	for (i = 0; i < SDMA_MAX_INSTANCE; i++) {
+	for (i = 0; i < adev->sdma.num_instances; i++) {
 		f32_cntl = RREG32(mmSDMA0_CNTL + sdma_offsets[i]);
 		if (enable)
 			f32_cntl = REG_SET_FIELD(f32_cntl, SDMA0_CNTL,
@@ -530,7 +530,7 @@ static void sdma_v3_0_enable(struct amdgpu_device *adev, bool enable)
 		sdma_v3_0_rlc_stop(adev);
 	}
 
-	for (i = 0; i < SDMA_MAX_INSTANCE; i++) {
+	for (i = 0; i < adev->sdma.num_instances; i++) {
 		f32_cntl = RREG32(mmSDMA0_F32_CNTL + sdma_offsets[i]);
 		if (enable)
 			f32_cntl = REG_SET_FIELD(f32_cntl, SDMA0_F32_CNTL, HALT, 0);
@@ -557,8 +557,8 @@ static int sdma_v3_0_gfx_resume(struct amdgpu_device *adev)
 	u32 doorbell;
 	int i, j, r;
 
-	for (i = 0; i < SDMA_MAX_INSTANCE; i++) {
-		ring = &adev->sdma[i].ring;
+	for (i = 0; i < adev->sdma.num_instances; i++) {
+		ring = &adev->sdma.instance[i].ring;
 		wb_offset = (ring->rptr_offs * 4);
 
 		mutex_lock(&adev->srbm_mutex);
@@ -669,23 +669,22 @@ static int sdma_v3_0_load_microcode(struct amdgpu_device *adev)
 	u32 fw_size;
 	int i, j;
 
-	if (!adev->sdma[0].fw || !adev->sdma[1].fw)
-		return -EINVAL;
-
 	/* halt the MEs */
 	sdma_v3_0_enable(adev, false);
 
-	for (i = 0; i < SDMA_MAX_INSTANCE; i++) {
-		hdr = (const struct sdma_firmware_header_v1_0 *)adev->sdma[i].fw->data;
+	for (i = 0; i < adev->sdma.num_instances; i++) {
+		if (!adev->sdma.instance[i].fw)
+			return -EINVAL;
+		hdr = (const struct sdma_firmware_header_v1_0 *)adev->sdma.instance[i].fw->data;
 		amdgpu_ucode_print_sdma_hdr(&hdr->header);
 		fw_size = le32_to_cpu(hdr->header.ucode_size_bytes) / 4;
 		fw_data = (const __le32 *)
-			(adev->sdma[i].fw->data +
+			(adev->sdma.instance[i].fw->data +
 				le32_to_cpu(hdr->header.ucode_array_offset_bytes));
 		WREG32(mmSDMA0_UCODE_ADDR + sdma_offsets[i], 0);
 		for (j = 0; j < fw_size; j++)
 			WREG32(mmSDMA0_UCODE_DATA + sdma_offsets[i], le32_to_cpup(fw_data++));
-		WREG32(mmSDMA0_UCODE_ADDR + sdma_offsets[i], adev->sdma[i].fw_version);
+		WREG32(mmSDMA0_UCODE_ADDR + sdma_offsets[i], adev->sdma.instance[i].fw_version);
 	}
 
 	return 0;
@@ -701,21 +700,21 @@ static int sdma_v3_0_load_microcode(struct amdgpu_device *adev)
  */
 static int sdma_v3_0_start(struct amdgpu_device *adev)
 {
-	int r;
+	int r, i;
 
 	if (!adev->firmware.smu_load) {
 		r = sdma_v3_0_load_microcode(adev);
 		if (r)
 			return r;
 	} else {
-		r = adev->smu.smumgr_funcs->check_fw_load_finish(adev,
-						AMDGPU_UCODE_ID_SDMA0);
-		if (r)
-			return -EINVAL;
-		r = adev->smu.smumgr_funcs->check_fw_load_finish(adev,
-						AMDGPU_UCODE_ID_SDMA1);
-		if (r)
-			return -EINVAL;
+		for (i = 0; i < adev->sdma.num_instances; i++) {
+			r = adev->smu.smumgr_funcs->check_fw_load_finish(adev,
+									 (i == 0) ?
+									 AMDGPU_UCODE_ID_SDMA0 :
+									 AMDGPU_UCODE_ID_SDMA1);
+			if (r)
+				return -EINVAL;
+		}
 	}
 
 	/* unhalt the MEs */
@@ -1013,7 +1012,7 @@ static void sdma_v3_0_vm_set_pte_pde(struct amdgpu_ib *ib,
  */
 static void sdma_v3_0_vm_pad_ib(struct amdgpu_ib *ib)
 {
-	struct amdgpu_sdma *sdma = amdgpu_get_sdma_instance(ib->ring);
+	struct amdgpu_sdma_instance *sdma = amdgpu_get_sdma_instance(ib->ring);
 	u32 pad_count;
 	int i;
 
@@ -1071,6 +1070,12 @@ static int sdma_v3_0_early_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
+	switch (adev->asic_type) {
+	default:
+		adev->sdma.num_instances = SDMA_MAX_INSTANCE;
+		break;
+	}
+
 	sdma_v3_0_set_ring_funcs(adev);
 	sdma_v3_0_set_buffer_funcs(adev);
 	sdma_v3_0_set_vm_pte_funcs(adev);
@@ -1082,21 +1087,21 @@ static int sdma_v3_0_early_init(void *handle)
 static int sdma_v3_0_sw_init(void *handle)
 {
 	struct amdgpu_ring *ring;
-	int r;
+	int r, i;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
 	/* SDMA trap event */
-	r = amdgpu_irq_add_id(adev, 224, &adev->sdma_trap_irq);
+	r = amdgpu_irq_add_id(adev, 224, &adev->sdma.trap_irq);
 	if (r)
 		return r;
 
 	/* SDMA Privileged inst */
-	r = amdgpu_irq_add_id(adev, 241, &adev->sdma_illegal_inst_irq);
+	r = amdgpu_irq_add_id(adev, 241, &adev->sdma.illegal_inst_irq);
 	if (r)
 		return r;
 
 	/* SDMA Privileged inst */
-	r = amdgpu_irq_add_id(adev, 247, &adev->sdma_illegal_inst_irq);
+	r = amdgpu_irq_add_id(adev, 247, &adev->sdma.illegal_inst_irq);
 	if (r)
 		return r;
 
@@ -1106,33 +1111,23 @@ static int sdma_v3_0_sw_init(void *handle)
 		return r;
 	}
 
-	ring = &adev->sdma[0].ring;
-	ring->ring_obj = NULL;
-	ring->use_doorbell = true;
-	ring->doorbell_index = AMDGPU_DOORBELL_sDMA_ENGINE0;
-
-	ring = &adev->sdma[1].ring;
-	ring->ring_obj = NULL;
-	ring->use_doorbell = true;
-	ring->doorbell_index = AMDGPU_DOORBELL_sDMA_ENGINE1;
-
-	ring = &adev->sdma[0].ring;
-	sprintf(ring->name, "sdma0");
-	r = amdgpu_ring_init(adev, ring, 256 * 1024,
-			     SDMA_PKT_NOP_HEADER_OP(SDMA_OP_NOP), 0xf,
-			     &adev->sdma_trap_irq, AMDGPU_SDMA_IRQ_TRAP0,
-			     AMDGPU_RING_TYPE_SDMA);
-	if (r)
-		return r;
-
-	ring = &adev->sdma[1].ring;
-	sprintf(ring->name, "sdma1");
-	r = amdgpu_ring_init(adev, ring, 256 * 1024,
-			     SDMA_PKT_NOP_HEADER_OP(SDMA_OP_NOP), 0xf,
-			     &adev->sdma_trap_irq, AMDGPU_SDMA_IRQ_TRAP1,
-			     AMDGPU_RING_TYPE_SDMA);
-	if (r)
-		return r;
+	for (i = 0; i < adev->sdma.num_instances; i++) {
+		ring = &adev->sdma.instance[i].ring;
+		ring->ring_obj = NULL;
+		ring->use_doorbell = true;
+		ring->doorbell_index = (i == 0) ?
+			AMDGPU_DOORBELL_sDMA_ENGINE0 : AMDGPU_DOORBELL_sDMA_ENGINE1;
+
+		sprintf(ring->name, "sdma%d", i);
+		r = amdgpu_ring_init(adev, ring, 256 * 1024,
+				     SDMA_PKT_NOP_HEADER_OP(SDMA_OP_NOP), 0xf,
+				     &adev->sdma.trap_irq,
+				     (i == 0) ?
+				     AMDGPU_SDMA_IRQ_TRAP0 : AMDGPU_SDMA_IRQ_TRAP1,
+				     AMDGPU_RING_TYPE_SDMA);
+		if (r)
+			return r;
+	}
 
 	return r;
 }
@@ -1140,9 +1135,10 @@ static int sdma_v3_0_sw_init(void *handle)
 static int sdma_v3_0_sw_fini(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	int i;
 
-	amdgpu_ring_fini(&adev->sdma[0].ring);
-	amdgpu_ring_fini(&adev->sdma[1].ring);
+	for (i = 0; i < adev->sdma.num_instances; i++)
+		amdgpu_ring_fini(&adev->sdma.instance[i].ring);
 
 	return 0;
 }
@@ -1222,7 +1218,7 @@ static void sdma_v3_0_print_status(void *handle)
 	dev_info(adev->dev, "VI SDMA registers\n");
 	dev_info(adev->dev, "  SRBM_STATUS2=0x%08X\n",
 		 RREG32(mmSRBM_STATUS2));
-	for (i = 0; i < SDMA_MAX_INSTANCE; i++) {
+	for (i = 0; i < adev->sdma.num_instances; i++) {
 		dev_info(adev->dev, "  SDMA%d_STATUS_REG=0x%08X\n",
 			 i, RREG32(mmSDMA0_STATUS_REG + sdma_offsets[i]));
 		dev_info(adev->dev, "  SDMA%d_F32_CNTL=0x%08X\n",
@@ -1367,7 +1363,7 @@ static int sdma_v3_0_process_trap_irq(struct amdgpu_device *adev,
 	case 0:
 		switch (queue_id) {
 		case 0:
-			amdgpu_fence_process(&adev->sdma[0].ring);
+			amdgpu_fence_process(&adev->sdma.instance[0].ring);
 			break;
 		case 1:
 			/* XXX compute */
@@ -1380,7 +1376,7 @@ static int sdma_v3_0_process_trap_irq(struct amdgpu_device *adev,
 	case 1:
 		switch (queue_id) {
 		case 0:
-			amdgpu_fence_process(&adev->sdma[1].ring);
+			amdgpu_fence_process(&adev->sdma.instance[1].ring);
 			break;
 		case 1:
 			/* XXX compute */
@@ -1468,8 +1464,10 @@ static const struct amdgpu_ring_funcs sdma_v3_0_ring_funcs = {
 
 static void sdma_v3_0_set_ring_funcs(struct amdgpu_device *adev)
 {
-	adev->sdma[0].ring.funcs = &sdma_v3_0_ring_funcs;
-	adev->sdma[1].ring.funcs = &sdma_v3_0_ring_funcs;
+	int i;
+
+	for (i = 0; i < adev->sdma.num_instances; i++)
+		adev->sdma.instance[i].ring.funcs = &sdma_v3_0_ring_funcs;
 }
 
 static const struct amdgpu_irq_src_funcs sdma_v3_0_trap_irq_funcs = {
@@ -1483,9 +1481,9 @@ static const struct amdgpu_irq_src_funcs sdma_v3_0_illegal_inst_irq_funcs = {
 
 static void sdma_v3_0_set_irq_funcs(struct amdgpu_device *adev)
 {
-	adev->sdma_trap_irq.num_types = AMDGPU_SDMA_IRQ_LAST;
-	adev->sdma_trap_irq.funcs = &sdma_v3_0_trap_irq_funcs;
-	adev->sdma_illegal_inst_irq.funcs = &sdma_v3_0_illegal_inst_irq_funcs;
+	adev->sdma.trap_irq.num_types = AMDGPU_SDMA_IRQ_LAST;
+	adev->sdma.trap_irq.funcs = &sdma_v3_0_trap_irq_funcs;
+	adev->sdma.illegal_inst_irq.funcs = &sdma_v3_0_illegal_inst_irq_funcs;
 }
 
 /**
@@ -1551,7 +1549,7 @@ static void sdma_v3_0_set_buffer_funcs(struct amdgpu_device *adev)
 {
 	if (adev->mman.buffer_funcs == NULL) {
 		adev->mman.buffer_funcs = &sdma_v3_0_buffer_funcs;
-		adev->mman.buffer_funcs_ring = &adev->sdma[0].ring;
+		adev->mman.buffer_funcs_ring = &adev->sdma.instance[0].ring;
 	}
 }
 
@@ -1566,7 +1564,7 @@ static void sdma_v3_0_set_vm_pte_funcs(struct amdgpu_device *adev)
 {
 	if (adev->vm_manager.vm_pte_funcs == NULL) {
 		adev->vm_manager.vm_pte_funcs = &sdma_v3_0_vm_pte_funcs;
-		adev->vm_manager.vm_pte_funcs_ring = &adev->sdma[0].ring;
+		adev->vm_manager.vm_pte_funcs_ring = &adev->sdma.instance[0].ring;
 		adev->vm_manager.vm_pte_funcs_ring->is_pte_ring = true;
 	}
 }
-- 
cgit v0.10.2


From 7dfac8965f8e108985bee38c746db2fe53314cc9 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Fri, 9 Oct 2015 11:38:49 -0400
Subject: drm/amdgpu: clean up pageflip interrupt handling
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Check to make sure we aren't touching a non-existent
display controller and simplify the code.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
index 57e1232..3707393 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
@@ -3305,37 +3305,20 @@ static int dce_v10_0_set_pageflip_irq_state(struct amdgpu_device *adev,
 					    unsigned type,
 					    enum amdgpu_interrupt_state state)
 {
-	u32 reg, reg_block;
-	/* now deal with page flip IRQ */
-	switch (type) {
-		case AMDGPU_PAGEFLIP_IRQ_D1:
-			reg_block = CRTC0_REGISTER_OFFSET;
-			break;
-		case AMDGPU_PAGEFLIP_IRQ_D2:
-			reg_block = CRTC1_REGISTER_OFFSET;
-			break;
-		case AMDGPU_PAGEFLIP_IRQ_D3:
-			reg_block = CRTC2_REGISTER_OFFSET;
-			break;
-		case AMDGPU_PAGEFLIP_IRQ_D4:
-			reg_block = CRTC3_REGISTER_OFFSET;
-			break;
-		case AMDGPU_PAGEFLIP_IRQ_D5:
-			reg_block = CRTC4_REGISTER_OFFSET;
-			break;
-		case AMDGPU_PAGEFLIP_IRQ_D6:
-			reg_block = CRTC5_REGISTER_OFFSET;
-			break;
-		default:
-			DRM_ERROR("invalid pageflip crtc %d\n", type);
-			return -EINVAL;
+	u32 reg;
+
+	if (type >= adev->mode_info.num_crtc) {
+		DRM_ERROR("invalid pageflip crtc %d\n", type);
+		return -EINVAL;
 	}
 
-	reg = RREG32(mmGRPH_INTERRUPT_CONTROL + reg_block);
+	reg = RREG32(mmGRPH_INTERRUPT_CONTROL + crtc_offsets[type]);
 	if (state == AMDGPU_IRQ_STATE_DISABLE)
-		WREG32(mmGRPH_INTERRUPT_CONTROL + reg_block, reg & ~GRPH_INTERRUPT_CONTROL__GRPH_PFLIP_INT_MASK_MASK);
+		WREG32(mmGRPH_INTERRUPT_CONTROL + crtc_offsets[type],
+		       reg & ~GRPH_INTERRUPT_CONTROL__GRPH_PFLIP_INT_MASK_MASK);
 	else
-		WREG32(mmGRPH_INTERRUPT_CONTROL + reg_block, reg | GRPH_INTERRUPT_CONTROL__GRPH_PFLIP_INT_MASK_MASK);
+		WREG32(mmGRPH_INTERRUPT_CONTROL + crtc_offsets[type],
+		       reg | GRPH_INTERRUPT_CONTROL__GRPH_PFLIP_INT_MASK_MASK);
 
 	return 0;
 }
@@ -3344,7 +3327,6 @@ static int dce_v10_0_pageflip_irq(struct amdgpu_device *adev,
 				  struct amdgpu_irq_src *source,
 				  struct amdgpu_iv_entry *entry)
 {
-	int reg_block;
 	unsigned long flags;
 	unsigned crtc_id;
 	struct amdgpu_crtc *amdgpu_crtc;
@@ -3353,33 +3335,15 @@ static int dce_v10_0_pageflip_irq(struct amdgpu_device *adev,
 	crtc_id = (entry->src_id - 8) >> 1;
 	amdgpu_crtc = adev->mode_info.crtcs[crtc_id];
 
-	/* ack the interrupt */
-	switch(crtc_id){
-		case AMDGPU_PAGEFLIP_IRQ_D1:
-			reg_block = CRTC0_REGISTER_OFFSET;
-			break;
-		case AMDGPU_PAGEFLIP_IRQ_D2:
-			reg_block = CRTC1_REGISTER_OFFSET;
-			break;
-		case AMDGPU_PAGEFLIP_IRQ_D3:
-			reg_block = CRTC2_REGISTER_OFFSET;
-			break;
-		case AMDGPU_PAGEFLIP_IRQ_D4:
-			reg_block = CRTC3_REGISTER_OFFSET;
-			break;
-		case AMDGPU_PAGEFLIP_IRQ_D5:
-			reg_block = CRTC4_REGISTER_OFFSET;
-			break;
-		case AMDGPU_PAGEFLIP_IRQ_D6:
-			reg_block = CRTC5_REGISTER_OFFSET;
-			break;
-		default:
-			DRM_ERROR("invalid pageflip crtc %d\n", crtc_id);
-			return -EINVAL;
+	if (crtc_id >= adev->mode_info.num_crtc) {
+		DRM_ERROR("invalid pageflip crtc %d\n", crtc_id);
+		return -EINVAL;
 	}
 
-	if (RREG32(mmGRPH_INTERRUPT_STATUS + reg_block) & GRPH_INTERRUPT_STATUS__GRPH_PFLIP_INT_OCCURRED_MASK)
-		WREG32(mmGRPH_INTERRUPT_STATUS + reg_block, GRPH_INTERRUPT_STATUS__GRPH_PFLIP_INT_CLEAR_MASK);
+	if (RREG32(mmGRPH_INTERRUPT_STATUS + crtc_offsets[crtc_id]) &
+	    GRPH_INTERRUPT_STATUS__GRPH_PFLIP_INT_OCCURRED_MASK)
+		WREG32(mmGRPH_INTERRUPT_STATUS + crtc_offsets[crtc_id],
+		       GRPH_INTERRUPT_STATUS__GRPH_PFLIP_INT_CLEAR_MASK);
 
 	/* IRQ could occur when in initial stage */
 	if (amdgpu_crtc == NULL)
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
index 378932a..c1147ec 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
@@ -3281,37 +3281,20 @@ static int dce_v11_0_set_pageflip_irq_state(struct amdgpu_device *adev,
 					    unsigned type,
 					    enum amdgpu_interrupt_state state)
 {
-	u32 reg, reg_block;
-	/* now deal with page flip IRQ */
-	switch (type) {
-		case AMDGPU_PAGEFLIP_IRQ_D1:
-			reg_block = CRTC0_REGISTER_OFFSET;
-			break;
-		case AMDGPU_PAGEFLIP_IRQ_D2:
-			reg_block = CRTC1_REGISTER_OFFSET;
-			break;
-		case AMDGPU_PAGEFLIP_IRQ_D3:
-			reg_block = CRTC2_REGISTER_OFFSET;
-			break;
-		case AMDGPU_PAGEFLIP_IRQ_D4:
-			reg_block = CRTC3_REGISTER_OFFSET;
-			break;
-		case AMDGPU_PAGEFLIP_IRQ_D5:
-			reg_block = CRTC4_REGISTER_OFFSET;
-			break;
-		case AMDGPU_PAGEFLIP_IRQ_D6:
-			reg_block = CRTC5_REGISTER_OFFSET;
-			break;
-		default:
-			DRM_ERROR("invalid pageflip crtc %d\n", type);
-			return -EINVAL;
+	u32 reg;
+
+	if (type >= adev->mode_info.num_crtc) {
+		DRM_ERROR("invalid pageflip crtc %d\n", type);
+		return -EINVAL;
 	}
 
-	reg = RREG32(mmGRPH_INTERRUPT_CONTROL + reg_block);
+	reg = RREG32(mmGRPH_INTERRUPT_CONTROL + crtc_offsets[type]);
 	if (state == AMDGPU_IRQ_STATE_DISABLE)
-		WREG32(mmGRPH_INTERRUPT_CONTROL + reg_block, reg & ~GRPH_INTERRUPT_CONTROL__GRPH_PFLIP_INT_MASK_MASK);
+		WREG32(mmGRPH_INTERRUPT_CONTROL + crtc_offsets[type],
+		       reg & ~GRPH_INTERRUPT_CONTROL__GRPH_PFLIP_INT_MASK_MASK);
 	else
-		WREG32(mmGRPH_INTERRUPT_CONTROL + reg_block, reg | GRPH_INTERRUPT_CONTROL__GRPH_PFLIP_INT_MASK_MASK);
+		WREG32(mmGRPH_INTERRUPT_CONTROL + crtc_offsets[type],
+		       reg | GRPH_INTERRUPT_CONTROL__GRPH_PFLIP_INT_MASK_MASK);
 
 	return 0;
 }
@@ -3320,7 +3303,6 @@ static int dce_v11_0_pageflip_irq(struct amdgpu_device *adev,
 				  struct amdgpu_irq_src *source,
 				  struct amdgpu_iv_entry *entry)
 {
-	int reg_block;
 	unsigned long flags;
 	unsigned crtc_id;
 	struct amdgpu_crtc *amdgpu_crtc;
@@ -3329,33 +3311,15 @@ static int dce_v11_0_pageflip_irq(struct amdgpu_device *adev,
 	crtc_id = (entry->src_id - 8) >> 1;
 	amdgpu_crtc = adev->mode_info.crtcs[crtc_id];
 
-	/* ack the interrupt */
-	switch(crtc_id){
-		case AMDGPU_PAGEFLIP_IRQ_D1:
-			reg_block = CRTC0_REGISTER_OFFSET;
-			break;
-		case AMDGPU_PAGEFLIP_IRQ_D2:
-			reg_block = CRTC1_REGISTER_OFFSET;
-			break;
-		case AMDGPU_PAGEFLIP_IRQ_D3:
-			reg_block = CRTC2_REGISTER_OFFSET;
-			break;
-		case AMDGPU_PAGEFLIP_IRQ_D4:
-			reg_block = CRTC3_REGISTER_OFFSET;
-			break;
-		case AMDGPU_PAGEFLIP_IRQ_D5:
-			reg_block = CRTC4_REGISTER_OFFSET;
-			break;
-		case AMDGPU_PAGEFLIP_IRQ_D6:
-			reg_block = CRTC5_REGISTER_OFFSET;
-			break;
-		default:
-			DRM_ERROR("invalid pageflip crtc %d\n", crtc_id);
-			return -EINVAL;
+	if (crtc_id >= adev->mode_info.num_crtc) {
+		DRM_ERROR("invalid pageflip crtc %d\n", crtc_id);
+		return -EINVAL;
 	}
 
-	if (RREG32(mmGRPH_INTERRUPT_STATUS + reg_block) & GRPH_INTERRUPT_STATUS__GRPH_PFLIP_INT_OCCURRED_MASK)
-		WREG32(mmGRPH_INTERRUPT_STATUS + reg_block, GRPH_INTERRUPT_STATUS__GRPH_PFLIP_INT_CLEAR_MASK);
+	if (RREG32(mmGRPH_INTERRUPT_STATUS + crtc_offsets[crtc_id]) &
+	    GRPH_INTERRUPT_STATUS__GRPH_PFLIP_INT_OCCURRED_MASK)
+		WREG32(mmGRPH_INTERRUPT_STATUS + crtc_offsets[crtc_id],
+		       GRPH_INTERRUPT_STATUS__GRPH_PFLIP_INT_CLEAR_MASK);
 
 	/* IRQ could occur when in initial stage */
 	if(amdgpu_crtc == NULL)
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
index 0d2c50d..d784fb4 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
@@ -3312,37 +3312,20 @@ static int dce_v8_0_set_pageflip_interrupt_state(struct amdgpu_device *adev,
 						 unsigned type,
 						 enum amdgpu_interrupt_state state)
 {
-	u32 reg, reg_block;
-	/* now deal with page flip IRQ */
-	switch (type) {
-		case AMDGPU_PAGEFLIP_IRQ_D1:
-			reg_block = CRTC0_REGISTER_OFFSET;
-			break;
-		case AMDGPU_PAGEFLIP_IRQ_D2:
-			reg_block = CRTC1_REGISTER_OFFSET;
-			break;
-		case AMDGPU_PAGEFLIP_IRQ_D3:
-			reg_block = CRTC2_REGISTER_OFFSET;
-			break;
-		case AMDGPU_PAGEFLIP_IRQ_D4:
-			reg_block = CRTC3_REGISTER_OFFSET;
-			break;
-		case AMDGPU_PAGEFLIP_IRQ_D5:
-			reg_block = CRTC4_REGISTER_OFFSET;
-			break;
-		case AMDGPU_PAGEFLIP_IRQ_D6:
-			reg_block = CRTC5_REGISTER_OFFSET;
-			break;
-		default:
-			DRM_ERROR("invalid pageflip crtc %d\n", type);
-			return -EINVAL;
+	u32 reg;
+
+	if (type >= adev->mode_info.num_crtc) {
+		DRM_ERROR("invalid pageflip crtc %d\n", type);
+		return -EINVAL;
 	}
 
-	reg = RREG32(mmGRPH_INTERRUPT_CONTROL + reg_block);
+	reg = RREG32(mmGRPH_INTERRUPT_CONTROL + crtc_offsets[type]);
 	if (state == AMDGPU_IRQ_STATE_DISABLE)
-		WREG32(mmGRPH_INTERRUPT_CONTROL + reg_block, reg & ~GRPH_INTERRUPT_CONTROL__GRPH_PFLIP_INT_MASK_MASK);
+		WREG32(mmGRPH_INTERRUPT_CONTROL + crtc_offsets[type],
+		       reg & ~GRPH_INTERRUPT_CONTROL__GRPH_PFLIP_INT_MASK_MASK);
 	else
-		WREG32(mmGRPH_INTERRUPT_CONTROL + reg_block, reg | GRPH_INTERRUPT_CONTROL__GRPH_PFLIP_INT_MASK_MASK);
+		WREG32(mmGRPH_INTERRUPT_CONTROL + crtc_offsets[type],
+		       reg | GRPH_INTERRUPT_CONTROL__GRPH_PFLIP_INT_MASK_MASK);
 
 	return 0;
 }
@@ -3351,7 +3334,6 @@ static int dce_v8_0_pageflip_irq(struct amdgpu_device *adev,
 				struct amdgpu_irq_src *source,
 				struct amdgpu_iv_entry *entry)
 {
-	int reg_block;
 	unsigned long flags;
 	unsigned crtc_id;
 	struct amdgpu_crtc *amdgpu_crtc;
@@ -3360,33 +3342,15 @@ static int dce_v8_0_pageflip_irq(struct amdgpu_device *adev,
 	crtc_id = (entry->src_id - 8) >> 1;
 	amdgpu_crtc = adev->mode_info.crtcs[crtc_id];
 
-	/* ack the interrupt */
-	switch(crtc_id){
-		case AMDGPU_PAGEFLIP_IRQ_D1:
-			reg_block = CRTC0_REGISTER_OFFSET;
-			break;
-		case AMDGPU_PAGEFLIP_IRQ_D2:
-			reg_block = CRTC1_REGISTER_OFFSET;
-			break;
-		case AMDGPU_PAGEFLIP_IRQ_D3:
-			reg_block = CRTC2_REGISTER_OFFSET;
-			break;
-		case AMDGPU_PAGEFLIP_IRQ_D4:
-			reg_block = CRTC3_REGISTER_OFFSET;
-			break;
-		case AMDGPU_PAGEFLIP_IRQ_D5:
-			reg_block = CRTC4_REGISTER_OFFSET;
-			break;
-		case AMDGPU_PAGEFLIP_IRQ_D6:
-			reg_block = CRTC5_REGISTER_OFFSET;
-			break;
-		default:
-			DRM_ERROR("invalid pageflip crtc %d\n", crtc_id);
-			return -EINVAL;
+	if (crtc_id >= adev->mode_info.num_crtc) {
+		DRM_ERROR("invalid pageflip crtc %d\n", crtc_id);
+		return -EINVAL;
 	}
 
-	if (RREG32(mmGRPH_INTERRUPT_STATUS + reg_block) & GRPH_INTERRUPT_STATUS__GRPH_PFLIP_INT_OCCURRED_MASK)
-		WREG32(mmGRPH_INTERRUPT_STATUS + reg_block, GRPH_INTERRUPT_STATUS__GRPH_PFLIP_INT_CLEAR_MASK);
+	if (RREG32(mmGRPH_INTERRUPT_STATUS + crtc_offsets[crtc_id]) &
+	    GRPH_INTERRUPT_STATUS__GRPH_PFLIP_INT_OCCURRED_MASK)
+		WREG32(mmGRPH_INTERRUPT_STATUS + crtc_offsets[crtc_id],
+		       GRPH_INTERRUPT_STATUS__GRPH_PFLIP_INT_CLEAR_MASK);
 
 	/* IRQ could occur when in initial stage */
 	if (amdgpu_crtc == NULL)
-- 
cgit v0.10.2


From 471577447a9af5f33435496c121343c5282ff3f5 Mon Sep 17 00:00:00 2001
From: Lukas Wunner <lukas@wunner.de>
Date: Mon, 12 Oct 2015 11:15:17 +0200
Subject: drm/radeon: Drop unnecessary #include <linux/vga_switcheroo.h>

This was added to three files even though they don't use any
vga_switcheroo symbols:

Added to radeon_acpi.c by commit d7a2952f1ade ("drm/radeon: Add
support for the ATIF ACPI method to the radeon driver").

Added to radeon_asic.c by commit 0a10c85129c2 ("drm/radeon: create
radeon_asic.c").

Added to radeon_bios.c by commit 6a9ee8af344e ("vga_switcheroo:
initial implementation (v15)").

Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/radeon/radeon_acpi.c b/drivers/gpu/drm/radeon/radeon_acpi.c
index 77e9d07..59acd0e 100644
--- a/drivers/gpu/drm/radeon/radeon_acpi.c
+++ b/drivers/gpu/drm/radeon/radeon_acpi.c
@@ -25,7 +25,6 @@
 #include <linux/acpi.h>
 #include <linux/slab.h>
 #include <linux/power_supply.h>
-#include <linux/vga_switcheroo.h>
 #include <acpi/video.h>
 #include <drm/drmP.h>
 #include <drm/drm_crtc_helper.h>
diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c
index f2421bc..1d4d452 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/drivers/gpu/drm/radeon/radeon_asic.c
@@ -31,7 +31,6 @@
 #include <drm/drm_crtc_helper.h>
 #include <drm/radeon_drm.h>
 #include <linux/vgaarb.h>
-#include <linux/vga_switcheroo.h>
 #include "radeon_reg.h"
 #include "radeon.h"
 #include "radeon_asic.h"
diff --git a/drivers/gpu/drm/radeon/radeon_bios.c b/drivers/gpu/drm/radeon/radeon_bios.c
index d27e4cc..21b6732 100644
--- a/drivers/gpu/drm/radeon/radeon_bios.c
+++ b/drivers/gpu/drm/radeon/radeon_bios.c
@@ -30,7 +30,6 @@
 #include "radeon.h"
 #include "atom.h"
 
-#include <linux/vga_switcheroo.h>
 #include <linux/slab.h>
 #include <linux/acpi.h>
 /*
-- 
cgit v0.10.2


From 3d2ce0d2394c82ac2ab7be8746eaa75042f2c997 Mon Sep 17 00:00:00 2001
From: Lukas Wunner <lukas@wunner.de>
Date: Mon, 12 Oct 2015 11:54:18 +0200
Subject: drm/amdgpu: Drop unnecessary #include <linux/vga_switcheroo.h>

This was added to two radeon files even though they don't use any
vga_switcheroo symbols, the amdgpu fork inherited them:

Added to amdgpu_acpi.c by commit d7a2952f1ade ("drm/radeon: Add
support for the ATIF ACPI method to the radeon driver").

Added to amdgpu_bios.c by commit 6a9ee8af344e ("vga_switcheroo:
initial implementation (v15)").

Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
index aef4a7a..a142d5a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c
@@ -25,7 +25,6 @@
 #include <linux/acpi.h>
 #include <linux/slab.h>
 #include <linux/power_supply.h>
-#include <linux/vga_switcheroo.h>
 #include <acpi/video.h>
 #include <drm/drmP.h>
 #include <drm/drm_crtc_helper.h>
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c
index 02add0a..c44c0c6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c
@@ -29,7 +29,6 @@
 #include "amdgpu.h"
 #include "atom.h"
 
-#include <linux/vga_switcheroo.h>
 #include <linux/slab.h>
 #include <linux/acpi.h>
 /*
-- 
cgit v0.10.2


From 39ff8449c5680466f568be57fbf79845ff5529ab Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Mon, 28 Sep 2015 12:01:20 +0200
Subject: drm/amdgpu: also trace already allocated VMIDs
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index 1e14531..2e15f2d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -147,8 +147,10 @@ int amdgpu_vm_grab_id(struct amdgpu_vm *vm, struct amdgpu_ring *ring,
 
 	/* check if the id is still valid */
 	if (vm_id->id && vm_id->last_id_use &&
-	    vm_id->last_id_use == adev->vm_manager.active[vm_id->id])
+	    vm_id->last_id_use == adev->vm_manager.active[vm_id->id]) {
+		trace_amdgpu_vm_grab_id(vm_id->id, ring->idx);
 		return 0;
+	}
 
 	/* we definately need to flush */
 	vm_id->pd_gpu_addr = ~0ll;
-- 
cgit v0.10.2


From ce0c6bcda6d0d9fc9df65dccb46a68b79ba6f018 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Thu, 10 Sep 2015 15:00:39 +0200
Subject: drm/amdgpu: only print meaningful VM faults
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Port of radeon commit 9b7d786b900baf7c0d1a7e211570aef1cb27590f.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
index 774528a..fab5471 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
@@ -1262,6 +1262,12 @@ static int gmc_v7_0_process_interrupt(struct amdgpu_device *adev,
 	addr = RREG32(mmVM_CONTEXT1_PROTECTION_FAULT_ADDR);
 	status = RREG32(mmVM_CONTEXT1_PROTECTION_FAULT_STATUS);
 	mc_client = RREG32(mmVM_CONTEXT1_PROTECTION_FAULT_MCCLIENT);
+	/* reset addr and status */
+	WREG32_P(mmVM_CONTEXT1_CNTL2, 1, ~1);
+
+	if (!addr && !status)
+		return 0;
+
 	dev_err(adev->dev, "GPU fault detected: %d 0x%08x\n",
 		entry->src_id, entry->src_data);
 	dev_err(adev->dev, "  VM_CONTEXT1_PROTECTION_FAULT_ADDR   0x%08X\n",
@@ -1269,8 +1275,6 @@ static int gmc_v7_0_process_interrupt(struct amdgpu_device *adev,
 	dev_err(adev->dev, "  VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x%08X\n",
 		status);
 	gmc_v7_0_vm_decode_fault(adev, status, addr, mc_client);
-	/* reset addr and status */
-	WREG32_P(mmVM_CONTEXT1_CNTL2, 1, ~1);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
index 9a07742..7bc9e9f 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
@@ -1262,6 +1262,12 @@ static int gmc_v8_0_process_interrupt(struct amdgpu_device *adev,
 	addr = RREG32(mmVM_CONTEXT1_PROTECTION_FAULT_ADDR);
 	status = RREG32(mmVM_CONTEXT1_PROTECTION_FAULT_STATUS);
 	mc_client = RREG32(mmVM_CONTEXT1_PROTECTION_FAULT_MCCLIENT);
+	/* reset addr and status */
+	WREG32_P(mmVM_CONTEXT1_CNTL2, 1, ~1);
+
+	if (!addr && !status)
+		return 0;
+
 	dev_err(adev->dev, "GPU fault detected: %d 0x%08x\n",
 		entry->src_id, entry->src_data);
 	dev_err(adev->dev, "  VM_CONTEXT1_PROTECTION_FAULT_ADDR   0x%08X\n",
@@ -1269,8 +1275,6 @@ static int gmc_v8_0_process_interrupt(struct amdgpu_device *adev,
 	dev_err(adev->dev, "  VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x%08X\n",
 		status);
 	gmc_v8_0_vm_decode_fault(adev, status, addr, mc_client);
-	/* reset addr and status */
-	WREG32_P(mmVM_CONTEXT1_CNTL2, 1, ~1);
 
 	return 0;
 }
-- 
cgit v0.10.2


From d9c13156a628cc9f8f062f2c10e2bff55b92aaab Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Mon, 28 Sep 2015 12:31:26 +0200
Subject: drm/amdgpu: add option to stop on VM fault
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index afc9848..a3dbbd1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -79,6 +79,7 @@ extern int amdgpu_bapm;
 extern int amdgpu_deep_color;
 extern int amdgpu_vm_size;
 extern int amdgpu_vm_block_size;
+extern int amdgpu_vm_fault_stop;
 extern int amdgpu_enable_scheduler;
 extern int amdgpu_sched_jobs;
 extern int amdgpu_sched_hw_submission;
@@ -960,6 +961,11 @@ struct amdgpu_ring {
 #define AMDGPU_PTE_FRAG_64KB	(4 << 7)
 #define AMDGPU_LOG2_PAGES_PER_FRAG 4
 
+/* How to programm VM fault handling */
+#define AMDGPU_VM_FAULT_STOP_NEVER	0
+#define AMDGPU_VM_FAULT_STOP_FIRST	1
+#define AMDGPU_VM_FAULT_STOP_ALWAYS	2
+
 struct amdgpu_vm_pt {
 	struct amdgpu_bo		*bo;
 	uint64_t			addr;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index c183772..bec0916 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -75,6 +75,7 @@ int amdgpu_bapm = -1;
 int amdgpu_deep_color = 0;
 int amdgpu_vm_size = 8;
 int amdgpu_vm_block_size = -1;
+int amdgpu_vm_fault_stop = 0;
 int amdgpu_exp_hw_support = 0;
 int amdgpu_enable_scheduler = 1;
 int amdgpu_sched_jobs = 16;
@@ -141,6 +142,9 @@ module_param_named(vm_size, amdgpu_vm_size, int, 0444);
 MODULE_PARM_DESC(vm_block_size, "VM page table size in bits (default depending on vm_size)");
 module_param_named(vm_block_size, amdgpu_vm_block_size, int, 0444);
 
+MODULE_PARM_DESC(vm_fault_stop, "Stop on VM fault (0 = never (default), 1 = print first, 2 = always)");
+module_param_named(vm_fault_stop, amdgpu_vm_fault_stop, int, 0444);
+
 MODULE_PARM_DESC(exp_hw_support, "experimental hw support (1 = enable, 0 = disable (default))");
 module_param_named(exp_hw_support, amdgpu_exp_hw_support, int, 0444);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
index fab5471..4883482 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
@@ -436,6 +436,33 @@ static int gmc_v7_0_gart_set_pte_pde(struct amdgpu_device *adev,
 }
 
 /**
+ * gmc_v8_0_set_fault_enable_default - update VM fault handling
+ *
+ * @adev: amdgpu_device pointer
+ * @value: true redirects VM faults to the default page
+ */
+static void gmc_v7_0_set_fault_enable_default(struct amdgpu_device *adev,
+					      bool value)
+{
+	u32 tmp;
+
+	tmp = RREG32(mmVM_CONTEXT1_CNTL);
+	tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+			    RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+			    DUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+			    PDE0_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+			    VALID_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+			    READ_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+			    WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	WREG32(mmVM_CONTEXT1_CNTL, tmp);
+}
+
+/**
  * gmc_v7_0_gart_enable - gart enable
  *
  * @adev: amdgpu_device pointer
@@ -523,15 +550,13 @@ static int gmc_v7_0_gart_enable(struct amdgpu_device *adev)
 	tmp = RREG32(mmVM_CONTEXT1_CNTL);
 	tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, ENABLE_CONTEXT, 1);
 	tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, PAGE_TABLE_DEPTH, 1);
-	tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
-	tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, DUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
-	tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, PDE0_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
-	tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, VALID_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
-	tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, READ_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
-	tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, 1);
 	tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, PAGE_TABLE_BLOCK_SIZE,
 			    amdgpu_vm_block_size - 9);
 	WREG32(mmVM_CONTEXT1_CNTL, tmp);
+	if (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_ALWAYS)
+		gmc_v7_0_set_fault_enable_default(adev, false);
+	else
+		gmc_v7_0_set_fault_enable_default(adev, true);
 
 	if (adev->asic_type == CHIP_KAVERI) {
 		tmp = RREG32(mmCHUB_CONTROL);
@@ -1268,6 +1293,9 @@ static int gmc_v7_0_process_interrupt(struct amdgpu_device *adev,
 	if (!addr && !status)
 		return 0;
 
+	if (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_FIRST)
+		gmc_v7_0_set_fault_enable_default(adev, false);
+
 	dev_err(adev->dev, "GPU fault detected: %d 0x%08x\n",
 		entry->src_id, entry->src_data);
 	dev_err(adev->dev, "  VM_CONTEXT1_PROTECTION_FAULT_ADDR   0x%08X\n",
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
index 7bc9e9f..42b5ff8 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
@@ -550,6 +550,35 @@ static int gmc_v8_0_gart_set_pte_pde(struct amdgpu_device *adev,
 }
 
 /**
+ * gmc_v8_0_set_fault_enable_default - update VM fault handling
+ *
+ * @adev: amdgpu_device pointer
+ * @value: true redirects VM faults to the default page
+ */
+static void gmc_v8_0_set_fault_enable_default(struct amdgpu_device *adev,
+					      bool value)
+{
+	u32 tmp;
+
+	tmp = RREG32(mmVM_CONTEXT1_CNTL);
+	tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+			    RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+			    DUMMY_PAGE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+			    PDE0_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+			    VALID_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+			    READ_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+			    WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL,
+			    EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, value);
+	WREG32(mmVM_CONTEXT1_CNTL, tmp);
+}
+
+/**
  * gmc_v8_0_gart_enable - gart enable
  *
  * @adev: amdgpu_device pointer
@@ -663,6 +692,10 @@ static int gmc_v8_0_gart_enable(struct amdgpu_device *adev)
 	tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, PAGE_TABLE_BLOCK_SIZE,
 			    amdgpu_vm_block_size - 9);
 	WREG32(mmVM_CONTEXT1_CNTL, tmp);
+	if (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_ALWAYS)
+		gmc_v8_0_set_fault_enable_default(adev, false);
+	else
+		gmc_v8_0_set_fault_enable_default(adev, true);
 
 	gmc_v8_0_gart_flush_gpu_tlb(adev, 0);
 	DRM_INFO("PCIE GART of %uM enabled (table at 0x%016llX).\n",
@@ -1268,6 +1301,9 @@ static int gmc_v8_0_process_interrupt(struct amdgpu_device *adev,
 	if (!addr && !status)
 		return 0;
 
+	if (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_FIRST)
+		gmc_v8_0_set_fault_enable_default(adev, false);
+
 	dev_err(adev->dev, "GPU fault detected: %d 0x%08x\n",
 		entry->src_id, entry->src_data);
 	dev_err(adev->dev, "  VM_CONTEXT1_PROTECTION_FAULT_ADDR   0x%08X\n",
-- 
cgit v0.10.2


From b495bd3a54e732e846bfde49ba2ea93b62b1de91 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Thu, 10 Sep 2015 14:00:35 +0200
Subject: drm/amdgpu: add option to clear VM page tables after every submit
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This makes it much easier to find when userspace misses to send some buffers.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index a3dbbd1..736eb45 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -80,6 +80,7 @@ extern int amdgpu_deep_color;
 extern int amdgpu_vm_size;
 extern int amdgpu_vm_block_size;
 extern int amdgpu_vm_fault_stop;
+extern int amdgpu_vm_debug;
 extern int amdgpu_enable_scheduler;
 extern int amdgpu_sched_jobs;
 extern int amdgpu_sched_hw_submission;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index 29fc45c..25012c7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -567,9 +567,24 @@ static int amdgpu_bo_vm_update_pte(struct amdgpu_cs_parser *p,
 			if (r)
 				return r;
 		}
+
+	}
+
+	r = amdgpu_vm_clear_invalids(adev, vm, &p->ibs[0].sync);
+
+	if (amdgpu_vm_debug && p->bo_list) {
+		/* Invalidate all BOs to test for userspace bugs */
+		for (i = 0; i < p->bo_list->num_entries; i++) {
+			/* ignore duplicates */
+			bo = p->bo_list->array[i].robj;
+			if (!bo)
+				continue;
+
+			amdgpu_vm_bo_invalidate(adev, bo);
+		}
 	}
 
-	return amdgpu_vm_clear_invalids(adev, vm, &p->ibs[0].sync);
+	return r;
 }
 
 static int amdgpu_cs_ib_vm_chunk(struct amdgpu_device *adev,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index bec0916..6134b1e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -76,6 +76,7 @@ int amdgpu_deep_color = 0;
 int amdgpu_vm_size = 8;
 int amdgpu_vm_block_size = -1;
 int amdgpu_vm_fault_stop = 0;
+int amdgpu_vm_debug = 0;
 int amdgpu_exp_hw_support = 0;
 int amdgpu_enable_scheduler = 1;
 int amdgpu_sched_jobs = 16;
@@ -145,6 +146,9 @@ module_param_named(vm_block_size, amdgpu_vm_block_size, int, 0444);
 MODULE_PARM_DESC(vm_fault_stop, "Stop on VM fault (0 = never (default), 1 = print first, 2 = always)");
 module_param_named(vm_fault_stop, amdgpu_vm_fault_stop, int, 0444);
 
+MODULE_PARM_DESC(vm_debug, "Debug VM handling (0 = disabled (default), 1 = enabled)");
+module_param_named(vm_debug, amdgpu_vm_debug, int, 0644);
+
 MODULE_PARM_DESC(exp_hw_support, "experimental hw support (1 = enable, 0 = disable (default))");
 module_param_named(exp_hw_support, amdgpu_exp_hw_support, int, 0444);
 
-- 
cgit v0.10.2


From d6c10f6b81e340e7d2d337183517a6da739d5ba2 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Mon, 28 Sep 2015 12:00:23 +0200
Subject: drm/amdgpu: add VM CS mapping trace point
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Output all VM mappings a command submission uses.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h
index 961d726..76ecbaf 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h
@@ -111,7 +111,7 @@ TRACE_EVENT(amdgpu_vm_bo_unmap,
 		      __entry->offset, __entry->flags)
 );
 
-TRACE_EVENT(amdgpu_vm_bo_update,
+DECLARE_EVENT_CLASS(amdgpu_vm_mapping,
 	    TP_PROTO(struct amdgpu_bo_va_mapping *mapping),
 	    TP_ARGS(mapping),
 	    TP_STRUCT__entry(
@@ -129,6 +129,16 @@ TRACE_EVENT(amdgpu_vm_bo_update,
 		      __entry->soffset, __entry->eoffset, __entry->flags)
 );
 
+DEFINE_EVENT(amdgpu_vm_mapping, amdgpu_vm_bo_update,
+	    TP_PROTO(struct amdgpu_bo_va_mapping *mapping),
+	    TP_ARGS(mapping)
+);
+
+DEFINE_EVENT(amdgpu_vm_mapping, amdgpu_vm_bo_mapping,
+	    TP_PROTO(struct amdgpu_bo_va_mapping *mapping),
+	    TP_ARGS(mapping)
+);
+
 TRACE_EVENT(amdgpu_vm_set_page,
 	    TP_PROTO(uint64_t pe, uint64_t addr, unsigned count,
 		     uint32_t incr, uint32_t flags),
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index 2e15f2d..0675524 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -852,6 +852,14 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev,
 			return r;
 	}
 
+	if (trace_amdgpu_vm_bo_mapping_enabled()) {
+		list_for_each_entry(mapping, &bo_va->valids, list)
+			trace_amdgpu_vm_bo_mapping(mapping);
+
+		list_for_each_entry(mapping, &bo_va->invalids, list)
+			trace_amdgpu_vm_bo_mapping(mapping);
+	}
+
 	spin_lock(&vm->status_lock);
 	list_splice_init(&bo_va->invalids, &bo_va->valids);
 	list_del_init(&bo_va->vm_status);
-- 
cgit v0.10.2


From 2440ff2c9151120c8ae27de6565b11831ee07e08 Mon Sep 17 00:00:00 2001
From: Junwei Zhang <Jerry.Zhang@amd.com>
Date: Sat, 10 Oct 2015 08:48:42 +0800
Subject: drm/amdgpu: add timer to fence to detect scheduler lockup
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Change-Id: I67e987db0efdca28faa80b332b75571192130d33
Signed-off-by: Junwei Zhang <Jerry.Zhang@amd.com>
Reviewed-by: David Zhou <david1.zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
index b3fc26c..fcad7e0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
@@ -628,8 +628,20 @@ int amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring)
 	init_waitqueue_head(&ring->fence_drv.fence_queue);
 
 	if (amdgpu_enable_scheduler) {
+		long timeout = msecs_to_jiffies(amdgpu_lockup_timeout);
+		if (timeout == 0) {
+			/*
+			 * FIXME:
+			 * Delayed workqueue cannot use it directly,
+			 * so the scheduler will not use delayed workqueue if
+			 * MAX_SCHEDULE_TIMEOUT is set.
+			 * Currently keep it simple and silly.
+			 */
+			timeout = MAX_SCHEDULE_TIMEOUT;
+		}
 		r = amd_sched_init(&ring->sched, &amdgpu_sched_ops,
-				   amdgpu_sched_hw_submission, ring->name);
+				   amdgpu_sched_hw_submission,
+				   timeout, ring->name);
 		if (r) {
 			DRM_ERROR("Failed to create scheduler on ring %s.\n",
 				  ring->name);
diff --git a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c
index 3697eee..a413dee 100644
--- a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c
+++ b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c
@@ -327,19 +327,49 @@ static void amd_sched_process_job(struct fence *f, struct fence_cb *cb)
 	struct amd_sched_fence *s_fence =
 		container_of(cb, struct amd_sched_fence, cb);
 	struct amd_gpu_scheduler *sched = s_fence->sched;
+	unsigned long flags;
 
 	atomic_dec(&sched->hw_rq_count);
 	amd_sched_fence_signal(s_fence);
+	if (sched->timeout != MAX_SCHEDULE_TIMEOUT) {
+		cancel_delayed_work_sync(&s_fence->dwork);
+		spin_lock_irqsave(&sched->fence_list_lock, flags);
+		list_del_init(&s_fence->list);
+		spin_unlock_irqrestore(&sched->fence_list_lock, flags);
+	}
 	fence_put(&s_fence->base);
 	wake_up_interruptible(&sched->wake_up_worker);
 }
 
+static void amd_sched_fence_work_func(struct work_struct *work)
+{
+	struct amd_sched_fence *s_fence =
+		container_of(work, struct amd_sched_fence, dwork.work);
+	struct amd_gpu_scheduler *sched = s_fence->sched;
+	struct amd_sched_fence *entity, *tmp;
+	unsigned long flags;
+
+	DRM_ERROR("[%s] scheduler is timeout!\n", sched->name);
+
+	/* Clean all pending fences */
+	list_for_each_entry_safe(entity, tmp, &sched->fence_list, list) {
+		DRM_ERROR("  fence no %d\n", entity->base.seqno);
+		cancel_delayed_work_sync(&entity->dwork);
+		spin_lock_irqsave(&sched->fence_list_lock, flags);
+		list_del_init(&entity->list);
+		spin_unlock_irqrestore(&sched->fence_list_lock, flags);
+		fence_put(&entity->base);
+	}
+}
+
 static int amd_sched_main(void *param)
 {
 	struct sched_param sparam = {.sched_priority = 1};
 	struct amd_gpu_scheduler *sched = (struct amd_gpu_scheduler *)param;
 	int r, count;
 
+	spin_lock_init(&sched->fence_list_lock);
+	INIT_LIST_HEAD(&sched->fence_list);
 	sched_setscheduler(current, SCHED_FIFO, &sparam);
 
 	while (!kthread_should_stop()) {
@@ -347,6 +377,7 @@ static int amd_sched_main(void *param)
 		struct amd_sched_fence *s_fence;
 		struct amd_sched_job *sched_job;
 		struct fence *fence;
+		unsigned long flags;
 
 		wait_event_interruptible(sched->wake_up_worker,
 			kthread_should_stop() ||
@@ -357,6 +388,15 @@ static int amd_sched_main(void *param)
 
 		entity = sched_job->s_entity;
 		s_fence = sched_job->s_fence;
+
+		if (sched->timeout != MAX_SCHEDULE_TIMEOUT) {
+			INIT_DELAYED_WORK(&s_fence->dwork, amd_sched_fence_work_func);
+			schedule_delayed_work(&s_fence->dwork, sched->timeout);
+			spin_lock_irqsave(&sched->fence_list_lock, flags);
+			list_add_tail(&s_fence->list, &sched->fence_list);
+			spin_unlock_irqrestore(&sched->fence_list_lock, flags);
+		}
+
 		atomic_inc(&sched->hw_rq_count);
 		fence = sched->ops->run_job(sched_job);
 		if (fence) {
@@ -392,11 +432,12 @@ static int amd_sched_main(void *param)
 */
 int amd_sched_init(struct amd_gpu_scheduler *sched,
 		   struct amd_sched_backend_ops *ops,
-		   unsigned hw_submission, const char *name)
+		   unsigned hw_submission, long timeout, const char *name)
 {
 	sched->ops = ops;
 	sched->hw_submission_limit = hw_submission;
 	sched->name = name;
+	sched->timeout = timeout;
 	amd_sched_rq_init(&sched->sched_rq);
 	amd_sched_rq_init(&sched->kernel_rq);
 
diff --git a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h
index 80b64dc..929e9ac 100644
--- a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h
+++ b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.h
@@ -68,6 +68,8 @@ struct amd_sched_fence {
 	struct amd_gpu_scheduler	*sched;
 	spinlock_t			lock;
 	void                            *owner;
+	struct delayed_work		dwork;
+	struct list_head		list;
 };
 
 struct amd_sched_job {
@@ -103,18 +105,21 @@ struct amd_sched_backend_ops {
 struct amd_gpu_scheduler {
 	struct amd_sched_backend_ops	*ops;
 	uint32_t			hw_submission_limit;
+	long				timeout;
 	const char			*name;
 	struct amd_sched_rq		sched_rq;
 	struct amd_sched_rq		kernel_rq;
 	wait_queue_head_t		wake_up_worker;
 	wait_queue_head_t		job_scheduled;
 	atomic_t			hw_rq_count;
+	struct list_head		fence_list;
+	spinlock_t			fence_list_lock;
 	struct task_struct		*thread;
 };
 
 int amd_sched_init(struct amd_gpu_scheduler *sched,
 		   struct amd_sched_backend_ops *ops,
-		   uint32_t hw_submission, const char *name);
+		   uint32_t hw_submission, long timeout, const char *name);
 void amd_sched_fini(struct amd_gpu_scheduler *sched);
 
 int amd_sched_entity_init(struct amd_gpu_scheduler *sched,
-- 
cgit v0.10.2


From 2fcef6ec87a044221fc3c2f16873f7c02b9ae991 Mon Sep 17 00:00:00 2001
From: Junwei Zhang <Jerry.Zhang@amd.com>
Date: Tue, 13 Oct 2015 11:14:23 +0800
Subject: drm/amdgpu: fix lockup when clean pending fences
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The first lockup fence will lock the fence list of scheduler.
Then cancel the delayed workqueues for all clean pending fences
without waiting the workqueues to finish.

Change-Id: I9bec826de1aa49d587b0662f3fb4a95333979429
Signed-off-by: Junwei Zhang <Jerry.Zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>

diff --git a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c
index a413dee..7fa1d7a 100644
--- a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c
+++ b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c
@@ -352,14 +352,14 @@ static void amd_sched_fence_work_func(struct work_struct *work)
 	DRM_ERROR("[%s] scheduler is timeout!\n", sched->name);
 
 	/* Clean all pending fences */
+	spin_lock_irqsave(&sched->fence_list_lock, flags);
 	list_for_each_entry_safe(entity, tmp, &sched->fence_list, list) {
 		DRM_ERROR("  fence no %d\n", entity->base.seqno);
-		cancel_delayed_work_sync(&entity->dwork);
-		spin_lock_irqsave(&sched->fence_list_lock, flags);
+		cancel_delayed_work(&entity->dwork);
 		list_del_init(&entity->list);
-		spin_unlock_irqrestore(&sched->fence_list_lock, flags);
 		fence_put(&entity->base);
 	}
+	spin_unlock_irqrestore(&sched->fence_list_lock, flags);
 }
 
 static int amd_sched_main(void *param)
-- 
cgit v0.10.2


From fc6f93bce582ccf76335843584e6a797ac72813c Mon Sep 17 00:00:00 2001
From: Bob Paauwe <bob.j.paauwe@intel.com>
Date: Mon, 31 Aug 2015 14:03:30 -0700
Subject: drm/i915/skl+: Enable pipe CSC on cursor planes. (v2)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Extend this to SKL and BXT as it's needed for these platforms as well.

v2: Change if condition to HAS_DDI() instead of listing each platform
Signed-off-by: Bob Paauwe <bob.j.paauwe@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index d37b7a1..82128b9 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -9927,7 +9927,7 @@ static void i9xx_update_cursor(struct drm_crtc *crtc, u32 base)
 		}
 		cntl |= pipe << 28; /* Connect to correct pipe */
 
-		if (IS_HASWELL(dev) || IS_BROADWELL(dev))
+		if (HAS_DDI(dev))
 			cntl |= CURSOR_PIPE_CSC_ENABLE;
 	}
 
-- 
cgit v0.10.2


From e12c8ce8c91fe4fa999d3b4cae8eb4af4fab49d5 Mon Sep 17 00:00:00 2001
From: Bob Paauwe <bob.j.paauwe@intel.com>
Date: Thu, 27 Aug 2015 13:46:30 -0700
Subject: drm/i915/skl: Enable pipe gamma for sprite planes.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Since SKL has universal planes, we should configure the sprite planes
and the primary plane the same.  For the primary plane we do enable
the pipe gamma on the plane so do the same for the non-primary planes.

Without this, the pipe CRC values will be different for something
displayed on the primary plane and something displayed on a sprite
plane when the ARGB8888 format is used.

Signed-off-by: Bob Paauwe <bob.j.paauwe@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c
index b229c67..e70b6fc 100644
--- a/drivers/gpu/drm/i915/intel_sprite.c
+++ b/drivers/gpu/drm/i915/intel_sprite.c
@@ -203,6 +203,7 @@ skl_update_plane(struct drm_plane *drm_plane, struct drm_crtc *crtc,
 	int scaler_id;
 
 	plane_ctl = PLANE_CTL_ENABLE |
+		PLANE_CTL_PIPE_GAMMA_ENABLE |
 		PLANE_CTL_PIPE_CSC_ENABLE;
 
 	plane_ctl |= skl_plane_ctl_format(fb->pixel_format);
-- 
cgit v0.10.2


From 0a878716265e9af9f697264dc2e858fcc060d833 Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Thu, 15 Oct 2015 14:23:01 +0200
Subject: drm/i915: restore ggtt double-bind avoidance

This was accidentally lost in

commit 75d04a3773ecee617847de963ae4195d6aa74c28
Author: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Date:   Tue Apr 28 17:56:17 2015 +0300

    drm/i915/gtt: Allocate va range only if vma is not bound

While at it implement an improved version suggested by Chris which
avoids the double-bind irrespective of what type of bind is done
first.

Note that this exact bug was already addressed in

commit d0e30adc42d979e4adc36b6c112b57337423b70c
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Wed Jul 29 20:02:48 2015 +0100

    drm/i915: Mark PIN_USER binding as GLOBAL_BIND without the aliasing ppgtt

but the problem is still that originally in

commit 0875546c5318c85c13d07014af5350e9000bc9e9
Author: Daniel Vetter <daniel.vetter@ffwll.ch>
Date:   Mon Apr 20 09:04:05 2015 -0700

    drm/i915: Fix up the vma aliasing ppgtt binding

if forgotten to take into account there case where we have a
GLOBAL_BIND before a LOCAL_BIND. This patch here fixes that.

v2: Pimp commit message and revert the partial fix.

v3: Split into two functions to specialize on aliasing_ppgtt y/n.

v4: WARN_ON for paranoia in the init sequence, since the ggtt probe
and aliasing ppgtt setup are far apart.

v5: Style nits.

Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Michel Thierry <michel.thierry@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Link: http://mid.gmane.org/1444911781-32607-1-git-send-email-daniel.vetter@ffwll.ch
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 620d57e..43f35d1 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -2502,6 +2502,36 @@ static int ggtt_bind_vma(struct i915_vma *vma,
 			 enum i915_cache_level cache_level,
 			 u32 flags)
 {
+	struct drm_i915_gem_object *obj = vma->obj;
+	u32 pte_flags = 0;
+	int ret;
+
+	ret = i915_get_ggtt_vma_pages(vma);
+	if (ret)
+		return ret;
+
+	/* Currently applicable only to VLV */
+	if (obj->gt_ro)
+		pte_flags |= PTE_READ_ONLY;
+
+	vma->vm->insert_entries(vma->vm, vma->ggtt_view.pages,
+				vma->node.start,
+				cache_level, pte_flags);
+
+	/*
+	 * Without aliasing PPGTT there's no difference between
+	 * GLOBAL/LOCAL_BIND, it's all the same ptes. Hence unconditionally
+	 * upgrade to both bound if we bind either to avoid double-binding.
+	 */
+	vma->bound |= GLOBAL_BIND | LOCAL_BIND;
+
+	return 0;
+}
+
+static int aliasing_gtt_bind_vma(struct i915_vma *vma,
+				 enum i915_cache_level cache_level,
+				 u32 flags)
+{
 	struct drm_device *dev = vma->vm->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_gem_object *obj = vma->obj;
@@ -2519,23 +2549,13 @@ static int ggtt_bind_vma(struct i915_vma *vma,
 		pte_flags |= PTE_READ_ONLY;
 
 
-	if (!dev_priv->mm.aliasing_ppgtt || flags & GLOBAL_BIND) {
+	if (flags & GLOBAL_BIND) {
 		vma->vm->insert_entries(vma->vm, pages,
 					vma->node.start,
 					cache_level, pte_flags);
-
-		/* Note the inconsistency here is due to absence of the
-		 * aliasing ppgtt on gen4 and earlier. Though we always
-		 * request PIN_USER for execbuffer (translated to LOCAL_BIND),
-		 * without the appgtt, we cannot honour that request and so
-		 * must substitute it with a global binding. Since we do this
-		 * behind the upper layers back, we need to explicitly set
-		 * the bound flag ourselves.
-		 */
-		vma->bound |= GLOBAL_BIND;
 	}
 
-	if (dev_priv->mm.aliasing_ppgtt && flags & LOCAL_BIND) {
+	if (flags & LOCAL_BIND) {
 		struct i915_hw_ppgtt *appgtt = dev_priv->mm.aliasing_ppgtt;
 		appgtt->base.insert_entries(&appgtt->base, pages,
 					    vma->node.start,
@@ -2699,6 +2719,8 @@ static int i915_gem_setup_global_gtt(struct drm_device *dev,
 					true);
 
 		dev_priv->mm.aliasing_ppgtt = ppgtt;
+		WARN_ON(dev_priv->gtt.base.bind_vma != ggtt_bind_vma);
+		dev_priv->gtt.base.bind_vma = aliasing_gtt_bind_vma;
 	}
 
 	return 0;
-- 
cgit v0.10.2


From 45c1da501fc5051fb04f579692f5f0a7d3b0ffbe Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Thu, 15 Oct 2015 09:36:34 +0200
Subject: drm/radeon: Use rdev->gem.mutex to protect hyperz/cmask owners
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This removes the last depency of radeon for dev->struct_mutex!

Also the locking scheme for hyperz/cmask owners seems a bit unsound,
there's no protection in the preclose handler (and that never did hold
dev->struct_mutex while being called). So grab the same lock there,
too.

There's also all the checks in the cs checker, but since the overall
design seems to never stall for the previous owner I figured it's ok
if I leave this racy. It was racy even before I touched it after all
too.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c
index 977f1a5..0a68e8d 100644
--- a/drivers/gpu/drm/radeon/radeon_kms.c
+++ b/drivers/gpu/drm/radeon/radeon_kms.c
@@ -181,7 +181,9 @@ static void radeon_set_filp_rights(struct drm_device *dev,
 				   struct drm_file *applier,
 				   uint32_t *value)
 {
-	mutex_lock(&dev->struct_mutex);
+	struct radeon_device *rdev = dev->dev_private;
+
+	mutex_lock(&rdev->gem.mutex);
 	if (*value == 1) {
 		/* wants rights */
 		if (!*owner)
@@ -192,7 +194,7 @@ static void radeon_set_filp_rights(struct drm_device *dev,
 			*owner = NULL;
 	}
 	*value = *owner == applier ? 1 : 0;
-	mutex_unlock(&dev->struct_mutex);
+	mutex_unlock(&rdev->gem.mutex);
 }
 
 /*
@@ -724,10 +726,14 @@ void radeon_driver_preclose_kms(struct drm_device *dev,
 				struct drm_file *file_priv)
 {
 	struct radeon_device *rdev = dev->dev_private;
+
+	mutex_lock(&rdev->gem.mutex);
 	if (rdev->hyperz_filp == file_priv)
 		rdev->hyperz_filp = NULL;
 	if (rdev->cmask_filp == file_priv)
 		rdev->cmask_filp = NULL;
+	mutex_unlock(&rdev->gem.mutex);
+
 	radeon_uvd_free_handles(rdev, file_priv);
 	radeon_vce_free_handles(rdev, file_priv);
 }
-- 
cgit v0.10.2


From c562657a75282afb00498ea82949ba7a9944ed15 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 15 Oct 2015 17:04:04 +0300
Subject: drm/i915: Kill the leftover RMW from ivb_sprite_disable()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We still had one lingering RMW in ivb_sprite_disable(), all the other
RMWs were killed off from the sprite code some time ago. Kill the
straggler too.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c
index e70b6fc..56dc132 100644
--- a/drivers/gpu/drm/i915/intel_sprite.c
+++ b/drivers/gpu/drm/i915/intel_sprite.c
@@ -613,7 +613,7 @@ ivb_disable_plane(struct drm_plane *plane, struct drm_crtc *crtc)
 	struct intel_plane *intel_plane = to_intel_plane(plane);
 	int pipe = intel_plane->pipe;
 
-	I915_WRITE(SPRCTL(pipe), I915_READ(SPRCTL(pipe)) & ~SPRITE_ENABLE);
+	I915_WRITE(SPRCTL(pipe), 0);
 	/* Can't leave the scaler enabled... */
 	if (intel_plane->can_scale)
 		I915_WRITE(SPRSCALE(pipe), 0);
-- 
cgit v0.10.2


From 7a6007c309c9b0e573dc17ac9d81e8b5f173847d Mon Sep 17 00:00:00 2001
From: Mikko Rapeli <mikko.rapeli@iki.fi>
Date: Thu, 15 Oct 2015 07:55:50 +0200
Subject: include/uapi/drm/sis_drm.h: move sis_file_private to
 drivers/gpu/drm/sis/sis_drv.h
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fixes userspace compile error:

drm/sis_drm.h:68:19: error: field ‘obj_list’ has incomplete type
  struct list_head obj_list;

Suggested by Emil Velikov <emil.l.velikov@gmail.com> at
https://lkml.org/lkml/2015/6/3/792

Signed-off-by: Mikko Rapeli <mikko.rapeli@iki.fi>
Signed-off-by: Dave Airlie <airlied@redhat.com>

diff --git a/drivers/gpu/drm/sis/sis_drv.h b/drivers/gpu/drm/sis/sis_drv.h
index 16f972b..328f8a7 100644
--- a/drivers/gpu/drm/sis/sis_drv.h
+++ b/drivers/gpu/drm/sis/sis_drv.h
@@ -67,6 +67,10 @@ typedef struct drm_sis_private {
 	struct idr object_idr;
 } drm_sis_private_t;
 
+struct sis_file_private {
+	struct list_head obj_list;
+};
+
 extern int sis_idle(struct drm_device *dev);
 extern void sis_reclaim_buffers_locked(struct drm_device *dev,
 				       struct drm_file *file_priv);
diff --git a/include/uapi/drm/sis_drm.h b/include/uapi/drm/sis_drm.h
index df37632..374858c 100644
--- a/include/uapi/drm/sis_drm.h
+++ b/include/uapi/drm/sis_drm.h
@@ -64,8 +64,4 @@ typedef struct {
 	unsigned long offset, size;
 } drm_sis_fb_t;
 
-struct sis_file_private {
-	struct list_head obj_list;
-};
-
 #endif				/* __SIS_DRM_H__ */
-- 
cgit v0.10.2


From c76af02d90ee9e9d2ef478fc6f874ad2abcf3ec9 Mon Sep 17 00:00:00 2001
From: Mikko Rapeli <mikko.rapeli@iki.fi>
Date: Thu, 15 Oct 2015 07:55:48 +0200
Subject: via_drm.h: move struct via_file_private definition to
 drivers/gpu/drm/via/via_drv.h

Fixes userspace compile error since list_head is not exported to userspace
headers.

Suggested by Emil Velikov <emil.l.velikov@gmail.com> at
https://lkml.org/lkml/2015/6/3/792

Signed-off-by: Mikko Rapeli <mikko.rapeli@iki.fi>
Signed-off-by: Dave Airlie <airlied@redhat.com>

diff --git a/drivers/gpu/drm/via/via_drv.h b/drivers/gpu/drm/via/via_drv.h
index 644093f..286a785 100644
--- a/drivers/gpu/drm/via/via_drv.h
+++ b/drivers/gpu/drm/via/via_drv.h
@@ -102,6 +102,10 @@ typedef struct drm_via_private {
 	uint32_t dma_diff;
 } drm_via_private_t;
 
+struct via_file_private {
+	struct list_head obj_list;
+};
+
 enum via_family {
   VIA_OTHER = 0,     /* Baseline */
   VIA_PRO_GROUP_A,   /* Another video engine and DMA commands */
diff --git a/include/uapi/drm/via_drm.h b/include/uapi/drm/via_drm.h
index 8b0533c..45bc80c 100644
--- a/include/uapi/drm/via_drm.h
+++ b/include/uapi/drm/via_drm.h
@@ -274,8 +274,4 @@ typedef struct drm_via_dmablit {
 	drm_via_blitsync_t sync;
 } drm_via_dmablit_t;
 
-struct via_file_private {
-	struct list_head obj_list;
-};
-
 #endif				/* _VIA_DRM_H_ */
-- 
cgit v0.10.2


From 9c73f4782642c785569ad50e01324002d160bd09 Mon Sep 17 00:00:00 2001
From: Gerd Hoffmann <kraxel@redhat.com>
Date: Wed, 19 Aug 2015 23:35:57 +0200
Subject: virtio-gpu: add virtio_gpu_queue_ctrl_buffer_locked

Add virtio_gpu_queue_ctrl_buffer_locked function, which does the same as
virtio_gpu_queue_ctrl_buffer but does not take the virtqueue lock.  The
caller must hold the lock instead.

Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>

diff --git a/drivers/gpu/drm/virtio/virtgpu_vq.c b/drivers/gpu/drm/virtio/virtgpu_vq.c
index 1698669f..5b9bc24 100644
--- a/drivers/gpu/drm/virtio/virtgpu_vq.c
+++ b/drivers/gpu/drm/virtio/virtgpu_vq.c
@@ -293,8 +293,8 @@ void virtio_gpu_dequeue_cursor_func(struct work_struct *work)
 	wake_up(&vgdev->cursorq.ack_queue);
 }
 
-static int virtio_gpu_queue_ctrl_buffer(struct virtio_gpu_device *vgdev,
-					struct virtio_gpu_vbuffer *vbuf)
+static int virtio_gpu_queue_ctrl_buffer_locked(struct virtio_gpu_device *vgdev,
+					       struct virtio_gpu_vbuffer *vbuf)
 {
 	struct virtqueue *vq = vgdev->ctrlq.vq;
 	struct scatterlist *sgs[3], vcmd, vout, vresp;
@@ -320,7 +320,6 @@ static int virtio_gpu_queue_ctrl_buffer(struct virtio_gpu_device *vgdev,
 		incnt++;
 	}
 
-	spin_lock(&vgdev->ctrlq.qlock);
 retry:
 	ret = virtqueue_add_sgs(vq, sgs, outcnt, incnt, vbuf, GFP_ATOMIC);
 	if (ret == -ENOSPC) {
@@ -331,13 +330,23 @@ retry:
 	} else {
 		virtqueue_kick(vq);
 	}
-	spin_unlock(&vgdev->ctrlq.qlock);
 
 	if (!ret)
 		ret = vq->num_free;
 	return ret;
 }
 
+static int virtio_gpu_queue_ctrl_buffer(struct virtio_gpu_device *vgdev,
+					struct virtio_gpu_vbuffer *vbuf)
+{
+	int rc;
+
+	spin_lock(&vgdev->ctrlq.qlock);
+	rc = virtio_gpu_queue_ctrl_buffer_locked(vgdev, vbuf);
+	spin_unlock(&vgdev->ctrlq.qlock);
+	return rc;
+}
+
 static int virtio_gpu_queue_cursor(struct virtio_gpu_device *vgdev,
 				   struct virtio_gpu_vbuffer *vbuf)
 {
-- 
cgit v0.10.2


From ec2f0577c7b1fccc7a5d7ee8002a4f162061498f Mon Sep 17 00:00:00 2001
From: Gerd Hoffmann <kraxel@redhat.com>
Date: Wed, 19 Aug 2015 23:44:15 +0200
Subject: virtio-gpu: add & use virtio_gpu_queue_fenced_ctrl_buffer

Add helper function to handle the submission of fenced control requests.
Make sure we initialize the fence while holding the virtqueue lock, so
requests can't be reordered.

Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>

diff --git a/drivers/gpu/drm/virtio/virtgpu_fence.c b/drivers/gpu/drm/virtio/virtgpu_fence.c
index 1da6326..793ad9f 100644
--- a/drivers/gpu/drm/virtio/virtgpu_fence.c
+++ b/drivers/gpu/drm/virtio/virtgpu_fence.c
@@ -81,7 +81,7 @@ int virtio_gpu_fence_emit(struct virtio_gpu_device *vgdev,
 	struct virtio_gpu_fence_driver *drv = &vgdev->fence_drv;
 	unsigned long irq_flags;
 
-	*fence = kmalloc(sizeof(struct virtio_gpu_fence), GFP_KERNEL);
+	*fence = kmalloc(sizeof(struct virtio_gpu_fence), GFP_ATOMIC);
 	if ((*fence) == NULL)
 		return -ENOMEM;
 
diff --git a/drivers/gpu/drm/virtio/virtgpu_vq.c b/drivers/gpu/drm/virtio/virtgpu_vq.c
index 5b9bc24..ee25e9a 100644
--- a/drivers/gpu/drm/virtio/virtgpu_vq.c
+++ b/drivers/gpu/drm/virtio/virtgpu_vq.c
@@ -347,6 +347,38 @@ static int virtio_gpu_queue_ctrl_buffer(struct virtio_gpu_device *vgdev,
 	return rc;
 }
 
+static int virtio_gpu_queue_fenced_ctrl_buffer(struct virtio_gpu_device *vgdev,
+					       struct virtio_gpu_vbuffer *vbuf,
+					       struct virtio_gpu_ctrl_hdr *hdr,
+					       struct virtio_gpu_fence **fence)
+{
+	struct virtqueue *vq = vgdev->ctrlq.vq;
+	int rc;
+
+again:
+	spin_lock(&vgdev->ctrlq.qlock);
+
+	/*
+	 * Make sure we have enouth space in the virtqueue.  If not
+	 * wait here until we have.
+	 *
+	 * Without that virtio_gpu_queue_ctrl_buffer_nolock might have
+	 * to wait for free space, which can result in fence ids being
+	 * submitted out-of-order.
+	 */
+	if (vq->num_free < 3) {
+		spin_unlock(&vgdev->ctrlq.qlock);
+		wait_event(vgdev->ctrlq.ack_queue, vq->num_free >= 3);
+		goto again;
+	}
+
+	if (fence)
+		virtio_gpu_fence_emit(vgdev, hdr, fence);
+	rc = virtio_gpu_queue_ctrl_buffer_locked(vgdev, vbuf);
+	spin_unlock(&vgdev->ctrlq.qlock);
+	return rc;
+}
+
 static int virtio_gpu_queue_cursor(struct virtio_gpu_device *vgdev,
 				   struct virtio_gpu_vbuffer *vbuf)
 {
@@ -499,9 +531,7 @@ void virtio_gpu_cmd_transfer_to_host_2d(struct virtio_gpu_device *vgdev,
 	cmd_p->r.x = x;
 	cmd_p->r.y = y;
 
-	if (fence)
-		virtio_gpu_fence_emit(vgdev, &cmd_p->hdr, fence);
-	virtio_gpu_queue_ctrl_buffer(vgdev, vbuf);
+	virtio_gpu_queue_fenced_ctrl_buffer(vgdev, vbuf, &cmd_p->hdr, fence);
 }
 
 static void
@@ -524,9 +554,7 @@ virtio_gpu_cmd_resource_attach_backing(struct virtio_gpu_device *vgdev,
 	vbuf->data_buf = ents;
 	vbuf->data_size = sizeof(*ents) * nents;
 
-	if (fence)
-		virtio_gpu_fence_emit(vgdev, &cmd_p->hdr, fence);
-	virtio_gpu_queue_ctrl_buffer(vgdev, vbuf);
+	virtio_gpu_queue_fenced_ctrl_buffer(vgdev, vbuf, &cmd_p->hdr, fence);
 }
 
 static void virtio_gpu_cmd_get_display_info_cb(struct virtio_gpu_device *vgdev,
-- 
cgit v0.10.2


From 6d41533433e2eff26005ee39d103948fa65dd891 Mon Sep 17 00:00:00 2001
From: Gerd Hoffmann <kraxel@redhat.com>
Date: Tue, 15 Sep 2015 08:20:46 +0200
Subject: virtio-gpu: wait for cursor updates finish

Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>

diff --git a/drivers/gpu/drm/virtio/virtgpu_display.c b/drivers/gpu/drm/virtio/virtgpu_display.c
index 4e160ef..c9c1427 100644
--- a/drivers/gpu/drm/virtio/virtgpu_display.c
+++ b/drivers/gpu/drm/virtio/virtgpu_display.c
@@ -90,6 +90,14 @@ static int virtio_gpu_crtc_cursor_set(struct drm_crtc *crtc,
 					   cpu_to_le32(64),
 					   cpu_to_le32(64),
 					   0, 0, &fence);
+	ret = virtio_gpu_object_reserve(qobj, false);
+	if (!ret) {
+		reservation_object_add_excl_fence(qobj->tbo.resv,
+						  &fence->f);
+		fence_put(&fence->f);
+		virtio_gpu_object_unreserve(qobj);
+		virtio_gpu_object_wait(qobj, false);
+	}
 
 	output->cursor.hdr.type = cpu_to_le32(VIRTIO_GPU_CMD_UPDATE_CURSOR);
 	output->cursor.resource_id = cpu_to_le32(qobj->hw_res_handle);
-- 
cgit v0.10.2


From 7552ed8a1a810552e16664d7020d61d2b01b9199 Mon Sep 17 00:00:00 2001
From: Dave Airlie <airlied@redhat.com>
Date: Tue, 13 Oct 2015 16:55:48 +1000
Subject: virtio-gpu: don't free things on ttm_bo_init failure

ttm_bo_init will call the destroy callback which will do all
this, this was causing a double free.

Signed-off-by: Dave Airlie <airlied@redhat.com>

diff --git a/drivers/gpu/drm/virtio/virtgpu_object.c b/drivers/gpu/drm/virtio/virtgpu_object.c
index 2c624c7..f300eba 100644
--- a/drivers/gpu/drm/virtio/virtgpu_object.c
+++ b/drivers/gpu/drm/virtio/virtgpu_object.c
@@ -82,24 +82,19 @@ int virtio_gpu_object_create(struct virtio_gpu_device *vgdev,
 	size = roundup(size, PAGE_SIZE);
 	ret = drm_gem_object_init(vgdev->ddev, &bo->gem_base, size);
 	if (ret != 0)
-		goto err_gem_init;
+		return ret;
 	bo->dumb = false;
 	virtio_gpu_init_ttm_placement(bo, pinned);
 
 	ret = ttm_bo_init(&vgdev->mman.bdev, &bo->tbo, size, type,
 			  &bo->placement, 0, !kernel, NULL, acc_size,
 			  NULL, NULL, &virtio_gpu_ttm_bo_destroy);
+	/* ttm_bo_init failure will call the destroy */
 	if (ret != 0)
-		goto err_ttm_init;
+		return ret;
 
 	*bo_ptr = bo;
 	return 0;
-
-err_ttm_init:
-	drm_gem_object_release(&bo->gem_base);
-err_gem_init:
-	kfree(bo);
-	return ret;
 }
 
 int virtio_gpu_object_kmap(struct virtio_gpu_object *bo, void **ptr)
-- 
cgit v0.10.2


From 62fb7a5e10962ac6ae2a2d2dbd3aedcb2a3e3257 Mon Sep 17 00:00:00 2001
From: Gerd Hoffmann <kraxel@redhat.com>
Date: Tue, 28 Oct 2014 12:48:00 +0100
Subject: virtio-gpu: add 3d/virgl support

Add the bits needed for opengl rendering support: query
capabilities, new virtio commands, drm ioctls.

Signed-off-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>

diff --git a/drivers/gpu/drm/virtio/Makefile b/drivers/gpu/drm/virtio/Makefile
index 2ee1602..da7bf19 100644
--- a/drivers/gpu/drm/virtio/Makefile
+++ b/drivers/gpu/drm/virtio/Makefile
@@ -6,6 +6,7 @@ ccflags-y := -Iinclude/drm
 
 virtio-gpu-y := virtgpu_drv.o virtgpu_kms.o virtgpu_drm_bus.o virtgpu_gem.o \
 	virtgpu_fb.o virtgpu_display.o virtgpu_vq.o virtgpu_ttm.o \
-	virtgpu_fence.o virtgpu_object.o virtgpu_debugfs.o virtgpu_plane.o
+	virtgpu_fence.o virtgpu_object.o virtgpu_debugfs.o virtgpu_plane.o \
+	virtgpu_ioctl.o
 
 obj-$(CONFIG_DRM_VIRTIO_GPU) += virtio-gpu.o
diff --git a/drivers/gpu/drm/virtio/virtgpu_drv.c b/drivers/gpu/drm/virtio/virtgpu_drv.c
index 7d9610a..c27d3a3 100644
--- a/drivers/gpu/drm/virtio/virtgpu_drv.c
+++ b/drivers/gpu/drm/virtio/virtgpu_drv.c
@@ -73,6 +73,14 @@ static struct virtio_device_id id_table[] = {
 };
 
 static unsigned int features[] = {
+#ifdef __LITTLE_ENDIAN
+	/*
+	 * Gallium command stream send by virgl is native endian.
+	 * Because of that we only support little endian guests on
+	 * little endian hosts.
+	 */
+	VIRTIO_GPU_F_VIRGL,
+#endif
 };
 static struct virtio_driver virtio_gpu_driver = {
 	.feature_table = features,
@@ -114,6 +122,8 @@ static struct drm_driver driver = {
 	.set_busid = drm_virtio_set_busid,
 	.load = virtio_gpu_driver_load,
 	.unload = virtio_gpu_driver_unload,
+	.open = virtio_gpu_driver_open,
+	.postclose = virtio_gpu_driver_postclose,
 
 	.dumb_create = virtio_gpu_mode_dumb_create,
 	.dumb_map_offset = virtio_gpu_mode_dumb_mmap,
@@ -125,8 +135,13 @@ static struct drm_driver driver = {
 #endif
 
 	.gem_free_object = virtio_gpu_gem_free_object,
+	.gem_open_object = virtio_gpu_gem_object_open,
+	.gem_close_object = virtio_gpu_gem_object_close,
 	.fops = &virtio_gpu_driver_fops,
 
+	.ioctls = virtio_gpu_ioctls,
+	.num_ioctls = DRM_VIRTIO_NUM_IOCTLS,
+
 	.name = DRIVER_NAME,
 	.desc = DRIVER_DESC,
 	.date = DRIVER_DATE,
diff --git a/drivers/gpu/drm/virtio/virtgpu_drv.h b/drivers/gpu/drm/virtio/virtgpu_drv.h
index 6d4db2d..2719108 100644
--- a/drivers/gpu/drm/virtio/virtgpu_drv.h
+++ b/drivers/gpu/drm/virtio/virtgpu_drv.h
@@ -146,6 +146,21 @@ struct virtio_gpu_queue {
 	struct work_struct dequeue_work;
 };
 
+struct virtio_gpu_drv_capset {
+	uint32_t id;
+	uint32_t max_version;
+	uint32_t max_size;
+};
+
+struct virtio_gpu_drv_cap_cache {
+	struct list_head head;
+	void *caps_cache;
+	uint32_t id;
+	uint32_t version;
+	uint32_t size;
+	atomic_t is_valid;
+};
+
 struct virtio_gpu_device {
 	struct device *dev;
 	struct drm_device *ddev;
@@ -179,7 +194,13 @@ struct virtio_gpu_device {
 	struct idr	ctx_id_idr;
 	spinlock_t ctx_id_idr_lock;
 
+	bool has_virgl_3d;
+
 	struct work_struct config_changed_work;
+
+	struct virtio_gpu_drv_capset *capsets;
+	uint32_t num_capsets;
+	struct list_head cap_cache;
 };
 
 struct virtio_gpu_fpriv {
@@ -193,6 +214,8 @@ extern struct drm_ioctl_desc virtio_gpu_ioctls[DRM_VIRTIO_NUM_IOCTLS];
 /* virtio_kms.c */
 int virtio_gpu_driver_load(struct drm_device *dev, unsigned long flags);
 int virtio_gpu_driver_unload(struct drm_device *dev);
+int virtio_gpu_driver_open(struct drm_device *dev, struct drm_file *file);
+void virtio_gpu_driver_postclose(struct drm_device *dev, struct drm_file *file);
 
 /* virtio_gem.c */
 void virtio_gpu_gem_free_object(struct drm_gem_object *gem_obj);
@@ -203,6 +226,10 @@ int virtio_gpu_gem_create(struct drm_file *file,
 			  uint64_t size,
 			  struct drm_gem_object **obj_p,
 			  uint32_t *handle_p);
+int virtio_gpu_gem_object_open(struct drm_gem_object *obj,
+			       struct drm_file *file);
+void virtio_gpu_gem_object_close(struct drm_gem_object *obj,
+				 struct drm_file *file);
 struct virtio_gpu_object *virtio_gpu_alloc_object(struct drm_device *dev,
 						  size_t size, bool kernel,
 						  bool pinned);
@@ -260,10 +287,43 @@ void virtio_gpu_cursor_ping(struct virtio_gpu_device *vgdev,
 int virtio_gpu_cmd_get_display_info(struct virtio_gpu_device *vgdev);
 void virtio_gpu_cmd_resource_inval_backing(struct virtio_gpu_device *vgdev,
 					   uint32_t resource_id);
+int virtio_gpu_cmd_get_capset_info(struct virtio_gpu_device *vgdev, int idx);
+int virtio_gpu_cmd_get_capset(struct virtio_gpu_device *vgdev,
+			      int idx, int version,
+			      struct virtio_gpu_drv_cap_cache **cache_p);
+void virtio_gpu_cmd_context_create(struct virtio_gpu_device *vgdev, uint32_t id,
+				   uint32_t nlen, const char *name);
+void virtio_gpu_cmd_context_destroy(struct virtio_gpu_device *vgdev,
+				    uint32_t id);
+void virtio_gpu_cmd_context_attach_resource(struct virtio_gpu_device *vgdev,
+					    uint32_t ctx_id,
+					    uint32_t resource_id);
+void virtio_gpu_cmd_context_detach_resource(struct virtio_gpu_device *vgdev,
+					    uint32_t ctx_id,
+					    uint32_t resource_id);
+void virtio_gpu_cmd_submit(struct virtio_gpu_device *vgdev,
+			   void *data, uint32_t data_size,
+			   uint32_t ctx_id, struct virtio_gpu_fence **fence);
+void virtio_gpu_cmd_transfer_from_host_3d(struct virtio_gpu_device *vgdev,
+					  uint32_t resource_id, uint32_t ctx_id,
+					  uint64_t offset, uint32_t level,
+					  struct virtio_gpu_box *box,
+					  struct virtio_gpu_fence **fence);
+void virtio_gpu_cmd_transfer_to_host_3d(struct virtio_gpu_device *vgdev,
+					uint32_t resource_id, uint32_t ctx_id,
+					uint64_t offset, uint32_t level,
+					struct virtio_gpu_box *box,
+					struct virtio_gpu_fence **fence);
+void
+virtio_gpu_cmd_resource_create_3d(struct virtio_gpu_device *vgdev,
+				  struct virtio_gpu_resource_create_3d *rc_3d,
+				  struct virtio_gpu_fence **fence);
 void virtio_gpu_ctrl_ack(struct virtqueue *vq);
 void virtio_gpu_cursor_ack(struct virtqueue *vq);
+void virtio_gpu_fence_ack(struct virtqueue *vq);
 void virtio_gpu_dequeue_ctrl_func(struct work_struct *work);
 void virtio_gpu_dequeue_cursor_func(struct work_struct *work);
+void virtio_gpu_dequeue_fence_func(struct work_struct *work);
 
 /* virtio_gpu_display.c */
 int virtio_gpu_framebuffer_init(struct drm_device *dev,
diff --git a/drivers/gpu/drm/virtio/virtgpu_gem.c b/drivers/gpu/drm/virtio/virtgpu_gem.c
index cfa0d27..1feb7ce 100644
--- a/drivers/gpu/drm/virtio/virtgpu_gem.c
+++ b/drivers/gpu/drm/virtio/virtgpu_gem.c
@@ -138,3 +138,44 @@ int virtio_gpu_mode_dumb_mmap(struct drm_file *file_priv,
 	drm_gem_object_unreference_unlocked(gobj);
 	return 0;
 }
+
+int virtio_gpu_gem_object_open(struct drm_gem_object *obj,
+			       struct drm_file *file)
+{
+	struct virtio_gpu_device *vgdev = obj->dev->dev_private;
+	struct virtio_gpu_fpriv *vfpriv = file->driver_priv;
+	struct virtio_gpu_object *qobj = gem_to_virtio_gpu_obj(obj);
+	int r;
+
+	if (!vgdev->has_virgl_3d)
+		return 0;
+
+	r = virtio_gpu_object_reserve(qobj, false);
+	if (r)
+		return r;
+
+	virtio_gpu_cmd_context_attach_resource(vgdev, vfpriv->ctx_id,
+					       qobj->hw_res_handle);
+	virtio_gpu_object_unreserve(qobj);
+	return 0;
+}
+
+void virtio_gpu_gem_object_close(struct drm_gem_object *obj,
+				 struct drm_file *file)
+{
+	struct virtio_gpu_device *vgdev = obj->dev->dev_private;
+	struct virtio_gpu_fpriv *vfpriv = file->driver_priv;
+	struct virtio_gpu_object *qobj = gem_to_virtio_gpu_obj(obj);
+	int r;
+
+	if (!vgdev->has_virgl_3d)
+		return;
+
+	r = virtio_gpu_object_reserve(qobj, false);
+	if (r)
+		return;
+
+	virtio_gpu_cmd_context_detach_resource(vgdev, vfpriv->ctx_id,
+						qobj->hw_res_handle);
+	virtio_gpu_object_unreserve(qobj);
+}
diff --git a/drivers/gpu/drm/virtio/virtgpu_ioctl.c b/drivers/gpu/drm/virtio/virtgpu_ioctl.c
new file mode 100644
index 0000000..4ef672b
--- /dev/null
+++ b/drivers/gpu/drm/virtio/virtgpu_ioctl.c
@@ -0,0 +1,573 @@
+/*
+ * Copyright (C) 2015 Red Hat, Inc.
+ * All Rights Reserved.
+ *
+ * Authors:
+ *    Dave Airlie
+ *    Alon Levy
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#include <drm/drmP.h>
+#include "virtgpu_drv.h"
+#include <drm/virtgpu_drm.h>
+#include "ttm/ttm_execbuf_util.h"
+
+static void convert_to_hw_box(struct virtio_gpu_box *dst,
+			      const struct drm_virtgpu_3d_box *src)
+{
+	dst->x = cpu_to_le32(src->x);
+	dst->y = cpu_to_le32(src->y);
+	dst->z = cpu_to_le32(src->z);
+	dst->w = cpu_to_le32(src->w);
+	dst->h = cpu_to_le32(src->h);
+	dst->d = cpu_to_le32(src->d);
+}
+
+static int virtio_gpu_map_ioctl(struct drm_device *dev, void *data,
+				struct drm_file *file_priv)
+{
+	struct virtio_gpu_device *vgdev = dev->dev_private;
+	struct drm_virtgpu_map *virtio_gpu_map = data;
+
+	return virtio_gpu_mode_dumb_mmap(file_priv, vgdev->ddev,
+					 virtio_gpu_map->handle,
+					 &virtio_gpu_map->offset);
+}
+
+static int virtio_gpu_object_list_validate(struct ww_acquire_ctx *ticket,
+					   struct list_head *head)
+{
+	struct ttm_validate_buffer *buf;
+	struct ttm_buffer_object *bo;
+	struct virtio_gpu_object *qobj;
+	int ret;
+
+	ret = ttm_eu_reserve_buffers(ticket, head, true, NULL);
+	if (ret != 0)
+		return ret;
+
+	list_for_each_entry(buf, head, head) {
+		bo = buf->bo;
+		qobj = container_of(bo, struct virtio_gpu_object, tbo);
+		ret = ttm_bo_validate(bo, &qobj->placement, false, false);
+		if (ret) {
+			ttm_eu_backoff_reservation(ticket, head);
+			return ret;
+		}
+	}
+	return 0;
+}
+
+static void virtio_gpu_unref_list(struct list_head *head)
+{
+	struct ttm_validate_buffer *buf;
+	struct ttm_buffer_object *bo;
+	struct virtio_gpu_object *qobj;
+	list_for_each_entry(buf, head, head) {
+		bo = buf->bo;
+		qobj = container_of(bo, struct virtio_gpu_object, tbo);
+
+		drm_gem_object_unreference_unlocked(&qobj->gem_base);
+	}
+}
+
+static int virtio_gpu_execbuffer(struct drm_device *dev,
+				 struct drm_virtgpu_execbuffer *exbuf,
+				 struct drm_file *drm_file)
+{
+	struct virtio_gpu_device *vgdev = dev->dev_private;
+	struct virtio_gpu_fpriv *vfpriv = drm_file->driver_priv;
+	struct drm_gem_object *gobj;
+	struct virtio_gpu_fence *fence;
+	struct virtio_gpu_object *qobj;
+	int ret;
+	uint32_t *bo_handles = NULL;
+	void __user *user_bo_handles = NULL;
+	struct list_head validate_list;
+	struct ttm_validate_buffer *buflist = NULL;
+	int i;
+	struct ww_acquire_ctx ticket;
+	void *buf;
+
+	if (vgdev->has_virgl_3d == false)
+		return -ENOSYS;
+
+	INIT_LIST_HEAD(&validate_list);
+	if (exbuf->num_bo_handles) {
+
+		bo_handles = drm_malloc_ab(exbuf->num_bo_handles,
+					   sizeof(uint32_t));
+		buflist = drm_calloc_large(exbuf->num_bo_handles,
+					   sizeof(struct ttm_validate_buffer));
+		if (!bo_handles || !buflist) {
+			drm_free_large(bo_handles);
+			drm_free_large(buflist);
+			return -ENOMEM;
+		}
+
+		user_bo_handles = (void __user *)(uintptr_t)exbuf->bo_handles;
+		if (copy_from_user(bo_handles, user_bo_handles,
+				   exbuf->num_bo_handles * sizeof(uint32_t))) {
+			ret = -EFAULT;
+			drm_free_large(bo_handles);
+			drm_free_large(buflist);
+			return ret;
+		}
+
+		for (i = 0; i < exbuf->num_bo_handles; i++) {
+			gobj = drm_gem_object_lookup(dev,
+						     drm_file, bo_handles[i]);
+			if (!gobj) {
+				drm_free_large(bo_handles);
+				drm_free_large(buflist);
+				return -ENOENT;
+			}
+
+			qobj = gem_to_virtio_gpu_obj(gobj);
+			buflist[i].bo = &qobj->tbo;
+
+			list_add(&buflist[i].head, &validate_list);
+		}
+		drm_free_large(bo_handles);
+	}
+
+	ret = virtio_gpu_object_list_validate(&ticket, &validate_list);
+	if (ret)
+		goto out_free;
+
+	buf = kmalloc(exbuf->size, GFP_KERNEL);
+	if (!buf) {
+		ret = -ENOMEM;
+		goto out_unresv;
+	}
+	if (copy_from_user(buf, (void __user *)(uintptr_t)exbuf->command,
+			   exbuf->size)) {
+		kfree(buf);
+		ret = -EFAULT;
+		goto out_unresv;
+	}
+	virtio_gpu_cmd_submit(vgdev, buf, exbuf->size,
+			      vfpriv->ctx_id, &fence);
+
+	ttm_eu_fence_buffer_objects(&ticket, &validate_list, &fence->f);
+
+	/* fence the command bo */
+	virtio_gpu_unref_list(&validate_list);
+	drm_free_large(buflist);
+	fence_put(&fence->f);
+	return 0;
+
+out_unresv:
+	ttm_eu_backoff_reservation(&ticket, &validate_list);
+out_free:
+	virtio_gpu_unref_list(&validate_list);
+	drm_free_large(buflist);
+	return ret;
+}
+
+/*
+ * Usage of execbuffer:
+ * Relocations need to take into account the full VIRTIO_GPUDrawable size.
+ * However, the command as passed from user space must *not* contain the initial
+ * VIRTIO_GPUReleaseInfo struct (first XXX bytes)
+ */
+static int virtio_gpu_execbuffer_ioctl(struct drm_device *dev, void *data,
+				       struct drm_file *file_priv)
+{
+	struct drm_virtgpu_execbuffer *execbuffer = data;
+	return virtio_gpu_execbuffer(dev, execbuffer, file_priv);
+}
+
+
+static int virtio_gpu_getparam_ioctl(struct drm_device *dev, void *data,
+				     struct drm_file *file_priv)
+{
+	struct virtio_gpu_device *vgdev = dev->dev_private;
+	struct drm_virtgpu_getparam *param = data;
+	int value;
+
+	switch (param->param) {
+	case VIRTGPU_PARAM_3D_FEATURES:
+		value = vgdev->has_virgl_3d == true ? 1 : 0;
+		break;
+	default:
+		return -EINVAL;
+	}
+	if (copy_to_user((void __user *)(unsigned long)param->value,
+			 &value, sizeof(int))) {
+		return -EFAULT;
+	}
+	return 0;
+}
+
+static int virtio_gpu_resource_create_ioctl(struct drm_device *dev, void *data,
+					    struct drm_file *file_priv)
+{
+	struct virtio_gpu_device *vgdev = dev->dev_private;
+	struct drm_virtgpu_resource_create *rc = data;
+	int ret;
+	uint32_t res_id;
+	struct virtio_gpu_object *qobj;
+	struct drm_gem_object *obj;
+	uint32_t handle = 0;
+	uint32_t size;
+	struct list_head validate_list;
+	struct ttm_validate_buffer mainbuf;
+	struct virtio_gpu_fence *fence = NULL;
+	struct ww_acquire_ctx ticket;
+	struct virtio_gpu_resource_create_3d rc_3d;
+
+	if (vgdev->has_virgl_3d == false) {
+		if (rc->depth > 1)
+			return -EINVAL;
+		if (rc->nr_samples > 1)
+			return -EINVAL;
+		if (rc->last_level > 1)
+			return -EINVAL;
+		if (rc->target != 2)
+			return -EINVAL;
+		if (rc->array_size > 1)
+			return -EINVAL;
+	}
+
+	INIT_LIST_HEAD(&validate_list);
+	memset(&mainbuf, 0, sizeof(struct ttm_validate_buffer));
+
+	virtio_gpu_resource_id_get(vgdev, &res_id);
+
+	size = rc->size;
+
+	/* allocate a single page size object */
+	if (size == 0)
+		size = PAGE_SIZE;
+
+	qobj = virtio_gpu_alloc_object(dev, size, false, false);
+	if (IS_ERR(qobj)) {
+		ret = PTR_ERR(qobj);
+		goto fail_id;
+	}
+	obj = &qobj->gem_base;
+
+	if (!vgdev->has_virgl_3d) {
+		virtio_gpu_cmd_create_resource(vgdev, res_id, rc->format,
+					       rc->width, rc->height);
+
+		ret = virtio_gpu_object_attach(vgdev, qobj, res_id, NULL);
+	} else {
+		/* use a gem reference since unref list undoes them */
+		drm_gem_object_reference(&qobj->gem_base);
+		mainbuf.bo = &qobj->tbo;
+		list_add(&mainbuf.head, &validate_list);
+
+		ret = virtio_gpu_object_list_validate(&ticket, &validate_list);
+		if (ret) {
+			DRM_DEBUG("failed to validate\n");
+			goto fail_unref;
+		}
+
+		rc_3d.resource_id = cpu_to_le32(res_id);
+		rc_3d.target = cpu_to_le32(rc->target);
+		rc_3d.format = cpu_to_le32(rc->format);
+		rc_3d.bind = cpu_to_le32(rc->bind);
+		rc_3d.width = cpu_to_le32(rc->width);
+		rc_3d.height = cpu_to_le32(rc->height);
+		rc_3d.depth = cpu_to_le32(rc->depth);
+		rc_3d.array_size = cpu_to_le32(rc->array_size);
+		rc_3d.last_level = cpu_to_le32(rc->last_level);
+		rc_3d.nr_samples = cpu_to_le32(rc->nr_samples);
+		rc_3d.flags = cpu_to_le32(rc->flags);
+
+		virtio_gpu_cmd_resource_create_3d(vgdev, &rc_3d, NULL);
+		ret = virtio_gpu_object_attach(vgdev, qobj, res_id, &fence);
+		if (ret) {
+			ttm_eu_backoff_reservation(&ticket, &validate_list);
+			goto fail_unref;
+		}
+		ttm_eu_fence_buffer_objects(&ticket, &validate_list, &fence->f);
+	}
+
+	qobj->hw_res_handle = res_id;
+
+	ret = drm_gem_handle_create(file_priv, obj, &handle);
+	if (ret) {
+
+		drm_gem_object_release(obj);
+		if (vgdev->has_virgl_3d) {
+			virtio_gpu_unref_list(&validate_list);
+			fence_put(&fence->f);
+		}
+		return ret;
+	}
+	drm_gem_object_unreference_unlocked(obj);
+
+	rc->res_handle = res_id; /* similiar to a VM address */
+	rc->bo_handle = handle;
+
+	if (vgdev->has_virgl_3d) {
+		virtio_gpu_unref_list(&validate_list);
+		fence_put(&fence->f);
+	}
+	return 0;
+fail_unref:
+	if (vgdev->has_virgl_3d) {
+		virtio_gpu_unref_list(&validate_list);
+		fence_put(&fence->f);
+	}
+//fail_obj:
+//	drm_gem_object_handle_unreference_unlocked(obj);
+fail_id:
+	virtio_gpu_resource_id_put(vgdev, res_id);
+	return ret;
+}
+
+static int virtio_gpu_resource_info_ioctl(struct drm_device *dev, void *data,
+					  struct drm_file *file_priv)
+{
+	struct drm_virtgpu_resource_info *ri = data;
+	struct drm_gem_object *gobj = NULL;
+	struct virtio_gpu_object *qobj = NULL;
+
+	gobj = drm_gem_object_lookup(dev, file_priv, ri->bo_handle);
+	if (gobj == NULL)
+		return -ENOENT;
+
+	qobj = gem_to_virtio_gpu_obj(gobj);
+
+	ri->size = qobj->gem_base.size;
+	ri->res_handle = qobj->hw_res_handle;
+	drm_gem_object_unreference_unlocked(gobj);
+	return 0;
+}
+
+static int virtio_gpu_transfer_from_host_ioctl(struct drm_device *dev,
+					       void *data,
+					       struct drm_file *file)
+{
+	struct virtio_gpu_device *vgdev = dev->dev_private;
+	struct virtio_gpu_fpriv *vfpriv = file->driver_priv;
+	struct drm_virtgpu_3d_transfer_from_host *args = data;
+	struct drm_gem_object *gobj = NULL;
+	struct virtio_gpu_object *qobj = NULL;
+	struct virtio_gpu_fence *fence;
+	int ret;
+	u32 offset = args->offset;
+	struct virtio_gpu_box box;
+
+	if (vgdev->has_virgl_3d == false)
+		return -ENOSYS;
+
+	gobj = drm_gem_object_lookup(dev, file, args->bo_handle);
+	if (gobj == NULL)
+		return -ENOENT;
+
+	qobj = gem_to_virtio_gpu_obj(gobj);
+
+	ret = virtio_gpu_object_reserve(qobj, false);
+	if (ret)
+		goto out;
+
+	ret = ttm_bo_validate(&qobj->tbo, &qobj->placement,
+			      true, false);
+	if (unlikely(ret))
+		goto out_unres;
+
+	convert_to_hw_box(&box, &args->box);
+	virtio_gpu_cmd_transfer_from_host_3d
+		(vgdev, qobj->hw_res_handle,
+		 vfpriv->ctx_id, offset, args->level,
+		 &box, &fence);
+	reservation_object_add_excl_fence(qobj->tbo.resv,
+					  &fence->f);
+
+	fence_put(&fence->f);
+out_unres:
+	virtio_gpu_object_unreserve(qobj);
+out:
+	drm_gem_object_unreference_unlocked(gobj);
+	return ret;
+}
+
+static int virtio_gpu_transfer_to_host_ioctl(struct drm_device *dev, void *data,
+					     struct drm_file *file)
+{
+	struct virtio_gpu_device *vgdev = dev->dev_private;
+	struct virtio_gpu_fpriv *vfpriv = file->driver_priv;
+	struct drm_virtgpu_3d_transfer_to_host *args = data;
+	struct drm_gem_object *gobj = NULL;
+	struct virtio_gpu_object *qobj = NULL;
+	struct virtio_gpu_fence *fence;
+	struct virtio_gpu_box box;
+	int ret;
+	u32 offset = args->offset;
+
+	gobj = drm_gem_object_lookup(dev, file, args->bo_handle);
+	if (gobj == NULL)
+		return -ENOENT;
+
+	qobj = gem_to_virtio_gpu_obj(gobj);
+
+	ret = virtio_gpu_object_reserve(qobj, false);
+	if (ret)
+		goto out;
+
+	ret = ttm_bo_validate(&qobj->tbo, &qobj->placement,
+			      true, false);
+	if (unlikely(ret))
+		goto out_unres;
+
+	convert_to_hw_box(&box, &args->box);
+	if (!vgdev->has_virgl_3d) {
+		virtio_gpu_cmd_transfer_to_host_2d
+			(vgdev, qobj->hw_res_handle, offset,
+			 box.w, box.h, box.x, box.y, NULL);
+	} else {
+		virtio_gpu_cmd_transfer_to_host_3d
+			(vgdev, qobj->hw_res_handle,
+			 vfpriv ? vfpriv->ctx_id : 0, offset,
+			 args->level, &box, &fence);
+		reservation_object_add_excl_fence(qobj->tbo.resv,
+						  &fence->f);
+		fence_put(&fence->f);
+	}
+
+out_unres:
+	virtio_gpu_object_unreserve(qobj);
+out:
+	drm_gem_object_unreference_unlocked(gobj);
+	return ret;
+}
+
+static int virtio_gpu_wait_ioctl(struct drm_device *dev, void *data,
+			    struct drm_file *file)
+{
+	struct drm_virtgpu_3d_wait *args = data;
+	struct drm_gem_object *gobj = NULL;
+	struct virtio_gpu_object *qobj = NULL;
+	int ret;
+	bool nowait = false;
+
+	gobj = drm_gem_object_lookup(dev, file, args->handle);
+	if (gobj == NULL)
+		return -ENOENT;
+
+	qobj = gem_to_virtio_gpu_obj(gobj);
+
+	if (args->flags & VIRTGPU_WAIT_NOWAIT)
+		nowait = true;
+	ret = virtio_gpu_object_wait(qobj, nowait);
+
+	drm_gem_object_unreference_unlocked(gobj);
+	return ret;
+}
+
+static int virtio_gpu_get_caps_ioctl(struct drm_device *dev,
+				void *data, struct drm_file *file)
+{
+	struct virtio_gpu_device *vgdev = dev->dev_private;
+	struct drm_virtgpu_get_caps *args = data;
+	int size;
+	int i;
+	int found_valid = -1;
+	int ret;
+	struct virtio_gpu_drv_cap_cache *cache_ent;
+	void *ptr;
+	if (vgdev->num_capsets == 0)
+		return -ENOSYS;
+
+	spin_lock(&vgdev->display_info_lock);
+	for (i = 0; i < vgdev->num_capsets; i++) {
+		if (vgdev->capsets[i].id == args->cap_set_id) {
+			if (vgdev->capsets[i].max_version >= args->cap_set_ver) {
+				found_valid = i;
+				break;
+			}
+		}
+	}
+
+	if (found_valid == -1) {
+		spin_unlock(&vgdev->display_info_lock);
+		return -EINVAL;
+	}
+
+	size = vgdev->capsets[found_valid].max_size;
+	if (args->size > size) {
+		spin_unlock(&vgdev->display_info_lock);
+		return -EINVAL;
+	}
+
+	list_for_each_entry(cache_ent, &vgdev->cap_cache, head) {
+		if (cache_ent->id == args->cap_set_id &&
+		    cache_ent->version == args->cap_set_ver) {
+			ptr = cache_ent->caps_cache;
+			spin_unlock(&vgdev->display_info_lock);
+			goto copy_exit;
+		}
+	}
+	spin_unlock(&vgdev->display_info_lock);
+
+	/* not in cache - need to talk to hw */
+	virtio_gpu_cmd_get_capset(vgdev, found_valid, args->cap_set_ver,
+				  &cache_ent);
+
+	ret = wait_event_timeout(vgdev->resp_wq,
+				 atomic_read(&cache_ent->is_valid), 5 * HZ);
+
+	ptr = cache_ent->caps_cache;
+
+copy_exit:
+	if (copy_to_user((void __user *)(unsigned long)args->addr, ptr, size))
+		return -EFAULT;
+
+	return 0;
+}
+
+struct drm_ioctl_desc virtio_gpu_ioctls[DRM_VIRTIO_NUM_IOCTLS] = {
+	DRM_IOCTL_DEF_DRV(VIRTGPU_MAP, virtio_gpu_map_ioctl,
+			  DRM_AUTH|DRM_UNLOCKED),
+
+	DRM_IOCTL_DEF_DRV(VIRTGPU_EXECBUFFER, virtio_gpu_execbuffer_ioctl,
+			  DRM_AUTH|DRM_UNLOCKED),
+
+	DRM_IOCTL_DEF_DRV(VIRTGPU_GETPARAM, virtio_gpu_getparam_ioctl,
+			  DRM_AUTH|DRM_UNLOCKED),
+
+	DRM_IOCTL_DEF_DRV(VIRTGPU_RESOURCE_CREATE,
+			  virtio_gpu_resource_create_ioctl,
+			  DRM_AUTH|DRM_UNLOCKED),
+
+	DRM_IOCTL_DEF_DRV(VIRTGPU_RESOURCE_INFO, virtio_gpu_resource_info_ioctl,
+			  DRM_AUTH|DRM_UNLOCKED),
+
+	/* make transfer async to the main ring? - no sure, can we
+	   thread these in the underlying GL */
+	DRM_IOCTL_DEF_DRV(VIRTGPU_TRANSFER_FROM_HOST,
+			  virtio_gpu_transfer_from_host_ioctl,
+			  DRM_AUTH|DRM_UNLOCKED),
+	DRM_IOCTL_DEF_DRV(VIRTGPU_TRANSFER_TO_HOST,
+			  virtio_gpu_transfer_to_host_ioctl,
+			  DRM_AUTH|DRM_UNLOCKED),
+
+	DRM_IOCTL_DEF_DRV(VIRTGPU_WAIT, virtio_gpu_wait_ioctl,
+			  DRM_AUTH|DRM_UNLOCKED),
+
+	DRM_IOCTL_DEF_DRV(VIRTGPU_GET_CAPS, virtio_gpu_get_caps_ioctl,
+			  DRM_AUTH|DRM_UNLOCKED),
+};
diff --git a/drivers/gpu/drm/virtio/virtgpu_kms.c b/drivers/gpu/drm/virtio/virtgpu_kms.c
index 782766c..06496a1 100644
--- a/drivers/gpu/drm/virtio/virtgpu_kms.c
+++ b/drivers/gpu/drm/virtio/virtgpu_kms.c
@@ -52,6 +52,41 @@ static void virtio_gpu_config_changed_work_func(struct work_struct *work)
 		      events_clear, &events_clear);
 }
 
+static void virtio_gpu_ctx_id_get(struct virtio_gpu_device *vgdev,
+				  uint32_t *resid)
+{
+	int handle;
+
+	idr_preload(GFP_KERNEL);
+	spin_lock(&vgdev->ctx_id_idr_lock);
+	handle = idr_alloc(&vgdev->ctx_id_idr, NULL, 1, 0, 0);
+	spin_unlock(&vgdev->ctx_id_idr_lock);
+	idr_preload_end();
+	*resid = handle;
+}
+
+static void virtio_gpu_ctx_id_put(struct virtio_gpu_device *vgdev, uint32_t id)
+{
+	spin_lock(&vgdev->ctx_id_idr_lock);
+	idr_remove(&vgdev->ctx_id_idr, id);
+	spin_unlock(&vgdev->ctx_id_idr_lock);
+}
+
+static void virtio_gpu_context_create(struct virtio_gpu_device *vgdev,
+				      uint32_t nlen, const char *name,
+				      uint32_t *ctx_id)
+{
+	virtio_gpu_ctx_id_get(vgdev, ctx_id);
+	virtio_gpu_cmd_context_create(vgdev, *ctx_id, nlen, name);
+}
+
+static void virtio_gpu_context_destroy(struct virtio_gpu_device *vgdev,
+				      uint32_t ctx_id)
+{
+	virtio_gpu_cmd_context_destroy(vgdev, ctx_id);
+	virtio_gpu_ctx_id_put(vgdev, ctx_id);
+}
+
 static void virtio_gpu_init_vq(struct virtio_gpu_queue *vgvq,
 			       void (*work_func)(struct work_struct *work))
 {
@@ -60,6 +95,36 @@ static void virtio_gpu_init_vq(struct virtio_gpu_queue *vgvq,
 	INIT_WORK(&vgvq->dequeue_work, work_func);
 }
 
+static void virtio_gpu_get_capsets(struct virtio_gpu_device *vgdev,
+				   int num_capsets)
+{
+	int i, ret;
+
+	vgdev->capsets = kcalloc(num_capsets,
+				 sizeof(struct virtio_gpu_drv_capset),
+				 GFP_KERNEL);
+	if (!vgdev->capsets) {
+		DRM_ERROR("failed to allocate cap sets\n");
+		return;
+	}
+	for (i = 0; i < num_capsets; i++) {
+		virtio_gpu_cmd_get_capset_info(vgdev, i);
+		ret = wait_event_timeout(vgdev->resp_wq,
+					 vgdev->capsets[i].id > 0, 5 * HZ);
+		if (ret == 0) {
+			DRM_ERROR("timed out waiting for cap set %d\n", i);
+			kfree(vgdev->capsets);
+			vgdev->capsets = NULL;
+			return;
+		}
+		DRM_INFO("cap set %d: id %d, max-version %d, max-size %d\n",
+			 i, vgdev->capsets[i].id,
+			 vgdev->capsets[i].max_version,
+			 vgdev->capsets[i].max_size);
+	}
+	vgdev->num_capsets = num_capsets;
+}
+
 int virtio_gpu_driver_load(struct drm_device *dev, unsigned long flags)
 {
 	static vq_callback_t *callbacks[] = {
@@ -70,7 +135,7 @@ int virtio_gpu_driver_load(struct drm_device *dev, unsigned long flags)
 	struct virtio_gpu_device *vgdev;
 	/* this will expand later */
 	struct virtqueue *vqs[2];
-	u32 num_scanouts;
+	u32 num_scanouts, num_capsets;
 	int ret;
 
 	if (!virtio_has_feature(dev->virtdev, VIRTIO_F_VERSION_1))
@@ -96,9 +161,15 @@ int virtio_gpu_driver_load(struct drm_device *dev, unsigned long flags)
 
 	spin_lock_init(&vgdev->fence_drv.lock);
 	INIT_LIST_HEAD(&vgdev->fence_drv.fences);
+	INIT_LIST_HEAD(&vgdev->cap_cache);
 	INIT_WORK(&vgdev->config_changed_work,
 		  virtio_gpu_config_changed_work_func);
 
+	if (virtio_has_feature(vgdev->vdev, VIRTIO_GPU_F_VIRGL))
+		vgdev->has_virgl_3d = true;
+	DRM_INFO("virgl 3d acceleration %s\n",
+		 vgdev->has_virgl_3d ? "enabled" : "not available");
+
 	ret = vgdev->vdev->config->find_vqs(vgdev->vdev, 2, vqs,
 					    callbacks, names);
 	if (ret) {
@@ -129,6 +200,11 @@ int virtio_gpu_driver_load(struct drm_device *dev, unsigned long flags)
 		ret = -EINVAL;
 		goto err_scanouts;
 	}
+	DRM_INFO("number of scanouts: %d\n", num_scanouts);
+
+	virtio_cread(vgdev->vdev, struct virtio_gpu_config,
+		     num_capsets, &num_capsets);
+	DRM_INFO("number of cap sets: %d\n", num_capsets);
 
 	ret = virtio_gpu_modeset_init(vgdev);
 	if (ret)
@@ -137,6 +213,8 @@ int virtio_gpu_driver_load(struct drm_device *dev, unsigned long flags)
 	virtio_device_ready(vgdev->vdev);
 	vgdev->vqs_ready = true;
 
+	if (num_capsets)
+		virtio_gpu_get_capsets(vgdev, num_capsets);
 	virtio_gpu_cmd_get_display_info(vgdev);
 	wait_event_timeout(vgdev->resp_wq, !vgdev->display_info_pending,
 			   5 * HZ);
@@ -157,6 +235,16 @@ err_vqs:
 	return ret;
 }
 
+static void virtio_gpu_cleanup_cap_cache(struct virtio_gpu_device *vgdev)
+{
+	struct virtio_gpu_drv_cap_cache *cache_ent, *tmp;
+
+	list_for_each_entry_safe(cache_ent, tmp, &vgdev->cap_cache, head) {
+		kfree(cache_ent->caps_cache);
+		kfree(cache_ent);
+	}
+}
+
 int virtio_gpu_driver_unload(struct drm_device *dev)
 {
 	struct virtio_gpu_device *vgdev = dev->dev_private;
@@ -170,6 +258,49 @@ int virtio_gpu_driver_unload(struct drm_device *dev)
 	virtio_gpu_modeset_fini(vgdev);
 	virtio_gpu_ttm_fini(vgdev);
 	virtio_gpu_free_vbufs(vgdev);
+	virtio_gpu_cleanup_cap_cache(vgdev);
+	kfree(vgdev->capsets);
 	kfree(vgdev);
 	return 0;
 }
+
+int virtio_gpu_driver_open(struct drm_device *dev, struct drm_file *file)
+{
+	struct virtio_gpu_device *vgdev = dev->dev_private;
+	struct virtio_gpu_fpriv *vfpriv;
+	uint32_t id;
+	char dbgname[64], tmpname[TASK_COMM_LEN];
+
+	/* can't create contexts without 3d renderer */
+	if (!vgdev->has_virgl_3d)
+		return 0;
+
+	get_task_comm(tmpname, current);
+	snprintf(dbgname, sizeof(dbgname), "%s", tmpname);
+	dbgname[63] = 0;
+	/* allocate a virt GPU context for this opener */
+	vfpriv = kzalloc(sizeof(*vfpriv), GFP_KERNEL);
+	if (!vfpriv)
+		return -ENOMEM;
+
+	virtio_gpu_context_create(vgdev, strlen(dbgname), dbgname, &id);
+
+	vfpriv->ctx_id = id;
+	file->driver_priv = vfpriv;
+	return 0;
+}
+
+void virtio_gpu_driver_postclose(struct drm_device *dev, struct drm_file *file)
+{
+	struct virtio_gpu_device *vgdev = dev->dev_private;
+	struct virtio_gpu_fpriv *vfpriv;
+
+	if (!vgdev->has_virgl_3d)
+		return;
+
+	vfpriv = file->driver_priv;
+
+	virtio_gpu_context_destroy(vgdev, vfpriv->ctx_id);
+	kfree(vfpriv);
+	file->driver_priv = NULL;
+}
diff --git a/drivers/gpu/drm/virtio/virtgpu_ttm.c b/drivers/gpu/drm/virtio/virtgpu_ttm.c
index b092d7b..9fd924c 100644
--- a/drivers/gpu/drm/virtio/virtgpu_ttm.c
+++ b/drivers/gpu/drm/virtio/virtgpu_ttm.c
@@ -32,6 +32,7 @@
 #include <ttm/ttm_module.h>
 #include <drm/drmP.h>
 #include <drm/drm.h>
+#include <drm/virtgpu_drm.h>
 #include "virtgpu_drv.h"
 
 #include <linux/delay.h>
diff --git a/drivers/gpu/drm/virtio/virtgpu_vq.c b/drivers/gpu/drm/virtio/virtgpu_vq.c
index ee25e9a..5a0f8a7 100644
--- a/drivers/gpu/drm/virtio/virtgpu_vq.c
+++ b/drivers/gpu/drm/virtio/virtgpu_vq.c
@@ -586,6 +586,47 @@ static void virtio_gpu_cmd_get_display_info_cb(struct virtio_gpu_device *vgdev,
 		drm_kms_helper_hotplug_event(vgdev->ddev);
 }
 
+static void virtio_gpu_cmd_get_capset_info_cb(struct virtio_gpu_device *vgdev,
+					      struct virtio_gpu_vbuffer *vbuf)
+{
+	struct virtio_gpu_get_capset_info *cmd =
+		(struct virtio_gpu_get_capset_info *)vbuf->buf;
+	struct virtio_gpu_resp_capset_info *resp =
+		(struct virtio_gpu_resp_capset_info *)vbuf->resp_buf;
+	int i = le32_to_cpu(cmd->capset_index);
+
+	spin_lock(&vgdev->display_info_lock);
+	vgdev->capsets[i].id = le32_to_cpu(resp->capset_id);
+	vgdev->capsets[i].max_version = le32_to_cpu(resp->capset_max_version);
+	vgdev->capsets[i].max_size = le32_to_cpu(resp->capset_max_size);
+	spin_unlock(&vgdev->display_info_lock);
+	wake_up(&vgdev->resp_wq);
+}
+
+static void virtio_gpu_cmd_capset_cb(struct virtio_gpu_device *vgdev,
+				     struct virtio_gpu_vbuffer *vbuf)
+{
+	struct virtio_gpu_get_capset *cmd =
+		(struct virtio_gpu_get_capset *)vbuf->buf;
+	struct virtio_gpu_resp_capset *resp =
+		(struct virtio_gpu_resp_capset *)vbuf->resp_buf;
+	struct virtio_gpu_drv_cap_cache *cache_ent;
+
+	spin_lock(&vgdev->display_info_lock);
+	list_for_each_entry(cache_ent, &vgdev->cap_cache, head) {
+		if (cache_ent->version == le32_to_cpu(cmd->capset_version) &&
+		    cache_ent->id == le32_to_cpu(cmd->capset_id)) {
+			memcpy(cache_ent->caps_cache, resp->capset_data,
+			       cache_ent->size);
+			atomic_set(&cache_ent->is_valid, 1);
+			break;
+		}
+	}
+	spin_unlock(&vgdev->display_info_lock);
+	wake_up(&vgdev->resp_wq);
+}
+
+
 int virtio_gpu_cmd_get_display_info(struct virtio_gpu_device *vgdev)
 {
 	struct virtio_gpu_ctrl_hdr *cmd_p;
@@ -609,6 +650,230 @@ int virtio_gpu_cmd_get_display_info(struct virtio_gpu_device *vgdev)
 	return 0;
 }
 
+int virtio_gpu_cmd_get_capset_info(struct virtio_gpu_device *vgdev, int idx)
+{
+	struct virtio_gpu_get_capset_info *cmd_p;
+	struct virtio_gpu_vbuffer *vbuf;
+	void *resp_buf;
+
+	resp_buf = kzalloc(sizeof(struct virtio_gpu_resp_capset_info),
+			   GFP_KERNEL);
+	if (!resp_buf)
+		return -ENOMEM;
+
+	cmd_p = virtio_gpu_alloc_cmd_resp
+		(vgdev, &virtio_gpu_cmd_get_capset_info_cb, &vbuf,
+		 sizeof(*cmd_p), sizeof(struct virtio_gpu_resp_capset_info),
+		 resp_buf);
+	memset(cmd_p, 0, sizeof(*cmd_p));
+
+	cmd_p->hdr.type = cpu_to_le32(VIRTIO_GPU_CMD_GET_CAPSET_INFO);
+	cmd_p->capset_index = cpu_to_le32(idx);
+	virtio_gpu_queue_ctrl_buffer(vgdev, vbuf);
+	return 0;
+}
+
+int virtio_gpu_cmd_get_capset(struct virtio_gpu_device *vgdev,
+			      int idx, int version,
+			      struct virtio_gpu_drv_cap_cache **cache_p)
+{
+	struct virtio_gpu_get_capset *cmd_p;
+	struct virtio_gpu_vbuffer *vbuf;
+	int max_size = vgdev->capsets[idx].max_size;
+	struct virtio_gpu_drv_cap_cache *cache_ent;
+	void *resp_buf;
+
+	if (idx > vgdev->num_capsets)
+		return -EINVAL;
+
+	if (version > vgdev->capsets[idx].max_version)
+		return -EINVAL;
+
+	cache_ent = kzalloc(sizeof(*cache_ent), GFP_KERNEL);
+	if (!cache_ent)
+		return -ENOMEM;
+
+	cache_ent->caps_cache = kmalloc(max_size, GFP_KERNEL);
+	if (!cache_ent->caps_cache) {
+		kfree(cache_ent);
+		return -ENOMEM;
+	}
+
+	resp_buf = kzalloc(sizeof(struct virtio_gpu_resp_capset) + max_size,
+			   GFP_KERNEL);
+	if (!resp_buf) {
+		kfree(cache_ent->caps_cache);
+		kfree(cache_ent);
+		return -ENOMEM;
+	}
+
+	cache_ent->version = version;
+	cache_ent->id = vgdev->capsets[idx].id;
+	atomic_set(&cache_ent->is_valid, 0);
+	cache_ent->size = max_size;
+	spin_lock(&vgdev->display_info_lock);
+	list_add_tail(&cache_ent->head, &vgdev->cap_cache);
+	spin_unlock(&vgdev->display_info_lock);
+
+	cmd_p = virtio_gpu_alloc_cmd_resp
+		(vgdev, &virtio_gpu_cmd_capset_cb, &vbuf, sizeof(*cmd_p),
+		 sizeof(struct virtio_gpu_resp_capset) + max_size,
+		 resp_buf);
+	cmd_p->hdr.type = cpu_to_le32(VIRTIO_GPU_CMD_GET_CAPSET);
+	cmd_p->capset_id = cpu_to_le32(vgdev->capsets[idx].id);
+	cmd_p->capset_version = cpu_to_le32(version);
+	*cache_p = cache_ent;
+	virtio_gpu_queue_ctrl_buffer(vgdev, vbuf);
+
+	return 0;
+}
+
+void virtio_gpu_cmd_context_create(struct virtio_gpu_device *vgdev, uint32_t id,
+				   uint32_t nlen, const char *name)
+{
+	struct virtio_gpu_ctx_create *cmd_p;
+	struct virtio_gpu_vbuffer *vbuf;
+
+	cmd_p = virtio_gpu_alloc_cmd(vgdev, &vbuf, sizeof(*cmd_p));
+	memset(cmd_p, 0, sizeof(*cmd_p));
+
+	cmd_p->hdr.type = cpu_to_le32(VIRTIO_GPU_CMD_CTX_CREATE);
+	cmd_p->hdr.ctx_id = cpu_to_le32(id);
+	cmd_p->nlen = cpu_to_le32(nlen);
+	strncpy(cmd_p->debug_name, name, sizeof(cmd_p->debug_name)-1);
+	cmd_p->debug_name[sizeof(cmd_p->debug_name)-1] = 0;
+	virtio_gpu_queue_ctrl_buffer(vgdev, vbuf);
+}
+
+void virtio_gpu_cmd_context_destroy(struct virtio_gpu_device *vgdev,
+				    uint32_t id)
+{
+	struct virtio_gpu_ctx_destroy *cmd_p;
+	struct virtio_gpu_vbuffer *vbuf;
+
+	cmd_p = virtio_gpu_alloc_cmd(vgdev, &vbuf, sizeof(*cmd_p));
+	memset(cmd_p, 0, sizeof(*cmd_p));
+
+	cmd_p->hdr.type = cpu_to_le32(VIRTIO_GPU_CMD_CTX_DESTROY);
+	cmd_p->hdr.ctx_id = cpu_to_le32(id);
+	virtio_gpu_queue_ctrl_buffer(vgdev, vbuf);
+}
+
+void virtio_gpu_cmd_context_attach_resource(struct virtio_gpu_device *vgdev,
+					    uint32_t ctx_id,
+					    uint32_t resource_id)
+{
+	struct virtio_gpu_ctx_resource *cmd_p;
+	struct virtio_gpu_vbuffer *vbuf;
+
+	cmd_p = virtio_gpu_alloc_cmd(vgdev, &vbuf, sizeof(*cmd_p));
+	memset(cmd_p, 0, sizeof(*cmd_p));
+
+	cmd_p->hdr.type = cpu_to_le32(VIRTIO_GPU_CMD_CTX_ATTACH_RESOURCE);
+	cmd_p->hdr.ctx_id = cpu_to_le32(ctx_id);
+	cmd_p->resource_id = cpu_to_le32(resource_id);
+	virtio_gpu_queue_ctrl_buffer(vgdev, vbuf);
+
+}
+
+void virtio_gpu_cmd_context_detach_resource(struct virtio_gpu_device *vgdev,
+					    uint32_t ctx_id,
+					    uint32_t resource_id)
+{
+	struct virtio_gpu_ctx_resource *cmd_p;
+	struct virtio_gpu_vbuffer *vbuf;
+
+	cmd_p = virtio_gpu_alloc_cmd(vgdev, &vbuf, sizeof(*cmd_p));
+	memset(cmd_p, 0, sizeof(*cmd_p));
+
+	cmd_p->hdr.type = cpu_to_le32(VIRTIO_GPU_CMD_CTX_DETACH_RESOURCE);
+	cmd_p->hdr.ctx_id = cpu_to_le32(ctx_id);
+	cmd_p->resource_id = cpu_to_le32(resource_id);
+	virtio_gpu_queue_ctrl_buffer(vgdev, vbuf);
+}
+
+void
+virtio_gpu_cmd_resource_create_3d(struct virtio_gpu_device *vgdev,
+				  struct virtio_gpu_resource_create_3d *rc_3d,
+				  struct virtio_gpu_fence **fence)
+{
+	struct virtio_gpu_resource_create_3d *cmd_p;
+	struct virtio_gpu_vbuffer *vbuf;
+
+	cmd_p = virtio_gpu_alloc_cmd(vgdev, &vbuf, sizeof(*cmd_p));
+	memset(cmd_p, 0, sizeof(*cmd_p));
+
+	*cmd_p = *rc_3d;
+	cmd_p->hdr.type = cpu_to_le32(VIRTIO_GPU_CMD_RESOURCE_CREATE_3D);
+	cmd_p->hdr.flags = 0;
+
+	virtio_gpu_queue_fenced_ctrl_buffer(vgdev, vbuf, &cmd_p->hdr, fence);
+}
+
+void virtio_gpu_cmd_transfer_to_host_3d(struct virtio_gpu_device *vgdev,
+					uint32_t resource_id, uint32_t ctx_id,
+					uint64_t offset, uint32_t level,
+					struct virtio_gpu_box *box,
+					struct virtio_gpu_fence **fence)
+{
+	struct virtio_gpu_transfer_host_3d *cmd_p;
+	struct virtio_gpu_vbuffer *vbuf;
+
+	cmd_p = virtio_gpu_alloc_cmd(vgdev, &vbuf, sizeof(*cmd_p));
+	memset(cmd_p, 0, sizeof(*cmd_p));
+
+	cmd_p->hdr.type = cpu_to_le32(VIRTIO_GPU_CMD_TRANSFER_TO_HOST_3D);
+	cmd_p->hdr.ctx_id = cpu_to_le32(ctx_id);
+	cmd_p->resource_id = cpu_to_le32(resource_id);
+	cmd_p->box = *box;
+	cmd_p->offset = cpu_to_le64(offset);
+	cmd_p->level = cpu_to_le32(level);
+
+	virtio_gpu_queue_fenced_ctrl_buffer(vgdev, vbuf, &cmd_p->hdr, fence);
+}
+
+void virtio_gpu_cmd_transfer_from_host_3d(struct virtio_gpu_device *vgdev,
+					  uint32_t resource_id, uint32_t ctx_id,
+					  uint64_t offset, uint32_t level,
+					  struct virtio_gpu_box *box,
+					  struct virtio_gpu_fence **fence)
+{
+	struct virtio_gpu_transfer_host_3d *cmd_p;
+	struct virtio_gpu_vbuffer *vbuf;
+
+	cmd_p = virtio_gpu_alloc_cmd(vgdev, &vbuf, sizeof(*cmd_p));
+	memset(cmd_p, 0, sizeof(*cmd_p));
+
+	cmd_p->hdr.type = cpu_to_le32(VIRTIO_GPU_CMD_TRANSFER_FROM_HOST_3D);
+	cmd_p->hdr.ctx_id = cpu_to_le32(ctx_id);
+	cmd_p->resource_id = cpu_to_le32(resource_id);
+	cmd_p->box = *box;
+	cmd_p->offset = cpu_to_le64(offset);
+	cmd_p->level = cpu_to_le32(level);
+
+	virtio_gpu_queue_fenced_ctrl_buffer(vgdev, vbuf, &cmd_p->hdr, fence);
+}
+
+void virtio_gpu_cmd_submit(struct virtio_gpu_device *vgdev,
+			   void *data, uint32_t data_size,
+			   uint32_t ctx_id, struct virtio_gpu_fence **fence)
+{
+	struct virtio_gpu_cmd_submit *cmd_p;
+	struct virtio_gpu_vbuffer *vbuf;
+
+	cmd_p = virtio_gpu_alloc_cmd(vgdev, &vbuf, sizeof(*cmd_p));
+	memset(cmd_p, 0, sizeof(*cmd_p));
+
+	vbuf->data_buf = data;
+	vbuf->data_size = data_size;
+
+	cmd_p->hdr.type = cpu_to_le32(VIRTIO_GPU_CMD_SUBMIT_3D);
+	cmd_p->hdr.ctx_id = cpu_to_le32(ctx_id);
+	cmd_p->size = cpu_to_le32(data_size);
+
+	virtio_gpu_queue_fenced_ctrl_buffer(vgdev, vbuf, &cmd_p->hdr, fence);
+}
+
 int virtio_gpu_object_attach(struct virtio_gpu_device *vgdev,
 			     struct virtio_gpu_object *obj,
 			     uint32_t resource_id,
diff --git a/include/uapi/drm/Kbuild b/include/uapi/drm/Kbuild
index 2d9a25d..38d4370 100644
--- a/include/uapi/drm/Kbuild
+++ b/include/uapi/drm/Kbuild
@@ -17,3 +17,4 @@ header-y += tegra_drm.h
 header-y += via_drm.h
 header-y += vmwgfx_drm.h
 header-y += msm_drm.h
+header-y += virtgpu_drm.h
diff --git a/include/uapi/drm/virtgpu_drm.h b/include/uapi/drm/virtgpu_drm.h
new file mode 100644
index 0000000..fc9e2d6
--- /dev/null
+++ b/include/uapi/drm/virtgpu_drm.h
@@ -0,0 +1,167 @@
+/*
+ * Copyright 2013 Red Hat
+ * All Rights Reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ */
+#ifndef VIRTGPU_DRM_H
+#define VIRTGPU_DRM_H
+
+#include <stddef.h>
+#include "drm/drm.h"
+
+/* Please note that modifications to all structs defined here are
+ * subject to backwards-compatibility constraints.
+ *
+ * Do not use pointers, use uint64_t instead for 32 bit / 64 bit user/kernel
+ * compatibility Keep fields aligned to their size
+ */
+
+#define DRM_VIRTGPU_MAP         0x01
+#define DRM_VIRTGPU_EXECBUFFER  0x02
+#define DRM_VIRTGPU_GETPARAM    0x03
+#define DRM_VIRTGPU_RESOURCE_CREATE 0x04
+#define DRM_VIRTGPU_RESOURCE_INFO     0x05
+#define DRM_VIRTGPU_TRANSFER_FROM_HOST 0x06
+#define DRM_VIRTGPU_TRANSFER_TO_HOST 0x07
+#define DRM_VIRTGPU_WAIT     0x08
+#define DRM_VIRTGPU_GET_CAPS  0x09
+
+struct drm_virtgpu_map {
+	uint64_t offset; /* use for mmap system call */
+	uint32_t handle;
+	uint32_t pad;
+};
+
+struct drm_virtgpu_execbuffer {
+	uint32_t		flags;		/* for future use */
+	uint32_t size;
+	uint64_t command; /* void* */
+	uint64_t bo_handles;
+	uint32_t num_bo_handles;
+	uint32_t pad;
+};
+
+#define VIRTGPU_PARAM_3D_FEATURES 1 /* do we have 3D features in the hw */
+
+struct drm_virtgpu_getparam {
+	uint64_t param;
+	uint64_t value;
+};
+
+/* NO_BO flags? NO resource flag? */
+/* resource flag for y_0_top */
+struct drm_virtgpu_resource_create {
+	uint32_t target;
+	uint32_t format;
+	uint32_t bind;
+	uint32_t width;
+	uint32_t height;
+	uint32_t depth;
+	uint32_t array_size;
+	uint32_t last_level;
+	uint32_t nr_samples;
+	uint32_t flags;
+	uint32_t bo_handle; /* if this is set - recreate a new resource attached to this bo ? */
+	uint32_t res_handle;  /* returned by kernel */
+	uint32_t size;        /* validate transfer in the host */
+	uint32_t stride;      /* validate transfer in the host */
+};
+
+struct drm_virtgpu_resource_info {
+	uint32_t bo_handle;
+	uint32_t res_handle;
+	uint32_t size;
+	uint32_t stride;
+};
+
+struct drm_virtgpu_3d_box {
+	uint32_t x;
+	uint32_t y;
+	uint32_t z;
+	uint32_t w;
+	uint32_t h;
+	uint32_t d;
+};
+
+struct drm_virtgpu_3d_transfer_to_host {
+	uint32_t bo_handle;
+	struct drm_virtgpu_3d_box box;
+	uint32_t level;
+	uint32_t offset;
+};
+
+struct drm_virtgpu_3d_transfer_from_host {
+	uint32_t bo_handle;
+	struct drm_virtgpu_3d_box box;
+	uint32_t level;
+	uint32_t offset;
+};
+
+#define VIRTGPU_WAIT_NOWAIT 1 /* like it */
+struct drm_virtgpu_3d_wait {
+	uint32_t handle; /* 0 is an invalid handle */
+	uint32_t flags;
+};
+
+struct drm_virtgpu_get_caps {
+	uint32_t cap_set_id;
+	uint32_t cap_set_ver;
+	uint64_t addr;
+	uint32_t size;
+	uint32_t pad;
+};
+
+#define DRM_IOCTL_VIRTGPU_MAP \
+	DRM_IOWR(DRM_COMMAND_BASE + DRM_VIRTGPU_MAP, struct drm_virtgpu_map)
+
+#define DRM_IOCTL_VIRTGPU_EXECBUFFER \
+	DRM_IOW(DRM_COMMAND_BASE + DRM_VIRTGPU_EXECBUFFER,\
+		struct drm_virtgpu_execbuffer)
+
+#define DRM_IOCTL_VIRTGPU_GETPARAM \
+	DRM_IOWR(DRM_COMMAND_BASE + DRM_VIRTGPU_GETPARAM,\
+		struct drm_virtgpu_getparam)
+
+#define DRM_IOCTL_VIRTGPU_RESOURCE_CREATE			\
+	DRM_IOWR(DRM_COMMAND_BASE + DRM_VIRTGPU_RESOURCE_CREATE,	\
+		struct drm_virtgpu_resource_create)
+
+#define DRM_IOCTL_VIRTGPU_RESOURCE_INFO \
+	DRM_IOWR(DRM_COMMAND_BASE + DRM_VIRTGPU_RESOURCE_INFO, \
+		 struct drm_virtgpu_resource_info)
+
+#define DRM_IOCTL_VIRTGPU_TRANSFER_FROM_HOST \
+	DRM_IOWR(DRM_COMMAND_BASE + DRM_VIRTGPU_TRANSFER_FROM_HOST,	\
+		struct drm_virtgpu_3d_transfer_from_host)
+
+#define DRM_IOCTL_VIRTGPU_TRANSFER_TO_HOST \
+	DRM_IOWR(DRM_COMMAND_BASE + DRM_VIRTGPU_TRANSFER_TO_HOST,	\
+		struct drm_virtgpu_3d_transfer_to_host)
+
+#define DRM_IOCTL_VIRTGPU_WAIT				\
+	DRM_IOWR(DRM_COMMAND_BASE + DRM_VIRTGPU_WAIT,	\
+		struct drm_virtgpu_3d_wait)
+
+#define DRM_IOCTL_VIRTGPU_GET_CAPS \
+	DRM_IOWR(DRM_COMMAND_BASE + DRM_VIRTGPU_GET_CAPS, \
+	struct drm_virtgpu_get_caps)
+
+#endif
diff --git a/include/uapi/linux/virtio_gpu.h b/include/uapi/linux/virtio_gpu.h
index 478be52..7a63faa 100644
--- a/include/uapi/linux/virtio_gpu.h
+++ b/include/uapi/linux/virtio_gpu.h
@@ -40,6 +40,8 @@
 
 #include <linux/types.h>
 
+#define VIRTIO_GPU_F_VIRGL 0
+
 enum virtio_gpu_ctrl_type {
 	VIRTIO_GPU_UNDEFINED = 0,
 
@@ -52,6 +54,18 @@ enum virtio_gpu_ctrl_type {
 	VIRTIO_GPU_CMD_TRANSFER_TO_HOST_2D,
 	VIRTIO_GPU_CMD_RESOURCE_ATTACH_BACKING,
 	VIRTIO_GPU_CMD_RESOURCE_DETACH_BACKING,
+	VIRTIO_GPU_CMD_GET_CAPSET_INFO,
+	VIRTIO_GPU_CMD_GET_CAPSET,
+
+	/* 3d commands */
+	VIRTIO_GPU_CMD_CTX_CREATE = 0x0200,
+	VIRTIO_GPU_CMD_CTX_DESTROY,
+	VIRTIO_GPU_CMD_CTX_ATTACH_RESOURCE,
+	VIRTIO_GPU_CMD_CTX_DETACH_RESOURCE,
+	VIRTIO_GPU_CMD_RESOURCE_CREATE_3D,
+	VIRTIO_GPU_CMD_TRANSFER_TO_HOST_3D,
+	VIRTIO_GPU_CMD_TRANSFER_FROM_HOST_3D,
+	VIRTIO_GPU_CMD_SUBMIT_3D,
 
 	/* cursor commands */
 	VIRTIO_GPU_CMD_UPDATE_CURSOR = 0x0300,
@@ -60,6 +74,8 @@ enum virtio_gpu_ctrl_type {
 	/* success responses */
 	VIRTIO_GPU_RESP_OK_NODATA = 0x1100,
 	VIRTIO_GPU_RESP_OK_DISPLAY_INFO,
+	VIRTIO_GPU_RESP_OK_CAPSET_INFO,
+	VIRTIO_GPU_RESP_OK_CAPSET,
 
 	/* error responses */
 	VIRTIO_GPU_RESP_ERR_UNSPEC = 0x1200,
@@ -180,13 +196,107 @@ struct virtio_gpu_resp_display_info {
 	} pmodes[VIRTIO_GPU_MAX_SCANOUTS];
 };
 
+/* data passed in the control vq, 3d related */
+
+struct virtio_gpu_box {
+	__le32 x, y, z;
+	__le32 w, h, d;
+};
+
+/* VIRTIO_GPU_CMD_TRANSFER_TO_HOST_3D, VIRTIO_GPU_CMD_TRANSFER_FROM_HOST_3D */
+struct virtio_gpu_transfer_host_3d {
+	struct virtio_gpu_ctrl_hdr hdr;
+	struct virtio_gpu_box box;
+	__le64 offset;
+	__le32 resource_id;
+	__le32 level;
+	__le32 stride;
+	__le32 layer_stride;
+};
+
+/* VIRTIO_GPU_CMD_RESOURCE_CREATE_3D */
+#define VIRTIO_GPU_RESOURCE_FLAG_Y_0_TOP (1 << 0)
+struct virtio_gpu_resource_create_3d {
+	struct virtio_gpu_ctrl_hdr hdr;
+	__le32 resource_id;
+	__le32 target;
+	__le32 format;
+	__le32 bind;
+	__le32 width;
+	__le32 height;
+	__le32 depth;
+	__le32 array_size;
+	__le32 last_level;
+	__le32 nr_samples;
+	__le32 flags;
+	__le32 padding;
+};
+
+/* VIRTIO_GPU_CMD_CTX_CREATE */
+struct virtio_gpu_ctx_create {
+	struct virtio_gpu_ctrl_hdr hdr;
+	__le32 nlen;
+	__le32 padding;
+	char debug_name[64];
+};
+
+/* VIRTIO_GPU_CMD_CTX_DESTROY */
+struct virtio_gpu_ctx_destroy {
+	struct virtio_gpu_ctrl_hdr hdr;
+};
+
+/* VIRTIO_GPU_CMD_CTX_ATTACH_RESOURCE, VIRTIO_GPU_CMD_CTX_DETACH_RESOURCE */
+struct virtio_gpu_ctx_resource {
+	struct virtio_gpu_ctrl_hdr hdr;
+	__le32 resource_id;
+	__le32 padding;
+};
+
+/* VIRTIO_GPU_CMD_SUBMIT_3D */
+struct virtio_gpu_cmd_submit {
+	struct virtio_gpu_ctrl_hdr hdr;
+	__le32 size;
+	__le32 padding;
+};
+
+#define VIRTIO_GPU_CAPSET_VIRGL 1
+
+/* VIRTIO_GPU_CMD_GET_CAPSET_INFO */
+struct virtio_gpu_get_capset_info {
+	struct virtio_gpu_ctrl_hdr hdr;
+	__le32 capset_index;
+	__le32 padding;
+};
+
+/* VIRTIO_GPU_RESP_OK_CAPSET_INFO */
+struct virtio_gpu_resp_capset_info {
+	struct virtio_gpu_ctrl_hdr hdr;
+	__le32 capset_id;
+	__le32 capset_max_version;
+	__le32 capset_max_size;
+	__le32 padding;
+};
+
+/* VIRTIO_GPU_CMD_GET_CAPSET */
+struct virtio_gpu_get_capset {
+	struct virtio_gpu_ctrl_hdr hdr;
+	__le32 capset_id;
+	__le32 capset_version;
+};
+
+/* VIRTIO_GPU_RESP_OK_CAPSET */
+struct virtio_gpu_resp_capset {
+	struct virtio_gpu_ctrl_hdr hdr;
+	uint8_t capset_data[];
+};
+
 #define VIRTIO_GPU_EVENT_DISPLAY (1 << 0)
 
 struct virtio_gpu_config {
 	__u32 events_read;
 	__u32 events_clear;
 	__u32 num_scanouts;
-	__u32 reserved;
+	__u32 num_capsets;
 };
 
 /* simple formats for fbcon/X use */
-- 
cgit v0.10.2


From 11a8f2805d4fd8a9c340b3f1b77990fd69b2dc0c Mon Sep 17 00:00:00 2001
From: Dave Airlie <airlied@redhat.com>
Date: Fri, 23 Jan 2015 13:04:11 +1000
Subject: virtio-gpu: add basic prime support

This is enough to enable DRI3.

Signed-off-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>

diff --git a/drivers/gpu/drm/virtio/Makefile b/drivers/gpu/drm/virtio/Makefile
index da7bf19..3fb8eac 100644
--- a/drivers/gpu/drm/virtio/Makefile
+++ b/drivers/gpu/drm/virtio/Makefile
@@ -7,6 +7,6 @@ ccflags-y := -Iinclude/drm
 virtio-gpu-y := virtgpu_drv.o virtgpu_kms.o virtgpu_drm_bus.o virtgpu_gem.o \
 	virtgpu_fb.o virtgpu_display.o virtgpu_vq.o virtgpu_ttm.o \
 	virtgpu_fence.o virtgpu_object.o virtgpu_debugfs.o virtgpu_plane.o \
-	virtgpu_ioctl.o
+	virtgpu_ioctl.o virtgpu_prime.o
 
 obj-$(CONFIG_DRM_VIRTIO_GPU) += virtio-gpu.o
diff --git a/drivers/gpu/drm/virtio/virtgpu_drv.c b/drivers/gpu/drm/virtio/virtgpu_drv.c
index c27d3a3..1f15ef8 100644
--- a/drivers/gpu/drm/virtio/virtgpu_drv.c
+++ b/drivers/gpu/drm/virtio/virtgpu_drv.c
@@ -118,7 +118,7 @@ static const struct file_operations virtio_gpu_driver_fops = {
 
 
 static struct drm_driver driver = {
-	.driver_features = DRIVER_MODESET | DRIVER_GEM,
+	.driver_features = DRIVER_MODESET | DRIVER_GEM | DRIVER_PRIME,
 	.set_busid = drm_virtio_set_busid,
 	.load = virtio_gpu_driver_load,
 	.unload = virtio_gpu_driver_unload,
@@ -133,6 +133,17 @@ static struct drm_driver driver = {
 	.debugfs_init = virtio_gpu_debugfs_init,
 	.debugfs_cleanup = virtio_gpu_debugfs_takedown,
 #endif
+	.prime_handle_to_fd = drm_gem_prime_handle_to_fd,
+	.prime_fd_to_handle = drm_gem_prime_fd_to_handle,
+	.gem_prime_export = drm_gem_prime_export,
+	.gem_prime_import = drm_gem_prime_import,
+	.gem_prime_pin = virtgpu_gem_prime_pin,
+	.gem_prime_unpin = virtgpu_gem_prime_unpin,
+	.gem_prime_get_sg_table = virtgpu_gem_prime_get_sg_table,
+	.gem_prime_import_sg_table = virtgpu_gem_prime_import_sg_table,
+	.gem_prime_vmap = virtgpu_gem_prime_vmap,
+	.gem_prime_vunmap = virtgpu_gem_prime_vunmap,
+	.gem_prime_mmap = virtgpu_gem_prime_mmap,
 
 	.gem_free_object = virtio_gpu_gem_free_object,
 	.gem_open_object = virtio_gpu_gem_object_open,
diff --git a/drivers/gpu/drm/virtio/virtgpu_drv.h b/drivers/gpu/drm/virtio/virtgpu_drv.h
index 2719108..79f0abe 100644
--- a/drivers/gpu/drm/virtio/virtgpu_drv.h
+++ b/drivers/gpu/drm/virtio/virtgpu_drv.h
@@ -359,6 +359,18 @@ int virtio_gpu_object_get_sg_table(struct virtio_gpu_device *qdev,
 void virtio_gpu_object_free_sg_table(struct virtio_gpu_object *bo);
 int virtio_gpu_object_wait(struct virtio_gpu_object *bo, bool no_wait);
 
+/* virtgpu_prime.c */
+int virtgpu_gem_prime_pin(struct drm_gem_object *obj);
+void virtgpu_gem_prime_unpin(struct drm_gem_object *obj);
+struct sg_table *virtgpu_gem_prime_get_sg_table(struct drm_gem_object *obj);
+struct drm_gem_object *virtgpu_gem_prime_import_sg_table(
+        struct drm_device *dev, struct dma_buf_attachment *attach,
+        struct sg_table *sgt);
+void *virtgpu_gem_prime_vmap(struct drm_gem_object *obj);
+void virtgpu_gem_prime_vunmap(struct drm_gem_object *obj, void *vaddr);
+int virtgpu_gem_prime_mmap(struct drm_gem_object *obj,
+                                struct vm_area_struct *vma);
+
 static inline struct virtio_gpu_object*
 virtio_gpu_object_ref(struct virtio_gpu_object *bo)
 {
diff --git a/drivers/gpu/drm/virtio/virtgpu_prime.c b/drivers/gpu/drm/virtio/virtgpu_prime.c
new file mode 100644
index 0000000..385e0eb
--- /dev/null
+++ b/drivers/gpu/drm/virtio/virtgpu_prime.c
@@ -0,0 +1,71 @@
+/*
+ * Copyright 2014 Canonical
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Andreas Pokorny
+ */
+
+#include "virtgpu_drv.h"
+
+/* Empty Implementations as there should not be any other driver for a virtual
+ * device that might share buffers with virtgpu */
+
+int virtgpu_gem_prime_pin(struct drm_gem_object *obj)
+{
+	WARN_ONCE(1, "not implemented");
+	return -ENODEV;
+}
+
+void virtgpu_gem_prime_unpin(struct drm_gem_object *obj)
+{
+	WARN_ONCE(1, "not implemented");
+}
+
+
+struct sg_table *virtgpu_gem_prime_get_sg_table(struct drm_gem_object *obj)
+{
+	WARN_ONCE(1, "not implemented");
+	return ERR_PTR(-ENODEV);
+}
+
+struct drm_gem_object *virtgpu_gem_prime_import_sg_table(
+	struct drm_device *dev, struct dma_buf_attachment *attach,
+	struct sg_table *table)
+{
+	WARN_ONCE(1, "not implemented");
+	return ERR_PTR(-ENODEV);
+}
+
+void *virtgpu_gem_prime_vmap(struct drm_gem_object *obj)
+{
+	WARN_ONCE(1, "not implemented");
+	return ERR_PTR(-ENODEV);
+}
+
+void virtgpu_gem_prime_vunmap(struct drm_gem_object *obj, void *vaddr)
+{
+	WARN_ONCE(1, "not implemented");
+}
+
+int virtgpu_gem_prime_mmap(struct drm_gem_object *obj,
+		       struct vm_area_struct *area)
+{
+	return -ENODEV;
+}
-- 
cgit v0.10.2


From f3380a3015541f13c25c980011c046e7114dff3d Mon Sep 17 00:00:00 2001
From: Dave Airlie <airlied@redhat.com>
Date: Tue, 16 Jun 2015 15:41:56 +1000
Subject: virtio-gpu: mark as a render gpu

Also add DRM_RENDER_ALLOW to the ioctls.

Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>

diff --git a/drivers/gpu/drm/virtio/virtgpu_drv.c b/drivers/gpu/drm/virtio/virtgpu_drv.c
index 1f15ef8..b40ed60 100644
--- a/drivers/gpu/drm/virtio/virtgpu_drv.c
+++ b/drivers/gpu/drm/virtio/virtgpu_drv.c
@@ -118,7 +118,7 @@ static const struct file_operations virtio_gpu_driver_fops = {
 
 
 static struct drm_driver driver = {
-	.driver_features = DRIVER_MODESET | DRIVER_GEM | DRIVER_PRIME,
+	.driver_features = DRIVER_MODESET | DRIVER_GEM | DRIVER_PRIME | DRIVER_RENDER,
 	.set_busid = drm_virtio_set_busid,
 	.load = virtio_gpu_driver_load,
 	.unload = virtio_gpu_driver_unload,
diff --git a/drivers/gpu/drm/virtio/virtgpu_ioctl.c b/drivers/gpu/drm/virtio/virtgpu_ioctl.c
index 4ef672b..b4de18e 100644
--- a/drivers/gpu/drm/virtio/virtgpu_ioctl.c
+++ b/drivers/gpu/drm/virtio/virtgpu_ioctl.c
@@ -541,33 +541,33 @@ copy_exit:
 
 struct drm_ioctl_desc virtio_gpu_ioctls[DRM_VIRTIO_NUM_IOCTLS] = {
 	DRM_IOCTL_DEF_DRV(VIRTGPU_MAP, virtio_gpu_map_ioctl,
-			  DRM_AUTH|DRM_UNLOCKED),
+			  DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
 
 	DRM_IOCTL_DEF_DRV(VIRTGPU_EXECBUFFER, virtio_gpu_execbuffer_ioctl,
-			  DRM_AUTH|DRM_UNLOCKED),
+			  DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
 
 	DRM_IOCTL_DEF_DRV(VIRTGPU_GETPARAM, virtio_gpu_getparam_ioctl,
-			  DRM_AUTH|DRM_UNLOCKED),
+			  DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
 
 	DRM_IOCTL_DEF_DRV(VIRTGPU_RESOURCE_CREATE,
 			  virtio_gpu_resource_create_ioctl,
-			  DRM_AUTH|DRM_UNLOCKED),
+			  DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
 
 	DRM_IOCTL_DEF_DRV(VIRTGPU_RESOURCE_INFO, virtio_gpu_resource_info_ioctl,
-			  DRM_AUTH|DRM_UNLOCKED),
+			  DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
 
 	/* make transfer async to the main ring? - no sure, can we
 	   thread these in the underlying GL */
 	DRM_IOCTL_DEF_DRV(VIRTGPU_TRANSFER_FROM_HOST,
 			  virtio_gpu_transfer_from_host_ioctl,
-			  DRM_AUTH|DRM_UNLOCKED),
+			  DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(VIRTGPU_TRANSFER_TO_HOST,
 			  virtio_gpu_transfer_to_host_ioctl,
-			  DRM_AUTH|DRM_UNLOCKED),
+			  DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
 
 	DRM_IOCTL_DEF_DRV(VIRTGPU_WAIT, virtio_gpu_wait_ioctl,
-			  DRM_AUTH|DRM_UNLOCKED),
+			  DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
 
 	DRM_IOCTL_DEF_DRV(VIRTGPU_GET_CAPS, virtio_gpu_get_caps_ioctl,
-			  DRM_AUTH|DRM_UNLOCKED),
+			  DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
 };
-- 
cgit v0.10.2


From d24796a4a13a6c6a37b9e7145d36f983a27ad2e2 Mon Sep 17 00:00:00 2001
From: Gerd Hoffmann <kraxel@redhat.com>
Date: Wed, 23 Sep 2015 12:19:11 +0200
Subject: virtio-gpu: add page flip support

Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>

diff --git a/drivers/gpu/drm/virtio/virtgpu_display.c b/drivers/gpu/drm/virtio/virtgpu_display.c
index c9c1427..f545913 100644
--- a/drivers/gpu/drm/virtio/virtgpu_display.c
+++ b/drivers/gpu/drm/virtio/virtgpu_display.c
@@ -125,6 +125,51 @@ static int virtio_gpu_crtc_cursor_move(struct drm_crtc *crtc,
 	return 0;
 }
 
+static int virtio_gpu_page_flip(struct drm_crtc *crtc,
+				struct drm_framebuffer *fb,
+				struct drm_pending_vblank_event *event,
+				uint32_t flags)
+{
+	struct virtio_gpu_device *vgdev = crtc->dev->dev_private;
+	struct virtio_gpu_output *output =
+		container_of(crtc, struct virtio_gpu_output, crtc);
+	struct drm_plane *plane = crtc->primary;
+	struct virtio_gpu_framebuffer *vgfb;
+	struct virtio_gpu_object *bo;
+	unsigned long irqflags;
+	uint32_t handle;
+
+	plane->fb = fb;
+	vgfb = to_virtio_gpu_framebuffer(plane->fb);
+	bo = gem_to_virtio_gpu_obj(vgfb->obj);
+	handle = bo->hw_res_handle;
+
+	DRM_DEBUG("handle 0x%x%s, crtc %dx%d\n", handle,
+		  bo->dumb ? ", dumb" : "",
+		  crtc->mode.hdisplay, crtc->mode.vdisplay);
+	if (bo->dumb) {
+		virtio_gpu_cmd_transfer_to_host_2d
+			(vgdev, handle, 0,
+			 cpu_to_le32(crtc->mode.hdisplay),
+			 cpu_to_le32(crtc->mode.vdisplay),
+			 0, 0, NULL);
+	}
+	virtio_gpu_cmd_set_scanout(vgdev, output->index, handle,
+				   crtc->mode.hdisplay,
+				   crtc->mode.vdisplay, 0, 0);
+	virtio_gpu_cmd_resource_flush(vgdev, handle, 0, 0,
+				      crtc->mode.hdisplay,
+				      crtc->mode.vdisplay);
+
+	if (event) {
+		spin_lock_irqsave(&crtc->dev->event_lock, irqflags);
+		drm_send_vblank_event(crtc->dev, -1, event);
+		spin_unlock_irqrestore(&crtc->dev->event_lock, irqflags);
+	}
+
+	return 0;
+}
+
 static const struct drm_crtc_funcs virtio_gpu_crtc_funcs = {
 	.cursor_set2            = virtio_gpu_crtc_cursor_set,
 	.cursor_move            = virtio_gpu_crtc_cursor_move,
@@ -132,9 +177,7 @@ static const struct drm_crtc_funcs virtio_gpu_crtc_funcs = {
 	.set_config             = drm_atomic_helper_set_config,
 	.destroy                = drm_crtc_cleanup,
 
-#if 0 /* not (yet) working without vblank support according to docs */
-	.page_flip              = drm_atomic_helper_page_flip,
-#endif
+	.page_flip              = virtio_gpu_page_flip,
 	.reset                  = drm_atomic_helper_crtc_reset,
 	.atomic_duplicate_state = drm_atomic_helper_crtc_duplicate_state,
 	.atomic_destroy_state   = drm_atomic_helper_crtc_destroy_state,
-- 
cgit v0.10.2


From 1c806a6e76795b72b09e2bf967d23b4ed791d0a2 Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Tue, 8 Sep 2015 13:56:25 +0200
Subject: drm/i915: Mark getparam ioctl as DRM_UNLOCKED

With kms all the data getparam looks at is actually invariant, and
certainly not protected by the global kms mutex. With ums all the
setup code is already racy as hell, so this won't make things any
worse.

I've done this change so that all ioctl still used by kms drivers
are marked as DRM_UNLOCKED, besides that we obviously don't need
it any more in kms mode.

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index d1454b5..656cbf2 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -1196,7 +1196,7 @@ const struct drm_ioctl_desc i915_ioctls[] = {
 	DRM_IOCTL_DEF_DRV(I915_BATCHBUFFER, drm_noop, DRM_AUTH),
 	DRM_IOCTL_DEF_DRV(I915_IRQ_EMIT, drm_noop, DRM_AUTH),
 	DRM_IOCTL_DEF_DRV(I915_IRQ_WAIT, drm_noop, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(I915_GETPARAM, i915_getparam, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(I915_GETPARAM, i915_getparam, DRM_AUTH|DRM_RENDER_ALLOW|DRM_UNLOCKED),
 	DRM_IOCTL_DEF_DRV(I915_SETPARAM, drm_noop, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
 	DRM_IOCTL_DEF_DRV(I915_ALLOC, drm_noop, DRM_AUTH),
 	DRM_IOCTL_DEF_DRV(I915_FREE, drm_noop, DRM_AUTH),
-- 
cgit v0.10.2


From ea487835e8876abf7ad909636e308c801a2bcda6 Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Mon, 28 Sep 2015 21:42:40 +0200
Subject: drm: Enforce unlocked ioctl operation for kms driver ioctls
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

With the prep patches for i915 all kms drivers either have
DRM_UNLOCKED on all their ioctls. Or the ioctl always directly returns
with an invariant return value when in modeset mode. But that's only
the case for i915 and radeon. The drm core ioctls are unfortunately
too much a mess still to dare this.

Follow-up patches will remove DRM_UNLOCKED from all kms drivers to
prove that this is indeed the case.

Also update the documentation.

v2: Really only do this for driver ioctls, spotted by David Herrmann.
And drop spurious whitespace change.

Cc: David Herrmann <dh.herrmann@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Reviewed-by: David Herrmann <dh.herrmann@gmail.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/Documentation/DocBook/drm.tmpl b/Documentation/DocBook/drm.tmpl
index 8d0cbf1..2bc6c80 100644
--- a/Documentation/DocBook/drm.tmpl
+++ b/Documentation/DocBook/drm.tmpl
@@ -3684,7 +3684,9 @@ int num_ioctls;</synopsis>
 	    </para></listitem>
             <listitem><para>
 	      DRM_UNLOCKED - The ioctl handler will be called without locking
-	      the DRM global mutex
+	      the DRM global mutex. This is the enforced default for kms drivers
+	      (i.e. using the DRIVER_MODESET flag) and hence shouldn't be used
+	      any more for new drivers.
 	    </para></listitem>
 	  </itemizedlist>
 	</para>
diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c
index 530c501..8ce2a0c 100644
--- a/drivers/gpu/drm/drm_ioctl.c
+++ b/drivers/gpu/drm/drm_ioctl.c
@@ -691,13 +691,16 @@ long drm_ioctl(struct file *filp,
 	char stack_kdata[128];
 	char *kdata = NULL;
 	unsigned int usize, asize, drv_size;
+	bool is_driver_ioctl;
 
 	dev = file_priv->minor->dev;
 
 	if (drm_device_is_unplugged(dev))
 		return -ENODEV;
 
-	if (nr >= DRM_COMMAND_BASE && nr < DRM_COMMAND_END) {
+	is_driver_ioctl = nr >= DRM_COMMAND_BASE && nr < DRM_COMMAND_END;
+
+	if (is_driver_ioctl) {
 		/* driver ioctl */
 		if (nr - DRM_COMMAND_BASE >= dev->driver->num_ioctls)
 			goto err_i1;
@@ -756,7 +759,10 @@ long drm_ioctl(struct file *filp,
 		memset(kdata, 0, usize);
 	}
 
-	if (ioctl->flags & DRM_UNLOCKED)
+	/* Enforce sane locking for kms driver ioctls. Core ioctls are
+	 * too messy still. */
+	if ((drm_core_check_feature(dev, DRIVER_MODESET) && is_driver_ioctl) ||
+	    (ioctl->flags & DRM_UNLOCKED))
 		retcode = func(dev, kdata, file_priv);
 	else {
 		mutex_lock(&drm_global_mutex);
-- 
cgit v0.10.2


From f8c47144bf2964c8599ccce350ef71b62c2cbe28 Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Tue, 8 Sep 2015 13:56:30 +0200
Subject: drm/<drivers>: Drop DRM_UNLOCKED from modeset drivers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Just one special case (since i915 lost its ums code, yay):
- radeon: Has slots for the old ums ioctls which don't have
  DRM_UNLOCKED, but all filled with drm_invalid_op. So ok to drop it
  everywhere.

Every other kms driver just has DRM_UNLOCKED for all their ioctls, as
they should.

v2: admgpu happened, include that one too. And i915 lost its UMS
support which means we can change all the i915 ioctls too.

v3: Rebased on top of new vmwgfx DX interface extensions.

v4: Rebase on top of render-node support in exynos.

Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Reviewed-by: David Herrmann <dh.herrmann@gmail.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
index b9faaf8..371f015 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
@@ -686,18 +686,18 @@ int amdgpu_get_vblank_timestamp_kms(struct drm_device *dev, unsigned int pipe,
 }
 
 const struct drm_ioctl_desc amdgpu_ioctls_kms[] = {
-	DRM_IOCTL_DEF_DRV(AMDGPU_GEM_CREATE, amdgpu_gem_create_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(AMDGPU_CTX, amdgpu_ctx_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(AMDGPU_BO_LIST, amdgpu_bo_list_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(AMDGPU_GEM_CREATE, amdgpu_gem_create_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(AMDGPU_CTX, amdgpu_ctx_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(AMDGPU_BO_LIST, amdgpu_bo_list_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
 	/* KMS */
-	DRM_IOCTL_DEF_DRV(AMDGPU_GEM_MMAP, amdgpu_gem_mmap_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(AMDGPU_GEM_WAIT_IDLE, amdgpu_gem_wait_idle_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(AMDGPU_CS, amdgpu_cs_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(AMDGPU_INFO, amdgpu_info_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(AMDGPU_WAIT_CS, amdgpu_cs_wait_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(AMDGPU_GEM_METADATA, amdgpu_gem_metadata_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(AMDGPU_GEM_VA, amdgpu_gem_va_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(AMDGPU_GEM_OP, amdgpu_gem_op_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(AMDGPU_GEM_USERPTR, amdgpu_gem_userptr_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(AMDGPU_GEM_MMAP, amdgpu_gem_mmap_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(AMDGPU_GEM_WAIT_IDLE, amdgpu_gem_wait_idle_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(AMDGPU_CS, amdgpu_cs_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(AMDGPU_INFO, amdgpu_info_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(AMDGPU_WAIT_CS, amdgpu_cs_wait_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(AMDGPU_GEM_METADATA, amdgpu_gem_metadata_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(AMDGPU_GEM_VA, amdgpu_gem_va_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(AMDGPU_GEM_OP, amdgpu_gem_op_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(AMDGPU_GEM_USERPTR, amdgpu_gem_userptr_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
 };
 int amdgpu_max_kms_ioctl = ARRAY_SIZE(amdgpu_ioctls_kms);
diff --git a/drivers/gpu/drm/armada/armada_drv.c b/drivers/gpu/drm/armada/armada_drv.c
index f91a496..5646b54 100644
--- a/drivers/gpu/drm/armada/armada_drv.c
+++ b/drivers/gpu/drm/armada/armada_drv.c
@@ -268,12 +268,9 @@ static void armada_drm_disable_vblank(struct drm_device *dev, unsigned int pipe)
 }
 
 static struct drm_ioctl_desc armada_ioctls[] = {
-	DRM_IOCTL_DEF_DRV(ARMADA_GEM_CREATE, armada_gem_create_ioctl,
-		DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(ARMADA_GEM_MMAP, armada_gem_mmap_ioctl,
-		DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(ARMADA_GEM_PWRITE, armada_gem_pwrite_ioctl,
-		DRM_UNLOCKED),
+	DRM_IOCTL_DEF_DRV(ARMADA_GEM_CREATE, armada_gem_create_ioctl,0),
+	DRM_IOCTL_DEF_DRV(ARMADA_GEM_MMAP, armada_gem_mmap_ioctl, 0),
+	DRM_IOCTL_DEF_DRV(ARMADA_GEM_PWRITE, armada_gem_pwrite_ioctl, 0),
 };
 
 static void armada_drm_lastclose(struct drm_device *dev)
diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.c b/drivers/gpu/drm/exynos/exynos_drm_drv.c
index fb9cfc5..3d565be 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_drv.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_drv.c
@@ -403,25 +403,25 @@ static const struct vm_operations_struct exynos_drm_gem_vm_ops = {
 
 static const struct drm_ioctl_desc exynos_ioctls[] = {
 	DRM_IOCTL_DEF_DRV(EXYNOS_GEM_CREATE, exynos_drm_gem_create_ioctl,
-			DRM_UNLOCKED | DRM_AUTH | DRM_RENDER_ALLOW),
+			DRM_AUTH | DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(EXYNOS_GEM_GET, exynos_drm_gem_get_ioctl,
-			DRM_UNLOCKED | DRM_RENDER_ALLOW),
+			DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(EXYNOS_VIDI_CONNECTION, vidi_connection_ioctl,
-			DRM_UNLOCKED | DRM_AUTH),
+			DRM_AUTH),
 	DRM_IOCTL_DEF_DRV(EXYNOS_G2D_GET_VER, exynos_g2d_get_ver_ioctl,
-			DRM_UNLOCKED | DRM_AUTH | DRM_RENDER_ALLOW),
+			DRM_AUTH | DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(EXYNOS_G2D_SET_CMDLIST, exynos_g2d_set_cmdlist_ioctl,
-			DRM_UNLOCKED | DRM_AUTH | DRM_RENDER_ALLOW),
+			DRM_AUTH | DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(EXYNOS_G2D_EXEC, exynos_g2d_exec_ioctl,
-			DRM_UNLOCKED | DRM_AUTH | DRM_RENDER_ALLOW),
+			DRM_AUTH | DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(EXYNOS_IPP_GET_PROPERTY, exynos_drm_ipp_get_property,
-			DRM_UNLOCKED | DRM_AUTH | DRM_RENDER_ALLOW),
+			DRM_AUTH | DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(EXYNOS_IPP_SET_PROPERTY, exynos_drm_ipp_set_property,
-			DRM_UNLOCKED | DRM_AUTH | DRM_RENDER_ALLOW),
+			DRM_AUTH | DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(EXYNOS_IPP_QUEUE_BUF, exynos_drm_ipp_queue_buf,
-			DRM_UNLOCKED | DRM_AUTH | DRM_RENDER_ALLOW),
+			DRM_AUTH | DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(EXYNOS_IPP_CMD_CTRL, exynos_drm_ipp_cmd_ctrl,
-			DRM_UNLOCKED | DRM_AUTH | DRM_RENDER_ALLOW),
+			DRM_AUTH | DRM_RENDER_ALLOW),
 };
 
 static const struct file_operations exynos_drm_driver_fops = {
diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index 656cbf2..68b0c9e 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -1196,7 +1196,7 @@ const struct drm_ioctl_desc i915_ioctls[] = {
 	DRM_IOCTL_DEF_DRV(I915_BATCHBUFFER, drm_noop, DRM_AUTH),
 	DRM_IOCTL_DEF_DRV(I915_IRQ_EMIT, drm_noop, DRM_AUTH),
 	DRM_IOCTL_DEF_DRV(I915_IRQ_WAIT, drm_noop, DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(I915_GETPARAM, i915_getparam, DRM_AUTH|DRM_RENDER_ALLOW|DRM_UNLOCKED),
+	DRM_IOCTL_DEF_DRV(I915_GETPARAM, i915_getparam, DRM_AUTH|DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(I915_SETPARAM, drm_noop, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
 	DRM_IOCTL_DEF_DRV(I915_ALLOC, drm_noop, DRM_AUTH),
 	DRM_IOCTL_DEF_DRV(I915_FREE, drm_noop, DRM_AUTH),
@@ -1207,41 +1207,41 @@ const struct drm_ioctl_desc i915_ioctls[] = {
 	DRM_IOCTL_DEF_DRV(I915_GET_VBLANK_PIPE,  drm_noop, DRM_AUTH),
 	DRM_IOCTL_DEF_DRV(I915_VBLANK_SWAP, drm_noop, DRM_AUTH),
 	DRM_IOCTL_DEF_DRV(I915_HWS_ADDR, drm_noop, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
-	DRM_IOCTL_DEF_DRV(I915_GEM_INIT, drm_noop, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY|DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(I915_GEM_EXECBUFFER, i915_gem_execbuffer, DRM_AUTH|DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(I915_GEM_EXECBUFFER2, i915_gem_execbuffer2, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(I915_GEM_PIN, i915_gem_reject_pin_ioctl, DRM_AUTH|DRM_ROOT_ONLY|DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(I915_GEM_UNPIN, i915_gem_reject_pin_ioctl, DRM_AUTH|DRM_ROOT_ONLY|DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(I915_GEM_BUSY, i915_gem_busy_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(I915_GEM_SET_CACHING, i915_gem_set_caching_ioctl, DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(I915_GEM_GET_CACHING, i915_gem_get_caching_ioctl, DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(I915_GEM_THROTTLE, i915_gem_throttle_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(I915_GEM_ENTERVT, drm_noop, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY|DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(I915_GEM_LEAVEVT, drm_noop, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY|DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(I915_GEM_CREATE, i915_gem_create_ioctl, DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(I915_GEM_PREAD, i915_gem_pread_ioctl, DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(I915_GEM_PWRITE, i915_gem_pwrite_ioctl, DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(I915_GEM_MMAP, i915_gem_mmap_ioctl, DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(I915_GEM_MMAP_GTT, i915_gem_mmap_gtt_ioctl, DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(I915_GEM_SET_DOMAIN, i915_gem_set_domain_ioctl, DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(I915_GEM_SW_FINISH, i915_gem_sw_finish_ioctl, DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(I915_GEM_SET_TILING, i915_gem_set_tiling, DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(I915_GEM_GET_TILING, i915_gem_get_tiling, DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(I915_GEM_GET_APERTURE, i915_gem_get_aperture_ioctl, DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(I915_GET_PIPE_FROM_CRTC_ID, intel_get_pipe_from_crtc_id, DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(I915_GEM_MADVISE, i915_gem_madvise_ioctl, DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(I915_OVERLAY_PUT_IMAGE, intel_overlay_put_image, DRM_MASTER|DRM_CONTROL_ALLOW|DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(I915_OVERLAY_ATTRS, intel_overlay_attrs, DRM_MASTER|DRM_CONTROL_ALLOW|DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(I915_SET_SPRITE_COLORKEY, intel_sprite_set_colorkey, DRM_MASTER|DRM_CONTROL_ALLOW|DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(I915_GET_SPRITE_COLORKEY, drm_noop, DRM_MASTER|DRM_CONTROL_ALLOW|DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(I915_GEM_WAIT, i915_gem_wait_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(I915_GEM_CONTEXT_CREATE, i915_gem_context_create_ioctl, DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(I915_GEM_CONTEXT_DESTROY, i915_gem_context_destroy_ioctl, DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(I915_REG_READ, i915_reg_read_ioctl, DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(I915_GET_RESET_STATS, i915_get_reset_stats_ioctl, DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(I915_GEM_USERPTR, i915_gem_userptr_ioctl, DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(I915_GEM_CONTEXT_GETPARAM, i915_gem_context_getparam_ioctl, DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(I915_GEM_CONTEXT_SETPARAM, i915_gem_context_setparam_ioctl, DRM_UNLOCKED|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(I915_GEM_INIT, drm_noop, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
+	DRM_IOCTL_DEF_DRV(I915_GEM_EXECBUFFER, i915_gem_execbuffer, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(I915_GEM_EXECBUFFER2, i915_gem_execbuffer2, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(I915_GEM_PIN, i915_gem_reject_pin_ioctl, DRM_AUTH|DRM_ROOT_ONLY),
+	DRM_IOCTL_DEF_DRV(I915_GEM_UNPIN, i915_gem_reject_pin_ioctl, DRM_AUTH|DRM_ROOT_ONLY),
+	DRM_IOCTL_DEF_DRV(I915_GEM_BUSY, i915_gem_busy_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(I915_GEM_SET_CACHING, i915_gem_set_caching_ioctl, DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(I915_GEM_GET_CACHING, i915_gem_get_caching_ioctl, DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(I915_GEM_THROTTLE, i915_gem_throttle_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(I915_GEM_ENTERVT, drm_noop, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
+	DRM_IOCTL_DEF_DRV(I915_GEM_LEAVEVT, drm_noop, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
+	DRM_IOCTL_DEF_DRV(I915_GEM_CREATE, i915_gem_create_ioctl, DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(I915_GEM_PREAD, i915_gem_pread_ioctl, DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(I915_GEM_PWRITE, i915_gem_pwrite_ioctl, DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(I915_GEM_MMAP, i915_gem_mmap_ioctl, DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(I915_GEM_MMAP_GTT, i915_gem_mmap_gtt_ioctl, DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(I915_GEM_SET_DOMAIN, i915_gem_set_domain_ioctl, DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(I915_GEM_SW_FINISH, i915_gem_sw_finish_ioctl, DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(I915_GEM_SET_TILING, i915_gem_set_tiling, DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(I915_GEM_GET_TILING, i915_gem_get_tiling, DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(I915_GEM_GET_APERTURE, i915_gem_get_aperture_ioctl, DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(I915_GET_PIPE_FROM_CRTC_ID, intel_get_pipe_from_crtc_id, 0),
+	DRM_IOCTL_DEF_DRV(I915_GEM_MADVISE, i915_gem_madvise_ioctl, DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(I915_OVERLAY_PUT_IMAGE, intel_overlay_put_image, DRM_MASTER|DRM_CONTROL_ALLOW),
+	DRM_IOCTL_DEF_DRV(I915_OVERLAY_ATTRS, intel_overlay_attrs, DRM_MASTER|DRM_CONTROL_ALLOW),
+	DRM_IOCTL_DEF_DRV(I915_SET_SPRITE_COLORKEY, intel_sprite_set_colorkey, DRM_MASTER|DRM_CONTROL_ALLOW),
+	DRM_IOCTL_DEF_DRV(I915_GET_SPRITE_COLORKEY, drm_noop, DRM_MASTER|DRM_CONTROL_ALLOW),
+	DRM_IOCTL_DEF_DRV(I915_GEM_WAIT, i915_gem_wait_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(I915_GEM_CONTEXT_CREATE, i915_gem_context_create_ioctl, DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(I915_GEM_CONTEXT_DESTROY, i915_gem_context_destroy_ioctl, DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(I915_REG_READ, i915_reg_read_ioctl, DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(I915_GET_RESET_STATS, i915_get_reset_stats_ioctl, DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(I915_GEM_USERPTR, i915_gem_userptr_ioctl, DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(I915_GEM_CONTEXT_GETPARAM, i915_gem_context_getparam_ioctl, DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(I915_GEM_CONTEXT_SETPARAM, i915_gem_context_setparam_ioctl, DRM_RENDER_ALLOW),
 };
 
 int i915_max_ioctl = ARRAY_SIZE(i915_ioctls);
diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c
index a06ec71..d170131 100644
--- a/drivers/gpu/drm/msm/msm_drv.c
+++ b/drivers/gpu/drm/msm/msm_drv.c
@@ -932,13 +932,13 @@ static int msm_ioctl_wait_fence(struct drm_device *dev, void *data,
 }
 
 static const struct drm_ioctl_desc msm_ioctls[] = {
-	DRM_IOCTL_DEF_DRV(MSM_GET_PARAM,    msm_ioctl_get_param,    DRM_UNLOCKED|DRM_AUTH|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(MSM_GEM_NEW,      msm_ioctl_gem_new,      DRM_UNLOCKED|DRM_AUTH|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(MSM_GEM_INFO,     msm_ioctl_gem_info,     DRM_UNLOCKED|DRM_AUTH|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(MSM_GEM_CPU_PREP, msm_ioctl_gem_cpu_prep, DRM_UNLOCKED|DRM_AUTH|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(MSM_GEM_CPU_FINI, msm_ioctl_gem_cpu_fini, DRM_UNLOCKED|DRM_AUTH|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(MSM_GEM_SUBMIT,   msm_ioctl_gem_submit,   DRM_UNLOCKED|DRM_AUTH|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(MSM_WAIT_FENCE,   msm_ioctl_wait_fence,   DRM_UNLOCKED|DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(MSM_GET_PARAM,    msm_ioctl_get_param,    DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(MSM_GEM_NEW,      msm_ioctl_gem_new,      DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(MSM_GEM_INFO,     msm_ioctl_gem_info,     DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(MSM_GEM_CPU_PREP, msm_ioctl_gem_cpu_prep, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(MSM_GEM_CPU_FINI, msm_ioctl_gem_cpu_fini, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(MSM_GEM_SUBMIT,   msm_ioctl_gem_submit,   DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(MSM_WAIT_FENCE,   msm_ioctl_wait_fence,   DRM_AUTH|DRM_RENDER_ALLOW),
 };
 
 static const struct vm_operations_struct vm_ops = {
diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c
index 2416c7d..45ba678 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_drm.c
@@ -862,18 +862,18 @@ nouveau_drm_postclose(struct drm_device *dev, struct drm_file *fpriv)
 
 static const struct drm_ioctl_desc
 nouveau_ioctls[] = {
-	DRM_IOCTL_DEF_DRV(NOUVEAU_GETPARAM, nouveau_abi16_ioctl_getparam, DRM_UNLOCKED|DRM_AUTH|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(NOUVEAU_SETPARAM, nouveau_abi16_ioctl_setparam, DRM_UNLOCKED|DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
-	DRM_IOCTL_DEF_DRV(NOUVEAU_CHANNEL_ALLOC, nouveau_abi16_ioctl_channel_alloc, DRM_UNLOCKED|DRM_AUTH|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(NOUVEAU_CHANNEL_FREE, nouveau_abi16_ioctl_channel_free, DRM_UNLOCKED|DRM_AUTH|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(NOUVEAU_GROBJ_ALLOC, nouveau_abi16_ioctl_grobj_alloc, DRM_UNLOCKED|DRM_AUTH|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(NOUVEAU_NOTIFIEROBJ_ALLOC, nouveau_abi16_ioctl_notifierobj_alloc, DRM_UNLOCKED|DRM_AUTH|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(NOUVEAU_GPUOBJ_FREE, nouveau_abi16_ioctl_gpuobj_free, DRM_UNLOCKED|DRM_AUTH|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(NOUVEAU_GEM_NEW, nouveau_gem_ioctl_new, DRM_UNLOCKED|DRM_AUTH|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(NOUVEAU_GEM_PUSHBUF, nouveau_gem_ioctl_pushbuf, DRM_UNLOCKED|DRM_AUTH|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(NOUVEAU_GEM_CPU_PREP, nouveau_gem_ioctl_cpu_prep, DRM_UNLOCKED|DRM_AUTH|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(NOUVEAU_GEM_CPU_FINI, nouveau_gem_ioctl_cpu_fini, DRM_UNLOCKED|DRM_AUTH|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(NOUVEAU_GEM_INFO, nouveau_gem_ioctl_info, DRM_UNLOCKED|DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(NOUVEAU_GETPARAM, nouveau_abi16_ioctl_getparam, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(NOUVEAU_SETPARAM, nouveau_abi16_ioctl_setparam, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
+	DRM_IOCTL_DEF_DRV(NOUVEAU_CHANNEL_ALLOC, nouveau_abi16_ioctl_channel_alloc, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(NOUVEAU_CHANNEL_FREE, nouveau_abi16_ioctl_channel_free, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(NOUVEAU_GROBJ_ALLOC, nouveau_abi16_ioctl_grobj_alloc, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(NOUVEAU_NOTIFIEROBJ_ALLOC, nouveau_abi16_ioctl_notifierobj_alloc, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(NOUVEAU_GPUOBJ_FREE, nouveau_abi16_ioctl_gpuobj_free, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(NOUVEAU_GEM_NEW, nouveau_gem_ioctl_new, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(NOUVEAU_GEM_PUSHBUF, nouveau_gem_ioctl_pushbuf, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(NOUVEAU_GEM_CPU_PREP, nouveau_gem_ioctl_cpu_prep, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(NOUVEAU_GEM_CPU_FINI, nouveau_gem_ioctl_cpu_fini, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(NOUVEAU_GEM_INFO, nouveau_gem_ioctl_info, DRM_AUTH|DRM_RENDER_ALLOW),
 };
 
 long
diff --git a/drivers/gpu/drm/omapdrm/omap_drv.c b/drivers/gpu/drm/omapdrm/omap_drv.c
index 4d58934..5c6609c 100644
--- a/drivers/gpu/drm/omapdrm/omap_drv.c
+++ b/drivers/gpu/drm/omapdrm/omap_drv.c
@@ -626,12 +626,12 @@ static int ioctl_gem_info(struct drm_device *dev, void *data,
 }
 
 static const struct drm_ioctl_desc ioctls[DRM_COMMAND_END - DRM_COMMAND_BASE] = {
-	DRM_IOCTL_DEF_DRV(OMAP_GET_PARAM, ioctl_get_param, DRM_UNLOCKED|DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(OMAP_SET_PARAM, ioctl_set_param, DRM_UNLOCKED|DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
-	DRM_IOCTL_DEF_DRV(OMAP_GEM_NEW, ioctl_gem_new, DRM_UNLOCKED|DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(OMAP_GEM_CPU_PREP, ioctl_gem_cpu_prep, DRM_UNLOCKED|DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(OMAP_GEM_CPU_FINI, ioctl_gem_cpu_fini, DRM_UNLOCKED|DRM_AUTH),
-	DRM_IOCTL_DEF_DRV(OMAP_GEM_INFO, ioctl_gem_info, DRM_UNLOCKED|DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(OMAP_GET_PARAM, ioctl_get_param, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(OMAP_SET_PARAM, ioctl_set_param, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY),
+	DRM_IOCTL_DEF_DRV(OMAP_GEM_NEW, ioctl_gem_new, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(OMAP_GEM_CPU_PREP, ioctl_gem_cpu_prep, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(OMAP_GEM_CPU_FINI, ioctl_gem_cpu_fini, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(OMAP_GEM_INFO, ioctl_gem_info, DRM_AUTH),
 };
 
 /*
diff --git a/drivers/gpu/drm/qxl/qxl_ioctl.c b/drivers/gpu/drm/qxl/qxl_ioctl.c
index bda5c5f..2ae8577 100644
--- a/drivers/gpu/drm/qxl/qxl_ioctl.c
+++ b/drivers/gpu/drm/qxl/qxl_ioctl.c
@@ -422,21 +422,21 @@ static int qxl_alloc_surf_ioctl(struct drm_device *dev, void *data,
 }
 
 const struct drm_ioctl_desc qxl_ioctls[] = {
-	DRM_IOCTL_DEF_DRV(QXL_ALLOC, qxl_alloc_ioctl, DRM_AUTH|DRM_UNLOCKED),
+	DRM_IOCTL_DEF_DRV(QXL_ALLOC, qxl_alloc_ioctl, DRM_AUTH),
 
-	DRM_IOCTL_DEF_DRV(QXL_MAP, qxl_map_ioctl, DRM_AUTH|DRM_UNLOCKED),
+	DRM_IOCTL_DEF_DRV(QXL_MAP, qxl_map_ioctl, DRM_AUTH),
 
 	DRM_IOCTL_DEF_DRV(QXL_EXECBUFFER, qxl_execbuffer_ioctl,
-							DRM_AUTH|DRM_UNLOCKED),
+							DRM_AUTH),
 	DRM_IOCTL_DEF_DRV(QXL_UPDATE_AREA, qxl_update_area_ioctl,
-							DRM_AUTH|DRM_UNLOCKED),
+							DRM_AUTH),
 	DRM_IOCTL_DEF_DRV(QXL_GETPARAM, qxl_getparam_ioctl,
-							DRM_AUTH|DRM_UNLOCKED),
+							DRM_AUTH),
 	DRM_IOCTL_DEF_DRV(QXL_CLIENTCAP, qxl_clientcap_ioctl,
-							DRM_AUTH|DRM_UNLOCKED),
+							DRM_AUTH),
 
 	DRM_IOCTL_DEF_DRV(QXL_ALLOC_SURF, qxl_alloc_surf_ioctl,
-			  DRM_AUTH|DRM_UNLOCKED),
+			  DRM_AUTH),
 };
 
 int qxl_max_ioctls = ARRAY_SIZE(qxl_ioctls);
diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c
index 2773403..4e2780f 100644
--- a/drivers/gpu/drm/radeon/radeon_kms.c
+++ b/drivers/gpu/drm/radeon/radeon_kms.c
@@ -873,20 +873,20 @@ const struct drm_ioctl_desc radeon_ioctls_kms[] = {
 	DRM_IOCTL_DEF_DRV(RADEON_SURF_ALLOC, drm_invalid_op, DRM_AUTH),
 	DRM_IOCTL_DEF_DRV(RADEON_SURF_FREE, drm_invalid_op, DRM_AUTH),
 	/* KMS */
-	DRM_IOCTL_DEF_DRV(RADEON_GEM_INFO, radeon_gem_info_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(RADEON_GEM_CREATE, radeon_gem_create_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(RADEON_GEM_MMAP, radeon_gem_mmap_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(RADEON_GEM_SET_DOMAIN, radeon_gem_set_domain_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(RADEON_GEM_PREAD, radeon_gem_pread_ioctl, DRM_AUTH|DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(RADEON_GEM_PWRITE, radeon_gem_pwrite_ioctl, DRM_AUTH|DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(RADEON_GEM_WAIT_IDLE, radeon_gem_wait_idle_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(RADEON_CS, radeon_cs_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(RADEON_INFO, radeon_info_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(RADEON_GEM_SET_TILING, radeon_gem_set_tiling_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(RADEON_GEM_GET_TILING, radeon_gem_get_tiling_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(RADEON_GEM_BUSY, radeon_gem_busy_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(RADEON_GEM_VA, radeon_gem_va_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(RADEON_GEM_OP, radeon_gem_op_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
-	DRM_IOCTL_DEF_DRV(RADEON_GEM_USERPTR, radeon_gem_userptr_ioctl, DRM_AUTH|DRM_UNLOCKED|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(RADEON_GEM_INFO, radeon_gem_info_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(RADEON_GEM_CREATE, radeon_gem_create_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(RADEON_GEM_MMAP, radeon_gem_mmap_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(RADEON_GEM_SET_DOMAIN, radeon_gem_set_domain_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(RADEON_GEM_PREAD, radeon_gem_pread_ioctl, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_GEM_PWRITE, radeon_gem_pwrite_ioctl, DRM_AUTH),
+	DRM_IOCTL_DEF_DRV(RADEON_GEM_WAIT_IDLE, radeon_gem_wait_idle_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(RADEON_CS, radeon_cs_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(RADEON_INFO, radeon_info_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(RADEON_GEM_SET_TILING, radeon_gem_set_tiling_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(RADEON_GEM_GET_TILING, radeon_gem_get_tiling_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(RADEON_GEM_BUSY, radeon_gem_busy_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(RADEON_GEM_VA, radeon_gem_va_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(RADEON_GEM_OP, radeon_gem_op_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(RADEON_GEM_USERPTR, radeon_gem_userptr_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
 };
 int radeon_max_kms_ioctl = ARRAY_SIZE(radeon_ioctls_kms);
diff --git a/drivers/gpu/drm/tegra/drm.c b/drivers/gpu/drm/tegra/drm.c
index 759e6af..159ef51 100644
--- a/drivers/gpu/drm/tegra/drm.c
+++ b/drivers/gpu/drm/tegra/drm.c
@@ -778,20 +778,20 @@ static int tegra_gem_get_flags(struct drm_device *drm, void *data,
 
 static const struct drm_ioctl_desc tegra_drm_ioctls[] = {
 #ifdef CONFIG_DRM_TEGRA_STAGING
-	DRM_IOCTL_DEF_DRV(TEGRA_GEM_CREATE, tegra_gem_create, DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(TEGRA_GEM_MMAP, tegra_gem_mmap, DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(TEGRA_SYNCPT_READ, tegra_syncpt_read, DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(TEGRA_SYNCPT_INCR, tegra_syncpt_incr, DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(TEGRA_SYNCPT_WAIT, tegra_syncpt_wait, DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(TEGRA_OPEN_CHANNEL, tegra_open_channel, DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(TEGRA_CLOSE_CHANNEL, tegra_close_channel, DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(TEGRA_GET_SYNCPT, tegra_get_syncpt, DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(TEGRA_SUBMIT, tegra_submit, DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(TEGRA_GET_SYNCPT_BASE, tegra_get_syncpt_base, DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(TEGRA_GEM_SET_TILING, tegra_gem_set_tiling, DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(TEGRA_GEM_GET_TILING, tegra_gem_get_tiling, DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(TEGRA_GEM_SET_FLAGS, tegra_gem_set_flags, DRM_UNLOCKED),
-	DRM_IOCTL_DEF_DRV(TEGRA_GEM_GET_FLAGS, tegra_gem_get_flags, DRM_UNLOCKED),
+	DRM_IOCTL_DEF_DRV(TEGRA_GEM_CREATE, tegra_gem_create, 0),
+	DRM_IOCTL_DEF_DRV(TEGRA_GEM_MMAP, tegra_gem_mmap, 0),
+	DRM_IOCTL_DEF_DRV(TEGRA_SYNCPT_READ, tegra_syncpt_read, 0),
+	DRM_IOCTL_DEF_DRV(TEGRA_SYNCPT_INCR, tegra_syncpt_incr, 0),
+	DRM_IOCTL_DEF_DRV(TEGRA_SYNCPT_WAIT, tegra_syncpt_wait, 0),
+	DRM_IOCTL_DEF_DRV(TEGRA_OPEN_CHANNEL, tegra_open_channel, 0),
+	DRM_IOCTL_DEF_DRV(TEGRA_CLOSE_CHANNEL, tegra_close_channel, 0),
+	DRM_IOCTL_DEF_DRV(TEGRA_GET_SYNCPT, tegra_get_syncpt, 0),
+	DRM_IOCTL_DEF_DRV(TEGRA_SUBMIT, tegra_submit, 0),
+	DRM_IOCTL_DEF_DRV(TEGRA_GET_SYNCPT_BASE, tegra_get_syncpt_base, 0),
+	DRM_IOCTL_DEF_DRV(TEGRA_GEM_SET_TILING, tegra_gem_set_tiling, 0),
+	DRM_IOCTL_DEF_DRV(TEGRA_GEM_GET_TILING, tegra_gem_get_tiling, 0),
+	DRM_IOCTL_DEF_DRV(TEGRA_GEM_SET_FLAGS, tegra_gem_set_flags, 0),
+	DRM_IOCTL_DEF_DRV(TEGRA_GEM_GET_FLAGS, tegra_gem_get_flags, 0),
 #endif
 };
 
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
index 8e7493d..b7525f7 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
@@ -146,73 +146,73 @@
 
 static const struct drm_ioctl_desc vmw_ioctls[] = {
 	VMW_IOCTL_DEF(VMW_GET_PARAM, vmw_getparam_ioctl,
-		      DRM_AUTH | DRM_UNLOCKED | DRM_RENDER_ALLOW),
+		      DRM_AUTH | DRM_RENDER_ALLOW),
 	VMW_IOCTL_DEF(VMW_ALLOC_DMABUF, vmw_dmabuf_alloc_ioctl,
-		      DRM_AUTH | DRM_UNLOCKED | DRM_RENDER_ALLOW),
+		      DRM_AUTH | DRM_RENDER_ALLOW),
 	VMW_IOCTL_DEF(VMW_UNREF_DMABUF, vmw_dmabuf_unref_ioctl,
-		      DRM_UNLOCKED | DRM_RENDER_ALLOW),
+		      DRM_RENDER_ALLOW),
 	VMW_IOCTL_DEF(VMW_CURSOR_BYPASS,
 		      vmw_kms_cursor_bypass_ioctl,
-		      DRM_MASTER | DRM_CONTROL_ALLOW | DRM_UNLOCKED),
+		      DRM_MASTER | DRM_CONTROL_ALLOW),
 
 	VMW_IOCTL_DEF(VMW_CONTROL_STREAM, vmw_overlay_ioctl,
-		      DRM_MASTER | DRM_CONTROL_ALLOW | DRM_UNLOCKED),
+		      DRM_MASTER | DRM_CONTROL_ALLOW),
 	VMW_IOCTL_DEF(VMW_CLAIM_STREAM, vmw_stream_claim_ioctl,
-		      DRM_MASTER | DRM_CONTROL_ALLOW | DRM_UNLOCKED),
+		      DRM_MASTER | DRM_CONTROL_ALLOW),
 	VMW_IOCTL_DEF(VMW_UNREF_STREAM, vmw_stream_unref_ioctl,
-		      DRM_MASTER | DRM_CONTROL_ALLOW | DRM_UNLOCKED),
+		      DRM_MASTER | DRM_CONTROL_ALLOW),
 
 	VMW_IOCTL_DEF(VMW_CREATE_CONTEXT, vmw_context_define_ioctl,
-		      DRM_AUTH | DRM_UNLOCKED | DRM_RENDER_ALLOW),
+		      DRM_AUTH | DRM_RENDER_ALLOW),
 	VMW_IOCTL_DEF(VMW_UNREF_CONTEXT, vmw_context_destroy_ioctl,
-		      DRM_UNLOCKED | DRM_RENDER_ALLOW),
+		      DRM_RENDER_ALLOW),
 	VMW_IOCTL_DEF(VMW_CREATE_SURFACE, vmw_surface_define_ioctl,
-		      DRM_AUTH | DRM_UNLOCKED | DRM_RENDER_ALLOW),
+		      DRM_AUTH | DRM_RENDER_ALLOW),
 	VMW_IOCTL_DEF(VMW_UNREF_SURFACE, vmw_surface_destroy_ioctl,
-		      DRM_UNLOCKED | DRM_RENDER_ALLOW),
+		      DRM_RENDER_ALLOW),
 	VMW_IOCTL_DEF(VMW_REF_SURFACE, vmw_surface_reference_ioctl,
-		      DRM_AUTH | DRM_UNLOCKED | DRM_RENDER_ALLOW),
-	VMW_IOCTL_DEF(VMW_EXECBUF, NULL, DRM_AUTH | DRM_UNLOCKED |
+		      DRM_AUTH | DRM_RENDER_ALLOW),
+	VMW_IOCTL_DEF(VMW_EXECBUF, NULL, DRM_AUTH |
 		      DRM_RENDER_ALLOW),
 	VMW_IOCTL_DEF(VMW_FENCE_WAIT, vmw_fence_obj_wait_ioctl,
-		      DRM_UNLOCKED | DRM_RENDER_ALLOW),
+		      DRM_RENDER_ALLOW),
 	VMW_IOCTL_DEF(VMW_FENCE_SIGNALED,
 		      vmw_fence_obj_signaled_ioctl,
-		      DRM_UNLOCKED | DRM_RENDER_ALLOW),
+		      DRM_RENDER_ALLOW),
 	VMW_IOCTL_DEF(VMW_FENCE_UNREF, vmw_fence_obj_unref_ioctl,
-		      DRM_UNLOCKED | DRM_RENDER_ALLOW),
+		      DRM_RENDER_ALLOW),
 	VMW_IOCTL_DEF(VMW_FENCE_EVENT, vmw_fence_event_ioctl,
-		      DRM_AUTH | DRM_UNLOCKED | DRM_RENDER_ALLOW),
+		      DRM_AUTH | DRM_RENDER_ALLOW),
 	VMW_IOCTL_DEF(VMW_GET_3D_CAP, vmw_get_cap_3d_ioctl,
-		      DRM_AUTH | DRM_UNLOCKED | DRM_RENDER_ALLOW),
+		      DRM_AUTH | DRM_RENDER_ALLOW),
 
 	/* these allow direct access to the framebuffers mark as master only */
 	VMW_IOCTL_DEF(VMW_PRESENT, vmw_present_ioctl,
-		      DRM_MASTER | DRM_AUTH | DRM_UNLOCKED),
+		      DRM_MASTER | DRM_AUTH),
 	VMW_IOCTL_DEF(VMW_PRESENT_READBACK,
 		      vmw_present_readback_ioctl,
-		      DRM_MASTER | DRM_AUTH | DRM_UNLOCKED),
+		      DRM_MASTER | DRM_AUTH),
 	VMW_IOCTL_DEF(VMW_UPDATE_LAYOUT,
 		      vmw_kms_update_layout_ioctl,
-		      DRM_MASTER | DRM_UNLOCKED),
+		      DRM_MASTER),
 	VMW_IOCTL_DEF(VMW_CREATE_SHADER,
 		      vmw_shader_define_ioctl,
-		      DRM_AUTH | DRM_UNLOCKED | DRM_RENDER_ALLOW),
+		      DRM_AUTH | DRM_RENDER_ALLOW),
 	VMW_IOCTL_DEF(VMW_UNREF_SHADER,
 		      vmw_shader_destroy_ioctl,
-		      DRM_UNLOCKED | DRM_RENDER_ALLOW),
+		      DRM_RENDER_ALLOW),
 	VMW_IOCTL_DEF(VMW_GB_SURFACE_CREATE,
 		      vmw_gb_surface_define_ioctl,
-		      DRM_AUTH | DRM_UNLOCKED | DRM_RENDER_ALLOW),
+		      DRM_AUTH | DRM_RENDER_ALLOW),
 	VMW_IOCTL_DEF(VMW_GB_SURFACE_REF,
 		      vmw_gb_surface_reference_ioctl,
-		      DRM_AUTH | DRM_UNLOCKED | DRM_RENDER_ALLOW),
+		      DRM_AUTH | DRM_RENDER_ALLOW),
 	VMW_IOCTL_DEF(VMW_SYNCCPU,
 		      vmw_user_dmabuf_synccpu_ioctl,
-		      DRM_UNLOCKED | DRM_RENDER_ALLOW),
+		      DRM_RENDER_ALLOW),
 	VMW_IOCTL_DEF(VMW_CREATE_EXTENDED_CONTEXT,
 		      vmw_extended_context_define_ioctl,
-		      DRM_AUTH | DRM_UNLOCKED | DRM_RENDER_ALLOW),
+		      DRM_AUTH | DRM_RENDER_ALLOW),
 };
 
 static struct pci_device_id vmw_pci_id_list[] = {
-- 
cgit v0.10.2


From 44844892cb94c4a6a550c0e7bfa9c667f213ee21 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 9 Oct 2015 22:57:36 +0300
Subject: drm: Don't use '\' for string literal concatenation
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

String literals get concatenated just fine on their own,
no need to use '\'.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c
index 7bdf247..bc2e7c6 100644
--- a/drivers/gpu/drm/drm_irq.c
+++ b/drivers/gpu/drm/drm_irq.c
@@ -1272,8 +1272,8 @@ void drm_vblank_off(struct drm_device *dev, unsigned int pipe)
 	list_for_each_entry_safe(e, t, &dev->vblank_event_list, base.link) {
 		if (e->pipe != pipe)
 			continue;
-		DRM_DEBUG("Sending premature vblank event on disable: \
-			  wanted %d, current %d\n",
+		DRM_DEBUG("Sending premature vblank event on disable: "
+			  "wanted %d, current %d\n",
 			  e->event.sequence, seq);
 		list_del(&e->base.link);
 		drm_vblank_put(dev, pipe);
-- 
cgit v0.10.2


From 235fabe09b46469adad2c9e4cb0563758155187c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 9 Oct 2015 22:57:37 +0300
Subject: drm: Add DRM_DEBUG_VBL()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add a new debug class for _verbose_ debug message from the vblank code.
That is message we spew out potentially for every vblank interrupt.
Thierry already got annoyed at the spew, and now I managed to lock up
my box with these debug prints (seems serial console + a few debug
prints every vblank aren't a good combination).

Or should I maybe call it DRM_DEBUG_IRQ?

Cc: Thierry Reding <thierry.reding@gmail.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c
index bc2e7c6..eba6337 100644
--- a/drivers/gpu/drm/drm_irq.c
+++ b/drivers/gpu/drm/drm_irq.c
@@ -213,17 +213,17 @@ static void drm_update_vblank_count(struct drm_device *dev, unsigned int pipe,
 		diff = DIV_ROUND_CLOSEST_ULL(diff_ns, framedur_ns);
 
 		if (diff == 0 && flags & DRM_CALLED_FROM_VBLIRQ)
-			DRM_DEBUG("crtc %u: Redundant vblirq ignored."
-				  " diff_ns = %lld, framedur_ns = %d)\n",
-				  pipe, (long long) diff_ns, framedur_ns);
+			DRM_DEBUG_VBL("crtc %u: Redundant vblirq ignored."
+				      " diff_ns = %lld, framedur_ns = %d)\n",
+				      pipe, (long long) diff_ns, framedur_ns);
 	} else {
 		/* some kind of default for drivers w/o accurate vbl timestamping */
 		diff = (flags & DRM_CALLED_FROM_VBLIRQ) != 0;
 	}
 
-	DRM_DEBUG("updating vblank count on crtc %u:"
-		  " current=%u, diff=%u, hw=%u hw_last=%u\n",
-		  pipe, vblank->count, diff, cur_vblank, vblank->last);
+	DRM_DEBUG_VBL("updating vblank count on crtc %u:"
+		      " current=%u, diff=%u, hw=%u hw_last=%u\n",
+		      pipe, vblank->count, diff, cur_vblank, vblank->last);
 
 	if (diff == 0) {
 		WARN_ON_ONCE(cur_vblank != vblank->last);
@@ -800,11 +800,11 @@ int drm_calc_vbltimestamp_from_scanoutpos(struct drm_device *dev,
 		etime = ktime_sub_ns(etime, delta_ns);
 	*vblank_time = ktime_to_timeval(etime);
 
-	DRM_DEBUG("crtc %u : v 0x%x p(%d,%d)@ %ld.%ld -> %ld.%ld [e %d us, %d rep]\n",
-		  pipe, vbl_status, hpos, vpos,
-		  (long)tv_etime.tv_sec, (long)tv_etime.tv_usec,
-		  (long)vblank_time->tv_sec, (long)vblank_time->tv_usec,
-		  duration_ns/1000, i);
+	DRM_DEBUG_VBL("crtc %u : v 0x%x p(%d,%d)@ %ld.%ld -> %ld.%ld [e %d us, %d rep]\n",
+		      pipe, vbl_status, hpos, vpos,
+		      (long)tv_etime.tv_sec, (long)tv_etime.tv_usec,
+		      (long)vblank_time->tv_sec, (long)vblank_time->tv_usec,
+		      duration_ns/1000, i);
 
 	return ret;
 }
diff --git a/include/drm/drmP.h b/include/drm/drmP.h
index 3dc56d3..4d3b842 100644
--- a/include/drm/drmP.h
+++ b/include/drm/drmP.h
@@ -107,6 +107,9 @@ struct dma_buf_attachment;
  * ATOMIC: used in the atomic code.
  *	  This is the category used by the DRM_DEBUG_ATOMIC() macro.
  *
+ * VBL: used for verbose debug message in the vblank code
+ *	  This is the category used by the DRM_DEBUG_VBL() macro.
+ *
  * Enabling verbose debug messages is done through the drm.debug parameter,
  * each category being enabled by a bit.
  *
@@ -114,7 +117,7 @@ struct dma_buf_attachment;
  * drm.debug=0x2 will enable DRIVER messages
  * drm.debug=0x3 will enable CORE and DRIVER messages
  * ...
- * drm.debug=0xf will enable all messages
+ * drm.debug=0x3f will enable all messages
  *
  * An interesting feature is that it's possible to enable verbose logging at
  * run-time by echoing the debug value in its sysfs node:
@@ -125,6 +128,7 @@ struct dma_buf_attachment;
 #define DRM_UT_KMS		0x04
 #define DRM_UT_PRIME		0x08
 #define DRM_UT_ATOMIC		0x10
+#define DRM_UT_VBL		0x20
 
 extern __printf(2, 3)
 void drm_ut_debug_printk(const char *function_name,
@@ -217,6 +221,11 @@ void drm_err(const char *format, ...);
 		if (unlikely(drm_debug & DRM_UT_ATOMIC))		\
 			drm_ut_debug_printk(__func__, fmt, ##args);	\
 	} while (0)
+#define DRM_DEBUG_VBL(fmt, args...)					\
+	do {								\
+		if (unlikely(drm_debug & DRM_UT_VBL))			\
+			drm_ut_debug_printk(__func__, fmt, ##args);	\
+	} while (0)
 
 /*@}*/
 
-- 
cgit v0.10.2


From 203d027de4d7068c607b60d4310a1599dec8839f Mon Sep 17 00:00:00 2001
From: Lukas Wunner <lukas@wunner.de>
Date: Fri, 28 Aug 2015 11:56:26 +0200
Subject: vga_switcheroo: Use enum vga_switcheroo_state instead of int

Signed-off-by: Lukas Wunner <lukas@wunner.de>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/vga/vga_switcheroo.c b/drivers/gpu/vga/vga_switcheroo.c
index 1acbe20..a7870d2 100644
--- a/drivers/gpu/vga/vga_switcheroo.c
+++ b/drivers/gpu/vga/vga_switcheroo.c
@@ -100,7 +100,7 @@
 struct vga_switcheroo_client {
 	struct pci_dev *pdev;
 	struct fb_info *fb_info;
-	int pwr_state;
+	enum vga_switcheroo_state pwr_state;
 	const struct vga_switcheroo_client_ops *ops;
 	int id;
 	bool active;
@@ -344,7 +344,7 @@ find_active_client(struct list_head *head)
  *
  * Return: Power state.
  */
-int vga_switcheroo_get_client_state(struct pci_dev *pdev)
+enum vga_switcheroo_state vga_switcheroo_get_client_state(struct pci_dev *pdev)
 {
 	struct vga_switcheroo_client *client;
 	enum vga_switcheroo_state ret;
@@ -496,7 +496,7 @@ static int vga_switchoff(struct vga_switcheroo_client *client)
 	return 0;
 }
 
-static void set_audio_state(int id, int state)
+static void set_audio_state(int id, enum vga_switcheroo_state state)
 {
 	struct vga_switcheroo_client *client;
 
diff --git a/include/linux/vga_switcheroo.h b/include/linux/vga_switcheroo.h
index 3764991..e636617 100644
--- a/include/linux/vga_switcheroo.h
+++ b/include/linux/vga_switcheroo.h
@@ -138,7 +138,7 @@ void vga_switcheroo_unregister_handler(void);
 
 int vga_switcheroo_process_delayed_switch(void);
 
-int vga_switcheroo_get_client_state(struct pci_dev *dev);
+enum vga_switcheroo_state vga_switcheroo_get_client_state(struct pci_dev *dev);
 
 void vga_switcheroo_set_dynamic_switch(struct pci_dev *pdev, enum vga_switcheroo_state dynamic);
 
@@ -157,7 +157,7 @@ static inline int vga_switcheroo_register_audio_client(struct pci_dev *pdev,
 	int id) { return 0; }
 static inline void vga_switcheroo_unregister_handler(void) {}
 static inline int vga_switcheroo_process_delayed_switch(void) { return 0; }
-static inline int vga_switcheroo_get_client_state(struct pci_dev *dev) { return VGA_SWITCHEROO_ON; }
+static inline enum vga_switcheroo_state vga_switcheroo_get_client_state(struct pci_dev *dev) { return VGA_SWITCHEROO_ON; }
 
 static inline void vga_switcheroo_set_dynamic_switch(struct pci_dev *pdev, enum vga_switcheroo_state dynamic) {}
 
-- 
cgit v0.10.2


From 21c5ba8c1ee02f204e556c26703cebaf9c4019e0 Mon Sep 17 00:00:00 2001
From: Lukas Wunner <lukas@wunner.de>
Date: Fri, 28 Aug 2015 13:30:32 +0200
Subject: vga_switcheroo: Use VGA_SWITCHEROO_UNKNOWN_ID instead of -1

Signed-off-by: Lukas Wunner <lukas@wunner.de>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/vga/vga_switcheroo.c b/drivers/gpu/vga/vga_switcheroo.c
index a7870d2..9896305 100644
--- a/drivers/gpu/vga/vga_switcheroo.c
+++ b/drivers/gpu/vga/vga_switcheroo.c
@@ -84,9 +84,9 @@
  * @fb_info: framebuffer to which console is remapped on switching
  * @pwr_state: current power state
  * @ops: client callbacks
- * @id: client identifier, see enum vga_switcheroo_client_id.
- * 	Determining the id requires the handler, so GPUs are initially
- * 	assigned -1 and later given their true id in vga_switcheroo_enable()
+ * @id: client identifier. Determining the id requires the handler,
+ * 	so gpus are initially assigned VGA_SWITCHEROO_UNKNOWN_ID
+ * 	and later given their true id in vga_switcheroo_enable()
  * @active: whether the outputs are currently switched to this client
  * @driver_power_control: whether power state is controlled by the driver's
  * 	runtime pm. If true, writing ON and OFF to the vga_switcheroo debugfs
@@ -145,7 +145,8 @@ struct vgasr_priv {
 
 #define ID_BIT_AUDIO		0x100
 #define client_is_audio(c)	((c)->id & ID_BIT_AUDIO)
-#define client_is_vga(c)	((c)->id == -1 || !client_is_audio(c))
+#define client_is_vga(c)	((c)->id == VGA_SWITCHEROO_UNKNOWN_ID || \
+				 !client_is_audio(c))
 #define client_id(c)		((c)->id & ~ID_BIT_AUDIO)
 
 static int vga_switcheroo_debugfs_init(struct vgasr_priv *priv);
@@ -173,7 +174,7 @@ static void vga_switcheroo_enable(void)
 		vgasr_priv.handler->init();
 
 	list_for_each_entry(client, &vgasr_priv.clients, list) {
-		if (client->id != -1)
+		if (client->id != VGA_SWITCHEROO_UNKNOWN_ID)
 			continue;
 		ret = vgasr_priv.handler->get_client_id(client->pdev);
 		if (ret < 0)
@@ -277,7 +278,7 @@ int vga_switcheroo_register_client(struct pci_dev *pdev,
 				   const struct vga_switcheroo_client_ops *ops,
 				   bool driver_power_control)
 {
-	return register_client(pdev, ops, -1,
+	return register_client(pdev, ops, VGA_SWITCHEROO_UNKNOWN_ID,
 			       pdev == vga_default_device(),
 			       driver_power_control);
 }
@@ -583,7 +584,7 @@ vga_switcheroo_debugfs_write(struct file *filp, const char __user *ubuf,
 	int ret;
 	bool delay = false, can_switch;
 	bool just_mux = false;
-	int client_id = -1;
+	int client_id = VGA_SWITCHEROO_UNKNOWN_ID;
 	struct vga_switcheroo_client *client = NULL;
 
 	if (cnt > 63)
@@ -652,7 +653,7 @@ vga_switcheroo_debugfs_write(struct file *filp, const char __user *ubuf,
 		client_id = VGA_SWITCHEROO_DIS;
 	}
 
-	if (client_id == -1)
+	if (client_id == VGA_SWITCHEROO_UNKNOWN_ID)
 		goto out;
 	client = find_client_from_id(&vgasr_priv.clients, client_id);
 	if (!client)
diff --git a/include/linux/vga_switcheroo.h b/include/linux/vga_switcheroo.h
index e636617..88909a8 100644
--- a/include/linux/vga_switcheroo.h
+++ b/include/linux/vga_switcheroo.h
@@ -59,6 +59,9 @@ enum vga_switcheroo_state {
 
 /**
  * enum vga_switcheroo_client_id - client identifier
+ * @VGA_SWITCHEROO_UNKNOWN_ID: initial identifier assigned to vga clients.
+ * 	Determining the id requires the handler, so GPUs are given their
+ * 	true id in a delayed fashion in vga_switcheroo_enable()
  * @VGA_SWITCHEROO_IGD: integrated graphics device
  * @VGA_SWITCHEROO_DIS: discrete graphics device
  * @VGA_SWITCHEROO_MAX_CLIENTS: currently no more than two GPUs are supported
@@ -66,6 +69,7 @@ enum vga_switcheroo_state {
  * Client identifier. Audio clients use the same identifier & 0x100.
  */
 enum vga_switcheroo_client_id {
+	VGA_SWITCHEROO_UNKNOWN_ID = -1,
 	VGA_SWITCHEROO_IGD,
 	VGA_SWITCHEROO_DIS,
 	VGA_SWITCHEROO_MAX_CLIENTS,
-- 
cgit v0.10.2


From fa3e967fffaf267ccab7959429722da34e45ad77 Mon Sep 17 00:00:00 2001
From: Lukas Wunner <lukas@wunner.de>
Date: Fri, 28 Aug 2015 12:54:07 +0200
Subject: vga_switcheroo: Use enum vga_switcheroo_client_id instead of int

Signed-off-by: Lukas Wunner <lukas@wunner.de>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/vga/vga_switcheroo.c b/drivers/gpu/vga/vga_switcheroo.c
index 9896305..af0d372 100644
--- a/drivers/gpu/vga/vga_switcheroo.c
+++ b/drivers/gpu/vga/vga_switcheroo.c
@@ -102,7 +102,7 @@ struct vga_switcheroo_client {
 	struct fb_info *fb_info;
 	enum vga_switcheroo_state pwr_state;
 	const struct vga_switcheroo_client_ops *ops;
-	int id;
+	enum vga_switcheroo_client_id id;
 	bool active;
 	bool driver_power_control;
 	struct list_head list;
@@ -233,7 +233,8 @@ EXPORT_SYMBOL(vga_switcheroo_unregister_handler);
 
 static int register_client(struct pci_dev *pdev,
 			   const struct vga_switcheroo_client_ops *ops,
-			   int id, bool active, bool driver_power_control)
+			   enum vga_switcheroo_client_id id, bool active,
+			   bool driver_power_control)
 {
 	struct vga_switcheroo_client *client;
 
@@ -288,7 +289,7 @@ EXPORT_SYMBOL(vga_switcheroo_register_client);
  * vga_switcheroo_register_audio_client - register audio client
  * @pdev: client pci device
  * @ops: client callbacks
- * @id: client identifier, see enum vga_switcheroo_client_id
+ * @id: client identifier
  *
  * Register audio client (audio device on a GPU). The power state of the
  * client is assumed to be ON.
@@ -297,7 +298,7 @@ EXPORT_SYMBOL(vga_switcheroo_register_client);
  */
 int vga_switcheroo_register_audio_client(struct pci_dev *pdev,
 					 const struct vga_switcheroo_client_ops *ops,
-					 int id)
+					 enum vga_switcheroo_client_id id)
 {
 	return register_client(pdev, ops, id | ID_BIT_AUDIO, false, false);
 }
@@ -315,7 +316,8 @@ find_client_from_pci(struct list_head *head, struct pci_dev *pdev)
 }
 
 static struct vga_switcheroo_client *
-find_client_from_id(struct list_head *head, int client_id)
+find_client_from_id(struct list_head *head,
+		    enum vga_switcheroo_client_id client_id)
 {
 	struct vga_switcheroo_client *client;
 
@@ -497,7 +499,8 @@ static int vga_switchoff(struct vga_switcheroo_client *client)
 	return 0;
 }
 
-static void set_audio_state(int id, enum vga_switcheroo_state state)
+static void set_audio_state(enum vga_switcheroo_client_id id,
+			    enum vga_switcheroo_state state)
 {
 	struct vga_switcheroo_client *client;
 
@@ -584,7 +587,7 @@ vga_switcheroo_debugfs_write(struct file *filp, const char __user *ubuf,
 	int ret;
 	bool delay = false, can_switch;
 	bool just_mux = false;
-	int client_id = VGA_SWITCHEROO_UNKNOWN_ID;
+	enum vga_switcheroo_client_id client_id = VGA_SWITCHEROO_UNKNOWN_ID;
 	struct vga_switcheroo_client *client = NULL;
 
 	if (cnt > 63)
diff --git a/include/linux/vga_switcheroo.h b/include/linux/vga_switcheroo.h
index 88909a8..c557511 100644
--- a/include/linux/vga_switcheroo.h
+++ b/include/linux/vga_switcheroo.h
@@ -100,7 +100,7 @@ struct vga_switcheroo_handler {
 	int (*switchto)(enum vga_switcheroo_client_id id);
 	int (*power_state)(enum vga_switcheroo_client_id id,
 			   enum vga_switcheroo_state state);
-	int (*get_client_id)(struct pci_dev *pdev);
+	enum vga_switcheroo_client_id (*get_client_id)(struct pci_dev *pdev);
 };
 
 /**
@@ -132,7 +132,7 @@ int vga_switcheroo_register_client(struct pci_dev *dev,
 				   bool driver_power_control);
 int vga_switcheroo_register_audio_client(struct pci_dev *pdev,
 					 const struct vga_switcheroo_client_ops *ops,
-					 int id);
+					 enum vga_switcheroo_client_id id);
 
 void vga_switcheroo_client_fb_set(struct pci_dev *dev,
 				  struct fb_info *info);
@@ -158,7 +158,7 @@ static inline void vga_switcheroo_client_fb_set(struct pci_dev *dev, struct fb_i
 static inline int vga_switcheroo_register_handler(struct vga_switcheroo_handler *handler) { return 0; }
 static inline int vga_switcheroo_register_audio_client(struct pci_dev *pdev,
 	const struct vga_switcheroo_client_ops *ops,
-	int id) { return 0; }
+	enum vga_switcheroo_client_id id) { return 0; }
 static inline void vga_switcheroo_unregister_handler(void) {}
 static inline int vga_switcheroo_process_delayed_switch(void) { return 0; }
 static inline enum vga_switcheroo_state vga_switcheroo_get_client_state(struct pci_dev *dev) { return VGA_SWITCHEROO_ON; }
-- 
cgit v0.10.2


From 02d0a493f3627df8c008417ba11e5f93f049f14c Mon Sep 17 00:00:00 2001
From: Rob Clark <robdclark@gmail.com>
Date: Tue, 13 Oct 2015 10:13:28 -0400
Subject: drm: misc cleanup

Drop unused drm_atomic and fix comment for drm_debug.

Signed-off-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c
index d01f8d6..9362609 100644
--- a/drivers/gpu/drm/drm_drv.c
+++ b/drivers/gpu/drm/drm_drv.c
@@ -37,11 +37,9 @@
 #include "drm_legacy.h"
 #include "drm_internal.h"
 
-unsigned int drm_debug = 0;	/* 1 to enable debug output */
+unsigned int drm_debug = 0;	/* bitmask of DRM_UT_x */
 EXPORT_SYMBOL(drm_debug);
 
-bool drm_atomic = 0;
-
 MODULE_AUTHOR(CORE_AUTHOR);
 MODULE_DESCRIPTION(CORE_DESC);
 MODULE_LICENSE("GPL and additional rights");
-- 
cgit v0.10.2


From 3a4579b41ccd1e786cadb720b55e06ca6c9d2594 Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Wed, 7 Oct 2015 09:55:28 +0200
Subject: drm/doc: Rename docbook to gpu.tmpl

DRM is a lot more than a direct rendering manager nowadays, and there's
also a bunch of things worth documenting for gpu driver developers
outside of drivers/gpu/drm, like vgaarb, vga_switcheroo or the various
hardware buses like host1x and ipu-v3.

To avoid further confusion let's rename the top-level to reflect
reality.

And yes I'm already looking forward to when we need to replace the G
in GPU with a * ;-)

Inspired by a thread with Lukas since he refused to include the
vga_switcheroo docs into the drm docs because it's not drm.

Cc: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
[Lukas: Drop BUG() easter egg in i915_gem_execbuffer.c spotted by Jani
and fix typos in commit message.]
Signed-off-by: Lukas Wunner <lukas@wunner.de>
Acked-by: Dave Airlie <airlied@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/Documentation/DocBook/Makefile b/Documentation/DocBook/Makefile
index 93eff64..810f466 100644
--- a/Documentation/DocBook/Makefile
+++ b/Documentation/DocBook/Makefile
@@ -14,7 +14,7 @@ DOCBOOKS := z8530book.xml device-drivers.xml \
 	    genericirq.xml s390-drivers.xml uio-howto.xml scsi.xml \
 	    80211.xml debugobjects.xml sh.xml regulator.xml \
 	    alsa-driver-api.xml writing-an-alsa-driver.xml \
-	    tracepoint.xml drm.xml media_api.xml w1.xml \
+	    tracepoint.xml gpu.xml media_api.xml w1.xml \
 	    writing_musb_glue_layer.xml crypto-API.xml iio.xml
 
 include Documentation/DocBook/media/Makefile
diff --git a/Documentation/DocBook/drm.tmpl b/Documentation/DocBook/drm.tmpl
deleted file mode 100644
index 2bc6c80..0000000
--- a/Documentation/DocBook/drm.tmpl
+++ /dev/null
@@ -1,4203 +0,0 @@
-<?xml version="1.0" encoding="UTF-8"?>
-<!DOCTYPE book PUBLIC "-//OASIS//DTD DocBook XML V4.1.2//EN"
-	"http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd" []>
-
-<book id="drmDevelopersGuide">
-  <bookinfo>
-    <title>Linux DRM Developer's Guide</title>
-
-    <authorgroup>
-      <author>
-	<firstname>Jesse</firstname>
-	<surname>Barnes</surname>
-	<contrib>Initial version</contrib>
-	<affiliation>
-	  <orgname>Intel Corporation</orgname>
-	  <address>
-	    <email>jesse.barnes@intel.com</email>
-	  </address>
-	</affiliation>
-      </author>
-      <author>
-	<firstname>Laurent</firstname>
-	<surname>Pinchart</surname>
-	<contrib>Driver internals</contrib>
-	<affiliation>
-	  <orgname>Ideas on board SPRL</orgname>
-	  <address>
-	    <email>laurent.pinchart@ideasonboard.com</email>
-	  </address>
-	</affiliation>
-      </author>
-      <author>
-	<firstname>Daniel</firstname>
-	<surname>Vetter</surname>
-	<contrib>Contributions all over the place</contrib>
-	<affiliation>
-	  <orgname>Intel Corporation</orgname>
-	  <address>
-	    <email>daniel.vetter@ffwll.ch</email>
-	  </address>
-	</affiliation>
-      </author>
-    </authorgroup>
-
-    <copyright>
-      <year>2008-2009</year>
-      <year>2013-2014</year>
-      <holder>Intel Corporation</holder>
-    </copyright>
-    <copyright>
-      <year>2012</year>
-      <holder>Laurent Pinchart</holder>
-    </copyright>
-
-    <legalnotice>
-      <para>
-	The contents of this file may be used under the terms of the GNU
-	General Public License version 2 (the "GPL") as distributed in
-	the kernel source COPYING file.
-      </para>
-    </legalnotice>
-
-    <revhistory>
-      <!-- Put document revisions here, newest first. -->
-      <revision>
-	<revnumber>1.0</revnumber>
-	<date>2012-07-13</date>
-	<authorinitials>LP</authorinitials>
-	<revremark>Added extensive documentation about driver internals.
-	</revremark>
-      </revision>
-    </revhistory>
-  </bookinfo>
-
-<toc></toc>
-
-<part id="drmCore">
-  <title>DRM Core</title>
-  <partintro>
-    <para>
-      This first part of the DRM Developer's Guide documents core DRM code,
-      helper libraries for writing drivers and generic userspace interfaces
-      exposed by DRM drivers.
-    </para>
-  </partintro>
-
-  <chapter id="drmIntroduction">
-    <title>Introduction</title>
-    <para>
-      The Linux DRM layer contains code intended to support the needs
-      of complex graphics devices, usually containing programmable
-      pipelines well suited to 3D graphics acceleration.  Graphics
-      drivers in the kernel may make use of DRM functions to make
-      tasks like memory management, interrupt handling and DMA easier,
-      and provide a uniform interface to applications.
-    </para>
-    <para>
-      A note on versions: this guide covers features found in the DRM
-      tree, including the TTM memory manager, output configuration and
-      mode setting, and the new vblank internals, in addition to all
-      the regular features found in current kernels.
-    </para>
-    <para>
-      [Insert diagram of typical DRM stack here]
-    </para>
-  </chapter>
-
-  <!-- Internals -->
-
-  <chapter id="drmInternals">
-    <title>DRM Internals</title>
-    <para>
-      This chapter documents DRM internals relevant to driver authors
-      and developers working to add support for the latest features to
-      existing drivers.
-    </para>
-    <para>
-      First, we go over some typical driver initialization
-      requirements, like setting up command buffers, creating an
-      initial output configuration, and initializing core services.
-      Subsequent sections cover core internals in more detail,
-      providing implementation notes and examples.
-    </para>
-    <para>
-      The DRM layer provides several services to graphics drivers,
-      many of them driven by the application interfaces it provides
-      through libdrm, the library that wraps most of the DRM ioctls.
-      These include vblank event handling, memory
-      management, output management, framebuffer management, command
-      submission &amp; fencing, suspend/resume support, and DMA
-      services.
-    </para>
-
-  <!-- Internals: driver init -->
-
-  <sect1>
-    <title>Driver Initialization</title>
-    <para>
-      At the core of every DRM driver is a <structname>drm_driver</structname>
-      structure. Drivers typically statically initialize a drm_driver structure,
-      and then pass it to <function>drm_dev_alloc()</function> to allocate a
-      device instance. After the device instance is fully initialized it can be
-      registered (which makes it accessible from userspace) using
-      <function>drm_dev_register()</function>.
-    </para>
-    <para>
-      The <structname>drm_driver</structname> structure contains static
-      information that describes the driver and features it supports, and
-      pointers to methods that the DRM core will call to implement the DRM API.
-      We will first go through the <structname>drm_driver</structname> static
-      information fields, and will then describe individual operations in
-      details as they get used in later sections.
-    </para>
-    <sect2>
-      <title>Driver Information</title>
-      <sect3>
-        <title>Driver Features</title>
-        <para>
-          Drivers inform the DRM core about their requirements and supported
-          features by setting appropriate flags in the
-          <structfield>driver_features</structfield> field. Since those flags
-          influence the DRM core behaviour since registration time, most of them
-          must be set to registering the <structname>drm_driver</structname>
-          instance.
-        </para>
-        <synopsis>u32 driver_features;</synopsis>
-        <variablelist>
-          <title>Driver Feature Flags</title>
-          <varlistentry>
-            <term>DRIVER_USE_AGP</term>
-            <listitem><para>
-              Driver uses AGP interface, the DRM core will manage AGP resources.
-            </para></listitem>
-          </varlistentry>
-          <varlistentry>
-            <term>DRIVER_REQUIRE_AGP</term>
-            <listitem><para>
-              Driver needs AGP interface to function. AGP initialization failure
-              will become a fatal error.
-            </para></listitem>
-          </varlistentry>
-          <varlistentry>
-            <term>DRIVER_PCI_DMA</term>
-            <listitem><para>
-              Driver is capable of PCI DMA, mapping of PCI DMA buffers to
-              userspace will be enabled. Deprecated.
-            </para></listitem>
-          </varlistentry>
-          <varlistentry>
-            <term>DRIVER_SG</term>
-            <listitem><para>
-              Driver can perform scatter/gather DMA, allocation and mapping of
-              scatter/gather buffers will be enabled. Deprecated.
-            </para></listitem>
-          </varlistentry>
-          <varlistentry>
-            <term>DRIVER_HAVE_DMA</term>
-            <listitem><para>
-              Driver supports DMA, the userspace DMA API will be supported.
-              Deprecated.
-            </para></listitem>
-          </varlistentry>
-          <varlistentry>
-            <term>DRIVER_HAVE_IRQ</term><term>DRIVER_IRQ_SHARED</term>
-            <listitem><para>
-              DRIVER_HAVE_IRQ indicates whether the driver has an IRQ handler
-              managed by the DRM Core. The core will support simple IRQ handler
-              installation when the flag is set. The installation process is
-              described in <xref linkend="drm-irq-registration"/>.</para>
-              <para>DRIVER_IRQ_SHARED indicates whether the device &amp; handler
-              support shared IRQs (note that this is required of PCI  drivers).
-            </para></listitem>
-          </varlistentry>
-          <varlistentry>
-            <term>DRIVER_GEM</term>
-            <listitem><para>
-              Driver use the GEM memory manager.
-            </para></listitem>
-          </varlistentry>
-          <varlistentry>
-            <term>DRIVER_MODESET</term>
-            <listitem><para>
-              Driver supports mode setting interfaces (KMS).
-            </para></listitem>
-          </varlistentry>
-          <varlistentry>
-            <term>DRIVER_PRIME</term>
-            <listitem><para>
-              Driver implements DRM PRIME buffer sharing.
-            </para></listitem>
-          </varlistentry>
-          <varlistentry>
-            <term>DRIVER_RENDER</term>
-            <listitem><para>
-              Driver supports dedicated render nodes.
-            </para></listitem>
-          </varlistentry>
-          <varlistentry>
-            <term>DRIVER_ATOMIC</term>
-            <listitem><para>
-              Driver supports atomic properties.  In this case the driver
-              must implement appropriate obj->atomic_get_property() vfuncs
-              for any modeset objects with driver specific properties.
-            </para></listitem>
-          </varlistentry>
-        </variablelist>
-      </sect3>
-      <sect3>
-        <title>Major, Minor and Patchlevel</title>
-        <synopsis>int major;
-int minor;
-int patchlevel;</synopsis>
-        <para>
-          The DRM core identifies driver versions by a major, minor and patch
-          level triplet. The information is printed to the kernel log at
-          initialization time and passed to userspace through the
-          DRM_IOCTL_VERSION ioctl.
-        </para>
-        <para>
-          The major and minor numbers are also used to verify the requested driver
-          API version passed to DRM_IOCTL_SET_VERSION. When the driver API changes
-          between minor versions, applications can call DRM_IOCTL_SET_VERSION to
-          select a specific version of the API. If the requested major isn't equal
-          to the driver major, or the requested minor is larger than the driver
-          minor, the DRM_IOCTL_SET_VERSION call will return an error. Otherwise
-          the driver's set_version() method will be called with the requested
-          version.
-        </para>
-      </sect3>
-      <sect3>
-        <title>Name, Description and Date</title>
-        <synopsis>char *name;
-char *desc;
-char *date;</synopsis>
-        <para>
-          The driver name is printed to the kernel log at initialization time,
-          used for IRQ registration and passed to userspace through
-          DRM_IOCTL_VERSION.
-        </para>
-        <para>
-          The driver description is a purely informative string passed to
-          userspace through the DRM_IOCTL_VERSION ioctl and otherwise unused by
-          the kernel.
-        </para>
-        <para>
-          The driver date, formatted as YYYYMMDD, is meant to identify the date of
-          the latest modification to the driver. However, as most drivers fail to
-          update it, its value is mostly useless. The DRM core prints it to the
-          kernel log at initialization time and passes it to userspace through the
-          DRM_IOCTL_VERSION ioctl.
-        </para>
-      </sect3>
-    </sect2>
-    <sect2>
-      <title>Device Instance and Driver Handling</title>
-!Pdrivers/gpu/drm/drm_drv.c driver instance overview
-!Edrivers/gpu/drm/drm_drv.c
-    </sect2>
-    <sect2>
-      <title>Driver Load</title>
-      <sect3 id="drm-irq-registration">
-        <title>IRQ Registration</title>
-        <para>
-          The DRM core tries to facilitate IRQ handler registration and
-          unregistration by providing <function>drm_irq_install</function> and
-          <function>drm_irq_uninstall</function> functions. Those functions only
-          support a single interrupt per device, devices that use more than one
-          IRQs need to be handled manually.
-        </para>
-        <sect4>
-          <title>Managed IRQ Registration</title>
-          <para>
-            <function>drm_irq_install</function> starts by calling the
-            <methodname>irq_preinstall</methodname> driver operation. The operation
-            is optional and must make sure that the interrupt will not get fired by
-            clearing all pending interrupt flags or disabling the interrupt.
-          </para>
-          <para>
-            The passed-in IRQ will then be requested by a call to
-            <function>request_irq</function>. If the DRIVER_IRQ_SHARED driver
-            feature flag is set, a shared (IRQF_SHARED) IRQ handler will be
-            requested.
-          </para>
-          <para>
-            The IRQ handler function must be provided as the mandatory irq_handler
-            driver operation. It will get passed directly to
-            <function>request_irq</function> and thus has the same prototype as all
-            IRQ handlers. It will get called with a pointer to the DRM device as the
-            second argument.
-          </para>
-          <para>
-            Finally the function calls the optional
-            <methodname>irq_postinstall</methodname> driver operation. The operation
-            usually enables interrupts (excluding the vblank interrupt, which is
-            enabled separately), but drivers may choose to enable/disable interrupts
-            at a different time.
-          </para>
-          <para>
-            <function>drm_irq_uninstall</function> is similarly used to uninstall an
-            IRQ handler. It starts by waking up all processes waiting on a vblank
-            interrupt to make sure they don't hang, and then calls the optional
-            <methodname>irq_uninstall</methodname> driver operation. The operation
-            must disable all hardware interrupts. Finally the function frees the IRQ
-            by calling <function>free_irq</function>.
-          </para>
-        </sect4>
-        <sect4>
-          <title>Manual IRQ Registration</title>
-          <para>
-            Drivers that require multiple interrupt handlers can't use the managed
-            IRQ registration functions. In that case IRQs must be registered and
-            unregistered manually (usually with the <function>request_irq</function>
-            and <function>free_irq</function> functions, or their devm_* equivalent).
-          </para>
-          <para>
-            When manually registering IRQs, drivers must not set the DRIVER_HAVE_IRQ
-            driver feature flag, and must not provide the
-	    <methodname>irq_handler</methodname> driver operation. They must set the
-	    <structname>drm_device</structname> <structfield>irq_enabled</structfield>
-	    field to 1 upon registration of the IRQs, and clear it to 0 after
-	    unregistering the IRQs.
-          </para>
-        </sect4>
-      </sect3>
-      <sect3>
-        <title>Memory Manager Initialization</title>
-        <para>
-          Every DRM driver requires a memory manager which must be initialized at
-          load time. DRM currently contains two memory managers, the Translation
-          Table Manager (TTM) and the Graphics Execution Manager (GEM).
-          This document describes the use of the GEM memory manager only. See
-          <xref linkend="drm-memory-management"/> for details.
-        </para>
-      </sect3>
-      <sect3>
-        <title>Miscellaneous Device Configuration</title>
-        <para>
-          Another task that may be necessary for PCI devices during configuration
-          is mapping the video BIOS. On many devices, the VBIOS describes device
-          configuration, LCD panel timings (if any), and contains flags indicating
-          device state. Mapping the BIOS can be done using the pci_map_rom() call,
-          a convenience function that takes care of mapping the actual ROM,
-          whether it has been shadowed into memory (typically at address 0xc0000)
-          or exists on the PCI device in the ROM BAR. Note that after the ROM has
-          been mapped and any necessary information has been extracted, it should
-          be unmapped; on many devices, the ROM address decoder is shared with
-          other BARs, so leaving it mapped could cause undesired behaviour like
-          hangs or memory corruption.
-  <!--!Fdrivers/pci/rom.c pci_map_rom-->
-        </para>
-      </sect3>
-    </sect2>
-    <sect2>
-      <title>Bus-specific Device Registration and PCI Support</title>
-      <para>
-        A number of functions are provided to help with device registration.
-	The functions deal with PCI and platform devices respectively and are
-	only provided for historical reasons. These are all deprecated and
-	shouldn't be used in new drivers. Besides that there's a few
-	helpers for pci drivers.
-      </para>
-!Edrivers/gpu/drm/drm_pci.c
-!Edrivers/gpu/drm/drm_platform.c
-    </sect2>
-  </sect1>
-
-  <!-- Internals: memory management -->
-
-  <sect1 id="drm-memory-management">
-    <title>Memory management</title>
-    <para>
-      Modern Linux systems require large amount of graphics memory to store
-      frame buffers, textures, vertices and other graphics-related data. Given
-      the very dynamic nature of many of that data, managing graphics memory
-      efficiently is thus crucial for the graphics stack and plays a central
-      role in the DRM infrastructure.
-    </para>
-    <para>
-      The DRM core includes two memory managers, namely Translation Table Maps
-      (TTM) and Graphics Execution Manager (GEM). TTM was the first DRM memory
-      manager to be developed and tried to be a one-size-fits-them all
-      solution. It provides a single userspace API to accommodate the need of
-      all hardware, supporting both Unified Memory Architecture (UMA) devices
-      and devices with dedicated video RAM (i.e. most discrete video cards).
-      This resulted in a large, complex piece of code that turned out to be
-      hard to use for driver development.
-    </para>
-    <para>
-      GEM started as an Intel-sponsored project in reaction to TTM's
-      complexity. Its design philosophy is completely different: instead of
-      providing a solution to every graphics memory-related problems, GEM
-      identified common code between drivers and created a support library to
-      share it. GEM has simpler initialization and execution requirements than
-      TTM, but has no video RAM management capabilities and is thus limited to
-      UMA devices.
-    </para>
-    <sect2>
-      <title>The Translation Table Manager (TTM)</title>
-      <para>
-        TTM design background and information belongs here.
-      </para>
-      <sect3>
-        <title>TTM initialization</title>
-        <warning><para>This section is outdated.</para></warning>
-        <para>
-          Drivers wishing to support TTM must fill out a drm_bo_driver
-          structure. The structure contains several fields with function
-          pointers for initializing the TTM, allocating and freeing memory,
-          waiting for command completion and fence synchronization, and memory
-          migration. See the radeon_ttm.c file for an example of usage.
-        </para>
-        <para>
-          The ttm_global_reference structure is made up of several fields:
-        </para>
-        <programlisting>
-          struct ttm_global_reference {
-                  enum ttm_global_types global_type;
-                  size_t size;
-                  void *object;
-                  int (*init) (struct ttm_global_reference *);
-                  void (*release) (struct ttm_global_reference *);
-          };
-        </programlisting>
-        <para>
-          There should be one global reference structure for your memory
-          manager as a whole, and there will be others for each object
-          created by the memory manager at runtime.  Your global TTM should
-          have a type of TTM_GLOBAL_TTM_MEM.  The size field for the global
-          object should be sizeof(struct ttm_mem_global), and the init and
-          release hooks should point at your driver-specific init and
-          release routines, which probably eventually call
-          ttm_mem_global_init and ttm_mem_global_release, respectively.
-        </para>
-        <para>
-          Once your global TTM accounting structure is set up and initialized
-          by calling ttm_global_item_ref() on it,
-          you need to create a buffer object TTM to
-          provide a pool for buffer object allocation by clients and the
-          kernel itself.  The type of this object should be TTM_GLOBAL_TTM_BO,
-          and its size should be sizeof(struct ttm_bo_global).  Again,
-          driver-specific init and release functions may be provided,
-          likely eventually calling ttm_bo_global_init() and
-          ttm_bo_global_release(), respectively.  Also, like the previous
-          object, ttm_global_item_ref() is used to create an initial reference
-          count for the TTM, which will call your initialization function.
-        </para>
-      </sect3>
-    </sect2>
-    <sect2 id="drm-gem">
-      <title>The Graphics Execution Manager (GEM)</title>
-      <para>
-        The GEM design approach has resulted in a memory manager that doesn't
-        provide full coverage of all (or even all common) use cases in its
-        userspace or kernel API. GEM exposes a set of standard memory-related
-        operations to userspace and a set of helper functions to drivers, and let
-        drivers implement hardware-specific operations with their own private API.
-      </para>
-      <para>
-        The GEM userspace API is described in the
-        <ulink url="http://lwn.net/Articles/283798/"><citetitle>GEM - the Graphics
-        Execution Manager</citetitle></ulink> article on LWN. While slightly
-        outdated, the document provides a good overview of the GEM API principles.
-        Buffer allocation and read and write operations, described as part of the
-        common GEM API, are currently implemented using driver-specific ioctls.
-      </para>
-      <para>
-        GEM is data-agnostic. It manages abstract buffer objects without knowing
-        what individual buffers contain. APIs that require knowledge of buffer
-        contents or purpose, such as buffer allocation or synchronization
-        primitives, are thus outside of the scope of GEM and must be implemented
-        using driver-specific ioctls.
-      </para>
-      <para>
-        On a fundamental level, GEM involves several operations:
-        <itemizedlist>
-          <listitem>Memory allocation and freeing</listitem>
-          <listitem>Command execution</listitem>
-          <listitem>Aperture management at command execution time</listitem>
-        </itemizedlist>
-        Buffer object allocation is relatively straightforward and largely
-        provided by Linux's shmem layer, which provides memory to back each
-        object.
-      </para>
-      <para>
-        Device-specific operations, such as command execution, pinning, buffer
-        read &amp; write, mapping, and domain ownership transfers are left to
-        driver-specific ioctls.
-      </para>
-      <sect3>
-        <title>GEM Initialization</title>
-        <para>
-          Drivers that use GEM must set the DRIVER_GEM bit in the struct
-          <structname>drm_driver</structname>
-          <structfield>driver_features</structfield> field. The DRM core will
-          then automatically initialize the GEM core before calling the
-          <methodname>load</methodname> operation. Behind the scene, this will
-          create a DRM Memory Manager object which provides an address space
-          pool for object allocation.
-        </para>
-        <para>
-          In a KMS configuration, drivers need to allocate and initialize a
-          command ring buffer following core GEM initialization if required by
-          the hardware. UMA devices usually have what is called a "stolen"
-          memory region, which provides space for the initial framebuffer and
-          large, contiguous memory regions required by the device. This space is
-          typically not managed by GEM, and must be initialized separately into
-          its own DRM MM object.
-        </para>
-      </sect3>
-      <sect3>
-        <title>GEM Objects Creation</title>
-        <para>
-          GEM splits creation of GEM objects and allocation of the memory that
-          backs them in two distinct operations.
-        </para>
-        <para>
-          GEM objects are represented by an instance of struct
-          <structname>drm_gem_object</structname>. Drivers usually need to extend
-          GEM objects with private information and thus create a driver-specific
-          GEM object structure type that embeds an instance of struct
-          <structname>drm_gem_object</structname>.
-        </para>
-        <para>
-          To create a GEM object, a driver allocates memory for an instance of its
-          specific GEM object type and initializes the embedded struct
-          <structname>drm_gem_object</structname> with a call to
-          <function>drm_gem_object_init</function>. The function takes a pointer to
-          the DRM device, a pointer to the GEM object and the buffer object size
-          in bytes.
-        </para>
-        <para>
-          GEM uses shmem to allocate anonymous pageable memory.
-          <function>drm_gem_object_init</function> will create an shmfs file of
-          the requested size and store it into the struct
-          <structname>drm_gem_object</structname> <structfield>filp</structfield>
-          field. The memory is used as either main storage for the object when the
-          graphics hardware uses system memory directly or as a backing store
-          otherwise.
-        </para>
-        <para>
-          Drivers are responsible for the actual physical pages allocation by
-          calling <function>shmem_read_mapping_page_gfp</function> for each page.
-          Note that they can decide to allocate pages when initializing the GEM
-          object, or to delay allocation until the memory is needed (for instance
-          when a page fault occurs as a result of a userspace memory access or
-          when the driver needs to start a DMA transfer involving the memory).
-        </para>
-        <para>
-          Anonymous pageable memory allocation is not always desired, for instance
-          when the hardware requires physically contiguous system memory as is
-          often the case in embedded devices. Drivers can create GEM objects with
-          no shmfs backing (called private GEM objects) by initializing them with
-          a call to <function>drm_gem_private_object_init</function> instead of
-          <function>drm_gem_object_init</function>. Storage for private GEM
-          objects must be managed by drivers.
-        </para>
-        <para>
-          Drivers that do not need to extend GEM objects with private information
-          can call the <function>drm_gem_object_alloc</function> function to
-          allocate and initialize a struct <structname>drm_gem_object</structname>
-          instance. The GEM core will call the optional driver
-          <methodname>gem_init_object</methodname> operation after initializing
-          the GEM object with <function>drm_gem_object_init</function>.
-          <synopsis>int (*gem_init_object) (struct drm_gem_object *obj);</synopsis>
-        </para>
-        <para>
-          No alloc-and-init function exists for private GEM objects.
-        </para>
-      </sect3>
-      <sect3>
-        <title>GEM Objects Lifetime</title>
-        <para>
-          All GEM objects are reference-counted by the GEM core. References can be
-          acquired and release by <function>calling drm_gem_object_reference</function>
-          and <function>drm_gem_object_unreference</function> respectively. The
-          caller must hold the <structname>drm_device</structname>
-          <structfield>struct_mutex</structfield> lock. As a convenience, GEM
-          provides the <function>drm_gem_object_reference_unlocked</function> and
-          <function>drm_gem_object_unreference_unlocked</function> functions that
-          can be called without holding the lock.
-        </para>
-        <para>
-          When the last reference to a GEM object is released the GEM core calls
-          the <structname>drm_driver</structname>
-          <methodname>gem_free_object</methodname> operation. That operation is
-          mandatory for GEM-enabled drivers and must free the GEM object and all
-          associated resources.
-        </para>
-        <para>
-          <synopsis>void (*gem_free_object) (struct drm_gem_object *obj);</synopsis>
-          Drivers are responsible for freeing all GEM object resources, including
-          the resources created by the GEM core. If an mmap offset has been
-          created for the object (in which case
-          <structname>drm_gem_object</structname>::<structfield>map_list</structfield>::<structfield>map</structfield>
-          is not NULL) it must be freed by a call to
-          <function>drm_gem_free_mmap_offset</function>. The shmfs backing store
-          must be released by calling <function>drm_gem_object_release</function>
-          (that function can safely be called if no shmfs backing store has been
-          created).
-        </para>
-      </sect3>
-      <sect3>
-        <title>GEM Objects Naming</title>
-        <para>
-          Communication between userspace and the kernel refers to GEM objects
-          using local handles, global names or, more recently, file descriptors.
-          All of those are 32-bit integer values; the usual Linux kernel limits
-          apply to the file descriptors.
-        </para>
-        <para>
-          GEM handles are local to a DRM file. Applications get a handle to a GEM
-          object through a driver-specific ioctl, and can use that handle to refer
-          to the GEM object in other standard or driver-specific ioctls. Closing a
-          DRM file handle frees all its GEM handles and dereferences the
-          associated GEM objects.
-        </para>
-        <para>
-          To create a handle for a GEM object drivers call
-          <function>drm_gem_handle_create</function>. The function takes a pointer
-          to the DRM file and the GEM object and returns a locally unique handle.
-          When the handle is no longer needed drivers delete it with a call to
-          <function>drm_gem_handle_delete</function>. Finally the GEM object
-          associated with a handle can be retrieved by a call to
-          <function>drm_gem_object_lookup</function>.
-        </para>
-        <para>
-          Handles don't take ownership of GEM objects, they only take a reference
-          to the object that will be dropped when the handle is destroyed. To
-          avoid leaking GEM objects, drivers must make sure they drop the
-          reference(s) they own (such as the initial reference taken at object
-          creation time) as appropriate, without any special consideration for the
-          handle. For example, in the particular case of combined GEM object and
-          handle creation in the implementation of the
-          <methodname>dumb_create</methodname> operation, drivers must drop the
-          initial reference to the GEM object before returning the handle.
-        </para>
-        <para>
-          GEM names are similar in purpose to handles but are not local to DRM
-          files. They can be passed between processes to reference a GEM object
-          globally. Names can't be used directly to refer to objects in the DRM
-          API, applications must convert handles to names and names to handles
-          using the DRM_IOCTL_GEM_FLINK and DRM_IOCTL_GEM_OPEN ioctls
-          respectively. The conversion is handled by the DRM core without any
-          driver-specific support.
-        </para>
-        <para>
-          GEM also supports buffer sharing with dma-buf file descriptors through
-          PRIME. GEM-based drivers must use the provided helpers functions to
-          implement the exporting and importing correctly. See <xref linkend="drm-prime-support" />.
-          Since sharing file descriptors is inherently more secure than the
-          easily guessable and global GEM names it is the preferred buffer
-          sharing mechanism. Sharing buffers through GEM names is only supported
-          for legacy userspace. Furthermore PRIME also allows cross-device
-          buffer sharing since it is based on dma-bufs.
-        </para>
-      </sect3>
-      <sect3 id="drm-gem-objects-mapping">
-        <title>GEM Objects Mapping</title>
-        <para>
-          Because mapping operations are fairly heavyweight GEM favours
-          read/write-like access to buffers, implemented through driver-specific
-          ioctls, over mapping buffers to userspace. However, when random access
-          to the buffer is needed (to perform software rendering for instance),
-          direct access to the object can be more efficient.
-        </para>
-        <para>
-          The mmap system call can't be used directly to map GEM objects, as they
-          don't have their own file handle. Two alternative methods currently
-          co-exist to map GEM objects to userspace. The first method uses a
-          driver-specific ioctl to perform the mapping operation, calling
-          <function>do_mmap</function> under the hood. This is often considered
-          dubious, seems to be discouraged for new GEM-enabled drivers, and will
-          thus not be described here.
-        </para>
-        <para>
-          The second method uses the mmap system call on the DRM file handle.
-          <synopsis>void *mmap(void *addr, size_t length, int prot, int flags, int fd,
-             off_t offset);</synopsis>
-          DRM identifies the GEM object to be mapped by a fake offset passed
-          through the mmap offset argument. Prior to being mapped, a GEM object
-          must thus be associated with a fake offset. To do so, drivers must call
-          <function>drm_gem_create_mmap_offset</function> on the object. The
-          function allocates a fake offset range from a pool and stores the
-          offset divided by PAGE_SIZE in
-          <literal>obj-&gt;map_list.hash.key</literal>. Care must be taken not to
-          call <function>drm_gem_create_mmap_offset</function> if a fake offset
-          has already been allocated for the object. This can be tested by
-          <literal>obj-&gt;map_list.map</literal> being non-NULL.
-        </para>
-        <para>
-          Once allocated, the fake offset value
-          (<literal>obj-&gt;map_list.hash.key &lt;&lt; PAGE_SHIFT</literal>)
-          must be passed to the application in a driver-specific way and can then
-          be used as the mmap offset argument.
-        </para>
-        <para>
-          The GEM core provides a helper method <function>drm_gem_mmap</function>
-          to handle object mapping. The method can be set directly as the mmap
-          file operation handler. It will look up the GEM object based on the
-          offset value and set the VMA operations to the
-          <structname>drm_driver</structname> <structfield>gem_vm_ops</structfield>
-          field. Note that <function>drm_gem_mmap</function> doesn't map memory to
-          userspace, but relies on the driver-provided fault handler to map pages
-          individually.
-        </para>
-        <para>
-          To use <function>drm_gem_mmap</function>, drivers must fill the struct
-          <structname>drm_driver</structname> <structfield>gem_vm_ops</structfield>
-          field with a pointer to VM operations.
-        </para>
-        <para>
-          <synopsis>struct vm_operations_struct *gem_vm_ops
-
-  struct vm_operations_struct {
-          void (*open)(struct vm_area_struct * area);
-          void (*close)(struct vm_area_struct * area);
-          int (*fault)(struct vm_area_struct *vma, struct vm_fault *vmf);
-  };</synopsis>
-        </para>
-        <para>
-          The <methodname>open</methodname> and <methodname>close</methodname>
-          operations must update the GEM object reference count. Drivers can use
-          the <function>drm_gem_vm_open</function> and
-          <function>drm_gem_vm_close</function> helper functions directly as open
-          and close handlers.
-        </para>
-        <para>
-          The fault operation handler is responsible for mapping individual pages
-          to userspace when a page fault occurs. Depending on the memory
-          allocation scheme, drivers can allocate pages at fault time, or can
-          decide to allocate memory for the GEM object at the time the object is
-          created.
-        </para>
-        <para>
-          Drivers that want to map the GEM object upfront instead of handling page
-          faults can implement their own mmap file operation handler.
-        </para>
-      </sect3>
-      <sect3>
-        <title>Memory Coherency</title>
-        <para>
-          When mapped to the device or used in a command buffer, backing pages
-          for an object are flushed to memory and marked write combined so as to
-          be coherent with the GPU. Likewise, if the CPU accesses an object
-          after the GPU has finished rendering to the object, then the object
-          must be made coherent with the CPU's view of memory, usually involving
-          GPU cache flushing of various kinds. This core CPU&lt;-&gt;GPU
-          coherency management is provided by a device-specific ioctl, which
-          evaluates an object's current domain and performs any necessary
-          flushing or synchronization to put the object into the desired
-          coherency domain (note that the object may be busy, i.e. an active
-          render target; in that case, setting the domain blocks the client and
-          waits for rendering to complete before performing any necessary
-          flushing operations).
-        </para>
-      </sect3>
-      <sect3>
-        <title>Command Execution</title>
-        <para>
-          Perhaps the most important GEM function for GPU devices is providing a
-          command execution interface to clients. Client programs construct
-          command buffers containing references to previously allocated memory
-          objects, and then submit them to GEM. At that point, GEM takes care to
-          bind all the objects into the GTT, execute the buffer, and provide
-          necessary synchronization between clients accessing the same buffers.
-          This often involves evicting some objects from the GTT and re-binding
-          others (a fairly expensive operation), and providing relocation
-          support which hides fixed GTT offsets from clients. Clients must take
-          care not to submit command buffers that reference more objects than
-          can fit in the GTT; otherwise, GEM will reject them and no rendering
-          will occur. Similarly, if several objects in the buffer require fence
-          registers to be allocated for correct rendering (e.g. 2D blits on
-          pre-965 chips), care must be taken not to require more fence registers
-          than are available to the client. Such resource management should be
-          abstracted from the client in libdrm.
-        </para>
-      </sect3>
-      <sect3>
-        <title>GEM Function Reference</title>
-!Edrivers/gpu/drm/drm_gem.c
-      </sect3>
-    </sect2>
-    <sect2>
-      <title>VMA Offset Manager</title>
-!Pdrivers/gpu/drm/drm_vma_manager.c vma offset manager
-!Edrivers/gpu/drm/drm_vma_manager.c
-!Iinclude/drm/drm_vma_manager.h
-    </sect2>
-    <sect2 id="drm-prime-support">
-      <title>PRIME Buffer Sharing</title>
-      <para>
-        PRIME is the cross device buffer sharing framework in drm, originally
-        created for the OPTIMUS range of multi-gpu platforms. To userspace
-        PRIME buffers are dma-buf based file descriptors.
-      </para>
-      <sect3>
-        <title>Overview and Driver Interface</title>
-        <para>
-          Similar to GEM global names, PRIME file descriptors are
-          also used to share buffer objects across processes. They offer
-          additional security: as file descriptors must be explicitly sent over
-          UNIX domain sockets to be shared between applications, they can't be
-          guessed like the globally unique GEM names.
-        </para>
-        <para>
-          Drivers that support the PRIME
-          API must set the DRIVER_PRIME bit in the struct
-          <structname>drm_driver</structname>
-          <structfield>driver_features</structfield> field, and implement the
-          <methodname>prime_handle_to_fd</methodname> and
-          <methodname>prime_fd_to_handle</methodname> operations.
-        </para>
-        <para>
-          <synopsis>int (*prime_handle_to_fd)(struct drm_device *dev,
-                          struct drm_file *file_priv, uint32_t handle,
-                          uint32_t flags, int *prime_fd);
-int (*prime_fd_to_handle)(struct drm_device *dev,
-                          struct drm_file *file_priv, int prime_fd,
-                          uint32_t *handle);</synopsis>
-            Those two operations convert a handle to a PRIME file descriptor and
-            vice versa. Drivers must use the kernel dma-buf buffer sharing framework
-            to manage the PRIME file descriptors. Similar to the mode setting
-            API PRIME is agnostic to the underlying buffer object manager, as
-            long as handles are 32bit unsigned integers.
-          </para>
-          <para>
-            While non-GEM drivers must implement the operations themselves, GEM
-            drivers must use the <function>drm_gem_prime_handle_to_fd</function>
-            and <function>drm_gem_prime_fd_to_handle</function> helper functions.
-            Those helpers rely on the driver
-            <methodname>gem_prime_export</methodname> and
-            <methodname>gem_prime_import</methodname> operations to create a dma-buf
-            instance from a GEM object (dma-buf exporter role) and to create a GEM
-            object from a dma-buf instance (dma-buf importer role).
-          </para>
-          <para>
-            <synopsis>struct dma_buf * (*gem_prime_export)(struct drm_device *dev,
-                             struct drm_gem_object *obj,
-                             int flags);
-struct drm_gem_object * (*gem_prime_import)(struct drm_device *dev,
-                                            struct dma_buf *dma_buf);</synopsis>
-            These two operations are mandatory for GEM drivers that support
-            PRIME.
-          </para>
-        </sect3>
-      <sect3>
-        <title>PRIME Helper Functions</title>
-!Pdrivers/gpu/drm/drm_prime.c PRIME Helpers
-      </sect3>
-    </sect2>
-    <sect2>
-      <title>PRIME Function References</title>
-!Edrivers/gpu/drm/drm_prime.c
-    </sect2>
-    <sect2>
-      <title>DRM MM Range Allocator</title>
-      <sect3>
-        <title>Overview</title>
-!Pdrivers/gpu/drm/drm_mm.c Overview
-      </sect3>
-      <sect3>
-        <title>LRU Scan/Eviction Support</title>
-!Pdrivers/gpu/drm/drm_mm.c lru scan roaster
-      </sect3>
-      </sect2>
-    <sect2>
-      <title>DRM MM Range Allocator Function References</title>
-!Edrivers/gpu/drm/drm_mm.c
-!Iinclude/drm/drm_mm.h
-    </sect2>
-    <sect2>
-      <title>CMA Helper Functions Reference</title>
-!Pdrivers/gpu/drm/drm_gem_cma_helper.c cma helpers
-!Edrivers/gpu/drm/drm_gem_cma_helper.c
-!Iinclude/drm/drm_gem_cma_helper.h
-    </sect2>
-  </sect1>
-
-  <!-- Internals: mode setting -->
-
-  <sect1 id="drm-mode-setting">
-    <title>Mode Setting</title>
-    <para>
-      Drivers must initialize the mode setting core by calling
-      <function>drm_mode_config_init</function> on the DRM device. The function
-      initializes the <structname>drm_device</structname>
-      <structfield>mode_config</structfield> field and never fails. Once done,
-      mode configuration must be setup by initializing the following fields.
-    </para>
-    <itemizedlist>
-      <listitem>
-        <synopsis>int min_width, min_height;
-int max_width, max_height;</synopsis>
-        <para>
-	  Minimum and maximum width and height of the frame buffers in pixel
-	  units.
-	</para>
-      </listitem>
-      <listitem>
-        <synopsis>struct drm_mode_config_funcs *funcs;</synopsis>
-	<para>Mode setting functions.</para>
-      </listitem>
-    </itemizedlist>
-    <sect2>
-      <title>Display Modes Function Reference</title>
-!Iinclude/drm/drm_modes.h
-!Edrivers/gpu/drm/drm_modes.c
-    </sect2>
-    <sect2>
-      <title>Atomic Mode Setting Function Reference</title>
-!Edrivers/gpu/drm/drm_atomic.c
-    </sect2>
-    <sect2>
-      <title>Frame Buffer Creation</title>
-      <synopsis>struct drm_framebuffer *(*fb_create)(struct drm_device *dev,
-				     struct drm_file *file_priv,
-				     struct drm_mode_fb_cmd2 *mode_cmd);</synopsis>
-      <para>
-        Frame buffers are abstract memory objects that provide a source of
-        pixels to scanout to a CRTC. Applications explicitly request the
-        creation of frame buffers through the DRM_IOCTL_MODE_ADDFB(2) ioctls and
-        receive an opaque handle that can be passed to the KMS CRTC control,
-        plane configuration and page flip functions.
-      </para>
-      <para>
-        Frame buffers rely on the underneath memory manager for low-level memory
-        operations. When creating a frame buffer applications pass a memory
-        handle (or a list of memory handles for multi-planar formats) through
-	the <parameter>drm_mode_fb_cmd2</parameter> argument. For drivers using
-	GEM as their userspace buffer management interface this would be a GEM
-	handle.  Drivers are however free to use their own backing storage object
-	handles, e.g. vmwgfx directly exposes special TTM handles to userspace
-	and so expects TTM handles in the create ioctl and not GEM handles.
-      </para>
-      <para>
-        Drivers must first validate the requested frame buffer parameters passed
-        through the mode_cmd argument. In particular this is where invalid
-        sizes, pixel formats or pitches can be caught.
-      </para>
-      <para>
-        If the parameters are deemed valid, drivers then create, initialize and
-        return an instance of struct <structname>drm_framebuffer</structname>.
-        If desired the instance can be embedded in a larger driver-specific
-	structure. Drivers must fill its <structfield>width</structfield>,
-	<structfield>height</structfield>, <structfield>pitches</structfield>,
-        <structfield>offsets</structfield>, <structfield>depth</structfield>,
-        <structfield>bits_per_pixel</structfield> and
-        <structfield>pixel_format</structfield> fields from the values passed
-        through the <parameter>drm_mode_fb_cmd2</parameter> argument. They
-        should call the <function>drm_helper_mode_fill_fb_struct</function>
-        helper function to do so.
-      </para>
-
-      <para>
-	The initialization of the new framebuffer instance is finalized with a
-	call to <function>drm_framebuffer_init</function> which takes a pointer
-	to DRM frame buffer operations (struct
-	<structname>drm_framebuffer_funcs</structname>). Note that this function
-	publishes the framebuffer and so from this point on it can be accessed
-	concurrently from other threads. Hence it must be the last step in the
-	driver's framebuffer initialization sequence. Frame buffer operations
-	are
-        <itemizedlist>
-          <listitem>
-            <synopsis>int (*create_handle)(struct drm_framebuffer *fb,
-		     struct drm_file *file_priv, unsigned int *handle);</synopsis>
-            <para>
-              Create a handle to the frame buffer underlying memory object. If
-              the frame buffer uses a multi-plane format, the handle will
-              reference the memory object associated with the first plane.
-            </para>
-            <para>
-              Drivers call <function>drm_gem_handle_create</function> to create
-              the handle.
-            </para>
-          </listitem>
-          <listitem>
-            <synopsis>void (*destroy)(struct drm_framebuffer *framebuffer);</synopsis>
-            <para>
-              Destroy the frame buffer object and frees all associated
-              resources. Drivers must call
-              <function>drm_framebuffer_cleanup</function> to free resources
-              allocated by the DRM core for the frame buffer object, and must
-              make sure to unreference all memory objects associated with the
-              frame buffer. Handles created by the
-              <methodname>create_handle</methodname> operation are released by
-              the DRM core.
-            </para>
-          </listitem>
-          <listitem>
-            <synopsis>int (*dirty)(struct drm_framebuffer *framebuffer,
-	     struct drm_file *file_priv, unsigned flags, unsigned color,
-	     struct drm_clip_rect *clips, unsigned num_clips);</synopsis>
-            <para>
-              This optional operation notifies the driver that a region of the
-              frame buffer has changed in response to a DRM_IOCTL_MODE_DIRTYFB
-              ioctl call.
-            </para>
-          </listitem>
-        </itemizedlist>
-      </para>
-      <para>
-	The lifetime of a drm framebuffer is controlled with a reference count,
-	drivers can grab additional references with
-	<function>drm_framebuffer_reference</function>and drop them
-	again with <function>drm_framebuffer_unreference</function>. For
-	driver-private framebuffers for which the last reference is never
-	dropped (e.g. for the fbdev framebuffer when the struct
-	<structname>drm_framebuffer</structname> is embedded into the fbdev
-	helper struct) drivers can manually clean up a framebuffer at module
-	unload time with
-	<function>drm_framebuffer_unregister_private</function>.
-      </para>
-    </sect2>
-    <sect2>
-      <title>Dumb Buffer Objects</title>
-      <para>
-	The KMS API doesn't standardize backing storage object creation and
-	leaves it to driver-specific ioctls. Furthermore actually creating a
-	buffer object even for GEM-based drivers is done through a
-	driver-specific ioctl - GEM only has a common userspace interface for
-	sharing and destroying objects. While not an issue for full-fledged
-	graphics stacks that include device-specific userspace components (in
-	libdrm for instance), this limit makes DRM-based early boot graphics
-	unnecessarily complex.
-      </para>
-      <para>
-        Dumb objects partly alleviate the problem by providing a standard
-        API to create dumb buffers suitable for scanout, which can then be used
-        to create KMS frame buffers.
-      </para>
-      <para>
-        To support dumb objects drivers must implement the
-        <methodname>dumb_create</methodname>,
-        <methodname>dumb_destroy</methodname> and
-        <methodname>dumb_map_offset</methodname> operations.
-      </para>
-      <itemizedlist>
-        <listitem>
-          <synopsis>int (*dumb_create)(struct drm_file *file_priv, struct drm_device *dev,
-                   struct drm_mode_create_dumb *args);</synopsis>
-          <para>
-            The <methodname>dumb_create</methodname> operation creates a driver
-	    object (GEM or TTM handle) suitable for scanout based on the
-	    width, height and depth from the struct
-	    <structname>drm_mode_create_dumb</structname> argument. It fills the
-	    argument's <structfield>handle</structfield>,
-	    <structfield>pitch</structfield> and <structfield>size</structfield>
-	    fields with a handle for the newly created object and its line
-            pitch and size in bytes.
-          </para>
-        </listitem>
-        <listitem>
-          <synopsis>int (*dumb_destroy)(struct drm_file *file_priv, struct drm_device *dev,
-                    uint32_t handle);</synopsis>
-          <para>
-            The <methodname>dumb_destroy</methodname> operation destroys a dumb
-            object created by <methodname>dumb_create</methodname>.
-          </para>
-        </listitem>
-        <listitem>
-          <synopsis>int (*dumb_map_offset)(struct drm_file *file_priv, struct drm_device *dev,
-                       uint32_t handle, uint64_t *offset);</synopsis>
-          <para>
-            The <methodname>dumb_map_offset</methodname> operation associates an
-            mmap fake offset with the object given by the handle and returns
-            it. Drivers must use the
-            <function>drm_gem_create_mmap_offset</function> function to
-            associate the fake offset as described in
-            <xref linkend="drm-gem-objects-mapping"/>.
-          </para>
-        </listitem>
-      </itemizedlist>
-      <para>
-        Note that dumb objects may not be used for gpu acceleration, as has been
-	attempted on some ARM embedded platforms. Such drivers really must have
-	a hardware-specific ioctl to allocate suitable buffer objects.
-      </para>
-    </sect2>
-    <sect2>
-      <title>Output Polling</title>
-      <synopsis>void (*output_poll_changed)(struct drm_device *dev);</synopsis>
-      <para>
-        This operation notifies the driver that the status of one or more
-        connectors has changed. Drivers that use the fb helper can just call the
-        <function>drm_fb_helper_hotplug_event</function> function to handle this
-        operation.
-      </para>
-    </sect2>
-    <sect2>
-      <title>Locking</title>
-      <para>
-        Beside some lookup structures with their own locking (which is hidden
-	behind the interface functions) most of the modeset state is protected
-	by the <code>dev-&lt;mode_config.lock</code> mutex and additionally
-	per-crtc locks to allow cursor updates, pageflips and similar operations
-	to occur concurrently with background tasks like output detection.
-	Operations which cross domains like a full modeset always grab all
-	locks. Drivers there need to protect resources shared between crtcs with
-	additional locking. They also need to be careful to always grab the
-	relevant crtc locks if a modset functions touches crtc state, e.g. for
-	load detection (which does only grab the <code>mode_config.lock</code>
-	to allow concurrent screen updates on live crtcs).
-      </para>
-    </sect2>
-  </sect1>
-
-  <!-- Internals: kms initialization and cleanup -->
-
-  <sect1 id="drm-kms-init">
-    <title>KMS Initialization and Cleanup</title>
-    <para>
-      A KMS device is abstracted and exposed as a set of planes, CRTCs, encoders
-      and connectors. KMS drivers must thus create and initialize all those
-      objects at load time after initializing mode setting.
-    </para>
-    <sect2>
-      <title>CRTCs (struct <structname>drm_crtc</structname>)</title>
-      <para>
-        A CRTC is an abstraction representing a part of the chip that contains a
-	pointer to a scanout buffer. Therefore, the number of CRTCs available
-	determines how many independent scanout buffers can be active at any
-	given time. The CRTC structure contains several fields to support this:
-	a pointer to some video memory (abstracted as a frame buffer object), a
-	display mode, and an (x, y) offset into the video memory to support
-	panning or configurations where one piece of video memory spans multiple
-	CRTCs.
-      </para>
-      <sect3>
-        <title>CRTC Initialization</title>
-        <para>
-          A KMS device must create and register at least one struct
-          <structname>drm_crtc</structname> instance. The instance is allocated
-          and zeroed by the driver, possibly as part of a larger structure, and
-          registered with a call to <function>drm_crtc_init</function> with a
-          pointer to CRTC functions.
-        </para>
-      </sect3>
-      <sect3 id="drm-kms-crtcops">
-        <title>CRTC Operations</title>
-        <sect4>
-          <title>Set Configuration</title>
-          <synopsis>int (*set_config)(struct drm_mode_set *set);</synopsis>
-          <para>
-            Apply a new CRTC configuration to the device. The configuration
-            specifies a CRTC, a frame buffer to scan out from, a (x,y) position in
-            the frame buffer, a display mode and an array of connectors to drive
-            with the CRTC if possible.
-          </para>
-          <para>
-            If the frame buffer specified in the configuration is NULL, the driver
-            must detach all encoders connected to the CRTC and all connectors
-            attached to those encoders and disable them.
-          </para>
-          <para>
-            This operation is called with the mode config lock held.
-          </para>
-          <note><para>
-	    Note that the drm core has no notion of restoring the mode setting
-	    state after resume, since all resume handling is in the full
-	    responsibility of the driver. The common mode setting helper library
-	    though provides a helper which can be used for this:
-	    <function>drm_helper_resume_force_mode</function>.
-          </para></note>
-        </sect4>
-        <sect4>
-          <title>Page Flipping</title>
-          <synopsis>int (*page_flip)(struct drm_crtc *crtc, struct drm_framebuffer *fb,
-                   struct drm_pending_vblank_event *event);</synopsis>
-          <para>
-            Schedule a page flip to the given frame buffer for the CRTC. This
-            operation is called with the mode config mutex held.
-          </para>
-          <para>
-            Page flipping is a synchronization mechanism that replaces the frame
-            buffer being scanned out by the CRTC with a new frame buffer during
-            vertical blanking, avoiding tearing. When an application requests a page
-            flip the DRM core verifies that the new frame buffer is large enough to
-            be scanned out by  the CRTC in the currently configured mode and then
-            calls the CRTC <methodname>page_flip</methodname> operation with a
-            pointer to the new frame buffer.
-          </para>
-          <para>
-            The <methodname>page_flip</methodname> operation schedules a page flip.
-            Once any pending rendering targeting the new frame buffer has
-            completed, the CRTC will be reprogrammed to display that frame buffer
-            after the next vertical refresh. The operation must return immediately
-            without waiting for rendering or page flip to complete and must block
-            any new rendering to the frame buffer until the page flip completes.
-          </para>
-          <para>
-            If a page flip can be successfully scheduled the driver must set the
-            <code>drm_crtc-&gt;fb</code> field to the new framebuffer pointed to
-            by <code>fb</code>. This is important so that the reference counting
-            on framebuffers stays balanced.
-          </para>
-          <para>
-            If a page flip is already pending, the
-            <methodname>page_flip</methodname> operation must return
-            -<errorname>EBUSY</errorname>.
-          </para>
-          <para>
-            To synchronize page flip to vertical blanking the driver will likely
-            need to enable vertical blanking interrupts. It should call
-            <function>drm_vblank_get</function> for that purpose, and call
-            <function>drm_vblank_put</function> after the page flip completes.
-          </para>
-          <para>
-            If the application has requested to be notified when page flip completes
-            the <methodname>page_flip</methodname> operation will be called with a
-            non-NULL <parameter>event</parameter> argument pointing to a
-            <structname>drm_pending_vblank_event</structname> instance. Upon page
-            flip completion the driver must call <methodname>drm_send_vblank_event</methodname>
-            to fill in the event and send to wake up any waiting processes.
-            This can be performed with
-            <programlisting><![CDATA[
-            spin_lock_irqsave(&dev->event_lock, flags);
-            ...
-            drm_send_vblank_event(dev, pipe, event);
-            spin_unlock_irqrestore(&dev->event_lock, flags);
-            ]]></programlisting>
-          </para>
-          <note><para>
-            FIXME: Could drivers that don't need to wait for rendering to complete
-            just add the event to <literal>dev-&gt;vblank_event_list</literal> and
-            let the DRM core handle everything, as for "normal" vertical blanking
-            events?
-          </para></note>
-          <para>
-            While waiting for the page flip to complete, the
-            <literal>event-&gt;base.link</literal> list head can be used freely by
-            the driver to store the pending event in a driver-specific list.
-          </para>
-          <para>
-            If the file handle is closed before the event is signaled, drivers must
-            take care to destroy the event in their
-            <methodname>preclose</methodname> operation (and, if needed, call
-            <function>drm_vblank_put</function>).
-          </para>
-        </sect4>
-        <sect4>
-          <title>Miscellaneous</title>
-          <itemizedlist>
-            <listitem>
-              <synopsis>void (*set_property)(struct drm_crtc *crtc,
-                     struct drm_property *property, uint64_t value);</synopsis>
-              <para>
-                Set the value of the given CRTC property to
-                <parameter>value</parameter>. See <xref linkend="drm-kms-properties"/>
-                for more information about properties.
-              </para>
-            </listitem>
-            <listitem>
-              <synopsis>void (*gamma_set)(struct drm_crtc *crtc, u16 *r, u16 *g, u16 *b,
-                        uint32_t start, uint32_t size);</synopsis>
-              <para>
-                Apply a gamma table to the device. The operation is optional.
-              </para>
-            </listitem>
-            <listitem>
-              <synopsis>void (*destroy)(struct drm_crtc *crtc);</synopsis>
-              <para>
-                Destroy the CRTC when not needed anymore. See
-                <xref linkend="drm-kms-init"/>.
-              </para>
-            </listitem>
-          </itemizedlist>
-        </sect4>
-      </sect3>
-    </sect2>
-    <sect2>
-      <title>Planes (struct <structname>drm_plane</structname>)</title>
-      <para>
-        A plane represents an image source that can be blended with or overlayed
-	on top of a CRTC during the scanout process. Planes are associated with
-	a frame buffer to crop a portion of the image memory (source) and
-	optionally scale it to a destination size. The result is then blended
-	with or overlayed on top of a CRTC.
-      </para>
-      <para>
-      The DRM core recognizes three types of planes:
-      <itemizedlist>
-        <listitem>
-        DRM_PLANE_TYPE_PRIMARY represents a "main" plane for a CRTC.  Primary
-        planes are the planes operated upon by CRTC modesetting and flipping
-        operations described in <xref linkend="drm-kms-crtcops"/>.
-        </listitem>
-        <listitem>
-        DRM_PLANE_TYPE_CURSOR represents a "cursor" plane for a CRTC.  Cursor
-        planes are the planes operated upon by the DRM_IOCTL_MODE_CURSOR and
-        DRM_IOCTL_MODE_CURSOR2 ioctls.
-        </listitem>
-        <listitem>
-        DRM_PLANE_TYPE_OVERLAY represents all non-primary, non-cursor planes.
-        Some drivers refer to these types of planes as "sprites" internally.
-        </listitem>
-      </itemizedlist>
-      For compatibility with legacy userspace, only overlay planes are made
-      available to userspace by default.  Userspace clients may set the
-      DRM_CLIENT_CAP_UNIVERSAL_PLANES client capability bit to indicate that
-      they wish to receive a universal plane list containing all plane types.
-      </para>
-      <sect3>
-        <title>Plane Initialization</title>
-        <para>
-          To create a plane, a KMS drivers allocates and
-          zeroes an instances of struct <structname>drm_plane</structname>
-          (possibly as part of a larger structure) and registers it with a call
-          to <function>drm_universal_plane_init</function>. The function takes a bitmask
-          of the CRTCs that can be associated with the plane, a pointer to the
-          plane functions, a list of format supported formats, and the type of
-          plane (primary, cursor, or overlay) being initialized.
-        </para>
-        <para>
-          Cursor and overlay planes are optional.  All drivers should provide
-          one primary plane per CRTC (although this requirement may change in
-          the future); drivers that do not wish to provide special handling for
-          primary planes may make use of the helper functions described in
-          <xref linkend="drm-kms-planehelpers"/> to create and register a
-          primary plane with standard capabilities.
-        </para>
-      </sect3>
-      <sect3>
-        <title>Plane Operations</title>
-        <itemizedlist>
-          <listitem>
-            <synopsis>int (*update_plane)(struct drm_plane *plane, struct drm_crtc *crtc,
-                        struct drm_framebuffer *fb, int crtc_x, int crtc_y,
-                        unsigned int crtc_w, unsigned int crtc_h,
-                        uint32_t src_x, uint32_t src_y,
-                        uint32_t src_w, uint32_t src_h);</synopsis>
-            <para>
-              Enable and configure the plane to use the given CRTC and frame buffer.
-            </para>
-            <para>
-              The source rectangle in frame buffer memory coordinates is given by
-              the <parameter>src_x</parameter>, <parameter>src_y</parameter>,
-              <parameter>src_w</parameter> and <parameter>src_h</parameter>
-              parameters (as 16.16 fixed point values). Devices that don't support
-              subpixel plane coordinates can ignore the fractional part.
-            </para>
-            <para>
-              The destination rectangle in CRTC coordinates is given by the
-              <parameter>crtc_x</parameter>, <parameter>crtc_y</parameter>,
-              <parameter>crtc_w</parameter> and <parameter>crtc_h</parameter>
-              parameters (as integer values). Devices scale the source rectangle to
-              the destination rectangle. If scaling is not supported, and the source
-              rectangle size doesn't match the destination rectangle size, the
-              driver must return a -<errorname>EINVAL</errorname> error.
-            </para>
-          </listitem>
-          <listitem>
-            <synopsis>int (*disable_plane)(struct drm_plane *plane);</synopsis>
-            <para>
-              Disable the plane. The DRM core calls this method in response to a
-              DRM_IOCTL_MODE_SETPLANE ioctl call with the frame buffer ID set to 0.
-              Disabled planes must not be processed by the CRTC.
-            </para>
-          </listitem>
-          <listitem>
-            <synopsis>void (*destroy)(struct drm_plane *plane);</synopsis>
-            <para>
-              Destroy the plane when not needed anymore. See
-              <xref linkend="drm-kms-init"/>.
-            </para>
-          </listitem>
-        </itemizedlist>
-      </sect3>
-    </sect2>
-    <sect2>
-      <title>Encoders (struct <structname>drm_encoder</structname>)</title>
-      <para>
-        An encoder takes pixel data from a CRTC and converts it to a format
-	suitable for any attached connectors. On some devices, it may be
-	possible to have a CRTC send data to more than one encoder. In that
-	case, both encoders would receive data from the same scanout buffer,
-	resulting in a "cloned" display configuration across the connectors
-	attached to each encoder.
-      </para>
-      <sect3>
-        <title>Encoder Initialization</title>
-        <para>
-          As for CRTCs, a KMS driver must create, initialize and register at
-          least one struct <structname>drm_encoder</structname> instance. The
-          instance is allocated and zeroed by the driver, possibly as part of a
-          larger structure.
-        </para>
-        <para>
-          Drivers must initialize the struct <structname>drm_encoder</structname>
-          <structfield>possible_crtcs</structfield> and
-          <structfield>possible_clones</structfield> fields before registering the
-          encoder. Both fields are bitmasks of respectively the CRTCs that the
-          encoder can be connected to, and sibling encoders candidate for cloning.
-        </para>
-        <para>
-          After being initialized, the encoder must be registered with a call to
-          <function>drm_encoder_init</function>. The function takes a pointer to
-          the encoder functions and an encoder type. Supported types are
-          <itemizedlist>
-            <listitem>
-              DRM_MODE_ENCODER_DAC for VGA and analog on DVI-I/DVI-A
-              </listitem>
-            <listitem>
-              DRM_MODE_ENCODER_TMDS for DVI, HDMI and (embedded) DisplayPort
-            </listitem>
-            <listitem>
-              DRM_MODE_ENCODER_LVDS for display panels
-            </listitem>
-            <listitem>
-              DRM_MODE_ENCODER_TVDAC for TV output (Composite, S-Video, Component,
-              SCART)
-            </listitem>
-            <listitem>
-              DRM_MODE_ENCODER_VIRTUAL for virtual machine displays
-            </listitem>
-          </itemizedlist>
-        </para>
-        <para>
-          Encoders must be attached to a CRTC to be used. DRM drivers leave
-          encoders unattached at initialization time. Applications (or the fbdev
-          compatibility layer when implemented) are responsible for attaching the
-          encoders they want to use to a CRTC.
-        </para>
-      </sect3>
-      <sect3>
-        <title>Encoder Operations</title>
-        <itemizedlist>
-          <listitem>
-            <synopsis>void (*destroy)(struct drm_encoder *encoder);</synopsis>
-            <para>
-              Called to destroy the encoder when not needed anymore. See
-              <xref linkend="drm-kms-init"/>.
-            </para>
-          </listitem>
-          <listitem>
-            <synopsis>void (*set_property)(struct drm_plane *plane,
-                     struct drm_property *property, uint64_t value);</synopsis>
-            <para>
-              Set the value of the given plane property to
-              <parameter>value</parameter>. See <xref linkend="drm-kms-properties"/>
-              for more information about properties.
-            </para>
-          </listitem>
-        </itemizedlist>
-      </sect3>
-    </sect2>
-    <sect2>
-      <title>Connectors (struct <structname>drm_connector</structname>)</title>
-      <para>
-        A connector is the final destination for pixel data on a device, and
-	usually connects directly to an external display device like a monitor
-	or laptop panel. A connector can only be attached to one encoder at a
-	time. The connector is also the structure where information about the
-	attached display is kept, so it contains fields for display data, EDID
-	data, DPMS &amp; connection status, and information about modes
-	supported on the attached displays.
-      </para>
-      <sect3>
-        <title>Connector Initialization</title>
-        <para>
-          Finally a KMS driver must create, initialize, register and attach at
-          least one struct <structname>drm_connector</structname> instance. The
-          instance is created as other KMS objects and initialized by setting the
-          following fields.
-        </para>
-        <variablelist>
-          <varlistentry>
-            <term><structfield>interlace_allowed</structfield></term>
-            <listitem><para>
-              Whether the connector can handle interlaced modes.
-            </para></listitem>
-          </varlistentry>
-          <varlistentry>
-            <term><structfield>doublescan_allowed</structfield></term>
-            <listitem><para>
-              Whether the connector can handle doublescan.
-            </para></listitem>
-          </varlistentry>
-          <varlistentry>
-            <term><structfield>display_info
-            </structfield></term>
-            <listitem><para>
-              Display information is filled from EDID information when a display
-              is detected. For non hot-pluggable displays such as flat panels in
-              embedded systems, the driver should initialize the
-              <structfield>display_info</structfield>.<structfield>width_mm</structfield>
-              and
-              <structfield>display_info</structfield>.<structfield>height_mm</structfield>
-              fields with the physical size of the display.
-            </para></listitem>
-          </varlistentry>
-          <varlistentry>
-            <term id="drm-kms-connector-polled"><structfield>polled</structfield></term>
-            <listitem><para>
-              Connector polling mode, a combination of
-              <variablelist>
-                <varlistentry>
-                  <term>DRM_CONNECTOR_POLL_HPD</term>
-                  <listitem><para>
-                    The connector generates hotplug events and doesn't need to be
-                    periodically polled. The CONNECT and DISCONNECT flags must not
-                    be set together with the HPD flag.
-                  </para></listitem>
-                </varlistentry>
-                <varlistentry>
-                  <term>DRM_CONNECTOR_POLL_CONNECT</term>
-                  <listitem><para>
-                    Periodically poll the connector for connection.
-                  </para></listitem>
-                </varlistentry>
-                <varlistentry>
-                  <term>DRM_CONNECTOR_POLL_DISCONNECT</term>
-                  <listitem><para>
-                    Periodically poll the connector for disconnection.
-                  </para></listitem>
-                </varlistentry>
-              </variablelist>
-              Set to 0 for connectors that don't support connection status
-              discovery.
-            </para></listitem>
-          </varlistentry>
-        </variablelist>
-        <para>
-          The connector is then registered with a call to
-          <function>drm_connector_init</function> with a pointer to the connector
-          functions and a connector type, and exposed through sysfs with a call to
-          <function>drm_connector_register</function>.
-        </para>
-        <para>
-          Supported connector types are
-          <itemizedlist>
-            <listitem>DRM_MODE_CONNECTOR_VGA</listitem>
-            <listitem>DRM_MODE_CONNECTOR_DVII</listitem>
-            <listitem>DRM_MODE_CONNECTOR_DVID</listitem>
-            <listitem>DRM_MODE_CONNECTOR_DVIA</listitem>
-            <listitem>DRM_MODE_CONNECTOR_Composite</listitem>
-            <listitem>DRM_MODE_CONNECTOR_SVIDEO</listitem>
-            <listitem>DRM_MODE_CONNECTOR_LVDS</listitem>
-            <listitem>DRM_MODE_CONNECTOR_Component</listitem>
-            <listitem>DRM_MODE_CONNECTOR_9PinDIN</listitem>
-            <listitem>DRM_MODE_CONNECTOR_DisplayPort</listitem>
-            <listitem>DRM_MODE_CONNECTOR_HDMIA</listitem>
-            <listitem>DRM_MODE_CONNECTOR_HDMIB</listitem>
-            <listitem>DRM_MODE_CONNECTOR_TV</listitem>
-            <listitem>DRM_MODE_CONNECTOR_eDP</listitem>
-            <listitem>DRM_MODE_CONNECTOR_VIRTUAL</listitem>
-          </itemizedlist>
-        </para>
-        <para>
-          Connectors must be attached to an encoder to be used. For devices that
-          map connectors to encoders 1:1, the connector should be attached at
-          initialization time with a call to
-          <function>drm_mode_connector_attach_encoder</function>. The driver must
-          also set the <structname>drm_connector</structname>
-          <structfield>encoder</structfield> field to point to the attached
-          encoder.
-        </para>
-        <para>
-          Finally, drivers must initialize the connectors state change detection
-          with a call to <function>drm_kms_helper_poll_init</function>. If at
-          least one connector is pollable but can't generate hotplug interrupts
-          (indicated by the DRM_CONNECTOR_POLL_CONNECT and
-          DRM_CONNECTOR_POLL_DISCONNECT connector flags), a delayed work will
-          automatically be queued to periodically poll for changes. Connectors
-          that can generate hotplug interrupts must be marked with the
-          DRM_CONNECTOR_POLL_HPD flag instead, and their interrupt handler must
-          call <function>drm_helper_hpd_irq_event</function>. The function will
-          queue a delayed work to check the state of all connectors, but no
-          periodic polling will be done.
-        </para>
-      </sect3>
-      <sect3>
-        <title>Connector Operations</title>
-        <note><para>
-          Unless otherwise state, all operations are mandatory.
-        </para></note>
-        <sect4>
-          <title>DPMS</title>
-          <synopsis>void (*dpms)(struct drm_connector *connector, int mode);</synopsis>
-          <para>
-            The DPMS operation sets the power state of a connector. The mode
-            argument is one of
-            <itemizedlist>
-              <listitem><para>DRM_MODE_DPMS_ON</para></listitem>
-              <listitem><para>DRM_MODE_DPMS_STANDBY</para></listitem>
-              <listitem><para>DRM_MODE_DPMS_SUSPEND</para></listitem>
-              <listitem><para>DRM_MODE_DPMS_OFF</para></listitem>
-            </itemizedlist>
-          </para>
-          <para>
-            In all but DPMS_ON mode the encoder to which the connector is attached
-            should put the display in low-power mode by driving its signals
-            appropriately. If more than one connector is attached to the encoder
-            care should be taken not to change the power state of other displays as
-            a side effect. Low-power mode should be propagated to the encoders and
-            CRTCs when all related connectors are put in low-power mode.
-          </para>
-        </sect4>
-        <sect4>
-          <title>Modes</title>
-          <synopsis>int (*fill_modes)(struct drm_connector *connector, uint32_t max_width,
-                      uint32_t max_height);</synopsis>
-          <para>
-            Fill the mode list with all supported modes for the connector. If the
-            <parameter>max_width</parameter> and <parameter>max_height</parameter>
-            arguments are non-zero, the implementation must ignore all modes wider
-            than <parameter>max_width</parameter> or higher than
-            <parameter>max_height</parameter>.
-          </para>
-          <para>
-            The connector must also fill in this operation its
-            <structfield>display_info</structfield>
-            <structfield>width_mm</structfield> and
-            <structfield>height_mm</structfield> fields with the connected display
-            physical size in millimeters. The fields should be set to 0 if the value
-            isn't known or is not applicable (for instance for projector devices).
-          </para>
-        </sect4>
-        <sect4>
-          <title>Connection Status</title>
-          <para>
-            The connection status is updated through polling or hotplug events when
-            supported (see <xref linkend="drm-kms-connector-polled"/>). The status
-            value is reported to userspace through ioctls and must not be used
-            inside the driver, as it only gets initialized by a call to
-            <function>drm_mode_getconnector</function> from userspace.
-          </para>
-          <synopsis>enum drm_connector_status (*detect)(struct drm_connector *connector,
-                                        bool force);</synopsis>
-          <para>
-            Check to see if anything is attached to the connector. The
-            <parameter>force</parameter> parameter is set to false whilst polling or
-            to true when checking the connector due to user request.
-            <parameter>force</parameter> can be used by the driver to avoid
-            expensive, destructive operations during automated probing.
-          </para>
-          <para>
-            Return connector_status_connected if something is connected to the
-            connector, connector_status_disconnected if nothing is connected and
-            connector_status_unknown if the connection state isn't known.
-          </para>
-          <para>
-            Drivers should only return connector_status_connected if the connection
-            status has really been probed as connected. Connectors that can't detect
-            the connection status, or failed connection status probes, should return
-            connector_status_unknown.
-          </para>
-        </sect4>
-        <sect4>
-          <title>Miscellaneous</title>
-          <itemizedlist>
-            <listitem>
-              <synopsis>void (*set_property)(struct drm_connector *connector,
-                     struct drm_property *property, uint64_t value);</synopsis>
-              <para>
-                Set the value of the given connector property to
-                <parameter>value</parameter>. See <xref linkend="drm-kms-properties"/>
-                for more information about properties.
-              </para>
-            </listitem>
-            <listitem>
-              <synopsis>void (*destroy)(struct drm_connector *connector);</synopsis>
-              <para>
-                Destroy the connector when not needed anymore. See
-                <xref linkend="drm-kms-init"/>.
-              </para>
-            </listitem>
-          </itemizedlist>
-        </sect4>
-      </sect3>
-    </sect2>
-    <sect2>
-      <title>Cleanup</title>
-      <para>
-        The DRM core manages its objects' lifetime. When an object is not needed
-	anymore the core calls its destroy function, which must clean up and
-	free every resource allocated for the object. Every
-	<function>drm_*_init</function> call must be matched with a
-	corresponding <function>drm_*_cleanup</function> call to cleanup CRTCs
-	(<function>drm_crtc_cleanup</function>), planes
-	(<function>drm_plane_cleanup</function>), encoders
-	(<function>drm_encoder_cleanup</function>) and connectors
-	(<function>drm_connector_cleanup</function>). Furthermore, connectors
-	that have been added to sysfs must be removed by a call to
-	<function>drm_connector_unregister</function> before calling
-	<function>drm_connector_cleanup</function>.
-      </para>
-      <para>
-        Connectors state change detection must be cleanup up with a call to
-	<function>drm_kms_helper_poll_fini</function>.
-      </para>
-    </sect2>
-    <sect2>
-      <title>Output discovery and initialization example</title>
-      <programlisting><![CDATA[
-void intel_crt_init(struct drm_device *dev)
-{
-	struct drm_connector *connector;
-	struct intel_output *intel_output;
-
-	intel_output = kzalloc(sizeof(struct intel_output), GFP_KERNEL);
-	if (!intel_output)
-		return;
-
-	connector = &intel_output->base;
-	drm_connector_init(dev, &intel_output->base,
-			   &intel_crt_connector_funcs, DRM_MODE_CONNECTOR_VGA);
-
-	drm_encoder_init(dev, &intel_output->enc, &intel_crt_enc_funcs,
-			 DRM_MODE_ENCODER_DAC);
-
-	drm_mode_connector_attach_encoder(&intel_output->base,
-					  &intel_output->enc);
-
-	/* Set up the DDC bus. */
-	intel_output->ddc_bus = intel_i2c_create(dev, GPIOA, "CRTDDC_A");
-	if (!intel_output->ddc_bus) {
-		dev_printk(KERN_ERR, &dev->pdev->dev, "DDC bus registration "
-			   "failed.\n");
-		return;
-	}
-
-	intel_output->type = INTEL_OUTPUT_ANALOG;
-	connector->interlace_allowed = 0;
-	connector->doublescan_allowed = 0;
-
-	drm_encoder_helper_add(&intel_output->enc, &intel_crt_helper_funcs);
-	drm_connector_helper_add(connector, &intel_crt_connector_helper_funcs);
-
-	drm_connector_register(connector);
-}]]></programlisting>
-      <para>
-        In the example above (taken from the i915 driver), a CRTC, connector and
-        encoder combination is created. A device-specific i2c bus is also
-        created for fetching EDID data and performing monitor detection. Once
-        the process is complete, the new connector is registered with sysfs to
-        make its properties available to applications.
-      </para>
-    </sect2>
-    <sect2>
-      <title>KMS API Functions</title>
-!Edrivers/gpu/drm/drm_crtc.c
-    </sect2>
-    <sect2>
-      <title>KMS Data Structures</title>
-!Iinclude/drm/drm_crtc.h
-    </sect2>
-    <sect2>
-      <title>KMS Locking</title>
-!Pdrivers/gpu/drm/drm_modeset_lock.c kms locking
-!Iinclude/drm/drm_modeset_lock.h
-!Edrivers/gpu/drm/drm_modeset_lock.c
-    </sect2>
-  </sect1>
-
-  <!-- Internals: kms helper functions -->
-
-  <sect1>
-    <title>Mode Setting Helper Functions</title>
-    <para>
-      The plane, CRTC, encoder and connector functions provided by the drivers
-      implement the DRM API. They're called by the DRM core and ioctl handlers
-      to handle device state changes and configuration request. As implementing
-      those functions often requires logic not specific to drivers, mid-layer
-      helper functions are available to avoid duplicating boilerplate code.
-    </para>
-    <para>
-      The DRM core contains one mid-layer implementation. The mid-layer provides
-      implementations of several plane, CRTC, encoder and connector functions
-      (called from the top of the mid-layer) that pre-process requests and call
-      lower-level functions provided by the driver (at the bottom of the
-      mid-layer). For instance, the
-      <function>drm_crtc_helper_set_config</function> function can be used to
-      fill the struct <structname>drm_crtc_funcs</structname>
-      <structfield>set_config</structfield> field. When called, it will split
-      the <methodname>set_config</methodname> operation in smaller, simpler
-      operations and call the driver to handle them.
-    </para>
-    <para>
-      To use the mid-layer, drivers call <function>drm_crtc_helper_add</function>,
-      <function>drm_encoder_helper_add</function> and
-      <function>drm_connector_helper_add</function> functions to install their
-      mid-layer bottom operations handlers, and fill the
-      <structname>drm_crtc_funcs</structname>,
-      <structname>drm_encoder_funcs</structname> and
-      <structname>drm_connector_funcs</structname> structures with pointers to
-      the mid-layer top API functions. Installing the mid-layer bottom operation
-      handlers is best done right after registering the corresponding KMS object.
-    </para>
-    <para>
-      The mid-layer is not split between CRTC, encoder and connector operations.
-      To use it, a driver must provide bottom functions for all of the three KMS
-      entities.
-    </para>
-    <sect2>
-      <title>Helper Functions</title>
-      <itemizedlist>
-        <listitem>
-          <synopsis>int drm_crtc_helper_set_config(struct drm_mode_set *set);</synopsis>
-          <para>
-            The <function>drm_crtc_helper_set_config</function> helper function
-            is a CRTC <methodname>set_config</methodname> implementation. It
-            first tries to locate the best encoder for each connector by calling
-            the connector <methodname>best_encoder</methodname> helper
-            operation.
-          </para>
-          <para>
-            After locating the appropriate encoders, the helper function will
-            call the <methodname>mode_fixup</methodname> encoder and CRTC helper
-            operations to adjust the requested mode, or reject it completely in
-            which case an error will be returned to the application. If the new
-            configuration after mode adjustment is identical to the current
-            configuration the helper function will return without performing any
-            other operation.
-          </para>
-          <para>
-            If the adjusted mode is identical to the current mode but changes to
-            the frame buffer need to be applied, the
-            <function>drm_crtc_helper_set_config</function> function will call
-            the CRTC <methodname>mode_set_base</methodname> helper operation. If
-            the adjusted mode differs from the current mode, or if the
-            <methodname>mode_set_base</methodname> helper operation is not
-            provided, the helper function performs a full mode set sequence by
-            calling the <methodname>prepare</methodname>,
-            <methodname>mode_set</methodname> and
-            <methodname>commit</methodname> CRTC and encoder helper operations,
-            in that order.
-          </para>
-        </listitem>
-        <listitem>
-          <synopsis>void drm_helper_connector_dpms(struct drm_connector *connector, int mode);</synopsis>
-          <para>
-            The <function>drm_helper_connector_dpms</function> helper function
-            is a connector <methodname>dpms</methodname> implementation that
-            tracks power state of connectors. To use the function, drivers must
-            provide <methodname>dpms</methodname> helper operations for CRTCs
-            and encoders to apply the DPMS state to the device.
-          </para>
-          <para>
-            The mid-layer doesn't track the power state of CRTCs and encoders.
-            The <methodname>dpms</methodname> helper operations can thus be
-            called with a mode identical to the currently active mode.
-          </para>
-        </listitem>
-        <listitem>
-          <synopsis>int drm_helper_probe_single_connector_modes(struct drm_connector *connector,
-                                            uint32_t maxX, uint32_t maxY);</synopsis>
-          <para>
-            The <function>drm_helper_probe_single_connector_modes</function> helper
-            function is a connector <methodname>fill_modes</methodname>
-            implementation that updates the connection status for the connector
-            and then retrieves a list of modes by calling the connector
-            <methodname>get_modes</methodname> helper operation.
-          </para>
-         <para>
-            If the helper operation returns no mode, and if the connector status
-            is connector_status_connected, standard VESA DMT modes up to
-            1024x768 are automatically added to the modes list by a call to
-            <function>drm_add_modes_noedid</function>.
-          </para>
-          <para>
-            The function then filters out modes larger than
-            <parameter>max_width</parameter> and <parameter>max_height</parameter>
-            if specified. It finally calls the optional connector
-            <methodname>mode_valid</methodname> helper operation for each mode in
-            the probed list to check whether the mode is valid for the connector.
-          </para>
-        </listitem>
-      </itemizedlist>
-    </sect2>
-    <sect2>
-      <title>CRTC Helper Operations</title>
-      <itemizedlist>
-        <listitem id="drm-helper-crtc-mode-fixup">
-          <synopsis>bool (*mode_fixup)(struct drm_crtc *crtc,
-                       const struct drm_display_mode *mode,
-                       struct drm_display_mode *adjusted_mode);</synopsis>
-          <para>
-            Let CRTCs adjust the requested mode or reject it completely. This
-            operation returns true if the mode is accepted (possibly after being
-            adjusted) or false if it is rejected.
-          </para>
-          <para>
-            The <methodname>mode_fixup</methodname> operation should reject the
-            mode if it can't reasonably use it. The definition of "reasonable"
-            is currently fuzzy in this context. One possible behaviour would be
-            to set the adjusted mode to the panel timings when a fixed-mode
-            panel is used with hardware capable of scaling. Another behaviour
-            would be to accept any input mode and adjust it to the closest mode
-            supported by the hardware (FIXME: This needs to be clarified).
-          </para>
-        </listitem>
-        <listitem>
-          <synopsis>int (*mode_set_base)(struct drm_crtc *crtc, int x, int y,
-                     struct drm_framebuffer *old_fb)</synopsis>
-          <para>
-            Move the CRTC on the current frame buffer (stored in
-            <literal>crtc-&gt;fb</literal>) to position (x,y). Any of the frame
-            buffer, x position or y position may have been modified.
-          </para>
-          <para>
-            This helper operation is optional. If not provided, the
-            <function>drm_crtc_helper_set_config</function> function will fall
-            back to the <methodname>mode_set</methodname> helper operation.
-          </para>
-          <note><para>
-            FIXME: Why are x and y passed as arguments, as they can be accessed
-            through <literal>crtc-&gt;x</literal> and
-            <literal>crtc-&gt;y</literal>?
-          </para></note>
-        </listitem>
-        <listitem>
-          <synopsis>void (*prepare)(struct drm_crtc *crtc);</synopsis>
-          <para>
-            Prepare the CRTC for mode setting. This operation is called after
-            validating the requested mode. Drivers use it to perform
-            device-specific operations required before setting the new mode.
-          </para>
-        </listitem>
-        <listitem>
-          <synopsis>int (*mode_set)(struct drm_crtc *crtc, struct drm_display_mode *mode,
-                struct drm_display_mode *adjusted_mode, int x, int y,
-                struct drm_framebuffer *old_fb);</synopsis>
-          <para>
-            Set a new mode, position and frame buffer. Depending on the device
-            requirements, the mode can be stored internally by the driver and
-            applied in the <methodname>commit</methodname> operation, or
-            programmed to the hardware immediately.
-          </para>
-          <para>
-            The <methodname>mode_set</methodname> operation returns 0 on success
-	    or a negative error code if an error occurs.
-          </para>
-        </listitem>
-        <listitem>
-          <synopsis>void (*commit)(struct drm_crtc *crtc);</synopsis>
-          <para>
-            Commit a mode. This operation is called after setting the new mode.
-            Upon return the device must use the new mode and be fully
-            operational.
-          </para>
-        </listitem>
-      </itemizedlist>
-    </sect2>
-    <sect2>
-      <title>Encoder Helper Operations</title>
-      <itemizedlist>
-        <listitem>
-          <synopsis>bool (*mode_fixup)(struct drm_encoder *encoder,
-                       const struct drm_display_mode *mode,
-                       struct drm_display_mode *adjusted_mode);</synopsis>
-          <para>
-            Let encoders adjust the requested mode or reject it completely. This
-            operation returns true if the mode is accepted (possibly after being
-            adjusted) or false if it is rejected. See the
-            <link linkend="drm-helper-crtc-mode-fixup">mode_fixup CRTC helper
-            operation</link> for an explanation of the allowed adjustments.
-          </para>
-        </listitem>
-        <listitem>
-          <synopsis>void (*prepare)(struct drm_encoder *encoder);</synopsis>
-          <para>
-            Prepare the encoder for mode setting. This operation is called after
-            validating the requested mode. Drivers use it to perform
-            device-specific operations required before setting the new mode.
-          </para>
-        </listitem>
-        <listitem>
-          <synopsis>void (*mode_set)(struct drm_encoder *encoder,
-                 struct drm_display_mode *mode,
-                 struct drm_display_mode *adjusted_mode);</synopsis>
-          <para>
-            Set a new mode. Depending on the device requirements, the mode can
-            be stored internally by the driver and applied in the
-            <methodname>commit</methodname> operation, or programmed to the
-            hardware immediately.
-          </para>
-        </listitem>
-        <listitem>
-          <synopsis>void (*commit)(struct drm_encoder *encoder);</synopsis>
-          <para>
-            Commit a mode. This operation is called after setting the new mode.
-            Upon return the device must use the new mode and be fully
-            operational.
-          </para>
-        </listitem>
-      </itemizedlist>
-    </sect2>
-    <sect2>
-      <title>Connector Helper Operations</title>
-      <itemizedlist>
-        <listitem>
-          <synopsis>struct drm_encoder *(*best_encoder)(struct drm_connector *connector);</synopsis>
-          <para>
-            Return a pointer to the best encoder for the connecter. Device that
-            map connectors to encoders 1:1 simply return the pointer to the
-            associated encoder. This operation is mandatory.
-          </para>
-        </listitem>
-        <listitem>
-          <synopsis>int (*get_modes)(struct drm_connector *connector);</synopsis>
-          <para>
-            Fill the connector's <structfield>probed_modes</structfield> list
-            by parsing EDID data with <function>drm_add_edid_modes</function>,
-            adding standard VESA DMT modes with <function>drm_add_modes_noedid</function>,
-            or calling <function>drm_mode_probed_add</function> directly for every
-            supported mode and return the number of modes it has detected. This
-            operation is mandatory.
-          </para>
-          <para>
-            Note that the caller function will automatically add standard VESA
-            DMT modes up to 1024x768 if the <methodname>get_modes</methodname>
-            helper operation returns no mode and if the connector status is
-            connector_status_connected. There is no need to call
-            <function>drm_add_edid_modes</function> manually in that case.
-          </para>
-          <para>
-            When adding modes manually the driver creates each mode with a call to
-            <function>drm_mode_create</function> and must fill the following fields.
-            <itemizedlist>
-              <listitem>
-                <synopsis>__u32 type;</synopsis>
-                <para>
-                  Mode type bitmask, a combination of
-                  <variablelist>
-                    <varlistentry>
-                      <term>DRM_MODE_TYPE_BUILTIN</term>
-                      <listitem><para>not used?</para></listitem>
-                    </varlistentry>
-                    <varlistentry>
-                      <term>DRM_MODE_TYPE_CLOCK_C</term>
-                      <listitem><para>not used?</para></listitem>
-                    </varlistentry>
-                    <varlistentry>
-                      <term>DRM_MODE_TYPE_CRTC_C</term>
-                      <listitem><para>not used?</para></listitem>
-                    </varlistentry>
-                    <varlistentry>
-                      <term>
-        DRM_MODE_TYPE_PREFERRED - The preferred mode for the connector
-                      </term>
-                      <listitem>
-                        <para>not used?</para>
-                      </listitem>
-                    </varlistentry>
-                    <varlistentry>
-                      <term>DRM_MODE_TYPE_DEFAULT</term>
-                      <listitem><para>not used?</para></listitem>
-                    </varlistentry>
-                    <varlistentry>
-                      <term>DRM_MODE_TYPE_USERDEF</term>
-                      <listitem><para>not used?</para></listitem>
-                    </varlistentry>
-                    <varlistentry>
-                      <term>DRM_MODE_TYPE_DRIVER</term>
-                      <listitem>
-                        <para>
-                          The mode has been created by the driver (as opposed to
-                          to user-created modes).
-                        </para>
-                      </listitem>
-                    </varlistentry>
-                  </variablelist>
-                  Drivers must set the DRM_MODE_TYPE_DRIVER bit for all modes they
-                  create, and set the DRM_MODE_TYPE_PREFERRED bit for the preferred
-                  mode.
-                </para>
-              </listitem>
-              <listitem>
-                <synopsis>__u32 clock;</synopsis>
-                <para>Pixel clock frequency in kHz unit</para>
-              </listitem>
-              <listitem>
-                <synopsis>__u16 hdisplay, hsync_start, hsync_end, htotal;
-    __u16 vdisplay, vsync_start, vsync_end, vtotal;</synopsis>
-                <para>Horizontal and vertical timing information</para>
-                <screen><![CDATA[
-             Active                 Front           Sync           Back
-             Region                 Porch                          Porch
-    <-----------------------><----------------><-------------><-------------->
-
-      //////////////////////|
-     ////////////////////// |
-    //////////////////////  |..................               ................
-                                               _______________
-
-    <----- [hv]display ----->
-    <------------- [hv]sync_start ------------>
-    <--------------------- [hv]sync_end --------------------->
-    <-------------------------------- [hv]total ----------------------------->
-]]></screen>
-              </listitem>
-              <listitem>
-                <synopsis>__u16 hskew;
-    __u16 vscan;</synopsis>
-                <para>Unknown</para>
-              </listitem>
-              <listitem>
-                <synopsis>__u32 flags;</synopsis>
-                <para>
-                  Mode flags, a combination of
-                  <variablelist>
-                    <varlistentry>
-                      <term>DRM_MODE_FLAG_PHSYNC</term>
-                      <listitem><para>
-                        Horizontal sync is active high
-                      </para></listitem>
-                    </varlistentry>
-                    <varlistentry>
-                      <term>DRM_MODE_FLAG_NHSYNC</term>
-                      <listitem><para>
-                        Horizontal sync is active low
-                      </para></listitem>
-                    </varlistentry>
-                    <varlistentry>
-                      <term>DRM_MODE_FLAG_PVSYNC</term>
-                      <listitem><para>
-                        Vertical sync is active high
-                      </para></listitem>
-                    </varlistentry>
-                    <varlistentry>
-                      <term>DRM_MODE_FLAG_NVSYNC</term>
-                      <listitem><para>
-                        Vertical sync is active low
-                      </para></listitem>
-                    </varlistentry>
-                    <varlistentry>
-                      <term>DRM_MODE_FLAG_INTERLACE</term>
-                      <listitem><para>
-                        Mode is interlaced
-                      </para></listitem>
-                    </varlistentry>
-                    <varlistentry>
-                      <term>DRM_MODE_FLAG_DBLSCAN</term>
-                      <listitem><para>
-                        Mode uses doublescan
-                      </para></listitem>
-                    </varlistentry>
-                    <varlistentry>
-                      <term>DRM_MODE_FLAG_CSYNC</term>
-                      <listitem><para>
-                        Mode uses composite sync
-                      </para></listitem>
-                    </varlistentry>
-                    <varlistentry>
-                      <term>DRM_MODE_FLAG_PCSYNC</term>
-                      <listitem><para>
-                        Composite sync is active high
-                      </para></listitem>
-                    </varlistentry>
-                    <varlistentry>
-                      <term>DRM_MODE_FLAG_NCSYNC</term>
-                      <listitem><para>
-                        Composite sync is active low
-                      </para></listitem>
-                    </varlistentry>
-                    <varlistentry>
-                      <term>DRM_MODE_FLAG_HSKEW</term>
-                      <listitem><para>
-                        hskew provided (not used?)
-                      </para></listitem>
-                    </varlistentry>
-                    <varlistentry>
-                      <term>DRM_MODE_FLAG_BCAST</term>
-                      <listitem><para>
-                        not used?
-                      </para></listitem>
-                    </varlistentry>
-                    <varlistentry>
-                      <term>DRM_MODE_FLAG_PIXMUX</term>
-                      <listitem><para>
-                        not used?
-                      </para></listitem>
-                    </varlistentry>
-                    <varlistentry>
-                      <term>DRM_MODE_FLAG_DBLCLK</term>
-                      <listitem><para>
-                        not used?
-                      </para></listitem>
-                    </varlistentry>
-                    <varlistentry>
-                      <term>DRM_MODE_FLAG_CLKDIV2</term>
-                      <listitem><para>
-                        ?
-                      </para></listitem>
-                    </varlistentry>
-                  </variablelist>
-                </para>
-                <para>
-                  Note that modes marked with the INTERLACE or DBLSCAN flags will be
-                  filtered out by
-                  <function>drm_helper_probe_single_connector_modes</function> if
-                  the connector's <structfield>interlace_allowed</structfield> or
-                  <structfield>doublescan_allowed</structfield> field is set to 0.
-                </para>
-              </listitem>
-              <listitem>
-                <synopsis>char name[DRM_DISPLAY_MODE_LEN];</synopsis>
-                <para>
-                  Mode name. The driver must call
-                  <function>drm_mode_set_name</function> to fill the mode name from
-                  <structfield>hdisplay</structfield>,
-                  <structfield>vdisplay</structfield> and interlace flag after
-                  filling the corresponding fields.
-                </para>
-              </listitem>
-            </itemizedlist>
-          </para>
-          <para>
-            The <structfield>vrefresh</structfield> value is computed by
-            <function>drm_helper_probe_single_connector_modes</function>.
-          </para>
-          <para>
-            When parsing EDID data, <function>drm_add_edid_modes</function> fills the
-            connector <structfield>display_info</structfield>
-            <structfield>width_mm</structfield> and
-            <structfield>height_mm</structfield> fields. When creating modes
-            manually the <methodname>get_modes</methodname> helper operation must
-            set the <structfield>display_info</structfield>
-            <structfield>width_mm</structfield> and
-            <structfield>height_mm</structfield> fields if they haven't been set
-            already (for instance at initialization time when a fixed-size panel is
-            attached to the connector). The mode <structfield>width_mm</structfield>
-            and <structfield>height_mm</structfield> fields are only used internally
-            during EDID parsing and should not be set when creating modes manually.
-          </para>
-        </listitem>
-        <listitem>
-          <synopsis>int (*mode_valid)(struct drm_connector *connector,
-		  struct drm_display_mode *mode);</synopsis>
-          <para>
-            Verify whether a mode is valid for the connector. Return MODE_OK for
-            supported modes and one of the enum drm_mode_status values (MODE_*)
-            for unsupported modes. This operation is optional.
-          </para>
-          <para>
-            As the mode rejection reason is currently not used beside for
-            immediately removing the unsupported mode, an implementation can
-            return MODE_BAD regardless of the exact reason why the mode is not
-            valid.
-          </para>
-          <note><para>
-            Note that the <methodname>mode_valid</methodname> helper operation is
-            only called for modes detected by the device, and
-            <emphasis>not</emphasis> for modes set by the user through the CRTC
-            <methodname>set_config</methodname> operation.
-          </para></note>
-        </listitem>
-      </itemizedlist>
-    </sect2>
-    <sect2>
-      <title>Atomic Modeset Helper Functions Reference</title>
-      <sect3>
-	<title>Overview</title>
-!Pdrivers/gpu/drm/drm_atomic_helper.c overview
-      </sect3>
-      <sect3>
-	<title>Implementing Asynchronous Atomic Commit</title>
-!Pdrivers/gpu/drm/drm_atomic_helper.c implementing async commit
-      </sect3>
-      <sect3>
-	<title>Atomic State Reset and Initialization</title>
-!Pdrivers/gpu/drm/drm_atomic_helper.c atomic state reset and initialization
-      </sect3>
-!Iinclude/drm/drm_atomic_helper.h
-!Edrivers/gpu/drm/drm_atomic_helper.c
-    </sect2>
-    <sect2>
-      <title>Modeset Helper Functions Reference</title>
-!Iinclude/drm/drm_crtc_helper.h
-!Edrivers/gpu/drm/drm_crtc_helper.c
-!Pdrivers/gpu/drm/drm_crtc_helper.c overview
-    </sect2>
-    <sect2>
-      <title>Output Probing Helper Functions Reference</title>
-!Pdrivers/gpu/drm/drm_probe_helper.c output probing helper overview
-!Edrivers/gpu/drm/drm_probe_helper.c
-    </sect2>
-    <sect2>
-      <title>fbdev Helper Functions Reference</title>
-!Pdrivers/gpu/drm/drm_fb_helper.c fbdev helpers
-!Edrivers/gpu/drm/drm_fb_helper.c
-!Iinclude/drm/drm_fb_helper.h
-    </sect2>
-    <sect2>
-      <title>Display Port Helper Functions Reference</title>
-!Pdrivers/gpu/drm/drm_dp_helper.c dp helpers
-!Iinclude/drm/drm_dp_helper.h
-!Edrivers/gpu/drm/drm_dp_helper.c
-    </sect2>
-    <sect2>
-      <title>Display Port MST Helper Functions Reference</title>
-!Pdrivers/gpu/drm/drm_dp_mst_topology.c dp mst helper
-!Iinclude/drm/drm_dp_mst_helper.h
-!Edrivers/gpu/drm/drm_dp_mst_topology.c
-    </sect2>
-    <sect2>
-      <title>MIPI DSI Helper Functions Reference</title>
-!Pdrivers/gpu/drm/drm_mipi_dsi.c dsi helpers
-!Iinclude/drm/drm_mipi_dsi.h
-!Edrivers/gpu/drm/drm_mipi_dsi.c
-    </sect2>
-    <sect2>
-      <title>EDID Helper Functions Reference</title>
-!Edrivers/gpu/drm/drm_edid.c
-    </sect2>
-    <sect2>
-      <title>Rectangle Utilities Reference</title>
-!Pinclude/drm/drm_rect.h rect utils
-!Iinclude/drm/drm_rect.h
-!Edrivers/gpu/drm/drm_rect.c
-    </sect2>
-    <sect2>
-      <title>Flip-work Helper Reference</title>
-!Pinclude/drm/drm_flip_work.h flip utils
-!Iinclude/drm/drm_flip_work.h
-!Edrivers/gpu/drm/drm_flip_work.c
-    </sect2>
-    <sect2>
-      <title>HDMI Infoframes Helper Reference</title>
-      <para>
-	Strictly speaking this is not a DRM helper library but generally useable
-	by any driver interfacing with HDMI outputs like v4l or alsa drivers.
-	But it nicely fits into the overall topic of mode setting helper
-	libraries and hence is also included here.
-      </para>
-!Iinclude/linux/hdmi.h
-!Edrivers/video/hdmi.c
-    </sect2>
-    <sect2>
-      <title id="drm-kms-planehelpers">Plane Helper Reference</title>
-!Edrivers/gpu/drm/drm_plane_helper.c
-!Pdrivers/gpu/drm/drm_plane_helper.c overview
-    </sect2>
-    <sect2>
-	  <title>Tile group</title>
-!Pdrivers/gpu/drm/drm_crtc.c Tile group
-    </sect2>
-    <sect2>
-	<title>Bridges</title>
-      <sect3>
-	 <title>Overview</title>
-!Pdrivers/gpu/drm/drm_bridge.c overview
-      </sect3>
-      <sect3>
-	 <title>Default bridge callback sequence</title>
-!Pdrivers/gpu/drm/drm_bridge.c bridge callbacks
-      </sect3>
-!Edrivers/gpu/drm/drm_bridge.c
-    </sect2>
-  </sect1>
-
-  <!-- Internals: kms properties -->
-
-  <sect1 id="drm-kms-properties">
-    <title>KMS Properties</title>
-    <para>
-      Drivers may need to expose additional parameters to applications than
-      those described in the previous sections. KMS supports attaching
-      properties to CRTCs, connectors and planes and offers a userspace API to
-      list, get and set the property values.
-    </para>
-    <para>
-      Properties are identified by a name that uniquely defines the property
-      purpose, and store an associated value. For all property types except blob
-      properties the value is a 64-bit unsigned integer.
-    </para>
-    <para>
-      KMS differentiates between properties and property instances. Drivers
-      first create properties and then create and associate individual instances
-      of those properties to objects. A property can be instantiated multiple
-      times and associated with different objects. Values are stored in property
-      instances, and all other property information are stored in the property
-      and shared between all instances of the property.
-    </para>
-    <para>
-      Every property is created with a type that influences how the KMS core
-      handles the property. Supported property types are
-      <variablelist>
-        <varlistentry>
-          <term>DRM_MODE_PROP_RANGE</term>
-          <listitem><para>Range properties report their minimum and maximum
-            admissible values. The KMS core verifies that values set by
-            application fit in that range.</para></listitem>
-        </varlistentry>
-        <varlistentry>
-          <term>DRM_MODE_PROP_ENUM</term>
-          <listitem><para>Enumerated properties take a numerical value that
-            ranges from 0 to the number of enumerated values defined by the
-            property minus one, and associate a free-formed string name to each
-            value. Applications can retrieve the list of defined value-name pairs
-            and use the numerical value to get and set property instance values.
-            </para></listitem>
-        </varlistentry>
-        <varlistentry>
-          <term>DRM_MODE_PROP_BITMASK</term>
-          <listitem><para>Bitmask properties are enumeration properties that
-            additionally restrict all enumerated values to the 0..63 range.
-            Bitmask property instance values combine one or more of the
-            enumerated bits defined by the property.</para></listitem>
-        </varlistentry>
-        <varlistentry>
-          <term>DRM_MODE_PROP_BLOB</term>
-          <listitem><para>Blob properties store a binary blob without any format
-            restriction. The binary blobs are created as KMS standalone objects,
-            and blob property instance values store the ID of their associated
-            blob object.</para>
-	    <para>Blob properties are only used for the connector EDID property
-	    and cannot be created by drivers.</para></listitem>
-        </varlistentry>
-      </variablelist>
-    </para>
-    <para>
-      To create a property drivers call one of the following functions depending
-      on the property type. All property creation functions take property flags
-      and name, as well as type-specific arguments.
-      <itemizedlist>
-        <listitem>
-          <synopsis>struct drm_property *drm_property_create_range(struct drm_device *dev, int flags,
-                                               const char *name,
-                                               uint64_t min, uint64_t max);</synopsis>
-          <para>Create a range property with the given minimum and maximum
-            values.</para>
-        </listitem>
-        <listitem>
-          <synopsis>struct drm_property *drm_property_create_enum(struct drm_device *dev, int flags,
-                                              const char *name,
-                                              const struct drm_prop_enum_list *props,
-                                              int num_values);</synopsis>
-          <para>Create an enumerated property. The <parameter>props</parameter>
-            argument points to an array of <parameter>num_values</parameter>
-            value-name pairs.</para>
-        </listitem>
-        <listitem>
-          <synopsis>struct drm_property *drm_property_create_bitmask(struct drm_device *dev,
-                                                 int flags, const char *name,
-                                                 const struct drm_prop_enum_list *props,
-                                                 int num_values);</synopsis>
-          <para>Create a bitmask property. The <parameter>props</parameter>
-            argument points to an array of <parameter>num_values</parameter>
-            value-name pairs.</para>
-        </listitem>
-      </itemizedlist>
-    </para>
-    <para>
-      Properties can additionally be created as immutable, in which case they
-      will be read-only for applications but can be modified by the driver. To
-      create an immutable property drivers must set the DRM_MODE_PROP_IMMUTABLE
-      flag at property creation time.
-    </para>
-    <para>
-      When no array of value-name pairs is readily available at property
-      creation time for enumerated or range properties, drivers can create
-      the property using the <function>drm_property_create</function> function
-      and manually add enumeration value-name pairs by calling the
-      <function>drm_property_add_enum</function> function. Care must be taken to
-      properly specify the property type through the <parameter>flags</parameter>
-      argument.
-    </para>
-    <para>
-      After creating properties drivers can attach property instances to CRTC,
-      connector and plane objects by calling the
-      <function>drm_object_attach_property</function>. The function takes a
-      pointer to the target object, a pointer to the previously created property
-      and an initial instance value.
-    </para>
-    <sect2>
-	<title>Existing KMS Properties</title>
-	<para>
-	The following table gives description of drm properties exposed by various
-	modules/drivers.
-	</para>
-	<table border="1" cellpadding="0" cellspacing="0">
-	<tbody>
-	<tr style="font-weight: bold;">
-	<td valign="top" >Owner Module/Drivers</td>
-	<td valign="top" >Group</td>
-	<td valign="top" >Property Name</td>
-	<td valign="top" >Type</td>
-	<td valign="top" >Property Values</td>
-	<td valign="top" >Object attached</td>
-	<td valign="top" >Description/Restrictions</td>
-	</tr>
-	<tr>
-	<td rowspan="37" valign="top" >DRM</td>
-	<td valign="top" >Generic</td>
-	<td valign="top" >“rotation”</td>
-	<td valign="top" >BITMASK</td>
-	<td valign="top" >{ 0, "rotate-0" },
-	{ 1, "rotate-90" },
-	{ 2, "rotate-180" },
-	{ 3, "rotate-270" },
-	{ 4, "reflect-x" },
-	{ 5, "reflect-y" }</td>
-	<td valign="top" >CRTC, Plane</td>
-	<td valign="top" >rotate-(degrees) rotates the image by the specified amount in degrees
-	in counter clockwise direction. reflect-x and reflect-y reflects the
-	image along the specified axis prior to rotation</td>
-	</tr>
-	<tr>
-	<td rowspan="5" valign="top" >Connector</td>
-	<td valign="top" >“EDID”</td>
-	<td valign="top" >BLOB | IMMUTABLE</td>
-	<td valign="top" >0</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >Contains id of edid blob ptr object.</td>
-	</tr>
-	<tr>
-	<td valign="top" >“DPMS”</td>
-	<td valign="top" >ENUM</td>
-	<td valign="top" >{ “On”, “Standby”, “Suspend”, “Off” }</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >Contains DPMS operation mode value.</td>
-	</tr>
-	<tr>
-	<td valign="top" >“PATH”</td>
-	<td valign="top" >BLOB | IMMUTABLE</td>
-	<td valign="top" >0</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >Contains topology path to a connector.</td>
-	</tr>
-	<tr>
-	<td valign="top" >“TILE”</td>
-	<td valign="top" >BLOB | IMMUTABLE</td>
-	<td valign="top" >0</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >Contains tiling information for a connector.</td>
-	</tr>
-	<tr>
-	<td valign="top" >“CRTC_ID”</td>
-	<td valign="top" >OBJECT</td>
-	<td valign="top" >DRM_MODE_OBJECT_CRTC</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >CRTC that connector is attached to (atomic)</td>
-	</tr>
-	<tr>
-	<td rowspan="11" valign="top" >Plane</td>
-	<td valign="top" >“type”</td>
-	<td valign="top" >ENUM | IMMUTABLE</td>
-	<td valign="top" >{ "Overlay", "Primary", "Cursor" }</td>
-	<td valign="top" >Plane</td>
-	<td valign="top" >Plane type</td>
-	</tr>
-	<tr>
-	<td valign="top" >“SRC_X”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=UINT_MAX</td>
-	<td valign="top" >Plane</td>
-	<td valign="top" >Scanout source x coordinate in 16.16 fixed point (atomic)</td>
-	</tr>
-	<tr>
-	<td valign="top" >“SRC_Y”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=UINT_MAX</td>
-	<td valign="top" >Plane</td>
-	<td valign="top" >Scanout source y coordinate in 16.16 fixed point (atomic)</td>
-	</tr>
-	<tr>
-	<td valign="top" >“SRC_W”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=UINT_MAX</td>
-	<td valign="top" >Plane</td>
-	<td valign="top" >Scanout source width in 16.16 fixed point (atomic)</td>
-	</tr>
-	<tr>
-	<td valign="top" >“SRC_H”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=UINT_MAX</td>
-	<td valign="top" >Plane</td>
-	<td valign="top" >Scanout source height in 16.16 fixed point (atomic)</td>
-	</tr>
-	<tr>
-	<td valign="top" >“CRTC_X”</td>
-	<td valign="top" >SIGNED_RANGE</td>
-	<td valign="top" >Min=INT_MIN, Max=INT_MAX</td>
-	<td valign="top" >Plane</td>
-	<td valign="top" >Scanout CRTC (destination) x coordinate (atomic)</td>
-	</tr>
-	<tr>
-	<td valign="top" >“CRTC_Y”</td>
-	<td valign="top" >SIGNED_RANGE</td>
-	<td valign="top" >Min=INT_MIN, Max=INT_MAX</td>
-	<td valign="top" >Plane</td>
-	<td valign="top" >Scanout CRTC (destination) y coordinate (atomic)</td>
-	</tr>
-	<tr>
-	<td valign="top" >“CRTC_W”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=UINT_MAX</td>
-	<td valign="top" >Plane</td>
-	<td valign="top" >Scanout CRTC (destination) width (atomic)</td>
-	</tr>
-	<tr>
-	<td valign="top" >“CRTC_H”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=UINT_MAX</td>
-	<td valign="top" >Plane</td>
-	<td valign="top" >Scanout CRTC (destination) height (atomic)</td>
-	</tr>
-	<tr>
-	<td valign="top" >“FB_ID”</td>
-	<td valign="top" >OBJECT</td>
-	<td valign="top" >DRM_MODE_OBJECT_FB</td>
-	<td valign="top" >Plane</td>
-	<td valign="top" >Scanout framebuffer (atomic)</td>
-	</tr>
-	<tr>
-	<td valign="top" >“CRTC_ID”</td>
-	<td valign="top" >OBJECT</td>
-	<td valign="top" >DRM_MODE_OBJECT_CRTC</td>
-	<td valign="top" >Plane</td>
-	<td valign="top" >CRTC that plane is attached to (atomic)</td>
-	</tr>
-	<tr>
-	<td rowspan="2" valign="top" >DVI-I</td>
-	<td valign="top" >“subconnector”</td>
-	<td valign="top" >ENUM</td>
-	<td valign="top" >{ “Unknown”, “DVI-D”, “DVI-A” }</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“select subconnector”</td>
-	<td valign="top" >ENUM</td>
-	<td valign="top" >{ “Automatic”, “DVI-D”, “DVI-A” }</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td rowspan="13" valign="top" >TV</td>
-	<td valign="top" >“subconnector”</td>
-	<td valign="top" >ENUM</td>
-	<td valign="top" >{ "Unknown", "Composite", "SVIDEO", "Component", "SCART" }</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“select subconnector”</td>
-	<td valign="top" >ENUM</td>
-	<td valign="top" >{ "Automatic", "Composite", "SVIDEO", "Component", "SCART" }</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“mode”</td>
-	<td valign="top" >ENUM</td>
-	<td valign="top" >{ "NTSC_M", "NTSC_J", "NTSC_443", "PAL_B" } etc.</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“left margin”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=100</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“right margin”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=100</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“top margin”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=100</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“bottom margin”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=100</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“brightness”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=100</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“contrast”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=100</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“flicker reduction”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=100</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“overscan”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=100</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“saturation”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=100</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“hue”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=100</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td rowspan="2" valign="top" >Virtual GPU</td>
-	<td valign="top" >“suggested X”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=0xffffffff</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >property to suggest an X offset for a connector</td>
-	</tr>
-	<tr>
-	<td valign="top" >“suggested Y”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=0xffffffff</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >property to suggest an Y offset for a connector</td>
-	</tr>
-	<tr>
-	<td rowspan="3" valign="top" >Optional</td>
-	<td valign="top" >“scaling mode”</td>
-	<td valign="top" >ENUM</td>
-	<td valign="top" >{ "None", "Full", "Center", "Full aspect" }</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >"aspect ratio"</td>
-	<td valign="top" >ENUM</td>
-	<td valign="top" >{ "None", "4:3", "16:9" }</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >DRM property to set aspect ratio from user space app.
-		This enum is made generic to allow addition of custom aspect
-		ratios.</td>
-	</tr>
-	<tr>
-	<td valign="top" >“dirty”</td>
-	<td valign="top" >ENUM | IMMUTABLE</td>
-	<td valign="top" >{ "Off", "On", "Annotate" }</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td rowspan="20" valign="top" >i915</td>
-	<td rowspan="2" valign="top" >Generic</td>
-	<td valign="top" >"Broadcast RGB"</td>
-	<td valign="top" >ENUM</td>
-	<td valign="top" >{ "Automatic", "Full", "Limited 16:235" }</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“audio”</td>
-	<td valign="top" >ENUM</td>
-	<td valign="top" >{ "force-dvi", "off", "auto", "on" }</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td rowspan="17" valign="top" >SDVO-TV</td>
-	<td valign="top" >“mode”</td>
-	<td valign="top" >ENUM</td>
-	<td valign="top" >{ "NTSC_M", "NTSC_J", "NTSC_443", "PAL_B" } etc.</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >"left_margin"</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max= SDVO dependent</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >"right_margin"</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max= SDVO dependent</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >"top_margin"</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max= SDVO dependent</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >"bottom_margin"</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max= SDVO dependent</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“hpos”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max= SDVO dependent</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“vpos”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max= SDVO dependent</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“contrast”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max= SDVO dependent</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“saturation”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max= SDVO dependent</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“hue”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max= SDVO dependent</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“sharpness”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max= SDVO dependent</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“flicker_filter”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max= SDVO dependent</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“flicker_filter_adaptive”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max= SDVO dependent</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“flicker_filter_2d”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max= SDVO dependent</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“tv_chroma_filter”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max= SDVO dependent</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“tv_luma_filter”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max= SDVO dependent</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“dot_crawl”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=1</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >SDVO-TV/LVDS</td>
-	<td valign="top" >“brightness”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max= SDVO dependent</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td rowspan="2" valign="top" >CDV gma-500</td>
-	<td rowspan="2" valign="top" >Generic</td>
-	<td valign="top" >"Broadcast RGB"</td>
-	<td valign="top" >ENUM</td>
-	<td valign="top" >{ “Full”, “Limited 16:235” }</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >"Broadcast RGB"</td>
-	<td valign="top" >ENUM</td>
-	<td valign="top" >{ “off”, “auto”, “on” }</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td rowspan="19" valign="top" >Poulsbo</td>
-	<td rowspan="1" valign="top" >Generic</td>
-	<td valign="top" >“backlight”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=100</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td rowspan="17" valign="top" >SDVO-TV</td>
-	<td valign="top" >“mode”</td>
-	<td valign="top" >ENUM</td>
-	<td valign="top" >{ "NTSC_M", "NTSC_J", "NTSC_443", "PAL_B" } etc.</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >"left_margin"</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max= SDVO dependent</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >"right_margin"</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max= SDVO dependent</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >"top_margin"</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max= SDVO dependent</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >"bottom_margin"</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max= SDVO dependent</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“hpos”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max= SDVO dependent</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“vpos”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max= SDVO dependent</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“contrast”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max= SDVO dependent</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“saturation”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max= SDVO dependent</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“hue”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max= SDVO dependent</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“sharpness”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max= SDVO dependent</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“flicker_filter”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max= SDVO dependent</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“flicker_filter_adaptive”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max= SDVO dependent</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“flicker_filter_2d”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max= SDVO dependent</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“tv_chroma_filter”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max= SDVO dependent</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“tv_luma_filter”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max= SDVO dependent</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“dot_crawl”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=1</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >SDVO-TV/LVDS</td>
-	<td valign="top" >“brightness”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max= SDVO dependent</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td rowspan="11" valign="top" >armada</td>
-	<td rowspan="2" valign="top" >CRTC</td>
-	<td valign="top" >"CSC_YUV"</td>
-	<td valign="top" >ENUM</td>
-	<td valign="top" >{ "Auto" , "CCIR601", "CCIR709" }</td>
-	<td valign="top" >CRTC</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >"CSC_RGB"</td>
-	<td valign="top" >ENUM</td>
-	<td valign="top" >{ "Auto", "Computer system", "Studio" }</td>
-	<td valign="top" >CRTC</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td rowspan="9" valign="top" >Overlay</td>
-	<td valign="top" >"colorkey"</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=0xffffff</td>
-	<td valign="top" >Plane</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >"colorkey_min"</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=0xffffff</td>
-	<td valign="top" >Plane</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >"colorkey_max"</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=0xffffff</td>
-	<td valign="top" >Plane</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >"colorkey_val"</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=0xffffff</td>
-	<td valign="top" >Plane</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >"colorkey_alpha"</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=0xffffff</td>
-	<td valign="top" >Plane</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >"colorkey_mode"</td>
-	<td valign="top" >ENUM</td>
-	<td valign="top" >{ "disabled", "Y component", "U component"
-	, "V component", "RGB", “R component", "G component", "B component" }</td>
-	<td valign="top" >Plane</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >"brightness"</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=256 + 255</td>
-	<td valign="top" >Plane</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >"contrast"</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=0x7fff</td>
-	<td valign="top" >Plane</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >"saturation"</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=0x7fff</td>
-	<td valign="top" >Plane</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td rowspan="2" valign="top" >exynos</td>
-	<td valign="top" >CRTC</td>
-	<td valign="top" >“mode”</td>
-	<td valign="top" >ENUM</td>
-	<td valign="top" >{ "normal", "blank" }</td>
-	<td valign="top" >CRTC</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >Overlay</td>
-	<td valign="top" >“zpos”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=MAX_PLANE-1</td>
-	<td valign="top" >Plane</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td rowspan="2" valign="top" >i2c/ch7006_drv</td>
-	<td valign="top" >Generic</td>
-	<td valign="top" >“scale”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=2</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td rowspan="1" valign="top" >TV</td>
-	<td valign="top" >“mode”</td>
-	<td valign="top" >ENUM</td>
-	<td valign="top" >{ "PAL", "PAL-M","PAL-N"}, ”PAL-Nc"
-	, "PAL-60", "NTSC-M", "NTSC-J" }</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td rowspan="15" valign="top" >nouveau</td>
-	<td rowspan="6" valign="top" >NV10 Overlay</td>
-	<td valign="top" >"colorkey"</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=0x01ffffff</td>
-	<td valign="top" >Plane</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“contrast”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=8192-1</td>
-	<td valign="top" >Plane</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“brightness”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=1024</td>
-	<td valign="top" >Plane</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“hue”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=359</td>
-	<td valign="top" >Plane</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“saturation”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=8192-1</td>
-	<td valign="top" >Plane</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“iturbt_709”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=1</td>
-	<td valign="top" >Plane</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td rowspan="2" valign="top" >Nv04 Overlay</td>
-	<td valign="top" >“colorkey”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=0x01ffffff</td>
-	<td valign="top" >Plane</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“brightness”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=1024</td>
-	<td valign="top" >Plane</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td rowspan="7" valign="top" >Display</td>
-	<td valign="top" >“dithering mode”</td>
-	<td valign="top" >ENUM</td>
-	<td valign="top" >{ "auto", "off", "on" }</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“dithering depth”</td>
-	<td valign="top" >ENUM</td>
-	<td valign="top" >{ "auto", "off", "on", "static 2x2", "dynamic 2x2", "temporal" }</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“underscan”</td>
-	<td valign="top" >ENUM</td>
-	<td valign="top" >{ "auto", "6 bpc", "8 bpc" }</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“underscan hborder”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=128</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“underscan vborder”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=128</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“vibrant hue”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=180</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >“color vibrance”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=200</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >omap</td>
-	<td valign="top" >Generic</td>
-	<td valign="top" >“zorder”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=3</td>
-	<td valign="top" >CRTC, Plane</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >qxl</td>
-	<td valign="top" >Generic</td>
-	<td valign="top" >“hotplug_mode_update"</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=1</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td rowspan="9" valign="top" >radeon</td>
-	<td valign="top" >DVI-I</td>
-	<td valign="top" >“coherent”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=1</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >DAC enable load detect</td>
-	<td valign="top" >“load detection”</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=1</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >TV Standard</td>
-	<td valign="top" >"tv standard"</td>
-	<td valign="top" >ENUM</td>
-	<td valign="top" >{ "ntsc", "pal", "pal-m", "pal-60", "ntsc-j"
-	, "scart-pal", "pal-cn", "secam" }</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >legacy TMDS PLL detect</td>
-	<td valign="top" >"tmds_pll"</td>
-	<td valign="top" >ENUM</td>
-	<td valign="top" >{ "driver", "bios" }</td>
-	<td valign="top" >-</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td rowspan="3" valign="top" >Underscan</td>
-	<td valign="top" >"underscan"</td>
-	<td valign="top" >ENUM</td>
-	<td valign="top" >{ "off", "on", "auto" }</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >"underscan hborder"</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=128</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >"underscan vborder"</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=128</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >Audio</td>
-	<td valign="top" >“audio”</td>
-	<td valign="top" >ENUM</td>
-	<td valign="top" >{ "off", "on", "auto" }</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >FMT Dithering</td>
-	<td valign="top" >“dither”</td>
-	<td valign="top" >ENUM</td>
-	<td valign="top" >{ "off", "on" }</td>
-	<td valign="top" >Connector</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td rowspan="3" valign="top" >rcar-du</td>
-	<td rowspan="3" valign="top" >Generic</td>
-	<td valign="top" >"alpha"</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=255</td>
-	<td valign="top" >Plane</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >"colorkey"</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=0, Max=0x01ffffff</td>
-	<td valign="top" >Plane</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	<tr>
-	<td valign="top" >"zpos"</td>
-	<td valign="top" >RANGE</td>
-	<td valign="top" >Min=1, Max=7</td>
-	<td valign="top" >Plane</td>
-	<td valign="top" >TBD</td>
-	</tr>
-	</tbody>
-	</table>
-    </sect2>
-  </sect1>
-
-  <!-- Internals: vertical blanking -->
-
-  <sect1 id="drm-vertical-blank">
-    <title>Vertical Blanking</title>
-    <para>
-      Vertical blanking plays a major role in graphics rendering. To achieve
-      tear-free display, users must synchronize page flips and/or rendering to
-      vertical blanking. The DRM API offers ioctls to perform page flips
-      synchronized to vertical blanking and wait for vertical blanking.
-    </para>
-    <para>
-      The DRM core handles most of the vertical blanking management logic, which
-      involves filtering out spurious interrupts, keeping race-free blanking
-      counters, coping with counter wrap-around and resets and keeping use
-      counts. It relies on the driver to generate vertical blanking interrupts
-      and optionally provide a hardware vertical blanking counter. Drivers must
-      implement the following operations.
-    </para>
-    <itemizedlist>
-      <listitem>
-        <synopsis>int (*enable_vblank) (struct drm_device *dev, int crtc);
-void (*disable_vblank) (struct drm_device *dev, int crtc);</synopsis>
-        <para>
-	  Enable or disable vertical blanking interrupts for the given CRTC.
-	</para>
-      </listitem>
-      <listitem>
-        <synopsis>u32 (*get_vblank_counter) (struct drm_device *dev, int crtc);</synopsis>
-        <para>
-	  Retrieve the value of the vertical blanking counter for the given
-	  CRTC. If the hardware maintains a vertical blanking counter its value
-	  should be returned. Otherwise drivers can use the
-	  <function>drm_vblank_count</function> helper function to handle this
-	  operation.
-	</para>
-      </listitem>
-    </itemizedlist>
-    <para>
-      Drivers must initialize the vertical blanking handling core with a call to
-      <function>drm_vblank_init</function> in their
-      <methodname>load</methodname> operation. The function will set the struct
-      <structname>drm_device</structname>
-      <structfield>vblank_disable_allowed</structfield> field to 0. This will
-      keep vertical blanking interrupts enabled permanently until the first mode
-      set operation, where <structfield>vblank_disable_allowed</structfield> is
-      set to 1. The reason behind this is not clear. Drivers can set the field
-      to 1 after <function>calling drm_vblank_init</function> to make vertical
-      blanking interrupts dynamically managed from the beginning.
-    </para>
-    <para>
-      Vertical blanking interrupts can be enabled by the DRM core or by drivers
-      themselves (for instance to handle page flipping operations). The DRM core
-      maintains a vertical blanking use count to ensure that the interrupts are
-      not disabled while a user still needs them. To increment the use count,
-      drivers call <function>drm_vblank_get</function>. Upon return vertical
-      blanking interrupts are guaranteed to be enabled.
-    </para>
-    <para>
-      To decrement the use count drivers call
-      <function>drm_vblank_put</function>. Only when the use count drops to zero
-      will the DRM core disable the vertical blanking interrupts after a delay
-      by scheduling a timer. The delay is accessible through the vblankoffdelay
-      module parameter or the <varname>drm_vblank_offdelay</varname> global
-      variable and expressed in milliseconds. Its default value is 5000 ms.
-      Zero means never disable, and a negative value means disable immediately.
-      Drivers may override the behaviour by setting the
-      <structname>drm_device</structname>
-      <structfield>vblank_disable_immediate</structfield> flag, which when set
-      causes vblank interrupts to be disabled immediately regardless of the
-      drm_vblank_offdelay value. The flag should only be set if there's a
-      properly working hardware vblank counter present.
-    </para>
-    <para>
-      When a vertical blanking interrupt occurs drivers only need to call the
-      <function>drm_handle_vblank</function> function to account for the
-      interrupt.
-    </para>
-    <para>
-      Resources allocated by <function>drm_vblank_init</function> must be freed
-      with a call to <function>drm_vblank_cleanup</function> in the driver
-      <methodname>unload</methodname> operation handler.
-    </para>
-    <sect2>
-      <title>Vertical Blanking and Interrupt Handling Functions Reference</title>
-!Edrivers/gpu/drm/drm_irq.c
-!Finclude/drm/drmP.h drm_crtc_vblank_waitqueue
-    </sect2>
-  </sect1>
-
-  <!-- Internals: open/close, file operations and ioctls -->
-
-  <sect1>
-    <title>Open/Close, File Operations and IOCTLs</title>
-    <sect2>
-      <title>Open and Close</title>
-      <synopsis>int (*firstopen) (struct drm_device *);
-void (*lastclose) (struct drm_device *);
-int (*open) (struct drm_device *, struct drm_file *);
-void (*preclose) (struct drm_device *, struct drm_file *);
-void (*postclose) (struct drm_device *, struct drm_file *);</synopsis>
-      <abstract>Open and close handlers. None of those methods are mandatory.
-      </abstract>
-      <para>
-        The <methodname>firstopen</methodname> method is called by the DRM core
-	for legacy UMS (User Mode Setting) drivers only when an application
-	opens a device that has no other opened file handle. UMS drivers can
-	implement it to acquire device resources. KMS drivers can't use the
-	method and must acquire resources in the <methodname>load</methodname>
-	method instead.
-      </para>
-      <para>
-	Similarly the <methodname>lastclose</methodname> method is called when
-	the last application holding a file handle opened on the device closes
-	it, for both UMS and KMS drivers. Additionally, the method is also
-	called at module unload time or, for hot-pluggable devices, when the
-	device is unplugged. The <methodname>firstopen</methodname> and
-	<methodname>lastclose</methodname> calls can thus be unbalanced.
-      </para>
-      <para>
-        The <methodname>open</methodname> method is called every time the device
-	is opened by an application. Drivers can allocate per-file private data
-	in this method and store them in the struct
-	<structname>drm_file</structname> <structfield>driver_priv</structfield>
-	field. Note that the <methodname>open</methodname> method is called
-	before <methodname>firstopen</methodname>.
-      </para>
-      <para>
-        The close operation is split into <methodname>preclose</methodname> and
-	<methodname>postclose</methodname> methods. Drivers must stop and
-	cleanup all per-file operations in the <methodname>preclose</methodname>
-	method. For instance pending vertical blanking and page flip events must
-	be cancelled. No per-file operation is allowed on the file handle after
-	returning from the <methodname>preclose</methodname> method.
-      </para>
-      <para>
-        Finally the <methodname>postclose</methodname> method is called as the
-	last step of the close operation, right before calling the
-	<methodname>lastclose</methodname> method if no other open file handle
-	exists for the device. Drivers that have allocated per-file private data
-	in the <methodname>open</methodname> method should free it here.
-      </para>
-      <para>
-        The <methodname>lastclose</methodname> method should restore CRTC and
-	plane properties to default value, so that a subsequent open of the
-	device will not inherit state from the previous user. It can also be
-	used to execute delayed power switching state changes, e.g. in
-	conjunction with the vga_switcheroo infrastructure. Beyond that KMS
-	drivers should not do any further cleanup. Only legacy UMS drivers might
-	need to clean up device state so that the vga console or an independent
-	fbdev driver could take over.
-      </para>
-    </sect2>
-    <sect2>
-      <title>File Operations</title>
-      <synopsis>const struct file_operations *fops</synopsis>
-      <abstract>File operations for the DRM device node.</abstract>
-      <para>
-        Drivers must define the file operations structure that forms the DRM
-	userspace API entry point, even though most of those operations are
-	implemented in the DRM core. The <methodname>open</methodname>,
-	<methodname>release</methodname> and <methodname>ioctl</methodname>
-	operations are handled by
-	<programlisting>
-	.owner = THIS_MODULE,
-	.open = drm_open,
-	.release = drm_release,
-	.unlocked_ioctl = drm_ioctl,
-  #ifdef CONFIG_COMPAT
-	.compat_ioctl = drm_compat_ioctl,
-  #endif
-        </programlisting>
-      </para>
-      <para>
-        Drivers that implement private ioctls that requires 32/64bit
-	compatibility support must provide their own
-	<methodname>compat_ioctl</methodname> handler that processes private
-	ioctls and calls <function>drm_compat_ioctl</function> for core ioctls.
-      </para>
-      <para>
-        The <methodname>read</methodname> and <methodname>poll</methodname>
-	operations provide support for reading DRM events and polling them. They
-	are implemented by
-	<programlisting>
-	.poll = drm_poll,
-	.read = drm_read,
-	.llseek = no_llseek,
-	</programlisting>
-      </para>
-      <para>
-        The memory mapping implementation varies depending on how the driver
-	manages memory. Pre-GEM drivers will use <function>drm_mmap</function>,
-	while GEM-aware drivers will use <function>drm_gem_mmap</function>. See
-	<xref linkend="drm-gem"/>.
-	<programlisting>
-	.mmap = drm_gem_mmap,
-	</programlisting>
-      </para>
-      <para>
-        No other file operation is supported by the DRM API.
-      </para>
-    </sect2>
-    <sect2>
-      <title>IOCTLs</title>
-      <synopsis>struct drm_ioctl_desc *ioctls;
-int num_ioctls;</synopsis>
-      <abstract>Driver-specific ioctls descriptors table.</abstract>
-      <para>
-        Driver-specific ioctls numbers start at DRM_COMMAND_BASE. The ioctls
-	descriptors table is indexed by the ioctl number offset from the base
-	value. Drivers can use the DRM_IOCTL_DEF_DRV() macro to initialize the
-	table entries.
-      </para>
-      <para>
-        <programlisting>DRM_IOCTL_DEF_DRV(ioctl, func, flags)</programlisting>
-	<para>
-	  <parameter>ioctl</parameter> is the ioctl name. Drivers must define
-	  the DRM_##ioctl and DRM_IOCTL_##ioctl macros to the ioctl number
-	  offset from DRM_COMMAND_BASE and the ioctl number respectively. The
-	  first macro is private to the device while the second must be exposed
-	  to userspace in a public header.
-	</para>
-	<para>
-	  <parameter>func</parameter> is a pointer to the ioctl handler function
-	  compatible with the <type>drm_ioctl_t</type> type.
-	  <programlisting>typedef int drm_ioctl_t(struct drm_device *dev, void *data,
-		struct drm_file *file_priv);</programlisting>
-	</para>
-	<para>
-	  <parameter>flags</parameter> is a bitmask combination of the following
-	  values. It restricts how the ioctl is allowed to be called.
-	  <itemizedlist>
-	    <listitem><para>
-	      DRM_AUTH - Only authenticated callers allowed
-	    </para></listitem>
-	    <listitem><para>
-	      DRM_MASTER - The ioctl can only be called on the master file
-	      handle
-	    </para></listitem>
-            <listitem><para>
-	      DRM_ROOT_ONLY - Only callers with the SYSADMIN capability allowed
-	    </para></listitem>
-            <listitem><para>
-	      DRM_CONTROL_ALLOW - The ioctl can only be called on a control
-	      device
-	    </para></listitem>
-            <listitem><para>
-	      DRM_UNLOCKED - The ioctl handler will be called without locking
-	      the DRM global mutex. This is the enforced default for kms drivers
-	      (i.e. using the DRIVER_MODESET flag) and hence shouldn't be used
-	      any more for new drivers.
-	    </para></listitem>
-	  </itemizedlist>
-	</para>
-      </para>
-!Edrivers/gpu/drm/drm_ioctl.c
-    </sect2>
-  </sect1>
-  <sect1>
-    <title>Legacy Support Code</title>
-    <para>
-      The section very briefly covers some of the old legacy support code which
-      is only used by old DRM drivers which have done a so-called shadow-attach
-      to the underlying device instead of registering as a real driver. This
-      also includes some of the old generic buffer management and command
-      submission code. Do not use any of this in new and modern drivers.
-    </para>
-
-    <sect2>
-      <title>Legacy Suspend/Resume</title>
-      <para>
-	The DRM core provides some suspend/resume code, but drivers wanting full
-	suspend/resume support should provide save() and restore() functions.
-	These are called at suspend, hibernate, or resume time, and should perform
-	any state save or restore required by your device across suspend or
-	hibernate states.
-      </para>
-      <synopsis>int (*suspend) (struct drm_device *, pm_message_t state);
-  int (*resume) (struct drm_device *);</synopsis>
-      <para>
-	Those are legacy suspend and resume methods which
-	<emphasis>only</emphasis> work with the legacy shadow-attach driver
-	registration functions. New driver should use the power management
-	interface provided by their bus type (usually through
-	the struct <structname>device_driver</structname> dev_pm_ops) and set
-	these methods to NULL.
-      </para>
-    </sect2>
-
-    <sect2>
-      <title>Legacy DMA Services</title>
-      <para>
-	This should cover how DMA mapping etc. is supported by the core.
-	These functions are deprecated and should not be used.
-      </para>
-    </sect2>
-  </sect1>
-  </chapter>
-
-<!-- TODO
-
-- Add a glossary
-- Document the struct_mutex catch-all lock
-- Document connector properties
-
-- Why is the load method optional?
-- What are drivers supposed to set the initial display state to, and how?
-  Connector's DPMS states are not initialized and are thus equal to
-  DRM_MODE_DPMS_ON. The fbcon compatibility layer calls
-  drm_helper_disable_unused_functions(), which disables unused encoders and
-  CRTCs, but doesn't touch the connectors' DPMS state, and
-  drm_helper_connector_dpms() in reaction to fbdev blanking events. Do drivers
-  that don't implement (or just don't use) fbcon compatibility need to call
-  those functions themselves?
-- KMS drivers must call drm_vblank_pre_modeset() and drm_vblank_post_modeset()
-  around mode setting. Should this be done in the DRM core?
-- vblank_disable_allowed is set to 1 in the first drm_vblank_post_modeset()
-  call and never set back to 0. It seems to be safe to permanently set it to 1
-  in drm_vblank_init() for KMS driver, and it might be safe for UMS drivers as
-  well. This should be investigated.
-- crtc and connector .save and .restore operations are only used internally in
-  drivers, should they be removed from the core?
-- encoder mid-layer .save and .restore operations are only used internally in
-  drivers, should they be removed from the core?
-- encoder mid-layer .detect operation is only used internally in drivers,
-  should it be removed from the core?
--->
-
-  <!-- External interfaces -->
-
-  <chapter id="drmExternals">
-    <title>Userland interfaces</title>
-    <para>
-      The DRM core exports several interfaces to applications,
-      generally intended to be used through corresponding libdrm
-      wrapper functions.  In addition, drivers export device-specific
-      interfaces for use by userspace drivers &amp; device-aware
-      applications through ioctls and sysfs files.
-    </para>
-    <para>
-      External interfaces include: memory mapping, context management,
-      DMA operations, AGP management, vblank control, fence
-      management, memory management, and output management.
-    </para>
-    <para>
-      Cover generic ioctls and sysfs layout here.  We only need high-level
-      info, since man pages should cover the rest.
-    </para>
-
-  <!-- External: render nodes -->
-
-    <sect1>
-      <title>Render nodes</title>
-      <para>
-        DRM core provides multiple character-devices for user-space to use.
-        Depending on which device is opened, user-space can perform a different
-        set of operations (mainly ioctls). The primary node is always created
-        and called card&lt;num&gt;. Additionally, a currently
-        unused control node, called controlD&lt;num&gt; is also
-        created. The primary node provides all legacy operations and
-        historically was the only interface used by userspace. With KMS, the
-        control node was introduced. However, the planned KMS control interface
-        has never been written and so the control node stays unused to date.
-      </para>
-      <para>
-        With the increased use of offscreen renderers and GPGPU applications,
-        clients no longer require running compositors or graphics servers to
-        make use of a GPU. But the DRM API required unprivileged clients to
-        authenticate to a DRM-Master prior to getting GPU access. To avoid this
-        step and to grant clients GPU access without authenticating, render
-        nodes were introduced. Render nodes solely serve render clients, that
-        is, no modesetting or privileged ioctls can be issued on render nodes.
-        Only non-global rendering commands are allowed. If a driver supports
-        render nodes, it must advertise it via the DRIVER_RENDER
-        DRM driver capability. If not supported, the primary node must be used
-        for render clients together with the legacy drmAuth authentication
-        procedure.
-      </para>
-      <para>
-        If a driver advertises render node support, DRM core will create a
-        separate render node called renderD&lt;num&gt;. There will
-        be one render node per device. No ioctls except  PRIME-related ioctls
-        will be allowed on this node. Especially GEM_OPEN will be
-        explicitly prohibited. Render nodes are designed to avoid the
-        buffer-leaks, which occur if clients guess the flink names or mmap
-        offsets on the legacy interface. Additionally to this basic interface,
-        drivers must mark their driver-dependent render-only ioctls as
-        DRM_RENDER_ALLOW so render clients can use them. Driver
-        authors must be careful not to allow any privileged ioctls on render
-        nodes.
-      </para>
-      <para>
-        With render nodes, user-space can now control access to the render node
-        via basic file-system access-modes. A running graphics server which
-        authenticates clients on the privileged primary/legacy node is no longer
-        required. Instead, a client can open the render node and is immediately
-        granted GPU access. Communication between clients (or servers) is done
-        via PRIME. FLINK from render node to legacy node is not supported. New
-        clients must not use the insecure FLINK interface.
-      </para>
-      <para>
-        Besides dropping all modeset/global ioctls, render nodes also drop the
-        DRM-Master concept. There is no reason to associate render clients with
-        a DRM-Master as they are independent of any graphics server. Besides,
-        they must work without any running master, anyway.
-        Drivers must be able to run without a master object if they support
-        render nodes. If, on the other hand, a driver requires shared state
-        between clients which is visible to user-space and accessible beyond
-        open-file boundaries, they cannot support render nodes.
-      </para>
-    </sect1>
-
-  <!-- External: vblank handling -->
-
-    <sect1>
-      <title>VBlank event handling</title>
-      <para>
-        The DRM core exposes two vertical blank related ioctls:
-        <variablelist>
-          <varlistentry>
-            <term>DRM_IOCTL_WAIT_VBLANK</term>
-            <listitem>
-              <para>
-                This takes a struct drm_wait_vblank structure as its argument,
-                and it is used to block or request a signal when a specified
-                vblank event occurs.
-              </para>
-            </listitem>
-          </varlistentry>
-          <varlistentry>
-            <term>DRM_IOCTL_MODESET_CTL</term>
-            <listitem>
-              <para>
-		This was only used for user-mode-settind drivers around
-		modesetting changes to allow the kernel to update the vblank
-		interrupt after mode setting, since on many devices the vertical
-		blank counter is reset to 0 at some point during modeset. Modern
-		drivers should not call this any more since with kernel mode
-		setting it is a no-op.
-              </para>
-            </listitem>
-          </varlistentry>
-        </variablelist>
-      </para>
-    </sect1>
-
-  </chapter>
-</part>
-<part id="drmDrivers">
-  <title>DRM Drivers</title>
-
-  <partintro>
-    <para>
-      This second part of the DRM Developer's Guide documents driver code,
-      implementation details and also all the driver-specific userspace
-      interfaces. Especially since all hardware-acceleration interfaces to
-      userspace are driver specific for efficiency and other reasons these
-      interfaces can be rather substantial. Hence every driver has its own
-      chapter.
-    </para>
-  </partintro>
-
-  <chapter id="drmI915">
-    <title>drm/i915 Intel GFX Driver</title>
-    <para>
-      The drm/i915 driver supports all (with the exception of some very early
-      models) integrated GFX chipsets with both Intel display and rendering
-      blocks. This excludes a set of SoC platforms with an SGX rendering unit,
-      those have basic support through the gma500 drm driver.
-    </para>
-    <sect1>
-      <title>Core Driver Infrastructure</title>
-      <para>
-	This section covers core driver infrastructure used by both the display
-	and the GEM parts of the driver.
-      </para>
-      <sect2>
-        <title>Runtime Power Management</title>
-!Pdrivers/gpu/drm/i915/intel_runtime_pm.c runtime pm
-!Idrivers/gpu/drm/i915/intel_runtime_pm.c
-!Idrivers/gpu/drm/i915/intel_uncore.c
-      </sect2>
-      <sect2>
-        <title>Interrupt Handling</title>
-!Pdrivers/gpu/drm/i915/i915_irq.c interrupt handling
-!Fdrivers/gpu/drm/i915/i915_irq.c intel_irq_init intel_irq_init_hw intel_hpd_init
-!Fdrivers/gpu/drm/i915/i915_irq.c intel_runtime_pm_disable_interrupts
-!Fdrivers/gpu/drm/i915/i915_irq.c intel_runtime_pm_enable_interrupts
-      </sect2>
-      <sect2>
-        <title>Intel GVT-g Guest Support(vGPU)</title>
-!Pdrivers/gpu/drm/i915/i915_vgpu.c Intel GVT-g guest support
-!Idrivers/gpu/drm/i915/i915_vgpu.c
-      </sect2>
-    </sect1>
-    <sect1>
-      <title>Display Hardware Handling</title>
-      <para>
-        This section covers everything related to the display hardware including
-        the mode setting infrastructure, plane, sprite and cursor handling and
-        display, output probing and related topics.
-      </para>
-      <sect2>
-        <title>Mode Setting Infrastructure</title>
-        <para>
-          The i915 driver is thus far the only DRM driver which doesn't use the
-          common DRM helper code to implement mode setting sequences. Thus it
-          has its own tailor-made infrastructure for executing a display
-          configuration change.
-        </para>
-      </sect2>
-      <sect2>
-        <title>Frontbuffer Tracking</title>
-!Pdrivers/gpu/drm/i915/intel_frontbuffer.c frontbuffer tracking
-!Idrivers/gpu/drm/i915/intel_frontbuffer.c
-!Fdrivers/gpu/drm/i915/i915_gem.c i915_gem_track_fb
-      </sect2>
-      <sect2>
-        <title>Display FIFO Underrun Reporting</title>
-!Pdrivers/gpu/drm/i915/intel_fifo_underrun.c fifo underrun handling
-!Idrivers/gpu/drm/i915/intel_fifo_underrun.c
-      </sect2>
-      <sect2>
-        <title>Plane Configuration</title>
-        <para>
-	  This section covers plane configuration and composition with the
-	  primary plane, sprites, cursors and overlays. This includes the
-	  infrastructure to do atomic vsync'ed updates of all this state and
-	  also tightly coupled topics like watermark setup and computation,
-	  framebuffer compression and panel self refresh.
-        </para>
-      </sect2>
-      <sect2>
-        <title>Atomic Plane Helpers</title>
-!Pdrivers/gpu/drm/i915/intel_atomic_plane.c atomic plane helpers
-!Idrivers/gpu/drm/i915/intel_atomic_plane.c
-      </sect2>
-      <sect2>
-        <title>Output Probing</title>
-        <para>
-	  This section covers output probing and related infrastructure like the
-	  hotplug interrupt storm detection and mitigation code. Note that the
-	  i915 driver still uses most of the common DRM helper code for output
-	  probing, so those sections fully apply.
-        </para>
-      </sect2>
-      <sect2>
-        <title>Hotplug</title>
-!Pdrivers/gpu/drm/i915/intel_hotplug.c Hotplug
-!Idrivers/gpu/drm/i915/intel_hotplug.c
-      </sect2>
-      <sect2>
-	<title>High Definition Audio</title>
-!Pdrivers/gpu/drm/i915/intel_audio.c High Definition Audio over HDMI and Display Port
-!Idrivers/gpu/drm/i915/intel_audio.c
-      </sect2>
-      <sect2>
-	<title>Panel Self Refresh PSR (PSR/SRD)</title>
-!Pdrivers/gpu/drm/i915/intel_psr.c Panel Self Refresh (PSR/SRD)
-!Idrivers/gpu/drm/i915/intel_psr.c
-      </sect2>
-      <sect2>
-	<title>Frame Buffer Compression (FBC)</title>
-!Pdrivers/gpu/drm/i915/intel_fbc.c Frame Buffer Compression (FBC)
-!Idrivers/gpu/drm/i915/intel_fbc.c
-      </sect2>
-      <sect2>
-        <title>Display Refresh Rate Switching (DRRS)</title>
-!Pdrivers/gpu/drm/i915/intel_dp.c Display Refresh Rate Switching (DRRS)
-!Fdrivers/gpu/drm/i915/intel_dp.c intel_dp_set_drrs_state
-!Fdrivers/gpu/drm/i915/intel_dp.c intel_edp_drrs_enable
-!Fdrivers/gpu/drm/i915/intel_dp.c intel_edp_drrs_disable
-!Fdrivers/gpu/drm/i915/intel_dp.c intel_edp_drrs_invalidate
-!Fdrivers/gpu/drm/i915/intel_dp.c intel_edp_drrs_flush
-!Fdrivers/gpu/drm/i915/intel_dp.c intel_dp_drrs_init
-
-      </sect2>
-      <sect2>
-        <title>DPIO</title>
-!Pdrivers/gpu/drm/i915/i915_reg.h DPIO
-	<table id="dpiox2">
-	  <title>Dual channel PHY (VLV/CHV/BXT)</title>
-	  <tgroup cols="8">
-	    <colspec colname="c0" />
-	    <colspec colname="c1" />
-	    <colspec colname="c2" />
-	    <colspec colname="c3" />
-	    <colspec colname="c4" />
-	    <colspec colname="c5" />
-	    <colspec colname="c6" />
-	    <colspec colname="c7" />
-	    <spanspec spanname="ch0" namest="c0" nameend="c3" />
-	    <spanspec spanname="ch1" namest="c4" nameend="c7" />
-	    <spanspec spanname="ch0pcs01" namest="c0" nameend="c1" />
-	    <spanspec spanname="ch0pcs23" namest="c2" nameend="c3" />
-	    <spanspec spanname="ch1pcs01" namest="c4" nameend="c5" />
-	    <spanspec spanname="ch1pcs23" namest="c6" nameend="c7" />
-	    <thead>
-	      <row>
-		<entry spanname="ch0">CH0</entry>
-		<entry spanname="ch1">CH1</entry>
-	      </row>
-	    </thead>
-	    <tbody valign="top" align="center">
-	      <row>
-		<entry spanname="ch0">CMN/PLL/REF</entry>
-		<entry spanname="ch1">CMN/PLL/REF</entry>
-	      </row>
-	      <row>
-		<entry spanname="ch0pcs01">PCS01</entry>
-		<entry spanname="ch0pcs23">PCS23</entry>
-		<entry spanname="ch1pcs01">PCS01</entry>
-		<entry spanname="ch1pcs23">PCS23</entry>
-	      </row>
-	      <row>
-		<entry>TX0</entry>
-		<entry>TX1</entry>
-		<entry>TX2</entry>
-		<entry>TX3</entry>
-		<entry>TX0</entry>
-		<entry>TX1</entry>
-		<entry>TX2</entry>
-		<entry>TX3</entry>
-	      </row>
-	      <row>
-		<entry spanname="ch0">DDI0</entry>
-		<entry spanname="ch1">DDI1</entry>
-	      </row>
-	    </tbody>
-	  </tgroup>
-	</table>
-	<table id="dpiox1">
-	  <title>Single channel PHY (CHV/BXT)</title>
-	  <tgroup cols="4">
-	    <colspec colname="c0" />
-	    <colspec colname="c1" />
-	    <colspec colname="c2" />
-	    <colspec colname="c3" />
-	    <spanspec spanname="ch0" namest="c0" nameend="c3" />
-	    <spanspec spanname="ch0pcs01" namest="c0" nameend="c1" />
-	    <spanspec spanname="ch0pcs23" namest="c2" nameend="c3" />
-	    <thead>
-	      <row>
-		<entry spanname="ch0">CH0</entry>
-	      </row>
-	    </thead>
-	    <tbody valign="top" align="center">
-	      <row>
-		<entry spanname="ch0">CMN/PLL/REF</entry>
-	      </row>
-	      <row>
-		<entry spanname="ch0pcs01">PCS01</entry>
-		<entry spanname="ch0pcs23">PCS23</entry>
-	      </row>
-	      <row>
-		<entry>TX0</entry>
-		<entry>TX1</entry>
-		<entry>TX2</entry>
-		<entry>TX3</entry>
-	      </row>
-	      <row>
-		<entry spanname="ch0">DDI2</entry>
-	      </row>
-	    </tbody>
-	  </tgroup>
-	</table>
-      </sect2>
-
-      <sect2>
-       <title>CSR firmware support for DMC</title>
-!Pdrivers/gpu/drm/i915/intel_csr.c csr support for dmc
-!Idrivers/gpu/drm/i915/intel_csr.c
-      </sect2>
-    </sect1>
-
-    <sect1>
-      <title>Memory Management and Command Submission</title>
-      <para>
-	This sections covers all things related to the GEM implementation in the
-	i915 driver.
-      </para>
-      <sect2>
-        <title>Batchbuffer Parsing</title>
-!Pdrivers/gpu/drm/i915/i915_cmd_parser.c batch buffer command parser
-!Idrivers/gpu/drm/i915/i915_cmd_parser.c
-      </sect2>
-      <sect2>
-        <title>Batchbuffer Pools</title>
-!Pdrivers/gpu/drm/i915/i915_gem_batch_pool.c batch pool
-!Idrivers/gpu/drm/i915/i915_gem_batch_pool.c
-      </sect2>
-      <sect2>
-        <title>Logical Rings, Logical Ring Contexts and Execlists</title>
-!Pdrivers/gpu/drm/i915/intel_lrc.c Logical Rings, Logical Ring Contexts and Execlists
-!Idrivers/gpu/drm/i915/intel_lrc.c
-      </sect2>
-      <sect2>
-        <title>Global GTT views</title>
-!Pdrivers/gpu/drm/i915/i915_gem_gtt.c Global GTT views
-!Idrivers/gpu/drm/i915/i915_gem_gtt.c
-      </sect2>
-      <sect2>
-        <title>GTT Fences and Swizzling</title>
-!Idrivers/gpu/drm/i915/i915_gem_fence.c
-        <sect3>
-          <title>Global GTT Fence Handling</title>
-!Pdrivers/gpu/drm/i915/i915_gem_fence.c fence register handling
-        </sect3>
-        <sect3>
-          <title>Hardware Tiling and Swizzling Details</title>
-!Pdrivers/gpu/drm/i915/i915_gem_fence.c tiling swizzling details
-        </sect3>
-      </sect2>
-      <sect2>
-        <title>Object Tiling IOCTLs</title>
-!Idrivers/gpu/drm/i915/i915_gem_tiling.c
-!Pdrivers/gpu/drm/i915/i915_gem_tiling.c buffer object tiling
-      </sect2>
-      <sect2>
-        <title>Buffer Object Eviction</title>
-	<para>
-	  This section documents the interface functions for evicting buffer
-	  objects to make space available in the virtual gpu address spaces.
-	  Note that this is mostly orthogonal to shrinking buffer objects
-	  caches, which has the goal to make main memory (shared with the gpu
-	  through the unified memory architecture) available.
-	</para>
-!Idrivers/gpu/drm/i915/i915_gem_evict.c
-      </sect2>
-      <sect2>
-        <title>Buffer Object Memory Shrinking</title>
-	<para>
-	  This section documents the interface function for shrinking memory
-	  usage of buffer object caches. Shrinking is used to make main memory
-	  available.  Note that this is mostly orthogonal to evicting buffer
-	  objects, which has the goal to make space in gpu virtual address
-	  spaces.
-	</para>
-!Idrivers/gpu/drm/i915/i915_gem_shrinker.c
-      </sect2>
-    </sect1>
-    <sect1>
-      <title> Tracing </title>
-      <para>
-    This sections covers all things related to the tracepoints implemented in
-    the i915 driver.
-      </para>
-      <sect2>
-        <title> i915_ppgtt_create and i915_ppgtt_release </title>
-!Pdrivers/gpu/drm/i915/i915_trace.h i915_ppgtt_create and i915_ppgtt_release tracepoints
-      </sect2>
-      <sect2>
-        <title> i915_context_create and i915_context_free </title>
-!Pdrivers/gpu/drm/i915/i915_trace.h i915_context_create and i915_context_free tracepoints
-      </sect2>
-      <sect2>
-        <title> switch_mm </title>
-!Pdrivers/gpu/drm/i915/i915_trace.h switch_mm tracepoint
-      </sect2>
-    </sect1>
-
-  </chapter>
-!Cdrivers/gpu/drm/i915/i915_irq.c
-</part>
-</book>
diff --git a/Documentation/DocBook/gpu.tmpl b/Documentation/DocBook/gpu.tmpl
new file mode 100644
index 0000000..48e3ab4
--- /dev/null
+++ b/Documentation/DocBook/gpu.tmpl
@@ -0,0 +1,4203 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE book PUBLIC "-//OASIS//DTD DocBook XML V4.1.2//EN"
+	"http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd" []>
+
+<book id="drmDevelopersGuide">
+  <bookinfo>
+    <title>Linux GPU Driver Developer's Guide</title>
+
+    <authorgroup>
+      <author>
+	<firstname>Jesse</firstname>
+	<surname>Barnes</surname>
+	<contrib>Initial version</contrib>
+	<affiliation>
+	  <orgname>Intel Corporation</orgname>
+	  <address>
+	    <email>jesse.barnes@intel.com</email>
+	  </address>
+	</affiliation>
+      </author>
+      <author>
+	<firstname>Laurent</firstname>
+	<surname>Pinchart</surname>
+	<contrib>Driver internals</contrib>
+	<affiliation>
+	  <orgname>Ideas on board SPRL</orgname>
+	  <address>
+	    <email>laurent.pinchart@ideasonboard.com</email>
+	  </address>
+	</affiliation>
+      </author>
+      <author>
+	<firstname>Daniel</firstname>
+	<surname>Vetter</surname>
+	<contrib>Contributions all over the place</contrib>
+	<affiliation>
+	  <orgname>Intel Corporation</orgname>
+	  <address>
+	    <email>daniel.vetter@ffwll.ch</email>
+	  </address>
+	</affiliation>
+      </author>
+    </authorgroup>
+
+    <copyright>
+      <year>2008-2009</year>
+      <year>2013-2014</year>
+      <holder>Intel Corporation</holder>
+    </copyright>
+    <copyright>
+      <year>2012</year>
+      <holder>Laurent Pinchart</holder>
+    </copyright>
+
+    <legalnotice>
+      <para>
+	The contents of this file may be used under the terms of the GNU
+	General Public License version 2 (the "GPL") as distributed in
+	the kernel source COPYING file.
+      </para>
+    </legalnotice>
+
+    <revhistory>
+      <!-- Put document revisions here, newest first. -->
+      <revision>
+	<revnumber>1.0</revnumber>
+	<date>2012-07-13</date>
+	<authorinitials>LP</authorinitials>
+	<revremark>Added extensive documentation about driver internals.
+	</revremark>
+      </revision>
+    </revhistory>
+  </bookinfo>
+
+<toc></toc>
+
+<part id="drmCore">
+  <title>DRM Core</title>
+  <partintro>
+    <para>
+      This first part of the DRM Developer's Guide documents core DRM code,
+      helper libraries for writing drivers and generic userspace interfaces
+      exposed by DRM drivers.
+    </para>
+  </partintro>
+
+  <chapter id="drmIntroduction">
+    <title>Introduction</title>
+    <para>
+      The Linux DRM layer contains code intended to support the needs
+      of complex graphics devices, usually containing programmable
+      pipelines well suited to 3D graphics acceleration.  Graphics
+      drivers in the kernel may make use of DRM functions to make
+      tasks like memory management, interrupt handling and DMA easier,
+      and provide a uniform interface to applications.
+    </para>
+    <para>
+      A note on versions: this guide covers features found in the DRM
+      tree, including the TTM memory manager, output configuration and
+      mode setting, and the new vblank internals, in addition to all
+      the regular features found in current kernels.
+    </para>
+    <para>
+      [Insert diagram of typical DRM stack here]
+    </para>
+  </chapter>
+
+  <!-- Internals -->
+
+  <chapter id="drmInternals">
+    <title>DRM Internals</title>
+    <para>
+      This chapter documents DRM internals relevant to driver authors
+      and developers working to add support for the latest features to
+      existing drivers.
+    </para>
+    <para>
+      First, we go over some typical driver initialization
+      requirements, like setting up command buffers, creating an
+      initial output configuration, and initializing core services.
+      Subsequent sections cover core internals in more detail,
+      providing implementation notes and examples.
+    </para>
+    <para>
+      The DRM layer provides several services to graphics drivers,
+      many of them driven by the application interfaces it provides
+      through libdrm, the library that wraps most of the DRM ioctls.
+      These include vblank event handling, memory
+      management, output management, framebuffer management, command
+      submission &amp; fencing, suspend/resume support, and DMA
+      services.
+    </para>
+
+  <!-- Internals: driver init -->
+
+  <sect1>
+    <title>Driver Initialization</title>
+    <para>
+      At the core of every DRM driver is a <structname>drm_driver</structname>
+      structure. Drivers typically statically initialize a drm_driver structure,
+      and then pass it to <function>drm_dev_alloc()</function> to allocate a
+      device instance. After the device instance is fully initialized it can be
+      registered (which makes it accessible from userspace) using
+      <function>drm_dev_register()</function>.
+    </para>
+    <para>
+      The <structname>drm_driver</structname> structure contains static
+      information that describes the driver and features it supports, and
+      pointers to methods that the DRM core will call to implement the DRM API.
+      We will first go through the <structname>drm_driver</structname> static
+      information fields, and will then describe individual operations in
+      details as they get used in later sections.
+    </para>
+    <sect2>
+      <title>Driver Information</title>
+      <sect3>
+        <title>Driver Features</title>
+        <para>
+          Drivers inform the DRM core about their requirements and supported
+          features by setting appropriate flags in the
+          <structfield>driver_features</structfield> field. Since those flags
+          influence the DRM core behaviour since registration time, most of them
+          must be set to registering the <structname>drm_driver</structname>
+          instance.
+        </para>
+        <synopsis>u32 driver_features;</synopsis>
+        <variablelist>
+          <title>Driver Feature Flags</title>
+          <varlistentry>
+            <term>DRIVER_USE_AGP</term>
+            <listitem><para>
+              Driver uses AGP interface, the DRM core will manage AGP resources.
+            </para></listitem>
+          </varlistentry>
+          <varlistentry>
+            <term>DRIVER_REQUIRE_AGP</term>
+            <listitem><para>
+              Driver needs AGP interface to function. AGP initialization failure
+              will become a fatal error.
+            </para></listitem>
+          </varlistentry>
+          <varlistentry>
+            <term>DRIVER_PCI_DMA</term>
+            <listitem><para>
+              Driver is capable of PCI DMA, mapping of PCI DMA buffers to
+              userspace will be enabled. Deprecated.
+            </para></listitem>
+          </varlistentry>
+          <varlistentry>
+            <term>DRIVER_SG</term>
+            <listitem><para>
+              Driver can perform scatter/gather DMA, allocation and mapping of
+              scatter/gather buffers will be enabled. Deprecated.
+            </para></listitem>
+          </varlistentry>
+          <varlistentry>
+            <term>DRIVER_HAVE_DMA</term>
+            <listitem><para>
+              Driver supports DMA, the userspace DMA API will be supported.
+              Deprecated.
+            </para></listitem>
+          </varlistentry>
+          <varlistentry>
+            <term>DRIVER_HAVE_IRQ</term><term>DRIVER_IRQ_SHARED</term>
+            <listitem><para>
+              DRIVER_HAVE_IRQ indicates whether the driver has an IRQ handler
+              managed by the DRM Core. The core will support simple IRQ handler
+              installation when the flag is set. The installation process is
+              described in <xref linkend="drm-irq-registration"/>.</para>
+              <para>DRIVER_IRQ_SHARED indicates whether the device &amp; handler
+              support shared IRQs (note that this is required of PCI  drivers).
+            </para></listitem>
+          </varlistentry>
+          <varlistentry>
+            <term>DRIVER_GEM</term>
+            <listitem><para>
+              Driver use the GEM memory manager.
+            </para></listitem>
+          </varlistentry>
+          <varlistentry>
+            <term>DRIVER_MODESET</term>
+            <listitem><para>
+              Driver supports mode setting interfaces (KMS).
+            </para></listitem>
+          </varlistentry>
+          <varlistentry>
+            <term>DRIVER_PRIME</term>
+            <listitem><para>
+              Driver implements DRM PRIME buffer sharing.
+            </para></listitem>
+          </varlistentry>
+          <varlistentry>
+            <term>DRIVER_RENDER</term>
+            <listitem><para>
+              Driver supports dedicated render nodes.
+            </para></listitem>
+          </varlistentry>
+          <varlistentry>
+            <term>DRIVER_ATOMIC</term>
+            <listitem><para>
+              Driver supports atomic properties.  In this case the driver
+              must implement appropriate obj->atomic_get_property() vfuncs
+              for any modeset objects with driver specific properties.
+            </para></listitem>
+          </varlistentry>
+        </variablelist>
+      </sect3>
+      <sect3>
+        <title>Major, Minor and Patchlevel</title>
+        <synopsis>int major;
+int minor;
+int patchlevel;</synopsis>
+        <para>
+          The DRM core identifies driver versions by a major, minor and patch
+          level triplet. The information is printed to the kernel log at
+          initialization time and passed to userspace through the
+          DRM_IOCTL_VERSION ioctl.
+        </para>
+        <para>
+          The major and minor numbers are also used to verify the requested driver
+          API version passed to DRM_IOCTL_SET_VERSION. When the driver API changes
+          between minor versions, applications can call DRM_IOCTL_SET_VERSION to
+          select a specific version of the API. If the requested major isn't equal
+          to the driver major, or the requested minor is larger than the driver
+          minor, the DRM_IOCTL_SET_VERSION call will return an error. Otherwise
+          the driver's set_version() method will be called with the requested
+          version.
+        </para>
+      </sect3>
+      <sect3>
+        <title>Name, Description and Date</title>
+        <synopsis>char *name;
+char *desc;
+char *date;</synopsis>
+        <para>
+          The driver name is printed to the kernel log at initialization time,
+          used for IRQ registration and passed to userspace through
+          DRM_IOCTL_VERSION.
+        </para>
+        <para>
+          The driver description is a purely informative string passed to
+          userspace through the DRM_IOCTL_VERSION ioctl and otherwise unused by
+          the kernel.
+        </para>
+        <para>
+          The driver date, formatted as YYYYMMDD, is meant to identify the date of
+          the latest modification to the driver. However, as most drivers fail to
+          update it, its value is mostly useless. The DRM core prints it to the
+          kernel log at initialization time and passes it to userspace through the
+          DRM_IOCTL_VERSION ioctl.
+        </para>
+      </sect3>
+    </sect2>
+    <sect2>
+      <title>Device Instance and Driver Handling</title>
+!Pdrivers/gpu/drm/drm_drv.c driver instance overview
+!Edrivers/gpu/drm/drm_drv.c
+    </sect2>
+    <sect2>
+      <title>Driver Load</title>
+      <sect3 id="drm-irq-registration">
+        <title>IRQ Registration</title>
+        <para>
+          The DRM core tries to facilitate IRQ handler registration and
+          unregistration by providing <function>drm_irq_install</function> and
+          <function>drm_irq_uninstall</function> functions. Those functions only
+          support a single interrupt per device, devices that use more than one
+          IRQs need to be handled manually.
+        </para>
+        <sect4>
+          <title>Managed IRQ Registration</title>
+          <para>
+            <function>drm_irq_install</function> starts by calling the
+            <methodname>irq_preinstall</methodname> driver operation. The operation
+            is optional and must make sure that the interrupt will not get fired by
+            clearing all pending interrupt flags or disabling the interrupt.
+          </para>
+          <para>
+            The passed-in IRQ will then be requested by a call to
+            <function>request_irq</function>. If the DRIVER_IRQ_SHARED driver
+            feature flag is set, a shared (IRQF_SHARED) IRQ handler will be
+            requested.
+          </para>
+          <para>
+            The IRQ handler function must be provided as the mandatory irq_handler
+            driver operation. It will get passed directly to
+            <function>request_irq</function> and thus has the same prototype as all
+            IRQ handlers. It will get called with a pointer to the DRM device as the
+            second argument.
+          </para>
+          <para>
+            Finally the function calls the optional
+            <methodname>irq_postinstall</methodname> driver operation. The operation
+            usually enables interrupts (excluding the vblank interrupt, which is
+            enabled separately), but drivers may choose to enable/disable interrupts
+            at a different time.
+          </para>
+          <para>
+            <function>drm_irq_uninstall</function> is similarly used to uninstall an
+            IRQ handler. It starts by waking up all processes waiting on a vblank
+            interrupt to make sure they don't hang, and then calls the optional
+            <methodname>irq_uninstall</methodname> driver operation. The operation
+            must disable all hardware interrupts. Finally the function frees the IRQ
+            by calling <function>free_irq</function>.
+          </para>
+        </sect4>
+        <sect4>
+          <title>Manual IRQ Registration</title>
+          <para>
+            Drivers that require multiple interrupt handlers can't use the managed
+            IRQ registration functions. In that case IRQs must be registered and
+            unregistered manually (usually with the <function>request_irq</function>
+            and <function>free_irq</function> functions, or their devm_* equivalent).
+          </para>
+          <para>
+            When manually registering IRQs, drivers must not set the DRIVER_HAVE_IRQ
+            driver feature flag, and must not provide the
+	    <methodname>irq_handler</methodname> driver operation. They must set the
+	    <structname>drm_device</structname> <structfield>irq_enabled</structfield>
+	    field to 1 upon registration of the IRQs, and clear it to 0 after
+	    unregistering the IRQs.
+          </para>
+        </sect4>
+      </sect3>
+      <sect3>
+        <title>Memory Manager Initialization</title>
+        <para>
+          Every DRM driver requires a memory manager which must be initialized at
+          load time. DRM currently contains two memory managers, the Translation
+          Table Manager (TTM) and the Graphics Execution Manager (GEM).
+          This document describes the use of the GEM memory manager only. See
+          <xref linkend="drm-memory-management"/> for details.
+        </para>
+      </sect3>
+      <sect3>
+        <title>Miscellaneous Device Configuration</title>
+        <para>
+          Another task that may be necessary for PCI devices during configuration
+          is mapping the video BIOS. On many devices, the VBIOS describes device
+          configuration, LCD panel timings (if any), and contains flags indicating
+          device state. Mapping the BIOS can be done using the pci_map_rom() call,
+          a convenience function that takes care of mapping the actual ROM,
+          whether it has been shadowed into memory (typically at address 0xc0000)
+          or exists on the PCI device in the ROM BAR. Note that after the ROM has
+          been mapped and any necessary information has been extracted, it should
+          be unmapped; on many devices, the ROM address decoder is shared with
+          other BARs, so leaving it mapped could cause undesired behaviour like
+          hangs or memory corruption.
+  <!--!Fdrivers/pci/rom.c pci_map_rom-->
+        </para>
+      </sect3>
+    </sect2>
+    <sect2>
+      <title>Bus-specific Device Registration and PCI Support</title>
+      <para>
+        A number of functions are provided to help with device registration.
+	The functions deal with PCI and platform devices respectively and are
+	only provided for historical reasons. These are all deprecated and
+	shouldn't be used in new drivers. Besides that there's a few
+	helpers for pci drivers.
+      </para>
+!Edrivers/gpu/drm/drm_pci.c
+!Edrivers/gpu/drm/drm_platform.c
+    </sect2>
+  </sect1>
+
+  <!-- Internals: memory management -->
+
+  <sect1 id="drm-memory-management">
+    <title>Memory management</title>
+    <para>
+      Modern Linux systems require large amount of graphics memory to store
+      frame buffers, textures, vertices and other graphics-related data. Given
+      the very dynamic nature of many of that data, managing graphics memory
+      efficiently is thus crucial for the graphics stack and plays a central
+      role in the DRM infrastructure.
+    </para>
+    <para>
+      The DRM core includes two memory managers, namely Translation Table Maps
+      (TTM) and Graphics Execution Manager (GEM). TTM was the first DRM memory
+      manager to be developed and tried to be a one-size-fits-them all
+      solution. It provides a single userspace API to accommodate the need of
+      all hardware, supporting both Unified Memory Architecture (UMA) devices
+      and devices with dedicated video RAM (i.e. most discrete video cards).
+      This resulted in a large, complex piece of code that turned out to be
+      hard to use for driver development.
+    </para>
+    <para>
+      GEM started as an Intel-sponsored project in reaction to TTM's
+      complexity. Its design philosophy is completely different: instead of
+      providing a solution to every graphics memory-related problems, GEM
+      identified common code between drivers and created a support library to
+      share it. GEM has simpler initialization and execution requirements than
+      TTM, but has no video RAM management capabilities and is thus limited to
+      UMA devices.
+    </para>
+    <sect2>
+      <title>The Translation Table Manager (TTM)</title>
+      <para>
+        TTM design background and information belongs here.
+      </para>
+      <sect3>
+        <title>TTM initialization</title>
+        <warning><para>This section is outdated.</para></warning>
+        <para>
+          Drivers wishing to support TTM must fill out a drm_bo_driver
+          structure. The structure contains several fields with function
+          pointers for initializing the TTM, allocating and freeing memory,
+          waiting for command completion and fence synchronization, and memory
+          migration. See the radeon_ttm.c file for an example of usage.
+        </para>
+        <para>
+          The ttm_global_reference structure is made up of several fields:
+        </para>
+        <programlisting>
+          struct ttm_global_reference {
+                  enum ttm_global_types global_type;
+                  size_t size;
+                  void *object;
+                  int (*init) (struct ttm_global_reference *);
+                  void (*release) (struct ttm_global_reference *);
+          };
+        </programlisting>
+        <para>
+          There should be one global reference structure for your memory
+          manager as a whole, and there will be others for each object
+          created by the memory manager at runtime.  Your global TTM should
+          have a type of TTM_GLOBAL_TTM_MEM.  The size field for the global
+          object should be sizeof(struct ttm_mem_global), and the init and
+          release hooks should point at your driver-specific init and
+          release routines, which probably eventually call
+          ttm_mem_global_init and ttm_mem_global_release, respectively.
+        </para>
+        <para>
+          Once your global TTM accounting structure is set up and initialized
+          by calling ttm_global_item_ref() on it,
+          you need to create a buffer object TTM to
+          provide a pool for buffer object allocation by clients and the
+          kernel itself.  The type of this object should be TTM_GLOBAL_TTM_BO,
+          and its size should be sizeof(struct ttm_bo_global).  Again,
+          driver-specific init and release functions may be provided,
+          likely eventually calling ttm_bo_global_init() and
+          ttm_bo_global_release(), respectively.  Also, like the previous
+          object, ttm_global_item_ref() is used to create an initial reference
+          count for the TTM, which will call your initialization function.
+        </para>
+      </sect3>
+    </sect2>
+    <sect2 id="drm-gem">
+      <title>The Graphics Execution Manager (GEM)</title>
+      <para>
+        The GEM design approach has resulted in a memory manager that doesn't
+        provide full coverage of all (or even all common) use cases in its
+        userspace or kernel API. GEM exposes a set of standard memory-related
+        operations to userspace and a set of helper functions to drivers, and let
+        drivers implement hardware-specific operations with their own private API.
+      </para>
+      <para>
+        The GEM userspace API is described in the
+        <ulink url="http://lwn.net/Articles/283798/"><citetitle>GEM - the Graphics
+        Execution Manager</citetitle></ulink> article on LWN. While slightly
+        outdated, the document provides a good overview of the GEM API principles.
+        Buffer allocation and read and write operations, described as part of the
+        common GEM API, are currently implemented using driver-specific ioctls.
+      </para>
+      <para>
+        GEM is data-agnostic. It manages abstract buffer objects without knowing
+        what individual buffers contain. APIs that require knowledge of buffer
+        contents or purpose, such as buffer allocation or synchronization
+        primitives, are thus outside of the scope of GEM and must be implemented
+        using driver-specific ioctls.
+      </para>
+      <para>
+        On a fundamental level, GEM involves several operations:
+        <itemizedlist>
+          <listitem>Memory allocation and freeing</listitem>
+          <listitem>Command execution</listitem>
+          <listitem>Aperture management at command execution time</listitem>
+        </itemizedlist>
+        Buffer object allocation is relatively straightforward and largely
+        provided by Linux's shmem layer, which provides memory to back each
+        object.
+      </para>
+      <para>
+        Device-specific operations, such as command execution, pinning, buffer
+        read &amp; write, mapping, and domain ownership transfers are left to
+        driver-specific ioctls.
+      </para>
+      <sect3>
+        <title>GEM Initialization</title>
+        <para>
+          Drivers that use GEM must set the DRIVER_GEM bit in the struct
+          <structname>drm_driver</structname>
+          <structfield>driver_features</structfield> field. The DRM core will
+          then automatically initialize the GEM core before calling the
+          <methodname>load</methodname> operation. Behind the scene, this will
+          create a DRM Memory Manager object which provides an address space
+          pool for object allocation.
+        </para>
+        <para>
+          In a KMS configuration, drivers need to allocate and initialize a
+          command ring buffer following core GEM initialization if required by
+          the hardware. UMA devices usually have what is called a "stolen"
+          memory region, which provides space for the initial framebuffer and
+          large, contiguous memory regions required by the device. This space is
+          typically not managed by GEM, and must be initialized separately into
+          its own DRM MM object.
+        </para>
+      </sect3>
+      <sect3>
+        <title>GEM Objects Creation</title>
+        <para>
+          GEM splits creation of GEM objects and allocation of the memory that
+          backs them in two distinct operations.
+        </para>
+        <para>
+          GEM objects are represented by an instance of struct
+          <structname>drm_gem_object</structname>. Drivers usually need to extend
+          GEM objects with private information and thus create a driver-specific
+          GEM object structure type that embeds an instance of struct
+          <structname>drm_gem_object</structname>.
+        </para>
+        <para>
+          To create a GEM object, a driver allocates memory for an instance of its
+          specific GEM object type and initializes the embedded struct
+          <structname>drm_gem_object</structname> with a call to
+          <function>drm_gem_object_init</function>. The function takes a pointer to
+          the DRM device, a pointer to the GEM object and the buffer object size
+          in bytes.
+        </para>
+        <para>
+          GEM uses shmem to allocate anonymous pageable memory.
+          <function>drm_gem_object_init</function> will create an shmfs file of
+          the requested size and store it into the struct
+          <structname>drm_gem_object</structname> <structfield>filp</structfield>
+          field. The memory is used as either main storage for the object when the
+          graphics hardware uses system memory directly or as a backing store
+          otherwise.
+        </para>
+        <para>
+          Drivers are responsible for the actual physical pages allocation by
+          calling <function>shmem_read_mapping_page_gfp</function> for each page.
+          Note that they can decide to allocate pages when initializing the GEM
+          object, or to delay allocation until the memory is needed (for instance
+          when a page fault occurs as a result of a userspace memory access or
+          when the driver needs to start a DMA transfer involving the memory).
+        </para>
+        <para>
+          Anonymous pageable memory allocation is not always desired, for instance
+          when the hardware requires physically contiguous system memory as is
+          often the case in embedded devices. Drivers can create GEM objects with
+          no shmfs backing (called private GEM objects) by initializing them with
+          a call to <function>drm_gem_private_object_init</function> instead of
+          <function>drm_gem_object_init</function>. Storage for private GEM
+          objects must be managed by drivers.
+        </para>
+        <para>
+          Drivers that do not need to extend GEM objects with private information
+          can call the <function>drm_gem_object_alloc</function> function to
+          allocate and initialize a struct <structname>drm_gem_object</structname>
+          instance. The GEM core will call the optional driver
+          <methodname>gem_init_object</methodname> operation after initializing
+          the GEM object with <function>drm_gem_object_init</function>.
+          <synopsis>int (*gem_init_object) (struct drm_gem_object *obj);</synopsis>
+        </para>
+        <para>
+          No alloc-and-init function exists for private GEM objects.
+        </para>
+      </sect3>
+      <sect3>
+        <title>GEM Objects Lifetime</title>
+        <para>
+          All GEM objects are reference-counted by the GEM core. References can be
+          acquired and release by <function>calling drm_gem_object_reference</function>
+          and <function>drm_gem_object_unreference</function> respectively. The
+          caller must hold the <structname>drm_device</structname>
+          <structfield>struct_mutex</structfield> lock. As a convenience, GEM
+          provides the <function>drm_gem_object_reference_unlocked</function> and
+          <function>drm_gem_object_unreference_unlocked</function> functions that
+          can be called without holding the lock.
+        </para>
+        <para>
+          When the last reference to a GEM object is released the GEM core calls
+          the <structname>drm_driver</structname>
+          <methodname>gem_free_object</methodname> operation. That operation is
+          mandatory for GEM-enabled drivers and must free the GEM object and all
+          associated resources.
+        </para>
+        <para>
+          <synopsis>void (*gem_free_object) (struct drm_gem_object *obj);</synopsis>
+          Drivers are responsible for freeing all GEM object resources, including
+          the resources created by the GEM core. If an mmap offset has been
+          created for the object (in which case
+          <structname>drm_gem_object</structname>::<structfield>map_list</structfield>::<structfield>map</structfield>
+          is not NULL) it must be freed by a call to
+          <function>drm_gem_free_mmap_offset</function>. The shmfs backing store
+          must be released by calling <function>drm_gem_object_release</function>
+          (that function can safely be called if no shmfs backing store has been
+          created).
+        </para>
+      </sect3>
+      <sect3>
+        <title>GEM Objects Naming</title>
+        <para>
+          Communication between userspace and the kernel refers to GEM objects
+          using local handles, global names or, more recently, file descriptors.
+          All of those are 32-bit integer values; the usual Linux kernel limits
+          apply to the file descriptors.
+        </para>
+        <para>
+          GEM handles are local to a DRM file. Applications get a handle to a GEM
+          object through a driver-specific ioctl, and can use that handle to refer
+          to the GEM object in other standard or driver-specific ioctls. Closing a
+          DRM file handle frees all its GEM handles and dereferences the
+          associated GEM objects.
+        </para>
+        <para>
+          To create a handle for a GEM object drivers call
+          <function>drm_gem_handle_create</function>. The function takes a pointer
+          to the DRM file and the GEM object and returns a locally unique handle.
+          When the handle is no longer needed drivers delete it with a call to
+          <function>drm_gem_handle_delete</function>. Finally the GEM object
+          associated with a handle can be retrieved by a call to
+          <function>drm_gem_object_lookup</function>.
+        </para>
+        <para>
+          Handles don't take ownership of GEM objects, they only take a reference
+          to the object that will be dropped when the handle is destroyed. To
+          avoid leaking GEM objects, drivers must make sure they drop the
+          reference(s) they own (such as the initial reference taken at object
+          creation time) as appropriate, without any special consideration for the
+          handle. For example, in the particular case of combined GEM object and
+          handle creation in the implementation of the
+          <methodname>dumb_create</methodname> operation, drivers must drop the
+          initial reference to the GEM object before returning the handle.
+        </para>
+        <para>
+          GEM names are similar in purpose to handles but are not local to DRM
+          files. They can be passed between processes to reference a GEM object
+          globally. Names can't be used directly to refer to objects in the DRM
+          API, applications must convert handles to names and names to handles
+          using the DRM_IOCTL_GEM_FLINK and DRM_IOCTL_GEM_OPEN ioctls
+          respectively. The conversion is handled by the DRM core without any
+          driver-specific support.
+        </para>
+        <para>
+          GEM also supports buffer sharing with dma-buf file descriptors through
+          PRIME. GEM-based drivers must use the provided helpers functions to
+          implement the exporting and importing correctly. See <xref linkend="drm-prime-support" />.
+          Since sharing file descriptors is inherently more secure than the
+          easily guessable and global GEM names it is the preferred buffer
+          sharing mechanism. Sharing buffers through GEM names is only supported
+          for legacy userspace. Furthermore PRIME also allows cross-device
+          buffer sharing since it is based on dma-bufs.
+        </para>
+      </sect3>
+      <sect3 id="drm-gem-objects-mapping">
+        <title>GEM Objects Mapping</title>
+        <para>
+          Because mapping operations are fairly heavyweight GEM favours
+          read/write-like access to buffers, implemented through driver-specific
+          ioctls, over mapping buffers to userspace. However, when random access
+          to the buffer is needed (to perform software rendering for instance),
+          direct access to the object can be more efficient.
+        </para>
+        <para>
+          The mmap system call can't be used directly to map GEM objects, as they
+          don't have their own file handle. Two alternative methods currently
+          co-exist to map GEM objects to userspace. The first method uses a
+          driver-specific ioctl to perform the mapping operation, calling
+          <function>do_mmap</function> under the hood. This is often considered
+          dubious, seems to be discouraged for new GEM-enabled drivers, and will
+          thus not be described here.
+        </para>
+        <para>
+          The second method uses the mmap system call on the DRM file handle.
+          <synopsis>void *mmap(void *addr, size_t length, int prot, int flags, int fd,
+             off_t offset);</synopsis>
+          DRM identifies the GEM object to be mapped by a fake offset passed
+          through the mmap offset argument. Prior to being mapped, a GEM object
+          must thus be associated with a fake offset. To do so, drivers must call
+          <function>drm_gem_create_mmap_offset</function> on the object. The
+          function allocates a fake offset range from a pool and stores the
+          offset divided by PAGE_SIZE in
+          <literal>obj-&gt;map_list.hash.key</literal>. Care must be taken not to
+          call <function>drm_gem_create_mmap_offset</function> if a fake offset
+          has already been allocated for the object. This can be tested by
+          <literal>obj-&gt;map_list.map</literal> being non-NULL.
+        </para>
+        <para>
+          Once allocated, the fake offset value
+          (<literal>obj-&gt;map_list.hash.key &lt;&lt; PAGE_SHIFT</literal>)
+          must be passed to the application in a driver-specific way and can then
+          be used as the mmap offset argument.
+        </para>
+        <para>
+          The GEM core provides a helper method <function>drm_gem_mmap</function>
+          to handle object mapping. The method can be set directly as the mmap
+          file operation handler. It will look up the GEM object based on the
+          offset value and set the VMA operations to the
+          <structname>drm_driver</structname> <structfield>gem_vm_ops</structfield>
+          field. Note that <function>drm_gem_mmap</function> doesn't map memory to
+          userspace, but relies on the driver-provided fault handler to map pages
+          individually.
+        </para>
+        <para>
+          To use <function>drm_gem_mmap</function>, drivers must fill the struct
+          <structname>drm_driver</structname> <structfield>gem_vm_ops</structfield>
+          field with a pointer to VM operations.
+        </para>
+        <para>
+          <synopsis>struct vm_operations_struct *gem_vm_ops
+
+  struct vm_operations_struct {
+          void (*open)(struct vm_area_struct * area);
+          void (*close)(struct vm_area_struct * area);
+          int (*fault)(struct vm_area_struct *vma, struct vm_fault *vmf);
+  };</synopsis>
+        </para>
+        <para>
+          The <methodname>open</methodname> and <methodname>close</methodname>
+          operations must update the GEM object reference count. Drivers can use
+          the <function>drm_gem_vm_open</function> and
+          <function>drm_gem_vm_close</function> helper functions directly as open
+          and close handlers.
+        </para>
+        <para>
+          The fault operation handler is responsible for mapping individual pages
+          to userspace when a page fault occurs. Depending on the memory
+          allocation scheme, drivers can allocate pages at fault time, or can
+          decide to allocate memory for the GEM object at the time the object is
+          created.
+        </para>
+        <para>
+          Drivers that want to map the GEM object upfront instead of handling page
+          faults can implement their own mmap file operation handler.
+        </para>
+      </sect3>
+      <sect3>
+        <title>Memory Coherency</title>
+        <para>
+          When mapped to the device or used in a command buffer, backing pages
+          for an object are flushed to memory and marked write combined so as to
+          be coherent with the GPU. Likewise, if the CPU accesses an object
+          after the GPU has finished rendering to the object, then the object
+          must be made coherent with the CPU's view of memory, usually involving
+          GPU cache flushing of various kinds. This core CPU&lt;-&gt;GPU
+          coherency management is provided by a device-specific ioctl, which
+          evaluates an object's current domain and performs any necessary
+          flushing or synchronization to put the object into the desired
+          coherency domain (note that the object may be busy, i.e. an active
+          render target; in that case, setting the domain blocks the client and
+          waits for rendering to complete before performing any necessary
+          flushing operations).
+        </para>
+      </sect3>
+      <sect3>
+        <title>Command Execution</title>
+        <para>
+          Perhaps the most important GEM function for GPU devices is providing a
+          command execution interface to clients. Client programs construct
+          command buffers containing references to previously allocated memory
+          objects, and then submit them to GEM. At that point, GEM takes care to
+          bind all the objects into the GTT, execute the buffer, and provide
+          necessary synchronization between clients accessing the same buffers.
+          This often involves evicting some objects from the GTT and re-binding
+          others (a fairly expensive operation), and providing relocation
+          support which hides fixed GTT offsets from clients. Clients must take
+          care not to submit command buffers that reference more objects than
+          can fit in the GTT; otherwise, GEM will reject them and no rendering
+          will occur. Similarly, if several objects in the buffer require fence
+          registers to be allocated for correct rendering (e.g. 2D blits on
+          pre-965 chips), care must be taken not to require more fence registers
+          than are available to the client. Such resource management should be
+          abstracted from the client in libdrm.
+        </para>
+      </sect3>
+      <sect3>
+        <title>GEM Function Reference</title>
+!Edrivers/gpu/drm/drm_gem.c
+      </sect3>
+    </sect2>
+    <sect2>
+      <title>VMA Offset Manager</title>
+!Pdrivers/gpu/drm/drm_vma_manager.c vma offset manager
+!Edrivers/gpu/drm/drm_vma_manager.c
+!Iinclude/drm/drm_vma_manager.h
+    </sect2>
+    <sect2 id="drm-prime-support">
+      <title>PRIME Buffer Sharing</title>
+      <para>
+        PRIME is the cross device buffer sharing framework in drm, originally
+        created for the OPTIMUS range of multi-gpu platforms. To userspace
+        PRIME buffers are dma-buf based file descriptors.
+      </para>
+      <sect3>
+        <title>Overview and Driver Interface</title>
+        <para>
+          Similar to GEM global names, PRIME file descriptors are
+          also used to share buffer objects across processes. They offer
+          additional security: as file descriptors must be explicitly sent over
+          UNIX domain sockets to be shared between applications, they can't be
+          guessed like the globally unique GEM names.
+        </para>
+        <para>
+          Drivers that support the PRIME
+          API must set the DRIVER_PRIME bit in the struct
+          <structname>drm_driver</structname>
+          <structfield>driver_features</structfield> field, and implement the
+          <methodname>prime_handle_to_fd</methodname> and
+          <methodname>prime_fd_to_handle</methodname> operations.
+        </para>
+        <para>
+          <synopsis>int (*prime_handle_to_fd)(struct drm_device *dev,
+                          struct drm_file *file_priv, uint32_t handle,
+                          uint32_t flags, int *prime_fd);
+int (*prime_fd_to_handle)(struct drm_device *dev,
+                          struct drm_file *file_priv, int prime_fd,
+                          uint32_t *handle);</synopsis>
+            Those two operations convert a handle to a PRIME file descriptor and
+            vice versa. Drivers must use the kernel dma-buf buffer sharing framework
+            to manage the PRIME file descriptors. Similar to the mode setting
+            API PRIME is agnostic to the underlying buffer object manager, as
+            long as handles are 32bit unsigned integers.
+          </para>
+          <para>
+            While non-GEM drivers must implement the operations themselves, GEM
+            drivers must use the <function>drm_gem_prime_handle_to_fd</function>
+            and <function>drm_gem_prime_fd_to_handle</function> helper functions.
+            Those helpers rely on the driver
+            <methodname>gem_prime_export</methodname> and
+            <methodname>gem_prime_import</methodname> operations to create a dma-buf
+            instance from a GEM object (dma-buf exporter role) and to create a GEM
+            object from a dma-buf instance (dma-buf importer role).
+          </para>
+          <para>
+            <synopsis>struct dma_buf * (*gem_prime_export)(struct drm_device *dev,
+                             struct drm_gem_object *obj,
+                             int flags);
+struct drm_gem_object * (*gem_prime_import)(struct drm_device *dev,
+                                            struct dma_buf *dma_buf);</synopsis>
+            These two operations are mandatory for GEM drivers that support
+            PRIME.
+          </para>
+        </sect3>
+      <sect3>
+        <title>PRIME Helper Functions</title>
+!Pdrivers/gpu/drm/drm_prime.c PRIME Helpers
+      </sect3>
+    </sect2>
+    <sect2>
+      <title>PRIME Function References</title>
+!Edrivers/gpu/drm/drm_prime.c
+    </sect2>
+    <sect2>
+      <title>DRM MM Range Allocator</title>
+      <sect3>
+        <title>Overview</title>
+!Pdrivers/gpu/drm/drm_mm.c Overview
+      </sect3>
+      <sect3>
+        <title>LRU Scan/Eviction Support</title>
+!Pdrivers/gpu/drm/drm_mm.c lru scan roaster
+      </sect3>
+      </sect2>
+    <sect2>
+      <title>DRM MM Range Allocator Function References</title>
+!Edrivers/gpu/drm/drm_mm.c
+!Iinclude/drm/drm_mm.h
+    </sect2>
+    <sect2>
+      <title>CMA Helper Functions Reference</title>
+!Pdrivers/gpu/drm/drm_gem_cma_helper.c cma helpers
+!Edrivers/gpu/drm/drm_gem_cma_helper.c
+!Iinclude/drm/drm_gem_cma_helper.h
+    </sect2>
+  </sect1>
+
+  <!-- Internals: mode setting -->
+
+  <sect1 id="drm-mode-setting">
+    <title>Mode Setting</title>
+    <para>
+      Drivers must initialize the mode setting core by calling
+      <function>drm_mode_config_init</function> on the DRM device. The function
+      initializes the <structname>drm_device</structname>
+      <structfield>mode_config</structfield> field and never fails. Once done,
+      mode configuration must be setup by initializing the following fields.
+    </para>
+    <itemizedlist>
+      <listitem>
+        <synopsis>int min_width, min_height;
+int max_width, max_height;</synopsis>
+        <para>
+	  Minimum and maximum width and height of the frame buffers in pixel
+	  units.
+	</para>
+      </listitem>
+      <listitem>
+        <synopsis>struct drm_mode_config_funcs *funcs;</synopsis>
+	<para>Mode setting functions.</para>
+      </listitem>
+    </itemizedlist>
+    <sect2>
+      <title>Display Modes Function Reference</title>
+!Iinclude/drm/drm_modes.h
+!Edrivers/gpu/drm/drm_modes.c
+    </sect2>
+    <sect2>
+      <title>Atomic Mode Setting Function Reference</title>
+!Edrivers/gpu/drm/drm_atomic.c
+    </sect2>
+    <sect2>
+      <title>Frame Buffer Creation</title>
+      <synopsis>struct drm_framebuffer *(*fb_create)(struct drm_device *dev,
+				     struct drm_file *file_priv,
+				     struct drm_mode_fb_cmd2 *mode_cmd);</synopsis>
+      <para>
+        Frame buffers are abstract memory objects that provide a source of
+        pixels to scanout to a CRTC. Applications explicitly request the
+        creation of frame buffers through the DRM_IOCTL_MODE_ADDFB(2) ioctls and
+        receive an opaque handle that can be passed to the KMS CRTC control,
+        plane configuration and page flip functions.
+      </para>
+      <para>
+        Frame buffers rely on the underneath memory manager for low-level memory
+        operations. When creating a frame buffer applications pass a memory
+        handle (or a list of memory handles for multi-planar formats) through
+	the <parameter>drm_mode_fb_cmd2</parameter> argument. For drivers using
+	GEM as their userspace buffer management interface this would be a GEM
+	handle.  Drivers are however free to use their own backing storage object
+	handles, e.g. vmwgfx directly exposes special TTM handles to userspace
+	and so expects TTM handles in the create ioctl and not GEM handles.
+      </para>
+      <para>
+        Drivers must first validate the requested frame buffer parameters passed
+        through the mode_cmd argument. In particular this is where invalid
+        sizes, pixel formats or pitches can be caught.
+      </para>
+      <para>
+        If the parameters are deemed valid, drivers then create, initialize and
+        return an instance of struct <structname>drm_framebuffer</structname>.
+        If desired the instance can be embedded in a larger driver-specific
+	structure. Drivers must fill its <structfield>width</structfield>,
+	<structfield>height</structfield>, <structfield>pitches</structfield>,
+        <structfield>offsets</structfield>, <structfield>depth</structfield>,
+        <structfield>bits_per_pixel</structfield> and
+        <structfield>pixel_format</structfield> fields from the values passed
+        through the <parameter>drm_mode_fb_cmd2</parameter> argument. They
+        should call the <function>drm_helper_mode_fill_fb_struct</function>
+        helper function to do so.
+      </para>
+
+      <para>
+	The initialization of the new framebuffer instance is finalized with a
+	call to <function>drm_framebuffer_init</function> which takes a pointer
+	to DRM frame buffer operations (struct
+	<structname>drm_framebuffer_funcs</structname>). Note that this function
+	publishes the framebuffer and so from this point on it can be accessed
+	concurrently from other threads. Hence it must be the last step in the
+	driver's framebuffer initialization sequence. Frame buffer operations
+	are
+        <itemizedlist>
+          <listitem>
+            <synopsis>int (*create_handle)(struct drm_framebuffer *fb,
+		     struct drm_file *file_priv, unsigned int *handle);</synopsis>
+            <para>
+              Create a handle to the frame buffer underlying memory object. If
+              the frame buffer uses a multi-plane format, the handle will
+              reference the memory object associated with the first plane.
+            </para>
+            <para>
+              Drivers call <function>drm_gem_handle_create</function> to create
+              the handle.
+            </para>
+          </listitem>
+          <listitem>
+            <synopsis>void (*destroy)(struct drm_framebuffer *framebuffer);</synopsis>
+            <para>
+              Destroy the frame buffer object and frees all associated
+              resources. Drivers must call
+              <function>drm_framebuffer_cleanup</function> to free resources
+              allocated by the DRM core for the frame buffer object, and must
+              make sure to unreference all memory objects associated with the
+              frame buffer. Handles created by the
+              <methodname>create_handle</methodname> operation are released by
+              the DRM core.
+            </para>
+          </listitem>
+          <listitem>
+            <synopsis>int (*dirty)(struct drm_framebuffer *framebuffer,
+	     struct drm_file *file_priv, unsigned flags, unsigned color,
+	     struct drm_clip_rect *clips, unsigned num_clips);</synopsis>
+            <para>
+              This optional operation notifies the driver that a region of the
+              frame buffer has changed in response to a DRM_IOCTL_MODE_DIRTYFB
+              ioctl call.
+            </para>
+          </listitem>
+        </itemizedlist>
+      </para>
+      <para>
+	The lifetime of a drm framebuffer is controlled with a reference count,
+	drivers can grab additional references with
+	<function>drm_framebuffer_reference</function>and drop them
+	again with <function>drm_framebuffer_unreference</function>. For
+	driver-private framebuffers for which the last reference is never
+	dropped (e.g. for the fbdev framebuffer when the struct
+	<structname>drm_framebuffer</structname> is embedded into the fbdev
+	helper struct) drivers can manually clean up a framebuffer at module
+	unload time with
+	<function>drm_framebuffer_unregister_private</function>.
+      </para>
+    </sect2>
+    <sect2>
+      <title>Dumb Buffer Objects</title>
+      <para>
+	The KMS API doesn't standardize backing storage object creation and
+	leaves it to driver-specific ioctls. Furthermore actually creating a
+	buffer object even for GEM-based drivers is done through a
+	driver-specific ioctl - GEM only has a common userspace interface for
+	sharing and destroying objects. While not an issue for full-fledged
+	graphics stacks that include device-specific userspace components (in
+	libdrm for instance), this limit makes DRM-based early boot graphics
+	unnecessarily complex.
+      </para>
+      <para>
+        Dumb objects partly alleviate the problem by providing a standard
+        API to create dumb buffers suitable for scanout, which can then be used
+        to create KMS frame buffers.
+      </para>
+      <para>
+        To support dumb objects drivers must implement the
+        <methodname>dumb_create</methodname>,
+        <methodname>dumb_destroy</methodname> and
+        <methodname>dumb_map_offset</methodname> operations.
+      </para>
+      <itemizedlist>
+        <listitem>
+          <synopsis>int (*dumb_create)(struct drm_file *file_priv, struct drm_device *dev,
+                   struct drm_mode_create_dumb *args);</synopsis>
+          <para>
+            The <methodname>dumb_create</methodname> operation creates a driver
+	    object (GEM or TTM handle) suitable for scanout based on the
+	    width, height and depth from the struct
+	    <structname>drm_mode_create_dumb</structname> argument. It fills the
+	    argument's <structfield>handle</structfield>,
+	    <structfield>pitch</structfield> and <structfield>size</structfield>
+	    fields with a handle for the newly created object and its line
+            pitch and size in bytes.
+          </para>
+        </listitem>
+        <listitem>
+          <synopsis>int (*dumb_destroy)(struct drm_file *file_priv, struct drm_device *dev,
+                    uint32_t handle);</synopsis>
+          <para>
+            The <methodname>dumb_destroy</methodname> operation destroys a dumb
+            object created by <methodname>dumb_create</methodname>.
+          </para>
+        </listitem>
+        <listitem>
+          <synopsis>int (*dumb_map_offset)(struct drm_file *file_priv, struct drm_device *dev,
+                       uint32_t handle, uint64_t *offset);</synopsis>
+          <para>
+            The <methodname>dumb_map_offset</methodname> operation associates an
+            mmap fake offset with the object given by the handle and returns
+            it. Drivers must use the
+            <function>drm_gem_create_mmap_offset</function> function to
+            associate the fake offset as described in
+            <xref linkend="drm-gem-objects-mapping"/>.
+          </para>
+        </listitem>
+      </itemizedlist>
+      <para>
+        Note that dumb objects may not be used for gpu acceleration, as has been
+	attempted on some ARM embedded platforms. Such drivers really must have
+	a hardware-specific ioctl to allocate suitable buffer objects.
+      </para>
+    </sect2>
+    <sect2>
+      <title>Output Polling</title>
+      <synopsis>void (*output_poll_changed)(struct drm_device *dev);</synopsis>
+      <para>
+        This operation notifies the driver that the status of one or more
+        connectors has changed. Drivers that use the fb helper can just call the
+        <function>drm_fb_helper_hotplug_event</function> function to handle this
+        operation.
+      </para>
+    </sect2>
+    <sect2>
+      <title>Locking</title>
+      <para>
+        Beside some lookup structures with their own locking (which is hidden
+	behind the interface functions) most of the modeset state is protected
+	by the <code>dev-&lt;mode_config.lock</code> mutex and additionally
+	per-crtc locks to allow cursor updates, pageflips and similar operations
+	to occur concurrently with background tasks like output detection.
+	Operations which cross domains like a full modeset always grab all
+	locks. Drivers there need to protect resources shared between crtcs with
+	additional locking. They also need to be careful to always grab the
+	relevant crtc locks if a modset functions touches crtc state, e.g. for
+	load detection (which does only grab the <code>mode_config.lock</code>
+	to allow concurrent screen updates on live crtcs).
+      </para>
+    </sect2>
+  </sect1>
+
+  <!-- Internals: kms initialization and cleanup -->
+
+  <sect1 id="drm-kms-init">
+    <title>KMS Initialization and Cleanup</title>
+    <para>
+      A KMS device is abstracted and exposed as a set of planes, CRTCs, encoders
+      and connectors. KMS drivers must thus create and initialize all those
+      objects at load time after initializing mode setting.
+    </para>
+    <sect2>
+      <title>CRTCs (struct <structname>drm_crtc</structname>)</title>
+      <para>
+        A CRTC is an abstraction representing a part of the chip that contains a
+	pointer to a scanout buffer. Therefore, the number of CRTCs available
+	determines how many independent scanout buffers can be active at any
+	given time. The CRTC structure contains several fields to support this:
+	a pointer to some video memory (abstracted as a frame buffer object), a
+	display mode, and an (x, y) offset into the video memory to support
+	panning or configurations where one piece of video memory spans multiple
+	CRTCs.
+      </para>
+      <sect3>
+        <title>CRTC Initialization</title>
+        <para>
+          A KMS device must create and register at least one struct
+          <structname>drm_crtc</structname> instance. The instance is allocated
+          and zeroed by the driver, possibly as part of a larger structure, and
+          registered with a call to <function>drm_crtc_init</function> with a
+          pointer to CRTC functions.
+        </para>
+      </sect3>
+      <sect3 id="drm-kms-crtcops">
+        <title>CRTC Operations</title>
+        <sect4>
+          <title>Set Configuration</title>
+          <synopsis>int (*set_config)(struct drm_mode_set *set);</synopsis>
+          <para>
+            Apply a new CRTC configuration to the device. The configuration
+            specifies a CRTC, a frame buffer to scan out from, a (x,y) position in
+            the frame buffer, a display mode and an array of connectors to drive
+            with the CRTC if possible.
+          </para>
+          <para>
+            If the frame buffer specified in the configuration is NULL, the driver
+            must detach all encoders connected to the CRTC and all connectors
+            attached to those encoders and disable them.
+          </para>
+          <para>
+            This operation is called with the mode config lock held.
+          </para>
+          <note><para>
+	    Note that the drm core has no notion of restoring the mode setting
+	    state after resume, since all resume handling is in the full
+	    responsibility of the driver. The common mode setting helper library
+	    though provides a helper which can be used for this:
+	    <function>drm_helper_resume_force_mode</function>.
+          </para></note>
+        </sect4>
+        <sect4>
+          <title>Page Flipping</title>
+          <synopsis>int (*page_flip)(struct drm_crtc *crtc, struct drm_framebuffer *fb,
+                   struct drm_pending_vblank_event *event);</synopsis>
+          <para>
+            Schedule a page flip to the given frame buffer for the CRTC. This
+            operation is called with the mode config mutex held.
+          </para>
+          <para>
+            Page flipping is a synchronization mechanism that replaces the frame
+            buffer being scanned out by the CRTC with a new frame buffer during
+            vertical blanking, avoiding tearing. When an application requests a page
+            flip the DRM core verifies that the new frame buffer is large enough to
+            be scanned out by  the CRTC in the currently configured mode and then
+            calls the CRTC <methodname>page_flip</methodname> operation with a
+            pointer to the new frame buffer.
+          </para>
+          <para>
+            The <methodname>page_flip</methodname> operation schedules a page flip.
+            Once any pending rendering targeting the new frame buffer has
+            completed, the CRTC will be reprogrammed to display that frame buffer
+            after the next vertical refresh. The operation must return immediately
+            without waiting for rendering or page flip to complete and must block
+            any new rendering to the frame buffer until the page flip completes.
+          </para>
+          <para>
+            If a page flip can be successfully scheduled the driver must set the
+            <code>drm_crtc-&gt;fb</code> field to the new framebuffer pointed to
+            by <code>fb</code>. This is important so that the reference counting
+            on framebuffers stays balanced.
+          </para>
+          <para>
+            If a page flip is already pending, the
+            <methodname>page_flip</methodname> operation must return
+            -<errorname>EBUSY</errorname>.
+          </para>
+          <para>
+            To synchronize page flip to vertical blanking the driver will likely
+            need to enable vertical blanking interrupts. It should call
+            <function>drm_vblank_get</function> for that purpose, and call
+            <function>drm_vblank_put</function> after the page flip completes.
+          </para>
+          <para>
+            If the application has requested to be notified when page flip completes
+            the <methodname>page_flip</methodname> operation will be called with a
+            non-NULL <parameter>event</parameter> argument pointing to a
+            <structname>drm_pending_vblank_event</structname> instance. Upon page
+            flip completion the driver must call <methodname>drm_send_vblank_event</methodname>
+            to fill in the event and send to wake up any waiting processes.
+            This can be performed with
+            <programlisting><![CDATA[
+            spin_lock_irqsave(&dev->event_lock, flags);
+            ...
+            drm_send_vblank_event(dev, pipe, event);
+            spin_unlock_irqrestore(&dev->event_lock, flags);
+            ]]></programlisting>
+          </para>
+          <note><para>
+            FIXME: Could drivers that don't need to wait for rendering to complete
+            just add the event to <literal>dev-&gt;vblank_event_list</literal> and
+            let the DRM core handle everything, as for "normal" vertical blanking
+            events?
+          </para></note>
+          <para>
+            While waiting for the page flip to complete, the
+            <literal>event-&gt;base.link</literal> list head can be used freely by
+            the driver to store the pending event in a driver-specific list.
+          </para>
+          <para>
+            If the file handle is closed before the event is signaled, drivers must
+            take care to destroy the event in their
+            <methodname>preclose</methodname> operation (and, if needed, call
+            <function>drm_vblank_put</function>).
+          </para>
+        </sect4>
+        <sect4>
+          <title>Miscellaneous</title>
+          <itemizedlist>
+            <listitem>
+              <synopsis>void (*set_property)(struct drm_crtc *crtc,
+                     struct drm_property *property, uint64_t value);</synopsis>
+              <para>
+                Set the value of the given CRTC property to
+                <parameter>value</parameter>. See <xref linkend="drm-kms-properties"/>
+                for more information about properties.
+              </para>
+            </listitem>
+            <listitem>
+              <synopsis>void (*gamma_set)(struct drm_crtc *crtc, u16 *r, u16 *g, u16 *b,
+                        uint32_t start, uint32_t size);</synopsis>
+              <para>
+                Apply a gamma table to the device. The operation is optional.
+              </para>
+            </listitem>
+            <listitem>
+              <synopsis>void (*destroy)(struct drm_crtc *crtc);</synopsis>
+              <para>
+                Destroy the CRTC when not needed anymore. See
+                <xref linkend="drm-kms-init"/>.
+              </para>
+            </listitem>
+          </itemizedlist>
+        </sect4>
+      </sect3>
+    </sect2>
+    <sect2>
+      <title>Planes (struct <structname>drm_plane</structname>)</title>
+      <para>
+        A plane represents an image source that can be blended with or overlayed
+	on top of a CRTC during the scanout process. Planes are associated with
+	a frame buffer to crop a portion of the image memory (source) and
+	optionally scale it to a destination size. The result is then blended
+	with or overlayed on top of a CRTC.
+      </para>
+      <para>
+      The DRM core recognizes three types of planes:
+      <itemizedlist>
+        <listitem>
+        DRM_PLANE_TYPE_PRIMARY represents a "main" plane for a CRTC.  Primary
+        planes are the planes operated upon by CRTC modesetting and flipping
+        operations described in <xref linkend="drm-kms-crtcops"/>.
+        </listitem>
+        <listitem>
+        DRM_PLANE_TYPE_CURSOR represents a "cursor" plane for a CRTC.  Cursor
+        planes are the planes operated upon by the DRM_IOCTL_MODE_CURSOR and
+        DRM_IOCTL_MODE_CURSOR2 ioctls.
+        </listitem>
+        <listitem>
+        DRM_PLANE_TYPE_OVERLAY represents all non-primary, non-cursor planes.
+        Some drivers refer to these types of planes as "sprites" internally.
+        </listitem>
+      </itemizedlist>
+      For compatibility with legacy userspace, only overlay planes are made
+      available to userspace by default.  Userspace clients may set the
+      DRM_CLIENT_CAP_UNIVERSAL_PLANES client capability bit to indicate that
+      they wish to receive a universal plane list containing all plane types.
+      </para>
+      <sect3>
+        <title>Plane Initialization</title>
+        <para>
+          To create a plane, a KMS drivers allocates and
+          zeroes an instances of struct <structname>drm_plane</structname>
+          (possibly as part of a larger structure) and registers it with a call
+          to <function>drm_universal_plane_init</function>. The function takes a bitmask
+          of the CRTCs that can be associated with the plane, a pointer to the
+          plane functions, a list of format supported formats, and the type of
+          plane (primary, cursor, or overlay) being initialized.
+        </para>
+        <para>
+          Cursor and overlay planes are optional.  All drivers should provide
+          one primary plane per CRTC (although this requirement may change in
+          the future); drivers that do not wish to provide special handling for
+          primary planes may make use of the helper functions described in
+          <xref linkend="drm-kms-planehelpers"/> to create and register a
+          primary plane with standard capabilities.
+        </para>
+      </sect3>
+      <sect3>
+        <title>Plane Operations</title>
+        <itemizedlist>
+          <listitem>
+            <synopsis>int (*update_plane)(struct drm_plane *plane, struct drm_crtc *crtc,
+                        struct drm_framebuffer *fb, int crtc_x, int crtc_y,
+                        unsigned int crtc_w, unsigned int crtc_h,
+                        uint32_t src_x, uint32_t src_y,
+                        uint32_t src_w, uint32_t src_h);</synopsis>
+            <para>
+              Enable and configure the plane to use the given CRTC and frame buffer.
+            </para>
+            <para>
+              The source rectangle in frame buffer memory coordinates is given by
+              the <parameter>src_x</parameter>, <parameter>src_y</parameter>,
+              <parameter>src_w</parameter> and <parameter>src_h</parameter>
+              parameters (as 16.16 fixed point values). Devices that don't support
+              subpixel plane coordinates can ignore the fractional part.
+            </para>
+            <para>
+              The destination rectangle in CRTC coordinates is given by the
+              <parameter>crtc_x</parameter>, <parameter>crtc_y</parameter>,
+              <parameter>crtc_w</parameter> and <parameter>crtc_h</parameter>
+              parameters (as integer values). Devices scale the source rectangle to
+              the destination rectangle. If scaling is not supported, and the source
+              rectangle size doesn't match the destination rectangle size, the
+              driver must return a -<errorname>EINVAL</errorname> error.
+            </para>
+          </listitem>
+          <listitem>
+            <synopsis>int (*disable_plane)(struct drm_plane *plane);</synopsis>
+            <para>
+              Disable the plane. The DRM core calls this method in response to a
+              DRM_IOCTL_MODE_SETPLANE ioctl call with the frame buffer ID set to 0.
+              Disabled planes must not be processed by the CRTC.
+            </para>
+          </listitem>
+          <listitem>
+            <synopsis>void (*destroy)(struct drm_plane *plane);</synopsis>
+            <para>
+              Destroy the plane when not needed anymore. See
+              <xref linkend="drm-kms-init"/>.
+            </para>
+          </listitem>
+        </itemizedlist>
+      </sect3>
+    </sect2>
+    <sect2>
+      <title>Encoders (struct <structname>drm_encoder</structname>)</title>
+      <para>
+        An encoder takes pixel data from a CRTC and converts it to a format
+	suitable for any attached connectors. On some devices, it may be
+	possible to have a CRTC send data to more than one encoder. In that
+	case, both encoders would receive data from the same scanout buffer,
+	resulting in a "cloned" display configuration across the connectors
+	attached to each encoder.
+      </para>
+      <sect3>
+        <title>Encoder Initialization</title>
+        <para>
+          As for CRTCs, a KMS driver must create, initialize and register at
+          least one struct <structname>drm_encoder</structname> instance. The
+          instance is allocated and zeroed by the driver, possibly as part of a
+          larger structure.
+        </para>
+        <para>
+          Drivers must initialize the struct <structname>drm_encoder</structname>
+          <structfield>possible_crtcs</structfield> and
+          <structfield>possible_clones</structfield> fields before registering the
+          encoder. Both fields are bitmasks of respectively the CRTCs that the
+          encoder can be connected to, and sibling encoders candidate for cloning.
+        </para>
+        <para>
+          After being initialized, the encoder must be registered with a call to
+          <function>drm_encoder_init</function>. The function takes a pointer to
+          the encoder functions and an encoder type. Supported types are
+          <itemizedlist>
+            <listitem>
+              DRM_MODE_ENCODER_DAC for VGA and analog on DVI-I/DVI-A
+              </listitem>
+            <listitem>
+              DRM_MODE_ENCODER_TMDS for DVI, HDMI and (embedded) DisplayPort
+            </listitem>
+            <listitem>
+              DRM_MODE_ENCODER_LVDS for display panels
+            </listitem>
+            <listitem>
+              DRM_MODE_ENCODER_TVDAC for TV output (Composite, S-Video, Component,
+              SCART)
+            </listitem>
+            <listitem>
+              DRM_MODE_ENCODER_VIRTUAL for virtual machine displays
+            </listitem>
+          </itemizedlist>
+        </para>
+        <para>
+          Encoders must be attached to a CRTC to be used. DRM drivers leave
+          encoders unattached at initialization time. Applications (or the fbdev
+          compatibility layer when implemented) are responsible for attaching the
+          encoders they want to use to a CRTC.
+        </para>
+      </sect3>
+      <sect3>
+        <title>Encoder Operations</title>
+        <itemizedlist>
+          <listitem>
+            <synopsis>void (*destroy)(struct drm_encoder *encoder);</synopsis>
+            <para>
+              Called to destroy the encoder when not needed anymore. See
+              <xref linkend="drm-kms-init"/>.
+            </para>
+          </listitem>
+          <listitem>
+            <synopsis>void (*set_property)(struct drm_plane *plane,
+                     struct drm_property *property, uint64_t value);</synopsis>
+            <para>
+              Set the value of the given plane property to
+              <parameter>value</parameter>. See <xref linkend="drm-kms-properties"/>
+              for more information about properties.
+            </para>
+          </listitem>
+        </itemizedlist>
+      </sect3>
+    </sect2>
+    <sect2>
+      <title>Connectors (struct <structname>drm_connector</structname>)</title>
+      <para>
+        A connector is the final destination for pixel data on a device, and
+	usually connects directly to an external display device like a monitor
+	or laptop panel. A connector can only be attached to one encoder at a
+	time. The connector is also the structure where information about the
+	attached display is kept, so it contains fields for display data, EDID
+	data, DPMS &amp; connection status, and information about modes
+	supported on the attached displays.
+      </para>
+      <sect3>
+        <title>Connector Initialization</title>
+        <para>
+          Finally a KMS driver must create, initialize, register and attach at
+          least one struct <structname>drm_connector</structname> instance. The
+          instance is created as other KMS objects and initialized by setting the
+          following fields.
+        </para>
+        <variablelist>
+          <varlistentry>
+            <term><structfield>interlace_allowed</structfield></term>
+            <listitem><para>
+              Whether the connector can handle interlaced modes.
+            </para></listitem>
+          </varlistentry>
+          <varlistentry>
+            <term><structfield>doublescan_allowed</structfield></term>
+            <listitem><para>
+              Whether the connector can handle doublescan.
+            </para></listitem>
+          </varlistentry>
+          <varlistentry>
+            <term><structfield>display_info
+            </structfield></term>
+            <listitem><para>
+              Display information is filled from EDID information when a display
+              is detected. For non hot-pluggable displays such as flat panels in
+              embedded systems, the driver should initialize the
+              <structfield>display_info</structfield>.<structfield>width_mm</structfield>
+              and
+              <structfield>display_info</structfield>.<structfield>height_mm</structfield>
+              fields with the physical size of the display.
+            </para></listitem>
+          </varlistentry>
+          <varlistentry>
+            <term id="drm-kms-connector-polled"><structfield>polled</structfield></term>
+            <listitem><para>
+              Connector polling mode, a combination of
+              <variablelist>
+                <varlistentry>
+                  <term>DRM_CONNECTOR_POLL_HPD</term>
+                  <listitem><para>
+                    The connector generates hotplug events and doesn't need to be
+                    periodically polled. The CONNECT and DISCONNECT flags must not
+                    be set together with the HPD flag.
+                  </para></listitem>
+                </varlistentry>
+                <varlistentry>
+                  <term>DRM_CONNECTOR_POLL_CONNECT</term>
+                  <listitem><para>
+                    Periodically poll the connector for connection.
+                  </para></listitem>
+                </varlistentry>
+                <varlistentry>
+                  <term>DRM_CONNECTOR_POLL_DISCONNECT</term>
+                  <listitem><para>
+                    Periodically poll the connector for disconnection.
+                  </para></listitem>
+                </varlistentry>
+              </variablelist>
+              Set to 0 for connectors that don't support connection status
+              discovery.
+            </para></listitem>
+          </varlistentry>
+        </variablelist>
+        <para>
+          The connector is then registered with a call to
+          <function>drm_connector_init</function> with a pointer to the connector
+          functions and a connector type, and exposed through sysfs with a call to
+          <function>drm_connector_register</function>.
+        </para>
+        <para>
+          Supported connector types are
+          <itemizedlist>
+            <listitem>DRM_MODE_CONNECTOR_VGA</listitem>
+            <listitem>DRM_MODE_CONNECTOR_DVII</listitem>
+            <listitem>DRM_MODE_CONNECTOR_DVID</listitem>
+            <listitem>DRM_MODE_CONNECTOR_DVIA</listitem>
+            <listitem>DRM_MODE_CONNECTOR_Composite</listitem>
+            <listitem>DRM_MODE_CONNECTOR_SVIDEO</listitem>
+            <listitem>DRM_MODE_CONNECTOR_LVDS</listitem>
+            <listitem>DRM_MODE_CONNECTOR_Component</listitem>
+            <listitem>DRM_MODE_CONNECTOR_9PinDIN</listitem>
+            <listitem>DRM_MODE_CONNECTOR_DisplayPort</listitem>
+            <listitem>DRM_MODE_CONNECTOR_HDMIA</listitem>
+            <listitem>DRM_MODE_CONNECTOR_HDMIB</listitem>
+            <listitem>DRM_MODE_CONNECTOR_TV</listitem>
+            <listitem>DRM_MODE_CONNECTOR_eDP</listitem>
+            <listitem>DRM_MODE_CONNECTOR_VIRTUAL</listitem>
+          </itemizedlist>
+        </para>
+        <para>
+          Connectors must be attached to an encoder to be used. For devices that
+          map connectors to encoders 1:1, the connector should be attached at
+          initialization time with a call to
+          <function>drm_mode_connector_attach_encoder</function>. The driver must
+          also set the <structname>drm_connector</structname>
+          <structfield>encoder</structfield> field to point to the attached
+          encoder.
+        </para>
+        <para>
+          Finally, drivers must initialize the connectors state change detection
+          with a call to <function>drm_kms_helper_poll_init</function>. If at
+          least one connector is pollable but can't generate hotplug interrupts
+          (indicated by the DRM_CONNECTOR_POLL_CONNECT and
+          DRM_CONNECTOR_POLL_DISCONNECT connector flags), a delayed work will
+          automatically be queued to periodically poll for changes. Connectors
+          that can generate hotplug interrupts must be marked with the
+          DRM_CONNECTOR_POLL_HPD flag instead, and their interrupt handler must
+          call <function>drm_helper_hpd_irq_event</function>. The function will
+          queue a delayed work to check the state of all connectors, but no
+          periodic polling will be done.
+        </para>
+      </sect3>
+      <sect3>
+        <title>Connector Operations</title>
+        <note><para>
+          Unless otherwise state, all operations are mandatory.
+        </para></note>
+        <sect4>
+          <title>DPMS</title>
+          <synopsis>void (*dpms)(struct drm_connector *connector, int mode);</synopsis>
+          <para>
+            The DPMS operation sets the power state of a connector. The mode
+            argument is one of
+            <itemizedlist>
+              <listitem><para>DRM_MODE_DPMS_ON</para></listitem>
+              <listitem><para>DRM_MODE_DPMS_STANDBY</para></listitem>
+              <listitem><para>DRM_MODE_DPMS_SUSPEND</para></listitem>
+              <listitem><para>DRM_MODE_DPMS_OFF</para></listitem>
+            </itemizedlist>
+          </para>
+          <para>
+            In all but DPMS_ON mode the encoder to which the connector is attached
+            should put the display in low-power mode by driving its signals
+            appropriately. If more than one connector is attached to the encoder
+            care should be taken not to change the power state of other displays as
+            a side effect. Low-power mode should be propagated to the encoders and
+            CRTCs when all related connectors are put in low-power mode.
+          </para>
+        </sect4>
+        <sect4>
+          <title>Modes</title>
+          <synopsis>int (*fill_modes)(struct drm_connector *connector, uint32_t max_width,
+                      uint32_t max_height);</synopsis>
+          <para>
+            Fill the mode list with all supported modes for the connector. If the
+            <parameter>max_width</parameter> and <parameter>max_height</parameter>
+            arguments are non-zero, the implementation must ignore all modes wider
+            than <parameter>max_width</parameter> or higher than
+            <parameter>max_height</parameter>.
+          </para>
+          <para>
+            The connector must also fill in this operation its
+            <structfield>display_info</structfield>
+            <structfield>width_mm</structfield> and
+            <structfield>height_mm</structfield> fields with the connected display
+            physical size in millimeters. The fields should be set to 0 if the value
+            isn't known or is not applicable (for instance for projector devices).
+          </para>
+        </sect4>
+        <sect4>
+          <title>Connection Status</title>
+          <para>
+            The connection status is updated through polling or hotplug events when
+            supported (see <xref linkend="drm-kms-connector-polled"/>). The status
+            value is reported to userspace through ioctls and must not be used
+            inside the driver, as it only gets initialized by a call to
+            <function>drm_mode_getconnector</function> from userspace.
+          </para>
+          <synopsis>enum drm_connector_status (*detect)(struct drm_connector *connector,
+                                        bool force);</synopsis>
+          <para>
+            Check to see if anything is attached to the connector. The
+            <parameter>force</parameter> parameter is set to false whilst polling or
+            to true when checking the connector due to user request.
+            <parameter>force</parameter> can be used by the driver to avoid
+            expensive, destructive operations during automated probing.
+          </para>
+          <para>
+            Return connector_status_connected if something is connected to the
+            connector, connector_status_disconnected if nothing is connected and
+            connector_status_unknown if the connection state isn't known.
+          </para>
+          <para>
+            Drivers should only return connector_status_connected if the connection
+            status has really been probed as connected. Connectors that can't detect
+            the connection status, or failed connection status probes, should return
+            connector_status_unknown.
+          </para>
+        </sect4>
+        <sect4>
+          <title>Miscellaneous</title>
+          <itemizedlist>
+            <listitem>
+              <synopsis>void (*set_property)(struct drm_connector *connector,
+                     struct drm_property *property, uint64_t value);</synopsis>
+              <para>
+                Set the value of the given connector property to
+                <parameter>value</parameter>. See <xref linkend="drm-kms-properties"/>
+                for more information about properties.
+              </para>
+            </listitem>
+            <listitem>
+              <synopsis>void (*destroy)(struct drm_connector *connector);</synopsis>
+              <para>
+                Destroy the connector when not needed anymore. See
+                <xref linkend="drm-kms-init"/>.
+              </para>
+            </listitem>
+          </itemizedlist>
+        </sect4>
+      </sect3>
+    </sect2>
+    <sect2>
+      <title>Cleanup</title>
+      <para>
+        The DRM core manages its objects' lifetime. When an object is not needed
+	anymore the core calls its destroy function, which must clean up and
+	free every resource allocated for the object. Every
+	<function>drm_*_init</function> call must be matched with a
+	corresponding <function>drm_*_cleanup</function> call to cleanup CRTCs
+	(<function>drm_crtc_cleanup</function>), planes
+	(<function>drm_plane_cleanup</function>), encoders
+	(<function>drm_encoder_cleanup</function>) and connectors
+	(<function>drm_connector_cleanup</function>). Furthermore, connectors
+	that have been added to sysfs must be removed by a call to
+	<function>drm_connector_unregister</function> before calling
+	<function>drm_connector_cleanup</function>.
+      </para>
+      <para>
+        Connectors state change detection must be cleanup up with a call to
+	<function>drm_kms_helper_poll_fini</function>.
+      </para>
+    </sect2>
+    <sect2>
+      <title>Output discovery and initialization example</title>
+      <programlisting><![CDATA[
+void intel_crt_init(struct drm_device *dev)
+{
+	struct drm_connector *connector;
+	struct intel_output *intel_output;
+
+	intel_output = kzalloc(sizeof(struct intel_output), GFP_KERNEL);
+	if (!intel_output)
+		return;
+
+	connector = &intel_output->base;
+	drm_connector_init(dev, &intel_output->base,
+			   &intel_crt_connector_funcs, DRM_MODE_CONNECTOR_VGA);
+
+	drm_encoder_init(dev, &intel_output->enc, &intel_crt_enc_funcs,
+			 DRM_MODE_ENCODER_DAC);
+
+	drm_mode_connector_attach_encoder(&intel_output->base,
+					  &intel_output->enc);
+
+	/* Set up the DDC bus. */
+	intel_output->ddc_bus = intel_i2c_create(dev, GPIOA, "CRTDDC_A");
+	if (!intel_output->ddc_bus) {
+		dev_printk(KERN_ERR, &dev->pdev->dev, "DDC bus registration "
+			   "failed.\n");
+		return;
+	}
+
+	intel_output->type = INTEL_OUTPUT_ANALOG;
+	connector->interlace_allowed = 0;
+	connector->doublescan_allowed = 0;
+
+	drm_encoder_helper_add(&intel_output->enc, &intel_crt_helper_funcs);
+	drm_connector_helper_add(connector, &intel_crt_connector_helper_funcs);
+
+	drm_connector_register(connector);
+}]]></programlisting>
+      <para>
+        In the example above (taken from the i915 driver), a CRTC, connector and
+        encoder combination is created. A device-specific i2c bus is also
+        created for fetching EDID data and performing monitor detection. Once
+        the process is complete, the new connector is registered with sysfs to
+        make its properties available to applications.
+      </para>
+    </sect2>
+    <sect2>
+      <title>KMS API Functions</title>
+!Edrivers/gpu/drm/drm_crtc.c
+    </sect2>
+    <sect2>
+      <title>KMS Data Structures</title>
+!Iinclude/drm/drm_crtc.h
+    </sect2>
+    <sect2>
+      <title>KMS Locking</title>
+!Pdrivers/gpu/drm/drm_modeset_lock.c kms locking
+!Iinclude/drm/drm_modeset_lock.h
+!Edrivers/gpu/drm/drm_modeset_lock.c
+    </sect2>
+  </sect1>
+
+  <!-- Internals: kms helper functions -->
+
+  <sect1>
+    <title>Mode Setting Helper Functions</title>
+    <para>
+      The plane, CRTC, encoder and connector functions provided by the drivers
+      implement the DRM API. They're called by the DRM core and ioctl handlers
+      to handle device state changes and configuration request. As implementing
+      those functions often requires logic not specific to drivers, mid-layer
+      helper functions are available to avoid duplicating boilerplate code.
+    </para>
+    <para>
+      The DRM core contains one mid-layer implementation. The mid-layer provides
+      implementations of several plane, CRTC, encoder and connector functions
+      (called from the top of the mid-layer) that pre-process requests and call
+      lower-level functions provided by the driver (at the bottom of the
+      mid-layer). For instance, the
+      <function>drm_crtc_helper_set_config</function> function can be used to
+      fill the struct <structname>drm_crtc_funcs</structname>
+      <structfield>set_config</structfield> field. When called, it will split
+      the <methodname>set_config</methodname> operation in smaller, simpler
+      operations and call the driver to handle them.
+    </para>
+    <para>
+      To use the mid-layer, drivers call <function>drm_crtc_helper_add</function>,
+      <function>drm_encoder_helper_add</function> and
+      <function>drm_connector_helper_add</function> functions to install their
+      mid-layer bottom operations handlers, and fill the
+      <structname>drm_crtc_funcs</structname>,
+      <structname>drm_encoder_funcs</structname> and
+      <structname>drm_connector_funcs</structname> structures with pointers to
+      the mid-layer top API functions. Installing the mid-layer bottom operation
+      handlers is best done right after registering the corresponding KMS object.
+    </para>
+    <para>
+      The mid-layer is not split between CRTC, encoder and connector operations.
+      To use it, a driver must provide bottom functions for all of the three KMS
+      entities.
+    </para>
+    <sect2>
+      <title>Helper Functions</title>
+      <itemizedlist>
+        <listitem>
+          <synopsis>int drm_crtc_helper_set_config(struct drm_mode_set *set);</synopsis>
+          <para>
+            The <function>drm_crtc_helper_set_config</function> helper function
+            is a CRTC <methodname>set_config</methodname> implementation. It
+            first tries to locate the best encoder for each connector by calling
+            the connector <methodname>best_encoder</methodname> helper
+            operation.
+          </para>
+          <para>
+            After locating the appropriate encoders, the helper function will
+            call the <methodname>mode_fixup</methodname> encoder and CRTC helper
+            operations to adjust the requested mode, or reject it completely in
+            which case an error will be returned to the application. If the new
+            configuration after mode adjustment is identical to the current
+            configuration the helper function will return without performing any
+            other operation.
+          </para>
+          <para>
+            If the adjusted mode is identical to the current mode but changes to
+            the frame buffer need to be applied, the
+            <function>drm_crtc_helper_set_config</function> function will call
+            the CRTC <methodname>mode_set_base</methodname> helper operation. If
+            the adjusted mode differs from the current mode, or if the
+            <methodname>mode_set_base</methodname> helper operation is not
+            provided, the helper function performs a full mode set sequence by
+            calling the <methodname>prepare</methodname>,
+            <methodname>mode_set</methodname> and
+            <methodname>commit</methodname> CRTC and encoder helper operations,
+            in that order.
+          </para>
+        </listitem>
+        <listitem>
+          <synopsis>void drm_helper_connector_dpms(struct drm_connector *connector, int mode);</synopsis>
+          <para>
+            The <function>drm_helper_connector_dpms</function> helper function
+            is a connector <methodname>dpms</methodname> implementation that
+            tracks power state of connectors. To use the function, drivers must
+            provide <methodname>dpms</methodname> helper operations for CRTCs
+            and encoders to apply the DPMS state to the device.
+          </para>
+          <para>
+            The mid-layer doesn't track the power state of CRTCs and encoders.
+            The <methodname>dpms</methodname> helper operations can thus be
+            called with a mode identical to the currently active mode.
+          </para>
+        </listitem>
+        <listitem>
+          <synopsis>int drm_helper_probe_single_connector_modes(struct drm_connector *connector,
+                                            uint32_t maxX, uint32_t maxY);</synopsis>
+          <para>
+            The <function>drm_helper_probe_single_connector_modes</function> helper
+            function is a connector <methodname>fill_modes</methodname>
+            implementation that updates the connection status for the connector
+            and then retrieves a list of modes by calling the connector
+            <methodname>get_modes</methodname> helper operation.
+          </para>
+         <para>
+            If the helper operation returns no mode, and if the connector status
+            is connector_status_connected, standard VESA DMT modes up to
+            1024x768 are automatically added to the modes list by a call to
+            <function>drm_add_modes_noedid</function>.
+          </para>
+          <para>
+            The function then filters out modes larger than
+            <parameter>max_width</parameter> and <parameter>max_height</parameter>
+            if specified. It finally calls the optional connector
+            <methodname>mode_valid</methodname> helper operation for each mode in
+            the probed list to check whether the mode is valid for the connector.
+          </para>
+        </listitem>
+      </itemizedlist>
+    </sect2>
+    <sect2>
+      <title>CRTC Helper Operations</title>
+      <itemizedlist>
+        <listitem id="drm-helper-crtc-mode-fixup">
+          <synopsis>bool (*mode_fixup)(struct drm_crtc *crtc,
+                       const struct drm_display_mode *mode,
+                       struct drm_display_mode *adjusted_mode);</synopsis>
+          <para>
+            Let CRTCs adjust the requested mode or reject it completely. This
+            operation returns true if the mode is accepted (possibly after being
+            adjusted) or false if it is rejected.
+          </para>
+          <para>
+            The <methodname>mode_fixup</methodname> operation should reject the
+            mode if it can't reasonably use it. The definition of "reasonable"
+            is currently fuzzy in this context. One possible behaviour would be
+            to set the adjusted mode to the panel timings when a fixed-mode
+            panel is used with hardware capable of scaling. Another behaviour
+            would be to accept any input mode and adjust it to the closest mode
+            supported by the hardware (FIXME: This needs to be clarified).
+          </para>
+        </listitem>
+        <listitem>
+          <synopsis>int (*mode_set_base)(struct drm_crtc *crtc, int x, int y,
+                     struct drm_framebuffer *old_fb)</synopsis>
+          <para>
+            Move the CRTC on the current frame buffer (stored in
+            <literal>crtc-&gt;fb</literal>) to position (x,y). Any of the frame
+            buffer, x position or y position may have been modified.
+          </para>
+          <para>
+            This helper operation is optional. If not provided, the
+            <function>drm_crtc_helper_set_config</function> function will fall
+            back to the <methodname>mode_set</methodname> helper operation.
+          </para>
+          <note><para>
+            FIXME: Why are x and y passed as arguments, as they can be accessed
+            through <literal>crtc-&gt;x</literal> and
+            <literal>crtc-&gt;y</literal>?
+          </para></note>
+        </listitem>
+        <listitem>
+          <synopsis>void (*prepare)(struct drm_crtc *crtc);</synopsis>
+          <para>
+            Prepare the CRTC for mode setting. This operation is called after
+            validating the requested mode. Drivers use it to perform
+            device-specific operations required before setting the new mode.
+          </para>
+        </listitem>
+        <listitem>
+          <synopsis>int (*mode_set)(struct drm_crtc *crtc, struct drm_display_mode *mode,
+                struct drm_display_mode *adjusted_mode, int x, int y,
+                struct drm_framebuffer *old_fb);</synopsis>
+          <para>
+            Set a new mode, position and frame buffer. Depending on the device
+            requirements, the mode can be stored internally by the driver and
+            applied in the <methodname>commit</methodname> operation, or
+            programmed to the hardware immediately.
+          </para>
+          <para>
+            The <methodname>mode_set</methodname> operation returns 0 on success
+	    or a negative error code if an error occurs.
+          </para>
+        </listitem>
+        <listitem>
+          <synopsis>void (*commit)(struct drm_crtc *crtc);</synopsis>
+          <para>
+            Commit a mode. This operation is called after setting the new mode.
+            Upon return the device must use the new mode and be fully
+            operational.
+          </para>
+        </listitem>
+      </itemizedlist>
+    </sect2>
+    <sect2>
+      <title>Encoder Helper Operations</title>
+      <itemizedlist>
+        <listitem>
+          <synopsis>bool (*mode_fixup)(struct drm_encoder *encoder,
+                       const struct drm_display_mode *mode,
+                       struct drm_display_mode *adjusted_mode);</synopsis>
+          <para>
+            Let encoders adjust the requested mode or reject it completely. This
+            operation returns true if the mode is accepted (possibly after being
+            adjusted) or false if it is rejected. See the
+            <link linkend="drm-helper-crtc-mode-fixup">mode_fixup CRTC helper
+            operation</link> for an explanation of the allowed adjustments.
+          </para>
+        </listitem>
+        <listitem>
+          <synopsis>void (*prepare)(struct drm_encoder *encoder);</synopsis>
+          <para>
+            Prepare the encoder for mode setting. This operation is called after
+            validating the requested mode. Drivers use it to perform
+            device-specific operations required before setting the new mode.
+          </para>
+        </listitem>
+        <listitem>
+          <synopsis>void (*mode_set)(struct drm_encoder *encoder,
+                 struct drm_display_mode *mode,
+                 struct drm_display_mode *adjusted_mode);</synopsis>
+          <para>
+            Set a new mode. Depending on the device requirements, the mode can
+            be stored internally by the driver and applied in the
+            <methodname>commit</methodname> operation, or programmed to the
+            hardware immediately.
+          </para>
+        </listitem>
+        <listitem>
+          <synopsis>void (*commit)(struct drm_encoder *encoder);</synopsis>
+          <para>
+            Commit a mode. This operation is called after setting the new mode.
+            Upon return the device must use the new mode and be fully
+            operational.
+          </para>
+        </listitem>
+      </itemizedlist>
+    </sect2>
+    <sect2>
+      <title>Connector Helper Operations</title>
+      <itemizedlist>
+        <listitem>
+          <synopsis>struct drm_encoder *(*best_encoder)(struct drm_connector *connector);</synopsis>
+          <para>
+            Return a pointer to the best encoder for the connecter. Device that
+            map connectors to encoders 1:1 simply return the pointer to the
+            associated encoder. This operation is mandatory.
+          </para>
+        </listitem>
+        <listitem>
+          <synopsis>int (*get_modes)(struct drm_connector *connector);</synopsis>
+          <para>
+            Fill the connector's <structfield>probed_modes</structfield> list
+            by parsing EDID data with <function>drm_add_edid_modes</function>,
+            adding standard VESA DMT modes with <function>drm_add_modes_noedid</function>,
+            or calling <function>drm_mode_probed_add</function> directly for every
+            supported mode and return the number of modes it has detected. This
+            operation is mandatory.
+          </para>
+          <para>
+            Note that the caller function will automatically add standard VESA
+            DMT modes up to 1024x768 if the <methodname>get_modes</methodname>
+            helper operation returns no mode and if the connector status is
+            connector_status_connected. There is no need to call
+            <function>drm_add_edid_modes</function> manually in that case.
+          </para>
+          <para>
+            When adding modes manually the driver creates each mode with a call to
+            <function>drm_mode_create</function> and must fill the following fields.
+            <itemizedlist>
+              <listitem>
+                <synopsis>__u32 type;</synopsis>
+                <para>
+                  Mode type bitmask, a combination of
+                  <variablelist>
+                    <varlistentry>
+                      <term>DRM_MODE_TYPE_BUILTIN</term>
+                      <listitem><para>not used?</para></listitem>
+                    </varlistentry>
+                    <varlistentry>
+                      <term>DRM_MODE_TYPE_CLOCK_C</term>
+                      <listitem><para>not used?</para></listitem>
+                    </varlistentry>
+                    <varlistentry>
+                      <term>DRM_MODE_TYPE_CRTC_C</term>
+                      <listitem><para>not used?</para></listitem>
+                    </varlistentry>
+                    <varlistentry>
+                      <term>
+        DRM_MODE_TYPE_PREFERRED - The preferred mode for the connector
+                      </term>
+                      <listitem>
+                        <para>not used?</para>
+                      </listitem>
+                    </varlistentry>
+                    <varlistentry>
+                      <term>DRM_MODE_TYPE_DEFAULT</term>
+                      <listitem><para>not used?</para></listitem>
+                    </varlistentry>
+                    <varlistentry>
+                      <term>DRM_MODE_TYPE_USERDEF</term>
+                      <listitem><para>not used?</para></listitem>
+                    </varlistentry>
+                    <varlistentry>
+                      <term>DRM_MODE_TYPE_DRIVER</term>
+                      <listitem>
+                        <para>
+                          The mode has been created by the driver (as opposed to
+                          to user-created modes).
+                        </para>
+                      </listitem>
+                    </varlistentry>
+                  </variablelist>
+                  Drivers must set the DRM_MODE_TYPE_DRIVER bit for all modes they
+                  create, and set the DRM_MODE_TYPE_PREFERRED bit for the preferred
+                  mode.
+                </para>
+              </listitem>
+              <listitem>
+                <synopsis>__u32 clock;</synopsis>
+                <para>Pixel clock frequency in kHz unit</para>
+              </listitem>
+              <listitem>
+                <synopsis>__u16 hdisplay, hsync_start, hsync_end, htotal;
+    __u16 vdisplay, vsync_start, vsync_end, vtotal;</synopsis>
+                <para>Horizontal and vertical timing information</para>
+                <screen><![CDATA[
+             Active                 Front           Sync           Back
+             Region                 Porch                          Porch
+    <-----------------------><----------------><-------------><-------------->
+
+      //////////////////////|
+     ////////////////////// |
+    //////////////////////  |..................               ................
+                                               _______________
+
+    <----- [hv]display ----->
+    <------------- [hv]sync_start ------------>
+    <--------------------- [hv]sync_end --------------------->
+    <-------------------------------- [hv]total ----------------------------->
+]]></screen>
+              </listitem>
+              <listitem>
+                <synopsis>__u16 hskew;
+    __u16 vscan;</synopsis>
+                <para>Unknown</para>
+              </listitem>
+              <listitem>
+                <synopsis>__u32 flags;</synopsis>
+                <para>
+                  Mode flags, a combination of
+                  <variablelist>
+                    <varlistentry>
+                      <term>DRM_MODE_FLAG_PHSYNC</term>
+                      <listitem><para>
+                        Horizontal sync is active high
+                      </para></listitem>
+                    </varlistentry>
+                    <varlistentry>
+                      <term>DRM_MODE_FLAG_NHSYNC</term>
+                      <listitem><para>
+                        Horizontal sync is active low
+                      </para></listitem>
+                    </varlistentry>
+                    <varlistentry>
+                      <term>DRM_MODE_FLAG_PVSYNC</term>
+                      <listitem><para>
+                        Vertical sync is active high
+                      </para></listitem>
+                    </varlistentry>
+                    <varlistentry>
+                      <term>DRM_MODE_FLAG_NVSYNC</term>
+                      <listitem><para>
+                        Vertical sync is active low
+                      </para></listitem>
+                    </varlistentry>
+                    <varlistentry>
+                      <term>DRM_MODE_FLAG_INTERLACE</term>
+                      <listitem><para>
+                        Mode is interlaced
+                      </para></listitem>
+                    </varlistentry>
+                    <varlistentry>
+                      <term>DRM_MODE_FLAG_DBLSCAN</term>
+                      <listitem><para>
+                        Mode uses doublescan
+                      </para></listitem>
+                    </varlistentry>
+                    <varlistentry>
+                      <term>DRM_MODE_FLAG_CSYNC</term>
+                      <listitem><para>
+                        Mode uses composite sync
+                      </para></listitem>
+                    </varlistentry>
+                    <varlistentry>
+                      <term>DRM_MODE_FLAG_PCSYNC</term>
+                      <listitem><para>
+                        Composite sync is active high
+                      </para></listitem>
+                    </varlistentry>
+                    <varlistentry>
+                      <term>DRM_MODE_FLAG_NCSYNC</term>
+                      <listitem><para>
+                        Composite sync is active low
+                      </para></listitem>
+                    </varlistentry>
+                    <varlistentry>
+                      <term>DRM_MODE_FLAG_HSKEW</term>
+                      <listitem><para>
+                        hskew provided (not used?)
+                      </para></listitem>
+                    </varlistentry>
+                    <varlistentry>
+                      <term>DRM_MODE_FLAG_BCAST</term>
+                      <listitem><para>
+                        not used?
+                      </para></listitem>
+                    </varlistentry>
+                    <varlistentry>
+                      <term>DRM_MODE_FLAG_PIXMUX</term>
+                      <listitem><para>
+                        not used?
+                      </para></listitem>
+                    </varlistentry>
+                    <varlistentry>
+                      <term>DRM_MODE_FLAG_DBLCLK</term>
+                      <listitem><para>
+                        not used?
+                      </para></listitem>
+                    </varlistentry>
+                    <varlistentry>
+                      <term>DRM_MODE_FLAG_CLKDIV2</term>
+                      <listitem><para>
+                        ?
+                      </para></listitem>
+                    </varlistentry>
+                  </variablelist>
+                </para>
+                <para>
+                  Note that modes marked with the INTERLACE or DBLSCAN flags will be
+                  filtered out by
+                  <function>drm_helper_probe_single_connector_modes</function> if
+                  the connector's <structfield>interlace_allowed</structfield> or
+                  <structfield>doublescan_allowed</structfield> field is set to 0.
+                </para>
+              </listitem>
+              <listitem>
+                <synopsis>char name[DRM_DISPLAY_MODE_LEN];</synopsis>
+                <para>
+                  Mode name. The driver must call
+                  <function>drm_mode_set_name</function> to fill the mode name from
+                  <structfield>hdisplay</structfield>,
+                  <structfield>vdisplay</structfield> and interlace flag after
+                  filling the corresponding fields.
+                </para>
+              </listitem>
+            </itemizedlist>
+          </para>
+          <para>
+            The <structfield>vrefresh</structfield> value is computed by
+            <function>drm_helper_probe_single_connector_modes</function>.
+          </para>
+          <para>
+            When parsing EDID data, <function>drm_add_edid_modes</function> fills the
+            connector <structfield>display_info</structfield>
+            <structfield>width_mm</structfield> and
+            <structfield>height_mm</structfield> fields. When creating modes
+            manually the <methodname>get_modes</methodname> helper operation must
+            set the <structfield>display_info</structfield>
+            <structfield>width_mm</structfield> and
+            <structfield>height_mm</structfield> fields if they haven't been set
+            already (for instance at initialization time when a fixed-size panel is
+            attached to the connector). The mode <structfield>width_mm</structfield>
+            and <structfield>height_mm</structfield> fields are only used internally
+            during EDID parsing and should not be set when creating modes manually.
+          </para>
+        </listitem>
+        <listitem>
+          <synopsis>int (*mode_valid)(struct drm_connector *connector,
+		  struct drm_display_mode *mode);</synopsis>
+          <para>
+            Verify whether a mode is valid for the connector. Return MODE_OK for
+            supported modes and one of the enum drm_mode_status values (MODE_*)
+            for unsupported modes. This operation is optional.
+          </para>
+          <para>
+            As the mode rejection reason is currently not used beside for
+            immediately removing the unsupported mode, an implementation can
+            return MODE_BAD regardless of the exact reason why the mode is not
+            valid.
+          </para>
+          <note><para>
+            Note that the <methodname>mode_valid</methodname> helper operation is
+            only called for modes detected by the device, and
+            <emphasis>not</emphasis> for modes set by the user through the CRTC
+            <methodname>set_config</methodname> operation.
+          </para></note>
+        </listitem>
+      </itemizedlist>
+    </sect2>
+    <sect2>
+      <title>Atomic Modeset Helper Functions Reference</title>
+      <sect3>
+	<title>Overview</title>
+!Pdrivers/gpu/drm/drm_atomic_helper.c overview
+      </sect3>
+      <sect3>
+	<title>Implementing Asynchronous Atomic Commit</title>
+!Pdrivers/gpu/drm/drm_atomic_helper.c implementing async commit
+      </sect3>
+      <sect3>
+	<title>Atomic State Reset and Initialization</title>
+!Pdrivers/gpu/drm/drm_atomic_helper.c atomic state reset and initialization
+      </sect3>
+!Iinclude/drm/drm_atomic_helper.h
+!Edrivers/gpu/drm/drm_atomic_helper.c
+    </sect2>
+    <sect2>
+      <title>Modeset Helper Functions Reference</title>
+!Iinclude/drm/drm_crtc_helper.h
+!Edrivers/gpu/drm/drm_crtc_helper.c
+!Pdrivers/gpu/drm/drm_crtc_helper.c overview
+    </sect2>
+    <sect2>
+      <title>Output Probing Helper Functions Reference</title>
+!Pdrivers/gpu/drm/drm_probe_helper.c output probing helper overview
+!Edrivers/gpu/drm/drm_probe_helper.c
+    </sect2>
+    <sect2>
+      <title>fbdev Helper Functions Reference</title>
+!Pdrivers/gpu/drm/drm_fb_helper.c fbdev helpers
+!Edrivers/gpu/drm/drm_fb_helper.c
+!Iinclude/drm/drm_fb_helper.h
+    </sect2>
+    <sect2>
+      <title>Display Port Helper Functions Reference</title>
+!Pdrivers/gpu/drm/drm_dp_helper.c dp helpers
+!Iinclude/drm/drm_dp_helper.h
+!Edrivers/gpu/drm/drm_dp_helper.c
+    </sect2>
+    <sect2>
+      <title>Display Port MST Helper Functions Reference</title>
+!Pdrivers/gpu/drm/drm_dp_mst_topology.c dp mst helper
+!Iinclude/drm/drm_dp_mst_helper.h
+!Edrivers/gpu/drm/drm_dp_mst_topology.c
+    </sect2>
+    <sect2>
+      <title>MIPI DSI Helper Functions Reference</title>
+!Pdrivers/gpu/drm/drm_mipi_dsi.c dsi helpers
+!Iinclude/drm/drm_mipi_dsi.h
+!Edrivers/gpu/drm/drm_mipi_dsi.c
+    </sect2>
+    <sect2>
+      <title>EDID Helper Functions Reference</title>
+!Edrivers/gpu/drm/drm_edid.c
+    </sect2>
+    <sect2>
+      <title>Rectangle Utilities Reference</title>
+!Pinclude/drm/drm_rect.h rect utils
+!Iinclude/drm/drm_rect.h
+!Edrivers/gpu/drm/drm_rect.c
+    </sect2>
+    <sect2>
+      <title>Flip-work Helper Reference</title>
+!Pinclude/drm/drm_flip_work.h flip utils
+!Iinclude/drm/drm_flip_work.h
+!Edrivers/gpu/drm/drm_flip_work.c
+    </sect2>
+    <sect2>
+      <title>HDMI Infoframes Helper Reference</title>
+      <para>
+	Strictly speaking this is not a DRM helper library but generally useable
+	by any driver interfacing with HDMI outputs like v4l or alsa drivers.
+	But it nicely fits into the overall topic of mode setting helper
+	libraries and hence is also included here.
+      </para>
+!Iinclude/linux/hdmi.h
+!Edrivers/video/hdmi.c
+    </sect2>
+    <sect2>
+      <title id="drm-kms-planehelpers">Plane Helper Reference</title>
+!Edrivers/gpu/drm/drm_plane_helper.c
+!Pdrivers/gpu/drm/drm_plane_helper.c overview
+    </sect2>
+    <sect2>
+	  <title>Tile group</title>
+!Pdrivers/gpu/drm/drm_crtc.c Tile group
+    </sect2>
+    <sect2>
+	<title>Bridges</title>
+      <sect3>
+	 <title>Overview</title>
+!Pdrivers/gpu/drm/drm_bridge.c overview
+      </sect3>
+      <sect3>
+	 <title>Default bridge callback sequence</title>
+!Pdrivers/gpu/drm/drm_bridge.c bridge callbacks
+      </sect3>
+!Edrivers/gpu/drm/drm_bridge.c
+    </sect2>
+  </sect1>
+
+  <!-- Internals: kms properties -->
+
+  <sect1 id="drm-kms-properties">
+    <title>KMS Properties</title>
+    <para>
+      Drivers may need to expose additional parameters to applications than
+      those described in the previous sections. KMS supports attaching
+      properties to CRTCs, connectors and planes and offers a userspace API to
+      list, get and set the property values.
+    </para>
+    <para>
+      Properties are identified by a name that uniquely defines the property
+      purpose, and store an associated value. For all property types except blob
+      properties the value is a 64-bit unsigned integer.
+    </para>
+    <para>
+      KMS differentiates between properties and property instances. Drivers
+      first create properties and then create and associate individual instances
+      of those properties to objects. A property can be instantiated multiple
+      times and associated with different objects. Values are stored in property
+      instances, and all other property information are stored in the property
+      and shared between all instances of the property.
+    </para>
+    <para>
+      Every property is created with a type that influences how the KMS core
+      handles the property. Supported property types are
+      <variablelist>
+        <varlistentry>
+          <term>DRM_MODE_PROP_RANGE</term>
+          <listitem><para>Range properties report their minimum and maximum
+            admissible values. The KMS core verifies that values set by
+            application fit in that range.</para></listitem>
+        </varlistentry>
+        <varlistentry>
+          <term>DRM_MODE_PROP_ENUM</term>
+          <listitem><para>Enumerated properties take a numerical value that
+            ranges from 0 to the number of enumerated values defined by the
+            property minus one, and associate a free-formed string name to each
+            value. Applications can retrieve the list of defined value-name pairs
+            and use the numerical value to get and set property instance values.
+            </para></listitem>
+        </varlistentry>
+        <varlistentry>
+          <term>DRM_MODE_PROP_BITMASK</term>
+          <listitem><para>Bitmask properties are enumeration properties that
+            additionally restrict all enumerated values to the 0..63 range.
+            Bitmask property instance values combine one or more of the
+            enumerated bits defined by the property.</para></listitem>
+        </varlistentry>
+        <varlistentry>
+          <term>DRM_MODE_PROP_BLOB</term>
+          <listitem><para>Blob properties store a binary blob without any format
+            restriction. The binary blobs are created as KMS standalone objects,
+            and blob property instance values store the ID of their associated
+            blob object.</para>
+	    <para>Blob properties are only used for the connector EDID property
+	    and cannot be created by drivers.</para></listitem>
+        </varlistentry>
+      </variablelist>
+    </para>
+    <para>
+      To create a property drivers call one of the following functions depending
+      on the property type. All property creation functions take property flags
+      and name, as well as type-specific arguments.
+      <itemizedlist>
+        <listitem>
+          <synopsis>struct drm_property *drm_property_create_range(struct drm_device *dev, int flags,
+                                               const char *name,
+                                               uint64_t min, uint64_t max);</synopsis>
+          <para>Create a range property with the given minimum and maximum
+            values.</para>
+        </listitem>
+        <listitem>
+          <synopsis>struct drm_property *drm_property_create_enum(struct drm_device *dev, int flags,
+                                              const char *name,
+                                              const struct drm_prop_enum_list *props,
+                                              int num_values);</synopsis>
+          <para>Create an enumerated property. The <parameter>props</parameter>
+            argument points to an array of <parameter>num_values</parameter>
+            value-name pairs.</para>
+        </listitem>
+        <listitem>
+          <synopsis>struct drm_property *drm_property_create_bitmask(struct drm_device *dev,
+                                                 int flags, const char *name,
+                                                 const struct drm_prop_enum_list *props,
+                                                 int num_values);</synopsis>
+          <para>Create a bitmask property. The <parameter>props</parameter>
+            argument points to an array of <parameter>num_values</parameter>
+            value-name pairs.</para>
+        </listitem>
+      </itemizedlist>
+    </para>
+    <para>
+      Properties can additionally be created as immutable, in which case they
+      will be read-only for applications but can be modified by the driver. To
+      create an immutable property drivers must set the DRM_MODE_PROP_IMMUTABLE
+      flag at property creation time.
+    </para>
+    <para>
+      When no array of value-name pairs is readily available at property
+      creation time for enumerated or range properties, drivers can create
+      the property using the <function>drm_property_create</function> function
+      and manually add enumeration value-name pairs by calling the
+      <function>drm_property_add_enum</function> function. Care must be taken to
+      properly specify the property type through the <parameter>flags</parameter>
+      argument.
+    </para>
+    <para>
+      After creating properties drivers can attach property instances to CRTC,
+      connector and plane objects by calling the
+      <function>drm_object_attach_property</function>. The function takes a
+      pointer to the target object, a pointer to the previously created property
+      and an initial instance value.
+    </para>
+    <sect2>
+	<title>Existing KMS Properties</title>
+	<para>
+	The following table gives description of drm properties exposed by various
+	modules/drivers.
+	</para>
+	<table border="1" cellpadding="0" cellspacing="0">
+	<tbody>
+	<tr style="font-weight: bold;">
+	<td valign="top" >Owner Module/Drivers</td>
+	<td valign="top" >Group</td>
+	<td valign="top" >Property Name</td>
+	<td valign="top" >Type</td>
+	<td valign="top" >Property Values</td>
+	<td valign="top" >Object attached</td>
+	<td valign="top" >Description/Restrictions</td>
+	</tr>
+	<tr>
+	<td rowspan="37" valign="top" >DRM</td>
+	<td valign="top" >Generic</td>
+	<td valign="top" >“rotation”</td>
+	<td valign="top" >BITMASK</td>
+	<td valign="top" >{ 0, "rotate-0" },
+	{ 1, "rotate-90" },
+	{ 2, "rotate-180" },
+	{ 3, "rotate-270" },
+	{ 4, "reflect-x" },
+	{ 5, "reflect-y" }</td>
+	<td valign="top" >CRTC, Plane</td>
+	<td valign="top" >rotate-(degrees) rotates the image by the specified amount in degrees
+	in counter clockwise direction. reflect-x and reflect-y reflects the
+	image along the specified axis prior to rotation</td>
+	</tr>
+	<tr>
+	<td rowspan="5" valign="top" >Connector</td>
+	<td valign="top" >“EDID”</td>
+	<td valign="top" >BLOB | IMMUTABLE</td>
+	<td valign="top" >0</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >Contains id of edid blob ptr object.</td>
+	</tr>
+	<tr>
+	<td valign="top" >“DPMS”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ “On”, “Standby”, “Suspend”, “Off” }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >Contains DPMS operation mode value.</td>
+	</tr>
+	<tr>
+	<td valign="top" >“PATH”</td>
+	<td valign="top" >BLOB | IMMUTABLE</td>
+	<td valign="top" >0</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >Contains topology path to a connector.</td>
+	</tr>
+	<tr>
+	<td valign="top" >“TILE”</td>
+	<td valign="top" >BLOB | IMMUTABLE</td>
+	<td valign="top" >0</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >Contains tiling information for a connector.</td>
+	</tr>
+	<tr>
+	<td valign="top" >“CRTC_ID”</td>
+	<td valign="top" >OBJECT</td>
+	<td valign="top" >DRM_MODE_OBJECT_CRTC</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >CRTC that connector is attached to (atomic)</td>
+	</tr>
+	<tr>
+	<td rowspan="11" valign="top" >Plane</td>
+	<td valign="top" >“type”</td>
+	<td valign="top" >ENUM | IMMUTABLE</td>
+	<td valign="top" >{ "Overlay", "Primary", "Cursor" }</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >Plane type</td>
+	</tr>
+	<tr>
+	<td valign="top" >“SRC_X”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=UINT_MAX</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >Scanout source x coordinate in 16.16 fixed point (atomic)</td>
+	</tr>
+	<tr>
+	<td valign="top" >“SRC_Y”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=UINT_MAX</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >Scanout source y coordinate in 16.16 fixed point (atomic)</td>
+	</tr>
+	<tr>
+	<td valign="top" >“SRC_W”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=UINT_MAX</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >Scanout source width in 16.16 fixed point (atomic)</td>
+	</tr>
+	<tr>
+	<td valign="top" >“SRC_H”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=UINT_MAX</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >Scanout source height in 16.16 fixed point (atomic)</td>
+	</tr>
+	<tr>
+	<td valign="top" >“CRTC_X”</td>
+	<td valign="top" >SIGNED_RANGE</td>
+	<td valign="top" >Min=INT_MIN, Max=INT_MAX</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >Scanout CRTC (destination) x coordinate (atomic)</td>
+	</tr>
+	<tr>
+	<td valign="top" >“CRTC_Y”</td>
+	<td valign="top" >SIGNED_RANGE</td>
+	<td valign="top" >Min=INT_MIN, Max=INT_MAX</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >Scanout CRTC (destination) y coordinate (atomic)</td>
+	</tr>
+	<tr>
+	<td valign="top" >“CRTC_W”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=UINT_MAX</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >Scanout CRTC (destination) width (atomic)</td>
+	</tr>
+	<tr>
+	<td valign="top" >“CRTC_H”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=UINT_MAX</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >Scanout CRTC (destination) height (atomic)</td>
+	</tr>
+	<tr>
+	<td valign="top" >“FB_ID”</td>
+	<td valign="top" >OBJECT</td>
+	<td valign="top" >DRM_MODE_OBJECT_FB</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >Scanout framebuffer (atomic)</td>
+	</tr>
+	<tr>
+	<td valign="top" >“CRTC_ID”</td>
+	<td valign="top" >OBJECT</td>
+	<td valign="top" >DRM_MODE_OBJECT_CRTC</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >CRTC that plane is attached to (atomic)</td>
+	</tr>
+	<tr>
+	<td rowspan="2" valign="top" >DVI-I</td>
+	<td valign="top" >“subconnector”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ “Unknown”, “DVI-D”, “DVI-A” }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“select subconnector”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ “Automatic”, “DVI-D”, “DVI-A” }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="13" valign="top" >TV</td>
+	<td valign="top" >“subconnector”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "Unknown", "Composite", "SVIDEO", "Component", "SCART" }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“select subconnector”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "Automatic", "Composite", "SVIDEO", "Component", "SCART" }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“mode”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "NTSC_M", "NTSC_J", "NTSC_443", "PAL_B" } etc.</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“left margin”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=100</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“right margin”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=100</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“top margin”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=100</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“bottom margin”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=100</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“brightness”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=100</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“contrast”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=100</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“flicker reduction”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=100</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“overscan”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=100</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“saturation”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=100</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“hue”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=100</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="2" valign="top" >Virtual GPU</td>
+	<td valign="top" >“suggested X”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=0xffffffff</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >property to suggest an X offset for a connector</td>
+	</tr>
+	<tr>
+	<td valign="top" >“suggested Y”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=0xffffffff</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >property to suggest an Y offset for a connector</td>
+	</tr>
+	<tr>
+	<td rowspan="3" valign="top" >Optional</td>
+	<td valign="top" >“scaling mode”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "None", "Full", "Center", "Full aspect" }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"aspect ratio"</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "None", "4:3", "16:9" }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >DRM property to set aspect ratio from user space app.
+		This enum is made generic to allow addition of custom aspect
+		ratios.</td>
+	</tr>
+	<tr>
+	<td valign="top" >“dirty”</td>
+	<td valign="top" >ENUM | IMMUTABLE</td>
+	<td valign="top" >{ "Off", "On", "Annotate" }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="20" valign="top" >i915</td>
+	<td rowspan="2" valign="top" >Generic</td>
+	<td valign="top" >"Broadcast RGB"</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "Automatic", "Full", "Limited 16:235" }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“audio”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "force-dvi", "off", "auto", "on" }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="17" valign="top" >SDVO-TV</td>
+	<td valign="top" >“mode”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "NTSC_M", "NTSC_J", "NTSC_443", "PAL_B" } etc.</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"left_margin"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"right_margin"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"top_margin"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"bottom_margin"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“hpos”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“vpos”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“contrast”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“saturation”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“hue”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“sharpness”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“flicker_filter”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“flicker_filter_adaptive”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“flicker_filter_2d”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“tv_chroma_filter”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“tv_luma_filter”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“dot_crawl”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=1</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >SDVO-TV/LVDS</td>
+	<td valign="top" >“brightness”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="2" valign="top" >CDV gma-500</td>
+	<td rowspan="2" valign="top" >Generic</td>
+	<td valign="top" >"Broadcast RGB"</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ “Full”, “Limited 16:235” }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"Broadcast RGB"</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ “off”, “auto”, “on” }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="19" valign="top" >Poulsbo</td>
+	<td rowspan="1" valign="top" >Generic</td>
+	<td valign="top" >“backlight”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=100</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="17" valign="top" >SDVO-TV</td>
+	<td valign="top" >“mode”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "NTSC_M", "NTSC_J", "NTSC_443", "PAL_B" } etc.</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"left_margin"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"right_margin"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"top_margin"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"bottom_margin"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“hpos”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“vpos”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“contrast”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“saturation”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“hue”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“sharpness”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“flicker_filter”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“flicker_filter_adaptive”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“flicker_filter_2d”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“tv_chroma_filter”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“tv_luma_filter”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“dot_crawl”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=1</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >SDVO-TV/LVDS</td>
+	<td valign="top" >“brightness”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max= SDVO dependent</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="11" valign="top" >armada</td>
+	<td rowspan="2" valign="top" >CRTC</td>
+	<td valign="top" >"CSC_YUV"</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "Auto" , "CCIR601", "CCIR709" }</td>
+	<td valign="top" >CRTC</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"CSC_RGB"</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "Auto", "Computer system", "Studio" }</td>
+	<td valign="top" >CRTC</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="9" valign="top" >Overlay</td>
+	<td valign="top" >"colorkey"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=0xffffff</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"colorkey_min"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=0xffffff</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"colorkey_max"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=0xffffff</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"colorkey_val"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=0xffffff</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"colorkey_alpha"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=0xffffff</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"colorkey_mode"</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "disabled", "Y component", "U component"
+	, "V component", "RGB", “R component", "G component", "B component" }</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"brightness"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=256 + 255</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"contrast"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=0x7fff</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"saturation"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=0x7fff</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="2" valign="top" >exynos</td>
+	<td valign="top" >CRTC</td>
+	<td valign="top" >“mode”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "normal", "blank" }</td>
+	<td valign="top" >CRTC</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >Overlay</td>
+	<td valign="top" >“zpos”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=MAX_PLANE-1</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="2" valign="top" >i2c/ch7006_drv</td>
+	<td valign="top" >Generic</td>
+	<td valign="top" >“scale”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=2</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="1" valign="top" >TV</td>
+	<td valign="top" >“mode”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "PAL", "PAL-M","PAL-N"}, ”PAL-Nc"
+	, "PAL-60", "NTSC-M", "NTSC-J" }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="15" valign="top" >nouveau</td>
+	<td rowspan="6" valign="top" >NV10 Overlay</td>
+	<td valign="top" >"colorkey"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=0x01ffffff</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“contrast”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=8192-1</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“brightness”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=1024</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“hue”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=359</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“saturation”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=8192-1</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“iturbt_709”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=1</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="2" valign="top" >Nv04 Overlay</td>
+	<td valign="top" >“colorkey”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=0x01ffffff</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“brightness”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=1024</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="7" valign="top" >Display</td>
+	<td valign="top" >“dithering mode”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "auto", "off", "on" }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“dithering depth”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "auto", "off", "on", "static 2x2", "dynamic 2x2", "temporal" }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“underscan”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "auto", "6 bpc", "8 bpc" }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“underscan hborder”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=128</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“underscan vborder”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=128</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“vibrant hue”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=180</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >“color vibrance”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=200</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >omap</td>
+	<td valign="top" >Generic</td>
+	<td valign="top" >“zorder”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=3</td>
+	<td valign="top" >CRTC, Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >qxl</td>
+	<td valign="top" >Generic</td>
+	<td valign="top" >“hotplug_mode_update"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=1</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="9" valign="top" >radeon</td>
+	<td valign="top" >DVI-I</td>
+	<td valign="top" >“coherent”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=1</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >DAC enable load detect</td>
+	<td valign="top" >“load detection”</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=1</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >TV Standard</td>
+	<td valign="top" >"tv standard"</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "ntsc", "pal", "pal-m", "pal-60", "ntsc-j"
+	, "scart-pal", "pal-cn", "secam" }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >legacy TMDS PLL detect</td>
+	<td valign="top" >"tmds_pll"</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "driver", "bios" }</td>
+	<td valign="top" >-</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="3" valign="top" >Underscan</td>
+	<td valign="top" >"underscan"</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "off", "on", "auto" }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"underscan hborder"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=128</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"underscan vborder"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=128</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >Audio</td>
+	<td valign="top" >“audio”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "off", "on", "auto" }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >FMT Dithering</td>
+	<td valign="top" >“dither”</td>
+	<td valign="top" >ENUM</td>
+	<td valign="top" >{ "off", "on" }</td>
+	<td valign="top" >Connector</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td rowspan="3" valign="top" >rcar-du</td>
+	<td rowspan="3" valign="top" >Generic</td>
+	<td valign="top" >"alpha"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=255</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"colorkey"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=0, Max=0x01ffffff</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	<tr>
+	<td valign="top" >"zpos"</td>
+	<td valign="top" >RANGE</td>
+	<td valign="top" >Min=1, Max=7</td>
+	<td valign="top" >Plane</td>
+	<td valign="top" >TBD</td>
+	</tr>
+	</tbody>
+	</table>
+    </sect2>
+  </sect1>
+
+  <!-- Internals: vertical blanking -->
+
+  <sect1 id="drm-vertical-blank">
+    <title>Vertical Blanking</title>
+    <para>
+      Vertical blanking plays a major role in graphics rendering. To achieve
+      tear-free display, users must synchronize page flips and/or rendering to
+      vertical blanking. The DRM API offers ioctls to perform page flips
+      synchronized to vertical blanking and wait for vertical blanking.
+    </para>
+    <para>
+      The DRM core handles most of the vertical blanking management logic, which
+      involves filtering out spurious interrupts, keeping race-free blanking
+      counters, coping with counter wrap-around and resets and keeping use
+      counts. It relies on the driver to generate vertical blanking interrupts
+      and optionally provide a hardware vertical blanking counter. Drivers must
+      implement the following operations.
+    </para>
+    <itemizedlist>
+      <listitem>
+        <synopsis>int (*enable_vblank) (struct drm_device *dev, int crtc);
+void (*disable_vblank) (struct drm_device *dev, int crtc);</synopsis>
+        <para>
+	  Enable or disable vertical blanking interrupts for the given CRTC.
+	</para>
+      </listitem>
+      <listitem>
+        <synopsis>u32 (*get_vblank_counter) (struct drm_device *dev, int crtc);</synopsis>
+        <para>
+	  Retrieve the value of the vertical blanking counter for the given
+	  CRTC. If the hardware maintains a vertical blanking counter its value
+	  should be returned. Otherwise drivers can use the
+	  <function>drm_vblank_count</function> helper function to handle this
+	  operation.
+	</para>
+      </listitem>
+    </itemizedlist>
+    <para>
+      Drivers must initialize the vertical blanking handling core with a call to
+      <function>drm_vblank_init</function> in their
+      <methodname>load</methodname> operation. The function will set the struct
+      <structname>drm_device</structname>
+      <structfield>vblank_disable_allowed</structfield> field to 0. This will
+      keep vertical blanking interrupts enabled permanently until the first mode
+      set operation, where <structfield>vblank_disable_allowed</structfield> is
+      set to 1. The reason behind this is not clear. Drivers can set the field
+      to 1 after <function>calling drm_vblank_init</function> to make vertical
+      blanking interrupts dynamically managed from the beginning.
+    </para>
+    <para>
+      Vertical blanking interrupts can be enabled by the DRM core or by drivers
+      themselves (for instance to handle page flipping operations). The DRM core
+      maintains a vertical blanking use count to ensure that the interrupts are
+      not disabled while a user still needs them. To increment the use count,
+      drivers call <function>drm_vblank_get</function>. Upon return vertical
+      blanking interrupts are guaranteed to be enabled.
+    </para>
+    <para>
+      To decrement the use count drivers call
+      <function>drm_vblank_put</function>. Only when the use count drops to zero
+      will the DRM core disable the vertical blanking interrupts after a delay
+      by scheduling a timer. The delay is accessible through the vblankoffdelay
+      module parameter or the <varname>drm_vblank_offdelay</varname> global
+      variable and expressed in milliseconds. Its default value is 5000 ms.
+      Zero means never disable, and a negative value means disable immediately.
+      Drivers may override the behaviour by setting the
+      <structname>drm_device</structname>
+      <structfield>vblank_disable_immediate</structfield> flag, which when set
+      causes vblank interrupts to be disabled immediately regardless of the
+      drm_vblank_offdelay value. The flag should only be set if there's a
+      properly working hardware vblank counter present.
+    </para>
+    <para>
+      When a vertical blanking interrupt occurs drivers only need to call the
+      <function>drm_handle_vblank</function> function to account for the
+      interrupt.
+    </para>
+    <para>
+      Resources allocated by <function>drm_vblank_init</function> must be freed
+      with a call to <function>drm_vblank_cleanup</function> in the driver
+      <methodname>unload</methodname> operation handler.
+    </para>
+    <sect2>
+      <title>Vertical Blanking and Interrupt Handling Functions Reference</title>
+!Edrivers/gpu/drm/drm_irq.c
+!Finclude/drm/drmP.h drm_crtc_vblank_waitqueue
+    </sect2>
+  </sect1>
+
+  <!-- Internals: open/close, file operations and ioctls -->
+
+  <sect1>
+    <title>Open/Close, File Operations and IOCTLs</title>
+    <sect2>
+      <title>Open and Close</title>
+      <synopsis>int (*firstopen) (struct drm_device *);
+void (*lastclose) (struct drm_device *);
+int (*open) (struct drm_device *, struct drm_file *);
+void (*preclose) (struct drm_device *, struct drm_file *);
+void (*postclose) (struct drm_device *, struct drm_file *);</synopsis>
+      <abstract>Open and close handlers. None of those methods are mandatory.
+      </abstract>
+      <para>
+        The <methodname>firstopen</methodname> method is called by the DRM core
+	for legacy UMS (User Mode Setting) drivers only when an application
+	opens a device that has no other opened file handle. UMS drivers can
+	implement it to acquire device resources. KMS drivers can't use the
+	method and must acquire resources in the <methodname>load</methodname>
+	method instead.
+      </para>
+      <para>
+	Similarly the <methodname>lastclose</methodname> method is called when
+	the last application holding a file handle opened on the device closes
+	it, for both UMS and KMS drivers. Additionally, the method is also
+	called at module unload time or, for hot-pluggable devices, when the
+	device is unplugged. The <methodname>firstopen</methodname> and
+	<methodname>lastclose</methodname> calls can thus be unbalanced.
+      </para>
+      <para>
+        The <methodname>open</methodname> method is called every time the device
+	is opened by an application. Drivers can allocate per-file private data
+	in this method and store them in the struct
+	<structname>drm_file</structname> <structfield>driver_priv</structfield>
+	field. Note that the <methodname>open</methodname> method is called
+	before <methodname>firstopen</methodname>.
+      </para>
+      <para>
+        The close operation is split into <methodname>preclose</methodname> and
+	<methodname>postclose</methodname> methods. Drivers must stop and
+	cleanup all per-file operations in the <methodname>preclose</methodname>
+	method. For instance pending vertical blanking and page flip events must
+	be cancelled. No per-file operation is allowed on the file handle after
+	returning from the <methodname>preclose</methodname> method.
+      </para>
+      <para>
+        Finally the <methodname>postclose</methodname> method is called as the
+	last step of the close operation, right before calling the
+	<methodname>lastclose</methodname> method if no other open file handle
+	exists for the device. Drivers that have allocated per-file private data
+	in the <methodname>open</methodname> method should free it here.
+      </para>
+      <para>
+        The <methodname>lastclose</methodname> method should restore CRTC and
+	plane properties to default value, so that a subsequent open of the
+	device will not inherit state from the previous user. It can also be
+	used to execute delayed power switching state changes, e.g. in
+	conjunction with the vga_switcheroo infrastructure. Beyond that KMS
+	drivers should not do any further cleanup. Only legacy UMS drivers might
+	need to clean up device state so that the vga console or an independent
+	fbdev driver could take over.
+      </para>
+    </sect2>
+    <sect2>
+      <title>File Operations</title>
+      <synopsis>const struct file_operations *fops</synopsis>
+      <abstract>File operations for the DRM device node.</abstract>
+      <para>
+        Drivers must define the file operations structure that forms the DRM
+	userspace API entry point, even though most of those operations are
+	implemented in the DRM core. The <methodname>open</methodname>,
+	<methodname>release</methodname> and <methodname>ioctl</methodname>
+	operations are handled by
+	<programlisting>
+	.owner = THIS_MODULE,
+	.open = drm_open,
+	.release = drm_release,
+	.unlocked_ioctl = drm_ioctl,
+  #ifdef CONFIG_COMPAT
+	.compat_ioctl = drm_compat_ioctl,
+  #endif
+        </programlisting>
+      </para>
+      <para>
+        Drivers that implement private ioctls that requires 32/64bit
+	compatibility support must provide their own
+	<methodname>compat_ioctl</methodname> handler that processes private
+	ioctls and calls <function>drm_compat_ioctl</function> for core ioctls.
+      </para>
+      <para>
+        The <methodname>read</methodname> and <methodname>poll</methodname>
+	operations provide support for reading DRM events and polling them. They
+	are implemented by
+	<programlisting>
+	.poll = drm_poll,
+	.read = drm_read,
+	.llseek = no_llseek,
+	</programlisting>
+      </para>
+      <para>
+        The memory mapping implementation varies depending on how the driver
+	manages memory. Pre-GEM drivers will use <function>drm_mmap</function>,
+	while GEM-aware drivers will use <function>drm_gem_mmap</function>. See
+	<xref linkend="drm-gem"/>.
+	<programlisting>
+	.mmap = drm_gem_mmap,
+	</programlisting>
+      </para>
+      <para>
+        No other file operation is supported by the DRM API.
+      </para>
+    </sect2>
+    <sect2>
+      <title>IOCTLs</title>
+      <synopsis>struct drm_ioctl_desc *ioctls;
+int num_ioctls;</synopsis>
+      <abstract>Driver-specific ioctls descriptors table.</abstract>
+      <para>
+        Driver-specific ioctls numbers start at DRM_COMMAND_BASE. The ioctls
+	descriptors table is indexed by the ioctl number offset from the base
+	value. Drivers can use the DRM_IOCTL_DEF_DRV() macro to initialize the
+	table entries.
+      </para>
+      <para>
+        <programlisting>DRM_IOCTL_DEF_DRV(ioctl, func, flags)</programlisting>
+	<para>
+	  <parameter>ioctl</parameter> is the ioctl name. Drivers must define
+	  the DRM_##ioctl and DRM_IOCTL_##ioctl macros to the ioctl number
+	  offset from DRM_COMMAND_BASE and the ioctl number respectively. The
+	  first macro is private to the device while the second must be exposed
+	  to userspace in a public header.
+	</para>
+	<para>
+	  <parameter>func</parameter> is a pointer to the ioctl handler function
+	  compatible with the <type>drm_ioctl_t</type> type.
+	  <programlisting>typedef int drm_ioctl_t(struct drm_device *dev, void *data,
+		struct drm_file *file_priv);</programlisting>
+	</para>
+	<para>
+	  <parameter>flags</parameter> is a bitmask combination of the following
+	  values. It restricts how the ioctl is allowed to be called.
+	  <itemizedlist>
+	    <listitem><para>
+	      DRM_AUTH - Only authenticated callers allowed
+	    </para></listitem>
+	    <listitem><para>
+	      DRM_MASTER - The ioctl can only be called on the master file
+	      handle
+	    </para></listitem>
+            <listitem><para>
+	      DRM_ROOT_ONLY - Only callers with the SYSADMIN capability allowed
+	    </para></listitem>
+            <listitem><para>
+	      DRM_CONTROL_ALLOW - The ioctl can only be called on a control
+	      device
+	    </para></listitem>
+            <listitem><para>
+	      DRM_UNLOCKED - The ioctl handler will be called without locking
+	      the DRM global mutex. This is the enforced default for kms drivers
+	      (i.e. using the DRIVER_MODESET flag) and hence shouldn't be used
+	      any more for new drivers.
+	    </para></listitem>
+	  </itemizedlist>
+	</para>
+      </para>
+!Edrivers/gpu/drm/drm_ioctl.c
+    </sect2>
+  </sect1>
+  <sect1>
+    <title>Legacy Support Code</title>
+    <para>
+      The section very briefly covers some of the old legacy support code which
+      is only used by old DRM drivers which have done a so-called shadow-attach
+      to the underlying device instead of registering as a real driver. This
+      also includes some of the old generic buffer management and command
+      submission code. Do not use any of this in new and modern drivers.
+    </para>
+
+    <sect2>
+      <title>Legacy Suspend/Resume</title>
+      <para>
+	The DRM core provides some suspend/resume code, but drivers wanting full
+	suspend/resume support should provide save() and restore() functions.
+	These are called at suspend, hibernate, or resume time, and should perform
+	any state save or restore required by your device across suspend or
+	hibernate states.
+      </para>
+      <synopsis>int (*suspend) (struct drm_device *, pm_message_t state);
+  int (*resume) (struct drm_device *);</synopsis>
+      <para>
+	Those are legacy suspend and resume methods which
+	<emphasis>only</emphasis> work with the legacy shadow-attach driver
+	registration functions. New driver should use the power management
+	interface provided by their bus type (usually through
+	the struct <structname>device_driver</structname> dev_pm_ops) and set
+	these methods to NULL.
+      </para>
+    </sect2>
+
+    <sect2>
+      <title>Legacy DMA Services</title>
+      <para>
+	This should cover how DMA mapping etc. is supported by the core.
+	These functions are deprecated and should not be used.
+      </para>
+    </sect2>
+  </sect1>
+  </chapter>
+
+<!-- TODO
+
+- Add a glossary
+- Document the struct_mutex catch-all lock
+- Document connector properties
+
+- Why is the load method optional?
+- What are drivers supposed to set the initial display state to, and how?
+  Connector's DPMS states are not initialized and are thus equal to
+  DRM_MODE_DPMS_ON. The fbcon compatibility layer calls
+  drm_helper_disable_unused_functions(), which disables unused encoders and
+  CRTCs, but doesn't touch the connectors' DPMS state, and
+  drm_helper_connector_dpms() in reaction to fbdev blanking events. Do drivers
+  that don't implement (or just don't use) fbcon compatibility need to call
+  those functions themselves?
+- KMS drivers must call drm_vblank_pre_modeset() and drm_vblank_post_modeset()
+  around mode setting. Should this be done in the DRM core?
+- vblank_disable_allowed is set to 1 in the first drm_vblank_post_modeset()
+  call and never set back to 0. It seems to be safe to permanently set it to 1
+  in drm_vblank_init() for KMS driver, and it might be safe for UMS drivers as
+  well. This should be investigated.
+- crtc and connector .save and .restore operations are only used internally in
+  drivers, should they be removed from the core?
+- encoder mid-layer .save and .restore operations are only used internally in
+  drivers, should they be removed from the core?
+- encoder mid-layer .detect operation is only used internally in drivers,
+  should it be removed from the core?
+-->
+
+  <!-- External interfaces -->
+
+  <chapter id="drmExternals">
+    <title>Userland interfaces</title>
+    <para>
+      The DRM core exports several interfaces to applications,
+      generally intended to be used through corresponding libdrm
+      wrapper functions.  In addition, drivers export device-specific
+      interfaces for use by userspace drivers &amp; device-aware
+      applications through ioctls and sysfs files.
+    </para>
+    <para>
+      External interfaces include: memory mapping, context management,
+      DMA operations, AGP management, vblank control, fence
+      management, memory management, and output management.
+    </para>
+    <para>
+      Cover generic ioctls and sysfs layout here.  We only need high-level
+      info, since man pages should cover the rest.
+    </para>
+
+  <!-- External: render nodes -->
+
+    <sect1>
+      <title>Render nodes</title>
+      <para>
+        DRM core provides multiple character-devices for user-space to use.
+        Depending on which device is opened, user-space can perform a different
+        set of operations (mainly ioctls). The primary node is always created
+        and called card&lt;num&gt;. Additionally, a currently
+        unused control node, called controlD&lt;num&gt; is also
+        created. The primary node provides all legacy operations and
+        historically was the only interface used by userspace. With KMS, the
+        control node was introduced. However, the planned KMS control interface
+        has never been written and so the control node stays unused to date.
+      </para>
+      <para>
+        With the increased use of offscreen renderers and GPGPU applications,
+        clients no longer require running compositors or graphics servers to
+        make use of a GPU. But the DRM API required unprivileged clients to
+        authenticate to a DRM-Master prior to getting GPU access. To avoid this
+        step and to grant clients GPU access without authenticating, render
+        nodes were introduced. Render nodes solely serve render clients, that
+        is, no modesetting or privileged ioctls can be issued on render nodes.
+        Only non-global rendering commands are allowed. If a driver supports
+        render nodes, it must advertise it via the DRIVER_RENDER
+        DRM driver capability. If not supported, the primary node must be used
+        for render clients together with the legacy drmAuth authentication
+        procedure.
+      </para>
+      <para>
+        If a driver advertises render node support, DRM core will create a
+        separate render node called renderD&lt;num&gt;. There will
+        be one render node per device. No ioctls except  PRIME-related ioctls
+        will be allowed on this node. Especially GEM_OPEN will be
+        explicitly prohibited. Render nodes are designed to avoid the
+        buffer-leaks, which occur if clients guess the flink names or mmap
+        offsets on the legacy interface. Additionally to this basic interface,
+        drivers must mark their driver-dependent render-only ioctls as
+        DRM_RENDER_ALLOW so render clients can use them. Driver
+        authors must be careful not to allow any privileged ioctls on render
+        nodes.
+      </para>
+      <para>
+        With render nodes, user-space can now control access to the render node
+        via basic file-system access-modes. A running graphics server which
+        authenticates clients on the privileged primary/legacy node is no longer
+        required. Instead, a client can open the render node and is immediately
+        granted GPU access. Communication between clients (or servers) is done
+        via PRIME. FLINK from render node to legacy node is not supported. New
+        clients must not use the insecure FLINK interface.
+      </para>
+      <para>
+        Besides dropping all modeset/global ioctls, render nodes also drop the
+        DRM-Master concept. There is no reason to associate render clients with
+        a DRM-Master as they are independent of any graphics server. Besides,
+        they must work without any running master, anyway.
+        Drivers must be able to run without a master object if they support
+        render nodes. If, on the other hand, a driver requires shared state
+        between clients which is visible to user-space and accessible beyond
+        open-file boundaries, they cannot support render nodes.
+      </para>
+    </sect1>
+
+  <!-- External: vblank handling -->
+
+    <sect1>
+      <title>VBlank event handling</title>
+      <para>
+        The DRM core exposes two vertical blank related ioctls:
+        <variablelist>
+          <varlistentry>
+            <term>DRM_IOCTL_WAIT_VBLANK</term>
+            <listitem>
+              <para>
+                This takes a struct drm_wait_vblank structure as its argument,
+                and it is used to block or request a signal when a specified
+                vblank event occurs.
+              </para>
+            </listitem>
+          </varlistentry>
+          <varlistentry>
+            <term>DRM_IOCTL_MODESET_CTL</term>
+            <listitem>
+              <para>
+		This was only used for user-mode-settind drivers around
+		modesetting changes to allow the kernel to update the vblank
+		interrupt after mode setting, since on many devices the vertical
+		blank counter is reset to 0 at some point during modeset. Modern
+		drivers should not call this any more since with kernel mode
+		setting it is a no-op.
+              </para>
+            </listitem>
+          </varlistentry>
+        </variablelist>
+      </para>
+    </sect1>
+
+  </chapter>
+</part>
+<part id="drmDrivers">
+  <title>DRM Drivers</title>
+
+  <partintro>
+    <para>
+      This second part of the DRM Developer's Guide documents driver code,
+      implementation details and also all the driver-specific userspace
+      interfaces. Especially since all hardware-acceleration interfaces to
+      userspace are driver specific for efficiency and other reasons these
+      interfaces can be rather substantial. Hence every driver has its own
+      chapter.
+    </para>
+  </partintro>
+
+  <chapter id="drmI915">
+    <title>drm/i915 Intel GFX Driver</title>
+    <para>
+      The drm/i915 driver supports all (with the exception of some very early
+      models) integrated GFX chipsets with both Intel display and rendering
+      blocks. This excludes a set of SoC platforms with an SGX rendering unit,
+      those have basic support through the gma500 drm driver.
+    </para>
+    <sect1>
+      <title>Core Driver Infrastructure</title>
+      <para>
+	This section covers core driver infrastructure used by both the display
+	and the GEM parts of the driver.
+      </para>
+      <sect2>
+        <title>Runtime Power Management</title>
+!Pdrivers/gpu/drm/i915/intel_runtime_pm.c runtime pm
+!Idrivers/gpu/drm/i915/intel_runtime_pm.c
+!Idrivers/gpu/drm/i915/intel_uncore.c
+      </sect2>
+      <sect2>
+        <title>Interrupt Handling</title>
+!Pdrivers/gpu/drm/i915/i915_irq.c interrupt handling
+!Fdrivers/gpu/drm/i915/i915_irq.c intel_irq_init intel_irq_init_hw intel_hpd_init
+!Fdrivers/gpu/drm/i915/i915_irq.c intel_runtime_pm_disable_interrupts
+!Fdrivers/gpu/drm/i915/i915_irq.c intel_runtime_pm_enable_interrupts
+      </sect2>
+      <sect2>
+        <title>Intel GVT-g Guest Support(vGPU)</title>
+!Pdrivers/gpu/drm/i915/i915_vgpu.c Intel GVT-g guest support
+!Idrivers/gpu/drm/i915/i915_vgpu.c
+      </sect2>
+    </sect1>
+    <sect1>
+      <title>Display Hardware Handling</title>
+      <para>
+        This section covers everything related to the display hardware including
+        the mode setting infrastructure, plane, sprite and cursor handling and
+        display, output probing and related topics.
+      </para>
+      <sect2>
+        <title>Mode Setting Infrastructure</title>
+        <para>
+          The i915 driver is thus far the only DRM driver which doesn't use the
+          common DRM helper code to implement mode setting sequences. Thus it
+          has its own tailor-made infrastructure for executing a display
+          configuration change.
+        </para>
+      </sect2>
+      <sect2>
+        <title>Frontbuffer Tracking</title>
+!Pdrivers/gpu/drm/i915/intel_frontbuffer.c frontbuffer tracking
+!Idrivers/gpu/drm/i915/intel_frontbuffer.c
+!Fdrivers/gpu/drm/i915/i915_gem.c i915_gem_track_fb
+      </sect2>
+      <sect2>
+        <title>Display FIFO Underrun Reporting</title>
+!Pdrivers/gpu/drm/i915/intel_fifo_underrun.c fifo underrun handling
+!Idrivers/gpu/drm/i915/intel_fifo_underrun.c
+      </sect2>
+      <sect2>
+        <title>Plane Configuration</title>
+        <para>
+	  This section covers plane configuration and composition with the
+	  primary plane, sprites, cursors and overlays. This includes the
+	  infrastructure to do atomic vsync'ed updates of all this state and
+	  also tightly coupled topics like watermark setup and computation,
+	  framebuffer compression and panel self refresh.
+        </para>
+      </sect2>
+      <sect2>
+        <title>Atomic Plane Helpers</title>
+!Pdrivers/gpu/drm/i915/intel_atomic_plane.c atomic plane helpers
+!Idrivers/gpu/drm/i915/intel_atomic_plane.c
+      </sect2>
+      <sect2>
+        <title>Output Probing</title>
+        <para>
+	  This section covers output probing and related infrastructure like the
+	  hotplug interrupt storm detection and mitigation code. Note that the
+	  i915 driver still uses most of the common DRM helper code for output
+	  probing, so those sections fully apply.
+        </para>
+      </sect2>
+      <sect2>
+        <title>Hotplug</title>
+!Pdrivers/gpu/drm/i915/intel_hotplug.c Hotplug
+!Idrivers/gpu/drm/i915/intel_hotplug.c
+      </sect2>
+      <sect2>
+	<title>High Definition Audio</title>
+!Pdrivers/gpu/drm/i915/intel_audio.c High Definition Audio over HDMI and Display Port
+!Idrivers/gpu/drm/i915/intel_audio.c
+      </sect2>
+      <sect2>
+	<title>Panel Self Refresh PSR (PSR/SRD)</title>
+!Pdrivers/gpu/drm/i915/intel_psr.c Panel Self Refresh (PSR/SRD)
+!Idrivers/gpu/drm/i915/intel_psr.c
+      </sect2>
+      <sect2>
+	<title>Frame Buffer Compression (FBC)</title>
+!Pdrivers/gpu/drm/i915/intel_fbc.c Frame Buffer Compression (FBC)
+!Idrivers/gpu/drm/i915/intel_fbc.c
+      </sect2>
+      <sect2>
+        <title>Display Refresh Rate Switching (DRRS)</title>
+!Pdrivers/gpu/drm/i915/intel_dp.c Display Refresh Rate Switching (DRRS)
+!Fdrivers/gpu/drm/i915/intel_dp.c intel_dp_set_drrs_state
+!Fdrivers/gpu/drm/i915/intel_dp.c intel_edp_drrs_enable
+!Fdrivers/gpu/drm/i915/intel_dp.c intel_edp_drrs_disable
+!Fdrivers/gpu/drm/i915/intel_dp.c intel_edp_drrs_invalidate
+!Fdrivers/gpu/drm/i915/intel_dp.c intel_edp_drrs_flush
+!Fdrivers/gpu/drm/i915/intel_dp.c intel_dp_drrs_init
+
+      </sect2>
+      <sect2>
+        <title>DPIO</title>
+!Pdrivers/gpu/drm/i915/i915_reg.h DPIO
+	<table id="dpiox2">
+	  <title>Dual channel PHY (VLV/CHV/BXT)</title>
+	  <tgroup cols="8">
+	    <colspec colname="c0" />
+	    <colspec colname="c1" />
+	    <colspec colname="c2" />
+	    <colspec colname="c3" />
+	    <colspec colname="c4" />
+	    <colspec colname="c5" />
+	    <colspec colname="c6" />
+	    <colspec colname="c7" />
+	    <spanspec spanname="ch0" namest="c0" nameend="c3" />
+	    <spanspec spanname="ch1" namest="c4" nameend="c7" />
+	    <spanspec spanname="ch0pcs01" namest="c0" nameend="c1" />
+	    <spanspec spanname="ch0pcs23" namest="c2" nameend="c3" />
+	    <spanspec spanname="ch1pcs01" namest="c4" nameend="c5" />
+	    <spanspec spanname="ch1pcs23" namest="c6" nameend="c7" />
+	    <thead>
+	      <row>
+		<entry spanname="ch0">CH0</entry>
+		<entry spanname="ch1">CH1</entry>
+	      </row>
+	    </thead>
+	    <tbody valign="top" align="center">
+	      <row>
+		<entry spanname="ch0">CMN/PLL/REF</entry>
+		<entry spanname="ch1">CMN/PLL/REF</entry>
+	      </row>
+	      <row>
+		<entry spanname="ch0pcs01">PCS01</entry>
+		<entry spanname="ch0pcs23">PCS23</entry>
+		<entry spanname="ch1pcs01">PCS01</entry>
+		<entry spanname="ch1pcs23">PCS23</entry>
+	      </row>
+	      <row>
+		<entry>TX0</entry>
+		<entry>TX1</entry>
+		<entry>TX2</entry>
+		<entry>TX3</entry>
+		<entry>TX0</entry>
+		<entry>TX1</entry>
+		<entry>TX2</entry>
+		<entry>TX3</entry>
+	      </row>
+	      <row>
+		<entry spanname="ch0">DDI0</entry>
+		<entry spanname="ch1">DDI1</entry>
+	      </row>
+	    </tbody>
+	  </tgroup>
+	</table>
+	<table id="dpiox1">
+	  <title>Single channel PHY (CHV/BXT)</title>
+	  <tgroup cols="4">
+	    <colspec colname="c0" />
+	    <colspec colname="c1" />
+	    <colspec colname="c2" />
+	    <colspec colname="c3" />
+	    <spanspec spanname="ch0" namest="c0" nameend="c3" />
+	    <spanspec spanname="ch0pcs01" namest="c0" nameend="c1" />
+	    <spanspec spanname="ch0pcs23" namest="c2" nameend="c3" />
+	    <thead>
+	      <row>
+		<entry spanname="ch0">CH0</entry>
+	      </row>
+	    </thead>
+	    <tbody valign="top" align="center">
+	      <row>
+		<entry spanname="ch0">CMN/PLL/REF</entry>
+	      </row>
+	      <row>
+		<entry spanname="ch0pcs01">PCS01</entry>
+		<entry spanname="ch0pcs23">PCS23</entry>
+	      </row>
+	      <row>
+		<entry>TX0</entry>
+		<entry>TX1</entry>
+		<entry>TX2</entry>
+		<entry>TX3</entry>
+	      </row>
+	      <row>
+		<entry spanname="ch0">DDI2</entry>
+	      </row>
+	    </tbody>
+	  </tgroup>
+	</table>
+      </sect2>
+
+      <sect2>
+       <title>CSR firmware support for DMC</title>
+!Pdrivers/gpu/drm/i915/intel_csr.c csr support for dmc
+!Idrivers/gpu/drm/i915/intel_csr.c
+      </sect2>
+    </sect1>
+
+    <sect1>
+      <title>Memory Management and Command Submission</title>
+      <para>
+	This sections covers all things related to the GEM implementation in the
+	i915 driver.
+      </para>
+      <sect2>
+        <title>Batchbuffer Parsing</title>
+!Pdrivers/gpu/drm/i915/i915_cmd_parser.c batch buffer command parser
+!Idrivers/gpu/drm/i915/i915_cmd_parser.c
+      </sect2>
+      <sect2>
+        <title>Batchbuffer Pools</title>
+!Pdrivers/gpu/drm/i915/i915_gem_batch_pool.c batch pool
+!Idrivers/gpu/drm/i915/i915_gem_batch_pool.c
+      </sect2>
+      <sect2>
+        <title>Logical Rings, Logical Ring Contexts and Execlists</title>
+!Pdrivers/gpu/drm/i915/intel_lrc.c Logical Rings, Logical Ring Contexts and Execlists
+!Idrivers/gpu/drm/i915/intel_lrc.c
+      </sect2>
+      <sect2>
+        <title>Global GTT views</title>
+!Pdrivers/gpu/drm/i915/i915_gem_gtt.c Global GTT views
+!Idrivers/gpu/drm/i915/i915_gem_gtt.c
+      </sect2>
+      <sect2>
+        <title>GTT Fences and Swizzling</title>
+!Idrivers/gpu/drm/i915/i915_gem_fence.c
+        <sect3>
+          <title>Global GTT Fence Handling</title>
+!Pdrivers/gpu/drm/i915/i915_gem_fence.c fence register handling
+        </sect3>
+        <sect3>
+          <title>Hardware Tiling and Swizzling Details</title>
+!Pdrivers/gpu/drm/i915/i915_gem_fence.c tiling swizzling details
+        </sect3>
+      </sect2>
+      <sect2>
+        <title>Object Tiling IOCTLs</title>
+!Idrivers/gpu/drm/i915/i915_gem_tiling.c
+!Pdrivers/gpu/drm/i915/i915_gem_tiling.c buffer object tiling
+      </sect2>
+      <sect2>
+        <title>Buffer Object Eviction</title>
+	<para>
+	  This section documents the interface functions for evicting buffer
+	  objects to make space available in the virtual gpu address spaces.
+	  Note that this is mostly orthogonal to shrinking buffer objects
+	  caches, which has the goal to make main memory (shared with the gpu
+	  through the unified memory architecture) available.
+	</para>
+!Idrivers/gpu/drm/i915/i915_gem_evict.c
+      </sect2>
+      <sect2>
+        <title>Buffer Object Memory Shrinking</title>
+	<para>
+	  This section documents the interface function for shrinking memory
+	  usage of buffer object caches. Shrinking is used to make main memory
+	  available.  Note that this is mostly orthogonal to evicting buffer
+	  objects, which has the goal to make space in gpu virtual address
+	  spaces.
+	</para>
+!Idrivers/gpu/drm/i915/i915_gem_shrinker.c
+      </sect2>
+    </sect1>
+    <sect1>
+      <title> Tracing </title>
+      <para>
+    This sections covers all things related to the tracepoints implemented in
+    the i915 driver.
+      </para>
+      <sect2>
+        <title> i915_ppgtt_create and i915_ppgtt_release </title>
+!Pdrivers/gpu/drm/i915/i915_trace.h i915_ppgtt_create and i915_ppgtt_release tracepoints
+      </sect2>
+      <sect2>
+        <title> i915_context_create and i915_context_free </title>
+!Pdrivers/gpu/drm/i915/i915_trace.h i915_context_create and i915_context_free tracepoints
+      </sect2>
+      <sect2>
+        <title> switch_mm </title>
+!Pdrivers/gpu/drm/i915/i915_trace.h switch_mm tracepoint
+      </sect2>
+    </sect1>
+
+  </chapter>
+!Cdrivers/gpu/drm/i915/i915_irq.c
+</part>
+</book>
-- 
cgit v0.10.2


From 7f81707429fe52b2842f0bfb6dac62329ea1d96f Mon Sep 17 00:00:00 2001
From: Lukas Wunner <lukas@wunner.de>
Date: Sun, 11 Oct 2015 11:26:26 +0200
Subject: gpu/doc: Fix up remaining occurrences of old document title

Following Daniel's renaming of the document.

Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/Documentation/DocBook/gpu.tmpl b/Documentation/DocBook/gpu.tmpl
index 48e3ab4..ab51cb3 100644
--- a/Documentation/DocBook/gpu.tmpl
+++ b/Documentation/DocBook/gpu.tmpl
@@ -2,7 +2,7 @@
 <!DOCTYPE book PUBLIC "-//OASIS//DTD DocBook XML V4.1.2//EN"
 	"http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd" []>
 
-<book id="drmDevelopersGuide">
+<book id="gpuDevelopersGuide">
   <bookinfo>
     <title>Linux GPU Driver Developer's Guide</title>
 
@@ -78,9 +78,9 @@
   <title>DRM Core</title>
   <partintro>
     <para>
-      This first part of the DRM Developer's Guide documents core DRM code,
-      helper libraries for writing drivers and generic userspace interfaces
-      exposed by DRM drivers.
+      This first part of the GPU Driver Developer's Guide documents core DRM
+      code, helper libraries for writing drivers and generic userspace
+      interfaces exposed by DRM drivers.
     </para>
   </partintro>
 
@@ -3889,8 +3889,8 @@ int num_ioctls;</synopsis>
 
   <partintro>
     <para>
-      This second part of the DRM Developer's Guide documents driver code,
-      implementation details and also all the driver-specific userspace
+      This second part of the GPU Driver Developer's Guide documents driver
+      code, implementation details and also all the driver-specific userspace
       interfaces. Especially since all hardware-acceleration interfaces to
       userspace are driver specific for efficiency and other reasons these
       interfaces can be rather substantial. Hence every driver has its own
-- 
cgit v0.10.2


From 6648f4879a12cd428c8584e03b41576521b7198f Mon Sep 17 00:00:00 2001
From: Lukas Wunner <lukas@wunner.de>
Date: Sun, 11 Oct 2015 11:55:00 +0200
Subject: gpu/doc: Add vga_switcheroo documentation

Requires Markdown support.

Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/Documentation/DocBook/gpu.tmpl b/Documentation/DocBook/gpu.tmpl
index ab51cb3..7f90c1e 100644
--- a/Documentation/DocBook/gpu.tmpl
+++ b/Documentation/DocBook/gpu.tmpl
@@ -40,6 +40,16 @@
 	  </address>
 	</affiliation>
       </author>
+      <author>
+	<firstname>Lukas</firstname>
+	<surname>Wunner</surname>
+	<contrib>vga_switcheroo documentation</contrib>
+	<affiliation>
+	  <address>
+	    <email>lukas@wunner.de</email>
+	  </address>
+	</affiliation>
+      </author>
     </authorgroup>
 
     <copyright>
@@ -51,6 +61,10 @@
       <year>2012</year>
       <holder>Laurent Pinchart</holder>
     </copyright>
+    <copyright>
+      <year>2015</year>
+      <holder>Lukas Wunner</holder>
+    </copyright>
 
     <legalnotice>
       <para>
@@ -69,6 +83,13 @@
 	<revremark>Added extensive documentation about driver internals.
 	</revremark>
       </revision>
+      <revision>
+	<revnumber>1.1</revnumber>
+	<date>2015-10-11</date>
+	<authorinitials>LW</authorinitials>
+	<revremark>Added vga_switcheroo documentation.
+	</revremark>
+      </revision>
     </revhistory>
   </bookinfo>
 
@@ -3583,10 +3604,11 @@ void (*postclose) (struct drm_device *, struct drm_file *);</synopsis>
 	plane properties to default value, so that a subsequent open of the
 	device will not inherit state from the previous user. It can also be
 	used to execute delayed power switching state changes, e.g. in
-	conjunction with the vga_switcheroo infrastructure. Beyond that KMS
-	drivers should not do any further cleanup. Only legacy UMS drivers might
-	need to clean up device state so that the vga console or an independent
-	fbdev driver could take over.
+	conjunction with the vga_switcheroo infrastructure (see
+	<xref linkend="vga_switcheroo"/>). Beyond that KMS drivers should not
+	do any further cleanup. Only legacy UMS drivers might need to clean up
+	device state so that the vga console or an independent fbdev driver
+	could take over.
       </para>
     </sect2>
     <sect2>
@@ -4200,4 +4222,50 @@ int num_ioctls;</synopsis>
   </chapter>
 !Cdrivers/gpu/drm/i915/i915_irq.c
 </part>
+
+<part id="vga_switcheroo">
+  <title>vga_switcheroo</title>
+  <partintro>
+!Pdrivers/gpu/vga/vga_switcheroo.c Overview
+  </partintro>
+
+  <chapter id="modes_of_use">
+    <title>Modes of Use</title>
+  <sect1>
+    <title>Manual switching and manual power control</title>
+!Pdrivers/gpu/vga/vga_switcheroo.c Manual switching and manual power control
+  </sect1>
+  <sect1>
+    <title>Driver power control</title>
+!Pdrivers/gpu/vga/vga_switcheroo.c Driver power control
+  </sect1>
+  </chapter>
+
+  <chapter id="pubfunctions">
+    <title>Public functions</title>
+!Edrivers/gpu/vga/vga_switcheroo.c
+  </chapter>
+
+  <chapter id="pubstructures">
+    <title>Public structures</title>
+!Finclude/linux/vga_switcheroo.h vga_switcheroo_handler
+!Finclude/linux/vga_switcheroo.h vga_switcheroo_client_ops
+  </chapter>
+
+  <chapter id="pubconstants">
+    <title>Public constants</title>
+!Finclude/linux/vga_switcheroo.h vga_switcheroo_client_id
+!Finclude/linux/vga_switcheroo.h vga_switcheroo_state
+  </chapter>
+
+  <chapter id="privstructures">
+    <title>Private structures</title>
+!Fdrivers/gpu/vga/vga_switcheroo.c vgasr_priv
+!Fdrivers/gpu/vga/vga_switcheroo.c vga_switcheroo_client
+  </chapter>
+
+!Cdrivers/gpu/vga/vga_switcheroo.c
+!Cinclude/linux/vga_switcheroo.h
+</part>
+
 </book>
-- 
cgit v0.10.2


From f71a6d6095782186c10c720d9ed813b68275d30d Mon Sep 17 00:00:00 2001
From: Lukas Wunner <lukas@wunner.de>
Date: Mon, 12 Oct 2015 09:10:33 +0200
Subject: gpu/doc: Convert to markdown harder

This snippet...

    * Lock VMA manager for extended lookups. Only *_locked() VMA function calls
    * are allowed while holding this lock. All other contexts are blocked from VMA
    * until the lock is released via drm_vma_offset_unlock_lookup().

...causes markdown-enabled kernel-doc to barf:

    debian/build/build-doc/Documentation/DocBook/gpu.aux.xml:3247: parser error : Opening and ending tag mismatch: emphasis line 3247 and function
       *<function><emphasis>locked</function> VMA function calls are allowed while
                                             ^
    /root/airlied/debian/build/build-doc/Documentation/DocBook/gpu.aux.xml:3249: parser error : Opening and ending tag mismatch: function line 3249 and emphasis
       released via <function>drm</emphasis>vma_offset_unlock_lookup</function>.
                                            ^
    unable to parse /root/airlied/debian/build/build-doc/Documentation/DocBook/gpu.aux.xml

A quick workaround is to replace *_locked() by X_locked().

Cc: Danilo Cesar Lemes de Paula <danilo.cesar@collabora.co.uk>
Signed-off-by: Lukas Wunner <lukas@wunner.de>
[danvet: Just drop the X_ too, the usual style is _unlocked, except
that _ seems to be what annoys markdown.]
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/include/drm/drm_vma_manager.h b/include/drm/drm_vma_manager.h
index 8cd402c..089cb73 100644
--- a/include/drm/drm_vma_manager.h
+++ b/include/drm/drm_vma_manager.h
@@ -97,7 +97,7 @@ drm_vma_offset_exact_lookup(struct drm_vma_offset_manager *mgr,
  * drm_vma_offset_lock_lookup() - Lock lookup for extended private use
  * @mgr: Manager object
  *
- * Lock VMA manager for extended lookups. Only *_locked() VMA function calls
+ * Lock VMA manager for extended lookups. Only locked VMA function calls
  * are allowed while holding this lock. All other contexts are blocked from VMA
  * until the lock is released via drm_vma_offset_unlock_lookup().
  *
-- 
cgit v0.10.2


From 22375f3e79b5b9e1dcbac998fd8473484d391d93 Mon Sep 17 00:00:00 2001
From: Mikko Rapeli <mikko.rapeli@iki.fi>
Date: Thu, 15 Oct 2015 07:55:49 +0200
Subject: savage_drm.h: include <drm/drm.h>

Fixes compiler error:

drm/savage_drm.h:50:24: error: array type has incomplete element type
  struct drm_tex_region texList[SAVAGE_NR_TEX_HEAPS][SAVAGE_NR_TEX_REGIONS +

Signed-off-by: Mikko Rapeli <mikko.rapeli@iki.fi>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/include/uapi/drm/savage_drm.h b/include/uapi/drm/savage_drm.h
index 818d49b..9dc9dc1 100644
--- a/include/uapi/drm/savage_drm.h
+++ b/include/uapi/drm/savage_drm.h
@@ -26,6 +26,8 @@
 #ifndef __SAVAGE_DRM_H__
 #define __SAVAGE_DRM_H__
 
+#include <drm/drm.h>
+
 #ifndef __SAVAGE_SAREA_DEFINES__
 #define __SAVAGE_SAREA_DEFINES__
 
-- 
cgit v0.10.2


From 0aa4a4b8200c9e47021e8bd2d54b3d20ca407640 Mon Sep 17 00:00:00 2001
From: Mikko Rapeli <mikko.rapeli@iki.fi>
Date: Thu, 15 Oct 2015 07:55:46 +0200
Subject: r128_drm.h: include drm/drm.h

Fixes compile error:

drm/r128_drm.h:156:23: error: array type has incomplete element type
  struct drm_clip_rect boxes[R128_NR_SAREA_CLIPRECTS];

Signed-off-by: Mikko Rapeli <mikko.rapeli@iki.fi>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/include/uapi/drm/r128_drm.h b/include/uapi/drm/r128_drm.h
index 8d8878b..76b0aa3 100644
--- a/include/uapi/drm/r128_drm.h
+++ b/include/uapi/drm/r128_drm.h
@@ -33,6 +33,8 @@
 #ifndef __R128_DRM_H__
 #define __R128_DRM_H__
 
+#include <drm/drm.h>
+
 /* WARNING: If you change any of these defines, make sure to change the
  * defines in the X server file (r128_sarea.h)
  */
-- 
cgit v0.10.2


From 70d994704e2c93a3d6be1de0c9f103fc7c186022 Mon Sep 17 00:00:00 2001
From: Mikko Rapeli <mikko.rapeli@iki.fi>
Date: Thu, 15 Oct 2015 07:55:51 +0200
Subject: drm/i810_drm.h: include drm/drm.h

Fixes userspace compilation error:

error: array type has incomplete element type
struct drm_clip_rect boxes[I810_NR_SAREA_CLIPRECTS];

Signed-off-by: Mikko Rapeli <mikko.rapeli@iki.fi>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/include/uapi/drm/i810_drm.h b/include/uapi/drm/i810_drm.h
index 7a10bb6..34736ef 100644
--- a/include/uapi/drm/i810_drm.h
+++ b/include/uapi/drm/i810_drm.h
@@ -1,6 +1,8 @@
 #ifndef _I810_DRM_H_
 #define _I810_DRM_H_
 
+#include <drm/drm.h>
+
 /* WARNING: These defines must be the same as what the Xserver uses.
  * if you change them, you must change the defines in the Xserver.
  */
-- 
cgit v0.10.2


From 4e270f088011c6954034d6c4b5453e5cd7e02c7a Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Thu, 15 Oct 2015 09:36:24 +0200
Subject: drm/gem: Drop struct_mutex requirement from drm_gem_mmap_obj

Since

commit 131e663bd6f1055caaff128f9aa5071d227eeb72
Author: Daniel Vetter <daniel.vetter@ffwll.ch>
Date:   Thu Jul 9 23:32:33 2015 +0200

    drm/gem: rip out drm vma accounting for gem mmaps

there is no need for this any more.

v2: Fixup compile noise spotted by 0-day build.

Link: http://mid.gmane.org/1444894601-5200-9-git-send-email-daniel.vetter@ffwll.ch
Reviewed-by: David Herrmann <dh.herrmann@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c
index 3c2d4ab..7dc4a8a 100644
--- a/drivers/gpu/drm/drm_gem.c
+++ b/drivers/gpu/drm/drm_gem.c
@@ -810,8 +810,6 @@ EXPORT_SYMBOL(drm_gem_vm_close);
  * drm_gem_mmap() prevents unprivileged users from mapping random objects. So
  * callers must verify access restrictions before calling this helper.
  *
- * NOTE: This function has to be protected with dev->struct_mutex
- *
  * Return 0 or success or -EINVAL if the object size is smaller than the VMA
  * size, or if no gem_vm_ops are provided.
  */
@@ -820,8 +818,6 @@ int drm_gem_mmap_obj(struct drm_gem_object *obj, unsigned long obj_size,
 {
 	struct drm_device *dev = obj->dev;
 
-	lockdep_assert_held(&dev->struct_mutex);
-
 	/* Check for valid size. */
 	if (obj_size < vma->vm_end - vma->vm_start)
 		return -EINVAL;
diff --git a/drivers/gpu/drm/drm_gem_cma_helper.c b/drivers/gpu/drm/drm_gem_cma_helper.c
index 86cc793..4fb4c45 100644
--- a/drivers/gpu/drm/drm_gem_cma_helper.c
+++ b/drivers/gpu/drm/drm_gem_cma_helper.c
@@ -484,9 +484,7 @@ int drm_gem_cma_prime_mmap(struct drm_gem_object *obj,
 	struct drm_device *dev = obj->dev;
 	int ret;
 
-	mutex_lock(&dev->struct_mutex);
 	ret = drm_gem_mmap_obj(obj, obj->size, vma);
-	mutex_unlock(&dev->struct_mutex);
 	if (ret < 0)
 		return ret;
 
diff --git a/drivers/gpu/drm/msm/msm_fbdev.c b/drivers/gpu/drm/msm/msm_fbdev.c
index f97a196..3f6ec07 100644
--- a/drivers/gpu/drm/msm/msm_fbdev.c
+++ b/drivers/gpu/drm/msm/msm_fbdev.c
@@ -68,12 +68,7 @@ static int msm_fbdev_mmap(struct fb_info *info, struct vm_area_struct *vma)
 	if (drm_device_is_unplugged(dev))
 		return -ENODEV;
 
-	mutex_lock(&dev->struct_mutex);
-
 	ret = drm_gem_mmap_obj(drm_obj, drm_obj->size, vma);
-
-	mutex_unlock(&dev->struct_mutex);
-
 	if (ret) {
 		pr_err("%s:drm_gem_mmap_obj fail\n", __func__);
 		return ret;
diff --git a/drivers/gpu/drm/msm/msm_gem_prime.c b/drivers/gpu/drm/msm/msm_gem_prime.c
index 831461b..121975b 100644
--- a/drivers/gpu/drm/msm/msm_gem_prime.c
+++ b/drivers/gpu/drm/msm/msm_gem_prime.c
@@ -45,9 +45,7 @@ int msm_gem_prime_mmap(struct drm_gem_object *obj, struct vm_area_struct *vma)
 {
 	int ret;
 
-	mutex_lock(&obj->dev->struct_mutex);
 	ret = drm_gem_mmap_obj(obj, obj->size, vma);
-	mutex_unlock(&obj->dev->struct_mutex);
 	if (ret < 0)
 		return ret;
 
diff --git a/drivers/gpu/drm/omapdrm/omap_gem_dmabuf.c b/drivers/gpu/drm/omapdrm/omap_gem_dmabuf.c
index 0cc71c9..27c2976 100644
--- a/drivers/gpu/drm/omapdrm/omap_gem_dmabuf.c
+++ b/drivers/gpu/drm/omapdrm/omap_gem_dmabuf.c
@@ -140,15 +140,12 @@ static int omap_gem_dmabuf_mmap(struct dma_buf *buffer,
 		struct vm_area_struct *vma)
 {
 	struct drm_gem_object *obj = buffer->priv;
-	struct drm_device *dev = obj->dev;
 	int ret = 0;
 
 	if (WARN_ON(!obj->filp))
 		return -EINVAL;
 
-	mutex_lock(&dev->struct_mutex);
 	ret = drm_gem_mmap_obj(obj, omap_gem_mmap_size(obj), vma);
-	mutex_unlock(&dev->struct_mutex);
 	if (ret < 0)
 		return ret;
 
diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_gem.c b/drivers/gpu/drm/rockchip/rockchip_drm_gem.c
index a6d9104..8caea0a 100644
--- a/drivers/gpu/drm/rockchip/rockchip_drm_gem.c
+++ b/drivers/gpu/drm/rockchip/rockchip_drm_gem.c
@@ -79,12 +79,9 @@ static int rockchip_drm_gem_object_mmap(struct drm_gem_object *obj,
 int rockchip_gem_mmap_buf(struct drm_gem_object *obj,
 			  struct vm_area_struct *vma)
 {
-	struct drm_device *drm = obj->dev;
 	int ret;
 
-	mutex_lock(&drm->struct_mutex);
 	ret = drm_gem_mmap_obj(obj, obj->size, vma);
-	mutex_unlock(&drm->struct_mutex);
 	if (ret)
 		return ret;
 
-- 
cgit v0.10.2


From ef4c6270bf2867e2f8032e9614d1a8cfc6c71663 Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Thu, 15 Oct 2015 09:36:25 +0200
Subject: drm/gem: Check locking in drm_gem_object_unreference

Pretty soon only some drivers will need dev->struct_mutex in their
gem_free_object callbacks. Hence it's really important to make sure
everything still keeps getting this right.

v2: Don't check for locking before we check for non-NULL obj. Spotted
by Dan Carpenter.

Link: http://mid.gmane.org/1444894601-5200-10-git-send-email-daniel.vetter@ffwll.ch
Reviewed-by: David Herrmann <dh.herrmann@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

diff --git a/include/drm/drm_gem.h b/include/drm/drm_gem.h
index 7a592d7..15e7f00 100644
--- a/include/drm/drm_gem.h
+++ b/include/drm/drm_gem.h
@@ -142,8 +142,11 @@ drm_gem_object_reference(struct drm_gem_object *obj)
 static inline void
 drm_gem_object_unreference(struct drm_gem_object *obj)
 {
-	if (obj != NULL)
+	if (obj != NULL) {
+		WARN_ON(!mutex_is_locked(&obj->dev->struct_mutex));
+
 		kref_put(&obj->refcount, drm_gem_object_free);
+	}
 }
 
 static inline void
-- 
cgit v0.10.2


From 6ff774bd472dcbe77df63ab8044cd9cf65535814 Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Thu, 15 Oct 2015 09:36:26 +0200
Subject: drm/gem: Use container_of in drm_gem_object_free

Just a random thing I spotted while reading code - better safe than
sorry.

Link: http://mid.gmane.org/1444894601-5200-11-git-send-email-daniel.vetter@ffwll.ch
Reviewed-by: David Herrmann <dh.herrmann@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c
index 7dc4a8a..ab8ea422 100644
--- a/drivers/gpu/drm/drm_gem.c
+++ b/drivers/gpu/drm/drm_gem.c
@@ -763,7 +763,8 @@ EXPORT_SYMBOL(drm_gem_object_release);
 void
 drm_gem_object_free(struct kref *kref)
 {
-	struct drm_gem_object *obj = (struct drm_gem_object *) kref;
+	struct drm_gem_object *obj =
+		container_of(kref, struct drm_gem_object, refcount);
 	struct drm_device *dev = obj->dev;
 
 	WARN_ON(!mutex_is_locked(&dev->struct_mutex));
-- 
cgit v0.10.2


From 3c67d839b30c7d6d6ab5c6fddac0f58ec8095d50 Mon Sep 17 00:00:00 2001
From: Lukas Wunner <lukas@wunner.de>
Date: Thu, 15 Oct 2015 11:56:56 +0200
Subject: drm: Fix return value of drm_framebuffer_init()

In its original version, drm_framebuffer_init() returned a negative int
if drm_mode_object_get() failed (f453ba046074, "DRM: add mode setting
support").

This was accidentally disabled by commit 4b096ac10da0 ("drm: revamp
locking around fb creation/destruction"). Thus, drm_framebuffer_init()
pretends success if drm_mode_object_get() failed.

Reinstate the original behaviour. Also fix erroneous kernel-doc of
drm_mode_object_get().

Fixes: 4b096ac10da0 ("drm: revamp locking around fb creation/
    destruction")
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c
index e7c8422..6058f4b 100644
--- a/drivers/gpu/drm/drm_crtc.c
+++ b/drivers/gpu/drm/drm_crtc.c
@@ -306,8 +306,7 @@ static int drm_mode_object_get_reg(struct drm_device *dev,
  * reference counted modeset objects like framebuffers.
  *
  * Returns:
- * New unique (relative to other objects in @dev) integer identifier for the
- * object.
+ * Zero on success, error code on failure.
  */
 int drm_mode_object_get(struct drm_device *dev,
 			struct drm_mode_object *obj, uint32_t obj_type)
@@ -423,7 +422,7 @@ int drm_framebuffer_init(struct drm_device *dev, struct drm_framebuffer *fb,
 out:
 	mutex_unlock(&dev->mode_config.fb_lock);
 
-	return 0;
+	return ret;
 }
 EXPORT_SYMBOL(drm_framebuffer_init);
 
-- 
cgit v0.10.2


From 3d57b42cabc8472ab63f0adc9529102314218f1e Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Thu, 15 Oct 2015 09:36:23 +0200
Subject: drm/vgem: Drop vgem_drm_gem_mmap

It's duplicating (without using some of the helpers) drm_gem_mmap with
the addition that it can redirect to drm-buf mmap support. But prime
import/export was dropped in

commit 990ed2720717173bbdea4cfb2bad37cc7aa91495
Author: Rob Clark <robdclark@gmail.com>
Date:   Thu May 21 11:58:30 2015 -0400

    drm/vgem: drop DRIVER_PRIME (v2)

for now, so this is dead code. And since I want to rework the locking
for drm_gem_mmap it seems simpler to de-dupe this code for now and
then start over with the reworked one again, if we want to resurrect
this all indeed.

Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Link: http://mid.gmane.org/1444894601-5200-8-git-send-email-daniel.vetter@ffwll.ch
Acked-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/vgem/vgem_drv.c b/drivers/gpu/drm/vgem/vgem_drv.c
index 860062e..c503a84 100644
--- a/drivers/gpu/drm/vgem/vgem_drv.c
+++ b/drivers/gpu/drm/vgem/vgem_drv.c
@@ -235,66 +235,13 @@ unlock:
 	return ret;
 }
 
-int vgem_drm_gem_mmap(struct file *filp, struct vm_area_struct *vma)
-{
-	struct drm_file *priv = filp->private_data;
-	struct drm_device *dev = priv->minor->dev;
-	struct drm_vma_offset_node *node;
-	struct drm_gem_object *obj;
-	struct drm_vgem_gem_object *vgem_obj;
-	int ret = 0;
-
-	mutex_lock(&dev->struct_mutex);
-
-	node = drm_vma_offset_exact_lookup(dev->vma_offset_manager,
-					   vma->vm_pgoff,
-					   vma_pages(vma));
-	if (!node) {
-		ret = -EINVAL;
-		goto out_unlock;
-	} else if (!drm_vma_node_is_allowed(node, filp)) {
-		ret = -EACCES;
-		goto out_unlock;
-	}
-
-	obj = container_of(node, struct drm_gem_object, vma_node);
-
-	vgem_obj = to_vgem_bo(obj);
-
-	if (obj->dma_buf && vgem_obj->use_dma_buf) {
-		ret = dma_buf_mmap(obj->dma_buf, vma, 0);
-		goto out_unlock;
-	}
-
-	if (!obj->dev->driver->gem_vm_ops) {
-		ret = -EINVAL;
-		goto out_unlock;
-	}
-
-	vma->vm_flags |= VM_IO | VM_MIXEDMAP | VM_DONTEXPAND | VM_DONTDUMP;
-	vma->vm_ops = obj->dev->driver->gem_vm_ops;
-	vma->vm_private_data = vgem_obj;
-	vma->vm_page_prot =
-		pgprot_writecombine(vm_get_page_prot(vma->vm_flags));
-
-	mutex_unlock(&dev->struct_mutex);
-	drm_gem_vm_open(vma);
-	return ret;
-
-out_unlock:
-	mutex_unlock(&dev->struct_mutex);
-
-	return ret;
-}
-
-
 static struct drm_ioctl_desc vgem_ioctls[] = {
 };
 
 static const struct file_operations vgem_driver_fops = {
 	.owner		= THIS_MODULE,
 	.open		= drm_open,
-	.mmap		= vgem_drm_gem_mmap,
+	.mmap		= drm_gem_mmap,
 	.poll		= drm_poll,
 	.read		= drm_read,
 	.unlocked_ioctl = drm_ioctl,
-- 
cgit v0.10.2


From 2225cfe46bcc7558d9e371d1bc117df2df1fbacd Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Thu, 15 Oct 2015 11:33:43 +0200
Subject: drm/gem: Use kref_get_unless_zero for the weak mmap references

Compared to wrapping the final kref_put with dev->struct_mutex this
allows us to only acquire the offset manager look both in the final
cleanup and in the lookup. Which has the upside that no locks leak out
of the core abstractions. But it means that we need to hold a
temporary reference to the object while checking mmap constraints, to
make sure the object doesn't disappear. Extended the critical region
would have worked too, but would result in more leaky locking.

Also, this is the final bit which required dev->struct_mutex in gem
core, now modern drivers can be completely struct_mutex free!

This needs a new drm_vma_offset_exact_lookup_locked and makes both
drm_vma_offset_exact_lookup and drm_vma_offset_lookup unused.

v2: Don't leak object references in failure paths (David).

v3: Add a comment from Chris explaining how the ordering works, with
the slight adjustment that I dropped any mention of struct_mutex since
with this patch it's now immaterial ot core gem.

Cc: David Herrmann <dh.herrmann@gmail.com>
Reviewed-by: David Herrmann <dh.herrmann@gmail.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: http://mid.gmane.org/1444901623-18918-1-git-send-email-daniel.vetter@ffwll.ch
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c
index ab8ea422..64353d4 100644
--- a/drivers/gpu/drm/drm_gem.c
+++ b/drivers/gpu/drm/drm_gem.c
@@ -862,30 +862,46 @@ int drm_gem_mmap(struct file *filp, struct vm_area_struct *vma)
 {
 	struct drm_file *priv = filp->private_data;
 	struct drm_device *dev = priv->minor->dev;
-	struct drm_gem_object *obj;
+	struct drm_gem_object *obj = NULL;
 	struct drm_vma_offset_node *node;
 	int ret;
 
 	if (drm_device_is_unplugged(dev))
 		return -ENODEV;
 
-	mutex_lock(&dev->struct_mutex);
+	drm_vma_offset_lock_lookup(dev->vma_offset_manager);
+	node = drm_vma_offset_exact_lookup_locked(dev->vma_offset_manager,
+						  vma->vm_pgoff,
+						  vma_pages(vma));
+	if (likely(node)) {
+		obj = container_of(node, struct drm_gem_object, vma_node);
+		/*
+		 * When the object is being freed, after it hits 0-refcnt it
+		 * proceeds to tear down the object. In the process it will
+		 * attempt to remove the VMA offset and so acquire this
+		 * mgr->vm_lock.  Therefore if we find an object with a 0-refcnt
+		 * that matches our range, we know it is in the process of being
+		 * destroyed and will be freed as soon as we release the lock -
+		 * so we have to check for the 0-refcnted object and treat it as
+		 * invalid.
+		 */
+		if (!kref_get_unless_zero(&obj->refcount))
+			obj = NULL;
+	}
+	drm_vma_offset_unlock_lookup(dev->vma_offset_manager);
 
-	node = drm_vma_offset_exact_lookup(dev->vma_offset_manager,
-					   vma->vm_pgoff,
-					   vma_pages(vma));
-	if (!node) {
-		mutex_unlock(&dev->struct_mutex);
+	if (!obj)
 		return -EINVAL;
-	} else if (!drm_vma_node_is_allowed(node, filp)) {
-		mutex_unlock(&dev->struct_mutex);
+
+	if (!drm_vma_node_is_allowed(node, filp)) {
+		drm_gem_object_unreference_unlocked(obj);
 		return -EACCES;
 	}
 
-	obj = container_of(node, struct drm_gem_object, vma_node);
-	ret = drm_gem_mmap_obj(obj, drm_vma_node_size(node) << PAGE_SHIFT, vma);
+	ret = drm_gem_mmap_obj(obj, drm_vma_node_size(node) << PAGE_SHIFT,
+			       vma);
 
-	mutex_unlock(&dev->struct_mutex);
+	drm_gem_object_unreference_unlocked(obj);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/drm_vma_manager.c b/drivers/gpu/drm/drm_vma_manager.c
index 68c1f32..2f2ecde 100644
--- a/drivers/gpu/drm/drm_vma_manager.c
+++ b/drivers/gpu/drm/drm_vma_manager.c
@@ -112,7 +112,7 @@ void drm_vma_offset_manager_destroy(struct drm_vma_offset_manager *mgr)
 EXPORT_SYMBOL(drm_vma_offset_manager_destroy);
 
 /**
- * drm_vma_offset_lookup() - Find node in offset space
+ * drm_vma_offset_lookup_locked() - Find node in offset space
  * @mgr: Manager object
  * @start: Start address for object (page-based)
  * @pages: Size of object (page-based)
@@ -122,37 +122,21 @@ EXPORT_SYMBOL(drm_vma_offset_manager_destroy);
  * region and the given node will be returned, as long as the node spans the
  * whole requested area (given the size in number of pages as @pages).
  *
- * RETURNS:
- * Returns NULL if no suitable node can be found. Otherwise, the best match
- * is returned. It's the caller's responsibility to make sure the node doesn't
- * get destroyed before the caller can access it.
- */
-struct drm_vma_offset_node *drm_vma_offset_lookup(struct drm_vma_offset_manager *mgr,
-						  unsigned long start,
-						  unsigned long pages)
-{
-	struct drm_vma_offset_node *node;
-
-	read_lock(&mgr->vm_lock);
-	node = drm_vma_offset_lookup_locked(mgr, start, pages);
-	read_unlock(&mgr->vm_lock);
-
-	return node;
-}
-EXPORT_SYMBOL(drm_vma_offset_lookup);
-
-/**
- * drm_vma_offset_lookup_locked() - Find node in offset space
- * @mgr: Manager object
- * @start: Start address for object (page-based)
- * @pages: Size of object (page-based)
+ * Note that before lookup the vma offset manager lookup lock must be acquired
+ * with drm_vma_offset_lock_lookup(). See there for an example. This can then be
+ * used to implement weakly referenced lookups using kref_get_unless_zero().
  *
- * Same as drm_vma_offset_lookup() but requires the caller to lock offset lookup
- * manually. See drm_vma_offset_lock_lookup() for an example.
+ * Example:
+ *     drm_vma_offset_lock_lookup(mgr);
+ *     node = drm_vma_offset_lookup_locked(mgr);
+ *     if (node)
+ *         kref_get_unless_zero(container_of(node, sth, entr));
+ *     drm_vma_offset_unlock_lookup(mgr);
  *
  * RETURNS:
  * Returns NULL if no suitable node can be found. Otherwise, the best match
- * is returned.
+ * is returned. It's the caller's responsibility to make sure the node doesn't
+ * get destroyed before the caller can access it.
  */
 struct drm_vma_offset_node *drm_vma_offset_lookup_locked(struct drm_vma_offset_manager *mgr,
 							 unsigned long start,
diff --git a/include/drm/drm_vma_manager.h b/include/drm/drm_vma_manager.h
index 089cb73..2f63dd5 100644
--- a/include/drm/drm_vma_manager.h
+++ b/include/drm/drm_vma_manager.h
@@ -54,9 +54,6 @@ void drm_vma_offset_manager_init(struct drm_vma_offset_manager *mgr,
 				 unsigned long page_offset, unsigned long size);
 void drm_vma_offset_manager_destroy(struct drm_vma_offset_manager *mgr);
 
-struct drm_vma_offset_node *drm_vma_offset_lookup(struct drm_vma_offset_manager *mgr,
-						  unsigned long start,
-						  unsigned long pages);
 struct drm_vma_offset_node *drm_vma_offset_lookup_locked(struct drm_vma_offset_manager *mgr,
 							   unsigned long start,
 							   unsigned long pages);
@@ -71,25 +68,25 @@ bool drm_vma_node_is_allowed(struct drm_vma_offset_node *node,
 			     struct file *filp);
 
 /**
- * drm_vma_offset_exact_lookup() - Look up node by exact address
+ * drm_vma_offset_exact_lookup_locked() - Look up node by exact address
  * @mgr: Manager object
  * @start: Start address (page-based, not byte-based)
  * @pages: Size of object (page-based)
  *
- * Same as drm_vma_offset_lookup() but does not allow any offset into the node.
+ * Same as drm_vma_offset_lookup_locked() but does not allow any offset into the node.
  * It only returns the exact object with the given start address.
  *
  * RETURNS:
  * Node at exact start address @start.
  */
 static inline struct drm_vma_offset_node *
-drm_vma_offset_exact_lookup(struct drm_vma_offset_manager *mgr,
-			    unsigned long start,
-			    unsigned long pages)
+drm_vma_offset_exact_lookup_locked(struct drm_vma_offset_manager *mgr,
+				   unsigned long start,
+				   unsigned long pages)
 {
 	struct drm_vma_offset_node *node;
 
-	node = drm_vma_offset_lookup(mgr, start, pages);
+	node = drm_vma_offset_lookup_locked(mgr, start, pages);
 	return (node && node->vm_node.start == start) ? node : NULL;
 }
 
@@ -108,13 +105,6 @@ drm_vma_offset_exact_lookup(struct drm_vma_offset_manager *mgr,
  * not call any other VMA helpers while holding this lock.
  *
  * Note: You're in atomic-context while holding this lock!
- *
- * Example:
- *   drm_vma_offset_lock_lookup(mgr);
- *   node = drm_vma_offset_lookup_locked(mgr);
- *   if (node)
- *       kref_get_unless_zero(container_of(node, sth, entr));
- *   drm_vma_offset_unlock_lookup(mgr);
  */
 static inline void drm_vma_offset_lock_lookup(struct drm_vma_offset_manager *mgr)
 {
-- 
cgit v0.10.2


From 2b760d88a0fcd85a526a5c1ce4556ceff5723baa Mon Sep 17 00:00:00 2001
From: Lukas Wunner <lukas@wunner.de>
Date: Fri, 4 Sep 2015 20:49:36 +0200
Subject: ALSA: hda - Spell vga_switcheroo consistently

Currently everyone and their dog has their own favourite spelling
for vga_switcheroo. This makes it hard to grep dmesg for log entries
relating to vga_switcheroo. It also makes it hard to find related
source files in the tree.

vga_switcheroo.c uses pr_fmt "vga_switcheroo". Use that everywhere.

Signed-off-by: Lukas Wunner <lukas@wunner.de>
Reviewed-by: Takashi Iwai <tiwai@suse.de>
Link: http://patchwork.freedesktop.org/patch/msgid/9b0175319ce78d831acfcf11e4c6c760f826b0e3.1444663039.git.lukas@wunner.de
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/sound/pci/hda/hda_controller.h b/sound/pci/hda/hda_controller.h
index 314105c..7b635d6 100644
--- a/sound/pci/hda/hda_controller.h
+++ b/sound/pci/hda/hda_controller.h
@@ -153,7 +153,7 @@ struct azx {
 	unsigned int snoop:1;
 	unsigned int align_buffer_size:1;
 	unsigned int region_requested:1;
-	unsigned int disabled:1; /* disabled by VGA-switcher */
+	unsigned int disabled:1; /* disabled by vga_switcheroo */
 
 #ifdef CONFIG_SND_HDA_DSP_LOADER
 	struct azx_dev saved_azx_dev;
diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c
index e819013..45233731 100644
--- a/sound/pci/hda/hda_intel.c
+++ b/sound/pci/hda/hda_intel.c
@@ -337,7 +337,7 @@ enum {
 	 AZX_DCAPS_4K_BDLE_BOUNDARY | AZX_DCAPS_SNOOP_OFF)
 
 /*
- * VGA-switcher support
+ * vga_switcheroo support
  */
 #ifdef SUPPORT_VGA_SWITCHEROO
 #define use_vga_switcheroo(chip)	((chip)->use_vga_switcheroo)
@@ -1076,12 +1076,12 @@ static void azx_vs_set_state(struct pci_dev *pci,
 			}
 		}
 	} else {
-		dev_info(chip->card->dev, "%s via VGA-switcheroo\n",
+		dev_info(chip->card->dev, "%s via vga_switcheroo\n",
 			 disabled ? "Disabling" : "Enabling");
 		if (disabled) {
 			pm_runtime_put_sync_suspend(card->dev);
 			azx_suspend(card->dev);
-			/* when we get suspended by vga switcheroo we end up in D3cold,
+			/* when we get suspended by vga_switcheroo we end up in D3cold,
 			 * however we have no ACPI handle, so pci/acpi can't put us there,
 			 * put ourselves there */
 			pci->current_state = PCI_D3cold;
@@ -1121,7 +1121,7 @@ static void init_vga_switcheroo(struct azx *chip)
 	struct pci_dev *p = get_bound_vga(chip->pci);
 	if (p) {
 		dev_info(chip->card->dev,
-			 "Handle VGA-switcheroo audio client\n");
+			 "Handle vga_switcheroo audio client\n");
 		hda->use_vga_switcheroo = 1;
 		pci_dev_put(p);
 	}
@@ -1232,7 +1232,7 @@ static int azx_dev_free(struct snd_device *device)
 
 #ifdef SUPPORT_VGA_SWITCHEROO
 /*
- * Check of disabled HDMI controller by vga-switcheroo
+ * Check of disabled HDMI controller by vga_switcheroo
  */
 static struct pci_dev *get_bound_vga(struct pci_dev *pci)
 {
@@ -1917,7 +1917,7 @@ static int azx_probe(struct pci_dev *pci,
 
 	err = register_vga_switcheroo(chip);
 	if (err < 0) {
-		dev_err(card->dev, "Error registering VGA-switcheroo client\n");
+		dev_err(card->dev, "Error registering vga_switcheroo client\n");
 		goto out_free;
 	}
 
diff --git a/sound/pci/hda/hda_intel.h b/sound/pci/hda/hda_intel.h
index 354f0bb..ff0c4d6 100644
--- a/sound/pci/hda/hda_intel.h
+++ b/sound/pci/hda/hda_intel.h
@@ -35,7 +35,7 @@ struct hda_intel {
 	unsigned int irq_pending_warned:1;
 	unsigned int probe_continued:1;
 
-	/* VGA-switcheroo setup */
+	/* vga_switcheroo setup */
 	unsigned int use_vga_switcheroo:1;
 	unsigned int vga_switcheroo_registered:1;
 	unsigned int init_failed:1; /* delayed init failed */
-- 
cgit v0.10.2


From c390eed025dbbd09fff6cf128fb4eee83c9008a3 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 15 Oct 2015 20:39:58 +0300
Subject: drm: Don't leak fb when plane crtc coodinates are bad
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c
index 6058f4b..ad7b7a7 100644
--- a/drivers/gpu/drm/drm_crtc.c
+++ b/drivers/gpu/drm/drm_crtc.c
@@ -2341,7 +2341,8 @@ static int __setplane_internal(struct drm_plane *plane,
 	    crtc_y > INT_MAX - (int32_t) crtc_h) {
 		DRM_DEBUG_KMS("Invalid CRTC coordinates %ux%u+%d+%d\n",
 			      crtc_w, crtc_h, crtc_x, crtc_y);
-		return -ERANGE;
+		ret = -ERANGE;
+		goto out;
 	}
 
 
-- 
cgit v0.10.2


From 41121248600f7f8c2a97b256bd9bcec436392a53 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 15 Oct 2015 20:39:59 +0300
Subject: drm: Swap w/h when converting the mode to src coordidates for a
 rotated primary plane
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

When converting the mode hdisplay/vdisplay to primary plane src
coordinates we need to take into account the current plane
rotation.

Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Cc: Daniel Vetter <daniel@ffwll.ch>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1444930802-8515-2-git-send-email-ville.syrjala@linux.intel.com
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c
index 87a2a44..0c6f621 100644
--- a/drivers/gpu/drm/drm_atomic_helper.c
+++ b/drivers/gpu/drm/drm_atomic_helper.c
@@ -1790,8 +1790,13 @@ int __drm_atomic_helper_set_config(struct drm_mode_set *set,
 	primary_state->crtc_w = set->mode->hdisplay;
 	primary_state->src_x = set->x << 16;
 	primary_state->src_y = set->y << 16;
-	primary_state->src_h = set->mode->vdisplay << 16;
-	primary_state->src_w = set->mode->hdisplay << 16;
+	if (primary_state->rotation & (BIT(DRM_ROTATE_90) | BIT(DRM_ROTATE_270))) {
+		primary_state->src_h = set->mode->hdisplay << 16;
+		primary_state->src_w = set->mode->vdisplay << 16;
+	} else {
+		primary_state->src_h = set->mode->vdisplay << 16;
+		primary_state->src_w = set->mode->hdisplay << 16;
+	}
 
 commit:
 	ret = update_output_state(state, set);
-- 
cgit v0.10.2


From ce8d9ecc1b9dafc593bac481a6fcc39cc913543a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 15 Oct 2015 20:40:00 +0300
Subject: drm: Refactor plane src coordinate checks
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Pull the plane src coordinate checks into a separate function so that we
can share them for the legacy and new stuff.

Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Cc: Daniel Vetter <daniel@ffwll.ch>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1444930802-8515-3-git-send-email-ville.syrjala@linux.intel.com
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c
index ad7b7a7..4003bdb 100644
--- a/drivers/gpu/drm/drm_crtc.c
+++ b/drivers/gpu/drm/drm_crtc.c
@@ -2285,6 +2285,32 @@ int drm_plane_check_pixel_format(const struct drm_plane *plane, u32 format)
 	return -EINVAL;
 }
 
+static int check_src_coords(uint32_t src_x, uint32_t src_y,
+			    uint32_t src_w, uint32_t src_h,
+			    const struct drm_framebuffer *fb)
+{
+	unsigned int fb_width, fb_height;
+
+	fb_width = fb->width << 16;
+	fb_height = fb->height << 16;
+
+	/* Make sure source coordinates are inside the fb. */
+	if (src_w > fb_width ||
+	    src_x > fb_width - src_w ||
+	    src_h > fb_height ||
+	    src_y > fb_height - src_h) {
+		DRM_DEBUG_KMS("Invalid source coordinates "
+			      "%u.%06ux%u.%06u+%u.%06u+%u.%06u\n",
+			      src_w >> 16, ((src_w & 0xffff) * 15625) >> 10,
+			      src_h >> 16, ((src_h & 0xffff) * 15625) >> 10,
+			      src_x >> 16, ((src_x & 0xffff) * 15625) >> 10,
+			      src_y >> 16, ((src_y & 0xffff) * 15625) >> 10);
+		return -ENOSPC;
+	}
+
+	return 0;
+}
+
 /*
  * setplane_internal - setplane handler for internal callers
  *
@@ -2304,7 +2330,6 @@ static int __setplane_internal(struct drm_plane *plane,
 			       uint32_t src_w, uint32_t src_h)
 {
 	int ret = 0;
-	unsigned int fb_width, fb_height;
 
 	/* No fb means shut it down */
 	if (!fb) {
@@ -2345,24 +2370,9 @@ static int __setplane_internal(struct drm_plane *plane,
 		goto out;
 	}
 
-
-	fb_width = fb->width << 16;
-	fb_height = fb->height << 16;
-
-	/* Make sure source coordinates are inside the fb. */
-	if (src_w > fb_width ||
-	    src_x > fb_width - src_w ||
-	    src_h > fb_height ||
-	    src_y > fb_height - src_h) {
-		DRM_DEBUG_KMS("Invalid source coordinates "
-			      "%u.%06ux%u.%06u+%u.%06u+%u.%06u\n",
-			      src_w >> 16, ((src_w & 0xffff) * 15625) >> 10,
-			      src_h >> 16, ((src_h & 0xffff) * 15625) >> 10,
-			      src_x >> 16, ((src_x & 0xffff) * 15625) >> 10,
-			      src_y >> 16, ((src_y & 0xffff) * 15625) >> 10);
-		ret = -ENOSPC;
+	ret = check_src_coords(src_x, src_y, src_w, src_h, fb);
+	if (ret)
 		goto out;
-	}
 
 	plane->old_fb = plane->fb;
 	ret = plane->funcs->update_plane(plane, crtc, fb,
@@ -2556,17 +2566,8 @@ int drm_crtc_check_viewport(const struct drm_crtc *crtc,
 	if (crtc->invert_dimensions)
 		swap(hdisplay, vdisplay);
 
-	if (hdisplay > fb->width ||
-	    vdisplay > fb->height ||
-	    x > fb->width - hdisplay ||
-	    y > fb->height - vdisplay) {
-		DRM_DEBUG_KMS("Invalid fb size %ux%u for CRTC viewport %ux%u+%d+%d%s.\n",
-			      fb->width, fb->height, hdisplay, vdisplay, x, y,
-			      crtc->invert_dimensions ? " (inverted)" : "");
-		return -ENOSPC;
-	}
-
-	return 0;
+	return check_src_coords(x << 16, y << 16,
+				hdisplay << 16, vdisplay << 16, fb);
 }
 EXPORT_SYMBOL(drm_crtc_check_viewport);
 
-- 
cgit v0.10.2


From 33e0be63759d472e7a6996d1277ff66e92a8685d Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 16 Oct 2015 18:38:39 +0300
Subject: drm: Check crtc viewport correctly with rotated primary plane on
 atomic drivers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

On atomic drivers we can dig out the primary plane rotation from the
plane state instead of looking at the legacy crtc->invert_dimensions
flag. The flag is not set by anyone except omapdrm, and it would be
racy to set it the same way in the atomic helpers.

v2: Kill crtc->invert_dimensions totally since omap is state based
    already and no one else ever used it (Matt)

Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: Tomi Valkeinen <tomi.valkeinen@ti.com>
Cc: Rob Clark <robdclark@gmail.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1445009919-22746-1-git-send-email-ville.syrjala@linux.intel.com
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c
index 4003bdb..8eec726 100644
--- a/drivers/gpu/drm/drm_crtc.c
+++ b/drivers/gpu/drm/drm_crtc.c
@@ -676,7 +676,6 @@ int drm_crtc_init_with_planes(struct drm_device *dev, struct drm_crtc *crtc,
 
 	crtc->dev = dev;
 	crtc->funcs = funcs;
-	crtc->invert_dimensions = false;
 
 	drm_modeset_lock_init(&crtc->mutex);
 	ret = drm_mode_object_get(dev, &crtc->base, DRM_MODE_OBJECT_CRTC);
@@ -2563,7 +2562,9 @@ int drm_crtc_check_viewport(const struct drm_crtc *crtc,
 
 	drm_crtc_get_hv_timing(mode, &hdisplay, &vdisplay);
 
-	if (crtc->invert_dimensions)
+	if (crtc->state &&
+	    crtc->primary->state->rotation & (BIT(DRM_ROTATE_90) |
+					      BIT(DRM_ROTATE_270)))
 		swap(hdisplay, vdisplay);
 
 	return check_src_coords(x << 16, y << 16,
diff --git a/drivers/gpu/drm/omapdrm/omap_crtc.c b/drivers/gpu/drm/omapdrm/omap_crtc.c
index 9a4ba4f..ad09590 100644
--- a/drivers/gpu/drm/omapdrm/omap_crtc.c
+++ b/drivers/gpu/drm/omapdrm/omap_crtc.c
@@ -412,9 +412,6 @@ static void omap_crtc_atomic_flush(struct drm_crtc *crtc,
 		dispc_mgr_go(omap_crtc->channel);
 		omap_irq_register(crtc->dev, &omap_crtc->vblank_irq);
 	}
-
-	crtc->invert_dimensions = !!(crtc->primary->state->rotation &
-				    (BIT(DRM_ROTATE_90) | BIT(DRM_ROTATE_270)));
 }
 
 static int omap_crtc_atomic_set_property(struct drm_crtc *crtc,
diff --git a/include/drm/drm_crtc.h b/include/drm/drm_crtc.h
index 33ddedd..3f0c690 100644
--- a/include/drm/drm_crtc.h
+++ b/include/drm/drm_crtc.h
@@ -407,9 +407,6 @@ struct drm_crtc_funcs {
  * @enabled: is this CRTC enabled?
  * @mode: current mode timings
  * @hwmode: mode timings as programmed to hw regs
- * @invert_dimensions: for purposes of error checking crtc vs fb sizes,
- *    invert the width/height of the crtc.  This is used if the driver
- *    is performing 90 or 270 degree rotated scanout
  * @x: x position on screen
  * @y: y position on screen
  * @funcs: CRTC control functions
@@ -458,8 +455,6 @@ struct drm_crtc {
 	 */
 	struct drm_display_mode hwmode;
 
-	bool invert_dimensions;
-
 	int x, y;
 	const struct drm_crtc_funcs *funcs;
 
-- 
cgit v0.10.2


From 2afa701d3e0bb9865130990b5e0cfe240ae3b605 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 15 Oct 2015 20:40:02 +0300
Subject: drm: Check plane src coordinates correctly during page flip for
 atomic drivers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Instead of relying on the old crtc-{x,y,mode} gunk, dig out the primary
plane coordinates from the plane state when checking them against the
new framebuffer during page flip.

Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Cc: Daniel Vetter <daniel@ffwll.ch>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1444930802-8515-5-git-send-email-ville.syrjala@linux.intel.com
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c
index 8eec726..e54660a 100644
--- a/drivers/gpu/drm/drm_crtc.c
+++ b/drivers/gpu/drm/drm_crtc.c
@@ -5183,7 +5183,14 @@ int drm_mode_page_flip_ioctl(struct drm_device *dev,
 		goto out;
 	}
 
-	ret = drm_crtc_check_viewport(crtc, crtc->x, crtc->y, &crtc->mode, fb);
+	if (crtc->state) {
+		const struct drm_plane_state *state = crtc->primary->state;
+
+		ret = check_src_coords(state->src_x, state->src_y,
+				       state->src_w, state->src_h, fb);
+	} else {
+		ret = drm_crtc_check_viewport(crtc, crtc->x, crtc->y, &crtc->mode, fb);
+	}
 	if (ret)
 		goto out;
 
-- 
cgit v0.10.2


From 6749c9f023d49adca3ca4773e93282b95ceebfb5 Mon Sep 17 00:00:00 2001
From: Adam Richter <adamrichter4@gmail.com>
Date: Fri, 16 Oct 2015 03:33:02 -0700
Subject: drm: fix mutex leak in drm_dp_get_mst_branch_device

In Linux 4.3-rc5, there is an error case in drm_dp_get_branch_device
that returns without releasing mgr->lock, resulting a spew of kernel
messages about a kernel work function possibly having leaked a mutex
and presumably more serious adverse consequences later.  This patch
changes the error to "goto out" to unlock the mutex before returning.

Signed-off-by: Adam J. Richter <adam_richter2004@yahoo.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_dp_mst_topology.c b/drivers/gpu/drm/drm_dp_mst_topology.c
index e23df5f..e4c9b4a 100644
--- a/drivers/gpu/drm/drm_dp_mst_topology.c
+++ b/drivers/gpu/drm/drm_dp_mst_topology.c
@@ -1182,17 +1182,18 @@ static struct drm_dp_mst_branch *drm_dp_get_mst_branch_device(struct drm_dp_mst_
 
 		list_for_each_entry(port, &mstb->ports, next) {
 			if (port->port_num == port_num) {
-				if (!port->mstb) {
+				mstb = port->mstb;
+				if (!mstb) {
 					DRM_ERROR("failed to lookup MSTB with lct %d, rad %02x\n", lct, rad[0]);
-					return NULL;
+					goto out;
 				}
 
-				mstb = port->mstb;
 				break;
 			}
 		}
 	}
 	kref_get(&mstb->kref);
+out:
 	mutex_unlock(&mgr->lock);
 	return mstb;
 }
-- 
cgit v0.10.2


From 16e910df19ae8aa2dc0f1e8502aea7a36888457b Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Fri, 16 Oct 2015 18:23:13 +0200
Subject: drm/fb-helper: Set plane rotation directly

The point behind standardizing properties into core drm state
structures is also that internal code looks prettiers. Take advantage
of that and set rotation directly in the fbdev atomic code.

Cc: Rob Clark <robdclark@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1445012594-25988-1-git-send-email-daniel.vetter@ffwll.ch
Acked-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c
index abe9793..80e0631 100644
--- a/drivers/gpu/drm/drm_fb_helper.c
+++ b/drivers/gpu/drm/drm_fb_helper.c
@@ -360,11 +360,7 @@ retry:
 			goto fail;
 		}
 
-		ret = drm_atomic_plane_set_property(plane, plane_state,
-				dev->mode_config.rotation_property,
-				BIT(DRM_ROTATE_0));
-		if (ret != 0)
-			goto fail;
+		plane_state->rotation = BIT(DRM_ROTATE_0);
 
 		/* disable non-primary: */
 		if (plane->type == DRM_PLANE_TYPE_PRIMARY)
-- 
cgit v0.10.2


From a0fb6ad7ae28a4dce34c010028dc070eeacae1d9 Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Fri, 16 Oct 2015 19:11:30 +0200
Subject: drm/fb-helper: Fix fb refcounting in pan_display_atomic

In

commit bbb1e52402b2a288b09ae37e8182599931c7e9df
Author: Rob Clark <robdclark@gmail.com>
Date:   Tue Aug 25 15:35:58 2015 -0400

    drm/fb-helper: atomic restore_fbdev_mode()..

we've forgotten to do the plane->old_fb refcount dance for
pan_display_atomic, which can result in refcount leaks if the current
configuration is not from fbcon. Which apparently can happen when
vt-switching - fbcon does a pan first before a set_par.

OCD-align function parameters while at it.

v2: Actually git add the OCD.

Cc: Rob Clark <robdclark@gmail.com>
Cc: Rodrigo Vivi <rodrigo.vivi@gmail.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=92483
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1445015490-27682-1-git-send-email-daniel.vetter@ffwll.ch
Tested-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c
index 80e0631..5b2de1b 100644
--- a/drivers/gpu/drm/drm_fb_helper.c
+++ b/drivers/gpu/drm/drm_fb_helper.c
@@ -1227,7 +1227,7 @@ int drm_fb_helper_set_par(struct fb_info *info)
 EXPORT_SYMBOL(drm_fb_helper_set_par);
 
 static int pan_display_atomic(struct fb_var_screeninfo *var,
-		struct fb_info *info)
+			      struct fb_info *info)
 {
 	struct drm_fb_helper *fb_helper = info->par;
 	struct drm_device *dev = fb_helper->dev;
@@ -1245,6 +1245,8 @@ retry:
 
 		mode_set = &fb_helper->crtc_info[i].mode_set;
 
+		mode_set->crtc->primary->old_fb = mode_set->crtc->primary->fb;
+
 		mode_set->x = var->xoffset;
 		mode_set->y = var->yoffset;
 
@@ -1260,13 +1262,34 @@ retry:
 	info->var.xoffset = var->xoffset;
 	info->var.yoffset = var->yoffset;
 
-	return 0;
 
 fail:
+	for(i = 0; i < fb_helper->crtc_count; i++) {
+		struct drm_mode_set *mode_set;
+		struct drm_plane *plane;
+
+		mode_set = &fb_helper->crtc_info[i].mode_set;
+		plane = mode_set->crtc->primary;
+
+		if (ret == 0) {
+			struct drm_framebuffer *new_fb = plane->state->fb;
+
+			if (new_fb)
+				drm_framebuffer_reference(new_fb);
+			plane->fb = new_fb;
+			plane->crtc = plane->state->crtc;
+
+			if (plane->old_fb)
+				drm_framebuffer_unreference(plane->old_fb);
+		}
+		plane->old_fb = NULL;
+	}
+
 	if (ret == -EDEADLK)
 		goto backoff;
 
-	drm_atomic_state_free(state);
+	if (ret != 0)
+		drm_atomic_state_free(state);
 
 	return ret;
 
-- 
cgit v0.10.2


From a3d5aaa836ed993747af7b53cfca1b3cd3c9fc46 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Thu, 15 Oct 2015 18:20:45 -0400
Subject: drm/amdgpu/gfx8: set TC_WB_ACTION_EN in RELEASE_MEM packet
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This is the recommended setting from the hw team for newer
versions of the firmware.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index 718250a..7dd7651 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -4114,6 +4114,7 @@ static void gfx_v8_0_ring_emit_fence_compute(struct amdgpu_ring *ring,
 	amdgpu_ring_write(ring, PACKET3(PACKET3_RELEASE_MEM, 5));
 	amdgpu_ring_write(ring, (EOP_TCL1_ACTION_EN |
 				 EOP_TC_ACTION_EN |
+				 EOP_TC_WB_ACTION_EN |
 				 EVENT_TYPE(CACHE_FLUSH_AND_INV_TS_EVENT) |
 				 EVENT_INDEX(5)));
 	amdgpu_ring_write(ring, DATA_SEL(write64bit ? 2 : 1) | INT_SEL(int_sel ? 2 : 0));
-- 
cgit v0.10.2


From 606bb5e0b28b540685fb94c22902cd9a948a3779 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 8 Oct 2015 11:43:34 +0300
Subject: drm/i915: Use round to closest when computing the CEA 1.001 pixel
 clocks
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

drm_edid.c now computes the alternate CEA clocks using
DIV_ROUND_CLOSEST(), so follow suit in the N/CTS setup to make sure we
pick the right setting for the mode.

Unfortunately we can't actually use DIV_ROUND_CLOSEST() here due to the
({}) construct used, so just stick in raw numbers instead.

Cc: Clint Taylor <clinton.a.taylor@intel.com>
Cc: Libin Yang <libin.yang@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/i915/intel_audio.c b/drivers/gpu/drm/i915/intel_audio.c
index 56c2f54..4dccd9b 100644
--- a/drivers/gpu/drm/i915/intel_audio.c
+++ b/drivers/gpu/drm/i915/intel_audio.c
@@ -61,21 +61,21 @@ static const struct {
 	int clock;
 	u32 config;
 } hdmi_audio_clock[] = {
-	{ DIV_ROUND_UP(25200 * 1000, 1001), AUD_CONFIG_PIXEL_CLOCK_HDMI_25175 },
+	{ 25175, AUD_CONFIG_PIXEL_CLOCK_HDMI_25175 },
 	{ 25200, AUD_CONFIG_PIXEL_CLOCK_HDMI_25200 }, /* default per bspec */
 	{ 27000, AUD_CONFIG_PIXEL_CLOCK_HDMI_27000 },
-	{ 27000 * 1001 / 1000, AUD_CONFIG_PIXEL_CLOCK_HDMI_27027 },
+	{ 27027, AUD_CONFIG_PIXEL_CLOCK_HDMI_27027 },
 	{ 54000, AUD_CONFIG_PIXEL_CLOCK_HDMI_54000 },
-	{ 54000 * 1001 / 1000, AUD_CONFIG_PIXEL_CLOCK_HDMI_54054 },
-	{ DIV_ROUND_UP(74250 * 1000, 1001), AUD_CONFIG_PIXEL_CLOCK_HDMI_74176 },
+	{ 54054, AUD_CONFIG_PIXEL_CLOCK_HDMI_54054 },
+	{ 74176, AUD_CONFIG_PIXEL_CLOCK_HDMI_74176 },
 	{ 74250, AUD_CONFIG_PIXEL_CLOCK_HDMI_74250 },
-	{ DIV_ROUND_UP(148500 * 1000, 1001), AUD_CONFIG_PIXEL_CLOCK_HDMI_148352 },
+	{ 148352, AUD_CONFIG_PIXEL_CLOCK_HDMI_148352 },
 	{ 148500, AUD_CONFIG_PIXEL_CLOCK_HDMI_148500 },
 };
 
 /* HDMI N/CTS table */
 #define TMDS_297M 297000
-#define TMDS_296M DIV_ROUND_UP(297000 * 1000, 1001)
+#define TMDS_296M 296703
 static const struct {
 	int sample_rate;
 	int clock;
-- 
cgit v0.10.2


From f9fff064bb83c55b3fc6291e3b59e3bc78fedddf Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Thu, 15 Oct 2015 16:53:27 -0400
Subject: drm/amdgpu/dce: simplify suspend/resume
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We were basically opencoding the same thing in both
hw_init and resume and hw_fini and suspend.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
index 3707393..a6ea2d8 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
@@ -3086,22 +3086,18 @@ static int dce_v10_0_suspend(void *handle)
 
 	amdgpu_atombios_scratch_regs_save(adev);
 
-	dce_v10_0_hpd_fini(adev);
-
-	return 0;
+	return dce_v10_0_hw_fini(handle);
 }
 
 static int dce_v10_0_resume(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	int ret;
 
-	dce_v10_0_init_golden_registers(adev);
+	ret = dce_v10_0_hw_init(handle);
 
 	amdgpu_atombios_scratch_regs_restore(adev);
 
-	/* init dig PHYs, disp eng pll */
-	amdgpu_atombios_encoder_init_dig(adev);
-	amdgpu_atombios_crtc_set_disp_eng_pll(adev, adev->clock.default_dispclk);
 	/* turn on the BL */
 	if (adev->mode_info.bl_encoder) {
 		u8 bl_level = amdgpu_display_backlight_get_level(adev,
@@ -3110,10 +3106,7 @@ static int dce_v10_0_resume(void *handle)
 						    bl_level);
 	}
 
-	/* initialize hpd */
-	dce_v10_0_hpd_init(adev);
-
-	return 0;
+	return ret;
 }
 
 static bool dce_v10_0_is_idle(void *handle)
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
index c1147ec..bdafeb2 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
@@ -3028,6 +3028,7 @@ static int dce_v11_0_hw_init(void *handle)
 	dce_v11_0_init_golden_registers(adev);
 
 	/* init dig PHYs, disp eng pll */
+	amdgpu_atombios_crtc_powergate_init(adev);
 	amdgpu_atombios_encoder_init_dig(adev);
 	amdgpu_atombios_crtc_set_disp_eng_pll(adev, adev->clock.default_dispclk);
 
@@ -3061,23 +3062,18 @@ static int dce_v11_0_suspend(void *handle)
 
 	amdgpu_atombios_scratch_regs_save(adev);
 
-	dce_v11_0_hpd_fini(adev);
-
-	return 0;
+	return dce_v11_0_hw_fini(handle);
 }
 
 static int dce_v11_0_resume(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	int ret;
 
-	dce_v11_0_init_golden_registers(adev);
+	ret = dce_v11_0_hw_init(handle);
 
 	amdgpu_atombios_scratch_regs_restore(adev);
 
-	/* init dig PHYs, disp eng pll */
-	amdgpu_atombios_crtc_powergate_init(adev);
-	amdgpu_atombios_encoder_init_dig(adev);
-	amdgpu_atombios_crtc_set_disp_eng_pll(adev, adev->clock.default_dispclk);
 	/* turn on the BL */
 	if (adev->mode_info.bl_encoder) {
 		u8 bl_level = amdgpu_display_backlight_get_level(adev,
@@ -3086,10 +3082,7 @@ static int dce_v11_0_resume(void *handle)
 						    bl_level);
 	}
 
-	/* initialize hpd */
-	dce_v11_0_hpd_init(adev);
-
-	return 0;
+	return ret;
 }
 
 static bool dce_v11_0_is_idle(void *handle)
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
index d784fb4..00c34f8 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
@@ -2994,20 +2994,18 @@ static int dce_v8_0_suspend(void *handle)
 
 	amdgpu_atombios_scratch_regs_save(adev);
 
-	dce_v8_0_hpd_fini(adev);
-
-	return 0;
+	return dce_v8_0_hw_fini(handle);
 }
 
 static int dce_v8_0_resume(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	int ret;
+
+	ret = dce_v8_0_hw_init(handle);
 
 	amdgpu_atombios_scratch_regs_restore(adev);
 
-	/* init dig PHYs, disp eng pll */
-	amdgpu_atombios_encoder_init_dig(adev);
-	amdgpu_atombios_crtc_set_disp_eng_pll(adev, adev->clock.default_dispclk);
 	/* turn on the BL */
 	if (adev->mode_info.bl_encoder) {
 		u8 bl_level = amdgpu_display_backlight_get_level(adev,
@@ -3016,10 +3014,7 @@ static int dce_v8_0_resume(void *handle)
 						    bl_level);
 	}
 
-	/* initialize hpd */
-	dce_v8_0_hpd_init(adev);
-
-	return 0;
+	return ret;
 }
 
 static bool dce_v8_0_is_idle(void *handle)
-- 
cgit v0.10.2


From d519432c8529efe65174cb683b3fb2f31faecffa Mon Sep 17 00:00:00 2001
From: Geert Uytterhoeven <geert+renesas@glider.be>
Date: Tue, 23 Jun 2015 14:58:52 +0200
Subject: drm: rcar-du: Remove obsolete rcar-du-r8a779x platform_device_id
 entries

Since commit a483dcbfa21f919c ("ARM: shmobile: lager: Remove legacy
board support"), R-Car Gen2 SoCs are only supported in generic DT-only
ARM multi-platform builds.  The driver doesn't need to match platform
devices by name anymore, hence remove the corresponding
platform_device_id entry.

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>

diff --git a/drivers/gpu/drm/rcar-du/rcar_du_drv.c b/drivers/gpu/drm/rcar-du/rcar_du_drv.c
index feddda0..a8dfca5 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_drv.c
+++ b/drivers/gpu/drm/rcar-du/rcar_du_drv.c
@@ -108,8 +108,6 @@ static const struct rcar_du_device_info rcar_du_r8a7791_info = {
 
 static const struct platform_device_id rcar_du_id_table[] = {
 	{ "rcar-du-r8a7779", (kernel_ulong_t)&rcar_du_r8a7779_info },
-	{ "rcar-du-r8a7790", (kernel_ulong_t)&rcar_du_r8a7790_info },
-	{ "rcar-du-r8a7791", (kernel_ulong_t)&rcar_du_r8a7791_info },
 	{ }
 };
 
-- 
cgit v0.10.2


From f3bda362b5fcab97509001dc51e64a99e30e6fb7 Mon Sep 17 00:00:00 2001
From: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Date: Thu, 16 Jul 2015 11:42:37 +0300
Subject: drm: rcar-du: Add dependency on OF

The driver requires OF support, add a dependency in Kconfig and remove
the platform_device_id table that isn't used anymore.

Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Acked-by: Geert Uytterhoeven <geert+renesas@glider.be>

diff --git a/drivers/gpu/drm/rcar-du/Kconfig b/drivers/gpu/drm/rcar-du/Kconfig
index 11485a4..d4e0a39 100644
--- a/drivers/gpu/drm/rcar-du/Kconfig
+++ b/drivers/gpu/drm/rcar-du/Kconfig
@@ -1,6 +1,6 @@
 config DRM_RCAR_DU
 	tristate "DRM Support for R-Car Display Unit"
-	depends on DRM && ARM && HAVE_DMA_ATTRS
+	depends on DRM && ARM && HAVE_DMA_ATTRS && OF
 	depends on ARCH_SHMOBILE || COMPILE_TEST
 	select DRM_KMS_HELPER
 	select DRM_KMS_CMA_HELPER
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_drv.c b/drivers/gpu/drm/rcar-du/rcar_du_drv.c
index a8dfca5..355d1e7 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_drv.c
+++ b/drivers/gpu/drm/rcar-du/rcar_du_drv.c
@@ -106,13 +106,6 @@ static const struct rcar_du_device_info rcar_du_r8a7791_info = {
 	.num_lvds = 1,
 };
 
-static const struct platform_device_id rcar_du_id_table[] = {
-	{ "rcar-du-r8a7779", (kernel_ulong_t)&rcar_du_r8a7779_info },
-	{ }
-};
-
-MODULE_DEVICE_TABLE(platform, rcar_du_id_table);
-
 static const struct of_device_id rcar_du_of_table[] = {
 	{ .compatible = "renesas,du-r8a7779", .data = &rcar_du_r8a7779_info },
 	{ .compatible = "renesas,du-r8a7790", .data = &rcar_du_r8a7790_info },
@@ -165,8 +158,7 @@ static int rcar_du_load(struct drm_device *dev, unsigned long flags)
 	init_waitqueue_head(&rcdu->commit.wait);
 
 	rcdu->dev = &pdev->dev;
-	rcdu->info = np ? of_match_device(rcar_du_of_table, rcdu->dev)->data
-		   : (void *)platform_get_device_id(pdev)->driver_data;
+	rcdu->info = of_match_device(rcar_du_of_table, rcdu->dev)->data;
 	rcdu->ddev = dev;
 	dev->dev_private = rcdu;
 
@@ -338,7 +330,6 @@ static struct platform_driver rcar_du_platform_driver = {
 		.pm	= &rcar_du_pm_ops,
 		.of_match_table = rcar_du_of_table,
 	},
-	.id_table	= rcar_du_id_table,
 };
 
 module_platform_driver(rcar_du_platform_driver);
-- 
cgit v0.10.2


From f4f0fb7949a9e55f4f3429381b8b92372038b6cd Mon Sep 17 00:00:00 2001
From: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Date: Tue, 28 Apr 2015 15:26:33 +0300
Subject: drm: rcar-du: Enable DU0 to DPAD connection on R8A7791

The DPAD RGB output can be driven by both DU0 and DU1 on R8A7791. Mark
the DU0 connection as possible.

Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>

diff --git a/drivers/gpu/drm/rcar-du/rcar_du_drv.c b/drivers/gpu/drm/rcar-du/rcar_du_drv.c
index 355d1e7..bebcc97 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_drv.c
+++ b/drivers/gpu/drm/rcar-du/rcar_du_drv.c
@@ -93,7 +93,7 @@ static const struct rcar_du_device_info rcar_du_r8a7791_info = {
 		 * (currently unsupported) TCON output.
 		 */
 		[RCAR_DU_OUTPUT_DPAD0] = {
-			.possible_crtcs = BIT(1),
+			.possible_crtcs = BIT(1) | BIT(0),
 			.encoder_type = DRM_MODE_ENCODER_NONE,
 			.port = 0,
 		},
-- 
cgit v0.10.2


From a32a3c800ca3ae822dfdd29f5b7ef81fbf970612 Mon Sep 17 00:00:00 2001
From: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Date: Mon, 27 Jul 2015 11:42:54 +0300
Subject: drm: rcar-du: Fix plane state free in plane reset handler

The plane reset handler frees the plane state and allocates a new
default state, but when doing so attempt to free the plane state using
the base plane state pointer instead of casting it to the
driver-specific state object that has been allocated. Fix it by using
the rcar_du_plane_atomic_destroy_state() function to destroy the plane
state instead of duplicating the code.

Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Reviewed-by: Thierry Reding <treding@nvidia.com>

diff --git a/drivers/gpu/drm/rcar-du/rcar_du_plane.c b/drivers/gpu/drm/rcar-du/rcar_du_plane.c
index c669864..ffa5837 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_plane.c
+++ b/drivers/gpu/drm/rcar-du/rcar_du_plane.c
@@ -273,29 +273,6 @@ static const struct drm_plane_helper_funcs rcar_du_plane_helper_funcs = {
 	.atomic_update = rcar_du_plane_atomic_update,
 };
 
-static void rcar_du_plane_reset(struct drm_plane *plane)
-{
-	struct rcar_du_plane_state *state;
-
-	if (plane->state && plane->state->fb)
-		drm_framebuffer_unreference(plane->state->fb);
-
-	kfree(plane->state);
-	plane->state = NULL;
-
-	state = kzalloc(sizeof(*state), GFP_KERNEL);
-	if (state == NULL)
-		return;
-
-	state->hwindex = -1;
-	state->alpha = 255;
-	state->colorkey = RCAR_DU_COLORKEY_NONE;
-	state->zpos = plane->type == DRM_PLANE_TYPE_PRIMARY ? 0 : 1;
-
-	plane->state = &state->state;
-	plane->state->plane = plane;
-}
-
 static struct drm_plane_state *
 rcar_du_plane_atomic_duplicate_state(struct drm_plane *plane)
 {
@@ -322,6 +299,28 @@ static void rcar_du_plane_atomic_destroy_state(struct drm_plane *plane,
 	kfree(to_rcar_plane_state(state));
 }
 
+static void rcar_du_plane_reset(struct drm_plane *plane)
+{
+	struct rcar_du_plane_state *state;
+
+	if (plane->state) {
+		rcar_du_plane_atomic_destroy_state(plane, plane->state);
+		plane->state = NULL;
+	}
+
+	state = kzalloc(sizeof(*state), GFP_KERNEL);
+	if (state == NULL)
+		return;
+
+	state->hwindex = -1;
+	state->alpha = 255;
+	state->colorkey = RCAR_DU_COLORKEY_NONE;
+	state->zpos = plane->type == DRM_PLANE_TYPE_PRIMARY ? 0 : 1;
+
+	plane->state = &state->state;
+	plane->state->plane = plane;
+}
+
 static int rcar_du_plane_atomic_set_property(struct drm_plane *plane,
 					     struct drm_plane_state *state,
 					     struct drm_property *property,
-- 
cgit v0.10.2


From fa3a7340eaa174a36c12c31747dc6eccd08530b5 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 8 Oct 2015 11:43:32 +0300
Subject: drm/edid: Fix up clock for CEA/HDMI modes specified via detailed
 timings
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

EDID detailed timings have a resolution of 10kHz for the pixel clock, so
they can't represent certain CEA/HDMI modes accurately. If we see a mode
coming in via detailed timings which otherwise matches one of the
CEA/HDMI modes except the clock is just a bit off, let's assume that the
intention was for that mode to be one of the CEA/HDMI modes and go ahead
and fix up the clock to match the CEA/HDMI spec exactly (well, as close
as we can get with the 1 kHz resolution we use).

This should help code that's looking for an exact clock match (eg. i915
audio N/CTS setup).

Cc: Adam Jackson <ajax@redhat.com>
Cc: Clint Taylor <clinton.a.taylor@intel.com>
Cc: Libin Yang <libin.yang@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c
index d895556..977915c 100644
--- a/drivers/gpu/drm/drm_edid.c
+++ b/drivers/gpu/drm/drm_edid.c
@@ -2418,6 +2418,8 @@ add_cvt_modes(struct drm_connector *connector, struct edid *edid)
 	return closure.modes;
 }
 
+static void fixup_detailed_cea_mode_clock(struct drm_display_mode *mode);
+
 static void
 do_detailed_mode(struct detailed_timing *timing, void *c)
 {
@@ -2434,6 +2436,13 @@ do_detailed_mode(struct detailed_timing *timing, void *c)
 		if (closure->preferred)
 			newmode->type |= DRM_MODE_TYPE_PREFERRED;
 
+		/*
+		 * Detailed modes are limited to 10kHz pixel clock resolution,
+		 * so fix up anything that looks like CEA/HDMI mode, but the clock
+		 * is just slightly off.
+		 */
+		fixup_detailed_cea_mode_clock(newmode);
+
 		drm_mode_probed_add(closure->connector, newmode);
 		closure->modes++;
 		closure->preferred = 0;
@@ -3103,6 +3112,45 @@ add_cea_modes(struct drm_connector *connector, struct edid *edid)
 	return modes;
 }
 
+static void fixup_detailed_cea_mode_clock(struct drm_display_mode *mode)
+{
+	const struct drm_display_mode *cea_mode;
+	int clock1, clock2, clock;
+	u8 mode_idx;
+	const char *type;
+
+	mode_idx = drm_match_cea_mode(mode) - 1;
+	if (mode_idx < ARRAY_SIZE(edid_cea_modes)) {
+		type = "CEA";
+		cea_mode = &edid_cea_modes[mode_idx];
+		clock1 = cea_mode->clock;
+		clock2 = cea_mode_alternate_clock(cea_mode);
+	} else {
+		mode_idx = drm_match_hdmi_mode(mode) - 1;
+		if (mode_idx < ARRAY_SIZE(edid_4k_modes)) {
+			type = "HDMI";
+			cea_mode = &edid_4k_modes[mode_idx];
+			clock1 = cea_mode->clock;
+			clock2 = hdmi_mode_alternate_clock(cea_mode);
+		} else {
+			return;
+		}
+	}
+
+	/* pick whichever is closest */
+	if (abs(mode->clock - clock1) < abs(mode->clock - clock2))
+		clock = clock1;
+	else
+		clock = clock2;
+
+	if (mode->clock == clock)
+		return;
+
+	DRM_DEBUG("detailed mode matches %s VIC %d, adjusting clock %d -> %d\n",
+		  type, mode_idx + 1, mode->clock, clock);
+	mode->clock = clock;
+}
+
 static void
 parse_hdmi_vsdb(struct drm_connector *connector, const u8 *db)
 {
-- 
cgit v0.10.2


From 9afd808cf3c45810492f2e05b7c422c99edc88cc Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 8 Oct 2015 11:43:33 +0300
Subject: drm/edid: Round to closest when computing the CEA/HDMI alternate
 clock
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Rounding to the closest kHz seems like the better option that round
down or up when computing the alternate clock for CEA/HDMI modes.
It'll give us a slightly more accurate clock in some cases.

Not sure why I went for the down+up approach originally. Perhaps
I was thinking we can go back and forth betwen the two frequencies
without introducing errors, but round to closest still maintains
that property.

Cc: Adam Jackson <ajax@redhat.com>
Cc: Clint Taylor <clinton.a.taylor@intel.com>
Cc: Libin Yang <libin.yang@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c
index 977915c..d5d2c03 100644
--- a/drivers/gpu/drm/drm_edid.c
+++ b/drivers/gpu/drm/drm_edid.c
@@ -2538,9 +2538,9 @@ cea_mode_alternate_clock(const struct drm_display_mode *cea_mode)
 	 * and the 60Hz variant otherwise.
 	 */
 	if (cea_mode->vdisplay == 240 || cea_mode->vdisplay == 480)
-		clock = clock * 1001 / 1000;
+		clock = DIV_ROUND_CLOSEST(clock * 1001, 1000);
 	else
-		clock = DIV_ROUND_UP(clock * 1000, 1001);
+		clock = DIV_ROUND_CLOSEST(clock * 1000, 1001);
 
 	return clock;
 }
-- 
cgit v0.10.2


From df785aa87f3a43d6784f1d59803646ad69447da8 Mon Sep 17 00:00:00 2001
From: Liviu Dudau <Liviu.Dudau@arm.com>
Date: Tue, 20 Oct 2015 10:23:12 +0100
Subject: drm: Introduce generic probe function for component based masters.

A lot of component based DRM drivers use a variant of the same code
as the probe function. They bind the crtc ports in the first iteration
and then scan through the child nodes and bind the encoders attached
to the remote endpoints. Factor the common code into a separate
function called drm_of_component_probe() in order to increase code
reuse.

Cc: David Airlie <airlied@linux.ie>
Signed-off-by: Liviu Dudau <Liviu.Dudau@arm.com>
Acked-by: Russell King <rmk+kernel@arm.linux.org.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/1445332995-11212-2-git-send-email-Liviu.Dudau@arm.com
Acked-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_of.c b/drivers/gpu/drm/drm_of.c
index be38840..493c05c 100644
--- a/drivers/gpu/drm/drm_of.c
+++ b/drivers/gpu/drm/drm_of.c
@@ -1,3 +1,4 @@
+#include <linux/component.h>
 #include <linux/export.h>
 #include <linux/list.h>
 #include <linux/of_graph.h>
@@ -61,3 +62,90 @@ uint32_t drm_of_find_possible_crtcs(struct drm_device *dev,
 	return possible_crtcs;
 }
 EXPORT_SYMBOL(drm_of_find_possible_crtcs);
+
+/**
+ * drm_of_component_probe - Generic probe function for a component based master
+ * @dev: master device containing the OF node
+ * @compare_of: compare function used for matching components
+ * @master_ops: component master ops to be used
+ *
+ * Parse the platform device OF node and bind all the components associated
+ * with the master. Interface ports are added before the encoders in order to
+ * satisfy their .bind requirements
+ * See Documentation/devicetree/bindings/graph.txt for the bindings.
+ *
+ * Returns zero if successful, or one of the standard error codes if it fails.
+ */
+int drm_of_component_probe(struct device *dev,
+			   int (*compare_of)(struct device *, void *),
+			   const struct component_master_ops *m_ops)
+{
+	struct device_node *ep, *port, *remote;
+	struct component_match *match = NULL;
+	int i;
+
+	if (!dev->of_node)
+		return -EINVAL;
+
+	/*
+	 * Bind the crtc's ports first, so that drm_of_find_possible_crtcs()
+	 * called from encoder's .bind callbacks works as expected
+	 */
+	for (i = 0; ; i++) {
+		port = of_parse_phandle(dev->of_node, "ports", i);
+		if (!port)
+			break;
+
+		if (!of_device_is_available(port->parent)) {
+			of_node_put(port);
+			continue;
+		}
+
+		component_match_add(dev, &match, compare_of, port);
+		of_node_put(port);
+	}
+
+	if (i == 0) {
+		dev_err(dev, "missing 'ports' property\n");
+		return -ENODEV;
+	}
+
+	if (!match) {
+		dev_err(dev, "no available port\n");
+		return -ENODEV;
+	}
+
+	/*
+	 * For bound crtcs, bind the encoders attached to their remote endpoint
+	 */
+	for (i = 0; ; i++) {
+		port = of_parse_phandle(dev->of_node, "ports", i);
+		if (!port)
+			break;
+
+		if (!of_device_is_available(port->parent)) {
+			of_node_put(port);
+			continue;
+		}
+
+		for_each_child_of_node(port, ep) {
+			remote = of_graph_get_remote_port_parent(ep);
+			if (!remote || !of_device_is_available(remote)) {
+				of_node_put(remote);
+				continue;
+			} else if (!of_device_is_available(remote->parent)) {
+				dev_warn(dev, "parent device of %s is not available\n",
+					 remote->full_name);
+				of_node_put(remote);
+				continue;
+			}
+
+			component_match_add(dev, &match, compare_of, remote);
+			of_node_put(remote);
+		}
+		of_node_put(port);
+	}
+
+	return component_master_add_with_match(dev, m_ops, match);
+}
+EXPORT_SYMBOL(drm_of_component_probe);
diff --git a/include/drm/drm_of.h b/include/drm/drm_of.h
index 2441f71..8544665 100644
--- a/include/drm/drm_of.h
+++ b/include/drm/drm_of.h
@@ -1,18 +1,31 @@
 #ifndef __DRM_OF_H__
 #define __DRM_OF_H__
 
+struct component_master_ops;
+struct device;
 struct drm_device;
 struct device_node;
 
 #ifdef CONFIG_OF
 extern uint32_t drm_of_find_possible_crtcs(struct drm_device *dev,
 					   struct device_node *port);
+extern int drm_of_component_probe(struct device *dev,
+				  int (*compare_of)(struct device *, void *),
+				  const struct component_master_ops *m_ops);
 #else
 static inline uint32_t drm_of_find_possible_crtcs(struct drm_device *dev,
 						  struct device_node *port)
 {
 	return 0;
 }
+
+static inline int
+drm_of_component_probe(struct device *dev,
+		       int (*compare_of)(struct device *, void *),
+		       const struct component_master_ops *m_ops)
+{
+	return -EINVAL;
+}
 #endif
 
 #endif /* __DRM_OF_H__ */
-- 
cgit v0.10.2


From 9cace32f7e9fb3d506bab2bfda2cafd11e0b5825 Mon Sep 17 00:00:00 2001
From: Liviu Dudau <Liviu.Dudau@arm.com>
Date: Tue, 20 Oct 2015 10:23:13 +0100
Subject: drm/imx: Convert the probe function to the generic
 drm_of_component_probe()

The generic function is functionally equivalent to the driver's
imx_drm_platform_probe(). Use the generic function and reduce the
overall code size.

Signed-off-by: Liviu Dudau <Liviu.Dudau@arm.com>
Acked-by: Russell King <rmk+kernel@arm.linux.org.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/1445332995-11212-3-git-send-email-Liviu.Dudau@arm.com
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/imx/imx-drm-core.c b/drivers/gpu/drm/imx/imx-drm-core.c
index de00a6c..64f16ea 100644
--- a/drivers/gpu/drm/imx/imx-drm-core.c
+++ b/drivers/gpu/drm/imx/imx-drm-core.c
@@ -531,59 +531,12 @@ static const struct component_master_ops imx_drm_ops = {
 
 static int imx_drm_platform_probe(struct platform_device *pdev)
 {
-	struct device_node *ep, *port, *remote;
-	struct component_match *match = NULL;
-	int ret;
-	int i;
-
-	/*
-	 * Bind the IPU display interface ports first, so that
-	 * imx_drm_encoder_parse_of called from encoder .bind callbacks
-	 * works as expected.
-	 */
-	for (i = 0; ; i++) {
-		port = of_parse_phandle(pdev->dev.of_node, "ports", i);
-		if (!port)
-			break;
-
-		component_match_add(&pdev->dev, &match, compare_of, port);
-	}
+	int ret = drm_of_component_probe(&pdev->dev, compare_of, &imx_drm_ops);
 
-	if (i == 0) {
-		dev_err(&pdev->dev, "missing 'ports' property\n");
-		return -ENODEV;
-	}
+	if (!ret)
+		ret = dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(32));
 
-	/* Then bind all encoders */
-	for (i = 0; ; i++) {
-		port = of_parse_phandle(pdev->dev.of_node, "ports", i);
-		if (!port)
-			break;
-
-		for_each_child_of_node(port, ep) {
-			remote = of_graph_get_remote_port_parent(ep);
-			if (!remote || !of_device_is_available(remote)) {
-				of_node_put(remote);
-				continue;
-			} else if (!of_device_is_available(remote->parent)) {
-				dev_warn(&pdev->dev, "parent device of %s is not available\n",
-					 remote->full_name);
-				of_node_put(remote);
-				continue;
-			}
-
-			component_match_add(&pdev->dev, &match, compare_of,
-					    remote);
-			of_node_put(remote);
-		}
-		of_node_put(port);
-	}
-
-	ret = dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(32));
-	if (ret)
-		return ret;
-
-	return component_master_add_with_match(&pdev->dev, &imx_drm_ops, match);
+	return ret;
 }
 
 static int imx_drm_platform_remove(struct platform_device *pdev)
-- 
cgit v0.10.2


From 52f5eb60940de889ce98a876f6933b574ead3225 Mon Sep 17 00:00:00 2001
From: Liviu Dudau <Liviu.Dudau@arm.com>
Date: Tue, 20 Oct 2015 10:23:14 +0100
Subject: drm/rockchip: Convert the probe function to the generic
 drm_of_component_probe()

Use the generic drm_of_component_probe() function to probe for components.

Signed-off-by: Liviu Dudau <Liviu.Dudau@arm.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1445332995-11212-4-git-send-email-Liviu.Dudau@arm.com
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_drv.c b/drivers/gpu/drm/rockchip/rockchip_drm_drv.c
index f22e1e1..d26e0cc 100644
--- a/drivers/gpu/drm/rockchip/rockchip_drm_drv.c
+++ b/drivers/gpu/drm/rockchip/rockchip_drm_drv.c
@@ -19,6 +19,7 @@
 #include <drm/drmP.h>
 #include <drm/drm_crtc_helper.h>
 #include <drm/drm_fb_helper.h>
+#include <drm/drm_of.h>
 #include <linux/dma-mapping.h>
 #include <linux/pm_runtime.h>
 #include <linux/module.h>
@@ -418,29 +419,6 @@ static int compare_of(struct device *dev, void *data)
 	return dev->of_node == np;
 }
 
-static void rockchip_add_endpoints(struct device *dev,
-				   struct component_match **match,
-				   struct device_node *port)
-{
-	struct device_node *ep, *remote;
-
-	for_each_child_of_node(port, ep) {
-		remote = of_graph_get_remote_port_parent(ep);
-		if (!remote || !of_device_is_available(remote)) {
-			of_node_put(remote);
-			continue;
-		} else if (!of_device_is_available(remote->parent)) {
-			dev_warn(dev, "parent device of %s is not available\n",
-				 remote->full_name);
-			of_node_put(remote);
-			continue;
-		}
-
-		component_match_add(dev, match, compare_of, remote);
-		of_node_put(remote);
-	}
-}
-
 static int rockchip_drm_bind(struct device *dev)
 {
 	struct drm_device *drm;
@@ -483,61 +461,14 @@ static const struct component_master_ops rockchip_drm_ops = {
 
 static int rockchip_drm_platform_probe(struct platform_device *pdev)
 {
-	struct device *dev = &pdev->dev;
-	struct component_match *match = NULL;
-	struct device_node *np = dev->of_node;
-	struct device_node *port;
-	int i;
-
-	if (!np)
-		return -ENODEV;
-	/*
-	 * Bind the crtc ports first, so that
-	 * drm_of_find_possible_crtcs called from encoder .bind callbacks
-	 * works as expected.
-	 */
-	for (i = 0;; i++) {
-		port = of_parse_phandle(np, "ports", i);
-		if (!port)
-			break;
-
-		if (!of_device_is_available(port->parent)) {
-			of_node_put(port);
-			continue;
-		}
-
-		component_match_add(dev, &match, compare_of, port->parent);
-		of_node_put(port);
-	}
+	int ret = drm_of_component_probe(&pdev->dev, compare_of,
+					 &rockchip_drm_ops);
 
-	if (i == 0) {
-		dev_err(dev, "missing 'ports' property\n");
+	/* keep compatibility with old code that was returning -ENODEV */
+	if (ret == -EINVAL)
 		return -ENODEV;
-	}
 
-	if (!match) {
-		dev_err(dev, "No available vop found for display-subsystem.\n");
-		return -ENODEV;
-	}
-	/*
-	 * For each bound crtc, bind the encoders attached to its
-	 * remote endpoint.
-	 */
-	for (i = 0;; i++) {
-		port = of_parse_phandle(np, "ports", i);
-		if (!port)
-			break;
-
-		if (!of_device_is_available(port->parent)) {
-			of_node_put(port);
-			continue;
-		}
-
-		rockchip_add_endpoints(dev, &match, port);
-		of_node_put(port);
-	}
-
-	return component_master_add_with_match(dev, &rockchip_drm_ops, match);
+	return ret;
 }
 
 static int rockchip_drm_platform_remove(struct platform_device *pdev)
-- 
cgit v0.10.2


From 7fc7e86d096f13cf88391b09a3a387172450ad79 Mon Sep 17 00:00:00 2001
From: Liviu Dudau <Liviu.Dudau@arm.com>
Date: Tue, 20 Oct 2015 10:23:15 +0100
Subject: drm/armada: Convert the probe function to the generic
 drm_of_component_probe()

The armada DRM driver keeps some old platform data compatibility in the
probe function that makes moving to the generic drm_of_component_probe()
a bit more complicated that it should. Refactor the probe function to do
the platform_data processing after the generic probe (and only if that
fails). This way future cleanup can further remove support for it.

Signed-off-by: Liviu Dudau <Liviu.Dudau@arm.com>
Acked-by: Russell King <rmk+kernel@arm.linux.org.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/1445332995-11212-5-git-send-email-Liviu.Dudau@arm.com
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/armada/armada_drv.c b/drivers/gpu/drm/armada/armada_drv.c
index 1cbb080..77ab93d 100644
--- a/drivers/gpu/drm/armada/armada_drv.c
+++ b/drivers/gpu/drm/armada/armada_drv.c
@@ -11,6 +11,7 @@
 #include <linux/of_graph.h>
 #include <drm/drmP.h>
 #include <drm/drm_crtc_helper.h>
+#include <drm/drm_of.h>
 #include "armada_crtc.h"
 #include "armada_drm.h"
 #include "armada_gem.h"
@@ -262,43 +263,29 @@ static void armada_add_endpoints(struct device *dev,
 	}
 }
 
-static int armada_drm_find_components(struct device *dev,
-	struct component_match **match)
-{
-	struct device_node *port;
-	int i;
-
-	if (dev->of_node) {
-		struct device_node *np = dev->of_node;
-
-		for (i = 0; ; i++) {
-			port = of_parse_phandle(np, "ports", i);
-			if (!port)
-				break;
-
-			component_match_add(dev, match, compare_of, port);
-			of_node_put(port);
-		}
+static const struct component_master_ops armada_master_ops = {
+	.bind = armada_drm_bind,
+	.unbind = armada_drm_unbind,
+};
 
-		if (i == 0) {
-			dev_err(dev, "missing 'ports' property\n");
-			return -ENODEV;
-		}
+static int armada_drm_probe(struct platform_device *pdev)
+{
+	struct component_match *match = NULL;
+	struct device *dev = &pdev->dev;
+	int ret;
 
-		for (i = 0; ; i++) {
-			port = of_parse_phandle(np, "ports", i);
-			if (!port)
-				break;
+	ret = drm_of_component_probe(dev, compare_dev_name, &armada_master_ops);
+	if (ret != -EINVAL)
+		return ret;
 
-			armada_add_endpoints(dev, match, port);
-			of_node_put(port);
-		}
-	} else if (dev->platform_data) {
+	if (dev->platform_data) {
 		char **devices = dev->platform_data;
+		struct device_node *port;
 		struct device *d;
+		int i;
 
 		for (i = 0; devices[i]; i++)
-			component_match_add(dev, match, compare_dev_name,
+			component_match_add(dev, &match, compare_dev_name,
 					    devices[i]);
 
 		if (i == 0) {
@@ -308,32 +295,15 @@ static int armada_drm_find_components(struct device *dev,
 
 		for (i = 0; devices[i]; i++) {
 			d = bus_find_device_by_name(&platform_bus_type, NULL,
-					devices[i]);
+						    devices[i]);
 			if (d && d->of_node) {
 				for_each_child_of_node(d->of_node, port)
-					armada_add_endpoints(dev, match, port);
+					armada_add_endpoints(dev, &match, port);
 			}
 			put_device(d);
 		}
 	}
 
-	return 0;
-}
-
-static const struct component_master_ops armada_master_ops = {
-	.bind = armada_drm_bind,
-	.unbind = armada_drm_unbind,
-};
-
-static int armada_drm_probe(struct platform_device *pdev)
-{
-	struct component_match *match = NULL;
-	int ret;
-
-	ret = armada_drm_find_components(&pdev->dev, &match);
-	if (ret < 0)
-		return ret;
-
 	return component_master_add_with_match(&pdev->dev, &armada_master_ops,
 					       match);
 }
-- 
cgit v0.10.2


From 5d170139eb10ae12e1bd076245c42b35453d8324 Mon Sep 17 00:00:00 2001
From: Lukas Wunner <lukas@wunner.de>
Date: Sun, 18 Oct 2015 13:05:40 +0200
Subject: vga_switcheroo: Constify vga_switcheroo_handler
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

vga_switcheroo_client_ops has always been declared const since its
introduction with 26ec685ff9d9 ("vga_switcheroo: Introduce struct
vga_switcheroo_client_ops").

Do so for vga_switcheroo_handler as well.

 drivers/gpu/drm/amd/amdgpu/amdgpu.ko:
   6 .rodata       00009888
- 19 .data         00001f00
+ 19 .data         00001ee0
 drivers/gpu/drm/nouveau/nouveau.ko:
   6 .rodata       000460b8
  17 .data         00018fe0
 drivers/gpu/drm/radeon/radeon.ko:
-  7 .rodata       00030944
+  7 .rodata       00030964
- 21 .data         0000d6a0
+ 21 .data         0000d678
 drivers/platform/x86/apple-gmux.ko:
-  7 .rodata       00000140
+  7 .rodata       00000160
- 11 .data         000000e0
+ 11 .data         000000b8

Cc: Ben Skeggs <bskeggs@redhat.com>
Cc: Darren Hart <dvhart@linux.intel.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Lukas Wunner <lukas@wunner.de>
Reviewed-by: Christian König <christian.koenig@amd.com>.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c
index 1a6b239..5a8fbad 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c
@@ -501,7 +501,7 @@ static int amdgpu_atpx_get_client_id(struct pci_dev *pdev)
 		return VGA_SWITCHEROO_DIS;
 }
 
-static struct vga_switcheroo_handler amdgpu_atpx_handler = {
+static const struct vga_switcheroo_handler amdgpu_atpx_handler = {
 	.switchto = amdgpu_atpx_switchto,
 	.power_state = amdgpu_atpx_power_state,
 	.init = amdgpu_atpx_init,
diff --git a/drivers/gpu/drm/nouveau/nouveau_acpi.c b/drivers/gpu/drm/nouveau/nouveau_acpi.c
index df2d981..8b8332e 100644
--- a/drivers/gpu/drm/nouveau/nouveau_acpi.c
+++ b/drivers/gpu/drm/nouveau/nouveau_acpi.c
@@ -206,7 +206,7 @@ static int nouveau_dsm_get_client_id(struct pci_dev *pdev)
 	return VGA_SWITCHEROO_DIS;
 }
 
-static struct vga_switcheroo_handler nouveau_dsm_handler = {
+static const struct vga_switcheroo_handler nouveau_dsm_handler = {
 	.switchto = nouveau_dsm_switchto,
 	.power_state = nouveau_dsm_power_state,
 	.get_client_id = nouveau_dsm_get_client_id,
diff --git a/drivers/gpu/drm/radeon/radeon_atpx_handler.c b/drivers/gpu/drm/radeon/radeon_atpx_handler.c
index a771b9f..c4b4f29 100644
--- a/drivers/gpu/drm/radeon/radeon_atpx_handler.c
+++ b/drivers/gpu/drm/radeon/radeon_atpx_handler.c
@@ -499,7 +499,7 @@ static int radeon_atpx_get_client_id(struct pci_dev *pdev)
 		return VGA_SWITCHEROO_DIS;
 }
 
-static struct vga_switcheroo_handler radeon_atpx_handler = {
+static const struct vga_switcheroo_handler radeon_atpx_handler = {
 	.switchto = radeon_atpx_switchto,
 	.power_state = radeon_atpx_power_state,
 	.init = radeon_atpx_init,
diff --git a/drivers/gpu/vga/vga_switcheroo.c b/drivers/gpu/vga/vga_switcheroo.c
index af0d372..56bbbd6 100644
--- a/drivers/gpu/vga/vga_switcheroo.c
+++ b/drivers/gpu/vga/vga_switcheroo.c
@@ -140,7 +140,7 @@ struct vgasr_priv {
 	int registered_clients;
 	struct list_head clients;
 
-	struct vga_switcheroo_handler *handler;
+	const struct vga_switcheroo_handler *handler;
 };
 
 #define ID_BIT_AUDIO		0x100
@@ -195,7 +195,7 @@ static void vga_switcheroo_enable(void)
  *
  * Return: 0 on success, -EINVAL if a handler was already registered.
  */
-int vga_switcheroo_register_handler(struct vga_switcheroo_handler *handler)
+int vga_switcheroo_register_handler(const struct vga_switcheroo_handler *handler)
 {
 	mutex_lock(&vgasr_mutex);
 	if (vgasr_priv.handler) {
diff --git a/drivers/platform/x86/apple-gmux.c b/drivers/platform/x86/apple-gmux.c
index 0dec3f5..976efeb 100644
--- a/drivers/platform/x86/apple-gmux.c
+++ b/drivers/platform/x86/apple-gmux.c
@@ -346,7 +346,7 @@ gmux_active_client(struct apple_gmux_data *gmux_data)
 	return VGA_SWITCHEROO_DIS;
 }
 
-static struct vga_switcheroo_handler gmux_handler = {
+static const struct vga_switcheroo_handler gmux_handler = {
 	.switchto = gmux_switchto,
 	.power_state = gmux_set_power_state,
 	.get_client_id = gmux_get_client_id,
diff --git a/include/linux/vga_switcheroo.h b/include/linux/vga_switcheroo.h
index c557511..786bc93 100644
--- a/include/linux/vga_switcheroo.h
+++ b/include/linux/vga_switcheroo.h
@@ -137,7 +137,7 @@ int vga_switcheroo_register_audio_client(struct pci_dev *pdev,
 void vga_switcheroo_client_fb_set(struct pci_dev *dev,
 				  struct fb_info *info);
 
-int vga_switcheroo_register_handler(struct vga_switcheroo_handler *handler);
+int vga_switcheroo_register_handler(const struct vga_switcheroo_handler *handler);
 void vga_switcheroo_unregister_handler(void);
 
 int vga_switcheroo_process_delayed_switch(void);
@@ -155,7 +155,7 @@ static inline void vga_switcheroo_unregister_client(struct pci_dev *dev) {}
 static inline int vga_switcheroo_register_client(struct pci_dev *dev,
 		const struct vga_switcheroo_client_ops *ops, bool driver_power_control) { return 0; }
 static inline void vga_switcheroo_client_fb_set(struct pci_dev *dev, struct fb_info *info) {}
-static inline int vga_switcheroo_register_handler(struct vga_switcheroo_handler *handler) { return 0; }
+static inline int vga_switcheroo_register_handler(const struct vga_switcheroo_handler *handler) { return 0; }
 static inline int vga_switcheroo_register_audio_client(struct pci_dev *pdev,
 	const struct vga_switcheroo_client_ops *ops,
 	enum vga_switcheroo_client_id id) { return 0; }
-- 
cgit v0.10.2


From 48aa1e748f29373fdcc2bc341eac08ef16bff269 Mon Sep 17 00:00:00 2001
From: Insu Yun <wuninsu@gmail.com>
Date: Mon, 19 Oct 2015 16:33:30 +0000
Subject: drm: correctly check failed allocation

drm_property_create_range can be failed in memory pressure
Therefore, check return value and handle an error

Signed-off-by: Insu Yun <wuninsu@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c
index e54660a..720a153 100644
--- a/drivers/gpu/drm/drm_crtc.c
+++ b/drivers/gpu/drm/drm_crtc.c
@@ -1533,6 +1533,9 @@ int drm_mode_create_tv_properties(struct drm_device *dev,
 					  "select subconnector",
 					  drm_tv_select_enum_list,
 					  ARRAY_SIZE(drm_tv_select_enum_list));
+	if (!tv_selector)
+		goto nomem;
+
 	dev->mode_config.tv_select_subconnector_property = tv_selector;
 
 	tv_subconnector =
@@ -1540,6 +1543,8 @@ int drm_mode_create_tv_properties(struct drm_device *dev,
 				    "subconnector",
 				    drm_tv_subconnector_enum_list,
 				    ARRAY_SIZE(drm_tv_subconnector_enum_list));
+	if (!tv_subconnector)
+		goto nomem;
 	dev->mode_config.tv_subconnector_property = tv_subconnector;
 
 	/*
@@ -1547,42 +1552,67 @@ int drm_mode_create_tv_properties(struct drm_device *dev,
 	 */
 	dev->mode_config.tv_left_margin_property =
 		drm_property_create_range(dev, 0, "left margin", 0, 100);
+	if (!dev->mode_config.tv_left_margin_property)
+		goto nomem;
 
 	dev->mode_config.tv_right_margin_property =
 		drm_property_create_range(dev, 0, "right margin", 0, 100);
+	if (!dev->mode_config.tv_right_margin_property)
+		goto nomem;
 
 	dev->mode_config.tv_top_margin_property =
 		drm_property_create_range(dev, 0, "top margin", 0, 100);
+	if (!dev->mode_config.tv_top_margin_property)
+		goto nomem;
 
 	dev->mode_config.tv_bottom_margin_property =
 		drm_property_create_range(dev, 0, "bottom margin", 0, 100);
+	if (!dev->mode_config.tv_bottom_margin_property)
+		goto nomem;
 
 	dev->mode_config.tv_mode_property =
 		drm_property_create(dev, DRM_MODE_PROP_ENUM,
 				    "mode", num_modes);
+	if (!dev->mode_config.tv_mode_property)
+		goto nomem;
+
 	for (i = 0; i < num_modes; i++)
 		drm_property_add_enum(dev->mode_config.tv_mode_property, i,
 				      i, modes[i]);
 
 	dev->mode_config.tv_brightness_property =
 		drm_property_create_range(dev, 0, "brightness", 0, 100);
+	if (!dev->mode_config.tv_brightness_property)
+		goto nomem;
 
 	dev->mode_config.tv_contrast_property =
 		drm_property_create_range(dev, 0, "contrast", 0, 100);
+	if (!dev->mode_config.tv_contrast_property)
+		goto nomem;
 
 	dev->mode_config.tv_flicker_reduction_property =
 		drm_property_create_range(dev, 0, "flicker reduction", 0, 100);
+	if (!dev->mode_config.tv_flicker_reduction_property)
+		goto nomem;
 
 	dev->mode_config.tv_overscan_property =
 		drm_property_create_range(dev, 0, "overscan", 0, 100);
+	if (!dev->mode_config.tv_overscan_property)
+		goto nomem;
 
 	dev->mode_config.tv_saturation_property =
 		drm_property_create_range(dev, 0, "saturation", 0, 100);
+	if (!dev->mode_config.tv_saturation_property)
+		goto nomem;
 
 	dev->mode_config.tv_hue_property =
 		drm_property_create_range(dev, 0, "hue", 0, 100);
+	if (!dev->mode_config.tv_hue_property)
+		goto nomem;
 
 	return 0;
+nomem:
+	return -ENOMEM;
 }
 EXPORT_SYMBOL(drm_mode_create_tv_properties);
 
-- 
cgit v0.10.2


From 1f9573241080449d221b6f8a9cc68e28fdeb7408 Mon Sep 17 00:00:00 2001
From: Eric Anholt <eric@anholt.net>
Date: Tue, 11 Aug 2015 13:17:11 -0700
Subject: drm/vc4: Add devicetree bindings for VC4.

VC4 is the GPU (display and 3D) subsystem present on the 2835 and some
other Broadcom SoCs.

This binding follows the model of msm, imx, sti, and others, where
there is a subsystem node for the whole GPU, with nodes for the
individual HW components within it.

v2: Extend the commit message, fix several nits from Stephen Warren.
v3: Rename the compatibility strings, clean up node names, drop the
    unnecessary lists of components.  Use compatibility strings for
    choosing CRTC HVS channel numbers.  Document the HDMI clock usage.
v4: Whitespace fix, expand acronyms, move to display/ instead of gpu/,
    rename "hpd-gpio" to "hpd-gpios".

Signed-off-by: Eric Anholt <eric@anholt.net>

diff --git a/Documentation/devicetree/bindings/display/brcm,bcm-vc4.txt b/Documentation/devicetree/bindings/display/brcm,bcm-vc4.txt
new file mode 100644
index 0000000..56a961a
--- /dev/null
+++ b/Documentation/devicetree/bindings/display/brcm,bcm-vc4.txt
@@ -0,0 +1,65 @@
+Broadcom VC4 (VideoCore4) GPU
+
+The VC4 device present on the Raspberry Pi includes a display system
+with HDMI output and the HVS (Hardware Video Scaler) for compositing
+display planes.
+
+Required properties for VC4:
+- compatible:	Should be "brcm,bcm2835-vc4"
+
+Required properties for Pixel Valve:
+- compatible:	Should be one of "brcm,bcm2835-pixelvalve0",
+		  "brcm,bcm2835-pixelvalve1", or "brcm,bcm2835-pixelvalve2"
+- reg:		Physical base address and length of the PV's registers
+- interrupts:	The interrupt number
+		  See bindings/interrupt-controller/brcm,bcm2835-armctrl-ic.txt
+
+Required properties for HVS:
+- compatible:	Should be "brcm,bcm2835-hvs"
+- reg:		Physical base address and length of the HVS's registers
+- interrupts:	The interrupt number
+		  See bindings/interrupt-controller/brcm,bcm2835-armctrl-ic.txt
+
+Required properties for HDMI
+- compatible:	Should be "brcm,bcm2835-hdmi"
+- reg:		Physical base address and length of the two register ranges
+		  ("HDMI" and "HD", in that order)
+- interrupts:	The interrupt numbers
+		  See bindings/interrupt-controller/brcm,bcm2835-armctrl-ic.txt
+- ddc:		phandle of the I2C controller used for DDC EDID probing
+- clocks:	a) hdmi: The HDMI state machine clock
+		b) pixel: The pixel clock.
+
+Optional properties for HDMI:
+- hpd-gpios:	The GPIO pin for HDMI hotplug detect (if it doesn't appear
+		  as an interrupt/status bit in the HDMI controller
+		  itself).  See bindings/pinctrl/brcm,bcm2835-gpio.txt
+
+Example:
+pixelvalve@7e807000 {
+	compatible = "brcm,bcm2835-pixelvalve2";
+	reg = <0x7e807000 0x100>;
+	interrupts = <2 10>; /* pixelvalve */
+};
+
+hvs@7e400000 {
+	compatible = "brcm,bcm2835-hvs";
+	reg = <0x7e400000 0x6000>;
+	interrupts = <2 1>;
+};
+
+hdmi: hdmi@7e902000 {
+	compatible = "brcm,bcm2835-hdmi";
+	reg = <0x7e902000 0x600>,
+	      <0x7e808000 0x100>;
+	interrupts = <2 8>, <2 9>;
+	ddc = <&i2c2>;
+	hpd-gpios = <&gpio 46 GPIO_ACTIVE_HIGH>;
+	clocks = <&clocks BCM2835_PLLH_PIX>,
+		 <&clocks BCM2835_CLOCK_HSM>;
+	clock-names = "pixel", "hdmi";
+};
+
+vc4: gpu {
+	compatible = "brcm,bcm2835-vc4";
+};
-- 
cgit v0.10.2


From c8b75bca92cbf064b9fa125fc74a85994452e935 Mon Sep 17 00:00:00 2001
From: Eric Anholt <eric@anholt.net>
Date: Mon, 2 Mar 2015 13:01:12 -0800
Subject: drm/vc4: Add KMS support for Raspberry Pi.

This is enough for fbcon and bringing up X using
xf86-video-modesetting.  It doesn't support the 3D accelerator or
power management yet.

v2: Drop FB_HELPER select thanks to Archit's patches.  Do manual init
    ordering instead of using the .load hook.  Structure registration
    more like tegra's, but still using the typical "component" code.
    Drop no-op hooks for atomic_begin and mode_fixup() now that
    they're optional.  Drop sentinel in Makefile.  Fix minor style
    nits I noticed on another reread.

v3: Use the new bcm2835 clk driver to manage pixel/HSM clocks instead
    of having a fixed video mode.  Use exynos-style component driver
    matching instead of devicetree nodes to list the component driver
    instances.  Rename compatibility strings to say bcm2835, and
    distinguish pv0/1/2.  Clean up some h/vsync code, and add in
    interlaced mode setup.  Fix up probe/bind error paths.  Use
    bitops.h macros for vc4_regs.h

v4: Include i2c.h, allow building under COMPILE_TEST, drop msleep now
    that other bugs have been fixed, add timeouts to cpu_relax()
    loops, rename hpd-gpio to hpd-gpios.

Signed-off-by: Eric Anholt <eric@anholt.net>
Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>

diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig
index 1a0a8df..c4bf9a1 100644
--- a/drivers/gpu/drm/Kconfig
+++ b/drivers/gpu/drm/Kconfig
@@ -264,3 +264,5 @@ source "drivers/gpu/drm/sti/Kconfig"
 source "drivers/gpu/drm/amd/amdkfd/Kconfig"
 
 source "drivers/gpu/drm/imx/Kconfig"
+
+source "drivers/gpu/drm/vc4/Kconfig"
diff --git a/drivers/gpu/drm/Makefile b/drivers/gpu/drm/Makefile
index 45e7719..0edc9e2 100644
--- a/drivers/gpu/drm/Makefile
+++ b/drivers/gpu/drm/Makefile
@@ -42,6 +42,7 @@ obj-$(CONFIG_DRM_MGA)	+= mga/
 obj-$(CONFIG_DRM_I810)	+= i810/
 obj-$(CONFIG_DRM_I915)  += i915/
 obj-$(CONFIG_DRM_MGAG200) += mgag200/
+obj-$(CONFIG_DRM_VC4)  += vc4/
 obj-$(CONFIG_DRM_CIRRUS_QEMU) += cirrus/
 obj-$(CONFIG_DRM_SIS)   += sis/
 obj-$(CONFIG_DRM_SAVAGE)+= savage/
diff --git a/drivers/gpu/drm/vc4/Kconfig b/drivers/gpu/drm/vc4/Kconfig
new file mode 100644
index 0000000..e502802
--- /dev/null
+++ b/drivers/gpu/drm/vc4/Kconfig
@@ -0,0 +1,13 @@
+config DRM_VC4
+	tristate "Broadcom VC4 Graphics"
+	depends on ARCH_BCM2835 || COMPILE_TEST
+	depends on DRM
+	select DRM_KMS_HELPER
+	select DRM_KMS_CMA_HELPER
+	help
+	  Choose this option if you have a system that has a Broadcom
+	  VC4 GPU, such as the Raspberry Pi or other BCM2708/BCM2835.
+
+	  This driver requires that "avoid_warnings=2" be present in
+	  the config.txt for the firmware, to keep it from smashing
+	  our display setup.
diff --git a/drivers/gpu/drm/vc4/Makefile b/drivers/gpu/drm/vc4/Makefile
new file mode 100644
index 0000000..32b4f9c
--- /dev/null
+++ b/drivers/gpu/drm/vc4/Makefile
@@ -0,0 +1,17 @@
+ccflags-y := -Iinclude/drm
+
+# Please keep these build lists sorted!
+
+# core driver code
+vc4-y := \
+	vc4_bo.o \
+	vc4_crtc.o \
+	vc4_drv.o \
+	vc4_kms.o \
+	vc4_hdmi.o \
+	vc4_hvs.o \
+	vc4_plane.o
+
+vc4-$(CONFIG_DEBUG_FS) += vc4_debugfs.o
+
+obj-$(CONFIG_DRM_VC4)  += vc4.o
diff --git a/drivers/gpu/drm/vc4/vc4_bo.c b/drivers/gpu/drm/vc4/vc4_bo.c
new file mode 100644
index 0000000..ab9f510
--- /dev/null
+++ b/drivers/gpu/drm/vc4/vc4_bo.c
@@ -0,0 +1,52 @@
+/*
+ *  Copyright © 2015 Broadcom
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+/* DOC: VC4 GEM BO management support.
+ *
+ * The VC4 GPU architecture (both scanout and rendering) has direct
+ * access to system memory with no MMU in between.  To support it, we
+ * use the GEM CMA helper functions to allocate contiguous ranges of
+ * physical memory for our BOs.
+ */
+
+#include "vc4_drv.h"
+
+struct vc4_bo *vc4_bo_create(struct drm_device *dev, size_t size)
+{
+	struct drm_gem_cma_object *cma_obj;
+
+	cma_obj = drm_gem_cma_create(dev, size);
+	if (IS_ERR(cma_obj))
+		return NULL;
+	else
+		return to_vc4_bo(&cma_obj->base);
+}
+
+int vc4_dumb_create(struct drm_file *file_priv,
+		    struct drm_device *dev,
+		    struct drm_mode_create_dumb *args)
+{
+	int min_pitch = DIV_ROUND_UP(args->width * args->bpp, 8);
+	struct vc4_bo *bo = NULL;
+	int ret;
+
+	if (args->pitch < min_pitch)
+		args->pitch = min_pitch;
+
+	if (args->size < args->pitch * args->height)
+		args->size = args->pitch * args->height;
+
+	bo = vc4_bo_create(dev, roundup(args->size, PAGE_SIZE));
+	if (!bo)
+		return -ENOMEM;
+
+	ret = drm_gem_handle_create(file_priv, &bo->base.base, &args->handle);
+	drm_gem_object_unreference_unlocked(&bo->base.base);
+
+	return ret;
+}
diff --git a/drivers/gpu/drm/vc4/vc4_crtc.c b/drivers/gpu/drm/vc4/vc4_crtc.c
new file mode 100644
index 0000000..8489d5b
--- /dev/null
+++ b/drivers/gpu/drm/vc4/vc4_crtc.c
@@ -0,0 +1,672 @@
+/*
+ * Copyright (C) 2015 Broadcom
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+/**
+ * DOC: VC4 CRTC module
+ *
+ * In VC4, the Pixel Valve is what most closely corresponds to the
+ * DRM's concept of a CRTC.  The PV generates video timings from the
+ * output's clock plus its configuration.  It pulls scaled pixels from
+ * the HVS at that timing, and feeds it to the encoder.
+ *
+ * However, the DRM CRTC also collects the configuration of all the
+ * DRM planes attached to it.  As a result, this file also manages
+ * setup of the VC4 HVS's display elements on the CRTC.
+ *
+ * The 2835 has 3 different pixel valves.  pv0 in the audio power
+ * domain feeds DSI0 or DPI, while pv1 feeds DS1 or SMI.  pv2 in the
+ * image domain can feed either HDMI or the SDTV controller.  The
+ * pixel valve chooses from the CPRMAN clocks (HSM for HDMI, VEC for
+ * SDTV, etc.) according to which output type is chosen in the mux.
+ *
+ * For power management, the pixel valve's registers are all clocked
+ * by the AXI clock, while the timings and FIFOs make use of the
+ * output-specific clock.  Since the encoders also directly consume
+ * the CPRMAN clocks, and know what timings they need, they are the
+ * ones that set the clock.
+ */
+
+#include "drm_atomic.h"
+#include "drm_atomic_helper.h"
+#include "drm_crtc_helper.h"
+#include "linux/clk.h"
+#include "linux/component.h"
+#include "linux/of_device.h"
+#include "vc4_drv.h"
+#include "vc4_regs.h"
+
+struct vc4_crtc {
+	struct drm_crtc base;
+	const struct vc4_crtc_data *data;
+	void __iomem *regs;
+
+	/* Which HVS channel we're using for our CRTC. */
+	int channel;
+
+	/* Pointer to the actual hardware display list memory for the
+	 * crtc.
+	 */
+	u32 __iomem *dlist;
+
+	u32 dlist_size; /* in dwords */
+
+	struct drm_pending_vblank_event *event;
+};
+
+static inline struct vc4_crtc *
+to_vc4_crtc(struct drm_crtc *crtc)
+{
+	return (struct vc4_crtc *)crtc;
+}
+
+struct vc4_crtc_data {
+	/* Which channel of the HVS this pixelvalve sources from. */
+	int hvs_channel;
+
+	enum vc4_encoder_type encoder0_type;
+	enum vc4_encoder_type encoder1_type;
+};
+
+#define CRTC_WRITE(offset, val) writel(val, vc4_crtc->regs + (offset))
+#define CRTC_READ(offset) readl(vc4_crtc->regs + (offset))
+
+#define CRTC_REG(reg) { reg, #reg }
+static const struct {
+	u32 reg;
+	const char *name;
+} crtc_regs[] = {
+	CRTC_REG(PV_CONTROL),
+	CRTC_REG(PV_V_CONTROL),
+	CRTC_REG(PV_VSYNCD),
+	CRTC_REG(PV_HORZA),
+	CRTC_REG(PV_HORZB),
+	CRTC_REG(PV_VERTA),
+	CRTC_REG(PV_VERTB),
+	CRTC_REG(PV_VERTA_EVEN),
+	CRTC_REG(PV_VERTB_EVEN),
+	CRTC_REG(PV_INTEN),
+	CRTC_REG(PV_INTSTAT),
+	CRTC_REG(PV_STAT),
+	CRTC_REG(PV_HACT_ACT),
+};
+
+static void vc4_crtc_dump_regs(struct vc4_crtc *vc4_crtc)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(crtc_regs); i++) {
+		DRM_INFO("0x%04x (%s): 0x%08x\n",
+			 crtc_regs[i].reg, crtc_regs[i].name,
+			 CRTC_READ(crtc_regs[i].reg));
+	}
+}
+
+#ifdef CONFIG_DEBUG_FS
+int vc4_crtc_debugfs_regs(struct seq_file *m, void *unused)
+{
+	struct drm_info_node *node = (struct drm_info_node *)m->private;
+	struct drm_device *dev = node->minor->dev;
+	int crtc_index = (uintptr_t)node->info_ent->data;
+	struct drm_crtc *crtc;
+	struct vc4_crtc *vc4_crtc;
+	int i;
+
+	i = 0;
+	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
+		if (i == crtc_index)
+			break;
+		i++;
+	}
+	if (!crtc)
+		return 0;
+	vc4_crtc = to_vc4_crtc(crtc);
+
+	for (i = 0; i < ARRAY_SIZE(crtc_regs); i++) {
+		seq_printf(m, "%s (0x%04x): 0x%08x\n",
+			   crtc_regs[i].name, crtc_regs[i].reg,
+			   CRTC_READ(crtc_regs[i].reg));
+	}
+
+	return 0;
+}
+#endif
+
+static void vc4_crtc_destroy(struct drm_crtc *crtc)
+{
+	drm_crtc_cleanup(crtc);
+}
+
+static u32 vc4_get_fifo_full_level(u32 format)
+{
+	static const u32 fifo_len_bytes = 64;
+	static const u32 hvs_latency_pix = 6;
+
+	switch (format) {
+	case PV_CONTROL_FORMAT_DSIV_16:
+	case PV_CONTROL_FORMAT_DSIC_16:
+		return fifo_len_bytes - 2 * hvs_latency_pix;
+	case PV_CONTROL_FORMAT_DSIV_18:
+		return fifo_len_bytes - 14;
+	case PV_CONTROL_FORMAT_24:
+	case PV_CONTROL_FORMAT_DSIV_24:
+	default:
+		return fifo_len_bytes - 3 * hvs_latency_pix;
+	}
+}
+
+/*
+ * Returns the clock select bit for the connector attached to the
+ * CRTC.
+ */
+static int vc4_get_clock_select(struct drm_crtc *crtc)
+{
+	struct drm_connector *connector;
+
+	drm_for_each_connector(connector, crtc->dev) {
+		if (connector && connector->state->crtc == crtc) {
+			struct drm_encoder *encoder = connector->encoder;
+			struct vc4_encoder *vc4_encoder =
+				to_vc4_encoder(encoder);
+
+			return vc4_encoder->clock_select;
+		}
+	}
+
+	return -1;
+}
+
+static void vc4_crtc_mode_set_nofb(struct drm_crtc *crtc)
+{
+	struct vc4_crtc *vc4_crtc = to_vc4_crtc(crtc);
+	struct drm_crtc_state *state = crtc->state;
+	struct drm_display_mode *mode = &state->adjusted_mode;
+	bool interlace = mode->flags & DRM_MODE_FLAG_INTERLACE;
+	u32 vactive = (mode->vdisplay >> (interlace ? 1 : 0));
+	u32 format = PV_CONTROL_FORMAT_24;
+	bool debug_dump_regs = false;
+	int clock_select = vc4_get_clock_select(crtc);
+
+	if (debug_dump_regs) {
+		DRM_INFO("CRTC %d regs before:\n", drm_crtc_index(crtc));
+		vc4_crtc_dump_regs(vc4_crtc);
+	}
+
+	/* Reset the PV fifo. */
+	CRTC_WRITE(PV_CONTROL, 0);
+	CRTC_WRITE(PV_CONTROL, PV_CONTROL_FIFO_CLR | PV_CONTROL_EN);
+	CRTC_WRITE(PV_CONTROL, 0);
+
+	CRTC_WRITE(PV_HORZA,
+		   VC4_SET_FIELD(mode->htotal - mode->hsync_end,
+				 PV_HORZA_HBP) |
+		   VC4_SET_FIELD(mode->hsync_end - mode->hsync_start,
+				 PV_HORZA_HSYNC));
+	CRTC_WRITE(PV_HORZB,
+		   VC4_SET_FIELD(mode->hsync_start - mode->hdisplay,
+				 PV_HORZB_HFP) |
+		   VC4_SET_FIELD(mode->hdisplay, PV_HORZB_HACTIVE));
+
+	if (interlace) {
+		CRTC_WRITE(PV_VERTA_EVEN,
+			   VC4_SET_FIELD(mode->vtotal - mode->vsync_end - 1,
+					 PV_VERTA_VBP) |
+			   VC4_SET_FIELD(mode->vsync_end - mode->vsync_start,
+					 PV_VERTA_VSYNC));
+		CRTC_WRITE(PV_VERTB_EVEN,
+			   VC4_SET_FIELD(mode->vsync_start - mode->vdisplay,
+					 PV_VERTB_VFP) |
+			   VC4_SET_FIELD(vactive, PV_VERTB_VACTIVE));
+	}
+
+	CRTC_WRITE(PV_HACT_ACT, mode->hdisplay);
+
+	CRTC_WRITE(PV_V_CONTROL,
+		   PV_VCONTROL_CONTINUOUS |
+		   (interlace ? PV_VCONTROL_INTERLACE : 0));
+
+	CRTC_WRITE(PV_CONTROL,
+		   VC4_SET_FIELD(format, PV_CONTROL_FORMAT) |
+		   VC4_SET_FIELD(vc4_get_fifo_full_level(format),
+				 PV_CONTROL_FIFO_LEVEL) |
+		   PV_CONTROL_CLR_AT_START |
+		   PV_CONTROL_TRIGGER_UNDERFLOW |
+		   PV_CONTROL_WAIT_HSTART |
+		   VC4_SET_FIELD(clock_select, PV_CONTROL_CLK_SELECT) |
+		   PV_CONTROL_FIFO_CLR |
+		   PV_CONTROL_EN);
+
+	if (debug_dump_regs) {
+		DRM_INFO("CRTC %d regs after:\n", drm_crtc_index(crtc));
+		vc4_crtc_dump_regs(vc4_crtc);
+	}
+}
+
+static void require_hvs_enabled(struct drm_device *dev)
+{
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+
+	WARN_ON_ONCE((HVS_READ(SCALER_DISPCTRL) & SCALER_DISPCTRL_ENABLE) !=
+		     SCALER_DISPCTRL_ENABLE);
+}
+
+static void vc4_crtc_disable(struct drm_crtc *crtc)
+{
+	struct drm_device *dev = crtc->dev;
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+	struct vc4_crtc *vc4_crtc = to_vc4_crtc(crtc);
+	u32 chan = vc4_crtc->channel;
+	int ret;
+	require_hvs_enabled(dev);
+
+	CRTC_WRITE(PV_V_CONTROL,
+		   CRTC_READ(PV_V_CONTROL) & ~PV_VCONTROL_VIDEN);
+	ret = wait_for(!(CRTC_READ(PV_V_CONTROL) & PV_VCONTROL_VIDEN), 1);
+	WARN_ONCE(ret, "Timeout waiting for !PV_VCONTROL_VIDEN\n");
+
+	if (HVS_READ(SCALER_DISPCTRLX(chan)) &
+	    SCALER_DISPCTRLX_ENABLE) {
+		HVS_WRITE(SCALER_DISPCTRLX(chan),
+			  SCALER_DISPCTRLX_RESET);
+
+		/* While the docs say that reset is self-clearing, it
+		 * seems it doesn't actually.
+		 */
+		HVS_WRITE(SCALER_DISPCTRLX(chan), 0);
+	}
+
+	/* Once we leave, the scaler should be disabled and its fifo empty. */
+
+	WARN_ON_ONCE(HVS_READ(SCALER_DISPCTRLX(chan)) & SCALER_DISPCTRLX_RESET);
+
+	WARN_ON_ONCE(VC4_GET_FIELD(HVS_READ(SCALER_DISPSTATX(chan)),
+				   SCALER_DISPSTATX_MODE) !=
+		     SCALER_DISPSTATX_MODE_DISABLED);
+
+	WARN_ON_ONCE((HVS_READ(SCALER_DISPSTATX(chan)) &
+		      (SCALER_DISPSTATX_FULL | SCALER_DISPSTATX_EMPTY)) !=
+		     SCALER_DISPSTATX_EMPTY);
+}
+
+static void vc4_crtc_enable(struct drm_crtc *crtc)
+{
+	struct drm_device *dev = crtc->dev;
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+	struct vc4_crtc *vc4_crtc = to_vc4_crtc(crtc);
+	struct drm_crtc_state *state = crtc->state;
+	struct drm_display_mode *mode = &state->adjusted_mode;
+
+	require_hvs_enabled(dev);
+
+	/* Turn on the scaler, which will wait for vstart to start
+	 * compositing.
+	 */
+	HVS_WRITE(SCALER_DISPCTRLX(vc4_crtc->channel),
+		  VC4_SET_FIELD(mode->hdisplay, SCALER_DISPCTRLX_WIDTH) |
+		  VC4_SET_FIELD(mode->vdisplay, SCALER_DISPCTRLX_HEIGHT) |
+		  SCALER_DISPCTRLX_ENABLE);
+
+	/* Turn on the pixel valve, which will emit the vstart signal. */
+	CRTC_WRITE(PV_V_CONTROL,
+		   CRTC_READ(PV_V_CONTROL) | PV_VCONTROL_VIDEN);
+}
+
+static int vc4_crtc_atomic_check(struct drm_crtc *crtc,
+				 struct drm_crtc_state *state)
+{
+	struct drm_device *dev = crtc->dev;
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+	struct drm_plane *plane;
+	struct vc4_crtc *vc4_crtc = to_vc4_crtc(crtc);
+	u32 dlist_count = 0;
+
+	/* The pixelvalve can only feed one encoder (and encoders are
+	 * 1:1 with connectors.)
+	 */
+	if (drm_atomic_connectors_for_crtc(state->state, crtc) > 1)
+		return -EINVAL;
+
+	drm_atomic_crtc_state_for_each_plane(plane, state) {
+		struct drm_plane_state *plane_state =
+			state->state->plane_states[drm_plane_index(plane)];
+
+		/* plane might not have changed, in which case take
+		 * current state:
+		 */
+		if (!plane_state)
+			plane_state = plane->state;
+
+		dlist_count += vc4_plane_dlist_size(plane_state);
+	}
+
+	dlist_count++; /* Account for SCALER_CTL0_END. */
+
+	if (!vc4_crtc->dlist || dlist_count > vc4_crtc->dlist_size) {
+		vc4_crtc->dlist = ((u32 __iomem *)vc4->hvs->dlist +
+				   HVS_BOOTLOADER_DLIST_END);
+		vc4_crtc->dlist_size = ((SCALER_DLIST_SIZE >> 2) -
+					HVS_BOOTLOADER_DLIST_END);
+
+		if (dlist_count > vc4_crtc->dlist_size) {
+			DRM_DEBUG_KMS("dlist too large for CRTC (%d > %d).\n",
+				      dlist_count, vc4_crtc->dlist_size);
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+
+static void vc4_crtc_atomic_flush(struct drm_crtc *crtc,
+				  struct drm_crtc_state *old_state)
+{
+	struct drm_device *dev = crtc->dev;
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+	struct vc4_crtc *vc4_crtc = to_vc4_crtc(crtc);
+	struct drm_plane *plane;
+	bool debug_dump_regs = false;
+	u32 __iomem *dlist_next = vc4_crtc->dlist;
+
+	if (debug_dump_regs) {
+		DRM_INFO("CRTC %d HVS before:\n", drm_crtc_index(crtc));
+		vc4_hvs_dump_state(dev);
+	}
+
+	/* Copy all the active planes' dlist contents to the hardware dlist.
+	 *
+	 * XXX: If the new display list was large enough that it
+	 * overlapped a currently-read display list, we need to do
+	 * something like disable scanout before putting in the new
+	 * list.  For now, we're safe because we only have the two
+	 * planes.
+	 */
+	drm_atomic_crtc_for_each_plane(plane, crtc) {
+		dlist_next += vc4_plane_write_dlist(plane, dlist_next);
+	}
+
+	if (dlist_next == vc4_crtc->dlist) {
+		/* If no planes were enabled, use the SCALER_CTL0_END
+		 * at the start of the display list memory (in the
+		 * bootloader section).  We'll rewrite that
+		 * SCALER_CTL0_END, just in case, though.
+		 */
+		writel(SCALER_CTL0_END, vc4->hvs->dlist);
+		HVS_WRITE(SCALER_DISPLISTX(vc4_crtc->channel), 0);
+	} else {
+		writel(SCALER_CTL0_END, dlist_next);
+		dlist_next++;
+
+		HVS_WRITE(SCALER_DISPLISTX(vc4_crtc->channel),
+			  (u32 *)vc4_crtc->dlist - (u32 *)vc4->hvs->dlist);
+
+		/* Make the next display list start after ours. */
+		vc4_crtc->dlist_size -= (dlist_next - vc4_crtc->dlist);
+		vc4_crtc->dlist = dlist_next;
+	}
+
+	if (debug_dump_regs) {
+		DRM_INFO("CRTC %d HVS after:\n", drm_crtc_index(crtc));
+		vc4_hvs_dump_state(dev);
+	}
+
+	if (crtc->state->event) {
+		unsigned long flags;
+
+		crtc->state->event->pipe = drm_crtc_index(crtc);
+
+		WARN_ON(drm_crtc_vblank_get(crtc) != 0);
+
+		spin_lock_irqsave(&dev->event_lock, flags);
+		vc4_crtc->event = crtc->state->event;
+		spin_unlock_irqrestore(&dev->event_lock, flags);
+		crtc->state->event = NULL;
+	}
+}
+
+int vc4_enable_vblank(struct drm_device *dev, int crtc_id)
+{
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+	struct vc4_crtc *vc4_crtc = vc4->crtc[crtc_id];
+
+	CRTC_WRITE(PV_INTEN, PV_INT_VFP_START);
+
+	return 0;
+}
+
+void vc4_disable_vblank(struct drm_device *dev, int crtc_id)
+{
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+	struct vc4_crtc *vc4_crtc = vc4->crtc[crtc_id];
+
+	CRTC_WRITE(PV_INTEN, 0);
+}
+
+static void vc4_crtc_handle_page_flip(struct vc4_crtc *vc4_crtc)
+{
+	struct drm_crtc *crtc = &vc4_crtc->base;
+	struct drm_device *dev = crtc->dev;
+	unsigned long flags;
+
+	spin_lock_irqsave(&dev->event_lock, flags);
+	if (vc4_crtc->event) {
+		drm_crtc_send_vblank_event(crtc, vc4_crtc->event);
+		vc4_crtc->event = NULL;
+	}
+	spin_unlock_irqrestore(&dev->event_lock, flags);
+}
+
+static irqreturn_t vc4_crtc_irq_handler(int irq, void *data)
+{
+	struct vc4_crtc *vc4_crtc = data;
+	u32 stat = CRTC_READ(PV_INTSTAT);
+	irqreturn_t ret = IRQ_NONE;
+
+	if (stat & PV_INT_VFP_START) {
+		CRTC_WRITE(PV_INTSTAT, PV_INT_VFP_START);
+		drm_crtc_handle_vblank(&vc4_crtc->base);
+		vc4_crtc_handle_page_flip(vc4_crtc);
+		ret = IRQ_HANDLED;
+	}
+
+	return ret;
+}
+
+static const struct drm_crtc_funcs vc4_crtc_funcs = {
+	.set_config = drm_atomic_helper_set_config,
+	.destroy = vc4_crtc_destroy,
+	.page_flip = drm_atomic_helper_page_flip,
+	.set_property = NULL,
+	.cursor_set = NULL, /* handled by drm_mode_cursor_universal */
+	.cursor_move = NULL, /* handled by drm_mode_cursor_universal */
+	.reset = drm_atomic_helper_crtc_reset,
+	.atomic_duplicate_state = drm_atomic_helper_crtc_duplicate_state,
+	.atomic_destroy_state = drm_atomic_helper_crtc_destroy_state,
+};
+
+static const struct drm_crtc_helper_funcs vc4_crtc_helper_funcs = {
+	.mode_set_nofb = vc4_crtc_mode_set_nofb,
+	.disable = vc4_crtc_disable,
+	.enable = vc4_crtc_enable,
+	.atomic_check = vc4_crtc_atomic_check,
+	.atomic_flush = vc4_crtc_atomic_flush,
+};
+
+/* Frees the page flip event when the DRM device is closed with the
+ * event still outstanding.
+ */
+void vc4_cancel_page_flip(struct drm_crtc *crtc, struct drm_file *file)
+{
+	struct vc4_crtc *vc4_crtc = to_vc4_crtc(crtc);
+	struct drm_device *dev = crtc->dev;
+	unsigned long flags;
+
+	spin_lock_irqsave(&dev->event_lock, flags);
+
+	if (vc4_crtc->event && vc4_crtc->event->base.file_priv == file) {
+		vc4_crtc->event->base.destroy(&vc4_crtc->event->base);
+		drm_crtc_vblank_put(crtc);
+		vc4_crtc->event = NULL;
+	}
+
+	spin_unlock_irqrestore(&dev->event_lock, flags);
+}
+
+static const struct vc4_crtc_data pv0_data = {
+	.hvs_channel = 0,
+	.encoder0_type = VC4_ENCODER_TYPE_DSI0,
+	.encoder1_type = VC4_ENCODER_TYPE_DPI,
+};
+
+static const struct vc4_crtc_data pv1_data = {
+	.hvs_channel = 2,
+	.encoder0_type = VC4_ENCODER_TYPE_DSI1,
+	.encoder1_type = VC4_ENCODER_TYPE_SMI,
+};
+
+static const struct vc4_crtc_data pv2_data = {
+	.hvs_channel = 1,
+	.encoder0_type = VC4_ENCODER_TYPE_VEC,
+	.encoder1_type = VC4_ENCODER_TYPE_HDMI,
+};
+
+static const struct of_device_id vc4_crtc_dt_match[] = {
+	{ .compatible = "brcm,bcm2835-pixelvalve0", .data = &pv0_data },
+	{ .compatible = "brcm,bcm2835-pixelvalve1", .data = &pv1_data },
+	{ .compatible = "brcm,bcm2835-pixelvalve2", .data = &pv2_data },
+	{}
+};
+
+static void vc4_set_crtc_possible_masks(struct drm_device *drm,
+					struct drm_crtc *crtc)
+{
+	struct vc4_crtc *vc4_crtc = to_vc4_crtc(crtc);
+	struct drm_encoder *encoder;
+
+	drm_for_each_encoder(encoder, drm) {
+		struct vc4_encoder *vc4_encoder = to_vc4_encoder(encoder);
+
+		if (vc4_encoder->type == vc4_crtc->data->encoder0_type) {
+			vc4_encoder->clock_select = 0;
+			encoder->possible_crtcs |= drm_crtc_mask(crtc);
+		} else if (vc4_encoder->type == vc4_crtc->data->encoder1_type) {
+			vc4_encoder->clock_select = 1;
+			encoder->possible_crtcs |= drm_crtc_mask(crtc);
+		}
+	}
+}
+
+static int vc4_crtc_bind(struct device *dev, struct device *master, void *data)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct drm_device *drm = dev_get_drvdata(master);
+	struct vc4_dev *vc4 = to_vc4_dev(drm);
+	struct vc4_crtc *vc4_crtc;
+	struct drm_crtc *crtc;
+	struct drm_plane *primary_plane, *cursor_plane;
+	const struct of_device_id *match;
+	int ret;
+
+	vc4_crtc = devm_kzalloc(dev, sizeof(*vc4_crtc), GFP_KERNEL);
+	if (!vc4_crtc)
+		return -ENOMEM;
+	crtc = &vc4_crtc->base;
+
+	match = of_match_device(vc4_crtc_dt_match, dev);
+	if (!match)
+		return -ENODEV;
+	vc4_crtc->data = match->data;
+
+	vc4_crtc->regs = vc4_ioremap_regs(pdev, 0);
+	if (IS_ERR(vc4_crtc->regs))
+		return PTR_ERR(vc4_crtc->regs);
+
+	/* For now, we create just the primary and the legacy cursor
+	 * planes.  We should be able to stack more planes on easily,
+	 * but to do that we would need to compute the bandwidth
+	 * requirement of the plane configuration, and reject ones
+	 * that will take too much.
+	 */
+	primary_plane = vc4_plane_init(drm, DRM_PLANE_TYPE_PRIMARY);
+	if (!primary_plane) {
+		dev_err(dev, "failed to construct primary plane\n");
+		ret = PTR_ERR(primary_plane);
+		goto err;
+	}
+
+	cursor_plane = vc4_plane_init(drm, DRM_PLANE_TYPE_CURSOR);
+	if (!cursor_plane) {
+		dev_err(dev, "failed to construct cursor plane\n");
+		ret = PTR_ERR(cursor_plane);
+		goto err_primary;
+	}
+
+	drm_crtc_init_with_planes(drm, crtc, primary_plane, cursor_plane,
+				  &vc4_crtc_funcs);
+	drm_crtc_helper_add(crtc, &vc4_crtc_helper_funcs);
+	primary_plane->crtc = crtc;
+	cursor_plane->crtc = crtc;
+	vc4->crtc[drm_crtc_index(crtc)] = vc4_crtc;
+	vc4_crtc->channel = vc4_crtc->data->hvs_channel;
+
+	CRTC_WRITE(PV_INTEN, 0);
+	CRTC_WRITE(PV_INTSTAT, PV_INT_VFP_START);
+	ret = devm_request_irq(dev, platform_get_irq(pdev, 0),
+			       vc4_crtc_irq_handler, 0, "vc4 crtc", vc4_crtc);
+	if (ret)
+		goto err_cursor;
+
+	vc4_set_crtc_possible_masks(drm, crtc);
+
+	platform_set_drvdata(pdev, vc4_crtc);
+
+	return 0;
+
+err_cursor:
+	cursor_plane->funcs->destroy(cursor_plane);
+err_primary:
+	primary_plane->funcs->destroy(primary_plane);
+err:
+	return ret;
+}
+
+static void vc4_crtc_unbind(struct device *dev, struct device *master,
+			    void *data)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct vc4_crtc *vc4_crtc = dev_get_drvdata(dev);
+
+	vc4_crtc_destroy(&vc4_crtc->base);
+
+	CRTC_WRITE(PV_INTEN, 0);
+
+	platform_set_drvdata(pdev, NULL);
+}
+
+static const struct component_ops vc4_crtc_ops = {
+	.bind   = vc4_crtc_bind,
+	.unbind = vc4_crtc_unbind,
+};
+
+static int vc4_crtc_dev_probe(struct platform_device *pdev)
+{
+	return component_add(&pdev->dev, &vc4_crtc_ops);
+}
+
+static int vc4_crtc_dev_remove(struct platform_device *pdev)
+{
+	component_del(&pdev->dev, &vc4_crtc_ops);
+	return 0;
+}
+
+struct platform_driver vc4_crtc_driver = {
+	.probe = vc4_crtc_dev_probe,
+	.remove = vc4_crtc_dev_remove,
+	.driver = {
+		.name = "vc4_crtc",
+		.of_match_table = vc4_crtc_dt_match,
+	},
+};
diff --git a/drivers/gpu/drm/vc4/vc4_debugfs.c b/drivers/gpu/drm/vc4/vc4_debugfs.c
new file mode 100644
index 0000000..4297b0a5
--- /dev/null
+++ b/drivers/gpu/drm/vc4/vc4_debugfs.c
@@ -0,0 +1,39 @@
+/*
+ *  Copyright © 2014 Broadcom
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/seq_file.h>
+#include <linux/circ_buf.h>
+#include <linux/ctype.h>
+#include <linux/debugfs.h>
+#include <drm/drmP.h>
+
+#include "vc4_drv.h"
+#include "vc4_regs.h"
+
+static const struct drm_info_list vc4_debugfs_list[] = {
+	{"hdmi_regs", vc4_hdmi_debugfs_regs, 0},
+	{"hvs_regs", vc4_hvs_debugfs_regs, 0},
+	{"crtc0_regs", vc4_crtc_debugfs_regs, 0, (void *)(uintptr_t)0},
+	{"crtc1_regs", vc4_crtc_debugfs_regs, 0, (void *)(uintptr_t)1},
+	{"crtc2_regs", vc4_crtc_debugfs_regs, 0, (void *)(uintptr_t)2},
+};
+
+#define VC4_DEBUGFS_ENTRIES ARRAY_SIZE(vc4_debugfs_list)
+
+int
+vc4_debugfs_init(struct drm_minor *minor)
+{
+	return drm_debugfs_create_files(vc4_debugfs_list, VC4_DEBUGFS_ENTRIES,
+					minor->debugfs_root, minor);
+}
+
+void
+vc4_debugfs_cleanup(struct drm_minor *minor)
+{
+	drm_debugfs_remove_files(vc4_debugfs_list, VC4_DEBUGFS_ENTRIES, minor);
+}
diff --git a/drivers/gpu/drm/vc4/vc4_drv.c b/drivers/gpu/drm/vc4/vc4_drv.c
new file mode 100644
index 0000000..3c3ccff
--- /dev/null
+++ b/drivers/gpu/drm/vc4/vc4_drv.c
@@ -0,0 +1,284 @@
+/*
+ * Copyright (C) 2014-2015 Broadcom
+ * Copyright (C) 2013 Red Hat
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/clk.h>
+#include <linux/component.h>
+#include <linux/device.h>
+#include <linux/io.h>
+#include <linux/module.h>
+#include <linux/of_platform.h>
+#include <linux/platform_device.h>
+
+#include "vc4_drv.h"
+#include "vc4_regs.h"
+
+#define DRIVER_NAME "vc4"
+#define DRIVER_DESC "Broadcom VC4 graphics"
+#define DRIVER_DATE "20140616"
+#define DRIVER_MAJOR 0
+#define DRIVER_MINOR 0
+#define DRIVER_PATCHLEVEL 0
+
+/* Helper function for mapping the regs on a platform device. */
+void __iomem *vc4_ioremap_regs(struct platform_device *dev, int index)
+{
+	struct resource *res;
+	void __iomem *map;
+
+	res = platform_get_resource(dev, IORESOURCE_MEM, index);
+	map = devm_ioremap_resource(&dev->dev, res);
+	if (IS_ERR(map)) {
+		DRM_ERROR("Failed to map registers: %ld\n", PTR_ERR(map));
+		return map;
+	}
+
+	return map;
+}
+
+static void vc4_drm_preclose(struct drm_device *dev, struct drm_file *file)
+{
+	struct drm_crtc *crtc;
+
+	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head)
+		vc4_cancel_page_flip(crtc, file);
+}
+
+static const struct file_operations vc4_drm_fops = {
+	.owner = THIS_MODULE,
+	.open = drm_open,
+	.release = drm_release,
+	.unlocked_ioctl = drm_ioctl,
+	.mmap = drm_gem_cma_mmap,
+	.poll = drm_poll,
+	.read = drm_read,
+#ifdef CONFIG_COMPAT
+	.compat_ioctl = drm_compat_ioctl,
+#endif
+	.llseek = noop_llseek,
+};
+
+static const struct drm_ioctl_desc vc4_drm_ioctls[] = {
+};
+
+static struct drm_driver vc4_drm_driver = {
+	.driver_features = (DRIVER_MODESET |
+			    DRIVER_ATOMIC |
+			    DRIVER_GEM |
+			    DRIVER_PRIME),
+	.preclose = vc4_drm_preclose,
+
+	.enable_vblank = vc4_enable_vblank,
+	.disable_vblank = vc4_disable_vblank,
+	.get_vblank_counter = drm_vblank_count,
+
+#if defined(CONFIG_DEBUG_FS)
+	.debugfs_init = vc4_debugfs_init,
+	.debugfs_cleanup = vc4_debugfs_cleanup,
+#endif
+
+	.gem_free_object = drm_gem_cma_free_object,
+	.gem_vm_ops = &drm_gem_cma_vm_ops,
+
+	.prime_handle_to_fd = drm_gem_prime_handle_to_fd,
+	.prime_fd_to_handle = drm_gem_prime_fd_to_handle,
+	.gem_prime_import = drm_gem_prime_import,
+	.gem_prime_export = drm_gem_prime_export,
+	.gem_prime_get_sg_table	= drm_gem_cma_prime_get_sg_table,
+	.gem_prime_import_sg_table = drm_gem_cma_prime_import_sg_table,
+	.gem_prime_vmap = drm_gem_cma_prime_vmap,
+	.gem_prime_vunmap = drm_gem_cma_prime_vunmap,
+	.gem_prime_mmap = drm_gem_cma_prime_mmap,
+
+	.dumb_create = vc4_dumb_create,
+	.dumb_map_offset = drm_gem_cma_dumb_map_offset,
+	.dumb_destroy = drm_gem_dumb_destroy,
+
+	.ioctls = vc4_drm_ioctls,
+	.num_ioctls = ARRAY_SIZE(vc4_drm_ioctls),
+	.fops = &vc4_drm_fops,
+
+	.name = DRIVER_NAME,
+	.desc = DRIVER_DESC,
+	.date = DRIVER_DATE,
+	.major = DRIVER_MAJOR,
+	.minor = DRIVER_MINOR,
+	.patchlevel = DRIVER_PATCHLEVEL,
+};
+
+static int compare_dev(struct device *dev, void *data)
+{
+	return dev == data;
+}
+
+static void vc4_match_add_drivers(struct device *dev,
+				  struct component_match **match,
+				  struct platform_driver *const *drivers,
+				  int count)
+{
+	int i;
+
+	for (i = 0; i < count; i++) {
+		struct device_driver *drv = &drivers[i]->driver;
+		struct device *p = NULL, *d;
+
+		while ((d = bus_find_device(&platform_bus_type, p, drv,
+					    (void *)platform_bus_type.match))) {
+			put_device(p);
+			component_match_add(dev, match, compare_dev, d);
+			p = d;
+		}
+		put_device(p);
+	}
+}
+
+static int vc4_drm_bind(struct device *dev)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct drm_device *drm;
+	struct drm_connector *connector;
+	struct vc4_dev *vc4;
+	int ret = 0;
+
+	dev->coherent_dma_mask = DMA_BIT_MASK(32);
+
+	vc4 = devm_kzalloc(dev, sizeof(*vc4), GFP_KERNEL);
+	if (!vc4)
+		return -ENOMEM;
+
+	drm = drm_dev_alloc(&vc4_drm_driver, dev);
+	if (!drm)
+		return -ENOMEM;
+	platform_set_drvdata(pdev, drm);
+	vc4->dev = drm;
+	drm->dev_private = vc4;
+
+	drm_dev_set_unique(drm, dev_name(dev));
+
+	drm_mode_config_init(drm);
+	if (ret)
+		goto unref;
+
+	ret = component_bind_all(dev, drm);
+	if (ret)
+		goto unref;
+
+	ret = drm_dev_register(drm, 0);
+	if (ret < 0)
+		goto unbind_all;
+
+	/* Connector registration has to occur after DRM device
+	 * registration, because it creates sysfs entries based on the
+	 * DRM device.
+	 */
+	list_for_each_entry(connector, &drm->mode_config.connector_list, head) {
+		ret = drm_connector_register(connector);
+		if (ret)
+			goto unregister;
+	}
+
+	vc4_kms_load(drm);
+
+	return 0;
+
+unregister:
+	drm_dev_unregister(drm);
+unbind_all:
+	component_unbind_all(dev, drm);
+unref:
+	drm_dev_unref(drm);
+	return ret;
+}
+
+static void vc4_drm_unbind(struct device *dev)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct drm_device *drm = platform_get_drvdata(pdev);
+
+	drm_mode_config_cleanup(drm);
+
+	drm_put_dev(drm);
+}
+
+static const struct component_master_ops vc4_drm_ops = {
+	.bind = vc4_drm_bind,
+	.unbind = vc4_drm_unbind,
+};
+
+static struct platform_driver *const component_drivers[] = {
+	&vc4_hdmi_driver,
+	&vc4_crtc_driver,
+	&vc4_hvs_driver,
+};
+
+static int vc4_platform_drm_probe(struct platform_device *pdev)
+{
+	struct component_match *match = NULL;
+	struct device *dev = &pdev->dev;
+
+	vc4_match_add_drivers(dev, &match,
+			      component_drivers, ARRAY_SIZE(component_drivers));
+
+	return component_master_add_with_match(dev, &vc4_drm_ops, match);
+}
+
+static int vc4_platform_drm_remove(struct platform_device *pdev)
+{
+	component_master_del(&pdev->dev, &vc4_drm_ops);
+
+	return 0;
+}
+
+static const struct of_device_id vc4_of_match[] = {
+	{ .compatible = "brcm,bcm2835-vc4", },
+	{},
+};
+MODULE_DEVICE_TABLE(of, vc4_of_match);
+
+static struct platform_driver vc4_platform_driver = {
+	.probe		= vc4_platform_drm_probe,
+	.remove		= vc4_platform_drm_remove,
+	.driver		= {
+		.name	= "vc4-drm",
+		.owner	= THIS_MODULE,
+		.of_match_table = vc4_of_match,
+	},
+};
+
+static int __init vc4_drm_register(void)
+{
+	int i, ret;
+
+	for (i = 0; i < ARRAY_SIZE(component_drivers); i++) {
+		ret = platform_driver_register(component_drivers[i]);
+		if (ret) {
+			while (--i >= 0)
+				platform_driver_unregister(component_drivers[i]);
+			return ret;
+		}
+	}
+	return platform_driver_register(&vc4_platform_driver);
+}
+
+static void __exit vc4_drm_unregister(void)
+{
+	int i;
+
+	for (i = ARRAY_SIZE(component_drivers) - 1; i >= 0; i--)
+		platform_driver_unregister(component_drivers[i]);
+
+	platform_driver_unregister(&vc4_platform_driver);
+}
+
+module_init(vc4_drm_register);
+module_exit(vc4_drm_unregister);
+
+MODULE_ALIAS("platform:vc4-drm");
+MODULE_DESCRIPTION("Broadcom VC4 DRM Driver");
+MODULE_AUTHOR("Eric Anholt <eric@anholt.net>");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/gpu/drm/vc4/vc4_drv.h b/drivers/gpu/drm/vc4/vc4_drv.h
new file mode 100644
index 0000000..04dd2e2
--- /dev/null
+++ b/drivers/gpu/drm/vc4/vc4_drv.h
@@ -0,0 +1,143 @@
+/*
+ * Copyright (C) 2015 Broadcom
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include "drmP.h"
+#include "drm_gem_cma_helper.h"
+
+struct vc4_dev {
+	struct drm_device *dev;
+
+	struct vc4_hdmi *hdmi;
+	struct vc4_hvs *hvs;
+	struct vc4_crtc *crtc[3];
+};
+
+static inline struct vc4_dev *
+to_vc4_dev(struct drm_device *dev)
+{
+	return (struct vc4_dev *)dev->dev_private;
+}
+
+struct vc4_bo {
+	struct drm_gem_cma_object base;
+};
+
+static inline struct vc4_bo *
+to_vc4_bo(struct drm_gem_object *bo)
+{
+	return (struct vc4_bo *)bo;
+}
+
+struct vc4_hvs {
+	struct platform_device *pdev;
+	void __iomem *regs;
+	void __iomem *dlist;
+};
+
+struct vc4_plane {
+	struct drm_plane base;
+};
+
+static inline struct vc4_plane *
+to_vc4_plane(struct drm_plane *plane)
+{
+	return (struct vc4_plane *)plane;
+}
+
+enum vc4_encoder_type {
+	VC4_ENCODER_TYPE_HDMI,
+	VC4_ENCODER_TYPE_VEC,
+	VC4_ENCODER_TYPE_DSI0,
+	VC4_ENCODER_TYPE_DSI1,
+	VC4_ENCODER_TYPE_SMI,
+	VC4_ENCODER_TYPE_DPI,
+};
+
+struct vc4_encoder {
+	struct drm_encoder base;
+	enum vc4_encoder_type type;
+	u32 clock_select;
+};
+
+static inline struct vc4_encoder *
+to_vc4_encoder(struct drm_encoder *encoder)
+{
+	return container_of(encoder, struct vc4_encoder, base);
+}
+
+#define HVS_READ(offset) readl(vc4->hvs->regs + offset)
+#define HVS_WRITE(offset, val) writel(val, vc4->hvs->regs + offset)
+
+/**
+ * _wait_for - magic (register) wait macro
+ *
+ * Does the right thing for modeset paths when run under kdgb or similar atomic
+ * contexts. Note that it's important that we check the condition again after
+ * having timed out, since the timeout could be due to preemption or similar and
+ * we've never had a chance to check the condition before the timeout.
+ */
+#define _wait_for(COND, MS, W) ({ \
+	unsigned long timeout__ = jiffies + msecs_to_jiffies(MS) + 1;	\
+	int ret__ = 0;							\
+	while (!(COND)) {						\
+		if (time_after(jiffies, timeout__)) {			\
+			if (!(COND))					\
+				ret__ = -ETIMEDOUT;			\
+			break;						\
+		}							\
+		if (W && drm_can_sleep())  {				\
+			msleep(W);					\
+		} else {						\
+			cpu_relax();					\
+		}							\
+	}								\
+	ret__;								\
+})
+
+#define wait_for(COND, MS) _wait_for(COND, MS, 1)
+
+/* vc4_bo.c */
+void vc4_free_object(struct drm_gem_object *gem_obj);
+struct vc4_bo *vc4_bo_create(struct drm_device *dev, size_t size);
+int vc4_dumb_create(struct drm_file *file_priv,
+		    struct drm_device *dev,
+		    struct drm_mode_create_dumb *args);
+struct dma_buf *vc4_prime_export(struct drm_device *dev,
+				 struct drm_gem_object *obj, int flags);
+
+/* vc4_crtc.c */
+extern struct platform_driver vc4_crtc_driver;
+int vc4_enable_vblank(struct drm_device *dev, int crtc_id);
+void vc4_disable_vblank(struct drm_device *dev, int crtc_id);
+void vc4_cancel_page_flip(struct drm_crtc *crtc, struct drm_file *file);
+int vc4_crtc_debugfs_regs(struct seq_file *m, void *arg);
+
+/* vc4_debugfs.c */
+int vc4_debugfs_init(struct drm_minor *minor);
+void vc4_debugfs_cleanup(struct drm_minor *minor);
+
+/* vc4_drv.c */
+void __iomem *vc4_ioremap_regs(struct platform_device *dev, int index);
+
+/* vc4_hdmi.c */
+extern struct platform_driver vc4_hdmi_driver;
+int vc4_hdmi_debugfs_regs(struct seq_file *m, void *unused);
+
+/* vc4_hvs.c */
+extern struct platform_driver vc4_hvs_driver;
+void vc4_hvs_dump_state(struct drm_device *dev);
+int vc4_hvs_debugfs_regs(struct seq_file *m, void *unused);
+
+/* vc4_kms.c */
+int vc4_kms_load(struct drm_device *dev);
+
+/* vc4_plane.c */
+struct drm_plane *vc4_plane_init(struct drm_device *dev,
+				 enum drm_plane_type type);
+u32 vc4_plane_write_dlist(struct drm_plane *plane, u32 __iomem *dlist);
+u32 vc4_plane_dlist_size(struct drm_plane_state *state);
diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c
new file mode 100644
index 0000000..da9a36d
--- /dev/null
+++ b/drivers/gpu/drm/vc4/vc4_hdmi.c
@@ -0,0 +1,590 @@
+/*
+ * Copyright (C) 2015 Broadcom
+ * Copyright (c) 2014 The Linux Foundation. All rights reserved.
+ * Copyright (C) 2013 Red Hat
+ * Author: Rob Clark <robdclark@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/**
+ * DOC: VC4 Falcon HDMI module
+ *
+ * The HDMI core has a state machine and a PHY.  Most of the unit
+ * operates off of the HSM clock from CPRMAN.  It also internally uses
+ * the PLLH_PIX clock for the PHY.
+ */
+
+#include "drm_atomic_helper.h"
+#include "drm_crtc_helper.h"
+#include "drm_edid.h"
+#include "linux/clk.h"
+#include "linux/component.h"
+#include "linux/i2c.h"
+#include "linux/of_gpio.h"
+#include "linux/of_platform.h"
+#include "vc4_drv.h"
+#include "vc4_regs.h"
+
+/* General HDMI hardware state. */
+struct vc4_hdmi {
+	struct platform_device *pdev;
+
+	struct drm_encoder *encoder;
+	struct drm_connector *connector;
+
+	struct i2c_adapter *ddc;
+	void __iomem *hdmicore_regs;
+	void __iomem *hd_regs;
+	int hpd_gpio;
+
+	struct clk *pixel_clock;
+	struct clk *hsm_clock;
+};
+
+#define HDMI_READ(offset) readl(vc4->hdmi->hdmicore_regs + offset)
+#define HDMI_WRITE(offset, val) writel(val, vc4->hdmi->hdmicore_regs + offset)
+#define HD_READ(offset) readl(vc4->hdmi->hd_regs + offset)
+#define HD_WRITE(offset, val) writel(val, vc4->hdmi->hd_regs + offset)
+
+/* VC4 HDMI encoder KMS struct */
+struct vc4_hdmi_encoder {
+	struct vc4_encoder base;
+	bool hdmi_monitor;
+};
+
+static inline struct vc4_hdmi_encoder *
+to_vc4_hdmi_encoder(struct drm_encoder *encoder)
+{
+	return container_of(encoder, struct vc4_hdmi_encoder, base.base);
+}
+
+/* VC4 HDMI connector KMS struct */
+struct vc4_hdmi_connector {
+	struct drm_connector base;
+
+	/* Since the connector is attached to just the one encoder,
+	 * this is the reference to it so we can do the best_encoder()
+	 * hook.
+	 */
+	struct drm_encoder *encoder;
+};
+
+static inline struct vc4_hdmi_connector *
+to_vc4_hdmi_connector(struct drm_connector *connector)
+{
+	return container_of(connector, struct vc4_hdmi_connector, base);
+}
+
+#define HDMI_REG(reg) { reg, #reg }
+static const struct {
+	u32 reg;
+	const char *name;
+} hdmi_regs[] = {
+	HDMI_REG(VC4_HDMI_CORE_REV),
+	HDMI_REG(VC4_HDMI_SW_RESET_CONTROL),
+	HDMI_REG(VC4_HDMI_HOTPLUG_INT),
+	HDMI_REG(VC4_HDMI_HOTPLUG),
+	HDMI_REG(VC4_HDMI_HORZA),
+	HDMI_REG(VC4_HDMI_HORZB),
+	HDMI_REG(VC4_HDMI_FIFO_CTL),
+	HDMI_REG(VC4_HDMI_SCHEDULER_CONTROL),
+	HDMI_REG(VC4_HDMI_VERTA0),
+	HDMI_REG(VC4_HDMI_VERTA1),
+	HDMI_REG(VC4_HDMI_VERTB0),
+	HDMI_REG(VC4_HDMI_VERTB1),
+	HDMI_REG(VC4_HDMI_TX_PHY_RESET_CTL),
+};
+
+static const struct {
+	u32 reg;
+	const char *name;
+} hd_regs[] = {
+	HDMI_REG(VC4_HD_M_CTL),
+	HDMI_REG(VC4_HD_MAI_CTL),
+	HDMI_REG(VC4_HD_VID_CTL),
+	HDMI_REG(VC4_HD_CSC_CTL),
+	HDMI_REG(VC4_HD_FRAME_COUNT),
+};
+
+#ifdef CONFIG_DEBUG_FS
+int vc4_hdmi_debugfs_regs(struct seq_file *m, void *unused)
+{
+	struct drm_info_node *node = (struct drm_info_node *)m->private;
+	struct drm_device *dev = node->minor->dev;
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(hdmi_regs); i++) {
+		seq_printf(m, "%s (0x%04x): 0x%08x\n",
+			   hdmi_regs[i].name, hdmi_regs[i].reg,
+			   HDMI_READ(hdmi_regs[i].reg));
+	}
+
+	for (i = 0; i < ARRAY_SIZE(hd_regs); i++) {
+		seq_printf(m, "%s (0x%04x): 0x%08x\n",
+			   hd_regs[i].name, hd_regs[i].reg,
+			   HD_READ(hd_regs[i].reg));
+	}
+
+	return 0;
+}
+#endif /* CONFIG_DEBUG_FS */
+
+static void vc4_hdmi_dump_regs(struct drm_device *dev)
+{
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(hdmi_regs); i++) {
+		DRM_INFO("0x%04x (%s): 0x%08x\n",
+			 hdmi_regs[i].reg, hdmi_regs[i].name,
+			 HDMI_READ(hdmi_regs[i].reg));
+	}
+	for (i = 0; i < ARRAY_SIZE(hd_regs); i++) {
+		DRM_INFO("0x%04x (%s): 0x%08x\n",
+			 hd_regs[i].reg, hd_regs[i].name,
+			 HD_READ(hd_regs[i].reg));
+	}
+}
+
+static enum drm_connector_status
+vc4_hdmi_connector_detect(struct drm_connector *connector, bool force)
+{
+	struct drm_device *dev = connector->dev;
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+
+	if (vc4->hdmi->hpd_gpio) {
+		if (gpio_get_value(vc4->hdmi->hpd_gpio))
+			return connector_status_connected;
+		else
+			return connector_status_disconnected;
+	}
+
+	if (HDMI_READ(VC4_HDMI_HOTPLUG) & VC4_HDMI_HOTPLUG_CONNECTED)
+		return connector_status_connected;
+	else
+		return connector_status_disconnected;
+}
+
+static void vc4_hdmi_connector_destroy(struct drm_connector *connector)
+{
+	drm_connector_unregister(connector);
+	drm_connector_cleanup(connector);
+}
+
+static int vc4_hdmi_connector_get_modes(struct drm_connector *connector)
+{
+	struct vc4_hdmi_connector *vc4_connector =
+		to_vc4_hdmi_connector(connector);
+	struct drm_encoder *encoder = vc4_connector->encoder;
+	struct vc4_hdmi_encoder *vc4_encoder = to_vc4_hdmi_encoder(encoder);
+	struct drm_device *dev = connector->dev;
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+	int ret = 0;
+	struct edid *edid;
+
+	edid = drm_get_edid(connector, vc4->hdmi->ddc);
+	if (!edid)
+		return -ENODEV;
+
+	vc4_encoder->hdmi_monitor = drm_detect_hdmi_monitor(edid);
+	drm_mode_connector_update_edid_property(connector, edid);
+	ret = drm_add_edid_modes(connector, edid);
+
+	return ret;
+}
+
+static struct drm_encoder *
+vc4_hdmi_connector_best_encoder(struct drm_connector *connector)
+{
+	struct vc4_hdmi_connector *hdmi_connector =
+		to_vc4_hdmi_connector(connector);
+	return hdmi_connector->encoder;
+}
+
+static const struct drm_connector_funcs vc4_hdmi_connector_funcs = {
+	.dpms = drm_atomic_helper_connector_dpms,
+	.detect = vc4_hdmi_connector_detect,
+	.fill_modes = drm_helper_probe_single_connector_modes,
+	.destroy = vc4_hdmi_connector_destroy,
+	.reset = drm_atomic_helper_connector_reset,
+	.atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state,
+	.atomic_destroy_state = drm_atomic_helper_connector_destroy_state,
+};
+
+static const struct drm_connector_helper_funcs vc4_hdmi_connector_helper_funcs = {
+	.get_modes = vc4_hdmi_connector_get_modes,
+	.best_encoder = vc4_hdmi_connector_best_encoder,
+};
+
+static struct drm_connector *vc4_hdmi_connector_init(struct drm_device *dev,
+						     struct drm_encoder *encoder)
+{
+	struct drm_connector *connector = NULL;
+	struct vc4_hdmi_connector *hdmi_connector;
+	int ret = 0;
+
+	hdmi_connector = devm_kzalloc(dev->dev, sizeof(*hdmi_connector),
+				      GFP_KERNEL);
+	if (!hdmi_connector) {
+		ret = -ENOMEM;
+		goto fail;
+	}
+	connector = &hdmi_connector->base;
+
+	hdmi_connector->encoder = encoder;
+
+	drm_connector_init(dev, connector, &vc4_hdmi_connector_funcs,
+			   DRM_MODE_CONNECTOR_HDMIA);
+	drm_connector_helper_add(connector, &vc4_hdmi_connector_helper_funcs);
+
+	connector->polled = (DRM_CONNECTOR_POLL_CONNECT |
+			     DRM_CONNECTOR_POLL_DISCONNECT);
+
+	connector->interlace_allowed = 0;
+	connector->doublescan_allowed = 0;
+
+	drm_mode_connector_attach_encoder(connector, encoder);
+
+	return connector;
+
+ fail:
+	if (connector)
+		vc4_hdmi_connector_destroy(connector);
+
+	return ERR_PTR(ret);
+}
+
+static void vc4_hdmi_encoder_destroy(struct drm_encoder *encoder)
+{
+	drm_encoder_cleanup(encoder);
+}
+
+static const struct drm_encoder_funcs vc4_hdmi_encoder_funcs = {
+	.destroy = vc4_hdmi_encoder_destroy,
+};
+
+static void vc4_hdmi_encoder_mode_set(struct drm_encoder *encoder,
+				      struct drm_display_mode *unadjusted_mode,
+				      struct drm_display_mode *mode)
+{
+	struct drm_device *dev = encoder->dev;
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+	bool debug_dump_regs = false;
+	bool hsync_pos = mode->flags & DRM_MODE_FLAG_PHSYNC;
+	bool vsync_pos = mode->flags & DRM_MODE_FLAG_PVSYNC;
+	u32 vactive = (mode->vdisplay >>
+		       ((mode->flags & DRM_MODE_FLAG_INTERLACE) ? 1 : 0));
+	u32 verta = (VC4_SET_FIELD(mode->vsync_end - mode->vsync_start,
+				   VC4_HDMI_VERTA_VSP) |
+		     VC4_SET_FIELD(mode->vsync_start - mode->vdisplay,
+				   VC4_HDMI_VERTA_VFP) |
+		     VC4_SET_FIELD(vactive, VC4_HDMI_VERTA_VAL));
+	u32 vertb = (VC4_SET_FIELD(0, VC4_HDMI_VERTB_VSPO) |
+		     VC4_SET_FIELD(mode->vtotal - mode->vsync_end,
+				   VC4_HDMI_VERTB_VBP));
+
+	if (debug_dump_regs) {
+		DRM_INFO("HDMI regs before:\n");
+		vc4_hdmi_dump_regs(dev);
+	}
+
+	HD_WRITE(VC4_HD_VID_CTL, 0);
+
+	clk_set_rate(vc4->hdmi->pixel_clock, mode->clock * 1000);
+
+	HDMI_WRITE(VC4_HDMI_SCHEDULER_CONTROL,
+		   HDMI_READ(VC4_HDMI_SCHEDULER_CONTROL) |
+		   VC4_HDMI_SCHEDULER_CONTROL_MANUAL_FORMAT |
+		   VC4_HDMI_SCHEDULER_CONTROL_IGNORE_VSYNC_PREDICTS);
+
+	HDMI_WRITE(VC4_HDMI_HORZA,
+		   (vsync_pos ? VC4_HDMI_HORZA_VPOS : 0) |
+		   (hsync_pos ? VC4_HDMI_HORZA_HPOS : 0) |
+		   VC4_SET_FIELD(mode->hdisplay, VC4_HDMI_HORZA_HAP));
+
+	HDMI_WRITE(VC4_HDMI_HORZB,
+		   VC4_SET_FIELD(mode->htotal - mode->hsync_end,
+				 VC4_HDMI_HORZB_HBP) |
+		   VC4_SET_FIELD(mode->hsync_end - mode->hsync_start,
+				 VC4_HDMI_HORZB_HSP) |
+		   VC4_SET_FIELD(mode->hsync_start - mode->hdisplay,
+				 VC4_HDMI_HORZB_HFP));
+
+	HDMI_WRITE(VC4_HDMI_VERTA0, verta);
+	HDMI_WRITE(VC4_HDMI_VERTA1, verta);
+
+	HDMI_WRITE(VC4_HDMI_VERTB0, vertb);
+	HDMI_WRITE(VC4_HDMI_VERTB1, vertb);
+
+	HD_WRITE(VC4_HD_VID_CTL,
+		 (vsync_pos ? 0 : VC4_HD_VID_CTL_VSYNC_LOW) |
+		 (hsync_pos ? 0 : VC4_HD_VID_CTL_HSYNC_LOW));
+
+	/* The RGB order applies even when CSC is disabled. */
+	HD_WRITE(VC4_HD_CSC_CTL, VC4_SET_FIELD(VC4_HD_CSC_CTL_ORDER_BGR,
+					       VC4_HD_CSC_CTL_ORDER));
+
+	HDMI_WRITE(VC4_HDMI_FIFO_CTL, VC4_HDMI_FIFO_CTL_MASTER_SLAVE_N);
+
+	if (debug_dump_regs) {
+		DRM_INFO("HDMI regs after:\n");
+		vc4_hdmi_dump_regs(dev);
+	}
+}
+
+static void vc4_hdmi_encoder_disable(struct drm_encoder *encoder)
+{
+	struct drm_device *dev = encoder->dev;
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+
+	HDMI_WRITE(VC4_HDMI_TX_PHY_RESET_CTL, 0xf << 16);
+	HD_WRITE(VC4_HD_VID_CTL,
+		 HD_READ(VC4_HD_VID_CTL) & ~VC4_HD_VID_CTL_ENABLE);
+}
+
+static void vc4_hdmi_encoder_enable(struct drm_encoder *encoder)
+{
+	struct vc4_hdmi_encoder *vc4_encoder = to_vc4_hdmi_encoder(encoder);
+	struct drm_device *dev = encoder->dev;
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+	int ret;
+
+	HDMI_WRITE(VC4_HDMI_TX_PHY_RESET_CTL, 0);
+
+	HD_WRITE(VC4_HD_VID_CTL,
+		 HD_READ(VC4_HD_VID_CTL) |
+		 VC4_HD_VID_CTL_ENABLE |
+		 VC4_HD_VID_CTL_UNDERFLOW_ENABLE |
+		 VC4_HD_VID_CTL_FRAME_COUNTER_RESET);
+
+	if (vc4_encoder->hdmi_monitor) {
+		HDMI_WRITE(VC4_HDMI_SCHEDULER_CONTROL,
+			   HDMI_READ(VC4_HDMI_SCHEDULER_CONTROL) |
+			   VC4_HDMI_SCHEDULER_CONTROL_MODE_HDMI);
+
+		ret = wait_for(HDMI_READ(VC4_HDMI_SCHEDULER_CONTROL) &
+			       VC4_HDMI_SCHEDULER_CONTROL_HDMI_ACTIVE, 1);
+		WARN_ONCE(ret, "Timeout waiting for "
+			  "VC4_HDMI_SCHEDULER_CONTROL_HDMI_ACTIVE\n");
+	} else {
+		HDMI_WRITE(VC4_HDMI_RAM_PACKET_CONFIG,
+			   HDMI_READ(VC4_HDMI_RAM_PACKET_CONFIG) &
+			   ~(VC4_HDMI_RAM_PACKET_ENABLE));
+		HDMI_WRITE(VC4_HDMI_SCHEDULER_CONTROL,
+			   HDMI_READ(VC4_HDMI_SCHEDULER_CONTROL) &
+			   ~VC4_HDMI_SCHEDULER_CONTROL_MODE_HDMI);
+
+		ret = wait_for(!(HDMI_READ(VC4_HDMI_SCHEDULER_CONTROL) &
+				 VC4_HDMI_SCHEDULER_CONTROL_HDMI_ACTIVE), 1);
+		WARN_ONCE(ret, "Timeout waiting for "
+			  "!VC4_HDMI_SCHEDULER_CONTROL_HDMI_ACTIVE\n");
+	}
+
+	if (vc4_encoder->hdmi_monitor) {
+		u32 drift;
+
+		WARN_ON(!(HDMI_READ(VC4_HDMI_SCHEDULER_CONTROL) &
+			  VC4_HDMI_SCHEDULER_CONTROL_HDMI_ACTIVE));
+		HDMI_WRITE(VC4_HDMI_SCHEDULER_CONTROL,
+			   HDMI_READ(VC4_HDMI_SCHEDULER_CONTROL) |
+			   VC4_HDMI_SCHEDULER_CONTROL_VERT_ALWAYS_KEEPOUT);
+
+		/* XXX: Set HDMI_RAM_PACKET_CONFIG (1 << 16) and set
+		 * up the infoframe.
+		 */
+
+		drift = HDMI_READ(VC4_HDMI_FIFO_CTL);
+		drift &= VC4_HDMI_FIFO_VALID_WRITE_MASK;
+
+		HDMI_WRITE(VC4_HDMI_FIFO_CTL,
+			   drift & ~VC4_HDMI_FIFO_CTL_RECENTER);
+		HDMI_WRITE(VC4_HDMI_FIFO_CTL,
+			   drift | VC4_HDMI_FIFO_CTL_RECENTER);
+		udelay(1000);
+		HDMI_WRITE(VC4_HDMI_FIFO_CTL,
+			   drift & ~VC4_HDMI_FIFO_CTL_RECENTER);
+		HDMI_WRITE(VC4_HDMI_FIFO_CTL,
+			   drift | VC4_HDMI_FIFO_CTL_RECENTER);
+
+		ret = wait_for(HDMI_READ(VC4_HDMI_FIFO_CTL) &
+			       VC4_HDMI_FIFO_CTL_RECENTER_DONE, 1);
+		WARN_ONCE(ret, "Timeout waiting for "
+			  "VC4_HDMI_FIFO_CTL_RECENTER_DONE");
+	}
+}
+
+static const struct drm_encoder_helper_funcs vc4_hdmi_encoder_helper_funcs = {
+	.mode_set = vc4_hdmi_encoder_mode_set,
+	.disable = vc4_hdmi_encoder_disable,
+	.enable = vc4_hdmi_encoder_enable,
+};
+
+static int vc4_hdmi_bind(struct device *dev, struct device *master, void *data)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct drm_device *drm = dev_get_drvdata(master);
+	struct vc4_dev *vc4 = drm->dev_private;
+	struct vc4_hdmi *hdmi;
+	struct vc4_hdmi_encoder *vc4_hdmi_encoder;
+	struct device_node *ddc_node;
+	u32 value;
+	int ret;
+
+	hdmi = devm_kzalloc(dev, sizeof(*hdmi), GFP_KERNEL);
+	if (!hdmi)
+		return -ENOMEM;
+
+	vc4_hdmi_encoder = devm_kzalloc(dev, sizeof(*vc4_hdmi_encoder),
+					GFP_KERNEL);
+	if (!vc4_hdmi_encoder)
+		return -ENOMEM;
+	vc4_hdmi_encoder->base.type = VC4_ENCODER_TYPE_HDMI;
+	hdmi->encoder = &vc4_hdmi_encoder->base.base;
+
+	hdmi->pdev = pdev;
+	hdmi->hdmicore_regs = vc4_ioremap_regs(pdev, 0);
+	if (IS_ERR(hdmi->hdmicore_regs))
+		return PTR_ERR(hdmi->hdmicore_regs);
+
+	hdmi->hd_regs = vc4_ioremap_regs(pdev, 1);
+	if (IS_ERR(hdmi->hd_regs))
+		return PTR_ERR(hdmi->hd_regs);
+
+	ddc_node = of_parse_phandle(dev->of_node, "ddc", 0);
+	if (!ddc_node) {
+		DRM_ERROR("Failed to find ddc node in device tree\n");
+		return -ENODEV;
+	}
+
+	hdmi->pixel_clock = devm_clk_get(dev, "pixel");
+	if (IS_ERR(hdmi->pixel_clock)) {
+		DRM_ERROR("Failed to get pixel clock\n");
+		return PTR_ERR(hdmi->pixel_clock);
+	}
+	hdmi->hsm_clock = devm_clk_get(dev, "hdmi");
+	if (IS_ERR(hdmi->hsm_clock)) {
+		DRM_ERROR("Failed to get HDMI state machine clock\n");
+		return PTR_ERR(hdmi->hsm_clock);
+	}
+
+	hdmi->ddc = of_find_i2c_adapter_by_node(ddc_node);
+	if (!hdmi->ddc) {
+		DRM_DEBUG("Failed to get ddc i2c adapter by node\n");
+		return -EPROBE_DEFER;
+	}
+
+	/* Enable the clocks at startup.  We can't quite recover from
+	 * turning off the pixel clock during disable/enables yet, so
+	 * it's always running.
+	 */
+	ret = clk_prepare_enable(hdmi->pixel_clock);
+	if (ret) {
+		DRM_ERROR("Failed to turn on pixel clock: %d\n", ret);
+		goto err_put_i2c;
+	}
+
+	ret = clk_prepare_enable(hdmi->hsm_clock);
+	if (ret) {
+		DRM_ERROR("Failed to turn on HDMI state machine clock: %d\n",
+			  ret);
+		goto err_unprepare_pix;
+	}
+
+	/* Only use the GPIO HPD pin if present in the DT, otherwise
+	 * we'll use the HDMI core's register.
+	 */
+	if (of_find_property(dev->of_node, "hpd-gpios", &value)) {
+		hdmi->hpd_gpio = of_get_named_gpio(dev->of_node, "hpd-gpios", 0);
+		if (hdmi->hpd_gpio < 0) {
+			ret = hdmi->hpd_gpio;
+			goto err_unprepare_hsm;
+		}
+	}
+
+	vc4->hdmi = hdmi;
+
+	/* HDMI core must be enabled. */
+	WARN_ON_ONCE((HD_READ(VC4_HD_M_CTL) & VC4_HD_M_ENABLE) == 0);
+
+	drm_encoder_init(drm, hdmi->encoder, &vc4_hdmi_encoder_funcs,
+			 DRM_MODE_ENCODER_TMDS);
+	drm_encoder_helper_add(hdmi->encoder, &vc4_hdmi_encoder_helper_funcs);
+
+	hdmi->connector = vc4_hdmi_connector_init(drm, hdmi->encoder);
+	if (IS_ERR(hdmi->connector)) {
+		ret = PTR_ERR(hdmi->connector);
+		goto err_destroy_encoder;
+	}
+
+	return 0;
+
+err_destroy_encoder:
+	vc4_hdmi_encoder_destroy(hdmi->encoder);
+err_unprepare_hsm:
+	clk_disable_unprepare(hdmi->hsm_clock);
+err_unprepare_pix:
+	clk_disable_unprepare(hdmi->pixel_clock);
+err_put_i2c:
+	put_device(&vc4->hdmi->ddc->dev);
+
+	return ret;
+}
+
+static void vc4_hdmi_unbind(struct device *dev, struct device *master,
+			    void *data)
+{
+	struct drm_device *drm = dev_get_drvdata(master);
+	struct vc4_dev *vc4 = drm->dev_private;
+	struct vc4_hdmi *hdmi = vc4->hdmi;
+
+	vc4_hdmi_connector_destroy(hdmi->connector);
+	vc4_hdmi_encoder_destroy(hdmi->encoder);
+
+	clk_disable_unprepare(hdmi->pixel_clock);
+	clk_disable_unprepare(hdmi->hsm_clock);
+	put_device(&hdmi->ddc->dev);
+
+	vc4->hdmi = NULL;
+}
+
+static const struct component_ops vc4_hdmi_ops = {
+	.bind   = vc4_hdmi_bind,
+	.unbind = vc4_hdmi_unbind,
+};
+
+static int vc4_hdmi_dev_probe(struct platform_device *pdev)
+{
+	return component_add(&pdev->dev, &vc4_hdmi_ops);
+}
+
+static int vc4_hdmi_dev_remove(struct platform_device *pdev)
+{
+	component_del(&pdev->dev, &vc4_hdmi_ops);
+	return 0;
+}
+
+static const struct of_device_id vc4_hdmi_dt_match[] = {
+	{ .compatible = "brcm,bcm2835-hdmi" },
+	{}
+};
+
+struct platform_driver vc4_hdmi_driver = {
+	.probe = vc4_hdmi_dev_probe,
+	.remove = vc4_hdmi_dev_remove,
+	.driver = {
+		.name = "vc4_hdmi",
+		.of_match_table = vc4_hdmi_dt_match,
+	},
+};
diff --git a/drivers/gpu/drm/vc4/vc4_hvs.c b/drivers/gpu/drm/vc4/vc4_hvs.c
new file mode 100644
index 0000000..ab1673f
--- /dev/null
+++ b/drivers/gpu/drm/vc4/vc4_hvs.c
@@ -0,0 +1,163 @@
+/*
+ * Copyright (C) 2015 Broadcom
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+/**
+ * DOC: VC4 HVS module.
+ *
+ * The HVS is the piece of hardware that does translation, scaling,
+ * colorspace conversion, and compositing of pixels stored in
+ * framebuffers into a FIFO of pixels going out to the Pixel Valve
+ * (CRTC).  It operates at the system clock rate (the system audio
+ * clock gate, specifically), which is much higher than the pixel
+ * clock rate.
+ *
+ * There is a single global HVS, with multiple output FIFOs that can
+ * be consumed by the PVs.  This file just manages the resources for
+ * the HVS, while the vc4_crtc.c code actually drives HVS setup for
+ * each CRTC.
+ */
+
+#include "linux/component.h"
+#include "vc4_drv.h"
+#include "vc4_regs.h"
+
+#define HVS_REG(reg) { reg, #reg }
+static const struct {
+	u32 reg;
+	const char *name;
+} hvs_regs[] = {
+	HVS_REG(SCALER_DISPCTRL),
+	HVS_REG(SCALER_DISPSTAT),
+	HVS_REG(SCALER_DISPID),
+	HVS_REG(SCALER_DISPECTRL),
+	HVS_REG(SCALER_DISPPROF),
+	HVS_REG(SCALER_DISPDITHER),
+	HVS_REG(SCALER_DISPEOLN),
+	HVS_REG(SCALER_DISPLIST0),
+	HVS_REG(SCALER_DISPLIST1),
+	HVS_REG(SCALER_DISPLIST2),
+	HVS_REG(SCALER_DISPLSTAT),
+	HVS_REG(SCALER_DISPLACT0),
+	HVS_REG(SCALER_DISPLACT1),
+	HVS_REG(SCALER_DISPLACT2),
+	HVS_REG(SCALER_DISPCTRL0),
+	HVS_REG(SCALER_DISPBKGND0),
+	HVS_REG(SCALER_DISPSTAT0),
+	HVS_REG(SCALER_DISPBASE0),
+	HVS_REG(SCALER_DISPCTRL1),
+	HVS_REG(SCALER_DISPBKGND1),
+	HVS_REG(SCALER_DISPSTAT1),
+	HVS_REG(SCALER_DISPBASE1),
+	HVS_REG(SCALER_DISPCTRL2),
+	HVS_REG(SCALER_DISPBKGND2),
+	HVS_REG(SCALER_DISPSTAT2),
+	HVS_REG(SCALER_DISPBASE2),
+	HVS_REG(SCALER_DISPALPHA2),
+};
+
+void vc4_hvs_dump_state(struct drm_device *dev)
+{
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(hvs_regs); i++) {
+		DRM_INFO("0x%04x (%s): 0x%08x\n",
+			 hvs_regs[i].reg, hvs_regs[i].name,
+			 HVS_READ(hvs_regs[i].reg));
+	}
+
+	DRM_INFO("HVS ctx:\n");
+	for (i = 0; i < 64; i += 4) {
+		DRM_INFO("0x%08x (%s): 0x%08x 0x%08x 0x%08x 0x%08x\n",
+			 i * 4, i < HVS_BOOTLOADER_DLIST_END ? "B" : "D",
+			 ((uint32_t *)vc4->hvs->dlist)[i + 0],
+			 ((uint32_t *)vc4->hvs->dlist)[i + 1],
+			 ((uint32_t *)vc4->hvs->dlist)[i + 2],
+			 ((uint32_t *)vc4->hvs->dlist)[i + 3]);
+	}
+}
+
+#ifdef CONFIG_DEBUG_FS
+int vc4_hvs_debugfs_regs(struct seq_file *m, void *unused)
+{
+	struct drm_info_node *node = (struct drm_info_node *)m->private;
+	struct drm_device *dev = node->minor->dev;
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(hvs_regs); i++) {
+		seq_printf(m, "%s (0x%04x): 0x%08x\n",
+			   hvs_regs[i].name, hvs_regs[i].reg,
+			   HVS_READ(hvs_regs[i].reg));
+	}
+
+	return 0;
+}
+#endif
+
+static int vc4_hvs_bind(struct device *dev, struct device *master, void *data)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct drm_device *drm = dev_get_drvdata(master);
+	struct vc4_dev *vc4 = drm->dev_private;
+	struct vc4_hvs *hvs = NULL;
+
+	hvs = devm_kzalloc(&pdev->dev, sizeof(*hvs), GFP_KERNEL);
+	if (!hvs)
+		return -ENOMEM;
+
+	hvs->pdev = pdev;
+
+	hvs->regs = vc4_ioremap_regs(pdev, 0);
+	if (IS_ERR(hvs->regs))
+		return PTR_ERR(hvs->regs);
+
+	hvs->dlist = hvs->regs + SCALER_DLIST_START;
+
+	vc4->hvs = hvs;
+	return 0;
+}
+
+static void vc4_hvs_unbind(struct device *dev, struct device *master,
+			   void *data)
+{
+	struct drm_device *drm = dev_get_drvdata(master);
+	struct vc4_dev *vc4 = drm->dev_private;
+
+	vc4->hvs = NULL;
+}
+
+static const struct component_ops vc4_hvs_ops = {
+	.bind   = vc4_hvs_bind,
+	.unbind = vc4_hvs_unbind,
+};
+
+static int vc4_hvs_dev_probe(struct platform_device *pdev)
+{
+	return component_add(&pdev->dev, &vc4_hvs_ops);
+}
+
+static int vc4_hvs_dev_remove(struct platform_device *pdev)
+{
+	component_del(&pdev->dev, &vc4_hvs_ops);
+	return 0;
+}
+
+static const struct of_device_id vc4_hvs_dt_match[] = {
+	{ .compatible = "brcm,bcm2835-hvs" },
+	{}
+};
+
+struct platform_driver vc4_hvs_driver = {
+	.probe = vc4_hvs_dev_probe,
+	.remove = vc4_hvs_dev_remove,
+	.driver = {
+		.name = "vc4_hvs",
+		.of_match_table = vc4_hvs_dt_match,
+	},
+};
diff --git a/drivers/gpu/drm/vc4/vc4_kms.c b/drivers/gpu/drm/vc4/vc4_kms.c
new file mode 100644
index 0000000..c241c8f
--- /dev/null
+++ b/drivers/gpu/drm/vc4/vc4_kms.c
@@ -0,0 +1,54 @@
+/*
+ * Copyright (C) 2015 Broadcom
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+/**
+ * DOC: VC4 KMS
+ *
+ * This is the general code for implementing KMS mode setting that
+ * doesn't clearly associate with any of the other objects (plane,
+ * crtc, HDMI encoder).
+ */
+
+#include "drm_crtc.h"
+#include "drm_atomic_helper.h"
+#include "drm_crtc_helper.h"
+#include "drm_plane_helper.h"
+#include "drm_fb_cma_helper.h"
+#include "vc4_drv.h"
+
+static const struct drm_mode_config_funcs vc4_mode_funcs = {
+	.atomic_check = drm_atomic_helper_check,
+	.atomic_commit = drm_atomic_helper_commit,
+	.fb_create = drm_fb_cma_create,
+};
+
+int vc4_kms_load(struct drm_device *dev)
+{
+	int ret;
+
+	ret = drm_vblank_init(dev, dev->mode_config.num_crtc);
+	if (ret < 0) {
+		dev_err(dev->dev, "failed to initialize vblank\n");
+		return ret;
+	}
+
+	dev->mode_config.max_width = 2048;
+	dev->mode_config.max_height = 2048;
+	dev->mode_config.funcs = &vc4_mode_funcs;
+	dev->mode_config.preferred_depth = 24;
+
+	drm_mode_config_reset(dev);
+
+	drm_fbdev_cma_init(dev, 32,
+			   dev->mode_config.num_crtc,
+			   dev->mode_config.num_connector);
+
+	drm_kms_helper_poll_init(dev);
+
+	return 0;
+}
diff --git a/drivers/gpu/drm/vc4/vc4_plane.c b/drivers/gpu/drm/vc4/vc4_plane.c
new file mode 100644
index 0000000..cdd8b10
--- /dev/null
+++ b/drivers/gpu/drm/vc4/vc4_plane.c
@@ -0,0 +1,320 @@
+/*
+ * Copyright (C) 2015 Broadcom
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+/**
+ * DOC: VC4 plane module
+ *
+ * Each DRM plane is a layer of pixels being scanned out by the HVS.
+ *
+ * At atomic modeset check time, we compute the HVS display element
+ * state that would be necessary for displaying the plane (giving us a
+ * chance to figure out if a plane configuration is invalid), then at
+ * atomic flush time the CRTC will ask us to write our element state
+ * into the region of the HVS that it has allocated for us.
+ */
+
+#include "vc4_drv.h"
+#include "vc4_regs.h"
+#include "drm_atomic_helper.h"
+#include "drm_fb_cma_helper.h"
+#include "drm_plane_helper.h"
+
+struct vc4_plane_state {
+	struct drm_plane_state base;
+	u32 *dlist;
+	u32 dlist_size; /* Number of dwords in allocated for the display list */
+	u32 dlist_count; /* Number of used dwords in the display list. */
+};
+
+static inline struct vc4_plane_state *
+to_vc4_plane_state(struct drm_plane_state *state)
+{
+	return (struct vc4_plane_state *)state;
+}
+
+static const struct hvs_format {
+	u32 drm; /* DRM_FORMAT_* */
+	u32 hvs; /* HVS_FORMAT_* */
+	u32 pixel_order;
+	bool has_alpha;
+} hvs_formats[] = {
+	{
+		.drm = DRM_FORMAT_XRGB8888, .hvs = HVS_PIXEL_FORMAT_RGBA8888,
+		.pixel_order = HVS_PIXEL_ORDER_ABGR, .has_alpha = false,
+	},
+	{
+		.drm = DRM_FORMAT_ARGB8888, .hvs = HVS_PIXEL_FORMAT_RGBA8888,
+		.pixel_order = HVS_PIXEL_ORDER_ABGR, .has_alpha = true,
+	},
+};
+
+static const struct hvs_format *vc4_get_hvs_format(u32 drm_format)
+{
+	unsigned i;
+
+	for (i = 0; i < ARRAY_SIZE(hvs_formats); i++) {
+		if (hvs_formats[i].drm == drm_format)
+			return &hvs_formats[i];
+	}
+
+	return NULL;
+}
+
+static bool plane_enabled(struct drm_plane_state *state)
+{
+	return state->fb && state->crtc;
+}
+
+struct drm_plane_state *vc4_plane_duplicate_state(struct drm_plane *plane)
+{
+	struct vc4_plane_state *vc4_state;
+
+	if (WARN_ON(!plane->state))
+		return NULL;
+
+	vc4_state = kmemdup(plane->state, sizeof(*vc4_state), GFP_KERNEL);
+	if (!vc4_state)
+		return NULL;
+
+	__drm_atomic_helper_plane_duplicate_state(plane, &vc4_state->base);
+
+	if (vc4_state->dlist) {
+		vc4_state->dlist = kmemdup(vc4_state->dlist,
+					   vc4_state->dlist_count * 4,
+					   GFP_KERNEL);
+		if (!vc4_state->dlist) {
+			kfree(vc4_state);
+			return NULL;
+		}
+		vc4_state->dlist_size = vc4_state->dlist_count;
+	}
+
+	return &vc4_state->base;
+}
+
+void vc4_plane_destroy_state(struct drm_plane *plane,
+			     struct drm_plane_state *state)
+{
+	struct vc4_plane_state *vc4_state = to_vc4_plane_state(state);
+
+	kfree(vc4_state->dlist);
+	__drm_atomic_helper_plane_destroy_state(plane, &vc4_state->base);
+	kfree(state);
+}
+
+/* Called during init to allocate the plane's atomic state. */
+void vc4_plane_reset(struct drm_plane *plane)
+{
+	struct vc4_plane_state *vc4_state;
+
+	WARN_ON(plane->state);
+
+	vc4_state = kzalloc(sizeof(*vc4_state), GFP_KERNEL);
+	if (!vc4_state)
+		return;
+
+	plane->state = &vc4_state->base;
+	vc4_state->base.plane = plane;
+}
+
+static void vc4_dlist_write(struct vc4_plane_state *vc4_state, u32 val)
+{
+	if (vc4_state->dlist_count == vc4_state->dlist_size) {
+		u32 new_size = max(4u, vc4_state->dlist_count * 2);
+		u32 *new_dlist = kmalloc(new_size * 4, GFP_KERNEL);
+
+		if (!new_dlist)
+			return;
+		memcpy(new_dlist, vc4_state->dlist, vc4_state->dlist_count * 4);
+
+		kfree(vc4_state->dlist);
+		vc4_state->dlist = new_dlist;
+		vc4_state->dlist_size = new_size;
+	}
+
+	vc4_state->dlist[vc4_state->dlist_count++] = val;
+}
+
+/* Writes out a full display list for an active plane to the plane's
+ * private dlist state.
+ */
+static int vc4_plane_mode_set(struct drm_plane *plane,
+			      struct drm_plane_state *state)
+{
+	struct vc4_plane_state *vc4_state = to_vc4_plane_state(state);
+	struct drm_framebuffer *fb = state->fb;
+	struct drm_gem_cma_object *bo = drm_fb_cma_get_gem_obj(fb, 0);
+	u32 ctl0_offset = vc4_state->dlist_count;
+	const struct hvs_format *format = vc4_get_hvs_format(fb->pixel_format);
+	uint32_t offset = fb->offsets[0];
+	int crtc_x = state->crtc_x;
+	int crtc_y = state->crtc_y;
+	int crtc_w = state->crtc_w;
+	int crtc_h = state->crtc_h;
+
+	if (crtc_x < 0) {
+		offset += drm_format_plane_cpp(fb->pixel_format, 0) * -crtc_x;
+		crtc_w += crtc_x;
+		crtc_x = 0;
+	}
+
+	if (crtc_y < 0) {
+		offset += fb->pitches[0] * -crtc_y;
+		crtc_h += crtc_y;
+		crtc_y = 0;
+	}
+
+	vc4_dlist_write(vc4_state,
+			SCALER_CTL0_VALID |
+			(format->pixel_order << SCALER_CTL0_ORDER_SHIFT) |
+			(format->hvs << SCALER_CTL0_PIXEL_FORMAT_SHIFT) |
+			SCALER_CTL0_UNITY);
+
+	/* Position Word 0: Image Positions and Alpha Value */
+	vc4_dlist_write(vc4_state,
+			VC4_SET_FIELD(0xff, SCALER_POS0_FIXED_ALPHA) |
+			VC4_SET_FIELD(crtc_x, SCALER_POS0_START_X) |
+			VC4_SET_FIELD(crtc_y, SCALER_POS0_START_Y));
+
+	/* Position Word 1: Scaled Image Dimensions.
+	 * Skipped due to SCALER_CTL0_UNITY scaling.
+	 */
+
+	/* Position Word 2: Source Image Size, Alpha Mode */
+	vc4_dlist_write(vc4_state,
+			VC4_SET_FIELD(format->has_alpha ?
+				      SCALER_POS2_ALPHA_MODE_PIPELINE :
+				      SCALER_POS2_ALPHA_MODE_FIXED,
+				      SCALER_POS2_ALPHA_MODE) |
+			VC4_SET_FIELD(crtc_w, SCALER_POS2_WIDTH) |
+			VC4_SET_FIELD(crtc_h, SCALER_POS2_HEIGHT));
+
+	/* Position Word 3: Context.  Written by the HVS. */
+	vc4_dlist_write(vc4_state, 0xc0c0c0c0);
+
+	/* Pointer Word 0: RGB / Y Pointer */
+	vc4_dlist_write(vc4_state, bo->paddr + offset);
+
+	/* Pointer Context Word 0: Written by the HVS */
+	vc4_dlist_write(vc4_state, 0xc0c0c0c0);
+
+	/* Pitch word 0: Pointer 0 Pitch */
+	vc4_dlist_write(vc4_state,
+			VC4_SET_FIELD(fb->pitches[0], SCALER_SRC_PITCH));
+
+	vc4_state->dlist[ctl0_offset] |=
+		VC4_SET_FIELD(vc4_state->dlist_count, SCALER_CTL0_SIZE);
+
+	return 0;
+}
+
+/* If a modeset involves changing the setup of a plane, the atomic
+ * infrastructure will call this to validate a proposed plane setup.
+ * However, if a plane isn't getting updated, this (and the
+ * corresponding vc4_plane_atomic_update) won't get called.  Thus, we
+ * compute the dlist here and have all active plane dlists get updated
+ * in the CRTC's flush.
+ */
+static int vc4_plane_atomic_check(struct drm_plane *plane,
+				  struct drm_plane_state *state)
+{
+	struct vc4_plane_state *vc4_state = to_vc4_plane_state(state);
+
+	vc4_state->dlist_count = 0;
+
+	if (plane_enabled(state))
+		return vc4_plane_mode_set(plane, state);
+	else
+		return 0;
+}
+
+static void vc4_plane_atomic_update(struct drm_plane *plane,
+				    struct drm_plane_state *old_state)
+{
+	/* No contents here.  Since we don't know where in the CRTC's
+	 * dlist we should be stored, our dlist is uploaded to the
+	 * hardware with vc4_plane_write_dlist() at CRTC atomic_flush
+	 * time.
+	 */
+}
+
+u32 vc4_plane_write_dlist(struct drm_plane *plane, u32 __iomem *dlist)
+{
+	struct vc4_plane_state *vc4_state = to_vc4_plane_state(plane->state);
+	int i;
+
+	/* Can't memcpy_toio() because it needs to be 32-bit writes. */
+	for (i = 0; i < vc4_state->dlist_count; i++)
+		writel(vc4_state->dlist[i], &dlist[i]);
+
+	return vc4_state->dlist_count;
+}
+
+u32 vc4_plane_dlist_size(struct drm_plane_state *state)
+{
+	struct vc4_plane_state *vc4_state = to_vc4_plane_state(state);
+
+	return vc4_state->dlist_count;
+}
+
+static const struct drm_plane_helper_funcs vc4_plane_helper_funcs = {
+	.prepare_fb = NULL,
+	.cleanup_fb = NULL,
+	.atomic_check = vc4_plane_atomic_check,
+	.atomic_update = vc4_plane_atomic_update,
+};
+
+static void vc4_plane_destroy(struct drm_plane *plane)
+{
+	drm_plane_helper_disable(plane);
+	drm_plane_cleanup(plane);
+}
+
+static const struct drm_plane_funcs vc4_plane_funcs = {
+	.update_plane = drm_atomic_helper_update_plane,
+	.disable_plane = drm_atomic_helper_disable_plane,
+	.destroy = vc4_plane_destroy,
+	.set_property = NULL,
+	.reset = vc4_plane_reset,
+	.atomic_duplicate_state = vc4_plane_duplicate_state,
+	.atomic_destroy_state = vc4_plane_destroy_state,
+};
+
+struct drm_plane *vc4_plane_init(struct drm_device *dev,
+				 enum drm_plane_type type)
+{
+	struct drm_plane *plane = NULL;
+	struct vc4_plane *vc4_plane;
+	u32 formats[ARRAY_SIZE(hvs_formats)];
+	int ret = 0;
+	unsigned i;
+
+	vc4_plane = devm_kzalloc(dev->dev, sizeof(*vc4_plane),
+				 GFP_KERNEL);
+	if (!vc4_plane) {
+		ret = -ENOMEM;
+		goto fail;
+	}
+
+	for (i = 0; i < ARRAY_SIZE(hvs_formats); i++)
+		formats[i] = hvs_formats[i].drm;
+	plane = &vc4_plane->base;
+	ret = drm_universal_plane_init(dev, plane, 0xff,
+				       &vc4_plane_funcs,
+				       formats, ARRAY_SIZE(formats),
+				       type);
+
+	drm_plane_helper_add(plane, &vc4_plane_helper_funcs);
+
+	return plane;
+fail:
+	if (plane)
+		vc4_plane_destroy(plane);
+
+	return ERR_PTR(ret);
+}
diff --git a/drivers/gpu/drm/vc4/vc4_regs.h b/drivers/gpu/drm/vc4/vc4_regs.h
new file mode 100644
index 0000000..9e4e904
--- /dev/null
+++ b/drivers/gpu/drm/vc4/vc4_regs.h
@@ -0,0 +1,570 @@
+/*
+ *  Copyright © 2014-2015 Broadcom
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#ifndef VC4_REGS_H
+#define VC4_REGS_H
+
+#include <linux/bitops.h>
+
+#define VC4_MASK(high, low) ((u32)GENMASK(high, low))
+/* Using the GNU statement expression extension */
+#define VC4_SET_FIELD(value, field)					\
+	({								\
+		uint32_t fieldval = (value) << field##_SHIFT;		\
+		WARN_ON((fieldval & ~field##_MASK) != 0);		\
+		fieldval & field##_MASK;				\
+	 })
+
+#define VC4_GET_FIELD(word, field) (((word) & field##_MASK) >>		\
+				    field##_SHIFT)
+
+#define V3D_IDENT0   0x00000
+# define V3D_EXPECTED_IDENT0 \
+	((2 << 24) | \
+	('V' << 0) | \
+	('3' << 8) | \
+	 ('D' << 16))
+
+#define V3D_IDENT1   0x00004
+/* Multiples of 1kb */
+# define V3D_IDENT1_VPM_SIZE_MASK                      VC4_MASK(31, 28)
+# define V3D_IDENT1_VPM_SIZE_SHIFT                     28
+# define V3D_IDENT1_NSEM_MASK                          VC4_MASK(23, 16)
+# define V3D_IDENT1_NSEM_SHIFT                         16
+# define V3D_IDENT1_TUPS_MASK                          VC4_MASK(15, 12)
+# define V3D_IDENT1_TUPS_SHIFT                         12
+# define V3D_IDENT1_QUPS_MASK                          VC4_MASK(11, 8)
+# define V3D_IDENT1_QUPS_SHIFT                         8
+# define V3D_IDENT1_NSLC_MASK                          VC4_MASK(7, 4)
+# define V3D_IDENT1_NSLC_SHIFT                         4
+# define V3D_IDENT1_REV_MASK                           VC4_MASK(3, 0)
+# define V3D_IDENT1_REV_SHIFT                          0
+
+#define V3D_IDENT2   0x00008
+#define V3D_SCRATCH  0x00010
+#define V3D_L2CACTL  0x00020
+# define V3D_L2CACTL_L2CCLR                            BIT(2)
+# define V3D_L2CACTL_L2CDIS                            BIT(1)
+# define V3D_L2CACTL_L2CENA                            BIT(0)
+
+#define V3D_SLCACTL  0x00024
+# define V3D_SLCACTL_T1CC_MASK                         VC4_MASK(27, 24)
+# define V3D_SLCACTL_T1CC_SHIFT                        24
+# define V3D_SLCACTL_T0CC_MASK                         VC4_MASK(19, 16)
+# define V3D_SLCACTL_T0CC_SHIFT                        16
+# define V3D_SLCACTL_UCC_MASK                          VC4_MASK(11, 8)
+# define V3D_SLCACTL_UCC_SHIFT                         8
+# define V3D_SLCACTL_ICC_MASK                          VC4_MASK(3, 0)
+# define V3D_SLCACTL_ICC_SHIFT                         0
+
+#define V3D_INTCTL   0x00030
+#define V3D_INTENA   0x00034
+#define V3D_INTDIS   0x00038
+# define V3D_INT_SPILLUSE                              BIT(3)
+# define V3D_INT_OUTOMEM                               BIT(2)
+# define V3D_INT_FLDONE                                BIT(1)
+# define V3D_INT_FRDONE                                BIT(0)
+
+#define V3D_CT0CS    0x00100
+#define V3D_CT1CS    0x00104
+#define V3D_CTNCS(n) (V3D_CT0CS + 4 * n)
+# define V3D_CTRSTA      BIT(15)
+# define V3D_CTSEMA      BIT(12)
+# define V3D_CTRTSD      BIT(8)
+# define V3D_CTRUN       BIT(5)
+# define V3D_CTSUBS      BIT(4)
+# define V3D_CTERR       BIT(3)
+# define V3D_CTMODE      BIT(0)
+
+#define V3D_CT0EA    0x00108
+#define V3D_CT1EA    0x0010c
+#define V3D_CTNEA(n) (V3D_CT0EA + 4 * (n))
+#define V3D_CT0CA    0x00110
+#define V3D_CT1CA    0x00114
+#define V3D_CTNCA(n) (V3D_CT0CA + 4 * (n))
+#define V3D_CT00RA0  0x00118
+#define V3D_CT01RA0  0x0011c
+#define V3D_CTNRA0(n) (V3D_CT00RA0 + 4 * (n))
+#define V3D_CT0LC    0x00120
+#define V3D_CT1LC    0x00124
+#define V3D_CTNLC(n) (V3D_CT0LC + 4 * (n))
+#define V3D_CT0PC    0x00128
+#define V3D_CT1PC    0x0012c
+#define V3D_CTNPC(n) (V3D_CT0PC + 4 * (n))
+
+#define V3D_PCS      0x00130
+# define V3D_BMOOM       BIT(8)
+# define V3D_RMBUSY      BIT(3)
+# define V3D_RMACTIVE    BIT(2)
+# define V3D_BMBUSY      BIT(1)
+# define V3D_BMACTIVE    BIT(0)
+
+#define V3D_BFC      0x00134
+#define V3D_RFC      0x00138
+#define V3D_BPCA     0x00300
+#define V3D_BPCS     0x00304
+#define V3D_BPOA     0x00308
+#define V3D_BPOS     0x0030c
+#define V3D_BXCF     0x00310
+#define V3D_SQRSV0   0x00410
+#define V3D_SQRSV1   0x00414
+#define V3D_SQCNTL   0x00418
+#define V3D_SRQPC    0x00430
+#define V3D_SRQUA    0x00434
+#define V3D_SRQUL    0x00438
+#define V3D_SRQCS    0x0043c
+#define V3D_VPACNTL  0x00500
+#define V3D_VPMBASE  0x00504
+#define V3D_PCTRC    0x00670
+#define V3D_PCTRE    0x00674
+#define V3D_PCTR0    0x00680
+#define V3D_PCTRS0   0x00684
+#define V3D_PCTR1    0x00688
+#define V3D_PCTRS1   0x0068c
+#define V3D_PCTR2    0x00690
+#define V3D_PCTRS2   0x00694
+#define V3D_PCTR3    0x00698
+#define V3D_PCTRS3   0x0069c
+#define V3D_PCTR4    0x006a0
+#define V3D_PCTRS4   0x006a4
+#define V3D_PCTR5    0x006a8
+#define V3D_PCTRS5   0x006ac
+#define V3D_PCTR6    0x006b0
+#define V3D_PCTRS6   0x006b4
+#define V3D_PCTR7    0x006b8
+#define V3D_PCTRS7   0x006bc
+#define V3D_PCTR8    0x006c0
+#define V3D_PCTRS8   0x006c4
+#define V3D_PCTR9    0x006c8
+#define V3D_PCTRS9   0x006cc
+#define V3D_PCTR10   0x006d0
+#define V3D_PCTRS10  0x006d4
+#define V3D_PCTR11   0x006d8
+#define V3D_PCTRS11  0x006dc
+#define V3D_PCTR12   0x006e0
+#define V3D_PCTRS12  0x006e4
+#define V3D_PCTR13   0x006e8
+#define V3D_PCTRS13  0x006ec
+#define V3D_PCTR14   0x006f0
+#define V3D_PCTRS14  0x006f4
+#define V3D_PCTR15   0x006f8
+#define V3D_PCTRS15  0x006fc
+#define V3D_BGE      0x00f00
+#define V3D_FDBGO    0x00f04
+#define V3D_FDBGB    0x00f08
+#define V3D_FDBGR    0x00f0c
+#define V3D_FDBGS    0x00f10
+#define V3D_ERRSTAT  0x00f20
+
+#define PV_CONTROL				0x00
+# define PV_CONTROL_FORMAT_MASK			VC4_MASK(23, 21)
+# define PV_CONTROL_FORMAT_SHIFT		21
+# define PV_CONTROL_FORMAT_24			0
+# define PV_CONTROL_FORMAT_DSIV_16		1
+# define PV_CONTROL_FORMAT_DSIC_16		2
+# define PV_CONTROL_FORMAT_DSIV_18		3
+# define PV_CONTROL_FORMAT_DSIV_24		4
+
+# define PV_CONTROL_FIFO_LEVEL_MASK		VC4_MASK(20, 15)
+# define PV_CONTROL_FIFO_LEVEL_SHIFT		15
+# define PV_CONTROL_CLR_AT_START		BIT(14)
+# define PV_CONTROL_TRIGGER_UNDERFLOW		BIT(13)
+# define PV_CONTROL_WAIT_HSTART			BIT(12)
+# define PV_CONTROL_CLK_SELECT_DSI_VEC		0
+# define PV_CONTROL_CLK_SELECT_DPI_SMI_HDMI	1
+# define PV_CONTROL_CLK_SELECT_MASK		VC4_MASK(3, 2)
+# define PV_CONTROL_CLK_SELECT_SHIFT		2
+# define PV_CONTROL_FIFO_CLR			BIT(1)
+# define PV_CONTROL_EN				BIT(0)
+
+#define PV_V_CONTROL				0x04
+# define PV_VCONTROL_INTERLACE			BIT(4)
+# define PV_VCONTROL_CONTINUOUS			BIT(1)
+# define PV_VCONTROL_VIDEN			BIT(0)
+
+#define PV_VSYNCD				0x08
+
+#define PV_HORZA				0x0c
+# define PV_HORZA_HBP_MASK			VC4_MASK(31, 16)
+# define PV_HORZA_HBP_SHIFT			16
+# define PV_HORZA_HSYNC_MASK			VC4_MASK(15, 0)
+# define PV_HORZA_HSYNC_SHIFT			0
+
+#define PV_HORZB				0x10
+# define PV_HORZB_HFP_MASK			VC4_MASK(31, 16)
+# define PV_HORZB_HFP_SHIFT			16
+# define PV_HORZB_HACTIVE_MASK			VC4_MASK(15, 0)
+# define PV_HORZB_HACTIVE_SHIFT			0
+
+#define PV_VERTA				0x14
+# define PV_VERTA_VBP_MASK			VC4_MASK(31, 16)
+# define PV_VERTA_VBP_SHIFT			16
+# define PV_VERTA_VSYNC_MASK			VC4_MASK(15, 0)
+# define PV_VERTA_VSYNC_SHIFT			0
+
+#define PV_VERTB				0x18
+# define PV_VERTB_VFP_MASK			VC4_MASK(31, 16)
+# define PV_VERTB_VFP_SHIFT			16
+# define PV_VERTB_VACTIVE_MASK			VC4_MASK(15, 0)
+# define PV_VERTB_VACTIVE_SHIFT			0
+
+#define PV_VERTA_EVEN				0x1c
+#define PV_VERTB_EVEN				0x20
+
+#define PV_INTEN				0x24
+#define PV_INTSTAT				0x28
+# define PV_INT_VID_IDLE			BIT(9)
+# define PV_INT_VFP_END				BIT(8)
+# define PV_INT_VFP_START			BIT(7)
+# define PV_INT_VACT_START			BIT(6)
+# define PV_INT_VBP_START			BIT(5)
+# define PV_INT_VSYNC_START			BIT(4)
+# define PV_INT_HFP_START			BIT(3)
+# define PV_INT_HACT_START			BIT(2)
+# define PV_INT_HBP_START			BIT(1)
+# define PV_INT_HSYNC_START			BIT(0)
+
+#define PV_STAT					0x2c
+
+#define PV_HACT_ACT				0x30
+
+#define SCALER_DISPCTRL                         0x00000000
+/* Global register for clock gating the HVS */
+# define SCALER_DISPCTRL_ENABLE			BIT(31)
+# define SCALER_DISPCTRL_DSP2EISLUR		BIT(15)
+# define SCALER_DISPCTRL_DSP1EISLUR		BIT(14)
+/* Enables Display 0 short line and underrun contribution to
+ * SCALER_DISPSTAT_IRQDISP0.  Note that short frame contributions are
+ * always enabled.
+ */
+# define SCALER_DISPCTRL_DSP0EISLUR		BIT(13)
+# define SCALER_DISPCTRL_DSP2EIEOLN		BIT(12)
+# define SCALER_DISPCTRL_DSP2EIEOF		BIT(11)
+# define SCALER_DISPCTRL_DSP1EIEOLN		BIT(10)
+# define SCALER_DISPCTRL_DSP1EIEOF		BIT(9)
+/* Enables Display 0 end-of-line-N contribution to
+ * SCALER_DISPSTAT_IRQDISP0
+ */
+# define SCALER_DISPCTRL_DSP0EIEOLN		BIT(8)
+/* Enables Display 0 EOF contribution to SCALER_DISPSTAT_IRQDISP0 */
+# define SCALER_DISPCTRL_DSP0EIEOF		BIT(7)
+
+# define SCALER_DISPCTRL_SLVRDEIRQ		BIT(6)
+# define SCALER_DISPCTRL_SLVWREIRQ		BIT(5)
+# define SCALER_DISPCTRL_DMAEIRQ		BIT(4)
+# define SCALER_DISPCTRL_DISP2EIRQ		BIT(3)
+# define SCALER_DISPCTRL_DISP1EIRQ		BIT(2)
+/* Enables interrupt generation on the enabled EOF/EOLN/EISLUR
+ * bits and short frames..
+ */
+# define SCALER_DISPCTRL_DISP0EIRQ		BIT(1)
+/* Enables interrupt generation on scaler profiler interrupt. */
+# define SCALER_DISPCTRL_SCLEIRQ		BIT(0)
+
+#define SCALER_DISPSTAT                         0x00000004
+# define SCALER_DISPSTAT_COBLOW2		BIT(29)
+# define SCALER_DISPSTAT_EOLN2			BIT(28)
+# define SCALER_DISPSTAT_ESFRAME2		BIT(27)
+# define SCALER_DISPSTAT_ESLINE2		BIT(26)
+# define SCALER_DISPSTAT_EUFLOW2		BIT(25)
+# define SCALER_DISPSTAT_EOF2			BIT(24)
+
+# define SCALER_DISPSTAT_COBLOW1		BIT(21)
+# define SCALER_DISPSTAT_EOLN1			BIT(20)
+# define SCALER_DISPSTAT_ESFRAME1		BIT(19)
+# define SCALER_DISPSTAT_ESLINE1		BIT(18)
+# define SCALER_DISPSTAT_EUFLOW1		BIT(17)
+# define SCALER_DISPSTAT_EOF1			BIT(16)
+
+# define SCALER_DISPSTAT_RESP_MASK		VC4_MASK(15, 14)
+# define SCALER_DISPSTAT_RESP_SHIFT		14
+# define SCALER_DISPSTAT_RESP_OKAY		0
+# define SCALER_DISPSTAT_RESP_EXOKAY		1
+# define SCALER_DISPSTAT_RESP_SLVERR		2
+# define SCALER_DISPSTAT_RESP_DECERR		3
+
+# define SCALER_DISPSTAT_COBLOW0		BIT(13)
+/* Set when the DISPEOLN line is done compositing. */
+# define SCALER_DISPSTAT_EOLN0			BIT(12)
+/* Set when VSTART is seen but there are still pixels in the current
+ * output line.
+ */
+# define SCALER_DISPSTAT_ESFRAME0		BIT(11)
+/* Set when HSTART is seen but there are still pixels in the current
+ * output line.
+ */
+# define SCALER_DISPSTAT_ESLINE0		BIT(10)
+/* Set when the the downstream tries to read from the display FIFO
+ * while it's empty.
+ */
+# define SCALER_DISPSTAT_EUFLOW0		BIT(9)
+/* Set when the display mode changes from RUN to EOF */
+# define SCALER_DISPSTAT_EOF0			BIT(8)
+
+/* Set on AXI invalid DMA ID error. */
+# define SCALER_DISPSTAT_DMA_ERROR		BIT(7)
+/* Set on AXI slave read decode error */
+# define SCALER_DISPSTAT_IRQSLVRD		BIT(6)
+/* Set on AXI slave write decode error */
+# define SCALER_DISPSTAT_IRQSLVWR		BIT(5)
+/* Set when SCALER_DISPSTAT_DMA_ERROR is set, or
+ * SCALER_DISPSTAT_RESP_ERROR is not SCALER_DISPSTAT_RESP_OKAY.
+ */
+# define SCALER_DISPSTAT_IRQDMA			BIT(4)
+# define SCALER_DISPSTAT_IRQDISP2		BIT(3)
+# define SCALER_DISPSTAT_IRQDISP1		BIT(2)
+/* Set when any of the EOF/EOLN/ESFRAME/ESLINE bits are set and their
+ * corresponding interrupt bit is enabled in DISPCTRL.
+ */
+# define SCALER_DISPSTAT_IRQDISP0		BIT(1)
+/* On read, the profiler interrupt.  On write, clear *all* interrupt bits. */
+# define SCALER_DISPSTAT_IRQSCL			BIT(0)
+
+#define SCALER_DISPID                           0x00000008
+#define SCALER_DISPECTRL                        0x0000000c
+#define SCALER_DISPPROF                         0x00000010
+#define SCALER_DISPDITHER                       0x00000014
+#define SCALER_DISPEOLN                         0x00000018
+#define SCALER_DISPLIST0                        0x00000020
+#define SCALER_DISPLIST1                        0x00000024
+#define SCALER_DISPLIST2                        0x00000028
+#define SCALER_DISPLSTAT                        0x0000002c
+#define SCALER_DISPLISTX(x)			(SCALER_DISPLIST0 +	\
+						 (x) * (SCALER_DISPLIST1 - \
+							SCALER_DISPLIST0))
+
+#define SCALER_DISPLACT0                        0x00000030
+#define SCALER_DISPLACT1                        0x00000034
+#define SCALER_DISPLACT2                        0x00000038
+#define SCALER_DISPCTRL0                        0x00000040
+# define SCALER_DISPCTRLX_ENABLE		BIT(31)
+# define SCALER_DISPCTRLX_RESET			BIT(30)
+# define SCALER_DISPCTRLX_WIDTH_MASK		VC4_MASK(23, 12)
+# define SCALER_DISPCTRLX_WIDTH_SHIFT		12
+# define SCALER_DISPCTRLX_HEIGHT_MASK		VC4_MASK(11, 0)
+# define SCALER_DISPCTRLX_HEIGHT_SHIFT		0
+
+#define SCALER_DISPBKGND0                       0x00000044
+#define SCALER_DISPSTAT0                        0x00000048
+#define SCALER_DISPBASE0                        0x0000004c
+# define SCALER_DISPSTATX_MODE_MASK		VC4_MASK(31, 30)
+# define SCALER_DISPSTATX_MODE_SHIFT		30
+# define SCALER_DISPSTATX_MODE_DISABLED		0
+# define SCALER_DISPSTATX_MODE_INIT		1
+# define SCALER_DISPSTATX_MODE_RUN		2
+# define SCALER_DISPSTATX_MODE_EOF		3
+# define SCALER_DISPSTATX_FULL			BIT(29)
+# define SCALER_DISPSTATX_EMPTY			BIT(28)
+#define SCALER_DISPCTRL1                        0x00000050
+#define SCALER_DISPBKGND1                       0x00000054
+#define SCALER_DISPSTAT1                        0x00000058
+#define SCALER_DISPSTATX(x)			(SCALER_DISPSTAT0 +        \
+						 (x) * (SCALER_DISPSTAT1 - \
+							SCALER_DISPSTAT0))
+#define SCALER_DISPBASE1                        0x0000005c
+#define SCALER_DISPCTRL2                        0x00000060
+#define SCALER_DISPCTRLX(x)			(SCALER_DISPCTRL0 +        \
+						 (x) * (SCALER_DISPCTRL1 - \
+							SCALER_DISPCTRL0))
+#define SCALER_DISPBKGND2                       0x00000064
+#define SCALER_DISPSTAT2                        0x00000068
+#define SCALER_DISPBASE2                        0x0000006c
+#define SCALER_DISPALPHA2                       0x00000070
+#define SCALER_GAMADDR                          0x00000078
+#define SCALER_GAMDATA                          0x000000e0
+#define SCALER_DLIST_START                      0x00002000
+#define SCALER_DLIST_SIZE                       0x00004000
+
+#define VC4_HDMI_CORE_REV			0x000
+
+#define VC4_HDMI_SW_RESET_CONTROL		0x004
+# define VC4_HDMI_SW_RESET_FORMAT_DETECT	BIT(1)
+# define VC4_HDMI_SW_RESET_HDMI			BIT(0)
+
+#define VC4_HDMI_HOTPLUG_INT			0x008
+
+#define VC4_HDMI_HOTPLUG			0x00c
+# define VC4_HDMI_HOTPLUG_CONNECTED		BIT(0)
+
+#define VC4_HDMI_RAM_PACKET_CONFIG		0x0a0
+# define VC4_HDMI_RAM_PACKET_ENABLE		BIT(16)
+
+#define VC4_HDMI_HORZA				0x0c4
+# define VC4_HDMI_HORZA_VPOS			BIT(14)
+# define VC4_HDMI_HORZA_HPOS			BIT(13)
+/* Horizontal active pixels (hdisplay). */
+# define VC4_HDMI_HORZA_HAP_MASK		VC4_MASK(12, 0)
+# define VC4_HDMI_HORZA_HAP_SHIFT		0
+
+#define VC4_HDMI_HORZB				0x0c8
+/* Horizontal pack porch (htotal - hsync_end). */
+# define VC4_HDMI_HORZB_HBP_MASK		VC4_MASK(29, 20)
+# define VC4_HDMI_HORZB_HBP_SHIFT		20
+/* Horizontal sync pulse (hsync_end - hsync_start). */
+# define VC4_HDMI_HORZB_HSP_MASK		VC4_MASK(19, 10)
+# define VC4_HDMI_HORZB_HSP_SHIFT		10
+/* Horizontal front porch (hsync_start - hdisplay). */
+# define VC4_HDMI_HORZB_HFP_MASK		VC4_MASK(9, 0)
+# define VC4_HDMI_HORZB_HFP_SHIFT		0
+
+#define VC4_HDMI_FIFO_CTL			0x05c
+# define VC4_HDMI_FIFO_CTL_RECENTER_DONE	BIT(14)
+# define VC4_HDMI_FIFO_CTL_USE_EMPTY		BIT(13)
+# define VC4_HDMI_FIFO_CTL_ON_VB		BIT(7)
+# define VC4_HDMI_FIFO_CTL_RECENTER		BIT(6)
+# define VC4_HDMI_FIFO_CTL_FIFO_RESET		BIT(5)
+# define VC4_HDMI_FIFO_CTL_USE_PLL_LOCK		BIT(4)
+# define VC4_HDMI_FIFO_CTL_INV_CLK_XFR		BIT(3)
+# define VC4_HDMI_FIFO_CTL_CAPTURE_PTR		BIT(2)
+# define VC4_HDMI_FIFO_CTL_USE_FULL		BIT(1)
+# define VC4_HDMI_FIFO_CTL_MASTER_SLAVE_N	BIT(0)
+# define VC4_HDMI_FIFO_VALID_WRITE_MASK		0xefff
+
+#define VC4_HDMI_SCHEDULER_CONTROL		0x0c0
+# define VC4_HDMI_SCHEDULER_CONTROL_MANUAL_FORMAT BIT(15)
+# define VC4_HDMI_SCHEDULER_CONTROL_IGNORE_VSYNC_PREDICTS BIT(5)
+# define VC4_HDMI_SCHEDULER_CONTROL_VERT_ALWAYS_KEEPOUT	BIT(3)
+# define VC4_HDMI_SCHEDULER_CONTROL_HDMI_ACTIVE	BIT(1)
+# define VC4_HDMI_SCHEDULER_CONTROL_MODE_HDMI	BIT(0)
+
+#define VC4_HDMI_VERTA0				0x0cc
+#define VC4_HDMI_VERTA1				0x0d4
+/* Vertical sync pulse (vsync_end - vsync_start). */
+# define VC4_HDMI_VERTA_VSP_MASK		VC4_MASK(24, 20)
+# define VC4_HDMI_VERTA_VSP_SHIFT		20
+/* Vertical front porch (vsync_start - vdisplay). */
+# define VC4_HDMI_VERTA_VFP_MASK		VC4_MASK(19, 13)
+# define VC4_HDMI_VERTA_VFP_SHIFT		13
+/* Vertical active lines (vdisplay). */
+# define VC4_HDMI_VERTA_VAL_MASK		VC4_MASK(12, 0)
+# define VC4_HDMI_VERTA_VAL_SHIFT		0
+
+#define VC4_HDMI_VERTB0				0x0d0
+#define VC4_HDMI_VERTB1				0x0d8
+/* Vertical sync pulse offset (for interlaced) */
+# define VC4_HDMI_VERTB_VSPO_MASK		VC4_MASK(21, 9)
+# define VC4_HDMI_VERTB_VSPO_SHIFT		9
+/* Vertical pack porch (vtotal - vsync_end). */
+# define VC4_HDMI_VERTB_VBP_MASK		VC4_MASK(8, 0)
+# define VC4_HDMI_VERTB_VBP_SHIFT		0
+
+#define VC4_HDMI_TX_PHY_RESET_CTL		0x2c0
+
+#define VC4_HD_M_CTL				0x00c
+# define VC4_HD_M_SW_RST			BIT(2)
+# define VC4_HD_M_ENABLE			BIT(0)
+
+#define VC4_HD_MAI_CTL				0x014
+
+#define VC4_HD_VID_CTL				0x038
+# define VC4_HD_VID_CTL_ENABLE			BIT(31)
+# define VC4_HD_VID_CTL_UNDERFLOW_ENABLE	BIT(30)
+# define VC4_HD_VID_CTL_FRAME_COUNTER_RESET	BIT(29)
+# define VC4_HD_VID_CTL_VSYNC_LOW		BIT(28)
+# define VC4_HD_VID_CTL_HSYNC_LOW		BIT(27)
+
+#define VC4_HD_CSC_CTL				0x040
+# define VC4_HD_CSC_CTL_ORDER_MASK		VC4_MASK(7, 5)
+# define VC4_HD_CSC_CTL_ORDER_SHIFT		5
+# define VC4_HD_CSC_CTL_ORDER_RGB		0
+# define VC4_HD_CSC_CTL_ORDER_BGR		1
+# define VC4_HD_CSC_CTL_ORDER_BRG		2
+# define VC4_HD_CSC_CTL_ORDER_GRB		3
+# define VC4_HD_CSC_CTL_ORDER_GBR		4
+# define VC4_HD_CSC_CTL_ORDER_RBG		5
+# define VC4_HD_CSC_CTL_PADMSB			BIT(4)
+# define VC4_HD_CSC_CTL_MODE_MASK		VC4_MASK(3, 2)
+# define VC4_HD_CSC_CTL_MODE_SHIFT		2
+# define VC4_HD_CSC_CTL_MODE_RGB_TO_SD_YPRPB	0
+# define VC4_HD_CSC_CTL_MODE_RGB_TO_HD_YPRPB	1
+# define VC4_HD_CSC_CTL_MODE_CUSTOM		2
+# define VC4_HD_CSC_CTL_RGB2YCC			BIT(1)
+# define VC4_HD_CSC_CTL_ENABLE			BIT(0)
+
+#define VC4_HD_FRAME_COUNT			0x068
+
+/* HVS display list information. */
+#define HVS_BOOTLOADER_DLIST_END                32
+
+enum hvs_pixel_format {
+	/* 8bpp */
+	HVS_PIXEL_FORMAT_RGB332 = 0,
+	/* 16bpp */
+	HVS_PIXEL_FORMAT_RGBA4444 = 1,
+	HVS_PIXEL_FORMAT_RGB555 = 2,
+	HVS_PIXEL_FORMAT_RGBA5551 = 3,
+	HVS_PIXEL_FORMAT_RGB565 = 4,
+	/* 24bpp */
+	HVS_PIXEL_FORMAT_RGB888 = 5,
+	HVS_PIXEL_FORMAT_RGBA6666 = 6,
+	/* 32bpp */
+	HVS_PIXEL_FORMAT_RGBA8888 = 7
+};
+
+/* Note: the LSB is the rightmost character shown.  Only valid for
+ * HVS_PIXEL_FORMAT_RGB8888, not RGB888.
+ */
+#define HVS_PIXEL_ORDER_RGBA			0
+#define HVS_PIXEL_ORDER_BGRA			1
+#define HVS_PIXEL_ORDER_ARGB			2
+#define HVS_PIXEL_ORDER_ABGR			3
+
+#define HVS_PIXEL_ORDER_XBRG			0
+#define HVS_PIXEL_ORDER_XRBG			1
+#define HVS_PIXEL_ORDER_XRGB			2
+#define HVS_PIXEL_ORDER_XBGR			3
+
+#define HVS_PIXEL_ORDER_XYCBCR			0
+#define HVS_PIXEL_ORDER_XYCRCB			1
+#define HVS_PIXEL_ORDER_YXCBCR			2
+#define HVS_PIXEL_ORDER_YXCRCB			3
+
+#define SCALER_CTL0_END				BIT(31)
+#define SCALER_CTL0_VALID			BIT(30)
+
+#define SCALER_CTL0_SIZE_MASK			VC4_MASK(29, 24)
+#define SCALER_CTL0_SIZE_SHIFT			24
+
+#define SCALER_CTL0_HFLIP                       BIT(16)
+#define SCALER_CTL0_VFLIP                       BIT(15)
+
+#define SCALER_CTL0_ORDER_MASK			VC4_MASK(14, 13)
+#define SCALER_CTL0_ORDER_SHIFT			13
+
+/* Set to indicate no scaling. */
+#define SCALER_CTL0_UNITY			BIT(4)
+
+#define SCALER_CTL0_PIXEL_FORMAT_MASK		VC4_MASK(3, 0)
+#define SCALER_CTL0_PIXEL_FORMAT_SHIFT		0
+
+#define SCALER_POS0_FIXED_ALPHA_MASK		VC4_MASK(31, 24)
+#define SCALER_POS0_FIXED_ALPHA_SHIFT		24
+
+#define SCALER_POS0_START_Y_MASK		VC4_MASK(23, 12)
+#define SCALER_POS0_START_Y_SHIFT		12
+
+#define SCALER_POS0_START_X_MASK		VC4_MASK(11, 0)
+#define SCALER_POS0_START_X_SHIFT		0
+
+#define SCALER_POS2_ALPHA_MODE_MASK		VC4_MASK(31, 30)
+#define SCALER_POS2_ALPHA_MODE_SHIFT		30
+#define SCALER_POS2_ALPHA_MODE_PIPELINE		0
+#define SCALER_POS2_ALPHA_MODE_FIXED		1
+#define SCALER_POS2_ALPHA_MODE_FIXED_NONZERO	2
+#define SCALER_POS2_ALPHA_MODE_FIXED_OVER_0x07	3
+
+#define SCALER_POS2_HEIGHT_MASK			VC4_MASK(27, 16)
+#define SCALER_POS2_HEIGHT_SHIFT		16
+
+#define SCALER_POS2_WIDTH_MASK			VC4_MASK(11, 0)
+#define SCALER_POS2_WIDTH_SHIFT			0
+
+#define SCALER_SRC_PITCH_MASK			VC4_MASK(15, 0)
+#define SCALER_SRC_PITCH_SHIFT			0
+
+#endif /* VC4_REGS_H */
-- 
cgit v0.10.2


From 48666d56312b337ff3f04451bf6c3f014fee1eeb Mon Sep 17 00:00:00 2001
From: Derek Foreman <derekf@osg.samsung.com>
Date: Thu, 2 Jul 2015 11:19:54 -0500
Subject: drm/vc4: Use the fbdev_cma helpers

Keep the fbdev_cma pointer around so we can use it on hotplog and close
to ensure the frame buffer console is in a useful state.

Signed-off-by: Derek Foreman <derekf@osg.samsung.com>
Signed-off-by: Eric Anholt <eric@anholt.net>

diff --git a/drivers/gpu/drm/vc4/vc4_drv.c b/drivers/gpu/drm/vc4/vc4_drv.c
index 3c3ccff..6e73060 100644
--- a/drivers/gpu/drm/vc4/vc4_drv.c
+++ b/drivers/gpu/drm/vc4/vc4_drv.c
@@ -14,6 +14,7 @@
 #include <linux/module.h>
 #include <linux/of_platform.h>
 #include <linux/platform_device.h>
+#include "drm_fb_cma_helper.h"
 
 #include "vc4_drv.h"
 #include "vc4_regs.h"
@@ -49,6 +50,14 @@ static void vc4_drm_preclose(struct drm_device *dev, struct drm_file *file)
 		vc4_cancel_page_flip(crtc, file);
 }
 
+static void vc4_lastclose(struct drm_device *dev)
+{
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+
+	if (vc4->fbdev)
+		drm_fbdev_cma_restore_mode(vc4->fbdev);
+}
+
 static const struct file_operations vc4_drm_fops = {
 	.owner = THIS_MODULE,
 	.open = drm_open,
@@ -71,6 +80,7 @@ static struct drm_driver vc4_drm_driver = {
 			    DRIVER_ATOMIC |
 			    DRIVER_GEM |
 			    DRIVER_PRIME),
+	.lastclose = vc4_lastclose,
 	.preclose = vc4_drm_preclose,
 
 	.enable_vblank = vc4_enable_vblank,
@@ -199,6 +209,10 @@ static void vc4_drm_unbind(struct device *dev)
 {
 	struct platform_device *pdev = to_platform_device(dev);
 	struct drm_device *drm = platform_get_drvdata(pdev);
+	struct vc4_dev *vc4 = to_vc4_dev(drm);
+
+	if (vc4->fbdev)
+		drm_fbdev_cma_fini(vc4->fbdev);
 
 	drm_mode_config_cleanup(drm);
 
diff --git a/drivers/gpu/drm/vc4/vc4_drv.h b/drivers/gpu/drm/vc4/vc4_drv.h
index 04dd2e2..57f54e0 100644
--- a/drivers/gpu/drm/vc4/vc4_drv.h
+++ b/drivers/gpu/drm/vc4/vc4_drv.h
@@ -15,6 +15,8 @@ struct vc4_dev {
 	struct vc4_hdmi *hdmi;
 	struct vc4_hvs *hvs;
 	struct vc4_crtc *crtc[3];
+
+	struct drm_fbdev_cma *fbdev;
 };
 
 static inline struct vc4_dev *
diff --git a/drivers/gpu/drm/vc4/vc4_kms.c b/drivers/gpu/drm/vc4/vc4_kms.c
index c241c8f..ee3e004 100644
--- a/drivers/gpu/drm/vc4/vc4_kms.c
+++ b/drivers/gpu/drm/vc4/vc4_kms.c
@@ -21,7 +21,16 @@
 #include "drm_fb_cma_helper.h"
 #include "vc4_drv.h"
 
+static void vc4_output_poll_changed(struct drm_device *dev)
+{
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+
+	if (vc4->fbdev)
+		drm_fbdev_cma_hotplug_event(vc4->fbdev);
+}
+
 static const struct drm_mode_config_funcs vc4_mode_funcs = {
+	.output_poll_changed = vc4_output_poll_changed,
 	.atomic_check = drm_atomic_helper_check,
 	.atomic_commit = drm_atomic_helper_commit,
 	.fb_create = drm_fb_cma_create,
@@ -29,6 +38,7 @@ static const struct drm_mode_config_funcs vc4_mode_funcs = {
 
 int vc4_kms_load(struct drm_device *dev)
 {
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
 	int ret;
 
 	ret = drm_vblank_init(dev, dev->mode_config.num_crtc);
@@ -44,9 +54,11 @@ int vc4_kms_load(struct drm_device *dev)
 
 	drm_mode_config_reset(dev);
 
-	drm_fbdev_cma_init(dev, 32,
-			   dev->mode_config.num_crtc,
-			   dev->mode_config.num_connector);
+	vc4->fbdev = drm_fbdev_cma_init(dev, 32,
+					dev->mode_config.num_crtc,
+					dev->mode_config.num_connector);
+	if (IS_ERR(vc4->fbdev))
+		vc4->fbdev = NULL;
 
 	drm_kms_helper_poll_init(dev);
 
-- 
cgit v0.10.2


From 98a44504541c6befb28366eb9ec432ba44070dd9 Mon Sep 17 00:00:00 2001
From: Derek Foreman <derekf@osg.samsung.com>
Date: Thu, 2 Jul 2015 11:20:21 -0500
Subject: drm/vc4: Allow vblank to be disabled

Signed-off-by: Derek Foreman <derekf@osg.samsung.com>
Signed-off-by: Eric Anholt <eric@anholt.net>

diff --git a/drivers/gpu/drm/vc4/vc4_kms.c b/drivers/gpu/drm/vc4/vc4_kms.c
index ee3e004..2e5597d 100644
--- a/drivers/gpu/drm/vc4/vc4_kms.c
+++ b/drivers/gpu/drm/vc4/vc4_kms.c
@@ -51,6 +51,7 @@ int vc4_kms_load(struct drm_device *dev)
 	dev->mode_config.max_height = 2048;
 	dev->mode_config.funcs = &vc4_mode_funcs;
 	dev->mode_config.preferred_depth = 24;
+	dev->vblank_disable_allowed = true;
 
 	drm_mode_config_reset(dev);
 
-- 
cgit v0.10.2


From d0be9f4ec1c4ad8a8b6be1efa5bc81e5bce957a3 Mon Sep 17 00:00:00 2001
From: Geliang Tang <geliangtang@163.com>
Date: Sun, 18 Oct 2015 23:29:48 +0800
Subject: drm: fix trivial typos

s/regsiter/register/

Signed-off-by: Geliang Tang <geliangtang@163.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/include/atombios.h b/drivers/gpu/drm/amd/include/atombios.h
index 44c5d4a..5526226 100644
--- a/drivers/gpu/drm/amd/include/atombios.h
+++ b/drivers/gpu/drm/amd/include/atombios.h
@@ -6784,7 +6784,7 @@ typedef struct _ATOM_MC_INIT_PARAM_TABLE_V2_1
   ULONG                         ulMCUcodeRomStartAddr;
   ULONG                         ulMCUcodeLength;
   USHORT                        usMcRegInitTableOffset;     // offset of ATOM_REG_INIT_SETTING array for MC core register settings.
-  USHORT                        usReserved;                 // offset of ATOM_INIT_REG_BLOCK for MC SEQ/PHY regsiter setting
+  USHORT                        usReserved;                 // offset of ATOM_INIT_REG_BLOCK for MC SEQ/PHY register setting
 }ATOM_MC_INIT_PARAM_TABLE_V2_1;
 
 
diff --git a/drivers/gpu/drm/radeon/cayman_blit_shaders.c b/drivers/gpu/drm/radeon/cayman_blit_shaders.c
index 98d009e..9fec4d0 100644
--- a/drivers/gpu/drm/radeon/cayman_blit_shaders.c
+++ b/drivers/gpu/drm/radeon/cayman_blit_shaders.c
@@ -32,7 +32,7 @@
  * evergreen cards need to use the 3D engine to blit data which requires
  * quite a bit of hw state setup.  Rather than pull the whole 3D driver
  * (which normally generates the 3D state) into the DRM, we opt to use
- * statically generated state tables.  The regsiter state and shaders
+ * statically generated state tables.  The register state and shaders
  * were hand generated to support blitting functionality.  See the 3D
  * driver or documentation for descriptions of the registers and
  * shader instructions.
diff --git a/drivers/gpu/drm/radeon/evergreen_blit_shaders.c b/drivers/gpu/drm/radeon/evergreen_blit_shaders.c
index d433834..1a96ddb 100644
--- a/drivers/gpu/drm/radeon/evergreen_blit_shaders.c
+++ b/drivers/gpu/drm/radeon/evergreen_blit_shaders.c
@@ -32,7 +32,7 @@
  * evergreen cards need to use the 3D engine to blit data which requires
  * quite a bit of hw state setup.  Rather than pull the whole 3D driver
  * (which normally generates the 3D state) into the DRM, we opt to use
- * statically generated state tables.  The regsiter state and shaders
+ * statically generated state tables.  The register state and shaders
  * were hand generated to support blitting functionality.  See the 3D
  * driver or documentation for descriptions of the registers and
  * shader instructions.
diff --git a/drivers/gpu/drm/radeon/r600_blit_shaders.c b/drivers/gpu/drm/radeon/r600_blit_shaders.c
index 34c8b23..443cbe5 100644
--- a/drivers/gpu/drm/radeon/r600_blit_shaders.c
+++ b/drivers/gpu/drm/radeon/r600_blit_shaders.c
@@ -32,7 +32,7 @@
  * R6xx+ cards need to use the 3D engine to blit data which requires
  * quite a bit of hw state setup.  Rather than pull the whole 3D driver
  * (which normally generates the 3D state) into the DRM, we opt to use
- * statically generated state tables.  The regsiter state and shaders
+ * statically generated state tables.  The register state and shaders
  * were hand generated to support blitting functionality.  See the 3D
  * driver or documentation for descriptions of the registers and
  * shader instructions.
-- 
cgit v0.10.2


From b7e4dad3e1fc5d3909737fb72e57aedeb0072c7f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Tue, 1 Sep 2015 10:50:26 +0200
Subject: drm/amdgpu: remove old lockup detection infrastructure
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

It didn't worked to well anyway.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Junwei Zhang <Jerry.Zhang@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 736eb45..911d670 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -345,7 +345,6 @@ struct amdgpu_ring_funcs {
 	/* testing functions */
 	int (*test_ring)(struct amdgpu_ring *ring);
 	int (*test_ib)(struct amdgpu_ring *ring);
-	bool (*is_lockup)(struct amdgpu_ring *ring);
 	/* insert NOP packets */
 	void (*insert_nop)(struct amdgpu_ring *ring, uint32_t count);
 };
@@ -907,8 +906,6 @@ struct amdgpu_ring {
 	unsigned		ring_size;
 	unsigned		ring_free_dw;
 	int			count_dw;
-	atomic_t		last_rptr;
-	atomic64_t		last_activity;
 	uint64_t		gpu_addr;
 	uint32_t		align_mask;
 	uint32_t		ptr_mask;
@@ -1230,8 +1227,6 @@ void amdgpu_ring_commit(struct amdgpu_ring *ring);
 void amdgpu_ring_unlock_commit(struct amdgpu_ring *ring);
 void amdgpu_ring_undo(struct amdgpu_ring *ring);
 void amdgpu_ring_unlock_undo(struct amdgpu_ring *ring);
-void amdgpu_ring_lockup_update(struct amdgpu_ring *ring);
-bool amdgpu_ring_test_lockup(struct amdgpu_ring *ring);
 unsigned amdgpu_ring_backup(struct amdgpu_ring *ring,
 			    uint32_t **data);
 int amdgpu_ring_restore(struct amdgpu_ring *ring,
@@ -1974,7 +1969,6 @@ struct amdgpu_device {
 	bool				suspend;
 	bool				need_dma32;
 	bool				accel_working;
-	bool				needs_reset;
 	struct work_struct 		reset_work;
 	struct notifier_block		acpi_nb;
 	struct amdgpu_i2c_chan		*i2c_bus[AMDGPU_MAX_I2C_BUS];
@@ -2253,7 +2247,6 @@ amdgpu_get_sdma_instance(struct amdgpu_ring *ring)
 #define amdgpu_ring_parse_cs(r, p, ib) ((r)->funcs->parse_cs((p), (ib)))
 #define amdgpu_ring_test_ring(r) (r)->funcs->test_ring((r))
 #define amdgpu_ring_test_ib(r) (r)->funcs->test_ib((r))
-#define amdgpu_ring_is_lockup(r) (r)->funcs->is_lockup((r))
 #define amdgpu_ring_get_rptr(r) (r)->funcs->get_rptr((r))
 #define amdgpu_ring_get_wptr(r) (r)->funcs->get_wptr((r))
 #define amdgpu_ring_set_wptr(r) (r)->funcs->set_wptr((r))
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 901a460..28c9ee3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -1816,12 +1816,6 @@ int amdgpu_gpu_reset(struct amdgpu_device *adev)
 
 	down_write(&adev->exclusive_lock);
 
-	if (!adev->needs_reset) {
-		up_write(&adev->exclusive_lock);
-		return 0;
-	}
-
-	adev->needs_reset = false;
 	atomic_inc(&adev->gpu_reset_counter);
 
 	/* block TTM */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
index fcad7e0..4010aa6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
@@ -269,17 +269,6 @@ static void amdgpu_fence_check_lockup(struct work_struct *work)
 	if (amdgpu_fence_activity(ring)) {
 		wake_up_all(&ring->fence_drv.fence_queue);
 	}
-	else if (amdgpu_ring_is_lockup(ring)) {
-		/* good news we believe it's a lockup */
-		dev_warn(ring->adev->dev, "GPU lockup (current fence id "
-			"0x%016llx last fence id 0x%016llx on ring %d)\n",
-			(uint64_t)atomic64_read(&fence_drv->last_seq),
-			fence_drv->sync_seq[ring->idx], ring->idx);
-
-		/* remember that we need an reset */
-		ring->adev->needs_reset = true;
-		wake_up_all(&ring->fence_drv.fence_queue);
-	}
 	up_read(&ring->adev->exclusive_lock);
 }
 
@@ -380,7 +369,6 @@ static bool amdgpu_fence_enable_signaling(struct fence *f)
  */
 static int amdgpu_fence_ring_wait_seq(struct amdgpu_ring *ring, uint64_t seq)
 {
-	struct amdgpu_device *adev = ring->adev;
 	bool signaled = false;
 
 	BUG_ON(!ring);
@@ -391,8 +379,7 @@ static int amdgpu_fence_ring_wait_seq(struct amdgpu_ring *ring, uint64_t seq)
 		return 0;
 
 	wait_event(ring->fence_drv.fence_queue, (
-		   (signaled = amdgpu_fence_seq_signaled(ring, seq))
-		   || adev->needs_reset));
+		   (signaled = amdgpu_fence_seq_signaled(ring, seq))));
 
 	if (signaled)
 		return 0;
@@ -939,11 +926,6 @@ signed long amdgpu_fence_wait_any(struct amdgpu_device *adev,
 		if (amdgpu_test_signaled_any(array, count))
 			break;
 
-		if (adev->needs_reset) {
-			t = -EDEADLK;
-			break;
-		}
-
 		t = schedule_timeout(t);
 
 		if (t > 0 && intr && signal_pending(current))
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
index c439735..aad4c1c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
@@ -298,7 +298,6 @@ int amdgpu_ib_ring_tests(struct amdgpu_device *adev)
 		r = amdgpu_ring_test_ib(ring);
 		if (r) {
 			ring->ready = false;
-			adev->needs_reset = false;
 
 			if (ring == &adev->gfx.gfx_ring[0]) {
 				/* oh, oh, that's really bad */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
index b13a74b..b2df348 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
@@ -67,8 +67,6 @@ void amdgpu_ring_free_size(struct amdgpu_ring *ring)
 	if (!ring->ring_free_dw) {
 		/* this is an empty ring */
 		ring->ring_free_dw = ring->ring_size / 4;
-		/*  update lockup info to avoid false positive */
-		amdgpu_ring_lockup_update(ring);
 	}
 }
 
@@ -209,46 +207,6 @@ void amdgpu_ring_unlock_undo(struct amdgpu_ring *ring)
 }
 
 /**
- * amdgpu_ring_lockup_update - update lockup variables
- *
- * @ring: amdgpu_ring structure holding ring information
- *
- * Update the last rptr value and timestamp (all asics).
- */
-void amdgpu_ring_lockup_update(struct amdgpu_ring *ring)
-{
-	atomic_set(&ring->last_rptr, amdgpu_ring_get_rptr(ring));
-	atomic64_set(&ring->last_activity, jiffies_64);
-}
-
-/**
- * amdgpu_ring_test_lockup() - check if ring is lockedup by recording information
- * @ring:       amdgpu_ring structure holding ring information
- *
- */
-bool amdgpu_ring_test_lockup(struct amdgpu_ring *ring)
-{
-	uint32_t rptr = amdgpu_ring_get_rptr(ring);
-	uint64_t last = atomic64_read(&ring->last_activity);
-	uint64_t elapsed;
-
-	if (rptr != atomic_read(&ring->last_rptr)) {
-		/* ring is still working, no lockup */
-		amdgpu_ring_lockup_update(ring);
-		return false;
-	}
-
-	elapsed = jiffies_to_msecs(jiffies_64 - last);
-	if (amdgpu_lockup_timeout && elapsed >= amdgpu_lockup_timeout) {
-		dev_err(ring->adev->dev, "ring %d stalled for more than %llumsec\n",
-			ring->idx, elapsed);
-		return true;
-	}
-	/* give a chance to the GPU ... */
-	return false;
-}
-
-/**
  * amdgpu_ring_backup - Back up the content of a ring
  *
  * @ring: the ring we want to back up
@@ -436,7 +394,6 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring,
 	if (amdgpu_debugfs_ring_init(adev, ring)) {
 		DRM_ERROR("Failed to register debugfs file for rings !\n");
 	}
-	amdgpu_ring_lockup_update(ring);
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
index 814598e..5f712ce 100644
--- a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
+++ b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
@@ -1290,24 +1290,6 @@ const struct amd_ip_funcs cik_sdma_ip_funcs = {
 	.set_powergating_state = cik_sdma_set_powergating_state,
 };
 
-/**
- * cik_sdma_ring_is_lockup - Check if the DMA engine is locked up
- *
- * @ring: amdgpu_ring structure holding ring information
- *
- * Check if the async DMA engine is locked up (CIK).
- * Returns true if the engine appears to be locked up, false if not.
- */
-static bool cik_sdma_ring_is_lockup(struct amdgpu_ring *ring)
-{
-
-	if (cik_sdma_is_idle(ring->adev)) {
-		amdgpu_ring_lockup_update(ring);
-		return false;
-	}
-	return amdgpu_ring_test_lockup(ring);
-}
-
 static const struct amdgpu_ring_funcs cik_sdma_ring_funcs = {
 	.get_rptr = cik_sdma_ring_get_rptr,
 	.get_wptr = cik_sdma_ring_get_wptr,
@@ -1320,7 +1302,6 @@ static const struct amdgpu_ring_funcs cik_sdma_ring_funcs = {
 	.emit_hdp_flush = cik_sdma_ring_emit_hdp_flush,
 	.test_ring = cik_sdma_ring_test_ring,
 	.test_ib = cik_sdma_ring_test_ib,
-	.is_lockup = cik_sdma_ring_is_lockup,
 	.insert_nop = cik_sdma_ring_insert_nop,
 };
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
index e992bf2..72793f9 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
@@ -5542,24 +5542,6 @@ const struct amd_ip_funcs gfx_v7_0_ip_funcs = {
 	.set_powergating_state = gfx_v7_0_set_powergating_state,
 };
 
-/**
- * gfx_v7_0_ring_is_lockup - check if the 3D engine is locked up
- *
- * @adev: amdgpu_device pointer
- * @ring: amdgpu_ring structure holding ring information
- *
- * Check if the 3D engine is locked up (CIK).
- * Returns true if the engine is locked, false if not.
- */
-static bool gfx_v7_0_ring_is_lockup(struct amdgpu_ring *ring)
-{
-	if (gfx_v7_0_is_idle(ring->adev)) {
-		amdgpu_ring_lockup_update(ring);
-		return false;
-	}
-	return amdgpu_ring_test_lockup(ring);
-}
-
 static const struct amdgpu_ring_funcs gfx_v7_0_ring_funcs_gfx = {
 	.get_rptr = gfx_v7_0_ring_get_rptr_gfx,
 	.get_wptr = gfx_v7_0_ring_get_wptr_gfx,
@@ -5573,7 +5555,6 @@ static const struct amdgpu_ring_funcs gfx_v7_0_ring_funcs_gfx = {
 	.emit_hdp_flush = gfx_v7_0_ring_emit_hdp_flush,
 	.test_ring = gfx_v7_0_ring_test_ring,
 	.test_ib = gfx_v7_0_ring_test_ib,
-	.is_lockup = gfx_v7_0_ring_is_lockup,
 	.insert_nop = amdgpu_ring_insert_nop,
 };
 
@@ -5590,7 +5571,6 @@ static const struct amdgpu_ring_funcs gfx_v7_0_ring_funcs_compute = {
 	.emit_hdp_flush = gfx_v7_0_ring_emit_hdp_flush,
 	.test_ring = gfx_v7_0_ring_test_ring,
 	.test_ib = gfx_v7_0_ring_test_ib,
-	.is_lockup = gfx_v7_0_ring_is_lockup,
 	.insert_nop = amdgpu_ring_insert_nop,
 };
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index 7dd7651..e0355ea 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -4075,15 +4075,6 @@ static void gfx_v8_0_ring_emit_vm_flush(struct amdgpu_ring *ring,
 	}
 }
 
-static bool gfx_v8_0_ring_is_lockup(struct amdgpu_ring *ring)
-{
-	if (gfx_v8_0_is_idle(ring->adev)) {
-		amdgpu_ring_lockup_update(ring);
-		return false;
-	}
-	return amdgpu_ring_test_lockup(ring);
-}
-
 static u32 gfx_v8_0_ring_get_rptr_compute(struct amdgpu_ring *ring)
 {
 	return ring->adev->wb.wb[ring->rptr_offs];
@@ -4365,7 +4356,6 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_gfx = {
 	.emit_hdp_flush = gfx_v8_0_ring_emit_hdp_flush,
 	.test_ring = gfx_v8_0_ring_test_ring,
 	.test_ib = gfx_v8_0_ring_test_ib,
-	.is_lockup = gfx_v8_0_ring_is_lockup,
 	.insert_nop = amdgpu_ring_insert_nop,
 };
 
@@ -4382,7 +4372,6 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_compute = {
 	.emit_hdp_flush = gfx_v8_0_ring_emit_hdp_flush,
 	.test_ring = gfx_v8_0_ring_test_ring,
 	.test_ib = gfx_v8_0_ring_test_ib,
-	.is_lockup = gfx_v8_0_ring_is_lockup,
 	.insert_nop = amdgpu_ring_insert_nop,
 };
 
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c
index f8b868c..2cf5018 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c
@@ -1295,24 +1295,6 @@ const struct amd_ip_funcs sdma_v2_4_ip_funcs = {
 	.set_powergating_state = sdma_v2_4_set_powergating_state,
 };
 
-/**
- * sdma_v2_4_ring_is_lockup - Check if the DMA engine is locked up
- *
- * @ring: amdgpu_ring structure holding ring information
- *
- * Check if the async DMA engine is locked up (VI).
- * Returns true if the engine appears to be locked up, false if not.
- */
-static bool sdma_v2_4_ring_is_lockup(struct amdgpu_ring *ring)
-{
-
-	if (sdma_v2_4_is_idle(ring->adev)) {
-		amdgpu_ring_lockup_update(ring);
-		return false;
-	}
-	return amdgpu_ring_test_lockup(ring);
-}
-
 static const struct amdgpu_ring_funcs sdma_v2_4_ring_funcs = {
 	.get_rptr = sdma_v2_4_ring_get_rptr,
 	.get_wptr = sdma_v2_4_ring_get_wptr,
@@ -1325,7 +1307,6 @@ static const struct amdgpu_ring_funcs sdma_v2_4_ring_funcs = {
 	.emit_hdp_flush = sdma_v2_4_ring_emit_hdp_flush,
 	.test_ring = sdma_v2_4_ring_test_ring,
 	.test_ib = sdma_v2_4_ring_test_ib,
-	.is_lockup = sdma_v2_4_ring_is_lockup,
 	.insert_nop = sdma_v2_4_ring_insert_nop,
 };
 
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
index 670555a..0271bf9 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
@@ -1428,24 +1428,6 @@ const struct amd_ip_funcs sdma_v3_0_ip_funcs = {
 	.set_powergating_state = sdma_v3_0_set_powergating_state,
 };
 
-/**
- * sdma_v3_0_ring_is_lockup - Check if the DMA engine is locked up
- *
- * @ring: amdgpu_ring structure holding ring information
- *
- * Check if the async DMA engine is locked up (VI).
- * Returns true if the engine appears to be locked up, false if not.
- */
-static bool sdma_v3_0_ring_is_lockup(struct amdgpu_ring *ring)
-{
-
-	if (sdma_v3_0_is_idle(ring->adev)) {
-		amdgpu_ring_lockup_update(ring);
-		return false;
-	}
-	return amdgpu_ring_test_lockup(ring);
-}
-
 static const struct amdgpu_ring_funcs sdma_v3_0_ring_funcs = {
 	.get_rptr = sdma_v3_0_ring_get_rptr,
 	.get_wptr = sdma_v3_0_ring_get_wptr,
@@ -1458,7 +1440,6 @@ static const struct amdgpu_ring_funcs sdma_v3_0_ring_funcs = {
 	.emit_hdp_flush = sdma_v3_0_ring_emit_hdp_flush,
 	.test_ring = sdma_v3_0_ring_test_ring,
 	.test_ib = sdma_v3_0_ring_test_ib,
-	.is_lockup = sdma_v3_0_ring_is_lockup,
 	.insert_nop = sdma_v3_0_ring_insert_nop,
 };
 
diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c b/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c
index ed50dd7..5e9f73a 100644
--- a/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c
@@ -885,7 +885,6 @@ static const struct amdgpu_ring_funcs uvd_v4_2_ring_funcs = {
 	.emit_semaphore = uvd_v4_2_ring_emit_semaphore,
 	.test_ring = uvd_v4_2_ring_test_ring,
 	.test_ib = uvd_v4_2_ring_test_ib,
-	.is_lockup = amdgpu_ring_test_lockup,
 	.insert_nop = amdgpu_ring_insert_nop,
 };
 
diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c
index 9ad8b99..38864f56 100644
--- a/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c
@@ -824,7 +824,6 @@ static const struct amdgpu_ring_funcs uvd_v5_0_ring_funcs = {
 	.emit_semaphore = uvd_v5_0_ring_emit_semaphore,
 	.test_ring = uvd_v5_0_ring_test_ring,
 	.test_ib = uvd_v5_0_ring_test_ib,
-	.is_lockup = amdgpu_ring_test_lockup,
 	.insert_nop = amdgpu_ring_insert_nop,
 };
 
diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c
index 7e9934f..121915b 100644
--- a/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c
@@ -808,7 +808,6 @@ static const struct amdgpu_ring_funcs uvd_v6_0_ring_funcs = {
 	.emit_semaphore = uvd_v6_0_ring_emit_semaphore,
 	.test_ring = uvd_v6_0_ring_test_ring,
 	.test_ib = uvd_v6_0_ring_test_ib,
-	.is_lockup = amdgpu_ring_test_lockup,
 	.insert_nop = amdgpu_ring_insert_nop,
 };
 
diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c
index cd16df5..52ac7a8 100644
--- a/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c
@@ -642,7 +642,6 @@ static const struct amdgpu_ring_funcs vce_v2_0_ring_funcs = {
 	.emit_semaphore = amdgpu_vce_ring_emit_semaphore,
 	.test_ring = amdgpu_vce_ring_test_ring,
 	.test_ib = amdgpu_vce_ring_test_ib,
-	.is_lockup = amdgpu_ring_test_lockup,
 	.insert_nop = amdgpu_ring_insert_nop,
 };
 
diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c
index f0656df..b1bf27b 100644
--- a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c
@@ -643,7 +643,6 @@ static const struct amdgpu_ring_funcs vce_v3_0_ring_funcs = {
 	.emit_semaphore = amdgpu_vce_ring_emit_semaphore,
 	.test_ring = amdgpu_vce_ring_test_ring,
 	.test_ib = amdgpu_vce_ring_test_ib,
-	.is_lockup = amdgpu_ring_test_lockup,
 	.insert_nop = amdgpu_ring_insert_nop,
 };
 
-- 
cgit v0.10.2


From 0c418f10104d4aa1d6b83698790898dc9ef1c12d Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Tue, 1 Sep 2015 15:13:53 +0200
Subject: drm/amdgpu: remove the exclusive lock
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Finally getting rid of it.

Signed-off-by: Christian König <christian.koenig@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 911d670..39d9608 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -1955,7 +1955,6 @@ struct amdgpu_device {
 	struct device			*dev;
 	struct drm_device		*ddev;
 	struct pci_dev			*pdev;
-	struct rw_semaphore		exclusive_lock;
 
 	/* ASIC */
 	enum amd_asic_type		asic_type;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index 25012c7..5fdc039 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -831,11 +831,8 @@ int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 	bool reserved_buffers = false;
 	int i, r;
 
-	down_read(&adev->exclusive_lock);
-	if (!adev->accel_working) {
-		up_read(&adev->exclusive_lock);
+	if (!adev->accel_working)
 		return -EBUSY;
-	}
 
 	parser = amdgpu_cs_parser_create(adev, filp, NULL, NULL, 0);
 	if (!parser)
@@ -843,8 +840,7 @@ int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 	r = amdgpu_cs_parser_init(parser, data);
 	if (r) {
 		DRM_ERROR("Failed to initialize parser !\n");
-		kfree(parser);
-		up_read(&adev->exclusive_lock);
+		amdgpu_cs_parser_fini(parser, r, false);
 		r = amdgpu_cs_handle_lockup(adev, r);
 		return r;
 	}
@@ -915,14 +911,12 @@ int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 
 		mutex_unlock(&job->job_lock);
 		amdgpu_cs_parser_fini_late(parser);
-		up_read(&adev->exclusive_lock);
 		return 0;
 	}
 
 	cs->out.handle = parser->ibs[parser->num_ibs - 1].sequence;
 out:
 	amdgpu_cs_parser_fini(parser, r, reserved_buffers);
-	up_read(&adev->exclusive_lock);
 	r = amdgpu_cs_handle_lockup(adev, r);
 	return r;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 28c9ee3..19e185e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -1418,7 +1418,6 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 	mutex_init(&adev->gfx.gpu_clock_mutex);
 	mutex_init(&adev->srbm_mutex);
 	mutex_init(&adev->grbm_idx_mutex);
-	init_rwsem(&adev->exclusive_lock);
 	mutex_init(&adev->mn_lock);
 	hash_init(adev->mn_hash);
 
@@ -1814,8 +1813,6 @@ int amdgpu_gpu_reset(struct amdgpu_device *adev)
 	int i, r;
 	int resched;
 
-	down_write(&adev->exclusive_lock);
-
 	atomic_inc(&adev->gpu_reset_counter);
 
 	/* block TTM */
@@ -1879,7 +1876,6 @@ retry:
 		dev_info(adev->dev, "GPU reset failed\n");
 	}
 
-	up_write(&adev->exclusive_lock);
 	return r;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
index 9b34a34..fdf7525 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
@@ -47,11 +47,8 @@ static void amdgpu_flip_wait_fence(struct amdgpu_device *adev,
 	fence = to_amdgpu_fence(*f);
 	if (fence) {
 		r = fence_wait(&fence->base, false);
-		if (r == -EDEADLK) {
-			up_read(&adev->exclusive_lock);
+		if (r == -EDEADLK)
 			r = amdgpu_gpu_reset(adev);
-			down_read(&adev->exclusive_lock);
-		}
 	} else
 		r = fence_wait(*f, false);
 
@@ -77,7 +74,6 @@ static void amdgpu_flip_work_func(struct work_struct *__work)
 	unsigned long flags;
 	unsigned i;
 
-	down_read(&adev->exclusive_lock);
 	amdgpu_flip_wait_fence(adev, &work->excl);
 	for (i = 0; i < work->shared_count; ++i)
 		amdgpu_flip_wait_fence(adev, &work->shared[i]);
@@ -93,7 +89,6 @@ static void amdgpu_flip_work_func(struct work_struct *__work)
 	amdgpuCrtc->pflip_status = AMDGPU_FLIP_SUBMITTED;
 
 	spin_unlock_irqrestore(&crtc->dev->event_lock, flags);
-	up_read(&adev->exclusive_lock);
 }
 
 /*
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
index 4010aa6..1fadc15 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
@@ -260,16 +260,8 @@ static void amdgpu_fence_check_lockup(struct work_struct *work)
 				lockup_work.work);
 	ring = fence_drv->ring;
 
-	if (!down_read_trylock(&ring->adev->exclusive_lock)) {
-		/* just reschedule the check if a reset is going on */
-		amdgpu_fence_schedule_check(ring);
-		return;
-	}
-
-	if (amdgpu_fence_activity(ring)) {
+	if (amdgpu_fence_activity(ring))
 		wake_up_all(&ring->fence_drv.fence_queue);
-	}
-	up_read(&ring->adev->exclusive_lock);
 }
 
 /**
@@ -317,18 +309,15 @@ static bool amdgpu_fence_is_signaled(struct fence *f)
 {
 	struct amdgpu_fence *fence = to_amdgpu_fence(f);
 	struct amdgpu_ring *ring = fence->ring;
-	struct amdgpu_device *adev = ring->adev;
 
 	if (atomic64_read(&ring->fence_drv.last_seq) >= fence->seq)
 		return true;
 
-	if (down_read_trylock(&adev->exclusive_lock)) {
-		amdgpu_fence_process(ring);
-		up_read(&adev->exclusive_lock);
+	amdgpu_fence_process(ring);
+
+	if (atomic64_read(&ring->fence_drv.last_seq) >= fence->seq)
+		return true;
 
-		if (atomic64_read(&ring->fence_drv.last_seq) >= fence->seq)
-			return true;
-	}
 	return false;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
index 7297ca3..d81ab78 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
@@ -181,7 +181,6 @@ int amdgpu_gem_create_ioctl(struct drm_device *dev, void *data,
 	bool kernel = false;
 	int r;
 
-	down_read(&adev->exclusive_lock);
 	/* create a gem object to contain this object in */
 	if (args->in.domains & (AMDGPU_GEM_DOMAIN_GDS |
 	    AMDGPU_GEM_DOMAIN_GWS | AMDGPU_GEM_DOMAIN_OA)) {
@@ -214,11 +213,9 @@ int amdgpu_gem_create_ioctl(struct drm_device *dev, void *data,
 
 	memset(args, 0, sizeof(*args));
 	args->out.handle = handle;
-	up_read(&adev->exclusive_lock);
 	return 0;
 
 error_unlock:
-	up_read(&adev->exclusive_lock);
 	r = amdgpu_gem_handle_lockup(adev, r);
 	return r;
 }
@@ -250,8 +247,6 @@ int amdgpu_gem_userptr_ioctl(struct drm_device *dev, void *data,
 		return -EACCES;
 	}
 
-	down_read(&adev->exclusive_lock);
-
 	/* create a gem object to contain this object in */
 	r = amdgpu_gem_object_create(adev, args->size, 0,
 				     AMDGPU_GEM_DOMAIN_CPU, 0,
@@ -293,14 +288,12 @@ int amdgpu_gem_userptr_ioctl(struct drm_device *dev, void *data,
 		goto handle_lockup;
 
 	args->handle = handle;
-	up_read(&adev->exclusive_lock);
 	return 0;
 
 release_object:
 	drm_gem_object_unreference_unlocked(gobj);
 
 handle_lockup:
-	up_read(&adev->exclusive_lock);
 	r = amdgpu_gem_handle_lockup(adev, r);
 
 	return r;
-- 
cgit v0.10.2


From ce16b0e5a32a157abd6446214e8b91c55064204e Mon Sep 17 00:00:00 2001
From: Junwei Zhang <Jerry.Zhang@amd.com>
Date: Sun, 6 Sep 2015 13:55:03 +0800
Subject: drm/amdgpu: remove the unnecessary parameter adev for
 amdgpu_fence_wait_any()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Junwei Zhang <Jerry.Zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 39d9608..a9c0def 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -447,8 +447,7 @@ int amdgpu_fence_wait_next(struct amdgpu_ring *ring);
 int amdgpu_fence_wait_empty(struct amdgpu_ring *ring);
 unsigned amdgpu_fence_count_emitted(struct amdgpu_ring *ring);
 
-signed long amdgpu_fence_wait_any(struct amdgpu_device *adev,
-				  struct fence **array,
+signed long amdgpu_fence_wait_any(struct fence **array,
 				  uint32_t count,
 				  bool intr,
 				  signed long t);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
index 1fadc15..003a219 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
@@ -857,16 +857,12 @@ static void amdgpu_fence_wait_cb(struct fence *fence, struct fence_cb *cb)
 static signed long amdgpu_fence_default_wait(struct fence *f, bool intr,
 					     signed long t)
 {
-	struct amdgpu_fence *fence = to_amdgpu_fence(f);
-	struct amdgpu_device *adev = fence->ring->adev;
-
-	return amdgpu_fence_wait_any(adev, &f, 1, intr, t);
+	return amdgpu_fence_wait_any(&f, 1, intr, t);
 }
 
 /**
  * Wait the fence array with timeout
  *
- * @adev:     amdgpu device
  * @array:    the fence array with amdgpu fence pointer
  * @count:    the number of the fence array
  * @intr:     when sleep, set the current task interruptable or not
@@ -874,8 +870,7 @@ static signed long amdgpu_fence_default_wait(struct fence *f, bool intr,
  *
  * It will return when any fence is signaled or timeout.
  */
-signed long amdgpu_fence_wait_any(struct amdgpu_device *adev,
-				  struct fence **array, uint32_t count,
+signed long amdgpu_fence_wait_any(struct fence **array, uint32_t count,
 				  bool intr, signed long t)
 {
 	struct amdgpu_wait_cb *cb;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c
index e907124..5cb27d5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c
@@ -372,7 +372,7 @@ int amdgpu_sa_bo_new(struct amdgpu_device *adev,
 		} while (amdgpu_sa_bo_next_hole(sa_manager, fences, tries));
 
 		spin_unlock(&sa_manager->wq.lock);
-		t = amdgpu_fence_wait_any(adev, fences, AMDGPU_MAX_RINGS,
+		t = amdgpu_fence_wait_any(fences, AMDGPU_MAX_RINGS,
 					  false, MAX_SCHEDULE_TIMEOUT);
 		r = (t > 0) ? 0 : t;
 		spin_lock(&sa_manager->wq.lock);
-- 
cgit v0.10.2


From f48b2659f521301753f9b3b67e308a79c6110346 Mon Sep 17 00:00:00 2001
From: Chunming Zhou <david1.zhou@amd.com>
Date: Fri, 16 Oct 2015 14:06:19 +0800
Subject: drm/amdgpu: fix the broken vm->mutex V2
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

fix the vm->mutex and ww_mutex confilcts.
vm->mutex is always token first, then ww_mutex.

V2: remove unneccessary checking for pt bo.

Change-Id: Iea56e183752c02831126d06d2f5b7a474a6e4743
Signed-off-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index 5fdc039..8727c30 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -608,7 +608,6 @@ static int amdgpu_cs_ib_vm_chunk(struct amdgpu_device *adev,
 		}
 	}
 
-	mutex_lock(&vm->mutex);
 	r = amdgpu_bo_vm_update_pte(parser, vm);
 	if (r) {
 		goto out;
@@ -619,7 +618,6 @@ static int amdgpu_cs_ib_vm_chunk(struct amdgpu_device *adev,
 				       parser->filp);
 
 out:
-	mutex_unlock(&vm->mutex);
 	return r;
 }
 
@@ -827,6 +825,8 @@ int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 {
 	struct amdgpu_device *adev = dev->dev_private;
 	union drm_amdgpu_cs *cs = data;
+	struct amdgpu_fpriv *fpriv = filp->driver_priv;
+	struct amdgpu_vm *vm = &fpriv->vm;
 	struct amdgpu_cs_parser *parser;
 	bool reserved_buffers = false;
 	int i, r;
@@ -844,7 +844,7 @@ int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 		r = amdgpu_cs_handle_lockup(adev, r);
 		return r;
 	}
-
+	mutex_lock(&vm->mutex);
 	r = amdgpu_cs_parser_relocs(parser);
 	if (r == -ENOMEM)
 		DRM_ERROR("Not enough memory for command submission!\n");
@@ -911,12 +911,14 @@ int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 
 		mutex_unlock(&job->job_lock);
 		amdgpu_cs_parser_fini_late(parser);
+		mutex_unlock(&vm->mutex);
 		return 0;
 	}
 
 	cs->out.handle = parser->ibs[parser->num_ibs - 1].sequence;
 out:
 	amdgpu_cs_parser_fini(parser, r, reserved_buffers);
+	mutex_unlock(&vm->mutex);
 	r = amdgpu_cs_handle_lockup(adev, r);
 	return r;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
index d81ab78..0873328 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
@@ -115,9 +115,10 @@ int amdgpu_gem_object_open(struct drm_gem_object *obj, struct drm_file *file_pri
 	struct amdgpu_vm *vm = &fpriv->vm;
 	struct amdgpu_bo_va *bo_va;
 	int r;
-
+	mutex_lock(&vm->mutex);
 	r = amdgpu_bo_reserve(rbo, false);
 	if (r) {
+		mutex_unlock(&vm->mutex);
 		return r;
 	}
 
@@ -128,7 +129,7 @@ int amdgpu_gem_object_open(struct drm_gem_object *obj, struct drm_file *file_pri
 		++bo_va->ref_count;
 	}
 	amdgpu_bo_unreserve(rbo);
-
+	mutex_unlock(&vm->mutex);
 	return 0;
 }
 
@@ -141,9 +142,10 @@ void amdgpu_gem_object_close(struct drm_gem_object *obj,
 	struct amdgpu_vm *vm = &fpriv->vm;
 	struct amdgpu_bo_va *bo_va;
 	int r;
-
+	mutex_lock(&vm->mutex);
 	r = amdgpu_bo_reserve(rbo, true);
 	if (r) {
+		mutex_unlock(&vm->mutex);
 		dev_err(adev->dev, "leaking bo va because "
 			"we fail to reserve bo (%d)\n", r);
 		return;
@@ -155,6 +157,7 @@ void amdgpu_gem_object_close(struct drm_gem_object *obj,
 		}
 	}
 	amdgpu_bo_unreserve(rbo);
+	mutex_unlock(&vm->mutex);
 }
 
 static int amdgpu_gem_handle_lockup(struct amdgpu_device *adev, int r)
@@ -481,18 +484,13 @@ static void amdgpu_gem_va_update_vm(struct amdgpu_device *adev,
 			goto error_unreserve;
 	}
 
-	mutex_lock(&bo_va->vm->mutex);
 	r = amdgpu_vm_clear_freed(adev, bo_va->vm);
 	if (r)
-		goto error_unlock;
-
+		goto error_unreserve;
 
 	if (operation == AMDGPU_VA_OP_MAP)
 		r = amdgpu_vm_bo_update(adev, bo_va, &bo_va->bo->tbo.mem);
 
-error_unlock:
-	mutex_unlock(&bo_va->vm->mutex);
-
 error_unreserve:
 	ttm_eu_backoff_reservation(&ticket, &list);
 
@@ -549,10 +547,11 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data,
 	gobj = drm_gem_object_lookup(dev, filp, args->handle);
 	if (gobj == NULL)
 		return -ENOENT;
-
+	mutex_lock(&fpriv->vm.mutex);
 	rbo = gem_to_amdgpu_bo(gobj);
 	r = amdgpu_bo_reserve(rbo, false);
 	if (r) {
+		mutex_unlock(&fpriv->vm.mutex);
 		drm_gem_object_unreference_unlocked(gobj);
 		return r;
 	}
@@ -560,6 +559,7 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data,
 	bo_va = amdgpu_vm_bo_find(&fpriv->vm, rbo);
 	if (!bo_va) {
 		amdgpu_bo_unreserve(rbo);
+		mutex_unlock(&fpriv->vm.mutex);
 		return -ENOENT;
 	}
 
@@ -584,7 +584,7 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data,
 
 	if (!r && !(args->flags & AMDGPU_VM_DELAY_UPDATE))
 		amdgpu_gem_va_update_vm(adev, bo_va, args->operation);
-
+	mutex_unlock(&fpriv->vm.mutex);
 	drm_gem_object_unreference_unlocked(gobj);
 	return r;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index 0675524..06e207f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -90,11 +90,9 @@ struct amdgpu_bo_list_entry *amdgpu_vm_get_bos(struct amdgpu_device *adev,
 	struct amdgpu_bo_list_entry *list;
 	unsigned i, idx;
 
-	mutex_lock(&vm->mutex);
 	list = drm_malloc_ab(vm->max_pde_used + 2,
 			     sizeof(struct amdgpu_bo_list_entry));
 	if (!list) {
-		mutex_unlock(&vm->mutex);
 		return NULL;
 	}
 
@@ -119,7 +117,6 @@ struct amdgpu_bo_list_entry *amdgpu_vm_get_bos(struct amdgpu_device *adev,
 		list[idx].tv.shared = true;
 		list_add(&list[idx++].tv.head, head);
 	}
-	mutex_unlock(&vm->mutex);
 
 	return list;
 }
@@ -970,9 +967,7 @@ struct amdgpu_bo_va *amdgpu_vm_bo_add(struct amdgpu_device *adev,
 	INIT_LIST_HEAD(&bo_va->invalids);
 	INIT_LIST_HEAD(&bo_va->vm_status);
 
-	mutex_lock(&vm->mutex);
 	list_add_tail(&bo_va->bo_list, &bo->va);
-	mutex_unlock(&vm->mutex);
 
 	return bo_va;
 }
@@ -1025,8 +1020,6 @@ int amdgpu_vm_bo_map(struct amdgpu_device *adev,
 		return -EINVAL;
 	}
 
-	mutex_lock(&vm->mutex);
-
 	saddr /= AMDGPU_GPU_PAGE_SIZE;
 	eaddr /= AMDGPU_GPU_PAGE_SIZE;
 
@@ -1040,14 +1033,14 @@ int amdgpu_vm_bo_map(struct amdgpu_device *adev,
 			tmp->it.start, tmp->it.last + 1);
 		amdgpu_bo_unreserve(bo_va->bo);
 		r = -EINVAL;
-		goto error_unlock;
+		goto error;
 	}
 
 	mapping = kmalloc(sizeof(*mapping), GFP_KERNEL);
 	if (!mapping) {
 		amdgpu_bo_unreserve(bo_va->bo);
 		r = -ENOMEM;
-		goto error_unlock;
+		goto error;
 	}
 
 	INIT_LIST_HEAD(&mapping->list);
@@ -1079,9 +1072,6 @@ int amdgpu_vm_bo_map(struct amdgpu_device *adev,
 		if (vm->page_tables[pt_idx].bo)
 			continue;
 
-		/* drop mutex to allocate and clear page table */
-		mutex_unlock(&vm->mutex);
-
 		ww_mutex_lock(&resv->lock, NULL);
 		r = amdgpu_bo_create(adev, AMDGPU_VM_PTE_COUNT * 8,
 				     AMDGPU_GPU_PAGE_SIZE, true,
@@ -1098,32 +1088,19 @@ int amdgpu_vm_bo_map(struct amdgpu_device *adev,
 			goto error_free;
 		}
 
-		/* aquire mutex again */
-		mutex_lock(&vm->mutex);
-		if (vm->page_tables[pt_idx].bo) {
-			/* someone else allocated the pt in the meantime */
-			mutex_unlock(&vm->mutex);
-			amdgpu_bo_unref(&pt);
-			mutex_lock(&vm->mutex);
-			continue;
-		}
-
 		vm->page_tables[pt_idx].addr = 0;
 		vm->page_tables[pt_idx].bo = pt;
 	}
 
-	mutex_unlock(&vm->mutex);
 	return 0;
 
 error_free:
-	mutex_lock(&vm->mutex);
 	list_del(&mapping->list);
 	interval_tree_remove(&mapping->it, &vm->va);
 	trace_amdgpu_vm_bo_unmap(bo_va, mapping);
 	kfree(mapping);
 
-error_unlock:
-	mutex_unlock(&vm->mutex);
+error:
 	return r;
 }
 
@@ -1168,7 +1145,6 @@ int amdgpu_vm_bo_unmap(struct amdgpu_device *adev,
 		}
 	}
 
-	mutex_lock(&vm->mutex);
 	list_del(&mapping->list);
 	interval_tree_remove(&mapping->it, &vm->va);
 	trace_amdgpu_vm_bo_unmap(bo_va, mapping);
@@ -1177,7 +1153,6 @@ int amdgpu_vm_bo_unmap(struct amdgpu_device *adev,
 		list_add(&mapping->list, &vm->freed);
 	else
 		kfree(mapping);
-	mutex_unlock(&vm->mutex);
 	amdgpu_bo_unreserve(bo_va->bo);
 
 	return 0;
@@ -1201,8 +1176,6 @@ void amdgpu_vm_bo_rmv(struct amdgpu_device *adev,
 
 	list_del(&bo_va->bo_list);
 
-	mutex_lock(&vm->mutex);
-
 	spin_lock(&vm->status_lock);
 	list_del(&bo_va->vm_status);
 	spin_unlock(&vm->status_lock);
@@ -1221,8 +1194,6 @@ void amdgpu_vm_bo_rmv(struct amdgpu_device *adev,
 
 	fence_put(bo_va->last_pt_update);
 	kfree(bo_va);
-
-	mutex_unlock(&vm->mutex);
 }
 
 /**
-- 
cgit v0.10.2


From f1ceb84ae0fba02dcb5aa0ff56b5546767376795 Mon Sep 17 00:00:00 2001
From: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Date: Fri, 17 Jul 2015 10:44:33 +0300
Subject: drm: rcar-du: Add support for the R8A7793 DU

The R8A7793 DU is identical to the R8A7791 and thus only requires a new
DT compatible string.

Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>

diff --git a/Documentation/devicetree/bindings/video/renesas,du.txt b/Documentation/devicetree/bindings/video/renesas,du.txt
index c902323..d05be12 100644
--- a/Documentation/devicetree/bindings/video/renesas,du.txt
+++ b/Documentation/devicetree/bindings/video/renesas,du.txt
@@ -5,7 +5,8 @@ Required Properties:
   - compatible: must be one of the following.
     - "renesas,du-r8a7779" for R8A7779 (R-Car H1) compatible DU
     - "renesas,du-r8a7790" for R8A7790 (R-Car H2) compatible DU
-    - "renesas,du-r8a7791" for R8A7791 (R-Car M2) compatible DU
+    - "renesas,du-r8a7791" for R8A7791 (R-Car M2-W) compatible DU
+    - "renesas,du-r8a7793" for R8A7793 (R-Car M2-N) compatible DU
 
   - reg: A list of base address and length of each memory resource, one for
     each entry in the reg-names property.
@@ -22,9 +23,9 @@ Required Properties:
   - clock-names: Name of the clocks. This property is model-dependent.
     - R8A7779 uses a single functional clock. The clock doesn't need to be
       named.
-    - R8A7790 and R8A7791 use one functional clock per channel and one clock
-      per LVDS encoder. The functional clocks must be named "du.x" with "x"
-      being the channel numerical index. The LVDS clocks must be named
+    - R8A779[013] use one functional clock per channel and one clock per LVDS
+      encoder (if available). The functional clocks must be named "du.x" with
+      "x" being the channel numerical index. The LVDS clocks must be named
       "lvds.x" with "x" being the LVDS encoder numerical index.
     - In addition to the functional and encoder clocks, all DU versions also
       support externally supplied pixel clocks. Those clocks are optional.
@@ -43,7 +44,8 @@ corresponding to each DU output.
 -----------------------------------------------------------------------------
  R8A7779 (H1)	DPAD 0		DPAD 1		-
  R8A7790 (H2)	DPAD		LVDS 0		LVDS 1
- R8A7791 (M2)	DPAD		LVDS 0		-
+ R8A7791 (M2-W)	DPAD		LVDS 0		-
+ R8A7793 (M2-N)	DPAD		LVDS 0		-
 
 
 Example: R8A7790 (R-Car H2) DU
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_drv.c b/drivers/gpu/drm/rcar-du/rcar_du_drv.c
index bebcc97..d8e1fd9 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_drv.c
+++ b/drivers/gpu/drm/rcar-du/rcar_du_drv.c
@@ -84,12 +84,13 @@ static const struct rcar_du_device_info rcar_du_r8a7790_info = {
 	.num_lvds = 2,
 };
 
+/* M2-W (r8a7791) and M2-N (r8a7793) are identical */
 static const struct rcar_du_device_info rcar_du_r8a7791_info = {
 	.features = RCAR_DU_FEATURE_CRTC_IRQ_CLOCK
 		  | RCAR_DU_FEATURE_EXT_CTRL_REGS,
 	.num_crtcs = 2,
 	.routes = {
-		/* R8A7791 has one RGB output, one LVDS output and one
+		/* R8A779[13] has one RGB output, one LVDS output and one
 		 * (currently unsupported) TCON output.
 		 */
 		[RCAR_DU_OUTPUT_DPAD0] = {
@@ -110,6 +111,7 @@ static const struct of_device_id rcar_du_of_table[] = {
 	{ .compatible = "renesas,du-r8a7779", .data = &rcar_du_r8a7779_info },
 	{ .compatible = "renesas,du-r8a7790", .data = &rcar_du_r8a7790_info },
 	{ .compatible = "renesas,du-r8a7791", .data = &rcar_du_r8a7791_info },
+	{ .compatible = "renesas,du-r8a7793", .data = &rcar_du_r8a7791_info },
 	{ }
 };
 
-- 
cgit v0.10.2


From 090425c45db3ec438c161f6a5dd05d1824e65dcd Mon Sep 17 00:00:00 2001
From: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Date: Fri, 17 Jul 2015 10:44:33 +0300
Subject: drm: rcar-du: Add support for the R8A7794 DU

The R8A7794 DU has a fixed output routing configuration with one RGB
output per CRTC and thus lacks the RGB output routing register field.

Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>

diff --git a/Documentation/devicetree/bindings/video/renesas,du.txt b/Documentation/devicetree/bindings/video/renesas,du.txt
index d05be12..eccd4f4 100644
--- a/Documentation/devicetree/bindings/video/renesas,du.txt
+++ b/Documentation/devicetree/bindings/video/renesas,du.txt
@@ -7,6 +7,7 @@ Required Properties:
     - "renesas,du-r8a7790" for R8A7790 (R-Car H2) compatible DU
     - "renesas,du-r8a7791" for R8A7791 (R-Car M2-W) compatible DU
     - "renesas,du-r8a7793" for R8A7793 (R-Car M2-N) compatible DU
+    - "renesas,du-r8a7794" for R8A7794 (R-Car E2) compatible DU
 
   - reg: A list of base address and length of each memory resource, one for
     each entry in the reg-names property.
@@ -23,7 +24,7 @@ Required Properties:
   - clock-names: Name of the clocks. This property is model-dependent.
     - R8A7779 uses a single functional clock. The clock doesn't need to be
       named.
-    - R8A779[013] use one functional clock per channel and one clock per LVDS
+    - R8A779[0134] use one functional clock per channel and one clock per LVDS
       encoder (if available). The functional clocks must be named "du.x" with
       "x" being the channel numerical index. The LVDS clocks must be named
       "lvds.x" with "x" being the LVDS encoder numerical index.
@@ -46,6 +47,7 @@ corresponding to each DU output.
  R8A7790 (H2)	DPAD		LVDS 0		LVDS 1
  R8A7791 (M2-W)	DPAD		LVDS 0		-
  R8A7793 (M2-N)	DPAD		LVDS 0		-
+ R8A7794 (E2)	DPAD 0		DPAD 1		-
 
 
 Example: R8A7790 (R-Car H2) DU
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_drv.c b/drivers/gpu/drm/rcar-du/rcar_du_drv.c
index d8e1fd9..40422f6 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_drv.c
+++ b/drivers/gpu/drm/rcar-du/rcar_du_drv.c
@@ -107,11 +107,34 @@ static const struct rcar_du_device_info rcar_du_r8a7791_info = {
 	.num_lvds = 1,
 };
 
+static const struct rcar_du_device_info rcar_du_r8a7794_info = {
+	.features = RCAR_DU_FEATURE_CRTC_IRQ_CLOCK
+		  | RCAR_DU_FEATURE_EXT_CTRL_REGS,
+	.num_crtcs = 2,
+	.routes = {
+		/* R8A7794 has two RGB outputs and one (currently unsupported)
+		 * TCON output.
+		 */
+		[RCAR_DU_OUTPUT_DPAD0] = {
+			.possible_crtcs = BIT(0),
+			.encoder_type = DRM_MODE_ENCODER_NONE,
+			.port = 0,
+		},
+		[RCAR_DU_OUTPUT_DPAD1] = {
+			.possible_crtcs = BIT(1),
+			.encoder_type = DRM_MODE_ENCODER_NONE,
+			.port = 1,
+		},
+	},
+	.num_lvds = 0,
+};
+
 static const struct of_device_id rcar_du_of_table[] = {
 	{ .compatible = "renesas,du-r8a7779", .data = &rcar_du_r8a7779_info },
 	{ .compatible = "renesas,du-r8a7790", .data = &rcar_du_r8a7790_info },
 	{ .compatible = "renesas,du-r8a7791", .data = &rcar_du_r8a7791_info },
 	{ .compatible = "renesas,du-r8a7793", .data = &rcar_du_r8a7791_info },
+	{ .compatible = "renesas,du-r8a7794", .data = &rcar_du_r8a7794_info },
 	{ }
 };
 
diff --git a/drivers/gpu/drm/rcar-du/rcar_du_group.c b/drivers/gpu/drm/rcar-du/rcar_du_group.c
index 7fd39a7..8e2ffe0 100644
--- a/drivers/gpu/drm/rcar-du/rcar_du_group.c
+++ b/drivers/gpu/drm/rcar-du/rcar_du_group.c
@@ -49,9 +49,10 @@ static void rcar_du_group_setup_defr8(struct rcar_du_group *rgrp)
 	u32 defr8 = DEFR8_CODE | DEFR8_DEFE8;
 
 	/* The DEFR8 register for the first group also controls RGB output
-	 * routing to DPAD0
+	 * routing to DPAD0 for DU instances that support it.
 	 */
-	if (rgrp->index == 0)
+	if (rgrp->dev->info->routes[RCAR_DU_OUTPUT_DPAD0].possible_crtcs > 1 &&
+	    rgrp->index == 0)
 		defr8 |= DEFR8_DRGBS_DU(rgrp->dev->dpad0_source);
 
 	rcar_du_group_write(rgrp, DEFR8, defr8);
-- 
cgit v0.10.2


From 139f491799c303c5269cf6d5600cec9bbf9853e6 Mon Sep 17 00:00:00 2001
From: Samuel Li <samuel.li@amd.com>
Date: Thu, 8 Oct 2015 14:50:27 -0400
Subject: drm/amdgpu: add Stoney chip family

Stoney is based on Carrizo with some IP upgrades.

Signed-off-by: Samuel Li <samuel.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 19e185e..645690b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -57,6 +57,7 @@ static const char *amdgpu_asic_name[] = {
 	"TONGA",
 	"FIJI",
 	"CARRIZO",
+	"STONEY",
 	"LAST",
 };
 
diff --git a/drivers/gpu/drm/amd/include/amd_shared.h b/drivers/gpu/drm/amd/include/amd_shared.h
index 68a8eaa..fe28fb3 100644
--- a/drivers/gpu/drm/amd/include/amd_shared.h
+++ b/drivers/gpu/drm/amd/include/amd_shared.h
@@ -47,6 +47,7 @@ enum amd_asic_type {
 	CHIP_TONGA,
 	CHIP_FIJI,
 	CHIP_CARRIZO,
+	CHIP_STONEY,
 	CHIP_LAST,
 };
 
-- 
cgit v0.10.2


From aade2f04f94aad4ed68a7a77893b300ef9cf25da Mon Sep 17 00:00:00 2001
From: Samuel Li <samuel.li@amd.com>
Date: Thu, 8 Oct 2015 16:26:41 -0400
Subject: drm/amdgpu: add GMC support for Stoney

Stoney is GMC 8.x.

Signed-off-by: Samuel Li <samuel.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
index 42b5ff8..72e977b 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
@@ -93,6 +93,12 @@ static const u32 cz_mgcg_cgcg_init[] =
 	mmMC_MEM_POWER_LS, 0xffffffff, 0x00000104
 };
 
+static const u32 stoney_mgcg_cgcg_init[] =
+{
+	mmMC_MEM_POWER_LS, 0xffffffff, 0x00000104
+};
+
+
 static void gmc_v8_0_init_golden_registers(struct amdgpu_device *adev)
 {
 	switch (adev->asic_type) {
@@ -125,6 +131,11 @@ static void gmc_v8_0_init_golden_registers(struct amdgpu_device *adev)
 						 cz_mgcg_cgcg_init,
 						 (const u32)ARRAY_SIZE(cz_mgcg_cgcg_init));
 		break;
+	case CHIP_STONEY:
+		amdgpu_program_register_sequence(adev,
+						 stoney_mgcg_cgcg_init,
+						 (const u32)ARRAY_SIZE(stoney_mgcg_cgcg_init));
+		break;
 	default:
 		break;
 	}
@@ -228,6 +239,7 @@ static int gmc_v8_0_init_microcode(struct amdgpu_device *adev)
 		chip_name = "fiji";
 		break;
 	case CHIP_CARRIZO:
+	case CHIP_STONEY:
 		return 0;
 	default: BUG();
 	}
-- 
cgit v0.10.2


From 7a753c3f3474458fa1d0b8f41617e41995bc4c40 Mon Sep 17 00:00:00 2001
From: Samuel Li <samuel.li@amd.com>
Date: Thu, 8 Oct 2015 16:28:41 -0400
Subject: drm/amdgpu: Update SMC/DPM for Stoney

Stoney is SMC 8.x.

Signed-off-by: Samuel Li <samuel.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/cz_dpm.c b/drivers/gpu/drm/amd/amdgpu/cz_dpm.c
index 44fa96a..6f244e9 100644
--- a/drivers/gpu/drm/amd/amdgpu/cz_dpm.c
+++ b/drivers/gpu/drm/amd/amdgpu/cz_dpm.c
@@ -1262,6 +1262,7 @@ static void cz_apply_state_adjust_rules(struct amdgpu_device *adev,
 
 static int cz_dpm_enable(struct amdgpu_device *adev)
 {
+	const char *chip_name;
 	int ret = 0;
 
 	/* renable will hang up SMU, so check first */
@@ -1270,21 +1271,33 @@ static int cz_dpm_enable(struct amdgpu_device *adev)
 
 	cz_program_voting_clients(adev);
 
+	switch (adev->asic_type) {
+	case CHIP_CARRIZO:
+		chip_name = "carrizo";
+		break;
+	case CHIP_STONEY:
+		chip_name = "stoney";
+		break;
+	default:
+		BUG();
+	}
+
+
 	ret = cz_start_dpm(adev);
 	if (ret) {
-		DRM_ERROR("Carrizo DPM enable failed\n");
+		DRM_ERROR("%s DPM enable failed\n", chip_name);
 		return -EINVAL;
 	}
 
 	ret = cz_program_bootup_state(adev);
 	if (ret) {
-		DRM_ERROR("Carrizo bootup state program failed\n");
+		DRM_ERROR("%s bootup state program failed\n", chip_name);
 		return -EINVAL;
 	}
 
 	ret = cz_enable_didt(adev, true);
 	if (ret) {
-		DRM_ERROR("Carrizo enable di/dt failed\n");
+		DRM_ERROR("%s enable di/dt failed\n", chip_name);
 		return -EINVAL;
 	}
 
@@ -1351,7 +1364,7 @@ static int cz_dpm_disable(struct amdgpu_device *adev)
 
 	ret = cz_enable_didt(adev, false);
 	if (ret) {
-		DRM_ERROR("Carrizo disable di/dt failed\n");
+		DRM_ERROR("disable di/dt failed\n");
 		return -EINVAL;
 	}
 
diff --git a/drivers/gpu/drm/amd/amdgpu/cz_smc.c b/drivers/gpu/drm/amd/amdgpu/cz_smc.c
index e33180d..ac7fee7 100644
--- a/drivers/gpu/drm/amd/amdgpu/cz_smc.c
+++ b/drivers/gpu/drm/amd/amdgpu/cz_smc.c
@@ -312,13 +312,16 @@ int cz_smu_start(struct amdgpu_device *adev)
 				UCODE_ID_CP_MEC_JT1_MASK |
 				UCODE_ID_CP_MEC_JT2_MASK;
 
+	if (adev->asic_type == CHIP_STONEY)
+		fw_to_check &= ~(UCODE_ID_SDMA1_MASK | UCODE_ID_CP_MEC_JT2_MASK);
+
 	cz_smu_request_load_fw(adev);
 	ret = cz_smu_check_fw_load_finish(adev, fw_to_check);
 	if (ret)
 		return ret;
 
 	/* manually load MEC firmware for CZ */
-	if (adev->asic_type == CHIP_CARRIZO) {
+	if (adev->asic_type == CHIP_CARRIZO || adev->asic_type == CHIP_STONEY) {
 		ret = cz_load_mec_firmware(adev);
 		if (ret) {
 			dev_err(adev->dev, "(%d) Mec Firmware load failed\n", ret);
@@ -336,6 +339,9 @@ int cz_smu_start(struct amdgpu_device *adev)
 				AMDGPU_CPMEC2_UCODE_LOADED |
 				AMDGPU_CPRLC_UCODE_LOADED;
 
+	if (adev->asic_type == CHIP_STONEY)
+		adev->smu.fw_flags &= ~(AMDGPU_SDMA1_UCODE_LOADED | AMDGPU_CPMEC2_UCODE_LOADED);
+
 	return ret;
 }
 
@@ -601,8 +607,13 @@ static int cz_smu_construct_toc_for_vddgfx_exit(struct amdgpu_device *adev)
 				CZ_SCRATCH_ENTRY_UCODE_ID_CP_ME, false);
 		cz_smu_populate_single_ucode_load_task(adev,
 				CZ_SCRATCH_ENTRY_UCODE_ID_CP_MEC_JT1, false);
-		cz_smu_populate_single_ucode_load_task(adev,
+		if (adev->asic_type == CHIP_STONEY) {
+			cz_smu_populate_single_ucode_load_task(adev,
+				CZ_SCRATCH_ENTRY_UCODE_ID_CP_MEC_JT1, false);
+		} else {
+			cz_smu_populate_single_ucode_load_task(adev,
 				CZ_SCRATCH_ENTRY_UCODE_ID_CP_MEC_JT2, false);
+		}
 		cz_smu_populate_single_ucode_load_task(adev,
 				CZ_SCRATCH_ENTRY_UCODE_ID_RLC_G, false);
 	}
@@ -642,8 +653,13 @@ static int cz_smu_construct_toc_for_bootup(struct amdgpu_device *adev)
 	if (adev->firmware.smu_load) {
 		cz_smu_populate_single_ucode_load_task(adev,
 				CZ_SCRATCH_ENTRY_UCODE_ID_SDMA0, false);
-		cz_smu_populate_single_ucode_load_task(adev,
+		if (adev->asic_type == CHIP_STONEY) {
+			cz_smu_populate_single_ucode_load_task(adev,
+				CZ_SCRATCH_ENTRY_UCODE_ID_SDMA0, false);
+		} else {
+			cz_smu_populate_single_ucode_load_task(adev,
 				CZ_SCRATCH_ENTRY_UCODE_ID_SDMA1, false);
+		}
 		cz_smu_populate_single_ucode_load_task(adev,
 				CZ_SCRATCH_ENTRY_UCODE_ID_CP_CE, false);
 		cz_smu_populate_single_ucode_load_task(adev,
@@ -652,8 +668,13 @@ static int cz_smu_construct_toc_for_bootup(struct amdgpu_device *adev)
 				CZ_SCRATCH_ENTRY_UCODE_ID_CP_ME, false);
 		cz_smu_populate_single_ucode_load_task(adev,
 				CZ_SCRATCH_ENTRY_UCODE_ID_CP_MEC_JT1, false);
-		cz_smu_populate_single_ucode_load_task(adev,
+		if (adev->asic_type == CHIP_STONEY) {
+			cz_smu_populate_single_ucode_load_task(adev,
+				CZ_SCRATCH_ENTRY_UCODE_ID_CP_MEC_JT1, false);
+		} else {
+			cz_smu_populate_single_ucode_load_task(adev,
 				CZ_SCRATCH_ENTRY_UCODE_ID_CP_MEC_JT2, false);
+		}
 		cz_smu_populate_single_ucode_load_task(adev,
 				CZ_SCRATCH_ENTRY_UCODE_ID_RLC_G, true);
 	}
@@ -888,10 +909,18 @@ int cz_smu_init(struct amdgpu_device *adev)
 				CZ_SCRATCH_ENTRY_UCODE_ID_SDMA0,
 				&priv->driver_buffer[priv->driver_buffer_length++]))
 			goto smu_init_failed;
-		if (cz_smu_populate_single_firmware_entry(adev,
-				CZ_SCRATCH_ENTRY_UCODE_ID_SDMA1,
-				&priv->driver_buffer[priv->driver_buffer_length++]))
-			goto smu_init_failed;
+
+		if (adev->asic_type == CHIP_STONEY) {
+			if (cz_smu_populate_single_firmware_entry(adev,
+					CZ_SCRATCH_ENTRY_UCODE_ID_SDMA0,
+					&priv->driver_buffer[priv->driver_buffer_length++]))
+				goto smu_init_failed;
+		} else {
+			if (cz_smu_populate_single_firmware_entry(adev,
+					CZ_SCRATCH_ENTRY_UCODE_ID_SDMA1,
+					&priv->driver_buffer[priv->driver_buffer_length++]))
+				goto smu_init_failed;
+		}
 		if (cz_smu_populate_single_firmware_entry(adev,
 				CZ_SCRATCH_ENTRY_UCODE_ID_CP_CE,
 				&priv->driver_buffer[priv->driver_buffer_length++]))
@@ -908,10 +937,17 @@ int cz_smu_init(struct amdgpu_device *adev)
 				CZ_SCRATCH_ENTRY_UCODE_ID_CP_MEC_JT1,
 				&priv->driver_buffer[priv->driver_buffer_length++]))
 			goto smu_init_failed;
-		if (cz_smu_populate_single_firmware_entry(adev,
-				CZ_SCRATCH_ENTRY_UCODE_ID_CP_MEC_JT2,
-				&priv->driver_buffer[priv->driver_buffer_length++]))
-			goto smu_init_failed;
+		if (adev->asic_type == CHIP_STONEY) {
+			if (cz_smu_populate_single_firmware_entry(adev,
+					CZ_SCRATCH_ENTRY_UCODE_ID_CP_MEC_JT1,
+					&priv->driver_buffer[priv->driver_buffer_length++]))
+				goto smu_init_failed;
+		} else {
+			if (cz_smu_populate_single_firmware_entry(adev,
+					CZ_SCRATCH_ENTRY_UCODE_ID_CP_MEC_JT2,
+					&priv->driver_buffer[priv->driver_buffer_length++]))
+				goto smu_init_failed;
+		}
 		if (cz_smu_populate_single_firmware_entry(adev,
 				CZ_SCRATCH_ENTRY_UCODE_ID_RLC_G,
 				&priv->driver_buffer[priv->driver_buffer_length++]))
-- 
cgit v0.10.2


From fa2f9befbbbe239bbc6aa1dc6284841a8bd15d49 Mon Sep 17 00:00:00 2001
From: Samuel Li <samuel.li@amd.com>
Date: Thu, 8 Oct 2015 16:29:06 -0400
Subject: drm/amdgpu: add DCE support for Stoney

Stoney is DCE 11.x.

Signed-off-by: Samuel Li <samuel.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
index bdafeb2..444d930 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
@@ -126,6 +126,13 @@ static const u32 cz_mgcg_cgcg_init[] =
 	mmXDMA_MEM_POWER_CNTL, 0x00000101, 0x00000000,
 };
 
+static const u32 stoney_golden_settings_a11[] =
+{
+	mmCRTC_DOUBLE_BUFFER_CONTROL, 0x00010101, 0x00010000,
+	mmFBC_MISC, 0x1f311fff, 0x14302000,
+};
+
+
 static void dce_v11_0_init_golden_registers(struct amdgpu_device *adev)
 {
 	switch (adev->asic_type) {
@@ -137,6 +144,11 @@ static void dce_v11_0_init_golden_registers(struct amdgpu_device *adev)
 						 cz_golden_settings_a11,
 						 (const u32)ARRAY_SIZE(cz_golden_settings_a11));
 		break;
+	case CHIP_STONEY:
+		amdgpu_program_register_sequence(adev,
+						 stoney_golden_settings_a11,
+						 (const u32)ARRAY_SIZE(stoney_golden_settings_a11));
+		break;
 	default:
 		break;
 	}
@@ -2425,7 +2437,7 @@ static u32 dce_v11_0_pick_pll(struct drm_crtc *crtc)
 
 	/* XXX need to determine what plls are available on each DCE11 part */
 	pll_in_use = amdgpu_pll_get_use_mask(crtc);
-	if (adev->asic_type == CHIP_CARRIZO) {
+	if (adev->asic_type == CHIP_CARRIZO || adev->asic_type == CHIP_STONEY) {
 		if (!(pll_in_use & (1 << ATOM_PPLL1)))
 			return ATOM_PPLL1;
 		if (!(pll_in_use & (1 << ATOM_PPLL0)))
@@ -2930,6 +2942,11 @@ static int dce_v11_0_early_init(void *handle)
 		adev->mode_info.num_hpd = 6;
 		adev->mode_info.num_dig = 9;
 		break;
+	case CHIP_STONEY:
+		adev->mode_info.num_crtc = 2;
+		adev->mode_info.num_hpd = 6;
+		adev->mode_info.num_dig = 9;
+		break;
 	default:
 		/* FIXME: not supported yet */
 		return -EINVAL;
-- 
cgit v0.10.2


From bb16e3b6c8641bdf4765c373ba41543861959cbb Mon Sep 17 00:00:00 2001
From: Samuel Li <samuel.li@amd.com>
Date: Thu, 8 Oct 2015 17:17:51 -0400
Subject: drm/amdgpu: add SDMA support for Stoney (v2)

Stoney is SDMA 3.x.

v2: update to latest golden register settings

Signed-off-by: Samuel Li <samuel.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
index 0271bf9..7253132 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
@@ -55,6 +55,7 @@ MODULE_FIRMWARE("amdgpu/carrizo_sdma.bin");
 MODULE_FIRMWARE("amdgpu/carrizo_sdma1.bin");
 MODULE_FIRMWARE("amdgpu/fiji_sdma.bin");
 MODULE_FIRMWARE("amdgpu/fiji_sdma1.bin");
+MODULE_FIRMWARE("amdgpu/stoney_sdma.bin");
 
 static const u32 sdma_offsets[SDMA_MAX_INSTANCE] =
 {
@@ -122,6 +123,19 @@ static const u32 cz_mgcg_cgcg_init[] =
 	mmSDMA1_CLK_CTRL, 0xff000ff0, 0x00000100
 };
 
+static const u32 stoney_golden_settings_a11[] =
+{
+	mmSDMA0_GFX_IB_CNTL, 0x00000100, 0x00000100,
+	mmSDMA0_POWER_CNTL, 0x00000800, 0x0003c800,
+	mmSDMA0_RLC0_IB_CNTL, 0x00000100, 0x00000100,
+	mmSDMA0_RLC1_IB_CNTL, 0x00000100, 0x00000100,
+};
+
+static const u32 stoney_mgcg_cgcg_init[] =
+{
+	mmSDMA0_CLK_CTRL, 0xffffffff, 0x00000100,
+};
+
 /*
  * sDMA - System DMA
  * Starting with CIK, the GPU has new asynchronous
@@ -166,6 +180,14 @@ static void sdma_v3_0_init_golden_registers(struct amdgpu_device *adev)
 						 cz_golden_settings_a11,
 						 (const u32)ARRAY_SIZE(cz_golden_settings_a11));
 		break;
+	case CHIP_STONEY:
+		amdgpu_program_register_sequence(adev,
+						 stoney_mgcg_cgcg_init,
+						 (const u32)ARRAY_SIZE(stoney_mgcg_cgcg_init));
+		amdgpu_program_register_sequence(adev,
+						 stoney_golden_settings_a11,
+						 (const u32)ARRAY_SIZE(stoney_golden_settings_a11));
+		break;
 	default:
 		break;
 	}
@@ -201,6 +223,9 @@ static int sdma_v3_0_init_microcode(struct amdgpu_device *adev)
 	case CHIP_CARRIZO:
 		chip_name = "carrizo";
 		break;
+	case CHIP_STONEY:
+		chip_name = "stoney";
+		break;
 	default: BUG();
 	}
 
@@ -1071,6 +1096,9 @@ static int sdma_v3_0_early_init(void *handle)
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
 	switch (adev->asic_type) {
+	case CHIP_STONEY:
+		adev->sdma.num_instances = 1;
+		break;
 	default:
 		adev->sdma.num_instances = SDMA_MAX_INSTANCE;
 		break;
-- 
cgit v0.10.2


From e3c7656c22697eeef46ce043e23417241844ab1c Mon Sep 17 00:00:00 2001
From: Samuel Li <samuel.li@amd.com>
Date: Thu, 8 Oct 2015 16:29:40 -0400
Subject: drm/amdgpu: add GFX support for Stoney (v2)

Stoney is GFX 8.1.

v2: update to latest golden settings

Signed-off-by: Samuel Li <samuel.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index e0355ea..cbc46a3 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -73,6 +73,12 @@ MODULE_FIRMWARE("amdgpu/carrizo_mec.bin");
 MODULE_FIRMWARE("amdgpu/carrizo_mec2.bin");
 MODULE_FIRMWARE("amdgpu/carrizo_rlc.bin");
 
+MODULE_FIRMWARE("amdgpu/stoney_ce.bin");
+MODULE_FIRMWARE("amdgpu/stoney_pfp.bin");
+MODULE_FIRMWARE("amdgpu/stoney_me.bin");
+MODULE_FIRMWARE("amdgpu/stoney_mec.bin");
+MODULE_FIRMWARE("amdgpu/stoney_rlc.bin");
+
 MODULE_FIRMWARE("amdgpu/tonga_ce.bin");
 MODULE_FIRMWARE("amdgpu/tonga_pfp.bin");
 MODULE_FIRMWARE("amdgpu/tonga_me.bin");
@@ -493,6 +499,42 @@ static const u32 cz_mgcg_cgcg_init[] =
 	mmCP_MEM_SLP_CNTL, 0x00000001, 0x00000001,
 };
 
+static const u32 stoney_golden_settings_a11[] =
+{
+	mmDB_DEBUG2, 0xf00fffff, 0x00000400,
+	mmGB_GPU_ID, 0x0000000f, 0x00000000,
+	mmPA_SC_ENHANCE, 0xffffffff, 0x20000001,
+	mmPA_SC_LINE_STIPPLE_STATE, 0x0000ff0f, 0x00000000,
+	mmRLC_CGCG_CGLS_CTRL, 0x00000003, 0x0001003c,
+	mmTA_CNTL_AUX, 0x000f000f, 0x000b0000,
+  	mmTCC_CTRL, 0x00100000, 0xf31fff7f,
+	mmTCC_EXE_DISABLE, 0x00000002, 0x00000002,
+	mmTCP_ADDR_CONFIG, 0x0000000f, 0x000000f1,
+	mmTCP_CHAN_STEER_LO, 0xffffffff, 0x10101010,
+};
+
+static const u32 stoney_golden_common_all[] =
+{
+	mmGRBM_GFX_INDEX, 0xffffffff, 0xe0000000,
+	mmPA_SC_RASTER_CONFIG, 0xffffffff, 0x00000000,
+	mmPA_SC_RASTER_CONFIG_1, 0xffffffff, 0x00000000,
+	mmGB_ADDR_CONFIG, 0xffffffff, 0x12010001,
+	mmSPI_RESOURCE_RESERVE_CU_0, 0xffffffff, 0x00000800,
+	mmSPI_RESOURCE_RESERVE_CU_1, 0xffffffff, 0x00000800,
+	mmSPI_RESOURCE_RESERVE_EN_CU_0, 0xffffffff, 0x00007FBF,
+	mmSPI_RESOURCE_RESERVE_EN_CU_1, 0xffffffff, 0x00007FAF,
+};
+
+static const u32 stoney_mgcg_cgcg_init[] =
+{
+	mmGRBM_GFX_INDEX, 0xffffffff, 0xe0000000,
+	mmRLC_CGCG_CGLS_CTRL, 0xffffffff, 0x0020003f,
+	mmCP_MEM_SLP_CNTL, 0xffffffff, 0x00020201,
+	mmRLC_MEM_SLP_CNTL, 0xffffffff, 0x00020201,
+	mmCGTS_SM_CTRL_REG, 0xffffffff, 0x96940200,
+	mmATC_MISC_CG, 0xffffffff, 0x000c0200,
+};
+
 static void gfx_v8_0_set_ring_funcs(struct amdgpu_device *adev);
 static void gfx_v8_0_set_irq_funcs(struct amdgpu_device *adev);
 static void gfx_v8_0_set_gds_init(struct amdgpu_device *adev);
@@ -545,6 +587,17 @@ static void gfx_v8_0_init_golden_registers(struct amdgpu_device *adev)
 						 cz_golden_common_all,
 						 (const u32)ARRAY_SIZE(cz_golden_common_all));
 		break;
+	case CHIP_STONEY:
+		amdgpu_program_register_sequence(adev,
+						 stoney_mgcg_cgcg_init,
+						 (const u32)ARRAY_SIZE(stoney_mgcg_cgcg_init));
+		amdgpu_program_register_sequence(adev,
+						 stoney_golden_settings_a11,
+						 (const u32)ARRAY_SIZE(stoney_golden_settings_a11));
+		amdgpu_program_register_sequence(adev,
+						 stoney_golden_common_all,
+						 (const u32)ARRAY_SIZE(stoney_golden_common_all));
+		break;
 	default:
 		break;
 	}
@@ -691,6 +744,9 @@ static int gfx_v8_0_init_microcode(struct amdgpu_device *adev)
 	case CHIP_FIJI:
 		chip_name = "fiji";
 		break;
+	case CHIP_STONEY:
+		chip_name = "stoney";
+		break;
 	default:
 		BUG();
 	}
@@ -748,21 +804,23 @@ static int gfx_v8_0_init_microcode(struct amdgpu_device *adev)
 	adev->gfx.mec_fw_version = le32_to_cpu(cp_hdr->header.ucode_version);
 	adev->gfx.mec_feature_version = le32_to_cpu(cp_hdr->ucode_feature_version);
 
-	snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_mec2.bin", chip_name);
-	err = request_firmware(&adev->gfx.mec2_fw, fw_name, adev->dev);
-	if (!err) {
-		err = amdgpu_ucode_validate(adev->gfx.mec2_fw);
-		if (err)
-			goto out;
-		cp_hdr = (const struct gfx_firmware_header_v1_0 *)
-						adev->gfx.mec2_fw->data;
-		adev->gfx.mec2_fw_version = le32_to_cpu(
-						cp_hdr->header.ucode_version);
-		adev->gfx.mec2_feature_version = le32_to_cpu(
-						cp_hdr->ucode_feature_version);
-	} else {
-		err = 0;
-		adev->gfx.mec2_fw = NULL;
+	if (adev->asic_type != CHIP_STONEY) {
+		snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_mec2.bin", chip_name);
+		err = request_firmware(&adev->gfx.mec2_fw, fw_name, adev->dev);
+		if (!err) {
+			err = amdgpu_ucode_validate(adev->gfx.mec2_fw);
+			if (err)
+				goto out;
+			cp_hdr = (const struct gfx_firmware_header_v1_0 *)
+				adev->gfx.mec2_fw->data;
+			adev->gfx.mec2_fw_version =
+				le32_to_cpu(cp_hdr->header.ucode_version);
+			adev->gfx.mec2_feature_version =
+				le32_to_cpu(cp_hdr->ucode_feature_version);
+		} else {
+			err = 0;
+			adev->gfx.mec2_fw = NULL;
+		}
 	}
 
 	if (adev->firmware.smu_load) {
@@ -1010,6 +1068,40 @@ static void gfx_v8_0_gpu_early_init(struct amdgpu_device *adev)
 		adev->gfx.config.sc_earlyz_tile_fifo_size = 0x130;
 		gb_addr_config = CARRIZO_GB_ADDR_CONFIG_GOLDEN;
 		break;
+	case CHIP_STONEY:
+		adev->gfx.config.max_shader_engines = 1;
+		adev->gfx.config.max_tile_pipes = 2;
+		adev->gfx.config.max_sh_per_se = 1;
+		adev->gfx.config.max_backends_per_se = 1;
+
+		switch (adev->pdev->revision) {
+		case 0xc0:
+		case 0xc1:
+		case 0xc2:
+		case 0xc4:
+		case 0xc8:
+		case 0xc9:
+			adev->gfx.config.max_cu_per_sh = 3;
+			break;
+		case 0xd0:
+		case 0xd1:
+		case 0xd2:
+		default:
+			adev->gfx.config.max_cu_per_sh = 2;
+			break;
+		}
+
+		adev->gfx.config.max_texture_channel_caches = 2;
+		adev->gfx.config.max_gprs = 256;
+		adev->gfx.config.max_gs_threads = 16;
+		adev->gfx.config.max_hw_contexts = 8;
+
+		adev->gfx.config.sc_prim_fifo_size_frontend = 0x20;
+		adev->gfx.config.sc_prim_fifo_size_backend = 0x100;
+		adev->gfx.config.sc_hiz_tile_fifo_size = 0x30;
+		adev->gfx.config.sc_earlyz_tile_fifo_size = 0x130;
+		gb_addr_config = CARRIZO_GB_ADDR_CONFIG_GOLDEN;
+		break;
 	default:
 		adev->gfx.config.max_shader_engines = 2;
 		adev->gfx.config.max_tile_pipes = 4;
@@ -1797,6 +1889,273 @@ static void gfx_v8_0_tiling_mode_table_init(struct amdgpu_device *adev)
 			WREG32(mmGB_MACROTILE_MODE0 + reg_offset, gb_tile_moden);
 		}
 		break;
+	case CHIP_STONEY:
+		for (reg_offset = 0; reg_offset < num_tile_mode_states; reg_offset++) {
+			switch (reg_offset) {
+			case 0:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+						PIPE_CONFIG(ADDR_SURF_P2) |
+						TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B) |
+						MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING));
+				break;
+			case 1:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+						PIPE_CONFIG(ADDR_SURF_P2) |
+						TILE_SPLIT(ADDR_SURF_TILE_SPLIT_128B) |
+						MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING));
+				break;
+			case 2:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+						PIPE_CONFIG(ADDR_SURF_P2) |
+						TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
+						MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING));
+				break;
+			case 3:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+						PIPE_CONFIG(ADDR_SURF_P2) |
+						TILE_SPLIT(ADDR_SURF_TILE_SPLIT_512B) |
+						MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING));
+				break;
+			case 4:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+						PIPE_CONFIG(ADDR_SURF_P2) |
+						TILE_SPLIT(ADDR_SURF_TILE_SPLIT_2KB) |
+						MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING));
+				break;
+			case 5:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+						PIPE_CONFIG(ADDR_SURF_P2) |
+						TILE_SPLIT(ADDR_SURF_TILE_SPLIT_2KB) |
+						MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING));
+				break;
+			case 6:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
+						PIPE_CONFIG(ADDR_SURF_P2) |
+						TILE_SPLIT(ADDR_SURF_TILE_SPLIT_2KB) |
+						MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING));
+				break;
+			case 8:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_LINEAR_ALIGNED) |
+						PIPE_CONFIG(ADDR_SURF_P2));
+				break;
+			case 9:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+						PIPE_CONFIG(ADDR_SURF_P2) |
+						MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
+						SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+				break;
+			case 10:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+						PIPE_CONFIG(ADDR_SURF_P2) |
+						MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
+						SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+				break;
+			case 11:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
+						PIPE_CONFIG(ADDR_SURF_P2) |
+						MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
+						SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_8));
+				break;
+			case 13:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+						PIPE_CONFIG(ADDR_SURF_P2) |
+						MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
+						SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+				break;
+			case 14:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+						PIPE_CONFIG(ADDR_SURF_P2) |
+						MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
+						SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+				break;
+			case 15:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_3D_TILED_THIN1) |
+						PIPE_CONFIG(ADDR_SURF_P2) |
+						MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
+						SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+				break;
+			case 16:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
+						PIPE_CONFIG(ADDR_SURF_P2) |
+						MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
+						SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_8));
+				break;
+			case 18:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THICK) |
+						PIPE_CONFIG(ADDR_SURF_P2) |
+						MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
+						SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_1));
+				break;
+			case 19:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THICK) |
+						PIPE_CONFIG(ADDR_SURF_P2) |
+						MICRO_TILE_MODE_NEW(ADDR_SURF_THICK_MICRO_TILING) |
+						SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_1));
+				break;
+			case 20:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THICK) |
+						PIPE_CONFIG(ADDR_SURF_P2) |
+						MICRO_TILE_MODE_NEW(ADDR_SURF_THICK_MICRO_TILING) |
+						SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_1));
+				break;
+			case 21:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_3D_TILED_THICK) |
+						PIPE_CONFIG(ADDR_SURF_P2) |
+						MICRO_TILE_MODE_NEW(ADDR_SURF_THICK_MICRO_TILING) |
+						SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_1));
+				break;
+			case 22:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_TILED_THICK) |
+						PIPE_CONFIG(ADDR_SURF_P2) |
+						MICRO_TILE_MODE_NEW(ADDR_SURF_THICK_MICRO_TILING) |
+						SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_1));
+				break;
+			case 24:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THICK) |
+						PIPE_CONFIG(ADDR_SURF_P2) |
+						MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
+						SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_1));
+				break;
+			case 25:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_XTHICK) |
+						PIPE_CONFIG(ADDR_SURF_P2) |
+						MICRO_TILE_MODE_NEW(ADDR_SURF_THICK_MICRO_TILING) |
+						SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_1));
+				break;
+			case 26:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_3D_TILED_XTHICK) |
+						PIPE_CONFIG(ADDR_SURF_P2) |
+						MICRO_TILE_MODE_NEW(ADDR_SURF_THICK_MICRO_TILING) |
+						SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_1));
+				break;
+			case 27:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+						PIPE_CONFIG(ADDR_SURF_P2) |
+						MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
+						SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+				break;
+			case 28:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+						PIPE_CONFIG(ADDR_SURF_P2) |
+						MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
+						SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+				break;
+			case 29:
+				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
+						PIPE_CONFIG(ADDR_SURF_P2) |
+						MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
+						SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_8));
+				break;
+			case 7:
+			case 12:
+			case 17:
+			case 23:
+				/* unused idx */
+				continue;
+			default:
+				gb_tile_moden = 0;
+				break;
+			};
+			adev->gfx.config.tile_mode_array[reg_offset] = gb_tile_moden;
+			WREG32(mmGB_TILE_MODE0 + reg_offset, gb_tile_moden);
+		}
+		for (reg_offset = 0; reg_offset < num_secondary_tile_mode_states; reg_offset++) {
+			switch (reg_offset) {
+			case 0:
+				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+						BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+						MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
+						NUM_BANKS(ADDR_SURF_8_BANK));
+				break;
+			case 1:
+				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+						BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
+						MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
+						NUM_BANKS(ADDR_SURF_8_BANK));
+				break;
+			case 2:
+				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+						BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+						MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
+						NUM_BANKS(ADDR_SURF_8_BANK));
+				break;
+			case 3:
+				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+						BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+						MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
+						NUM_BANKS(ADDR_SURF_8_BANK));
+				break;
+			case 4:
+				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+						BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+						MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
+						NUM_BANKS(ADDR_SURF_8_BANK));
+				break;
+			case 5:
+				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+						BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+						MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
+						NUM_BANKS(ADDR_SURF_8_BANK));
+				break;
+			case 6:
+				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+						BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+						MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
+						NUM_BANKS(ADDR_SURF_8_BANK));
+				break;
+			case 8:
+				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_4) |
+						BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_8) |
+						MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
+						NUM_BANKS(ADDR_SURF_16_BANK));
+				break;
+			case 9:
+				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_4) |
+						BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+						MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
+						NUM_BANKS(ADDR_SURF_16_BANK));
+				break;
+			case 10:
+				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_2) |
+						BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+						MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
+						NUM_BANKS(ADDR_SURF_16_BANK));
+				break;
+			case 11:
+				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_2) |
+						BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
+						MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
+						NUM_BANKS(ADDR_SURF_16_BANK));
+				break;
+			case 12:
+				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+						BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
+						MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
+						NUM_BANKS(ADDR_SURF_16_BANK));
+				break;
+			case 13:
+				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+						BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+						MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
+						NUM_BANKS(ADDR_SURF_16_BANK));
+				break;
+			case 14:
+				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+						BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+						MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
+						NUM_BANKS(ADDR_SURF_8_BANK));
+				break;
+			case 7:
+				/* unused idx */
+				continue;
+			default:
+				gb_tile_moden = 0;
+				break;
+			};
+			adev->gfx.config.macrotile_mode_array[reg_offset] = gb_tile_moden;
+			WREG32(mmGB_MACROTILE_MODE0 + reg_offset, gb_tile_moden);
+		}
+		break;
 	case CHIP_CARRIZO:
 	default:
 		for (reg_offset = 0; reg_offset < num_tile_mode_states; reg_offset++) {
@@ -2384,7 +2743,7 @@ static void gfx_v8_0_rlc_start(struct amdgpu_device *adev)
 	WREG32(mmRLC_CNTL, tmp);
 
 	/* carrizo do enable cp interrupt after cp inited */
-	if (adev->asic_type != CHIP_CARRIZO)
+	if (!(adev->flags & AMD_IS_APU))
 		gfx_v8_0_enable_gui_idle_interrupt(adev, true);
 
 	udelay(50);
@@ -2606,6 +2965,10 @@ static int gfx_v8_0_cp_gfx_start(struct amdgpu_device *adev)
 		amdgpu_ring_write(ring, 0x00000002);
 		amdgpu_ring_write(ring, 0x00000000);
 		break;
+	case CHIP_STONEY:
+		amdgpu_ring_write(ring, 0x00000000);
+		amdgpu_ring_write(ring, 0x00000000);
+		break;
 	default:
 		BUG();
 	}
@@ -3240,7 +3603,8 @@ static int gfx_v8_0_cp_compute_resume(struct amdgpu_device *adev)
 		/* enable the doorbell if requested */
 		if (use_doorbell) {
 			if ((adev->asic_type == CHIP_CARRIZO) ||
-			    (adev->asic_type == CHIP_FIJI)) {
+			    (adev->asic_type == CHIP_FIJI) ||
+			    (adev->asic_type == CHIP_STONEY)) {
 				WREG32(mmCP_MEC_DOORBELL_RANGE_LOWER,
 				       AMDGPU_DOORBELL_KIQ << 2);
 				WREG32(mmCP_MEC_DOORBELL_RANGE_UPPER,
@@ -3312,7 +3676,7 @@ static int gfx_v8_0_cp_resume(struct amdgpu_device *adev)
 {
 	int r;
 
-	if (adev->asic_type != CHIP_CARRIZO)
+	if (!(adev->flags & AMD_IS_APU))
 		gfx_v8_0_enable_gui_idle_interrupt(adev, false);
 
 	if (!adev->firmware.smu_load) {
-- 
cgit v0.10.2


From a39c8cea3512bdf9d002edc9142ff35d314a690e Mon Sep 17 00:00:00 2001
From: Samuel Li <samuel.li@amd.com>
Date: Thu, 8 Oct 2015 16:27:21 -0400
Subject: drm/amdgpu: add UVD support for Stoney

Stoney is UVD 6.x.

Signed-off-by: Samuel Li <samuel.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c
index d0312364..53f987a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c
@@ -53,6 +53,7 @@
 #define FIRMWARE_TONGA		"amdgpu/tonga_uvd.bin"
 #define FIRMWARE_CARRIZO	"amdgpu/carrizo_uvd.bin"
 #define FIRMWARE_FIJI		"amdgpu/fiji_uvd.bin"
+#define FIRMWARE_STONEY		"amdgpu/stoney_uvd.bin"
 
 /**
  * amdgpu_uvd_cs_ctx - Command submission parser context
@@ -83,6 +84,7 @@ MODULE_FIRMWARE(FIRMWARE_MULLINS);
 MODULE_FIRMWARE(FIRMWARE_TONGA);
 MODULE_FIRMWARE(FIRMWARE_CARRIZO);
 MODULE_FIRMWARE(FIRMWARE_FIJI);
+MODULE_FIRMWARE(FIRMWARE_STONEY);
 
 static void amdgpu_uvd_note_usage(struct amdgpu_device *adev);
 static void amdgpu_uvd_idle_work_handler(struct work_struct *work);
@@ -124,6 +126,9 @@ int amdgpu_uvd_sw_init(struct amdgpu_device *adev)
 	case CHIP_CARRIZO:
 		fw_name = FIRMWARE_CARRIZO;
 		break;
+	case CHIP_STONEY:
+		fw_name = FIRMWARE_STONEY;
+		break;
 	default:
 		return -EINVAL;
 	}
-- 
cgit v0.10.2


From cfaba566035d2f5a977b18d3287fd407bacf34bc Mon Sep 17 00:00:00 2001
From: Samuel Li <samuel.li@amd.com>
Date: Thu, 8 Oct 2015 16:27:55 -0400
Subject: drm/amdgpu: add VCE support for Stoney (v2)

Stoney is VCE 3.x single.

v2: Stoney is single pipe like Fiji

Signed-off-by: Samuel Li <samuel.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c
index 74f2038a..03f0c3b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c
@@ -49,6 +49,7 @@
 #define FIRMWARE_TONGA		"amdgpu/tonga_vce.bin"
 #define FIRMWARE_CARRIZO	"amdgpu/carrizo_vce.bin"
 #define FIRMWARE_FIJI		"amdgpu/fiji_vce.bin"
+#define FIRMWARE_STONEY		"amdgpu/stoney_vce.bin"
 
 #ifdef CONFIG_DRM_AMDGPU_CIK
 MODULE_FIRMWARE(FIRMWARE_BONAIRE);
@@ -60,6 +61,7 @@ MODULE_FIRMWARE(FIRMWARE_MULLINS);
 MODULE_FIRMWARE(FIRMWARE_TONGA);
 MODULE_FIRMWARE(FIRMWARE_CARRIZO);
 MODULE_FIRMWARE(FIRMWARE_FIJI);
+MODULE_FIRMWARE(FIRMWARE_STONEY);
 
 static void amdgpu_vce_idle_work_handler(struct work_struct *work);
 
@@ -106,6 +108,9 @@ int amdgpu_vce_sw_init(struct amdgpu_device *adev, unsigned long size)
 	case CHIP_FIJI:
 		fw_name = FIRMWARE_FIJI;
 		break;
+	case CHIP_STONEY:
+		fw_name = FIRMWARE_STONEY;
+		break;
 
 	default:
 		return -EINVAL;
diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c
index b1bf27b..6a52db6 100644
--- a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c
@@ -205,8 +205,9 @@ static unsigned vce_v3_0_get_harvest_config(struct amdgpu_device *adev)
 	u32 tmp;
 	unsigned ret;
 
-	/* Fiji is single pipe */
-	if (adev->asic_type == CHIP_FIJI) {
+	/* Fiji, Stoney are single pipe */
+	if ((adev->asic_type == CHIP_FIJI) ||
+	    (adev->asic_type == CHIP_STONEY)){
 		ret = AMDGPU_VCE_HARVEST_VCE1;
 		return ret;
 	}
-- 
cgit v0.10.2


From 39bb0c92829ad9d7525fa809aa89fc411c85a2c2 Mon Sep 17 00:00:00 2001
From: Samuel Li <samuel.li@amd.com>
Date: Thu, 8 Oct 2015 16:31:43 -0400
Subject: drm/amdgpu: update the core VI support for Stoney

Add core VI enablement for Stoney.

Signed-off-by: Samuel Li <samuel.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 645690b..d5b4213 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -1166,7 +1166,8 @@ static int amdgpu_early_init(struct amdgpu_device *adev)
 	case CHIP_TONGA:
 	case CHIP_FIJI:
 	case CHIP_CARRIZO:
-		if (adev->asic_type == CHIP_CARRIZO)
+	case CHIP_STONEY:
+		if (adev->asic_type == CHIP_CARRIZO || adev->asic_type == CHIP_STONEY)
 			adev->family = AMDGPU_FAMILY_CZ;
 		else
 			adev->family = AMDGPU_FAMILY_VI;
diff --git a/drivers/gpu/drm/amd/amdgpu/vi.c b/drivers/gpu/drm/amd/amdgpu/vi.c
index b55ceb1..9904761 100644
--- a/drivers/gpu/drm/amd/amdgpu/vi.c
+++ b/drivers/gpu/drm/amd/amdgpu/vi.c
@@ -232,6 +232,13 @@ static const u32 cz_mgcg_cgcg_init[] =
 	mmHDP_XDP_CGTT_BLK_CTRL, 0xc0000fff, 0x00000104,
 };
 
+static const u32 stoney_mgcg_cgcg_init[] =
+{
+	mmCGTT_DRM_CLK_CTRL0, 0xffffffff, 0x00000100,
+	mmHDP_XDP_CGTT_BLK_CTRL, 0xffffffff, 0x00000104,
+	mmHDP_HOST_PATH_CNTL, 0xffffffff, 0x0f000027,
+};
+
 static void vi_init_golden_registers(struct amdgpu_device *adev)
 {
 	/* Some of the registers might be dependent on GRBM_GFX_INDEX */
@@ -258,6 +265,11 @@ static void vi_init_golden_registers(struct amdgpu_device *adev)
 						 cz_mgcg_cgcg_init,
 						 (const u32)ARRAY_SIZE(cz_mgcg_cgcg_init));
 		break;
+	case CHIP_STONEY:
+		amdgpu_program_register_sequence(adev,
+						 stoney_mgcg_cgcg_init,
+						 (const u32)ARRAY_SIZE(stoney_mgcg_cgcg_init));
+		break;
 	default:
 		break;
 	}
@@ -488,6 +500,7 @@ static int vi_read_register(struct amdgpu_device *adev, u32 se_num,
 	case CHIP_FIJI:
 	case CHIP_TONGA:
 	case CHIP_CARRIZO:
+	case CHIP_STONEY:
 		asic_register_table = cz_allowed_read_registers;
 		size = ARRAY_SIZE(cz_allowed_read_registers);
 		break;
@@ -543,8 +556,10 @@ static void vi_print_gpu_status_regs(struct amdgpu_device *adev)
 		RREG32(mmSRBM_STATUS2));
 	dev_info(adev->dev, "  SDMA0_STATUS_REG   = 0x%08X\n",
 		RREG32(mmSDMA0_STATUS_REG + SDMA0_REGISTER_OFFSET));
-	dev_info(adev->dev, "  SDMA1_STATUS_REG   = 0x%08X\n",
-		 RREG32(mmSDMA0_STATUS_REG + SDMA1_REGISTER_OFFSET));
+	if (adev->sdma.num_instances > 1) {
+		dev_info(adev->dev, "  SDMA1_STATUS_REG   = 0x%08X\n",
+			RREG32(mmSDMA0_STATUS_REG + SDMA1_REGISTER_OFFSET));
+	}
 	dev_info(adev->dev, "  CP_STAT = 0x%08x\n", RREG32(mmCP_STAT));
 	dev_info(adev->dev, "  CP_STALLED_STAT1 = 0x%08x\n",
 		 RREG32(mmCP_STALLED_STAT1));
@@ -639,9 +654,11 @@ u32 vi_gpu_check_soft_reset(struct amdgpu_device *adev)
 		reset_mask |= AMDGPU_RESET_DMA;
 
 	/* SDMA1_STATUS_REG */
-	tmp = RREG32(mmSDMA0_STATUS_REG + SDMA1_REGISTER_OFFSET);
-	if (!(tmp & SDMA0_STATUS_REG__IDLE_MASK))
-		reset_mask |= AMDGPU_RESET_DMA1;
+	if (adev->sdma.num_instances > 1) {
+		tmp = RREG32(mmSDMA0_STATUS_REG + SDMA1_REGISTER_OFFSET);
+		if (!(tmp & SDMA0_STATUS_REG__IDLE_MASK))
+			reset_mask |= AMDGPU_RESET_DMA1;
+	}
 #if 0
 	/* VCE_STATUS */
 	if (adev->asic_type != CHIP_TOPAZ) {
@@ -1316,6 +1333,7 @@ int vi_set_ip_blocks(struct amdgpu_device *adev)
 		adev->num_ip_blocks = ARRAY_SIZE(tonga_ip_blocks);
 		break;
 	case CHIP_CARRIZO:
+	case CHIP_STONEY:
 		adev->ip_blocks = cz_ip_blocks;
 		adev->num_ip_blocks = ARRAY_SIZE(cz_ip_blocks);
 		break;
@@ -1327,11 +1345,18 @@ int vi_set_ip_blocks(struct amdgpu_device *adev)
 	return 0;
 }
 
+#define ATI_REV_ID_FUSE_MACRO__ADDRESS      0xC0014044
+#define ATI_REV_ID_FUSE_MACRO__SHIFT        9
+#define ATI_REV_ID_FUSE_MACRO__MASK         0x00001E00
+
 static uint32_t vi_get_rev_id(struct amdgpu_device *adev)
 {
 	if (adev->asic_type == CHIP_TOPAZ)
 		return (RREG32(mmPCIE_EFUSE4) & PCIE_EFUSE4__STRAP_BIF_ATI_REV_ID_MASK)
 			>> PCIE_EFUSE4__STRAP_BIF_ATI_REV_ID__SHIFT;
+	else if (adev->flags & AMD_IS_APU)
+		return (RREG32_SMC(ATI_REV_ID_FUSE_MACRO__ADDRESS) & ATI_REV_ID_FUSE_MACRO__MASK)
+			>> ATI_REV_ID_FUSE_MACRO__SHIFT;
 	else
 		return (RREG32(mmCC_DRM_ID_STRAPS) & CC_DRM_ID_STRAPS__ATI_REV_ID_MASK)
 			>> CC_DRM_ID_STRAPS__ATI_REV_ID__SHIFT;
@@ -1398,6 +1423,7 @@ static int vi_common_early_init(void *handle)
 			adev->firmware.smu_load = true;
 		break;
 	case CHIP_CARRIZO:
+	case CHIP_STONEY:
 		adev->has_uvd = true;
 		adev->cg_flags = 0;
 		/* Disable UVD pg */
diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h
index fbdd118..e52933a 100644
--- a/include/uapi/drm/amdgpu_drm.h
+++ b/include/uapi/drm/amdgpu_drm.h
@@ -640,6 +640,6 @@ struct drm_amdgpu_info_hw_ip {
 #define AMDGPU_FAMILY_CI			120 /* Bonaire, Hawaii */
 #define AMDGPU_FAMILY_KV			125 /* Kaveri, Kabini, Mullins */
 #define AMDGPU_FAMILY_VI			130 /* Iceland, Tonga */
-#define AMDGPU_FAMILY_CZ			135 /* Carrizo */
+#define AMDGPU_FAMILY_CZ			135 /* Carrizo, Stoney */
 
 #endif
-- 
cgit v0.10.2


From 81b1509aa99b968ac19b0aaa451ecd61a996db10 Mon Sep 17 00:00:00 2001
From: Samuel Li <samuel.li@amd.com>
Date: Thu, 8 Oct 2015 16:32:03 -0400
Subject: drm/amdgpu: add Stoney pci ids

Signed-off-by: Samuel Li <samuel.li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index 6134b1e..c3216be 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -273,6 +273,8 @@ static struct pci_device_id pciidlist[] = {
 	{0x1002, 0x9875, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CARRIZO|AMD_IS_APU},
 	{0x1002, 0x9876, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CARRIZO|AMD_IS_APU},
 	{0x1002, 0x9877, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CARRIZO|AMD_IS_APU},
+	/* stoney */
+	{0x1002, 0x98E4, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_STONEY|AMD_IS_APU},
 
 	{0, 0, 0}
 };
-- 
cgit v0.10.2


From ed885b210752563b5e90dc0933e262f768ea9fa4 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Thu, 15 Oct 2015 17:34:20 +0200
Subject: drm/amdgpu: change VM size default to 64GB
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

That's still small enough to not waste to much memory on PD/PTs.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index c3216be..8d6668c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -73,7 +73,7 @@ int amdgpu_hard_reset = 0;
 unsigned amdgpu_ip_block_mask = 0xffffffff;
 int amdgpu_bapm = -1;
 int amdgpu_deep_color = 0;
-int amdgpu_vm_size = 8;
+int amdgpu_vm_size = 64;
 int amdgpu_vm_block_size = -1;
 int amdgpu_vm_fault_stop = 0;
 int amdgpu_vm_debug = 0;
@@ -137,7 +137,7 @@ module_param_named(bapm, amdgpu_bapm, int, 0444);
 MODULE_PARM_DESC(deep_color, "Deep Color support (1 = enable, 0 = disable (default))");
 module_param_named(deep_color, amdgpu_deep_color, int, 0444);
 
-MODULE_PARM_DESC(vm_size, "VM address space size in gigabytes (default 8GB)");
+MODULE_PARM_DESC(vm_size, "VM address space size in gigabytes (default 64GB)");
 module_param_named(vm_size, amdgpu_vm_size, int, 0444);
 
 MODULE_PARM_DESC(vm_block_size, "VM page table size in bits (default depending on vm_size)");
-- 
cgit v0.10.2


From f9e3dc786a585e1ec371c446c62dcc77d0548629 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Wed, 21 Oct 2015 17:22:43 +0300
Subject: drm/i915: add hotplug activation period to hotplug update mask

commit 0706f17c307b056ff6f1848320ba82d76945a6ff
Author: Egbert Eich <eich@suse.de>
Date:   Wed Sep 23 16:15:27 2015 +0200

    drm/i915: Avoid race of intel_crt_detect_hotplug() with HPD interrupt, v2

added a check with WARN to ensure only bits within the mask are
enabled. Turns out that doesn't hold for G4X, which spits out:

[    2.641439] ------------[ cut here ]------------
[    2.641444] WARNING: CPU: 0 PID: 1 at drivers/gpu/drm/i915/i915_irq.c:182 i915_hotplug_interrupt_update_locked+0x45/0x83()
[    2.641446] WARN_ON(bits & ~mask)
etc.

Add CRT_HOTPLUG_ACTIVATION_PERIOD_64 to the mask to fix the warning.

Reported-and-tested-by: Oleksij Rempel <linux@rempel-privat.de>
References: https://bugzilla.kernel.org/show_bug.cgi?id=104991
Fixes: 0706f17c307b ("drm/i915: Avoid race of intel_crt_detect_hotplug() with HPD interrupt, v2")
Cc: Egbert Eich <eich@suse.de>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/1445437363-3030-1-git-send-email-jani.nikula@intel.com

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index e24378ee..5d8ec7a 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -4236,9 +4236,10 @@ static void i915_hpd_irq_setup(struct drm_device *dev)
 
 	/* Ignore TV since it's buggy */
 	i915_hotplug_interrupt_update_locked(dev_priv,
-				      (HOTPLUG_INT_EN_MASK
-				       | CRT_HOTPLUG_VOLTAGE_COMPARE_MASK),
-				      hotplug_en);
+					     HOTPLUG_INT_EN_MASK |
+					     CRT_HOTPLUG_VOLTAGE_COMPARE_MASK |
+					     CRT_HOTPLUG_ACTIVATION_PERIOD_64,
+					     hotplug_en);
 }
 
 static irqreturn_t i965_irq_handler(int irq, void *arg)
-- 
cgit v0.10.2


From 8217e97ab9888651f4a5b24c682047457965349a Mon Sep 17 00:00:00 2001
From: Rob Clark <robdclark@gmail.com>
Date: Thu, 22 Oct 2015 12:36:57 -0400
Subject: drm/msm: update generated headers

Signed-off-by: Rob Clark <robdclark@gmail.com>

diff --git a/drivers/gpu/drm/msm/adreno/a2xx.xml.h b/drivers/gpu/drm/msm/adreno/a2xx.xml.h
index 0261f0d..9e2aceb 100644
--- a/drivers/gpu/drm/msm/adreno/a2xx.xml.h
+++ b/drivers/gpu/drm/msm/adreno/a2xx.xml.h
@@ -8,13 +8,14 @@ http://github.com/freedreno/envytools/
 git clone https://github.com/freedreno/envytools.git
 
 The rules-ng-ng source files this header was generated from are:
-- /home/robclark/src/freedreno/envytools/rnndb/adreno.xml               (    364 bytes, from 2015-05-20 20:03:07)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno.xml               (    398 bytes, from 2015-09-24 17:25:31)
 - /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml  (   1453 bytes, from 2015-05-20 20:03:07)
 - /home/robclark/src/freedreno/envytools/rnndb/adreno/a2xx.xml          (  32901 bytes, from 2015-05-20 20:03:14)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_common.xml (  10551 bytes, from 2015-05-20 20:03:14)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_common.xml (  10755 bytes, from 2015-09-14 20:46:55)
 - /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_pm4.xml    (  14968 bytes, from 2015-05-20 20:12:27)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/a3xx.xml          (  67120 bytes, from 2015-08-14 23:22:03)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/a4xx.xml          (  63785 bytes, from 2015-08-14 18:27:06)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/a3xx.xml          (  67771 bytes, from 2015-09-14 20:46:55)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/a4xx.xml          (  63970 bytes, from 2015-09-14 20:50:12)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/ocmem.xml         (   1773 bytes, from 2015-09-24 17:30:00)
 
 Copyright (C) 2013-2015 by the following authors:
 - Rob Clark <robdclark@gmail.com> (robclark)
diff --git a/drivers/gpu/drm/msm/adreno/a3xx.xml.h b/drivers/gpu/drm/msm/adreno/a3xx.xml.h
index 48d1337..97dc1c6 100644
--- a/drivers/gpu/drm/msm/adreno/a3xx.xml.h
+++ b/drivers/gpu/drm/msm/adreno/a3xx.xml.h
@@ -8,13 +8,14 @@ http://github.com/freedreno/envytools/
 git clone https://github.com/freedreno/envytools.git
 
 The rules-ng-ng source files this header was generated from are:
-- /home/robclark/src/freedreno/envytools/rnndb/adreno.xml               (    364 bytes, from 2015-05-20 20:03:07)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno.xml               (    398 bytes, from 2015-09-24 17:25:31)
 - /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml  (   1453 bytes, from 2015-05-20 20:03:07)
 - /home/robclark/src/freedreno/envytools/rnndb/adreno/a2xx.xml          (  32901 bytes, from 2015-05-20 20:03:14)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_common.xml (  10551 bytes, from 2015-05-20 20:03:14)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_common.xml (  10755 bytes, from 2015-09-14 20:46:55)
 - /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_pm4.xml    (  14968 bytes, from 2015-05-20 20:12:27)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/a3xx.xml          (  67120 bytes, from 2015-08-14 23:22:03)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/a4xx.xml          (  63785 bytes, from 2015-08-14 18:27:06)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/a3xx.xml          (  67771 bytes, from 2015-09-14 20:46:55)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/a4xx.xml          (  63970 bytes, from 2015-09-14 20:50:12)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/ocmem.xml         (   1773 bytes, from 2015-09-24 17:30:00)
 
 Copyright (C) 2013-2015 by the following authors:
 - Rob Clark <robdclark@gmail.com> (robclark)
@@ -280,6 +281,8 @@ enum a3xx_rb_blend_opcode {
 enum a3xx_intp_mode {
 	SMOOTH = 0,
 	FLAT = 1,
+	ZERO = 2,
+	ONE = 3,
 };
 
 enum a3xx_repl_mode {
@@ -680,9 +683,16 @@ static inline uint32_t REG_A3XX_CP_PROTECT_REG(uint32_t i0) { return 0x00000460
 #define A3XX_GRAS_CL_CLIP_CNTL_VP_CLIP_CODE_IGNORE		0x00080000
 #define A3XX_GRAS_CL_CLIP_CNTL_VP_XFORM_DISABLE			0x00100000
 #define A3XX_GRAS_CL_CLIP_CNTL_PERSP_DIVISION_DISABLE		0x00200000
+#define A3XX_GRAS_CL_CLIP_CNTL_ZERO_GB_SCALE_Z			0x00400000
 #define A3XX_GRAS_CL_CLIP_CNTL_ZCOORD				0x00800000
 #define A3XX_GRAS_CL_CLIP_CNTL_WCOORD				0x01000000
 #define A3XX_GRAS_CL_CLIP_CNTL_ZCLIP_DISABLE			0x02000000
+#define A3XX_GRAS_CL_CLIP_CNTL_NUM_USER_CLIP_PLANES__MASK	0x1c000000
+#define A3XX_GRAS_CL_CLIP_CNTL_NUM_USER_CLIP_PLANES__SHIFT	26
+static inline uint32_t A3XX_GRAS_CL_CLIP_CNTL_NUM_USER_CLIP_PLANES(uint32_t val)
+{
+	return ((val) << A3XX_GRAS_CL_CLIP_CNTL_NUM_USER_CLIP_PLANES__SHIFT) & A3XX_GRAS_CL_CLIP_CNTL_NUM_USER_CLIP_PLANES__MASK;
+}
 
 #define REG_A3XX_GRAS_CL_GB_CLIP_ADJ				0x00002044
 #define A3XX_GRAS_CL_GB_CLIP_ADJ_HORZ__MASK			0x000003ff
@@ -773,7 +783,7 @@ static inline uint32_t A3XX_GRAS_SU_POINT_SIZE(float val)
 #define A3XX_GRAS_SU_POLY_OFFSET_SCALE_VAL__SHIFT		0
 static inline uint32_t A3XX_GRAS_SU_POLY_OFFSET_SCALE_VAL(float val)
 {
-	return ((((int32_t)(val * 16384.0))) << A3XX_GRAS_SU_POLY_OFFSET_SCALE_VAL__SHIFT) & A3XX_GRAS_SU_POLY_OFFSET_SCALE_VAL__MASK;
+	return ((((int32_t)(val * 1048576.0))) << A3XX_GRAS_SU_POLY_OFFSET_SCALE_VAL__SHIFT) & A3XX_GRAS_SU_POLY_OFFSET_SCALE_VAL__MASK;
 }
 
 #define REG_A3XX_GRAS_SU_POLY_OFFSET_OFFSET			0x0000206d
@@ -894,6 +904,9 @@ static inline uint32_t A3XX_RB_MODE_CONTROL_MRT(uint32_t val)
 #define A3XX_RB_MODE_CONTROL_PACKER_TIMER_ENABLE		0x00010000
 
 #define REG_A3XX_RB_RENDER_CONTROL				0x000020c1
+#define A3XX_RB_RENDER_CONTROL_DUAL_COLOR_IN_ENABLE		0x00000001
+#define A3XX_RB_RENDER_CONTROL_YUV_IN_ENABLE			0x00000002
+#define A3XX_RB_RENDER_CONTROL_COV_VALUE_INPUT_ENABLE		0x00000004
 #define A3XX_RB_RENDER_CONTROL_FACENESS				0x00000008
 #define A3XX_RB_RENDER_CONTROL_BIN_WIDTH__MASK			0x00000ff0
 #define A3XX_RB_RENDER_CONTROL_BIN_WIDTH__SHIFT			4
@@ -907,6 +920,8 @@ static inline uint32_t A3XX_RB_RENDER_CONTROL_BIN_WIDTH(uint32_t val)
 #define A3XX_RB_RENDER_CONTROL_YCOORD				0x00008000
 #define A3XX_RB_RENDER_CONTROL_ZCOORD				0x00010000
 #define A3XX_RB_RENDER_CONTROL_WCOORD				0x00020000
+#define A3XX_RB_RENDER_CONTROL_I_CLAMP_ENABLE			0x00080000
+#define A3XX_RB_RENDER_CONTROL_COV_VALUE_OUTPUT_ENABLE		0x00100000
 #define A3XX_RB_RENDER_CONTROL_ALPHA_TEST			0x00400000
 #define A3XX_RB_RENDER_CONTROL_ALPHA_TEST_FUNC__MASK		0x07000000
 #define A3XX_RB_RENDER_CONTROL_ALPHA_TEST_FUNC__SHIFT		24
@@ -914,6 +929,8 @@ static inline uint32_t A3XX_RB_RENDER_CONTROL_ALPHA_TEST_FUNC(enum adreno_compar
 {
 	return ((val) << A3XX_RB_RENDER_CONTROL_ALPHA_TEST_FUNC__SHIFT) & A3XX_RB_RENDER_CONTROL_ALPHA_TEST_FUNC__MASK;
 }
+#define A3XX_RB_RENDER_CONTROL_ALPHA_TO_COVERAGE		0x40000000
+#define A3XX_RB_RENDER_CONTROL_ALPHA_TO_ONE			0x80000000
 
 #define REG_A3XX_RB_MSAA_CONTROL				0x000020c2
 #define A3XX_RB_MSAA_CONTROL_DISABLE				0x00000400
diff --git a/drivers/gpu/drm/msm/adreno/a4xx.xml.h b/drivers/gpu/drm/msm/adreno/a4xx.xml.h
index ac55066..99de827 100644
--- a/drivers/gpu/drm/msm/adreno/a4xx.xml.h
+++ b/drivers/gpu/drm/msm/adreno/a4xx.xml.h
@@ -8,13 +8,14 @@ http://github.com/freedreno/envytools/
 git clone https://github.com/freedreno/envytools.git
 
 The rules-ng-ng source files this header was generated from are:
-- /home/robclark/src/freedreno/envytools/rnndb/adreno.xml               (    364 bytes, from 2015-05-20 20:03:07)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno.xml               (    398 bytes, from 2015-09-24 17:25:31)
 - /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml  (   1453 bytes, from 2015-05-20 20:03:07)
 - /home/robclark/src/freedreno/envytools/rnndb/adreno/a2xx.xml          (  32901 bytes, from 2015-05-20 20:03:14)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_common.xml (  10551 bytes, from 2015-05-20 20:03:14)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_common.xml (  10755 bytes, from 2015-09-14 20:46:55)
 - /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_pm4.xml    (  14968 bytes, from 2015-05-20 20:12:27)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/a3xx.xml          (  67120 bytes, from 2015-08-14 23:22:03)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/a4xx.xml          (  63785 bytes, from 2015-08-14 18:27:06)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/a3xx.xml          (  67771 bytes, from 2015-09-14 20:46:55)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/a4xx.xml          (  63970 bytes, from 2015-09-14 20:50:12)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/ocmem.xml         (   1773 bytes, from 2015-09-24 17:30:00)
 
 Copyright (C) 2013-2015 by the following authors:
 - Rob Clark <robdclark@gmail.com> (robclark)
@@ -162,10 +163,13 @@ enum a4xx_tex_fmt {
 	TFMT4_8_UNORM = 4,
 	TFMT4_8_8_UNORM = 14,
 	TFMT4_8_8_8_8_UNORM = 28,
+	TFMT4_8_SNORM = 5,
 	TFMT4_8_8_SNORM = 15,
 	TFMT4_8_8_8_8_SNORM = 29,
+	TFMT4_8_UINT = 6,
 	TFMT4_8_8_UINT = 16,
 	TFMT4_8_8_8_8_UINT = 30,
+	TFMT4_8_SINT = 7,
 	TFMT4_8_8_SINT = 17,
 	TFMT4_8_8_8_8_SINT = 31,
 	TFMT4_16_UINT = 21,
@@ -246,7 +250,8 @@ enum a4xx_tex_clamp {
 	A4XX_TEX_REPEAT = 0,
 	A4XX_TEX_CLAMP_TO_EDGE = 1,
 	A4XX_TEX_MIRROR_REPEAT = 2,
-	A4XX_TEX_CLAMP_NONE = 3,
+	A4XX_TEX_CLAMP_TO_BORDER = 3,
+	A4XX_TEX_MIRROR_CLAMP = 4,
 };
 
 enum a4xx_tex_aniso {
diff --git a/drivers/gpu/drm/msm/adreno/adreno_common.xml.h b/drivers/gpu/drm/msm/adreno/adreno_common.xml.h
index 399a9e5..c304468 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_common.xml.h
+++ b/drivers/gpu/drm/msm/adreno/adreno_common.xml.h
@@ -8,13 +8,14 @@ http://github.com/freedreno/envytools/
 git clone https://github.com/freedreno/envytools.git
 
 The rules-ng-ng source files this header was generated from are:
-- /home/robclark/src/freedreno/envytools/rnndb/adreno.xml               (    364 bytes, from 2015-05-20 20:03:07)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno.xml               (    398 bytes, from 2015-09-24 17:25:31)
 - /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml  (   1453 bytes, from 2015-05-20 20:03:07)
 - /home/robclark/src/freedreno/envytools/rnndb/adreno/a2xx.xml          (  32901 bytes, from 2015-05-20 20:03:14)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_common.xml (  10551 bytes, from 2015-05-20 20:03:14)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_common.xml (  10755 bytes, from 2015-09-14 20:46:55)
 - /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_pm4.xml    (  14968 bytes, from 2015-05-20 20:12:27)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/a3xx.xml          (  67120 bytes, from 2015-08-14 23:22:03)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/a4xx.xml          (  63785 bytes, from 2015-08-14 18:27:06)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/a3xx.xml          (  67771 bytes, from 2015-09-14 20:46:55)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/a4xx.xml          (  63970 bytes, from 2015-09-14 20:50:12)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/ocmem.xml         (   1773 bytes, from 2015-09-24 17:30:00)
 
 Copyright (C) 2013-2015 by the following authors:
 - Rob Clark <robdclark@gmail.com> (robclark)
@@ -85,6 +86,10 @@ enum adreno_rb_blend_factor {
 	FACTOR_CONSTANT_ALPHA = 14,
 	FACTOR_ONE_MINUS_CONSTANT_ALPHA = 15,
 	FACTOR_SRC_ALPHA_SATURATE = 16,
+	FACTOR_SRC1_COLOR = 20,
+	FACTOR_ONE_MINUS_SRC1_COLOR = 21,
+	FACTOR_SRC1_ALPHA = 22,
+	FACTOR_ONE_MINUS_SRC1_ALPHA = 23,
 };
 
 enum adreno_rb_surface_endian {
diff --git a/drivers/gpu/drm/msm/adreno/adreno_pm4.xml.h b/drivers/gpu/drm/msm/adreno/adreno_pm4.xml.h
index 41904fe..a22fef5 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_pm4.xml.h
+++ b/drivers/gpu/drm/msm/adreno/adreno_pm4.xml.h
@@ -8,13 +8,14 @@ http://github.com/freedreno/envytools/
 git clone https://github.com/freedreno/envytools.git
 
 The rules-ng-ng source files this header was generated from are:
-- /home/robclark/src/freedreno/envytools/rnndb/adreno.xml               (    364 bytes, from 2015-05-20 20:03:07)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno.xml               (    398 bytes, from 2015-09-24 17:25:31)
 - /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml  (   1453 bytes, from 2015-05-20 20:03:07)
 - /home/robclark/src/freedreno/envytools/rnndb/adreno/a2xx.xml          (  32901 bytes, from 2015-05-20 20:03:14)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_common.xml (  10551 bytes, from 2015-05-20 20:03:14)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_common.xml (  10755 bytes, from 2015-09-14 20:46:55)
 - /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_pm4.xml    (  14968 bytes, from 2015-05-20 20:12:27)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/a3xx.xml          (  67120 bytes, from 2015-08-14 23:22:03)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/a4xx.xml          (  63785 bytes, from 2015-08-14 18:27:06)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/a3xx.xml          (  67771 bytes, from 2015-09-14 20:46:55)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/a4xx.xml          (  63970 bytes, from 2015-09-14 20:50:12)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/ocmem.xml         (   1773 bytes, from 2015-09-24 17:30:00)
 
 Copyright (C) 2013-2015 by the following authors:
 - Rob Clark <robdclark@gmail.com> (robclark)
diff --git a/drivers/gpu/drm/msm/dsi/dsi.xml.h b/drivers/gpu/drm/msm/dsi/dsi.xml.h
index 1d2e32f..b2b5f3d 100644
--- a/drivers/gpu/drm/msm/dsi/dsi.xml.h
+++ b/drivers/gpu/drm/msm/dsi/dsi.xml.h
@@ -11,10 +11,10 @@ The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/msm.xml                 (    676 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1453 bytes, from 2015-05-20 20:03:07)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp4.xml            (  20915 bytes, from 2015-05-20 20:03:14)
-- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp_common.xml      (   2576 bytes, from 2015-07-09 22:10:24)
-- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp5.xml            (  36021 bytes, from 2015-07-09 22:10:24)
-- /home/robclark/src/freedreno/envytools/rnndb/dsi/dsi.xml             (  26057 bytes, from 2015-08-14 21:47:57)
-- /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    344 bytes, from 2015-05-20 20:03:07)
+- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp_common.xml      (   2849 bytes, from 2015-09-18 12:07:28)
+- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp5.xml            (  37194 bytes, from 2015-09-18 12:07:28)
+- /home/robclark/src/freedreno/envytools/rnndb/dsi/dsi.xml             (  27887 bytes, from 2015-10-22 16:34:52)
+- /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    602 bytes, from 2015-10-22 16:35:02)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/mmss_cc.xml         (   1686 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/hdmi/qfprom.xml         (    600 bytes, from 2015-05-20 20:03:07)
 - /home/robclark/src/freedreno/envytools/rnndb/hdmi/hdmi.xml           (  29154 bytes, from 2015-08-10 21:25:43)
@@ -567,114 +567,234 @@ static inline uint32_t DSI_VERSION_MAJOR(uint32_t val)
 #define REG_DSI_8x60_PHY_CAL_STATUS				0x000000fc
 #define DSI_8x60_PHY_CAL_STATUS_CAL_BUSY			0x10000000
 
-static inline uint32_t REG_DSI_8960_LN(uint32_t i0) { return 0x00000300 + 0x40*i0; }
+static inline uint32_t REG_DSI_28nm_8960_PHY_LN(uint32_t i0) { return 0x00000000 + 0x40*i0; }
 
-static inline uint32_t REG_DSI_8960_LN_CFG_0(uint32_t i0) { return 0x00000300 + 0x40*i0; }
+static inline uint32_t REG_DSI_28nm_8960_PHY_LN_CFG_0(uint32_t i0) { return 0x00000000 + 0x40*i0; }
 
-static inline uint32_t REG_DSI_8960_LN_CFG_1(uint32_t i0) { return 0x00000304 + 0x40*i0; }
+static inline uint32_t REG_DSI_28nm_8960_PHY_LN_CFG_1(uint32_t i0) { return 0x00000004 + 0x40*i0; }
 
-static inline uint32_t REG_DSI_8960_LN_CFG_2(uint32_t i0) { return 0x00000308 + 0x40*i0; }
+static inline uint32_t REG_DSI_28nm_8960_PHY_LN_CFG_2(uint32_t i0) { return 0x00000008 + 0x40*i0; }
 
-static inline uint32_t REG_DSI_8960_LN_TEST_DATAPATH(uint32_t i0) { return 0x0000030c + 0x40*i0; }
+static inline uint32_t REG_DSI_28nm_8960_PHY_LN_TEST_DATAPATH(uint32_t i0) { return 0x0000000c + 0x40*i0; }
 
-static inline uint32_t REG_DSI_8960_LN_TEST_STR_0(uint32_t i0) { return 0x00000314 + 0x40*i0; }
+static inline uint32_t REG_DSI_28nm_8960_PHY_LN_TEST_STR_0(uint32_t i0) { return 0x00000014 + 0x40*i0; }
 
-static inline uint32_t REG_DSI_8960_LN_TEST_STR_1(uint32_t i0) { return 0x00000318 + 0x40*i0; }
+static inline uint32_t REG_DSI_28nm_8960_PHY_LN_TEST_STR_1(uint32_t i0) { return 0x00000018 + 0x40*i0; }
 
-#define REG_DSI_8960_PHY_LNCK_CFG_0				0x00000400
+#define REG_DSI_28nm_8960_PHY_LNCK_CFG_0			0x00000100
 
-#define REG_DSI_8960_PHY_LNCK_CFG_1				0x00000404
+#define REG_DSI_28nm_8960_PHY_LNCK_CFG_1			0x00000104
 
-#define REG_DSI_8960_PHY_LNCK_CFG_2				0x00000408
+#define REG_DSI_28nm_8960_PHY_LNCK_CFG_2			0x00000108
 
-#define REG_DSI_8960_PHY_LNCK_TEST_DATAPATH			0x0000040c
+#define REG_DSI_28nm_8960_PHY_LNCK_TEST_DATAPATH		0x0000010c
 
-#define REG_DSI_8960_PHY_LNCK_TEST_STR0				0x00000414
+#define REG_DSI_28nm_8960_PHY_LNCK_TEST_STR0			0x00000114
 
-#define REG_DSI_8960_PHY_LNCK_TEST_STR1				0x00000418
+#define REG_DSI_28nm_8960_PHY_LNCK_TEST_STR1			0x00000118
 
-#define REG_DSI_8960_PHY_TIMING_CTRL_0				0x00000440
+#define REG_DSI_28nm_8960_PHY_TIMING_CTRL_0			0x00000140
+#define DSI_28nm_8960_PHY_TIMING_CTRL_0_CLK_ZERO__MASK		0x000000ff
+#define DSI_28nm_8960_PHY_TIMING_CTRL_0_CLK_ZERO__SHIFT		0
+static inline uint32_t DSI_28nm_8960_PHY_TIMING_CTRL_0_CLK_ZERO(uint32_t val)
+{
+	return ((val) << DSI_28nm_8960_PHY_TIMING_CTRL_0_CLK_ZERO__SHIFT) & DSI_28nm_8960_PHY_TIMING_CTRL_0_CLK_ZERO__MASK;
+}
+
+#define REG_DSI_28nm_8960_PHY_TIMING_CTRL_1			0x00000144
+#define DSI_28nm_8960_PHY_TIMING_CTRL_1_CLK_TRAIL__MASK		0x000000ff
+#define DSI_28nm_8960_PHY_TIMING_CTRL_1_CLK_TRAIL__SHIFT	0
+static inline uint32_t DSI_28nm_8960_PHY_TIMING_CTRL_1_CLK_TRAIL(uint32_t val)
+{
+	return ((val) << DSI_28nm_8960_PHY_TIMING_CTRL_1_CLK_TRAIL__SHIFT) & DSI_28nm_8960_PHY_TIMING_CTRL_1_CLK_TRAIL__MASK;
+}
+
+#define REG_DSI_28nm_8960_PHY_TIMING_CTRL_2			0x00000148
+#define DSI_28nm_8960_PHY_TIMING_CTRL_2_CLK_PREPARE__MASK	0x000000ff
+#define DSI_28nm_8960_PHY_TIMING_CTRL_2_CLK_PREPARE__SHIFT	0
+static inline uint32_t DSI_28nm_8960_PHY_TIMING_CTRL_2_CLK_PREPARE(uint32_t val)
+{
+	return ((val) << DSI_28nm_8960_PHY_TIMING_CTRL_2_CLK_PREPARE__SHIFT) & DSI_28nm_8960_PHY_TIMING_CTRL_2_CLK_PREPARE__MASK;
+}
+
+#define REG_DSI_28nm_8960_PHY_TIMING_CTRL_3			0x0000014c
+
+#define REG_DSI_28nm_8960_PHY_TIMING_CTRL_4			0x00000150
+#define DSI_28nm_8960_PHY_TIMING_CTRL_4_HS_EXIT__MASK		0x000000ff
+#define DSI_28nm_8960_PHY_TIMING_CTRL_4_HS_EXIT__SHIFT		0
+static inline uint32_t DSI_28nm_8960_PHY_TIMING_CTRL_4_HS_EXIT(uint32_t val)
+{
+	return ((val) << DSI_28nm_8960_PHY_TIMING_CTRL_4_HS_EXIT__SHIFT) & DSI_28nm_8960_PHY_TIMING_CTRL_4_HS_EXIT__MASK;
+}
+
+#define REG_DSI_28nm_8960_PHY_TIMING_CTRL_5			0x00000154
+#define DSI_28nm_8960_PHY_TIMING_CTRL_5_HS_ZERO__MASK		0x000000ff
+#define DSI_28nm_8960_PHY_TIMING_CTRL_5_HS_ZERO__SHIFT		0
+static inline uint32_t DSI_28nm_8960_PHY_TIMING_CTRL_5_HS_ZERO(uint32_t val)
+{
+	return ((val) << DSI_28nm_8960_PHY_TIMING_CTRL_5_HS_ZERO__SHIFT) & DSI_28nm_8960_PHY_TIMING_CTRL_5_HS_ZERO__MASK;
+}
+
+#define REG_DSI_28nm_8960_PHY_TIMING_CTRL_6			0x00000158
+#define DSI_28nm_8960_PHY_TIMING_CTRL_6_HS_PREPARE__MASK	0x000000ff
+#define DSI_28nm_8960_PHY_TIMING_CTRL_6_HS_PREPARE__SHIFT	0
+static inline uint32_t DSI_28nm_8960_PHY_TIMING_CTRL_6_HS_PREPARE(uint32_t val)
+{
+	return ((val) << DSI_28nm_8960_PHY_TIMING_CTRL_6_HS_PREPARE__SHIFT) & DSI_28nm_8960_PHY_TIMING_CTRL_6_HS_PREPARE__MASK;
+}
+
+#define REG_DSI_28nm_8960_PHY_TIMING_CTRL_7			0x0000015c
+#define DSI_28nm_8960_PHY_TIMING_CTRL_7_HS_TRAIL__MASK		0x000000ff
+#define DSI_28nm_8960_PHY_TIMING_CTRL_7_HS_TRAIL__SHIFT		0
+static inline uint32_t DSI_28nm_8960_PHY_TIMING_CTRL_7_HS_TRAIL(uint32_t val)
+{
+	return ((val) << DSI_28nm_8960_PHY_TIMING_CTRL_7_HS_TRAIL__SHIFT) & DSI_28nm_8960_PHY_TIMING_CTRL_7_HS_TRAIL__MASK;
+}
+
+#define REG_DSI_28nm_8960_PHY_TIMING_CTRL_8			0x00000160
+#define DSI_28nm_8960_PHY_TIMING_CTRL_8_HS_RQST__MASK		0x000000ff
+#define DSI_28nm_8960_PHY_TIMING_CTRL_8_HS_RQST__SHIFT		0
+static inline uint32_t DSI_28nm_8960_PHY_TIMING_CTRL_8_HS_RQST(uint32_t val)
+{
+	return ((val) << DSI_28nm_8960_PHY_TIMING_CTRL_8_HS_RQST__SHIFT) & DSI_28nm_8960_PHY_TIMING_CTRL_8_HS_RQST__MASK;
+}
+
+#define REG_DSI_28nm_8960_PHY_TIMING_CTRL_9			0x00000164
+#define DSI_28nm_8960_PHY_TIMING_CTRL_9_TA_GO__MASK		0x00000007
+#define DSI_28nm_8960_PHY_TIMING_CTRL_9_TA_GO__SHIFT		0
+static inline uint32_t DSI_28nm_8960_PHY_TIMING_CTRL_9_TA_GO(uint32_t val)
+{
+	return ((val) << DSI_28nm_8960_PHY_TIMING_CTRL_9_TA_GO__SHIFT) & DSI_28nm_8960_PHY_TIMING_CTRL_9_TA_GO__MASK;
+}
+#define DSI_28nm_8960_PHY_TIMING_CTRL_9_TA_SURE__MASK		0x00000070
+#define DSI_28nm_8960_PHY_TIMING_CTRL_9_TA_SURE__SHIFT		4
+static inline uint32_t DSI_28nm_8960_PHY_TIMING_CTRL_9_TA_SURE(uint32_t val)
+{
+	return ((val) << DSI_28nm_8960_PHY_TIMING_CTRL_9_TA_SURE__SHIFT) & DSI_28nm_8960_PHY_TIMING_CTRL_9_TA_SURE__MASK;
+}
+
+#define REG_DSI_28nm_8960_PHY_TIMING_CTRL_10			0x00000168
+#define DSI_28nm_8960_PHY_TIMING_CTRL_10_TA_GET__MASK		0x00000007
+#define DSI_28nm_8960_PHY_TIMING_CTRL_10_TA_GET__SHIFT		0
+static inline uint32_t DSI_28nm_8960_PHY_TIMING_CTRL_10_TA_GET(uint32_t val)
+{
+	return ((val) << DSI_28nm_8960_PHY_TIMING_CTRL_10_TA_GET__SHIFT) & DSI_28nm_8960_PHY_TIMING_CTRL_10_TA_GET__MASK;
+}
+
+#define REG_DSI_28nm_8960_PHY_TIMING_CTRL_11			0x0000016c
+#define DSI_28nm_8960_PHY_TIMING_CTRL_11_TRIG3_CMD__MASK	0x000000ff
+#define DSI_28nm_8960_PHY_TIMING_CTRL_11_TRIG3_CMD__SHIFT	0
+static inline uint32_t DSI_28nm_8960_PHY_TIMING_CTRL_11_TRIG3_CMD(uint32_t val)
+{
+	return ((val) << DSI_28nm_8960_PHY_TIMING_CTRL_11_TRIG3_CMD__SHIFT) & DSI_28nm_8960_PHY_TIMING_CTRL_11_TRIG3_CMD__MASK;
+}
+
+#define REG_DSI_28nm_8960_PHY_CTRL_0				0x00000170
+
+#define REG_DSI_28nm_8960_PHY_CTRL_1				0x00000174
+
+#define REG_DSI_28nm_8960_PHY_CTRL_2				0x00000178
+
+#define REG_DSI_28nm_8960_PHY_CTRL_3				0x0000017c
+
+#define REG_DSI_28nm_8960_PHY_STRENGTH_0			0x00000180
+
+#define REG_DSI_28nm_8960_PHY_STRENGTH_1			0x00000184
+
+#define REG_DSI_28nm_8960_PHY_STRENGTH_2			0x00000188
+
+#define REG_DSI_28nm_8960_PHY_BIST_CTRL_0			0x0000018c
+
+#define REG_DSI_28nm_8960_PHY_BIST_CTRL_1			0x00000190
+
+#define REG_DSI_28nm_8960_PHY_BIST_CTRL_2			0x00000194
+
+#define REG_DSI_28nm_8960_PHY_BIST_CTRL_3			0x00000198
+
+#define REG_DSI_28nm_8960_PHY_BIST_CTRL_4			0x0000019c
 
-#define REG_DSI_8960_PHY_TIMING_CTRL_1				0x00000444
+#define REG_DSI_28nm_8960_PHY_LDO_CTRL				0x000001b0
 
-#define REG_DSI_8960_PHY_TIMING_CTRL_2				0x00000448
+#define REG_DSI_28nm_8960_PHY_MISC_REGULATOR_CTRL_0		0x00000000
 
-#define REG_DSI_8960_PHY_TIMING_CTRL_3				0x0000044c
+#define REG_DSI_28nm_8960_PHY_MISC_REGULATOR_CTRL_1		0x00000004
 
-#define REG_DSI_8960_PHY_TIMING_CTRL_4				0x00000450
+#define REG_DSI_28nm_8960_PHY_MISC_REGULATOR_CTRL_2		0x00000008
 
-#define REG_DSI_8960_PHY_TIMING_CTRL_5				0x00000454
+#define REG_DSI_28nm_8960_PHY_MISC_REGULATOR_CTRL_3		0x0000000c
 
-#define REG_DSI_8960_PHY_TIMING_CTRL_6				0x00000458
+#define REG_DSI_28nm_8960_PHY_MISC_REGULATOR_CTRL_4		0x00000010
 
-#define REG_DSI_8960_PHY_TIMING_CTRL_7				0x0000045c
+#define REG_DSI_28nm_8960_PHY_MISC_REGULATOR_CTRL_5		0x00000014
 
-#define REG_DSI_8960_PHY_TIMING_CTRL_8				0x00000460
+#define REG_DSI_28nm_8960_PHY_MISC_REGULATOR_CAL_PWR_CFG	0x00000018
 
-#define REG_DSI_8960_PHY_TIMING_CTRL_9				0x00000464
+#define REG_DSI_28nm_8960_PHY_MISC_CAL_HW_TRIGGER		0x00000028
 
-#define REG_DSI_8960_PHY_TIMING_CTRL_10				0x00000468
+#define REG_DSI_28nm_8960_PHY_MISC_CAL_SW_CFG_0			0x0000002c
 
-#define REG_DSI_8960_PHY_TIMING_CTRL_11				0x0000046c
+#define REG_DSI_28nm_8960_PHY_MISC_CAL_SW_CFG_1			0x00000030
 
-#define REG_DSI_8960_PHY_CTRL_0					0x00000470
+#define REG_DSI_28nm_8960_PHY_MISC_CAL_SW_CFG_2			0x00000034
 
-#define REG_DSI_8960_PHY_CTRL_1					0x00000474
+#define REG_DSI_28nm_8960_PHY_MISC_CAL_HW_CFG_0			0x00000038
 
-#define REG_DSI_8960_PHY_CTRL_2					0x00000478
+#define REG_DSI_28nm_8960_PHY_MISC_CAL_HW_CFG_1			0x0000003c
 
-#define REG_DSI_8960_PHY_CTRL_3					0x0000047c
+#define REG_DSI_28nm_8960_PHY_MISC_CAL_HW_CFG_2			0x00000040
 
-#define REG_DSI_8960_PHY_STRENGTH_0				0x00000480
+#define REG_DSI_28nm_8960_PHY_MISC_CAL_HW_CFG_3			0x00000044
 
-#define REG_DSI_8960_PHY_STRENGTH_1				0x00000484
+#define REG_DSI_28nm_8960_PHY_MISC_CAL_HW_CFG_4			0x00000048
 
-#define REG_DSI_8960_PHY_STRENGTH_2				0x00000488
+#define REG_DSI_28nm_8960_PHY_MISC_CAL_STATUS			0x00000050
+#define DSI_28nm_8960_PHY_MISC_CAL_STATUS_CAL_BUSY		0x00000010
 
-#define REG_DSI_8960_PHY_BIST_CTRL_0				0x0000048c
+#define REG_DSI_28nm_8960_PHY_PLL_CTRL_0			0x00000000
+#define DSI_28nm_8960_PHY_PLL_CTRL_0_ENABLE			0x00000001
 
-#define REG_DSI_8960_PHY_BIST_CTRL_1				0x00000490
+#define REG_DSI_28nm_8960_PHY_PLL_CTRL_1			0x00000004
 
-#define REG_DSI_8960_PHY_BIST_CTRL_2				0x00000494
+#define REG_DSI_28nm_8960_PHY_PLL_CTRL_2			0x00000008
 
-#define REG_DSI_8960_PHY_BIST_CTRL_3				0x00000498
+#define REG_DSI_28nm_8960_PHY_PLL_CTRL_3			0x0000000c
 
-#define REG_DSI_8960_PHY_BIST_CTRL_4				0x0000049c
+#define REG_DSI_28nm_8960_PHY_PLL_CTRL_4			0x00000010
 
-#define REG_DSI_8960_PHY_LDO_CTRL				0x000004b0
+#define REG_DSI_28nm_8960_PHY_PLL_CTRL_5			0x00000014
 
-#define REG_DSI_8960_PHY_REGULATOR_CTRL_0			0x00000500
+#define REG_DSI_28nm_8960_PHY_PLL_CTRL_6			0x00000018
 
-#define REG_DSI_8960_PHY_REGULATOR_CTRL_1			0x00000504
+#define REG_DSI_28nm_8960_PHY_PLL_CTRL_7			0x0000001c
 
-#define REG_DSI_8960_PHY_REGULATOR_CTRL_2			0x00000508
+#define REG_DSI_28nm_8960_PHY_PLL_CTRL_8			0x00000020
 
-#define REG_DSI_8960_PHY_REGULATOR_CTRL_3			0x0000050c
+#define REG_DSI_28nm_8960_PHY_PLL_CTRL_9			0x00000024
 
-#define REG_DSI_8960_PHY_REGULATOR_CTRL_4			0x00000510
+#define REG_DSI_28nm_8960_PHY_PLL_CTRL_10			0x00000028
 
-#define REG_DSI_8960_PHY_REGULATOR_CAL_PWR_CFG			0x00000518
+#define REG_DSI_28nm_8960_PHY_PLL_CTRL_11			0x0000002c
 
-#define REG_DSI_8960_PHY_CAL_HW_TRIGGER				0x00000528
+#define REG_DSI_28nm_8960_PHY_PLL_CTRL_12			0x00000030
 
-#define REG_DSI_8960_PHY_CAL_SW_CFG_0				0x0000052c
+#define REG_DSI_28nm_8960_PHY_PLL_CTRL_13			0x00000034
 
-#define REG_DSI_8960_PHY_CAL_SW_CFG_1				0x00000530
+#define REG_DSI_28nm_8960_PHY_PLL_CTRL_14			0x00000038
 
-#define REG_DSI_8960_PHY_CAL_SW_CFG_2				0x00000534
+#define REG_DSI_28nm_8960_PHY_PLL_CTRL_15			0x0000003c
 
-#define REG_DSI_8960_PHY_CAL_HW_CFG_0				0x00000538
+#define REG_DSI_28nm_8960_PHY_PLL_CTRL_16			0x00000040
 
-#define REG_DSI_8960_PHY_CAL_HW_CFG_1				0x0000053c
+#define REG_DSI_28nm_8960_PHY_PLL_CTRL_17			0x00000044
 
-#define REG_DSI_8960_PHY_CAL_HW_CFG_2				0x00000540
+#define REG_DSI_28nm_8960_PHY_PLL_CTRL_18			0x00000048
 
-#define REG_DSI_8960_PHY_CAL_HW_CFG_3				0x00000544
+#define REG_DSI_28nm_8960_PHY_PLL_CTRL_19			0x0000004c
 
-#define REG_DSI_8960_PHY_CAL_HW_CFG_4				0x00000548
+#define REG_DSI_28nm_8960_PHY_PLL_CTRL_20			0x00000050
 
-#define REG_DSI_8960_PHY_CAL_STATUS				0x00000550
-#define DSI_8960_PHY_CAL_STATUS_CAL_BUSY			0x00000010
+#define REG_DSI_28nm_8960_PHY_PLL_RDY				0x00000080
+#define DSI_28nm_8960_PHY_PLL_RDY_PLL_RDY			0x00000001
 
 static inline uint32_t REG_DSI_28nm_PHY_LN(uint32_t i0) { return 0x00000000 + 0x40*i0; }
 
diff --git a/drivers/gpu/drm/msm/dsi/mmss_cc.xml.h b/drivers/gpu/drm/msm/dsi/mmss_cc.xml.h
index 5de505e..80ec65e 100644
--- a/drivers/gpu/drm/msm/dsi/mmss_cc.xml.h
+++ b/drivers/gpu/drm/msm/dsi/mmss_cc.xml.h
@@ -11,10 +11,10 @@ The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/msm.xml                 (    676 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1453 bytes, from 2015-05-20 20:03:07)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp4.xml            (  20915 bytes, from 2015-05-20 20:03:14)
-- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp_common.xml      (   2576 bytes, from 2015-07-09 22:10:24)
-- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp5.xml            (  36021 bytes, from 2015-07-09 22:10:24)
-- /home/robclark/src/freedreno/envytools/rnndb/dsi/dsi.xml             (  26057 bytes, from 2015-08-14 21:47:57)
-- /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    344 bytes, from 2015-05-20 20:03:07)
+- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp_common.xml      (   2849 bytes, from 2015-09-18 12:07:28)
+- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp5.xml            (  37194 bytes, from 2015-09-18 12:07:28)
+- /home/robclark/src/freedreno/envytools/rnndb/dsi/dsi.xml             (  27887 bytes, from 2015-10-22 16:34:52)
+- /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    602 bytes, from 2015-10-22 16:35:02)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/mmss_cc.xml         (   1686 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/hdmi/qfprom.xml         (    600 bytes, from 2015-05-20 20:03:07)
 - /home/robclark/src/freedreno/envytools/rnndb/hdmi/hdmi.xml           (  29154 bytes, from 2015-08-10 21:25:43)
diff --git a/drivers/gpu/drm/msm/dsi/sfpb.xml.h b/drivers/gpu/drm/msm/dsi/sfpb.xml.h
index 06cbddf..7d7662e 100644
--- a/drivers/gpu/drm/msm/dsi/sfpb.xml.h
+++ b/drivers/gpu/drm/msm/dsi/sfpb.xml.h
@@ -11,10 +11,10 @@ The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/msm.xml                 (    676 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1453 bytes, from 2015-05-20 20:03:07)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp4.xml            (  20915 bytes, from 2015-05-20 20:03:14)
-- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp_common.xml      (   2576 bytes, from 2015-07-09 22:10:24)
-- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp5.xml            (  36021 bytes, from 2015-07-09 22:10:24)
-- /home/robclark/src/freedreno/envytools/rnndb/dsi/dsi.xml             (  26057 bytes, from 2015-08-14 21:47:57)
-- /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    344 bytes, from 2015-05-20 20:03:07)
+- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp_common.xml      (   2849 bytes, from 2015-09-18 12:07:28)
+- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp5.xml            (  37194 bytes, from 2015-09-18 12:07:28)
+- /home/robclark/src/freedreno/envytools/rnndb/dsi/dsi.xml             (  27887 bytes, from 2015-10-22 16:34:52)
+- /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    602 bytes, from 2015-10-22 16:35:02)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/mmss_cc.xml         (   1686 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/hdmi/qfprom.xml         (    600 bytes, from 2015-05-20 20:03:07)
 - /home/robclark/src/freedreno/envytools/rnndb/hdmi/hdmi.xml           (  29154 bytes, from 2015-08-10 21:25:43)
@@ -45,7 +45,18 @@ WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
 */
 
 
-#define REG_SFPB_CFG						0x00000058
+enum sfpb_ahb_arb_master_port_en {
+	SFPB_MASTER_PORT_ENABLE = 3,
+	SFPB_MASTER_PORT_DISABLE = 0,
+};
+
+#define REG_SFPB_GPREG						0x00000058
+#define SFPB_GPREG_MASTER_PORT_EN__MASK				0x00001800
+#define SFPB_GPREG_MASTER_PORT_EN__SHIFT			11
+static inline uint32_t SFPB_GPREG_MASTER_PORT_EN(enum sfpb_ahb_arb_master_port_en val)
+{
+	return ((val) << SFPB_GPREG_MASTER_PORT_EN__SHIFT) & SFPB_GPREG_MASTER_PORT_EN__MASK;
+}
 
 
 #endif /* SFPB_XML */
diff --git a/drivers/gpu/drm/msm/edp/edp.xml.h b/drivers/gpu/drm/msm/edp/edp.xml.h
index bef1d65..90bf5ed 100644
--- a/drivers/gpu/drm/msm/edp/edp.xml.h
+++ b/drivers/gpu/drm/msm/edp/edp.xml.h
@@ -11,10 +11,10 @@ The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/msm.xml                 (    676 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1453 bytes, from 2015-05-20 20:03:07)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp4.xml            (  20915 bytes, from 2015-05-20 20:03:14)
-- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp_common.xml      (   2576 bytes, from 2015-07-09 22:10:24)
-- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp5.xml            (  36021 bytes, from 2015-07-09 22:10:24)
-- /home/robclark/src/freedreno/envytools/rnndb/dsi/dsi.xml             (  26057 bytes, from 2015-08-14 21:47:57)
-- /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    344 bytes, from 2015-05-20 20:03:07)
+- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp_common.xml      (   2849 bytes, from 2015-09-18 12:07:28)
+- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp5.xml            (  37194 bytes, from 2015-09-18 12:07:28)
+- /home/robclark/src/freedreno/envytools/rnndb/dsi/dsi.xml             (  27887 bytes, from 2015-10-22 16:34:52)
+- /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    602 bytes, from 2015-10-22 16:35:02)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/mmss_cc.xml         (   1686 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/hdmi/qfprom.xml         (    600 bytes, from 2015-05-20 20:03:07)
 - /home/robclark/src/freedreno/envytools/rnndb/hdmi/hdmi.xml           (  29154 bytes, from 2015-08-10 21:25:43)
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi.xml.h b/drivers/gpu/drm/msm/hdmi/hdmi.xml.h
index 0b1b558..10c4570 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi.xml.h
+++ b/drivers/gpu/drm/msm/hdmi/hdmi.xml.h
@@ -11,10 +11,10 @@ The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/msm.xml                 (    676 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1453 bytes, from 2015-05-20 20:03:07)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp4.xml            (  20915 bytes, from 2015-05-20 20:03:14)
-- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp_common.xml      (   2576 bytes, from 2015-07-09 22:10:24)
-- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp5.xml            (  36021 bytes, from 2015-07-09 22:10:24)
-- /home/robclark/src/freedreno/envytools/rnndb/dsi/dsi.xml             (  26057 bytes, from 2015-08-14 21:47:57)
-- /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    344 bytes, from 2015-05-20 20:03:07)
+- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp_common.xml      (   2849 bytes, from 2015-09-18 12:07:28)
+- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp5.xml            (  37194 bytes, from 2015-09-18 12:07:28)
+- /home/robclark/src/freedreno/envytools/rnndb/dsi/dsi.xml             (  27887 bytes, from 2015-10-22 16:34:52)
+- /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    602 bytes, from 2015-10-22 16:35:02)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/mmss_cc.xml         (   1686 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/hdmi/qfprom.xml         (    600 bytes, from 2015-05-20 20:03:07)
 - /home/robclark/src/freedreno/envytools/rnndb/hdmi/hdmi.xml           (  29154 bytes, from 2015-08-10 21:25:43)
diff --git a/drivers/gpu/drm/msm/hdmi/qfprom.xml.h b/drivers/gpu/drm/msm/hdmi/qfprom.xml.h
index 2aa23b9..dbd9cc4 100644
--- a/drivers/gpu/drm/msm/hdmi/qfprom.xml.h
+++ b/drivers/gpu/drm/msm/hdmi/qfprom.xml.h
@@ -11,10 +11,10 @@ The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/msm.xml                 (    676 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1453 bytes, from 2015-05-20 20:03:07)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp4.xml            (  20915 bytes, from 2015-05-20 20:03:14)
-- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp_common.xml      (   2576 bytes, from 2015-07-09 22:10:24)
-- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp5.xml            (  36021 bytes, from 2015-07-09 22:10:24)
-- /home/robclark/src/freedreno/envytools/rnndb/dsi/dsi.xml             (  26057 bytes, from 2015-08-14 21:47:57)
-- /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    344 bytes, from 2015-05-20 20:03:07)
+- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp_common.xml      (   2849 bytes, from 2015-09-18 12:07:28)
+- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp5.xml            (  37194 bytes, from 2015-09-18 12:07:28)
+- /home/robclark/src/freedreno/envytools/rnndb/dsi/dsi.xml             (  27887 bytes, from 2015-10-22 16:34:52)
+- /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    602 bytes, from 2015-10-22 16:35:02)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/mmss_cc.xml         (   1686 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/hdmi/qfprom.xml         (    600 bytes, from 2015-05-20 20:03:07)
 - /home/robclark/src/freedreno/envytools/rnndb/hdmi/hdmi.xml           (  29154 bytes, from 2015-08-10 21:25:43)
diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4.xml.h b/drivers/gpu/drm/msm/mdp/mdp4/mdp4.xml.h
index 74b8673..d5d9457 100644
--- a/drivers/gpu/drm/msm/mdp/mdp4/mdp4.xml.h
+++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4.xml.h
@@ -11,10 +11,10 @@ The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/msm.xml                 (    676 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1453 bytes, from 2015-05-20 20:03:07)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp4.xml            (  20915 bytes, from 2015-05-20 20:03:14)
-- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp_common.xml      (   2576 bytes, from 2015-07-09 22:10:24)
-- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp5.xml            (  36021 bytes, from 2015-07-09 22:10:24)
-- /home/robclark/src/freedreno/envytools/rnndb/dsi/dsi.xml             (  26057 bytes, from 2015-08-14 21:47:57)
-- /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    344 bytes, from 2015-05-20 20:03:07)
+- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp_common.xml      (   2849 bytes, from 2015-09-18 12:07:28)
+- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp5.xml            (  37194 bytes, from 2015-09-18 12:07:28)
+- /home/robclark/src/freedreno/envytools/rnndb/dsi/dsi.xml             (  27887 bytes, from 2015-10-22 16:34:52)
+- /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    602 bytes, from 2015-10-22 16:35:02)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/mmss_cc.xml         (   1686 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/hdmi/qfprom.xml         (    600 bytes, from 2015-05-20 20:03:07)
 - /home/robclark/src/freedreno/envytools/rnndb/hdmi/hdmi.xml           (  29154 bytes, from 2015-08-10 21:25:43)
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5.xml.h b/drivers/gpu/drm/msm/mdp/mdp5/mdp5.xml.h
index 3469f50..c37da9c 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5.xml.h
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5.xml.h
@@ -11,10 +11,10 @@ The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/msm.xml                 (    676 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1453 bytes, from 2015-05-20 20:03:07)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp4.xml            (  20915 bytes, from 2015-05-20 20:03:14)
-- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp_common.xml      (   2576 bytes, from 2015-07-09 22:10:24)
-- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp5.xml            (  36021 bytes, from 2015-07-09 22:10:24)
-- /home/robclark/src/freedreno/envytools/rnndb/dsi/dsi.xml             (  26057 bytes, from 2015-08-14 21:47:57)
-- /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    344 bytes, from 2015-05-20 20:03:07)
+- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp_common.xml      (   2849 bytes, from 2015-09-18 12:07:28)
+- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp5.xml            (  37194 bytes, from 2015-09-18 12:07:28)
+- /home/robclark/src/freedreno/envytools/rnndb/dsi/dsi.xml             (  27887 bytes, from 2015-10-22 16:34:52)
+- /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    602 bytes, from 2015-10-22 16:35:02)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/mmss_cc.xml         (   1686 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/hdmi/qfprom.xml         (    600 bytes, from 2015-05-20 20:03:07)
 - /home/robclark/src/freedreno/envytools/rnndb/hdmi/hdmi.xml           (  29154 bytes, from 2015-08-10 21:25:43)
@@ -895,6 +895,7 @@ static inline uint32_t MDP5_PIPE_SRC_OP_MODE_BWC(enum mdp5_pipe_bwc val)
 #define MDP5_PIPE_SRC_OP_MODE_IGC_ROM_1				0x00040000
 #define MDP5_PIPE_SRC_OP_MODE_DEINTERLACE			0x00400000
 #define MDP5_PIPE_SRC_OP_MODE_DEINTERLACE_ODD			0x00800000
+#define MDP5_PIPE_SRC_OP_MODE_SW_PIX_EXT_OVERRIDE		0x80000000
 
 static inline uint32_t REG_MDP5_PIPE_SRC_CONSTANT_COLOR(enum mdp5_pipe i0) { return 0x0000003c + __offset_PIPE(i0); }
 
@@ -932,6 +933,83 @@ static inline uint32_t MDP5_PIPE_DECIMATION_HORZ(uint32_t val)
 	return ((val) << MDP5_PIPE_DECIMATION_HORZ__SHIFT) & MDP5_PIPE_DECIMATION_HORZ__MASK;
 }
 
+static inline uint32_t __offset_SW_PIX_EXT(enum mdp_component_type idx)
+{
+	switch (idx) {
+		case COMP_0: return 0x00000100;
+		case COMP_1_2: return 0x00000110;
+		case COMP_3: return 0x00000120;
+		default: return INVALID_IDX(idx);
+	}
+}
+static inline uint32_t REG_MDP5_PIPE_SW_PIX_EXT(enum mdp5_pipe i0, enum mdp_component_type i1) { return 0x00000000 + __offset_PIPE(i0) + __offset_SW_PIX_EXT(i1); }
+
+static inline uint32_t REG_MDP5_PIPE_SW_PIX_EXT_LR(enum mdp5_pipe i0, enum mdp_component_type i1) { return 0x00000000 + __offset_PIPE(i0) + __offset_SW_PIX_EXT(i1); }
+#define MDP5_PIPE_SW_PIX_EXT_LR_LEFT_RPT__MASK			0x000000ff
+#define MDP5_PIPE_SW_PIX_EXT_LR_LEFT_RPT__SHIFT			0
+static inline uint32_t MDP5_PIPE_SW_PIX_EXT_LR_LEFT_RPT(uint32_t val)
+{
+	return ((val) << MDP5_PIPE_SW_PIX_EXT_LR_LEFT_RPT__SHIFT) & MDP5_PIPE_SW_PIX_EXT_LR_LEFT_RPT__MASK;
+}
+#define MDP5_PIPE_SW_PIX_EXT_LR_LEFT_OVF__MASK			0x0000ff00
+#define MDP5_PIPE_SW_PIX_EXT_LR_LEFT_OVF__SHIFT			8
+static inline uint32_t MDP5_PIPE_SW_PIX_EXT_LR_LEFT_OVF(int32_t val)
+{
+	return ((val) << MDP5_PIPE_SW_PIX_EXT_LR_LEFT_OVF__SHIFT) & MDP5_PIPE_SW_PIX_EXT_LR_LEFT_OVF__MASK;
+}
+#define MDP5_PIPE_SW_PIX_EXT_LR_RIGHT_RPT__MASK			0x00ff0000
+#define MDP5_PIPE_SW_PIX_EXT_LR_RIGHT_RPT__SHIFT		16
+static inline uint32_t MDP5_PIPE_SW_PIX_EXT_LR_RIGHT_RPT(uint32_t val)
+{
+	return ((val) << MDP5_PIPE_SW_PIX_EXT_LR_RIGHT_RPT__SHIFT) & MDP5_PIPE_SW_PIX_EXT_LR_RIGHT_RPT__MASK;
+}
+#define MDP5_PIPE_SW_PIX_EXT_LR_RIGHT_OVF__MASK			0xff000000
+#define MDP5_PIPE_SW_PIX_EXT_LR_RIGHT_OVF__SHIFT		24
+static inline uint32_t MDP5_PIPE_SW_PIX_EXT_LR_RIGHT_OVF(int32_t val)
+{
+	return ((val) << MDP5_PIPE_SW_PIX_EXT_LR_RIGHT_OVF__SHIFT) & MDP5_PIPE_SW_PIX_EXT_LR_RIGHT_OVF__MASK;
+}
+
+static inline uint32_t REG_MDP5_PIPE_SW_PIX_EXT_TB(enum mdp5_pipe i0, enum mdp_component_type i1) { return 0x00000004 + __offset_PIPE(i0) + __offset_SW_PIX_EXT(i1); }
+#define MDP5_PIPE_SW_PIX_EXT_TB_TOP_RPT__MASK			0x000000ff
+#define MDP5_PIPE_SW_PIX_EXT_TB_TOP_RPT__SHIFT			0
+static inline uint32_t MDP5_PIPE_SW_PIX_EXT_TB_TOP_RPT(uint32_t val)
+{
+	return ((val) << MDP5_PIPE_SW_PIX_EXT_TB_TOP_RPT__SHIFT) & MDP5_PIPE_SW_PIX_EXT_TB_TOP_RPT__MASK;
+}
+#define MDP5_PIPE_SW_PIX_EXT_TB_TOP_OVF__MASK			0x0000ff00
+#define MDP5_PIPE_SW_PIX_EXT_TB_TOP_OVF__SHIFT			8
+static inline uint32_t MDP5_PIPE_SW_PIX_EXT_TB_TOP_OVF(int32_t val)
+{
+	return ((val) << MDP5_PIPE_SW_PIX_EXT_TB_TOP_OVF__SHIFT) & MDP5_PIPE_SW_PIX_EXT_TB_TOP_OVF__MASK;
+}
+#define MDP5_PIPE_SW_PIX_EXT_TB_BOTTOM_RPT__MASK		0x00ff0000
+#define MDP5_PIPE_SW_PIX_EXT_TB_BOTTOM_RPT__SHIFT		16
+static inline uint32_t MDP5_PIPE_SW_PIX_EXT_TB_BOTTOM_RPT(uint32_t val)
+{
+	return ((val) << MDP5_PIPE_SW_PIX_EXT_TB_BOTTOM_RPT__SHIFT) & MDP5_PIPE_SW_PIX_EXT_TB_BOTTOM_RPT__MASK;
+}
+#define MDP5_PIPE_SW_PIX_EXT_TB_BOTTOM_OVF__MASK		0xff000000
+#define MDP5_PIPE_SW_PIX_EXT_TB_BOTTOM_OVF__SHIFT		24
+static inline uint32_t MDP5_PIPE_SW_PIX_EXT_TB_BOTTOM_OVF(int32_t val)
+{
+	return ((val) << MDP5_PIPE_SW_PIX_EXT_TB_BOTTOM_OVF__SHIFT) & MDP5_PIPE_SW_PIX_EXT_TB_BOTTOM_OVF__MASK;
+}
+
+static inline uint32_t REG_MDP5_PIPE_SW_PIX_EXT_REQ_PIXELS(enum mdp5_pipe i0, enum mdp_component_type i1) { return 0x00000008 + __offset_PIPE(i0) + __offset_SW_PIX_EXT(i1); }
+#define MDP5_PIPE_SW_PIX_EXT_REQ_PIXELS_LEFT_RIGHT__MASK	0x0000ffff
+#define MDP5_PIPE_SW_PIX_EXT_REQ_PIXELS_LEFT_RIGHT__SHIFT	0
+static inline uint32_t MDP5_PIPE_SW_PIX_EXT_REQ_PIXELS_LEFT_RIGHT(uint32_t val)
+{
+	return ((val) << MDP5_PIPE_SW_PIX_EXT_REQ_PIXELS_LEFT_RIGHT__SHIFT) & MDP5_PIPE_SW_PIX_EXT_REQ_PIXELS_LEFT_RIGHT__MASK;
+}
+#define MDP5_PIPE_SW_PIX_EXT_REQ_PIXELS_TOP_BOTTOM__MASK	0xffff0000
+#define MDP5_PIPE_SW_PIX_EXT_REQ_PIXELS_TOP_BOTTOM__SHIFT	16
+static inline uint32_t MDP5_PIPE_SW_PIX_EXT_REQ_PIXELS_TOP_BOTTOM(uint32_t val)
+{
+	return ((val) << MDP5_PIPE_SW_PIX_EXT_REQ_PIXELS_TOP_BOTTOM__SHIFT) & MDP5_PIPE_SW_PIX_EXT_REQ_PIXELS_TOP_BOTTOM__MASK;
+}
+
 static inline uint32_t REG_MDP5_PIPE_SCALE_CONFIG(enum mdp5_pipe i0) { return 0x00000204 + __offset_PIPE(i0); }
 #define MDP5_PIPE_SCALE_CONFIG_SCALEX_EN			0x00000001
 #define MDP5_PIPE_SCALE_CONFIG_SCALEY_EN			0x00000002
diff --git a/drivers/gpu/drm/msm/mdp/mdp_common.xml.h b/drivers/gpu/drm/msm/mdp/mdp_common.xml.h
index 4f792c4..0aec1ac 100644
--- a/drivers/gpu/drm/msm/mdp/mdp_common.xml.h
+++ b/drivers/gpu/drm/msm/mdp/mdp_common.xml.h
@@ -11,10 +11,10 @@ The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/msm.xml                 (    676 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1453 bytes, from 2015-05-20 20:03:07)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp4.xml            (  20915 bytes, from 2015-05-20 20:03:14)
-- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp_common.xml      (   2576 bytes, from 2015-07-09 22:10:24)
-- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp5.xml            (  36021 bytes, from 2015-07-09 22:10:24)
-- /home/robclark/src/freedreno/envytools/rnndb/dsi/dsi.xml             (  26057 bytes, from 2015-08-14 21:47:57)
-- /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    344 bytes, from 2015-05-20 20:03:07)
+- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp_common.xml      (   2849 bytes, from 2015-09-18 12:07:28)
+- /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp5.xml            (  37194 bytes, from 2015-09-18 12:07:28)
+- /home/robclark/src/freedreno/envytools/rnndb/dsi/dsi.xml             (  27887 bytes, from 2015-10-22 16:34:52)
+- /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    602 bytes, from 2015-10-22 16:35:02)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/mmss_cc.xml         (   1686 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/hdmi/qfprom.xml         (    600 bytes, from 2015-05-20 20:03:07)
 - /home/robclark/src/freedreno/envytools/rnndb/hdmi/hdmi.xml           (  29154 bytes, from 2015-08-10 21:25:43)
@@ -78,6 +78,13 @@ enum mdp_alpha_type {
 	BG_PIXEL = 3,
 };
 
+enum mdp_component_type {
+	COMP_0 = 0,
+	COMP_1_2 = 1,
+	COMP_3 = 2,
+	COMP_MAX = 3,
+};
+
 enum mdp_bpc {
 	BPC1 = 0,
 	BPC5 = 1,
-- 
cgit v0.10.2


From 556a76e51b5c8e16986e2cc0a5e14306a4e2505a Mon Sep 17 00:00:00 2001
From: Bjorn Andersson <bjorn.andersson@sonymobile.com>
Date: Tue, 18 Aug 2015 10:34:32 -0700
Subject: drm: msm: dsi: Don't attempt changing voltage of switches

In some configurations the supplies are voltage switches and not LDOs,
making the set voltage call to fail. Check with the regulator framework
if the supply can change voltage before attempting.

Signed-off-by: Bjorn Andersson <bjorn.andersson@sonymobile.com>
Reviewed-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>

diff --git a/drivers/gpu/drm/msm/dsi/dsi_host.c b/drivers/gpu/drm/msm/dsi/dsi_host.c
index 8d82973..4c49868 100644
--- a/drivers/gpu/drm/msm/dsi/dsi_host.c
+++ b/drivers/gpu/drm/msm/dsi/dsi_host.c
@@ -278,7 +278,7 @@ static int dsi_regulator_init(struct msm_dsi_host *msm_host)
 	}
 
 	for (i = 0; i < num; i++) {
-		if ((regs[i].min_voltage >= 0) && (regs[i].max_voltage >= 0)) {
+		if (regulator_can_change_voltage(s[i].consumer)) {
 			ret = regulator_set_voltage(s[i].consumer,
 				regs[i].min_voltage, regs[i].max_voltage);
 			if (ret < 0) {
diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c
index 401ff58..f1f955f 100644
--- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c
+++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c
@@ -178,7 +178,7 @@ static int dsi_phy_regulator_init(struct msm_dsi_phy *phy)
 	}
 
 	for (i = 0; i < num; i++) {
-		if ((regs[i].min_voltage >= 0) && (regs[i].max_voltage >= 0)) {
+		if (regulator_can_change_voltage(s[i].consumer)) {
 			ret = regulator_set_voltage(s[i].consumer,
 				regs[i].min_voltage, regs[i].max_voltage);
 			if (ret < 0) {
-- 
cgit v0.10.2


From e01b1bfd88f9c5ec32b471a5a696a79f45740e63 Mon Sep 17 00:00:00 2001
From: Hai Li <hali@codeaurora.org>
Date: Fri, 11 Sep 2015 15:56:09 -0400
Subject: drm/msm/dsi: Updata LNn_CFG4 register settings for 28nm PHY

The current settings for 28nm PHY data lane CFG4 registers do
not work with certain panels. This change is to modify them to
hw recommended values.

Signed-off-by: Hai Li <hali@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>

diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm.c
index f1a7c7b..edf7411 100644
--- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm.c
+++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm.c
@@ -99,16 +99,14 @@ static int dsi_28nm_phy_enable(struct msm_dsi_phy *phy, int src_pll_id,
 		dsi_phy_write(base + REG_DSI_28nm_PHY_LN_CFG_1(i), 0);
 		dsi_phy_write(base + REG_DSI_28nm_PHY_LN_CFG_2(i), 0);
 		dsi_phy_write(base + REG_DSI_28nm_PHY_LN_CFG_3(i), 0);
+		dsi_phy_write(base + REG_DSI_28nm_PHY_LN_CFG_4(i), 0);
 		dsi_phy_write(base + REG_DSI_28nm_PHY_LN_TEST_DATAPATH(i), 0);
 		dsi_phy_write(base + REG_DSI_28nm_PHY_LN_DEBUG_SEL(i), 0);
 		dsi_phy_write(base + REG_DSI_28nm_PHY_LN_TEST_STR_0(i), 0x1);
 		dsi_phy_write(base + REG_DSI_28nm_PHY_LN_TEST_STR_1(i), 0x97);
 	}
-	dsi_phy_write(base + REG_DSI_28nm_PHY_LN_CFG_4(0), 0);
-	dsi_phy_write(base + REG_DSI_28nm_PHY_LN_CFG_4(1), 0x5);
-	dsi_phy_write(base + REG_DSI_28nm_PHY_LN_CFG_4(2), 0xa);
-	dsi_phy_write(base + REG_DSI_28nm_PHY_LN_CFG_4(3), 0xf);
 
+	dsi_phy_write(base + REG_DSI_28nm_PHY_LNCK_CFG_4, 0);
 	dsi_phy_write(base + REG_DSI_28nm_PHY_LNCK_CFG_1, 0xc0);
 	dsi_phy_write(base + REG_DSI_28nm_PHY_LNCK_TEST_STR0, 0x1);
 	dsi_phy_write(base + REG_DSI_28nm_PHY_LNCK_TEST_STR1, 0xbb);
-- 
cgit v0.10.2


From 60fb49ca52ec304a937bcee14ae6acbe1e75562b Mon Sep 17 00:00:00 2001
From: Stephane Viau <sviau@codeaurora.org>
Date: Tue, 15 Sep 2015 08:41:44 -0400
Subject: drm/msm/mdp5: remove the cfg pointer from SMP struct

We want to make sure we control all the information being passed
down to SMP block. Having access to the cfg pointer here may create
bad things in the future.

Signed-off-by: Stephane Viau <sviau@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>

diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_cfg.h b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_cfg.h
index efb918d..a6facaf 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_cfg.h
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_cfg.h
@@ -61,7 +61,7 @@ struct mdp5_smp_block {
 	int mmb_size;			/* MMB: size in bytes */
 	uint32_t clients[MAX_CLIENTS];	/* SMP port allocation /pipe */
 	mdp5_smp_state_t reserved_state;/* SMP MMBs statically allocated */
-	int reserved[MAX_CLIENTS];	/* # of MMBs allocated per client */
+	uint8_t reserved[MAX_CLIENTS];	/* # of MMBs allocated per client */
 };
 
 #define MDP5_INTF_NUM_MAX	5
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.c
index 563cca9..6f425c2 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.c
@@ -90,7 +90,7 @@
 struct mdp5_smp {
 	struct drm_device *dev;
 
-	const struct mdp5_smp_block *cfg;
+	uint8_t reserved[MAX_CLIENTS]; /* fixed MMBs allocation per client */
 
 	int blk_cnt;
 	int blk_size;
@@ -141,10 +141,10 @@ static int smp_request_block(struct mdp5_smp *smp,
 	struct mdp5_kms *mdp5_kms = get_kms(smp);
 	struct mdp5_client_smp_state *ps = &smp->client_state[cid];
 	int i, ret, avail, cur_nblks, cnt = smp->blk_cnt;
-	int reserved;
+	uint8_t reserved;
 	unsigned long flags;
 
-	reserved = smp->cfg->reserved[cid];
+	reserved = smp->reserved[cid];
 
 	spin_lock_irqsave(&smp->state_lock, flags);
 
@@ -405,12 +405,12 @@ struct mdp5_smp *mdp5_smp_init(struct drm_device *dev, const struct mdp5_smp_blo
 	}
 
 	smp->dev = dev;
-	smp->cfg = cfg;
 	smp->blk_cnt = cfg->mmb_count;
 	smp->blk_size = cfg->mmb_size;
 
 	/* statically tied MMBs cannot be re-allocated: */
 	bitmap_copy(smp->state, cfg->reserved_state, smp->blk_cnt);
+	memcpy(smp->reserved, cfg->reserved, sizeof(smp->reserved));
 	spin_lock_init(&smp->state_lock);
 
 	return smp;
-- 
cgit v0.10.2


From 5e921b19268d87a0c64f3181c0d3ea9df9eea972 Mon Sep 17 00:00:00 2001
From: Stephane Viau <sviau@codeaurora.org>
Date: Tue, 15 Sep 2015 08:41:46 -0400
Subject: drm/msm: Fix IOMMU clean up path in case msm_iommu_new() fails

msm_iommu_new() can fail and this change makes sure that we
detect the failure and free the allocated domain before going
any further.

Signed-off-by: Stephane Viau <sviau@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>

diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c
index 047cb04..7dbdced 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c
@@ -586,6 +586,7 @@ struct msm_kms *mdp5_kms_init(struct drm_device *dev)
 		if (IS_ERR(mmu)) {
 			ret = PTR_ERR(mmu);
 			dev_err(dev->dev, "failed to init iommu: %d\n", ret);
+			iommu_domain_free(config->platform.iommu);
 			goto fail;
 		}
 
diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c
index 8f70d92..6b02ada 100644
--- a/drivers/gpu/drm/msm/msm_gpu.c
+++ b/drivers/gpu/drm/msm/msm_gpu.c
@@ -651,6 +651,14 @@ int msm_gpu_init(struct drm_device *drm, struct platform_device *pdev,
 	if (iommu) {
 		dev_info(drm->dev, "%s: using IOMMU\n", name);
 		gpu->mmu = msm_iommu_new(&pdev->dev, iommu);
+		if (IS_ERR(gpu->mmu)) {
+			ret = PTR_ERR(gpu->mmu);
+			dev_err(drm->dev, "failed to init iommu: %d\n", ret);
+			gpu->mmu = NULL;
+			iommu_domain_free(iommu);
+			goto fail;
+		}
+
 	} else {
 		dev_info(drm->dev, "%s: no IOMMU, fallback to VRAM carveout!\n", name);
 	}
-- 
cgit v0.10.2


From d40325b43750eb4a8a81240e36fe38e9ccbcc9eb Mon Sep 17 00:00:00 2001
From: Stephane Viau <sviau@codeaurora.org>
Date: Tue, 15 Sep 2015 08:41:47 -0400
Subject: drm/msm/mdp5: Avoid printing error messages for optional clocks

The current behavior is to try to get optional clocks and print a
dev_err message in case of failure. This looks rather confusing
and may increase with the amount of optional clocks.

We may need a cleaner way to handle per-device clocks but in the
meantime, let's reduce the amount of dev_err messages during the
probe.

Signed-off-by: Stephane Viau <sviau@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>

diff --git a/Documentation/devicetree/bindings/drm/msm/mdp.txt b/Documentation/devicetree/bindings/drm/msm/mdp.txt
index 1a0598e..0833eda 100644
--- a/Documentation/devicetree/bindings/drm/msm/mdp.txt
+++ b/Documentation/devicetree/bindings/drm/msm/mdp.txt
@@ -11,13 +11,14 @@ Required properties:
 - clock-names: the following clocks are required:
   * "core_clk"
   * "iface_clk"
-  * "lut_clk"
   * "src_clk"
   * "hdmi_clk"
   * "mpd_clk"
 
 Optional properties:
 - gpus: phandle for gpu device
+- clock-names: the following clocks are optional:
+  * "lut_clk"
 
 Example:
 
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c
index 7dbdced..34c4ad5 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c
@@ -452,15 +452,19 @@ static void read_hw_revision(struct mdp5_kms *mdp5_kms,
 }
 
 static int get_clk(struct platform_device *pdev, struct clk **clkp,
-		const char *name)
+		const char *name, bool mandatory)
 {
 	struct device *dev = &pdev->dev;
 	struct clk *clk = devm_clk_get(dev, name);
-	if (IS_ERR(clk)) {
+	if (IS_ERR(clk) && mandatory) {
 		dev_err(dev, "failed to get %s (%ld)\n", name, PTR_ERR(clk));
 		return PTR_ERR(clk);
 	}
-	*clkp = clk;
+	if (IS_ERR(clk))
+		DBG("skipping %s", name);
+	else
+		*clkp = clk;
+
 	return 0;
 }
 
@@ -514,25 +518,26 @@ struct msm_kms *mdp5_kms_init(struct drm_device *dev)
 		goto fail;
 	}
 
-	ret = get_clk(pdev, &mdp5_kms->axi_clk, "bus_clk");
+	/* mandatory clocks: */
+	ret = get_clk(pdev, &mdp5_kms->axi_clk, "bus_clk", true);
 	if (ret)
 		goto fail;
-	ret = get_clk(pdev, &mdp5_kms->ahb_clk, "iface_clk");
+	ret = get_clk(pdev, &mdp5_kms->ahb_clk, "iface_clk", true);
 	if (ret)
 		goto fail;
-	ret = get_clk(pdev, &mdp5_kms->src_clk, "core_clk_src");
+	ret = get_clk(pdev, &mdp5_kms->src_clk, "core_clk_src", true);
 	if (ret)
 		goto fail;
-	ret = get_clk(pdev, &mdp5_kms->core_clk, "core_clk");
+	ret = get_clk(pdev, &mdp5_kms->core_clk, "core_clk", true);
 	if (ret)
 		goto fail;
-	ret = get_clk(pdev, &mdp5_kms->lut_clk, "lut_clk");
-	if (ret)
-		DBG("failed to get (optional) lut_clk clock");
-	ret = get_clk(pdev, &mdp5_kms->vsync_clk, "vsync_clk");
+	ret = get_clk(pdev, &mdp5_kms->vsync_clk, "vsync_clk", true);
 	if (ret)
 		goto fail;
 
+	/* optional clocks: */
+	get_clk(pdev, &mdp5_kms->lut_clk, "lut_clk", false);
+
 	/* we need to set a default rate before enabling.  Set a safe
 	 * rate first, then figure out hw revision, and then set a
 	 * more optimal rate:
-- 
cgit v0.10.2


From 0afbe59edd3fd3618da67b4835b57ebcc4ac3abd Mon Sep 17 00:00:00 2001
From: Stephane Viau <sviau@codeaurora.org>
Date: Tue, 15 Sep 2015 08:41:49 -0400
Subject: drm/msm/hdmi: Add basic HDMI support for msm8996

The HDMI controller is new in MDP5 v1.7. As of now, this change
doesn't reflect the novelty and only adds the basics so the probe
gets triggered.

Signed-off-by: Stephane Viau <sviau@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>

diff --git a/Documentation/devicetree/bindings/drm/msm/hdmi.txt b/Documentation/devicetree/bindings/drm/msm/hdmi.txt
index e926239..379ee2e 100644
--- a/Documentation/devicetree/bindings/drm/msm/hdmi.txt
+++ b/Documentation/devicetree/bindings/drm/msm/hdmi.txt
@@ -2,6 +2,7 @@ Qualcomm adreno/snapdragon hdmi output
 
 Required properties:
 - compatible: one of the following
+   * "qcom,hdmi-tx-8996"
    * "qcom,hdmi-tx-8994"
    * "qcom,hdmi-tx-8084"
    * "qcom,hdmi-tx-8974"
@@ -21,6 +22,7 @@ Required properties:
 Optional properties:
 - qcom,hdmi-tx-mux-en-gpio: hdmi mux enable pin
 - qcom,hdmi-tx-mux-sel-gpio: hdmi mux select pin
+- power-domains: reference to the power domain(s), if available.
 - pinctrl-names: the pin control state names; should contain "default"
 - pinctrl-0: the default pinctrl state (active)
 - pinctrl-1: the "sleep" pinctrl state
@@ -35,6 +37,7 @@ Example:
 		reg-names = "core_physical";
 		reg = <0x04a00000 0x1000>;
 		interrupts = <GIC_SPI 79 0>;
+		power-domains = <&mmcc MDSS_GDSC>;
 		clock-names =
 		    "core_clk",
 		    "master_iface_clk",
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi.c b/drivers/gpu/drm/msm/hdmi/hdmi.c
index 101b324..1f4a95e 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi.c
+++ b/drivers/gpu/drm/msm/hdmi/hdmi.c
@@ -328,6 +328,9 @@ fail:
 	.item ## _names = item ##_names_ ## entry, \
 	.item ## _cnt   = ARRAY_SIZE(item ## _names_ ## entry)
 
+static const char *pwr_reg_names_none[] = {};
+static const char *hpd_reg_names_none[] = {};
+
 static struct hdmi_platform_config hdmi_tx_8660_config = {
 		.phy_init = hdmi_phy_8x60_init,
 };
@@ -367,18 +370,26 @@ static struct hdmi_platform_config hdmi_tx_8084_config = {
 		.hpd_freq      = hpd_clk_freq_8x74,
 };
 
-static const char *hpd_reg_names_8x94[] = {};
-
 static struct hdmi_platform_config hdmi_tx_8994_config = {
 		.phy_init = NULL, /* nothing to do for this HDMI PHY 20nm */
 		HDMI_CFG(pwr_reg, 8x74),
-		HDMI_CFG(hpd_reg, 8x94),
+		HDMI_CFG(hpd_reg, none),
+		HDMI_CFG(pwr_clk, 8x74),
+		HDMI_CFG(hpd_clk, 8x74),
+		.hpd_freq      = hpd_clk_freq_8x74,
+};
+
+static struct hdmi_platform_config hdmi_tx_8996_config = {
+		.phy_init = NULL,
+		HDMI_CFG(pwr_reg, none),
+		HDMI_CFG(hpd_reg, none),
 		HDMI_CFG(pwr_clk, 8x74),
 		HDMI_CFG(hpd_clk, 8x74),
 		.hpd_freq      = hpd_clk_freq_8x74,
 };
 
 static const struct of_device_id dt_match[] = {
+	{ .compatible = "qcom,hdmi-tx-8996", .data = &hdmi_tx_8996_config },
 	{ .compatible = "qcom,hdmi-tx-8994", .data = &hdmi_tx_8994_config },
 	{ .compatible = "qcom,hdmi-tx-8084", .data = &hdmi_tx_8084_config },
 	{ .compatible = "qcom,hdmi-tx-8974", .data = &hdmi_tx_8974_config },
-- 
cgit v0.10.2


From 95651cd926bfa37036d7011a65421ab0db1b4959 Mon Sep 17 00:00:00 2001
From: Stephane Viau <sviau@codeaurora.org>
Date: Tue, 15 Sep 2015 08:41:51 -0400
Subject: drm/msm/mdp5: Use the newly introduced enum mdp_component_type

When calculating phase steps, let's use the same enum
mdp_component_type in order to ease the readability; 0/1 indexes
are a bit confusing and we now have explicit values to index
this type of arrays.

Signed-off-by: Stephane Viau <sviau@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>

diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c
index a0f5ff0..5a11fc4 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c
@@ -500,7 +500,7 @@ static int calc_phase_step(uint32_t src, uint32_t dst, uint32_t *out_phase)
 
 static int calc_scalex_steps(struct drm_plane *plane,
 		uint32_t pixel_format, uint32_t src, uint32_t dest,
-		uint32_t phasex_steps[2])
+		uint32_t phasex_steps[COMP_MAX])
 {
 	struct mdp5_kms *mdp5_kms = get_kms(plane);
 	struct device *dev = mdp5_kms->dev->dev;
@@ -516,15 +516,16 @@ static int calc_scalex_steps(struct drm_plane *plane,
 
 	hsub = drm_format_horz_chroma_subsampling(pixel_format);
 
-	phasex_steps[0] = phasex_step;
-	phasex_steps[1] = phasex_step / hsub;
+	phasex_steps[COMP_0]   = phasex_step;
+	phasex_steps[COMP_3]   = phasex_step;
+	phasex_steps[COMP_1_2] = phasex_step / hsub;
 
 	return 0;
 }
 
 static int calc_scaley_steps(struct drm_plane *plane,
 		uint32_t pixel_format, uint32_t src, uint32_t dest,
-		uint32_t phasey_steps[2])
+		uint32_t phasey_steps[COMP_MAX])
 {
 	struct mdp5_kms *mdp5_kms = get_kms(plane);
 	struct device *dev = mdp5_kms->dev->dev;
@@ -540,8 +541,9 @@ static int calc_scaley_steps(struct drm_plane *plane,
 
 	vsub = drm_format_vert_chroma_subsampling(pixel_format);
 
-	phasey_steps[0] = phasey_step;
-	phasey_steps[1] = phasey_step / vsub;
+	phasey_steps[COMP_0]   = phasey_step;
+	phasey_steps[COMP_3]   = phasey_step;
+	phasey_steps[COMP_1_2] = phasey_step / vsub;
 
 	return 0;
 }
@@ -593,8 +595,7 @@ static int mdp5_plane_mode_set(struct drm_plane *plane,
 	enum mdp5_pipe pipe = mdp5_plane->pipe;
 	const struct mdp_format *format;
 	uint32_t nplanes, config = 0;
-	/* below array -> index 0: comp 0/3 ; index 1: comp 1/2 */
-	uint32_t phasex_step[2] = {0,}, phasey_step[2] = {0,};
+	uint32_t phasex_step[COMP_MAX] = {0,}, phasey_step[COMP_MAX] = {0,};
 	uint32_t hdecm = 0, vdecm = 0;
 	uint32_t pix_format;
 	bool vflip, hflip;
@@ -702,13 +703,13 @@ static int mdp5_plane_mode_set(struct drm_plane *plane,
 
 	if (mdp5_plane->caps & MDP_PIPE_CAP_SCALE) {
 		mdp5_write(mdp5_kms, REG_MDP5_PIPE_SCALE_PHASE_STEP_X(pipe),
-				phasex_step[0]);
+				phasex_step[COMP_0]);
 		mdp5_write(mdp5_kms, REG_MDP5_PIPE_SCALE_PHASE_STEP_Y(pipe),
-				phasey_step[0]);
+				phasey_step[COMP_0]);
 		mdp5_write(mdp5_kms, REG_MDP5_PIPE_SCALE_CR_PHASE_STEP_X(pipe),
-				phasex_step[1]);
+				phasex_step[COMP_1_2]);
 		mdp5_write(mdp5_kms, REG_MDP5_PIPE_SCALE_CR_PHASE_STEP_Y(pipe),
-				phasey_step[1]);
+				phasey_step[COMP_1_2]);
 		mdp5_write(mdp5_kms, REG_MDP5_PIPE_DECIMATION(pipe),
 				MDP5_PIPE_DECIMATION_VERT(vdecm) |
 				MDP5_PIPE_DECIMATION_HORZ(hdecm));
-- 
cgit v0.10.2


From 8e2930c6cfb45acc00da9bd6779f5fdd6ec4cb03 Mon Sep 17 00:00:00 2001
From: Stephane Viau <sviau@codeaurora.org>
Date: Tue, 15 Sep 2015 08:41:52 -0400
Subject: drm/msm/mdp: Add Software Pixel Extension support

In order to produce an image, the scalar needs to be fed extra
pixels. These top/bottom/left/right values depend on a various of
factors, including resolution, scaling type, phase step and
initial phase.

Pixel Extension are programmed by hardware in most targets - and
can be overwritten by software. For some targets (e.g.: msm8996),
software *must* program those registers.

In order to ease this computation, let's always use bilinear
filters, which are easier to program from kernel. Eventually,
all of these values will come down from user space for better
quality.

Signed-off-by: Stephane Viau <sviau@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>

diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c
index 5a11fc4..d93185e 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c
@@ -548,40 +548,120 @@ static int calc_scaley_steps(struct drm_plane *plane,
 	return 0;
 }
 
-static uint32_t get_scale_config(enum mdp_chroma_samp_type chroma_sample,
-		uint32_t src, uint32_t dest, bool hor)
+static uint32_t get_scale_config(const struct mdp_format *format,
+		uint32_t src, uint32_t dst, bool horz)
 {
-	uint32_t y_filter =   (src <= dest) ? SCALE_FILTER_CA  : SCALE_FILTER_PCMN;
-	uint32_t y_a_filter = (src <= dest) ? SCALE_FILTER_BIL : SCALE_FILTER_PCMN;
-	uint32_t uv_filter = ((src / 2) <= dest) ? /* 2x upsample */
-			SCALE_FILTER_BIL : SCALE_FILTER_PCMN;
-	uint32_t value = 0;
-
-	if (chroma_sample == CHROMA_420 || chroma_sample == CHROMA_H2V1) {
-		if (hor)
-			value = MDP5_PIPE_SCALE_CONFIG_SCALEX_EN |
-				MDP5_PIPE_SCALE_CONFIG_SCALEX_FILTER_COMP_0(y_filter) |
-				MDP5_PIPE_SCALE_CONFIG_SCALEX_FILTER_COMP_3(y_a_filter) |
-				MDP5_PIPE_SCALE_CONFIG_SCALEX_FILTER_COMP_1_2(uv_filter);
-		else
-			value = MDP5_PIPE_SCALE_CONFIG_SCALEY_EN |
-				MDP5_PIPE_SCALE_CONFIG_SCALEY_FILTER_COMP_0(y_filter) |
-				MDP5_PIPE_SCALE_CONFIG_SCALEY_FILTER_COMP_3(y_a_filter) |
-				MDP5_PIPE_SCALE_CONFIG_SCALEY_FILTER_COMP_1_2(uv_filter);
-	} else if (src != dest) {
-		if (hor)
-			value = MDP5_PIPE_SCALE_CONFIG_SCALEX_EN |
-				MDP5_PIPE_SCALE_CONFIG_SCALEX_FILTER_COMP_0(y_a_filter) |
-				MDP5_PIPE_SCALE_CONFIG_SCALEX_FILTER_COMP_3(y_a_filter);
-		else
-			value = MDP5_PIPE_SCALE_CONFIG_SCALEY_EN |
-				MDP5_PIPE_SCALE_CONFIG_SCALEY_FILTER_COMP_0(y_a_filter) |
-				MDP5_PIPE_SCALE_CONFIG_SCALEY_FILTER_COMP_3(y_a_filter);
+	bool scaling = format->is_yuv ? true : (src != dst);
+	uint32_t sub, pix_fmt = format->base.pixel_format;
+	uint32_t ya_filter, uv_filter;
+	bool yuv = format->is_yuv;
+
+	if (!scaling)
+		return 0;
+
+	if (yuv) {
+		sub = horz ? drm_format_horz_chroma_subsampling(pix_fmt) :
+			     drm_format_vert_chroma_subsampling(pix_fmt);
+		uv_filter = ((src / sub) <= dst) ?
+				   SCALE_FILTER_BIL : SCALE_FILTER_PCMN;
+	}
+	ya_filter = (src <= dst) ? SCALE_FILTER_BIL : SCALE_FILTER_PCMN;
+
+	if (horz)
+		return  MDP5_PIPE_SCALE_CONFIG_SCALEX_EN |
+			MDP5_PIPE_SCALE_CONFIG_SCALEX_FILTER_COMP_0(ya_filter) |
+			MDP5_PIPE_SCALE_CONFIG_SCALEX_FILTER_COMP_3(ya_filter) |
+			COND(yuv, MDP5_PIPE_SCALE_CONFIG_SCALEX_FILTER_COMP_1_2(uv_filter));
+	else
+		return  MDP5_PIPE_SCALE_CONFIG_SCALEY_EN |
+			MDP5_PIPE_SCALE_CONFIG_SCALEY_FILTER_COMP_0(ya_filter) |
+			MDP5_PIPE_SCALE_CONFIG_SCALEY_FILTER_COMP_3(ya_filter) |
+			COND(yuv, MDP5_PIPE_SCALE_CONFIG_SCALEY_FILTER_COMP_1_2(uv_filter));
+}
+
+static void calc_pixel_ext(const struct mdp_format *format,
+		uint32_t src, uint32_t dst, uint32_t phase_step[2],
+		int pix_ext_edge1[COMP_MAX], int pix_ext_edge2[COMP_MAX],
+		bool horz)
+{
+	bool scaling = format->is_yuv ? true : (src != dst);
+	int i;
+
+	/*
+	 * Note:
+	 * We assume here that:
+	 *     1. PCMN filter is used for downscale
+	 *     2. bilinear filter is used for upscale
+	 *     3. we are in a single pipe configuration
+	 */
+
+	for (i = 0; i < COMP_MAX; i++) {
+		pix_ext_edge1[i] = 0;
+		pix_ext_edge2[i] = scaling ? 1 : 0;
 	}
+}
 
-	return value;
+static void mdp5_write_pixel_ext(struct mdp5_kms *mdp5_kms, enum mdp5_pipe pipe,
+	const struct mdp_format *format,
+	uint32_t src_w, int pe_left[COMP_MAX], int pe_right[COMP_MAX],
+	uint32_t src_h, int pe_top[COMP_MAX], int pe_bottom[COMP_MAX])
+{
+	uint32_t pix_fmt = format->base.pixel_format;
+	uint32_t lr, tb, req;
+	int i;
+
+	for (i = 0; i < COMP_MAX; i++) {
+		uint32_t roi_w = src_w;
+		uint32_t roi_h = src_h;
+
+		if (format->is_yuv && i == COMP_1_2) {
+			roi_w /= drm_format_horz_chroma_subsampling(pix_fmt);
+			roi_h /= drm_format_vert_chroma_subsampling(pix_fmt);
+		}
+
+		lr  = (pe_left[i] >= 0) ?
+			MDP5_PIPE_SW_PIX_EXT_LR_LEFT_RPT(pe_left[i]) :
+			MDP5_PIPE_SW_PIX_EXT_LR_LEFT_OVF(pe_left[i]);
+
+		lr |= (pe_right[i] >= 0) ?
+			MDP5_PIPE_SW_PIX_EXT_LR_RIGHT_RPT(pe_right[i]) :
+			MDP5_PIPE_SW_PIX_EXT_LR_RIGHT_OVF(pe_right[i]);
+
+		tb  = (pe_top[i] >= 0) ?
+			MDP5_PIPE_SW_PIX_EXT_TB_TOP_RPT(pe_top[i]) :
+			MDP5_PIPE_SW_PIX_EXT_TB_TOP_OVF(pe_top[i]);
+
+		tb |= (pe_bottom[i] >= 0) ?
+			MDP5_PIPE_SW_PIX_EXT_TB_BOTTOM_RPT(pe_bottom[i]) :
+			MDP5_PIPE_SW_PIX_EXT_TB_BOTTOM_OVF(pe_bottom[i]);
+
+		req  = MDP5_PIPE_SW_PIX_EXT_REQ_PIXELS_LEFT_RIGHT(roi_w +
+				pe_left[i] + pe_right[i]);
+
+		req |= MDP5_PIPE_SW_PIX_EXT_REQ_PIXELS_TOP_BOTTOM(roi_h +
+				pe_top[i] + pe_bottom[i]);
+
+		mdp5_write(mdp5_kms, REG_MDP5_PIPE_SW_PIX_EXT_LR(pipe, i), lr);
+		mdp5_write(mdp5_kms, REG_MDP5_PIPE_SW_PIX_EXT_TB(pipe, i), tb);
+		mdp5_write(mdp5_kms, REG_MDP5_PIPE_SW_PIX_EXT_REQ_PIXELS(pipe, i), req);
+
+		DBG("comp-%d (L/R): rpt=%d/%d, ovf=%d/%d, req=%d", i,
+			FIELD(lr,  MDP5_PIPE_SW_PIX_EXT_LR_LEFT_RPT),
+			FIELD(lr,  MDP5_PIPE_SW_PIX_EXT_LR_RIGHT_RPT),
+			FIELD(lr,  MDP5_PIPE_SW_PIX_EXT_LR_LEFT_OVF),
+			FIELD(lr,  MDP5_PIPE_SW_PIX_EXT_LR_RIGHT_OVF),
+			FIELD(req, MDP5_PIPE_SW_PIX_EXT_REQ_PIXELS_LEFT_RIGHT));
+
+		DBG("comp-%d (T/B): rpt=%d/%d, ovf=%d/%d, req=%d", i,
+			FIELD(tb,  MDP5_PIPE_SW_PIX_EXT_TB_TOP_RPT),
+			FIELD(tb,  MDP5_PIPE_SW_PIX_EXT_TB_BOTTOM_RPT),
+			FIELD(tb,  MDP5_PIPE_SW_PIX_EXT_TB_TOP_OVF),
+			FIELD(tb,  MDP5_PIPE_SW_PIX_EXT_TB_BOTTOM_OVF),
+			FIELD(req, MDP5_PIPE_SW_PIX_EXT_REQ_PIXELS_TOP_BOTTOM));
+	}
 }
 
+
 static int mdp5_plane_mode_set(struct drm_plane *plane,
 		struct drm_crtc *crtc, struct drm_framebuffer *fb,
 		int crtc_x, int crtc_y,
@@ -596,6 +676,9 @@ static int mdp5_plane_mode_set(struct drm_plane *plane,
 	const struct mdp_format *format;
 	uint32_t nplanes, config = 0;
 	uint32_t phasex_step[COMP_MAX] = {0,}, phasey_step[COMP_MAX] = {0,};
+	bool pe = mdp5_plane->caps & MDP_PIPE_CAP_SW_PIX_EXT;
+	int pe_left[COMP_MAX], pe_right[COMP_MAX];
+	int pe_top[COMP_MAX], pe_bottom[COMP_MAX];
 	uint32_t hdecm = 0, vdecm = 0;
 	uint32_t pix_format;
 	bool vflip, hflip;
@@ -643,11 +726,18 @@ static int mdp5_plane_mode_set(struct drm_plane *plane,
 	if (ret)
 		return ret;
 
+	if (mdp5_plane->caps & MDP_PIPE_CAP_SW_PIX_EXT) {
+		calc_pixel_ext(format, src_w, crtc_w, phasex_step,
+					 pe_left, pe_right, true);
+		calc_pixel_ext(format, src_h, crtc_h, phasey_step,
+					pe_top, pe_bottom, false);
+	}
+
 	/* TODO calc hdecm, vdecm */
 
 	/* SCALE is used to both scale and up-sample chroma components */
-	config |= get_scale_config(format->chroma_sample, src_w, crtc_w, true);
-	config |= get_scale_config(format->chroma_sample, src_h, crtc_h, false);
+	config |= get_scale_config(format, src_w, crtc_w, true);
+	config |= get_scale_config(format, src_h, crtc_h, false);
 	DBG("scale config = %x", config);
 
 	hflip = !!(pstate->rotation & BIT(DRM_REFLECT_X));
@@ -696,11 +786,17 @@ static int mdp5_plane_mode_set(struct drm_plane *plane,
 	mdp5_write(mdp5_kms, REG_MDP5_PIPE_SRC_OP_MODE(pipe),
 			(hflip ? MDP5_PIPE_SRC_OP_MODE_FLIP_LR : 0) |
 			(vflip ? MDP5_PIPE_SRC_OP_MODE_FLIP_UD : 0) |
+			COND(pe, MDP5_PIPE_SRC_OP_MODE_SW_PIX_EXT_OVERRIDE) |
 			MDP5_PIPE_SRC_OP_MODE_BWC(BWC_LOSSLESS));
 
 	/* not using secure mode: */
 	mdp5_write(mdp5_kms, REG_MDP5_PIPE_SRC_ADDR_SW_STATUS(pipe), 0);
 
+	if (mdp5_plane->caps & MDP_PIPE_CAP_SW_PIX_EXT)
+		mdp5_write_pixel_ext(mdp5_kms, pipe, format,
+				src_w, pe_left, pe_right,
+				src_h, pe_top, pe_bottom);
+
 	if (mdp5_plane->caps & MDP_PIPE_CAP_SCALE) {
 		mdp5_write(mdp5_kms, REG_MDP5_PIPE_SCALE_PHASE_STEP_X(pipe),
 				phasex_step[COMP_0]);
diff --git a/drivers/gpu/drm/msm/mdp/mdp_kms.h b/drivers/gpu/drm/msm/mdp/mdp_kms.h
index 46a94e7..0af2a54 100644
--- a/drivers/gpu/drm/msm/mdp/mdp_kms.h
+++ b/drivers/gpu/drm/msm/mdp/mdp_kms.h
@@ -106,6 +106,7 @@ const struct msm_format *mdp_get_format(struct msm_kms *kms, uint32_t format);
 #define MDP_PIPE_CAP_SCALE			BIT(2)
 #define MDP_PIPE_CAP_CSC			BIT(3)
 #define MDP_PIPE_CAP_DECIMATION			BIT(4)
+#define MDP_PIPE_CAP_SW_PIX_EXT			BIT(5)
 
 static inline bool pipe_supports_yuv(uint32_t pipe_caps)
 {
-- 
cgit v0.10.2


From d879eb5ad5d9e80cfc818c22fdd4335ebb5de063 Mon Sep 17 00:00:00 2001
From: Stephane Viau <sviau@codeaurora.org>
Date: Tue, 15 Sep 2015 08:41:53 -0400
Subject: drm/msm/mdp5: Basic support for MDP5 v1.7 (MSM8996)

This change adds the basic MDP5 support for MSM8996.

Signed-off-by: Stephane Viau <sviau@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>

diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_cfg.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_cfg.c
index a1e26f2..bb1225a 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_cfg.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_cfg.c
@@ -27,6 +27,8 @@ const struct mdp5_cfg_hw msm8x74v1_config = {
 	.mdp = {
 		.count = 1,
 		.base = { 0x00100 },
+		.caps = MDP_CAP_SMP |
+			0,
 	},
 	.smp = {
 		.mmb_count = 22,
@@ -96,6 +98,8 @@ const struct mdp5_cfg_hw msm8x74v2_config = {
 	.mdp = {
 		.count = 1,
 		.base = { 0x00100 },
+		.caps = MDP_CAP_SMP |
+			0,
 	},
 	.smp = {
 		.mmb_count = 22,
@@ -165,6 +169,8 @@ const struct mdp5_cfg_hw apq8084_config = {
 	.mdp = {
 		.count = 1,
 		.base = { 0x00100 },
+		.caps = MDP_CAP_SMP |
+			0,
 	},
 	.smp = {
 		.mmb_count = 44,
@@ -242,6 +248,8 @@ const struct mdp5_cfg_hw msm8x16_config = {
 	.mdp = {
 		.count = 1,
 		.base = { 0x01000 },
+		.caps = MDP_CAP_SMP |
+			0,
 	},
 	.smp = {
 		.mmb_count = 8,
@@ -301,6 +309,8 @@ const struct mdp5_cfg_hw msm8x94_config = {
 	.mdp = {
 		.count = 1,
 		.base = { 0x01000 },
+		.caps = MDP_CAP_SMP |
+			0,
 	},
 	.smp = {
 		.mmb_count = 44,
@@ -370,7 +380,89 @@ const struct mdp5_cfg_hw msm8x94_config = {
 			[3] = INTF_HDMI,
 		},
 	},
-	.max_clk = 320000000,
+	.max_clk = 400000000,
+};
+
+const struct mdp5_cfg_hw msm8x96_config = {
+	.name = "msm8x96",
+	.mdp = {
+		.count = 1,
+		.base = { 0x01000 },
+		.caps = MDP_CAP_DSC |
+			MDP_CAP_CDM |
+			0,
+	},
+	.ctl = {
+		.count = 5,
+		.base = { 0x02000, 0x02200, 0x02400, 0x02600, 0x02800 },
+		.flush_hw_mask = 0xf4ffffff,
+	},
+	.pipe_vig = {
+		.count = 4,
+		.base = { 0x05000, 0x07000, 0x09000, 0x0b000 },
+		.caps = MDP_PIPE_CAP_HFLIP	|
+			MDP_PIPE_CAP_VFLIP	|
+			MDP_PIPE_CAP_SCALE	|
+			MDP_PIPE_CAP_CSC	|
+			MDP_PIPE_CAP_DECIMATION	|
+			MDP_PIPE_CAP_SW_PIX_EXT	|
+			0,
+	},
+	.pipe_rgb = {
+		.count = 4,
+		.base = { 0x15000, 0x17000, 0x19000, 0x1b000 },
+		.caps = MDP_PIPE_CAP_HFLIP	|
+			MDP_PIPE_CAP_VFLIP	|
+			MDP_PIPE_CAP_SCALE	|
+			MDP_PIPE_CAP_DECIMATION	|
+			MDP_PIPE_CAP_SW_PIX_EXT	|
+			0,
+	},
+	.pipe_dma = {
+		.count = 2,
+		.base = { 0x25000, 0x27000 },
+		.caps = MDP_PIPE_CAP_HFLIP	|
+			MDP_PIPE_CAP_VFLIP	|
+			MDP_PIPE_CAP_SW_PIX_EXT	|
+			0,
+	},
+	.lm = {
+		.count = 6,
+		.base = { 0x45000, 0x46000, 0x47000, 0x48000, 0x49000, 0x4a000 },
+		.nb_stages = 8,
+		.max_width = 2560,
+		.max_height = 0xFFFF,
+	},
+	.dspp = {
+		.count = 2,
+		.base = { 0x55000, 0x57000 },
+	},
+	.ad = {
+		.count = 3,
+		.base = { 0x79000, 0x79800, 0x7a000 },
+	},
+	.pp = {
+		.count = 4,
+		.base = { 0x71000, 0x71800, 0x72000, 0x72800 },
+	},
+	.cdm = {
+		.count = 1,
+		.base = { 0x7a200 },
+	},
+	.dsc = {
+		.count = 2,
+		.base = { 0x81000, 0x81400 },
+	},
+	.intf = {
+		.base = { 0x6b000, 0x6b800, 0x6c000, 0x6c800, 0x6d000 },
+		.connect = {
+			[0] = INTF_DISABLED,
+			[1] = INTF_DSI,
+			[2] = INTF_DSI,
+			[3] = INTF_HDMI,
+		},
+	},
+	.max_clk = 412500000,
 };
 
 static const struct mdp5_cfg_handler cfg_handlers[] = {
@@ -379,6 +471,7 @@ static const struct mdp5_cfg_handler cfg_handlers[] = {
 	{ .revision = 3, .config = { .hw = &apq8084_config } },
 	{ .revision = 6, .config = { .hw = &msm8x16_config } },
 	{ .revision = 9, .config = { .hw = &msm8x94_config } },
+	{ .revision = 7, .config = { .hw = &msm8x96_config } },
 };
 
 static struct mdp5_cfg_platform *mdp5_get_config(struct platform_device *dev);
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_cfg.h b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_cfg.h
index a6facaf..050e161 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_cfg.h
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_cfg.h
@@ -64,6 +64,11 @@ struct mdp5_smp_block {
 	uint8_t reserved[MAX_CLIENTS];	/* # of MMBs allocated per client */
 };
 
+struct mdp5_mdp_block {
+	MDP5_SUB_BLOCK_DEFINITION;
+	uint32_t caps;			/* MDP capabilities: MDP_CAP_xxx bits */
+};
+
 #define MDP5_INTF_NUM_MAX	5
 
 struct mdp5_intf_block {
@@ -74,7 +79,7 @@ struct mdp5_intf_block {
 struct mdp5_cfg_hw {
 	char  *name;
 
-	struct mdp5_sub_block mdp;
+	struct mdp5_mdp_block mdp;
 	struct mdp5_smp_block smp;
 	struct mdp5_ctl_block ctl;
 	struct mdp5_pipe_block pipe_vig;
@@ -84,6 +89,8 @@ struct mdp5_cfg_hw {
 	struct mdp5_sub_block dspp;
 	struct mdp5_sub_block ad;
 	struct mdp5_sub_block pp;
+	struct mdp5_sub_block dsc;
+	struct mdp5_sub_block cdm;
 	struct mdp5_intf_block intf;
 
 	uint32_t max_clk;
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c
index 34c4ad5..b532faa 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c
@@ -554,15 +554,23 @@ struct msm_kms *mdp5_kms_init(struct drm_device *dev)
 	}
 
 	config = mdp5_cfg_get_config(mdp5_kms->cfg);
+	mdp5_kms->caps = config->hw->mdp.caps;
 
 	/* TODO: compute core clock rate at runtime */
 	clk_set_rate(mdp5_kms->src_clk, config->hw->max_clk);
 
-	mdp5_kms->smp = mdp5_smp_init(mdp5_kms->dev, &config->hw->smp);
-	if (IS_ERR(mdp5_kms->smp)) {
-		ret = PTR_ERR(mdp5_kms->smp);
-		mdp5_kms->smp = NULL;
-		goto fail;
+	/*
+	 * Some chipsets have a Shared Memory Pool (SMP), while others
+	 * have dedicated latency buffering per source pipe instead;
+	 * this section initializes the SMP:
+	 */
+	if (mdp5_kms->caps & MDP_CAP_SMP) {
+		mdp5_kms->smp = mdp5_smp_init(mdp5_kms->dev, &config->hw->smp);
+		if (IS_ERR(mdp5_kms->smp)) {
+			ret = PTR_ERR(mdp5_kms->smp);
+			mdp5_kms->smp = NULL;
+			goto fail;
+		}
 	}
 
 	mdp5_kms->ctlm = mdp5_ctlm_init(dev, mdp5_kms->mmio, mdp5_kms->cfg);
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h
index 0bb6242..84f65d4 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h
@@ -32,6 +32,8 @@ struct mdp5_kms {
 	struct drm_device *dev;
 
 	struct mdp5_cfg_handler *cfg;
+	uint32_t caps;	/* MDP capabilities (MDP_CAP_XXX bits) */
+
 
 	/* mapper-id used to request GEM buffer mapped for scanout: */
 	int id;
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c
index d93185e..81cd490 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c
@@ -705,10 +705,12 @@ static int mdp5_plane_mode_set(struct drm_plane *plane,
 			crtc->base.id, crtc_x, crtc_y, crtc_w, crtc_h);
 
 	/* Request some memory from the SMP: */
-	ret = mdp5_smp_request(mdp5_kms->smp,
-			mdp5_plane->pipe, format, src_w, false);
-	if (ret)
-		return ret;
+	if (mdp5_kms->smp) {
+		ret = mdp5_smp_request(mdp5_kms->smp,
+				mdp5_plane->pipe, format, src_w, false);
+		if (ret)
+			return ret;
+	}
 
 	/*
 	 * Currently we update the hw for allocations/requests immediately,
@@ -716,7 +718,8 @@ static int mdp5_plane_mode_set(struct drm_plane *plane,
 	 * would move into atomic->check_plane_state(), while updating the
 	 * hw would remain here:
 	 */
-	mdp5_smp_configure(mdp5_kms->smp, pipe);
+	if (mdp5_kms->smp)
+		mdp5_smp_configure(mdp5_kms->smp, pipe);
 
 	ret = calc_scalex_steps(plane, pix_format, src_w, crtc_w, phasex_step);
 	if (ret)
@@ -835,7 +838,8 @@ void mdp5_plane_complete_flip(struct drm_plane *plane)
 
 	DBG("%s: complete flip", mdp5_plane->name);
 
-	mdp5_smp_commit(mdp5_kms->smp, pipe);
+	if (mdp5_kms->smp)
+		mdp5_smp_commit(mdp5_kms->smp, pipe);
 
 	to_mdp5_plane_state(plane->state)->pending = false;
 }
@@ -861,7 +865,7 @@ void mdp5_plane_complete_commit(struct drm_plane *plane,
 	struct mdp5_plane *mdp5_plane = to_mdp5_plane(plane);
 	enum mdp5_pipe pipe = mdp5_plane->pipe;
 
-	if (!plane_enabled(plane->state)) {
+	if (!plane_enabled(plane->state) && mdp5_kms->smp) {
 		DBG("%s: free SMP", mdp5_plane->name);
 		mdp5_smp_release(mdp5_kms->smp, pipe);
 	}
diff --git a/drivers/gpu/drm/msm/mdp/mdp_kms.h b/drivers/gpu/drm/msm/mdp/mdp_kms.h
index 0af2a54..3031303 100644
--- a/drivers/gpu/drm/msm/mdp/mdp_kms.h
+++ b/drivers/gpu/drm/msm/mdp/mdp_kms.h
@@ -100,6 +100,11 @@ struct mdp_format {
 uint32_t mdp_get_formats(uint32_t *formats, uint32_t max_formats, bool rgb_only);
 const struct msm_format *mdp_get_format(struct msm_kms *kms, uint32_t format);
 
+/* MDP capabilities */
+#define MDP_CAP_SMP		BIT(0)	/* Shared Memory Pool                 */
+#define MDP_CAP_DSC		BIT(1)	/* VESA Display Stream Compression    */
+#define MDP_CAP_CDM		BIT(2)	/* Chroma Down Module (HDMI 2.0 YUV)  */
+
 /* MDP pipe capabilities */
 #define MDP_PIPE_CAP_HFLIP			BIT(0)
 #define MDP_PIPE_CAP_VFLIP			BIT(1)
-- 
cgit v0.10.2


From a9ee34b70e07ce942806eda154e48ea3f754a76f Mon Sep 17 00:00:00 2001
From: Archit Taneja <architt@codeaurora.org>
Date: Mon, 13 Jul 2015 12:12:07 +0530
Subject: drm/msm: Remove local fbdev emulation Kconfig option

DRM_MSM_FBDEV config is used to enable/disable fbdev emulation for the
msm kms driver.

Replace this with the top level DRM_FBDEV_EMULATION config option where
applicable. This also prevents build breaks caused by undefined
drm_fb_helper_* functions when legacy fbdev support was disabled.

Signed-off-by: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Rob Clark <robdclark@gmail.com>

diff --git a/drivers/gpu/drm/msm/Kconfig b/drivers/gpu/drm/msm/Kconfig
index 8e6c7c6..84d3ec9 100644
--- a/drivers/gpu/drm/msm/Kconfig
+++ b/drivers/gpu/drm/msm/Kconfig
@@ -14,20 +14,6 @@ config DRM_MSM
 	help
 	  DRM/KMS driver for MSM/snapdragon.
 
-config DRM_MSM_FBDEV
-	bool "Enable legacy fbdev support for MSM modesetting driver"
-	depends on DRM_MSM
-	select DRM_KMS_FB_HELPER
-	select FB_SYS_FILLRECT
-	select FB_SYS_COPYAREA
-	select FB_SYS_IMAGEBLIT
-	select FB_SYS_FOPS
-	default y
-	help
-	  Choose this option if you have a need for the legacy fbdev
-	  support. Note that this support also provide the linux console
-	  support on top of the MSM modesetting driver.
-
 config DRM_MSM_REGISTER_LOGGING
 	bool "MSM DRM register logging"
 	depends on DRM_MSM
diff --git a/drivers/gpu/drm/msm/Makefile b/drivers/gpu/drm/msm/Makefile
index 0a543eb..1c90290 100644
--- a/drivers/gpu/drm/msm/Makefile
+++ b/drivers/gpu/drm/msm/Makefile
@@ -50,7 +50,7 @@ msm-y := \
 	msm_rd.o \
 	msm_ringbuffer.o
 
-msm-$(CONFIG_DRM_MSM_FBDEV) += msm_fbdev.o
+msm-$(CONFIG_DRM_FBDEV_EMULATION) += msm_fbdev.o
 msm-$(CONFIG_COMMON_CLK) += mdp/mdp4/mdp4_lvds_pll.o
 
 msm-$(CONFIG_DRM_MSM_DSI) += dsi/dsi.o \
diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c
index d170131..b88ce51 100644
--- a/drivers/gpu/drm/msm/msm_drv.c
+++ b/drivers/gpu/drm/msm/msm_drv.c
@@ -21,11 +21,9 @@
 
 static void msm_fb_output_poll_changed(struct drm_device *dev)
 {
-#ifdef CONFIG_DRM_MSM_FBDEV
 	struct msm_drm_private *priv = dev->dev_private;
 	if (priv->fbdev)
 		drm_fb_helper_hotplug_event(priv->fbdev);
-#endif
 }
 
 static const struct drm_mode_config_funcs mode_config_funcs = {
@@ -56,7 +54,7 @@ module_param(reglog, bool, 0600);
 #define reglog 0
 #endif
 
-#ifdef CONFIG_DRM_MSM_FBDEV
+#ifdef CONFIG_DRM_FBDEV_EMULATION
 static bool fbdev = true;
 MODULE_PARM_DESC(fbdev, "Enable fbdev compat layer");
 module_param(fbdev, bool, 0600);
@@ -423,7 +421,7 @@ static int msm_load(struct drm_device *dev, unsigned long flags)
 
 	drm_mode_config_reset(dev);
 
-#ifdef CONFIG_DRM_MSM_FBDEV
+#ifdef CONFIG_DRM_FBDEV_EMULATION
 	if (fbdev)
 		priv->fbdev = msm_fbdev_init(dev);
 #endif
@@ -491,11 +489,9 @@ static void msm_preclose(struct drm_device *dev, struct drm_file *file)
 
 static void msm_lastclose(struct drm_device *dev)
 {
-#ifdef CONFIG_DRM_MSM_FBDEV
 	struct msm_drm_private *priv = dev->dev_private;
 	if (priv->fbdev)
 		drm_fb_helper_restore_fbdev_mode_unlocked(priv->fbdev);
-#endif
 }
 
 static irqreturn_t msm_irq(int irq, void *arg)
-- 
cgit v0.10.2


From 82326860b3ee58db33536a3e29675e28603f661c Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Thu, 22 Oct 2015 14:08:35 -0400
Subject: drm/amdgpu/dce8: optimize pageflip
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Taking the grph update lock is only necessary when
updating the the secondary address (for single pipe stereo).

Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
index 00c34f8..b17abbe 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
@@ -211,46 +211,22 @@ static u32 dce_v8_0_vblank_get_counter(struct amdgpu_device *adev, int crtc)
  * @crtc_id: crtc to cleanup pageflip on
  * @crtc_base: new address of the crtc (GPU MC address)
  *
- * Does the actual pageflip (evergreen+).
- * During vblank we take the crtc lock and wait for the update_pending
- * bit to go high, when it does, we release the lock, and allow the
- * double buffered update to take place.
- * Returns the current update pending status.
+ * Triggers the actual pageflip by updating the primary
+ * surface base address.
  */
 static void dce_v8_0_page_flip(struct amdgpu_device *adev,
 			      int crtc_id, u64 crtc_base)
 {
 	struct amdgpu_crtc *amdgpu_crtc = adev->mode_info.crtcs[crtc_id];
-	u32 tmp = RREG32(mmGRPH_UPDATE + amdgpu_crtc->crtc_offset);
-	int i;
-
-	/* Lock the graphics update lock */
-	tmp |= GRPH_UPDATE__GRPH_UPDATE_LOCK_MASK;
-	WREG32(mmGRPH_UPDATE + amdgpu_crtc->crtc_offset, tmp);
-
-	/* update the scanout addresses */
-	WREG32(mmGRPH_SECONDARY_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset,
-	       upper_32_bits(crtc_base));
-	WREG32(mmGRPH_SECONDARY_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset,
-	       (u32)crtc_base);
 
+	/* update the primary scanout addresses */
 	WREG32(mmGRPH_PRIMARY_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset,
 	       upper_32_bits(crtc_base));
+	/* writing to the low address triggers the update */
 	WREG32(mmGRPH_PRIMARY_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset,
-	       (u32)crtc_base);
-
-	/* Wait for update_pending to go high. */
-	for (i = 0; i < adev->usec_timeout; i++) {
-		if (RREG32(mmGRPH_UPDATE + amdgpu_crtc->crtc_offset) &
-				GRPH_UPDATE__GRPH_SURFACE_UPDATE_PENDING_MASK)
-			break;
-		udelay(1);
-	}
-	DRM_DEBUG("Update pending now high. Unlocking vupdate_lock.\n");
-
-	/* Unlock the lock, so double-buffering can take place inside vblank */
-	tmp &= ~GRPH_UPDATE__GRPH_UPDATE_LOCK_MASK;
-	WREG32(mmGRPH_UPDATE + amdgpu_crtc->crtc_offset, tmp);
+	       lower_32_bits(crtc_base));
+	/* post the write */
+	RREG32(mmGRPH_PRIMARY_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset);
 }
 
 static int dce_v8_0_crtc_get_scanoutpos(struct amdgpu_device *adev, int crtc,
-- 
cgit v0.10.2


From 0eaaacab038ec03c010cd820554cfa6d1a93cd78 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Thu, 22 Oct 2015 14:16:57 -0400
Subject: drm/amdgpu/dce10: optimize pageflip
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Taking the grph update lock is only necessary when
updating the the secondary address (for single pipe stereo).

Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
index a6ea2d8..6132051 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
@@ -262,46 +262,22 @@ static u32 dce_v10_0_vblank_get_counter(struct amdgpu_device *adev, int crtc)
  * @crtc_id: crtc to cleanup pageflip on
  * @crtc_base: new address of the crtc (GPU MC address)
  *
- * Does the actual pageflip (evergreen+).
- * During vblank we take the crtc lock and wait for the update_pending
- * bit to go high, when it does, we release the lock, and allow the
- * double buffered update to take place.
- * Returns the current update pending status.
+ * Triggers the actual pageflip by updating the primary
+ * surface base address.
  */
 static void dce_v10_0_page_flip(struct amdgpu_device *adev,
 			      int crtc_id, u64 crtc_base)
 {
 	struct amdgpu_crtc *amdgpu_crtc = adev->mode_info.crtcs[crtc_id];
-	u32 tmp = RREG32(mmGRPH_UPDATE + amdgpu_crtc->crtc_offset);
-	int i;
-
-	/* Lock the graphics update lock */
-	tmp = REG_SET_FIELD(tmp, GRPH_UPDATE, GRPH_UPDATE_LOCK, 1);
-	WREG32(mmGRPH_UPDATE + amdgpu_crtc->crtc_offset, tmp);
-
-	/* update the scanout addresses */
-	WREG32(mmGRPH_SECONDARY_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset,
-	       upper_32_bits(crtc_base));
-	WREG32(mmGRPH_SECONDARY_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset,
-	       lower_32_bits(crtc_base));
 
+	/* update the primary scanout address */
 	WREG32(mmGRPH_PRIMARY_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset,
 	       upper_32_bits(crtc_base));
+	/* writing to the low address triggers the update */
 	WREG32(mmGRPH_PRIMARY_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset,
 	       lower_32_bits(crtc_base));
-
-	/* Wait for update_pending to go high. */
-	for (i = 0; i < adev->usec_timeout; i++) {
-		if (RREG32(mmGRPH_UPDATE + amdgpu_crtc->crtc_offset) &
-				GRPH_UPDATE__GRPH_SURFACE_UPDATE_PENDING_MASK)
-			break;
-		udelay(1);
-	}
-	DRM_DEBUG("Update pending now high. Unlocking vupdate_lock.\n");
-
-	/* Unlock the lock, so double-buffering can take place inside vblank */
-	tmp = REG_SET_FIELD(tmp, GRPH_UPDATE, GRPH_UPDATE_LOCK, 0);
-	WREG32(mmGRPH_UPDATE + amdgpu_crtc->crtc_offset, tmp);
+	/* post the write */
+	RREG32(mmGRPH_PRIMARY_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset);
 }
 
 static int dce_v10_0_crtc_get_scanoutpos(struct amdgpu_device *adev, int crtc,
-- 
cgit v0.10.2


From ce055fe3329e9dece4fb676c8cf7331748b07d32 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Thu, 22 Oct 2015 14:17:21 -0400
Subject: drm/amdgpu/dce11: optimize pageflip
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Taking the grph update lock is only necessary when
updating the the secondary address (for single pipe stereo).

Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
index 444d930..4898c46 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
@@ -252,46 +252,22 @@ static u32 dce_v11_0_vblank_get_counter(struct amdgpu_device *adev, int crtc)
  * @crtc_id: crtc to cleanup pageflip on
  * @crtc_base: new address of the crtc (GPU MC address)
  *
- * Does the actual pageflip (evergreen+).
- * During vblank we take the crtc lock and wait for the update_pending
- * bit to go high, when it does, we release the lock, and allow the
- * double buffered update to take place.
- * Returns the current update pending status.
+ * Triggers the actual pageflip by updating the primary
+ * surface base address.
  */
 static void dce_v11_0_page_flip(struct amdgpu_device *adev,
 			      int crtc_id, u64 crtc_base)
 {
 	struct amdgpu_crtc *amdgpu_crtc = adev->mode_info.crtcs[crtc_id];
-	u32 tmp = RREG32(mmGRPH_UPDATE + amdgpu_crtc->crtc_offset);
-	int i;
-
-	/* Lock the graphics update lock */
-	tmp = REG_SET_FIELD(tmp, GRPH_UPDATE, GRPH_UPDATE_LOCK, 1);
-	WREG32(mmGRPH_UPDATE + amdgpu_crtc->crtc_offset, tmp);
 
 	/* update the scanout addresses */
-	WREG32(mmGRPH_SECONDARY_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset,
-	       upper_32_bits(crtc_base));
-	WREG32(mmGRPH_SECONDARY_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset,
-	       lower_32_bits(crtc_base));
-
 	WREG32(mmGRPH_PRIMARY_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset,
 	       upper_32_bits(crtc_base));
+	/* writing to the low address triggers the update */
 	WREG32(mmGRPH_PRIMARY_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset,
 	       lower_32_bits(crtc_base));
-
-	/* Wait for update_pending to go high. */
-	for (i = 0; i < adev->usec_timeout; i++) {
-		if (RREG32(mmGRPH_UPDATE + amdgpu_crtc->crtc_offset) &
-				GRPH_UPDATE__GRPH_SURFACE_UPDATE_PENDING_MASK)
-			break;
-		udelay(1);
-	}
-	DRM_DEBUG("Update pending now high. Unlocking vupdate_lock.\n");
-
-	/* Unlock the lock, so double-buffering can take place inside vblank */
-	tmp = REG_SET_FIELD(tmp, GRPH_UPDATE, GRPH_UPDATE_LOCK, 0);
-	WREG32(mmGRPH_UPDATE + amdgpu_crtc->crtc_offset, tmp);	
+	/* post the write */
+	RREG32(mmGRPH_PRIMARY_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset);
 }
 
 static int dce_v11_0_crtc_get_scanoutpos(struct amdgpu_device *adev, int crtc,
-- 
cgit v0.10.2


From e68d547b8ae109c7ecaaaad665c8de8373a215bc Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <a.hajda@samsung.com>
Date: Fri, 25 Sep 2015 14:48:14 +0200
Subject: drm/exynos/hdmi: remove support for deprecated compatible

This compatible was marked as deprecated in Jun 2013 and it is not used since
then. Additionally its driver data points to wrong pll settings, so it
cannot work anyway.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c
index 932f7fa..c25b892 100644
--- a/drivers/gpu/drm/exynos/exynos_hdmi.c
+++ b/drivers/gpu/drm/exynos/exynos_hdmi.c
@@ -507,13 +507,6 @@ static struct hdmi_driver_data exynos4210_hdmi_driver_data = {
 	.is_apb_phy	= 0,
 };
 
-static struct hdmi_driver_data exynos5_hdmi_driver_data = {
-	.type		= HDMI_TYPE14,
-	.phy_confs	= hdmiphy_v13_configs,
-	.phy_conf_count	= ARRAY_SIZE(hdmiphy_v13_configs),
-	.is_apb_phy	= 0,
-};
-
 static inline u32 hdmi_reg_read(struct hdmi_context *hdata, u32 reg_id)
 {
 	return readl(hdata->regs + reg_id);
@@ -1909,9 +1902,6 @@ fail:
 
 static struct of_device_id hdmi_match_types[] = {
 	{
-		.compatible = "samsung,exynos5-hdmi",
-		.data = &exynos5_hdmi_driver_data,
-	}, {
 		.compatible = "samsung,exynos4210-hdmi",
 		.data = &exynos4210_hdmi_driver_data,
 	}, {
-- 
cgit v0.10.2


From 633d00b327c0ef5ec0113e27d2d5e36073ebca7d Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <a.hajda@samsung.com>
Date: Fri, 25 Sep 2015 14:48:16 +0200
Subject: drm/exynos/hdmi: use mappings for registers with IP dependent address

Some registers resides at different offsets depending on device version.
This patch adds infrastructure for mapping such registers to proper address
based on hdmi_type. It adds also mappings to some registers.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c
index c25b892..6221803 100644
--- a/drivers/gpu/drm/exynos/exynos_hdmi.c
+++ b/drivers/gpu/drm/exynos/exynos_hdmi.c
@@ -66,6 +66,21 @@
 enum hdmi_type {
 	HDMI_TYPE13,
 	HDMI_TYPE14,
+	HDMI_TYPE_COUNT
+};
+
+#define HDMI_MAPPED_BASE 0xffff0000
+
+enum hdmi_mapped_regs {
+	HDMI_PHY_STATUS = HDMI_MAPPED_BASE,
+	HDMI_PHY_RSTOUT,
+	HDMI_ACR_CON,
+};
+
+static const u32 hdmi_reg_map[][HDMI_TYPE_COUNT] = {
+	{ HDMI_V13_PHY_STATUS, HDMI_PHY_STATUS_0 },
+	{ HDMI_V13_PHY_RSTOUT, HDMI_V14_PHY_RSTOUT },
+	{ HDMI_V13_ACR_CON, HDMI_V14_ACR_CON },
 };
 
 struct hdmi_driver_data {
@@ -507,20 +522,29 @@ static struct hdmi_driver_data exynos4210_hdmi_driver_data = {
 	.is_apb_phy	= 0,
 };
 
+static inline u32 hdmi_map_reg(struct hdmi_context *hdata, u32 reg_id)
+{
+	if ((reg_id & 0xffff0000) == HDMI_MAPPED_BASE)
+		return hdmi_reg_map[reg_id & 0xffff][hdata->drv_data->type];
+	return reg_id;
+}
+
 static inline u32 hdmi_reg_read(struct hdmi_context *hdata, u32 reg_id)
 {
-	return readl(hdata->regs + reg_id);
+	return readl(hdata->regs + hdmi_map_reg(hdata, reg_id));
 }
 
 static inline void hdmi_reg_writeb(struct hdmi_context *hdata,
 				 u32 reg_id, u8 value)
 {
-	writeb(value, hdata->regs + reg_id);
+	writeb(value, hdata->regs + hdmi_map_reg(hdata, reg_id));
 }
 
 static inline void hdmi_reg_writev(struct hdmi_context *hdata, u32 reg_id,
 				   int bytes, u32 val)
 {
+	reg_id = hdmi_map_reg(hdata, reg_id);
+
 	while (--bytes >= 0) {
 		writeb(val & 0xff, hdata->regs + reg_id);
 		val >>= 8;
@@ -531,7 +555,10 @@ static inline void hdmi_reg_writev(struct hdmi_context *hdata, u32 reg_id,
 static inline void hdmi_reg_writemask(struct hdmi_context *hdata,
 				 u32 reg_id, u32 value, u32 mask)
 {
-	u32 old = readl(hdata->regs + reg_id);
+	u32 old;
+
+	reg_id = hdmi_map_reg(hdata, reg_id);
+	old = readl(hdata->regs + reg_id);
 	value = (value & mask) | (old & ~mask);
 	writel(value, hdata->regs + reg_id);
 }
@@ -682,7 +709,7 @@ static void hdmi_v14_regs_dump(struct hdmi_context *hdata, char *prefix)
 	DUMPREG(HDMI_PHY_STATUS_0);
 	DUMPREG(HDMI_PHY_STATUS_PLL);
 	DUMPREG(HDMI_PHY_CON_0);
-	DUMPREG(HDMI_PHY_RSTOUT);
+	DUMPREG(HDMI_V14_PHY_RSTOUT);
 	DUMPREG(HDMI_PHY_VPLL);
 	DUMPREG(HDMI_PHY_CMU);
 	DUMPREG(HDMI_CORE_RSTOUT);
@@ -1162,11 +1189,7 @@ static void hdmi_reg_acr(struct hdmi_context *hdata, u8 *acr)
 	hdmi_reg_writeb(hdata, HDMI_ACR_CTS0, acr[3]);
 	hdmi_reg_writeb(hdata, HDMI_ACR_CTS1, acr[2]);
 	hdmi_reg_writeb(hdata, HDMI_ACR_CTS2, acr[1]);
-
-	if (hdata->drv_data->type == HDMI_TYPE13)
-		hdmi_reg_writeb(hdata, HDMI_V13_ACR_CON, 4);
-	else
-		hdmi_reg_writeb(hdata, HDMI_ACR_CON, 4);
+	hdmi_reg_writeb(hdata, HDMI_ACR_CON, 4);
 }
 
 static void hdmi_audio_init(struct hdmi_context *hdata)
@@ -1421,7 +1444,7 @@ static void hdmi_v13_mode_apply(struct hdmi_context *hdata)
 
 	/* waiting for HDMIPHY's PLL to get to steady state */
 	for (tries = 100; tries; --tries) {
-		u32 val = hdmi_reg_read(hdata, HDMI_V13_PHY_STATUS);
+		u32 val = hdmi_reg_read(hdata, HDMI_PHY_STATUS);
 		if (val & HDMI_PHY_STATUS_READY)
 			break;
 		usleep_range(1000, 2000);
@@ -1558,7 +1581,7 @@ static void hdmi_v14_mode_apply(struct hdmi_context *hdata)
 
 	/* waiting for HDMIPHY's PLL to get to steady state */
 	for (tries = 100; tries; --tries) {
-		u32 val = hdmi_reg_read(hdata, HDMI_PHY_STATUS_0);
+		u32 val = hdmi_reg_read(hdata, HDMI_PHY_STATUS);
 		if (val & HDMI_PHY_STATUS_READY)
 			break;
 		usleep_range(1000, 2000);
@@ -1587,8 +1610,6 @@ static void hdmi_mode_apply(struct hdmi_context *hdata)
 
 static void hdmiphy_conf_reset(struct hdmi_context *hdata)
 {
-	u32 reg;
-
 	clk_disable_unprepare(hdata->res.sclk_hdmi);
 	clk_set_parent(hdata->res.mout_hdmi, hdata->res.sclk_pixel);
 	clk_prepare_enable(hdata->res.sclk_hdmi);
@@ -1597,15 +1618,10 @@ static void hdmiphy_conf_reset(struct hdmi_context *hdata)
 	hdmiphy_reg_writeb(hdata, HDMIPHY_MODE_SET_DONE,
 				HDMI_PHY_ENABLE_MODE_SET);
 
-	if (hdata->drv_data->type == HDMI_TYPE13)
-		reg = HDMI_V13_PHY_RSTOUT;
-	else
-		reg = HDMI_PHY_RSTOUT;
-
 	/* reset hdmiphy */
-	hdmi_reg_writemask(hdata, reg, ~0, HDMI_PHY_SW_RSTOUT);
+	hdmi_reg_writemask(hdata, HDMI_PHY_RSTOUT, ~0, HDMI_PHY_SW_RSTOUT);
 	usleep_range(10000, 12000);
-	hdmi_reg_writemask(hdata, reg,  0, HDMI_PHY_SW_RSTOUT);
+	hdmi_reg_writemask(hdata, HDMI_PHY_RSTOUT,  0, HDMI_PHY_SW_RSTOUT);
 	usleep_range(10000, 12000);
 }
 
diff --git a/drivers/gpu/drm/exynos/regs-hdmi.h b/drivers/gpu/drm/exynos/regs-hdmi.h
index 3f35ac6..8b2d337 100644
--- a/drivers/gpu/drm/exynos/regs-hdmi.h
+++ b/drivers/gpu/drm/exynos/regs-hdmi.h
@@ -171,7 +171,7 @@
 #define HDMI_HPD_ST			HDMI_CTRL_BASE(0x0044)
 #define HDMI_HPD_TH_X			HDMI_CTRL_BASE(0x0050)
 #define HDMI_AUDIO_CLKSEL		HDMI_CTRL_BASE(0x0070)
-#define HDMI_PHY_RSTOUT			HDMI_CTRL_BASE(0x0074)
+#define HDMI_V14_PHY_RSTOUT		HDMI_CTRL_BASE(0x0074)
 #define HDMI_PHY_VPLL			HDMI_CTRL_BASE(0x0078)
 #define HDMI_PHY_CMU			HDMI_CTRL_BASE(0x007C)
 #define HDMI_CORE_RSTOUT		HDMI_CTRL_BASE(0x0080)
@@ -277,7 +277,7 @@
 #define HDMI_ASP_CHCFG2			HDMI_CORE_BASE(0x0318)
 #define HDMI_ASP_CHCFG3			HDMI_CORE_BASE(0x031C)
 
-#define HDMI_ACR_CON			HDMI_CORE_BASE(0x0400)
+#define HDMI_V14_ACR_CON		HDMI_CORE_BASE(0x0400)
 #define HDMI_ACR_MCTS0			HDMI_CORE_BASE(0x0410)
 #define HDMI_ACR_MCTS1			HDMI_CORE_BASE(0x0414)
 #define HDMI_ACR_MCTS2			HDMI_CORE_BASE(0x0418)
-- 
cgit v0.10.2


From 8eb6d4ec99bf4c5004d4b87f4f9dc539b708bdb1 Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <a.hajda@samsung.com>
Date: Fri, 25 Sep 2015 14:48:17 +0200
Subject: drm/exynos/hdmi: move PLL stabilization check code to separate
 function

The patch moves PLL stabilization check to separate function, adjust timeout
parameters and de-duplicates code common for both HW variants.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c
index 6221803..414ea2d 100644
--- a/drivers/gpu/drm/exynos/exynos_hdmi.c
+++ b/drivers/gpu/drm/exynos/exynos_hdmi.c
@@ -1351,11 +1351,27 @@ static void hdmi_conf_init(struct hdmi_context *hdata)
 	}
 }
 
+static void hdmiphy_wait_for_pll(struct hdmi_context *hdata)
+{
+	int tries;
+
+	for (tries = 0; tries < 10; ++tries) {
+		u32 val = hdmi_reg_read(hdata, HDMI_PHY_STATUS);
+
+		if (val & HDMI_PHY_STATUS_READY) {
+			DRM_DEBUG_KMS("PLL stabilized after %d tries\n", tries);
+			return;
+		}
+		usleep_range(10, 20);
+	}
+
+	DRM_ERROR("PLL could not reach steady state\n");
+}
+
 static void hdmi_v13_mode_apply(struct hdmi_context *hdata)
 {
 	struct drm_display_mode *m = &hdata->current_mode;
 	unsigned int val;
-	int tries;
 
 	hdmi_reg_writev(hdata, HDMI_H_BLANK_0, 2, m->htotal - m->hdisplay);
 	hdmi_reg_writev(hdata, HDMI_V13_H_V_LINE_0, 3,
@@ -1441,32 +1457,11 @@ static void hdmi_v13_mode_apply(struct hdmi_context *hdata)
 	hdmi_reg_writev(hdata, HDMI_TG_VSYNC_BOT_HDMI_L, 2, 0x233);
 	hdmi_reg_writev(hdata, HDMI_TG_FIELD_TOP_HDMI_L, 2, 0x1);
 	hdmi_reg_writev(hdata, HDMI_TG_FIELD_BOT_HDMI_L, 2, 0x233);
-
-	/* waiting for HDMIPHY's PLL to get to steady state */
-	for (tries = 100; tries; --tries) {
-		u32 val = hdmi_reg_read(hdata, HDMI_PHY_STATUS);
-		if (val & HDMI_PHY_STATUS_READY)
-			break;
-		usleep_range(1000, 2000);
-	}
-	/* steady state not achieved */
-	if (tries == 0) {
-		DRM_ERROR("hdmiphy's pll could not reach steady state.\n");
-		hdmi_regs_dump(hdata, "timing apply");
-	}
-
-	clk_disable_unprepare(hdata->res.sclk_hdmi);
-	clk_set_parent(hdata->res.mout_hdmi, hdata->res.sclk_hdmiphy);
-	clk_prepare_enable(hdata->res.sclk_hdmi);
-
-	/* enable HDMI and timing generator */
-	hdmi_start(hdata, true);
 }
 
 static void hdmi_v14_mode_apply(struct hdmi_context *hdata)
 {
 	struct drm_display_mode *m = &hdata->current_mode;
-	int tries;
 
 	hdmi_reg_writev(hdata, HDMI_H_BLANK_0, 2, m->htotal - m->hdisplay);
 	hdmi_reg_writev(hdata, HDMI_V_LINE_0, 2, m->vtotal);
@@ -1578,26 +1573,6 @@ static void hdmi_v14_mode_apply(struct hdmi_context *hdata)
 	hdmi_reg_writev(hdata, HDMI_TG_VSYNC_TOP_HDMI_L, 2, 0x1);
 	hdmi_reg_writev(hdata, HDMI_TG_FIELD_TOP_HDMI_L, 2, 0x1);
 	hdmi_reg_writev(hdata, HDMI_TG_3D, 1, 0x0);
-
-	/* waiting for HDMIPHY's PLL to get to steady state */
-	for (tries = 100; tries; --tries) {
-		u32 val = hdmi_reg_read(hdata, HDMI_PHY_STATUS);
-		if (val & HDMI_PHY_STATUS_READY)
-			break;
-		usleep_range(1000, 2000);
-	}
-	/* steady state not achieved */
-	if (tries == 0) {
-		DRM_ERROR("hdmiphy's pll could not reach steady state.\n");
-		hdmi_regs_dump(hdata, "timing apply");
-	}
-
-	clk_disable_unprepare(hdata->res.sclk_hdmi);
-	clk_set_parent(hdata->res.mout_hdmi, hdata->res.sclk_hdmiphy);
-	clk_prepare_enable(hdata->res.sclk_hdmi);
-
-	/* enable HDMI and timing generator */
-	hdmi_start(hdata, true);
 }
 
 static void hdmi_mode_apply(struct hdmi_context *hdata)
@@ -1606,6 +1581,15 @@ static void hdmi_mode_apply(struct hdmi_context *hdata)
 		hdmi_v13_mode_apply(hdata);
 	else
 		hdmi_v14_mode_apply(hdata);
+
+	hdmiphy_wait_for_pll(hdata);
+
+	clk_disable_unprepare(hdata->res.sclk_hdmi);
+	clk_set_parent(hdata->res.mout_hdmi, hdata->res.sclk_hdmiphy);
+	clk_prepare_enable(hdata->res.sclk_hdmi);
+
+	/* enable HDMI and timing generator */
+	hdmi_start(hdata, true);
 }
 
 static void hdmiphy_conf_reset(struct hdmi_context *hdata)
-- 
cgit v0.10.2


From 74a74ff4e6fa55d9bdeba57187c3622a55ec2204 Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <a.hajda@samsung.com>
Date: Fri, 25 Sep 2015 14:48:18 +0200
Subject: drm/exynos/hdmi: simplify HDMI-PHY power sequence

Currently driver tries to set specific HDMI-PHY registers in three situations:
- before reset,
- before power off,
- after applying HDMI-PHY configuration.

First two cases seems to be unnecessary - register contents will be lost
anyway. The third case can be merged with HDMI-PHY configuration by fixing
the last byte of configuration data.

The patch has been tested with following platforms:
- exynos4210-universal_c210,
- exynos4412-odroidu3,
- exynos5422-odroidxu3.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c
index 414ea2d..814dd8c 100644
--- a/drivers/gpu/drm/exynos/exynos_hdmi.c
+++ b/drivers/gpu/drm/exynos/exynos_hdmi.c
@@ -148,7 +148,7 @@ static const struct hdmiphy_config hdmiphy_v13_configs[] = {
 			0x01, 0x05, 0x00, 0xD8, 0x10, 0x1C, 0x30, 0x40,
 			0x6B, 0x10, 0x02, 0x51, 0xDF, 0xF2, 0x54, 0x87,
 			0x84, 0x00, 0x30, 0x38, 0x00, 0x08, 0x10, 0xE0,
-			0x22, 0x40, 0xE3, 0x26, 0x00, 0x00, 0x00, 0x00,
+			0x22, 0x40, 0xE3, 0x26, 0x00, 0x00, 0x00, 0x80,
 		},
 	},
 	{
@@ -157,7 +157,7 @@ static const struct hdmiphy_config hdmiphy_v13_configs[] = {
 			0x01, 0x05, 0x00, 0xD4, 0x10, 0x9C, 0x09, 0x64,
 			0x6B, 0x10, 0x02, 0x51, 0xDF, 0xF2, 0x54, 0x87,
 			0x84, 0x00, 0x30, 0x38, 0x00, 0x08, 0x10, 0xE0,
-			0x22, 0x40, 0xE3, 0x26, 0x00, 0x00, 0x00, 0x00,
+			0x22, 0x40, 0xE3, 0x26, 0x00, 0x00, 0x00, 0x80,
 		},
 	},
 	{
@@ -166,7 +166,7 @@ static const struct hdmiphy_config hdmiphy_v13_configs[] = {
 			0x01, 0x05, 0x00, 0xD8, 0x10, 0x9C, 0xef, 0x5B,
 			0x6D, 0x10, 0x01, 0x51, 0xef, 0xF3, 0x54, 0xb9,
 			0x84, 0x00, 0x30, 0x38, 0x00, 0x08, 0x10, 0xE0,
-			0x22, 0x40, 0xa5, 0x26, 0x01, 0x00, 0x00, 0x00,
+			0x22, 0x40, 0xa5, 0x26, 0x01, 0x00, 0x00, 0x80,
 		},
 	},
 	{
@@ -175,7 +175,7 @@ static const struct hdmiphy_config hdmiphy_v13_configs[] = {
 			0x01, 0x05, 0x00, 0xd8, 0x10, 0x9c, 0xf8, 0x40,
 			0x6a, 0x10, 0x01, 0x51, 0xff, 0xf1, 0x54, 0xba,
 			0x84, 0x00, 0x10, 0x38, 0x00, 0x08, 0x10, 0xe0,
-			0x22, 0x40, 0xa4, 0x26, 0x01, 0x00, 0x00, 0x00,
+			0x22, 0x40, 0xa4, 0x26, 0x01, 0x00, 0x00, 0x80,
 		},
 	},
 	{
@@ -184,7 +184,7 @@ static const struct hdmiphy_config hdmiphy_v13_configs[] = {
 			0x01, 0x05, 0x00, 0xD8, 0x10, 0x9C, 0xf8, 0x40,
 			0x6A, 0x18, 0x00, 0x51, 0xff, 0xF1, 0x54, 0xba,
 			0x84, 0x00, 0x10, 0x38, 0x00, 0x08, 0x10, 0xE0,
-			0x22, 0x40, 0xa4, 0x26, 0x02, 0x00, 0x00, 0x00,
+			0x22, 0x40, 0xa4, 0x26, 0x02, 0x00, 0x00, 0x80,
 		},
 	},
 };
@@ -214,7 +214,7 @@ static const struct hdmiphy_config hdmiphy_v14_configs[] = {
 			0x01, 0xd1, 0x2d, 0x72, 0x40, 0x64, 0x12, 0x08,
 			0x43, 0xa0, 0x0e, 0xd9, 0x45, 0xa0, 0xac, 0x80,
 			0x08, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
-			0x54, 0xe3, 0x24, 0x00, 0x00, 0x00, 0x01, 0x00,
+			0x54, 0xe3, 0x24, 0x00, 0x00, 0x00, 0x01, 0x80,
 		},
 	},
 	{
@@ -277,7 +277,7 @@ static const struct hdmiphy_config hdmiphy_v14_configs[] = {
 			0x01, 0xd1, 0x1f, 0x10, 0x40, 0x40, 0xf8, 0x08,
 			0x81, 0xa0, 0xba, 0xd8, 0x45, 0xa0, 0xac, 0x80,
 			0x3c, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
-			0x54, 0xa5, 0x24, 0x01, 0x00, 0x00, 0x01, 0x00,
+			0x54, 0xa5, 0x24, 0x01, 0x00, 0x00, 0x01, 0x80,
 		},
 	},
 	{
@@ -340,7 +340,7 @@ static const struct hdmiphy_config hdmiphy_v14_configs[] = {
 			0x01, 0xd1, 0x1f, 0x00, 0x40, 0x40, 0xf8, 0x08,
 			0x81, 0xa0, 0xba, 0xd8, 0x45, 0xa0, 0xac, 0x80,
 			0x3c, 0x80, 0x11, 0x04, 0x02, 0x22, 0x44, 0x86,
-			0x54, 0x4b, 0x25, 0x03, 0x00, 0x00, 0x01, 0x00,
+			0x54, 0x4b, 0x25, 0x03, 0x00, 0x00, 0x01, 0x80,
 		},
 	},
 };
@@ -563,26 +563,6 @@ static inline void hdmi_reg_writemask(struct hdmi_context *hdata,
 	writel(value, hdata->regs + reg_id);
 }
 
-static int hdmiphy_reg_writeb(struct hdmi_context *hdata,
-			u32 reg_offset, u8 value)
-{
-	if (hdata->hdmiphy_port) {
-		u8 buffer[2];
-		int ret;
-
-		buffer[0] = reg_offset;
-		buffer[1] = value;
-
-		ret = i2c_master_send(hdata->hdmiphy_port, buffer, 2);
-		if (ret == 2)
-			return 0;
-		return ret;
-	} else {
-		writeb(value, hdata->regs_hdmiphy + (reg_offset<<2));
-		return 0;
-	}
-}
-
 static int hdmiphy_reg_write_buf(struct hdmi_context *hdata,
 			u32 reg_offset, const u8 *buf, u32 len)
 {
@@ -1598,10 +1578,6 @@ static void hdmiphy_conf_reset(struct hdmi_context *hdata)
 	clk_set_parent(hdata->res.mout_hdmi, hdata->res.sclk_pixel);
 	clk_prepare_enable(hdata->res.sclk_hdmi);
 
-	/* operation mode */
-	hdmiphy_reg_writeb(hdata, HDMIPHY_MODE_SET_DONE,
-				HDMI_PHY_ENABLE_MODE_SET);
-
 	/* reset hdmiphy */
 	hdmi_reg_writemask(hdata, HDMI_PHY_RSTOUT, ~0, HDMI_PHY_SW_RSTOUT);
 	usleep_range(10000, 12000);
@@ -1609,48 +1585,6 @@ static void hdmiphy_conf_reset(struct hdmi_context *hdata)
 	usleep_range(10000, 12000);
 }
 
-static void hdmiphy_poweron(struct hdmi_context *hdata)
-{
-	if (hdata->drv_data->type != HDMI_TYPE14)
-		return;
-
-	DRM_DEBUG_KMS("\n");
-
-	/* For PHY Mode Setting */
-	hdmiphy_reg_writeb(hdata, HDMIPHY_MODE_SET_DONE,
-				HDMI_PHY_ENABLE_MODE_SET);
-	/* Phy Power On */
-	hdmiphy_reg_writeb(hdata, HDMIPHY_POWER,
-				HDMI_PHY_POWER_ON);
-	/* For PHY Mode Setting */
-	hdmiphy_reg_writeb(hdata, HDMIPHY_MODE_SET_DONE,
-				HDMI_PHY_DISABLE_MODE_SET);
-	/* PHY SW Reset */
-	hdmiphy_conf_reset(hdata);
-}
-
-static void hdmiphy_poweroff(struct hdmi_context *hdata)
-{
-	if (hdata->drv_data->type != HDMI_TYPE14)
-		return;
-
-	DRM_DEBUG_KMS("\n");
-
-	/* PHY SW Reset */
-	hdmiphy_conf_reset(hdata);
-	/* For PHY Mode Setting */
-	hdmiphy_reg_writeb(hdata, HDMIPHY_MODE_SET_DONE,
-				HDMI_PHY_ENABLE_MODE_SET);
-
-	/* PHY Power Off */
-	hdmiphy_reg_writeb(hdata, HDMIPHY_POWER,
-				HDMI_PHY_POWER_OFF);
-
-	/* For PHY Mode Setting */
-	hdmiphy_reg_writeb(hdata, HDMIPHY_MODE_SET_DONE,
-				HDMI_PHY_DISABLE_MODE_SET);
-}
-
 static void hdmiphy_conf_apply(struct hdmi_context *hdata)
 {
 	int ret;
@@ -1671,14 +1605,6 @@ static void hdmiphy_conf_apply(struct hdmi_context *hdata)
 	}
 
 	usleep_range(10000, 12000);
-
-	ret = hdmiphy_reg_writeb(hdata, HDMIPHY_MODE_SET_DONE,
-				HDMI_PHY_DISABLE_MODE_SET);
-	if (ret) {
-		DRM_ERROR("failed to enable hdmiphy\n");
-		return;
-	}
-
 }
 
 static void hdmi_conf_apply(struct hdmi_context *hdata)
@@ -1736,7 +1662,6 @@ static void hdmi_enable(struct drm_encoder *encoder)
 	clk_prepare_enable(res->hdmi);
 	clk_prepare_enable(res->sclk_hdmi);
 
-	hdmiphy_poweron(hdata);
 	hdmi_conf_apply(hdata);
 }
 
@@ -1767,8 +1692,6 @@ static void hdmi_disable(struct drm_encoder *encoder)
 	/* HDMI System Disable */
 	hdmi_reg_writemask(hdata, HDMI_CON_0, 0, HDMI_EN);
 
-	hdmiphy_poweroff(hdata);
-
 	cancel_delayed_work(&hdata->hotplug_work);
 
 	clk_disable_unprepare(res->sclk_hdmi);
-- 
cgit v0.10.2


From 1993c33942c7518ceeb77117e5f0e20d5ef269df Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <a.hajda@samsung.com>
Date: Fri, 25 Sep 2015 14:48:19 +0200
Subject: drm/exynos/hdmi: replace all writeb with writel

Registers are 32-bit, even if only lower 8-bits are used.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c
index 814dd8c..50ced709 100644
--- a/drivers/gpu/drm/exynos/exynos_hdmi.c
+++ b/drivers/gpu/drm/exynos/exynos_hdmi.c
@@ -537,7 +537,7 @@ static inline u32 hdmi_reg_read(struct hdmi_context *hdata, u32 reg_id)
 static inline void hdmi_reg_writeb(struct hdmi_context *hdata,
 				 u32 reg_id, u8 value)
 {
-	writeb(value, hdata->regs + hdmi_map_reg(hdata, reg_id));
+	writel(value, hdata->regs + hdmi_map_reg(hdata, reg_id));
 }
 
 static inline void hdmi_reg_writev(struct hdmi_context *hdata, u32 reg_id,
@@ -546,7 +546,7 @@ static inline void hdmi_reg_writev(struct hdmi_context *hdata, u32 reg_id,
 	reg_id = hdmi_map_reg(hdata, reg_id);
 
 	while (--bytes >= 0) {
-		writeb(val & 0xff, hdata->regs + reg_id);
+		writel(val & 0xff, hdata->regs + reg_id);
 		val >>= 8;
 		reg_id += 4;
 	}
@@ -579,7 +579,7 @@ static int hdmiphy_reg_write_buf(struct hdmi_context *hdata,
 	} else {
 		int i;
 		for (i = 0; i < len; i++)
-			writeb(buf[i], hdata->regs_hdmiphy +
+			writel(buf[i], hdata->regs_hdmiphy +
 				((reg_offset + i)<<2));
 		return 0;
 	}
-- 
cgit v0.10.2


From 2445c4a44f200a3137dbaaa3603f2f095c9fdd45 Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <a.hajda@samsung.com>
Date: Fri, 25 Sep 2015 14:48:20 +0200
Subject: drm/exynos/hdmi: fix removal order

DRM device should be destroyed before releasing resources.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c
index 50ced709..725ddb5 100644
--- a/drivers/gpu/drm/exynos/exynos_hdmi.c
+++ b/drivers/gpu/drm/exynos/exynos_hdmi.c
@@ -2046,15 +2046,17 @@ static int hdmi_remove(struct platform_device *pdev)
 
 	cancel_delayed_work_sync(&hdata->hotplug_work);
 
+	component_del(&pdev->dev, &hdmi_component_ops);
+
+	pm_runtime_disable(&pdev->dev);
+
 	if (hdata->res.reg_hdmi_en)
 		regulator_disable(hdata->res.reg_hdmi_en);
 
 	if (hdata->hdmiphy_port)
 		put_device(&hdata->hdmiphy_port->dev);
-	put_device(&hdata->ddc_adpt->dev);
 
-	pm_runtime_disable(&pdev->dev);
-	component_del(&pdev->dev, &hdmi_component_ops);
+	put_device(&hdata->ddc_adpt->dev);
 
 	return 0;
 }
-- 
cgit v0.10.2


From 498d5a373ac4aa20d7ad63770f2196922ae2f26c Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <a.hajda@samsung.com>
Date: Fri, 25 Sep 2015 14:48:21 +0200
Subject: drm/exynos/hdmi: use optional regulator_get for hdmi-en

hdmi-en is an optional regulator so it should be better handled by
devm_regulator_get_optional call.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c
index 725ddb5..841470c 100644
--- a/drivers/gpu/drm/exynos/exynos_hdmi.c
+++ b/drivers/gpu/drm/exynos/exynos_hdmi.c
@@ -1803,19 +1803,17 @@ static int hdmi_resources_init(struct hdmi_context *hdata)
 	}
 	res->regul_count = ARRAY_SIZE(supply);
 
-	res->reg_hdmi_en = devm_regulator_get(dev, "hdmi-en");
-	if (IS_ERR(res->reg_hdmi_en) && PTR_ERR(res->reg_hdmi_en) != -ENOENT) {
-		DRM_ERROR("failed to get hdmi-en regulator\n");
+	res->reg_hdmi_en = devm_regulator_get_optional(dev, "hdmi-en");
+
+	if (PTR_ERR(res->reg_hdmi_en) == -ENODEV)
+		return 0;
+
+	if (IS_ERR(res->reg_hdmi_en))
 		return PTR_ERR(res->reg_hdmi_en);
-	}
-	if (!IS_ERR(res->reg_hdmi_en)) {
-		ret = regulator_enable(res->reg_hdmi_en);
-		if (ret) {
-			DRM_ERROR("failed to enable hdmi-en regulator\n");
-			return ret;
-		}
-	} else
-		res->reg_hdmi_en = NULL;
+
+	ret = regulator_enable(res->reg_hdmi_en);
+	if (ret)
+		DRM_ERROR("failed to enable hdmi-en regulator\n");
 
 	return ret;
 fail:
@@ -2050,7 +2048,7 @@ static int hdmi_remove(struct platform_device *pdev)
 
 	pm_runtime_disable(&pdev->dev);
 
-	if (hdata->res.reg_hdmi_en)
+	if (!IS_ERR(hdata->res.reg_hdmi_en))
 		regulator_disable(hdata->res.reg_hdmi_en);
 
 	if (hdata->hdmiphy_port)
-- 
cgit v0.10.2


From 1ab739db8ed86ab2815d903c0e53d88a88e315c4 Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <a.hajda@samsung.com>
Date: Fri, 25 Sep 2015 14:48:22 +0200
Subject: drm/exynos/hdmi: use constant size array for regulators

Driver always uses the same number of regulators, so there is no point in
dynamic allocation.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c
index 841470c..583dd5e 100644
--- a/drivers/gpu/drm/exynos/exynos_hdmi.c
+++ b/drivers/gpu/drm/exynos/exynos_hdmi.c
@@ -83,6 +83,12 @@ static const u32 hdmi_reg_map[][HDMI_TYPE_COUNT] = {
 	{ HDMI_V13_ACR_CON, HDMI_V14_ACR_CON },
 };
 
+static const char * const supply[] = {
+	"vdd",
+	"vdd_osc",
+	"vdd_pll",
+};
+
 struct hdmi_driver_data {
 	unsigned int type;
 	const struct hdmiphy_config *phy_confs;
@@ -96,9 +102,8 @@ struct hdmi_resources {
 	struct clk			*sclk_pixel;
 	struct clk			*sclk_hdmiphy;
 	struct clk			*mout_hdmi;
-	struct regulator_bulk_data	*regul_bulk;
+	struct regulator_bulk_data	regul_bulk[ARRAY_SIZE(supply)];
 	struct regulator		*reg_hdmi_en;
-	int				regul_count;
 };
 
 struct hdmi_context {
@@ -1652,7 +1657,7 @@ static void hdmi_enable(struct drm_encoder *encoder)
 
 	pm_runtime_get_sync(hdata->dev);
 
-	if (regulator_bulk_enable(res->regul_count, res->regul_bulk))
+	if (regulator_bulk_enable(ARRAY_SIZE(supply), res->regul_bulk))
 		DRM_DEBUG_KMS("failed to enable regulator bulk\n");
 
 	/* set pmu hdmiphy control bit to enable hdmiphy */
@@ -1701,7 +1706,7 @@ static void hdmi_disable(struct drm_encoder *encoder)
 	regmap_update_bits(hdata->pmureg, PMU_HDMI_PHY_CONTROL,
 			PMU_HDMI_PHY_ENABLE_BIT, 0);
 
-	regulator_bulk_disable(res->regul_count, res->regul_bulk);
+	regulator_bulk_disable(ARRAY_SIZE(supply), res->regul_bulk);
 
 	pm_runtime_put_sync(hdata->dev);
 
@@ -1743,11 +1748,6 @@ static int hdmi_resources_init(struct hdmi_context *hdata)
 {
 	struct device *dev = hdata->dev;
 	struct hdmi_resources *res = &hdata->res;
-	static char *supply[] = {
-		"vdd",
-		"vdd_osc",
-		"vdd_pll",
-	};
 	int i, ret;
 
 	DRM_DEBUG_KMS("HDMI resource init\n");
@@ -1786,12 +1786,6 @@ static int hdmi_resources_init(struct hdmi_context *hdata)
 
 	clk_set_parent(res->mout_hdmi, res->sclk_pixel);
 
-	res->regul_bulk = devm_kzalloc(dev, ARRAY_SIZE(supply) *
-		sizeof(res->regul_bulk[0]), GFP_KERNEL);
-	if (!res->regul_bulk) {
-		ret = -ENOMEM;
-		goto fail;
-	}
 	for (i = 0; i < ARRAY_SIZE(supply); ++i) {
 		res->regul_bulk[i].supply = supply[i];
 		res->regul_bulk[i].consumer = NULL;
@@ -1801,7 +1795,6 @@ static int hdmi_resources_init(struct hdmi_context *hdata)
 		DRM_ERROR("failed to get regulators\n");
 		return ret;
 	}
-	res->regul_count = ARRAY_SIZE(supply);
 
 	res->reg_hdmi_en = devm_regulator_get_optional(dev, "hdmi-en");
 
-- 
cgit v0.10.2


From da5e36ae963b24318e0d6b9f226abb795999106a Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <a.hajda@samsung.com>
Date: Fri, 25 Sep 2015 14:48:23 +0200
Subject: drm/exynos/hdmi: simplify clock re-parenting

Driver tries to disable sclk_hdmi during re-parenting, to avoid possible
glitches. It is ineffective as the clock is used also by other devices (mixer).
Anyway driver works without disabling sclk_hdmi.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c
index 583dd5e..475341f 100644
--- a/drivers/gpu/drm/exynos/exynos_hdmi.c
+++ b/drivers/gpu/drm/exynos/exynos_hdmi.c
@@ -1569,9 +1569,7 @@ static void hdmi_mode_apply(struct hdmi_context *hdata)
 
 	hdmiphy_wait_for_pll(hdata);
 
-	clk_disable_unprepare(hdata->res.sclk_hdmi);
 	clk_set_parent(hdata->res.mout_hdmi, hdata->res.sclk_hdmiphy);
-	clk_prepare_enable(hdata->res.sclk_hdmi);
 
 	/* enable HDMI and timing generator */
 	hdmi_start(hdata, true);
@@ -1579,9 +1577,7 @@ static void hdmi_mode_apply(struct hdmi_context *hdata)
 
 static void hdmiphy_conf_reset(struct hdmi_context *hdata)
 {
-	clk_disable_unprepare(hdata->res.sclk_hdmi);
 	clk_set_parent(hdata->res.mout_hdmi, hdata->res.sclk_pixel);
-	clk_prepare_enable(hdata->res.sclk_hdmi);
 
 	/* reset hdmiphy */
 	hdmi_reg_writemask(hdata, HDMI_PHY_RSTOUT, ~0, HDMI_PHY_SW_RSTOUT);
-- 
cgit v0.10.2


From 2228b7cda25b79f106411b97bd039ed0b97067a6 Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <a.hajda@samsung.com>
Date: Fri, 25 Sep 2015 14:48:24 +0200
Subject: drm/exynos/hdmi: convert to gpiod API

The patch converts API to gpiod and moves initialization code
to hdmi_resources_init.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c
index 475341f..c0664d53 100644
--- a/drivers/gpu/drm/exynos/exynos_hdmi.c
+++ b/drivers/gpu/drm/exynos/exynos_hdmi.c
@@ -30,11 +30,11 @@
 #include <linux/delay.h>
 #include <linux/pm_runtime.h>
 #include <linux/clk.h>
+#include <linux/gpio/consumer.h>
 #include <linux/regulator/consumer.h>
 #include <linux/io.h>
 #include <linux/of_address.h>
 #include <linux/of_device.h>
-#include <linux/of_gpio.h>
 #include <linux/hdmi.h>
 #include <linux/component.h>
 #include <linux/mfd/syscon.h>
@@ -46,8 +46,6 @@
 #include "exynos_drm_crtc.h"
 #include "exynos_mixer.h"
 
-#include <linux/gpio.h>
-
 #define ctx_from_connector(c)	container_of(c, struct hdmi_context, connector)
 
 #define HOTPLUG_DEBOUNCE_MS		1100
@@ -129,7 +127,7 @@ struct hdmi_context {
 	struct hdmi_resources		res;
 	const struct hdmi_driver_data	*drv_data;
 
-	int				hpd_gpio;
+	struct gpio_desc 		*hpd_gpio;
 	void __iomem			*regs_hdmiphy;
 
 	struct regmap			*pmureg;
@@ -949,7 +947,7 @@ static enum drm_connector_status hdmi_detect(struct drm_connector *connector,
 {
 	struct hdmi_context *hdata = ctx_from_connector(connector);
 
-	if (gpio_get_value(hdata->hpd_gpio))
+	if (gpiod_get_value(hdata->hpd_gpio))
 		return connector_status_connected;
 
 	return connector_status_disconnected;
@@ -1748,6 +1746,17 @@ static int hdmi_resources_init(struct hdmi_context *hdata)
 
 	DRM_DEBUG_KMS("HDMI resource init\n");
 
+	hdata->hpd_gpio = devm_gpiod_get(dev, "hpd", GPIOD_IN);
+	if (IS_ERR(hdata->hpd_gpio)) {
+		DRM_ERROR("cannot get hpd gpio property\n");
+		return PTR_ERR(hdata->hpd_gpio);
+	}
+
+	hdata->irq = gpiod_to_irq(hdata->hpd_gpio);
+	if (hdata->irq < 0) {
+		DRM_ERROR("failed to get GPIO irq\n");
+		return  hdata->irq;
+	}
 	/* get clocks, power */
 	res->hdmi = devm_clk_get(dev, "hdmi");
 	if (IS_ERR(res->hdmi)) {
@@ -1909,11 +1918,6 @@ static int hdmi_probe(struct platform_device *pdev)
 	platform_set_drvdata(pdev, hdata);
 
 	hdata->dev = dev;
-	hdata->hpd_gpio = of_get_named_gpio(dev->of_node, "hpd-gpio", 0);
-	if (hdata->hpd_gpio < 0) {
-		DRM_ERROR("cannot get hpd gpio property\n");
-		return hdata->hpd_gpio;
-	}
 
 	ret = hdmi_resources_init(hdata);
 	if (ret) {
@@ -1928,12 +1932,6 @@ static int hdmi_probe(struct platform_device *pdev)
 		return ret;
 	}
 
-	ret = devm_gpio_request(dev, hdata->hpd_gpio, "HPD");
-	if (ret) {
-		DRM_ERROR("failed to request HPD gpio\n");
-		return ret;
-	}
-
 	ddc_node = hdmi_legacy_ddc_dt_binding(dev);
 	if (ddc_node)
 		goto out_get_ddc_adpt;
@@ -1981,13 +1979,6 @@ out_get_phy_port:
 		}
 	}
 
-	hdata->irq = gpio_to_irq(hdata->hpd_gpio);
-	if (hdata->irq < 0) {
-		DRM_ERROR("failed to get GPIO irq\n");
-		ret = hdata->irq;
-		goto err_hdmiphy;
-	}
-
 	INIT_DELAYED_WORK(&hdata->hotplug_work, hdmi_hotplug_work_func);
 
 	ret = devm_request_threaded_irq(dev, hdata->irq, NULL,
-- 
cgit v0.10.2


From af1f7c24d619f2c21f7f6cbac140ef469840e6d3 Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <a.hajda@samsung.com>
Date: Fri, 25 Sep 2015 14:48:25 +0200
Subject: drm/exynos/hdmi: remove deprecated hdmi_resources structure

hdmi_resources structure was filled by old platform data code and is not
necessary anymore. The patch removes it at groups together resource related
fields in hdmi_context.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c
index c0664d53..0d538d3c 100644
--- a/drivers/gpu/drm/exynos/exynos_hdmi.c
+++ b/drivers/gpu/drm/exynos/exynos_hdmi.c
@@ -94,16 +94,6 @@ struct hdmi_driver_data {
 	unsigned int is_apb_phy:1;
 };
 
-struct hdmi_resources {
-	struct clk			*hdmi;
-	struct clk			*sclk_hdmi;
-	struct clk			*sclk_pixel;
-	struct clk			*sclk_hdmiphy;
-	struct clk			*mout_hdmi;
-	struct regulator_bulk_data	regul_bulk[ARRAY_SIZE(supply)];
-	struct regulator		*reg_hdmi_en;
-};
-
 struct hdmi_context {
 	struct drm_encoder		encoder;
 	struct device			*dev;
@@ -112,25 +102,25 @@ struct hdmi_context {
 	bool				hpd;
 	bool				powered;
 	bool				dvi_mode;
-
-	void __iomem			*regs;
-	int				irq;
 	struct delayed_work		hotplug_work;
-
-	struct i2c_adapter		*ddc_adpt;
-	struct i2c_client		*hdmiphy_port;
-
-	/* current hdmiphy conf regs */
 	struct drm_display_mode		current_mode;
 	u8				cea_video_id;
-
-	struct hdmi_resources		res;
 	const struct hdmi_driver_data	*drv_data;
 
-	struct gpio_desc 		*hpd_gpio;
+	void __iomem			*regs;
 	void __iomem			*regs_hdmiphy;
-
+	struct i2c_client		*hdmiphy_port;
+	struct i2c_adapter		*ddc_adpt;
+	struct gpio_desc 		*hpd_gpio;
+	int				irq;
 	struct regmap			*pmureg;
+	struct clk			*hdmi;
+	struct clk			*sclk_hdmi;
+	struct clk			*sclk_pixel;
+	struct clk			*sclk_hdmiphy;
+	struct clk			*mout_hdmi;
+	struct regulator_bulk_data	regul_bulk[ARRAY_SIZE(supply)];
+	struct regulator		*reg_hdmi_en;
 };
 
 static inline struct hdmi_context *encoder_to_hdmi(struct drm_encoder *e)
@@ -1567,7 +1557,7 @@ static void hdmi_mode_apply(struct hdmi_context *hdata)
 
 	hdmiphy_wait_for_pll(hdata);
 
-	clk_set_parent(hdata->res.mout_hdmi, hdata->res.sclk_hdmiphy);
+	clk_set_parent(hdata->mout_hdmi, hdata->sclk_hdmiphy);
 
 	/* enable HDMI and timing generator */
 	hdmi_start(hdata, true);
@@ -1575,7 +1565,7 @@ static void hdmi_mode_apply(struct hdmi_context *hdata)
 
 static void hdmiphy_conf_reset(struct hdmi_context *hdata)
 {
-	clk_set_parent(hdata->res.mout_hdmi, hdata->res.sclk_pixel);
+	clk_set_parent(hdata->mout_hdmi, hdata->sclk_pixel);
 
 	/* reset hdmiphy */
 	hdmi_reg_writemask(hdata, HDMI_PHY_RSTOUT, ~0, HDMI_PHY_SW_RSTOUT);
@@ -1642,7 +1632,6 @@ static void hdmi_mode_set(struct drm_encoder *encoder,
 static void hdmi_enable(struct drm_encoder *encoder)
 {
 	struct hdmi_context *hdata = encoder_to_hdmi(encoder);
-	struct hdmi_resources *res = &hdata->res;
 
 	if (hdata->powered)
 		return;
@@ -1651,15 +1640,15 @@ static void hdmi_enable(struct drm_encoder *encoder)
 
 	pm_runtime_get_sync(hdata->dev);
 
-	if (regulator_bulk_enable(ARRAY_SIZE(supply), res->regul_bulk))
+	if (regulator_bulk_enable(ARRAY_SIZE(supply), hdata->regul_bulk))
 		DRM_DEBUG_KMS("failed to enable regulator bulk\n");
 
 	/* set pmu hdmiphy control bit to enable hdmiphy */
 	regmap_update_bits(hdata->pmureg, PMU_HDMI_PHY_CONTROL,
 			PMU_HDMI_PHY_ENABLE_BIT, 1);
 
-	clk_prepare_enable(res->hdmi);
-	clk_prepare_enable(res->sclk_hdmi);
+	clk_prepare_enable(hdata->hdmi);
+	clk_prepare_enable(hdata->sclk_hdmi);
 
 	hdmi_conf_apply(hdata);
 }
@@ -1667,7 +1656,6 @@ static void hdmi_enable(struct drm_encoder *encoder)
 static void hdmi_disable(struct drm_encoder *encoder)
 {
 	struct hdmi_context *hdata = encoder_to_hdmi(encoder);
-	struct hdmi_resources *res = &hdata->res;
 	struct drm_crtc *crtc = encoder->crtc;
 	const struct drm_crtc_helper_funcs *funcs = NULL;
 
@@ -1693,14 +1681,14 @@ static void hdmi_disable(struct drm_encoder *encoder)
 
 	cancel_delayed_work(&hdata->hotplug_work);
 
-	clk_disable_unprepare(res->sclk_hdmi);
-	clk_disable_unprepare(res->hdmi);
+	clk_disable_unprepare(hdata->sclk_hdmi);
+	clk_disable_unprepare(hdata->hdmi);
 
 	/* reset pmu hdmiphy control bit to disable hdmiphy */
 	regmap_update_bits(hdata->pmureg, PMU_HDMI_PHY_CONTROL,
 			PMU_HDMI_PHY_ENABLE_BIT, 0);
 
-	regulator_bulk_disable(ARRAY_SIZE(supply), res->regul_bulk);
+	regulator_bulk_disable(ARRAY_SIZE(supply), hdata->regul_bulk);
 
 	pm_runtime_put_sync(hdata->dev);
 
@@ -1741,7 +1729,6 @@ static irqreturn_t hdmi_irq_thread(int irq, void *arg)
 static int hdmi_resources_init(struct hdmi_context *hdata)
 {
 	struct device *dev = hdata->dev;
-	struct hdmi_resources *res = &hdata->res;
 	int i, ret;
 
 	DRM_DEBUG_KMS("HDMI resource init\n");
@@ -1758,58 +1745,58 @@ static int hdmi_resources_init(struct hdmi_context *hdata)
 		return  hdata->irq;
 	}
 	/* get clocks, power */
-	res->hdmi = devm_clk_get(dev, "hdmi");
-	if (IS_ERR(res->hdmi)) {
+	hdata->hdmi = devm_clk_get(dev, "hdmi");
+	if (IS_ERR(hdata->hdmi)) {
 		DRM_ERROR("failed to get clock 'hdmi'\n");
-		ret = PTR_ERR(res->hdmi);
+		ret = PTR_ERR(hdata->hdmi);
 		goto fail;
 	}
-	res->sclk_hdmi = devm_clk_get(dev, "sclk_hdmi");
-	if (IS_ERR(res->sclk_hdmi)) {
+	hdata->sclk_hdmi = devm_clk_get(dev, "sclk_hdmi");
+	if (IS_ERR(hdata->sclk_hdmi)) {
 		DRM_ERROR("failed to get clock 'sclk_hdmi'\n");
-		ret = PTR_ERR(res->sclk_hdmi);
+		ret = PTR_ERR(hdata->sclk_hdmi);
 		goto fail;
 	}
-	res->sclk_pixel = devm_clk_get(dev, "sclk_pixel");
-	if (IS_ERR(res->sclk_pixel)) {
+	hdata->sclk_pixel = devm_clk_get(dev, "sclk_pixel");
+	if (IS_ERR(hdata->sclk_pixel)) {
 		DRM_ERROR("failed to get clock 'sclk_pixel'\n");
-		ret = PTR_ERR(res->sclk_pixel);
+		ret = PTR_ERR(hdata->sclk_pixel);
 		goto fail;
 	}
-	res->sclk_hdmiphy = devm_clk_get(dev, "sclk_hdmiphy");
-	if (IS_ERR(res->sclk_hdmiphy)) {
+	hdata->sclk_hdmiphy = devm_clk_get(dev, "sclk_hdmiphy");
+	if (IS_ERR(hdata->sclk_hdmiphy)) {
 		DRM_ERROR("failed to get clock 'sclk_hdmiphy'\n");
-		ret = PTR_ERR(res->sclk_hdmiphy);
+		ret = PTR_ERR(hdata->sclk_hdmiphy);
 		goto fail;
 	}
-	res->mout_hdmi = devm_clk_get(dev, "mout_hdmi");
-	if (IS_ERR(res->mout_hdmi)) {
+	hdata->mout_hdmi = devm_clk_get(dev, "mout_hdmi");
+	if (IS_ERR(hdata->mout_hdmi)) {
 		DRM_ERROR("failed to get clock 'mout_hdmi'\n");
-		ret = PTR_ERR(res->mout_hdmi);
+		ret = PTR_ERR(hdata->mout_hdmi);
 		goto fail;
 	}
 
-	clk_set_parent(res->mout_hdmi, res->sclk_pixel);
+	clk_set_parent(hdata->mout_hdmi, hdata->sclk_pixel);
 
 	for (i = 0; i < ARRAY_SIZE(supply); ++i) {
-		res->regul_bulk[i].supply = supply[i];
-		res->regul_bulk[i].consumer = NULL;
+		hdata->regul_bulk[i].supply = supply[i];
+		hdata->regul_bulk[i].consumer = NULL;
 	}
-	ret = devm_regulator_bulk_get(dev, ARRAY_SIZE(supply), res->regul_bulk);
+	ret = devm_regulator_bulk_get(dev, ARRAY_SIZE(supply), hdata->regul_bulk);
 	if (ret) {
 		DRM_ERROR("failed to get regulators\n");
 		return ret;
 	}
 
-	res->reg_hdmi_en = devm_regulator_get_optional(dev, "hdmi-en");
+	hdata->reg_hdmi_en = devm_regulator_get_optional(dev, "hdmi-en");
 
-	if (PTR_ERR(res->reg_hdmi_en) == -ENODEV)
+	if (PTR_ERR(hdata->reg_hdmi_en) == -ENODEV)
 		return 0;
 
-	if (IS_ERR(res->reg_hdmi_en))
-		return PTR_ERR(res->reg_hdmi_en);
+	if (IS_ERR(hdata->reg_hdmi_en))
+		return PTR_ERR(hdata->reg_hdmi_en);
 
-	ret = regulator_enable(res->reg_hdmi_en);
+	ret = regulator_enable(hdata->reg_hdmi_en);
 	if (ret)
 		DRM_ERROR("failed to enable hdmi-en regulator\n");
 
@@ -2028,8 +2015,8 @@ static int hdmi_remove(struct platform_device *pdev)
 
 	pm_runtime_disable(&pdev->dev);
 
-	if (!IS_ERR(hdata->res.reg_hdmi_en))
-		regulator_disable(hdata->res.reg_hdmi_en);
+	if (!IS_ERR(hdata->reg_hdmi_en))
+		regulator_disable(hdata->reg_hdmi_en);
 
 	if (hdata->hdmiphy_port)
 		put_device(&hdata->hdmiphy_port->dev);
-- 
cgit v0.10.2


From 185f22d222fb12d571eed43fc6d5d13e27813f05 Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <a.hajda@samsung.com>
Date: Fri, 25 Sep 2015 14:48:26 +0200
Subject: drm/exynos/hdmi: convert container_of macro to inline function

Inline function is safer than macro, also the name has been changed to
be consistent with other inline function encoder_to_hdmi.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c
index 0d538d3c..c5a0617 100644
--- a/drivers/gpu/drm/exynos/exynos_hdmi.c
+++ b/drivers/gpu/drm/exynos/exynos_hdmi.c
@@ -46,8 +46,6 @@
 #include "exynos_drm_crtc.h"
 #include "exynos_mixer.h"
 
-#define ctx_from_connector(c)	container_of(c, struct hdmi_context, connector)
-
 #define HOTPLUG_DEBOUNCE_MS		1100
 
 /* AVI header and aspect ratio */
@@ -128,6 +126,11 @@ static inline struct hdmi_context *encoder_to_hdmi(struct drm_encoder *e)
 	return container_of(e, struct hdmi_context, encoder);
 }
 
+static inline struct hdmi_context *connector_to_hdmi(struct drm_connector *c)
+{
+	return container_of(c, struct hdmi_context, connector);
+}
+
 struct hdmiphy_config {
 	int pixel_clock;
 	u8 conf[32];
@@ -935,7 +938,7 @@ static void hdmi_reg_infoframe(struct hdmi_context *hdata,
 static enum drm_connector_status hdmi_detect(struct drm_connector *connector,
 				bool force)
 {
-	struct hdmi_context *hdata = ctx_from_connector(connector);
+	struct hdmi_context *hdata = connector_to_hdmi(connector);
 
 	if (gpiod_get_value(hdata->hpd_gpio))
 		return connector_status_connected;
@@ -961,7 +964,7 @@ static struct drm_connector_funcs hdmi_connector_funcs = {
 
 static int hdmi_get_modes(struct drm_connector *connector)
 {
-	struct hdmi_context *hdata = ctx_from_connector(connector);
+	struct hdmi_context *hdata = connector_to_hdmi(connector);
 	struct edid *edid;
 	int ret;
 
@@ -1001,7 +1004,7 @@ static int hdmi_find_phy_conf(struct hdmi_context *hdata, u32 pixel_clock)
 static int hdmi_mode_valid(struct drm_connector *connector,
 			struct drm_display_mode *mode)
 {
-	struct hdmi_context *hdata = ctx_from_connector(connector);
+	struct hdmi_context *hdata = connector_to_hdmi(connector);
 	int ret;
 
 	DRM_DEBUG_KMS("xres=%d, yres=%d, refresh=%d, intl=%d clock=%d\n",
@@ -1022,7 +1025,7 @@ static int hdmi_mode_valid(struct drm_connector *connector,
 
 static struct drm_encoder *hdmi_best_encoder(struct drm_connector *connector)
 {
-	struct hdmi_context *hdata = ctx_from_connector(connector);
+	struct hdmi_context *hdata = connector_to_hdmi(connector);
 
 	return &hdata->encoder;
 }
-- 
cgit v0.10.2


From d24bb3e59d3863304f81e71e10c0e9a1ae13c16d Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <a.hajda@samsung.com>
Date: Fri, 25 Sep 2015 14:48:27 +0200
Subject: drm/exynos/hdmi: improve HDMI/ACR related code

Simple formula can be used to calculate CTS and N coefficients.
Additionaly ACR registers have different offsets for different versions
of IP.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c
index c5a0617..a4ec8b9 100644
--- a/drivers/gpu/drm/exynos/exynos_hdmi.c
+++ b/drivers/gpu/drm/exynos/exynos_hdmi.c
@@ -71,12 +71,18 @@ enum hdmi_mapped_regs {
 	HDMI_PHY_STATUS = HDMI_MAPPED_BASE,
 	HDMI_PHY_RSTOUT,
 	HDMI_ACR_CON,
+	HDMI_ACR_MCTS0,
+	HDMI_ACR_CTS0,
+	HDMI_ACR_N0
 };
 
 static const u32 hdmi_reg_map[][HDMI_TYPE_COUNT] = {
 	{ HDMI_V13_PHY_STATUS, HDMI_PHY_STATUS_0 },
 	{ HDMI_V13_PHY_RSTOUT, HDMI_V14_PHY_RSTOUT },
 	{ HDMI_V13_ACR_CON, HDMI_V14_ACR_CON },
+	{ HDMI_V13_ACR_MCTS0, HDMI_V14_ACR_MCTS0 },
+	{ HDMI_V13_ACR_CTS0, HDMI_V14_ACR_CTS0 },
+	{ HDMI_V13_ACR_N0, HDMI_V14_ACR_N0 },
 };
 
 static const char * const supply[] = {
@@ -1106,65 +1112,16 @@ static bool hdmi_mode_fixup(struct drm_encoder *encoder,
 	return true;
 }
 
-static void hdmi_set_acr(u32 freq, u8 *acr)
+static void hdmi_reg_acr(struct hdmi_context *hdata, u32 freq)
 {
 	u32 n, cts;
 
-	switch (freq) {
-	case 32000:
-		n = 4096;
-		cts = 27000;
-		break;
-	case 44100:
-		n = 6272;
-		cts = 30000;
-		break;
-	case 88200:
-		n = 12544;
-		cts = 30000;
-		break;
-	case 176400:
-		n = 25088;
-		cts = 30000;
-		break;
-	case 48000:
-		n = 6144;
-		cts = 27000;
-		break;
-	case 96000:
-		n = 12288;
-		cts = 27000;
-		break;
-	case 192000:
-		n = 24576;
-		cts = 27000;
-		break;
-	default:
-		n = 0;
-		cts = 0;
-		break;
-	}
-
-	acr[1] = cts >> 16;
-	acr[2] = cts >> 8 & 0xff;
-	acr[3] = cts & 0xff;
+	cts = (freq % 9) ? 27000 : 30000;
+	n = 128 * freq / (27000000 / cts);
 
-	acr[4] = n >> 16;
-	acr[5] = n >> 8 & 0xff;
-	acr[6] = n & 0xff;
-}
-
-static void hdmi_reg_acr(struct hdmi_context *hdata, u8 *acr)
-{
-	hdmi_reg_writeb(hdata, HDMI_ACR_N0, acr[6]);
-	hdmi_reg_writeb(hdata, HDMI_ACR_N1, acr[5]);
-	hdmi_reg_writeb(hdata, HDMI_ACR_N2, acr[4]);
-	hdmi_reg_writeb(hdata, HDMI_ACR_MCTS0, acr[3]);
-	hdmi_reg_writeb(hdata, HDMI_ACR_MCTS1, acr[2]);
-	hdmi_reg_writeb(hdata, HDMI_ACR_MCTS2, acr[1]);
-	hdmi_reg_writeb(hdata, HDMI_ACR_CTS0, acr[3]);
-	hdmi_reg_writeb(hdata, HDMI_ACR_CTS1, acr[2]);
-	hdmi_reg_writeb(hdata, HDMI_ACR_CTS2, acr[1]);
+	hdmi_reg_writev(hdata, HDMI_ACR_N0, 3, n);
+	hdmi_reg_writev(hdata, HDMI_ACR_MCTS0, 3, cts);
+	hdmi_reg_writev(hdata, HDMI_ACR_CTS0, 3, cts);
 	hdmi_reg_writeb(hdata, HDMI_ACR_CON, 4);
 }
 
@@ -1173,7 +1130,6 @@ static void hdmi_audio_init(struct hdmi_context *hdata)
 	u32 sample_rate, bits_per_sample;
 	u32 data_num, bit_ch, sample_frq;
 	u32 val;
-	u8 acr[7];
 
 	sample_rate = 44100;
 	bits_per_sample = 16;
@@ -1193,8 +1149,7 @@ static void hdmi_audio_init(struct hdmi_context *hdata)
 		break;
 	}
 
-	hdmi_set_acr(sample_rate, acr);
-	hdmi_reg_acr(hdata, acr);
+	hdmi_reg_acr(hdata, sample_rate);
 
 	hdmi_reg_writeb(hdata, HDMI_I2S_MUX_CON, HDMI_I2S_IN_DISABLE
 				| HDMI_I2S_AUD_I2S | HDMI_I2S_CUV_I2S_ENABLE
diff --git a/drivers/gpu/drm/exynos/regs-hdmi.h b/drivers/gpu/drm/exynos/regs-hdmi.h
index 8b2d337..8c891e5 100644
--- a/drivers/gpu/drm/exynos/regs-hdmi.h
+++ b/drivers/gpu/drm/exynos/regs-hdmi.h
@@ -72,7 +72,6 @@
 #define HDMI_V13_V_SYNC_GEN_3_0		HDMI_CORE_BASE(0x0150)
 #define HDMI_V13_V_SYNC_GEN_3_1		HDMI_CORE_BASE(0x0154)
 #define HDMI_V13_V_SYNC_GEN_3_2		HDMI_CORE_BASE(0x0158)
-#define HDMI_V13_ACR_CON		HDMI_CORE_BASE(0x0180)
 #define HDMI_V13_AVI_CON		HDMI_CORE_BASE(0x0300)
 #define HDMI_V13_AVI_BYTE(n)		HDMI_CORE_BASE(0x0320 + 4 * (n))
 #define HDMI_V13_DC_CONTROL		HDMI_CORE_BASE(0x05C0)
@@ -277,16 +276,26 @@
 #define HDMI_ASP_CHCFG2			HDMI_CORE_BASE(0x0318)
 #define HDMI_ASP_CHCFG3			HDMI_CORE_BASE(0x031C)
 
+#define HDMI_V13_ACR_CON		HDMI_CORE_BASE(0x0180)
+#define HDMI_V13_ACR_MCTS0		HDMI_CORE_BASE(0x0184)
+#define HDMI_V13_ACR_MCTS1		HDMI_CORE_BASE(0x0188)
+#define HDMI_V13_ACR_MCTS2		HDMI_CORE_BASE(0x018C)
+#define HDMI_V13_ACR_CTS0		HDMI_CORE_BASE(0x0190)
+#define HDMI_V13_ACR_CTS1		HDMI_CORE_BASE(0x0194)
+#define HDMI_V13_ACR_CTS2		HDMI_CORE_BASE(0x0198)
+#define HDMI_V13_ACR_N0			HDMI_CORE_BASE(0x01A0)
+#define HDMI_V13_ACR_N1			HDMI_CORE_BASE(0x01A4)
+#define HDMI_V13_ACR_N2			HDMI_CORE_BASE(0x01A8)
 #define HDMI_V14_ACR_CON		HDMI_CORE_BASE(0x0400)
-#define HDMI_ACR_MCTS0			HDMI_CORE_BASE(0x0410)
-#define HDMI_ACR_MCTS1			HDMI_CORE_BASE(0x0414)
-#define HDMI_ACR_MCTS2			HDMI_CORE_BASE(0x0418)
-#define HDMI_ACR_CTS0			HDMI_CORE_BASE(0x0420)
-#define HDMI_ACR_CTS1			HDMI_CORE_BASE(0x0424)
-#define HDMI_ACR_CTS2			HDMI_CORE_BASE(0x0428)
-#define HDMI_ACR_N0			HDMI_CORE_BASE(0x0430)
-#define HDMI_ACR_N1			HDMI_CORE_BASE(0x0434)
-#define HDMI_ACR_N2			HDMI_CORE_BASE(0x0438)
+#define HDMI_V14_ACR_MCTS0		HDMI_CORE_BASE(0x0410)
+#define HDMI_V14_ACR_MCTS1		HDMI_CORE_BASE(0x0414)
+#define HDMI_V14_ACR_MCTS2		HDMI_CORE_BASE(0x0418)
+#define HDMI_V14_ACR_CTS0		HDMI_CORE_BASE(0x0420)
+#define HDMI_V14_ACR_CTS1		HDMI_CORE_BASE(0x0424)
+#define HDMI_V14_ACR_CTS2		HDMI_CORE_BASE(0x0428)
+#define HDMI_V14_ACR_N0			HDMI_CORE_BASE(0x0430)
+#define HDMI_V14_ACR_N1			HDMI_CORE_BASE(0x0434)
+#define HDMI_V14_ACR_N2			HDMI_CORE_BASE(0x0438)
 
 /* Packet related registers */
 #define HDMI_ACP_CON			HDMI_CORE_BASE(0x0500)
-- 
cgit v0.10.2


From cf3e0fdd54f7d792e158cbb87ead1d9e6a5c58f7 Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <a.hajda@samsung.com>
Date: Fri, 25 Sep 2015 14:48:28 +0200
Subject: drm/exynos/hdmi: remove unused field

The patch removes unused hdmi_context field.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c
index a4ec8b9..b0f5ff4 100644
--- a/drivers/gpu/drm/exynos/exynos_hdmi.c
+++ b/drivers/gpu/drm/exynos/exynos_hdmi.c
@@ -103,7 +103,6 @@ struct hdmi_context {
 	struct device			*dev;
 	struct drm_device		*drm_dev;
 	struct drm_connector		connector;
-	bool				hpd;
 	bool				powered;
 	bool				dvi_mode;
 	struct delayed_work		hotplug_work;
-- 
cgit v0.10.2


From 02b3de431111014643c48a326b46d127df061658 Mon Sep 17 00:00:00 2001
From: Tomasz Stanislawski <t.stanislaws@samsung.com>
Date: Fri, 25 Sep 2015 14:48:29 +0200
Subject: drm: exynos: mixer: fix using usleep() in atomic context

This patch fixes calling usleep_range() after taking reg_slock
using spin_lock_irqsave(). The mdelay() is used instead.
Waiting in atomic context is not the best idea in general.
Hopefully, waiting occurs only when Video Processor fails
to reset correctly.

Signed-off-by: Tomasz Stanislawski <t.stanislaws@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/exynos_mixer.c b/drivers/gpu/drm/exynos/exynos_mixer.c
index 7f81cce..ecf956c 100644
--- a/drivers/gpu/drm/exynos/exynos_mixer.c
+++ b/drivers/gpu/drm/exynos/exynos_mixer.c
@@ -652,7 +652,7 @@ static void vp_win_reset(struct mixer_context *ctx)
 		/* waiting until VP_SRESET_PROCESSING is 0 */
 		if (~vp_reg_read(res, VP_SRESET) & VP_SRESET_PROCESSING)
 			break;
-		usleep_range(10000, 12000);
+		mdelay(10);
 	}
 	WARN(tries == 0, "failed to reset Video Processor\n");
 }
-- 
cgit v0.10.2


From c691349ca429c3a57487eca857f87d2b2442c359 Mon Sep 17 00:00:00 2001
From: Ingi Kim <ingi2.kim@samsung.com>
Date: Fri, 2 Oct 2015 17:59:26 +0900
Subject: drm/exynos: fix spelling errors

This patch fixes spelling errors in drm fimc/gsc
inavild -> invaild

Signed-off-by: Ingi Kim <ingi2.kim@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimc.c b/drivers/gpu/drm/exynos/exynos_drm_fimc.c
index dd3a5e6..c747824 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fimc.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fimc.c
@@ -466,7 +466,7 @@ static int fimc_src_set_fmt_order(struct fimc_context *ctx, u32 fmt)
 			EXYNOS_MSCTRL_C_INT_IN_2PLANE);
 		break;
 	default:
-		dev_err(ippdrv->dev, "inavlid source yuv order 0x%x.\n", fmt);
+		dev_err(ippdrv->dev, "invalid source yuv order 0x%x.\n", fmt);
 		return -EINVAL;
 	}
 
@@ -513,7 +513,7 @@ static int fimc_src_set_fmt(struct device *dev, u32 fmt)
 		cfg |= EXYNOS_MSCTRL_INFORMAT_YCBCR420;
 		break;
 	default:
-		dev_err(ippdrv->dev, "inavlid source format 0x%x.\n", fmt);
+		dev_err(ippdrv->dev, "invalid source format 0x%x.\n", fmt);
 		return -EINVAL;
 	}
 
@@ -578,7 +578,7 @@ static int fimc_src_set_transf(struct device *dev,
 			cfg1 &= ~EXYNOS_MSCTRL_FLIP_Y_MIRROR;
 		break;
 	default:
-		dev_err(ippdrv->dev, "inavlid degree value %d.\n", degree);
+		dev_err(ippdrv->dev, "invalid degree value %d.\n", degree);
 		return -EINVAL;
 	}
 
@@ -701,7 +701,7 @@ static int fimc_src_set_addr(struct device *dev,
 		property->prop_id, buf_id, buf_type);
 
 	if (buf_id > FIMC_MAX_SRC) {
-		dev_info(ippdrv->dev, "inavlid buf_id %d.\n", buf_id);
+		dev_info(ippdrv->dev, "invalid buf_id %d.\n", buf_id);
 		return -ENOMEM;
 	}
 
@@ -812,7 +812,7 @@ static int fimc_dst_set_fmt_order(struct fimc_context *ctx, u32 fmt)
 		cfg |= EXYNOS_CIOCTRL_YCBCR_2PLANE;
 		break;
 	default:
-		dev_err(ippdrv->dev, "inavlid target yuv order 0x%x.\n", fmt);
+		dev_err(ippdrv->dev, "invalid target yuv order 0x%x.\n", fmt);
 		return -EINVAL;
 	}
 
@@ -865,7 +865,7 @@ static int fimc_dst_set_fmt(struct device *dev, u32 fmt)
 			cfg |= EXYNOS_CITRGFMT_OUTFORMAT_YCBCR420;
 			break;
 		default:
-			dev_err(ippdrv->dev, "inavlid target format 0x%x.\n",
+			dev_err(ippdrv->dev, "invalid target format 0x%x.\n",
 				fmt);
 			return -EINVAL;
 		}
@@ -929,7 +929,7 @@ static int fimc_dst_set_transf(struct device *dev,
 			cfg &= ~EXYNOS_CITRGFMT_FLIP_Y_MIRROR;
 		break;
 	default:
-		dev_err(ippdrv->dev, "inavlid degree value %d.\n", degree);
+		dev_err(ippdrv->dev, "invalid degree value %d.\n", degree);
 		return -EINVAL;
 	}
 
@@ -1160,7 +1160,7 @@ static int fimc_dst_set_addr(struct device *dev,
 		property->prop_id, buf_id, buf_type);
 
 	if (buf_id > FIMC_MAX_DST) {
-		dev_info(ippdrv->dev, "inavlid buf_id %d.\n", buf_id);
+		dev_info(ippdrv->dev, "invalid buf_id %d.\n", buf_id);
 		return -ENOMEM;
 	}
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_gsc.c b/drivers/gpu/drm/exynos/exynos_drm_gsc.c
index 808a0a0..11b87d2 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_gsc.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_gsc.c
@@ -543,7 +543,7 @@ static int gsc_src_set_fmt(struct device *dev, u32 fmt)
 			GSC_IN_YUV420_2P);
 		break;
 	default:
-		dev_err(ippdrv->dev, "inavlid target yuv order 0x%x.\n", fmt);
+		dev_err(ippdrv->dev, "invalid target yuv order 0x%x.\n", fmt);
 		return -EINVAL;
 	}
 
@@ -595,7 +595,7 @@ static int gsc_src_set_transf(struct device *dev,
 			cfg &= ~GSC_IN_ROT_YFLIP;
 		break;
 	default:
-		dev_err(ippdrv->dev, "inavlid degree value %d.\n", degree);
+		dev_err(ippdrv->dev, "invalid degree value %d.\n", degree);
 		return -EINVAL;
 	}
 
@@ -721,7 +721,7 @@ static int gsc_src_set_addr(struct device *dev,
 		property->prop_id, buf_id, buf_type);
 
 	if (buf_id > GSC_MAX_SRC) {
-		dev_info(ippdrv->dev, "inavlid buf_id %d.\n", buf_id);
+		dev_info(ippdrv->dev, "invalid buf_id %d.\n", buf_id);
 		return -EINVAL;
 	}
 
@@ -814,7 +814,7 @@ static int gsc_dst_set_fmt(struct device *dev, u32 fmt)
 			GSC_OUT_YUV420_2P);
 		break;
 	default:
-		dev_err(ippdrv->dev, "inavlid target yuv order 0x%x.\n", fmt);
+		dev_err(ippdrv->dev, "invalid target yuv order 0x%x.\n", fmt);
 		return -EINVAL;
 	}
 
@@ -866,7 +866,7 @@ static int gsc_dst_set_transf(struct device *dev,
 			cfg &= ~GSC_IN_ROT_YFLIP;
 		break;
 	default:
-		dev_err(ippdrv->dev, "inavlid degree value %d.\n", degree);
+		dev_err(ippdrv->dev, "invalid degree value %d.\n", degree);
 		return -EINVAL;
 	}
 
@@ -1176,7 +1176,7 @@ static int gsc_dst_set_addr(struct device *dev,
 		property->prop_id, buf_id, buf_type);
 
 	if (buf_id > GSC_MAX_DST) {
-		dev_info(ippdrv->dev, "inavlid buf_id %d.\n", buf_id);
+		dev_info(ippdrv->dev, "invalid buf_id %d.\n", buf_id);
 		return -EINVAL;
 	}
 
-- 
cgit v0.10.2


From 5d3d099574ba6319998ae9275090e42140d6b37a Mon Sep 17 00:00:00 2001
From: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Date: Mon, 12 Oct 2015 22:07:48 +0900
Subject: drm/exynos: add global macro for the default primary plane

Define DEFAULT_WIN as zero to help set the primary plane on all CRTCs.
Some CRTCs were defining a variable to store the default window, but that
is not necessary as the default (primary) window is always the window zero.

Signed-off-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
index b3c7307..f467ce0 100644
--- a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
+++ b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
@@ -33,7 +33,6 @@ struct decon_context {
 	struct exynos_drm_plane		planes[WINDOWS_NR];
 	void __iomem			*addr;
 	struct clk			*clks[6];
-	unsigned int			default_win;
 	unsigned long			irq_flags;
 	int				pipe;
 	bool				suspended;
@@ -501,7 +500,7 @@ static int decon_bind(struct device *dev, struct device *master, void *data)
 	ctx->pipe = priv->pipe++;
 
 	for (zpos = 0; zpos < WINDOWS_NR; zpos++) {
-		type = (zpos == ctx->default_win) ? DRM_PLANE_TYPE_PRIMARY :
+		type = (zpos == DEFAULT_WIN) ? DRM_PLANE_TYPE_PRIMARY :
 							DRM_PLANE_TYPE_OVERLAY;
 		ret = exynos_plane_init(drm_dev, &ctx->planes[zpos],
 				1 << ctx->pipe, type, decon_formats,
@@ -510,7 +509,7 @@ static int decon_bind(struct device *dev, struct device *master, void *data)
 			return ret;
 	}
 
-	exynos_plane = &ctx->planes[ctx->default_win];
+	exynos_plane = &ctx->planes[DEFAULT_WIN];
 	ctx->crtc = exynos_drm_crtc_create(drm_dev, &exynos_plane->base,
 					ctx->pipe, EXYNOS_DISPLAY_TYPE_LCD,
 					&decon_crtc_ops, ctx);
@@ -607,7 +606,6 @@ static int exynos5433_decon_probe(struct platform_device *pdev)
 	if (!ctx)
 		return -ENOMEM;
 
-	ctx->default_win = 0;
 	ctx->suspended = true;
 	ctx->dev = dev;
 	if (of_get_child_by_name(dev->of_node, "i80-if-timings"))
diff --git a/drivers/gpu/drm/exynos/exynos7_drm_decon.c b/drivers/gpu/drm/exynos/exynos7_drm_decon.c
index e6cbaca..bd0f96b 100644
--- a/drivers/gpu/drm/exynos/exynos7_drm_decon.c
+++ b/drivers/gpu/drm/exynos/exynos7_drm_decon.c
@@ -51,7 +51,6 @@ struct decon_context {
 	struct clk			*eclk;
 	struct clk			*vclk;
 	void __iomem			*regs;
-	unsigned int			default_win;
 	unsigned long			irq_flags;
 	bool				i80_if;
 	bool				suspended;
@@ -690,7 +689,7 @@ static int decon_bind(struct device *dev, struct device *master, void *data)
 	}
 
 	for (zpos = 0; zpos < WINDOWS_NR; zpos++) {
-		type = (zpos == ctx->default_win) ? DRM_PLANE_TYPE_PRIMARY :
+		type = (zpos == DEFAULT_WIN) ? DRM_PLANE_TYPE_PRIMARY :
 						DRM_PLANE_TYPE_OVERLAY;
 		ret = exynos_plane_init(drm_dev, &ctx->planes[zpos],
 					1 << ctx->pipe, type, decon_formats,
@@ -699,7 +698,7 @@ static int decon_bind(struct device *dev, struct device *master, void *data)
 			return ret;
 	}
 
-	exynos_plane = &ctx->planes[ctx->default_win];
+	exynos_plane = &ctx->planes[DEFAULT_WIN];
 	ctx->crtc = exynos_drm_crtc_create(drm_dev, &exynos_plane->base,
 					   ctx->pipe, EXYNOS_DISPLAY_TYPE_LCD,
 					   &decon_crtc_ops, ctx);
diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.h b/drivers/gpu/drm/exynos/exynos_drm_drv.h
index 6c717ba..638fc43 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_drv.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_drv.h
@@ -22,6 +22,8 @@
 #define MAX_PLANE	5
 #define MAX_FB_BUFFER	4
 
+#define DEFAULT_WIN	0
+
 #define to_exynos_crtc(x)	container_of(x, struct exynos_drm_crtc, base)
 #define to_exynos_plane(x)	container_of(x, struct exynos_drm_plane, base)
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimd.c b/drivers/gpu/drm/exynos/exynos_drm_fimd.c
index 3d1aba6..7505a71 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fimd.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fimd.c
@@ -153,7 +153,6 @@ struct fimd_context {
 	struct clk			*lcd_clk;
 	void __iomem			*regs;
 	struct regmap			*sysreg;
-	unsigned int			default_win;
 	unsigned long			irq_flags;
 	u32				vidcon0;
 	u32				vidcon1;
@@ -949,7 +948,7 @@ static int fimd_bind(struct device *dev, struct device *master, void *data)
 	ctx->pipe = priv->pipe++;
 
 	for (zpos = 0; zpos < WINDOWS_NR; zpos++) {
-		type = (zpos == ctx->default_win) ? DRM_PLANE_TYPE_PRIMARY :
+		type = (zpos == DEFAULT_WIN) ? DRM_PLANE_TYPE_PRIMARY :
 						DRM_PLANE_TYPE_OVERLAY;
 		ret = exynos_plane_init(drm_dev, &ctx->planes[zpos],
 					1 << ctx->pipe, type, fimd_formats,
@@ -958,7 +957,7 @@ static int fimd_bind(struct device *dev, struct device *master, void *data)
 			return ret;
 	}
 
-	exynos_plane = &ctx->planes[ctx->default_win];
+	exynos_plane = &ctx->planes[DEFAULT_WIN];
 	ctx->crtc = exynos_drm_crtc_create(drm_dev, &exynos_plane->base,
 					   ctx->pipe, EXYNOS_DISPLAY_TYPE_LCD,
 					   &fimd_crtc_ops, ctx);
diff --git a/drivers/gpu/drm/exynos/exynos_drm_vidi.c b/drivers/gpu/drm/exynos/exynos_drm_vidi.c
index 75718e1..fc57687 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_vidi.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_vidi.c
@@ -42,7 +42,6 @@ struct vidi_context {
 	struct exynos_drm_plane		planes[WINDOWS_NR];
 	struct edid			*raw_edid;
 	unsigned int			clkdiv;
-	unsigned int			default_win;
 	unsigned long			irq_flags;
 	unsigned int			connected;
 	bool				vblank_on;
@@ -446,7 +445,7 @@ static int vidi_bind(struct device *dev, struct device *master, void *data)
 	vidi_ctx_initialize(ctx, drm_dev);
 
 	for (zpos = 0; zpos < WINDOWS_NR; zpos++) {
-		type = (zpos == ctx->default_win) ? DRM_PLANE_TYPE_PRIMARY :
+		type = (zpos == DEFAULT_WIN) ? DRM_PLANE_TYPE_PRIMARY :
 						DRM_PLANE_TYPE_OVERLAY;
 		ret = exynos_plane_init(drm_dev, &ctx->planes[zpos],
 					1 << ctx->pipe, type, formats,
@@ -455,7 +454,7 @@ static int vidi_bind(struct device *dev, struct device *master, void *data)
 			return ret;
 	}
 
-	exynos_plane = &ctx->planes[ctx->default_win];
+	exynos_plane = &ctx->planes[DEFAULT_WIN];
 	ctx->crtc = exynos_drm_crtc_create(drm_dev, &exynos_plane->base,
 					   ctx->pipe, EXYNOS_DISPLAY_TYPE_VIDI,
 					   &vidi_crtc_ops, ctx);
@@ -507,7 +506,6 @@ static int vidi_probe(struct platform_device *pdev)
 	if (!ctx)
 		return -ENOMEM;
 
-	ctx->default_win = 0;
 	ctx->pdev = pdev;
 
 	INIT_WORK(&ctx->work, vidi_fake_vblank_handler);
diff --git a/drivers/gpu/drm/exynos/exynos_mixer.c b/drivers/gpu/drm/exynos/exynos_mixer.c
index ecf956c..f83e27d 100644
--- a/drivers/gpu/drm/exynos/exynos_mixer.c
+++ b/drivers/gpu/drm/exynos/exynos_mixer.c
@@ -42,7 +42,6 @@
 #include "exynos_mixer.h"
 
 #define MIXER_WIN_NR		3
-#define MIXER_DEFAULT_WIN	0
 #define VP_DEFAULT_WIN		2
 
 /* The pixelformats that are natively supported by the mixer. */
@@ -600,7 +599,7 @@ static void mixer_graph_buffer(struct mixer_context *ctx,
 
 	/* setup display size */
 	if (ctx->mxr_ver == MXR_VER_128_0_0_184 &&
-		win == MIXER_DEFAULT_WIN) {
+		win == DEFAULT_WIN) {
 		val  = MXR_MXR_RES_HEIGHT(mode->vdisplay);
 		val |= MXR_MXR_RES_WIDTH(mode->hdisplay);
 		mixer_reg_write(res, MXR_RESOLUTION, val);
@@ -1197,7 +1196,7 @@ static int mixer_bind(struct device *dev, struct device *manager, void *data)
 		const uint32_t *formats;
 		unsigned int fcount;
 
-		type = (zpos == MIXER_DEFAULT_WIN) ? DRM_PLANE_TYPE_PRIMARY :
+		type = (zpos == DEFAULT_WIN) ? DRM_PLANE_TYPE_PRIMARY :
 						DRM_PLANE_TYPE_OVERLAY;
 		if (zpos < VP_DEFAULT_WIN) {
 			formats = mixer_formats;
@@ -1214,7 +1213,7 @@ static int mixer_bind(struct device *dev, struct device *manager, void *data)
 			return ret;
 	}
 
-	exynos_plane = &ctx->planes[MIXER_DEFAULT_WIN];
+	exynos_plane = &ctx->planes[DEFAULT_WIN];
 	ctx->crtc = exynos_drm_crtc_create(drm_dev, &exynos_plane->base,
 					   ctx->pipe, EXYNOS_DISPLAY_TYPE_HDMI,
 					   &mixer_crtc_ops, ctx);
-- 
cgit v0.10.2


From 323db0ed7de21dff93affd51c4c742a3dec1c103 Mon Sep 17 00:00:00 2001
From: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Date: Fri, 4 Sep 2015 19:05:57 -0300
Subject: drm/exynos: add cursor plane support

Set one of the planes for each crtc driver as a cursor plane enabled
window managers to fully work on exynos.

Signed-off-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
index f467ce0..1ea26dbb 100644
--- a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
+++ b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
@@ -24,6 +24,7 @@
 #include "exynos_drm_iommu.h"
 
 #define WINDOWS_NR	3
+#define CURSOR_WIN	2
 #define MIN_FB_WIDTH_FOR_16WORD_BURST	128
 
 struct decon_context {
@@ -500,8 +501,7 @@ static int decon_bind(struct device *dev, struct device *master, void *data)
 	ctx->pipe = priv->pipe++;
 
 	for (zpos = 0; zpos < WINDOWS_NR; zpos++) {
-		type = (zpos == DEFAULT_WIN) ? DRM_PLANE_TYPE_PRIMARY :
-							DRM_PLANE_TYPE_OVERLAY;
+		type = exynos_plane_get_type(zpos, CURSOR_WIN);
 		ret = exynos_plane_init(drm_dev, &ctx->planes[zpos],
 				1 << ctx->pipe, type, decon_formats,
 				ARRAY_SIZE(decon_formats), zpos);
diff --git a/drivers/gpu/drm/exynos/exynos7_drm_decon.c b/drivers/gpu/drm/exynos/exynos7_drm_decon.c
index bd0f96b..ead2b16 100644
--- a/drivers/gpu/drm/exynos/exynos7_drm_decon.c
+++ b/drivers/gpu/drm/exynos/exynos7_drm_decon.c
@@ -40,6 +40,7 @@
 #define MIN_FB_WIDTH_FOR_16WORD_BURST 128
 
 #define WINDOWS_NR	2
+#define CURSOR_WIN	1
 
 struct decon_context {
 	struct device			*dev;
@@ -689,8 +690,7 @@ static int decon_bind(struct device *dev, struct device *master, void *data)
 	}
 
 	for (zpos = 0; zpos < WINDOWS_NR; zpos++) {
-		type = (zpos == DEFAULT_WIN) ? DRM_PLANE_TYPE_PRIMARY :
-						DRM_PLANE_TYPE_OVERLAY;
+		type = exynos_plane_get_type(zpos, CURSOR_WIN);
 		ret = exynos_plane_init(drm_dev, &ctx->planes[zpos],
 					1 << ctx->pipe, type, decon_formats,
 					ARRAY_SIZE(decon_formats), zpos);
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimd.c b/drivers/gpu/drm/exynos/exynos_drm_fimd.c
index 7505a71..bd75c15 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fimd.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fimd.c
@@ -87,6 +87,7 @@
 
 /* FIMD has totally five hardware windows. */
 #define WINDOWS_NR	5
+#define CURSOR_WIN	4
 
 struct fimd_driver_data {
 	unsigned int timing_base;
@@ -948,8 +949,7 @@ static int fimd_bind(struct device *dev, struct device *master, void *data)
 	ctx->pipe = priv->pipe++;
 
 	for (zpos = 0; zpos < WINDOWS_NR; zpos++) {
-		type = (zpos == DEFAULT_WIN) ? DRM_PLANE_TYPE_PRIMARY :
-						DRM_PLANE_TYPE_OVERLAY;
+		type = exynos_plane_get_type(zpos, CURSOR_WIN);
 		ret = exynos_plane_init(drm_dev, &ctx->planes[zpos],
 					1 << ctx->pipe, type, fimd_formats,
 					ARRAY_SIZE(fimd_formats), zpos);
diff --git a/drivers/gpu/drm/exynos/exynos_drm_plane.c b/drivers/gpu/drm/exynos/exynos_drm_plane.c
index 7148224..80b2151 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_plane.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_plane.c
@@ -208,6 +208,17 @@ static void exynos_plane_attach_zpos_property(struct drm_plane *plane,
 	drm_object_attach_property(&plane->base, prop, zpos);
 }
 
+enum drm_plane_type exynos_plane_get_type(unsigned int zpos,
+					  unsigned int cursor_win)
+{
+		if (zpos == DEFAULT_WIN)
+			return DRM_PLANE_TYPE_PRIMARY;
+		else if (zpos == cursor_win)
+			return DRM_PLANE_TYPE_CURSOR;
+		else
+			return DRM_PLANE_TYPE_OVERLAY;
+}
+
 int exynos_plane_init(struct drm_device *dev,
 		      struct exynos_drm_plane *exynos_plane,
 		      unsigned long possible_crtcs, enum drm_plane_type type,
diff --git a/drivers/gpu/drm/exynos/exynos_drm_plane.h b/drivers/gpu/drm/exynos/exynos_drm_plane.h
index 476c934..abb641e 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_plane.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_plane.h
@@ -9,6 +9,8 @@
  *
  */
 
+enum drm_plane_type exynos_plane_get_type(unsigned int zpos,
+					  unsigned int cursor_win);
 int exynos_plane_init(struct drm_device *dev,
 		      struct exynos_drm_plane *exynos_plane,
 		      unsigned long possible_crtcs, enum drm_plane_type type,
diff --git a/drivers/gpu/drm/exynos/exynos_drm_vidi.c b/drivers/gpu/drm/exynos/exynos_drm_vidi.c
index fc57687..669362c 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_vidi.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_vidi.c
@@ -29,6 +29,7 @@
 
 /* vidi has totally three virtual windows. */
 #define WINDOWS_NR		3
+#define CURSOR_WIN		2
 
 #define ctx_from_connector(c)	container_of(c, struct vidi_context, \
 					connector)
@@ -445,8 +446,7 @@ static int vidi_bind(struct device *dev, struct device *master, void *data)
 	vidi_ctx_initialize(ctx, drm_dev);
 
 	for (zpos = 0; zpos < WINDOWS_NR; zpos++) {
-		type = (zpos == DEFAULT_WIN) ? DRM_PLANE_TYPE_PRIMARY :
-						DRM_PLANE_TYPE_OVERLAY;
+		type = exynos_plane_get_type(zpos, CURSOR_WIN);
 		ret = exynos_plane_init(drm_dev, &ctx->planes[zpos],
 					1 << ctx->pipe, type, formats,
 					ARRAY_SIZE(formats), zpos);
diff --git a/drivers/gpu/drm/exynos/exynos_mixer.c b/drivers/gpu/drm/exynos/exynos_mixer.c
index f83e27d..3f9f072 100644
--- a/drivers/gpu/drm/exynos/exynos_mixer.c
+++ b/drivers/gpu/drm/exynos/exynos_mixer.c
@@ -43,6 +43,7 @@
 
 #define MIXER_WIN_NR		3
 #define VP_DEFAULT_WIN		2
+#define CURSOR_WIN		1
 
 /* The pixelformats that are natively supported by the mixer. */
 #define MXR_FORMAT_RGB565	4
@@ -1196,8 +1197,6 @@ static int mixer_bind(struct device *dev, struct device *manager, void *data)
 		const uint32_t *formats;
 		unsigned int fcount;
 
-		type = (zpos == DEFAULT_WIN) ? DRM_PLANE_TYPE_PRIMARY :
-						DRM_PLANE_TYPE_OVERLAY;
 		if (zpos < VP_DEFAULT_WIN) {
 			formats = mixer_formats;
 			fcount = ARRAY_SIZE(mixer_formats);
@@ -1206,6 +1205,7 @@ static int mixer_bind(struct device *dev, struct device *manager, void *data)
 			fcount = ARRAY_SIZE(vp_formats);
 		}
 
+		type = exynos_plane_get_type(zpos, CURSOR_WIN);
 		ret = exynos_plane_init(drm_dev, &ctx->planes[zpos],
 					1 << ctx->pipe, type, formats, fcount,
 					zpos);
-- 
cgit v0.10.2


From 4d91a855989af53b81f1ad7a82250d973f9b7b01 Mon Sep 17 00:00:00 2001
From: Joonyoung Shim <jy0922.shim@samsung.com>
Date: Fri, 2 Oct 2015 09:30:38 +0900
Subject: drm/exynos: fix to detach device of iommu

The arm_iommu_detach_device() is a function to detach device of iommu
attached by arm_iommu_attach_device(). The exynos-drm uses
arm_iommu_attach_device() so it should use arm_iommu_detach_device() to
detach device of iommu, not iommu_detach_device().

The drm_release_iommu_mapping() is a function to release mapping of
iommu created by arm_iommu_create_mapping(). It is called by
exynos_drm_unload() so shouldn't be called by drm_iommu_detach_device().

Signed-off-by: Joonyoung Shim <jy0922.shim@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/exynos_drm_iommu.c b/drivers/gpu/drm/exynos/exynos_drm_iommu.c
index 055e8ec..d73b9ad 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_iommu.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_iommu.c
@@ -139,6 +139,5 @@ void drm_iommu_detach_device(struct drm_device *drm_dev,
 	if (!mapping || !mapping->domain)
 		return;
 
-	iommu_detach_device(mapping->domain, subdrv_dev);
-	drm_release_iommu_mapping(drm_dev);
+	arm_iommu_detach_device(subdrv_dev);
 }
-- 
cgit v0.10.2


From 813fd67b57ffca0dd19fe1834f894a4558cf5e0b Mon Sep 17 00:00:00 2001
From: Joonyoung Shim <jy0922.shim@samsung.com>
Date: Fri, 2 Oct 2015 09:33:47 +0900
Subject: drm/exynos: cleanup name of gem object for exynos_drm

Struct of gem object in exynos_drm driver is struct exynos_drm_gem_obj.
It's too long and we can know its meaning of name without _obj postfix.

We use several names to variable name of gem object for exynos_drm -
exynos_gem_obj, gem_obj and obj. Especially "obj" name can cause
misunderstanding with variable name "obj" of struct drm_gem_object.

This will clean about name of gem object for exynos_drm as follows.
s/struct exynos_drm_gem_obj/struct exynos_drm_gem
s/exynos_gem_obj or gem_obj or obj/exynos_gem

Signed-off-by: Joonyoung Shim <jy0922.shim@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/exynos_drm_fb.c b/drivers/gpu/drm/exynos/exynos_drm_fb.c
index 0842808..fcea28b 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fb.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fb.c
@@ -32,15 +32,15 @@
  * exynos specific framebuffer structure.
  *
  * @fb: drm framebuffer obejct.
- * @exynos_gem_obj: array of exynos specific gem object containing a gem object.
+ * @exynos_gem: array of exynos specific gem object containing a gem object.
  */
 struct exynos_drm_fb {
-	struct drm_framebuffer		fb;
-	struct exynos_drm_gem_obj	*exynos_gem_obj[MAX_FB_BUFFER];
+	struct drm_framebuffer	fb;
+	struct exynos_drm_gem	*exynos_gem[MAX_FB_BUFFER];
 };
 
 static int check_fb_gem_memory_type(struct drm_device *drm_dev,
-				struct exynos_drm_gem_obj *exynos_gem_obj)
+				    struct exynos_drm_gem *exynos_gem)
 {
 	unsigned int flags;
 
@@ -51,7 +51,7 @@ static int check_fb_gem_memory_type(struct drm_device *drm_dev,
 	if (is_drm_iommu_supported(drm_dev))
 		return 0;
 
-	flags = exynos_gem_obj->flags;
+	flags = exynos_gem->flags;
 
 	/*
 	 * without iommu support, not support physically non-continuous memory
@@ -75,13 +75,13 @@ static void exynos_drm_fb_destroy(struct drm_framebuffer *fb)
 
 	drm_framebuffer_cleanup(fb);
 
-	for (i = 0; i < ARRAY_SIZE(exynos_fb->exynos_gem_obj); i++) {
+	for (i = 0; i < ARRAY_SIZE(exynos_fb->exynos_gem); i++) {
 		struct drm_gem_object *obj;
 
-		if (exynos_fb->exynos_gem_obj[i] == NULL)
+		if (exynos_fb->exynos_gem[i] == NULL)
 			continue;
 
-		obj = &exynos_fb->exynos_gem_obj[i]->base;
+		obj = &exynos_fb->exynos_gem[i]->base;
 		drm_gem_object_unreference_unlocked(obj);
 	}
 
@@ -96,7 +96,7 @@ static int exynos_drm_fb_create_handle(struct drm_framebuffer *fb,
 	struct exynos_drm_fb *exynos_fb = to_exynos_fb(fb);
 
 	return drm_gem_handle_create(file_priv,
-			&exynos_fb->exynos_gem_obj[0]->base, handle);
+				     &exynos_fb->exynos_gem[0]->base, handle);
 }
 
 static int exynos_drm_fb_dirty(struct drm_framebuffer *fb,
@@ -118,7 +118,7 @@ static struct drm_framebuffer_funcs exynos_drm_fb_funcs = {
 struct drm_framebuffer *
 exynos_drm_framebuffer_init(struct drm_device *dev,
 			    struct drm_mode_fb_cmd2 *mode_cmd,
-			    struct exynos_drm_gem_obj **gem_obj,
+			    struct exynos_drm_gem **exynos_gem,
 			    int count)
 {
 	struct exynos_drm_fb *exynos_fb;
@@ -130,11 +130,11 @@ exynos_drm_framebuffer_init(struct drm_device *dev,
 		return ERR_PTR(-ENOMEM);
 
 	for (i = 0; i < count; i++) {
-		ret = check_fb_gem_memory_type(dev, gem_obj[i]);
+		ret = check_fb_gem_memory_type(dev, exynos_gem[i]);
 		if (ret < 0)
 			goto err;
 
-		exynos_fb->exynos_gem_obj[i] = gem_obj[i];
+		exynos_fb->exynos_gem[i] = exynos_gem[i];
 	}
 
 	drm_helper_mode_fill_fb_struct(&exynos_fb->fb, mode_cmd);
@@ -156,7 +156,7 @@ static struct drm_framebuffer *
 exynos_user_fb_create(struct drm_device *dev, struct drm_file *file_priv,
 		      struct drm_mode_fb_cmd2 *mode_cmd)
 {
-	struct exynos_drm_gem_obj *gem_objs[MAX_FB_BUFFER];
+	struct exynos_drm_gem *exynos_gem[MAX_FB_BUFFER];
 	struct drm_gem_object *obj;
 	struct drm_framebuffer *fb;
 	int i;
@@ -171,10 +171,10 @@ exynos_user_fb_create(struct drm_device *dev, struct drm_file *file_priv,
 			goto err;
 		}
 
-		gem_objs[i] = to_exynos_gem_obj(obj);
+		exynos_gem[i] = to_exynos_gem(obj);
 	}
 
-	fb = exynos_drm_framebuffer_init(dev, mode_cmd, gem_objs, i);
+	fb = exynos_drm_framebuffer_init(dev, mode_cmd, exynos_gem, i);
 	if (IS_ERR(fb)) {
 		ret = PTR_ERR(fb);
 		goto err;
@@ -184,27 +184,26 @@ exynos_user_fb_create(struct drm_device *dev, struct drm_file *file_priv,
 
 err:
 	while (i--)
-		drm_gem_object_unreference_unlocked(&gem_objs[i]->base);
+		drm_gem_object_unreference_unlocked(&exynos_gem[i]->base);
 
 	return ERR_PTR(ret);
 }
 
-struct exynos_drm_gem_obj *exynos_drm_fb_gem_obj(struct drm_framebuffer *fb,
-						 int index)
+struct exynos_drm_gem *exynos_drm_fb_gem(struct drm_framebuffer *fb, int index)
 {
 	struct exynos_drm_fb *exynos_fb = to_exynos_fb(fb);
-	struct exynos_drm_gem_obj *obj;
+	struct exynos_drm_gem *exynos_gem;
 
 	if (index >= MAX_FB_BUFFER)
 		return NULL;
 
-	obj = exynos_fb->exynos_gem_obj[index];
-	if (!obj)
+	exynos_gem = exynos_fb->exynos_gem[index];
+	if (!exynos_gem)
 		return NULL;
 
-	DRM_DEBUG_KMS("dma_addr = 0x%lx\n", (unsigned long)obj->dma_addr);
+	DRM_DEBUG_KMS("dma_addr: 0x%lx\n", (unsigned long)exynos_gem->dma_addr);
 
-	return obj;
+	return exynos_gem;
 }
 
 static void exynos_drm_output_poll_changed(struct drm_device *dev)
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fb.h b/drivers/gpu/drm/exynos/exynos_drm_fb.h
index 85e4445..726a2d4 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fb.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_fb.h
@@ -19,12 +19,11 @@
 struct drm_framebuffer *
 exynos_drm_framebuffer_init(struct drm_device *dev,
 			    struct drm_mode_fb_cmd2 *mode_cmd,
-			    struct exynos_drm_gem_obj **gem_obj,
+			    struct exynos_drm_gem **exynos_gem,
 			    int count);
 
 /* get gem object of a drm framebuffer */
-struct exynos_drm_gem_obj *exynos_drm_fb_gem_obj(struct drm_framebuffer *fb,
-						 int index);
+struct exynos_drm_gem *exynos_drm_fb_gem(struct drm_framebuffer *fb, int index);
 
 void exynos_drm_mode_config_init(struct drm_device *dev);
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
index a221f75..f6118ba 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
@@ -30,8 +30,8 @@
 				drm_fb_helper)
 
 struct exynos_drm_fbdev {
-	struct drm_fb_helper		drm_fb_helper;
-	struct exynos_drm_gem_obj	*obj;
+	struct drm_fb_helper	drm_fb_helper;
+	struct exynos_drm_gem	*exynos_gem;
 };
 
 static int exynos_drm_fb_mmap(struct fb_info *info,
@@ -39,7 +39,7 @@ static int exynos_drm_fb_mmap(struct fb_info *info,
 {
 	struct drm_fb_helper *helper = info->par;
 	struct exynos_drm_fbdev *exynos_fbd = to_exynos_fbdev(helper);
-	struct exynos_drm_gem_obj *obj = exynos_fbd->obj;
+	struct exynos_drm_gem *exynos_gem = exynos_fbd->exynos_gem;
 	unsigned long vm_size;
 	int ret;
 
@@ -47,11 +47,12 @@ static int exynos_drm_fb_mmap(struct fb_info *info,
 
 	vm_size = vma->vm_end - vma->vm_start;
 
-	if (vm_size > obj->size)
+	if (vm_size > exynos_gem->size)
 		return -EINVAL;
 
-	ret = dma_mmap_attrs(helper->dev->dev, vma, obj->pages, obj->dma_addr,
-			     obj->size, &obj->dma_attrs);
+	ret = dma_mmap_attrs(helper->dev->dev, vma, exynos_gem->pages,
+			     exynos_gem->dma_addr, exynos_gem->size,
+			     &exynos_gem->dma_attrs);
 	if (ret < 0) {
 		DRM_ERROR("failed to mmap.\n");
 		return ret;
@@ -75,7 +76,7 @@ static struct fb_ops exynos_drm_fb_ops = {
 
 static int exynos_drm_fbdev_update(struct drm_fb_helper *helper,
 				   struct drm_fb_helper_surface_size *sizes,
-				   struct exynos_drm_gem_obj *obj)
+				   struct exynos_drm_gem *exynos_gem)
 {
 	struct fb_info *fbi;
 	struct drm_framebuffer *fb = helper->fb;
@@ -96,11 +97,11 @@ static int exynos_drm_fbdev_update(struct drm_fb_helper *helper,
 	drm_fb_helper_fill_fix(fbi, fb->pitches[0], fb->depth);
 	drm_fb_helper_fill_var(fbi, helper, sizes->fb_width, sizes->fb_height);
 
-	nr_pages = obj->size >> PAGE_SHIFT;
+	nr_pages = exynos_gem->size >> PAGE_SHIFT;
 
-	obj->kvaddr = (void __iomem *) vmap(obj->pages, nr_pages, VM_MAP,
-			pgprot_writecombine(PAGE_KERNEL));
-	if (!obj->kvaddr) {
+	exynos_gem->kvaddr = (void __iomem *) vmap(exynos_gem->pages, nr_pages,
+				VM_MAP, pgprot_writecombine(PAGE_KERNEL));
+	if (!exynos_gem->kvaddr) {
 		DRM_ERROR("failed to map pages to kernel space.\n");
 		drm_fb_helper_release_fbi(helper);
 		return -EIO;
@@ -109,7 +110,7 @@ static int exynos_drm_fbdev_update(struct drm_fb_helper *helper,
 	offset = fbi->var.xoffset * (fb->bits_per_pixel >> 3);
 	offset += fbi->var.yoffset * fb->pitches[0];
 
-	fbi->screen_base = obj->kvaddr + offset;
+	fbi->screen_base = exynos_gem->kvaddr + offset;
 	fbi->screen_size = size;
 	fbi->fix.smem_len = size;
 
@@ -120,7 +121,7 @@ static int exynos_drm_fbdev_create(struct drm_fb_helper *helper,
 				    struct drm_fb_helper_surface_size *sizes)
 {
 	struct exynos_drm_fbdev *exynos_fbdev = to_exynos_fbdev(helper);
-	struct exynos_drm_gem_obj *obj;
+	struct exynos_drm_gem *exynos_gem;
 	struct drm_device *dev = helper->dev;
 	struct drm_mode_fb_cmd2 mode_cmd = { 0 };
 	struct platform_device *pdev = dev->platformdev;
@@ -141,32 +142,34 @@ static int exynos_drm_fbdev_create(struct drm_fb_helper *helper,
 
 	size = mode_cmd.pitches[0] * mode_cmd.height;
 
-	obj = exynos_drm_gem_create(dev, EXYNOS_BO_CONTIG, size);
+	exynos_gem = exynos_drm_gem_create(dev, EXYNOS_BO_CONTIG, size);
 	/*
 	 * If physically contiguous memory allocation fails and if IOMMU is
 	 * supported then try to get buffer from non physically contiguous
 	 * memory area.
 	 */
-	if (IS_ERR(obj) && is_drm_iommu_supported(dev)) {
+	if (IS_ERR(exynos_gem) && is_drm_iommu_supported(dev)) {
 		dev_warn(&pdev->dev, "contiguous FB allocation failed, falling back to non-contiguous\n");
-		obj = exynos_drm_gem_create(dev, EXYNOS_BO_NONCONTIG, size);
+		exynos_gem = exynos_drm_gem_create(dev, EXYNOS_BO_NONCONTIG,
+						   size);
 	}
 
-	if (IS_ERR(obj)) {
-		ret = PTR_ERR(obj);
+	if (IS_ERR(exynos_gem)) {
+		ret = PTR_ERR(exynos_gem);
 		goto out;
 	}
 
-	exynos_fbdev->obj = obj;
+	exynos_fbdev->exynos_gem = exynos_gem;
 
-	helper->fb = exynos_drm_framebuffer_init(dev, &mode_cmd, &obj, 1);
+	helper->fb =
+		exynos_drm_framebuffer_init(dev, &mode_cmd, &exynos_gem, 1);
 	if (IS_ERR(helper->fb)) {
 		DRM_ERROR("failed to create drm framebuffer.\n");
 		ret = PTR_ERR(helper->fb);
 		goto err_destroy_gem;
 	}
 
-	ret = exynos_drm_fbdev_update(helper, sizes, obj);
+	ret = exynos_drm_fbdev_update(helper, sizes, exynos_gem);
 	if (ret < 0)
 		goto err_destroy_framebuffer;
 
@@ -176,7 +179,7 @@ static int exynos_drm_fbdev_create(struct drm_fb_helper *helper,
 err_destroy_framebuffer:
 	drm_framebuffer_cleanup(helper->fb);
 err_destroy_gem:
-	exynos_drm_gem_destroy(obj);
+	exynos_drm_gem_destroy(exynos_gem);
 
 /*
  * if failed, all resources allocated above would be released by
@@ -269,11 +272,11 @@ static void exynos_drm_fbdev_destroy(struct drm_device *dev,
 				      struct drm_fb_helper *fb_helper)
 {
 	struct exynos_drm_fbdev *exynos_fbd = to_exynos_fbdev(fb_helper);
-	struct exynos_drm_gem_obj *obj = exynos_fbd->obj;
+	struct exynos_drm_gem *exynos_gem = exynos_fbd->exynos_gem;
 	struct drm_framebuffer *fb;
 
-	if (obj->kvaddr)
-		vunmap(obj->kvaddr);
+	if (exynos_gem->kvaddr)
+		vunmap(exynos_gem->kvaddr);
 
 	/* release drm framebuffer and real buffer */
 	if (fb_helper->fb && fb_helper->fb->funcs) {
diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.c b/drivers/gpu/drm/exynos/exynos_drm_gem.c
index 407afed..f1dcdd0 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_gem.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_gem.c
@@ -20,97 +20,100 @@
 #include "exynos_drm_gem.h"
 #include "exynos_drm_iommu.h"
 
-static int exynos_drm_alloc_buf(struct exynos_drm_gem_obj *obj)
+static int exynos_drm_alloc_buf(struct exynos_drm_gem *exynos_gem)
 {
-	struct drm_device *dev = obj->base.dev;
+	struct drm_device *dev = exynos_gem->base.dev;
 	enum dma_attr attr;
 	unsigned int nr_pages;
 
-	if (obj->dma_addr) {
+	if (exynos_gem->dma_addr) {
 		DRM_DEBUG_KMS("already allocated.\n");
 		return 0;
 	}
 
-	init_dma_attrs(&obj->dma_attrs);
+	init_dma_attrs(&exynos_gem->dma_attrs);
 
 	/*
 	 * if EXYNOS_BO_CONTIG, fully physically contiguous memory
 	 * region will be allocated else physically contiguous
 	 * as possible.
 	 */
-	if (!(obj->flags & EXYNOS_BO_NONCONTIG))
-		dma_set_attr(DMA_ATTR_FORCE_CONTIGUOUS, &obj->dma_attrs);
+	if (!(exynos_gem->flags & EXYNOS_BO_NONCONTIG))
+		dma_set_attr(DMA_ATTR_FORCE_CONTIGUOUS, &exynos_gem->dma_attrs);
 
 	/*
 	 * if EXYNOS_BO_WC or EXYNOS_BO_NONCACHABLE, writecombine mapping
 	 * else cachable mapping.
 	 */
-	if (obj->flags & EXYNOS_BO_WC || !(obj->flags & EXYNOS_BO_CACHABLE))
+	if (exynos_gem->flags & EXYNOS_BO_WC ||
+			!(exynos_gem->flags & EXYNOS_BO_CACHABLE))
 		attr = DMA_ATTR_WRITE_COMBINE;
 	else
 		attr = DMA_ATTR_NON_CONSISTENT;
 
-	dma_set_attr(attr, &obj->dma_attrs);
-	dma_set_attr(DMA_ATTR_NO_KERNEL_MAPPING, &obj->dma_attrs);
+	dma_set_attr(attr, &exynos_gem->dma_attrs);
+	dma_set_attr(DMA_ATTR_NO_KERNEL_MAPPING, &exynos_gem->dma_attrs);
 
-	nr_pages = obj->size >> PAGE_SHIFT;
+	nr_pages = exynos_gem->size >> PAGE_SHIFT;
 
 	if (!is_drm_iommu_supported(dev)) {
-		obj->pages = drm_calloc_large(nr_pages, sizeof(struct page *));
-		if (!obj->pages) {
+		exynos_gem->pages = drm_calloc_large(nr_pages,
+						     sizeof(struct page *));
+		if (!exynos_gem->pages) {
 			DRM_ERROR("failed to allocate pages.\n");
 			return -ENOMEM;
 		}
 	}
 
-	obj->cookie = dma_alloc_attrs(dev->dev, obj->size, &obj->dma_addr,
-				      GFP_KERNEL, &obj->dma_attrs);
-	if (!obj->cookie) {
+	exynos_gem->cookie = dma_alloc_attrs(dev->dev, exynos_gem->size,
+					     &exynos_gem->dma_addr, GFP_KERNEL,
+					     &exynos_gem->dma_attrs);
+	if (!exynos_gem->cookie) {
 		DRM_ERROR("failed to allocate buffer.\n");
-		if (obj->pages)
-			drm_free_large(obj->pages);
+		if (exynos_gem->pages)
+			drm_free_large(exynos_gem->pages);
 		return -ENOMEM;
 	}
 
-	if (obj->pages) {
+	if (exynos_gem->pages) {
 		dma_addr_t start_addr;
 		unsigned int i = 0;
 
-		start_addr = obj->dma_addr;
+		start_addr = exynos_gem->dma_addr;
 		while (i < nr_pages) {
-			obj->pages[i] = pfn_to_page(dma_to_pfn(dev->dev,
-							       start_addr));
+			exynos_gem->pages[i] =
+				pfn_to_page(dma_to_pfn(dev->dev, start_addr));
 			start_addr += PAGE_SIZE;
 			i++;
 		}
 	} else {
-		obj->pages = obj->cookie;
+		exynos_gem->pages = exynos_gem->cookie;
 	}
 
 	DRM_DEBUG_KMS("dma_addr(0x%lx), size(0x%lx)\n",
-			(unsigned long)obj->dma_addr,
-			obj->size);
+			(unsigned long)exynos_gem->dma_addr, exynos_gem->size);
 
 	return 0;
 }
 
-static void exynos_drm_free_buf(struct exynos_drm_gem_obj *obj)
+static void exynos_drm_free_buf(struct exynos_drm_gem *exynos_gem)
 {
-	struct drm_device *dev = obj->base.dev;
+	struct drm_device *dev = exynos_gem->base.dev;
 
-	if (!obj->dma_addr) {
+	if (!exynos_gem->dma_addr) {
 		DRM_DEBUG_KMS("dma_addr is invalid.\n");
 		return;
 	}
 
 	DRM_DEBUG_KMS("dma_addr(0x%lx), size(0x%lx)\n",
-			(unsigned long)obj->dma_addr, obj->size);
+			(unsigned long)exynos_gem->dma_addr, exynos_gem->size);
 
-	dma_free_attrs(dev->dev, obj->size, obj->cookie,
-			(dma_addr_t)obj->dma_addr, &obj->dma_attrs);
+	dma_free_attrs(dev->dev, exynos_gem->size, exynos_gem->cookie,
+			(dma_addr_t)exynos_gem->dma_addr,
+			&exynos_gem->dma_attrs);
 
 	if (!is_drm_iommu_supported(dev))
-		drm_free_large(obj->pages);
+		drm_free_large(exynos_gem->pages);
 }
 
 static int exynos_drm_gem_handle_create(struct drm_gem_object *obj,
@@ -135,9 +138,9 @@ static int exynos_drm_gem_handle_create(struct drm_gem_object *obj,
 	return 0;
 }
 
-void exynos_drm_gem_destroy(struct exynos_drm_gem_obj *exynos_gem_obj)
+void exynos_drm_gem_destroy(struct exynos_drm_gem *exynos_gem)
 {
-	struct drm_gem_object *obj = &exynos_gem_obj->base;
+	struct drm_gem_object *obj = &exynos_gem->base;
 
 	DRM_DEBUG_KMS("handle count = %d\n", obj->handle_count);
 
@@ -148,21 +151,21 @@ void exynos_drm_gem_destroy(struct exynos_drm_gem_obj *exynos_gem_obj)
 	 * once dmabuf's refcount becomes 0.
 	 */
 	if (obj->import_attach)
-		drm_prime_gem_destroy(obj, exynos_gem_obj->sgt);
+		drm_prime_gem_destroy(obj, exynos_gem->sgt);
 	else
-		exynos_drm_free_buf(exynos_gem_obj);
+		exynos_drm_free_buf(exynos_gem);
 
 	/* release file pointer to gem object. */
 	drm_gem_object_release(obj);
 
-	kfree(exynos_gem_obj);
+	kfree(exynos_gem);
 }
 
 unsigned long exynos_drm_gem_get_size(struct drm_device *dev,
 						unsigned int gem_handle,
 						struct drm_file *file_priv)
 {
-	struct exynos_drm_gem_obj *exynos_gem_obj;
+	struct exynos_drm_gem *exynos_gem;
 	struct drm_gem_object *obj;
 
 	obj = drm_gem_object_lookup(dev, file_priv, gem_handle);
@@ -171,51 +174,51 @@ unsigned long exynos_drm_gem_get_size(struct drm_device *dev,
 		return 0;
 	}
 
-	exynos_gem_obj = to_exynos_gem_obj(obj);
+	exynos_gem = to_exynos_gem(obj);
 
 	drm_gem_object_unreference_unlocked(obj);
 
-	return exynos_gem_obj->size;
+	return exynos_gem->size;
 }
 
-static struct exynos_drm_gem_obj *exynos_drm_gem_init(struct drm_device *dev,
-						      unsigned long size)
+static struct exynos_drm_gem *exynos_drm_gem_init(struct drm_device *dev,
+						  unsigned long size)
 {
-	struct exynos_drm_gem_obj *exynos_gem_obj;
+	struct exynos_drm_gem *exynos_gem;
 	struct drm_gem_object *obj;
 	int ret;
 
-	exynos_gem_obj = kzalloc(sizeof(*exynos_gem_obj), GFP_KERNEL);
-	if (!exynos_gem_obj)
+	exynos_gem = kzalloc(sizeof(*exynos_gem), GFP_KERNEL);
+	if (!exynos_gem)
 		return ERR_PTR(-ENOMEM);
 
-	exynos_gem_obj->size = size;
-	obj = &exynos_gem_obj->base;
+	exynos_gem->size = size;
+	obj = &exynos_gem->base;
 
 	ret = drm_gem_object_init(dev, obj, size);
 	if (ret < 0) {
 		DRM_ERROR("failed to initialize gem object\n");
-		kfree(exynos_gem_obj);
+		kfree(exynos_gem);
 		return ERR_PTR(ret);
 	}
 
 	ret = drm_gem_create_mmap_offset(obj);
 	if (ret < 0) {
 		drm_gem_object_release(obj);
-		kfree(exynos_gem_obj);
+		kfree(exynos_gem);
 		return ERR_PTR(ret);
 	}
 
 	DRM_DEBUG_KMS("created file object = 0x%x\n", (unsigned int)obj->filp);
 
-	return exynos_gem_obj;
+	return exynos_gem;
 }
 
-struct exynos_drm_gem_obj *exynos_drm_gem_create(struct drm_device *dev,
-						unsigned int flags,
-						unsigned long size)
+struct exynos_drm_gem *exynos_drm_gem_create(struct drm_device *dev,
+					     unsigned int flags,
+					     unsigned long size)
 {
-	struct exynos_drm_gem_obj *exynos_gem_obj;
+	struct exynos_drm_gem *exynos_gem;
 	int ret;
 
 	if (flags & ~(EXYNOS_BO_MASK)) {
@@ -230,38 +233,38 @@ struct exynos_drm_gem_obj *exynos_drm_gem_create(struct drm_device *dev,
 
 	size = roundup(size, PAGE_SIZE);
 
-	exynos_gem_obj = exynos_drm_gem_init(dev, size);
-	if (IS_ERR(exynos_gem_obj))
-		return exynos_gem_obj;
+	exynos_gem = exynos_drm_gem_init(dev, size);
+	if (IS_ERR(exynos_gem))
+		return exynos_gem;
 
 	/* set memory type and cache attribute from user side. */
-	exynos_gem_obj->flags = flags;
+	exynos_gem->flags = flags;
 
-	ret = exynos_drm_alloc_buf(exynos_gem_obj);
+	ret = exynos_drm_alloc_buf(exynos_gem);
 	if (ret < 0) {
-		drm_gem_object_release(&exynos_gem_obj->base);
-		kfree(exynos_gem_obj);
+		drm_gem_object_release(&exynos_gem->base);
+		kfree(exynos_gem);
 		return ERR_PTR(ret);
 	}
 
-	return exynos_gem_obj;
+	return exynos_gem;
 }
 
 int exynos_drm_gem_create_ioctl(struct drm_device *dev, void *data,
 				struct drm_file *file_priv)
 {
 	struct drm_exynos_gem_create *args = data;
-	struct exynos_drm_gem_obj *exynos_gem_obj;
+	struct exynos_drm_gem *exynos_gem;
 	int ret;
 
-	exynos_gem_obj = exynos_drm_gem_create(dev, args->flags, args->size);
-	if (IS_ERR(exynos_gem_obj))
-		return PTR_ERR(exynos_gem_obj);
+	exynos_gem = exynos_drm_gem_create(dev, args->flags, args->size);
+	if (IS_ERR(exynos_gem))
+		return PTR_ERR(exynos_gem);
 
-	ret = exynos_drm_gem_handle_create(&exynos_gem_obj->base, file_priv,
-			&args->handle);
+	ret = exynos_drm_gem_handle_create(&exynos_gem->base, file_priv,
+					   &args->handle);
 	if (ret) {
-		exynos_drm_gem_destroy(exynos_gem_obj);
+		exynos_drm_gem_destroy(exynos_gem);
 		return ret;
 	}
 
@@ -272,7 +275,7 @@ dma_addr_t *exynos_drm_gem_get_dma_addr(struct drm_device *dev,
 					unsigned int gem_handle,
 					struct drm_file *filp)
 {
-	struct exynos_drm_gem_obj *exynos_gem_obj;
+	struct exynos_drm_gem *exynos_gem;
 	struct drm_gem_object *obj;
 
 	obj = drm_gem_object_lookup(dev, filp, gem_handle);
@@ -281,9 +284,9 @@ dma_addr_t *exynos_drm_gem_get_dma_addr(struct drm_device *dev,
 		return ERR_PTR(-EINVAL);
 	}
 
-	exynos_gem_obj = to_exynos_gem_obj(obj);
+	exynos_gem = to_exynos_gem(obj);
 
-	return &exynos_gem_obj->dma_addr;
+	return &exynos_gem->dma_addr;
 }
 
 void exynos_drm_gem_put_dma_addr(struct drm_device *dev,
@@ -307,10 +310,10 @@ void exynos_drm_gem_put_dma_addr(struct drm_device *dev,
 	drm_gem_object_unreference_unlocked(obj);
 }
 
-static int exynos_drm_gem_mmap_buffer(struct exynos_drm_gem_obj *exynos_gem_obj,
+static int exynos_drm_gem_mmap_buffer(struct exynos_drm_gem *exynos_gem,
 				      struct vm_area_struct *vma)
 {
-	struct drm_device *drm_dev = exynos_gem_obj->base.dev;
+	struct drm_device *drm_dev = exynos_gem->base.dev;
 	unsigned long vm_size;
 	int ret;
 
@@ -320,12 +323,12 @@ static int exynos_drm_gem_mmap_buffer(struct exynos_drm_gem_obj *exynos_gem_obj,
 	vm_size = vma->vm_end - vma->vm_start;
 
 	/* check if user-requested size is valid. */
-	if (vm_size > exynos_gem_obj->size)
+	if (vm_size > exynos_gem->size)
 		return -EINVAL;
 
-	ret = dma_mmap_attrs(drm_dev->dev, vma, exynos_gem_obj->pages,
-				exynos_gem_obj->dma_addr, exynos_gem_obj->size,
-				&exynos_gem_obj->dma_attrs);
+	ret = dma_mmap_attrs(drm_dev->dev, vma, exynos_gem->pages,
+			     exynos_gem->dma_addr, exynos_gem->size,
+			     &exynos_gem->dma_attrs);
 	if (ret < 0) {
 		DRM_ERROR("failed to mmap.\n");
 		return ret;
@@ -337,7 +340,7 @@ static int exynos_drm_gem_mmap_buffer(struct exynos_drm_gem_obj *exynos_gem_obj,
 int exynos_drm_gem_get_ioctl(struct drm_device *dev, void *data,
 				      struct drm_file *file_priv)
 {
-	struct exynos_drm_gem_obj *exynos_gem_obj;
+	struct exynos_drm_gem *exynos_gem;
 	struct drm_exynos_gem_info *args = data;
 	struct drm_gem_object *obj;
 
@@ -350,10 +353,10 @@ int exynos_drm_gem_get_ioctl(struct drm_device *dev, void *data,
 		return -EINVAL;
 	}
 
-	exynos_gem_obj = to_exynos_gem_obj(obj);
+	exynos_gem = to_exynos_gem(obj);
 
-	args->flags = exynos_gem_obj->flags;
-	args->size = exynos_gem_obj->size;
+	args->flags = exynos_gem->flags;
+	args->size = exynos_gem->size;
 
 	drm_gem_object_unreference(obj);
 	mutex_unlock(&dev->struct_mutex);
@@ -389,14 +392,14 @@ void exynos_gem_unmap_sgt_from_dma(struct drm_device *drm_dev,
 
 void exynos_drm_gem_free_object(struct drm_gem_object *obj)
 {
-	exynos_drm_gem_destroy(to_exynos_gem_obj(obj));
+	exynos_drm_gem_destroy(to_exynos_gem(obj));
 }
 
 int exynos_drm_gem_dumb_create(struct drm_file *file_priv,
 			       struct drm_device *dev,
 			       struct drm_mode_create_dumb *args)
 {
-	struct exynos_drm_gem_obj *exynos_gem_obj;
+	struct exynos_drm_gem *exynos_gem;
 	unsigned int flags;
 	int ret;
 
@@ -414,16 +417,16 @@ int exynos_drm_gem_dumb_create(struct drm_file *file_priv,
 	else
 		flags = EXYNOS_BO_CONTIG | EXYNOS_BO_WC;
 
-	exynos_gem_obj = exynos_drm_gem_create(dev, flags, args->size);
-	if (IS_ERR(exynos_gem_obj)) {
+	exynos_gem = exynos_drm_gem_create(dev, flags, args->size);
+	if (IS_ERR(exynos_gem)) {
 		dev_warn(dev->dev, "FB allocation failed.\n");
-		return PTR_ERR(exynos_gem_obj);
+		return PTR_ERR(exynos_gem);
 	}
 
-	ret = exynos_drm_gem_handle_create(&exynos_gem_obj->base, file_priv,
-			&args->handle);
+	ret = exynos_drm_gem_handle_create(&exynos_gem->base, file_priv,
+					   &args->handle);
 	if (ret) {
-		exynos_drm_gem_destroy(exynos_gem_obj);
+		exynos_drm_gem_destroy(exynos_gem);
 		return ret;
 	}
 
@@ -464,7 +467,7 @@ unlock:
 int exynos_drm_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 {
 	struct drm_gem_object *obj = vma->vm_private_data;
-	struct exynos_drm_gem_obj *exynos_gem_obj = to_exynos_gem_obj(obj);
+	struct exynos_drm_gem *exynos_gem = to_exynos_gem(obj);
 	unsigned long pfn;
 	pgoff_t page_offset;
 	int ret;
@@ -472,13 +475,13 @@ int exynos_drm_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 	page_offset = ((unsigned long)vmf->virtual_address -
 			vma->vm_start) >> PAGE_SHIFT;
 
-	if (page_offset >= (exynos_gem_obj->size >> PAGE_SHIFT)) {
+	if (page_offset >= (exynos_gem->size >> PAGE_SHIFT)) {
 		DRM_ERROR("invalid page offset\n");
 		ret = -EINVAL;
 		goto out;
 	}
 
-	pfn = page_to_pfn(exynos_gem_obj->pages[page_offset]);
+	pfn = page_to_pfn(exynos_gem->pages[page_offset]);
 	ret = vm_insert_mixed(vma, (unsigned long)vmf->virtual_address, pfn);
 
 out:
@@ -496,7 +499,7 @@ out:
 
 int exynos_drm_gem_mmap(struct file *filp, struct vm_area_struct *vma)
 {
-	struct exynos_drm_gem_obj *exynos_gem_obj;
+	struct exynos_drm_gem *exynos_gem;
 	struct drm_gem_object *obj;
 	int ret;
 
@@ -508,21 +511,21 @@ int exynos_drm_gem_mmap(struct file *filp, struct vm_area_struct *vma)
 	}
 
 	obj = vma->vm_private_data;
-	exynos_gem_obj = to_exynos_gem_obj(obj);
+	exynos_gem = to_exynos_gem(obj);
 
-	DRM_DEBUG_KMS("flags = 0x%x\n", exynos_gem_obj->flags);
+	DRM_DEBUG_KMS("flags = 0x%x\n", exynos_gem->flags);
 
 	/* non-cachable as default. */
-	if (exynos_gem_obj->flags & EXYNOS_BO_CACHABLE)
+	if (exynos_gem->flags & EXYNOS_BO_CACHABLE)
 		vma->vm_page_prot = vm_get_page_prot(vma->vm_flags);
-	else if (exynos_gem_obj->flags & EXYNOS_BO_WC)
+	else if (exynos_gem->flags & EXYNOS_BO_WC)
 		vma->vm_page_prot =
 			pgprot_writecombine(vm_get_page_prot(vma->vm_flags));
 	else
 		vma->vm_page_prot =
 			pgprot_noncached(vm_get_page_prot(vma->vm_flags));
 
-	ret = exynos_drm_gem_mmap_buffer(exynos_gem_obj, vma);
+	ret = exynos_drm_gem_mmap_buffer(exynos_gem, vma);
 	if (ret)
 		goto err_close_vm;
 
@@ -537,12 +540,12 @@ err_close_vm:
 /* low-level interface prime helpers */
 struct sg_table *exynos_drm_gem_prime_get_sg_table(struct drm_gem_object *obj)
 {
-	struct exynos_drm_gem_obj *exynos_gem_obj = to_exynos_gem_obj(obj);
+	struct exynos_drm_gem *exynos_gem = to_exynos_gem(obj);
 	int npages;
 
-	npages = exynos_gem_obj->size >> PAGE_SHIFT;
+	npages = exynos_gem->size >> PAGE_SHIFT;
 
-	return drm_prime_pages_to_sg(exynos_gem_obj->pages, npages);
+	return drm_prime_pages_to_sg(exynos_gem->pages, npages);
 }
 
 struct drm_gem_object *
@@ -550,35 +553,35 @@ exynos_drm_gem_prime_import_sg_table(struct drm_device *dev,
 				     struct dma_buf_attachment *attach,
 				     struct sg_table *sgt)
 {
-	struct exynos_drm_gem_obj *exynos_gem_obj;
+	struct exynos_drm_gem *exynos_gem;
 	int npages;
 	int ret;
 
-	exynos_gem_obj = exynos_drm_gem_init(dev, attach->dmabuf->size);
-	if (IS_ERR(exynos_gem_obj)) {
-		ret = PTR_ERR(exynos_gem_obj);
+	exynos_gem = exynos_drm_gem_init(dev, attach->dmabuf->size);
+	if (IS_ERR(exynos_gem)) {
+		ret = PTR_ERR(exynos_gem);
 		return ERR_PTR(ret);
 	}
 
-	exynos_gem_obj->dma_addr = sg_dma_address(sgt->sgl);
+	exynos_gem->dma_addr = sg_dma_address(sgt->sgl);
 
-	npages = exynos_gem_obj->size >> PAGE_SHIFT;
-	exynos_gem_obj->pages = drm_malloc_ab(npages, sizeof(struct page *));
-	if (!exynos_gem_obj->pages) {
+	npages = exynos_gem->size >> PAGE_SHIFT;
+	exynos_gem->pages = drm_malloc_ab(npages, sizeof(struct page *));
+	if (!exynos_gem->pages) {
 		ret = -ENOMEM;
 		goto err;
 	}
 
-	ret = drm_prime_sg_to_page_addr_arrays(sgt, exynos_gem_obj->pages, NULL,
-			npages);
+	ret = drm_prime_sg_to_page_addr_arrays(sgt, exynos_gem->pages, NULL,
+					       npages);
 	if (ret < 0)
 		goto err_free_large;
 
-	exynos_gem_obj->sgt = sgt;
+	exynos_gem->sgt = sgt;
 
 	if (sgt->nents == 1) {
 		/* always physically continuous memory if sgt->nents is 1. */
-		exynos_gem_obj->flags |= EXYNOS_BO_CONTIG;
+		exynos_gem->flags |= EXYNOS_BO_CONTIG;
 	} else {
 		/*
 		 * this case could be CONTIG or NONCONTIG type but for now
@@ -586,16 +589,16 @@ exynos_drm_gem_prime_import_sg_table(struct drm_device *dev,
 		 * TODO. we have to find a way that exporter can notify
 		 * the type of its own buffer to importer.
 		 */
-		exynos_gem_obj->flags |= EXYNOS_BO_NONCONTIG;
+		exynos_gem->flags |= EXYNOS_BO_NONCONTIG;
 	}
 
-	return &exynos_gem_obj->base;
+	return &exynos_gem->base;
 
 err_free_large:
-	drm_free_large(exynos_gem_obj->pages);
+	drm_free_large(exynos_gem->pages);
 err:
-	drm_gem_object_release(&exynos_gem_obj->base);
-	kfree(exynos_gem_obj);
+	drm_gem_object_release(&exynos_gem->base);
+	kfree(exynos_gem);
 	return ERR_PTR(ret);
 }
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.h b/drivers/gpu/drm/exynos/exynos_drm_gem.h
index b62d100..37ab8b2 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_gem.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_gem.h
@@ -14,8 +14,7 @@
 
 #include <drm/drm_gem.h>
 
-#define to_exynos_gem_obj(x)	container_of(x,\
-			struct exynos_drm_gem_obj, base)
+#define to_exynos_gem(x)	container_of(x, struct exynos_drm_gem, base)
 
 #define IS_NONCONTIG_BUFFER(f)		(f & EXYNOS_BO_NONCONTIG)
 
@@ -44,7 +43,7 @@
  * P.S. this object would be transferred to user as kms_bo.handle so
  *	user can access the buffer through kms_bo.handle.
  */
-struct exynos_drm_gem_obj {
+struct exynos_drm_gem {
 	struct drm_gem_object	base;
 	unsigned int		flags;
 	unsigned long		size;
@@ -59,12 +58,12 @@ struct exynos_drm_gem_obj {
 struct page **exynos_gem_get_pages(struct drm_gem_object *obj, gfp_t gfpmask);
 
 /* destroy a buffer with gem object */
-void exynos_drm_gem_destroy(struct exynos_drm_gem_obj *exynos_gem_obj);
+void exynos_drm_gem_destroy(struct exynos_drm_gem *exynos_gem);
 
 /* create a new buffer with gem object */
-struct exynos_drm_gem_obj *exynos_drm_gem_create(struct drm_device *dev,
-						unsigned int flags,
-						unsigned long size);
+struct exynos_drm_gem *exynos_drm_gem_create(struct drm_device *dev,
+					     unsigned int flags,
+					     unsigned long size);
 
 /*
  * request gem object creation and buffer allocation as the size
@@ -106,7 +105,7 @@ unsigned long exynos_drm_gem_get_size(struct drm_device *dev,
 						struct drm_file *file_priv);
 
 /* free gem object. */
-void exynos_drm_gem_free_object(struct drm_gem_object *gem_obj);
+void exynos_drm_gem_free_object(struct drm_gem_object *obj);
 
 /* create memory region for drm framebuffer. */
 int exynos_drm_gem_dumb_create(struct drm_file *file_priv,
diff --git a/drivers/gpu/drm/exynos/exynos_drm_plane.c b/drivers/gpu/drm/exynos/exynos_drm_plane.c
index 80b2151..1793117 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_plane.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_plane.c
@@ -128,15 +128,14 @@ static int exynos_plane_atomic_check(struct drm_plane *plane,
 
 	nr = drm_format_num_planes(state->fb->pixel_format);
 	for (i = 0; i < nr; i++) {
-		struct exynos_drm_gem_obj *obj =
-					exynos_drm_fb_gem_obj(state->fb, i);
-
-		if (!obj) {
+		struct exynos_drm_gem *exynos_gem =
+					exynos_drm_fb_gem(state->fb, i);
+		if (!exynos_gem) {
 			DRM_DEBUG_KMS("gem object is null\n");
 			return -EFAULT;
 		}
 
-		exynos_plane->dma_addr[i] = obj->dma_addr +
+		exynos_plane->dma_addr[i] = exynos_gem->dma_addr +
 					    state->fb->offsets[i];
 
 		DRM_DEBUG_KMS("buffer: %d, dma_addr = 0x%lx\n",
-- 
cgit v0.10.2


From cbd3e2493ca359131f412133bc57407a974285e6 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Fri, 23 Oct 2015 12:56:51 -0400
Subject: drm/radeon/dce4+: optimize pageflip (v2)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Taking the grph update lock is only necessary when
updating the the secondary address (for single pipe stereo).

v2: fix comment as per Michel's suggestion

Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c
index 0acde19..7f33767 100644
--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -1404,44 +1404,20 @@ void dce4_wait_for_vblank(struct radeon_device *rdev, int crtc)
  * @crtc_id: crtc to cleanup pageflip on
  * @crtc_base: new address of the crtc (GPU MC address)
  *
- * Does the actual pageflip (evergreen+).
- * During vblank we take the crtc lock and wait for the update_pending
- * bit to go high, when it does, we release the lock, and allow the
- * double buffered update to take place.
- * Returns the current update pending status.
+ * Triggers the actual pageflip by updating the primary
+ * surface base address (evergreen+).
  */
 void evergreen_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 {
 	struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
-	u32 tmp = RREG32(EVERGREEN_GRPH_UPDATE + radeon_crtc->crtc_offset);
-	int i;
-
-	/* Lock the graphics update lock */
-	tmp |= EVERGREEN_GRPH_UPDATE_LOCK;
-	WREG32(EVERGREEN_GRPH_UPDATE + radeon_crtc->crtc_offset, tmp);
 
 	/* update the scanout addresses */
-	WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS_HIGH + radeon_crtc->crtc_offset,
-	       upper_32_bits(crtc_base));
-	WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS + radeon_crtc->crtc_offset,
-	       (u32)crtc_base);
-
 	WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS_HIGH + radeon_crtc->crtc_offset,
 	       upper_32_bits(crtc_base));
 	WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS + radeon_crtc->crtc_offset,
 	       (u32)crtc_base);
-
-	/* Wait for update_pending to go high. */
-	for (i = 0; i < rdev->usec_timeout; i++) {
-		if (RREG32(EVERGREEN_GRPH_UPDATE + radeon_crtc->crtc_offset) & EVERGREEN_GRPH_SURFACE_UPDATE_PENDING)
-			break;
-		udelay(1);
-	}
-	DRM_DEBUG("Update pending now high. Unlocking vupdate_lock.\n");
-
-	/* Unlock the lock, so double-buffering can take place inside vblank */
-	tmp &= ~EVERGREEN_GRPH_UPDATE_LOCK;
-	WREG32(EVERGREEN_GRPH_UPDATE + radeon_crtc->crtc_offset, tmp);
+	/* post the write */
+	RREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS + radeon_crtc->crtc_offset);
 }
 
 /**
-- 
cgit v0.10.2


From 5a37ed0a9eda5fd77a4cc92c6e46b6f6e8db873d Mon Sep 17 00:00:00 2001
From: Thomas Daniel <thomas.daniel@intel.com>
Date: Fri, 23 Oct 2015 14:55:38 +0100
Subject: drm/i915: Fix maxfifo watermark calc on vlv cursor planes
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

A typo resulted in the watermarks for cursor planes not being calculated
correctly.  Fixed the typo.

Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Thomas Daniel <thomas.daniel@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>

diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index 9dda3ea..80175a3 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -1135,7 +1135,7 @@ static void vlv_compute_wm(struct intel_crtc *crtc)
 		case DRM_PLANE_TYPE_CURSOR:
 			for (level = 0; level < wm_state->num_levels; level++)
 				wm_state->sr[level].cursor =
-					wm_state->sr[level].cursor;
+					wm_state->wm[level].cursor;
 			break;
 		case DRM_PLANE_TYPE_PRIMARY:
 			for (level = 0; level < wm_state->num_levels; level++)
-- 
cgit v0.10.2


From b8b339ea3b76392b1be7445f5ce57958fa6539f3 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Wed, 28 Oct 2015 00:51:48 -0400
Subject: drm/amdgpu: add some additional CZ revisions

Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index cbc46a3..f092400 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -1031,6 +1031,8 @@ static void gfx_v8_0_gpu_early_init(struct amdgpu_device *adev)
 		case 0x84:
 		case 0xc8:
 		case 0xcc:
+		case 0xe1:
+		case 0xe3:
 			/* B10 */
 			adev->gfx.config.max_cu_per_sh = 8;
 			break;
@@ -1039,18 +1041,23 @@ static void gfx_v8_0_gpu_early_init(struct amdgpu_device *adev)
 		case 0x85:
 		case 0xc9:
 		case 0xcd:
+		case 0xe2:
+		case 0xe4:
 			/* B8 */
 			adev->gfx.config.max_cu_per_sh = 6;
 			break;
 		case 0xc6:
 		case 0xca:
 		case 0xce:
+		case 0x88:
 			/* B6 */
 			adev->gfx.config.max_cu_per_sh = 6;
 			break;
 		case 0xc7:
 		case 0x87:
 		case 0xcb:
+		case 0xe5:
+		case 0x89:
 		default:
 			/* B4 */
 			adev->gfx.config.max_cu_per_sh = 4;
-- 
cgit v0.10.2


From 6675bce251b6ec20ec09c07e941d9f664be4ae58 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Wed, 28 Oct 2015 13:57:09 +0200
Subject: drm/i915: disable CPU PWM also on LPT/SPT backlight disable
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Although we don't support or enable CPU PWM with LPT/SPT based systems,
it may have been enabled prior to loading the driver. Disable the CPU
PWM on LPT/SPT backlight disable to avoid warnings on LCPLL disable.

The issue has been present on BDW since BDW enabling, but was recently
introduced on HSW with

commit 437b15b8017e0d946453c10794b0c5d4591cf180
Author: Jani Nikula <jani.nikula@intel.com>
Date:   Fri Sep 4 16:55:13 2015 +0300

    drm/i915: use pch backlight override on hsw too

Cc: Paulo Zanoni <paulo.r.zanoni@intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reference: http://mid.gmane.org/87y4frhwsn.fsf@yhuang-dev.intel.com
Reported-by: kernel test robot <ying.huang@intel.com>
Tested-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1446033429-8006-1-git-send-email-jani.nikula@intel.com

diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c
index b05c6d9..a24df35 100644
--- a/drivers/gpu/drm/i915/intel_panel.c
+++ b/drivers/gpu/drm/i915/intel_panel.c
@@ -732,6 +732,20 @@ static void lpt_disable_backlight(struct intel_connector *connector)
 
 	intel_panel_actually_set_backlight(connector, 0);
 
+	/*
+	 * Although we don't support or enable CPU PWM with LPT/SPT based
+	 * systems, it may have been enabled prior to loading the
+	 * driver. Disable to avoid warnings on LCPLL disable.
+	 *
+	 * This needs rework if we need to add support for CPU PWM on PCH split
+	 * platforms.
+	 */
+	tmp = I915_READ(BLC_PWM_CPU_CTL2);
+	if (tmp & BLM_PWM_ENABLE) {
+		DRM_DEBUG_KMS("cpu backlight was enabled, disabling\n");
+		I915_WRITE(BLC_PWM_CPU_CTL2, tmp & ~BLM_PWM_ENABLE);
+	}
+
 	tmp = I915_READ(BLC_PWM_PCH_CTL1);
 	I915_WRITE(BLC_PWM_PCH_CTL1, tmp & ~BLM_PCH_PWM_ENABLE);
 }
-- 
cgit v0.10.2


From 6bd53c4125e545a495fba63024d5522e33c600f5 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Fri, 23 Oct 2015 18:53:36 -0400
Subject: drm/amdgpu: add GFX 8.1 register headers

Minor differences compared to GFX 8.0

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/include/asic_reg/gca/gfx_8_1_d.h b/drivers/gpu/drm/amd/include/asic_reg/gca/gfx_8_1_d.h
new file mode 100644
index 0000000..2d672b3
--- /dev/null
+++ b/drivers/gpu/drm/amd/include/asic_reg/gca/gfx_8_1_d.h
@@ -0,0 +1,2791 @@
+/*
+ * GFX_8_1 Register documentation
+ *
+ * Copyright (C) 2014  Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included
+ * in all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
+ * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN
+ * AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#ifndef GFX_8_1_D_H
+#define GFX_8_1_D_H
+
+#define mmCB_BLEND_RED                                                          0xa105
+#define mmCB_BLEND_GREEN                                                        0xa106
+#define mmCB_BLEND_BLUE                                                         0xa107
+#define mmCB_BLEND_ALPHA                                                        0xa108
+#define mmCB_DCC_CONTROL                                                        0xa109
+#define mmCB_COLOR_CONTROL                                                      0xa202
+#define mmCB_BLEND0_CONTROL                                                     0xa1e0
+#define mmCB_BLEND1_CONTROL                                                     0xa1e1
+#define mmCB_BLEND2_CONTROL                                                     0xa1e2
+#define mmCB_BLEND3_CONTROL                                                     0xa1e3
+#define mmCB_BLEND4_CONTROL                                                     0xa1e4
+#define mmCB_BLEND5_CONTROL                                                     0xa1e5
+#define mmCB_BLEND6_CONTROL                                                     0xa1e6
+#define mmCB_BLEND7_CONTROL                                                     0xa1e7
+#define mmCB_COLOR0_BASE                                                        0xa318
+#define mmCB_COLOR1_BASE                                                        0xa327
+#define mmCB_COLOR2_BASE                                                        0xa336
+#define mmCB_COLOR3_BASE                                                        0xa345
+#define mmCB_COLOR4_BASE                                                        0xa354
+#define mmCB_COLOR5_BASE                                                        0xa363
+#define mmCB_COLOR6_BASE                                                        0xa372
+#define mmCB_COLOR7_BASE                                                        0xa381
+#define mmCB_COLOR0_PITCH                                                       0xa319
+#define mmCB_COLOR1_PITCH                                                       0xa328
+#define mmCB_COLOR2_PITCH                                                       0xa337
+#define mmCB_COLOR3_PITCH                                                       0xa346
+#define mmCB_COLOR4_PITCH                                                       0xa355
+#define mmCB_COLOR5_PITCH                                                       0xa364
+#define mmCB_COLOR6_PITCH                                                       0xa373
+#define mmCB_COLOR7_PITCH                                                       0xa382
+#define mmCB_COLOR0_SLICE                                                       0xa31a
+#define mmCB_COLOR1_SLICE                                                       0xa329
+#define mmCB_COLOR2_SLICE                                                       0xa338
+#define mmCB_COLOR3_SLICE                                                       0xa347
+#define mmCB_COLOR4_SLICE                                                       0xa356
+#define mmCB_COLOR5_SLICE                                                       0xa365
+#define mmCB_COLOR6_SLICE                                                       0xa374
+#define mmCB_COLOR7_SLICE                                                       0xa383
+#define mmCB_COLOR0_VIEW                                                        0xa31b
+#define mmCB_COLOR1_VIEW                                                        0xa32a
+#define mmCB_COLOR2_VIEW                                                        0xa339
+#define mmCB_COLOR3_VIEW                                                        0xa348
+#define mmCB_COLOR4_VIEW                                                        0xa357
+#define mmCB_COLOR5_VIEW                                                        0xa366
+#define mmCB_COLOR6_VIEW                                                        0xa375
+#define mmCB_COLOR7_VIEW                                                        0xa384
+#define mmCB_COLOR0_INFO                                                        0xa31c
+#define mmCB_COLOR1_INFO                                                        0xa32b
+#define mmCB_COLOR2_INFO                                                        0xa33a
+#define mmCB_COLOR3_INFO                                                        0xa349
+#define mmCB_COLOR4_INFO                                                        0xa358
+#define mmCB_COLOR5_INFO                                                        0xa367
+#define mmCB_COLOR6_INFO                                                        0xa376
+#define mmCB_COLOR7_INFO                                                        0xa385
+#define mmCB_COLOR0_ATTRIB                                                      0xa31d
+#define mmCB_COLOR1_ATTRIB                                                      0xa32c
+#define mmCB_COLOR2_ATTRIB                                                      0xa33b
+#define mmCB_COLOR3_ATTRIB                                                      0xa34a
+#define mmCB_COLOR4_ATTRIB                                                      0xa359
+#define mmCB_COLOR5_ATTRIB                                                      0xa368
+#define mmCB_COLOR6_ATTRIB                                                      0xa377
+#define mmCB_COLOR7_ATTRIB                                                      0xa386
+#define mmCB_COLOR0_DCC_CONTROL                                                 0xa31e
+#define mmCB_COLOR1_DCC_CONTROL                                                 0xa32d
+#define mmCB_COLOR2_DCC_CONTROL                                                 0xa33c
+#define mmCB_COLOR3_DCC_CONTROL                                                 0xa34b
+#define mmCB_COLOR4_DCC_CONTROL                                                 0xa35a
+#define mmCB_COLOR5_DCC_CONTROL                                                 0xa369
+#define mmCB_COLOR6_DCC_CONTROL                                                 0xa378
+#define mmCB_COLOR7_DCC_CONTROL                                                 0xa387
+#define mmCB_COLOR0_CMASK                                                       0xa31f
+#define mmCB_COLOR1_CMASK                                                       0xa32e
+#define mmCB_COLOR2_CMASK                                                       0xa33d
+#define mmCB_COLOR3_CMASK                                                       0xa34c
+#define mmCB_COLOR4_CMASK                                                       0xa35b
+#define mmCB_COLOR5_CMASK                                                       0xa36a
+#define mmCB_COLOR6_CMASK                                                       0xa379
+#define mmCB_COLOR7_CMASK                                                       0xa388
+#define mmCB_COLOR0_CMASK_SLICE                                                 0xa320
+#define mmCB_COLOR1_CMASK_SLICE                                                 0xa32f
+#define mmCB_COLOR2_CMASK_SLICE                                                 0xa33e
+#define mmCB_COLOR3_CMASK_SLICE                                                 0xa34d
+#define mmCB_COLOR4_CMASK_SLICE                                                 0xa35c
+#define mmCB_COLOR5_CMASK_SLICE                                                 0xa36b
+#define mmCB_COLOR6_CMASK_SLICE                                                 0xa37a
+#define mmCB_COLOR7_CMASK_SLICE                                                 0xa389
+#define mmCB_COLOR0_FMASK                                                       0xa321
+#define mmCB_COLOR1_FMASK                                                       0xa330
+#define mmCB_COLOR2_FMASK                                                       0xa33f
+#define mmCB_COLOR3_FMASK                                                       0xa34e
+#define mmCB_COLOR4_FMASK                                                       0xa35d
+#define mmCB_COLOR5_FMASK                                                       0xa36c
+#define mmCB_COLOR6_FMASK                                                       0xa37b
+#define mmCB_COLOR7_FMASK                                                       0xa38a
+#define mmCB_COLOR0_FMASK_SLICE                                                 0xa322
+#define mmCB_COLOR1_FMASK_SLICE                                                 0xa331
+#define mmCB_COLOR2_FMASK_SLICE                                                 0xa340
+#define mmCB_COLOR3_FMASK_SLICE                                                 0xa34f
+#define mmCB_COLOR4_FMASK_SLICE                                                 0xa35e
+#define mmCB_COLOR5_FMASK_SLICE                                                 0xa36d
+#define mmCB_COLOR6_FMASK_SLICE                                                 0xa37c
+#define mmCB_COLOR7_FMASK_SLICE                                                 0xa38b
+#define mmCB_COLOR0_CLEAR_WORD0                                                 0xa323
+#define mmCB_COLOR1_CLEAR_WORD0                                                 0xa332
+#define mmCB_COLOR2_CLEAR_WORD0                                                 0xa341
+#define mmCB_COLOR3_CLEAR_WORD0                                                 0xa350
+#define mmCB_COLOR4_CLEAR_WORD0                                                 0xa35f
+#define mmCB_COLOR5_CLEAR_WORD0                                                 0xa36e
+#define mmCB_COLOR6_CLEAR_WORD0                                                 0xa37d
+#define mmCB_COLOR7_CLEAR_WORD0                                                 0xa38c
+#define mmCB_COLOR0_CLEAR_WORD1                                                 0xa324
+#define mmCB_COLOR1_CLEAR_WORD1                                                 0xa333
+#define mmCB_COLOR2_CLEAR_WORD1                                                 0xa342
+#define mmCB_COLOR3_CLEAR_WORD1                                                 0xa351
+#define mmCB_COLOR4_CLEAR_WORD1                                                 0xa360
+#define mmCB_COLOR5_CLEAR_WORD1                                                 0xa36f
+#define mmCB_COLOR6_CLEAR_WORD1                                                 0xa37e
+#define mmCB_COLOR7_CLEAR_WORD1                                                 0xa38d
+#define mmCB_COLOR0_DCC_BASE                                                    0xa325
+#define mmCB_COLOR1_DCC_BASE                                                    0xa334
+#define mmCB_COLOR2_DCC_BASE                                                    0xa343
+#define mmCB_COLOR3_DCC_BASE                                                    0xa352
+#define mmCB_COLOR4_DCC_BASE                                                    0xa361
+#define mmCB_COLOR5_DCC_BASE                                                    0xa370
+#define mmCB_COLOR6_DCC_BASE                                                    0xa37f
+#define mmCB_COLOR7_DCC_BASE                                                    0xa38e
+#define mmCB_TARGET_MASK                                                        0xa08e
+#define mmCB_SHADER_MASK                                                        0xa08f
+#define mmCB_HW_CONTROL                                                         0x2684
+#define mmCB_HW_CONTROL_1                                                       0x2685
+#define mmCB_HW_CONTROL_2                                                       0x2686
+#define mmCB_HW_CONTROL_3                                                       0x2683
+#define mmCB_DCC_CONFIG                                                         0x2687
+#define mmCB_PERFCOUNTER_FILTER                                                 0xdc00
+#define mmCB_PERFCOUNTER0_SELECT                                                0xdc01
+#define mmCB_PERFCOUNTER0_SELECT1                                               0xdc02
+#define mmCB_PERFCOUNTER1_SELECT                                                0xdc03
+#define mmCB_PERFCOUNTER2_SELECT                                                0xdc04
+#define mmCB_PERFCOUNTER3_SELECT                                                0xdc05
+#define mmCB_PERFCOUNTER0_LO                                                    0xd406
+#define mmCB_PERFCOUNTER1_LO                                                    0xd408
+#define mmCB_PERFCOUNTER2_LO                                                    0xd40a
+#define mmCB_PERFCOUNTER3_LO                                                    0xd40c
+#define mmCB_PERFCOUNTER0_HI                                                    0xd407
+#define mmCB_PERFCOUNTER1_HI                                                    0xd409
+#define mmCB_PERFCOUNTER2_HI                                                    0xd40b
+#define mmCB_PERFCOUNTER3_HI                                                    0xd40d
+#define mmCB_CGTT_SCLK_CTRL                                                     0xf0a8
+#define mmCB_DEBUG_BUS_1                                                        0x2699
+#define mmCB_DEBUG_BUS_2                                                        0x269a
+#define mmCB_DEBUG_BUS_3                                                        0x269b
+#define mmCB_DEBUG_BUS_4                                                        0x269c
+#define mmCB_DEBUG_BUS_5                                                        0x269d
+#define mmCB_DEBUG_BUS_6                                                        0x269e
+#define mmCB_DEBUG_BUS_7                                                        0x269f
+#define mmCB_DEBUG_BUS_8                                                        0x26a0
+#define mmCB_DEBUG_BUS_9                                                        0x26a1
+#define mmCB_DEBUG_BUS_10                                                       0x26a2
+#define mmCB_DEBUG_BUS_11                                                       0x26a3
+#define mmCB_DEBUG_BUS_12                                                       0x26a4
+#define mmCB_DEBUG_BUS_13                                                       0x26a5
+#define mmCB_DEBUG_BUS_14                                                       0x26a6
+#define mmCB_DEBUG_BUS_15                                                       0x26a7
+#define mmCB_DEBUG_BUS_16                                                       0x26a8
+#define mmCB_DEBUG_BUS_17                                                       0x26a9
+#define mmCB_DEBUG_BUS_18                                                       0x26aa
+#define mmCB_DEBUG_BUS_19                                                       0x26ab
+#define mmCB_DEBUG_BUS_20                                                       0x26ac
+#define mmCB_DEBUG_BUS_21                                                       0x26ad
+#define mmCB_DEBUG_BUS_22                                                       0x26ae
+#define mmCP_DFY_CNTL                                                           0x3020
+#define mmCP_DFY_STAT                                                           0x3021
+#define mmCP_DFY_ADDR_HI                                                        0x3022
+#define mmCP_DFY_ADDR_LO                                                        0x3023
+#define mmCP_DFY_DATA_0                                                         0x3024
+#define mmCP_DFY_DATA_1                                                         0x3025
+#define mmCP_DFY_DATA_2                                                         0x3026
+#define mmCP_DFY_DATA_3                                                         0x3027
+#define mmCP_DFY_DATA_4                                                         0x3028
+#define mmCP_DFY_DATA_5                                                         0x3029
+#define mmCP_DFY_DATA_6                                                         0x302a
+#define mmCP_DFY_DATA_7                                                         0x302b
+#define mmCP_DFY_DATA_8                                                         0x302c
+#define mmCP_DFY_DATA_9                                                         0x302d
+#define mmCP_DFY_DATA_10                                                        0x302e
+#define mmCP_DFY_DATA_11                                                        0x302f
+#define mmCP_DFY_DATA_12                                                        0x3030
+#define mmCP_DFY_DATA_13                                                        0x3031
+#define mmCP_DFY_DATA_14                                                        0x3032
+#define mmCP_DFY_DATA_15                                                        0x3033
+#define mmCP_DFY_CMD                                                            0x3034
+#define mmCP_CPC_MGCG_SYNC_CNTL                                                 0x3036
+#define mmCP_ATCL1_CNTL                                                         0x303c
+#define mmCP_RB0_BASE                                                           0x3040
+#define mmCP_RB0_BASE_HI                                                        0x30b1
+#define mmCP_RB_BASE                                                            0x3040
+#define mmCP_RB1_BASE                                                           0x3060
+#define mmCP_RB1_BASE_HI                                                        0x30b2
+#define mmCP_RB2_BASE                                                           0x3065
+#define mmCP_RB0_CNTL                                                           0x3041
+#define mmCP_RB_CNTL                                                            0x3041
+#define mmCP_RB1_CNTL                                                           0x3061
+#define mmCP_RB2_CNTL                                                           0x3066
+#define mmCP_RB_RPTR_WR                                                         0x3042
+#define mmCP_RB0_RPTR_ADDR                                                      0x3043
+#define mmCP_RB_RPTR_ADDR                                                       0x3043
+#define mmCP_RB1_RPTR_ADDR                                                      0x3062
+#define mmCP_RB2_RPTR_ADDR                                                      0x3067
+#define mmCP_RB0_RPTR_ADDR_HI                                                   0x3044
+#define mmCP_RB_RPTR_ADDR_HI                                                    0x3044
+#define mmCP_RB1_RPTR_ADDR_HI                                                   0x3063
+#define mmCP_RB2_RPTR_ADDR_HI                                                   0x3068
+#define mmCP_RB0_WPTR                                                           0x3045
+#define mmCP_RB_WPTR                                                            0x3045
+#define mmCP_RB1_WPTR                                                           0x3064
+#define mmCP_RB2_WPTR                                                           0x3069
+#define mmCP_RB_WPTR_POLL_ADDR_LO                                               0x3046
+#define mmCP_RB_WPTR_POLL_ADDR_HI                                               0x3047
+#define mmGC_PRIV_MODE                                                          0x3048
+#define mmCP_INT_CNTL                                                           0x3049
+#define mmCP_INT_CNTL_RING0                                                     0x306a
+#define mmCP_INT_CNTL_RING1                                                     0x306b
+#define mmCP_INT_CNTL_RING2                                                     0x306c
+#define mmCP_INT_STATUS                                                         0x304a
+#define mmCP_INT_STATUS_RING0                                                   0x306d
+#define mmCP_INT_STATUS_RING1                                                   0x306e
+#define mmCP_INT_STATUS_RING2                                                   0x306f
+#define mmCP_DEVICE_ID                                                          0x304b
+#define mmCP_RING_PRIORITY_CNTS                                                 0x304c
+#define mmCP_ME0_PIPE_PRIORITY_CNTS                                             0x304c
+#define mmCP_RING0_PRIORITY                                                     0x304d
+#define mmCP_ME0_PIPE0_PRIORITY                                                 0x304d
+#define mmCP_RING1_PRIORITY                                                     0x304e
+#define mmCP_ME0_PIPE1_PRIORITY                                                 0x304e
+#define mmCP_RING2_PRIORITY                                                     0x304f
+#define mmCP_ME0_PIPE2_PRIORITY                                                 0x304f
+#define mmCP_ENDIAN_SWAP                                                        0x3050
+#define mmCP_RB_VMID                                                            0x3051
+#define mmCP_ME0_PIPE0_VMID                                                     0x3052
+#define mmCP_ME0_PIPE1_VMID                                                     0x3053
+#define mmCP_RB_DOORBELL_CONTROL                                                0x3059
+#define mmCP_RB_DOORBELL_RANGE_LOWER                                            0x305a
+#define mmCP_RB_DOORBELL_RANGE_UPPER                                            0x305b
+#define mmCP_MEC_DOORBELL_RANGE_LOWER                                           0x305c
+#define mmCP_MEC_DOORBELL_RANGE_UPPER                                           0x305d
+#define mmCP_PFP_UCODE_ADDR                                                     0xf814
+#define mmCP_PFP_UCODE_DATA                                                     0xf815
+#define mmCP_ME_RAM_RADDR                                                       0xf816
+#define mmCP_ME_RAM_WADDR                                                       0xf816
+#define mmCP_ME_RAM_DATA                                                        0xf817
+#define mmCGTT_CPC_CLK_CTRL                                                     0xf0b2
+#define mmCGTT_CPF_CLK_CTRL                                                     0xf0b1
+#define mmCGTT_CP_CLK_CTRL                                                      0xf0b0
+#define mmCP_CE_UCODE_ADDR                                                      0xf818
+#define mmCP_CE_UCODE_DATA                                                      0xf819
+#define mmCP_MEC_ME1_UCODE_ADDR                                                 0xf81a
+#define mmCP_MEC_ME1_UCODE_DATA                                                 0xf81b
+#define mmCP_MEC_ME2_UCODE_ADDR                                                 0xf81c
+#define mmCP_MEC_ME2_UCODE_DATA                                                 0xf81d
+#define mmCP_MEC1_F32_INT_DIS                                                   0x30bd
+#define mmCP_MEC2_F32_INT_DIS                                                   0x30be
+#define mmCP_PWR_CNTL                                                           0x3078
+#define mmCP_MEM_SLP_CNTL                                                       0x3079
+#define mmCP_ECC_FIRSTOCCURRENCE                                                0x307a
+#define mmCP_ECC_FIRSTOCCURRENCE_RING0                                          0x307b
+#define mmCP_ECC_FIRSTOCCURRENCE_RING1                                          0x307c
+#define mmCP_ECC_FIRSTOCCURRENCE_RING2                                          0x307d
+#define mmCP_CPF_DEBUG                                                          0x3080
+#define mmCP_PQ_WPTR_POLL_CNTL                                                  0x3083
+#define mmCP_PQ_WPTR_POLL_CNTL1                                                 0x3084
+#define mmCPC_INT_CNTL                                                          0x30b4
+#define mmCP_ME1_PIPE0_INT_CNTL                                                 0x3085
+#define mmCP_ME1_PIPE1_INT_CNTL                                                 0x3086
+#define mmCP_ME1_PIPE2_INT_CNTL                                                 0x3087
+#define mmCP_ME1_PIPE3_INT_CNTL                                                 0x3088
+#define mmCP_ME2_PIPE0_INT_CNTL                                                 0x3089
+#define mmCP_ME2_PIPE1_INT_CNTL                                                 0x308a
+#define mmCP_ME2_PIPE2_INT_CNTL                                                 0x308b
+#define mmCP_ME2_PIPE3_INT_CNTL                                                 0x308c
+#define mmCPC_INT_STATUS                                                        0x30b5
+#define mmCP_ME1_PIPE0_INT_STATUS                                               0x308d
+#define mmCP_ME1_PIPE1_INT_STATUS                                               0x308e
+#define mmCP_ME1_PIPE2_INT_STATUS                                               0x308f
+#define mmCP_ME1_PIPE3_INT_STATUS                                               0x3090
+#define mmCP_ME2_PIPE0_INT_STATUS                                               0x3091
+#define mmCP_ME2_PIPE1_INT_STATUS                                               0x3092
+#define mmCP_ME2_PIPE2_INT_STATUS                                               0x3093
+#define mmCP_ME2_PIPE3_INT_STATUS                                               0x3094
+#define mmCP_ME1_INT_STAT_DEBUG                                                 0x3095
+#define mmCP_ME2_INT_STAT_DEBUG                                                 0x3096
+#define mmCP_ME1_PIPE_PRIORITY_CNTS                                             0x3099
+#define mmCP_ME1_PIPE0_PRIORITY                                                 0x309a
+#define mmCP_ME1_PIPE1_PRIORITY                                                 0x309b
+#define mmCP_ME1_PIPE2_PRIORITY                                                 0x309c
+#define mmCP_ME1_PIPE3_PRIORITY                                                 0x309d
+#define mmCP_ME2_PIPE_PRIORITY_CNTS                                             0x309e
+#define mmCP_ME2_PIPE0_PRIORITY                                                 0x309f
+#define mmCP_ME2_PIPE1_PRIORITY                                                 0x30a0
+#define mmCP_ME2_PIPE2_PRIORITY                                                 0x30a1
+#define mmCP_ME2_PIPE3_PRIORITY                                                 0x30a2
+#define mmCP_CE_PRGRM_CNTR_START                                                0x30a3
+#define mmCP_PFP_PRGRM_CNTR_START                                               0x30a4
+#define mmCP_ME_PRGRM_CNTR_START                                                0x30a5
+#define mmCP_MEC1_PRGRM_CNTR_START                                              0x30a6
+#define mmCP_MEC2_PRGRM_CNTR_START                                              0x30a7
+#define mmCP_CE_INTR_ROUTINE_START                                              0x30a8
+#define mmCP_PFP_INTR_ROUTINE_START                                             0x30a9
+#define mmCP_ME_INTR_ROUTINE_START                                              0x30aa
+#define mmCP_MEC1_INTR_ROUTINE_START                                            0x30ab
+#define mmCP_MEC2_INTR_ROUTINE_START                                            0x30ac
+#define mmCP_CONTEXT_CNTL                                                       0x30ad
+#define mmCP_MAX_CONTEXT                                                        0x30ae
+#define mmCP_IQ_WAIT_TIME1                                                      0x30af
+#define mmCP_IQ_WAIT_TIME2                                                      0x30b0
+#define mmCP_VMID_RESET                                                         0x30b3
+#define mmCP_VMID_PREEMPT                                                       0x30b6
+#define mmCP_VMID_STATUS                                                        0x30bf
+#define mmCPC_INT_CNTX_ID                                                       0x30b7
+#define mmCP_PQ_STATUS                                                          0x30b8
+#define mmCP_CPC_IC_BASE_LO                                                     0x30b9
+#define mmCP_CPC_IC_BASE_HI                                                     0x30ba
+#define mmCP_CPC_IC_BASE_CNTL                                                   0x30bb
+#define mmCP_CPC_IC_OP_CNTL                                                     0x30bc
+#define mmCP_CPC_STATUS                                                         0x2084
+#define mmCP_CPC_BUSY_STAT                                                      0x2085
+#define mmCP_CPC_STALLED_STAT1                                                  0x2086
+#define mmCP_CPF_STATUS                                                         0x2087
+#define mmCP_CPF_BUSY_STAT                                                      0x2088
+#define mmCP_CPF_STALLED_STAT1                                                  0x2089
+#define mmCP_CPC_GRBM_FREE_COUNT                                                0x208b
+#define mmCP_MEC_CNTL                                                           0x208d
+#define mmCP_MEC_ME1_HEADER_DUMP                                                0x208e
+#define mmCP_MEC_ME2_HEADER_DUMP                                                0x208f
+#define mmCP_CPC_SCRATCH_INDEX                                                  0x2090
+#define mmCP_CPC_SCRATCH_DATA                                                   0x2091
+#define mmCPG_PERFCOUNTER1_SELECT                                               0xd800
+#define mmCPG_PERFCOUNTER1_LO                                                   0xd000
+#define mmCPG_PERFCOUNTER1_HI                                                   0xd001
+#define mmCPG_PERFCOUNTER0_SELECT1                                              0xd801
+#define mmCPG_PERFCOUNTER0_SELECT                                               0xd802
+#define mmCPG_PERFCOUNTER0_LO                                                   0xd002
+#define mmCPG_PERFCOUNTER0_HI                                                   0xd003
+#define mmCPC_PERFCOUNTER1_SELECT                                               0xd803
+#define mmCPC_PERFCOUNTER1_LO                                                   0xd004
+#define mmCPC_PERFCOUNTER1_HI                                                   0xd005
+#define mmCPC_PERFCOUNTER0_SELECT1                                              0xd804
+#define mmCPC_PERFCOUNTER0_SELECT                                               0xd809
+#define mmCPC_PERFCOUNTER0_LO                                                   0xd006
+#define mmCPC_PERFCOUNTER0_HI                                                   0xd007
+#define mmCPF_PERFCOUNTER1_SELECT                                               0xd805
+#define mmCPF_PERFCOUNTER1_LO                                                   0xd008
+#define mmCPF_PERFCOUNTER1_HI                                                   0xd009
+#define mmCPF_PERFCOUNTER0_SELECT1                                              0xd806
+#define mmCPF_PERFCOUNTER0_SELECT                                               0xd807
+#define mmCPF_PERFCOUNTER0_LO                                                   0xd00a
+#define mmCPF_PERFCOUNTER0_HI                                                   0xd00b
+#define mmCP_CPC_HALT_HYST_COUNT                                                0x20a7
+#define mmCP_DRAW_OBJECT                                                        0xd810
+#define mmCP_DRAW_OBJECT_COUNTER                                                0xd811
+#define mmCP_DRAW_WINDOW_MASK_HI                                                0xd812
+#define mmCP_DRAW_WINDOW_HI                                                     0xd813
+#define mmCP_DRAW_WINDOW_LO                                                     0xd814
+#define mmCP_DRAW_WINDOW_CNTL                                                   0xd815
+#define mmCP_PRT_LOD_STATS_CNTL0                                                0x20ad
+#define mmCP_PRT_LOD_STATS_CNTL1                                                0x20ae
+#define mmCP_PRT_LOD_STATS_CNTL2                                                0x20af
+#define mmCP_CE_COMPARE_COUNT                                                   0x20c0
+#define mmCP_CE_DE_COUNT                                                        0x20c1
+#define mmCP_DE_CE_COUNT                                                        0x20c2
+#define mmCP_DE_LAST_INVAL_COUNT                                                0x20c3
+#define mmCP_DE_DE_COUNT                                                        0x20c4
+#define mmCP_EOP_DONE_EVENT_CNTL                                                0xc0d5
+#define mmCP_EOP_DONE_DATA_CNTL                                                 0xc0d6
+#define mmCP_EOP_DONE_CNTX_ID                                                   0xc0d7
+#define mmCP_EOP_DONE_ADDR_LO                                                   0xc000
+#define mmCP_EOP_DONE_ADDR_HI                                                   0xc001
+#define mmCP_EOP_DONE_DATA_LO                                                   0xc002
+#define mmCP_EOP_DONE_DATA_HI                                                   0xc003
+#define mmCP_EOP_LAST_FENCE_LO                                                  0xc004
+#define mmCP_EOP_LAST_FENCE_HI                                                  0xc005
+#define mmCP_STREAM_OUT_ADDR_LO                                                 0xc006
+#define mmCP_STREAM_OUT_ADDR_HI                                                 0xc007
+#define mmCP_NUM_PRIM_WRITTEN_COUNT0_LO                                         0xc008
+#define mmCP_NUM_PRIM_WRITTEN_COUNT0_HI                                         0xc009
+#define mmCP_NUM_PRIM_NEEDED_COUNT0_LO                                          0xc00a
+#define mmCP_NUM_PRIM_NEEDED_COUNT0_HI                                          0xc00b
+#define mmCP_NUM_PRIM_WRITTEN_COUNT1_LO                                         0xc00c
+#define mmCP_NUM_PRIM_WRITTEN_COUNT1_HI                                         0xc00d
+#define mmCP_NUM_PRIM_NEEDED_COUNT1_LO                                          0xc00e
+#define mmCP_NUM_PRIM_NEEDED_COUNT1_HI                                          0xc00f
+#define mmCP_NUM_PRIM_WRITTEN_COUNT2_LO                                         0xc010
+#define mmCP_NUM_PRIM_WRITTEN_COUNT2_HI                                         0xc011
+#define mmCP_NUM_PRIM_NEEDED_COUNT2_LO                                          0xc012
+#define mmCP_NUM_PRIM_NEEDED_COUNT2_HI                                          0xc013
+#define mmCP_NUM_PRIM_WRITTEN_COUNT3_LO                                         0xc014
+#define mmCP_NUM_PRIM_WRITTEN_COUNT3_HI                                         0xc015
+#define mmCP_NUM_PRIM_NEEDED_COUNT3_LO                                          0xc016
+#define mmCP_NUM_PRIM_NEEDED_COUNT3_HI                                          0xc017
+#define mmCP_PIPE_STATS_ADDR_LO                                                 0xc018
+#define mmCP_PIPE_STATS_ADDR_HI                                                 0xc019
+#define mmCP_VGT_IAVERT_COUNT_LO                                                0xc01a
+#define mmCP_VGT_IAVERT_COUNT_HI                                                0xc01b
+#define mmCP_VGT_IAPRIM_COUNT_LO                                                0xc01c
+#define mmCP_VGT_IAPRIM_COUNT_HI                                                0xc01d
+#define mmCP_VGT_GSPRIM_COUNT_LO                                                0xc01e
+#define mmCP_VGT_GSPRIM_COUNT_HI                                                0xc01f
+#define mmCP_VGT_VSINVOC_COUNT_LO                                               0xc020
+#define mmCP_VGT_VSINVOC_COUNT_HI                                               0xc021
+#define mmCP_VGT_GSINVOC_COUNT_LO                                               0xc022
+#define mmCP_VGT_GSINVOC_COUNT_HI                                               0xc023
+#define mmCP_VGT_HSINVOC_COUNT_LO                                               0xc024
+#define mmCP_VGT_HSINVOC_COUNT_HI                                               0xc025
+#define mmCP_VGT_DSINVOC_COUNT_LO                                               0xc026
+#define mmCP_VGT_DSINVOC_COUNT_HI                                               0xc027
+#define mmCP_PA_CINVOC_COUNT_LO                                                 0xc028
+#define mmCP_PA_CINVOC_COUNT_HI                                                 0xc029
+#define mmCP_PA_CPRIM_COUNT_LO                                                  0xc02a
+#define mmCP_PA_CPRIM_COUNT_HI                                                  0xc02b
+#define mmCP_SC_PSINVOC_COUNT0_LO                                               0xc02c
+#define mmCP_SC_PSINVOC_COUNT0_HI                                               0xc02d
+#define mmCP_SC_PSINVOC_COUNT1_LO                                               0xc02e
+#define mmCP_SC_PSINVOC_COUNT1_HI                                               0xc02f
+#define mmCP_VGT_CSINVOC_COUNT_LO                                               0xc030
+#define mmCP_VGT_CSINVOC_COUNT_HI                                               0xc031
+#define mmCP_PIPE_STATS_CONTROL                                                 0xc03d
+#define mmCP_STREAM_OUT_CONTROL                                                 0xc03e
+#define mmCP_STRMOUT_CNTL                                                       0xc03f
+#define mmSCRATCH_REG0                                                          0xc040
+#define mmSCRATCH_REG1                                                          0xc041
+#define mmSCRATCH_REG2                                                          0xc042
+#define mmSCRATCH_REG3                                                          0xc043
+#define mmSCRATCH_REG4                                                          0xc044
+#define mmSCRATCH_REG5                                                          0xc045
+#define mmSCRATCH_REG6                                                          0xc046
+#define mmSCRATCH_REG7                                                          0xc047
+#define mmSCRATCH_UMSK                                                          0xc050
+#define mmSCRATCH_ADDR                                                          0xc051
+#define mmCP_PFP_ATOMIC_PREOP_LO                                                0xc052
+#define mmCP_PFP_ATOMIC_PREOP_HI                                                0xc053
+#define mmCP_PFP_GDS_ATOMIC0_PREOP_LO                                           0xc054
+#define mmCP_PFP_GDS_ATOMIC0_PREOP_HI                                           0xc055
+#define mmCP_PFP_GDS_ATOMIC1_PREOP_LO                                           0xc056
+#define mmCP_PFP_GDS_ATOMIC1_PREOP_HI                                           0xc057
+#define mmCP_APPEND_ADDR_LO                                                     0xc058
+#define mmCP_APPEND_ADDR_HI                                                     0xc059
+#define mmCP_APPEND_DATA                                                        0xc05a
+#define mmCP_APPEND_LAST_CS_FENCE                                               0xc05b
+#define mmCP_APPEND_LAST_PS_FENCE                                               0xc05c
+#define mmCP_ATOMIC_PREOP_LO                                                    0xc05d
+#define mmCP_ME_ATOMIC_PREOP_LO                                                 0xc05d
+#define mmCP_ATOMIC_PREOP_HI                                                    0xc05e
+#define mmCP_ME_ATOMIC_PREOP_HI                                                 0xc05e
+#define mmCP_GDS_ATOMIC0_PREOP_LO                                               0xc05f
+#define mmCP_ME_GDS_ATOMIC0_PREOP_LO                                            0xc05f
+#define mmCP_GDS_ATOMIC0_PREOP_HI                                               0xc060
+#define mmCP_ME_GDS_ATOMIC0_PREOP_HI                                            0xc060
+#define mmCP_GDS_ATOMIC1_PREOP_LO                                               0xc061
+#define mmCP_ME_GDS_ATOMIC1_PREOP_LO                                            0xc061
+#define mmCP_GDS_ATOMIC1_PREOP_HI                                               0xc062
+#define mmCP_ME_GDS_ATOMIC1_PREOP_HI                                            0xc062
+#define mmCP_ME_MC_WADDR_LO                                                     0xc069
+#define mmCP_ME_MC_WADDR_HI                                                     0xc06a
+#define mmCP_ME_MC_WDATA_LO                                                     0xc06b
+#define mmCP_ME_MC_WDATA_HI                                                     0xc06c
+#define mmCP_ME_MC_RADDR_LO                                                     0xc06d
+#define mmCP_ME_MC_RADDR_HI                                                     0xc06e
+#define mmCP_SEM_WAIT_TIMER                                                     0xc06f
+#define mmCP_SIG_SEM_ADDR_LO                                                    0xc070
+#define mmCP_SIG_SEM_ADDR_HI                                                    0xc071
+#define mmCP_WAIT_SEM_ADDR_LO                                                   0xc075
+#define mmCP_WAIT_SEM_ADDR_HI                                                   0xc076
+#define mmCP_WAIT_REG_MEM_TIMEOUT                                               0xc074
+#define mmCP_COHER_START_DELAY                                                  0xc07b
+#define mmCP_COHER_CNTL                                                         0xc07c
+#define mmCP_COHER_SIZE                                                         0xc07d
+#define mmCP_COHER_SIZE_HI                                                      0xc08c
+#define mmCP_COHER_BASE                                                         0xc07e
+#define mmCP_COHER_BASE_HI                                                      0xc079
+#define mmCP_COHER_STATUS                                                       0xc07f
+#define mmCOHER_DEST_BASE_0                                                     0xa092
+#define mmCOHER_DEST_BASE_1                                                     0xa093
+#define mmCOHER_DEST_BASE_2                                                     0xa07e
+#define mmCOHER_DEST_BASE_3                                                     0xa07f
+#define mmCOHER_DEST_BASE_HI_0                                                  0xa07a
+#define mmCOHER_DEST_BASE_HI_1                                                  0xa07b
+#define mmCOHER_DEST_BASE_HI_2                                                  0xa07c
+#define mmCOHER_DEST_BASE_HI_3                                                  0xa07d
+#define mmCP_DMA_ME_SRC_ADDR                                                    0xc080
+#define mmCP_DMA_ME_SRC_ADDR_HI                                                 0xc081
+#define mmCP_DMA_ME_DST_ADDR                                                    0xc082
+#define mmCP_DMA_ME_DST_ADDR_HI                                                 0xc083
+#define mmCP_DMA_ME_CONTROL                                                     0xc078
+#define mmCP_DMA_ME_COMMAND                                                     0xc084
+#define mmCP_DMA_PFP_SRC_ADDR                                                   0xc085
+#define mmCP_DMA_PFP_SRC_ADDR_HI                                                0xc086
+#define mmCP_DMA_PFP_DST_ADDR                                                   0xc087
+#define mmCP_DMA_PFP_DST_ADDR_HI                                                0xc088
+#define mmCP_DMA_PFP_CONTROL                                                    0xc077
+#define mmCP_DMA_PFP_COMMAND                                                    0xc089
+#define mmCP_DMA_CNTL                                                           0xc08a
+#define mmCP_DMA_READ_TAGS                                                      0xc08b
+#define mmCP_PFP_IB_CONTROL                                                     0xc08d
+#define mmCP_PFP_LOAD_CONTROL                                                   0xc08e
+#define mmCP_SCRATCH_INDEX                                                      0xc08f
+#define mmCP_SCRATCH_DATA                                                       0xc090
+#define mmCP_RB_OFFSET                                                          0xc091
+#define mmCP_IB1_OFFSET                                                         0xc092
+#define mmCP_IB2_OFFSET                                                         0xc093
+#define mmCP_IB1_PREAMBLE_BEGIN                                                 0xc094
+#define mmCP_IB1_PREAMBLE_END                                                   0xc095
+#define mmCP_IB2_PREAMBLE_BEGIN                                                 0xc096
+#define mmCP_IB2_PREAMBLE_END                                                   0xc097
+#define mmCP_CE_IB1_OFFSET                                                      0xc098
+#define mmCP_CE_IB2_OFFSET                                                      0xc099
+#define mmCP_CE_COUNTER                                                         0xc09a
+#define mmCP_CE_RB_OFFSET                                                       0xc09b
+#define mmCP_PFP_COMPLETION_STATUS                                              0xc0ec
+#define mmCP_CE_COMPLETION_STATUS                                               0xc0ed
+#define mmCP_PRED_NOT_VISIBLE                                                   0xc0ee
+#define mmCP_PFP_METADATA_BASE_ADDR                                             0xc0f0
+#define mmCP_PFP_METADATA_BASE_ADDR_HI                                          0xc0f1
+#define mmCP_CE_METADATA_BASE_ADDR                                              0xc0f2
+#define mmCP_CE_METADATA_BASE_ADDR_HI                                           0xc0f3
+#define mmCP_DRAW_INDX_INDR_ADDR                                                0xc0f4
+#define mmCP_DRAW_INDX_INDR_ADDR_HI                                             0xc0f5
+#define mmCP_DISPATCH_INDR_ADDR                                                 0xc0f6
+#define mmCP_DISPATCH_INDR_ADDR_HI                                              0xc0f7
+#define mmCP_INDEX_BASE_ADDR                                                    0xc0f8
+#define mmCP_INDEX_BASE_ADDR_HI                                                 0xc0f9
+#define mmCP_INDEX_TYPE                                                         0xc0fa
+#define mmCP_GDS_BKUP_ADDR                                                      0xc0fb
+#define mmCP_GDS_BKUP_ADDR_HI                                                   0xc0fc
+#define mmCP_SAMPLE_STATUS                                                      0xc0fd
+#define mmCP_STALLED_STAT1                                                      0x219d
+#define mmCP_STALLED_STAT2                                                      0x219e
+#define mmCP_STALLED_STAT3                                                      0x219c
+#define mmCP_BUSY_STAT                                                          0x219f
+#define mmCP_STAT                                                               0x21a0
+#define mmCP_ME_HEADER_DUMP                                                     0x21a1
+#define mmCP_PFP_HEADER_DUMP                                                    0x21a2
+#define mmCP_GRBM_FREE_COUNT                                                    0x21a3
+#define mmCP_CE_HEADER_DUMP                                                     0x21a4
+#define mmCP_CSF_STAT                                                           0x21b4
+#define mmCP_CSF_CNTL                                                           0x21b5
+#define mmCP_ME_CNTL                                                            0x21b6
+#define mmCP_CNTX_STAT                                                          0x21b8
+#define mmCP_ME_PREEMPTION                                                      0x21b9
+#define mmCP_RB0_RPTR                                                           0x21c0
+#define mmCP_RB_RPTR                                                            0x21c0
+#define mmCP_RB1_RPTR                                                           0x21bf
+#define mmCP_RB2_RPTR                                                           0x21be
+#define mmCP_RB_WPTR_DELAY                                                      0x21c1
+#define mmCP_RB_WPTR_POLL_CNTL                                                  0x21c2
+#define mmCP_CE_INIT_BASE_LO                                                    0xc0c3
+#define mmCP_CE_INIT_BASE_HI                                                    0xc0c4
+#define mmCP_CE_INIT_BUFSZ                                                      0xc0c5
+#define mmCP_CE_IB1_BASE_LO                                                     0xc0c6
+#define mmCP_CE_IB1_BASE_HI                                                     0xc0c7
+#define mmCP_CE_IB1_BUFSZ                                                       0xc0c8
+#define mmCP_CE_IB2_BASE_LO                                                     0xc0c9
+#define mmCP_CE_IB2_BASE_HI                                                     0xc0ca
+#define mmCP_CE_IB2_BUFSZ                                                       0xc0cb
+#define mmCP_IB1_BASE_LO                                                        0xc0cc
+#define mmCP_IB1_BASE_HI                                                        0xc0cd
+#define mmCP_IB1_BUFSZ                                                          0xc0ce
+#define mmCP_IB2_BASE_LO                                                        0xc0cf
+#define mmCP_IB2_BASE_HI                                                        0xc0d0
+#define mmCP_IB2_BUFSZ                                                          0xc0d1
+#define mmCP_ST_BASE_LO                                                         0xc0d2
+#define mmCP_ST_BASE_HI                                                         0xc0d3
+#define mmCP_ST_BUFSZ                                                           0xc0d4
+#define mmCP_ROQ_THRESHOLDS                                                     0x21bc
+#define mmCP_MEQ_STQ_THRESHOLD                                                  0x21bd
+#define mmCP_ROQ1_THRESHOLDS                                                    0x21d5
+#define mmCP_ROQ2_THRESHOLDS                                                    0x21d6
+#define mmCP_STQ_THRESHOLDS                                                     0x21d7
+#define mmCP_QUEUE_THRESHOLDS                                                   0x21d8
+#define mmCP_MEQ_THRESHOLDS                                                     0x21d9
+#define mmCP_ROQ_AVAIL                                                          0x21da
+#define mmCP_STQ_AVAIL                                                          0x21db
+#define mmCP_ROQ2_AVAIL                                                         0x21dc
+#define mmCP_MEQ_AVAIL                                                          0x21dd
+#define mmCP_CMD_INDEX                                                          0x21de
+#define mmCP_CMD_DATA                                                           0x21df
+#define mmCP_ROQ_RB_STAT                                                        0x21e0
+#define mmCP_ROQ_IB1_STAT                                                       0x21e1
+#define mmCP_ROQ_IB2_STAT                                                       0x21e2
+#define mmCP_STQ_STAT                                                           0x21e3
+#define mmCP_STQ_WR_STAT                                                        0x21e4
+#define mmCP_MEQ_STAT                                                           0x21e5
+#define mmCP_CEQ1_AVAIL                                                         0x21e6
+#define mmCP_CEQ2_AVAIL                                                         0x21e7
+#define mmCP_CE_ROQ_RB_STAT                                                     0x21e8
+#define mmCP_CE_ROQ_IB1_STAT                                                    0x21e9
+#define mmCP_CE_ROQ_IB2_STAT                                                    0x21ea
+#define mmCP_INT_STAT_DEBUG                                                     0x21f7
+#define mmCP_PERFMON_CNTL                                                       0xd808
+#define mmCP_PERFMON_CNTX_CNTL                                                  0xa0d8
+#define mmCP_RINGID                                                             0xa0d9
+#define mmCP_PIPEID                                                             0xa0d9
+#define mmCP_VMID                                                               0xa0da
+#define mmCP_HPD_ROQ_OFFSETS                                                    0x3240
+#define mmCP_HPD_STATUS0                                                        0x3241
+#define mmCP_MQD_BASE_ADDR                                                      0x3245
+#define mmCP_MQD_BASE_ADDR_HI                                                   0x3246
+#define mmCP_HQD_ACTIVE                                                         0x3247
+#define mmCP_HQD_VMID                                                           0x3248
+#define mmCP_HQD_PERSISTENT_STATE                                               0x3249
+#define mmCP_HQD_PIPE_PRIORITY                                                  0x324a
+#define mmCP_HQD_QUEUE_PRIORITY                                                 0x324b
+#define mmCP_HQD_QUANTUM                                                        0x324c
+#define mmCP_HQD_PQ_BASE                                                        0x324d
+#define mmCP_HQD_PQ_BASE_HI                                                     0x324e
+#define mmCP_HQD_PQ_RPTR                                                        0x324f
+#define mmCP_HQD_PQ_RPTR_REPORT_ADDR                                            0x3250
+#define mmCP_HQD_PQ_RPTR_REPORT_ADDR_HI                                         0x3251
+#define mmCP_HQD_PQ_WPTR_POLL_ADDR                                              0x3252
+#define mmCP_HQD_PQ_WPTR_POLL_ADDR_HI                                           0x3253
+#define mmCP_HQD_PQ_DOORBELL_CONTROL                                            0x3254
+#define mmCP_HQD_PQ_WPTR                                                        0x3255
+#define mmCP_HQD_PQ_CONTROL                                                     0x3256
+#define mmCP_HQD_IB_BASE_ADDR                                                   0x3257
+#define mmCP_HQD_IB_BASE_ADDR_HI                                                0x3258
+#define mmCP_HQD_IB_RPTR                                                        0x3259
+#define mmCP_HQD_IB_CONTROL                                                     0x325a
+#define mmCP_HQD_IQ_TIMER                                                       0x325b
+#define mmCP_HQD_IQ_RPTR                                                        0x325c
+#define mmCP_HQD_DEQUEUE_REQUEST                                                0x325d
+#define mmCP_HQD_DMA_OFFLOAD                                                    0x325e
+#define mmCP_HQD_OFFLOAD                                                        0x325e
+#define mmCP_HQD_SEMA_CMD                                                       0x325f
+#define mmCP_HQD_MSG_TYPE                                                       0x3260
+#define mmCP_HQD_ATOMIC0_PREOP_LO                                               0x3261
+#define mmCP_HQD_ATOMIC0_PREOP_HI                                               0x3262
+#define mmCP_HQD_ATOMIC1_PREOP_LO                                               0x3263
+#define mmCP_HQD_ATOMIC1_PREOP_HI                                               0x3264
+#define mmCP_HQD_HQ_SCHEDULER0                                                  0x3265
+#define mmCP_HQD_HQ_STATUS0                                                     0x3265
+#define mmCP_HQD_HQ_SCHEDULER1                                                  0x3266
+#define mmCP_HQD_HQ_CONTROL0                                                    0x3266
+#define mmCP_MQD_CONTROL                                                        0x3267
+#define mmCP_HQD_HQ_STATUS1                                                     0x3268
+#define mmCP_HQD_HQ_CONTROL1                                                    0x3269
+#define mmCP_HQD_EOP_BASE_ADDR                                                  0x326a
+#define mmCP_HQD_EOP_BASE_ADDR_HI                                               0x326b
+#define mmCP_HQD_EOP_CONTROL                                                    0x326c
+#define mmCP_HQD_EOP_RPTR                                                       0x326d
+#define mmCP_HQD_EOP_WPTR                                                       0x326e
+#define mmCP_HQD_EOP_EVENTS                                                     0x326f
+#define mmCP_HQD_CTX_SAVE_BASE_ADDR_LO                                          0x3270
+#define mmCP_HQD_CTX_SAVE_BASE_ADDR_HI                                          0x3271
+#define mmCP_HQD_CTX_SAVE_CONTROL                                               0x3272
+#define mmCP_HQD_CNTL_STACK_OFFSET                                              0x3273
+#define mmCP_HQD_CNTL_STACK_SIZE                                                0x3274
+#define mmCP_HQD_WG_STATE_OFFSET                                                0x3275
+#define mmCP_HQD_CTX_SAVE_SIZE                                                  0x3276
+#define mmCP_HQD_GDS_RESOURCE_STATE                                             0x3277
+#define mmCP_HQD_ERROR                                                          0x3278
+#define mmCP_HQD_EOP_WPTR_MEM                                                   0x3279
+#define mmCP_HQD_EOP_DONES                                                      0x327a
+#define mmDB_Z_READ_BASE                                                        0xa012
+#define mmDB_STENCIL_READ_BASE                                                  0xa013
+#define mmDB_Z_WRITE_BASE                                                       0xa014
+#define mmDB_STENCIL_WRITE_BASE                                                 0xa015
+#define mmDB_DEPTH_INFO                                                         0xa00f
+#define mmDB_Z_INFO                                                             0xa010
+#define mmDB_STENCIL_INFO                                                       0xa011
+#define mmDB_DEPTH_SIZE                                                         0xa016
+#define mmDB_DEPTH_SLICE                                                        0xa017
+#define mmDB_DEPTH_VIEW                                                         0xa002
+#define mmDB_RENDER_CONTROL                                                     0xa000
+#define mmDB_COUNT_CONTROL                                                      0xa001
+#define mmDB_RENDER_OVERRIDE                                                    0xa003
+#define mmDB_RENDER_OVERRIDE2                                                   0xa004
+#define mmDB_EQAA                                                               0xa201
+#define mmDB_SHADER_CONTROL                                                     0xa203
+#define mmDB_DEPTH_BOUNDS_MIN                                                   0xa008
+#define mmDB_DEPTH_BOUNDS_MAX                                                   0xa009
+#define mmDB_STENCIL_CLEAR                                                      0xa00a
+#define mmDB_DEPTH_CLEAR                                                        0xa00b
+#define mmDB_HTILE_DATA_BASE                                                    0xa005
+#define mmDB_HTILE_SURFACE                                                      0xa2af
+#define mmDB_PRELOAD_CONTROL                                                    0xa2b2
+#define mmDB_STENCILREFMASK                                                     0xa10c
+#define mmDB_STENCILREFMASK_BF                                                  0xa10d
+#define mmDB_SRESULTS_COMPARE_STATE0                                            0xa2b0
+#define mmDB_SRESULTS_COMPARE_STATE1                                            0xa2b1
+#define mmDB_DEPTH_CONTROL                                                      0xa200
+#define mmDB_STENCIL_CONTROL                                                    0xa10b
+#define mmDB_ALPHA_TO_MASK                                                      0xa2dc
+#define mmDB_PERFCOUNTER0_SELECT                                                0xdc40
+#define mmDB_PERFCOUNTER1_SELECT                                                0xdc42
+#define mmDB_PERFCOUNTER2_SELECT                                                0xdc44
+#define mmDB_PERFCOUNTER3_SELECT                                                0xdc46
+#define mmDB_PERFCOUNTER0_SELECT1                                               0xdc41
+#define mmDB_PERFCOUNTER1_SELECT1                                               0xdc43
+#define mmDB_PERFCOUNTER0_LO                                                    0xd440
+#define mmDB_PERFCOUNTER1_LO                                                    0xd442
+#define mmDB_PERFCOUNTER2_LO                                                    0xd444
+#define mmDB_PERFCOUNTER3_LO                                                    0xd446
+#define mmDB_PERFCOUNTER0_HI                                                    0xd441
+#define mmDB_PERFCOUNTER1_HI                                                    0xd443
+#define mmDB_PERFCOUNTER2_HI                                                    0xd445
+#define mmDB_PERFCOUNTER3_HI                                                    0xd447
+#define mmDB_DEBUG                                                              0x260c
+#define mmDB_DEBUG2                                                             0x260d
+#define mmDB_DEBUG3                                                             0x260e
+#define mmDB_DEBUG4                                                             0x260f
+#define mmDB_CREDIT_LIMIT                                                       0x2614
+#define mmDB_WATERMARKS                                                         0x2615
+#define mmDB_SUBTILE_CONTROL                                                    0x2616
+#define mmDB_FREE_CACHELINES                                                    0x2617
+#define mmDB_FIFO_DEPTH1                                                        0x2618
+#define mmDB_FIFO_DEPTH2                                                        0x2619
+#define mmDB_CGTT_CLK_CTRL_0                                                    0xf0a4
+#define mmDB_ZPASS_COUNT_LOW                                                    0xc3fe
+#define mmDB_ZPASS_COUNT_HI                                                     0xc3ff
+#define mmDB_RING_CONTROL                                                       0x261b
+#define mmDB_READ_DEBUG_0                                                       0x2620
+#define mmDB_READ_DEBUG_1                                                       0x2621
+#define mmDB_READ_DEBUG_2                                                       0x2622
+#define mmDB_READ_DEBUG_3                                                       0x2623
+#define mmDB_READ_DEBUG_4                                                       0x2624
+#define mmDB_READ_DEBUG_5                                                       0x2625
+#define mmDB_READ_DEBUG_6                                                       0x2626
+#define mmDB_READ_DEBUG_7                                                       0x2627
+#define mmDB_READ_DEBUG_8                                                       0x2628
+#define mmDB_READ_DEBUG_9                                                       0x2629
+#define mmDB_READ_DEBUG_A                                                       0x262a
+#define mmDB_READ_DEBUG_B                                                       0x262b
+#define mmDB_READ_DEBUG_C                                                       0x262c
+#define mmDB_READ_DEBUG_D                                                       0x262d
+#define mmDB_READ_DEBUG_E                                                       0x262e
+#define mmDB_READ_DEBUG_F                                                       0x262f
+#define mmDB_OCCLUSION_COUNT0_LOW                                               0xc3c0
+#define mmDB_OCCLUSION_COUNT0_HI                                                0xc3c1
+#define mmDB_OCCLUSION_COUNT1_LOW                                               0xc3c2
+#define mmDB_OCCLUSION_COUNT1_HI                                                0xc3c3
+#define mmDB_OCCLUSION_COUNT2_LOW                                               0xc3c4
+#define mmDB_OCCLUSION_COUNT2_HI                                                0xc3c5
+#define mmDB_OCCLUSION_COUNT3_LOW                                               0xc3c6
+#define mmDB_OCCLUSION_COUNT3_HI                                                0xc3c7
+#define mmCC_RB_REDUNDANCY                                                      0x263c
+#define mmCC_RB_BACKEND_DISABLE                                                 0x263d
+#define mmGC_USER_RB_REDUNDANCY                                                 0x26de
+#define mmGC_USER_RB_BACKEND_DISABLE                                            0x26df
+#define mmGB_ADDR_CONFIG                                                        0x263e
+#define mmGB_BACKEND_MAP                                                        0x263f
+#define mmGB_GPU_ID                                                             0x2640
+#define mmCC_RB_DAISY_CHAIN                                                     0x2641
+#define mmGB_TILE_MODE0                                                         0x2644
+#define mmGB_TILE_MODE1                                                         0x2645
+#define mmGB_TILE_MODE2                                                         0x2646
+#define mmGB_TILE_MODE3                                                         0x2647
+#define mmGB_TILE_MODE4                                                         0x2648
+#define mmGB_TILE_MODE5                                                         0x2649
+#define mmGB_TILE_MODE6                                                         0x264a
+#define mmGB_TILE_MODE7                                                         0x264b
+#define mmGB_TILE_MODE8                                                         0x264c
+#define mmGB_TILE_MODE9                                                         0x264d
+#define mmGB_TILE_MODE10                                                        0x264e
+#define mmGB_TILE_MODE11                                                        0x264f
+#define mmGB_TILE_MODE12                                                        0x2650
+#define mmGB_TILE_MODE13                                                        0x2651
+#define mmGB_TILE_MODE14                                                        0x2652
+#define mmGB_TILE_MODE15                                                        0x2653
+#define mmGB_TILE_MODE16                                                        0x2654
+#define mmGB_TILE_MODE17                                                        0x2655
+#define mmGB_TILE_MODE18                                                        0x2656
+#define mmGB_TILE_MODE19                                                        0x2657
+#define mmGB_TILE_MODE20                                                        0x2658
+#define mmGB_TILE_MODE21                                                        0x2659
+#define mmGB_TILE_MODE22                                                        0x265a
+#define mmGB_TILE_MODE23                                                        0x265b
+#define mmGB_TILE_MODE24                                                        0x265c
+#define mmGB_TILE_MODE25                                                        0x265d
+#define mmGB_TILE_MODE26                                                        0x265e
+#define mmGB_TILE_MODE27                                                        0x265f
+#define mmGB_TILE_MODE28                                                        0x2660
+#define mmGB_TILE_MODE29                                                        0x2661
+#define mmGB_TILE_MODE30                                                        0x2662
+#define mmGB_TILE_MODE31                                                        0x2663
+#define mmGB_MACROTILE_MODE0                                                    0x2664
+#define mmGB_MACROTILE_MODE1                                                    0x2665
+#define mmGB_MACROTILE_MODE2                                                    0x2666
+#define mmGB_MACROTILE_MODE3                                                    0x2667
+#define mmGB_MACROTILE_MODE4                                                    0x2668
+#define mmGB_MACROTILE_MODE5                                                    0x2669
+#define mmGB_MACROTILE_MODE6                                                    0x266a
+#define mmGB_MACROTILE_MODE7                                                    0x266b
+#define mmGB_MACROTILE_MODE8                                                    0x266c
+#define mmGB_MACROTILE_MODE9                                                    0x266d
+#define mmGB_MACROTILE_MODE10                                                   0x266e
+#define mmGB_MACROTILE_MODE11                                                   0x266f
+#define mmGB_MACROTILE_MODE12                                                   0x2670
+#define mmGB_MACROTILE_MODE13                                                   0x2671
+#define mmGB_MACROTILE_MODE14                                                   0x2672
+#define mmGB_MACROTILE_MODE15                                                   0x2673
+#define mmGB_EDC_MODE                                                           0x307e
+#define mmCC_GC_EDC_CONFIG                                                      0x3098
+#define mmRAS_SIGNATURE_CONTROL                                                 0x3380
+#define mmRAS_SIGNATURE_MASK                                                    0x3381
+#define mmRAS_SX_SIGNATURE0                                                     0x3382
+#define mmRAS_SX_SIGNATURE1                                                     0x3383
+#define mmRAS_SX_SIGNATURE2                                                     0x3384
+#define mmRAS_SX_SIGNATURE3                                                     0x3385
+#define mmRAS_DB_SIGNATURE0                                                     0x338b
+#define mmRAS_PA_SIGNATURE0                                                     0x338c
+#define mmRAS_VGT_SIGNATURE0                                                    0x338d
+#define mmRAS_SC_SIGNATURE0                                                     0x338f
+#define mmRAS_SC_SIGNATURE1                                                     0x3390
+#define mmRAS_SC_SIGNATURE2                                                     0x3391
+#define mmRAS_SC_SIGNATURE3                                                     0x3392
+#define mmRAS_SC_SIGNATURE4                                                     0x3393
+#define mmRAS_SC_SIGNATURE5                                                     0x3394
+#define mmRAS_SC_SIGNATURE6                                                     0x3395
+#define mmRAS_SC_SIGNATURE7                                                     0x3396
+#define mmRAS_IA_SIGNATURE0                                                     0x3397
+#define mmRAS_IA_SIGNATURE1                                                     0x3398
+#define mmRAS_SPI_SIGNATURE0                                                    0x3399
+#define mmRAS_SPI_SIGNATURE1                                                    0x339a
+#define mmRAS_TA_SIGNATURE0                                                     0x339b
+#define mmRAS_TD_SIGNATURE0                                                     0x339c
+#define mmRAS_CB_SIGNATURE0                                                     0x339d
+#define mmRAS_BCI_SIGNATURE0                                                    0x339e
+#define mmRAS_BCI_SIGNATURE1                                                    0x339f
+#define mmRAS_TA_SIGNATURE1                                                     0x33a0
+#define mmGRBM_HYP_CAM_INDEX                                                    0xf83e
+#define mmGRBM_CAM_INDEX                                                        0xf83e
+#define mmGRBM_HYP_CAM_DATA                                                     0xf83f
+#define mmGRBM_CAM_DATA                                                         0xf83f
+#define mmGRBM_CNTL                                                             0x2000
+#define mmGRBM_SKEW_CNTL                                                        0x2001
+#define mmGRBM_PWR_CNTL                                                         0x2003
+#define mmGRBM_STATUS                                                           0x2004
+#define mmGRBM_STATUS2                                                          0x2002
+#define mmGRBM_STATUS_SE0                                                       0x2005
+#define mmGRBM_STATUS_SE1                                                       0x2006
+#define mmGRBM_STATUS_SE2                                                       0x200e
+#define mmGRBM_STATUS_SE3                                                       0x200f
+#define mmGRBM_SOFT_RESET                                                       0x2008
+#define mmGRBM_DEBUG_CNTL                                                       0x2009
+#define mmGRBM_DEBUG_DATA                                                       0x200a
+#define mmGRBM_CGTT_CLK_CNTL                                                    0x200b
+#define mmGRBM_GFX_INDEX                                                        0xc200
+#define mmGRBM_GFX_CLKEN_CNTL                                                   0x200c
+#define mmGRBM_WAIT_IDLE_CLOCKS                                                 0x200d
+#define mmGRBM_DEBUG                                                            0x2014
+#define mmGRBM_DEBUG_SNAPSHOT                                                   0x2015
+#define mmGRBM_READ_ERROR                                                       0x2016
+#define mmGRBM_READ_ERROR2                                                      0x2017
+#define mmGRBM_INT_CNTL                                                         0x2018
+#define mmGRBM_TRAP_OP                                                          0x2019
+#define mmGRBM_TRAP_ADDR                                                        0x201a
+#define mmGRBM_TRAP_ADDR_MSK                                                    0x201b
+#define mmGRBM_TRAP_WD                                                          0x201c
+#define mmGRBM_TRAP_WD_MSK                                                      0x201d
+#define mmGRBM_DSM_BYPASS                                                       0x201e
+#define mmGRBM_WRITE_ERROR                                                      0x201f
+#define mmGRBM_PERFCOUNTER0_SELECT                                              0xd840
+#define mmGRBM_PERFCOUNTER1_SELECT                                              0xd841
+#define mmGRBM_SE0_PERFCOUNTER_SELECT                                           0xd842
+#define mmGRBM_SE1_PERFCOUNTER_SELECT                                           0xd843
+#define mmGRBM_SE2_PERFCOUNTER_SELECT                                           0xd844
+#define mmGRBM_SE3_PERFCOUNTER_SELECT                                           0xd845
+#define mmGRBM_PERFCOUNTER0_LO                                                  0xd040
+#define mmGRBM_PERFCOUNTER0_HI                                                  0xd041
+#define mmGRBM_PERFCOUNTER1_LO                                                  0xd043
+#define mmGRBM_PERFCOUNTER1_HI                                                  0xd044
+#define mmGRBM_SE0_PERFCOUNTER_LO                                               0xd045
+#define mmGRBM_SE0_PERFCOUNTER_HI                                               0xd046
+#define mmGRBM_SE1_PERFCOUNTER_LO                                               0xd047
+#define mmGRBM_SE1_PERFCOUNTER_HI                                               0xd048
+#define mmGRBM_SE2_PERFCOUNTER_LO                                               0xd049
+#define mmGRBM_SE2_PERFCOUNTER_HI                                               0xd04a
+#define mmGRBM_SE3_PERFCOUNTER_LO                                               0xd04b
+#define mmGRBM_SE3_PERFCOUNTER_HI                                               0xd04c
+#define mmGRBM_SCRATCH_REG0                                                     0x2040
+#define mmGRBM_SCRATCH_REG1                                                     0x2041
+#define mmGRBM_SCRATCH_REG2                                                     0x2042
+#define mmGRBM_SCRATCH_REG3                                                     0x2043
+#define mmGRBM_SCRATCH_REG4                                                     0x2044
+#define mmGRBM_SCRATCH_REG5                                                     0x2045
+#define mmGRBM_SCRATCH_REG6                                                     0x2046
+#define mmGRBM_SCRATCH_REG7                                                     0x2047
+#define mmDEBUG_INDEX                                                           0x203c
+#define mmDEBUG_DATA                                                            0x203d
+#define mmGRBM_NOWHERE                                                          0x203f
+#define mmPA_CL_VPORT_XSCALE                                                    0xa10f
+#define mmPA_CL_VPORT_XOFFSET                                                   0xa110
+#define mmPA_CL_VPORT_YSCALE                                                    0xa111
+#define mmPA_CL_VPORT_YOFFSET                                                   0xa112
+#define mmPA_CL_VPORT_ZSCALE                                                    0xa113
+#define mmPA_CL_VPORT_ZOFFSET                                                   0xa114
+#define mmPA_CL_VPORT_XSCALE_1                                                  0xa115
+#define mmPA_CL_VPORT_XSCALE_2                                                  0xa11b
+#define mmPA_CL_VPORT_XSCALE_3                                                  0xa121
+#define mmPA_CL_VPORT_XSCALE_4                                                  0xa127
+#define mmPA_CL_VPORT_XSCALE_5                                                  0xa12d
+#define mmPA_CL_VPORT_XSCALE_6                                                  0xa133
+#define mmPA_CL_VPORT_XSCALE_7                                                  0xa139
+#define mmPA_CL_VPORT_XSCALE_8                                                  0xa13f
+#define mmPA_CL_VPORT_XSCALE_9                                                  0xa145
+#define mmPA_CL_VPORT_XSCALE_10                                                 0xa14b
+#define mmPA_CL_VPORT_XSCALE_11                                                 0xa151
+#define mmPA_CL_VPORT_XSCALE_12                                                 0xa157
+#define mmPA_CL_VPORT_XSCALE_13                                                 0xa15d
+#define mmPA_CL_VPORT_XSCALE_14                                                 0xa163
+#define mmPA_CL_VPORT_XSCALE_15                                                 0xa169
+#define mmPA_CL_VPORT_XOFFSET_1                                                 0xa116
+#define mmPA_CL_VPORT_XOFFSET_2                                                 0xa11c
+#define mmPA_CL_VPORT_XOFFSET_3                                                 0xa122
+#define mmPA_CL_VPORT_XOFFSET_4                                                 0xa128
+#define mmPA_CL_VPORT_XOFFSET_5                                                 0xa12e
+#define mmPA_CL_VPORT_XOFFSET_6                                                 0xa134
+#define mmPA_CL_VPORT_XOFFSET_7                                                 0xa13a
+#define mmPA_CL_VPORT_XOFFSET_8                                                 0xa140
+#define mmPA_CL_VPORT_XOFFSET_9                                                 0xa146
+#define mmPA_CL_VPORT_XOFFSET_10                                                0xa14c
+#define mmPA_CL_VPORT_XOFFSET_11                                                0xa152
+#define mmPA_CL_VPORT_XOFFSET_12                                                0xa158
+#define mmPA_CL_VPORT_XOFFSET_13                                                0xa15e
+#define mmPA_CL_VPORT_XOFFSET_14                                                0xa164
+#define mmPA_CL_VPORT_XOFFSET_15                                                0xa16a
+#define mmPA_CL_VPORT_YSCALE_1                                                  0xa117
+#define mmPA_CL_VPORT_YSCALE_2                                                  0xa11d
+#define mmPA_CL_VPORT_YSCALE_3                                                  0xa123
+#define mmPA_CL_VPORT_YSCALE_4                                                  0xa129
+#define mmPA_CL_VPORT_YSCALE_5                                                  0xa12f
+#define mmPA_CL_VPORT_YSCALE_6                                                  0xa135
+#define mmPA_CL_VPORT_YSCALE_7                                                  0xa13b
+#define mmPA_CL_VPORT_YSCALE_8                                                  0xa141
+#define mmPA_CL_VPORT_YSCALE_9                                                  0xa147
+#define mmPA_CL_VPORT_YSCALE_10                                                 0xa14d
+#define mmPA_CL_VPORT_YSCALE_11                                                 0xa153
+#define mmPA_CL_VPORT_YSCALE_12                                                 0xa159
+#define mmPA_CL_VPORT_YSCALE_13                                                 0xa15f
+#define mmPA_CL_VPORT_YSCALE_14                                                 0xa165
+#define mmPA_CL_VPORT_YSCALE_15                                                 0xa16b
+#define mmPA_CL_VPORT_YOFFSET_1                                                 0xa118
+#define mmPA_CL_VPORT_YOFFSET_2                                                 0xa11e
+#define mmPA_CL_VPORT_YOFFSET_3                                                 0xa124
+#define mmPA_CL_VPORT_YOFFSET_4                                                 0xa12a
+#define mmPA_CL_VPORT_YOFFSET_5                                                 0xa130
+#define mmPA_CL_VPORT_YOFFSET_6                                                 0xa136
+#define mmPA_CL_VPORT_YOFFSET_7                                                 0xa13c
+#define mmPA_CL_VPORT_YOFFSET_8                                                 0xa142
+#define mmPA_CL_VPORT_YOFFSET_9                                                 0xa148
+#define mmPA_CL_VPORT_YOFFSET_10                                                0xa14e
+#define mmPA_CL_VPORT_YOFFSET_11                                                0xa154
+#define mmPA_CL_VPORT_YOFFSET_12                                                0xa15a
+#define mmPA_CL_VPORT_YOFFSET_13                                                0xa160
+#define mmPA_CL_VPORT_YOFFSET_14                                                0xa166
+#define mmPA_CL_VPORT_YOFFSET_15                                                0xa16c
+#define mmPA_CL_VPORT_ZSCALE_1                                                  0xa119
+#define mmPA_CL_VPORT_ZSCALE_2                                                  0xa11f
+#define mmPA_CL_VPORT_ZSCALE_3                                                  0xa125
+#define mmPA_CL_VPORT_ZSCALE_4                                                  0xa12b
+#define mmPA_CL_VPORT_ZSCALE_5                                                  0xa131
+#define mmPA_CL_VPORT_ZSCALE_6                                                  0xa137
+#define mmPA_CL_VPORT_ZSCALE_7                                                  0xa13d
+#define mmPA_CL_VPORT_ZSCALE_8                                                  0xa143
+#define mmPA_CL_VPORT_ZSCALE_9                                                  0xa149
+#define mmPA_CL_VPORT_ZSCALE_10                                                 0xa14f
+#define mmPA_CL_VPORT_ZSCALE_11                                                 0xa155
+#define mmPA_CL_VPORT_ZSCALE_12                                                 0xa15b
+#define mmPA_CL_VPORT_ZSCALE_13                                                 0xa161
+#define mmPA_CL_VPORT_ZSCALE_14                                                 0xa167
+#define mmPA_CL_VPORT_ZSCALE_15                                                 0xa16d
+#define mmPA_CL_VPORT_ZOFFSET_1                                                 0xa11a
+#define mmPA_CL_VPORT_ZOFFSET_2                                                 0xa120
+#define mmPA_CL_VPORT_ZOFFSET_3                                                 0xa126
+#define mmPA_CL_VPORT_ZOFFSET_4                                                 0xa12c
+#define mmPA_CL_VPORT_ZOFFSET_5                                                 0xa132
+#define mmPA_CL_VPORT_ZOFFSET_6                                                 0xa138
+#define mmPA_CL_VPORT_ZOFFSET_7                                                 0xa13e
+#define mmPA_CL_VPORT_ZOFFSET_8                                                 0xa144
+#define mmPA_CL_VPORT_ZOFFSET_9                                                 0xa14a
+#define mmPA_CL_VPORT_ZOFFSET_10                                                0xa150
+#define mmPA_CL_VPORT_ZOFFSET_11                                                0xa156
+#define mmPA_CL_VPORT_ZOFFSET_12                                                0xa15c
+#define mmPA_CL_VPORT_ZOFFSET_13                                                0xa162
+#define mmPA_CL_VPORT_ZOFFSET_14                                                0xa168
+#define mmPA_CL_VPORT_ZOFFSET_15                                                0xa16e
+#define mmPA_CL_VTE_CNTL                                                        0xa206
+#define mmPA_CL_VS_OUT_CNTL                                                     0xa207
+#define mmPA_CL_NANINF_CNTL                                                     0xa208
+#define mmPA_CL_CLIP_CNTL                                                       0xa204
+#define mmPA_CL_GB_VERT_CLIP_ADJ                                                0xa2fa
+#define mmPA_CL_GB_VERT_DISC_ADJ                                                0xa2fb
+#define mmPA_CL_GB_HORZ_CLIP_ADJ                                                0xa2fc
+#define mmPA_CL_GB_HORZ_DISC_ADJ                                                0xa2fd
+#define mmPA_CL_UCP_0_X                                                         0xa16f
+#define mmPA_CL_UCP_0_Y                                                         0xa170
+#define mmPA_CL_UCP_0_Z                                                         0xa171
+#define mmPA_CL_UCP_0_W                                                         0xa172
+#define mmPA_CL_UCP_1_X                                                         0xa173
+#define mmPA_CL_UCP_1_Y                                                         0xa174
+#define mmPA_CL_UCP_1_Z                                                         0xa175
+#define mmPA_CL_UCP_1_W                                                         0xa176
+#define mmPA_CL_UCP_2_X                                                         0xa177
+#define mmPA_CL_UCP_2_Y                                                         0xa178
+#define mmPA_CL_UCP_2_Z                                                         0xa179
+#define mmPA_CL_UCP_2_W                                                         0xa17a
+#define mmPA_CL_UCP_3_X                                                         0xa17b
+#define mmPA_CL_UCP_3_Y                                                         0xa17c
+#define mmPA_CL_UCP_3_Z                                                         0xa17d
+#define mmPA_CL_UCP_3_W                                                         0xa17e
+#define mmPA_CL_UCP_4_X                                                         0xa17f
+#define mmPA_CL_UCP_4_Y                                                         0xa180
+#define mmPA_CL_UCP_4_Z                                                         0xa181
+#define mmPA_CL_UCP_4_W                                                         0xa182
+#define mmPA_CL_UCP_5_X                                                         0xa183
+#define mmPA_CL_UCP_5_Y                                                         0xa184
+#define mmPA_CL_UCP_5_Z                                                         0xa185
+#define mmPA_CL_UCP_5_W                                                         0xa186
+#define mmPA_CL_POINT_X_RAD                                                     0xa1f5
+#define mmPA_CL_POINT_Y_RAD                                                     0xa1f6
+#define mmPA_CL_POINT_SIZE                                                      0xa1f7
+#define mmPA_CL_POINT_CULL_RAD                                                  0xa1f8
+#define mmPA_CL_ENHANCE                                                         0x2285
+#define mmPA_CL_RESET_DEBUG                                                     0x2286
+#define mmPA_SU_VTX_CNTL                                                        0xa2f9
+#define mmPA_SU_POINT_SIZE                                                      0xa280
+#define mmPA_SU_POINT_MINMAX                                                    0xa281
+#define mmPA_SU_LINE_CNTL                                                       0xa282
+#define mmPA_SU_LINE_STIPPLE_CNTL                                               0xa209
+#define mmPA_SU_LINE_STIPPLE_SCALE                                              0xa20a
+#define mmPA_SU_PRIM_FILTER_CNTL                                                0xa20b
+#define mmPA_SU_SC_MODE_CNTL                                                    0xa205
+#define mmPA_SU_POLY_OFFSET_DB_FMT_CNTL                                         0xa2de
+#define mmPA_SU_POLY_OFFSET_CLAMP                                               0xa2df
+#define mmPA_SU_POLY_OFFSET_FRONT_SCALE                                         0xa2e0
+#define mmPA_SU_POLY_OFFSET_FRONT_OFFSET                                        0xa2e1
+#define mmPA_SU_POLY_OFFSET_BACK_SCALE                                          0xa2e2
+#define mmPA_SU_POLY_OFFSET_BACK_OFFSET                                         0xa2e3
+#define mmPA_SU_HARDWARE_SCREEN_OFFSET                                          0xa08d
+#define mmPA_SU_LINE_STIPPLE_VALUE                                              0xc280
+#define mmPA_SU_PERFCOUNTER0_SELECT                                             0xd900
+#define mmPA_SU_PERFCOUNTER0_SELECT1                                            0xd901
+#define mmPA_SU_PERFCOUNTER1_SELECT                                             0xd902
+#define mmPA_SU_PERFCOUNTER1_SELECT1                                            0xd903
+#define mmPA_SU_PERFCOUNTER2_SELECT                                             0xd904
+#define mmPA_SU_PERFCOUNTER3_SELECT                                             0xd905
+#define mmPA_SU_PERFCOUNTER0_LO                                                 0xd100
+#define mmPA_SU_PERFCOUNTER0_HI                                                 0xd101
+#define mmPA_SU_PERFCOUNTER1_LO                                                 0xd102
+#define mmPA_SU_PERFCOUNTER1_HI                                                 0xd103
+#define mmPA_SU_PERFCOUNTER2_LO                                                 0xd104
+#define mmPA_SU_PERFCOUNTER2_HI                                                 0xd105
+#define mmPA_SU_PERFCOUNTER3_LO                                                 0xd106
+#define mmPA_SU_PERFCOUNTER3_HI                                                 0xd107
+#define mmPA_SC_AA_CONFIG                                                       0xa2f8
+#define mmPA_SC_AA_MASK_X0Y0_X1Y0                                               0xa30e
+#define mmPA_SC_AA_MASK_X0Y1_X1Y1                                               0xa30f
+#define mmPA_SC_SHADER_CONTROL                                                  0xa310
+#define mmPA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_0                                     0xa2fe
+#define mmPA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_1                                     0xa2ff
+#define mmPA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_2                                     0xa300
+#define mmPA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_3                                     0xa301
+#define mmPA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_0                                     0xa302
+#define mmPA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_1                                     0xa303
+#define mmPA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_2                                     0xa304
+#define mmPA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_3                                     0xa305
+#define mmPA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_0                                     0xa306
+#define mmPA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_1                                     0xa307
+#define mmPA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_2                                     0xa308
+#define mmPA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_3                                     0xa309
+#define mmPA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_0                                     0xa30a
+#define mmPA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_1                                     0xa30b
+#define mmPA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_2                                     0xa30c
+#define mmPA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_3                                     0xa30d
+#define mmPA_SC_CENTROID_PRIORITY_0                                             0xa2f5
+#define mmPA_SC_CENTROID_PRIORITY_1                                             0xa2f6
+#define mmPA_SC_CLIPRECT_0_TL                                                   0xa084
+#define mmPA_SC_CLIPRECT_0_BR                                                   0xa085
+#define mmPA_SC_CLIPRECT_1_TL                                                   0xa086
+#define mmPA_SC_CLIPRECT_1_BR                                                   0xa087
+#define mmPA_SC_CLIPRECT_2_TL                                                   0xa088
+#define mmPA_SC_CLIPRECT_2_BR                                                   0xa089
+#define mmPA_SC_CLIPRECT_3_TL                                                   0xa08a
+#define mmPA_SC_CLIPRECT_3_BR                                                   0xa08b
+#define mmPA_SC_CLIPRECT_RULE                                                   0xa083
+#define mmPA_SC_EDGERULE                                                        0xa08c
+#define mmPA_SC_LINE_CNTL                                                       0xa2f7
+#define mmPA_SC_LINE_STIPPLE                                                    0xa283
+#define mmPA_SC_MODE_CNTL_0                                                     0xa292
+#define mmPA_SC_MODE_CNTL_1                                                     0xa293
+#define mmPA_SC_RASTER_CONFIG                                                   0xa0d4
+#define mmPA_SC_RASTER_CONFIG_1                                                 0xa0d5
+#define mmPA_SC_SCREEN_EXTENT_CONTROL                                           0xa0d6
+#define mmPA_SC_GENERIC_SCISSOR_TL                                              0xa090
+#define mmPA_SC_GENERIC_SCISSOR_BR                                              0xa091
+#define mmPA_SC_SCREEN_SCISSOR_TL                                               0xa00c
+#define mmPA_SC_SCREEN_SCISSOR_BR                                               0xa00d
+#define mmPA_SC_WINDOW_OFFSET                                                   0xa080
+#define mmPA_SC_WINDOW_SCISSOR_TL                                               0xa081
+#define mmPA_SC_WINDOW_SCISSOR_BR                                               0xa082
+#define mmPA_SC_VPORT_SCISSOR_0_TL                                              0xa094
+#define mmPA_SC_VPORT_SCISSOR_1_TL                                              0xa096
+#define mmPA_SC_VPORT_SCISSOR_2_TL                                              0xa098
+#define mmPA_SC_VPORT_SCISSOR_3_TL                                              0xa09a
+#define mmPA_SC_VPORT_SCISSOR_4_TL                                              0xa09c
+#define mmPA_SC_VPORT_SCISSOR_5_TL                                              0xa09e
+#define mmPA_SC_VPORT_SCISSOR_6_TL                                              0xa0a0
+#define mmPA_SC_VPORT_SCISSOR_7_TL                                              0xa0a2
+#define mmPA_SC_VPORT_SCISSOR_8_TL                                              0xa0a4
+#define mmPA_SC_VPORT_SCISSOR_9_TL                                              0xa0a6
+#define mmPA_SC_VPORT_SCISSOR_10_TL                                             0xa0a8
+#define mmPA_SC_VPORT_SCISSOR_11_TL                                             0xa0aa
+#define mmPA_SC_VPORT_SCISSOR_12_TL                                             0xa0ac
+#define mmPA_SC_VPORT_SCISSOR_13_TL                                             0xa0ae
+#define mmPA_SC_VPORT_SCISSOR_14_TL                                             0xa0b0
+#define mmPA_SC_VPORT_SCISSOR_15_TL                                             0xa0b2
+#define mmPA_SC_VPORT_SCISSOR_0_BR                                              0xa095
+#define mmPA_SC_VPORT_SCISSOR_1_BR                                              0xa097
+#define mmPA_SC_VPORT_SCISSOR_2_BR                                              0xa099
+#define mmPA_SC_VPORT_SCISSOR_3_BR                                              0xa09b
+#define mmPA_SC_VPORT_SCISSOR_4_BR                                              0xa09d
+#define mmPA_SC_VPORT_SCISSOR_5_BR                                              0xa09f
+#define mmPA_SC_VPORT_SCISSOR_6_BR                                              0xa0a1
+#define mmPA_SC_VPORT_SCISSOR_7_BR                                              0xa0a3
+#define mmPA_SC_VPORT_SCISSOR_8_BR                                              0xa0a5
+#define mmPA_SC_VPORT_SCISSOR_9_BR                                              0xa0a7
+#define mmPA_SC_VPORT_SCISSOR_10_BR                                             0xa0a9
+#define mmPA_SC_VPORT_SCISSOR_11_BR                                             0xa0ab
+#define mmPA_SC_VPORT_SCISSOR_12_BR                                             0xa0ad
+#define mmPA_SC_VPORT_SCISSOR_13_BR                                             0xa0af
+#define mmPA_SC_VPORT_SCISSOR_14_BR                                             0xa0b1
+#define mmPA_SC_VPORT_SCISSOR_15_BR                                             0xa0b3
+#define mmPA_SC_VPORT_ZMIN_0                                                    0xa0b4
+#define mmPA_SC_VPORT_ZMIN_1                                                    0xa0b6
+#define mmPA_SC_VPORT_ZMIN_2                                                    0xa0b8
+#define mmPA_SC_VPORT_ZMIN_3                                                    0xa0ba
+#define mmPA_SC_VPORT_ZMIN_4                                                    0xa0bc
+#define mmPA_SC_VPORT_ZMIN_5                                                    0xa0be
+#define mmPA_SC_VPORT_ZMIN_6                                                    0xa0c0
+#define mmPA_SC_VPORT_ZMIN_7                                                    0xa0c2
+#define mmPA_SC_VPORT_ZMIN_8                                                    0xa0c4
+#define mmPA_SC_VPORT_ZMIN_9                                                    0xa0c6
+#define mmPA_SC_VPORT_ZMIN_10                                                   0xa0c8
+#define mmPA_SC_VPORT_ZMIN_11                                                   0xa0ca
+#define mmPA_SC_VPORT_ZMIN_12                                                   0xa0cc
+#define mmPA_SC_VPORT_ZMIN_13                                                   0xa0ce
+#define mmPA_SC_VPORT_ZMIN_14                                                   0xa0d0
+#define mmPA_SC_VPORT_ZMIN_15                                                   0xa0d2
+#define mmPA_SC_VPORT_ZMAX_0                                                    0xa0b5
+#define mmPA_SC_VPORT_ZMAX_1                                                    0xa0b7
+#define mmPA_SC_VPORT_ZMAX_2                                                    0xa0b9
+#define mmPA_SC_VPORT_ZMAX_3                                                    0xa0bb
+#define mmPA_SC_VPORT_ZMAX_4                                                    0xa0bd
+#define mmPA_SC_VPORT_ZMAX_5                                                    0xa0bf
+#define mmPA_SC_VPORT_ZMAX_6                                                    0xa0c1
+#define mmPA_SC_VPORT_ZMAX_7                                                    0xa0c3
+#define mmPA_SC_VPORT_ZMAX_8                                                    0xa0c5
+#define mmPA_SC_VPORT_ZMAX_9                                                    0xa0c7
+#define mmPA_SC_VPORT_ZMAX_10                                                   0xa0c9
+#define mmPA_SC_VPORT_ZMAX_11                                                   0xa0cb
+#define mmPA_SC_VPORT_ZMAX_12                                                   0xa0cd
+#define mmPA_SC_VPORT_ZMAX_13                                                   0xa0cf
+#define mmPA_SC_VPORT_ZMAX_14                                                   0xa0d1
+#define mmPA_SC_VPORT_ZMAX_15                                                   0xa0d3
+#define mmPA_SC_ENHANCE                                                         0x22fc
+#define mmPA_SC_ENHANCE_1                                                       0x22fd
+#define mmPA_SC_DSM_CNTL                                                        0x22fe
+#define mmPA_SC_FIFO_SIZE                                                       0x22f3
+#define mmPA_SC_IF_FIFO_SIZE                                                    0x22f5
+#define mmPA_SC_FORCE_EOV_MAX_CNTS                                              0x22c9
+#define mmPA_SC_LINE_STIPPLE_STATE                                              0xc281
+#define mmPA_SC_SCREEN_EXTENT_MIN_0                                             0xc284
+#define mmPA_SC_SCREEN_EXTENT_MAX_0                                             0xc285
+#define mmPA_SC_SCREEN_EXTENT_MIN_1                                             0xc286
+#define mmPA_SC_SCREEN_EXTENT_MAX_1                                             0xc28b
+#define mmPA_SC_PERFCOUNTER0_SELECT                                             0xd940
+#define mmPA_SC_PERFCOUNTER0_SELECT1                                            0xd941
+#define mmPA_SC_PERFCOUNTER1_SELECT                                             0xd942
+#define mmPA_SC_PERFCOUNTER2_SELECT                                             0xd943
+#define mmPA_SC_PERFCOUNTER3_SELECT                                             0xd944
+#define mmPA_SC_PERFCOUNTER4_SELECT                                             0xd945
+#define mmPA_SC_PERFCOUNTER5_SELECT                                             0xd946
+#define mmPA_SC_PERFCOUNTER6_SELECT                                             0xd947
+#define mmPA_SC_PERFCOUNTER7_SELECT                                             0xd948
+#define mmPA_SC_PERFCOUNTER0_LO                                                 0xd140
+#define mmPA_SC_PERFCOUNTER0_HI                                                 0xd141
+#define mmPA_SC_PERFCOUNTER1_LO                                                 0xd142
+#define mmPA_SC_PERFCOUNTER1_HI                                                 0xd143
+#define mmPA_SC_PERFCOUNTER2_LO                                                 0xd144
+#define mmPA_SC_PERFCOUNTER2_HI                                                 0xd145
+#define mmPA_SC_PERFCOUNTER3_LO                                                 0xd146
+#define mmPA_SC_PERFCOUNTER3_HI                                                 0xd147
+#define mmPA_SC_PERFCOUNTER4_LO                                                 0xd148
+#define mmPA_SC_PERFCOUNTER4_HI                                                 0xd149
+#define mmPA_SC_PERFCOUNTER5_LO                                                 0xd14a
+#define mmPA_SC_PERFCOUNTER5_HI                                                 0xd14b
+#define mmPA_SC_PERFCOUNTER6_LO                                                 0xd14c
+#define mmPA_SC_PERFCOUNTER6_HI                                                 0xd14d
+#define mmPA_SC_PERFCOUNTER7_LO                                                 0xd14e
+#define mmPA_SC_PERFCOUNTER7_HI                                                 0xd14f
+#define mmPA_SC_P3D_TRAP_SCREEN_HV_EN                                           0xc2a0
+#define mmPA_SC_P3D_TRAP_SCREEN_H                                               0xc2a1
+#define mmPA_SC_P3D_TRAP_SCREEN_V                                               0xc2a2
+#define mmPA_SC_P3D_TRAP_SCREEN_OCCURRENCE                                      0xc2a3
+#define mmPA_SC_P3D_TRAP_SCREEN_COUNT                                           0xc2a4
+#define mmPA_SC_HP3D_TRAP_SCREEN_HV_EN                                          0xc2a8
+#define mmPA_SC_HP3D_TRAP_SCREEN_H                                              0xc2a9
+#define mmPA_SC_HP3D_TRAP_SCREEN_V                                              0xc2aa
+#define mmPA_SC_HP3D_TRAP_SCREEN_OCCURRENCE                                     0xc2ab
+#define mmPA_SC_HP3D_TRAP_SCREEN_COUNT                                          0xc2ac
+#define mmPA_SC_TRAP_SCREEN_HV_EN                                               0xc2b0
+#define mmPA_SC_TRAP_SCREEN_H                                                   0xc2b1
+#define mmPA_SC_TRAP_SCREEN_V                                                   0xc2b2
+#define mmPA_SC_TRAP_SCREEN_OCCURRENCE                                          0xc2b3
+#define mmPA_SC_TRAP_SCREEN_COUNT                                               0xc2b4
+#define mmPA_SC_P3D_TRAP_SCREEN_HV_LOCK                                         0x22c0
+#define mmPA_SC_HP3D_TRAP_SCREEN_HV_LOCK                                        0x22c1
+#define mmPA_SC_TRAP_SCREEN_HV_LOCK                                             0x22c2
+#define mmPA_CL_CNTL_STATUS                                                     0x2284
+#define mmPA_SU_CNTL_STATUS                                                     0x2294
+#define mmPA_SC_FIFO_DEPTH_CNTL                                                 0x2295
+#define mmCGTT_PA_CLK_CTRL                                                      0xf088
+#define mmCGTT_SC_CLK_CTRL                                                      0xf089
+#define mmPA_SU_DEBUG_CNTL                                                      0x2280
+#define mmPA_SU_DEBUG_DATA                                                      0x2281
+#define mmPA_SC_DEBUG_CNTL                                                      0x22f6
+#define mmPA_SC_DEBUG_DATA                                                      0x22f7
+#define ixCLIPPER_DEBUG_REG00                                                   0x0
+#define ixCLIPPER_DEBUG_REG01                                                   0x1
+#define ixCLIPPER_DEBUG_REG02                                                   0x2
+#define ixCLIPPER_DEBUG_REG03                                                   0x3
+#define ixCLIPPER_DEBUG_REG04                                                   0x4
+#define ixCLIPPER_DEBUG_REG05                                                   0x5
+#define ixCLIPPER_DEBUG_REG06                                                   0x6
+#define ixCLIPPER_DEBUG_REG07                                                   0x7
+#define ixCLIPPER_DEBUG_REG08                                                   0x8
+#define ixCLIPPER_DEBUG_REG09                                                   0x9
+#define ixCLIPPER_DEBUG_REG10                                                   0xa
+#define ixCLIPPER_DEBUG_REG11                                                   0xb
+#define ixCLIPPER_DEBUG_REG12                                                   0xc
+#define ixCLIPPER_DEBUG_REG13                                                   0xd
+#define ixCLIPPER_DEBUG_REG14                                                   0xe
+#define ixCLIPPER_DEBUG_REG15                                                   0xf
+#define ixCLIPPER_DEBUG_REG16                                                   0x10
+#define ixCLIPPER_DEBUG_REG17                                                   0x11
+#define ixCLIPPER_DEBUG_REG18                                                   0x12
+#define ixCLIPPER_DEBUG_REG19                                                   0x13
+#define ixSXIFCCG_DEBUG_REG0                                                    0x14
+#define ixSXIFCCG_DEBUG_REG1                                                    0x15
+#define ixSXIFCCG_DEBUG_REG2                                                    0x16
+#define ixSXIFCCG_DEBUG_REG3                                                    0x17
+#define ixSETUP_DEBUG_REG0                                                      0x18
+#define ixSETUP_DEBUG_REG1                                                      0x19
+#define ixSETUP_DEBUG_REG2                                                      0x1a
+#define ixSETUP_DEBUG_REG3                                                      0x1b
+#define ixSETUP_DEBUG_REG4                                                      0x1c
+#define ixSETUP_DEBUG_REG5                                                      0x1d
+#define ixPA_SC_DEBUG_REG0                                                      0x0
+#define ixPA_SC_DEBUG_REG1                                                      0x1
+#define mmCOMPUTE_DISPATCH_INITIATOR                                            0x2e00
+#define mmCOMPUTE_DIM_X                                                         0x2e01
+#define mmCOMPUTE_DIM_Y                                                         0x2e02
+#define mmCOMPUTE_DIM_Z                                                         0x2e03
+#define mmCOMPUTE_START_X                                                       0x2e04
+#define mmCOMPUTE_START_Y                                                       0x2e05
+#define mmCOMPUTE_START_Z                                                       0x2e06
+#define mmCOMPUTE_NUM_THREAD_X                                                  0x2e07
+#define mmCOMPUTE_NUM_THREAD_Y                                                  0x2e08
+#define mmCOMPUTE_NUM_THREAD_Z                                                  0x2e09
+#define mmCOMPUTE_PIPELINESTAT_ENABLE                                           0x2e0a
+#define mmCOMPUTE_PERFCOUNT_ENABLE                                              0x2e0b
+#define mmCOMPUTE_PGM_LO                                                        0x2e0c
+#define mmCOMPUTE_PGM_HI                                                        0x2e0d
+#define mmCOMPUTE_TBA_LO                                                        0x2e0e
+#define mmCOMPUTE_TBA_HI                                                        0x2e0f
+#define mmCOMPUTE_TMA_LO                                                        0x2e10
+#define mmCOMPUTE_TMA_HI                                                        0x2e11
+#define mmCOMPUTE_PGM_RSRC1                                                     0x2e12
+#define mmCOMPUTE_PGM_RSRC2                                                     0x2e13
+#define mmCOMPUTE_VMID                                                          0x2e14
+#define mmCOMPUTE_RESOURCE_LIMITS                                               0x2e15
+#define mmCOMPUTE_STATIC_THREAD_MGMT_SE0                                        0x2e16
+#define mmCOMPUTE_STATIC_THREAD_MGMT_SE1                                        0x2e17
+#define mmCOMPUTE_TMPRING_SIZE                                                  0x2e18
+#define mmCOMPUTE_STATIC_THREAD_MGMT_SE2                                        0x2e19
+#define mmCOMPUTE_STATIC_THREAD_MGMT_SE3                                        0x2e1a
+#define mmCOMPUTE_RESTART_X                                                     0x2e1b
+#define mmCOMPUTE_RESTART_Y                                                     0x2e1c
+#define mmCOMPUTE_RESTART_Z                                                     0x2e1d
+#define mmCOMPUTE_THREAD_TRACE_ENABLE                                           0x2e1e
+#define mmCOMPUTE_MISC_RESERVED                                                 0x2e1f
+#define mmCOMPUTE_DISPATCH_ID                                                   0x2e20
+#define mmCOMPUTE_THREADGROUP_ID                                                0x2e21
+#define mmCOMPUTE_RELAUNCH                                                      0x2e22
+#define mmCOMPUTE_WAVE_RESTORE_ADDR_LO                                          0x2e23
+#define mmCOMPUTE_WAVE_RESTORE_ADDR_HI                                          0x2e24
+#define mmCOMPUTE_WAVE_RESTORE_CONTROL                                          0x2e25
+#define mmCOMPUTE_USER_DATA_0                                                   0x2e40
+#define mmCOMPUTE_USER_DATA_1                                                   0x2e41
+#define mmCOMPUTE_USER_DATA_2                                                   0x2e42
+#define mmCOMPUTE_USER_DATA_3                                                   0x2e43
+#define mmCOMPUTE_USER_DATA_4                                                   0x2e44
+#define mmCOMPUTE_USER_DATA_5                                                   0x2e45
+#define mmCOMPUTE_USER_DATA_6                                                   0x2e46
+#define mmCOMPUTE_USER_DATA_7                                                   0x2e47
+#define mmCOMPUTE_USER_DATA_8                                                   0x2e48
+#define mmCOMPUTE_USER_DATA_9                                                   0x2e49
+#define mmCOMPUTE_USER_DATA_10                                                  0x2e4a
+#define mmCOMPUTE_USER_DATA_11                                                  0x2e4b
+#define mmCOMPUTE_USER_DATA_12                                                  0x2e4c
+#define mmCOMPUTE_USER_DATA_13                                                  0x2e4d
+#define mmCOMPUTE_USER_DATA_14                                                  0x2e4e
+#define mmCOMPUTE_USER_DATA_15                                                  0x2e4f
+#define mmCOMPUTE_NOWHERE                                                       0x2e7f
+#define mmCSPRIV_CONNECT                                                        0x0
+#define mmCSPRIV_THREAD_TRACE_TG0                                               0x1e
+#define mmCSPRIV_THREAD_TRACE_TG1                                               0x1e
+#define mmCSPRIV_THREAD_TRACE_TG2                                               0x1e
+#define mmCSPRIV_THREAD_TRACE_TG3                                               0x1e
+#define mmCSPRIV_THREAD_TRACE_EVENT                                             0x1f
+#define mmRLC_CNTL                                                              0xec00
+#define mmRLC_DEBUG_SELECT                                                      0xec01
+#define mmRLC_DEBUG                                                             0xec02
+#define mmRLC_MC_CNTL                                                           0xec03
+#define mmRLC_STAT                                                              0xec04
+#define mmRLC_SAFE_MODE                                                         0xec05
+#define mmRLC_MEM_SLP_CNTL                                                      0xec06
+#define mmSMU_RLC_RESPONSE                                                      0xec07
+#define mmRLC_RLCV_SAFE_MODE                                                    0xec08
+#define mmRLC_SMU_SAFE_MODE                                                     0xec09
+#define mmRLC_RLCV_COMMAND                                                      0xec0a
+#define mmRLC_CLK_CNTL                                                          0xec0b
+#define mmRLC_PERFMON_CLK_CNTL                                                  0xdcbf
+#define mmRLC_PERFMON_CNTL                                                      0xdcc0
+#define mmRLC_PERFCOUNTER0_SELECT                                               0xdcc1
+#define mmRLC_PERFCOUNTER1_SELECT                                               0xdcc2
+#define mmRLC_PERFCOUNTER0_LO                                                   0xd480
+#define mmRLC_PERFCOUNTER1_LO                                                   0xd482
+#define mmRLC_PERFCOUNTER0_HI                                                   0xd481
+#define mmRLC_PERFCOUNTER1_HI                                                   0xd483
+#define mmCGTT_RLC_CLK_CTRL                                                     0xf0b8
+#define mmRLC_LB_CNTL                                                           0xec19
+#define mmRLC_LB_CNTR_MAX                                                       0xec12
+#define mmRLC_LB_CNTR_INIT                                                      0xec1b
+#define mmRLC_LOAD_BALANCE_CNTR                                                 0xec1c
+#define mmRLC_JUMP_TABLE_RESTORE                                                0xec1e
+#define mmRLC_PG_DELAY_2                                                        0xec1f
+#define mmRLC_GPM_DEBUG_SELECT                                                  0xec20
+#define mmRLC_GPM_DEBUG                                                         0xec21
+#define mmRLC_GPM_DEBUG_INST_A                                                  0xec22
+#define mmRLC_GPM_DEBUG_INST_B                                                  0xec23
+#define mmRLC_GPM_DEBUG_INST_ADDR                                               0xec1d
+#define mmRLC_GPM_UCODE_ADDR                                                    0xf83c
+#define mmRLC_GPM_UCODE_DATA                                                    0xf83d
+#define mmGPU_BIST_CONTROL                                                      0xf835
+#define mmRLC_ROM_CNTL                                                          0xf836
+#define mmRLC_GPU_CLOCK_COUNT_LSB                                               0xec24
+#define mmRLC_GPU_CLOCK_COUNT_MSB                                               0xec25
+#define mmRLC_CAPTURE_GPU_CLOCK_COUNT                                           0xec26
+#define mmRLC_UCODE_CNTL                                                        0xec27
+#define mmRLC_GPM_STAT                                                          0xec40
+#define mmRLC_GPU_CLOCK_32_RES_SEL                                              0xec41
+#define mmRLC_GPU_CLOCK_32                                                      0xec42
+#define mmRLC_PG_CNTL                                                           0xec43
+#define mmRLC_GPM_THREAD_PRIORITY                                               0xec44
+#define mmRLC_GPM_THREAD_ENABLE                                                 0xec45
+#define mmRLC_GPM_VMID_THREAD0                                                  0xec46
+#define mmRLC_GPM_VMID_THREAD1                                                  0xec47
+#define mmRLC_CGTT_MGCG_OVERRIDE                                                0xec48
+#define mmRLC_CGCG_CGLS_CTRL                                                    0xec49
+#define mmRLC_CGCG_RAMP_CTRL                                                    0xec4a
+#define mmRLC_DYN_PG_STATUS                                                     0xec4b
+#define mmRLC_DYN_PG_REQUEST                                                    0xec4c
+#define mmRLC_PG_DELAY                                                          0xec4d
+#define mmRLC_CU_STATUS                                                         0xec4e
+#define mmRLC_LB_INIT_CU_MASK                                                   0xec4f
+#define mmRLC_LB_ALWAYS_ACTIVE_CU_MASK                                          0xec50
+#define mmRLC_LB_PARAMS                                                         0xec51
+#define mmRLC_THREAD1_DELAY                                                     0xec52
+#define mmRLC_PG_ALWAYS_ON_CU_MASK                                              0xec53
+#define mmRLC_MAX_PG_CU                                                         0xec54
+#define mmRLC_AUTO_PG_CTRL                                                      0xec55
+#define mmRLC_SMU_GRBM_REG_SAVE_CTRL                                            0xec56
+#define mmRLC_SERDES_RD_MASTER_INDEX                                            0xec59
+#define mmRLC_SERDES_RD_DATA_0                                                  0xec5a
+#define mmRLC_SERDES_RD_DATA_1                                                  0xec5b
+#define mmRLC_SERDES_RD_DATA_2                                                  0xec5c
+#define mmRLC_SERDES_WR_CU_MASTER_MASK                                          0xec5d
+#define mmRLC_SERDES_WR_NONCU_MASTER_MASK                                       0xec5e
+#define mmRLC_SERDES_WR_CTRL                                                    0xec5f
+#define mmRLC_SERDES_WR_DATA                                                    0xec60
+#define mmRLC_SERDES_CU_MASTER_BUSY                                             0xec61
+#define mmRLC_SERDES_NONCU_MASTER_BUSY                                          0xec62
+#define mmRLC_GPM_GENERAL_0                                                     0xec63
+#define mmRLC_GPM_GENERAL_1                                                     0xec64
+#define mmRLC_GPM_GENERAL_2                                                     0xec65
+#define mmRLC_GPM_GENERAL_3                                                     0xec66
+#define mmRLC_GPM_GENERAL_4                                                     0xec67
+#define mmRLC_GPM_GENERAL_5                                                     0xec68
+#define mmRLC_GPM_GENERAL_6                                                     0xec69
+#define mmRLC_GPM_GENERAL_7                                                     0xec6a
+#define mmRLC_GPM_SCRATCH_ADDR                                                  0xec6c
+#define mmRLC_GPM_SCRATCH_DATA                                                  0xec6d
+#define mmRLC_STATIC_PG_STATUS                                                  0xec6e
+#define mmRLC_GPM_PERF_COUNT_0                                                  0xec6f
+#define mmRLC_GPM_PERF_COUNT_1                                                  0xec70
+#define mmRLC_GPR_REG1                                                          0xec79
+#define mmRLC_GPR_REG2                                                          0xec7a
+#define mmRLC_MGCG_CTRL                                                         0xec1a
+#define mmRLC_GPM_THREAD_RESET                                                  0xec28
+#define mmRLC_SPM_VMID                                                          0xec71
+#define mmRLC_SPM_INT_CNTL                                                      0xec72
+#define mmRLC_SPM_INT_STATUS                                                    0xec73
+#define mmRLC_SPM_DEBUG_SELECT                                                  0xec74
+#define mmRLC_SPM_DEBUG                                                         0xec75
+#define mmRLC_SMU_MESSAGE                                                       0xec76
+#define mmRLC_GPM_LOG_SIZE                                                      0xec77
+#define mmRLC_GPM_LOG_CONT                                                      0xec7b
+#define mmRLC_PG_DELAY_3                                                        0xec78
+#define mmRLC_GPM_INT_DISABLE_TH0                                               0xec7c
+#define mmRLC_GPM_INT_DISABLE_TH1                                               0xec7d
+#define mmRLC_GPM_INT_FORCE_TH0                                                 0xec7e
+#define mmRLC_GPM_INT_FORCE_TH1                                                 0xec7f
+#define mmRLC_SRM_CNTL                                                          0xec80
+#define mmRLC_SRM_DEBUG_SELECT                                                  0xec81
+#define mmRLC_SRM_DEBUG                                                         0xec82
+#define mmRLC_SRM_ARAM_ADDR                                                     0xec83
+#define mmRLC_SRM_ARAM_DATA                                                     0xec84
+#define mmRLC_SRM_DRAM_ADDR                                                     0xec85
+#define mmRLC_SRM_DRAM_DATA                                                     0xec86
+#define mmRLC_SRM_GPM_COMMAND                                                   0xec87
+#define mmRLC_SRM_GPM_COMMAND_STATUS                                            0xec88
+#define mmRLC_SRM_RLCV_COMMAND                                                  0xec89
+#define mmRLC_SRM_RLCV_COMMAND_STATUS                                           0xec8a
+#define mmRLC_SRM_INDEX_CNTL_ADDR_0                                             0xec8b
+#define mmRLC_SRM_INDEX_CNTL_ADDR_1                                             0xec8c
+#define mmRLC_SRM_INDEX_CNTL_ADDR_2                                             0xec8d
+#define mmRLC_SRM_INDEX_CNTL_ADDR_3                                             0xec8e
+#define mmRLC_SRM_INDEX_CNTL_ADDR_4                                             0xec8f
+#define mmRLC_SRM_INDEX_CNTL_ADDR_5                                             0xec90
+#define mmRLC_SRM_INDEX_CNTL_ADDR_6                                             0xec91
+#define mmRLC_SRM_INDEX_CNTL_ADDR_7                                             0xec92
+#define mmRLC_SRM_INDEX_CNTL_DATA_0                                             0xec93
+#define mmRLC_SRM_INDEX_CNTL_DATA_1                                             0xec94
+#define mmRLC_SRM_INDEX_CNTL_DATA_2                                             0xec95
+#define mmRLC_SRM_INDEX_CNTL_DATA_3                                             0xec96
+#define mmRLC_SRM_INDEX_CNTL_DATA_4                                             0xec97
+#define mmRLC_SRM_INDEX_CNTL_DATA_5                                             0xec98
+#define mmRLC_SRM_INDEX_CNTL_DATA_6                                             0xec99
+#define mmRLC_SRM_INDEX_CNTL_DATA_7                                             0xec9a
+#define mmRLC_SRM_STAT                                                          0xec9b
+#define mmRLC_SRM_GPM_ABORT                                                     0xec9c
+#define mmRLC_CSIB_ADDR_LO                                                      0xeca2
+#define mmRLC_CSIB_ADDR_HI                                                      0xeca3
+#define mmRLC_CSIB_LENGTH                                                       0xeca4
+#define mmRLC_CP_RESPONSE0                                                      0xeca5
+#define mmRLC_CP_RESPONSE1                                                      0xeca6
+#define mmRLC_CP_RESPONSE2                                                      0xeca7
+#define mmRLC_CP_RESPONSE3                                                      0xeca8
+#define mmRLC_SMU_COMMAND                                                       0xeca9
+#define mmRLC_CP_SCHEDULERS                                                     0xecaa
+#define mmRLC_SMU_ARGUMENT_1                                                    0xecab
+#define mmRLC_SMU_ARGUMENT_2                                                    0xecac
+#define mmRLC_GPM_GENERAL_8                                                     0xecad
+#define mmRLC_GPM_GENERAL_9                                                     0xecae
+#define mmRLC_GPM_GENERAL_10                                                    0xecaf
+#define mmRLC_GPM_GENERAL_11                                                    0xecb0
+#define mmRLC_GPM_GENERAL_12                                                    0xecb1
+#define mmRLC_SPM_PERFMON_CNTL                                                  0xdc80
+#define mmRLC_SPM_PERFMON_RING_BASE_LO                                          0xdc81
+#define mmRLC_SPM_PERFMON_RING_BASE_HI                                          0xdc82
+#define mmRLC_SPM_PERFMON_RING_SIZE                                             0xdc83
+#define mmRLC_SPM_PERFMON_SEGMENT_SIZE                                          0xdc84
+#define mmRLC_SPM_SE_MUXSEL_ADDR                                                0xdc85
+#define mmRLC_SPM_SE_MUXSEL_DATA                                                0xdc86
+#define mmRLC_SPM_CPG_PERFMON_SAMPLE_DELAY                                      0xdc87
+#define mmRLC_SPM_CPC_PERFMON_SAMPLE_DELAY                                      0xdc88
+#define mmRLC_SPM_CPF_PERFMON_SAMPLE_DELAY                                      0xdc89
+#define mmRLC_SPM_CB_PERFMON_SAMPLE_DELAY                                       0xdc8a
+#define mmRLC_SPM_DB_PERFMON_SAMPLE_DELAY                                       0xdc8b
+#define mmRLC_SPM_PA_PERFMON_SAMPLE_DELAY                                       0xdc8c
+#define mmRLC_SPM_GDS_PERFMON_SAMPLE_DELAY                                      0xdc8d
+#define mmRLC_SPM_IA_PERFMON_SAMPLE_DELAY                                       0xdc8e
+#define mmRLC_SPM_SC_PERFMON_SAMPLE_DELAY                                       0xdc90
+#define mmRLC_SPM_TCC_PERFMON_SAMPLE_DELAY                                      0xdc91
+#define mmRLC_SPM_TCA_PERFMON_SAMPLE_DELAY                                      0xdc92
+#define mmRLC_SPM_TCP_PERFMON_SAMPLE_DELAY                                      0xdc93
+#define mmRLC_SPM_TA_PERFMON_SAMPLE_DELAY                                       0xdc94
+#define mmRLC_SPM_TD_PERFMON_SAMPLE_DELAY                                       0xdc95
+#define mmRLC_SPM_VGT_PERFMON_SAMPLE_DELAY                                      0xdc96
+#define mmRLC_SPM_SPI_PERFMON_SAMPLE_DELAY                                      0xdc97
+#define mmRLC_SPM_SQG_PERFMON_SAMPLE_DELAY                                      0xdc98
+#define mmRLC_SPM_SX_PERFMON_SAMPLE_DELAY                                       0xdc9a
+#define mmRLC_SPM_GLOBAL_MUXSEL_ADDR                                            0xdc9b
+#define mmRLC_SPM_GLOBAL_MUXSEL_DATA                                            0xdc9c
+#define mmRLC_SPM_RING_RDPTR                                                    0xdc9d
+#define mmRLC_SPM_SEGMENT_THRESHOLD                                             0xdc9e
+#define mmRLC_GPU_IOV_VF_ENABLE                                                 0xfb00
+#define mmRLC_GPU_IOV_RLC_RESPONSE                                              0xfb4d
+#define mmRLC_GPU_IOV_ACTIVE_FCN_ID                                             0xfb40
+#define mmSPI_PS_INPUT_CNTL_0                                                   0xa191
+#define mmSPI_PS_INPUT_CNTL_1                                                   0xa192
+#define mmSPI_PS_INPUT_CNTL_2                                                   0xa193
+#define mmSPI_PS_INPUT_CNTL_3                                                   0xa194
+#define mmSPI_PS_INPUT_CNTL_4                                                   0xa195
+#define mmSPI_PS_INPUT_CNTL_5                                                   0xa196
+#define mmSPI_PS_INPUT_CNTL_6                                                   0xa197
+#define mmSPI_PS_INPUT_CNTL_7                                                   0xa198
+#define mmSPI_PS_INPUT_CNTL_8                                                   0xa199
+#define mmSPI_PS_INPUT_CNTL_9                                                   0xa19a
+#define mmSPI_PS_INPUT_CNTL_10                                                  0xa19b
+#define mmSPI_PS_INPUT_CNTL_11                                                  0xa19c
+#define mmSPI_PS_INPUT_CNTL_12                                                  0xa19d
+#define mmSPI_PS_INPUT_CNTL_13                                                  0xa19e
+#define mmSPI_PS_INPUT_CNTL_14                                                  0xa19f
+#define mmSPI_PS_INPUT_CNTL_15                                                  0xa1a0
+#define mmSPI_PS_INPUT_CNTL_16                                                  0xa1a1
+#define mmSPI_PS_INPUT_CNTL_17                                                  0xa1a2
+#define mmSPI_PS_INPUT_CNTL_18                                                  0xa1a3
+#define mmSPI_PS_INPUT_CNTL_19                                                  0xa1a4
+#define mmSPI_PS_INPUT_CNTL_20                                                  0xa1a5
+#define mmSPI_PS_INPUT_CNTL_21                                                  0xa1a6
+#define mmSPI_PS_INPUT_CNTL_22                                                  0xa1a7
+#define mmSPI_PS_INPUT_CNTL_23                                                  0xa1a8
+#define mmSPI_PS_INPUT_CNTL_24                                                  0xa1a9
+#define mmSPI_PS_INPUT_CNTL_25                                                  0xa1aa
+#define mmSPI_PS_INPUT_CNTL_26                                                  0xa1ab
+#define mmSPI_PS_INPUT_CNTL_27                                                  0xa1ac
+#define mmSPI_PS_INPUT_CNTL_28                                                  0xa1ad
+#define mmSPI_PS_INPUT_CNTL_29                                                  0xa1ae
+#define mmSPI_PS_INPUT_CNTL_30                                                  0xa1af
+#define mmSPI_PS_INPUT_CNTL_31                                                  0xa1b0
+#define mmSPI_VS_OUT_CONFIG                                                     0xa1b1
+#define mmSPI_PS_INPUT_ENA                                                      0xa1b3
+#define mmSPI_PS_INPUT_ADDR                                                     0xa1b4
+#define mmSPI_INTERP_CONTROL_0                                                  0xa1b5
+#define mmSPI_PS_IN_CONTROL                                                     0xa1b6
+#define mmSPI_BARYC_CNTL                                                        0xa1b8
+#define mmSPI_TMPRING_SIZE                                                      0xa1ba
+#define mmSPI_SHADER_POS_FORMAT                                                 0xa1c3
+#define mmSPI_SHADER_Z_FORMAT                                                   0xa1c4
+#define mmSPI_SHADER_COL_FORMAT                                                 0xa1c5
+#define mmSPI_ARB_PRIORITY                                                      0x31c0
+#define mmSPI_ARB_CYCLES_0                                                      0x31c1
+#define mmSPI_ARB_CYCLES_1                                                      0x31c2
+#define mmSPI_CDBG_SYS_GFX                                                      0x31c3
+#define mmSPI_CDBG_SYS_HP3D                                                     0x31c4
+#define mmSPI_CDBG_SYS_CS0                                                      0x31c5
+#define mmSPI_CDBG_SYS_CS1                                                      0x31c6
+#define mmSPI_WCL_PIPE_PERCENT_GFX                                              0x31c7
+#define mmSPI_WCL_PIPE_PERCENT_HP3D                                             0x31c8
+#define mmSPI_WCL_PIPE_PERCENT_CS0                                              0x31c9
+#define mmSPI_WCL_PIPE_PERCENT_CS1                                              0x31ca
+#define mmSPI_WCL_PIPE_PERCENT_CS2                                              0x31cb
+#define mmSPI_WCL_PIPE_PERCENT_CS3                                              0x31cc
+#define mmSPI_WCL_PIPE_PERCENT_CS4                                              0x31cd
+#define mmSPI_WCL_PIPE_PERCENT_CS5                                              0x31ce
+#define mmSPI_WCL_PIPE_PERCENT_CS6                                              0x31cf
+#define mmSPI_WCL_PIPE_PERCENT_CS7                                              0x31d0
+#define mmSPI_GDBG_WAVE_CNTL                                                    0x31d1
+#define mmSPI_GDBG_TRAP_CONFIG                                                  0x31d2
+#define mmSPI_GDBG_TRAP_MASK                                                    0x31d3
+#define mmSPI_GDBG_TBA_LO                                                       0x31d4
+#define mmSPI_GDBG_TBA_HI                                                       0x31d5
+#define mmSPI_GDBG_TMA_LO                                                       0x31d6
+#define mmSPI_GDBG_TMA_HI                                                       0x31d7
+#define mmSPI_GDBG_TRAP_DATA0                                                   0x31d8
+#define mmSPI_GDBG_TRAP_DATA1                                                   0x31d9
+#define mmSPI_RESET_DEBUG                                                       0x31da
+#define mmSPI_COMPUTE_QUEUE_RESET                                               0x31db
+#define mmSPI_RESOURCE_RESERVE_CU_0                                             0x31dc
+#define mmSPI_RESOURCE_RESERVE_CU_1                                             0x31dd
+#define mmSPI_RESOURCE_RESERVE_CU_2                                             0x31de
+#define mmSPI_RESOURCE_RESERVE_CU_3                                             0x31df
+#define mmSPI_RESOURCE_RESERVE_CU_4                                             0x31e0
+#define mmSPI_RESOURCE_RESERVE_CU_5                                             0x31e1
+#define mmSPI_RESOURCE_RESERVE_CU_6                                             0x31e2
+#define mmSPI_RESOURCE_RESERVE_CU_7                                             0x31e3
+#define mmSPI_RESOURCE_RESERVE_CU_8                                             0x31e4
+#define mmSPI_RESOURCE_RESERVE_CU_9                                             0x31e5
+#define mmSPI_RESOURCE_RESERVE_CU_10                                            0x31f0
+#define mmSPI_RESOURCE_RESERVE_CU_11                                            0x31f1
+#define mmSPI_RESOURCE_RESERVE_CU_12                                            0x31f4
+#define mmSPI_RESOURCE_RESERVE_CU_13                                            0x31f5
+#define mmSPI_RESOURCE_RESERVE_CU_14                                            0x31f6
+#define mmSPI_RESOURCE_RESERVE_CU_15                                            0x31f7
+#define mmSPI_RESOURCE_RESERVE_EN_CU_0                                          0x31e6
+#define mmSPI_RESOURCE_RESERVE_EN_CU_1                                          0x31e7
+#define mmSPI_RESOURCE_RESERVE_EN_CU_2                                          0x31e8
+#define mmSPI_RESOURCE_RESERVE_EN_CU_3                                          0x31e9
+#define mmSPI_RESOURCE_RESERVE_EN_CU_4                                          0x31ea
+#define mmSPI_RESOURCE_RESERVE_EN_CU_5                                          0x31eb
+#define mmSPI_RESOURCE_RESERVE_EN_CU_6                                          0x31ec
+#define mmSPI_RESOURCE_RESERVE_EN_CU_7                                          0x31ed
+#define mmSPI_RESOURCE_RESERVE_EN_CU_8                                          0x31ee
+#define mmSPI_RESOURCE_RESERVE_EN_CU_9                                          0x31ef
+#define mmSPI_RESOURCE_RESERVE_EN_CU_10                                         0x31f2
+#define mmSPI_RESOURCE_RESERVE_EN_CU_11                                         0x31f3
+#define mmSPI_RESOURCE_RESERVE_EN_CU_12                                         0x31f8
+#define mmSPI_RESOURCE_RESERVE_EN_CU_13                                         0x31f9
+#define mmSPI_RESOURCE_RESERVE_EN_CU_14                                         0x31fa
+#define mmSPI_RESOURCE_RESERVE_EN_CU_15                                         0x31fb
+#define mmSPI_COMPUTE_WF_CTX_SAVE                                               0x31fc
+#define mmSPI_PS_MAX_WAVE_ID                                                    0x243a
+#define mmSPI_START_PHASE                                                       0x243b
+#define mmSPI_GFX_CNTL                                                          0x243c
+#define mmSPI_CONFIG_CNTL                                                       0x2440
+#define mmSPI_DEBUG_CNTL                                                        0x2441
+#define mmSPI_DEBUG_READ                                                        0x2442
+#define mmSPI_DSM_CNTL                                                          0x2443
+#define mmSPI_EDC_CNT                                                           0x2444
+#define mmSPI_PERFCOUNTER0_SELECT                                               0xd980
+#define mmSPI_PERFCOUNTER1_SELECT                                               0xd981
+#define mmSPI_PERFCOUNTER2_SELECT                                               0xd982
+#define mmSPI_PERFCOUNTER3_SELECT                                               0xd983
+#define mmSPI_PERFCOUNTER0_SELECT1                                              0xd984
+#define mmSPI_PERFCOUNTER1_SELECT1                                              0xd985
+#define mmSPI_PERFCOUNTER2_SELECT1                                              0xd986
+#define mmSPI_PERFCOUNTER3_SELECT1                                              0xd987
+#define mmSPI_PERFCOUNTER4_SELECT                                               0xd988
+#define mmSPI_PERFCOUNTER5_SELECT                                               0xd989
+#define mmSPI_PERFCOUNTER_BINS                                                  0xd98a
+#define mmSPI_PERFCOUNTER0_HI                                                   0xd180
+#define mmSPI_PERFCOUNTER0_LO                                                   0xd181
+#define mmSPI_PERFCOUNTER1_HI                                                   0xd182
+#define mmSPI_PERFCOUNTER1_LO                                                   0xd183
+#define mmSPI_PERFCOUNTER2_HI                                                   0xd184
+#define mmSPI_PERFCOUNTER2_LO                                                   0xd185
+#define mmSPI_PERFCOUNTER3_HI                                                   0xd186
+#define mmSPI_PERFCOUNTER3_LO                                                   0xd187
+#define mmSPI_PERFCOUNTER4_HI                                                   0xd188
+#define mmSPI_PERFCOUNTER4_LO                                                   0xd189
+#define mmSPI_PERFCOUNTER5_HI                                                   0xd18a
+#define mmSPI_PERFCOUNTER5_LO                                                   0xd18b
+#define mmSPI_CONFIG_CNTL_1                                                     0x244f
+#define mmSPI_DEBUG_BUSY                                                        0x2450
+#define mmSPI_CONFIG_CNTL_2                                                     0x2451
+#define mmCGTS_SM_CTRL_REG                                                      0xf000
+#define mmCGTS_RD_CTRL_REG                                                      0xf001
+#define mmCGTS_RD_REG                                                           0xf002
+#define mmCGTS_TCC_DISABLE                                                      0xf003
+#define mmCGTS_USER_TCC_DISABLE                                                 0xf004
+#define mmCGTS_CU0_SP0_CTRL_REG                                                 0xf008
+#define mmCGTS_CU0_LDS_SQ_CTRL_REG                                              0xf009
+#define mmCGTS_CU0_TA_SQC_CTRL_REG                                              0xf00a
+#define mmCGTS_CU0_SP1_CTRL_REG                                                 0xf00b
+#define mmCGTS_CU0_TD_TCP_CTRL_REG                                              0xf00c
+#define mmCGTS_CU1_SP0_CTRL_REG                                                 0xf00d
+#define mmCGTS_CU1_LDS_SQ_CTRL_REG                                              0xf00e
+#define mmCGTS_CU1_TA_CTRL_REG                                                  0xf00f
+#define mmCGTS_CU1_SP1_CTRL_REG                                                 0xf010
+#define mmCGTS_CU1_TD_TCP_CTRL_REG                                              0xf011
+#define mmCGTS_CU2_SP0_CTRL_REG                                                 0xf012
+#define mmCGTS_CU2_LDS_SQ_CTRL_REG                                              0xf013
+#define mmCGTS_CU2_TA_CTRL_REG                                                  0xf014
+#define mmCGTS_CU2_SP1_CTRL_REG                                                 0xf015
+#define mmCGTS_CU2_TD_TCP_CTRL_REG                                              0xf016
+#define mmCGTS_CU3_SP0_CTRL_REG                                                 0xf017
+#define mmCGTS_CU3_LDS_SQ_CTRL_REG                                              0xf018
+#define mmCGTS_CU3_TA_CTRL_REG                                                  0xf019
+#define mmCGTS_CU3_SP1_CTRL_REG                                                 0xf01a
+#define mmCGTS_CU3_TD_TCP_CTRL_REG                                              0xf01b
+#define mmCGTS_CU4_SP0_CTRL_REG                                                 0xf01c
+#define mmCGTS_CU4_LDS_SQ_CTRL_REG                                              0xf01d
+#define mmCGTS_CU4_TA_SQC_CTRL_REG                                              0xf01e
+#define mmCGTS_CU4_SP1_CTRL_REG                                                 0xf01f
+#define mmCGTS_CU4_TD_TCP_CTRL_REG                                              0xf020
+#define mmCGTS_CU5_SP0_CTRL_REG                                                 0xf021
+#define mmCGTS_CU5_LDS_SQ_CTRL_REG                                              0xf022
+#define mmCGTS_CU5_TA_CTRL_REG                                                  0xf023
+#define mmCGTS_CU5_SP1_CTRL_REG                                                 0xf024
+#define mmCGTS_CU5_TD_TCP_CTRL_REG                                              0xf025
+#define mmCGTS_CU6_SP0_CTRL_REG                                                 0xf026
+#define mmCGTS_CU6_LDS_SQ_CTRL_REG                                              0xf027
+#define mmCGTS_CU6_TA_CTRL_REG                                                  0xf028
+#define mmCGTS_CU6_SP1_CTRL_REG                                                 0xf029
+#define mmCGTS_CU6_TD_TCP_CTRL_REG                                              0xf02a
+#define mmCGTS_CU7_SP0_CTRL_REG                                                 0xf02b
+#define mmCGTS_CU7_LDS_SQ_CTRL_REG                                              0xf02c
+#define mmCGTS_CU7_TA_CTRL_REG                                                  0xf02d
+#define mmCGTS_CU7_SP1_CTRL_REG                                                 0xf02e
+#define mmCGTS_CU7_TD_TCP_CTRL_REG                                              0xf02f
+#define mmCGTS_CU8_SP0_CTRL_REG                                                 0xf030
+#define mmCGTS_CU8_LDS_SQ_CTRL_REG                                              0xf031
+#define mmCGTS_CU8_TA_SQC_CTRL_REG                                              0xf032
+#define mmCGTS_CU8_SP1_CTRL_REG                                                 0xf033
+#define mmCGTS_CU8_TD_TCP_CTRL_REG                                              0xf034
+#define mmCGTS_CU9_SP0_CTRL_REG                                                 0xf035
+#define mmCGTS_CU9_LDS_SQ_CTRL_REG                                              0xf036
+#define mmCGTS_CU9_TA_CTRL_REG                                                  0xf037
+#define mmCGTS_CU9_SP1_CTRL_REG                                                 0xf038
+#define mmCGTS_CU9_TD_TCP_CTRL_REG                                              0xf039
+#define mmCGTS_CU10_SP0_CTRL_REG                                                0xf03a
+#define mmCGTS_CU10_LDS_SQ_CTRL_REG                                             0xf03b
+#define mmCGTS_CU10_TA_CTRL_REG                                                 0xf03c
+#define mmCGTS_CU10_SP1_CTRL_REG                                                0xf03d
+#define mmCGTS_CU10_TD_TCP_CTRL_REG                                             0xf03e
+#define mmCGTS_CU11_SP0_CTRL_REG                                                0xf03f
+#define mmCGTS_CU11_LDS_SQ_CTRL_REG                                             0xf040
+#define mmCGTS_CU11_TA_CTRL_REG                                                 0xf041
+#define mmCGTS_CU11_SP1_CTRL_REG                                                0xf042
+#define mmCGTS_CU11_TD_TCP_CTRL_REG                                             0xf043
+#define mmCGTS_CU12_SP0_CTRL_REG                                                0xf044
+#define mmCGTS_CU12_LDS_SQ_CTRL_REG                                             0xf045
+#define mmCGTS_CU12_TA_SQC_CTRL_REG                                             0xf046
+#define mmCGTS_CU12_SP1_CTRL_REG                                                0xf047
+#define mmCGTS_CU12_TD_TCP_CTRL_REG                                             0xf048
+#define mmCGTS_CU13_SP0_CTRL_REG                                                0xf049
+#define mmCGTS_CU13_LDS_SQ_CTRL_REG                                             0xf04a
+#define mmCGTS_CU13_TA_CTRL_REG                                                 0xf04b
+#define mmCGTS_CU13_SP1_CTRL_REG                                                0xf04c
+#define mmCGTS_CU13_TD_TCP_CTRL_REG                                             0xf04d
+#define mmCGTS_CU14_SP0_CTRL_REG                                                0xf04e
+#define mmCGTS_CU14_LDS_SQ_CTRL_REG                                             0xf04f
+#define mmCGTS_CU14_TA_CTRL_REG                                                 0xf050
+#define mmCGTS_CU14_SP1_CTRL_REG                                                0xf051
+#define mmCGTS_CU14_TD_TCP_CTRL_REG                                             0xf052
+#define mmCGTS_CU15_SP0_CTRL_REG                                                0xf053
+#define mmCGTS_CU15_LDS_SQ_CTRL_REG                                             0xf054
+#define mmCGTS_CU15_TA_CTRL_REG                                                 0xf055
+#define mmCGTS_CU15_SP1_CTRL_REG                                                0xf056
+#define mmCGTS_CU15_TD_TCP_CTRL_REG                                             0xf057
+#define mmCGTT_SPI_CLK_CTRL                                                     0xf080
+#define mmCGTT_PC_CLK_CTRL                                                      0xf081
+#define mmCGTT_BCI_CLK_CTRL                                                     0xf082
+#define mmSPI_WF_LIFETIME_CNTL                                                  0x24aa
+#define mmSPI_WF_LIFETIME_LIMIT_0                                               0x24ab
+#define mmSPI_WF_LIFETIME_LIMIT_1                                               0x24ac
+#define mmSPI_WF_LIFETIME_LIMIT_2                                               0x24ad
+#define mmSPI_WF_LIFETIME_LIMIT_3                                               0x24ae
+#define mmSPI_WF_LIFETIME_LIMIT_4                                               0x24af
+#define mmSPI_WF_LIFETIME_LIMIT_5                                               0x24b0
+#define mmSPI_WF_LIFETIME_LIMIT_6                                               0x24b1
+#define mmSPI_WF_LIFETIME_LIMIT_7                                               0x24b2
+#define mmSPI_WF_LIFETIME_LIMIT_8                                               0x24b3
+#define mmSPI_WF_LIFETIME_LIMIT_9                                               0x24b4
+#define mmSPI_WF_LIFETIME_STATUS_0                                              0x24b5
+#define mmSPI_WF_LIFETIME_STATUS_1                                              0x24b6
+#define mmSPI_WF_LIFETIME_STATUS_2                                              0x24b7
+#define mmSPI_WF_LIFETIME_STATUS_3                                              0x24b8
+#define mmSPI_WF_LIFETIME_STATUS_4                                              0x24b9
+#define mmSPI_WF_LIFETIME_STATUS_5                                              0x24ba
+#define mmSPI_WF_LIFETIME_STATUS_6                                              0x24bb
+#define mmSPI_WF_LIFETIME_STATUS_7                                              0x24bc
+#define mmSPI_WF_LIFETIME_STATUS_8                                              0x24bd
+#define mmSPI_WF_LIFETIME_STATUS_9                                              0x24be
+#define mmSPI_WF_LIFETIME_STATUS_10                                             0x24bf
+#define mmSPI_WF_LIFETIME_STATUS_11                                             0x24c0
+#define mmSPI_WF_LIFETIME_STATUS_12                                             0x24c1
+#define mmSPI_WF_LIFETIME_STATUS_13                                             0x24c2
+#define mmSPI_WF_LIFETIME_STATUS_14                                             0x24c3
+#define mmSPI_WF_LIFETIME_STATUS_15                                             0x24c4
+#define mmSPI_WF_LIFETIME_STATUS_16                                             0x24c5
+#define mmSPI_WF_LIFETIME_STATUS_17                                             0x24c6
+#define mmSPI_WF_LIFETIME_STATUS_18                                             0x24c7
+#define mmSPI_WF_LIFETIME_STATUS_19                                             0x24c8
+#define mmSPI_WF_LIFETIME_STATUS_20                                             0x24c9
+#define mmSPI_WF_LIFETIME_DEBUG                                                 0x24ca
+#define mmSPI_SLAVE_DEBUG_BUSY                                                  0x24d3
+#define mmSPI_LB_CTR_CTRL                                                       0x24d4
+#define mmSPI_LB_CU_MASK                                                        0x24d5
+#define mmSPI_LB_DATA_REG                                                       0x24d6
+#define mmSPI_PG_ENABLE_STATIC_CU_MASK                                          0x24d7
+#define mmSPI_GDS_CREDITS                                                       0x24d8
+#define mmSPI_SX_EXPORT_BUFFER_SIZES                                            0x24d9
+#define mmSPI_SX_SCOREBOARD_BUFFER_SIZES                                        0x24da
+#define mmSPI_CSQ_WF_ACTIVE_STATUS                                              0x24db
+#define mmSPI_CSQ_WF_ACTIVE_COUNT_0                                             0x24dc
+#define mmSPI_CSQ_WF_ACTIVE_COUNT_1                                             0x24dd
+#define mmSPI_CSQ_WF_ACTIVE_COUNT_2                                             0x24de
+#define mmSPI_CSQ_WF_ACTIVE_COUNT_3                                             0x24df
+#define mmSPI_CSQ_WF_ACTIVE_COUNT_4                                             0x24e0
+#define mmSPI_CSQ_WF_ACTIVE_COUNT_5                                             0x24e1
+#define mmSPI_CSQ_WF_ACTIVE_COUNT_6                                             0x24e2
+#define mmSPI_CSQ_WF_ACTIVE_COUNT_7                                             0x24e3
+#define mmBCI_DEBUG_READ                                                        0x24eb
+#define mmSPI_P0_TRAP_SCREEN_PSBA_LO                                            0x24ec
+#define mmSPI_P0_TRAP_SCREEN_PSBA_HI                                            0x24ed
+#define mmSPI_P0_TRAP_SCREEN_PSMA_LO                                            0x24ee
+#define mmSPI_P0_TRAP_SCREEN_PSMA_HI                                            0x24ef
+#define mmSPI_P0_TRAP_SCREEN_GPR_MIN                                            0x24f0
+#define mmSPI_P1_TRAP_SCREEN_PSBA_LO                                            0x24f1
+#define mmSPI_P1_TRAP_SCREEN_PSBA_HI                                            0x24f2
+#define mmSPI_P1_TRAP_SCREEN_PSMA_LO                                            0x24f3
+#define mmSPI_P1_TRAP_SCREEN_PSMA_HI                                            0x24f4
+#define mmSPI_P1_TRAP_SCREEN_GPR_MIN                                            0x24f5
+#define mmSPI_SHADER_TBA_LO_PS                                                  0x2c00
+#define mmSPI_SHADER_TBA_HI_PS                                                  0x2c01
+#define mmSPI_SHADER_TMA_LO_PS                                                  0x2c02
+#define mmSPI_SHADER_TMA_HI_PS                                                  0x2c03
+#define mmSPI_SHADER_PGM_LO_PS                                                  0x2c08
+#define mmSPI_SHADER_PGM_HI_PS                                                  0x2c09
+#define mmSPI_SHADER_PGM_RSRC1_PS                                               0x2c0a
+#define mmSPI_SHADER_PGM_RSRC2_PS                                               0x2c0b
+#define mmSPI_SHADER_PGM_RSRC3_PS                                               0x2c07
+#define mmSPI_SHADER_USER_DATA_PS_0                                             0x2c0c
+#define mmSPI_SHADER_USER_DATA_PS_1                                             0x2c0d
+#define mmSPI_SHADER_USER_DATA_PS_2                                             0x2c0e
+#define mmSPI_SHADER_USER_DATA_PS_3                                             0x2c0f
+#define mmSPI_SHADER_USER_DATA_PS_4                                             0x2c10
+#define mmSPI_SHADER_USER_DATA_PS_5                                             0x2c11
+#define mmSPI_SHADER_USER_DATA_PS_6                                             0x2c12
+#define mmSPI_SHADER_USER_DATA_PS_7                                             0x2c13
+#define mmSPI_SHADER_USER_DATA_PS_8                                             0x2c14
+#define mmSPI_SHADER_USER_DATA_PS_9                                             0x2c15
+#define mmSPI_SHADER_USER_DATA_PS_10                                            0x2c16
+#define mmSPI_SHADER_USER_DATA_PS_11                                            0x2c17
+#define mmSPI_SHADER_USER_DATA_PS_12                                            0x2c18
+#define mmSPI_SHADER_USER_DATA_PS_13                                            0x2c19
+#define mmSPI_SHADER_USER_DATA_PS_14                                            0x2c1a
+#define mmSPI_SHADER_USER_DATA_PS_15                                            0x2c1b
+#define mmSPI_SHADER_TBA_LO_VS                                                  0x2c40
+#define mmSPI_SHADER_TBA_HI_VS                                                  0x2c41
+#define mmSPI_SHADER_TMA_LO_VS                                                  0x2c42
+#define mmSPI_SHADER_TMA_HI_VS                                                  0x2c43
+#define mmSPI_SHADER_PGM_LO_VS                                                  0x2c48
+#define mmSPI_SHADER_PGM_HI_VS                                                  0x2c49
+#define mmSPI_SHADER_PGM_RSRC1_VS                                               0x2c4a
+#define mmSPI_SHADER_PGM_RSRC2_VS                                               0x2c4b
+#define mmSPI_SHADER_PGM_RSRC3_VS                                               0x2c46
+#define mmSPI_SHADER_LATE_ALLOC_VS                                              0x2c47
+#define mmSPI_SHADER_USER_DATA_VS_0                                             0x2c4c
+#define mmSPI_SHADER_USER_DATA_VS_1                                             0x2c4d
+#define mmSPI_SHADER_USER_DATA_VS_2                                             0x2c4e
+#define mmSPI_SHADER_USER_DATA_VS_3                                             0x2c4f
+#define mmSPI_SHADER_USER_DATA_VS_4                                             0x2c50
+#define mmSPI_SHADER_USER_DATA_VS_5                                             0x2c51
+#define mmSPI_SHADER_USER_DATA_VS_6                                             0x2c52
+#define mmSPI_SHADER_USER_DATA_VS_7                                             0x2c53
+#define mmSPI_SHADER_USER_DATA_VS_8                                             0x2c54
+#define mmSPI_SHADER_USER_DATA_VS_9                                             0x2c55
+#define mmSPI_SHADER_USER_DATA_VS_10                                            0x2c56
+#define mmSPI_SHADER_USER_DATA_VS_11                                            0x2c57
+#define mmSPI_SHADER_USER_DATA_VS_12                                            0x2c58
+#define mmSPI_SHADER_USER_DATA_VS_13                                            0x2c59
+#define mmSPI_SHADER_USER_DATA_VS_14                                            0x2c5a
+#define mmSPI_SHADER_USER_DATA_VS_15                                            0x2c5b
+#define mmSPI_SHADER_PGM_RSRC2_ES_VS                                            0x2c7c
+#define mmSPI_SHADER_PGM_RSRC2_LS_VS                                            0x2c7d
+#define mmSPI_SHADER_TBA_LO_GS                                                  0x2c80
+#define mmSPI_SHADER_TBA_HI_GS                                                  0x2c81
+#define mmSPI_SHADER_TMA_LO_GS                                                  0x2c82
+#define mmSPI_SHADER_TMA_HI_GS                                                  0x2c83
+#define mmSPI_SHADER_PGM_LO_GS                                                  0x2c88
+#define mmSPI_SHADER_PGM_HI_GS                                                  0x2c89
+#define mmSPI_SHADER_PGM_RSRC1_GS                                               0x2c8a
+#define mmSPI_SHADER_PGM_RSRC2_GS                                               0x2c8b
+#define mmSPI_SHADER_PGM_RSRC3_GS                                               0x2c87
+#define mmSPI_SHADER_USER_DATA_GS_0                                             0x2c8c
+#define mmSPI_SHADER_USER_DATA_GS_1                                             0x2c8d
+#define mmSPI_SHADER_USER_DATA_GS_2                                             0x2c8e
+#define mmSPI_SHADER_USER_DATA_GS_3                                             0x2c8f
+#define mmSPI_SHADER_USER_DATA_GS_4                                             0x2c90
+#define mmSPI_SHADER_USER_DATA_GS_5                                             0x2c91
+#define mmSPI_SHADER_USER_DATA_GS_6                                             0x2c92
+#define mmSPI_SHADER_USER_DATA_GS_7                                             0x2c93
+#define mmSPI_SHADER_USER_DATA_GS_8                                             0x2c94
+#define mmSPI_SHADER_USER_DATA_GS_9                                             0x2c95
+#define mmSPI_SHADER_USER_DATA_GS_10                                            0x2c96
+#define mmSPI_SHADER_USER_DATA_GS_11                                            0x2c97
+#define mmSPI_SHADER_USER_DATA_GS_12                                            0x2c98
+#define mmSPI_SHADER_USER_DATA_GS_13                                            0x2c99
+#define mmSPI_SHADER_USER_DATA_GS_14                                            0x2c9a
+#define mmSPI_SHADER_USER_DATA_GS_15                                            0x2c9b
+#define mmSPI_SHADER_PGM_RSRC2_ES_GS                                            0x2cbc
+#define mmSPI_SHADER_TBA_LO_ES                                                  0x2cc0
+#define mmSPI_SHADER_TBA_HI_ES                                                  0x2cc1
+#define mmSPI_SHADER_TMA_LO_ES                                                  0x2cc2
+#define mmSPI_SHADER_TMA_HI_ES                                                  0x2cc3
+#define mmSPI_SHADER_PGM_LO_ES                                                  0x2cc8
+#define mmSPI_SHADER_PGM_HI_ES                                                  0x2cc9
+#define mmSPI_SHADER_PGM_RSRC1_ES                                               0x2cca
+#define mmSPI_SHADER_PGM_RSRC2_ES                                               0x2ccb
+#define mmSPI_SHADER_PGM_RSRC3_ES                                               0x2cc7
+#define mmSPI_SHADER_USER_DATA_ES_0                                             0x2ccc
+#define mmSPI_SHADER_USER_DATA_ES_1                                             0x2ccd
+#define mmSPI_SHADER_USER_DATA_ES_2                                             0x2cce
+#define mmSPI_SHADER_USER_DATA_ES_3                                             0x2ccf
+#define mmSPI_SHADER_USER_DATA_ES_4                                             0x2cd0
+#define mmSPI_SHADER_USER_DATA_ES_5                                             0x2cd1
+#define mmSPI_SHADER_USER_DATA_ES_6                                             0x2cd2
+#define mmSPI_SHADER_USER_DATA_ES_7                                             0x2cd3
+#define mmSPI_SHADER_USER_DATA_ES_8                                             0x2cd4
+#define mmSPI_SHADER_USER_DATA_ES_9                                             0x2cd5
+#define mmSPI_SHADER_USER_DATA_ES_10                                            0x2cd6
+#define mmSPI_SHADER_USER_DATA_ES_11                                            0x2cd7
+#define mmSPI_SHADER_USER_DATA_ES_12                                            0x2cd8
+#define mmSPI_SHADER_USER_DATA_ES_13                                            0x2cd9
+#define mmSPI_SHADER_USER_DATA_ES_14                                            0x2cda
+#define mmSPI_SHADER_USER_DATA_ES_15                                            0x2cdb
+#define mmSPI_SHADER_PGM_RSRC2_LS_ES                                            0x2cfd
+#define mmSPI_SHADER_TBA_LO_HS                                                  0x2d00
+#define mmSPI_SHADER_TBA_HI_HS                                                  0x2d01
+#define mmSPI_SHADER_TMA_LO_HS                                                  0x2d02
+#define mmSPI_SHADER_TMA_HI_HS                                                  0x2d03
+#define mmSPI_SHADER_PGM_LO_HS                                                  0x2d08
+#define mmSPI_SHADER_PGM_HI_HS                                                  0x2d09
+#define mmSPI_SHADER_PGM_RSRC1_HS                                               0x2d0a
+#define mmSPI_SHADER_PGM_RSRC2_HS                                               0x2d0b
+#define mmSPI_SHADER_PGM_RSRC3_HS                                               0x2d07
+#define mmSPI_SHADER_USER_DATA_HS_0                                             0x2d0c
+#define mmSPI_SHADER_USER_DATA_HS_1                                             0x2d0d
+#define mmSPI_SHADER_USER_DATA_HS_2                                             0x2d0e
+#define mmSPI_SHADER_USER_DATA_HS_3                                             0x2d0f
+#define mmSPI_SHADER_USER_DATA_HS_4                                             0x2d10
+#define mmSPI_SHADER_USER_DATA_HS_5                                             0x2d11
+#define mmSPI_SHADER_USER_DATA_HS_6                                             0x2d12
+#define mmSPI_SHADER_USER_DATA_HS_7                                             0x2d13
+#define mmSPI_SHADER_USER_DATA_HS_8                                             0x2d14
+#define mmSPI_SHADER_USER_DATA_HS_9                                             0x2d15
+#define mmSPI_SHADER_USER_DATA_HS_10                                            0x2d16
+#define mmSPI_SHADER_USER_DATA_HS_11                                            0x2d17
+#define mmSPI_SHADER_USER_DATA_HS_12                                            0x2d18
+#define mmSPI_SHADER_USER_DATA_HS_13                                            0x2d19
+#define mmSPI_SHADER_USER_DATA_HS_14                                            0x2d1a
+#define mmSPI_SHADER_USER_DATA_HS_15                                            0x2d1b
+#define mmSPI_SHADER_PGM_RSRC2_LS_HS                                            0x2d3d
+#define mmSPI_SHADER_TBA_LO_LS                                                  0x2d40
+#define mmSPI_SHADER_TBA_HI_LS                                                  0x2d41
+#define mmSPI_SHADER_TMA_LO_LS                                                  0x2d42
+#define mmSPI_SHADER_TMA_HI_LS                                                  0x2d43
+#define mmSPI_SHADER_PGM_LO_LS                                                  0x2d48
+#define mmSPI_SHADER_PGM_HI_LS                                                  0x2d49
+#define mmSPI_SHADER_PGM_RSRC1_LS                                               0x2d4a
+#define mmSPI_SHADER_PGM_RSRC2_LS                                               0x2d4b
+#define mmSPI_SHADER_PGM_RSRC3_LS                                               0x2d47
+#define mmSPI_SHADER_USER_DATA_LS_0                                             0x2d4c
+#define mmSPI_SHADER_USER_DATA_LS_1                                             0x2d4d
+#define mmSPI_SHADER_USER_DATA_LS_2                                             0x2d4e
+#define mmSPI_SHADER_USER_DATA_LS_3                                             0x2d4f
+#define mmSPI_SHADER_USER_DATA_LS_4                                             0x2d50
+#define mmSPI_SHADER_USER_DATA_LS_5                                             0x2d51
+#define mmSPI_SHADER_USER_DATA_LS_6                                             0x2d52
+#define mmSPI_SHADER_USER_DATA_LS_7                                             0x2d53
+#define mmSPI_SHADER_USER_DATA_LS_8                                             0x2d54
+#define mmSPI_SHADER_USER_DATA_LS_9                                             0x2d55
+#define mmSPI_SHADER_USER_DATA_LS_10                                            0x2d56
+#define mmSPI_SHADER_USER_DATA_LS_11                                            0x2d57
+#define mmSPI_SHADER_USER_DATA_LS_12                                            0x2d58
+#define mmSPI_SHADER_USER_DATA_LS_13                                            0x2d59
+#define mmSPI_SHADER_USER_DATA_LS_14                                            0x2d5a
+#define mmSPI_SHADER_USER_DATA_LS_15                                            0x2d5b
+#define mmSQ_CONFIG                                                             0x2300
+#define mmSQC_CONFIG                                                            0x2301
+#define mmSQC_CACHES                                                            0xc348
+#define mmSQC_WRITEBACK                                                         0xc349
+#define mmSQC_DSM_CNTL                                                          0x230f
+#define mmSQ_RANDOM_WAVE_PRI                                                    0x2303
+#define mmSQ_REG_CREDITS                                                        0x2304
+#define mmSQ_FIFO_SIZES                                                         0x2305
+#define mmSQ_DSM_CNTL                                                           0x2306
+#define mmCC_GC_SHADER_RATE_CONFIG                                              0x2312
+#define mmGC_USER_SHADER_RATE_CONFIG                                            0x2313
+#define mmSQ_INTERRUPT_AUTO_MASK                                                0x2314
+#define mmSQ_INTERRUPT_MSG_CTRL                                                 0x2315
+#define mmSQ_PERFCOUNTER_CTRL                                                   0xd9e0
+#define mmSQ_PERFCOUNTER_MASK                                                   0xd9e1
+#define mmSQ_PERFCOUNTER_CTRL2                                                  0xd9e2
+#define mmCC_SQC_BANK_DISABLE                                                   0x2307
+#define mmUSER_SQC_BANK_DISABLE                                                 0x2308
+#define mmSQ_PERFCOUNTER0_LO                                                    0xd1c0
+#define mmSQ_PERFCOUNTER1_LO                                                    0xd1c2
+#define mmSQ_PERFCOUNTER2_LO                                                    0xd1c4
+#define mmSQ_PERFCOUNTER3_LO                                                    0xd1c6
+#define mmSQ_PERFCOUNTER4_LO                                                    0xd1c8
+#define mmSQ_PERFCOUNTER5_LO                                                    0xd1ca
+#define mmSQ_PERFCOUNTER6_LO                                                    0xd1cc
+#define mmSQ_PERFCOUNTER7_LO                                                    0xd1ce
+#define mmSQ_PERFCOUNTER8_LO                                                    0xd1d0
+#define mmSQ_PERFCOUNTER9_LO                                                    0xd1d2
+#define mmSQ_PERFCOUNTER10_LO                                                   0xd1d4
+#define mmSQ_PERFCOUNTER11_LO                                                   0xd1d6
+#define mmSQ_PERFCOUNTER12_LO                                                   0xd1d8
+#define mmSQ_PERFCOUNTER13_LO                                                   0xd1da
+#define mmSQ_PERFCOUNTER14_LO                                                   0xd1dc
+#define mmSQ_PERFCOUNTER15_LO                                                   0xd1de
+#define mmSQ_PERFCOUNTER0_HI                                                    0xd1c1
+#define mmSQ_PERFCOUNTER1_HI                                                    0xd1c3
+#define mmSQ_PERFCOUNTER2_HI                                                    0xd1c5
+#define mmSQ_PERFCOUNTER3_HI                                                    0xd1c7
+#define mmSQ_PERFCOUNTER4_HI                                                    0xd1c9
+#define mmSQ_PERFCOUNTER5_HI                                                    0xd1cb
+#define mmSQ_PERFCOUNTER6_HI                                                    0xd1cd
+#define mmSQ_PERFCOUNTER7_HI                                                    0xd1cf
+#define mmSQ_PERFCOUNTER8_HI                                                    0xd1d1
+#define mmSQ_PERFCOUNTER9_HI                                                    0xd1d3
+#define mmSQ_PERFCOUNTER10_HI                                                   0xd1d5
+#define mmSQ_PERFCOUNTER11_HI                                                   0xd1d7
+#define mmSQ_PERFCOUNTER12_HI                                                   0xd1d9
+#define mmSQ_PERFCOUNTER13_HI                                                   0xd1db
+#define mmSQ_PERFCOUNTER14_HI                                                   0xd1dd
+#define mmSQ_PERFCOUNTER15_HI                                                   0xd1df
+#define mmSQ_PERFCOUNTER0_SELECT                                                0xd9c0
+#define mmSQ_PERFCOUNTER1_SELECT                                                0xd9c1
+#define mmSQ_PERFCOUNTER2_SELECT                                                0xd9c2
+#define mmSQ_PERFCOUNTER3_SELECT                                                0xd9c3
+#define mmSQ_PERFCOUNTER4_SELECT                                                0xd9c4
+#define mmSQ_PERFCOUNTER5_SELECT                                                0xd9c5
+#define mmSQ_PERFCOUNTER6_SELECT                                                0xd9c6
+#define mmSQ_PERFCOUNTER7_SELECT                                                0xd9c7
+#define mmSQ_PERFCOUNTER8_SELECT                                                0xd9c8
+#define mmSQ_PERFCOUNTER9_SELECT                                                0xd9c9
+#define mmSQ_PERFCOUNTER10_SELECT                                               0xd9ca
+#define mmSQ_PERFCOUNTER11_SELECT                                               0xd9cb
+#define mmSQ_PERFCOUNTER12_SELECT                                               0xd9cc
+#define mmSQ_PERFCOUNTER13_SELECT                                               0xd9cd
+#define mmSQ_PERFCOUNTER14_SELECT                                               0xd9ce
+#define mmSQ_PERFCOUNTER15_SELECT                                               0xd9cf
+#define mmCGTT_SQ_CLK_CTRL                                                      0xf08c
+#define mmCGTT_SQG_CLK_CTRL                                                     0xf08d
+#define mmSQ_ALU_CLK_CTRL                                                       0xf08e
+#define mmSQ_TEX_CLK_CTRL                                                       0xf08f
+#define mmSQ_LDS_CLK_CTRL                                                       0xf090
+#define mmSQ_POWER_THROTTLE                                                     0xf091
+#define mmSQ_POWER_THROTTLE2                                                    0xf092
+#define mmSQ_TIME_HI                                                            0x237c
+#define mmSQ_TIME_LO                                                            0x237d
+#define mmSQ_THREAD_TRACE_BASE                                                  0xc330
+#define mmSQ_THREAD_TRACE_BASE2                                                 0xc337
+#define mmSQ_THREAD_TRACE_SIZE                                                  0xc331
+#define mmSQ_THREAD_TRACE_MASK                                                  0xc332
+#define mmSQ_THREAD_TRACE_USERDATA_0                                            0xc340
+#define mmSQ_THREAD_TRACE_USERDATA_1                                            0xc341
+#define mmSQ_THREAD_TRACE_USERDATA_2                                            0xc342
+#define mmSQ_THREAD_TRACE_USERDATA_3                                            0xc343
+#define mmSQ_THREAD_TRACE_MODE                                                  0xc336
+#define mmSQ_THREAD_TRACE_CTRL                                                  0xc335
+#define mmSQ_THREAD_TRACE_TOKEN_MASK                                            0xc333
+#define mmSQ_THREAD_TRACE_TOKEN_MASK2                                           0xc338
+#define mmSQ_THREAD_TRACE_PERF_MASK                                             0xc334
+#define mmSQ_THREAD_TRACE_WPTR                                                  0xc339
+#define mmSQ_THREAD_TRACE_STATUS                                                0xc33a
+#define mmSQ_THREAD_TRACE_CNTR                                                  0x2390
+#define mmSQ_THREAD_TRACE_HIWATER                                               0xc33b
+#define mmSQ_LB_CTR_CTRL                                                        0x2398
+#define mmSQ_LB_DATA_ALU_CYCLES                                                 0x2399
+#define mmSQ_LB_DATA_TEX_CYCLES                                                 0x239a
+#define mmSQ_LB_DATA_ALU_STALLS                                                 0x239b
+#define mmSQ_LB_DATA_TEX_STALLS                                                 0x239c
+#define mmSQC_EDC_CNT                                                           0x23a0
+#define mmSQ_EDC_SEC_CNT                                                        0x23a1
+#define mmSQ_EDC_DED_CNT                                                        0x23a2
+#define mmSQ_EDC_INFO                                                           0x23a3
+#define mmSQ_BUF_RSRC_WORD0                                                     0x23c0
+#define mmSQ_BUF_RSRC_WORD1                                                     0x23c1
+#define mmSQ_BUF_RSRC_WORD2                                                     0x23c2
+#define mmSQ_BUF_RSRC_WORD3                                                     0x23c3
+#define mmSQ_IMG_RSRC_WORD0                                                     0x23c4
+#define mmSQ_IMG_RSRC_WORD1                                                     0x23c5
+#define mmSQ_IMG_RSRC_WORD2                                                     0x23c6
+#define mmSQ_IMG_RSRC_WORD3                                                     0x23c7
+#define mmSQ_IMG_RSRC_WORD4                                                     0x23c8
+#define mmSQ_IMG_RSRC_WORD5                                                     0x23c9
+#define mmSQ_IMG_RSRC_WORD6                                                     0x23ca
+#define mmSQ_IMG_RSRC_WORD7                                                     0x23cb
+#define mmSQ_IMG_SAMP_WORD0                                                     0x23cc
+#define mmSQ_IMG_SAMP_WORD1                                                     0x23cd
+#define mmSQ_IMG_SAMP_WORD2                                                     0x23ce
+#define mmSQ_IMG_SAMP_WORD3                                                     0x23cf
+#define mmSQ_FLAT_SCRATCH_WORD0                                                 0x23d0
+#define mmSQ_FLAT_SCRATCH_WORD1                                                 0x23d1
+#define mmSQ_M0_GPR_IDX_WORD                                                    0x23d2
+#define mmSQ_IND_INDEX                                                          0x2378
+#define mmSQ_CMD                                                                0x237b
+#define mmSQ_IND_DATA                                                           0x2379
+#define mmSQ_REG_TIMESTAMP                                                      0x2374
+#define mmSQ_CMD_TIMESTAMP                                                      0x2375
+#define mmSQ_HV_VMID_CTRL                                                       0xf840
+#define ixSQ_WAVE_INST_DW0                                                      0x1a
+#define ixSQ_WAVE_INST_DW1                                                      0x1b
+#define ixSQ_WAVE_PC_LO                                                         0x18
+#define ixSQ_WAVE_PC_HI                                                         0x19
+#define ixSQ_WAVE_IB_DBG0                                                       0x1c
+#define ixSQ_WAVE_IB_DBG1                                                       0x1d
+#define ixSQ_WAVE_EXEC_LO                                                       0x27e
+#define ixSQ_WAVE_EXEC_HI                                                       0x27f
+#define ixSQ_WAVE_STATUS                                                        0x12
+#define ixSQ_WAVE_MODE                                                          0x11
+#define ixSQ_WAVE_TRAPSTS                                                       0x13
+#define ixSQ_WAVE_HW_ID                                                         0x14
+#define ixSQ_WAVE_GPR_ALLOC                                                     0x15
+#define ixSQ_WAVE_LDS_ALLOC                                                     0x16
+#define ixSQ_WAVE_IB_STS                                                        0x17
+#define ixSQ_WAVE_M0                                                            0x27c
+#define ixSQ_WAVE_TBA_LO                                                        0x26c
+#define ixSQ_WAVE_TBA_HI                                                        0x26d
+#define ixSQ_WAVE_TMA_LO                                                        0x26e
+#define ixSQ_WAVE_TMA_HI                                                        0x26f
+#define ixSQ_WAVE_TTMP0                                                         0x270
+#define ixSQ_WAVE_TTMP1                                                         0x271
+#define ixSQ_WAVE_TTMP2                                                         0x272
+#define ixSQ_WAVE_TTMP3                                                         0x273
+#define ixSQ_WAVE_TTMP4                                                         0x274
+#define ixSQ_WAVE_TTMP5                                                         0x275
+#define ixSQ_WAVE_TTMP6                                                         0x276
+#define ixSQ_WAVE_TTMP7                                                         0x277
+#define ixSQ_WAVE_TTMP8                                                         0x278
+#define ixSQ_WAVE_TTMP9                                                         0x279
+#define ixSQ_WAVE_TTMP10                                                        0x27a
+#define ixSQ_WAVE_TTMP11                                                        0x27b
+#define mmSQ_DEBUG_STS_GLOBAL                                                   0x2309
+#define mmSQ_DEBUG_STS_GLOBAL2                                                  0x2310
+#define mmSQ_DEBUG_STS_GLOBAL3                                                  0x2311
+#define ixSQ_DEBUG_STS_LOCAL                                                    0x8
+#define ixSQ_DEBUG_CTRL_LOCAL                                                   0x9
+#define mmSH_MEM_BASES                                                          0x230a
+#define mmSH_MEM_APE1_BASE                                                      0x230b
+#define mmSH_MEM_APE1_LIMIT                                                     0x230c
+#define mmSH_MEM_CONFIG                                                         0x230d
+#define mmSQ_THREAD_TRACE_WORD_CMN                                              0x23b0
+#define mmSQ_THREAD_TRACE_WORD_INST                                             0x23b0
+#define mmSQ_THREAD_TRACE_WORD_INST_PC_1_OF_2                                   0x23b0
+#define mmSQ_THREAD_TRACE_WORD_INST_PC_2_OF_2                                   0x23b1
+#define mmSQ_THREAD_TRACE_WORD_INST_USERDATA_1_OF_2                             0x23b0
+#define mmSQ_THREAD_TRACE_WORD_INST_USERDATA_2_OF_2                             0x23b1
+#define mmSQ_THREAD_TRACE_WORD_TIMESTAMP_1_OF_2                                 0x23b0
+#define mmSQ_THREAD_TRACE_WORD_TIMESTAMP_2_OF_2                                 0x23b1
+#define mmSQ_THREAD_TRACE_WORD_WAVE                                             0x23b0
+#define mmSQ_THREAD_TRACE_WORD_MISC                                             0x23b0
+#define mmSQ_THREAD_TRACE_WORD_WAVE_START                                       0x23b0
+#define mmSQ_THREAD_TRACE_WORD_REG_1_OF_2                                       0x23b0
+#define mmSQ_THREAD_TRACE_WORD_REG_2_OF_2                                       0x23b0
+#define mmSQ_THREAD_TRACE_WORD_REG_CS_1_OF_2                                    0x23b0
+#define mmSQ_THREAD_TRACE_WORD_REG_CS_2_OF_2                                    0x23b0
+#define mmSQ_THREAD_TRACE_WORD_EVENT                                            0x23b0
+#define mmSQ_THREAD_TRACE_WORD_ISSUE                                            0x23b0
+#define mmSQ_THREAD_TRACE_WORD_PERF_1_OF_2                                      0x23b0
+#define mmSQ_THREAD_TRACE_WORD_PERF_2_OF_2                                      0x23b1
+#define mmSQ_WREXEC_EXEC_LO                                                     0x23b1
+#define mmSQ_WREXEC_EXEC_HI                                                     0x23b1
+#define mmSQC_GATCL1_CNTL                                                       0x23b2
+#define mmSQC_ATC_EDC_GATCL1_CNT                                                0x23b3
+#define ixSQ_INTERRUPT_WORD_CMN                                                 0x20c0
+#define ixSQ_INTERRUPT_WORD_AUTO                                                0x20c0
+#define ixSQ_INTERRUPT_WORD_WAVE                                                0x20c0
+#define mmSQ_SOP2                                                               0x237f
+#define mmSQ_VOP1                                                               0x237f
+#define mmSQ_MTBUF_1                                                            0x237f
+#define mmSQ_EXP_1                                                              0x237f
+#define mmSQ_MUBUF_1                                                            0x237f
+#define mmSQ_SMEM_1                                                             0x237f
+#define mmSQ_INST                                                               0x237f
+#define mmSQ_EXP_0                                                              0x237f
+#define mmSQ_MUBUF_0                                                            0x237f
+#define mmSQ_VOP_SDWA                                                           0x237f
+#define mmSQ_VOP3_0                                                             0x237f
+#define mmSQ_VOP2                                                               0x237f
+#define mmSQ_MTBUF_0                                                            0x237f
+#define mmSQ_SOPP                                                               0x237f
+#define mmSQ_FLAT_0                                                             0x237f
+#define mmSQ_VOP3_0_SDST_ENC                                                    0x237f
+#define mmSQ_MIMG_1                                                             0x237f
+#define mmSQ_SOP1                                                               0x237f
+#define mmSQ_SOPC                                                               0x237f
+#define mmSQ_FLAT_1                                                             0x237f
+#define mmSQ_DS_1                                                               0x237f
+#define mmSQ_VOP3_1                                                             0x237f
+#define mmSQ_SMEM_0                                                             0x237f
+#define mmSQ_MIMG_0                                                             0x237f
+#define mmSQ_SOPK                                                               0x237f
+#define mmSQ_DS_0                                                               0x237f
+#define mmSQ_VOP_DPP                                                            0x237f
+#define mmSQ_VOPC                                                               0x237f
+#define mmSQ_VINTRP                                                             0x237f
+#define mmCGTT_SX_CLK_CTRL0                                                     0xf094
+#define mmCGTT_SX_CLK_CTRL1                                                     0xf095
+#define mmCGTT_SX_CLK_CTRL2                                                     0xf096
+#define mmCGTT_SX_CLK_CTRL3                                                     0xf097
+#define mmCGTT_SX_CLK_CTRL4                                                     0xf098
+#define mmSX_DEBUG_BUSY                                                         0x2414
+#define mmSX_DEBUG_BUSY_2                                                       0x2415
+#define mmSX_DEBUG_BUSY_3                                                       0x2416
+#define mmSX_DEBUG_BUSY_4                                                       0x2417
+#define mmSX_DEBUG_1                                                            0x2418
+#define mmSX_PERFCOUNTER0_SELECT                                                0xda40
+#define mmSX_PERFCOUNTER1_SELECT                                                0xda41
+#define mmSX_PERFCOUNTER2_SELECT                                                0xda42
+#define mmSX_PERFCOUNTER3_SELECT                                                0xda43
+#define mmSX_PERFCOUNTER0_SELECT1                                               0xda44
+#define mmSX_PERFCOUNTER1_SELECT1                                               0xda45
+#define mmSX_PERFCOUNTER0_LO                                                    0xd240
+#define mmSX_PERFCOUNTER0_HI                                                    0xd241
+#define mmSX_PERFCOUNTER1_LO                                                    0xd242
+#define mmSX_PERFCOUNTER1_HI                                                    0xd243
+#define mmSX_PERFCOUNTER2_LO                                                    0xd244
+#define mmSX_PERFCOUNTER2_HI                                                    0xd245
+#define mmSX_PERFCOUNTER3_LO                                                    0xd246
+#define mmSX_PERFCOUNTER3_HI                                                    0xd247
+#define mmSX_PS_DOWNCONVERT                                                     0xa1d5
+#define mmSX_BLEND_OPT_EPSILON                                                  0xa1d6
+#define mmSX_BLEND_OPT_CONTROL                                                  0xa1d7
+#define mmSX_MRT0_BLEND_OPT                                                     0xa1d8
+#define mmSX_MRT1_BLEND_OPT                                                     0xa1d9
+#define mmSX_MRT2_BLEND_OPT                                                     0xa1da
+#define mmSX_MRT3_BLEND_OPT                                                     0xa1db
+#define mmSX_MRT4_BLEND_OPT                                                     0xa1dc
+#define mmSX_MRT5_BLEND_OPT                                                     0xa1dd
+#define mmSX_MRT6_BLEND_OPT                                                     0xa1de
+#define mmSX_MRT7_BLEND_OPT                                                     0xa1df
+#define mmTCC_CTRL                                                              0x2b80
+#define mmTCC_EDC_CNT                                                           0x2b82
+#define mmTCC_REDUNDANCY                                                        0x2b83
+#define mmTCC_EXE_DISABLE                                                       0x2b84
+#define mmTCC_DSM_CNTL                                                          0x2b85
+#define mmTCC_CGTT_SCLK_CTRL                                                    0xf0ac
+#define mmTCA_CGTT_SCLK_CTRL                                                    0xf0ad
+#define mmTCC_PERFCOUNTER0_SELECT                                               0xdb80
+#define mmTCC_PERFCOUNTER1_SELECT                                               0xdb82
+#define mmTCC_PERFCOUNTER0_SELECT1                                              0xdb81
+#define mmTCC_PERFCOUNTER1_SELECT1                                              0xdb83
+#define mmTCC_PERFCOUNTER2_SELECT                                               0xdb84
+#define mmTCC_PERFCOUNTER3_SELECT                                               0xdb85
+#define mmTCC_PERFCOUNTER0_LO                                                   0xd380
+#define mmTCC_PERFCOUNTER1_LO                                                   0xd382
+#define mmTCC_PERFCOUNTER2_LO                                                   0xd384
+#define mmTCC_PERFCOUNTER3_LO                                                   0xd386
+#define mmTCC_PERFCOUNTER0_HI                                                   0xd381
+#define mmTCC_PERFCOUNTER1_HI                                                   0xd383
+#define mmTCC_PERFCOUNTER2_HI                                                   0xd385
+#define mmTCC_PERFCOUNTER3_HI                                                   0xd387
+#define mmTCA_CTRL                                                              0x2bc0
+#define mmTCA_PERFCOUNTER0_SELECT                                               0xdb90
+#define mmTCA_PERFCOUNTER1_SELECT                                               0xdb92
+#define mmTCA_PERFCOUNTER0_SELECT1                                              0xdb91
+#define mmTCA_PERFCOUNTER1_SELECT1                                              0xdb93
+#define mmTCA_PERFCOUNTER2_SELECT                                               0xdb94
+#define mmTCA_PERFCOUNTER3_SELECT                                               0xdb95
+#define mmTCA_PERFCOUNTER0_LO                                                   0xd390
+#define mmTCA_PERFCOUNTER1_LO                                                   0xd392
+#define mmTCA_PERFCOUNTER2_LO                                                   0xd394
+#define mmTCA_PERFCOUNTER3_LO                                                   0xd396
+#define mmTCA_PERFCOUNTER0_HI                                                   0xd391
+#define mmTCA_PERFCOUNTER1_HI                                                   0xd393
+#define mmTCA_PERFCOUNTER2_HI                                                   0xd395
+#define mmTCA_PERFCOUNTER3_HI                                                   0xd397
+#define mmTA_BC_BASE_ADDR                                                       0xa020
+#define mmTA_BC_BASE_ADDR_HI                                                    0xa021
+#define mmTD_CNTL                                                               0x2525
+#define mmTD_STATUS                                                             0x2526
+#define mmTD_DEBUG_INDEX                                                        0x2528
+#define mmTD_DEBUG_DATA                                                         0x2529
+#define mmTD_DSM_CNTL                                                           0x252f
+#define mmTD_PERFCOUNTER0_SELECT                                                0xdb00
+#define mmTD_PERFCOUNTER1_SELECT                                                0xdb02
+#define mmTD_PERFCOUNTER0_SELECT1                                               0xdb01
+#define mmTD_PERFCOUNTER0_LO                                                    0xd300
+#define mmTD_PERFCOUNTER1_LO                                                    0xd302
+#define mmTD_PERFCOUNTER0_HI                                                    0xd301
+#define mmTD_PERFCOUNTER1_HI                                                    0xd303
+#define mmTD_SCRATCH                                                            0x2533
+#define mmTA_CNTL                                                               0x2541
+#define mmTA_CNTL_AUX                                                           0x2542
+#define mmTA_RESERVED_010C                                                      0x2543
+#define mmTA_CS_BC_BASE_ADDR                                                    0xc380
+#define mmTA_CS_BC_BASE_ADDR_HI                                                 0xc381
+#define mmTA_STATUS                                                             0x2548
+#define mmTA_DEBUG_INDEX                                                        0x254c
+#define mmTA_DEBUG_DATA                                                         0x254d
+#define mmTA_PERFCOUNTER0_SELECT                                                0xdac0
+#define mmTA_PERFCOUNTER1_SELECT                                                0xdac2
+#define mmTA_PERFCOUNTER0_SELECT1                                               0xdac1
+#define mmTA_PERFCOUNTER0_LO                                                    0xd2c0
+#define mmTA_PERFCOUNTER1_LO                                                    0xd2c2
+#define mmTA_PERFCOUNTER0_HI                                                    0xd2c1
+#define mmTA_PERFCOUNTER1_HI                                                    0xd2c3
+#define mmTA_SCRATCH                                                            0x2564
+#define mmSH_HIDDEN_PRIVATE_BASE_VMID                                           0x2580
+#define mmSH_STATIC_MEM_CONFIG                                                  0x2581
+#define mmTCP_INVALIDATE                                                        0x2b00
+#define mmTCP_STATUS                                                            0x2b01
+#define mmTCP_CNTL                                                              0x2b02
+#define mmTCP_CHAN_STEER_LO                                                     0x2b03
+#define mmTCP_CHAN_STEER_HI                                                     0x2b04
+#define mmTCP_ADDR_CONFIG                                                       0x2b05
+#define mmTCP_CREDIT                                                            0x2b06
+#define mmTCP_PERFCOUNTER0_SELECT                                               0xdb40
+#define mmTCP_PERFCOUNTER1_SELECT                                               0xdb42
+#define mmTCP_PERFCOUNTER0_SELECT1                                              0xdb41
+#define mmTCP_PERFCOUNTER1_SELECT1                                              0xdb43
+#define mmTCP_PERFCOUNTER2_SELECT                                               0xdb44
+#define mmTCP_PERFCOUNTER3_SELECT                                               0xdb45
+#define mmTCP_PERFCOUNTER0_LO                                                   0xd340
+#define mmTCP_PERFCOUNTER1_LO                                                   0xd342
+#define mmTCP_PERFCOUNTER2_LO                                                   0xd344
+#define mmTCP_PERFCOUNTER3_LO                                                   0xd346
+#define mmTCP_PERFCOUNTER0_HI                                                   0xd341
+#define mmTCP_PERFCOUNTER1_HI                                                   0xd343
+#define mmTCP_PERFCOUNTER2_HI                                                   0xd345
+#define mmTCP_PERFCOUNTER3_HI                                                   0xd347
+#define mmTCP_BUFFER_ADDR_HASH_CNTL                                             0x2b16
+#define mmTCP_EDC_CNT                                                           0x2b17
+#define mmTC_CFG_L1_LOAD_POLICY0                                                0x2b1a
+#define mmTC_CFG_L1_LOAD_POLICY1                                                0x2b1b
+#define mmTC_CFG_L1_STORE_POLICY                                                0x2b1c
+#define mmTC_CFG_L2_LOAD_POLICY0                                                0x2b1d
+#define mmTC_CFG_L2_LOAD_POLICY1                                                0x2b1e
+#define mmTC_CFG_L2_STORE_POLICY0                                               0x2b1f
+#define mmTC_CFG_L2_STORE_POLICY1                                               0x2b20
+#define mmTC_CFG_L2_ATOMIC_POLICY                                               0x2b21
+#define mmTC_CFG_L1_VOLATILE                                                    0x2b22
+#define mmTC_CFG_L2_VOLATILE                                                    0x2b23
+#define mmTCP_WATCH0_ADDR_H                                                     0x32a0
+#define mmTCP_WATCH1_ADDR_H                                                     0x32a3
+#define mmTCP_WATCH2_ADDR_H                                                     0x32a6
+#define mmTCP_WATCH3_ADDR_H                                                     0x32a9
+#define mmTCP_WATCH0_ADDR_L                                                     0x32a1
+#define mmTCP_WATCH1_ADDR_L                                                     0x32a4
+#define mmTCP_WATCH2_ADDR_L                                                     0x32a7
+#define mmTCP_WATCH3_ADDR_L                                                     0x32aa
+#define mmTCP_WATCH0_CNTL                                                       0x32a2
+#define mmTCP_WATCH1_CNTL                                                       0x32a5
+#define mmTCP_WATCH2_CNTL                                                       0x32a8
+#define mmTCP_WATCH3_CNTL                                                       0x32ab
+#define mmTCP_GATCL1_CNTL                                                       0x32b0
+#define mmTCP_ATC_EDC_GATCL1_CNT                                                0x32b1
+#define mmTCP_GATCL1_DSM_CNTL                                                   0x32b2
+#define mmTCP_DSM_CNTL                                                          0x32b3
+#define mmTCP_CNTL2                                                             0x32b4
+#define mmTD_CGTT_CTRL                                                          0xf09c
+#define mmTA_CGTT_CTRL                                                          0xf09d
+#define mmCGTT_TCP_CLK_CTRL                                                     0xf09e
+#define mmCGTT_TCI_CLK_CTRL                                                     0xf09f
+#define mmTCI_STATUS                                                            0x2b61
+#define mmTCI_CNTL_1                                                            0x2b62
+#define mmTCI_CNTL_2                                                            0x2b63
+#define mmGDS_CONFIG                                                            0x25c0
+#define mmGDS_CNTL_STATUS                                                       0x25c1
+#define mmGDS_ENHANCE2                                                          0x25c2
+#define mmGDS_PROTECTION_FAULT                                                  0x25c3
+#define mmGDS_VM_PROTECTION_FAULT                                               0x25c4
+#define mmGDS_EDC_CNT                                                           0x25c5
+#define mmGDS_EDC_GRBM_CNT                                                      0x25c6
+#define mmGDS_EDC_OA_DED                                                        0x25c7
+#define mmGDS_DEBUG_CNTL                                                        0x25c8
+#define mmGDS_DEBUG_DATA                                                        0x25c9
+#define mmGDS_DSM_CNTL                                                          0x25ca
+#define mmCGTT_GDS_CLK_CTRL                                                     0xf0a0
+#define mmGDS_RD_ADDR                                                           0xc400
+#define mmGDS_RD_DATA                                                           0xc401
+#define mmGDS_RD_BURST_ADDR                                                     0xc402
+#define mmGDS_RD_BURST_COUNT                                                    0xc403
+#define mmGDS_RD_BURST_DATA                                                     0xc404
+#define mmGDS_WR_ADDR                                                           0xc405
+#define mmGDS_WR_DATA                                                           0xc406
+#define mmGDS_WR_BURST_ADDR                                                     0xc407
+#define mmGDS_WR_BURST_DATA                                                     0xc408
+#define mmGDS_WRITE_COMPLETE                                                    0xc409
+#define mmGDS_ATOM_CNTL                                                         0xc40a
+#define mmGDS_ATOM_COMPLETE                                                     0xc40b
+#define mmGDS_ATOM_BASE                                                         0xc40c
+#define mmGDS_ATOM_SIZE                                                         0xc40d
+#define mmGDS_ATOM_OFFSET0                                                      0xc40e
+#define mmGDS_ATOM_OFFSET1                                                      0xc40f
+#define mmGDS_ATOM_DST                                                          0xc410
+#define mmGDS_ATOM_OP                                                           0xc411
+#define mmGDS_ATOM_SRC0                                                         0xc412
+#define mmGDS_ATOM_SRC0_U                                                       0xc413
+#define mmGDS_ATOM_SRC1                                                         0xc414
+#define mmGDS_ATOM_SRC1_U                                                       0xc415
+#define mmGDS_ATOM_READ0                                                        0xc416
+#define mmGDS_ATOM_READ0_U                                                      0xc417
+#define mmGDS_ATOM_READ1                                                        0xc418
+#define mmGDS_ATOM_READ1_U                                                      0xc419
+#define mmGDS_GWS_RESOURCE_CNTL                                                 0xc41a
+#define mmGDS_GWS_RESOURCE                                                      0xc41b
+#define mmGDS_GWS_RESOURCE_CNT                                                  0xc41c
+#define mmGDS_OA_CNTL                                                           0xc41d
+#define mmGDS_OA_COUNTER                                                        0xc41e
+#define mmGDS_OA_ADDRESS                                                        0xc41f
+#define mmGDS_OA_INCDEC                                                         0xc420
+#define mmGDS_OA_RING_SIZE                                                      0xc421
+#define ixGDS_DEBUG_REG0                                                        0x0
+#define ixGDS_DEBUG_REG1                                                        0x1
+#define ixGDS_DEBUG_REG2                                                        0x2
+#define ixGDS_DEBUG_REG3                                                        0x3
+#define ixGDS_DEBUG_REG4                                                        0x4
+#define ixGDS_DEBUG_REG5                                                        0x5
+#define ixGDS_DEBUG_REG6                                                        0x6
+#define mmGDS_PERFCOUNTER0_SELECT                                               0xda80
+#define mmGDS_PERFCOUNTER1_SELECT                                               0xda81
+#define mmGDS_PERFCOUNTER2_SELECT                                               0xda82
+#define mmGDS_PERFCOUNTER3_SELECT                                               0xda83
+#define mmGDS_PERFCOUNTER0_LO                                                   0xd280
+#define mmGDS_PERFCOUNTER1_LO                                                   0xd282
+#define mmGDS_PERFCOUNTER2_LO                                                   0xd284
+#define mmGDS_PERFCOUNTER3_LO                                                   0xd286
+#define mmGDS_PERFCOUNTER0_HI                                                   0xd281
+#define mmGDS_PERFCOUNTER1_HI                                                   0xd283
+#define mmGDS_PERFCOUNTER2_HI                                                   0xd285
+#define mmGDS_PERFCOUNTER3_HI                                                   0xd287
+#define mmGDS_PERFCOUNTER0_SELECT1                                              0xda84
+#define mmGDS_VMID0_BASE                                                        0x3300
+#define mmGDS_VMID1_BASE                                                        0x3302
+#define mmGDS_VMID2_BASE                                                        0x3304
+#define mmGDS_VMID3_BASE                                                        0x3306
+#define mmGDS_VMID4_BASE                                                        0x3308
+#define mmGDS_VMID5_BASE                                                        0x330a
+#define mmGDS_VMID6_BASE                                                        0x330c
+#define mmGDS_VMID7_BASE                                                        0x330e
+#define mmGDS_VMID8_BASE                                                        0x3310
+#define mmGDS_VMID9_BASE                                                        0x3312
+#define mmGDS_VMID10_BASE                                                       0x3314
+#define mmGDS_VMID11_BASE                                                       0x3316
+#define mmGDS_VMID12_BASE                                                       0x3318
+#define mmGDS_VMID13_BASE                                                       0x331a
+#define mmGDS_VMID14_BASE                                                       0x331c
+#define mmGDS_VMID15_BASE                                                       0x331e
+#define mmGDS_VMID0_SIZE                                                        0x3301
+#define mmGDS_VMID1_SIZE                                                        0x3303
+#define mmGDS_VMID2_SIZE                                                        0x3305
+#define mmGDS_VMID3_SIZE                                                        0x3307
+#define mmGDS_VMID4_SIZE                                                        0x3309
+#define mmGDS_VMID5_SIZE                                                        0x330b
+#define mmGDS_VMID6_SIZE                                                        0x330d
+#define mmGDS_VMID7_SIZE                                                        0x330f
+#define mmGDS_VMID8_SIZE                                                        0x3311
+#define mmGDS_VMID9_SIZE                                                        0x3313
+#define mmGDS_VMID10_SIZE                                                       0x3315
+#define mmGDS_VMID11_SIZE                                                       0x3317
+#define mmGDS_VMID12_SIZE                                                       0x3319
+#define mmGDS_VMID13_SIZE                                                       0x331b
+#define mmGDS_VMID14_SIZE                                                       0x331d
+#define mmGDS_VMID15_SIZE                                                       0x331f
+#define mmGDS_GWS_VMID0                                                         0x3320
+#define mmGDS_GWS_VMID1                                                         0x3321
+#define mmGDS_GWS_VMID2                                                         0x3322
+#define mmGDS_GWS_VMID3                                                         0x3323
+#define mmGDS_GWS_VMID4                                                         0x3324
+#define mmGDS_GWS_VMID5                                                         0x3325
+#define mmGDS_GWS_VMID6                                                         0x3326
+#define mmGDS_GWS_VMID7                                                         0x3327
+#define mmGDS_GWS_VMID8                                                         0x3328
+#define mmGDS_GWS_VMID9                                                         0x3329
+#define mmGDS_GWS_VMID10                                                        0x332a
+#define mmGDS_GWS_VMID11                                                        0x332b
+#define mmGDS_GWS_VMID12                                                        0x332c
+#define mmGDS_GWS_VMID13                                                        0x332d
+#define mmGDS_GWS_VMID14                                                        0x332e
+#define mmGDS_GWS_VMID15                                                        0x332f
+#define mmGDS_OA_VMID0                                                          0x3330
+#define mmGDS_OA_VMID1                                                          0x3331
+#define mmGDS_OA_VMID2                                                          0x3332
+#define mmGDS_OA_VMID3                                                          0x3333
+#define mmGDS_OA_VMID4                                                          0x3334
+#define mmGDS_OA_VMID5                                                          0x3335
+#define mmGDS_OA_VMID6                                                          0x3336
+#define mmGDS_OA_VMID7                                                          0x3337
+#define mmGDS_OA_VMID8                                                          0x3338
+#define mmGDS_OA_VMID9                                                          0x3339
+#define mmGDS_OA_VMID10                                                         0x333a
+#define mmGDS_OA_VMID11                                                         0x333b
+#define mmGDS_OA_VMID12                                                         0x333c
+#define mmGDS_OA_VMID13                                                         0x333d
+#define mmGDS_OA_VMID14                                                         0x333e
+#define mmGDS_OA_VMID15                                                         0x333f
+#define mmGDS_GWS_RESET0                                                        0x3344
+#define mmGDS_GWS_RESET1                                                        0x3345
+#define mmGDS_GWS_RESOURCE_RESET                                                0x3346
+#define mmGDS_COMPUTE_MAX_WAVE_ID                                               0x3348
+#define mmGDS_OA_RESET_MASK                                                     0x3349
+#define mmGDS_OA_RESET                                                          0x334a
+#define mmGDS_ENHANCE                                                           0x334b
+#define mmGDS_OA_CGPG_RESTORE                                                   0x334c
+#define mmGDS_CS_CTXSW_STATUS                                                   0x334d
+#define mmGDS_CS_CTXSW_CNT0                                                     0x334e
+#define mmGDS_CS_CTXSW_CNT1                                                     0x334f
+#define mmGDS_CS_CTXSW_CNT2                                                     0x3350
+#define mmGDS_CS_CTXSW_CNT3                                                     0x3351
+#define mmGDS_GFX_CTXSW_STATUS                                                  0x3352
+#define mmGDS_VS_CTXSW_CNT0                                                     0x3353
+#define mmGDS_VS_CTXSW_CNT1                                                     0x3354
+#define mmGDS_VS_CTXSW_CNT2                                                     0x3355
+#define mmGDS_VS_CTXSW_CNT3                                                     0x3356
+#define mmGDS_PS0_CTXSW_CNT0                                                    0x3357
+#define mmGDS_PS1_CTXSW_CNT0                                                    0x335b
+#define mmGDS_PS2_CTXSW_CNT0                                                    0x335f
+#define mmGDS_PS3_CTXSW_CNT0                                                    0x3363
+#define mmGDS_PS4_CTXSW_CNT0                                                    0x3367
+#define mmGDS_PS5_CTXSW_CNT0                                                    0x336b
+#define mmGDS_PS6_CTXSW_CNT0                                                    0x336f
+#define mmGDS_PS7_CTXSW_CNT0                                                    0x3373
+#define mmGDS_PS0_CTXSW_CNT1                                                    0x3358
+#define mmGDS_PS1_CTXSW_CNT1                                                    0x335c
+#define mmGDS_PS2_CTXSW_CNT1                                                    0x3360
+#define mmGDS_PS3_CTXSW_CNT1                                                    0x3364
+#define mmGDS_PS4_CTXSW_CNT1                                                    0x3368
+#define mmGDS_PS5_CTXSW_CNT1                                                    0x336c
+#define mmGDS_PS6_CTXSW_CNT1                                                    0x3370
+#define mmGDS_PS7_CTXSW_CNT1                                                    0x3374
+#define mmGDS_PS0_CTXSW_CNT2                                                    0x3359
+#define mmGDS_PS1_CTXSW_CNT2                                                    0x335d
+#define mmGDS_PS2_CTXSW_CNT2                                                    0x3361
+#define mmGDS_PS3_CTXSW_CNT2                                                    0x3365
+#define mmGDS_PS4_CTXSW_CNT2                                                    0x3369
+#define mmGDS_PS5_CTXSW_CNT2                                                    0x336d
+#define mmGDS_PS6_CTXSW_CNT2                                                    0x3371
+#define mmGDS_PS7_CTXSW_CNT2                                                    0x3375
+#define mmGDS_PS0_CTXSW_CNT3                                                    0x335a
+#define mmGDS_PS1_CTXSW_CNT3                                                    0x335e
+#define mmGDS_PS2_CTXSW_CNT3                                                    0x3362
+#define mmGDS_PS3_CTXSW_CNT3                                                    0x3366
+#define mmGDS_PS4_CTXSW_CNT3                                                    0x336a
+#define mmGDS_PS5_CTXSW_CNT3                                                    0x336e
+#define mmGDS_PS6_CTXSW_CNT3                                                    0x3372
+#define mmGDS_PS7_CTXSW_CNT3                                                    0x3376
+#define mmCS_COPY_STATE                                                         0xa1f3
+#define mmGFX_COPY_STATE                                                        0xa1f4
+#define mmVGT_DRAW_INITIATOR                                                    0xa1fc
+#define mmVGT_EVENT_INITIATOR                                                   0xa2a4
+#define mmVGT_EVENT_ADDRESS_REG                                                 0xa1fe
+#define mmVGT_DMA_BASE_HI                                                       0xa1f9
+#define mmVGT_DMA_BASE                                                          0xa1fa
+#define mmVGT_DMA_INDEX_TYPE                                                    0xa29f
+#define mmVGT_DMA_NUM_INSTANCES                                                 0xa2a2
+#define mmIA_ENHANCE                                                            0xa29c
+#define mmVGT_DMA_SIZE                                                          0xa29d
+#define mmVGT_DMA_MAX_SIZE                                                      0xa29e
+#define mmVGT_DMA_PRIMITIVE_TYPE                                                0x2271
+#define mmVGT_DMA_CONTROL                                                       0x2272
+#define mmVGT_IMMED_DATA                                                        0xa1fd
+#define mmVGT_INDEX_TYPE                                                        0xc243
+#define mmVGT_NUM_INDICES                                                       0xc24c
+#define mmVGT_NUM_INSTANCES                                                     0xc24d
+#define mmVGT_PRIMITIVE_TYPE                                                    0xc242
+#define mmVGT_PRIMITIVEID_EN                                                    0xa2a1
+#define mmVGT_PRIMITIVEID_RESET                                                 0xa2a3
+#define mmVGT_VTX_CNT_EN                                                        0xa2ae
+#define mmVGT_REUSE_OFF                                                         0xa2ad
+#define mmVGT_INSTANCE_STEP_RATE_0                                              0xa2a8
+#define mmVGT_INSTANCE_STEP_RATE_1                                              0xa2a9
+#define mmVGT_MAX_VTX_INDX                                                      0xa100
+#define mmVGT_MIN_VTX_INDX                                                      0xa101
+#define mmVGT_INDX_OFFSET                                                       0xa102
+#define mmVGT_VERTEX_REUSE_BLOCK_CNTL                                           0xa316
+#define mmVGT_OUT_DEALLOC_CNTL                                                  0xa317
+#define mmVGT_MULTI_PRIM_IB_RESET_INDX                                          0xa103
+#define mmVGT_MULTI_PRIM_IB_RESET_EN                                            0xa2a5
+#define mmVGT_ENHANCE                                                           0xa294
+#define mmVGT_OUTPUT_PATH_CNTL                                                  0xa284
+#define mmVGT_HOS_CNTL                                                          0xa285
+#define mmVGT_HOS_MAX_TESS_LEVEL                                                0xa286
+#define mmVGT_HOS_MIN_TESS_LEVEL                                                0xa287
+#define mmVGT_HOS_REUSE_DEPTH                                                   0xa288
+#define mmVGT_GROUP_PRIM_TYPE                                                   0xa289
+#define mmVGT_GROUP_FIRST_DECR                                                  0xa28a
+#define mmVGT_GROUP_DECR                                                        0xa28b
+#define mmVGT_GROUP_VECT_0_CNTL                                                 0xa28c
+#define mmVGT_GROUP_VECT_1_CNTL                                                 0xa28d
+#define mmVGT_GROUP_VECT_0_FMT_CNTL                                             0xa28e
+#define mmVGT_GROUP_VECT_1_FMT_CNTL                                             0xa28f
+#define mmVGT_VTX_VECT_EJECT_REG                                                0x222c
+#define mmVGT_DMA_DATA_FIFO_DEPTH                                               0x222d
+#define mmVGT_DMA_REQ_FIFO_DEPTH                                                0x222e
+#define mmVGT_DRAW_INIT_FIFO_DEPTH                                              0x222f
+#define mmVGT_LAST_COPY_STATE                                                   0x2230
+#define mmCC_GC_SHADER_ARRAY_CONFIG                                             0x226f
+#define mmGC_USER_SHADER_ARRAY_CONFIG                                           0x2270
+#define mmVGT_GS_MODE                                                           0xa290
+#define mmVGT_GS_ONCHIP_CNTL                                                    0xa291
+#define mmVGT_GS_OUT_PRIM_TYPE                                                  0xa29b
+#define mmVGT_CACHE_INVALIDATION                                                0x2231
+#define mmVGT_RESET_DEBUG                                                       0x2232
+#define mmVGT_STRMOUT_DELAY                                                     0x2233
+#define mmVGT_FIFO_DEPTHS                                                       0x2234
+#define mmVGT_GS_PER_ES                                                         0xa295
+#define mmVGT_ES_PER_GS                                                         0xa296
+#define mmVGT_GS_PER_VS                                                         0xa297
+#define mmVGT_GS_VERTEX_REUSE                                                   0x2235
+#define mmVGT_MC_LAT_CNTL                                                       0x2236
+#define mmIA_CNTL_STATUS                                                        0x2237
+#define mmVGT_STRMOUT_CONFIG                                                    0xa2e5
+#define mmVGT_STRMOUT_BUFFER_SIZE_0                                             0xa2b4
+#define mmVGT_STRMOUT_BUFFER_SIZE_1                                             0xa2b8
+#define mmVGT_STRMOUT_BUFFER_SIZE_2                                             0xa2bc
+#define mmVGT_STRMOUT_BUFFER_SIZE_3                                             0xa2c0
+#define mmVGT_STRMOUT_BUFFER_OFFSET_0                                           0xa2b7
+#define mmVGT_STRMOUT_BUFFER_OFFSET_1                                           0xa2bb
+#define mmVGT_STRMOUT_BUFFER_OFFSET_2                                           0xa2bf
+#define mmVGT_STRMOUT_BUFFER_OFFSET_3                                           0xa2c3
+#define mmVGT_STRMOUT_VTX_STRIDE_0                                              0xa2b5
+#define mmVGT_STRMOUT_VTX_STRIDE_1                                              0xa2b9
+#define mmVGT_STRMOUT_VTX_STRIDE_2                                              0xa2bd
+#define mmVGT_STRMOUT_VTX_STRIDE_3                                              0xa2c1
+#define mmVGT_STRMOUT_BUFFER_CONFIG                                             0xa2e6
+#define mmVGT_STRMOUT_BUFFER_FILLED_SIZE_0                                      0xc244
+#define mmVGT_STRMOUT_BUFFER_FILLED_SIZE_1                                      0xc245
+#define mmVGT_STRMOUT_BUFFER_FILLED_SIZE_2                                      0xc246
+#define mmVGT_STRMOUT_BUFFER_FILLED_SIZE_3                                      0xc247
+#define mmVGT_STRMOUT_DRAW_OPAQUE_OFFSET                                        0xa2ca
+#define mmVGT_STRMOUT_DRAW_OPAQUE_BUFFER_FILLED_SIZE                            0xa2cb
+#define mmVGT_STRMOUT_DRAW_OPAQUE_VERTEX_STRIDE                                 0xa2cc
+#define mmVGT_GS_MAX_VERT_OUT                                                   0xa2ce
+#define mmVGT_SHADER_STAGES_EN                                                  0xa2d5
+#define mmVGT_DISPATCH_DRAW_INDEX                                               0xa2dd
+#define mmVGT_LS_HS_CONFIG                                                      0xa2d6
+#define mmVGT_DMA_LS_HS_CONFIG                                                  0x2273
+#define mmVGT_TF_PARAM                                                          0xa2db
+#define mmVGT_TESS_DISTRIBUTION                                                 0xa2d4
+#define mmVGT_TF_RING_SIZE                                                      0xc24e
+#define mmVGT_SYS_CONFIG                                                        0x2263
+#define mmVGT_HS_OFFCHIP_PARAM                                                  0xc24f
+#define mmVGT_TF_MEMORY_BASE                                                    0xc250
+#define mmVGT_GS_INSTANCE_CNT                                                   0xa2e4
+#define mmIA_MULTI_VGT_PARAM                                                    0xa2aa
+#define mmVGT_VS_MAX_WAVE_ID                                                    0x2268
+#define mmVGT_ESGS_RING_SIZE                                                    0xc240
+#define mmVGT_GSVS_RING_SIZE                                                    0xc241
+#define mmVGT_GSVS_RING_OFFSET_1                                                0xa298
+#define mmVGT_GSVS_RING_OFFSET_2                                                0xa299
+#define mmVGT_GSVS_RING_OFFSET_3                                                0xa29a
+#define mmVGT_ESGS_RING_ITEMSIZE                                                0xa2ab
+#define mmVGT_GSVS_RING_ITEMSIZE                                                0xa2ac
+#define mmVGT_GS_VERT_ITEMSIZE                                                  0xa2d7
+#define mmVGT_GS_VERT_ITEMSIZE_1                                                0xa2d8
+#define mmVGT_GS_VERT_ITEMSIZE_2                                                0xa2d9
+#define mmVGT_GS_VERT_ITEMSIZE_3                                                0xa2da
+#define mmWD_CNTL_STATUS                                                        0x223f
+#define mmWD_ENHANCE                                                            0xa2a0
+#define mmGFX_PIPE_CONTROL                                                      0x226d
+#define mmCGTT_VGT_CLK_CTRL                                                     0xf084
+#define mmCGTT_IA_CLK_CTRL                                                      0xf085
+#define mmCGTT_WD_CLK_CTRL                                                      0xf086
+#define mmVGT_DEBUG_CNTL                                                        0x2238
+#define mmVGT_DEBUG_DATA                                                        0x2239
+#define mmIA_DEBUG_CNTL                                                         0x223a
+#define mmIA_DEBUG_DATA                                                         0x223b
+#define mmVGT_CNTL_STATUS                                                       0x223c
+#define mmWD_DEBUG_CNTL                                                         0x223d
+#define mmWD_DEBUG_DATA                                                         0x223e
+#define mmWD_QOS                                                                0x2242
+#define mmCC_GC_PRIM_CONFIG                                                     0x2240
+#define mmGC_USER_PRIM_CONFIG                                                   0x2241
+#define ixWD_DEBUG_REG0                                                         0x0
+#define ixWD_DEBUG_REG1                                                         0x1
+#define ixWD_DEBUG_REG2                                                         0x2
+#define ixWD_DEBUG_REG3                                                         0x3
+#define ixWD_DEBUG_REG4                                                         0x4
+#define ixWD_DEBUG_REG5                                                         0x5
+#define ixWD_DEBUG_REG6                                                         0x6
+#define ixWD_DEBUG_REG7                                                         0x7
+#define ixWD_DEBUG_REG8                                                         0x8
+#define ixWD_DEBUG_REG9                                                         0x9
+#define ixWD_DEBUG_REG10                                                        0xa
+#define ixIA_DEBUG_REG0                                                         0x0
+#define ixIA_DEBUG_REG1                                                         0x1
+#define ixIA_DEBUG_REG2                                                         0x2
+#define ixIA_DEBUG_REG3                                                         0x3
+#define ixIA_DEBUG_REG4                                                         0x4
+#define ixIA_DEBUG_REG5                                                         0x5
+#define ixIA_DEBUG_REG6                                                         0x6
+#define ixIA_DEBUG_REG7                                                         0x7
+#define ixIA_DEBUG_REG8                                                         0x8
+#define ixIA_DEBUG_REG9                                                         0x9
+#define ixVGT_DEBUG_REG0                                                        0x0
+#define ixVGT_DEBUG_REG1                                                        0x1
+#define ixVGT_DEBUG_REG2                                                        0x1e
+#define ixVGT_DEBUG_REG3                                                        0x1f
+#define ixVGT_DEBUG_REG4                                                        0x20
+#define ixVGT_DEBUG_REG5                                                        0x21
+#define ixVGT_DEBUG_REG6                                                        0x22
+#define ixVGT_DEBUG_REG7                                                        0x23
+#define ixVGT_DEBUG_REG8                                                        0x8
+#define ixVGT_DEBUG_REG9                                                        0x9
+#define ixVGT_DEBUG_REG10                                                       0xa
+#define ixVGT_DEBUG_REG11                                                       0xb
+#define ixVGT_DEBUG_REG12                                                       0xc
+#define ixVGT_DEBUG_REG13                                                       0xd
+#define ixVGT_DEBUG_REG14                                                       0xe
+#define ixVGT_DEBUG_REG15                                                       0xf
+#define ixVGT_DEBUG_REG16                                                       0x10
+#define ixVGT_DEBUG_REG17                                                       0x11
+#define ixVGT_DEBUG_REG18                                                       0x7
+#define ixVGT_DEBUG_REG19                                                       0x13
+#define ixVGT_DEBUG_REG20                                                       0x14
+#define ixVGT_DEBUG_REG21                                                       0x15
+#define ixVGT_DEBUG_REG22                                                       0x16
+#define ixVGT_DEBUG_REG23                                                       0x17
+#define ixVGT_DEBUG_REG24                                                       0x18
+#define ixVGT_DEBUG_REG25                                                       0x19
+#define ixVGT_DEBUG_REG26                                                       0x24
+#define ixVGT_DEBUG_REG27                                                       0x1b
+#define ixVGT_DEBUG_REG28                                                       0x1c
+#define ixVGT_DEBUG_REG29                                                       0x1d
+#define ixVGT_DEBUG_REG31                                                       0x26
+#define ixVGT_DEBUG_REG32                                                       0x27
+#define ixVGT_DEBUG_REG33                                                       0x28
+#define ixVGT_DEBUG_REG34                                                       0x29
+#define ixVGT_DEBUG_REG36                                                       0x2b
+#define mmVGT_PERFCOUNTER_SEID_MASK                                             0xd894
+#define mmVGT_PERFCOUNTER0_SELECT                                               0xd88c
+#define mmVGT_PERFCOUNTER1_SELECT                                               0xd88d
+#define mmVGT_PERFCOUNTER2_SELECT                                               0xd88e
+#define mmVGT_PERFCOUNTER3_SELECT                                               0xd88f
+#define mmVGT_PERFCOUNTER0_SELECT1                                              0xd890
+#define mmVGT_PERFCOUNTER1_SELECT1                                              0xd891
+#define mmVGT_PERFCOUNTER0_LO                                                   0xd090
+#define mmVGT_PERFCOUNTER1_LO                                                   0xd092
+#define mmVGT_PERFCOUNTER2_LO                                                   0xd094
+#define mmVGT_PERFCOUNTER3_LO                                                   0xd096
+#define mmVGT_PERFCOUNTER0_HI                                                   0xd091
+#define mmVGT_PERFCOUNTER1_HI                                                   0xd093
+#define mmVGT_PERFCOUNTER2_HI                                                   0xd095
+#define mmVGT_PERFCOUNTER3_HI                                                   0xd097
+#define mmIA_PERFCOUNTER0_SELECT                                                0xd884
+#define mmIA_PERFCOUNTER1_SELECT                                                0xd885
+#define mmIA_PERFCOUNTER2_SELECT                                                0xd886
+#define mmIA_PERFCOUNTER3_SELECT                                                0xd887
+#define mmIA_PERFCOUNTER0_SELECT1                                               0xd888
+#define mmIA_PERFCOUNTER0_LO                                                    0xd088
+#define mmIA_PERFCOUNTER1_LO                                                    0xd08a
+#define mmIA_PERFCOUNTER2_LO                                                    0xd08c
+#define mmIA_PERFCOUNTER3_LO                                                    0xd08e
+#define mmIA_PERFCOUNTER0_HI                                                    0xd089
+#define mmIA_PERFCOUNTER1_HI                                                    0xd08b
+#define mmIA_PERFCOUNTER2_HI                                                    0xd08d
+#define mmIA_PERFCOUNTER3_HI                                                    0xd08f
+#define mmWD_PERFCOUNTER0_SELECT                                                0xd880
+#define mmWD_PERFCOUNTER1_SELECT                                                0xd881
+#define mmWD_PERFCOUNTER2_SELECT                                                0xd882
+#define mmWD_PERFCOUNTER3_SELECT                                                0xd883
+#define mmWD_PERFCOUNTER0_LO                                                    0xd080
+#define mmWD_PERFCOUNTER1_LO                                                    0xd082
+#define mmWD_PERFCOUNTER2_LO                                                    0xd084
+#define mmWD_PERFCOUNTER3_LO                                                    0xd086
+#define mmWD_PERFCOUNTER0_HI                                                    0xd081
+#define mmWD_PERFCOUNTER1_HI                                                    0xd083
+#define mmWD_PERFCOUNTER2_HI                                                    0xd085
+#define mmWD_PERFCOUNTER3_HI                                                    0xd087
+#define mmDIDT_IND_INDEX                                                        0x3280
+#define mmDIDT_IND_DATA                                                         0x3281
+#define ixDIDT_SQ_CTRL0                                                         0x0
+#define ixDIDT_SQ_CTRL1                                                         0x1
+#define ixDIDT_SQ_CTRL2                                                         0x2
+#define ixDIDT_SQ_CTRL_OCP                                                      0x3
+#define ixDIDT_SQ_WEIGHT0_3                                                     0x10
+#define ixDIDT_SQ_WEIGHT4_7                                                     0x11
+#define ixDIDT_SQ_WEIGHT8_11                                                    0x12
+#define ixDIDT_DB_CTRL0                                                         0x20
+#define ixDIDT_DB_CTRL1                                                         0x21
+#define ixDIDT_DB_CTRL2                                                         0x22
+#define ixDIDT_DB_CTRL_OCP                                                      0x23
+#define ixDIDT_DB_WEIGHT0_3                                                     0x30
+#define ixDIDT_DB_WEIGHT4_7                                                     0x31
+#define ixDIDT_DB_WEIGHT8_11                                                    0x32
+#define ixDIDT_TD_CTRL0                                                         0x40
+#define ixDIDT_TD_CTRL1                                                         0x41
+#define ixDIDT_TD_CTRL2                                                         0x42
+#define ixDIDT_TD_CTRL_OCP                                                      0x43
+#define ixDIDT_TD_WEIGHT0_3                                                     0x50
+#define ixDIDT_TD_WEIGHT4_7                                                     0x51
+#define ixDIDT_TD_WEIGHT8_11                                                    0x52
+#define ixDIDT_TCP_CTRL0                                                        0x60
+#define ixDIDT_TCP_CTRL1                                                        0x61
+#define ixDIDT_TCP_CTRL2                                                        0x62
+#define ixDIDT_TCP_CTRL_OCP                                                     0x63
+#define ixDIDT_TCP_WEIGHT0_3                                                    0x70
+#define ixDIDT_TCP_WEIGHT4_7                                                    0x71
+#define ixDIDT_TCP_WEIGHT8_11                                                   0x72
+#define ixDIDT_DBR_CTRL0                                                        0x80
+#define ixDIDT_DBR_CTRL1                                                        0x81
+#define ixDIDT_DBR_CTRL2                                                        0x82
+#define ixDIDT_DBR_CTRL_OCP                                                     0x83
+#define ixDIDT_DBR_WEIGHT0_3                                                    0x90
+#define ixDIDT_DBR_WEIGHT4_7                                                    0x91
+#define ixDIDT_DBR_WEIGHT8_11                                                   0x92
+
+#endif /* GFX_8_1_D_H */
diff --git a/drivers/gpu/drm/amd/include/asic_reg/gca/gfx_8_1_enum.h b/drivers/gpu/drm/amd/include/asic_reg/gca/gfx_8_1_enum.h
new file mode 100644
index 0000000..f902209
--- /dev/null
+++ b/drivers/gpu/drm/amd/include/asic_reg/gca/gfx_8_1_enum.h
@@ -0,0 +1,6808 @@
+/*
+ * GFX_8_1 Register documentation
+ *
+ * Copyright (C) 2014  Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included
+ * in all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
+ * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN
+ * AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#ifndef GFX_8_1_ENUM_H
+#define GFX_8_1_ENUM_H
+
+typedef enum SurfaceNumber {
+	NUMBER_UNORM                                     = 0x0,
+	NUMBER_SNORM                                     = 0x1,
+	NUMBER_USCALED                                   = 0x2,
+	NUMBER_SSCALED                                   = 0x3,
+	NUMBER_UINT                                      = 0x4,
+	NUMBER_SINT                                      = 0x5,
+	NUMBER_SRGB                                      = 0x6,
+	NUMBER_FLOAT                                     = 0x7,
+} SurfaceNumber;
+typedef enum SurfaceSwap {
+	SWAP_STD                                         = 0x0,
+	SWAP_ALT                                         = 0x1,
+	SWAP_STD_REV                                     = 0x2,
+	SWAP_ALT_REV                                     = 0x3,
+} SurfaceSwap;
+typedef enum CBMode {
+	CB_DISABLE                                       = 0x0,
+	CB_NORMAL                                        = 0x1,
+	CB_ELIMINATE_FAST_CLEAR                          = 0x2,
+	CB_RESOLVE                                       = 0x3,
+	CB_DECOMPRESS                                    = 0x4,
+	CB_FMASK_DECOMPRESS                              = 0x5,
+	CB_DCC_DECOMPRESS                                = 0x6,
+} CBMode;
+typedef enum RoundMode {
+	ROUND_BY_HALF                                    = 0x0,
+	ROUND_TRUNCATE                                   = 0x1,
+} RoundMode;
+typedef enum SourceFormat {
+	EXPORT_4C_32BPC                                  = 0x0,
+	EXPORT_4C_16BPC                                  = 0x1,
+	EXPORT_2C_32BPC_GR                               = 0x2,
+	EXPORT_2C_32BPC_AR                               = 0x3,
+} SourceFormat;
+typedef enum BlendOp {
+	BLEND_ZERO                                       = 0x0,
+	BLEND_ONE                                        = 0x1,
+	BLEND_SRC_COLOR                                  = 0x2,
+	BLEND_ONE_MINUS_SRC_COLOR                        = 0x3,
+	BLEND_SRC_ALPHA                                  = 0x4,
+	BLEND_ONE_MINUS_SRC_ALPHA                        = 0x5,
+	BLEND_DST_ALPHA                                  = 0x6,
+	BLEND_ONE_MINUS_DST_ALPHA                        = 0x7,
+	BLEND_DST_COLOR                                  = 0x8,
+	BLEND_ONE_MINUS_DST_COLOR                        = 0x9,
+	BLEND_SRC_ALPHA_SATURATE                         = 0xa,
+	BLEND_BOTH_SRC_ALPHA                             = 0xb,
+	BLEND_BOTH_INV_SRC_ALPHA                         = 0xc,
+	BLEND_CONSTANT_COLOR                             = 0xd,
+	BLEND_ONE_MINUS_CONSTANT_COLOR                   = 0xe,
+	BLEND_SRC1_COLOR                                 = 0xf,
+	BLEND_INV_SRC1_COLOR                             = 0x10,
+	BLEND_SRC1_ALPHA                                 = 0x11,
+	BLEND_INV_SRC1_ALPHA                             = 0x12,
+	BLEND_CONSTANT_ALPHA                             = 0x13,
+	BLEND_ONE_MINUS_CONSTANT_ALPHA                   = 0x14,
+} BlendOp;
+typedef enum CombFunc {
+	COMB_DST_PLUS_SRC                                = 0x0,
+	COMB_SRC_MINUS_DST                               = 0x1,
+	COMB_MIN_DST_SRC                                 = 0x2,
+	COMB_MAX_DST_SRC                                 = 0x3,
+	COMB_DST_MINUS_SRC                               = 0x4,
+} CombFunc;
+typedef enum BlendOpt {
+	FORCE_OPT_AUTO                                   = 0x0,
+	FORCE_OPT_DISABLE                                = 0x1,
+	FORCE_OPT_ENABLE_IF_SRC_A_0                      = 0x2,
+	FORCE_OPT_ENABLE_IF_SRC_RGB_0                    = 0x3,
+	FORCE_OPT_ENABLE_IF_SRC_ARGB_0                   = 0x4,
+	FORCE_OPT_ENABLE_IF_SRC_A_1                      = 0x5,
+	FORCE_OPT_ENABLE_IF_SRC_RGB_1                    = 0x6,
+	FORCE_OPT_ENABLE_IF_SRC_ARGB_1                   = 0x7,
+} BlendOpt;
+typedef enum CmaskCode {
+	CMASK_CLR00_F0                                   = 0x0,
+	CMASK_CLR00_F1                                   = 0x1,
+	CMASK_CLR00_F2                                   = 0x2,
+	CMASK_CLR00_FX                                   = 0x3,
+	CMASK_CLR01_F0                                   = 0x4,
+	CMASK_CLR01_F1                                   = 0x5,
+	CMASK_CLR01_F2                                   = 0x6,
+	CMASK_CLR01_FX                                   = 0x7,
+	CMASK_CLR10_F0                                   = 0x8,
+	CMASK_CLR10_F1                                   = 0x9,
+	CMASK_CLR10_F2                                   = 0xa,
+	CMASK_CLR10_FX                                   = 0xb,
+	CMASK_CLR11_F0                                   = 0xc,
+	CMASK_CLR11_F1                                   = 0xd,
+	CMASK_CLR11_F2                                   = 0xe,
+	CMASK_CLR11_FX                                   = 0xf,
+} CmaskCode;
+typedef enum CmaskAddr {
+	CMASK_ADDR_TILED                                 = 0x0,
+	CMASK_ADDR_LINEAR                                = 0x1,
+	CMASK_ADDR_COMPATIBLE                            = 0x2,
+} CmaskAddr;
+typedef enum CBPerfSel {
+	CB_PERF_SEL_NONE                                 = 0x0,
+	CB_PERF_SEL_BUSY                                 = 0x1,
+	CB_PERF_SEL_CORE_SCLK_VLD                        = 0x2,
+	CB_PERF_SEL_REG_SCLK0_VLD                        = 0x3,
+	CB_PERF_SEL_REG_SCLK1_VLD                        = 0x4,
+	CB_PERF_SEL_DRAWN_QUAD                           = 0x5,
+	CB_PERF_SEL_DRAWN_PIXEL                          = 0x6,
+	CB_PERF_SEL_DRAWN_QUAD_FRAGMENT                  = 0x7,
+	CB_PERF_SEL_DRAWN_TILE                           = 0x8,
+	CB_PERF_SEL_DB_CB_TILE_VALID_READY               = 0x9,
+	CB_PERF_SEL_DB_CB_TILE_VALID_READYB              = 0xa,
+	CB_PERF_SEL_DB_CB_TILE_VALIDB_READY              = 0xb,
+	CB_PERF_SEL_DB_CB_TILE_VALIDB_READYB             = 0xc,
+	CB_PERF_SEL_CM_FC_TILE_VALID_READY               = 0xd,
+	CB_PERF_SEL_CM_FC_TILE_VALID_READYB              = 0xe,
+	CB_PERF_SEL_CM_FC_TILE_VALIDB_READY              = 0xf,
+	CB_PERF_SEL_CM_FC_TILE_VALIDB_READYB             = 0x10,
+	CB_PERF_SEL_MERGE_TILE_ONLY_VALID_READY          = 0x11,
+	CB_PERF_SEL_MERGE_TILE_ONLY_VALID_READYB         = 0x12,
+	CB_PERF_SEL_DB_CB_LQUAD_VALID_READY              = 0x13,
+	CB_PERF_SEL_DB_CB_LQUAD_VALID_READYB             = 0x14,
+	CB_PERF_SEL_DB_CB_LQUAD_VALIDB_READY             = 0x15,
+	CB_PERF_SEL_DB_CB_LQUAD_VALIDB_READYB            = 0x16,
+	CB_PERF_SEL_LQUAD_NO_TILE                        = 0x17,
+	CB_PERF_SEL_LQUAD_FORMAT_IS_EXPORT_32_R          = 0x18,
+	CB_PERF_SEL_LQUAD_FORMAT_IS_EXPORT_32_AR         = 0x19,
+	CB_PERF_SEL_LQUAD_FORMAT_IS_EXPORT_32_GR         = 0x1a,
+	CB_PERF_SEL_LQUAD_FORMAT_IS_EXPORT_32_ABGR       = 0x1b,
+	CB_PERF_SEL_LQUAD_FORMAT_IS_EXPORT_FP16_ABGR     = 0x1c,
+	CB_PERF_SEL_LQUAD_FORMAT_IS_EXPORT_SIGNED16_ABGR = 0x1d,
+	CB_PERF_SEL_LQUAD_FORMAT_IS_EXPORT_UNSIGNED16_ABGR= 0x1e,
+	CB_PERF_SEL_QUAD_KILLED_BY_EXTRA_PIXEL_EXPORT    = 0x1f,
+	CB_PERF_SEL_QUAD_KILLED_BY_COLOR_INVALID         = 0x20,
+	CB_PERF_SEL_QUAD_KILLED_BY_NULL_TARGET_SHADER_MASK= 0x21,
+	CB_PERF_SEL_QUAD_KILLED_BY_NULL_SAMPLE_MASK      = 0x22,
+	CB_PERF_SEL_QUAD_KILLED_BY_DISCARD_PIXEL         = 0x23,
+	CB_PERF_SEL_FC_CLEAR_QUAD_VALID_READY            = 0x24,
+	CB_PERF_SEL_FC_CLEAR_QUAD_VALID_READYB           = 0x25,
+	CB_PERF_SEL_FC_CLEAR_QUAD_VALIDB_READY           = 0x26,
+	CB_PERF_SEL_FC_CLEAR_QUAD_VALIDB_READYB          = 0x27,
+	CB_PERF_SEL_FOP_IN_VALID_READY                   = 0x28,
+	CB_PERF_SEL_FOP_IN_VALID_READYB                  = 0x29,
+	CB_PERF_SEL_FOP_IN_VALIDB_READY                  = 0x2a,
+	CB_PERF_SEL_FOP_IN_VALIDB_READYB                 = 0x2b,
+	CB_PERF_SEL_FC_CC_QUADFRAG_VALID_READY           = 0x2c,
+	CB_PERF_SEL_FC_CC_QUADFRAG_VALID_READYB          = 0x2d,
+	CB_PERF_SEL_FC_CC_QUADFRAG_VALIDB_READY          = 0x2e,
+	CB_PERF_SEL_FC_CC_QUADFRAG_VALIDB_READYB         = 0x2f,
+	CB_PERF_SEL_CC_IB_SR_FRAG_VALID_READY            = 0x30,
+	CB_PERF_SEL_CC_IB_SR_FRAG_VALID_READYB           = 0x31,
+	CB_PERF_SEL_CC_IB_SR_FRAG_VALIDB_READY           = 0x32,
+	CB_PERF_SEL_CC_IB_SR_FRAG_VALIDB_READYB          = 0x33,
+	CB_PERF_SEL_CC_IB_TB_FRAG_VALID_READY            = 0x34,
+	CB_PERF_SEL_CC_IB_TB_FRAG_VALID_READYB           = 0x35,
+	CB_PERF_SEL_CC_IB_TB_FRAG_VALIDB_READY           = 0x36,
+	CB_PERF_SEL_CC_IB_TB_FRAG_VALIDB_READYB          = 0x37,
+	CB_PERF_SEL_CC_RB_BC_EVENFRAG_VALID_READY        = 0x38,
+	CB_PERF_SEL_CC_RB_BC_EVENFRAG_VALID_READYB       = 0x39,
+	CB_PERF_SEL_CC_RB_BC_EVENFRAG_VALIDB_READY       = 0x3a,
+	CB_PERF_SEL_CC_RB_BC_EVENFRAG_VALIDB_READYB      = 0x3b,
+	CB_PERF_SEL_CC_RB_BC_ODDFRAG_VALID_READY         = 0x3c,
+	CB_PERF_SEL_CC_RB_BC_ODDFRAG_VALID_READYB        = 0x3d,
+	CB_PERF_SEL_CC_RB_BC_ODDFRAG_VALIDB_READY        = 0x3e,
+	CB_PERF_SEL_CC_RB_BC_ODDFRAG_VALIDB_READYB       = 0x3f,
+	CB_PERF_SEL_CC_BC_CS_FRAG_VALID                  = 0x40,
+	CB_PERF_SEL_CM_CACHE_HIT                         = 0x41,
+	CB_PERF_SEL_CM_CACHE_TAG_MISS                    = 0x42,
+	CB_PERF_SEL_CM_CACHE_SECTOR_MISS                 = 0x43,
+	CB_PERF_SEL_CM_CACHE_REEVICTION_STALL            = 0x44,
+	CB_PERF_SEL_CM_CACHE_EVICT_NONZERO_INFLIGHT_STALL= 0x45,
+	CB_PERF_SEL_CM_CACHE_REPLACE_PENDING_EVICT_STALL = 0x46,
+	CB_PERF_SEL_CM_CACHE_INFLIGHT_COUNTER_MAXIMUM_STALL= 0x47,
+	CB_PERF_SEL_CM_CACHE_READ_OUTPUT_STALL           = 0x48,
+	CB_PERF_SEL_CM_CACHE_WRITE_OUTPUT_STALL          = 0x49,
+	CB_PERF_SEL_CM_CACHE_ACK_OUTPUT_STALL            = 0x4a,
+	CB_PERF_SEL_CM_CACHE_STALL                       = 0x4b,
+	CB_PERF_SEL_CM_CACHE_FLUSH                       = 0x4c,
+	CB_PERF_SEL_CM_CACHE_TAGS_FLUSHED                = 0x4d,
+	CB_PERF_SEL_CM_CACHE_SECTORS_FLUSHED             = 0x4e,
+	CB_PERF_SEL_CM_CACHE_DIRTY_SECTORS_FLUSHED       = 0x4f,
+	CB_PERF_SEL_FC_CACHE_HIT                         = 0x50,
+	CB_PERF_SEL_FC_CACHE_TAG_MISS                    = 0x51,
+	CB_PERF_SEL_FC_CACHE_SECTOR_MISS                 = 0x52,
+	CB_PERF_SEL_FC_CACHE_REEVICTION_STALL            = 0x53,
+	CB_PERF_SEL_FC_CACHE_EVICT_NONZERO_INFLIGHT_STALL= 0x54,
+	CB_PERF_SEL_FC_CACHE_REPLACE_PENDING_EVICT_STALL = 0x55,
+	CB_PERF_SEL_FC_CACHE_INFLIGHT_COUNTER_MAXIMUM_STALL= 0x56,
+	CB_PERF_SEL_FC_CACHE_READ_OUTPUT_STALL           = 0x57,
+	CB_PERF_SEL_FC_CACHE_WRITE_OUTPUT_STALL          = 0x58,
+	CB_PERF_SEL_FC_CACHE_ACK_OUTPUT_STALL            = 0x59,
+	CB_PERF_SEL_FC_CACHE_STALL                       = 0x5a,
+	CB_PERF_SEL_FC_CACHE_FLUSH                       = 0x5b,
+	CB_PERF_SEL_FC_CACHE_TAGS_FLUSHED                = 0x5c,
+	CB_PERF_SEL_FC_CACHE_SECTORS_FLUSHED             = 0x5d,
+	CB_PERF_SEL_FC_CACHE_DIRTY_SECTORS_FLUSHED       = 0x5e,
+	CB_PERF_SEL_CC_CACHE_HIT                         = 0x5f,
+	CB_PERF_SEL_CC_CACHE_TAG_MISS                    = 0x60,
+	CB_PERF_SEL_CC_CACHE_SECTOR_MISS                 = 0x61,
+	CB_PERF_SEL_CC_CACHE_REEVICTION_STALL            = 0x62,
+	CB_PERF_SEL_CC_CACHE_EVICT_NONZERO_INFLIGHT_STALL= 0x63,
+	CB_PERF_SEL_CC_CACHE_REPLACE_PENDING_EVICT_STALL = 0x64,
+	CB_PERF_SEL_CC_CACHE_INFLIGHT_COUNTER_MAXIMUM_STALL= 0x65,
+	CB_PERF_SEL_CC_CACHE_READ_OUTPUT_STALL           = 0x66,
+	CB_PERF_SEL_CC_CACHE_WRITE_OUTPUT_STALL          = 0x67,
+	CB_PERF_SEL_CC_CACHE_ACK_OUTPUT_STALL            = 0x68,
+	CB_PERF_SEL_CC_CACHE_STALL                       = 0x69,
+	CB_PERF_SEL_CC_CACHE_FLUSH                       = 0x6a,
+	CB_PERF_SEL_CC_CACHE_TAGS_FLUSHED                = 0x6b,
+	CB_PERF_SEL_CC_CACHE_SECTORS_FLUSHED             = 0x6c,
+	CB_PERF_SEL_CC_CACHE_DIRTY_SECTORS_FLUSHED       = 0x6d,
+	CB_PERF_SEL_CC_CACHE_WA_TO_RMW_CONVERSION        = 0x6e,
+	CB_PERF_SEL_CC_CACHE_READS_SAVED_DUE_TO_DCC      = 0x6f,
+	CB_PERF_SEL_CB_TAP_WRREQ_VALID_READY             = 0x70,
+	CB_PERF_SEL_CB_TAP_WRREQ_VALID_READYB            = 0x71,
+	CB_PERF_SEL_CB_TAP_WRREQ_VALIDB_READY            = 0x72,
+	CB_PERF_SEL_CB_TAP_WRREQ_VALIDB_READYB           = 0x73,
+	CB_PERF_SEL_CM_MC_WRITE_REQUEST                  = 0x74,
+	CB_PERF_SEL_FC_MC_WRITE_REQUEST                  = 0x75,
+	CB_PERF_SEL_CC_MC_WRITE_REQUEST                  = 0x76,
+	CB_PERF_SEL_CM_MC_WRITE_REQUESTS_IN_FLIGHT       = 0x77,
+	CB_PERF_SEL_FC_MC_WRITE_REQUESTS_IN_FLIGHT       = 0x78,
+	CB_PERF_SEL_CC_MC_WRITE_REQUESTS_IN_FLIGHT       = 0x79,
+	CB_PERF_SEL_CB_TAP_RDREQ_VALID_READY             = 0x7a,
+	CB_PERF_SEL_CB_TAP_RDREQ_VALID_READYB            = 0x7b,
+	CB_PERF_SEL_CB_TAP_RDREQ_VALIDB_READY            = 0x7c,
+	CB_PERF_SEL_CB_TAP_RDREQ_VALIDB_READYB           = 0x7d,
+	CB_PERF_SEL_CM_MC_READ_REQUEST                   = 0x7e,
+	CB_PERF_SEL_FC_MC_READ_REQUEST                   = 0x7f,
+	CB_PERF_SEL_CC_MC_READ_REQUEST                   = 0x80,
+	CB_PERF_SEL_CM_MC_READ_REQUESTS_IN_FLIGHT        = 0x81,
+	CB_PERF_SEL_FC_MC_READ_REQUESTS_IN_FLIGHT        = 0x82,
+	CB_PERF_SEL_CC_MC_READ_REQUESTS_IN_FLIGHT        = 0x83,
+	CB_PERF_SEL_CM_TQ_FULL                           = 0x84,
+	CB_PERF_SEL_CM_TQ_FIFO_TILE_RESIDENCY_STALL      = 0x85,
+	CB_PERF_SEL_FC_QUAD_RDLAT_FIFO_FULL              = 0x86,
+	CB_PERF_SEL_FC_TILE_RDLAT_FIFO_FULL              = 0x87,
+	CB_PERF_SEL_FC_RDLAT_FIFO_QUAD_RESIDENCY_STALL   = 0x88,
+	CB_PERF_SEL_FOP_FMASK_RAW_STALL                  = 0x89,
+	CB_PERF_SEL_FOP_FMASK_BYPASS_STALL               = 0x8a,
+	CB_PERF_SEL_CC_SF_FULL                           = 0x8b,
+	CB_PERF_SEL_CC_RB_FULL                           = 0x8c,
+	CB_PERF_SEL_CC_EVENFIFO_QUAD_RESIDENCY_STALL     = 0x8d,
+	CB_PERF_SEL_CC_ODDFIFO_QUAD_RESIDENCY_STALL      = 0x8e,
+	CB_PERF_SEL_BLENDER_RAW_HAZARD_STALL             = 0x8f,
+	CB_PERF_SEL_EVENT                                = 0x90,
+	CB_PERF_SEL_EVENT_CACHE_FLUSH_TS                 = 0x91,
+	CB_PERF_SEL_EVENT_CONTEXT_DONE                   = 0x92,
+	CB_PERF_SEL_EVENT_CACHE_FLUSH                    = 0x93,
+	CB_PERF_SEL_EVENT_CACHE_FLUSH_AND_INV_TS_EVENT   = 0x94,
+	CB_PERF_SEL_EVENT_CACHE_FLUSH_AND_INV_EVENT      = 0x95,
+	CB_PERF_SEL_EVENT_FLUSH_AND_INV_CB_DATA_TS       = 0x96,
+	CB_PERF_SEL_EVENT_FLUSH_AND_INV_CB_META          = 0x97,
+	CB_PERF_SEL_CC_SURFACE_SYNC                      = 0x98,
+	CB_PERF_SEL_CMASK_READ_DATA_0xC                  = 0x99,
+	CB_PERF_SEL_CMASK_READ_DATA_0xD                  = 0x9a,
+	CB_PERF_SEL_CMASK_READ_DATA_0xE                  = 0x9b,
+	CB_PERF_SEL_CMASK_READ_DATA_0xF                  = 0x9c,
+	CB_PERF_SEL_CMASK_WRITE_DATA_0xC                 = 0x9d,
+	CB_PERF_SEL_CMASK_WRITE_DATA_0xD                 = 0x9e,
+	CB_PERF_SEL_CMASK_WRITE_DATA_0xE                 = 0x9f,
+	CB_PERF_SEL_CMASK_WRITE_DATA_0xF                 = 0xa0,
+	CB_PERF_SEL_TWO_PROBE_QUAD_FRAGMENT              = 0xa1,
+	CB_PERF_SEL_EXPORT_32_ABGR_QUAD_FRAGMENT         = 0xa2,
+	CB_PERF_SEL_DUAL_SOURCE_COLOR_QUAD_FRAGMENT      = 0xa3,
+	CB_PERF_SEL_QUAD_HAS_1_FRAGMENT_BEFORE_UPDATE    = 0xa4,
+	CB_PERF_SEL_QUAD_HAS_2_FRAGMENTS_BEFORE_UPDATE   = 0xa5,
+	CB_PERF_SEL_QUAD_HAS_3_FRAGMENTS_BEFORE_UPDATE   = 0xa6,
+	CB_PERF_SEL_QUAD_HAS_4_FRAGMENTS_BEFORE_UPDATE   = 0xa7,
+	CB_PERF_SEL_QUAD_HAS_5_FRAGMENTS_BEFORE_UPDATE   = 0xa8,
+	CB_PERF_SEL_QUAD_HAS_6_FRAGMENTS_BEFORE_UPDATE   = 0xa9,
+	CB_PERF_SEL_QUAD_HAS_7_FRAGMENTS_BEFORE_UPDATE   = 0xaa,
+	CB_PERF_SEL_QUAD_HAS_8_FRAGMENTS_BEFORE_UPDATE   = 0xab,
+	CB_PERF_SEL_QUAD_HAS_1_FRAGMENT_AFTER_UPDATE     = 0xac,
+	CB_PERF_SEL_QUAD_HAS_2_FRAGMENTS_AFTER_UPDATE    = 0xad,
+	CB_PERF_SEL_QUAD_HAS_3_FRAGMENTS_AFTER_UPDATE    = 0xae,
+	CB_PERF_SEL_QUAD_HAS_4_FRAGMENTS_AFTER_UPDATE    = 0xaf,
+	CB_PERF_SEL_QUAD_HAS_5_FRAGMENTS_AFTER_UPDATE    = 0xb0,
+	CB_PERF_SEL_QUAD_HAS_6_FRAGMENTS_AFTER_UPDATE    = 0xb1,
+	CB_PERF_SEL_QUAD_HAS_7_FRAGMENTS_AFTER_UPDATE    = 0xb2,
+	CB_PERF_SEL_QUAD_HAS_8_FRAGMENTS_AFTER_UPDATE    = 0xb3,
+	CB_PERF_SEL_QUAD_ADDED_1_FRAGMENT                = 0xb4,
+	CB_PERF_SEL_QUAD_ADDED_2_FRAGMENTS               = 0xb5,
+	CB_PERF_SEL_QUAD_ADDED_3_FRAGMENTS               = 0xb6,
+	CB_PERF_SEL_QUAD_ADDED_4_FRAGMENTS               = 0xb7,
+	CB_PERF_SEL_QUAD_ADDED_5_FRAGMENTS               = 0xb8,
+	CB_PERF_SEL_QUAD_ADDED_6_FRAGMENTS               = 0xb9,
+	CB_PERF_SEL_QUAD_ADDED_7_FRAGMENTS               = 0xba,
+	CB_PERF_SEL_QUAD_REMOVED_1_FRAGMENT              = 0xbb,
+	CB_PERF_SEL_QUAD_REMOVED_2_FRAGMENTS             = 0xbc,
+	CB_PERF_SEL_QUAD_REMOVED_3_FRAGMENTS             = 0xbd,
+	CB_PERF_SEL_QUAD_REMOVED_4_FRAGMENTS             = 0xbe,
+	CB_PERF_SEL_QUAD_REMOVED_5_FRAGMENTS             = 0xbf,
+	CB_PERF_SEL_QUAD_REMOVED_6_FRAGMENTS             = 0xc0,
+	CB_PERF_SEL_QUAD_REMOVED_7_FRAGMENTS             = 0xc1,
+	CB_PERF_SEL_QUAD_READS_FRAGMENT_0                = 0xc2,
+	CB_PERF_SEL_QUAD_READS_FRAGMENT_1                = 0xc3,
+	CB_PERF_SEL_QUAD_READS_FRAGMENT_2                = 0xc4,
+	CB_PERF_SEL_QUAD_READS_FRAGMENT_3                = 0xc5,
+	CB_PERF_SEL_QUAD_READS_FRAGMENT_4                = 0xc6,
+	CB_PERF_SEL_QUAD_READS_FRAGMENT_5                = 0xc7,
+	CB_PERF_SEL_QUAD_READS_FRAGMENT_6                = 0xc8,
+	CB_PERF_SEL_QUAD_READS_FRAGMENT_7                = 0xc9,
+	CB_PERF_SEL_QUAD_WRITES_FRAGMENT_0               = 0xca,
+	CB_PERF_SEL_QUAD_WRITES_FRAGMENT_1               = 0xcb,
+	CB_PERF_SEL_QUAD_WRITES_FRAGMENT_2               = 0xcc,
+	CB_PERF_SEL_QUAD_WRITES_FRAGMENT_3               = 0xcd,
+	CB_PERF_SEL_QUAD_WRITES_FRAGMENT_4               = 0xce,
+	CB_PERF_SEL_QUAD_WRITES_FRAGMENT_5               = 0xcf,
+	CB_PERF_SEL_QUAD_WRITES_FRAGMENT_6               = 0xd0,
+	CB_PERF_SEL_QUAD_WRITES_FRAGMENT_7               = 0xd1,
+	CB_PERF_SEL_QUAD_BLEND_OPT_DONT_READ_DST         = 0xd2,
+	CB_PERF_SEL_QUAD_BLEND_OPT_BLEND_BYPASS          = 0xd3,
+	CB_PERF_SEL_QUAD_BLEND_OPT_DISCARD_PIXELS        = 0xd4,
+	CB_PERF_SEL_QUAD_DST_READ_COULD_HAVE_BEEN_OPTIMIZED= 0xd5,
+	CB_PERF_SEL_QUAD_BLENDING_COULD_HAVE_BEEN_BYPASSED= 0xd6,
+	CB_PERF_SEL_QUAD_COULD_HAVE_BEEN_DISCARDED       = 0xd7,
+	CB_PERF_SEL_BLEND_OPT_PIXELS_RESULT_EQ_DEST      = 0xd8,
+	CB_PERF_SEL_DRAWN_BUSY                           = 0xd9,
+	CB_PERF_SEL_TILE_TO_CMR_REGION_BUSY              = 0xda,
+	CB_PERF_SEL_CMR_TO_FCR_REGION_BUSY               = 0xdb,
+	CB_PERF_SEL_FCR_TO_CCR_REGION_BUSY               = 0xdc,
+	CB_PERF_SEL_CCR_TO_CCW_REGION_BUSY               = 0xdd,
+	CB_PERF_SEL_FC_PF_SLOW_MODE_QUAD_EMPTY_HALF_DROPPED= 0xde,
+	CB_PERF_SEL_FC_SEQUENCER_CLEAR                   = 0xdf,
+	CB_PERF_SEL_FC_SEQUENCER_ELIMINATE_FAST_CLEAR    = 0xe0,
+	CB_PERF_SEL_FC_SEQUENCER_FMASK_DECOMPRESS        = 0xe1,
+	CB_PERF_SEL_FC_SEQUENCER_FMASK_COMPRESSION_DISABLE= 0xe2,
+	CB_PERF_SEL_FC_KEYID_RDLAT_FIFO_FULL             = 0xe3,
+	CB_PERF_SEL_FC_DOC_IS_STALLED                    = 0xe4,
+	CB_PERF_SEL_FC_DOC_MRTS_NOT_COMBINED             = 0xe5,
+	CB_PERF_SEL_FC_DOC_MRTS_COMBINED                 = 0xe6,
+	CB_PERF_SEL_FC_DOC_QTILE_CAM_MISS                = 0xe7,
+	CB_PERF_SEL_FC_DOC_QTILE_CAM_HIT                 = 0xe8,
+	CB_PERF_SEL_FC_DOC_CLINE_CAM_MISS                = 0xe9,
+	CB_PERF_SEL_FC_DOC_CLINE_CAM_HIT                 = 0xea,
+	CB_PERF_SEL_FC_DOC_QUAD_PTR_FIFO_IS_FULL         = 0xeb,
+	CB_PERF_SEL_FC_DOC_OVERWROTE_1_SECTOR            = 0xec,
+	CB_PERF_SEL_FC_DOC_OVERWROTE_2_SECTORS           = 0xed,
+	CB_PERF_SEL_FC_DOC_OVERWROTE_3_SECTORS           = 0xee,
+	CB_PERF_SEL_FC_DOC_OVERWROTE_4_SECTORS           = 0xef,
+	CB_PERF_SEL_FC_DOC_TOTAL_OVERWRITTEN_SECTORS     = 0xf0,
+	CB_PERF_SEL_FC_DCC_CACHE_HIT                     = 0xf1,
+	CB_PERF_SEL_FC_DCC_CACHE_TAG_MISS                = 0xf2,
+	CB_PERF_SEL_FC_DCC_CACHE_SECTOR_MISS             = 0xf3,
+	CB_PERF_SEL_FC_DCC_CACHE_REEVICTION_STALL        = 0xf4,
+	CB_PERF_SEL_FC_DCC_CACHE_EVICT_NONZERO_INFLIGHT_STALL= 0xf5,
+	CB_PERF_SEL_FC_DCC_CACHE_REPLACE_PENDING_EVICT_STALL= 0xf6,
+	CB_PERF_SEL_FC_DCC_CACHE_INFLIGHT_COUNTER_MAXIMUM_STALL= 0xf7,
+	CB_PERF_SEL_FC_DCC_CACHE_READ_OUTPUT_STALL       = 0xf8,
+	CB_PERF_SEL_FC_DCC_CACHE_WRITE_OUTPUT_STALL      = 0xf9,
+	CB_PERF_SEL_FC_DCC_CACHE_ACK_OUTPUT_STALL        = 0xfa,
+	CB_PERF_SEL_FC_DCC_CACHE_STALL                   = 0xfb,
+	CB_PERF_SEL_FC_DCC_CACHE_FLUSH                   = 0xfc,
+	CB_PERF_SEL_FC_DCC_CACHE_TAGS_FLUSHED            = 0xfd,
+	CB_PERF_SEL_FC_DCC_CACHE_SECTORS_FLUSHED         = 0xfe,
+	CB_PERF_SEL_FC_DCC_CACHE_DIRTY_SECTORS_FLUSHED   = 0xff,
+	CB_PERF_SEL_CC_DCC_BEYOND_TILE_SPLIT             = 0x100,
+	CB_PERF_SEL_FC_MC_DCC_WRITE_REQUEST              = 0x101,
+	CB_PERF_SEL_FC_MC_DCC_WRITE_REQUESTS_IN_FLIGHT   = 0x102,
+	CB_PERF_SEL_FC_MC_DCC_READ_REQUEST               = 0x103,
+	CB_PERF_SEL_FC_MC_DCC_READ_REQUESTS_IN_FLIGHT    = 0x104,
+	CB_PERF_SEL_CC_DCC_RDREQ_STALL                   = 0x105,
+	CB_PERF_SEL_CC_DCC_DECOMPRESS_TIDS_IN            = 0x106,
+	CB_PERF_SEL_CC_DCC_DECOMPRESS_TIDS_OUT           = 0x107,
+	CB_PERF_SEL_CC_DCC_COMPRESS_TIDS_IN              = 0x108,
+	CB_PERF_SEL_CC_DCC_COMPRESS_TIDS_OUT             = 0x109,
+	CB_PERF_SEL_FC_DCC_KEY_VALUE__CLEAR              = 0x10a,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__4_BLOCKS__2TO1     = 0x10b,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__3BLOCKS_2TO1__1BLOCK_2TO2= 0x10c,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__2BLOCKS_2TO1__1BLOCK_2TO2__1BLOCK_2TO1= 0x10d,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO1__1BLOCK_2TO2__2BLOCKS_2TO1= 0x10e,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO2__3BLOCKS_2TO1= 0x10f,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__2BLOCKS_2TO1__2BLOCKS_2TO2= 0x110,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO1__2BLOCKS_2TO2__1BLOCK_2TO1= 0x111,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO1__1BLOCK_2TO2__1BLOCK_2TO1__1BLOCK_2TO2= 0x112,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO2__1BLOCK_2TO1__1BLOCK_2TO2__1BLOCK_2TO1= 0x113,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__2BLOCKS_2TO2__2BLOCKS_2TO1= 0x114,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO2__2BLOCKS_2TO1__1BLOCK_2TO2= 0x115,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO1__3BLOCKS_2TO2= 0x116,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO2__1BLOCK_2TO1__2BLOCKS_2TO2= 0x117,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__2BLOCKS_2TO2__1BLOCK_2TO1__1BLOCK_2TO2= 0x118,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__3BLOCKS_2TO2__1BLOCK_2TO1= 0x119,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__2BLOCKS_4TO1       = 0x11a,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_4TO1__1BLOCK_4TO2= 0x11b,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_4TO1__1BLOCK_4TO3= 0x11c,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_4TO1__1BLOCK_4TO4= 0x11d,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_4TO2__1BLOCK_4TO1= 0x11e,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__2BLOCKS_4TO2       = 0x11f,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_4TO2__1BLOCK_4TO3= 0x120,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_4TO2__1BLOCK_4TO4= 0x121,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_4TO3__1BLOCK_4TO1= 0x122,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_4TO3__1BLOCK_4TO2= 0x123,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__2BLOCKS_4TO3       = 0x124,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_4TO3__1BLOCK_4TO4= 0x125,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_4TO4__1BLOCK_4TO1= 0x126,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_4TO4__1BLOCK_4TO2= 0x127,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_4TO4__1BLOCK_4TO3= 0x128,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__2BLOCKS_2TO1__1BLOCK_4TO1= 0x129,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__2BLOCKS_2TO1__1BLOCK_4TO2= 0x12a,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__2BLOCKS_2TO1__1BLOCK_4TO3= 0x12b,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__2BLOCKS_2TO1__1BLOCK_4TO4= 0x12c,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO1__1BLOCK_2TO2__1BLOCK_4TO1= 0x12d,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO1__1BLOCK_2TO2__1BLOCK_4TO2= 0x12e,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO1__1BLOCK_2TO2__1BLOCK_4TO3= 0x12f,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO1__1BLOCK_2TO2__1BLOCK_4TO4= 0x130,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO2__1BLOCK_2TO1__1BLOCK_4TO1= 0x131,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO2__1BLOCK_2TO1__1BLOCK_4TO2= 0x132,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO2__1BLOCK_2TO1__1BLOCK_4TO3= 0x133,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO2__1BLOCK_2TO1__1BLOCK_4TO4= 0x134,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__2BLOCKS_2TO2__1BLOCK_4TO1= 0x135,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__2BLOCKS_2TO2__1BLOCK_4TO2= 0x136,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__2BLOCKS_2TO2__1BLOCK_4TO3= 0x137,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO1__1BLOCK_4TO1__1BLOCK_2TO1= 0x138,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO1__1BLOCK_4TO2__1BLOCK_2TO1= 0x139,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO1__1BLOCK_4TO3__1BLOCK_2TO1= 0x13a,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO1__1BLOCK_4TO4__1BLOCK_2TO1= 0x13b,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO2__1BLOCK_4TO1__1BLOCK_2TO1= 0x13c,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO2__1BLOCK_4TO2__1BLOCK_2TO1= 0x13d,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO2__1BLOCK_4TO3__1BLOCK_2TO1= 0x13e,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO2__1BLOCK_4TO4__1BLOCK_2TO1= 0x13f,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO1__1BLOCK_4TO1__1BLOCK_2TO2= 0x140,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO1__1BLOCK_4TO2__1BLOCK_2TO2= 0x141,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO1__1BLOCK_4TO3__1BLOCK_2TO2= 0x142,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO1__1BLOCK_4TO4__1BLOCK_2TO2= 0x143,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO2__1BLOCK_4TO1__1BLOCK_2TO2= 0x144,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO2__1BLOCK_4TO2__1BLOCK_2TO2= 0x145,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO2__1BLOCK_4TO3__1BLOCK_2TO2= 0x146,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_4TO1__2BLOCKS_2TO1= 0x147,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_4TO2__2BLOCKS_2TO1= 0x148,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_4TO3__2BLOCKS_2TO1= 0x149,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_4TO4__2BLOCKS_2TO1= 0x14a,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_4TO1__2BLOCKS_2TO2= 0x14b,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_4TO2__2BLOCKS_2TO2= 0x14c,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_4TO3__2BLOCKS_2TO2= 0x14d,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_4TO1__1BLOCK_2TO1__1BLOCK_2TO2= 0x14e,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_4TO2__1BLOCK_2TO1__1BLOCK_2TO2= 0x14f,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_4TO3__1BLOCK_2TO1__1BLOCK_2TO2= 0x150,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_4TO4__1BLOCK_2TO1__1BLOCK_2TO2= 0x151,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_4TO1__1BLOCK_2TO2__1BLOCK_2TO1= 0x152,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_4TO2__1BLOCK_2TO2__1BLOCK_2TO1= 0x153,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_4TO3__1BLOCK_2TO2__1BLOCK_2TO1= 0x154,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_4TO4__1BLOCK_2TO2__1BLOCK_2TO1= 0x155,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO1__1BLOCK_6TO1= 0x156,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO1__1BLOCK_6TO2= 0x157,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO1__1BLOCK_6TO3= 0x158,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO1__1BLOCK_6TO4= 0x159,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO1__1BLOCK_6TO5= 0x15a,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO1__1BLOCK_6TO6= 0x15b,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO1__INV0  = 0x15c,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO1__INV1  = 0x15d,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO2__1BLOCK_6TO1= 0x15e,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO2__1BLOCK_6TO2= 0x15f,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO2__1BLOCK_6TO3= 0x160,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO2__1BLOCK_6TO4= 0x161,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO2__1BLOCK_6TO5= 0x162,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO2__INV0  = 0x163,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_2TO2__INV1  = 0x164,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_6TO1__1BLOCK_2TO1= 0x165,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_6TO2__1BLOCK_2TO1= 0x166,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_6TO3__1BLOCK_2TO1= 0x167,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_6TO4__1BLOCK_2TO1= 0x168,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_6TO5__1BLOCK_2TO1= 0x169,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_6TO6__1BLOCK_2TO1= 0x16a,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__INV0__1BLOCK_2TO1  = 0x16b,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__INV1__1BLOCK_2TO1  = 0x16c,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_6TO1__1BLOCK_2TO2= 0x16d,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_6TO2__1BLOCK_2TO2= 0x16e,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_6TO3__1BLOCK_2TO2= 0x16f,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_6TO4__1BLOCK_2TO2= 0x170,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_6TO5__1BLOCK_2TO2= 0x171,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__INV0__1BLOCK_2TO2  = 0x172,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__INV1__1BLOCK_2TO2  = 0x173,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_8TO1        = 0x174,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_8TO2        = 0x175,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_8TO3        = 0x176,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_8TO4        = 0x177,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_8TO5        = 0x178,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_8TO6        = 0x179,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__1BLOCK_8TO7        = 0x17a,
+	CB_PERF_SEL_CC_DCC_KEY_VALUE__UNCOMPRESSED       = 0x17b,
+	CB_PERF_SEL_CC_DCC_COMPRESS_RATIO_2TO1           = 0x17c,
+	CB_PERF_SEL_CC_DCC_COMPRESS_RATIO_4TO1           = 0x17d,
+	CB_PERF_SEL_CC_DCC_COMPRESS_RATIO_4TO2           = 0x17e,
+	CB_PERF_SEL_CC_DCC_COMPRESS_RATIO_4TO3           = 0x17f,
+	CB_PERF_SEL_CC_DCC_COMPRESS_RATIO_6TO1           = 0x180,
+	CB_PERF_SEL_CC_DCC_COMPRESS_RATIO_6TO2           = 0x181,
+	CB_PERF_SEL_CC_DCC_COMPRESS_RATIO_6TO3           = 0x182,
+	CB_PERF_SEL_CC_DCC_COMPRESS_RATIO_6TO4           = 0x183,
+	CB_PERF_SEL_CC_DCC_COMPRESS_RATIO_6TO5           = 0x184,
+	CB_PERF_SEL_CC_DCC_COMPRESS_RATIO_8TO1           = 0x185,
+	CB_PERF_SEL_CC_DCC_COMPRESS_RATIO_8TO2           = 0x186,
+	CB_PERF_SEL_CC_DCC_COMPRESS_RATIO_8TO3           = 0x187,
+	CB_PERF_SEL_CC_DCC_COMPRESS_RATIO_8TO4           = 0x188,
+	CB_PERF_SEL_CC_DCC_COMPRESS_RATIO_8TO5           = 0x189,
+	CB_PERF_SEL_CC_DCC_COMPRESS_RATIO_8TO6           = 0x18a,
+	CB_PERF_SEL_CC_DCC_COMPRESS_RATIO_8TO7           = 0x18b,
+	CB_PERF_SEL_RBP_EXPORT_8PIX_LIT_BOTH             = 0x18c,
+	CB_PERF_SEL_RBP_EXPORT_8PIX_LIT_LEFT             = 0x18d,
+	CB_PERF_SEL_RBP_EXPORT_8PIX_LIT_RIGHT            = 0x18e,
+	CB_PERF_SEL_RBP_SPLIT_MICROTILE                  = 0x18f,
+	CB_PERF_SEL_RBP_SPLIT_AA_SAMPLE_MASK             = 0x190,
+	CB_PERF_SEL_RBP_SPLIT_PARTIAL_TARGET_MASK        = 0x191,
+	CB_PERF_SEL_RBP_SPLIT_LINEAR_ADDRESSING          = 0x192,
+	CB_PERF_SEL_RBP_SPLIT_AA_NO_FMASK_COMPRESS       = 0x193,
+	CB_PERF_SEL_RBP_INSERT_MISSING_LAST_QUAD         = 0x194,
+} CBPerfSel;
+typedef enum CBPerfOpFilterSel {
+	CB_PERF_OP_FILTER_SEL_WRITE_ONLY                 = 0x0,
+	CB_PERF_OP_FILTER_SEL_NEEDS_DESTINATION          = 0x1,
+	CB_PERF_OP_FILTER_SEL_RESOLVE                    = 0x2,
+	CB_PERF_OP_FILTER_SEL_DECOMPRESS                 = 0x3,
+	CB_PERF_OP_FILTER_SEL_FMASK_DECOMPRESS           = 0x4,
+	CB_PERF_OP_FILTER_SEL_ELIMINATE_FAST_CLEAR       = 0x5,
+} CBPerfOpFilterSel;
+typedef enum CBPerfClearFilterSel {
+	CB_PERF_CLEAR_FILTER_SEL_NONCLEAR                = 0x0,
+	CB_PERF_CLEAR_FILTER_SEL_CLEAR                   = 0x1,
+} CBPerfClearFilterSel;
+typedef enum CP_RING_ID {
+	RINGID0                                          = 0x0,
+	RINGID1                                          = 0x1,
+	RINGID2                                          = 0x2,
+	RINGID3                                          = 0x3,
+} CP_RING_ID;
+typedef enum CP_PIPE_ID {
+	PIPE_ID0                                         = 0x0,
+	PIPE_ID1                                         = 0x1,
+	PIPE_ID2                                         = 0x2,
+	PIPE_ID3                                         = 0x3,
+} CP_PIPE_ID;
+typedef enum CP_ME_ID {
+	ME_ID0                                           = 0x0,
+	ME_ID1                                           = 0x1,
+	ME_ID2                                           = 0x2,
+	ME_ID3                                           = 0x3,
+} CP_ME_ID;
+typedef enum SPM_PERFMON_STATE {
+	STRM_PERFMON_STATE_DISABLE_AND_RESET             = 0x0,
+	STRM_PERFMON_STATE_START_COUNTING                = 0x1,
+	STRM_PERFMON_STATE_STOP_COUNTING                 = 0x2,
+	STRM_PERFMON_STATE_RESERVED_3                    = 0x3,
+	STRM_PERFMON_STATE_DISABLE_AND_RESET_PHANTOM     = 0x4,
+	STRM_PERFMON_STATE_COUNT_AND_DUMP_PHANTOM        = 0x5,
+} SPM_PERFMON_STATE;
+typedef enum CP_PERFMON_STATE {
+	CP_PERFMON_STATE_DISABLE_AND_RESET               = 0x0,
+	CP_PERFMON_STATE_START_COUNTING                  = 0x1,
+	CP_PERFMON_STATE_STOP_COUNTING                   = 0x2,
+	CP_PERFMON_STATE_RESERVED_3                      = 0x3,
+	CP_PERFMON_STATE_DISABLE_AND_RESET_PHANTOM       = 0x4,
+	CP_PERFMON_STATE_COUNT_AND_DUMP_PHANTOM          = 0x5,
+} CP_PERFMON_STATE;
+typedef enum CP_PERFMON_ENABLE_MODE {
+	CP_PERFMON_ENABLE_MODE_ALWAYS_COUNT              = 0x0,
+	CP_PERFMON_ENABLE_MODE_RESERVED_1                = 0x1,
+	CP_PERFMON_ENABLE_MODE_COUNT_CONTEXT_TRUE        = 0x2,
+	CP_PERFMON_ENABLE_MODE_COUNT_CONTEXT_FALSE       = 0x3,
+} CP_PERFMON_ENABLE_MODE;
+typedef enum CPG_PERFCOUNT_SEL {
+	CPG_PERF_SEL_ALWAYS_COUNT                        = 0x0,
+	CPG_PERF_SEL_RBIU_FIFO_FULL                      = 0x1,
+	CPG_PERF_SEL_CSF_RTS_BUT_MIU_NOT_RTR             = 0x2,
+	CPG_PERF_SEL_CSF_ST_BASE_SIZE_FIFO_FULL          = 0x3,
+	CPG_PERF_SEL_CP_GRBM_DWORDS_SENT                 = 0x4,
+	CPG_PERF_SEL_ME_PARSER_BUSY                      = 0x5,
+	CPG_PERF_SEL_COUNT_TYPE0_PACKETS                 = 0x6,
+	CPG_PERF_SEL_COUNT_TYPE3_PACKETS                 = 0x7,
+	CPG_PERF_SEL_CSF_FETCHING_CMD_BUFFERS            = 0x8,
+	CPG_PERF_SEL_CP_GRBM_OUT_OF_CREDITS              = 0x9,
+	CPG_PERF_SEL_CP_PFP_GRBM_OUT_OF_CREDITS          = 0xa,
+	CPG_PERF_SEL_CP_GDS_GRBM_OUT_OF_CREDITS          = 0xb,
+	CPG_PERF_SEL_RCIU_STALLED_ON_ME_READ             = 0xc,
+	CPG_PERF_SEL_RCIU_STALLED_ON_DMA_READ            = 0xd,
+	CPG_PERF_SEL_SSU_STALLED_ON_ACTIVE_CNTX          = 0xe,
+	CPG_PERF_SEL_SSU_STALLED_ON_CLEAN_SIGNALS        = 0xf,
+	CPG_PERF_SEL_QU_STALLED_ON_EOP_DONE_PULSE        = 0x10,
+	CPG_PERF_SEL_QU_STALLED_ON_EOP_DONE_WR_CONFIRM   = 0x11,
+	CPG_PERF_SEL_PFP_STALLED_ON_CSF_READY            = 0x12,
+	CPG_PERF_SEL_PFP_STALLED_ON_MEQ_READY            = 0x13,
+	CPG_PERF_SEL_PFP_STALLED_ON_RCIU_READY           = 0x14,
+	CPG_PERF_SEL_PFP_STALLED_FOR_DATA_FROM_ROQ       = 0x15,
+	CPG_PERF_SEL_ME_STALLED_FOR_DATA_FROM_PFP        = 0x16,
+	CPG_PERF_SEL_ME_STALLED_FOR_DATA_FROM_STQ        = 0x17,
+	CPG_PERF_SEL_ME_STALLED_ON_NO_AVAIL_GFX_CNTX     = 0x18,
+	CPG_PERF_SEL_ME_STALLED_WRITING_TO_RCIU          = 0x19,
+	CPG_PERF_SEL_ME_STALLED_WRITING_CONSTANTS        = 0x1a,
+	CPG_PERF_SEL_ME_STALLED_ON_PARTIAL_FLUSH         = 0x1b,
+	CPG_PERF_SEL_ME_WAIT_ON_CE_COUNTER               = 0x1c,
+	CPG_PERF_SEL_ME_WAIT_ON_AVAIL_BUFFER             = 0x1d,
+	CPG_PERF_SEL_SEMAPHORE_BUSY_POLLING_FOR_PASS     = 0x1e,
+	CPG_PERF_SEL_LOAD_STALLED_ON_SET_COHERENCY       = 0x1f,
+	CPG_PERF_SEL_DYNAMIC_CLK_VALID                   = 0x20,
+	CPG_PERF_SEL_REGISTER_CLK_VALID                  = 0x21,
+	CPG_PERF_SEL_MIU_WRITE_REQUEST_SENT              = 0x22,
+	CPG_PERF_SEL_MIU_READ_REQUEST_SENT               = 0x23,
+	CPG_PERF_SEL_CE_STALL_RAM_DUMP                   = 0x24,
+	CPG_PERF_SEL_CE_STALL_RAM_WRITE                  = 0x25,
+	CPG_PERF_SEL_CE_STALL_ON_INC_FIFO                = 0x26,
+	CPG_PERF_SEL_CE_STALL_ON_WR_RAM_FIFO             = 0x27,
+	CPG_PERF_SEL_CE_STALL_ON_DATA_FROM_MIU           = 0x28,
+	CPG_PERF_SEL_CE_STALL_ON_DATA_FROM_ROQ           = 0x29,
+	CPG_PERF_SEL_CE_STALL_ON_CE_BUFFER_FLAG          = 0x2a,
+	CPG_PERF_SEL_CE_STALL_ON_DE_COUNTER              = 0x2b,
+	CPG_PERF_SEL_TCIU_STALL_WAIT_ON_FREE             = 0x2c,
+	CPG_PERF_SEL_TCIU_STALL_WAIT_ON_TAGS             = 0x2d,
+	CPG_PERF_SEL_ATCL2IU_STALL_WAIT_ON_FREE          = 0x2e,
+	CPG_PERF_SEL_ATCL2IU_STALL_WAIT_ON_TAGS          = 0x2f,
+	CPG_PERF_SEL_ATCL1_STALL_ON_TRANSLATION          = 0x30,
+} CPG_PERFCOUNT_SEL;
+typedef enum CPF_PERFCOUNT_SEL {
+	CPF_PERF_SEL_ALWAYS_COUNT                        = 0x0,
+	CPF_PERF_SEL_MIU_STALLED_WAITING_RDREQ_FREE      = 0x1,
+	CPF_PERF_SEL_TCIU_STALLED_WAITING_ON_FREE        = 0x2,
+	CPF_PERF_SEL_TCIU_STALLED_WAITING_ON_TAGS        = 0x3,
+	CPF_PERF_SEL_CSF_BUSY_FOR_FETCHING_RING          = 0x4,
+	CPF_PERF_SEL_CSF_BUSY_FOR_FETCHING_IB1           = 0x5,
+	CPF_PERF_SEL_CSF_BUSY_FOR_FETCHING_IB2           = 0x6,
+	CPF_PERF_SEL_CSF_BUSY_FOR_FECTHINC_STATE         = 0x7,
+	CPF_PERF_SEL_MIU_BUSY_FOR_OUTSTANDING_TAGS       = 0x8,
+	CPF_PERF_SEL_CSF_RTS_MIU_NOT_RTR                 = 0x9,
+	CPF_PERF_SEL_CSF_STATE_FIFO_NOT_RTR              = 0xa,
+	CPF_PERF_SEL_CSF_FETCHING_CMD_BUFFERS            = 0xb,
+	CPF_PERF_SEL_GRBM_DWORDS_SENT                    = 0xc,
+	CPF_PERF_SEL_DYNAMIC_CLOCK_VALID                 = 0xd,
+	CPF_PERF_SEL_REGISTER_CLOCK_VALID                = 0xe,
+	CPF_PERF_SEL_MIU_WRITE_REQUEST_SEND              = 0xf,
+	CPF_PERF_SEL_MIU_READ_REQUEST_SEND               = 0x10,
+	CPF_PERF_SEL_ATCL2IU_STALL_WAIT_ON_FREE          = 0x11,
+	CPF_PERF_SEL_ATCL2IU_STALL_WAIT_ON_TAGS          = 0x12,
+	CPF_PERF_SEL_ATCL1_STALL_ON_TRANSLATION          = 0x13,
+} CPF_PERFCOUNT_SEL;
+typedef enum CPC_PERFCOUNT_SEL {
+	CPC_PERF_SEL_ALWAYS_COUNT                        = 0x0,
+	CPC_PERF_SEL_RCIU_STALL_WAIT_ON_FREE             = 0x1,
+	CPC_PERF_SEL_RCIU_STALL_PRIV_VIOLATION           = 0x2,
+	CPC_PERF_SEL_MIU_STALL_ON_RDREQ_FREE             = 0x3,
+	CPC_PERF_SEL_MIU_STALL_ON_WRREQ_FREE             = 0x4,
+	CPC_PERF_SEL_TCIU_STALL_WAIT_ON_FREE             = 0x5,
+	CPC_PERF_SEL_ME1_STALL_WAIT_ON_RCIU_READY        = 0x6,
+	CPC_PERF_SEL_ME1_STALL_WAIT_ON_RCIU_READY_PERF   = 0x7,
+	CPC_PERF_SEL_ME1_STALL_WAIT_ON_RCIU_READ         = 0x8,
+	CPC_PERF_SEL_ME1_STALL_WAIT_ON_MIU_READ          = 0x9,
+	CPC_PERF_SEL_ME1_STALL_WAIT_ON_MIU_WRITE         = 0xa,
+	CPC_PERF_SEL_ME1_STALL_ON_DATA_FROM_ROQ          = 0xb,
+	CPC_PERF_SEL_ME1_STALL_ON_DATA_FROM_ROQ_PERF     = 0xc,
+	CPC_PERF_SEL_ME1_BUSY_FOR_PACKET_DECODE          = 0xd,
+	CPC_PERF_SEL_ME2_STALL_WAIT_ON_RCIU_READY        = 0xe,
+	CPC_PERF_SEL_ME2_STALL_WAIT_ON_RCIU_READY_PERF   = 0xf,
+	CPC_PERF_SEL_ME2_STALL_WAIT_ON_RCIU_READ         = 0x10,
+	CPC_PERF_SEL_ME2_STALL_WAIT_ON_MIU_READ          = 0x11,
+	CPC_PERF_SEL_ME2_STALL_WAIT_ON_MIU_WRITE         = 0x12,
+	CPC_PERF_SEL_ME2_STALL_ON_DATA_FROM_ROQ          = 0x13,
+	CPC_PERF_SEL_ME2_STALL_ON_DATA_FROM_ROQ_PERF     = 0x14,
+	CPC_PERF_SEL_ME2_BUSY_FOR_PACKET_DECODE          = 0x15,
+	CPC_PERF_SEL_ATCL2IU_STALL_WAIT_ON_FREE          = 0x16,
+	CPC_PERF_SEL_ATCL2IU_STALL_WAIT_ON_TAGS          = 0x17,
+	CPC_PERF_SEL_ATCL1_STALL_ON_TRANSLATION          = 0x18,
+} CPC_PERFCOUNT_SEL;
+typedef enum CP_ALPHA_TAG_RAM_SEL {
+	CPG_TAG_RAM                                      = 0x0,
+	CPC_TAG_RAM                                      = 0x1,
+	CPF_TAG_RAM                                      = 0x2,
+	RSV_TAG_RAM                                      = 0x3,
+} CP_ALPHA_TAG_RAM_SEL;
+#define SEM_ECC_ERROR                             0x0
+#define SEM_RESERVED                              0x1
+#define SEM_FAILED                                0x2
+#define SEM_PASSED                                0x3
+#define IQ_QUEUE_SLEEP                            0x0
+#define IQ_OFFLOAD_RETRY                          0x1
+#define IQ_SCH_WAVE_MSG                           0x2
+#define IQ_SEM_REARM                              0x3
+#define IQ_DEQUEUE_RETRY                          0x4
+#define IQ_INTR_TYPE_PQ                           0x0
+#define IQ_INTR_TYPE_IB                           0x1
+#define IQ_INTR_TYPE_MQD                          0x2
+#define VMID_SZ                                   0x4
+#define CONFIG_SPACE_START                        0x2000
+#define CONFIG_SPACE_END                          0x9fff
+#define CONFIG_SPACE1_START                       0x2000
+#define CONFIG_SPACE1_END                         0x2bff
+#define CONFIG_SPACE2_START                       0x3000
+#define CONFIG_SPACE2_END                         0x9fff
+#define UCONFIG_SPACE_START                       0xc000
+#define UCONFIG_SPACE_END                         0xffff
+#define PERSISTENT_SPACE_START                    0x2c00
+#define PERSISTENT_SPACE_END                      0x2fff
+#define CONTEXT_SPACE_START                       0xa000
+#define CONTEXT_SPACE_END                         0xbfff
+typedef enum ForceControl {
+	FORCE_OFF                                        = 0x0,
+	FORCE_ENABLE                                     = 0x1,
+	FORCE_DISABLE                                    = 0x2,
+	FORCE_RESERVED                                   = 0x3,
+} ForceControl;
+typedef enum ZSamplePosition {
+	Z_SAMPLE_CENTER                                  = 0x0,
+	Z_SAMPLE_CENTROID                                = 0x1,
+} ZSamplePosition;
+typedef enum ZOrder {
+	LATE_Z                                           = 0x0,
+	EARLY_Z_THEN_LATE_Z                              = 0x1,
+	RE_Z                                             = 0x2,
+	EARLY_Z_THEN_RE_Z                                = 0x3,
+} ZOrder;
+typedef enum ZpassControl {
+	ZPASS_DISABLE                                    = 0x0,
+	ZPASS_SAMPLES                                    = 0x1,
+	ZPASS_PIXELS                                     = 0x2,
+} ZpassControl;
+typedef enum ZModeForce {
+	NO_FORCE                                         = 0x0,
+	FORCE_EARLY_Z                                    = 0x1,
+	FORCE_LATE_Z                                     = 0x2,
+	FORCE_RE_Z                                       = 0x3,
+} ZModeForce;
+typedef enum ZLimitSumm {
+	FORCE_SUMM_OFF                                   = 0x0,
+	FORCE_SUMM_MINZ                                  = 0x1,
+	FORCE_SUMM_MAXZ                                  = 0x2,
+	FORCE_SUMM_BOTH                                  = 0x3,
+} ZLimitSumm;
+typedef enum CompareFrag {
+	FRAG_NEVER                                       = 0x0,
+	FRAG_LESS                                        = 0x1,
+	FRAG_EQUAL                                       = 0x2,
+	FRAG_LEQUAL                                      = 0x3,
+	FRAG_GREATER                                     = 0x4,
+	FRAG_NOTEQUAL                                    = 0x5,
+	FRAG_GEQUAL                                      = 0x6,
+	FRAG_ALWAYS                                      = 0x7,
+} CompareFrag;
+typedef enum StencilOp {
+	STENCIL_KEEP                                     = 0x0,
+	STENCIL_ZERO                                     = 0x1,
+	STENCIL_ONES                                     = 0x2,
+	STENCIL_REPLACE_TEST                             = 0x3,
+	STENCIL_REPLACE_OP                               = 0x4,
+	STENCIL_ADD_CLAMP                                = 0x5,
+	STENCIL_SUB_CLAMP                                = 0x6,
+	STENCIL_INVERT                                   = 0x7,
+	STENCIL_ADD_WRAP                                 = 0x8,
+	STENCIL_SUB_WRAP                                 = 0x9,
+	STENCIL_AND                                      = 0xa,
+	STENCIL_OR                                       = 0xb,
+	STENCIL_XOR                                      = 0xc,
+	STENCIL_NAND                                     = 0xd,
+	STENCIL_NOR                                      = 0xe,
+	STENCIL_XNOR                                     = 0xf,
+} StencilOp;
+typedef enum ConservativeZExport {
+	EXPORT_ANY_Z                                     = 0x0,
+	EXPORT_LESS_THAN_Z                               = 0x1,
+	EXPORT_GREATER_THAN_Z                            = 0x2,
+	EXPORT_RESERVED                                  = 0x3,
+} ConservativeZExport;
+typedef enum DbPSLControl {
+	PSLC_AUTO                                        = 0x0,
+	PSLC_ON_HANG_ONLY                                = 0x1,
+	PSLC_ASAP                                        = 0x2,
+	PSLC_COUNTDOWN                                   = 0x3,
+} DbPSLControl;
+typedef enum PerfCounter_Vals {
+	DB_PERF_SEL_SC_DB_tile_sends                     = 0x0,
+	DB_PERF_SEL_SC_DB_tile_busy                      = 0x1,
+	DB_PERF_SEL_SC_DB_tile_stalls                    = 0x2,
+	DB_PERF_SEL_SC_DB_tile_events                    = 0x3,
+	DB_PERF_SEL_SC_DB_tile_tiles                     = 0x4,
+	DB_PERF_SEL_SC_DB_tile_covered                   = 0x5,
+	DB_PERF_SEL_hiz_tc_read_starved                  = 0x6,
+	DB_PERF_SEL_hiz_tc_write_stall                   = 0x7,
+	DB_PERF_SEL_hiz_qtiles_culled                    = 0x8,
+	DB_PERF_SEL_his_qtiles_culled                    = 0x9,
+	DB_PERF_SEL_DB_SC_tile_sends                     = 0xa,
+	DB_PERF_SEL_DB_SC_tile_busy                      = 0xb,
+	DB_PERF_SEL_DB_SC_tile_stalls                    = 0xc,
+	DB_PERF_SEL_DB_SC_tile_df_stalls                 = 0xd,
+	DB_PERF_SEL_DB_SC_tile_tiles                     = 0xe,
+	DB_PERF_SEL_DB_SC_tile_culled                    = 0xf,
+	DB_PERF_SEL_DB_SC_tile_hier_kill                 = 0x10,
+	DB_PERF_SEL_DB_SC_tile_fast_ops                  = 0x11,
+	DB_PERF_SEL_DB_SC_tile_no_ops                    = 0x12,
+	DB_PERF_SEL_DB_SC_tile_tile_rate                 = 0x13,
+	DB_PERF_SEL_DB_SC_tile_ssaa_kill                 = 0x14,
+	DB_PERF_SEL_DB_SC_tile_fast_z_ops                = 0x15,
+	DB_PERF_SEL_DB_SC_tile_fast_stencil_ops          = 0x16,
+	DB_PERF_SEL_SC_DB_quad_sends                     = 0x17,
+	DB_PERF_SEL_SC_DB_quad_busy                      = 0x18,
+	DB_PERF_SEL_SC_DB_quad_squads                    = 0x19,
+	DB_PERF_SEL_SC_DB_quad_tiles                     = 0x1a,
+	DB_PERF_SEL_SC_DB_quad_pixels                    = 0x1b,
+	DB_PERF_SEL_SC_DB_quad_killed_tiles              = 0x1c,
+	DB_PERF_SEL_DB_SC_quad_sends                     = 0x1d,
+	DB_PERF_SEL_DB_SC_quad_busy                      = 0x1e,
+	DB_PERF_SEL_DB_SC_quad_stalls                    = 0x1f,
+	DB_PERF_SEL_DB_SC_quad_tiles                     = 0x20,
+	DB_PERF_SEL_DB_SC_quad_lit_quad                  = 0x21,
+	DB_PERF_SEL_DB_CB_tile_sends                     = 0x22,
+	DB_PERF_SEL_DB_CB_tile_busy                      = 0x23,
+	DB_PERF_SEL_DB_CB_tile_stalls                    = 0x24,
+	DB_PERF_SEL_SX_DB_quad_sends                     = 0x25,
+	DB_PERF_SEL_SX_DB_quad_busy                      = 0x26,
+	DB_PERF_SEL_SX_DB_quad_stalls                    = 0x27,
+	DB_PERF_SEL_SX_DB_quad_quads                     = 0x28,
+	DB_PERF_SEL_SX_DB_quad_pixels                    = 0x29,
+	DB_PERF_SEL_SX_DB_quad_exports                   = 0x2a,
+	DB_PERF_SEL_SH_quads_outstanding_sum             = 0x2b,
+	DB_PERF_SEL_DB_CB_lquad_sends                    = 0x2c,
+	DB_PERF_SEL_DB_CB_lquad_busy                     = 0x2d,
+	DB_PERF_SEL_DB_CB_lquad_stalls                   = 0x2e,
+	DB_PERF_SEL_DB_CB_lquad_quads                    = 0x2f,
+	DB_PERF_SEL_tile_rd_sends                        = 0x30,
+	DB_PERF_SEL_mi_tile_rd_outstanding_sum           = 0x31,
+	DB_PERF_SEL_quad_rd_sends                        = 0x32,
+	DB_PERF_SEL_quad_rd_busy                         = 0x33,
+	DB_PERF_SEL_quad_rd_mi_stall                     = 0x34,
+	DB_PERF_SEL_quad_rd_rw_collision                 = 0x35,
+	DB_PERF_SEL_quad_rd_tag_stall                    = 0x36,
+	DB_PERF_SEL_quad_rd_32byte_reqs                  = 0x37,
+	DB_PERF_SEL_quad_rd_panic                        = 0x38,
+	DB_PERF_SEL_mi_quad_rd_outstanding_sum           = 0x39,
+	DB_PERF_SEL_quad_rdret_sends                     = 0x3a,
+	DB_PERF_SEL_quad_rdret_busy                      = 0x3b,
+	DB_PERF_SEL_tile_wr_sends                        = 0x3c,
+	DB_PERF_SEL_tile_wr_acks                         = 0x3d,
+	DB_PERF_SEL_mi_tile_wr_outstanding_sum           = 0x3e,
+	DB_PERF_SEL_quad_wr_sends                        = 0x3f,
+	DB_PERF_SEL_quad_wr_busy                         = 0x40,
+	DB_PERF_SEL_quad_wr_mi_stall                     = 0x41,
+	DB_PERF_SEL_quad_wr_coherency_stall              = 0x42,
+	DB_PERF_SEL_quad_wr_acks                         = 0x43,
+	DB_PERF_SEL_mi_quad_wr_outstanding_sum           = 0x44,
+	DB_PERF_SEL_Tile_Cache_misses                    = 0x45,
+	DB_PERF_SEL_Tile_Cache_hits                      = 0x46,
+	DB_PERF_SEL_Tile_Cache_flushes                   = 0x47,
+	DB_PERF_SEL_Tile_Cache_surface_stall             = 0x48,
+	DB_PERF_SEL_Tile_Cache_starves                   = 0x49,
+	DB_PERF_SEL_Tile_Cache_mem_return_starve         = 0x4a,
+	DB_PERF_SEL_tcp_dispatcher_reads                 = 0x4b,
+	DB_PERF_SEL_tcp_prefetcher_reads                 = 0x4c,
+	DB_PERF_SEL_tcp_preloader_reads                  = 0x4d,
+	DB_PERF_SEL_tcp_dispatcher_flushes               = 0x4e,
+	DB_PERF_SEL_tcp_prefetcher_flushes               = 0x4f,
+	DB_PERF_SEL_tcp_preloader_flushes                = 0x50,
+	DB_PERF_SEL_Depth_Tile_Cache_sends               = 0x51,
+	DB_PERF_SEL_Depth_Tile_Cache_busy                = 0x52,
+	DB_PERF_SEL_Depth_Tile_Cache_starves             = 0x53,
+	DB_PERF_SEL_Depth_Tile_Cache_dtile_locked        = 0x54,
+	DB_PERF_SEL_Depth_Tile_Cache_alloc_stall         = 0x55,
+	DB_PERF_SEL_Depth_Tile_Cache_misses              = 0x56,
+	DB_PERF_SEL_Depth_Tile_Cache_hits                = 0x57,
+	DB_PERF_SEL_Depth_Tile_Cache_flushes             = 0x58,
+	DB_PERF_SEL_Depth_Tile_Cache_noop_tile           = 0x59,
+	DB_PERF_SEL_Depth_Tile_Cache_detailed_noop       = 0x5a,
+	DB_PERF_SEL_Depth_Tile_Cache_event               = 0x5b,
+	DB_PERF_SEL_Depth_Tile_Cache_tile_frees          = 0x5c,
+	DB_PERF_SEL_Depth_Tile_Cache_data_frees          = 0x5d,
+	DB_PERF_SEL_Depth_Tile_Cache_mem_return_starve   = 0x5e,
+	DB_PERF_SEL_Stencil_Cache_misses                 = 0x5f,
+	DB_PERF_SEL_Stencil_Cache_hits                   = 0x60,
+	DB_PERF_SEL_Stencil_Cache_flushes                = 0x61,
+	DB_PERF_SEL_Stencil_Cache_starves                = 0x62,
+	DB_PERF_SEL_Stencil_Cache_frees                  = 0x63,
+	DB_PERF_SEL_Z_Cache_separate_Z_misses            = 0x64,
+	DB_PERF_SEL_Z_Cache_separate_Z_hits              = 0x65,
+	DB_PERF_SEL_Z_Cache_separate_Z_flushes           = 0x66,
+	DB_PERF_SEL_Z_Cache_separate_Z_starves           = 0x67,
+	DB_PERF_SEL_Z_Cache_pmask_misses                 = 0x68,
+	DB_PERF_SEL_Z_Cache_pmask_hits                   = 0x69,
+	DB_PERF_SEL_Z_Cache_pmask_flushes                = 0x6a,
+	DB_PERF_SEL_Z_Cache_pmask_starves                = 0x6b,
+	DB_PERF_SEL_Z_Cache_frees                        = 0x6c,
+	DB_PERF_SEL_Plane_Cache_misses                   = 0x6d,
+	DB_PERF_SEL_Plane_Cache_hits                     = 0x6e,
+	DB_PERF_SEL_Plane_Cache_flushes                  = 0x6f,
+	DB_PERF_SEL_Plane_Cache_starves                  = 0x70,
+	DB_PERF_SEL_Plane_Cache_frees                    = 0x71,
+	DB_PERF_SEL_flush_expanded_stencil               = 0x72,
+	DB_PERF_SEL_flush_compressed_stencil             = 0x73,
+	DB_PERF_SEL_flush_single_stencil                 = 0x74,
+	DB_PERF_SEL_planes_flushed                       = 0x75,
+	DB_PERF_SEL_flush_1plane                         = 0x76,
+	DB_PERF_SEL_flush_2plane                         = 0x77,
+	DB_PERF_SEL_flush_3plane                         = 0x78,
+	DB_PERF_SEL_flush_4plane                         = 0x79,
+	DB_PERF_SEL_flush_5plane                         = 0x7a,
+	DB_PERF_SEL_flush_6plane                         = 0x7b,
+	DB_PERF_SEL_flush_7plane                         = 0x7c,
+	DB_PERF_SEL_flush_8plane                         = 0x7d,
+	DB_PERF_SEL_flush_9plane                         = 0x7e,
+	DB_PERF_SEL_flush_10plane                        = 0x7f,
+	DB_PERF_SEL_flush_11plane                        = 0x80,
+	DB_PERF_SEL_flush_12plane                        = 0x81,
+	DB_PERF_SEL_flush_13plane                        = 0x82,
+	DB_PERF_SEL_flush_14plane                        = 0x83,
+	DB_PERF_SEL_flush_15plane                        = 0x84,
+	DB_PERF_SEL_flush_16plane                        = 0x85,
+	DB_PERF_SEL_flush_expanded_z                     = 0x86,
+	DB_PERF_SEL_earlyZ_waiting_for_postZ_done        = 0x87,
+	DB_PERF_SEL_reZ_waiting_for_postZ_done           = 0x88,
+	DB_PERF_SEL_dk_tile_sends                        = 0x89,
+	DB_PERF_SEL_dk_tile_busy                         = 0x8a,
+	DB_PERF_SEL_dk_tile_quad_starves                 = 0x8b,
+	DB_PERF_SEL_dk_tile_stalls                       = 0x8c,
+	DB_PERF_SEL_dk_squad_sends                       = 0x8d,
+	DB_PERF_SEL_dk_squad_busy                        = 0x8e,
+	DB_PERF_SEL_dk_squad_stalls                      = 0x8f,
+	DB_PERF_SEL_Op_Pipe_Busy                         = 0x90,
+	DB_PERF_SEL_Op_Pipe_MC_Read_stall                = 0x91,
+	DB_PERF_SEL_qc_busy                              = 0x92,
+	DB_PERF_SEL_qc_xfc                               = 0x93,
+	DB_PERF_SEL_qc_conflicts                         = 0x94,
+	DB_PERF_SEL_qc_full_stall                        = 0x95,
+	DB_PERF_SEL_qc_in_preZ_tile_stalls_postZ         = 0x96,
+	DB_PERF_SEL_qc_in_postZ_tile_stalls_preZ         = 0x97,
+	DB_PERF_SEL_tsc_insert_summarize_stall           = 0x98,
+	DB_PERF_SEL_tl_busy                              = 0x99,
+	DB_PERF_SEL_tl_dtc_read_starved                  = 0x9a,
+	DB_PERF_SEL_tl_z_fetch_stall                     = 0x9b,
+	DB_PERF_SEL_tl_stencil_stall                     = 0x9c,
+	DB_PERF_SEL_tl_z_decompress_stall                = 0x9d,
+	DB_PERF_SEL_tl_stencil_locked_stall              = 0x9e,
+	DB_PERF_SEL_tl_events                            = 0x9f,
+	DB_PERF_SEL_tl_summarize_squads                  = 0xa0,
+	DB_PERF_SEL_tl_flush_expand_squads               = 0xa1,
+	DB_PERF_SEL_tl_expand_squads                     = 0xa2,
+	DB_PERF_SEL_tl_preZ_squads                       = 0xa3,
+	DB_PERF_SEL_tl_postZ_squads                      = 0xa4,
+	DB_PERF_SEL_tl_preZ_noop_squads                  = 0xa5,
+	DB_PERF_SEL_tl_postZ_noop_squads                 = 0xa6,
+	DB_PERF_SEL_tl_tile_ops                          = 0xa7,
+	DB_PERF_SEL_tl_in_xfc                            = 0xa8,
+	DB_PERF_SEL_tl_in_single_stencil_expand_stall    = 0xa9,
+	DB_PERF_SEL_tl_in_fast_z_stall                   = 0xaa,
+	DB_PERF_SEL_tl_out_xfc                           = 0xab,
+	DB_PERF_SEL_tl_out_squads                        = 0xac,
+	DB_PERF_SEL_zf_plane_multicycle                  = 0xad,
+	DB_PERF_SEL_PostZ_Samples_passing_Z              = 0xae,
+	DB_PERF_SEL_PostZ_Samples_failing_Z              = 0xaf,
+	DB_PERF_SEL_PostZ_Samples_failing_S              = 0xb0,
+	DB_PERF_SEL_PreZ_Samples_passing_Z               = 0xb1,
+	DB_PERF_SEL_PreZ_Samples_failing_Z               = 0xb2,
+	DB_PERF_SEL_PreZ_Samples_failing_S               = 0xb3,
+	DB_PERF_SEL_ts_tc_update_stall                   = 0xb4,
+	DB_PERF_SEL_sc_kick_start                        = 0xb5,
+	DB_PERF_SEL_sc_kick_end                          = 0xb6,
+	DB_PERF_SEL_clock_reg_active                     = 0xb7,
+	DB_PERF_SEL_clock_main_active                    = 0xb8,
+	DB_PERF_SEL_clock_mem_export_active              = 0xb9,
+	DB_PERF_SEL_esr_ps_out_busy                      = 0xba,
+	DB_PERF_SEL_esr_ps_lqf_busy                      = 0xbb,
+	DB_PERF_SEL_esr_ps_lqf_stall                     = 0xbc,
+	DB_PERF_SEL_etr_out_send                         = 0xbd,
+	DB_PERF_SEL_etr_out_busy                         = 0xbe,
+	DB_PERF_SEL_etr_out_ltile_probe_fifo_full_stall  = 0xbf,
+	DB_PERF_SEL_etr_out_cb_tile_stall                = 0xc0,
+	DB_PERF_SEL_etr_out_esr_stall                    = 0xc1,
+	DB_PERF_SEL_esr_ps_sqq_busy                      = 0xc2,
+	DB_PERF_SEL_esr_ps_sqq_stall                     = 0xc3,
+	DB_PERF_SEL_esr_eot_fwd_busy                     = 0xc4,
+	DB_PERF_SEL_esr_eot_fwd_holding_squad            = 0xc5,
+	DB_PERF_SEL_esr_eot_fwd_forward                  = 0xc6,
+	DB_PERF_SEL_esr_sqq_zi_busy                      = 0xc7,
+	DB_PERF_SEL_esr_sqq_zi_stall                     = 0xc8,
+	DB_PERF_SEL_postzl_sq_pt_busy                    = 0xc9,
+	DB_PERF_SEL_postzl_sq_pt_stall                   = 0xca,
+	DB_PERF_SEL_postzl_se_busy                       = 0xcb,
+	DB_PERF_SEL_postzl_se_stall                      = 0xcc,
+	DB_PERF_SEL_postzl_partial_launch                = 0xcd,
+	DB_PERF_SEL_postzl_full_launch                   = 0xce,
+	DB_PERF_SEL_postzl_partial_waiting               = 0xcf,
+	DB_PERF_SEL_postzl_tile_mem_stall                = 0xd0,
+	DB_PERF_SEL_postzl_tile_init_stall               = 0xd1,
+	DB_PEFF_SEL_prezl_tile_mem_stall                 = 0xd2,
+	DB_PERF_SEL_prezl_tile_init_stall                = 0xd3,
+	DB_PERF_SEL_dtt_sm_clash_stall                   = 0xd4,
+	DB_PERF_SEL_dtt_sm_slot_stall                    = 0xd5,
+	DB_PERF_SEL_dtt_sm_miss_stall                    = 0xd6,
+	DB_PERF_SEL_mi_rdreq_busy                        = 0xd7,
+	DB_PERF_SEL_mi_rdreq_stall                       = 0xd8,
+	DB_PERF_SEL_mi_wrreq_busy                        = 0xd9,
+	DB_PERF_SEL_mi_wrreq_stall                       = 0xda,
+	DB_PERF_SEL_recomp_tile_to_1zplane_no_fastop     = 0xdb,
+	DB_PERF_SEL_dkg_tile_rate_tile                   = 0xdc,
+	DB_PERF_SEL_prezl_src_in_sends                   = 0xdd,
+	DB_PERF_SEL_prezl_src_in_stall                   = 0xde,
+	DB_PERF_SEL_prezl_src_in_squads                  = 0xdf,
+	DB_PERF_SEL_prezl_src_in_squads_unrolled         = 0xe0,
+	DB_PERF_SEL_prezl_src_in_tile_rate               = 0xe1,
+	DB_PERF_SEL_prezl_src_in_tile_rate_unrolled      = 0xe2,
+	DB_PERF_SEL_prezl_src_out_stall                  = 0xe3,
+	DB_PERF_SEL_postzl_src_in_sends                  = 0xe4,
+	DB_PERF_SEL_postzl_src_in_stall                  = 0xe5,
+	DB_PERF_SEL_postzl_src_in_squads                 = 0xe6,
+	DB_PERF_SEL_postzl_src_in_squads_unrolled        = 0xe7,
+	DB_PERF_SEL_postzl_src_in_tile_rate              = 0xe8,
+	DB_PERF_SEL_postzl_src_in_tile_rate_unrolled     = 0xe9,
+	DB_PERF_SEL_postzl_src_out_stall                 = 0xea,
+	DB_PERF_SEL_esr_ps_src_in_sends                  = 0xeb,
+	DB_PERF_SEL_esr_ps_src_in_stall                  = 0xec,
+	DB_PERF_SEL_esr_ps_src_in_squads                 = 0xed,
+	DB_PERF_SEL_esr_ps_src_in_squads_unrolled        = 0xee,
+	DB_PERF_SEL_esr_ps_src_in_tile_rate              = 0xef,
+	DB_PERF_SEL_esr_ps_src_in_tile_rate_unrolled     = 0xf0,
+	DB_PERF_SEL_esr_ps_src_in_tile_rate_unrolled_to_pixel_rate= 0xf1,
+	DB_PERF_SEL_esr_ps_src_out_stall                 = 0xf2,
+	DB_PERF_SEL_depth_bounds_qtiles_culled           = 0xf3,
+	DB_PERF_SEL_PreZ_Samples_failing_DB              = 0xf4,
+	DB_PERF_SEL_PostZ_Samples_failing_DB             = 0xf5,
+	DB_PERF_SEL_flush_compressed                     = 0xf6,
+	DB_PERF_SEL_flush_plane_le4                      = 0xf7,
+	DB_PERF_SEL_tiles_z_fully_summarized             = 0xf8,
+	DB_PERF_SEL_tiles_stencil_fully_summarized       = 0xf9,
+	DB_PERF_SEL_tiles_z_clear_on_expclear            = 0xfa,
+	DB_PERF_SEL_tiles_s_clear_on_expclear            = 0xfb,
+	DB_PERF_SEL_tiles_decomp_on_expclear             = 0xfc,
+	DB_PERF_SEL_tiles_compressed_to_decompressed     = 0xfd,
+	DB_PERF_SEL_Op_Pipe_Prez_Busy                    = 0xfe,
+	DB_PERF_SEL_Op_Pipe_Postz_Busy                   = 0xff,
+	DB_PERF_SEL_di_dt_stall                          = 0x100,
+	DB_PERF_SEL_DB_SC_quad_double_quad               = 0x101,
+	DB_PERF_SEL_SX_DB_quad_export_quads              = 0x102,
+	DB_PERF_SEL_SX_DB_quad_double_format             = 0x103,
+	DB_PERF_SEL_SX_DB_quad_fast_format               = 0x104,
+	DB_PERF_SEL_SX_DB_quad_slow_format               = 0x105,
+	DB_PERF_SEL_DB_CB_lquad_export_quads             = 0x106,
+	DB_PERF_SEL_DB_CB_lquad_double_format            = 0x107,
+	DB_PERF_SEL_DB_CB_lquad_fast_format              = 0x108,
+	DB_PERF_SEL_DB_CB_lquad_slow_format              = 0x109,
+} PerfCounter_Vals;
+typedef enum RingCounterControl {
+	COUNTER_RING_SPLIT                               = 0x0,
+	COUNTER_RING_0                                   = 0x1,
+	COUNTER_RING_1                                   = 0x2,
+} RingCounterControl;
+typedef enum PixelPipeCounterId {
+	PIXEL_PIPE_OCCLUSION_COUNT_0                     = 0x0,
+	PIXEL_PIPE_OCCLUSION_COUNT_1                     = 0x1,
+	PIXEL_PIPE_OCCLUSION_COUNT_2                     = 0x2,
+	PIXEL_PIPE_OCCLUSION_COUNT_3                     = 0x3,
+	PIXEL_PIPE_SCREEN_MIN_EXTENTS_0                  = 0x4,
+	PIXEL_PIPE_SCREEN_MAX_EXTENTS_0                  = 0x5,
+	PIXEL_PIPE_SCREEN_MIN_EXTENTS_1                  = 0x6,
+	PIXEL_PIPE_SCREEN_MAX_EXTENTS_1                  = 0x7,
+} PixelPipeCounterId;
+typedef enum PixelPipeStride {
+	PIXEL_PIPE_STRIDE_32_BITS                        = 0x0,
+	PIXEL_PIPE_STRIDE_64_BITS                        = 0x1,
+	PIXEL_PIPE_STRIDE_128_BITS                       = 0x2,
+	PIXEL_PIPE_STRIDE_256_BITS                       = 0x3,
+} PixelPipeStride;
+typedef enum GB_EDC_DED_MODE {
+	GB_EDC_DED_MODE_LOG                              = 0x0,
+	GB_EDC_DED_MODE_HALT                             = 0x1,
+	GB_EDC_DED_MODE_INT_HALT                         = 0x2,
+} GB_EDC_DED_MODE;
+#define GB_TILING_CONFIG_TABLE_SIZE               0x20
+#define GB_TILING_CONFIG_MACROTABLE_SIZE          0x10
+typedef enum GRBM_PERF_SEL {
+	GRBM_PERF_SEL_COUNT                              = 0x0,
+	GRBM_PERF_SEL_USER_DEFINED                       = 0x1,
+	GRBM_PERF_SEL_GUI_ACTIVE                         = 0x2,
+	GRBM_PERF_SEL_CP_BUSY                            = 0x3,
+	GRBM_PERF_SEL_CP_COHER_BUSY                      = 0x4,
+	GRBM_PERF_SEL_CP_DMA_BUSY                        = 0x5,
+	GRBM_PERF_SEL_CB_BUSY                            = 0x6,
+	GRBM_PERF_SEL_DB_BUSY                            = 0x7,
+	GRBM_PERF_SEL_PA_BUSY                            = 0x8,
+	GRBM_PERF_SEL_SC_BUSY                            = 0x9,
+	GRBM_PERF_SEL_RESERVED_6                         = 0xa,
+	GRBM_PERF_SEL_SPI_BUSY                           = 0xb,
+	GRBM_PERF_SEL_SX_BUSY                            = 0xc,
+	GRBM_PERF_SEL_TA_BUSY                            = 0xd,
+	GRBM_PERF_SEL_CB_CLEAN                           = 0xe,
+	GRBM_PERF_SEL_DB_CLEAN                           = 0xf,
+	GRBM_PERF_SEL_RESERVED_5                         = 0x10,
+	GRBM_PERF_SEL_VGT_BUSY                           = 0x11,
+	GRBM_PERF_SEL_RESERVED_4                         = 0x12,
+	GRBM_PERF_SEL_RESERVED_3                         = 0x13,
+	GRBM_PERF_SEL_RESERVED_2                         = 0x14,
+	GRBM_PERF_SEL_RESERVED_1                         = 0x15,
+	GRBM_PERF_SEL_RESERVED_0                         = 0x16,
+	GRBM_PERF_SEL_IA_BUSY                            = 0x17,
+	GRBM_PERF_SEL_IA_NO_DMA_BUSY                     = 0x18,
+	GRBM_PERF_SEL_GDS_BUSY                           = 0x19,
+	GRBM_PERF_SEL_BCI_BUSY                           = 0x1a,
+	GRBM_PERF_SEL_RLC_BUSY                           = 0x1b,
+	GRBM_PERF_SEL_TC_BUSY                            = 0x1c,
+	GRBM_PERF_SEL_CPG_BUSY                           = 0x1d,
+	GRBM_PERF_SEL_CPC_BUSY                           = 0x1e,
+	GRBM_PERF_SEL_CPF_BUSY                           = 0x1f,
+	GRBM_PERF_SEL_WD_BUSY                            = 0x20,
+	GRBM_PERF_SEL_WD_NO_DMA_BUSY                     = 0x21,
+} GRBM_PERF_SEL;
+typedef enum GRBM_SE0_PERF_SEL {
+	GRBM_SE0_PERF_SEL_COUNT                          = 0x0,
+	GRBM_SE0_PERF_SEL_USER_DEFINED                   = 0x1,
+	GRBM_SE0_PERF_SEL_CB_BUSY                        = 0x2,
+	GRBM_SE0_PERF_SEL_DB_BUSY                        = 0x3,
+	GRBM_SE0_PERF_SEL_SC_BUSY                        = 0x4,
+	GRBM_SE0_PERF_SEL_RESERVED_1                     = 0x5,
+	GRBM_SE0_PERF_SEL_SPI_BUSY                       = 0x6,
+	GRBM_SE0_PERF_SEL_SX_BUSY                        = 0x7,
+	GRBM_SE0_PERF_SEL_TA_BUSY                        = 0x8,
+	GRBM_SE0_PERF_SEL_CB_CLEAN                       = 0x9,
+	GRBM_SE0_PERF_SEL_DB_CLEAN                       = 0xa,
+	GRBM_SE0_PERF_SEL_RESERVED_0                     = 0xb,
+	GRBM_SE0_PERF_SEL_PA_BUSY                        = 0xc,
+	GRBM_SE0_PERF_SEL_VGT_BUSY                       = 0xd,
+	GRBM_SE0_PERF_SEL_BCI_BUSY                       = 0xe,
+} GRBM_SE0_PERF_SEL;
+typedef enum GRBM_SE1_PERF_SEL {
+	GRBM_SE1_PERF_SEL_COUNT                          = 0x0,
+	GRBM_SE1_PERF_SEL_USER_DEFINED                   = 0x1,
+	GRBM_SE1_PERF_SEL_CB_BUSY                        = 0x2,
+	GRBM_SE1_PERF_SEL_DB_BUSY                        = 0x3,
+	GRBM_SE1_PERF_SEL_SC_BUSY                        = 0x4,
+	GRBM_SE1_PERF_SEL_RESERVED_1                     = 0x5,
+	GRBM_SE1_PERF_SEL_SPI_BUSY                       = 0x6,
+	GRBM_SE1_PERF_SEL_SX_BUSY                        = 0x7,
+	GRBM_SE1_PERF_SEL_TA_BUSY                        = 0x8,
+	GRBM_SE1_PERF_SEL_CB_CLEAN                       = 0x9,
+	GRBM_SE1_PERF_SEL_DB_CLEAN                       = 0xa,
+	GRBM_SE1_PERF_SEL_RESERVED_0                     = 0xb,
+	GRBM_SE1_PERF_SEL_PA_BUSY                        = 0xc,
+	GRBM_SE1_PERF_SEL_VGT_BUSY                       = 0xd,
+	GRBM_SE1_PERF_SEL_BCI_BUSY                       = 0xe,
+} GRBM_SE1_PERF_SEL;
+typedef enum GRBM_SE2_PERF_SEL {
+	GRBM_SE2_PERF_SEL_COUNT                          = 0x0,
+	GRBM_SE2_PERF_SEL_USER_DEFINED                   = 0x1,
+	GRBM_SE2_PERF_SEL_CB_BUSY                        = 0x2,
+	GRBM_SE2_PERF_SEL_DB_BUSY                        = 0x3,
+	GRBM_SE2_PERF_SEL_SC_BUSY                        = 0x4,
+	GRBM_SE2_PERF_SEL_RESERVED_1                     = 0x5,
+	GRBM_SE2_PERF_SEL_SPI_BUSY                       = 0x6,
+	GRBM_SE2_PERF_SEL_SX_BUSY                        = 0x7,
+	GRBM_SE2_PERF_SEL_TA_BUSY                        = 0x8,
+	GRBM_SE2_PERF_SEL_CB_CLEAN                       = 0x9,
+	GRBM_SE2_PERF_SEL_DB_CLEAN                       = 0xa,
+	GRBM_SE2_PERF_SEL_RESERVED_0                     = 0xb,
+	GRBM_SE2_PERF_SEL_PA_BUSY                        = 0xc,
+	GRBM_SE2_PERF_SEL_VGT_BUSY                       = 0xd,
+	GRBM_SE2_PERF_SEL_BCI_BUSY                       = 0xe,
+} GRBM_SE2_PERF_SEL;
+typedef enum GRBM_SE3_PERF_SEL {
+	GRBM_SE3_PERF_SEL_COUNT                          = 0x0,
+	GRBM_SE3_PERF_SEL_USER_DEFINED                   = 0x1,
+	GRBM_SE3_PERF_SEL_CB_BUSY                        = 0x2,
+	GRBM_SE3_PERF_SEL_DB_BUSY                        = 0x3,
+	GRBM_SE3_PERF_SEL_SC_BUSY                        = 0x4,
+	GRBM_SE3_PERF_SEL_RESERVED_1                     = 0x5,
+	GRBM_SE3_PERF_SEL_SPI_BUSY                       = 0x6,
+	GRBM_SE3_PERF_SEL_SX_BUSY                        = 0x7,
+	GRBM_SE3_PERF_SEL_TA_BUSY                        = 0x8,
+	GRBM_SE3_PERF_SEL_CB_CLEAN                       = 0x9,
+	GRBM_SE3_PERF_SEL_DB_CLEAN                       = 0xa,
+	GRBM_SE3_PERF_SEL_RESERVED_0                     = 0xb,
+	GRBM_SE3_PERF_SEL_PA_BUSY                        = 0xc,
+	GRBM_SE3_PERF_SEL_VGT_BUSY                       = 0xd,
+	GRBM_SE3_PERF_SEL_BCI_BUSY                       = 0xe,
+} GRBM_SE3_PERF_SEL;
+typedef enum SU_PERFCNT_SEL {
+	PERF_PAPC_PASX_REQ                               = 0x0,
+	PERF_PAPC_PASX_DISABLE_PIPE                      = 0x1,
+	PERF_PAPC_PASX_FIRST_VECTOR                      = 0x2,
+	PERF_PAPC_PASX_SECOND_VECTOR                     = 0x3,
+	PERF_PAPC_PASX_FIRST_DEAD                        = 0x4,
+	PERF_PAPC_PASX_SECOND_DEAD                       = 0x5,
+	PERF_PAPC_PASX_VTX_KILL_DISCARD                  = 0x6,
+	PERF_PAPC_PASX_VTX_NAN_DISCARD                   = 0x7,
+	PERF_PAPC_PA_INPUT_PRIM                          = 0x8,
+	PERF_PAPC_PA_INPUT_NULL_PRIM                     = 0x9,
+	PERF_PAPC_PA_INPUT_EVENT_FLAG                    = 0xa,
+	PERF_PAPC_PA_INPUT_FIRST_PRIM_SLOT               = 0xb,
+	PERF_PAPC_PA_INPUT_END_OF_PACKET                 = 0xc,
+	PERF_PAPC_PA_INPUT_EXTENDED_EVENT                = 0xd,
+	PERF_PAPC_CLPR_CULL_PRIM                         = 0xe,
+	PERF_PAPC_CLPR_VVUCP_CULL_PRIM                   = 0xf,
+	PERF_PAPC_CLPR_VV_CULL_PRIM                      = 0x10,
+	PERF_PAPC_CLPR_UCP_CULL_PRIM                     = 0x11,
+	PERF_PAPC_CLPR_VTX_KILL_CULL_PRIM                = 0x12,
+	PERF_PAPC_CLPR_VTX_NAN_CULL_PRIM                 = 0x13,
+	PERF_PAPC_CLPR_CULL_TO_NULL_PRIM                 = 0x14,
+	PERF_PAPC_CLPR_VVUCP_CLIP_PRIM                   = 0x15,
+	PERF_PAPC_CLPR_VV_CLIP_PRIM                      = 0x16,
+	PERF_PAPC_CLPR_UCP_CLIP_PRIM                     = 0x17,
+	PERF_PAPC_CLPR_POINT_CLIP_CANDIDATE              = 0x18,
+	PERF_PAPC_CLPR_CLIP_PLANE_CNT_1                  = 0x19,
+	PERF_PAPC_CLPR_CLIP_PLANE_CNT_2                  = 0x1a,
+	PERF_PAPC_CLPR_CLIP_PLANE_CNT_3                  = 0x1b,
+	PERF_PAPC_CLPR_CLIP_PLANE_CNT_4                  = 0x1c,
+	PERF_PAPC_CLPR_CLIP_PLANE_CNT_5_8                = 0x1d,
+	PERF_PAPC_CLPR_CLIP_PLANE_CNT_9_12               = 0x1e,
+	PERF_PAPC_CLPR_CLIP_PLANE_NEAR                   = 0x1f,
+	PERF_PAPC_CLPR_CLIP_PLANE_FAR                    = 0x20,
+	PERF_PAPC_CLPR_CLIP_PLANE_LEFT                   = 0x21,
+	PERF_PAPC_CLPR_CLIP_PLANE_RIGHT                  = 0x22,
+	PERF_PAPC_CLPR_CLIP_PLANE_TOP                    = 0x23,
+	PERF_PAPC_CLPR_CLIP_PLANE_BOTTOM                 = 0x24,
+	PERF_PAPC_CLPR_GSC_KILL_CULL_PRIM                = 0x25,
+	PERF_PAPC_CLPR_RASTER_KILL_CULL_PRIM             = 0x26,
+	PERF_PAPC_CLSM_NULL_PRIM                         = 0x27,
+	PERF_PAPC_CLSM_TOTALLY_VISIBLE_PRIM              = 0x28,
+	PERF_PAPC_CLSM_CULL_TO_NULL_PRIM                 = 0x29,
+	PERF_PAPC_CLSM_OUT_PRIM_CNT_1                    = 0x2a,
+	PERF_PAPC_CLSM_OUT_PRIM_CNT_2                    = 0x2b,
+	PERF_PAPC_CLSM_OUT_PRIM_CNT_3                    = 0x2c,
+	PERF_PAPC_CLSM_OUT_PRIM_CNT_4                    = 0x2d,
+	PERF_PAPC_CLSM_OUT_PRIM_CNT_5_8                  = 0x2e,
+	PERF_PAPC_CLSM_OUT_PRIM_CNT_9_13                 = 0x2f,
+	PERF_PAPC_CLIPGA_VTE_KILL_PRIM                   = 0x30,
+	PERF_PAPC_SU_INPUT_PRIM                          = 0x31,
+	PERF_PAPC_SU_INPUT_CLIP_PRIM                     = 0x32,
+	PERF_PAPC_SU_INPUT_NULL_PRIM                     = 0x33,
+	PERF_PAPC_SU_INPUT_PRIM_DUAL                     = 0x34,
+	PERF_PAPC_SU_INPUT_CLIP_PRIM_DUAL                = 0x35,
+	PERF_PAPC_SU_ZERO_AREA_CULL_PRIM                 = 0x36,
+	PERF_PAPC_SU_BACK_FACE_CULL_PRIM                 = 0x37,
+	PERF_PAPC_SU_FRONT_FACE_CULL_PRIM                = 0x38,
+	PERF_PAPC_SU_POLYMODE_FACE_CULL                  = 0x39,
+	PERF_PAPC_SU_POLYMODE_BACK_CULL                  = 0x3a,
+	PERF_PAPC_SU_POLYMODE_FRONT_CULL                 = 0x3b,
+	PERF_PAPC_SU_POLYMODE_INVALID_FILL               = 0x3c,
+	PERF_PAPC_SU_OUTPUT_PRIM                         = 0x3d,
+	PERF_PAPC_SU_OUTPUT_CLIP_PRIM                    = 0x3e,
+	PERF_PAPC_SU_OUTPUT_NULL_PRIM                    = 0x3f,
+	PERF_PAPC_SU_OUTPUT_EVENT_FLAG                   = 0x40,
+	PERF_PAPC_SU_OUTPUT_FIRST_PRIM_SLOT              = 0x41,
+	PERF_PAPC_SU_OUTPUT_END_OF_PACKET                = 0x42,
+	PERF_PAPC_SU_OUTPUT_POLYMODE_FACE                = 0x43,
+	PERF_PAPC_SU_OUTPUT_POLYMODE_BACK                = 0x44,
+	PERF_PAPC_SU_OUTPUT_POLYMODE_FRONT               = 0x45,
+	PERF_PAPC_SU_OUT_CLIP_POLYMODE_FACE              = 0x46,
+	PERF_PAPC_SU_OUT_CLIP_POLYMODE_BACK              = 0x47,
+	PERF_PAPC_SU_OUT_CLIP_POLYMODE_FRONT             = 0x48,
+	PERF_PAPC_SU_OUTPUT_PRIM_DUAL                    = 0x49,
+	PERF_PAPC_SU_OUTPUT_CLIP_PRIM_DUAL               = 0x4a,
+	PERF_PAPC_SU_OUTPUT_POLYMODE_DUAL                = 0x4b,
+	PERF_PAPC_SU_OUTPUT_CLIP_POLYMODE_DUAL           = 0x4c,
+	PERF_PAPC_PASX_REQ_IDLE                          = 0x4d,
+	PERF_PAPC_PASX_REQ_BUSY                          = 0x4e,
+	PERF_PAPC_PASX_REQ_STALLED                       = 0x4f,
+	PERF_PAPC_PASX_REC_IDLE                          = 0x50,
+	PERF_PAPC_PASX_REC_BUSY                          = 0x51,
+	PERF_PAPC_PASX_REC_STARVED_SX                    = 0x52,
+	PERF_PAPC_PASX_REC_STALLED                       = 0x53,
+	PERF_PAPC_PASX_REC_STALLED_POS_MEM               = 0x54,
+	PERF_PAPC_PASX_REC_STALLED_CCGSM_IN              = 0x55,
+	PERF_PAPC_CCGSM_IDLE                             = 0x56,
+	PERF_PAPC_CCGSM_BUSY                             = 0x57,
+	PERF_PAPC_CCGSM_STALLED                          = 0x58,
+	PERF_PAPC_CLPRIM_IDLE                            = 0x59,
+	PERF_PAPC_CLPRIM_BUSY                            = 0x5a,
+	PERF_PAPC_CLPRIM_STALLED                         = 0x5b,
+	PERF_PAPC_CLPRIM_STARVED_CCGSM                   = 0x5c,
+	PERF_PAPC_CLIPSM_IDLE                            = 0x5d,
+	PERF_PAPC_CLIPSM_BUSY                            = 0x5e,
+	PERF_PAPC_CLIPSM_WAIT_CLIP_VERT_ENGH             = 0x5f,
+	PERF_PAPC_CLIPSM_WAIT_HIGH_PRI_SEQ               = 0x60,
+	PERF_PAPC_CLIPSM_WAIT_CLIPGA                     = 0x61,
+	PERF_PAPC_CLIPSM_WAIT_AVAIL_VTE_CLIP             = 0x62,
+	PERF_PAPC_CLIPSM_WAIT_CLIP_OUTSM                 = 0x63,
+	PERF_PAPC_CLIPGA_IDLE                            = 0x64,
+	PERF_PAPC_CLIPGA_BUSY                            = 0x65,
+	PERF_PAPC_CLIPGA_STARVED_VTE_CLIP                = 0x66,
+	PERF_PAPC_CLIPGA_STALLED                         = 0x67,
+	PERF_PAPC_CLIP_IDLE                              = 0x68,
+	PERF_PAPC_CLIP_BUSY                              = 0x69,
+	PERF_PAPC_SU_IDLE                                = 0x6a,
+	PERF_PAPC_SU_BUSY                                = 0x6b,
+	PERF_PAPC_SU_STARVED_CLIP                        = 0x6c,
+	PERF_PAPC_SU_STALLED_SC                          = 0x6d,
+	PERF_PAPC_CL_DYN_SCLK_VLD                        = 0x6e,
+	PERF_PAPC_SU_DYN_SCLK_VLD                        = 0x6f,
+	PERF_PAPC_PA_REG_SCLK_VLD                        = 0x70,
+	PERF_PAPC_SU_MULTI_GPU_PRIM_FILTER_CULL          = 0x71,
+	PERF_PAPC_PASX_SE0_REQ                           = 0x72,
+	PERF_PAPC_PASX_SE1_REQ                           = 0x73,
+	PERF_PAPC_PASX_SE0_FIRST_VECTOR                  = 0x74,
+	PERF_PAPC_PASX_SE0_SECOND_VECTOR                 = 0x75,
+	PERF_PAPC_PASX_SE1_FIRST_VECTOR                  = 0x76,
+	PERF_PAPC_PASX_SE1_SECOND_VECTOR                 = 0x77,
+	PERF_PAPC_SU_SE0_PRIM_FILTER_CULL                = 0x78,
+	PERF_PAPC_SU_SE1_PRIM_FILTER_CULL                = 0x79,
+	PERF_PAPC_SU_SE01_PRIM_FILTER_CULL               = 0x7a,
+	PERF_PAPC_SU_SE0_OUTPUT_PRIM                     = 0x7b,
+	PERF_PAPC_SU_SE1_OUTPUT_PRIM                     = 0x7c,
+	PERF_PAPC_SU_SE01_OUTPUT_PRIM                    = 0x7d,
+	PERF_PAPC_SU_SE0_OUTPUT_NULL_PRIM                = 0x7e,
+	PERF_PAPC_SU_SE1_OUTPUT_NULL_PRIM                = 0x7f,
+	PERF_PAPC_SU_SE01_OUTPUT_NULL_PRIM               = 0x80,
+	PERF_PAPC_SU_SE0_OUTPUT_FIRST_PRIM_SLOT          = 0x81,
+	PERF_PAPC_SU_SE1_OUTPUT_FIRST_PRIM_SLOT          = 0x82,
+	PERF_PAPC_SU_SE0_STALLED_SC                      = 0x83,
+	PERF_PAPC_SU_SE1_STALLED_SC                      = 0x84,
+	PERF_PAPC_SU_SE01_STALLED_SC                     = 0x85,
+	PERF_PAPC_CLSM_CLIPPING_PRIM                     = 0x86,
+	PERF_PAPC_SU_CULLED_PRIM                         = 0x87,
+	PERF_PAPC_SU_OUTPUT_EOPG                         = 0x88,
+	PERF_PAPC_SU_SE2_PRIM_FILTER_CULL                = 0x89,
+	PERF_PAPC_SU_SE3_PRIM_FILTER_CULL                = 0x8a,
+	PERF_PAPC_SU_SE2_OUTPUT_PRIM                     = 0x8b,
+	PERF_PAPC_SU_SE3_OUTPUT_PRIM                     = 0x8c,
+	PERF_PAPC_SU_SE2_OUTPUT_NULL_PRIM                = 0x8d,
+	PERF_PAPC_SU_SE3_OUTPUT_NULL_PRIM                = 0x8e,
+	PERF_PAPC_SU_SE0_OUTPUT_END_OF_PACKET            = 0x8f,
+	PERF_PAPC_SU_SE1_OUTPUT_END_OF_PACKET            = 0x90,
+	PERF_PAPC_SU_SE2_OUTPUT_END_OF_PACKET            = 0x91,
+	PERF_PAPC_SU_SE3_OUTPUT_END_OF_PACKET            = 0x92,
+	PERF_PAPC_SU_SE0_OUTPUT_EOPG                     = 0x93,
+	PERF_PAPC_SU_SE1_OUTPUT_EOPG                     = 0x94,
+	PERF_PAPC_SU_SE2_OUTPUT_EOPG                     = 0x95,
+	PERF_PAPC_SU_SE3_OUTPUT_EOPG                     = 0x96,
+	PERF_PAPC_SU_SE2_STALLED_SC                      = 0x97,
+	PERF_PAPC_SU_SE3_STALLED_SC                      = 0x98,
+} SU_PERFCNT_SEL;
+typedef enum SC_PERFCNT_SEL {
+	SC_SRPS_WINDOW_VALID                             = 0x0,
+	SC_PSSW_WINDOW_VALID                             = 0x1,
+	SC_TPQZ_WINDOW_VALID                             = 0x2,
+	SC_QZQP_WINDOW_VALID                             = 0x3,
+	SC_TRPK_WINDOW_VALID                             = 0x4,
+	SC_SRPS_WINDOW_VALID_BUSY                        = 0x5,
+	SC_PSSW_WINDOW_VALID_BUSY                        = 0x6,
+	SC_TPQZ_WINDOW_VALID_BUSY                        = 0x7,
+	SC_QZQP_WINDOW_VALID_BUSY                        = 0x8,
+	SC_TRPK_WINDOW_VALID_BUSY                        = 0x9,
+	SC_STARVED_BY_PA                                 = 0xa,
+	SC_STALLED_BY_PRIMFIFO                           = 0xb,
+	SC_STALLED_BY_DB_TILE                            = 0xc,
+	SC_STARVED_BY_DB_TILE                            = 0xd,
+	SC_STALLED_BY_TILEORDERFIFO                      = 0xe,
+	SC_STALLED_BY_TILEFIFO                           = 0xf,
+	SC_STALLED_BY_DB_QUAD                            = 0x10,
+	SC_STARVED_BY_DB_QUAD                            = 0x11,
+	SC_STALLED_BY_QUADFIFO                           = 0x12,
+	SC_STALLED_BY_BCI                                = 0x13,
+	SC_STALLED_BY_SPI                                = 0x14,
+	SC_SCISSOR_DISCARD                               = 0x15,
+	SC_BB_DISCARD                                    = 0x16,
+	SC_SUPERTILE_COUNT                               = 0x17,
+	SC_SUPERTILE_PER_PRIM_H0                         = 0x18,
+	SC_SUPERTILE_PER_PRIM_H1                         = 0x19,
+	SC_SUPERTILE_PER_PRIM_H2                         = 0x1a,
+	SC_SUPERTILE_PER_PRIM_H3                         = 0x1b,
+	SC_SUPERTILE_PER_PRIM_H4                         = 0x1c,
+	SC_SUPERTILE_PER_PRIM_H5                         = 0x1d,
+	SC_SUPERTILE_PER_PRIM_H6                         = 0x1e,
+	SC_SUPERTILE_PER_PRIM_H7                         = 0x1f,
+	SC_SUPERTILE_PER_PRIM_H8                         = 0x20,
+	SC_SUPERTILE_PER_PRIM_H9                         = 0x21,
+	SC_SUPERTILE_PER_PRIM_H10                        = 0x22,
+	SC_SUPERTILE_PER_PRIM_H11                        = 0x23,
+	SC_SUPERTILE_PER_PRIM_H12                        = 0x24,
+	SC_SUPERTILE_PER_PRIM_H13                        = 0x25,
+	SC_SUPERTILE_PER_PRIM_H14                        = 0x26,
+	SC_SUPERTILE_PER_PRIM_H15                        = 0x27,
+	SC_SUPERTILE_PER_PRIM_H16                        = 0x28,
+	SC_TILE_PER_PRIM_H0                              = 0x29,
+	SC_TILE_PER_PRIM_H1                              = 0x2a,
+	SC_TILE_PER_PRIM_H2                              = 0x2b,
+	SC_TILE_PER_PRIM_H3                              = 0x2c,
+	SC_TILE_PER_PRIM_H4                              = 0x2d,
+	SC_TILE_PER_PRIM_H5                              = 0x2e,
+	SC_TILE_PER_PRIM_H6                              = 0x2f,
+	SC_TILE_PER_PRIM_H7                              = 0x30,
+	SC_TILE_PER_PRIM_H8                              = 0x31,
+	SC_TILE_PER_PRIM_H9                              = 0x32,
+	SC_TILE_PER_PRIM_H10                             = 0x33,
+	SC_TILE_PER_PRIM_H11                             = 0x34,
+	SC_TILE_PER_PRIM_H12                             = 0x35,
+	SC_TILE_PER_PRIM_H13                             = 0x36,
+	SC_TILE_PER_PRIM_H14                             = 0x37,
+	SC_TILE_PER_PRIM_H15                             = 0x38,
+	SC_TILE_PER_PRIM_H16                             = 0x39,
+	SC_TILE_PER_SUPERTILE_H0                         = 0x3a,
+	SC_TILE_PER_SUPERTILE_H1                         = 0x3b,
+	SC_TILE_PER_SUPERTILE_H2                         = 0x3c,
+	SC_TILE_PER_SUPERTILE_H3                         = 0x3d,
+	SC_TILE_PER_SUPERTILE_H4                         = 0x3e,
+	SC_TILE_PER_SUPERTILE_H5                         = 0x3f,
+	SC_TILE_PER_SUPERTILE_H6                         = 0x40,
+	SC_TILE_PER_SUPERTILE_H7                         = 0x41,
+	SC_TILE_PER_SUPERTILE_H8                         = 0x42,
+	SC_TILE_PER_SUPERTILE_H9                         = 0x43,
+	SC_TILE_PER_SUPERTILE_H10                        = 0x44,
+	SC_TILE_PER_SUPERTILE_H11                        = 0x45,
+	SC_TILE_PER_SUPERTILE_H12                        = 0x46,
+	SC_TILE_PER_SUPERTILE_H13                        = 0x47,
+	SC_TILE_PER_SUPERTILE_H14                        = 0x48,
+	SC_TILE_PER_SUPERTILE_H15                        = 0x49,
+	SC_TILE_PER_SUPERTILE_H16                        = 0x4a,
+	SC_TILE_PICKED_H1                                = 0x4b,
+	SC_TILE_PICKED_H2                                = 0x4c,
+	SC_TILE_PICKED_H3                                = 0x4d,
+	SC_TILE_PICKED_H4                                = 0x4e,
+	SC_QZ0_MULTI_GPU_TILE_DISCARD                    = 0x4f,
+	SC_QZ1_MULTI_GPU_TILE_DISCARD                    = 0x50,
+	SC_QZ2_MULTI_GPU_TILE_DISCARD                    = 0x51,
+	SC_QZ3_MULTI_GPU_TILE_DISCARD                    = 0x52,
+	SC_QZ0_TILE_COUNT                                = 0x53,
+	SC_QZ1_TILE_COUNT                                = 0x54,
+	SC_QZ2_TILE_COUNT                                = 0x55,
+	SC_QZ3_TILE_COUNT                                = 0x56,
+	SC_QZ0_TILE_COVERED_COUNT                        = 0x57,
+	SC_QZ1_TILE_COVERED_COUNT                        = 0x58,
+	SC_QZ2_TILE_COVERED_COUNT                        = 0x59,
+	SC_QZ3_TILE_COVERED_COUNT                        = 0x5a,
+	SC_QZ0_TILE_NOT_COVERED_COUNT                    = 0x5b,
+	SC_QZ1_TILE_NOT_COVERED_COUNT                    = 0x5c,
+	SC_QZ2_TILE_NOT_COVERED_COUNT                    = 0x5d,
+	SC_QZ3_TILE_NOT_COVERED_COUNT                    = 0x5e,
+	SC_QZ0_QUAD_PER_TILE_H0                          = 0x5f,
+	SC_QZ0_QUAD_PER_TILE_H1                          = 0x60,
+	SC_QZ0_QUAD_PER_TILE_H2                          = 0x61,
+	SC_QZ0_QUAD_PER_TILE_H3                          = 0x62,
+	SC_QZ0_QUAD_PER_TILE_H4                          = 0x63,
+	SC_QZ0_QUAD_PER_TILE_H5                          = 0x64,
+	SC_QZ0_QUAD_PER_TILE_H6                          = 0x65,
+	SC_QZ0_QUAD_PER_TILE_H7                          = 0x66,
+	SC_QZ0_QUAD_PER_TILE_H8                          = 0x67,
+	SC_QZ0_QUAD_PER_TILE_H9                          = 0x68,
+	SC_QZ0_QUAD_PER_TILE_H10                         = 0x69,
+	SC_QZ0_QUAD_PER_TILE_H11                         = 0x6a,
+	SC_QZ0_QUAD_PER_TILE_H12                         = 0x6b,
+	SC_QZ0_QUAD_PER_TILE_H13                         = 0x6c,
+	SC_QZ0_QUAD_PER_TILE_H14                         = 0x6d,
+	SC_QZ0_QUAD_PER_TILE_H15                         = 0x6e,
+	SC_QZ0_QUAD_PER_TILE_H16                         = 0x6f,
+	SC_QZ1_QUAD_PER_TILE_H0                          = 0x70,
+	SC_QZ1_QUAD_PER_TILE_H1                          = 0x71,
+	SC_QZ1_QUAD_PER_TILE_H2                          = 0x72,
+	SC_QZ1_QUAD_PER_TILE_H3                          = 0x73,
+	SC_QZ1_QUAD_PER_TILE_H4                          = 0x74,
+	SC_QZ1_QUAD_PER_TILE_H5                          = 0x75,
+	SC_QZ1_QUAD_PER_TILE_H6                          = 0x76,
+	SC_QZ1_QUAD_PER_TILE_H7                          = 0x77,
+	SC_QZ1_QUAD_PER_TILE_H8                          = 0x78,
+	SC_QZ1_QUAD_PER_TILE_H9                          = 0x79,
+	SC_QZ1_QUAD_PER_TILE_H10                         = 0x7a,
+	SC_QZ1_QUAD_PER_TILE_H11                         = 0x7b,
+	SC_QZ1_QUAD_PER_TILE_H12                         = 0x7c,
+	SC_QZ1_QUAD_PER_TILE_H13                         = 0x7d,
+	SC_QZ1_QUAD_PER_TILE_H14                         = 0x7e,
+	SC_QZ1_QUAD_PER_TILE_H15                         = 0x7f,
+	SC_QZ1_QUAD_PER_TILE_H16                         = 0x80,
+	SC_QZ2_QUAD_PER_TILE_H0                          = 0x81,
+	SC_QZ2_QUAD_PER_TILE_H1                          = 0x82,
+	SC_QZ2_QUAD_PER_TILE_H2                          = 0x83,
+	SC_QZ2_QUAD_PER_TILE_H3                          = 0x84,
+	SC_QZ2_QUAD_PER_TILE_H4                          = 0x85,
+	SC_QZ2_QUAD_PER_TILE_H5                          = 0x86,
+	SC_QZ2_QUAD_PER_TILE_H6                          = 0x87,
+	SC_QZ2_QUAD_PER_TILE_H7                          = 0x88,
+	SC_QZ2_QUAD_PER_TILE_H8                          = 0x89,
+	SC_QZ2_QUAD_PER_TILE_H9                          = 0x8a,
+	SC_QZ2_QUAD_PER_TILE_H10                         = 0x8b,
+	SC_QZ2_QUAD_PER_TILE_H11                         = 0x8c,
+	SC_QZ2_QUAD_PER_TILE_H12                         = 0x8d,
+	SC_QZ2_QUAD_PER_TILE_H13                         = 0x8e,
+	SC_QZ2_QUAD_PER_TILE_H14                         = 0x8f,
+	SC_QZ2_QUAD_PER_TILE_H15                         = 0x90,
+	SC_QZ2_QUAD_PER_TILE_H16                         = 0x91,
+	SC_QZ3_QUAD_PER_TILE_H0                          = 0x92,
+	SC_QZ3_QUAD_PER_TILE_H1                          = 0x93,
+	SC_QZ3_QUAD_PER_TILE_H2                          = 0x94,
+	SC_QZ3_QUAD_PER_TILE_H3                          = 0x95,
+	SC_QZ3_QUAD_PER_TILE_H4                          = 0x96,
+	SC_QZ3_QUAD_PER_TILE_H5                          = 0x97,
+	SC_QZ3_QUAD_PER_TILE_H6                          = 0x98,
+	SC_QZ3_QUAD_PER_TILE_H7                          = 0x99,
+	SC_QZ3_QUAD_PER_TILE_H8                          = 0x9a,
+	SC_QZ3_QUAD_PER_TILE_H9                          = 0x9b,
+	SC_QZ3_QUAD_PER_TILE_H10                         = 0x9c,
+	SC_QZ3_QUAD_PER_TILE_H11                         = 0x9d,
+	SC_QZ3_QUAD_PER_TILE_H12                         = 0x9e,
+	SC_QZ3_QUAD_PER_TILE_H13                         = 0x9f,
+	SC_QZ3_QUAD_PER_TILE_H14                         = 0xa0,
+	SC_QZ3_QUAD_PER_TILE_H15                         = 0xa1,
+	SC_QZ3_QUAD_PER_TILE_H16                         = 0xa2,
+	SC_QZ0_QUAD_COUNT                                = 0xa3,
+	SC_QZ1_QUAD_COUNT                                = 0xa4,
+	SC_QZ2_QUAD_COUNT                                = 0xa5,
+	SC_QZ3_QUAD_COUNT                                = 0xa6,
+	SC_P0_HIZ_TILE_COUNT                             = 0xa7,
+	SC_P1_HIZ_TILE_COUNT                             = 0xa8,
+	SC_P2_HIZ_TILE_COUNT                             = 0xa9,
+	SC_P3_HIZ_TILE_COUNT                             = 0xaa,
+	SC_P0_HIZ_QUAD_PER_TILE_H0                       = 0xab,
+	SC_P0_HIZ_QUAD_PER_TILE_H1                       = 0xac,
+	SC_P0_HIZ_QUAD_PER_TILE_H2                       = 0xad,
+	SC_P0_HIZ_QUAD_PER_TILE_H3                       = 0xae,
+	SC_P0_HIZ_QUAD_PER_TILE_H4                       = 0xaf,
+	SC_P0_HIZ_QUAD_PER_TILE_H5                       = 0xb0,
+	SC_P0_HIZ_QUAD_PER_TILE_H6                       = 0xb1,
+	SC_P0_HIZ_QUAD_PER_TILE_H7                       = 0xb2,
+	SC_P0_HIZ_QUAD_PER_TILE_H8                       = 0xb3,
+	SC_P0_HIZ_QUAD_PER_TILE_H9                       = 0xb4,
+	SC_P0_HIZ_QUAD_PER_TILE_H10                      = 0xb5,
+	SC_P0_HIZ_QUAD_PER_TILE_H11                      = 0xb6,
+	SC_P0_HIZ_QUAD_PER_TILE_H12                      = 0xb7,
+	SC_P0_HIZ_QUAD_PER_TILE_H13                      = 0xb8,
+	SC_P0_HIZ_QUAD_PER_TILE_H14                      = 0xb9,
+	SC_P0_HIZ_QUAD_PER_TILE_H15                      = 0xba,
+	SC_P0_HIZ_QUAD_PER_TILE_H16                      = 0xbb,
+	SC_P1_HIZ_QUAD_PER_TILE_H0                       = 0xbc,
+	SC_P1_HIZ_QUAD_PER_TILE_H1                       = 0xbd,
+	SC_P1_HIZ_QUAD_PER_TILE_H2                       = 0xbe,
+	SC_P1_HIZ_QUAD_PER_TILE_H3                       = 0xbf,
+	SC_P1_HIZ_QUAD_PER_TILE_H4                       = 0xc0,
+	SC_P1_HIZ_QUAD_PER_TILE_H5                       = 0xc1,
+	SC_P1_HIZ_QUAD_PER_TILE_H6                       = 0xc2,
+	SC_P1_HIZ_QUAD_PER_TILE_H7                       = 0xc3,
+	SC_P1_HIZ_QUAD_PER_TILE_H8                       = 0xc4,
+	SC_P1_HIZ_QUAD_PER_TILE_H9                       = 0xc5,
+	SC_P1_HIZ_QUAD_PER_TILE_H10                      = 0xc6,
+	SC_P1_HIZ_QUAD_PER_TILE_H11                      = 0xc7,
+	SC_P1_HIZ_QUAD_PER_TILE_H12                      = 0xc8,
+	SC_P1_HIZ_QUAD_PER_TILE_H13                      = 0xc9,
+	SC_P1_HIZ_QUAD_PER_TILE_H14                      = 0xca,
+	SC_P1_HIZ_QUAD_PER_TILE_H15                      = 0xcb,
+	SC_P1_HIZ_QUAD_PER_TILE_H16                      = 0xcc,
+	SC_P2_HIZ_QUAD_PER_TILE_H0                       = 0xcd,
+	SC_P2_HIZ_QUAD_PER_TILE_H1                       = 0xce,
+	SC_P2_HIZ_QUAD_PER_TILE_H2                       = 0xcf,
+	SC_P2_HIZ_QUAD_PER_TILE_H3                       = 0xd0,
+	SC_P2_HIZ_QUAD_PER_TILE_H4                       = 0xd1,
+	SC_P2_HIZ_QUAD_PER_TILE_H5                       = 0xd2,
+	SC_P2_HIZ_QUAD_PER_TILE_H6                       = 0xd3,
+	SC_P2_HIZ_QUAD_PER_TILE_H7                       = 0xd4,
+	SC_P2_HIZ_QUAD_PER_TILE_H8                       = 0xd5,
+	SC_P2_HIZ_QUAD_PER_TILE_H9                       = 0xd6,
+	SC_P2_HIZ_QUAD_PER_TILE_H10                      = 0xd7,
+	SC_P2_HIZ_QUAD_PER_TILE_H11                      = 0xd8,
+	SC_P2_HIZ_QUAD_PER_TILE_H12                      = 0xd9,
+	SC_P2_HIZ_QUAD_PER_TILE_H13                      = 0xda,
+	SC_P2_HIZ_QUAD_PER_TILE_H14                      = 0xdb,
+	SC_P2_HIZ_QUAD_PER_TILE_H15                      = 0xdc,
+	SC_P2_HIZ_QUAD_PER_TILE_H16                      = 0xdd,
+	SC_P3_HIZ_QUAD_PER_TILE_H0                       = 0xde,
+	SC_P3_HIZ_QUAD_PER_TILE_H1                       = 0xdf,
+	SC_P3_HIZ_QUAD_PER_TILE_H2                       = 0xe0,
+	SC_P3_HIZ_QUAD_PER_TILE_H3                       = 0xe1,
+	SC_P3_HIZ_QUAD_PER_TILE_H4                       = 0xe2,
+	SC_P3_HIZ_QUAD_PER_TILE_H5                       = 0xe3,
+	SC_P3_HIZ_QUAD_PER_TILE_H6                       = 0xe4,
+	SC_P3_HIZ_QUAD_PER_TILE_H7                       = 0xe5,
+	SC_P3_HIZ_QUAD_PER_TILE_H8                       = 0xe6,
+	SC_P3_HIZ_QUAD_PER_TILE_H9                       = 0xe7,
+	SC_P3_HIZ_QUAD_PER_TILE_H10                      = 0xe8,
+	SC_P3_HIZ_QUAD_PER_TILE_H11                      = 0xe9,
+	SC_P3_HIZ_QUAD_PER_TILE_H12                      = 0xea,
+	SC_P3_HIZ_QUAD_PER_TILE_H13                      = 0xeb,
+	SC_P3_HIZ_QUAD_PER_TILE_H14                      = 0xec,
+	SC_P3_HIZ_QUAD_PER_TILE_H15                      = 0xed,
+	SC_P3_HIZ_QUAD_PER_TILE_H16                      = 0xee,
+	SC_P0_HIZ_QUAD_COUNT                             = 0xef,
+	SC_P1_HIZ_QUAD_COUNT                             = 0xf0,
+	SC_P2_HIZ_QUAD_COUNT                             = 0xf1,
+	SC_P3_HIZ_QUAD_COUNT                             = 0xf2,
+	SC_P0_DETAIL_QUAD_COUNT                          = 0xf3,
+	SC_P1_DETAIL_QUAD_COUNT                          = 0xf4,
+	SC_P2_DETAIL_QUAD_COUNT                          = 0xf5,
+	SC_P3_DETAIL_QUAD_COUNT                          = 0xf6,
+	SC_P0_DETAIL_QUAD_WITH_1_PIX                     = 0xf7,
+	SC_P0_DETAIL_QUAD_WITH_2_PIX                     = 0xf8,
+	SC_P0_DETAIL_QUAD_WITH_3_PIX                     = 0xf9,
+	SC_P0_DETAIL_QUAD_WITH_4_PIX                     = 0xfa,
+	SC_P1_DETAIL_QUAD_WITH_1_PIX                     = 0xfb,
+	SC_P1_DETAIL_QUAD_WITH_2_PIX                     = 0xfc,
+	SC_P1_DETAIL_QUAD_WITH_3_PIX                     = 0xfd,
+	SC_P1_DETAIL_QUAD_WITH_4_PIX                     = 0xfe,
+	SC_P2_DETAIL_QUAD_WITH_1_PIX                     = 0xff,
+	SC_P2_DETAIL_QUAD_WITH_2_PIX                     = 0x100,
+	SC_P2_DETAIL_QUAD_WITH_3_PIX                     = 0x101,
+	SC_P2_DETAIL_QUAD_WITH_4_PIX                     = 0x102,
+	SC_P3_DETAIL_QUAD_WITH_1_PIX                     = 0x103,
+	SC_P3_DETAIL_QUAD_WITH_2_PIX                     = 0x104,
+	SC_P3_DETAIL_QUAD_WITH_3_PIX                     = 0x105,
+	SC_P3_DETAIL_QUAD_WITH_4_PIX                     = 0x106,
+	SC_EARLYZ_QUAD_COUNT                             = 0x107,
+	SC_EARLYZ_QUAD_WITH_1_PIX                        = 0x108,
+	SC_EARLYZ_QUAD_WITH_2_PIX                        = 0x109,
+	SC_EARLYZ_QUAD_WITH_3_PIX                        = 0x10a,
+	SC_EARLYZ_QUAD_WITH_4_PIX                        = 0x10b,
+	SC_PKR_QUAD_PER_ROW_H1                           = 0x10c,
+	SC_PKR_QUAD_PER_ROW_H2                           = 0x10d,
+	SC_PKR_4X2_QUAD_SPLIT                            = 0x10e,
+	SC_PKR_4X2_FILL_QUAD                             = 0x10f,
+	SC_PKR_END_OF_VECTOR                             = 0x110,
+	SC_PKR_CONTROL_XFER                              = 0x111,
+	SC_PKR_DBHANG_FORCE_EOV                          = 0x112,
+	SC_REG_SCLK_BUSY                                 = 0x113,
+	SC_GRP0_DYN_SCLK_BUSY                            = 0x114,
+	SC_GRP1_DYN_SCLK_BUSY                            = 0x115,
+	SC_GRP2_DYN_SCLK_BUSY                            = 0x116,
+	SC_GRP3_DYN_SCLK_BUSY                            = 0x117,
+	SC_GRP4_DYN_SCLK_BUSY                            = 0x118,
+	SC_PA0_SC_DATA_FIFO_RD                           = 0x119,
+	SC_PA0_SC_DATA_FIFO_WE                           = 0x11a,
+	SC_PA1_SC_DATA_FIFO_RD                           = 0x11b,
+	SC_PA1_SC_DATA_FIFO_WE                           = 0x11c,
+	SC_PS_ARB_XFC_ALL_EVENT_OR_PRIM_CYCLES           = 0x11d,
+	SC_PS_ARB_XFC_ONLY_PRIM_CYCLES                   = 0x11e,
+	SC_PS_ARB_XFC_ONLY_ONE_INC_PER_PRIM              = 0x11f,
+	SC_PS_ARB_STALLED_FROM_BELOW                     = 0x120,
+	SC_PS_ARB_STARVED_FROM_ABOVE                     = 0x121,
+	SC_PS_ARB_SC_BUSY                                = 0x122,
+	SC_PS_ARB_PA_SC_BUSY                             = 0x123,
+	SC_PA2_SC_DATA_FIFO_RD                           = 0x124,
+	SC_PA2_SC_DATA_FIFO_WE                           = 0x125,
+	SC_PA3_SC_DATA_FIFO_RD                           = 0x126,
+	SC_PA3_SC_DATA_FIFO_WE                           = 0x127,
+	SC_PA_SC_DEALLOC_0_0_WE                          = 0x128,
+	SC_PA_SC_DEALLOC_0_1_WE                          = 0x129,
+	SC_PA_SC_DEALLOC_1_0_WE                          = 0x12a,
+	SC_PA_SC_DEALLOC_1_1_WE                          = 0x12b,
+	SC_PA_SC_DEALLOC_2_0_WE                          = 0x12c,
+	SC_PA_SC_DEALLOC_2_1_WE                          = 0x12d,
+	SC_PA_SC_DEALLOC_3_0_WE                          = 0x12e,
+	SC_PA_SC_DEALLOC_3_1_WE                          = 0x12f,
+	SC_PA0_SC_EOP_WE                                 = 0x130,
+	SC_PA0_SC_EOPG_WE                                = 0x131,
+	SC_PA0_SC_EVENT_WE                               = 0x132,
+	SC_PA1_SC_EOP_WE                                 = 0x133,
+	SC_PA1_SC_EOPG_WE                                = 0x134,
+	SC_PA1_SC_EVENT_WE                               = 0x135,
+	SC_PA2_SC_EOP_WE                                 = 0x136,
+	SC_PA2_SC_EOPG_WE                                = 0x137,
+	SC_PA2_SC_EVENT_WE                               = 0x138,
+	SC_PA3_SC_EOP_WE                                 = 0x139,
+	SC_PA3_SC_EOPG_WE                                = 0x13a,
+	SC_PA3_SC_EVENT_WE                               = 0x13b,
+	SC_PS_ARB_OOO_THRESHOLD_SWITCH_TO_DESIRED_FIFO   = 0x13c,
+	SC_PS_ARB_OOO_FIFO_EMPTY_SWITCH                  = 0x13d,
+	SC_PS_ARB_NULL_PRIM_BUBBLE_POP                   = 0x13e,
+	SC_PS_ARB_EOP_POP_SYNC_POP                       = 0x13f,
+	SC_PS_ARB_EVENT_SYNC_POP                         = 0x140,
+	SC_SC_PS_ENG_MULTICYCLE_BUBBLE                   = 0x141,
+	SC_PA0_SC_FPOV_WE                                = 0x142,
+	SC_PA1_SC_FPOV_WE                                = 0x143,
+	SC_PA2_SC_FPOV_WE                                = 0x144,
+	SC_PA3_SC_FPOV_WE                                = 0x145,
+	SC_PA0_SC_LPOV_WE                                = 0x146,
+	SC_PA1_SC_LPOV_WE                                = 0x147,
+	SC_PA2_SC_LPOV_WE                                = 0x148,
+	SC_PA3_SC_LPOV_WE                                = 0x149,
+	SC_SC_SPI_DEALLOC_0_0                            = 0x14a,
+	SC_SC_SPI_DEALLOC_0_1                            = 0x14b,
+	SC_SC_SPI_DEALLOC_0_2                            = 0x14c,
+	SC_SC_SPI_DEALLOC_1_0                            = 0x14d,
+	SC_SC_SPI_DEALLOC_1_1                            = 0x14e,
+	SC_SC_SPI_DEALLOC_1_2                            = 0x14f,
+	SC_SC_SPI_DEALLOC_2_0                            = 0x150,
+	SC_SC_SPI_DEALLOC_2_1                            = 0x151,
+	SC_SC_SPI_DEALLOC_2_2                            = 0x152,
+	SC_SC_SPI_DEALLOC_3_0                            = 0x153,
+	SC_SC_SPI_DEALLOC_3_1                            = 0x154,
+	SC_SC_SPI_DEALLOC_3_2                            = 0x155,
+	SC_SC_SPI_FPOV_0                                 = 0x156,
+	SC_SC_SPI_FPOV_1                                 = 0x157,
+	SC_SC_SPI_FPOV_2                                 = 0x158,
+	SC_SC_SPI_FPOV_3                                 = 0x159,
+	SC_SC_SPI_EVENT                                  = 0x15a,
+	SC_PS_TS_EVENT_FIFO_PUSH                         = 0x15b,
+	SC_PS_TS_EVENT_FIFO_POP                          = 0x15c,
+	SC_PS_CTX_DONE_FIFO_PUSH                         = 0x15d,
+	SC_PS_CTX_DONE_FIFO_POP                          = 0x15e,
+	SC_MULTICYCLE_BUBBLE_FREEZE                      = 0x15f,
+	SC_EOP_SYNC_WINDOW                               = 0x160,
+	SC_PA0_SC_NULL_WE                                = 0x161,
+	SC_PA0_SC_NULL_DEALLOC_WE                        = 0x162,
+	SC_PA0_SC_DATA_FIFO_EOPG_RD                      = 0x163,
+	SC_PA0_SC_DATA_FIFO_EOP_RD                       = 0x164,
+	SC_PA0_SC_DEALLOC_0_RD                           = 0x165,
+	SC_PA0_SC_DEALLOC_1_RD                           = 0x166,
+	SC_PA1_SC_DATA_FIFO_EOPG_RD                      = 0x167,
+	SC_PA1_SC_DATA_FIFO_EOP_RD                       = 0x168,
+	SC_PA1_SC_DEALLOC_0_RD                           = 0x169,
+	SC_PA1_SC_DEALLOC_1_RD                           = 0x16a,
+	SC_PA1_SC_NULL_WE                                = 0x16b,
+	SC_PA1_SC_NULL_DEALLOC_WE                        = 0x16c,
+	SC_PA2_SC_DATA_FIFO_EOPG_RD                      = 0x16d,
+	SC_PA2_SC_DATA_FIFO_EOP_RD                       = 0x16e,
+	SC_PA2_SC_DEALLOC_0_RD                           = 0x16f,
+	SC_PA2_SC_DEALLOC_1_RD                           = 0x170,
+	SC_PA2_SC_NULL_WE                                = 0x171,
+	SC_PA2_SC_NULL_DEALLOC_WE                        = 0x172,
+	SC_PA3_SC_DATA_FIFO_EOPG_RD                      = 0x173,
+	SC_PA3_SC_DATA_FIFO_EOP_RD                       = 0x174,
+	SC_PA3_SC_DEALLOC_0_RD                           = 0x175,
+	SC_PA3_SC_DEALLOC_1_RD                           = 0x176,
+	SC_PA3_SC_NULL_WE                                = 0x177,
+	SC_PA3_SC_NULL_DEALLOC_WE                        = 0x178,
+	SC_PS_PA0_SC_FIFO_EMPTY                          = 0x179,
+	SC_PS_PA0_SC_FIFO_FULL                           = 0x17a,
+	SC_PA0_PS_DATA_SEND                              = 0x17b,
+	SC_PS_PA1_SC_FIFO_EMPTY                          = 0x17c,
+	SC_PS_PA1_SC_FIFO_FULL                           = 0x17d,
+	SC_PA1_PS_DATA_SEND                              = 0x17e,
+	SC_PS_PA2_SC_FIFO_EMPTY                          = 0x17f,
+	SC_PS_PA2_SC_FIFO_FULL                           = 0x180,
+	SC_PA2_PS_DATA_SEND                              = 0x181,
+	SC_PS_PA3_SC_FIFO_EMPTY                          = 0x182,
+	SC_PS_PA3_SC_FIFO_FULL                           = 0x183,
+	SC_PA3_PS_DATA_SEND                              = 0x184,
+	SC_BUSY_PROCESSING_MULTICYCLE_PRIM               = 0x185,
+	SC_BUSY_CNT_NOT_ZERO                             = 0x186,
+	SC_BM_BUSY                                       = 0x187,
+	SC_BACKEND_BUSY                                  = 0x188,
+	SC_SCF_SCB_INTERFACE_BUSY                        = 0x189,
+	SC_SCB_BUSY                                      = 0x18a,
+	SC_STARVED_BY_PA_WITH_UNSELECTED_PA_NOT_EMPTY    = 0x18b,
+	SC_STARVED_BY_PA_WITH_UNSELECTED_PA_FULL         = 0x18c,
+} SC_PERFCNT_SEL;
+typedef enum SePairXsel {
+	RASTER_CONFIG_SE_PAIR_XSEL_8_WIDE_TILE           = 0x0,
+	RASTER_CONFIG_SE_PAIR_XSEL_16_WIDE_TILE          = 0x1,
+	RASTER_CONFIG_SE_PAIR_XSEL_32_WIDE_TILE          = 0x2,
+	RASTER_CONFIG_SE_PAIR_XSEL_64_WIDE_TILE          = 0x3,
+} SePairXsel;
+typedef enum SePairYsel {
+	RASTER_CONFIG_SE_PAIR_YSEL_8_WIDE_TILE           = 0x0,
+	RASTER_CONFIG_SE_PAIR_YSEL_16_WIDE_TILE          = 0x1,
+	RASTER_CONFIG_SE_PAIR_YSEL_32_WIDE_TILE          = 0x2,
+	RASTER_CONFIG_SE_PAIR_YSEL_64_WIDE_TILE          = 0x3,
+} SePairYsel;
+typedef enum SePairMap {
+	RASTER_CONFIG_SE_PAIR_MAP_0                      = 0x0,
+	RASTER_CONFIG_SE_PAIR_MAP_1                      = 0x1,
+	RASTER_CONFIG_SE_PAIR_MAP_2                      = 0x2,
+	RASTER_CONFIG_SE_PAIR_MAP_3                      = 0x3,
+} SePairMap;
+typedef enum SeXsel {
+	RASTER_CONFIG_SE_XSEL_8_WIDE_TILE                = 0x0,
+	RASTER_CONFIG_SE_XSEL_16_WIDE_TILE               = 0x1,
+	RASTER_CONFIG_SE_XSEL_32_WIDE_TILE               = 0x2,
+	RASTER_CONFIG_SE_XSEL_64_WIDE_TILE               = 0x3,
+} SeXsel;
+typedef enum SeYsel {
+	RASTER_CONFIG_SE_YSEL_8_WIDE_TILE                = 0x0,
+	RASTER_CONFIG_SE_YSEL_16_WIDE_TILE               = 0x1,
+	RASTER_CONFIG_SE_YSEL_32_WIDE_TILE               = 0x2,
+	RASTER_CONFIG_SE_YSEL_64_WIDE_TILE               = 0x3,
+} SeYsel;
+typedef enum SeMap {
+	RASTER_CONFIG_SE_MAP_0                           = 0x0,
+	RASTER_CONFIG_SE_MAP_1                           = 0x1,
+	RASTER_CONFIG_SE_MAP_2                           = 0x2,
+	RASTER_CONFIG_SE_MAP_3                           = 0x3,
+} SeMap;
+typedef enum ScXsel {
+	RASTER_CONFIG_SC_XSEL_8_WIDE_TILE                = 0x0,
+	RASTER_CONFIG_SC_XSEL_16_WIDE_TILE               = 0x1,
+	RASTER_CONFIG_SC_XSEL_32_WIDE_TILE               = 0x2,
+	RASTER_CONFIG_SC_XSEL_64_WIDE_TILE               = 0x3,
+} ScXsel;
+typedef enum ScYsel {
+	RASTER_CONFIG_SC_YSEL_8_WIDE_TILE                = 0x0,
+	RASTER_CONFIG_SC_YSEL_16_WIDE_TILE               = 0x1,
+	RASTER_CONFIG_SC_YSEL_32_WIDE_TILE               = 0x2,
+	RASTER_CONFIG_SC_YSEL_64_WIDE_TILE               = 0x3,
+} ScYsel;
+typedef enum ScMap {
+	RASTER_CONFIG_SC_MAP_0                           = 0x0,
+	RASTER_CONFIG_SC_MAP_1                           = 0x1,
+	RASTER_CONFIG_SC_MAP_2                           = 0x2,
+	RASTER_CONFIG_SC_MAP_3                           = 0x3,
+} ScMap;
+typedef enum PkrXsel2 {
+	RASTER_CONFIG_PKR_XSEL2_0                        = 0x0,
+	RASTER_CONFIG_PKR_XSEL2_1                        = 0x1,
+	RASTER_CONFIG_PKR_XSEL2_2                        = 0x2,
+	RASTER_CONFIG_PKR_XSEL2_3                        = 0x3,
+} PkrXsel2;
+typedef enum PkrXsel {
+	RASTER_CONFIG_PKR_XSEL_0                         = 0x0,
+	RASTER_CONFIG_PKR_XSEL_1                         = 0x1,
+	RASTER_CONFIG_PKR_XSEL_2                         = 0x2,
+	RASTER_CONFIG_PKR_XSEL_3                         = 0x3,
+} PkrXsel;
+typedef enum PkrYsel {
+	RASTER_CONFIG_PKR_YSEL_0                         = 0x0,
+	RASTER_CONFIG_PKR_YSEL_1                         = 0x1,
+	RASTER_CONFIG_PKR_YSEL_2                         = 0x2,
+	RASTER_CONFIG_PKR_YSEL_3                         = 0x3,
+} PkrYsel;
+typedef enum PkrMap {
+	RASTER_CONFIG_PKR_MAP_0                          = 0x0,
+	RASTER_CONFIG_PKR_MAP_1                          = 0x1,
+	RASTER_CONFIG_PKR_MAP_2                          = 0x2,
+	RASTER_CONFIG_PKR_MAP_3                          = 0x3,
+} PkrMap;
+typedef enum RbXsel {
+	RASTER_CONFIG_RB_XSEL_0                          = 0x0,
+	RASTER_CONFIG_RB_XSEL_1                          = 0x1,
+} RbXsel;
+typedef enum RbYsel {
+	RASTER_CONFIG_RB_YSEL_0                          = 0x0,
+	RASTER_CONFIG_RB_YSEL_1                          = 0x1,
+} RbYsel;
+typedef enum RbXsel2 {
+	RASTER_CONFIG_RB_XSEL2_0                         = 0x0,
+	RASTER_CONFIG_RB_XSEL2_1                         = 0x1,
+	RASTER_CONFIG_RB_XSEL2_2                         = 0x2,
+	RASTER_CONFIG_RB_XSEL2_3                         = 0x3,
+} RbXsel2;
+typedef enum RbMap {
+	RASTER_CONFIG_RB_MAP_0                           = 0x0,
+	RASTER_CONFIG_RB_MAP_1                           = 0x1,
+	RASTER_CONFIG_RB_MAP_2                           = 0x2,
+	RASTER_CONFIG_RB_MAP_3                           = 0x3,
+} RbMap;
+typedef enum CSDATA_TYPE {
+	CSDATA_TYPE_TG                                   = 0x0,
+	CSDATA_TYPE_STATE                                = 0x1,
+	CSDATA_TYPE_EVENT                                = 0x2,
+	CSDATA_TYPE_PRIVATE                              = 0x3,
+} CSDATA_TYPE;
+#define CSDATA_TYPE_WIDTH                         0x2
+#define CSDATA_ADDR_WIDTH                         0x7
+#define CSDATA_DATA_WIDTH                         0x20
+typedef enum SPI_SAMPLE_CNTL {
+	CENTROIDS_ONLY                                   = 0x0,
+	CENTERS_ONLY                                     = 0x1,
+	CENTROIDS_AND_CENTERS                            = 0x2,
+	UNDEF                                            = 0x3,
+} SPI_SAMPLE_CNTL;
+typedef enum SPI_FOG_MODE {
+	SPI_FOG_NONE                                     = 0x0,
+	SPI_FOG_EXP                                      = 0x1,
+	SPI_FOG_EXP2                                     = 0x2,
+	SPI_FOG_LINEAR                                   = 0x3,
+} SPI_FOG_MODE;
+typedef enum SPI_PNT_SPRITE_OVERRIDE {
+	SPI_PNT_SPRITE_SEL_0                             = 0x0,
+	SPI_PNT_SPRITE_SEL_1                             = 0x1,
+	SPI_PNT_SPRITE_SEL_S                             = 0x2,
+	SPI_PNT_SPRITE_SEL_T                             = 0x3,
+	SPI_PNT_SPRITE_SEL_NONE                          = 0x4,
+} SPI_PNT_SPRITE_OVERRIDE;
+typedef enum SPI_PERFCNT_SEL {
+	SPI_PERF_VS_WINDOW_VALID                         = 0x0,
+	SPI_PERF_VS_BUSY                                 = 0x1,
+	SPI_PERF_VS_FIRST_WAVE                           = 0x2,
+	SPI_PERF_VS_LAST_WAVE                            = 0x3,
+	SPI_PERF_VS_LSHS_DEALLOC                         = 0x4,
+	SPI_PERF_VS_PC_STALL                             = 0x5,
+	SPI_PERF_VS_POS0_STALL                           = 0x6,
+	SPI_PERF_VS_POS1_STALL                           = 0x7,
+	SPI_PERF_VS_CRAWLER_STALL                        = 0x8,
+	SPI_PERF_VS_EVENT_WAVE                           = 0x9,
+	SPI_PERF_VS_WAVE                                 = 0xa,
+	SPI_PERF_VS_PERS_UPD_FULL0                       = 0xb,
+	SPI_PERF_VS_PERS_UPD_FULL1                       = 0xc,
+	SPI_PERF_VS_LATE_ALLOC_FULL                      = 0xd,
+	SPI_PERF_VS_FIRST_SUBGRP                         = 0xe,
+	SPI_PERF_VS_LAST_SUBGRP                          = 0xf,
+	SPI_PERF_GS_WINDOW_VALID                         = 0x10,
+	SPI_PERF_GS_BUSY                                 = 0x11,
+	SPI_PERF_GS_CRAWLER_STALL                        = 0x12,
+	SPI_PERF_GS_EVENT_WAVE                           = 0x13,
+	SPI_PERF_GS_WAVE                                 = 0x14,
+	SPI_PERF_GS_PERS_UPD_FULL0                       = 0x15,
+	SPI_PERF_GS_PERS_UPD_FULL1                       = 0x16,
+	SPI_PERF_GS_FIRST_SUBGRP                         = 0x17,
+	SPI_PERF_GS_LAST_SUBGRP                          = 0x18,
+	SPI_PERF_ES_WINDOW_VALID                         = 0x19,
+	SPI_PERF_ES_BUSY                                 = 0x1a,
+	SPI_PERF_ES_CRAWLER_STALL                        = 0x1b,
+	SPI_PERF_ES_FIRST_WAVE                           = 0x1c,
+	SPI_PERF_ES_LAST_WAVE                            = 0x1d,
+	SPI_PERF_ES_LSHS_DEALLOC                         = 0x1e,
+	SPI_PERF_ES_EVENT_WAVE                           = 0x1f,
+	SPI_PERF_ES_WAVE                                 = 0x20,
+	SPI_PERF_ES_PERS_UPD_FULL0                       = 0x21,
+	SPI_PERF_ES_PERS_UPD_FULL1                       = 0x22,
+	SPI_PERF_ES_FIRST_SUBGRP                         = 0x23,
+	SPI_PERF_ES_LAST_SUBGRP                          = 0x24,
+	SPI_PERF_HS_WINDOW_VALID                         = 0x25,
+	SPI_PERF_HS_BUSY                                 = 0x26,
+	SPI_PERF_HS_CRAWLER_STALL                        = 0x27,
+	SPI_PERF_HS_FIRST_WAVE                           = 0x28,
+	SPI_PERF_HS_LAST_WAVE                            = 0x29,
+	SPI_PERF_HS_LSHS_DEALLOC                         = 0x2a,
+	SPI_PERF_HS_EVENT_WAVE                           = 0x2b,
+	SPI_PERF_HS_WAVE                                 = 0x2c,
+	SPI_PERF_HS_PERS_UPD_FULL0                       = 0x2d,
+	SPI_PERF_HS_PERS_UPD_FULL1                       = 0x2e,
+	SPI_PERF_LS_WINDOW_VALID                         = 0x2f,
+	SPI_PERF_LS_BUSY                                 = 0x30,
+	SPI_PERF_LS_CRAWLER_STALL                        = 0x31,
+	SPI_PERF_LS_FIRST_WAVE                           = 0x32,
+	SPI_PERF_LS_LAST_WAVE                            = 0x33,
+	SPI_PERF_OFFCHIP_LDS_STALL_LS                    = 0x34,
+	SPI_PERF_LS_EVENT_WAVE                           = 0x35,
+	SPI_PERF_LS_WAVE                                 = 0x36,
+	SPI_PERF_LS_PERS_UPD_FULL0                       = 0x37,
+	SPI_PERF_LS_PERS_UPD_FULL1                       = 0x38,
+	SPI_PERF_CSG_WINDOW_VALID                        = 0x39,
+	SPI_PERF_CSG_BUSY                                = 0x3a,
+	SPI_PERF_CSG_NUM_THREADGROUPS                    = 0x3b,
+	SPI_PERF_CSG_CRAWLER_STALL                       = 0x3c,
+	SPI_PERF_CSG_EVENT_WAVE                          = 0x3d,
+	SPI_PERF_CSG_WAVE                                = 0x3e,
+	SPI_PERF_CSN_WINDOW_VALID                        = 0x3f,
+	SPI_PERF_CSN_BUSY                                = 0x40,
+	SPI_PERF_CSN_NUM_THREADGROUPS                    = 0x41,
+	SPI_PERF_CSN_CRAWLER_STALL                       = 0x42,
+	SPI_PERF_CSN_EVENT_WAVE                          = 0x43,
+	SPI_PERF_CSN_WAVE                                = 0x44,
+	SPI_PERF_PS_CTL_WINDOW_VALID                     = 0x45,
+	SPI_PERF_PS_CTL_BUSY                             = 0x46,
+	SPI_PERF_PS_CTL_ACTIVE                           = 0x47,
+	SPI_PERF_PS_CTL_DEALLOC_BIN0                     = 0x48,
+	SPI_PERF_PS_CTL_FPOS_BIN1_STALL                  = 0x49,
+	SPI_PERF_PS_CTL_EVENT_WAVE                       = 0x4a,
+	SPI_PERF_PS_CTL_WAVE                             = 0x4b,
+	SPI_PERF_PS_CTL_OPT_WAVE                         = 0x4c,
+	SPI_PERF_PS_CTL_PASS_BIN0                        = 0x4d,
+	SPI_PERF_PS_CTL_PASS_BIN1                        = 0x4e,
+	SPI_PERF_PS_CTL_FPOS_BIN2                        = 0x4f,
+	SPI_PERF_PS_CTL_PRIM_BIN0                        = 0x50,
+	SPI_PERF_PS_CTL_PRIM_BIN1                        = 0x51,
+	SPI_PERF_PS_CTL_CNF_BIN2                         = 0x52,
+	SPI_PERF_PS_CTL_CNF_BIN3                         = 0x53,
+	SPI_PERF_PS_CTL_CRAWLER_STALL                    = 0x54,
+	SPI_PERF_PS_CTL_LDS_RES_FULL                     = 0x55,
+	SPI_PERF_PS_PERS_UPD_FULL0                       = 0x56,
+	SPI_PERF_PS_PERS_UPD_FULL1                       = 0x57,
+	SPI_PERF_PIX_ALLOC_PEND_CNT                      = 0x58,
+	SPI_PERF_PIX_ALLOC_SCB_STALL                     = 0x59,
+	SPI_PERF_PIX_ALLOC_DB0_STALL                     = 0x5a,
+	SPI_PERF_PIX_ALLOC_DB1_STALL                     = 0x5b,
+	SPI_PERF_PIX_ALLOC_DB2_STALL                     = 0x5c,
+	SPI_PERF_PIX_ALLOC_DB3_STALL                     = 0x5d,
+	SPI_PERF_LDS0_PC_VALID                           = 0x5e,
+	SPI_PERF_LDS1_PC_VALID                           = 0x5f,
+	SPI_PERF_RA_PIPE_REQ_BIN2                        = 0x60,
+	SPI_PERF_RA_TASK_REQ_BIN3                        = 0x61,
+	SPI_PERF_RA_WR_CTL_FULL                          = 0x62,
+	SPI_PERF_RA_REQ_NO_ALLOC                         = 0x63,
+	SPI_PERF_RA_REQ_NO_ALLOC_PS                      = 0x64,
+	SPI_PERF_RA_REQ_NO_ALLOC_VS                      = 0x65,
+	SPI_PERF_RA_REQ_NO_ALLOC_GS                      = 0x66,
+	SPI_PERF_RA_REQ_NO_ALLOC_ES                      = 0x67,
+	SPI_PERF_RA_REQ_NO_ALLOC_HS                      = 0x68,
+	SPI_PERF_RA_REQ_NO_ALLOC_LS                      = 0x69,
+	SPI_PERF_RA_REQ_NO_ALLOC_CSG                     = 0x6a,
+	SPI_PERF_RA_REQ_NO_ALLOC_CSN                     = 0x6b,
+	SPI_PERF_RA_RES_STALL_PS                         = 0x6c,
+	SPI_PERF_RA_RES_STALL_VS                         = 0x6d,
+	SPI_PERF_RA_RES_STALL_GS                         = 0x6e,
+	SPI_PERF_RA_RES_STALL_ES                         = 0x6f,
+	SPI_PERF_RA_RES_STALL_HS                         = 0x70,
+	SPI_PERF_RA_RES_STALL_LS                         = 0x71,
+	SPI_PERF_RA_RES_STALL_CSG                        = 0x72,
+	SPI_PERF_RA_RES_STALL_CSN                        = 0x73,
+	SPI_PERF_RA_TMP_STALL_PS                         = 0x74,
+	SPI_PERF_RA_TMP_STALL_VS                         = 0x75,
+	SPI_PERF_RA_TMP_STALL_GS                         = 0x76,
+	SPI_PERF_RA_TMP_STALL_ES                         = 0x77,
+	SPI_PERF_RA_TMP_STALL_HS                         = 0x78,
+	SPI_PERF_RA_TMP_STALL_LS                         = 0x79,
+	SPI_PERF_RA_TMP_STALL_CSG                        = 0x7a,
+	SPI_PERF_RA_TMP_STALL_CSN                        = 0x7b,
+	SPI_PERF_RA_WAVE_SIMD_FULL_PS                    = 0x7c,
+	SPI_PERF_RA_WAVE_SIMD_FULL_VS                    = 0x7d,
+	SPI_PERF_RA_WAVE_SIMD_FULL_GS                    = 0x7e,
+	SPI_PERF_RA_WAVE_SIMD_FULL_ES                    = 0x7f,
+	SPI_PERF_RA_WAVE_SIMD_FULL_HS                    = 0x80,
+	SPI_PERF_RA_WAVE_SIMD_FULL_LS                    = 0x81,
+	SPI_PERF_RA_WAVE_SIMD_FULL_CSG                   = 0x82,
+	SPI_PERF_RA_WAVE_SIMD_FULL_CSN                   = 0x83,
+	SPI_PERF_RA_VGPR_SIMD_FULL_PS                    = 0x84,
+	SPI_PERF_RA_VGPR_SIMD_FULL_VS                    = 0x85,
+	SPI_PERF_RA_VGPR_SIMD_FULL_GS                    = 0x86,
+	SPI_PERF_RA_VGPR_SIMD_FULL_ES                    = 0x87,
+	SPI_PERF_RA_VGPR_SIMD_FULL_HS                    = 0x88,
+	SPI_PERF_RA_VGPR_SIMD_FULL_LS                    = 0x89,
+	SPI_PERF_RA_VGPR_SIMD_FULL_CSG                   = 0x8a,
+	SPI_PERF_RA_VGPR_SIMD_FULL_CSN                   = 0x8b,
+	SPI_PERF_RA_SGPR_SIMD_FULL_PS                    = 0x8c,
+	SPI_PERF_RA_SGPR_SIMD_FULL_VS                    = 0x8d,
+	SPI_PERF_RA_SGPR_SIMD_FULL_GS                    = 0x8e,
+	SPI_PERF_RA_SGPR_SIMD_FULL_ES                    = 0x8f,
+	SPI_PERF_RA_SGPR_SIMD_FULL_HS                    = 0x90,
+	SPI_PERF_RA_SGPR_SIMD_FULL_LS                    = 0x91,
+	SPI_PERF_RA_SGPR_SIMD_FULL_CSG                   = 0x92,
+	SPI_PERF_RA_SGPR_SIMD_FULL_CSN                   = 0x93,
+	SPI_PERF_RA_LDS_CU_FULL_PS                       = 0x94,
+	SPI_PERF_RA_LDS_CU_FULL_LS                       = 0x95,
+	SPI_PERF_RA_LDS_CU_FULL_ES                       = 0x96,
+	SPI_PERF_RA_LDS_CU_FULL_CSG                      = 0x97,
+	SPI_PERF_RA_LDS_CU_FULL_CSN                      = 0x98,
+	SPI_PERF_RA_BAR_CU_FULL_HS                       = 0x99,
+	SPI_PERF_RA_BAR_CU_FULL_CSG                      = 0x9a,
+	SPI_PERF_RA_BAR_CU_FULL_CSN                      = 0x9b,
+	SPI_PERF_RA_BULKY_CU_FULL_CSG                    = 0x9c,
+	SPI_PERF_RA_BULKY_CU_FULL_CSN                    = 0x9d,
+	SPI_PERF_RA_TGLIM_CU_FULL_CSG                    = 0x9e,
+	SPI_PERF_RA_TGLIM_CU_FULL_CSN                    = 0x9f,
+	SPI_PERF_RA_WVLIM_STALL_PS                       = 0xa0,
+	SPI_PERF_RA_WVLIM_STALL_VS                       = 0xa1,
+	SPI_PERF_RA_WVLIM_STALL_GS                       = 0xa2,
+	SPI_PERF_RA_WVLIM_STALL_ES                       = 0xa3,
+	SPI_PERF_RA_WVLIM_STALL_HS                       = 0xa4,
+	SPI_PERF_RA_WVLIM_STALL_LS                       = 0xa5,
+	SPI_PERF_RA_WVLIM_STALL_CSG                      = 0xa6,
+	SPI_PERF_RA_WVLIM_STALL_CSN                      = 0xa7,
+	SPI_PERF_RA_PS_LOCK_NA                           = 0xa8,
+	SPI_PERF_RA_VS_LOCK                              = 0xa9,
+	SPI_PERF_RA_GS_LOCK                              = 0xaa,
+	SPI_PERF_RA_ES_LOCK                              = 0xab,
+	SPI_PERF_RA_HS_LOCK                              = 0xac,
+	SPI_PERF_RA_LS_LOCK                              = 0xad,
+	SPI_PERF_RA_CSG_LOCK                             = 0xae,
+	SPI_PERF_RA_CSN_LOCK                             = 0xaf,
+	SPI_PERF_RA_RSV_UPD                              = 0xb0,
+	SPI_PERF_EXP_ARB_COL_CNT                         = 0xb1,
+	SPI_PERF_EXP_ARB_PAR_CNT                         = 0xb2,
+	SPI_PERF_EXP_ARB_POS_CNT                         = 0xb3,
+	SPI_PERF_EXP_ARB_GDS_CNT                         = 0xb4,
+	SPI_PERF_CLKGATE_BUSY_STALL                      = 0xb5,
+	SPI_PERF_CLKGATE_ACTIVE_STALL                    = 0xb6,
+	SPI_PERF_CLKGATE_ALL_CLOCKS_ON                   = 0xb7,
+	SPI_PERF_CLKGATE_CGTT_DYN_ON                     = 0xb8,
+	SPI_PERF_CLKGATE_CGTT_REG_ON                     = 0xb9,
+	SPI_PERF_NUM_VS_POS_EXPORTS                      = 0xba,
+	SPI_PERF_NUM_VS_PARAM_EXPORTS                    = 0xbb,
+	SPI_PERF_NUM_PS_COL_EXPORTS                      = 0xbc,
+	SPI_PERF_ES_GRP_FIFO_FULL                        = 0xbd,
+	SPI_PERF_GS_GRP_FIFO_FULL                        = 0xbe,
+	SPI_PERF_HS_GRP_FIFO_FULL                        = 0xbf,
+	SPI_PERF_LS_GRP_FIFO_FULL                        = 0xc0,
+	SPI_PERF_VS_ALLOC_CNT                            = 0xc1,
+	SPI_PERF_VS_LATE_ALLOC_ACCUM                     = 0xc2,
+	SPI_PERF_PC_ALLOC_CNT                            = 0xc3,
+	SPI_PERF_PC_ALLOC_ACCUM                          = 0xc4,
+} SPI_PERFCNT_SEL;
+typedef enum SPI_SHADER_FORMAT {
+	SPI_SHADER_NONE                                  = 0x0,
+	SPI_SHADER_1COMP                                 = 0x1,
+	SPI_SHADER_2COMP                                 = 0x2,
+	SPI_SHADER_4COMPRESS                             = 0x3,
+	SPI_SHADER_4COMP                                 = 0x4,
+} SPI_SHADER_FORMAT;
+typedef enum SPI_SHADER_EX_FORMAT {
+	SPI_SHADER_ZERO                                  = 0x0,
+	SPI_SHADER_32_R                                  = 0x1,
+	SPI_SHADER_32_GR                                 = 0x2,
+	SPI_SHADER_32_AR                                 = 0x3,
+	SPI_SHADER_FP16_ABGR                             = 0x4,
+	SPI_SHADER_UNORM16_ABGR                          = 0x5,
+	SPI_SHADER_SNORM16_ABGR                          = 0x6,
+	SPI_SHADER_UINT16_ABGR                           = 0x7,
+	SPI_SHADER_SINT16_ABGR                           = 0x8,
+	SPI_SHADER_32_ABGR                               = 0x9,
+} SPI_SHADER_EX_FORMAT;
+typedef enum CLKGATE_SM_MODE {
+	ON_SEQ                                           = 0x0,
+	OFF_SEQ                                          = 0x1,
+	PROG_SEQ                                         = 0x2,
+	READ_SEQ                                         = 0x3,
+	SM_MODE_RESERVED                                 = 0x4,
+} CLKGATE_SM_MODE;
+typedef enum CLKGATE_BASE_MODE {
+	MULT_8                                           = 0x0,
+	MULT_16                                          = 0x1,
+} CLKGATE_BASE_MODE;
+typedef enum SQ_TEX_CLAMP {
+	SQ_TEX_WRAP                                      = 0x0,
+	SQ_TEX_MIRROR                                    = 0x1,
+	SQ_TEX_CLAMP_LAST_TEXEL                          = 0x2,
+	SQ_TEX_MIRROR_ONCE_LAST_TEXEL                    = 0x3,
+	SQ_TEX_CLAMP_HALF_BORDER                         = 0x4,
+	SQ_TEX_MIRROR_ONCE_HALF_BORDER                   = 0x5,
+	SQ_TEX_CLAMP_BORDER                              = 0x6,
+	SQ_TEX_MIRROR_ONCE_BORDER                        = 0x7,
+} SQ_TEX_CLAMP;
+typedef enum SQ_TEX_XY_FILTER {
+	SQ_TEX_XY_FILTER_POINT                           = 0x0,
+	SQ_TEX_XY_FILTER_BILINEAR                        = 0x1,
+	SQ_TEX_XY_FILTER_ANISO_POINT                     = 0x2,
+	SQ_TEX_XY_FILTER_ANISO_BILINEAR                  = 0x3,
+} SQ_TEX_XY_FILTER;
+typedef enum SQ_TEX_Z_FILTER {
+	SQ_TEX_Z_FILTER_NONE                             = 0x0,
+	SQ_TEX_Z_FILTER_POINT                            = 0x1,
+	SQ_TEX_Z_FILTER_LINEAR                           = 0x2,
+} SQ_TEX_Z_FILTER;
+typedef enum SQ_TEX_MIP_FILTER {
+	SQ_TEX_MIP_FILTER_NONE                           = 0x0,
+	SQ_TEX_MIP_FILTER_POINT                          = 0x1,
+	SQ_TEX_MIP_FILTER_LINEAR                         = 0x2,
+	SQ_TEX_MIP_FILTER_POINT_ANISO_ADJ                = 0x3,
+} SQ_TEX_MIP_FILTER;
+typedef enum SQ_TEX_ANISO_RATIO {
+	SQ_TEX_ANISO_RATIO_1                             = 0x0,
+	SQ_TEX_ANISO_RATIO_2                             = 0x1,
+	SQ_TEX_ANISO_RATIO_4                             = 0x2,
+	SQ_TEX_ANISO_RATIO_8                             = 0x3,
+	SQ_TEX_ANISO_RATIO_16                            = 0x4,
+} SQ_TEX_ANISO_RATIO;
+typedef enum SQ_TEX_DEPTH_COMPARE {
+	SQ_TEX_DEPTH_COMPARE_NEVER                       = 0x0,
+	SQ_TEX_DEPTH_COMPARE_LESS                        = 0x1,
+	SQ_TEX_DEPTH_COMPARE_EQUAL                       = 0x2,
+	SQ_TEX_DEPTH_COMPARE_LESSEQUAL                   = 0x3,
+	SQ_TEX_DEPTH_COMPARE_GREATER                     = 0x4,
+	SQ_TEX_DEPTH_COMPARE_NOTEQUAL                    = 0x5,
+	SQ_TEX_DEPTH_COMPARE_GREATEREQUAL                = 0x6,
+	SQ_TEX_DEPTH_COMPARE_ALWAYS                      = 0x7,
+} SQ_TEX_DEPTH_COMPARE;
+typedef enum SQ_TEX_BORDER_COLOR {
+	SQ_TEX_BORDER_COLOR_TRANS_BLACK                  = 0x0,
+	SQ_TEX_BORDER_COLOR_OPAQUE_BLACK                 = 0x1,
+	SQ_TEX_BORDER_COLOR_OPAQUE_WHITE                 = 0x2,
+	SQ_TEX_BORDER_COLOR_REGISTER                     = 0x3,
+} SQ_TEX_BORDER_COLOR;
+typedef enum SQ_RSRC_BUF_TYPE {
+	SQ_RSRC_BUF                                      = 0x0,
+	SQ_RSRC_BUF_RSVD_1                               = 0x1,
+	SQ_RSRC_BUF_RSVD_2                               = 0x2,
+	SQ_RSRC_BUF_RSVD_3                               = 0x3,
+} SQ_RSRC_BUF_TYPE;
+typedef enum SQ_RSRC_IMG_TYPE {
+	SQ_RSRC_IMG_RSVD_0                               = 0x0,
+	SQ_RSRC_IMG_RSVD_1                               = 0x1,
+	SQ_RSRC_IMG_RSVD_2                               = 0x2,
+	SQ_RSRC_IMG_RSVD_3                               = 0x3,
+	SQ_RSRC_IMG_RSVD_4                               = 0x4,
+	SQ_RSRC_IMG_RSVD_5                               = 0x5,
+	SQ_RSRC_IMG_RSVD_6                               = 0x6,
+	SQ_RSRC_IMG_RSVD_7                               = 0x7,
+	SQ_RSRC_IMG_1D                                   = 0x8,
+	SQ_RSRC_IMG_2D                                   = 0x9,
+	SQ_RSRC_IMG_3D                                   = 0xa,
+	SQ_RSRC_IMG_CUBE                                 = 0xb,
+	SQ_RSRC_IMG_1D_ARRAY                             = 0xc,
+	SQ_RSRC_IMG_2D_ARRAY                             = 0xd,
+	SQ_RSRC_IMG_2D_MSAA                              = 0xe,
+	SQ_RSRC_IMG_2D_MSAA_ARRAY                        = 0xf,
+} SQ_RSRC_IMG_TYPE;
+typedef enum SQ_RSRC_FLAT_TYPE {
+	SQ_RSRC_FLAT_RSVD_0                              = 0x0,
+	SQ_RSRC_FLAT                                     = 0x1,
+	SQ_RSRC_FLAT_RSVD_2                              = 0x2,
+	SQ_RSRC_FLAT_RSVD_3                              = 0x3,
+} SQ_RSRC_FLAT_TYPE;
+typedef enum SQ_IMG_FILTER_TYPE {
+	SQ_IMG_FILTER_MODE_BLEND                         = 0x0,
+	SQ_IMG_FILTER_MODE_MIN                           = 0x1,
+	SQ_IMG_FILTER_MODE_MAX                           = 0x2,
+} SQ_IMG_FILTER_TYPE;
+typedef enum SQ_SEL_XYZW01 {
+	SQ_SEL_0                                         = 0x0,
+	SQ_SEL_1                                         = 0x1,
+	SQ_SEL_RESERVED_0                                = 0x2,
+	SQ_SEL_RESERVED_1                                = 0x3,
+	SQ_SEL_X                                         = 0x4,
+	SQ_SEL_Y                                         = 0x5,
+	SQ_SEL_Z                                         = 0x6,
+	SQ_SEL_W                                         = 0x7,
+} SQ_SEL_XYZW01;
+typedef enum SQ_WAVE_TYPE {
+	SQ_WAVE_TYPE_PS                                  = 0x0,
+	SQ_WAVE_TYPE_VS                                  = 0x1,
+	SQ_WAVE_TYPE_GS                                  = 0x2,
+	SQ_WAVE_TYPE_ES                                  = 0x3,
+	SQ_WAVE_TYPE_HS                                  = 0x4,
+	SQ_WAVE_TYPE_LS                                  = 0x5,
+	SQ_WAVE_TYPE_CS                                  = 0x6,
+	SQ_WAVE_TYPE_PS1                                 = 0x7,
+} SQ_WAVE_TYPE;
+typedef enum SQ_THREAD_TRACE_TOKEN_TYPE {
+	SQ_THREAD_TRACE_TOKEN_MISC                       = 0x0,
+	SQ_THREAD_TRACE_TOKEN_TIMESTAMP                  = 0x1,
+	SQ_THREAD_TRACE_TOKEN_REG                        = 0x2,
+	SQ_THREAD_TRACE_TOKEN_WAVE_START                 = 0x3,
+	SQ_THREAD_TRACE_TOKEN_WAVE_ALLOC                 = 0x4,
+	SQ_THREAD_TRACE_TOKEN_REG_CSPRIV                 = 0x5,
+	SQ_THREAD_TRACE_TOKEN_WAVE_END                   = 0x6,
+	SQ_THREAD_TRACE_TOKEN_EVENT                      = 0x7,
+	SQ_THREAD_TRACE_TOKEN_EVENT_CS                   = 0x8,
+	SQ_THREAD_TRACE_TOKEN_EVENT_GFX1                 = 0x9,
+	SQ_THREAD_TRACE_TOKEN_INST                       = 0xa,
+	SQ_THREAD_TRACE_TOKEN_INST_PC                    = 0xb,
+	SQ_THREAD_TRACE_TOKEN_INST_USERDATA              = 0xc,
+	SQ_THREAD_TRACE_TOKEN_ISSUE                      = 0xd,
+	SQ_THREAD_TRACE_TOKEN_PERF                       = 0xe,
+	SQ_THREAD_TRACE_TOKEN_REG_CS                     = 0xf,
+} SQ_THREAD_TRACE_TOKEN_TYPE;
+typedef enum SQ_THREAD_TRACE_MISC_TOKEN_TYPE {
+	SQ_THREAD_TRACE_MISC_TOKEN_TIME                  = 0x0,
+	SQ_THREAD_TRACE_MISC_TOKEN_TIME_RESET            = 0x1,
+	SQ_THREAD_TRACE_MISC_TOKEN_PACKET_LOST           = 0x2,
+	SQ_THREAD_TRACE_MISC_TOKEN_SURF_SYNC             = 0x3,
+	SQ_THREAD_TRACE_MISC_TOKEN_TTRACE_STALL_BEGIN    = 0x4,
+	SQ_THREAD_TRACE_MISC_TOKEN_TTRACE_STALL_END      = 0x5,
+	SQ_THREAD_TRACE_MISC_TOKEN_SAVECTX               = 0x6,
+	SQ_THREAD_TRACE_MISC_TOKEN_SHOOT_DOWN            = 0x7,
+} SQ_THREAD_TRACE_MISC_TOKEN_TYPE;
+typedef enum SQ_THREAD_TRACE_INST_TYPE {
+	SQ_THREAD_TRACE_INST_TYPE_SMEM_RD                = 0x0,
+	SQ_THREAD_TRACE_INST_TYPE_SALU_32                = 0x1,
+	SQ_THREAD_TRACE_INST_TYPE_VMEM_RD                = 0x2,
+	SQ_THREAD_TRACE_INST_TYPE_VMEM_WR                = 0x3,
+	SQ_THREAD_TRACE_INST_TYPE_FLAT_WR                = 0x4,
+	SQ_THREAD_TRACE_INST_TYPE_VALU_32                = 0x5,
+	SQ_THREAD_TRACE_INST_TYPE_LDS                    = 0x6,
+	SQ_THREAD_TRACE_INST_TYPE_PC                     = 0x7,
+	SQ_THREAD_TRACE_INST_TYPE_EXPREQ_GDS             = 0x8,
+	SQ_THREAD_TRACE_INST_TYPE_EXPREQ_GFX             = 0x9,
+	SQ_THREAD_TRACE_INST_TYPE_EXPGNT_PAR_COL         = 0xa,
+	SQ_THREAD_TRACE_INST_TYPE_EXPGNT_POS_GDS         = 0xb,
+	SQ_THREAD_TRACE_INST_TYPE_JUMP                   = 0xc,
+	SQ_THREAD_TRACE_INST_TYPE_NEXT                   = 0xd,
+	SQ_THREAD_TRACE_INST_TYPE_FLAT_RD                = 0xe,
+	SQ_THREAD_TRACE_INST_TYPE_OTHER_MSG              = 0xf,
+	SQ_THREAD_TRACE_INST_TYPE_SMEM_WR                = 0x10,
+	SQ_THREAD_TRACE_INST_TYPE_SALU_64                = 0x11,
+	SQ_THREAD_TRACE_INST_TYPE_VALU_64                = 0x12,
+	SQ_THREAD_TRACE_INST_TYPE_SMEM_RD_REPLAY         = 0x13,
+	SQ_THREAD_TRACE_INST_TYPE_SMEM_WR_REPLAY         = 0x14,
+	SQ_THREAD_TRACE_INST_TYPE_VMEM_RD_REPLAY         = 0x15,
+	SQ_THREAD_TRACE_INST_TYPE_VMEM_WR_REPLAY         = 0x16,
+	SQ_THREAD_TRACE_INST_TYPE_FLAT_RD_REPLAY         = 0x17,
+	SQ_THREAD_TRACE_INST_TYPE_FLAT_WR_REPLAY         = 0x18,
+} SQ_THREAD_TRACE_INST_TYPE;
+typedef enum SQ_THREAD_TRACE_REG_TYPE {
+	SQ_THREAD_TRACE_REG_TYPE_EVENT                   = 0x0,
+	SQ_THREAD_TRACE_REG_TYPE_DRAW                    = 0x1,
+	SQ_THREAD_TRACE_REG_TYPE_DISPATCH                = 0x2,
+	SQ_THREAD_TRACE_REG_TYPE_USERDATA                = 0x3,
+	SQ_THREAD_TRACE_REG_TYPE_MARKER                  = 0x4,
+	SQ_THREAD_TRACE_REG_TYPE_GFXDEC                  = 0x5,
+	SQ_THREAD_TRACE_REG_TYPE_SHDEC                   = 0x6,
+	SQ_THREAD_TRACE_REG_TYPE_OTHER                   = 0x7,
+} SQ_THREAD_TRACE_REG_TYPE;
+typedef enum SQ_THREAD_TRACE_REG_OP {
+	SQ_THREAD_TRACE_REG_OP_READ                      = 0x0,
+	SQ_THREAD_TRACE_REG_OP_WRITE                     = 0x1,
+} SQ_THREAD_TRACE_REG_OP;
+typedef enum SQ_THREAD_TRACE_MODE_SEL {
+	SQ_THREAD_TRACE_MODE_OFF                         = 0x0,
+	SQ_THREAD_TRACE_MODE_ON                          = 0x1,
+} SQ_THREAD_TRACE_MODE_SEL;
+typedef enum SQ_THREAD_TRACE_CAPTURE_MODE {
+	SQ_THREAD_TRACE_CAPTURE_MODE_ALL                 = 0x0,
+	SQ_THREAD_TRACE_CAPTURE_MODE_SELECT              = 0x1,
+	SQ_THREAD_TRACE_CAPTURE_MODE_SELECT_DETAIL       = 0x2,
+} SQ_THREAD_TRACE_CAPTURE_MODE;
+typedef enum SQ_THREAD_TRACE_VM_ID_MASK {
+	SQ_THREAD_TRACE_VM_ID_MASK_SINGLE                = 0x0,
+	SQ_THREAD_TRACE_VM_ID_MASK_ALL                   = 0x1,
+	SQ_THREAD_TRACE_VM_ID_MASK_SINGLE_DETAIL         = 0x2,
+} SQ_THREAD_TRACE_VM_ID_MASK;
+typedef enum SQ_THREAD_TRACE_WAVE_MASK {
+	SQ_THREAD_TRACE_WAVE_MASK_NONE                   = 0x0,
+	SQ_THREAD_TRACE_WAVE_MASK_ALL                    = 0x1,
+} SQ_THREAD_TRACE_WAVE_MASK;
+typedef enum SQ_THREAD_TRACE_ISSUE {
+	SQ_THREAD_TRACE_ISSUE_NULL                       = 0x0,
+	SQ_THREAD_TRACE_ISSUE_STALL                      = 0x1,
+	SQ_THREAD_TRACE_ISSUE_INST                       = 0x2,
+	SQ_THREAD_TRACE_ISSUE_IMMED                      = 0x3,
+} SQ_THREAD_TRACE_ISSUE;
+typedef enum SQ_THREAD_TRACE_ISSUE_MASK {
+	SQ_THREAD_TRACE_ISSUE_MASK_ALL                   = 0x0,
+	SQ_THREAD_TRACE_ISSUE_MASK_STALLED               = 0x1,
+	SQ_THREAD_TRACE_ISSUE_MASK_STALLED_AND_IMMED     = 0x2,
+	SQ_THREAD_TRACE_ISSUE_MASK_IMMED                 = 0x3,
+} SQ_THREAD_TRACE_ISSUE_MASK;
+typedef enum SQ_PERF_SEL {
+	SQ_PERF_SEL_NONE                                 = 0x0,
+	SQ_PERF_SEL_ACCUM_PREV                           = 0x1,
+	SQ_PERF_SEL_CYCLES                               = 0x2,
+	SQ_PERF_SEL_BUSY_CYCLES                          = 0x3,
+	SQ_PERF_SEL_WAVES                                = 0x4,
+	SQ_PERF_SEL_LEVEL_WAVES                          = 0x5,
+	SQ_PERF_SEL_WAVES_EQ_64                          = 0x6,
+	SQ_PERF_SEL_WAVES_LT_64                          = 0x7,
+	SQ_PERF_SEL_WAVES_LT_48                          = 0x8,
+	SQ_PERF_SEL_WAVES_LT_32                          = 0x9,
+	SQ_PERF_SEL_WAVES_LT_16                          = 0xa,
+	SQ_PERF_SEL_WAVES_CU                             = 0xb,
+	SQ_PERF_SEL_LEVEL_WAVES_CU                       = 0xc,
+	SQ_PERF_SEL_BUSY_CU_CYCLES                       = 0xd,
+	SQ_PERF_SEL_ITEMS                                = 0xe,
+	SQ_PERF_SEL_QUADS                                = 0xf,
+	SQ_PERF_SEL_EVENTS                               = 0x10,
+	SQ_PERF_SEL_SURF_SYNCS                           = 0x11,
+	SQ_PERF_SEL_TTRACE_REQS                          = 0x12,
+	SQ_PERF_SEL_TTRACE_INFLIGHT_REQS                 = 0x13,
+	SQ_PERF_SEL_TTRACE_STALL                         = 0x14,
+	SQ_PERF_SEL_MSG_CNTR                             = 0x15,
+	SQ_PERF_SEL_MSG_PERF                             = 0x16,
+	SQ_PERF_SEL_MSG_GSCNT                            = 0x17,
+	SQ_PERF_SEL_MSG_INTERRUPT                        = 0x18,
+	SQ_PERF_SEL_INSTS                                = 0x19,
+	SQ_PERF_SEL_INSTS_VALU                           = 0x1a,
+	SQ_PERF_SEL_INSTS_VMEM_WR                        = 0x1b,
+	SQ_PERF_SEL_INSTS_VMEM_RD                        = 0x1c,
+	SQ_PERF_SEL_INSTS_VMEM                           = 0x1d,
+	SQ_PERF_SEL_INSTS_SALU                           = 0x1e,
+	SQ_PERF_SEL_INSTS_SMEM                           = 0x1f,
+	SQ_PERF_SEL_INSTS_FLAT                           = 0x20,
+	SQ_PERF_SEL_INSTS_FLAT_LDS_ONLY                  = 0x21,
+	SQ_PERF_SEL_INSTS_LDS                            = 0x22,
+	SQ_PERF_SEL_INSTS_GDS                            = 0x23,
+	SQ_PERF_SEL_INSTS_EXP                            = 0x24,
+	SQ_PERF_SEL_INSTS_EXP_GDS                        = 0x25,
+	SQ_PERF_SEL_INSTS_BRANCH                         = 0x26,
+	SQ_PERF_SEL_INSTS_SENDMSG                        = 0x27,
+	SQ_PERF_SEL_INSTS_VSKIPPED                       = 0x28,
+	SQ_PERF_SEL_INST_LEVEL_VMEM                      = 0x29,
+	SQ_PERF_SEL_INST_LEVEL_SMEM                      = 0x2a,
+	SQ_PERF_SEL_INST_LEVEL_LDS                       = 0x2b,
+	SQ_PERF_SEL_INST_LEVEL_GDS                       = 0x2c,
+	SQ_PERF_SEL_INST_LEVEL_EXP                       = 0x2d,
+	SQ_PERF_SEL_WAVE_CYCLES                          = 0x2e,
+	SQ_PERF_SEL_WAVE_READY                           = 0x2f,
+	SQ_PERF_SEL_WAIT_CNT_VM                          = 0x30,
+	SQ_PERF_SEL_WAIT_CNT_LGKM                        = 0x31,
+	SQ_PERF_SEL_WAIT_CNT_EXP                         = 0x32,
+	SQ_PERF_SEL_WAIT_CNT_ANY                         = 0x33,
+	SQ_PERF_SEL_WAIT_BARRIER                         = 0x34,
+	SQ_PERF_SEL_WAIT_EXP_ALLOC                       = 0x35,
+	SQ_PERF_SEL_WAIT_SLEEP                           = 0x36,
+	SQ_PERF_SEL_WAIT_OTHER                           = 0x37,
+	SQ_PERF_SEL_WAIT_ANY                             = 0x38,
+	SQ_PERF_SEL_WAIT_TTRACE                          = 0x39,
+	SQ_PERF_SEL_WAIT_IFETCH                          = 0x3a,
+	SQ_PERF_SEL_WAIT_INST_VMEM                       = 0x3b,
+	SQ_PERF_SEL_WAIT_INST_SCA                        = 0x3c,
+	SQ_PERF_SEL_WAIT_INST_LDS                        = 0x3d,
+	SQ_PERF_SEL_WAIT_INST_VALU                       = 0x3e,
+	SQ_PERF_SEL_WAIT_INST_EXP_GDS                    = 0x3f,
+	SQ_PERF_SEL_WAIT_INST_MISC                       = 0x40,
+	SQ_PERF_SEL_WAIT_INST_FLAT                       = 0x41,
+	SQ_PERF_SEL_ACTIVE_INST_ANY                      = 0x42,
+	SQ_PERF_SEL_ACTIVE_INST_VMEM                     = 0x43,
+	SQ_PERF_SEL_ACTIVE_INST_LDS                      = 0x44,
+	SQ_PERF_SEL_ACTIVE_INST_VALU                     = 0x45,
+	SQ_PERF_SEL_ACTIVE_INST_SCA                      = 0x46,
+	SQ_PERF_SEL_ACTIVE_INST_EXP_GDS                  = 0x47,
+	SQ_PERF_SEL_ACTIVE_INST_MISC                     = 0x48,
+	SQ_PERF_SEL_ACTIVE_INST_FLAT                     = 0x49,
+	SQ_PERF_SEL_INST_CYCLES_VMEM_WR                  = 0x4a,
+	SQ_PERF_SEL_INST_CYCLES_VMEM_RD                  = 0x4b,
+	SQ_PERF_SEL_INST_CYCLES_VMEM_ADDR                = 0x4c,
+	SQ_PERF_SEL_INST_CYCLES_VMEM_DATA                = 0x4d,
+	SQ_PERF_SEL_INST_CYCLES_VMEM_CMD                 = 0x4e,
+	SQ_PERF_SEL_INST_CYCLES_VMEM                     = 0x4f,
+	SQ_PERF_SEL_INST_CYCLES_LDS                      = 0x50,
+	SQ_PERF_SEL_INST_CYCLES_VALU                     = 0x51,
+	SQ_PERF_SEL_INST_CYCLES_EXP                      = 0x52,
+	SQ_PERF_SEL_INST_CYCLES_GDS                      = 0x53,
+	SQ_PERF_SEL_INST_CYCLES_SCA                      = 0x54,
+	SQ_PERF_SEL_INST_CYCLES_SMEM                     = 0x55,
+	SQ_PERF_SEL_INST_CYCLES_SALU                     = 0x56,
+	SQ_PERF_SEL_INST_CYCLES_EXP_GDS                  = 0x57,
+	SQ_PERF_SEL_INST_CYCLES_MISC                     = 0x58,
+	SQ_PERF_SEL_THREAD_CYCLES_VALU                   = 0x59,
+	SQ_PERF_SEL_THREAD_CYCLES_VALU_MAX               = 0x5a,
+	SQ_PERF_SEL_IFETCH                               = 0x5b,
+	SQ_PERF_SEL_IFETCH_LEVEL                         = 0x5c,
+	SQ_PERF_SEL_CBRANCH_FORK                         = 0x5d,
+	SQ_PERF_SEL_CBRANCH_FORK_SPLIT                   = 0x5e,
+	SQ_PERF_SEL_VALU_LDS_DIRECT_RD                   = 0x5f,
+	SQ_PERF_SEL_VALU_LDS_INTERP_OP                   = 0x60,
+	SQ_PERF_SEL_LDS_BANK_CONFLICT                    = 0x61,
+	SQ_PERF_SEL_LDS_ADDR_CONFLICT                    = 0x62,
+	SQ_PERF_SEL_LDS_UNALIGNED_STALL                  = 0x63,
+	SQ_PERF_SEL_LDS_MEM_VIOLATIONS                   = 0x64,
+	SQ_PERF_SEL_LDS_ATOMIC_RETURN                    = 0x65,
+	SQ_PERF_SEL_LDS_IDX_ACTIVE                       = 0x66,
+	SQ_PERF_SEL_VALU_DEP_STALL                       = 0x67,
+	SQ_PERF_SEL_VALU_STARVE                          = 0x68,
+	SQ_PERF_SEL_EXP_REQ_FIFO_FULL                    = 0x69,
+	SQ_PERF_SEL_LDS_BACK2BACK_STALL                  = 0x6a,
+	SQ_PERF_SEL_LDS_DATA_FIFO_FULL                   = 0x6b,
+	SQ_PERF_SEL_LDS_CMD_FIFO_FULL                    = 0x6c,
+	SQ_PERF_SEL_VMEM_BACK2BACK_STALL                 = 0x6d,
+	SQ_PERF_SEL_VMEM_TA_ADDR_FIFO_FULL               = 0x6e,
+	SQ_PERF_SEL_VMEM_TA_CMD_FIFO_FULL                = 0x6f,
+	SQ_PERF_SEL_VMEM_EX_DATA_REG_BUSY                = 0x70,
+	SQ_PERF_SEL_VMEM_WR_BACK2BACK_STALL              = 0x71,
+	SQ_PERF_SEL_VMEM_WR_TA_DATA_FIFO_FULL            = 0x72,
+	SQ_PERF_SEL_VALU_SRC_C_CONFLICT                  = 0x73,
+	SQ_PERF_SEL_VMEM_RD_SRC_CD_CONFLICT              = 0x74,
+	SQ_PERF_SEL_VMEM_WR_SRC_CD_CONFLICT              = 0x75,
+	SQ_PERF_SEL_FLAT_SRC_CD_CONFLICT                 = 0x76,
+	SQ_PERF_SEL_LDS_SRC_CD_CONFLICT                  = 0x77,
+	SQ_PERF_SEL_SRC_CD_BUSY                          = 0x78,
+	SQ_PERF_SEL_PT_POWER_STALL                       = 0x79,
+	SQ_PERF_SEL_USER0                                = 0x7a,
+	SQ_PERF_SEL_USER1                                = 0x7b,
+	SQ_PERF_SEL_USER2                                = 0x7c,
+	SQ_PERF_SEL_USER3                                = 0x7d,
+	SQ_PERF_SEL_USER4                                = 0x7e,
+	SQ_PERF_SEL_USER5                                = 0x7f,
+	SQ_PERF_SEL_USER6                                = 0x80,
+	SQ_PERF_SEL_USER7                                = 0x81,
+	SQ_PERF_SEL_USER8                                = 0x82,
+	SQ_PERF_SEL_USER9                                = 0x83,
+	SQ_PERF_SEL_USER10                               = 0x84,
+	SQ_PERF_SEL_USER11                               = 0x85,
+	SQ_PERF_SEL_USER12                               = 0x86,
+	SQ_PERF_SEL_USER13                               = 0x87,
+	SQ_PERF_SEL_USER14                               = 0x88,
+	SQ_PERF_SEL_USER15                               = 0x89,
+	SQ_PERF_SEL_USER_LEVEL0                          = 0x8a,
+	SQ_PERF_SEL_USER_LEVEL1                          = 0x8b,
+	SQ_PERF_SEL_USER_LEVEL2                          = 0x8c,
+	SQ_PERF_SEL_USER_LEVEL3                          = 0x8d,
+	SQ_PERF_SEL_USER_LEVEL4                          = 0x8e,
+	SQ_PERF_SEL_USER_LEVEL5                          = 0x8f,
+	SQ_PERF_SEL_USER_LEVEL6                          = 0x90,
+	SQ_PERF_SEL_USER_LEVEL7                          = 0x91,
+	SQ_PERF_SEL_USER_LEVEL8                          = 0x92,
+	SQ_PERF_SEL_USER_LEVEL9                          = 0x93,
+	SQ_PERF_SEL_USER_LEVEL10                         = 0x94,
+	SQ_PERF_SEL_USER_LEVEL11                         = 0x95,
+	SQ_PERF_SEL_USER_LEVEL12                         = 0x96,
+	SQ_PERF_SEL_USER_LEVEL13                         = 0x97,
+	SQ_PERF_SEL_USER_LEVEL14                         = 0x98,
+	SQ_PERF_SEL_USER_LEVEL15                         = 0x99,
+	SQ_PERF_SEL_POWER_VALU                           = 0x9a,
+	SQ_PERF_SEL_POWER_VALU0                          = 0x9b,
+	SQ_PERF_SEL_POWER_VALU1                          = 0x9c,
+	SQ_PERF_SEL_POWER_VALU2                          = 0x9d,
+	SQ_PERF_SEL_POWER_GPR_RD                         = 0x9e,
+	SQ_PERF_SEL_POWER_GPR_WR                         = 0x9f,
+	SQ_PERF_SEL_POWER_LDS_BUSY                       = 0xa0,
+	SQ_PERF_SEL_POWER_ALU_BUSY                       = 0xa1,
+	SQ_PERF_SEL_POWER_TEX_BUSY                       = 0xa2,
+	SQ_PERF_SEL_ACCUM_PREV_HIRES                     = 0xa3,
+	SQ_PERF_SEL_WAVES_RESTORED                       = 0xa4,
+	SQ_PERF_SEL_WAVES_SAVED                          = 0xa5,
+	SQ_PERF_SEL_DUMMY_LAST                           = 0xa7,
+	SQC_PERF_SEL_ICACHE_INPUT_VALID_READY            = 0xa8,
+	SQC_PERF_SEL_ICACHE_INPUT_VALID_READYB           = 0xa9,
+	SQC_PERF_SEL_ICACHE_INPUT_VALIDB                 = 0xaa,
+	SQC_PERF_SEL_DCACHE_INPUT_VALID_READY            = 0xab,
+	SQC_PERF_SEL_DCACHE_INPUT_VALID_READYB           = 0xac,
+	SQC_PERF_SEL_DCACHE_INPUT_VALIDB                 = 0xad,
+	SQC_PERF_SEL_TC_REQ                              = 0xae,
+	SQC_PERF_SEL_TC_INST_REQ                         = 0xaf,
+	SQC_PERF_SEL_TC_DATA_READ_REQ                    = 0xb0,
+	SQC_PERF_SEL_TC_DATA_WRITE_REQ                   = 0xb1,
+	SQC_PERF_SEL_TC_DATA_ATOMIC_REQ                  = 0xb2,
+	SQC_PERF_SEL_TC_STALL                            = 0xb3,
+	SQC_PERF_SEL_TC_STARVE                           = 0xb4,
+	SQC_PERF_SEL_ICACHE_BUSY_CYCLES                  = 0xb5,
+	SQC_PERF_SEL_ICACHE_REQ                          = 0xb6,
+	SQC_PERF_SEL_ICACHE_HITS                         = 0xb7,
+	SQC_PERF_SEL_ICACHE_MISSES                       = 0xb8,
+	SQC_PERF_SEL_ICACHE_MISSES_DUPLICATE             = 0xb9,
+	SQC_PERF_SEL_ICACHE_INVAL_INST                   = 0xba,
+	SQC_PERF_SEL_ICACHE_INVAL_ASYNC                  = 0xbb,
+	SQC_PERF_SEL_ICACHE_INPUT_STALL_ARB_NO_GRANT     = 0xbc,
+	SQC_PERF_SEL_ICACHE_INPUT_STALL_BANK_READYB      = 0xbd,
+	SQC_PERF_SEL_ICACHE_CACHE_STALLED                = 0xbe,
+	SQC_PERF_SEL_ICACHE_CACHE_STALL_INFLIGHT_NONZERO = 0xbf,
+	SQC_PERF_SEL_ICACHE_CACHE_STALL_INFLIGHT_MAX     = 0xc0,
+	SQC_PERF_SEL_ICACHE_CACHE_STALL_OUTPUT           = 0xc1,
+	SQC_PERF_SEL_ICACHE_CACHE_STALL_OUTPUT_MISS_FIFO = 0xc2,
+	SQC_PERF_SEL_ICACHE_CACHE_STALL_OUTPUT_HIT_FIFO  = 0xc3,
+	SQC_PERF_SEL_ICACHE_CACHE_STALL_OUTPUT_TC_IF     = 0xc4,
+	SQC_PERF_SEL_ICACHE_STALL_OUTXBAR_ARB_NO_GRANT   = 0xc5,
+	SQC_PERF_SEL_DCACHE_BUSY_CYCLES                  = 0xc6,
+	SQC_PERF_SEL_DCACHE_REQ                          = 0xc7,
+	SQC_PERF_SEL_DCACHE_HITS                         = 0xc8,
+	SQC_PERF_SEL_DCACHE_MISSES                       = 0xc9,
+	SQC_PERF_SEL_DCACHE_MISSES_DUPLICATE             = 0xca,
+	SQC_PERF_SEL_DCACHE_HIT_LRU_READ                 = 0xcb,
+	SQC_PERF_SEL_DCACHE_MISS_EVICT_READ              = 0xcc,
+	SQC_PERF_SEL_DCACHE_WC_LRU_WRITE                 = 0xcd,
+	SQC_PERF_SEL_DCACHE_WT_EVICT_WRITE               = 0xce,
+	SQC_PERF_SEL_DCACHE_ATOMIC                       = 0xcf,
+	SQC_PERF_SEL_DCACHE_VOLATILE                     = 0xd0,
+	SQC_PERF_SEL_DCACHE_INVAL_INST                   = 0xd1,
+	SQC_PERF_SEL_DCACHE_INVAL_ASYNC                  = 0xd2,
+	SQC_PERF_SEL_DCACHE_INVAL_VOLATILE_INST          = 0xd3,
+	SQC_PERF_SEL_DCACHE_INVAL_VOLATILE_ASYNC         = 0xd4,
+	SQC_PERF_SEL_DCACHE_WB_INST                      = 0xd5,
+	SQC_PERF_SEL_DCACHE_WB_ASYNC                     = 0xd6,
+	SQC_PERF_SEL_DCACHE_WB_VOLATILE_INST             = 0xd7,
+	SQC_PERF_SEL_DCACHE_WB_VOLATILE_ASYNC            = 0xd8,
+	SQC_PERF_SEL_DCACHE_INPUT_STALL_ARB_NO_GRANT     = 0xd9,
+	SQC_PERF_SEL_DCACHE_INPUT_STALL_BANK_READYB      = 0xda,
+	SQC_PERF_SEL_DCACHE_CACHE_STALLED                = 0xdb,
+	SQC_PERF_SEL_DCACHE_CACHE_STALL_INFLIGHT_MAX     = 0xdc,
+	SQC_PERF_SEL_DCACHE_CACHE_STALL_OUTPUT           = 0xdd,
+	SQC_PERF_SEL_DCACHE_CACHE_STALL_EVICT            = 0xde,
+	SQC_PERF_SEL_DCACHE_CACHE_STALL_UNORDERED        = 0xdf,
+	SQC_PERF_SEL_DCACHE_CACHE_STALL_ALLOC_UNAVAILABLE= 0xe0,
+	SQC_PERF_SEL_DCACHE_CACHE_STALL_FORCE_EVICT      = 0xe1,
+	SQC_PERF_SEL_DCACHE_CACHE_STALL_MULTI_FLUSH      = 0xe2,
+	SQC_PERF_SEL_DCACHE_CACHE_STALL_FLUSH_DONE       = 0xe3,
+	SQC_PERF_SEL_DCACHE_CACHE_STALL_OUTPUT_MISS_FIFO = 0xe4,
+	SQC_PERF_SEL_DCACHE_CACHE_STALL_OUTPUT_HIT_FIFO  = 0xe5,
+	SQC_PERF_SEL_DCACHE_CACHE_STALL_OUTPUT_TC_IF     = 0xe6,
+	SQC_PERF_SEL_DCACHE_STALL_OUTXBAR_ARB_NO_GRANT   = 0xe7,
+	SQC_PERF_SEL_DCACHE_REQ_READ_1                   = 0xe8,
+	SQC_PERF_SEL_DCACHE_REQ_READ_2                   = 0xe9,
+	SQC_PERF_SEL_DCACHE_REQ_READ_4                   = 0xea,
+	SQC_PERF_SEL_DCACHE_REQ_READ_8                   = 0xeb,
+	SQC_PERF_SEL_DCACHE_REQ_READ_16                  = 0xec,
+	SQC_PERF_SEL_DCACHE_REQ_TIME                     = 0xed,
+	SQC_PERF_SEL_DCACHE_REQ_WRITE_1                  = 0xee,
+	SQC_PERF_SEL_DCACHE_REQ_WRITE_2                  = 0xef,
+	SQC_PERF_SEL_DCACHE_REQ_WRITE_4                  = 0xf0,
+	SQC_PERF_SEL_DCACHE_REQ_ATC_PROBE                = 0xf1,
+	SQC_PERF_SEL_SQ_DCACHE_REQS                      = 0xf2,
+	SQC_PERF_SEL_DCACHE_FLAT_REQ                     = 0xf3,
+	SQC_PERF_SEL_DCACHE_NONFLAT_REQ                  = 0xf4,
+	SQC_PERF_SEL_ICACHE_INFLIGHT_LEVEL               = 0xf5,
+	SQC_PERF_SEL_DCACHE_INFLIGHT_LEVEL               = 0xf6,
+	SQC_PERF_SEL_TC_INFLIGHT_LEVEL                   = 0xf7,
+	SQC_PERF_SEL_ICACHE_TC_INFLIGHT_LEVEL            = 0xf8,
+	SQC_PERF_SEL_DCACHE_TC_INFLIGHT_LEVEL            = 0xf9,
+	SQC_PERF_SEL_ICACHE_GATCL1_TRANSLATION_MISS      = 0xfa,
+	SQC_PERF_SEL_ICACHE_GATCL1_PERMISSION_MISS       = 0xfb,
+	SQC_PERF_SEL_ICACHE_GATCL1_REQUEST               = 0xfc,
+	SQC_PERF_SEL_ICACHE_GATCL1_STALL_INFLIGHT_MAX    = 0xfd,
+	SQC_PERF_SEL_ICACHE_GATCL1_STALL_LRU_INFLIGHT    = 0xfe,
+	SQC_PERF_SEL_ICACHE_GATCL1_LFIFO_FULL            = 0xff,
+	SQC_PERF_SEL_ICACHE_GATCL1_STALL_LFIFO_NOT_RES   = 0x100,
+	SQC_PERF_SEL_ICACHE_GATCL1_STALL_ATCL2_REQ_OUT_OF_CREDITS= 0x101,
+	SQC_PERF_SEL_ICACHE_GATCL1_ATCL2_INFLIGHT        = 0x102,
+	SQC_PERF_SEL_ICACHE_GATCL1_STALL_MISSFIFO_FULL   = 0x103,
+	SQC_PERF_SEL_DCACHE_GATCL1_TRANSLATION_MISS      = 0x104,
+	SQC_PERF_SEL_DCACHE_GATCL1_PERMISSION_MISS       = 0x105,
+	SQC_PERF_SEL_DCACHE_GATCL1_REQUEST               = 0x106,
+	SQC_PERF_SEL_DCACHE_GATCL1_STALL_INFLIGHT_MAX    = 0x107,
+	SQC_PERF_SEL_DCACHE_GATCL1_STALL_LRU_INFLIGHT    = 0x108,
+	SQC_PERF_SEL_DCACHE_GATCL1_LFIFO_FULL            = 0x109,
+	SQC_PERF_SEL_DCACHE_GATCL1_STALL_LFIFO_NOT_RES   = 0x10a,
+	SQC_PERF_SEL_DCACHE_GATCL1_STALL_ATCL2_REQ_OUT_OF_CREDITS= 0x10b,
+	SQC_PERF_SEL_DCACHE_GATCL1_ATCL2_INFLIGHT        = 0x10c,
+	SQC_PERF_SEL_DCACHE_GATCL1_STALL_MISSFIFO_FULL   = 0x10d,
+	SQC_PERF_SEL_DCACHE_GATCL1_STALL_MULTI_MISS      = 0x10e,
+	SQC_PERF_SEL_DCACHE_GATCL1_HIT_FIFO_FULL         = 0x10f,
+	SQC_PERF_SEL_DUMMY_LAST                          = 0x110,
+	SQ_PERF_SEL_INSTS_SMEM_NORM                      = 0x111,
+	SQ_PERF_SEL_ATC_INSTS_VMEM                       = 0x112,
+	SQ_PERF_SEL_ATC_INST_LEVEL_VMEM                  = 0x113,
+	SQ_PERF_SEL_ATC_XNACK_FIRST                      = 0x114,
+	SQ_PERF_SEL_ATC_XNACK_ALL                        = 0x115,
+	SQ_PERF_SEL_ATC_XNACK_FIFO_FULL                  = 0x116,
+	SQ_PERF_SEL_ATC_INSTS_SMEM                       = 0x117,
+	SQ_PERF_SEL_ATC_INST_LEVEL_SMEM                  = 0x118,
+	SQ_PERF_SEL_IFETCH_XNACK                         = 0x119,
+	SQ_PERF_SEL_TLB_SHOOTDOWN                        = 0x11a,
+	SQ_PERF_SEL_TLB_SHOOTDOWN_CYCLES                 = 0x11b,
+	SQ_PERF_SEL_INSTS_VMEM_WR_REPLAY                 = 0x11c,
+	SQ_PERF_SEL_INSTS_VMEM_RD_REPLAY                 = 0x11d,
+	SQ_PERF_SEL_INSTS_VMEM_REPLAY                    = 0x11e,
+	SQ_PERF_SEL_INSTS_SMEM_REPLAY                    = 0x11f,
+	SQ_PERF_SEL_INSTS_SMEM_NORM_REPLAY               = 0x120,
+	SQ_PERF_SEL_INSTS_FLAT_REPLAY                    = 0x121,
+	SQ_PERF_SEL_ATC_INSTS_VMEM_REPLAY                = 0x122,
+	SQ_PERF_SEL_ATC_INSTS_SMEM_REPLAY                = 0x123,
+	SQ_PERF_SEL_DUMMY_LAST1                          = 0x12a,
+} SQ_PERF_SEL;
+typedef enum SQ_CAC_POWER_SEL {
+	SQ_CAC_POWER_VALU                                = 0x0,
+	SQ_CAC_POWER_VALU0                               = 0x1,
+	SQ_CAC_POWER_VALU1                               = 0x2,
+	SQ_CAC_POWER_VALU2                               = 0x3,
+	SQ_CAC_POWER_GPR_RD                              = 0x4,
+	SQ_CAC_POWER_GPR_WR                              = 0x5,
+	SQ_CAC_POWER_LDS_BUSY                            = 0x6,
+	SQ_CAC_POWER_ALU_BUSY                            = 0x7,
+	SQ_CAC_POWER_TEX_BUSY                            = 0x8,
+} SQ_CAC_POWER_SEL;
+typedef enum SQ_IND_CMD_CMD {
+	SQ_IND_CMD_CMD_NULL                              = 0x0,
+	SQ_IND_CMD_CMD_SETHALT                           = 0x1,
+	SQ_IND_CMD_CMD_SAVECTX                           = 0x2,
+	SQ_IND_CMD_CMD_KILL                              = 0x3,
+	SQ_IND_CMD_CMD_DEBUG                             = 0x4,
+	SQ_IND_CMD_CMD_TRAP                              = 0x5,
+	SQ_IND_CMD_CMD_SET_SPI_PRIO                      = 0x6,
+} SQ_IND_CMD_CMD;
+typedef enum SQ_IND_CMD_MODE {
+	SQ_IND_CMD_MODE_SINGLE                           = 0x0,
+	SQ_IND_CMD_MODE_BROADCAST                        = 0x1,
+	SQ_IND_CMD_MODE_BROADCAST_QUEUE                  = 0x2,
+	SQ_IND_CMD_MODE_BROADCAST_PIPE                   = 0x3,
+	SQ_IND_CMD_MODE_BROADCAST_ME                     = 0x4,
+} SQ_IND_CMD_MODE;
+typedef enum SQ_EDC_INFO_SOURCE {
+	SQ_EDC_INFO_SOURCE_INVALID                       = 0x0,
+	SQ_EDC_INFO_SOURCE_INST                          = 0x1,
+	SQ_EDC_INFO_SOURCE_SGPR                          = 0x2,
+	SQ_EDC_INFO_SOURCE_VGPR                          = 0x3,
+	SQ_EDC_INFO_SOURCE_LDS                           = 0x4,
+	SQ_EDC_INFO_SOURCE_GDS                           = 0x5,
+	SQ_EDC_INFO_SOURCE_TA                            = 0x6,
+} SQ_EDC_INFO_SOURCE;
+typedef enum SQ_ROUND_MODE {
+	SQ_ROUND_NEAREST_EVEN                            = 0x0,
+	SQ_ROUND_PLUS_INFINITY                           = 0x1,
+	SQ_ROUND_MINUS_INFINITY                          = 0x2,
+	SQ_ROUND_TO_ZERO                                 = 0x3,
+} SQ_ROUND_MODE;
+typedef enum SQ_INTERRUPT_WORD_ENCODING {
+	SQ_INTERRUPT_WORD_ENCODING_AUTO                  = 0x0,
+	SQ_INTERRUPT_WORD_ENCODING_INST                  = 0x1,
+	SQ_INTERRUPT_WORD_ENCODING_ERROR                 = 0x2,
+} SQ_INTERRUPT_WORD_ENCODING;
+typedef enum ENUM_SQ_EXPORT_RAT_INST {
+	SQ_EXPORT_RAT_INST_NOP                           = 0x0,
+	SQ_EXPORT_RAT_INST_STORE_TYPED                   = 0x1,
+	SQ_EXPORT_RAT_INST_STORE_RAW                     = 0x2,
+	SQ_EXPORT_RAT_INST_STORE_RAW_FDENORM             = 0x3,
+	SQ_EXPORT_RAT_INST_CMPXCHG_INT                   = 0x4,
+	SQ_EXPORT_RAT_INST_CMPXCHG_FLT                   = 0x5,
+	SQ_EXPORT_RAT_INST_CMPXCHG_FDENORM               = 0x6,
+	SQ_EXPORT_RAT_INST_ADD                           = 0x7,
+	SQ_EXPORT_RAT_INST_SUB                           = 0x8,
+	SQ_EXPORT_RAT_INST_RSUB                          = 0x9,
+	SQ_EXPORT_RAT_INST_MIN_INT                       = 0xa,
+	SQ_EXPORT_RAT_INST_MIN_UINT                      = 0xb,
+	SQ_EXPORT_RAT_INST_MAX_INT                       = 0xc,
+	SQ_EXPORT_RAT_INST_MAX_UINT                      = 0xd,
+	SQ_EXPORT_RAT_INST_AND                           = 0xe,
+	SQ_EXPORT_RAT_INST_OR                            = 0xf,
+	SQ_EXPORT_RAT_INST_XOR                           = 0x10,
+	SQ_EXPORT_RAT_INST_MSKOR                         = 0x11,
+	SQ_EXPORT_RAT_INST_INC_UINT                      = 0x12,
+	SQ_EXPORT_RAT_INST_DEC_UINT                      = 0x13,
+	SQ_EXPORT_RAT_INST_STORE_DWORD                   = 0x14,
+	SQ_EXPORT_RAT_INST_STORE_SHORT                   = 0x15,
+	SQ_EXPORT_RAT_INST_STORE_BYTE                    = 0x16,
+	SQ_EXPORT_RAT_INST_NOP_RTN                       = 0x20,
+	SQ_EXPORT_RAT_INST_XCHG_RTN                      = 0x22,
+	SQ_EXPORT_RAT_INST_XCHG_FDENORM_RTN              = 0x23,
+	SQ_EXPORT_RAT_INST_CMPXCHG_INT_RTN               = 0x24,
+	SQ_EXPORT_RAT_INST_CMPXCHG_FLT_RTN               = 0x25,
+	SQ_EXPORT_RAT_INST_CMPXCHG_FDENORM_RTN           = 0x26,
+	SQ_EXPORT_RAT_INST_ADD_RTN                       = 0x27,
+	SQ_EXPORT_RAT_INST_SUB_RTN                       = 0x28,
+	SQ_EXPORT_RAT_INST_RSUB_RTN                      = 0x29,
+	SQ_EXPORT_RAT_INST_MIN_INT_RTN                   = 0x2a,
+	SQ_EXPORT_RAT_INST_MIN_UINT_RTN                  = 0x2b,
+	SQ_EXPORT_RAT_INST_MAX_INT_RTN                   = 0x2c,
+	SQ_EXPORT_RAT_INST_MAX_UINT_RTN                  = 0x2d,
+	SQ_EXPORT_RAT_INST_AND_RTN                       = 0x2e,
+	SQ_EXPORT_RAT_INST_OR_RTN                        = 0x2f,
+	SQ_EXPORT_RAT_INST_XOR_RTN                       = 0x30,
+	SQ_EXPORT_RAT_INST_MSKOR_RTN                     = 0x31,
+	SQ_EXPORT_RAT_INST_INC_UINT_RTN                  = 0x32,
+	SQ_EXPORT_RAT_INST_DEC_UINT_RTN                  = 0x33,
+} ENUM_SQ_EXPORT_RAT_INST;
+typedef enum SQ_IBUF_ST {
+	SQ_IBUF_IB_IDLE                                  = 0x0,
+	SQ_IBUF_IB_INI_WAIT_GNT                          = 0x1,
+	SQ_IBUF_IB_INI_WAIT_DRET                         = 0x2,
+	SQ_IBUF_IB_LE_4DW                                = 0x3,
+	SQ_IBUF_IB_WAIT_DRET                             = 0x4,
+	SQ_IBUF_IB_EMPTY_WAIT_DRET                       = 0x5,
+	SQ_IBUF_IB_DRET                                  = 0x6,
+	SQ_IBUF_IB_EMPTY_WAIT_GNT                        = 0x7,
+} SQ_IBUF_ST;
+typedef enum SQ_INST_STR_ST {
+	SQ_INST_STR_IB_WAVE_NORML                        = 0x0,
+	SQ_INST_STR_IB_WAVE2ID_NORMAL_INST_AV            = 0x1,
+	SQ_INST_STR_IB_WAVE_INTERNAL_INST_AV             = 0x2,
+	SQ_INST_STR_IB_WAVE_INST_SKIP_AV                 = 0x3,
+	SQ_INST_STR_IB_WAVE_SETVSKIP_ST0                 = 0x4,
+	SQ_INST_STR_IB_WAVE_SETVSKIP_ST1                 = 0x5,
+	SQ_INST_STR_IB_WAVE_NOP_SLEEP_WAIT               = 0x6,
+	SQ_INST_STR_IB_WAVE_PC_FROM_SGPR_MSG_WAIT        = 0x7,
+} SQ_INST_STR_ST;
+typedef enum SQ_WAVE_IB_ECC_ST {
+	SQ_WAVE_IB_ECC_CLEAN                             = 0x0,
+	SQ_WAVE_IB_ECC_ERR_CONTINUE                      = 0x1,
+	SQ_WAVE_IB_ECC_ERR_HALT                          = 0x2,
+	SQ_WAVE_IB_ECC_WITH_ERR_MSG                      = 0x3,
+} SQ_WAVE_IB_ECC_ST;
+typedef enum SH_MEM_ADDRESS_MODE {
+	SH_MEM_ADDRESS_MODE_GPUVM64                      = 0x0,
+	SH_MEM_ADDRESS_MODE_GPUVM32                      = 0x1,
+	SH_MEM_ADDRESS_MODE_HSA64                        = 0x2,
+	SH_MEM_ADDRESS_MODE_HSA32                        = 0x3,
+} SH_MEM_ADDRESS_MODE;
+typedef enum SH_MEM_ALIGNMENT_MODE {
+	SH_MEM_ALIGNMENT_MODE_DWORD                      = 0x0,
+	SH_MEM_ALIGNMENT_MODE_DWORD_STRICT               = 0x1,
+	SH_MEM_ALIGNMENT_MODE_STRICT                     = 0x2,
+	SH_MEM_ALIGNMENT_MODE_UNALIGNED                  = 0x3,
+} SH_MEM_ALIGNMENT_MODE;
+typedef enum SQ_THREAD_TRACE_WAVE_START_COUNT_PREFIX {
+	SQ_THREAD_TRACE_WAVE_START_COUNT_PREFIX_WREXEC   = 0x18,
+	SQ_THREAD_TRACE_WAVE_START_COUNT_PREFIX_RESTORE  = 0x19,
+} SQ_THREAD_TRACE_WAVE_START_COUNT_PREFIX;
+#define SQ_WAVE_TYPE_PS0                          0x0
+#define SQIND_GLOBAL_REGS_OFFSET                  0x0
+#define SQIND_GLOBAL_REGS_SIZE                    0x8
+#define SQIND_LOCAL_REGS_OFFSET                   0x8
+#define SQIND_LOCAL_REGS_SIZE                     0x8
+#define SQIND_WAVE_HWREGS_OFFSET                  0x10
+#define SQIND_WAVE_HWREGS_SIZE                    0x1f0
+#define SQIND_WAVE_SGPRS_OFFSET                   0x200
+#define SQIND_WAVE_SGPRS_SIZE                     0x200
+#define SQ_GFXDEC_BEGIN                           0xa000
+#define SQ_GFXDEC_END                             0xc000
+#define SQ_GFXDEC_STATE_ID_SHIFT                  0xa
+#define SQDEC_BEGIN                               0x2300
+#define SQDEC_END                                 0x23ff
+#define SQPERFSDEC_BEGIN                          0xd9c0
+#define SQPERFSDEC_END                            0xda40
+#define SQPERFDDEC_BEGIN                          0xd1c0
+#define SQPERFDDEC_END                            0xd240
+#define SQGFXUDEC_BEGIN                           0xc330
+#define SQGFXUDEC_END                             0xc380
+#define SQPWRDEC_BEGIN                            0xf08c
+#define SQPWRDEC_END                              0xf094
+#define SQ_DISPATCHER_GFX_MIN                     0x10
+#define SQ_DISPATCHER_GFX_CNT_PER_RING            0x8
+#define SQ_MAX_PGM_SGPRS                          0x68
+#define SQ_MAX_PGM_VGPRS                          0x100
+#define SQ_THREAD_TRACE_TIME_UNIT                 0x4
+#define SQ_EX_MODE_EXCP_VALU_BASE                 0x0
+#define SQ_EX_MODE_EXCP_VALU_SIZE                 0x7
+#define SQ_EX_MODE_EXCP_INVALID                   0x0
+#define SQ_EX_MODE_EXCP_INPUT_DENORM              0x1
+#define SQ_EX_MODE_EXCP_DIV0                      0x2
+#define SQ_EX_MODE_EXCP_OVERFLOW                  0x3
+#define SQ_EX_MODE_EXCP_UNDERFLOW                 0x4
+#define SQ_EX_MODE_EXCP_INEXACT                   0x5
+#define SQ_EX_MODE_EXCP_INT_DIV0                  0x6
+#define SQ_EX_MODE_EXCP_ADDR_WATCH                0x7
+#define SQ_EX_MODE_EXCP_MEM_VIOL                  0x8
+#define INST_ID_PRIV_START                        0x80000000
+#define INST_ID_ECC_INTERRUPT_MSG                 0xfffffff0
+#define INST_ID_TTRACE_NEW_PC_MSG                 0xfffffff1
+#define INST_ID_HW_TRAP                           0xfffffff2
+#define INST_ID_KILL_SEQ                          0xfffffff3
+#define INST_ID_SPI_WREXEC                        0xfffffff4
+#define INST_ID_HOST_REG_TRAP_MSG                 0xfffffffe
+#define SQ_ENC_SOP1_BITS                          0xbe800000
+#define SQ_ENC_SOP1_MASK                          0xff800000
+#define SQ_ENC_SOP1_FIELD                         0x17d
+#define SQ_ENC_SOPC_BITS                          0xbf000000
+#define SQ_ENC_SOPC_MASK                          0xff800000
+#define SQ_ENC_SOPC_FIELD                         0x17e
+#define SQ_ENC_SOPP_BITS                          0xbf800000
+#define SQ_ENC_SOPP_MASK                          0xff800000
+#define SQ_ENC_SOPP_FIELD                         0x17f
+#define SQ_ENC_SOPK_BITS                          0xb0000000
+#define SQ_ENC_SOPK_MASK                          0xf0000000
+#define SQ_ENC_SOPK_FIELD                         0xb
+#define SQ_ENC_SOP2_BITS                          0x80000000
+#define SQ_ENC_SOP2_MASK                          0xc0000000
+#define SQ_ENC_SOP2_FIELD                         0x2
+#define SQ_ENC_SMEM_BITS                          0xc0000000
+#define SQ_ENC_SMEM_MASK                          0xfc000000
+#define SQ_ENC_SMEM_FIELD                         0x30
+#define SQ_ENC_VOP1_BITS                          0x7e000000
+#define SQ_ENC_VOP1_MASK                          0xfe000000
+#define SQ_ENC_VOP1_FIELD                         0x3f
+#define SQ_ENC_VOPC_BITS                          0x7c000000
+#define SQ_ENC_VOPC_MASK                          0xfe000000
+#define SQ_ENC_VOPC_FIELD                         0x3e
+#define SQ_ENC_VOP2_BITS                          0x0
+#define SQ_ENC_VOP2_MASK                          0x80000000
+#define SQ_ENC_VOP2_FIELD                         0x0
+#define SQ_ENC_VINTRP_BITS                        0xd4000000
+#define SQ_ENC_VINTRP_MASK                        0xfc000000
+#define SQ_ENC_VINTRP_FIELD                       0x35
+#define SQ_ENC_VOP3_BITS                          0xd0000000
+#define SQ_ENC_VOP3_MASK                          0xfc000000
+#define SQ_ENC_VOP3_FIELD                         0x34
+#define SQ_ENC_DS_BITS                            0xd8000000
+#define SQ_ENC_DS_MASK                            0xfc000000
+#define SQ_ENC_DS_FIELD                           0x36
+#define SQ_ENC_MUBUF_BITS                         0xe0000000
+#define SQ_ENC_MUBUF_MASK                         0xfc000000
+#define SQ_ENC_MUBUF_FIELD                        0x38
+#define SQ_ENC_MTBUF_BITS                         0xe8000000
+#define SQ_ENC_MTBUF_MASK                         0xfc000000
+#define SQ_ENC_MTBUF_FIELD                        0x3a
+#define SQ_ENC_MIMG_BITS                          0xf0000000
+#define SQ_ENC_MIMG_MASK                          0xfc000000
+#define SQ_ENC_MIMG_FIELD                         0x3c
+#define SQ_ENC_EXP_BITS                           0xc4000000
+#define SQ_ENC_EXP_MASK                           0xfc000000
+#define SQ_ENC_EXP_FIELD                          0x31
+#define SQ_ENC_FLAT_BITS                          0xdc000000
+#define SQ_ENC_FLAT_MASK                          0xfc000000
+#define SQ_ENC_FLAT_FIELD                         0x37
+#define SQ_V_OP3_INTRP_OFFSET                     0x274
+#define SQ_WAITCNT_VM_SHIFT                       0x0
+#define SQ_SENDMSG_STREAMID_SIZE                  0x2
+#define SQ_V_OPC_COUNT                            0x100
+#define SQ_V_OP3_INTRP_COUNT                      0xc
+#define SQ_XLATE_VOP3_TO_VOP2_OFFSET              0x100
+#define SQ_HWREG_OFFSET_SIZE                      0x5
+#define SQ_HWREG_OFFSET_SHIFT                     0x6
+#define SQ_V_OP3_3IN_OFFSET                       0x1c0
+#define SQ_NUM_ATTR                               0x21
+#define SQ_NUM_VGPR                               0x100
+#define SQ_XLATE_VOP3_TO_VINTRP_COUNT             0x4
+#define SQ_SENDMSG_MSG_SIZE                       0x4
+#define SQ_NUM_TTMP                               0xc
+#define SQ_HWREG_ID_SIZE                          0x6
+#define SQ_SENDMSG_GSOP_SIZE                      0x2
+#define SQ_NUM_SGPR                               0x66
+#define SQ_EXP_NUM_MRT                            0x8
+#define SQ_SENDMSG_SYSTEM_SIZE                    0x3
+#define SQ_WAITCNT_LGKM_SHIFT                     0x8
+#define SQ_XLATE_VOP3_TO_VOP2_COUNT               0x40
+#define SQ_V_OP3_3IN_COUNT                        0xb0
+#define SQ_V_INTRP_COUNT                          0x4
+#define SQ_WAITCNT_EXP_SIZE                       0x3
+#define SQ_SENDMSG_SYSTEM_SHIFT                   0x4
+#define SQ_EXP_NUM_GDS                            0x5
+#define SQ_HWREG_SIZE_SHIFT                       0xb
+#define SQ_XLATE_VOP3_TO_VOPC_OFFSET              0x0
+#define SQ_V_OP3_2IN_COUNT                        0x80
+#define SQ_XLATE_VOP3_TO_VINTRP_OFFSET            0x270
+#define SQ_SENDMSG_MSG_SHIFT                      0x0
+#define SQ_WAITCNT_EXP_SHIFT                      0x4
+#define SQ_WAITCNT_VM_SIZE                        0x4
+#define SQ_XLATE_VOP3_TO_VOP1_OFFSET              0x140
+#define SQ_SENDMSG_GSOP_SHIFT                     0x4
+#define SQ_XLATE_VOP3_TO_VOP1_COUNT               0x80
+#define SQ_SRC_VGPR_BIT                           0x100
+#define SQ_V_OP2_COUNT                            0x40
+#define SQ_EXP_NUM_PARAM                          0x20
+#define SQ_V_OP1_COUNT                            0x80
+#define SQ_SENDMSG_STREAMID_SHIFT                 0x8
+#define SQ_V_OP3_2IN_OFFSET                       0x280
+#define SQ_WAITCNT_LGKM_SIZE                      0x4
+#define SQ_XLATE_VOP3_TO_VOPC_COUNT               0x100
+#define SQ_EXP_NUM_POS                            0x4
+#define SQ_HWREG_SIZE_SIZE                        0x5
+#define SQ_HWREG_ID_SHIFT                         0x0
+#define SQ_S_MOV_B32                              0x0
+#define SQ_S_MOV_B64                              0x1
+#define SQ_S_CMOV_B32                             0x2
+#define SQ_S_CMOV_B64                             0x3
+#define SQ_S_NOT_B32                              0x4
+#define SQ_S_NOT_B64                              0x5
+#define SQ_S_WQM_B32                              0x6
+#define SQ_S_WQM_B64                              0x7
+#define SQ_S_BREV_B32                             0x8
+#define SQ_S_BREV_B64                             0x9
+#define SQ_S_BCNT0_I32_B32                        0xa
+#define SQ_S_BCNT0_I32_B64                        0xb
+#define SQ_S_BCNT1_I32_B32                        0xc
+#define SQ_S_BCNT1_I32_B64                        0xd
+#define SQ_S_FF0_I32_B32                          0xe
+#define SQ_S_FF0_I32_B64                          0xf
+#define SQ_S_FF1_I32_B32                          0x10
+#define SQ_S_FF1_I32_B64                          0x11
+#define SQ_S_FLBIT_I32_B32                        0x12
+#define SQ_S_FLBIT_I32_B64                        0x13
+#define SQ_S_FLBIT_I32                            0x14
+#define SQ_S_FLBIT_I32_I64                        0x15
+#define SQ_S_SEXT_I32_I8                          0x16
+#define SQ_S_SEXT_I32_I16                         0x17
+#define SQ_S_BITSET0_B32                          0x18
+#define SQ_S_BITSET0_B64                          0x19
+#define SQ_S_BITSET1_B32                          0x1a
+#define SQ_S_BITSET1_B64                          0x1b
+#define SQ_S_GETPC_B64                            0x1c
+#define SQ_S_SETPC_B64                            0x1d
+#define SQ_S_SWAPPC_B64                           0x1e
+#define SQ_S_RFE_B64                              0x1f
+#define SQ_S_AND_SAVEEXEC_B64                     0x20
+#define SQ_S_OR_SAVEEXEC_B64                      0x21
+#define SQ_S_XOR_SAVEEXEC_B64                     0x22
+#define SQ_S_ANDN2_SAVEEXEC_B64                   0x23
+#define SQ_S_ORN2_SAVEEXEC_B64                    0x24
+#define SQ_S_NAND_SAVEEXEC_B64                    0x25
+#define SQ_S_NOR_SAVEEXEC_B64                     0x26
+#define SQ_S_XNOR_SAVEEXEC_B64                    0x27
+#define SQ_S_QUADMASK_B32                         0x28
+#define SQ_S_QUADMASK_B64                         0x29
+#define SQ_S_MOVRELS_B32                          0x2a
+#define SQ_S_MOVRELS_B64                          0x2b
+#define SQ_S_MOVRELD_B32                          0x2c
+#define SQ_S_MOVRELD_B64                          0x2d
+#define SQ_S_CBRANCH_JOIN                         0x2e
+#define SQ_S_MOV_REGRD_B32                        0x2f
+#define SQ_S_ABS_I32                              0x30
+#define SQ_S_MOV_FED_B32                          0x31
+#define SQ_S_SET_GPR_IDX_IDX                      0x32
+#define SQ_ATTR0                                  0x0
+#define SQ_S_MOVK_I32                             0x0
+#define SQ_S_CMOVK_I32                            0x1
+#define SQ_S_CMPK_EQ_I32                          0x2
+#define SQ_S_CMPK_LG_I32                          0x3
+#define SQ_S_CMPK_GT_I32                          0x4
+#define SQ_S_CMPK_GE_I32                          0x5
+#define SQ_S_CMPK_LT_I32                          0x6
+#define SQ_S_CMPK_LE_I32                          0x7
+#define SQ_S_CMPK_EQ_U32                          0x8
+#define SQ_S_CMPK_LG_U32                          0x9
+#define SQ_S_CMPK_GT_U32                          0xa
+#define SQ_S_CMPK_GE_U32                          0xb
+#define SQ_S_CMPK_LT_U32                          0xc
+#define SQ_S_CMPK_LE_U32                          0xd
+#define SQ_S_ADDK_I32                             0xe
+#define SQ_S_MULK_I32                             0xf
+#define SQ_S_CBRANCH_I_FORK                       0x10
+#define SQ_S_GETREG_B32                           0x11
+#define SQ_S_SETREG_B32                           0x12
+#define SQ_S_GETREG_REGRD_B32                     0x13
+#define SQ_S_SETREG_IMM32_B32                     0x14
+#define SQ_TBA_LO                                 0x6c
+#define SQ_TBA_HI                                 0x6d
+#define SQ_TMA_LO                                 0x6e
+#define SQ_TMA_HI                                 0x6f
+#define SQ_TTMP0                                  0x70
+#define SQ_TTMP1                                  0x71
+#define SQ_TTMP2                                  0x72
+#define SQ_TTMP3                                  0x73
+#define SQ_TTMP4                                  0x74
+#define SQ_TTMP5                                  0x75
+#define SQ_TTMP6                                  0x76
+#define SQ_TTMP7                                  0x77
+#define SQ_TTMP8                                  0x78
+#define SQ_TTMP9                                  0x79
+#define SQ_TTMP10                                 0x7a
+#define SQ_TTMP11                                 0x7b
+#define SQ_VGPR0                                  0x0
+#define SQ_EXP                                    0x0
+#define SQ_EXP_MRT0                               0x0
+#define SQ_EXP_MRTZ                               0x8
+#define SQ_EXP_NULL                               0x9
+#define SQ_EXP_POS0                               0xc
+#define SQ_EXP_PARAM0                             0x20
+#define SQ_CNT1                                   0x0
+#define SQ_CNT2                                   0x1
+#define SQ_CNT3                                   0x2
+#define SQ_CNT4                                   0x3
+#define SQ_S_LOAD_DWORD                           0x0
+#define SQ_S_LOAD_DWORDX2                         0x1
+#define SQ_S_LOAD_DWORDX4                         0x2
+#define SQ_S_LOAD_DWORDX8                         0x3
+#define SQ_S_LOAD_DWORDX16                        0x4
+#define SQ_S_BUFFER_LOAD_DWORD                    0x8
+#define SQ_S_BUFFER_LOAD_DWORDX2                  0x9
+#define SQ_S_BUFFER_LOAD_DWORDX4                  0xa
+#define SQ_S_BUFFER_LOAD_DWORDX8                  0xb
+#define SQ_S_BUFFER_LOAD_DWORDX16                 0xc
+#define SQ_S_STORE_DWORD                          0x10
+#define SQ_S_STORE_DWORDX2                        0x11
+#define SQ_S_STORE_DWORDX4                        0x12
+#define SQ_S_BUFFER_STORE_DWORD                   0x18
+#define SQ_S_BUFFER_STORE_DWORDX2                 0x19
+#define SQ_S_BUFFER_STORE_DWORDX4                 0x1a
+#define SQ_S_DCACHE_INV                           0x20
+#define SQ_S_DCACHE_WB                            0x21
+#define SQ_S_DCACHE_INV_VOL                       0x22
+#define SQ_S_DCACHE_WB_VOL                        0x23
+#define SQ_S_MEMTIME                              0x24
+#define SQ_S_MEMREALTIME                          0x25
+#define SQ_S_ATC_PROBE                            0x26
+#define SQ_S_ATC_PROBE_BUFFER                     0x27
+#define SQ_S_BUFFER_ATOMIC_SWAP                   0x40
+#define SQ_S_BUFFER_ATOMIC_CMPSWAP                0x41
+#define SQ_S_BUFFER_ATOMIC_ADD                    0x42
+#define SQ_S_BUFFER_ATOMIC_SUB                    0x43
+#define SQ_S_BUFFER_ATOMIC_SMIN                   0x44
+#define SQ_S_BUFFER_ATOMIC_UMIN                   0x45
+#define SQ_S_BUFFER_ATOMIC_SMAX                   0x46
+#define SQ_S_BUFFER_ATOMIC_UMAX                   0x47
+#define SQ_S_BUFFER_ATOMIC_AND                    0x48
+#define SQ_S_BUFFER_ATOMIC_OR                     0x49
+#define SQ_S_BUFFER_ATOMIC_XOR                    0x4a
+#define SQ_S_BUFFER_ATOMIC_INC                    0x4b
+#define SQ_S_BUFFER_ATOMIC_DEC                    0x4c
+#define SQ_S_BUFFER_ATOMIC_SWAP_X2                0x60
+#define SQ_S_BUFFER_ATOMIC_CMPSWAP_X2             0x61
+#define SQ_S_BUFFER_ATOMIC_ADD_X2                 0x62
+#define SQ_S_BUFFER_ATOMIC_SUB_X2                 0x63
+#define SQ_S_BUFFER_ATOMIC_SMIN_X2                0x64
+#define SQ_S_BUFFER_ATOMIC_UMIN_X2                0x65
+#define SQ_S_BUFFER_ATOMIC_SMAX_X2                0x66
+#define SQ_S_BUFFER_ATOMIC_UMAX_X2                0x67
+#define SQ_S_BUFFER_ATOMIC_AND_X2                 0x68
+#define SQ_S_BUFFER_ATOMIC_OR_X2                  0x69
+#define SQ_S_BUFFER_ATOMIC_XOR_X2                 0x6a
+#define SQ_S_BUFFER_ATOMIC_INC_X2                 0x6b
+#define SQ_S_BUFFER_ATOMIC_DEC_X2                 0x6c
+#define SQ_F                                      0x0
+#define SQ_LT                                     0x1
+#define SQ_EQ                                     0x2
+#define SQ_LE                                     0x3
+#define SQ_GT                                     0x4
+#define SQ_LG                                     0x5
+#define SQ_GE                                     0x6
+#define SQ_O                                      0x7
+#define SQ_U                                      0x8
+#define SQ_NGE                                    0x9
+#define SQ_NLG                                    0xa
+#define SQ_NGT                                    0xb
+#define SQ_NLE                                    0xc
+#define SQ_NEQ                                    0xd
+#define SQ_NLT                                    0xe
+#define SQ_TRU                                    0xf
+#define SQ_V_CMP_CLASS_F32                        0x10
+#define SQ_V_CMPX_CLASS_F32                       0x11
+#define SQ_V_CMP_CLASS_F64                        0x12
+#define SQ_V_CMPX_CLASS_F64                       0x13
+#define SQ_V_CMP_CLASS_F16                        0x14
+#define SQ_V_CMPX_CLASS_F16                       0x15
+#define SQ_V_CMP_F_F16                            0x20
+#define SQ_V_CMP_LT_F16                           0x21
+#define SQ_V_CMP_EQ_F16                           0x22
+#define SQ_V_CMP_LE_F16                           0x23
+#define SQ_V_CMP_GT_F16                           0x24
+#define SQ_V_CMP_LG_F16                           0x25
+#define SQ_V_CMP_GE_F16                           0x26
+#define SQ_V_CMP_O_F16                            0x27
+#define SQ_V_CMP_U_F16                            0x28
+#define SQ_V_CMP_NGE_F16                          0x29
+#define SQ_V_CMP_NLG_F16                          0x2a
+#define SQ_V_CMP_NGT_F16                          0x2b
+#define SQ_V_CMP_NLE_F16                          0x2c
+#define SQ_V_CMP_NEQ_F16                          0x2d
+#define SQ_V_CMP_NLT_F16                          0x2e
+#define SQ_V_CMP_TRU_F16                          0x2f
+#define SQ_V_CMPX_F_F16                           0x30
+#define SQ_V_CMPX_LT_F16                          0x31
+#define SQ_V_CMPX_EQ_F16                          0x32
+#define SQ_V_CMPX_LE_F16                          0x33
+#define SQ_V_CMPX_GT_F16                          0x34
+#define SQ_V_CMPX_LG_F16                          0x35
+#define SQ_V_CMPX_GE_F16                          0x36
+#define SQ_V_CMPX_O_F16                           0x37
+#define SQ_V_CMPX_U_F16                           0x38
+#define SQ_V_CMPX_NGE_F16                         0x39
+#define SQ_V_CMPX_NLG_F16                         0x3a
+#define SQ_V_CMPX_NGT_F16                         0x3b
+#define SQ_V_CMPX_NLE_F16                         0x3c
+#define SQ_V_CMPX_NEQ_F16                         0x3d
+#define SQ_V_CMPX_NLT_F16                         0x3e
+#define SQ_V_CMPX_TRU_F16                         0x3f
+#define SQ_V_CMP_F_F32                            0x40
+#define SQ_V_CMP_LT_F32                           0x41
+#define SQ_V_CMP_EQ_F32                           0x42
+#define SQ_V_CMP_LE_F32                           0x43
+#define SQ_V_CMP_GT_F32                           0x44
+#define SQ_V_CMP_LG_F32                           0x45
+#define SQ_V_CMP_GE_F32                           0x46
+#define SQ_V_CMP_O_F32                            0x47
+#define SQ_V_CMP_U_F32                            0x48
+#define SQ_V_CMP_NGE_F32                          0x49
+#define SQ_V_CMP_NLG_F32                          0x4a
+#define SQ_V_CMP_NGT_F32                          0x4b
+#define SQ_V_CMP_NLE_F32                          0x4c
+#define SQ_V_CMP_NEQ_F32                          0x4d
+#define SQ_V_CMP_NLT_F32                          0x4e
+#define SQ_V_CMP_TRU_F32                          0x4f
+#define SQ_V_CMPX_F_F32                           0x50
+#define SQ_V_CMPX_LT_F32                          0x51
+#define SQ_V_CMPX_EQ_F32                          0x52
+#define SQ_V_CMPX_LE_F32                          0x53
+#define SQ_V_CMPX_GT_F32                          0x54
+#define SQ_V_CMPX_LG_F32                          0x55
+#define SQ_V_CMPX_GE_F32                          0x56
+#define SQ_V_CMPX_O_F32                           0x57
+#define SQ_V_CMPX_U_F32                           0x58
+#define SQ_V_CMPX_NGE_F32                         0x59
+#define SQ_V_CMPX_NLG_F32                         0x5a
+#define SQ_V_CMPX_NGT_F32                         0x5b
+#define SQ_V_CMPX_NLE_F32                         0x5c
+#define SQ_V_CMPX_NEQ_F32                         0x5d
+#define SQ_V_CMPX_NLT_F32                         0x5e
+#define SQ_V_CMPX_TRU_F32                         0x5f
+#define SQ_V_CMP_F_F64                            0x60
+#define SQ_V_CMP_LT_F64                           0x61
+#define SQ_V_CMP_EQ_F64                           0x62
+#define SQ_V_CMP_LE_F64                           0x63
+#define SQ_V_CMP_GT_F64                           0x64
+#define SQ_V_CMP_LG_F64                           0x65
+#define SQ_V_CMP_GE_F64                           0x66
+#define SQ_V_CMP_O_F64                            0x67
+#define SQ_V_CMP_U_F64                            0x68
+#define SQ_V_CMP_NGE_F64                          0x69
+#define SQ_V_CMP_NLG_F64                          0x6a
+#define SQ_V_CMP_NGT_F64                          0x6b
+#define SQ_V_CMP_NLE_F64                          0x6c
+#define SQ_V_CMP_NEQ_F64                          0x6d
+#define SQ_V_CMP_NLT_F64                          0x6e
+#define SQ_V_CMP_TRU_F64                          0x6f
+#define SQ_V_CMPX_F_F64                           0x70
+#define SQ_V_CMPX_LT_F64                          0x71
+#define SQ_V_CMPX_EQ_F64                          0x72
+#define SQ_V_CMPX_LE_F64                          0x73
+#define SQ_V_CMPX_GT_F64                          0x74
+#define SQ_V_CMPX_LG_F64                          0x75
+#define SQ_V_CMPX_GE_F64                          0x76
+#define SQ_V_CMPX_O_F64                           0x77
+#define SQ_V_CMPX_U_F64                           0x78
+#define SQ_V_CMPX_NGE_F64                         0x79
+#define SQ_V_CMPX_NLG_F64                         0x7a
+#define SQ_V_CMPX_NGT_F64                         0x7b
+#define SQ_V_CMPX_NLE_F64                         0x7c
+#define SQ_V_CMPX_NEQ_F64                         0x7d
+#define SQ_V_CMPX_NLT_F64                         0x7e
+#define SQ_V_CMPX_TRU_F64                         0x7f
+#define SQ_V_CMP_F_I16                            0xa0
+#define SQ_V_CMP_LT_I16                           0xa1
+#define SQ_V_CMP_EQ_I16                           0xa2
+#define SQ_V_CMP_LE_I16                           0xa3
+#define SQ_V_CMP_GT_I16                           0xa4
+#define SQ_V_CMP_NE_I16                           0xa5
+#define SQ_V_CMP_GE_I16                           0xa6
+#define SQ_V_CMP_T_I16                            0xa7
+#define SQ_V_CMP_F_U16                            0xa8
+#define SQ_V_CMP_LT_U16                           0xa9
+#define SQ_V_CMP_EQ_U16                           0xaa
+#define SQ_V_CMP_LE_U16                           0xab
+#define SQ_V_CMP_GT_U16                           0xac
+#define SQ_V_CMP_NE_U16                           0xad
+#define SQ_V_CMP_GE_U16                           0xae
+#define SQ_V_CMP_T_U16                            0xaf
+#define SQ_V_CMPX_F_I16                           0xb0
+#define SQ_V_CMPX_LT_I16                          0xb1
+#define SQ_V_CMPX_EQ_I16                          0xb2
+#define SQ_V_CMPX_LE_I16                          0xb3
+#define SQ_V_CMPX_GT_I16                          0xb4
+#define SQ_V_CMPX_NE_I16                          0xb5
+#define SQ_V_CMPX_GE_I16                          0xb6
+#define SQ_V_CMPX_T_I16                           0xb7
+#define SQ_V_CMPX_F_U16                           0xb8
+#define SQ_V_CMPX_LT_U16                          0xb9
+#define SQ_V_CMPX_EQ_U16                          0xba
+#define SQ_V_CMPX_LE_U16                          0xbb
+#define SQ_V_CMPX_GT_U16                          0xbc
+#define SQ_V_CMPX_NE_U16                          0xbd
+#define SQ_V_CMPX_GE_U16                          0xbe
+#define SQ_V_CMPX_T_U16                           0xbf
+#define SQ_V_CMP_F_I32                            0xc0
+#define SQ_V_CMP_LT_I32                           0xc1
+#define SQ_V_CMP_EQ_I32                           0xc2
+#define SQ_V_CMP_LE_I32                           0xc3
+#define SQ_V_CMP_GT_I32                           0xc4
+#define SQ_V_CMP_NE_I32                           0xc5
+#define SQ_V_CMP_GE_I32                           0xc6
+#define SQ_V_CMP_T_I32                            0xc7
+#define SQ_V_CMP_F_U32                            0xc8
+#define SQ_V_CMP_LT_U32                           0xc9
+#define SQ_V_CMP_EQ_U32                           0xca
+#define SQ_V_CMP_LE_U32                           0xcb
+#define SQ_V_CMP_GT_U32                           0xcc
+#define SQ_V_CMP_NE_U32                           0xcd
+#define SQ_V_CMP_GE_U32                           0xce
+#define SQ_V_CMP_T_U32                            0xcf
+#define SQ_V_CMPX_F_I32                           0xd0
+#define SQ_V_CMPX_LT_I32                          0xd1
+#define SQ_V_CMPX_EQ_I32                          0xd2
+#define SQ_V_CMPX_LE_I32                          0xd3
+#define SQ_V_CMPX_GT_I32                          0xd4
+#define SQ_V_CMPX_NE_I32                          0xd5
+#define SQ_V_CMPX_GE_I32                          0xd6
+#define SQ_V_CMPX_T_I32                           0xd7
+#define SQ_V_CMPX_F_U32                           0xd8
+#define SQ_V_CMPX_LT_U32                          0xd9
+#define SQ_V_CMPX_EQ_U32                          0xda
+#define SQ_V_CMPX_LE_U32                          0xdb
+#define SQ_V_CMPX_GT_U32                          0xdc
+#define SQ_V_CMPX_NE_U32                          0xdd
+#define SQ_V_CMPX_GE_U32                          0xde
+#define SQ_V_CMPX_T_U32                           0xdf
+#define SQ_V_CMP_F_I64                            0xe0
+#define SQ_V_CMP_LT_I64                           0xe1
+#define SQ_V_CMP_EQ_I64                           0xe2
+#define SQ_V_CMP_LE_I64                           0xe3
+#define SQ_V_CMP_GT_I64                           0xe4
+#define SQ_V_CMP_NE_I64                           0xe5
+#define SQ_V_CMP_GE_I64                           0xe6
+#define SQ_V_CMP_T_I64                            0xe7
+#define SQ_V_CMP_F_U64                            0xe8
+#define SQ_V_CMP_LT_U64                           0xe9
+#define SQ_V_CMP_EQ_U64                           0xea
+#define SQ_V_CMP_LE_U64                           0xeb
+#define SQ_V_CMP_GT_U64                           0xec
+#define SQ_V_CMP_NE_U64                           0xed
+#define SQ_V_CMP_GE_U64                           0xee
+#define SQ_V_CMP_T_U64                            0xef
+#define SQ_V_CMPX_F_I64                           0xf0
+#define SQ_V_CMPX_LT_I64                          0xf1
+#define SQ_V_CMPX_EQ_I64                          0xf2
+#define SQ_V_CMPX_LE_I64                          0xf3
+#define SQ_V_CMPX_GT_I64                          0xf4
+#define SQ_V_CMPX_NE_I64                          0xf5
+#define SQ_V_CMPX_GE_I64                          0xf6
+#define SQ_V_CMPX_T_I64                           0xf7
+#define SQ_V_CMPX_F_U64                           0xf8
+#define SQ_V_CMPX_LT_U64                          0xf9
+#define SQ_V_CMPX_EQ_U64                          0xfa
+#define SQ_V_CMPX_LE_U64                          0xfb
+#define SQ_V_CMPX_GT_U64                          0xfc
+#define SQ_V_CMPX_NE_U64                          0xfd
+#define SQ_V_CMPX_GE_U64                          0xfe
+#define SQ_V_CMPX_T_U64                           0xff
+#define SQ_L1                                     0x1
+#define SQ_L2                                     0x2
+#define SQ_L3                                     0x3
+#define SQ_L4                                     0x4
+#define SQ_L5                                     0x5
+#define SQ_L6                                     0x6
+#define SQ_L7                                     0x7
+#define SQ_L8                                     0x8
+#define SQ_L9                                     0x9
+#define SQ_L10                                    0xa
+#define SQ_L11                                    0xb
+#define SQ_L12                                    0xc
+#define SQ_L13                                    0xd
+#define SQ_L14                                    0xe
+#define SQ_L15                                    0xf
+#define SQ_SGPR0                                  0x0
+#define SQ_SDWA_UNUSED_PAD                        0x0
+#define SQ_SDWA_UNUSED_SEXT                       0x1
+#define SQ_SDWA_UNUSED_PRESERVE                   0x2
+#define SQ_F                                      0x0
+#define SQ_LT                                     0x1
+#define SQ_EQ                                     0x2
+#define SQ_LE                                     0x3
+#define SQ_GT                                     0x4
+#define SQ_NE                                     0x5
+#define SQ_GE                                     0x6
+#define SQ_T                                      0x7
+#define SQ_SRC_64_INT                             0xc0
+#define SQ_SRC_M_1_INT                            0xc1
+#define SQ_SRC_M_2_INT                            0xc2
+#define SQ_SRC_M_3_INT                            0xc3
+#define SQ_SRC_M_4_INT                            0xc4
+#define SQ_SRC_M_5_INT                            0xc5
+#define SQ_SRC_M_6_INT                            0xc6
+#define SQ_SRC_M_7_INT                            0xc7
+#define SQ_SRC_M_8_INT                            0xc8
+#define SQ_SRC_M_9_INT                            0xc9
+#define SQ_SRC_M_10_INT                           0xca
+#define SQ_SRC_M_11_INT                           0xcb
+#define SQ_SRC_M_12_INT                           0xcc
+#define SQ_SRC_M_13_INT                           0xcd
+#define SQ_SRC_M_14_INT                           0xce
+#define SQ_SRC_M_15_INT                           0xcf
+#define SQ_SRC_M_16_INT                           0xd0
+#define SQ_SRC_0_5                                0xf0
+#define SQ_SRC_M_0_5                              0xf1
+#define SQ_SRC_1                                  0xf2
+#define SQ_SRC_M_1                                0xf3
+#define SQ_SRC_2                                  0xf4
+#define SQ_SRC_M_2                                0xf5
+#define SQ_SRC_4                                  0xf6
+#define SQ_SRC_M_4                                0xf7
+#define SQ_SRC_INV_2PI                            0xf8
+#define SQ_SRC_0                                  0x80
+#define SQ_SRC_1_INT                              0x81
+#define SQ_SRC_2_INT                              0x82
+#define SQ_SRC_3_INT                              0x83
+#define SQ_SRC_4_INT                              0x84
+#define SQ_SRC_5_INT                              0x85
+#define SQ_SRC_6_INT                              0x86
+#define SQ_SRC_7_INT                              0x87
+#define SQ_SRC_8_INT                              0x88
+#define SQ_SRC_9_INT                              0x89
+#define SQ_SRC_10_INT                             0x8a
+#define SQ_SRC_11_INT                             0x8b
+#define SQ_SRC_12_INT                             0x8c
+#define SQ_SRC_13_INT                             0x8d
+#define SQ_SRC_14_INT                             0x8e
+#define SQ_SRC_15_INT                             0x8f
+#define SQ_SRC_16_INT                             0x90
+#define SQ_SRC_17_INT                             0x91
+#define SQ_SRC_18_INT                             0x92
+#define SQ_SRC_19_INT                             0x93
+#define SQ_SRC_20_INT                             0x94
+#define SQ_SRC_21_INT                             0x95
+#define SQ_SRC_22_INT                             0x96
+#define SQ_SRC_23_INT                             0x97
+#define SQ_SRC_24_INT                             0x98
+#define SQ_SRC_25_INT                             0x99
+#define SQ_SRC_26_INT                             0x9a
+#define SQ_SRC_27_INT                             0x9b
+#define SQ_SRC_28_INT                             0x9c
+#define SQ_SRC_29_INT                             0x9d
+#define SQ_SRC_30_INT                             0x9e
+#define SQ_SRC_31_INT                             0x9f
+#define SQ_SRC_32_INT                             0xa0
+#define SQ_SRC_33_INT                             0xa1
+#define SQ_SRC_34_INT                             0xa2
+#define SQ_SRC_35_INT                             0xa3
+#define SQ_SRC_36_INT                             0xa4
+#define SQ_SRC_37_INT                             0xa5
+#define SQ_SRC_38_INT                             0xa6
+#define SQ_SRC_39_INT                             0xa7
+#define SQ_SRC_40_INT                             0xa8
+#define SQ_SRC_41_INT                             0xa9
+#define SQ_SRC_42_INT                             0xaa
+#define SQ_SRC_43_INT                             0xab
+#define SQ_SRC_44_INT                             0xac
+#define SQ_SRC_45_INT                             0xad
+#define SQ_SRC_46_INT                             0xae
+#define SQ_SRC_47_INT                             0xaf
+#define SQ_SRC_48_INT                             0xb0
+#define SQ_SRC_49_INT                             0xb1
+#define SQ_SRC_50_INT                             0xb2
+#define SQ_SRC_51_INT                             0xb3
+#define SQ_SRC_52_INT                             0xb4
+#define SQ_SRC_53_INT                             0xb5
+#define SQ_SRC_54_INT                             0xb6
+#define SQ_SRC_55_INT                             0xb7
+#define SQ_SRC_56_INT                             0xb8
+#define SQ_SRC_57_INT                             0xb9
+#define SQ_SRC_58_INT                             0xba
+#define SQ_SRC_59_INT                             0xbb
+#define SQ_SRC_60_INT                             0xbc
+#define SQ_SRC_61_INT                             0xbd
+#define SQ_SRC_62_INT                             0xbe
+#define SQ_SRC_63_INT                             0xbf
+#define SQ_DS_ADD_U32                             0x0
+#define SQ_DS_SUB_U32                             0x1
+#define SQ_DS_RSUB_U32                            0x2
+#define SQ_DS_INC_U32                             0x3
+#define SQ_DS_DEC_U32                             0x4
+#define SQ_DS_MIN_I32                             0x5
+#define SQ_DS_MAX_I32                             0x6
+#define SQ_DS_MIN_U32                             0x7
+#define SQ_DS_MAX_U32                             0x8
+#define SQ_DS_AND_B32                             0x9
+#define SQ_DS_OR_B32                              0xa
+#define SQ_DS_XOR_B32                             0xb
+#define SQ_DS_MSKOR_B32                           0xc
+#define SQ_DS_WRITE_B32                           0xd
+#define SQ_DS_WRITE2_B32                          0xe
+#define SQ_DS_WRITE2ST64_B32                      0xf
+#define SQ_DS_CMPST_B32                           0x10
+#define SQ_DS_CMPST_F32                           0x11
+#define SQ_DS_MIN_F32                             0x12
+#define SQ_DS_MAX_F32                             0x13
+#define SQ_DS_NOP                                 0x14
+#define SQ_DS_ADD_F32                             0x15
+#define SQ_DS_WRITE_B8                            0x1e
+#define SQ_DS_WRITE_B16                           0x1f
+#define SQ_DS_ADD_RTN_U32                         0x20
+#define SQ_DS_SUB_RTN_U32                         0x21
+#define SQ_DS_RSUB_RTN_U32                        0x22
+#define SQ_DS_INC_RTN_U32                         0x23
+#define SQ_DS_DEC_RTN_U32                         0x24
+#define SQ_DS_MIN_RTN_I32                         0x25
+#define SQ_DS_MAX_RTN_I32                         0x26
+#define SQ_DS_MIN_RTN_U32                         0x27
+#define SQ_DS_MAX_RTN_U32                         0x28
+#define SQ_DS_AND_RTN_B32                         0x29
+#define SQ_DS_OR_RTN_B32                          0x2a
+#define SQ_DS_XOR_RTN_B32                         0x2b
+#define SQ_DS_MSKOR_RTN_B32                       0x2c
+#define SQ_DS_WRXCHG_RTN_B32                      0x2d
+#define SQ_DS_WRXCHG2_RTN_B32                     0x2e
+#define SQ_DS_WRXCHG2ST64_RTN_B32                 0x2f
+#define SQ_DS_CMPST_RTN_B32                       0x30
+#define SQ_DS_CMPST_RTN_F32                       0x31
+#define SQ_DS_MIN_RTN_F32                         0x32
+#define SQ_DS_MAX_RTN_F32                         0x33
+#define SQ_DS_WRAP_RTN_B32                        0x34
+#define SQ_DS_ADD_RTN_F32                         0x35
+#define SQ_DS_READ_B32                            0x36
+#define SQ_DS_READ2_B32                           0x37
+#define SQ_DS_READ2ST64_B32                       0x38
+#define SQ_DS_READ_I8                             0x39
+#define SQ_DS_READ_U8                             0x3a
+#define SQ_DS_READ_I16                            0x3b
+#define SQ_DS_READ_U16                            0x3c
+#define SQ_DS_SWIZZLE_B32                         0x3d
+#define SQ_DS_PERMUTE_B32                         0x3e
+#define SQ_DS_BPERMUTE_B32                        0x3f
+#define SQ_DS_ADD_U64                             0x40
+#define SQ_DS_SUB_U64                             0x41
+#define SQ_DS_RSUB_U64                            0x42
+#define SQ_DS_INC_U64                             0x43
+#define SQ_DS_DEC_U64                             0x44
+#define SQ_DS_MIN_I64                             0x45
+#define SQ_DS_MAX_I64                             0x46
+#define SQ_DS_MIN_U64                             0x47
+#define SQ_DS_MAX_U64                             0x48
+#define SQ_DS_AND_B64                             0x49
+#define SQ_DS_OR_B64                              0x4a
+#define SQ_DS_XOR_B64                             0x4b
+#define SQ_DS_MSKOR_B64                           0x4c
+#define SQ_DS_WRITE_B64                           0x4d
+#define SQ_DS_WRITE2_B64                          0x4e
+#define SQ_DS_WRITE2ST64_B64                      0x4f
+#define SQ_DS_CMPST_B64                           0x50
+#define SQ_DS_CMPST_F64                           0x51
+#define SQ_DS_MIN_F64                             0x52
+#define SQ_DS_MAX_F64                             0x53
+#define SQ_DS_ADD_RTN_U64                         0x60
+#define SQ_DS_SUB_RTN_U64                         0x61
+#define SQ_DS_RSUB_RTN_U64                        0x62
+#define SQ_DS_INC_RTN_U64                         0x63
+#define SQ_DS_DEC_RTN_U64                         0x64
+#define SQ_DS_MIN_RTN_I64                         0x65
+#define SQ_DS_MAX_RTN_I64                         0x66
+#define SQ_DS_MIN_RTN_U64                         0x67
+#define SQ_DS_MAX_RTN_U64                         0x68
+#define SQ_DS_AND_RTN_B64                         0x69
+#define SQ_DS_OR_RTN_B64                          0x6a
+#define SQ_DS_XOR_RTN_B64                         0x6b
+#define SQ_DS_MSKOR_RTN_B64                       0x6c
+#define SQ_DS_WRXCHG_RTN_B64                      0x6d
+#define SQ_DS_WRXCHG2_RTN_B64                     0x6e
+#define SQ_DS_WRXCHG2ST64_RTN_B64                 0x6f
+#define SQ_DS_CMPST_RTN_B64                       0x70
+#define SQ_DS_CMPST_RTN_F64                       0x71
+#define SQ_DS_MIN_RTN_F64                         0x72
+#define SQ_DS_MAX_RTN_F64                         0x73
+#define SQ_DS_READ_B64                            0x76
+#define SQ_DS_READ2_B64                           0x77
+#define SQ_DS_READ2ST64_B64                       0x78
+#define SQ_DS_CONDXCHG32_RTN_B64                  0x7e
+#define SQ_DS_ADD_SRC2_U32                        0x80
+#define SQ_DS_SUB_SRC2_U32                        0x81
+#define SQ_DS_RSUB_SRC2_U32                       0x82
+#define SQ_DS_INC_SRC2_U32                        0x83
+#define SQ_DS_DEC_SRC2_U32                        0x84
+#define SQ_DS_MIN_SRC2_I32                        0x85
+#define SQ_DS_MAX_SRC2_I32                        0x86
+#define SQ_DS_MIN_SRC2_U32                        0x87
+#define SQ_DS_MAX_SRC2_U32                        0x88
+#define SQ_DS_AND_SRC2_B32                        0x89
+#define SQ_DS_OR_SRC2_B32                         0x8a
+#define SQ_DS_XOR_SRC2_B32                        0x8b
+#define SQ_DS_WRITE_SRC2_B32                      0x8d
+#define SQ_DS_MIN_SRC2_F32                        0x92
+#define SQ_DS_MAX_SRC2_F32                        0x93
+#define SQ_DS_ADD_SRC2_F32                        0x95
+#define SQ_DS_GWS_SEMA_RELEASE_ALL                0x98
+#define SQ_DS_GWS_INIT                            0x99
+#define SQ_DS_GWS_SEMA_V                          0x9a
+#define SQ_DS_GWS_SEMA_BR                         0x9b
+#define SQ_DS_GWS_SEMA_P                          0x9c
+#define SQ_DS_GWS_BARRIER                         0x9d
+#define SQ_DS_CONSUME                             0xbd
+#define SQ_DS_APPEND                              0xbe
+#define SQ_DS_ORDERED_COUNT                       0xbf
+#define SQ_DS_ADD_SRC2_U64                        0xc0
+#define SQ_DS_SUB_SRC2_U64                        0xc1
+#define SQ_DS_RSUB_SRC2_U64                       0xc2
+#define SQ_DS_INC_SRC2_U64                        0xc3
+#define SQ_DS_DEC_SRC2_U64                        0xc4
+#define SQ_DS_MIN_SRC2_I64                        0xc5
+#define SQ_DS_MAX_SRC2_I64                        0xc6
+#define SQ_DS_MIN_SRC2_U64                        0xc7
+#define SQ_DS_MAX_SRC2_U64                        0xc8
+#define SQ_DS_AND_SRC2_B64                        0xc9
+#define SQ_DS_OR_SRC2_B64                         0xca
+#define SQ_DS_XOR_SRC2_B64                        0xcb
+#define SQ_DS_WRITE_SRC2_B64                      0xcd
+#define SQ_DS_MIN_SRC2_F64                        0xd2
+#define SQ_DS_MAX_SRC2_F64                        0xd3
+#define SQ_DS_WRITE_B96                           0xde
+#define SQ_DS_WRITE_B128                          0xdf
+#define SQ_DS_CONDXCHG32_RTN_B128                 0xfd
+#define SQ_DS_READ_B96                            0xfe
+#define SQ_DS_READ_B128                           0xff
+#define SQ_BUFFER_LOAD_FORMAT_X                   0x0
+#define SQ_BUFFER_LOAD_FORMAT_XY                  0x1
+#define SQ_BUFFER_LOAD_FORMAT_XYZ                 0x2
+#define SQ_BUFFER_LOAD_FORMAT_XYZW                0x3
+#define SQ_BUFFER_STORE_FORMAT_X                  0x4
+#define SQ_BUFFER_STORE_FORMAT_XY                 0x5
+#define SQ_BUFFER_STORE_FORMAT_XYZ                0x6
+#define SQ_BUFFER_STORE_FORMAT_XYZW               0x7
+#define SQ_BUFFER_LOAD_FORMAT_D16_X               0x8
+#define SQ_BUFFER_LOAD_FORMAT_D16_XY              0x9
+#define SQ_BUFFER_LOAD_FORMAT_D16_XYZ             0xa
+#define SQ_BUFFER_LOAD_FORMAT_D16_XYZW            0xb
+#define SQ_BUFFER_STORE_FORMAT_D16_X              0xc
+#define SQ_BUFFER_STORE_FORMAT_D16_XY             0xd
+#define SQ_BUFFER_STORE_FORMAT_D16_XYZ            0xe
+#define SQ_BUFFER_STORE_FORMAT_D16_XYZW           0xf
+#define SQ_BUFFER_LOAD_UBYTE                      0x10
+#define SQ_BUFFER_LOAD_SBYTE                      0x11
+#define SQ_BUFFER_LOAD_USHORT                     0x12
+#define SQ_BUFFER_LOAD_SSHORT                     0x13
+#define SQ_BUFFER_LOAD_DWORD                      0x14
+#define SQ_BUFFER_LOAD_DWORDX2                    0x15
+#define SQ_BUFFER_LOAD_DWORDX3                    0x16
+#define SQ_BUFFER_LOAD_DWORDX4                    0x17
+#define SQ_BUFFER_STORE_BYTE                      0x18
+#define SQ_BUFFER_STORE_SHORT                     0x1a
+#define SQ_BUFFER_STORE_DWORD                     0x1c
+#define SQ_BUFFER_STORE_DWORDX2                   0x1d
+#define SQ_BUFFER_STORE_DWORDX3                   0x1e
+#define SQ_BUFFER_STORE_DWORDX4                   0x1f
+#define SQ_BUFFER_STORE_LDS_DWORD                 0x3d
+#define SQ_BUFFER_WBINVL1                         0x3e
+#define SQ_BUFFER_WBINVL1_VOL                     0x3f
+#define SQ_BUFFER_ATOMIC_SWAP                     0x40
+#define SQ_BUFFER_ATOMIC_CMPSWAP                  0x41
+#define SQ_BUFFER_ATOMIC_ADD                      0x42
+#define SQ_BUFFER_ATOMIC_SUB                      0x43
+#define SQ_BUFFER_ATOMIC_SMIN                     0x44
+#define SQ_BUFFER_ATOMIC_UMIN                     0x45
+#define SQ_BUFFER_ATOMIC_SMAX                     0x46
+#define SQ_BUFFER_ATOMIC_UMAX                     0x47
+#define SQ_BUFFER_ATOMIC_AND                      0x48
+#define SQ_BUFFER_ATOMIC_OR                       0x49
+#define SQ_BUFFER_ATOMIC_XOR                      0x4a
+#define SQ_BUFFER_ATOMIC_INC                      0x4b
+#define SQ_BUFFER_ATOMIC_DEC                      0x4c
+#define SQ_BUFFER_ATOMIC_SWAP_X2                  0x60
+#define SQ_BUFFER_ATOMIC_CMPSWAP_X2               0x61
+#define SQ_BUFFER_ATOMIC_ADD_X2                   0x62
+#define SQ_BUFFER_ATOMIC_SUB_X2                   0x63
+#define SQ_BUFFER_ATOMIC_SMIN_X2                  0x64
+#define SQ_BUFFER_ATOMIC_UMIN_X2                  0x65
+#define SQ_BUFFER_ATOMIC_SMAX_X2                  0x66
+#define SQ_BUFFER_ATOMIC_UMAX_X2                  0x67
+#define SQ_BUFFER_ATOMIC_AND_X2                   0x68
+#define SQ_BUFFER_ATOMIC_OR_X2                    0x69
+#define SQ_BUFFER_ATOMIC_XOR_X2                   0x6a
+#define SQ_BUFFER_ATOMIC_INC_X2                   0x6b
+#define SQ_BUFFER_ATOMIC_DEC_X2                   0x6c
+#define SQ_EXEC_LO                                0x7e
+#define SQ_EXEC_HI                                0x7f
+#define SQ_SRC_SCC                                0xfd
+#define SQ_OMOD_OFF                               0x0
+#define SQ_OMOD_M2                                0x1
+#define SQ_OMOD_M4                                0x2
+#define SQ_OMOD_D2                                0x3
+#define SQ_DPP_QUAD_PERM                          0x0
+#define SQ_DPP_ROW_SL1                            0x101
+#define SQ_DPP_ROW_SL2                            0x102
+#define SQ_DPP_ROW_SL3                            0x103
+#define SQ_DPP_ROW_SL4                            0x104
+#define SQ_DPP_ROW_SL5                            0x105
+#define SQ_DPP_ROW_SL6                            0x106
+#define SQ_DPP_ROW_SL7                            0x107
+#define SQ_DPP_ROW_SL8                            0x108
+#define SQ_DPP_ROW_SL9                            0x109
+#define SQ_DPP_ROW_SL10                           0x10a
+#define SQ_DPP_ROW_SL11                           0x10b
+#define SQ_DPP_ROW_SL12                           0x10c
+#define SQ_DPP_ROW_SL13                           0x10d
+#define SQ_DPP_ROW_SL14                           0x10e
+#define SQ_DPP_ROW_SL15                           0x10f
+#define SQ_DPP_ROW_SR1                            0x111
+#define SQ_DPP_ROW_SR2                            0x112
+#define SQ_DPP_ROW_SR3                            0x113
+#define SQ_DPP_ROW_SR4                            0x114
+#define SQ_DPP_ROW_SR5                            0x115
+#define SQ_DPP_ROW_SR6                            0x116
+#define SQ_DPP_ROW_SR7                            0x117
+#define SQ_DPP_ROW_SR8                            0x118
+#define SQ_DPP_ROW_SR9                            0x119
+#define SQ_DPP_ROW_SR10                           0x11a
+#define SQ_DPP_ROW_SR11                           0x11b
+#define SQ_DPP_ROW_SR12                           0x11c
+#define SQ_DPP_ROW_SR13                           0x11d
+#define SQ_DPP_ROW_SR14                           0x11e
+#define SQ_DPP_ROW_SR15                           0x11f
+#define SQ_DPP_ROW_RR1                            0x121
+#define SQ_DPP_ROW_RR2                            0x122
+#define SQ_DPP_ROW_RR3                            0x123
+#define SQ_DPP_ROW_RR4                            0x124
+#define SQ_DPP_ROW_RR5                            0x125
+#define SQ_DPP_ROW_RR6                            0x126
+#define SQ_DPP_ROW_RR7                            0x127
+#define SQ_DPP_ROW_RR8                            0x128
+#define SQ_DPP_ROW_RR9                            0x129
+#define SQ_DPP_ROW_RR10                           0x12a
+#define SQ_DPP_ROW_RR11                           0x12b
+#define SQ_DPP_ROW_RR12                           0x12c
+#define SQ_DPP_ROW_RR13                           0x12d
+#define SQ_DPP_ROW_RR14                           0x12e
+#define SQ_DPP_ROW_RR15                           0x12f
+#define SQ_DPP_WF_SL1                             0x130
+#define SQ_DPP_WF_RL1                             0x134
+#define SQ_DPP_WF_SR1                             0x138
+#define SQ_DPP_WF_RR1                             0x13c
+#define SQ_DPP_ROW_MIRROR                         0x140
+#define SQ_DPP_ROW_HALF_MIRROR                    0x141
+#define SQ_DPP_ROW_BCAST15                        0x142
+#define SQ_DPP_ROW_BCAST31                        0x143
+#define SQ_EXP_GDS0                               0x18
+#define SQ_GS_OP_NOP                              0x0
+#define SQ_GS_OP_CUT                              0x1
+#define SQ_GS_OP_EMIT                             0x2
+#define SQ_GS_OP_EMIT_CUT                         0x3
+#define SQ_IMAGE_LOAD                             0x0
+#define SQ_IMAGE_LOAD_MIP                         0x1
+#define SQ_IMAGE_LOAD_PCK                         0x2
+#define SQ_IMAGE_LOAD_PCK_SGN                     0x3
+#define SQ_IMAGE_LOAD_MIP_PCK                     0x4
+#define SQ_IMAGE_LOAD_MIP_PCK_SGN                 0x5
+#define SQ_IMAGE_STORE                            0x8
+#define SQ_IMAGE_STORE_MIP                        0x9
+#define SQ_IMAGE_STORE_PCK                        0xa
+#define SQ_IMAGE_STORE_MIP_PCK                    0xb
+#define SQ_IMAGE_GET_RESINFO                      0xe
+#define SQ_IMAGE_ATOMIC_SWAP                      0x10
+#define SQ_IMAGE_ATOMIC_CMPSWAP                   0x11
+#define SQ_IMAGE_ATOMIC_ADD                       0x12
+#define SQ_IMAGE_ATOMIC_SUB                       0x13
+#define SQ_IMAGE_ATOMIC_SMIN                      0x14
+#define SQ_IMAGE_ATOMIC_UMIN                      0x15
+#define SQ_IMAGE_ATOMIC_SMAX                      0x16
+#define SQ_IMAGE_ATOMIC_UMAX                      0x17
+#define SQ_IMAGE_ATOMIC_AND                       0x18
+#define SQ_IMAGE_ATOMIC_OR                        0x19
+#define SQ_IMAGE_ATOMIC_XOR                       0x1a
+#define SQ_IMAGE_ATOMIC_INC                       0x1b
+#define SQ_IMAGE_ATOMIC_DEC                       0x1c
+#define SQ_IMAGE_SAMPLE                           0x20
+#define SQ_IMAGE_SAMPLE_CL                        0x21
+#define SQ_IMAGE_SAMPLE_D                         0x22
+#define SQ_IMAGE_SAMPLE_D_CL                      0x23
+#define SQ_IMAGE_SAMPLE_L                         0x24
+#define SQ_IMAGE_SAMPLE_B                         0x25
+#define SQ_IMAGE_SAMPLE_B_CL                      0x26
+#define SQ_IMAGE_SAMPLE_LZ                        0x27
+#define SQ_IMAGE_SAMPLE_C                         0x28
+#define SQ_IMAGE_SAMPLE_C_CL                      0x29
+#define SQ_IMAGE_SAMPLE_C_D                       0x2a
+#define SQ_IMAGE_SAMPLE_C_D_CL                    0x2b
+#define SQ_IMAGE_SAMPLE_C_L                       0x2c
+#define SQ_IMAGE_SAMPLE_C_B                       0x2d
+#define SQ_IMAGE_SAMPLE_C_B_CL                    0x2e
+#define SQ_IMAGE_SAMPLE_C_LZ                      0x2f
+#define SQ_IMAGE_SAMPLE_O                         0x30
+#define SQ_IMAGE_SAMPLE_CL_O                      0x31
+#define SQ_IMAGE_SAMPLE_D_O                       0x32
+#define SQ_IMAGE_SAMPLE_D_CL_O                    0x33
+#define SQ_IMAGE_SAMPLE_L_O                       0x34
+#define SQ_IMAGE_SAMPLE_B_O                       0x35
+#define SQ_IMAGE_SAMPLE_B_CL_O                    0x36
+#define SQ_IMAGE_SAMPLE_LZ_O                      0x37
+#define SQ_IMAGE_SAMPLE_C_O                       0x38
+#define SQ_IMAGE_SAMPLE_C_CL_O                    0x39
+#define SQ_IMAGE_SAMPLE_C_D_O                     0x3a
+#define SQ_IMAGE_SAMPLE_C_D_CL_O                  0x3b
+#define SQ_IMAGE_SAMPLE_C_L_O                     0x3c
+#define SQ_IMAGE_SAMPLE_C_B_O                     0x3d
+#define SQ_IMAGE_SAMPLE_C_B_CL_O                  0x3e
+#define SQ_IMAGE_SAMPLE_C_LZ_O                    0x3f
+#define SQ_IMAGE_GATHER4                          0x40
+#define SQ_IMAGE_GATHER4_CL                       0x41
+#define SQ_IMAGE_GATHER4_L                        0x44
+#define SQ_IMAGE_GATHER4_B                        0x45
+#define SQ_IMAGE_GATHER4_B_CL                     0x46
+#define SQ_IMAGE_GATHER4_LZ                       0x47
+#define SQ_IMAGE_GATHER4_C                        0x48
+#define SQ_IMAGE_GATHER4_C_CL                     0x49
+#define SQ_IMAGE_GATHER4_C_L                      0x4c
+#define SQ_IMAGE_GATHER4_C_B                      0x4d
+#define SQ_IMAGE_GATHER4_C_B_CL                   0x4e
+#define SQ_IMAGE_GATHER4_C_LZ                     0x4f
+#define SQ_IMAGE_GATHER4_O                        0x50
+#define SQ_IMAGE_GATHER4_CL_O                     0x51
+#define SQ_IMAGE_GATHER4_L_O                      0x54
+#define SQ_IMAGE_GATHER4_B_O                      0x55
+#define SQ_IMAGE_GATHER4_B_CL_O                   0x56
+#define SQ_IMAGE_GATHER4_LZ_O                     0x57
+#define SQ_IMAGE_GATHER4_C_O                      0x58
+#define SQ_IMAGE_GATHER4_C_CL_O                   0x59
+#define SQ_IMAGE_GATHER4_C_L_O                    0x5c
+#define SQ_IMAGE_GATHER4_C_B_O                    0x5d
+#define SQ_IMAGE_GATHER4_C_B_CL_O                 0x5e
+#define SQ_IMAGE_GATHER4_C_LZ_O                   0x5f
+#define SQ_IMAGE_GET_LOD                          0x60
+#define SQ_IMAGE_SAMPLE_CD                        0x68
+#define SQ_IMAGE_SAMPLE_CD_CL                     0x69
+#define SQ_IMAGE_SAMPLE_C_CD                      0x6a
+#define SQ_IMAGE_SAMPLE_C_CD_CL                   0x6b
+#define SQ_IMAGE_SAMPLE_CD_O                      0x6c
+#define SQ_IMAGE_SAMPLE_CD_CL_O                   0x6d
+#define SQ_IMAGE_SAMPLE_C_CD_O                    0x6e
+#define SQ_IMAGE_SAMPLE_C_CD_CL_O                 0x6f
+#define SQ_IMAGE_RSRC256                          0x7e
+#define SQ_IMAGE_SAMPLER                          0x7f
+#define SQ_SRC_VCCZ                               0xfb
+#define SQ_SRC_VGPR0                              0x100
+#define SQ_SDWA_BYTE_0                            0x0
+#define SQ_SDWA_BYTE_1                            0x1
+#define SQ_SDWA_BYTE_2                            0x2
+#define SQ_SDWA_BYTE_3                            0x3
+#define SQ_SDWA_WORD_0                            0x4
+#define SQ_SDWA_WORD_1                            0x5
+#define SQ_SDWA_DWORD                             0x6
+#define SQ_XNACK_MASK_LO                          0x68
+#define SQ_XNACK_MASK_HI                          0x69
+#define SQ_TBUFFER_LOAD_FORMAT_X                  0x0
+#define SQ_TBUFFER_LOAD_FORMAT_XY                 0x1
+#define SQ_TBUFFER_LOAD_FORMAT_XYZ                0x2
+#define SQ_TBUFFER_LOAD_FORMAT_XYZW               0x3
+#define SQ_TBUFFER_STORE_FORMAT_X                 0x4
+#define SQ_TBUFFER_STORE_FORMAT_XY                0x5
+#define SQ_TBUFFER_STORE_FORMAT_XYZ               0x6
+#define SQ_TBUFFER_STORE_FORMAT_XYZW              0x7
+#define SQ_TBUFFER_LOAD_FORMAT_D16_X              0x8
+#define SQ_TBUFFER_LOAD_FORMAT_D16_XY             0x9
+#define SQ_TBUFFER_LOAD_FORMAT_D16_XYZ            0xa
+#define SQ_TBUFFER_LOAD_FORMAT_D16_XYZW           0xb
+#define SQ_TBUFFER_STORE_FORMAT_D16_X             0xc
+#define SQ_TBUFFER_STORE_FORMAT_D16_XY            0xd
+#define SQ_TBUFFER_STORE_FORMAT_D16_XYZ           0xe
+#define SQ_TBUFFER_STORE_FORMAT_D16_XYZW          0xf
+#define SQ_CHAN_X                                 0x0
+#define SQ_CHAN_Y                                 0x1
+#define SQ_CHAN_Z                                 0x2
+#define SQ_CHAN_W                                 0x3
+#define SQ_V_NOP                                  0x0
+#define SQ_V_MOV_B32                              0x1
+#define SQ_V_READFIRSTLANE_B32                    0x2
+#define SQ_V_CVT_I32_F64                          0x3
+#define SQ_V_CVT_F64_I32                          0x4
+#define SQ_V_CVT_F32_I32                          0x5
+#define SQ_V_CVT_F32_U32                          0x6
+#define SQ_V_CVT_U32_F32                          0x7
+#define SQ_V_CVT_I32_F32                          0x8
+#define SQ_V_MOV_FED_B32                          0x9
+#define SQ_V_CVT_F16_F32                          0xa
+#define SQ_V_CVT_F32_F16                          0xb
+#define SQ_V_CVT_RPI_I32_F32                      0xc
+#define SQ_V_CVT_FLR_I32_F32                      0xd
+#define SQ_V_CVT_OFF_F32_I4                       0xe
+#define SQ_V_CVT_F32_F64                          0xf
+#define SQ_V_CVT_F64_F32                          0x10
+#define SQ_V_CVT_F32_UBYTE0                       0x11
+#define SQ_V_CVT_F32_UBYTE1                       0x12
+#define SQ_V_CVT_F32_UBYTE2                       0x13
+#define SQ_V_CVT_F32_UBYTE3                       0x14
+#define SQ_V_CVT_U32_F64                          0x15
+#define SQ_V_CVT_F64_U32                          0x16
+#define SQ_V_TRUNC_F64                            0x17
+#define SQ_V_CEIL_F64                             0x18
+#define SQ_V_RNDNE_F64                            0x19
+#define SQ_V_FLOOR_F64                            0x1a
+#define SQ_V_FRACT_F32                            0x1b
+#define SQ_V_TRUNC_F32                            0x1c
+#define SQ_V_CEIL_F32                             0x1d
+#define SQ_V_RNDNE_F32                            0x1e
+#define SQ_V_FLOOR_F32                            0x1f
+#define SQ_V_EXP_F32                              0x20
+#define SQ_V_LOG_F32                              0x21
+#define SQ_V_RCP_F32                              0x22
+#define SQ_V_RCP_IFLAG_F32                        0x23
+#define SQ_V_RSQ_F32                              0x24
+#define SQ_V_RCP_F64                              0x25
+#define SQ_V_RSQ_F64                              0x26
+#define SQ_V_SQRT_F32                             0x27
+#define SQ_V_SQRT_F64                             0x28
+#define SQ_V_SIN_F32                              0x29
+#define SQ_V_COS_F32                              0x2a
+#define SQ_V_NOT_B32                              0x2b
+#define SQ_V_BFREV_B32                            0x2c
+#define SQ_V_FFBH_U32                             0x2d
+#define SQ_V_FFBL_B32                             0x2e
+#define SQ_V_FFBH_I32                             0x2f
+#define SQ_V_FREXP_EXP_I32_F64                    0x30
+#define SQ_V_FREXP_MANT_F64                       0x31
+#define SQ_V_FRACT_F64                            0x32
+#define SQ_V_FREXP_EXP_I32_F32                    0x33
+#define SQ_V_FREXP_MANT_F32                       0x34
+#define SQ_V_CLREXCP                              0x35
+#define SQ_V_MOVRELD_B32                          0x36
+#define SQ_V_MOVRELS_B32                          0x37
+#define SQ_V_MOVRELSD_B32                         0x38
+#define SQ_V_CVT_F16_U16                          0x39
+#define SQ_V_CVT_F16_I16                          0x3a
+#define SQ_V_CVT_U16_F16                          0x3b
+#define SQ_V_CVT_I16_F16                          0x3c
+#define SQ_V_RCP_F16                              0x3d
+#define SQ_V_SQRT_F16                             0x3e
+#define SQ_V_RSQ_F16                              0x3f
+#define SQ_V_LOG_F16                              0x40
+#define SQ_V_EXP_F16                              0x41
+#define SQ_V_FREXP_MANT_F16                       0x42
+#define SQ_V_FREXP_EXP_I16_F16                    0x43
+#define SQ_V_FLOOR_F16                            0x44
+#define SQ_V_CEIL_F16                             0x45
+#define SQ_V_TRUNC_F16                            0x46
+#define SQ_V_RNDNE_F16                            0x47
+#define SQ_V_FRACT_F16                            0x48
+#define SQ_V_SIN_F16                              0x49
+#define SQ_V_COS_F16                              0x4a
+#define SQ_V_EXP_LEGACY_F32                       0x4b
+#define SQ_V_LOG_LEGACY_F32                       0x4c
+#define SQ_V_CVT_NORM_I16_F16                     0x4d
+#define SQ_V_CVT_NORM_U16_F16                     0x4e
+#define SQ_SRC_SDWA                               0xf9
+#define SQ_V_OPC_OFFSET                           0x0
+#define SQ_V_OP2_OFFSET                           0x100
+#define SQ_V_OP1_OFFSET                           0x140
+#define SQ_V_INTRP_OFFSET                         0x270
+#define SQ_V_INTERP_P1_F32                        0x0
+#define SQ_V_INTERP_P2_F32                        0x1
+#define SQ_V_INTERP_MOV_F32                       0x2
+#define SQ_S_NOP                                  0x0
+#define SQ_S_ENDPGM                               0x1
+#define SQ_S_BRANCH                               0x2
+#define SQ_S_WAKEUP                               0x3
+#define SQ_S_CBRANCH_SCC0                         0x4
+#define SQ_S_CBRANCH_SCC1                         0x5
+#define SQ_S_CBRANCH_VCCZ                         0x6
+#define SQ_S_CBRANCH_VCCNZ                        0x7
+#define SQ_S_CBRANCH_EXECZ                        0x8
+#define SQ_S_CBRANCH_EXECNZ                       0x9
+#define SQ_S_BARRIER                              0xa
+#define SQ_S_SETKILL                              0xb
+#define SQ_S_WAITCNT                              0xc
+#define SQ_S_SETHALT                              0xd
+#define SQ_S_SLEEP                                0xe
+#define SQ_S_SETPRIO                              0xf
+#define SQ_S_SENDMSG                              0x10
+#define SQ_S_SENDMSGHALT                          0x11
+#define SQ_S_TRAP                                 0x12
+#define SQ_S_ICACHE_INV                           0x13
+#define SQ_S_INCPERFLEVEL                         0x14
+#define SQ_S_DECPERFLEVEL                         0x15
+#define SQ_S_TTRACEDATA                           0x16
+#define SQ_S_CBRANCH_CDBGSYS                      0x17
+#define SQ_S_CBRANCH_CDBGUSER                     0x18
+#define SQ_S_CBRANCH_CDBGSYS_OR_USER              0x19
+#define SQ_S_CBRANCH_CDBGSYS_AND_USER             0x1a
+#define SQ_S_ENDPGM_SAVED                         0x1b
+#define SQ_S_SET_GPR_IDX_OFF                      0x1c
+#define SQ_S_SET_GPR_IDX_MODE                     0x1d
+#define SQ_SRC_DPP                                0xfa
+#define SQ_SRC_LITERAL                            0xff
+#define SQ_VCC_LO                                 0x6a
+#define SQ_VCC_HI                                 0x6b
+#define SQ_PARAM_P10                              0x0
+#define SQ_PARAM_P20                              0x1
+#define SQ_PARAM_P0                               0x2
+#define SQ_SRC_LDS_DIRECT                         0xfe
+#define SQ_V_CNDMASK_B32                          0x0
+#define SQ_V_ADD_F32                              0x1
+#define SQ_V_SUB_F32                              0x2
+#define SQ_V_SUBREV_F32                           0x3
+#define SQ_V_MUL_LEGACY_F32                       0x4
+#define SQ_V_MUL_F32                              0x5
+#define SQ_V_MUL_I32_I24                          0x6
+#define SQ_V_MUL_HI_I32_I24                       0x7
+#define SQ_V_MUL_U32_U24                          0x8
+#define SQ_V_MUL_HI_U32_U24                       0x9
+#define SQ_V_MIN_F32                              0xa
+#define SQ_V_MAX_F32                              0xb
+#define SQ_V_MIN_I32                              0xc
+#define SQ_V_MAX_I32                              0xd
+#define SQ_V_MIN_U32                              0xe
+#define SQ_V_MAX_U32                              0xf
+#define SQ_V_LSHRREV_B32                          0x10
+#define SQ_V_ASHRREV_I32                          0x11
+#define SQ_V_LSHLREV_B32                          0x12
+#define SQ_V_AND_B32                              0x13
+#define SQ_V_OR_B32                               0x14
+#define SQ_V_XOR_B32                              0x15
+#define SQ_V_MAC_F32                              0x16
+#define SQ_V_MADMK_F32                            0x17
+#define SQ_V_MADAK_F32                            0x18
+#define SQ_V_ADD_U32                              0x19
+#define SQ_V_SUB_U32                              0x1a
+#define SQ_V_SUBREV_U32                           0x1b
+#define SQ_V_ADDC_U32                             0x1c
+#define SQ_V_SUBB_U32                             0x1d
+#define SQ_V_SUBBREV_U32                          0x1e
+#define SQ_V_ADD_F16                              0x1f
+#define SQ_V_SUB_F16                              0x20
+#define SQ_V_SUBREV_F16                           0x21
+#define SQ_V_MUL_F16                              0x22
+#define SQ_V_MAC_F16                              0x23
+#define SQ_V_MADMK_F16                            0x24
+#define SQ_V_MADAK_F16                            0x25
+#define SQ_V_ADD_U16                              0x26
+#define SQ_V_SUB_U16                              0x27
+#define SQ_V_SUBREV_U16                           0x28
+#define SQ_V_MUL_LO_U16                           0x29
+#define SQ_V_LSHLREV_B16                          0x2a
+#define SQ_V_LSHRREV_B16                          0x2b
+#define SQ_V_ASHRREV_I16                          0x2c
+#define SQ_V_MAX_F16                              0x2d
+#define SQ_V_MIN_F16                              0x2e
+#define SQ_V_MAX_U16                              0x2f
+#define SQ_V_MAX_I16                              0x30
+#define SQ_V_MIN_U16                              0x31
+#define SQ_V_MIN_I16                              0x32
+#define SQ_V_LDEXP_F16                            0x33
+#define SQ_FLAT_LOAD_UBYTE                        0x10
+#define SQ_FLAT_LOAD_SBYTE                        0x11
+#define SQ_FLAT_LOAD_USHORT                       0x12
+#define SQ_FLAT_LOAD_SSHORT                       0x13
+#define SQ_FLAT_LOAD_DWORD                        0x14
+#define SQ_FLAT_LOAD_DWORDX2                      0x15
+#define SQ_FLAT_LOAD_DWORDX3                      0x16
+#define SQ_FLAT_LOAD_DWORDX4                      0x17
+#define SQ_FLAT_STORE_BYTE                        0x18
+#define SQ_FLAT_STORE_SHORT                       0x1a
+#define SQ_FLAT_STORE_DWORD                       0x1c
+#define SQ_FLAT_STORE_DWORDX2                     0x1d
+#define SQ_FLAT_STORE_DWORDX3                     0x1e
+#define SQ_FLAT_STORE_DWORDX4                     0x1f
+#define SQ_FLAT_ATOMIC_SWAP                       0x40
+#define SQ_FLAT_ATOMIC_CMPSWAP                    0x41
+#define SQ_FLAT_ATOMIC_ADD                        0x42
+#define SQ_FLAT_ATOMIC_SUB                        0x43
+#define SQ_FLAT_ATOMIC_SMIN                       0x44
+#define SQ_FLAT_ATOMIC_UMIN                       0x45
+#define SQ_FLAT_ATOMIC_SMAX                       0x46
+#define SQ_FLAT_ATOMIC_UMAX                       0x47
+#define SQ_FLAT_ATOMIC_AND                        0x48
+#define SQ_FLAT_ATOMIC_OR                         0x49
+#define SQ_FLAT_ATOMIC_XOR                        0x4a
+#define SQ_FLAT_ATOMIC_INC                        0x4b
+#define SQ_FLAT_ATOMIC_DEC                        0x4c
+#define SQ_FLAT_ATOMIC_SWAP_X2                    0x60
+#define SQ_FLAT_ATOMIC_CMPSWAP_X2                 0x61
+#define SQ_FLAT_ATOMIC_ADD_X2                     0x62
+#define SQ_FLAT_ATOMIC_SUB_X2                     0x63
+#define SQ_FLAT_ATOMIC_SMIN_X2                    0x64
+#define SQ_FLAT_ATOMIC_UMIN_X2                    0x65
+#define SQ_FLAT_ATOMIC_SMAX_X2                    0x66
+#define SQ_FLAT_ATOMIC_UMAX_X2                    0x67
+#define SQ_FLAT_ATOMIC_AND_X2                     0x68
+#define SQ_FLAT_ATOMIC_OR_X2                      0x69
+#define SQ_FLAT_ATOMIC_XOR_X2                     0x6a
+#define SQ_FLAT_ATOMIC_INC_X2                     0x6b
+#define SQ_FLAT_ATOMIC_DEC_X2                     0x6c
+#define SQ_S_CMP_EQ_I32                           0x0
+#define SQ_S_CMP_LG_I32                           0x1
+#define SQ_S_CMP_GT_I32                           0x2
+#define SQ_S_CMP_GE_I32                           0x3
+#define SQ_S_CMP_LT_I32                           0x4
+#define SQ_S_CMP_LE_I32                           0x5
+#define SQ_S_CMP_EQ_U32                           0x6
+#define SQ_S_CMP_LG_U32                           0x7
+#define SQ_S_CMP_GT_U32                           0x8
+#define SQ_S_CMP_GE_U32                           0x9
+#define SQ_S_CMP_LT_U32                           0xa
+#define SQ_S_CMP_LE_U32                           0xb
+#define SQ_S_BITCMP0_B32                          0xc
+#define SQ_S_BITCMP1_B32                          0xd
+#define SQ_S_BITCMP0_B64                          0xe
+#define SQ_S_BITCMP1_B64                          0xf
+#define SQ_S_SETVSKIP                             0x10
+#define SQ_S_SET_GPR_IDX_ON                       0x11
+#define SQ_S_CMP_EQ_U64                           0x12
+#define SQ_S_CMP_LG_U64                           0x13
+#define SQ_M0                                     0x7c
+#define SQ_V_MAD_LEGACY_F32                       0x1c0
+#define SQ_V_MAD_F32                              0x1c1
+#define SQ_V_MAD_I32_I24                          0x1c2
+#define SQ_V_MAD_U32_U24                          0x1c3
+#define SQ_V_CUBEID_F32                           0x1c4
+#define SQ_V_CUBESC_F32                           0x1c5
+#define SQ_V_CUBETC_F32                           0x1c6
+#define SQ_V_CUBEMA_F32                           0x1c7
+#define SQ_V_BFE_U32                              0x1c8
+#define SQ_V_BFE_I32                              0x1c9
+#define SQ_V_BFI_B32                              0x1ca
+#define SQ_V_FMA_F32                              0x1cb
+#define SQ_V_FMA_F64                              0x1cc
+#define SQ_V_LERP_U8                              0x1cd
+#define SQ_V_ALIGNBIT_B32                         0x1ce
+#define SQ_V_ALIGNBYTE_B32                        0x1cf
+#define SQ_V_MIN3_F32                             0x1d0
+#define SQ_V_MIN3_I32                             0x1d1
+#define SQ_V_MIN3_U32                             0x1d2
+#define SQ_V_MAX3_F32                             0x1d3
+#define SQ_V_MAX3_I32                             0x1d4
+#define SQ_V_MAX3_U32                             0x1d5
+#define SQ_V_MED3_F32                             0x1d6
+#define SQ_V_MED3_I32                             0x1d7
+#define SQ_V_MED3_U32                             0x1d8
+#define SQ_V_SAD_U8                               0x1d9
+#define SQ_V_SAD_HI_U8                            0x1da
+#define SQ_V_SAD_U16                              0x1db
+#define SQ_V_SAD_U32                              0x1dc
+#define SQ_V_CVT_PK_U8_F32                        0x1dd
+#define SQ_V_DIV_FIXUP_F32                        0x1de
+#define SQ_V_DIV_FIXUP_F64                        0x1df
+#define SQ_V_DIV_SCALE_F32                        0x1e0
+#define SQ_V_DIV_SCALE_F64                        0x1e1
+#define SQ_V_DIV_FMAS_F32                         0x1e2
+#define SQ_V_DIV_FMAS_F64                         0x1e3
+#define SQ_V_MSAD_U8                              0x1e4
+#define SQ_V_QSAD_PK_U16_U8                       0x1e5
+#define SQ_V_MQSAD_PK_U16_U8                      0x1e6
+#define SQ_V_MQSAD_U32_U8                         0x1e7
+#define SQ_V_MAD_U64_U32                          0x1e8
+#define SQ_V_MAD_I64_I32                          0x1e9
+#define SQ_V_MAD_F16                              0x1ea
+#define SQ_V_MAD_U16                              0x1eb
+#define SQ_V_MAD_I16                              0x1ec
+#define SQ_V_PERM_B32                             0x1ed
+#define SQ_V_FMA_F16                              0x1ee
+#define SQ_V_DIV_FIXUP_F16                        0x1ef
+#define SQ_V_CVT_PKACCUM_U8_F32                   0x1f0
+#define SQ_V_INTERP_P1LL_F16                      0x274
+#define SQ_V_INTERP_P1LV_F16                      0x275
+#define SQ_V_INTERP_P2_F16                        0x276
+#define SQ_V_ADD_F64                              0x280
+#define SQ_V_MUL_F64                              0x281
+#define SQ_V_MIN_F64                              0x282
+#define SQ_V_MAX_F64                              0x283
+#define SQ_V_LDEXP_F64                            0x284
+#define SQ_V_MUL_LO_U32                           0x285
+#define SQ_V_MUL_HI_U32                           0x286
+#define SQ_V_MUL_HI_I32                           0x287
+#define SQ_V_LDEXP_F32                            0x288
+#define SQ_V_READLANE_B32                         0x289
+#define SQ_V_WRITELANE_B32                        0x28a
+#define SQ_V_BCNT_U32_B32                         0x28b
+#define SQ_V_MBCNT_LO_U32_B32                     0x28c
+#define SQ_V_MBCNT_HI_U32_B32                     0x28d
+#define SQ_V_MAC_LEGACY_F32                       0x28e
+#define SQ_V_LSHLREV_B64                          0x28f
+#define SQ_V_LSHRREV_B64                          0x290
+#define SQ_V_ASHRREV_I64                          0x291
+#define SQ_V_TRIG_PREOP_F64                       0x292
+#define SQ_V_BFM_B32                              0x293
+#define SQ_V_CVT_PKNORM_I16_F32                   0x294
+#define SQ_V_CVT_PKNORM_U16_F32                   0x295
+#define SQ_V_CVT_PKRTZ_F16_F32                    0x296
+#define SQ_V_CVT_PK_U16_U32                       0x297
+#define SQ_V_CVT_PK_I16_I32                       0x298
+#define SQ_V_CVT_PKNORM_I16_F16                   0x299
+#define SQ_V_CVT_PKNORM_U16_F16                   0x29a
+#define SQ_VCC_ALL                                0x0
+#define SQ_SRC_EXECZ                              0xfc
+#define SQ_FLAT_SCRATCH_LO                        0x66
+#define SQ_FLAT_SCRATCH_HI                        0x67
+#define SQ_SYSMSG_OP_ECC_ERR_INTERRUPT            0x1
+#define SQ_SYSMSG_OP_REG_RD                       0x2
+#define SQ_SYSMSG_OP_HOST_TRAP_ACK                0x3
+#define SQ_SYSMSG_OP_TTRACE_PC                    0x4
+#define SQ_HW_REG_MODE                            0x1
+#define SQ_HW_REG_STATUS                          0x2
+#define SQ_HW_REG_TRAPSTS                         0x3
+#define SQ_HW_REG_HW_ID                           0x4
+#define SQ_HW_REG_GPR_ALLOC                       0x5
+#define SQ_HW_REG_LDS_ALLOC                       0x6
+#define SQ_HW_REG_IB_STS                          0x7
+#define SQ_HW_REG_PC_LO                           0x8
+#define SQ_HW_REG_PC_HI                           0x9
+#define SQ_HW_REG_INST_DW0                        0xa
+#define SQ_HW_REG_INST_DW1                        0xb
+#define SQ_HW_REG_IB_DBG0                         0xc
+#define SQ_HW_REG_IB_DBG1                         0xd
+#define SQ_DPP_BOUND_OFF                          0x0
+#define SQ_DPP_BOUND_ZERO                         0x1
+#define SQ_R1                                     0x1
+#define SQ_R2                                     0x2
+#define SQ_R3                                     0x3
+#define SQ_R4                                     0x4
+#define SQ_R5                                     0x5
+#define SQ_R6                                     0x6
+#define SQ_R7                                     0x7
+#define SQ_R8                                     0x8
+#define SQ_R9                                     0x9
+#define SQ_R10                                    0xa
+#define SQ_R11                                    0xb
+#define SQ_R12                                    0xc
+#define SQ_R13                                    0xd
+#define SQ_R14                                    0xe
+#define SQ_R15                                    0xf
+#define SQ_S_ADD_U32                              0x0
+#define SQ_S_SUB_U32                              0x1
+#define SQ_S_ADD_I32                              0x2
+#define SQ_S_SUB_I32                              0x3
+#define SQ_S_ADDC_U32                             0x4
+#define SQ_S_SUBB_U32                             0x5
+#define SQ_S_MIN_I32                              0x6
+#define SQ_S_MIN_U32                              0x7
+#define SQ_S_MAX_I32                              0x8
+#define SQ_S_MAX_U32                              0x9
+#define SQ_S_CSELECT_B32                          0xa
+#define SQ_S_CSELECT_B64                          0xb
+#define SQ_S_AND_B32                              0xc
+#define SQ_S_AND_B64                              0xd
+#define SQ_S_OR_B32                               0xe
+#define SQ_S_OR_B64                               0xf
+#define SQ_S_XOR_B32                              0x10
+#define SQ_S_XOR_B64                              0x11
+#define SQ_S_ANDN2_B32                            0x12
+#define SQ_S_ANDN2_B64                            0x13
+#define SQ_S_ORN2_B32                             0x14
+#define SQ_S_ORN2_B64                             0x15
+#define SQ_S_NAND_B32                             0x16
+#define SQ_S_NAND_B64                             0x17
+#define SQ_S_NOR_B32                              0x18
+#define SQ_S_NOR_B64                              0x19
+#define SQ_S_XNOR_B32                             0x1a
+#define SQ_S_XNOR_B64                             0x1b
+#define SQ_S_LSHL_B32                             0x1c
+#define SQ_S_LSHL_B64                             0x1d
+#define SQ_S_LSHR_B32                             0x1e
+#define SQ_S_LSHR_B64                             0x1f
+#define SQ_S_ASHR_I32                             0x20
+#define SQ_S_ASHR_I64                             0x21
+#define SQ_S_BFM_B32                              0x22
+#define SQ_S_BFM_B64                              0x23
+#define SQ_S_MUL_I32                              0x24
+#define SQ_S_BFE_U32                              0x25
+#define SQ_S_BFE_I32                              0x26
+#define SQ_S_BFE_U64                              0x27
+#define SQ_S_BFE_I64                              0x28
+#define SQ_S_CBRANCH_G_FORK                       0x29
+#define SQ_S_ABSDIFF_I32                          0x2a
+#define SQ_S_RFE_RESTORE_B64                      0x2b
+#define SQ_MSG_INTERRUPT                          0x1
+#define SQ_MSG_GS                                 0x2
+#define SQ_MSG_GS_DONE                            0x3
+#define SQ_MSG_SAVEWAVE                           0x4
+#define SQ_MSG_SYSMSG                             0xf
+typedef enum SX_BLEND_OPT {
+	BLEND_OPT_PRESERVE_NONE_IGNORE_ALL               = 0x0,
+	BLEND_OPT_PRESERVE_ALL_IGNORE_NONE               = 0x1,
+	BLEND_OPT_PRESERVE_C1_IGNORE_C0                  = 0x2,
+	BLEND_OPT_PRESERVE_C0_IGNORE_C1                  = 0x3,
+	BLEND_OPT_PRESERVE_A1_IGNORE_A0                  = 0x4,
+	BLEND_OPT_PRESERVE_A0_IGNORE_A1                  = 0x5,
+	BLEND_OPT_PRESERVE_NONE_IGNORE_A0                = 0x6,
+	BLEND_OPT_PRESERVE_NONE_IGNORE_NONE              = 0x7,
+} SX_BLEND_OPT;
+typedef enum SX_OPT_COMB_FCN {
+	OPT_COMB_NONE                                    = 0x0,
+	OPT_COMB_ADD                                     = 0x1,
+	OPT_COMB_SUBTRACT                                = 0x2,
+	OPT_COMB_MIN                                     = 0x3,
+	OPT_COMB_MAX                                     = 0x4,
+	OPT_COMB_REVSUBTRACT                             = 0x5,
+	OPT_COMB_BLEND_DISABLED                          = 0x6,
+	OPT_COMB_SAFE_ADD                                = 0x7,
+} SX_OPT_COMB_FCN;
+typedef enum SX_DOWNCONVERT_FORMAT {
+	SX_RT_EXPORT_NO_CONVERSION                       = 0x0,
+	SX_RT_EXPORT_32_R                                = 0x1,
+	SX_RT_EXPORT_32_A                                = 0x2,
+	SX_RT_EXPORT_10_11_11                            = 0x3,
+	SX_RT_EXPORT_2_10_10_10                          = 0x4,
+	SX_RT_EXPORT_8_8_8_8                             = 0x5,
+	SX_RT_EXPORT_5_6_5                               = 0x6,
+	SX_RT_EXPORT_1_5_5_5                             = 0x7,
+	SX_RT_EXPORT_4_4_4_4                             = 0x8,
+	SX_RT_EXPORT_16_16_GR                            = 0x9,
+	SX_RT_EXPORT_16_16_AR                            = 0xa,
+} SX_DOWNCONVERT_FORMAT;
+typedef enum TEX_BORDER_COLOR_TYPE {
+	TEX_BorderColor_TransparentBlack                 = 0x0,
+	TEX_BorderColor_OpaqueBlack                      = 0x1,
+	TEX_BorderColor_OpaqueWhite                      = 0x2,
+	TEX_BorderColor_Register                         = 0x3,
+} TEX_BORDER_COLOR_TYPE;
+typedef enum TEX_CHROMA_KEY {
+	TEX_ChromaKey_Disabled                           = 0x0,
+	TEX_ChromaKey_Kill                               = 0x1,
+	TEX_ChromaKey_Blend                              = 0x2,
+	TEX_ChromaKey_RESERVED_3                         = 0x3,
+} TEX_CHROMA_KEY;
+typedef enum TEX_CLAMP {
+	TEX_Clamp_Repeat                                 = 0x0,
+	TEX_Clamp_Mirror                                 = 0x1,
+	TEX_Clamp_ClampToLast                            = 0x2,
+	TEX_Clamp_MirrorOnceToLast                       = 0x3,
+	TEX_Clamp_ClampHalfToBorder                      = 0x4,
+	TEX_Clamp_MirrorOnceHalfToBorder                 = 0x5,
+	TEX_Clamp_ClampToBorder                          = 0x6,
+	TEX_Clamp_MirrorOnceToBorder                     = 0x7,
+} TEX_CLAMP;
+typedef enum TEX_COORD_TYPE {
+	TEX_CoordType_Unnormalized                       = 0x0,
+	TEX_CoordType_Normalized                         = 0x1,
+} TEX_COORD_TYPE;
+typedef enum TEX_DEPTH_COMPARE_FUNCTION {
+	TEX_DepthCompareFunction_Never                   = 0x0,
+	TEX_DepthCompareFunction_Less                    = 0x1,
+	TEX_DepthCompareFunction_Equal                   = 0x2,
+	TEX_DepthCompareFunction_LessEqual               = 0x3,
+	TEX_DepthCompareFunction_Greater                 = 0x4,
+	TEX_DepthCompareFunction_NotEqual                = 0x5,
+	TEX_DepthCompareFunction_GreaterEqual            = 0x6,
+	TEX_DepthCompareFunction_Always                  = 0x7,
+} TEX_DEPTH_COMPARE_FUNCTION;
+typedef enum TEX_DIM {
+	TEX_Dim_1D                                       = 0x0,
+	TEX_Dim_2D                                       = 0x1,
+	TEX_Dim_3D                                       = 0x2,
+	TEX_Dim_CubeMap                                  = 0x3,
+	TEX_Dim_1DArray                                  = 0x4,
+	TEX_Dim_2DArray                                  = 0x5,
+	TEX_Dim_2D_MSAA                                  = 0x6,
+	TEX_Dim_2DArray_MSAA                             = 0x7,
+} TEX_DIM;
+typedef enum TEX_FORMAT_COMP {
+	TEX_FormatComp_Unsigned                          = 0x0,
+	TEX_FormatComp_Signed                            = 0x1,
+	TEX_FormatComp_UnsignedBiased                    = 0x2,
+	TEX_FormatComp_RESERVED_3                        = 0x3,
+} TEX_FORMAT_COMP;
+typedef enum TEX_MAX_ANISO_RATIO {
+	TEX_MaxAnisoRatio_1to1                           = 0x0,
+	TEX_MaxAnisoRatio_2to1                           = 0x1,
+	TEX_MaxAnisoRatio_4to1                           = 0x2,
+	TEX_MaxAnisoRatio_8to1                           = 0x3,
+	TEX_MaxAnisoRatio_16to1                          = 0x4,
+	TEX_MaxAnisoRatio_RESERVED_5                     = 0x5,
+	TEX_MaxAnisoRatio_RESERVED_6                     = 0x6,
+	TEX_MaxAnisoRatio_RESERVED_7                     = 0x7,
+} TEX_MAX_ANISO_RATIO;
+typedef enum TEX_MIP_FILTER {
+	TEX_MipFilter_None                               = 0x0,
+	TEX_MipFilter_Point                              = 0x1,
+	TEX_MipFilter_Linear                             = 0x2,
+	TEX_MipFilter_Point_Aniso_Adj                    = 0x3,
+} TEX_MIP_FILTER;
+typedef enum TEX_REQUEST_SIZE {
+	TEX_RequestSize_32B                              = 0x0,
+	TEX_RequestSize_64B                              = 0x1,
+	TEX_RequestSize_128B                             = 0x2,
+	TEX_RequestSize_2X64B                            = 0x3,
+} TEX_REQUEST_SIZE;
+typedef enum TEX_SAMPLER_TYPE {
+	TEX_SamplerType_Invalid                          = 0x0,
+	TEX_SamplerType_Valid                            = 0x1,
+} TEX_SAMPLER_TYPE;
+typedef enum TEX_XY_FILTER {
+	TEX_XYFilter_Point                               = 0x0,
+	TEX_XYFilter_Linear                              = 0x1,
+	TEX_XYFilter_AnisoPoint                          = 0x2,
+	TEX_XYFilter_AnisoLinear                         = 0x3,
+} TEX_XY_FILTER;
+typedef enum TEX_Z_FILTER {
+	TEX_ZFilter_None                                 = 0x0,
+	TEX_ZFilter_Point                                = 0x1,
+	TEX_ZFilter_Linear                               = 0x2,
+	TEX_ZFilter_RESERVED_3                           = 0x3,
+} TEX_Z_FILTER;
+typedef enum VTX_CLAMP {
+	VTX_Clamp_ClampToZero                            = 0x0,
+	VTX_Clamp_ClampToNAN                             = 0x1,
+} VTX_CLAMP;
+typedef enum VTX_FETCH_TYPE {
+	VTX_FetchType_VertexData                         = 0x0,
+	VTX_FetchType_InstanceData                       = 0x1,
+	VTX_FetchType_NoIndexOffset                      = 0x2,
+	VTX_FetchType_RESERVED_3                         = 0x3,
+} VTX_FETCH_TYPE;
+typedef enum VTX_FORMAT_COMP_ALL {
+	VTX_FormatCompAll_Unsigned                       = 0x0,
+	VTX_FormatCompAll_Signed                         = 0x1,
+} VTX_FORMAT_COMP_ALL;
+typedef enum VTX_MEM_REQUEST_SIZE {
+	VTX_MemRequestSize_32B                           = 0x0,
+	VTX_MemRequestSize_64B                           = 0x1,
+} VTX_MEM_REQUEST_SIZE;
+typedef enum TVX_DATA_FORMAT {
+	TVX_FMT_INVALID                                  = 0x0,
+	TVX_FMT_8                                        = 0x1,
+	TVX_FMT_4_4                                      = 0x2,
+	TVX_FMT_3_3_2                                    = 0x3,
+	TVX_FMT_RESERVED_4                               = 0x4,
+	TVX_FMT_16                                       = 0x5,
+	TVX_FMT_16_FLOAT                                 = 0x6,
+	TVX_FMT_8_8                                      = 0x7,
+	TVX_FMT_5_6_5                                    = 0x8,
+	TVX_FMT_6_5_5                                    = 0x9,
+	TVX_FMT_1_5_5_5                                  = 0xa,
+	TVX_FMT_4_4_4_4                                  = 0xb,
+	TVX_FMT_5_5_5_1                                  = 0xc,
+	TVX_FMT_32                                       = 0xd,
+	TVX_FMT_32_FLOAT                                 = 0xe,
+	TVX_FMT_16_16                                    = 0xf,
+	TVX_FMT_16_16_FLOAT                              = 0x10,
+	TVX_FMT_8_24                                     = 0x11,
+	TVX_FMT_8_24_FLOAT                               = 0x12,
+	TVX_FMT_24_8                                     = 0x13,
+	TVX_FMT_24_8_FLOAT                               = 0x14,
+	TVX_FMT_10_11_11                                 = 0x15,
+	TVX_FMT_10_11_11_FLOAT                           = 0x16,
+	TVX_FMT_11_11_10                                 = 0x17,
+	TVX_FMT_11_11_10_FLOAT                           = 0x18,
+	TVX_FMT_2_10_10_10                               = 0x19,
+	TVX_FMT_8_8_8_8                                  = 0x1a,
+	TVX_FMT_10_10_10_2                               = 0x1b,
+	TVX_FMT_X24_8_32_FLOAT                           = 0x1c,
+	TVX_FMT_32_32                                    = 0x1d,
+	TVX_FMT_32_32_FLOAT                              = 0x1e,
+	TVX_FMT_16_16_16_16                              = 0x1f,
+	TVX_FMT_16_16_16_16_FLOAT                        = 0x20,
+	TVX_FMT_RESERVED_33                              = 0x21,
+	TVX_FMT_32_32_32_32                              = 0x22,
+	TVX_FMT_32_32_32_32_FLOAT                        = 0x23,
+	TVX_FMT_RESERVED_36                              = 0x24,
+	TVX_FMT_1                                        = 0x25,
+	TVX_FMT_1_REVERSED                               = 0x26,
+	TVX_FMT_GB_GR                                    = 0x27,
+	TVX_FMT_BG_RG                                    = 0x28,
+	TVX_FMT_32_AS_8                                  = 0x29,
+	TVX_FMT_32_AS_8_8                                = 0x2a,
+	TVX_FMT_5_9_9_9_SHAREDEXP                        = 0x2b,
+	TVX_FMT_8_8_8                                    = 0x2c,
+	TVX_FMT_16_16_16                                 = 0x2d,
+	TVX_FMT_16_16_16_FLOAT                           = 0x2e,
+	TVX_FMT_32_32_32                                 = 0x2f,
+	TVX_FMT_32_32_32_FLOAT                           = 0x30,
+	TVX_FMT_BC1                                      = 0x31,
+	TVX_FMT_BC2                                      = 0x32,
+	TVX_FMT_BC3                                      = 0x33,
+	TVX_FMT_BC4                                      = 0x34,
+	TVX_FMT_BC5                                      = 0x35,
+	TVX_FMT_APC0                                     = 0x36,
+	TVX_FMT_APC1                                     = 0x37,
+	TVX_FMT_APC2                                     = 0x38,
+	TVX_FMT_APC3                                     = 0x39,
+	TVX_FMT_APC4                                     = 0x3a,
+	TVX_FMT_APC5                                     = 0x3b,
+	TVX_FMT_APC6                                     = 0x3c,
+	TVX_FMT_APC7                                     = 0x3d,
+	TVX_FMT_CTX1                                     = 0x3e,
+	TVX_FMT_RESERVED_63                              = 0x3f,
+} TVX_DATA_FORMAT;
+typedef enum TVX_DST_SEL {
+	TVX_DstSel_X                                     = 0x0,
+	TVX_DstSel_Y                                     = 0x1,
+	TVX_DstSel_Z                                     = 0x2,
+	TVX_DstSel_W                                     = 0x3,
+	TVX_DstSel_0f                                    = 0x4,
+	TVX_DstSel_1f                                    = 0x5,
+	TVX_DstSel_RESERVED_6                            = 0x6,
+	TVX_DstSel_Mask                                  = 0x7,
+} TVX_DST_SEL;
+typedef enum TVX_ENDIAN_SWAP {
+	TVX_EndianSwap_None                              = 0x0,
+	TVX_EndianSwap_8in16                             = 0x1,
+	TVX_EndianSwap_8in32                             = 0x2,
+	TVX_EndianSwap_8in64                             = 0x3,
+} TVX_ENDIAN_SWAP;
+typedef enum TVX_INST {
+	TVX_Inst_NormalVertexFetch                       = 0x0,
+	TVX_Inst_SemanticVertexFetch                     = 0x1,
+	TVX_Inst_RESERVED_2                              = 0x2,
+	TVX_Inst_LD                                      = 0x3,
+	TVX_Inst_GetTextureResInfo                       = 0x4,
+	TVX_Inst_GetNumberOfSamples                      = 0x5,
+	TVX_Inst_GetLOD                                  = 0x6,
+	TVX_Inst_GetGradientsH                           = 0x7,
+	TVX_Inst_GetGradientsV                           = 0x8,
+	TVX_Inst_SetTextureOffsets                       = 0x9,
+	TVX_Inst_KeepGradients                           = 0xa,
+	TVX_Inst_SetGradientsH                           = 0xb,
+	TVX_Inst_SetGradientsV                           = 0xc,
+	TVX_Inst_Pass                                    = 0xd,
+	TVX_Inst_GetBufferResInfo                        = 0xe,
+	TVX_Inst_RESERVED_15                             = 0xf,
+	TVX_Inst_Sample                                  = 0x10,
+	TVX_Inst_Sample_L                                = 0x11,
+	TVX_Inst_Sample_LB                               = 0x12,
+	TVX_Inst_Sample_LZ                               = 0x13,
+	TVX_Inst_Sample_G                                = 0x14,
+	TVX_Inst_Gather4                                 = 0x15,
+	TVX_Inst_Sample_G_LB                             = 0x16,
+	TVX_Inst_Gather4_O                               = 0x17,
+	TVX_Inst_Sample_C                                = 0x18,
+	TVX_Inst_Sample_C_L                              = 0x19,
+	TVX_Inst_Sample_C_LB                             = 0x1a,
+	TVX_Inst_Sample_C_LZ                             = 0x1b,
+	TVX_Inst_Sample_C_G                              = 0x1c,
+	TVX_Inst_Gather4_C                               = 0x1d,
+	TVX_Inst_Sample_C_G_LB                           = 0x1e,
+	TVX_Inst_Gather4_C_O                             = 0x1f,
+} TVX_INST;
+typedef enum TVX_NUM_FORMAT_ALL {
+	TVX_NumFormatAll_Norm                            = 0x0,
+	TVX_NumFormatAll_Int                             = 0x1,
+	TVX_NumFormatAll_Scaled                          = 0x2,
+	TVX_NumFormatAll_RESERVED_3                      = 0x3,
+} TVX_NUM_FORMAT_ALL;
+typedef enum TVX_SRC_SEL {
+	TVX_SrcSel_X                                     = 0x0,
+	TVX_SrcSel_Y                                     = 0x1,
+	TVX_SrcSel_Z                                     = 0x2,
+	TVX_SrcSel_W                                     = 0x3,
+	TVX_SrcSel_0f                                    = 0x4,
+	TVX_SrcSel_1f                                    = 0x5,
+} TVX_SRC_SEL;
+typedef enum TVX_SRF_MODE_ALL {
+	TVX_SRFModeAll_ZCMO                              = 0x0,
+	TVX_SRFModeAll_NZ                                = 0x1,
+} TVX_SRF_MODE_ALL;
+typedef enum TVX_TYPE {
+	TVX_Type_InvalidTextureResource                  = 0x0,
+	TVX_Type_InvalidVertexBuffer                     = 0x1,
+	TVX_Type_ValidTextureResource                    = 0x2,
+	TVX_Type_ValidVertexBuffer                       = 0x3,
+} TVX_TYPE;
+typedef enum TC_OP_MASKS {
+	TC_OP_MASK_FLUSH_DENROM                          = 0x8,
+	TC_OP_MASK_64                                    = 0x20,
+	TC_OP_MASK_NO_RTN                                = 0x40,
+} TC_OP_MASKS;
+typedef enum TC_OP {
+	TC_OP_READ                                       = 0x0,
+	TC_OP_ATOMIC_FCMPSWAP_RTN_32                     = 0x1,
+	TC_OP_ATOMIC_FMIN_RTN_32                         = 0x2,
+	TC_OP_ATOMIC_FMAX_RTN_32                         = 0x3,
+	TC_OP_RESERVED_FOP_RTN_32_0                      = 0x4,
+	TC_OP_RESERVED_FOP_RTN_32_1                      = 0x5,
+	TC_OP_RESERVED_FOP_RTN_32_2                      = 0x6,
+	TC_OP_ATOMIC_SWAP_RTN_32                         = 0x7,
+	TC_OP_ATOMIC_CMPSWAP_RTN_32                      = 0x8,
+	TC_OP_ATOMIC_FCMPSWAP_FLUSH_DENORM_RTN_32        = 0x9,
+	TC_OP_ATOMIC_FMIN_FLUSH_DENORM_RTN_32            = 0xa,
+	TC_OP_ATOMIC_FMAX_FLUSH_DENORM_RTN_32            = 0xb,
+	TC_OP_RESERVED_FOP_FLUSH_DENORM_RTN_32_0         = 0xc,
+	TC_OP_RESERVED_FOP_FLUSH_DENORM_RTN_32_1         = 0xd,
+	TC_OP_RESERVED_FOP_FLUSH_DENORM_RTN_32_2         = 0xe,
+	TC_OP_ATOMIC_ADD_RTN_32                          = 0xf,
+	TC_OP_ATOMIC_SUB_RTN_32                          = 0x10,
+	TC_OP_ATOMIC_SMIN_RTN_32                         = 0x11,
+	TC_OP_ATOMIC_UMIN_RTN_32                         = 0x12,
+	TC_OP_ATOMIC_SMAX_RTN_32                         = 0x13,
+	TC_OP_ATOMIC_UMAX_RTN_32                         = 0x14,
+	TC_OP_ATOMIC_AND_RTN_32                          = 0x15,
+	TC_OP_ATOMIC_OR_RTN_32                           = 0x16,
+	TC_OP_ATOMIC_XOR_RTN_32                          = 0x17,
+	TC_OP_ATOMIC_INC_RTN_32                          = 0x18,
+	TC_OP_ATOMIC_DEC_RTN_32                          = 0x19,
+	TC_OP_WBINVL1_VOL                                = 0x1a,
+	TC_OP_WBINVL1_SD                                 = 0x1b,
+	TC_OP_RESERVED_NON_FLOAT_RTN_32_0                = 0x1c,
+	TC_OP_RESERVED_NON_FLOAT_RTN_32_1                = 0x1d,
+	TC_OP_RESERVED_NON_FLOAT_RTN_32_2                = 0x1e,
+	TC_OP_RESERVED_NON_FLOAT_RTN_32_3                = 0x1f,
+	TC_OP_WRITE                                      = 0x20,
+	TC_OP_ATOMIC_FCMPSWAP_RTN_64                     = 0x21,
+	TC_OP_ATOMIC_FMIN_RTN_64                         = 0x22,
+	TC_OP_ATOMIC_FMAX_RTN_64                         = 0x23,
+	TC_OP_RESERVED_FOP_RTN_64_0                      = 0x24,
+	TC_OP_RESERVED_FOP_RTN_64_1                      = 0x25,
+	TC_OP_RESERVED_FOP_RTN_64_2                      = 0x26,
+	TC_OP_ATOMIC_SWAP_RTN_64                         = 0x27,
+	TC_OP_ATOMIC_CMPSWAP_RTN_64                      = 0x28,
+	TC_OP_ATOMIC_FCMPSWAP_FLUSH_DENORM_RTN_64        = 0x29,
+	TC_OP_ATOMIC_FMIN_FLUSH_DENORM_RTN_64            = 0x2a,
+	TC_OP_ATOMIC_FMAX_FLUSH_DENORM_RTN_64            = 0x2b,
+	TC_OP_WBINVL2_SD                                 = 0x2c,
+	TC_OP_RESERVED_FOP_FLUSH_DENORM_RTN_64_0         = 0x2d,
+	TC_OP_RESERVED_FOP_FLUSH_DENORM_RTN_64_1         = 0x2e,
+	TC_OP_ATOMIC_ADD_RTN_64                          = 0x2f,
+	TC_OP_ATOMIC_SUB_RTN_64                          = 0x30,
+	TC_OP_ATOMIC_SMIN_RTN_64                         = 0x31,
+	TC_OP_ATOMIC_UMIN_RTN_64                         = 0x32,
+	TC_OP_ATOMIC_SMAX_RTN_64                         = 0x33,
+	TC_OP_ATOMIC_UMAX_RTN_64                         = 0x34,
+	TC_OP_ATOMIC_AND_RTN_64                          = 0x35,
+	TC_OP_ATOMIC_OR_RTN_64                           = 0x36,
+	TC_OP_ATOMIC_XOR_RTN_64                          = 0x37,
+	TC_OP_ATOMIC_INC_RTN_64                          = 0x38,
+	TC_OP_ATOMIC_DEC_RTN_64                          = 0x39,
+	TC_OP_WBL2_NC                                    = 0x3a,
+	TC_OP_RESERVED_NON_FLOAT_RTN_64_0                = 0x3b,
+	TC_OP_RESERVED_NON_FLOAT_RTN_64_1                = 0x3c,
+	TC_OP_RESERVED_NON_FLOAT_RTN_64_2                = 0x3d,
+	TC_OP_RESERVED_NON_FLOAT_RTN_64_3                = 0x3e,
+	TC_OP_RESERVED_NON_FLOAT_RTN_64_4                = 0x3f,
+	TC_OP_WBINVL1                                    = 0x40,
+	TC_OP_ATOMIC_FCMPSWAP_32                         = 0x41,
+	TC_OP_ATOMIC_FMIN_32                             = 0x42,
+	TC_OP_ATOMIC_FMAX_32                             = 0x43,
+	TC_OP_RESERVED_FOP_32_0                          = 0x44,
+	TC_OP_RESERVED_FOP_32_1                          = 0x45,
+	TC_OP_RESERVED_FOP_32_2                          = 0x46,
+	TC_OP_ATOMIC_SWAP_32                             = 0x47,
+	TC_OP_ATOMIC_CMPSWAP_32                          = 0x48,
+	TC_OP_ATOMIC_FCMPSWAP_FLUSH_DENORM_32            = 0x49,
+	TC_OP_ATOMIC_FMIN_FLUSH_DENORM_32                = 0x4a,
+	TC_OP_ATOMIC_FMAX_FLUSH_DENORM_32                = 0x4b,
+	TC_OP_RESERVED_FOP_FLUSH_DENORM_32_0             = 0x4c,
+	TC_OP_RESERVED_FOP_FLUSH_DENORM_32_1             = 0x4d,
+	TC_OP_RESERVED_FOP_FLUSH_DENORM_32_2             = 0x4e,
+	TC_OP_ATOMIC_ADD_32                              = 0x4f,
+	TC_OP_ATOMIC_SUB_32                              = 0x50,
+	TC_OP_ATOMIC_SMIN_32                             = 0x51,
+	TC_OP_ATOMIC_UMIN_32                             = 0x52,
+	TC_OP_ATOMIC_SMAX_32                             = 0x53,
+	TC_OP_ATOMIC_UMAX_32                             = 0x54,
+	TC_OP_ATOMIC_AND_32                              = 0x55,
+	TC_OP_ATOMIC_OR_32                               = 0x56,
+	TC_OP_ATOMIC_XOR_32                              = 0x57,
+	TC_OP_ATOMIC_INC_32                              = 0x58,
+	TC_OP_ATOMIC_DEC_32                              = 0x59,
+	TC_OP_INVL2_NC                                   = 0x5a,
+	TC_OP_RESERVED_NON_FLOAT_32_0                    = 0x5b,
+	TC_OP_RESERVED_NON_FLOAT_32_1                    = 0x5c,
+	TC_OP_RESERVED_NON_FLOAT_32_2                    = 0x5d,
+	TC_OP_RESERVED_NON_FLOAT_32_3                    = 0x5e,
+	TC_OP_RESERVED_NON_FLOAT_32_4                    = 0x5f,
+	TC_OP_WBINVL2                                    = 0x60,
+	TC_OP_ATOMIC_FCMPSWAP_64                         = 0x61,
+	TC_OP_ATOMIC_FMIN_64                             = 0x62,
+	TC_OP_ATOMIC_FMAX_64                             = 0x63,
+	TC_OP_RESERVED_FOP_64_0                          = 0x64,
+	TC_OP_RESERVED_FOP_64_1                          = 0x65,
+	TC_OP_RESERVED_FOP_64_2                          = 0x66,
+	TC_OP_ATOMIC_SWAP_64                             = 0x67,
+	TC_OP_ATOMIC_CMPSWAP_64                          = 0x68,
+	TC_OP_ATOMIC_FCMPSWAP_FLUSH_DENORM_64            = 0x69,
+	TC_OP_ATOMIC_FMIN_FLUSH_DENORM_64                = 0x6a,
+	TC_OP_ATOMIC_FMAX_FLUSH_DENORM_64                = 0x6b,
+	TC_OP_RESERVED_FOP_FLUSH_DENORM_64_0             = 0x6c,
+	TC_OP_RESERVED_FOP_FLUSH_DENORM_64_1             = 0x6d,
+	TC_OP_RESERVED_FOP_FLUSH_DENORM_64_2             = 0x6e,
+	TC_OP_ATOMIC_ADD_64                              = 0x6f,
+	TC_OP_ATOMIC_SUB_64                              = 0x70,
+	TC_OP_ATOMIC_SMIN_64                             = 0x71,
+	TC_OP_ATOMIC_UMIN_64                             = 0x72,
+	TC_OP_ATOMIC_SMAX_64                             = 0x73,
+	TC_OP_ATOMIC_UMAX_64                             = 0x74,
+	TC_OP_ATOMIC_AND_64                              = 0x75,
+	TC_OP_ATOMIC_OR_64                               = 0x76,
+	TC_OP_ATOMIC_XOR_64                              = 0x77,
+	TC_OP_ATOMIC_INC_64                              = 0x78,
+	TC_OP_ATOMIC_DEC_64                              = 0x79,
+	TC_OP_WBINVL2_NC                                 = 0x7a,
+	TC_OP_RESERVED_NON_FLOAT_64_0                    = 0x7b,
+	TC_OP_RESERVED_NON_FLOAT_64_1                    = 0x7c,
+	TC_OP_RESERVED_NON_FLOAT_64_2                    = 0x7d,
+	TC_OP_RESERVED_NON_FLOAT_64_3                    = 0x7e,
+	TC_OP_RESERVED_NON_FLOAT_64_4                    = 0x7f,
+} TC_OP;
+typedef enum TC_CHUB_REQ_CREDITS_ENUM {
+	TC_CHUB_REQ_CREDITS                              = 0x10,
+} TC_CHUB_REQ_CREDITS_ENUM;
+typedef enum CHUB_TC_RET_CREDITS_ENUM {
+	CHUB_TC_RET_CREDITS                              = 0x20,
+} CHUB_TC_RET_CREDITS_ENUM;
+typedef enum TC_NACKS {
+	TC_NACK_NO_FAULT                                 = 0x0,
+	TC_NACK_PAGE_FAULT                               = 0x1,
+	TC_NACK_PROTECTION_FAULT                         = 0x2,
+	TC_NACK_DATA_ERROR                               = 0x3,
+} TC_NACKS;
+typedef enum TCC_PERF_SEL {
+	TCC_PERF_SEL_NONE                                = 0x0,
+	TCC_PERF_SEL_CYCLE                               = 0x1,
+	TCC_PERF_SEL_BUSY                                = 0x2,
+	TCC_PERF_SEL_REQ                                 = 0x3,
+	TCC_PERF_SEL_STREAMING_REQ                       = 0x4,
+	TCC_PERF_SEL_EXE_REQ                             = 0x5,
+	TCC_PERF_SEL_COMPRESSED_REQ                      = 0x6,
+	TCC_PERF_SEL_COMPRESSED_0_REQ                    = 0x7,
+	TCC_PERF_SEL_METADATA_REQ                        = 0x8,
+	TCC_PERF_SEL_NC_VIRTUAL_REQ                      = 0x9,
+	TCC_PERF_SEL_NC_PHYSICAL_REQ                     = 0xa,
+	TCC_PERF_SEL_UC_VIRTUAL_REQ                      = 0xb,
+	TCC_PERF_SEL_UC_PHYSICAL_REQ                     = 0xc,
+	TCC_PERF_SEL_CC_PHYSICAL_REQ                     = 0xd,
+	TCC_PERF_SEL_PROBE                               = 0xe,
+	TCC_PERF_SEL_READ                                = 0xf,
+	TCC_PERF_SEL_WRITE                               = 0x10,
+	TCC_PERF_SEL_ATOMIC                              = 0x11,
+	TCC_PERF_SEL_HIT                                 = 0x12,
+	TCC_PERF_SEL_MISS                                = 0x13,
+	TCC_PERF_SEL_DEWRITE_ALLOCATE_HIT                = 0x14,
+	TCC_PERF_SEL_FULLY_WRITTEN_HIT                   = 0x15,
+	TCC_PERF_SEL_WRITEBACK                           = 0x16,
+	TCC_PERF_SEL_LATENCY_FIFO_FULL                   = 0x17,
+	TCC_PERF_SEL_SRC_FIFO_FULL                       = 0x18,
+	TCC_PERF_SEL_HOLE_FIFO_FULL                      = 0x19,
+	TCC_PERF_SEL_MC_WRREQ                            = 0x1a,
+	TCC_PERF_SEL_MC_WRREQ_UNCACHED                   = 0x1b,
+	TCC_PERF_SEL_MC_WRREQ_STALL                      = 0x1c,
+	TCC_PERF_SEL_MC_WRREQ_CREDIT_STALL               = 0x1d,
+	TCC_PERF_SEL_MC_WRREQ_MC_HALT_STALL              = 0x1e,
+	TCC_PERF_SEL_TOO_MANY_MC_WRREQS_STALL            = 0x1f,
+	TCC_PERF_SEL_MC_WRREQ_LEVEL                      = 0x20,
+	TCC_PERF_SEL_MC_ATOMIC                           = 0x21,
+	TCC_PERF_SEL_MC_ATOMIC_LEVEL                     = 0x22,
+	TCC_PERF_SEL_MC_RDREQ                            = 0x23,
+	TCC_PERF_SEL_MC_RDREQ_UNCACHED                   = 0x24,
+	TCC_PERF_SEL_MC_RDREQ_MDC                        = 0x25,
+	TCC_PERF_SEL_MC_RDREQ_COMPRESSED                 = 0x26,
+	TCC_PERF_SEL_MC_RDREQ_CREDIT_STALL               = 0x27,
+	TCC_PERF_SEL_MC_RDREQ_MC_HALT_STALL              = 0x28,
+	TCC_PERF_SEL_MC_RDREQ_LEVEL                      = 0x29,
+	TCC_PERF_SEL_TAG_STALL                           = 0x2a,
+	TCC_PERF_SEL_TAG_WRITEBACK_FIFO_FULL_STALL       = 0x2b,
+	TCC_PERF_SEL_TAG_MISS_NOTHING_REPLACEABLE_STALL  = 0x2c,
+	TCC_PERF_SEL_TAG_UNCACHED_WRITE_ATOMIC_FIFO_FULL_STALL= 0x2d,
+	TCC_PERF_SEL_TAG_NO_UNCACHED_WRITE_ATOMIC_ENTRIES_STALL= 0x2e,
+	TCC_PERF_SEL_TAG_PROBE_STALL                     = 0x2f,
+	TCC_PERF_SEL_TAG_PROBE_FILTER_STALL              = 0x30,
+	TCC_PERF_SEL_READ_RETURN_TIMEOUT                 = 0x31,
+	TCC_PERF_SEL_WRITEBACK_READ_TIMEOUT              = 0x32,
+	TCC_PERF_SEL_READ_RETURN_FULL_BUBBLE             = 0x33,
+	TCC_PERF_SEL_BUBBLE                              = 0x34,
+	TCC_PERF_SEL_RETURN_ACK                          = 0x35,
+	TCC_PERF_SEL_RETURN_DATA                         = 0x36,
+	TCC_PERF_SEL_RETURN_HOLE                         = 0x37,
+	TCC_PERF_SEL_RETURN_ACK_HOLE                     = 0x38,
+	TCC_PERF_SEL_IB_REQ                              = 0x39,
+	TCC_PERF_SEL_IB_STALL                            = 0x3a,
+	TCC_PERF_SEL_IB_TAG_STALL                        = 0x3b,
+	TCC_PERF_SEL_IB_MDC_STALL                        = 0x3c,
+	TCC_PERF_SEL_TCA_LEVEL                           = 0x3d,
+	TCC_PERF_SEL_HOLE_LEVEL                          = 0x3e,
+	TCC_PERF_SEL_MC_RDRET_NACK                       = 0x3f,
+	TCC_PERF_SEL_MC_WRRET_NACK                       = 0x40,
+	TCC_PERF_SEL_NORMAL_WRITEBACK                    = 0x41,
+	TCC_PERF_SEL_TC_OP_WBL2_NC_WRITEBACK             = 0x42,
+	TCC_PERF_SEL_TC_OP_WBINVL2_WRITEBACK             = 0x43,
+	TCC_PERF_SEL_TC_OP_WBINVL2_NC_WRITEBACK          = 0x44,
+	TCC_PERF_SEL_TC_OP_WBINVL2_SD_WRITEBACK          = 0x45,
+	TCC_PERF_SEL_ALL_TC_OP_WB_WRITEBACK              = 0x46,
+	TCC_PERF_SEL_NORMAL_EVICT                        = 0x47,
+	TCC_PERF_SEL_TC_OP_WBL2_NC_EVICT                 = 0x48,
+	TCC_PERF_SEL_TC_OP_INVL2_NC_EVICT                = 0x49,
+	TCC_PERF_SEL_TC_OP_WBINVL2_EVICT                 = 0x4a,
+	TCC_PERF_SEL_TC_OP_WBINVL2_NC_EVICT              = 0x4b,
+	TCC_PERF_SEL_TC_OP_WBINVL2_SD_EVICT              = 0x4c,
+	TCC_PERF_SEL_ALL_TC_OP_INV_EVICT                 = 0x4d,
+	TCC_PERF_SEL_PROBE_EVICT                         = 0x4e,
+	TCC_PERF_SEL_TC_OP_WBL2_NC_CYCLE                 = 0x4f,
+	TCC_PERF_SEL_TC_OP_INVL2_NC_CYCLE                = 0x50,
+	TCC_PERF_SEL_TC_OP_WBINVL2_CYCLE                 = 0x51,
+	TCC_PERF_SEL_TC_OP_WBINVL2_NC_CYCLE              = 0x52,
+	TCC_PERF_SEL_TC_OP_WBINVL2_SD_CYCLE              = 0x53,
+	TCC_PERF_SEL_ALL_TC_OP_WB_OR_INV_CYCLE           = 0x54,
+	TCC_PERF_SEL_TC_OP_WBL2_NC_START                 = 0x55,
+	TCC_PERF_SEL_TC_OP_INVL2_NC_START                = 0x56,
+	TCC_PERF_SEL_TC_OP_WBINVL2_START                 = 0x57,
+	TCC_PERF_SEL_TC_OP_WBINVL2_NC_START              = 0x58,
+	TCC_PERF_SEL_TC_OP_WBINVL2_SD_START              = 0x59,
+	TCC_PERF_SEL_ALL_TC_OP_WB_OR_INV_START           = 0x5a,
+	TCC_PERF_SEL_TC_OP_WBL2_NC_FINISH                = 0x5b,
+	TCC_PERF_SEL_TC_OP_INVL2_NC_FINISH               = 0x5c,
+	TCC_PERF_SEL_TC_OP_WBINVL2_FINISH                = 0x5d,
+	TCC_PERF_SEL_TC_OP_WBINVL2_NC_FINISH             = 0x5e,
+	TCC_PERF_SEL_TC_OP_WBINVL2_SD_FINISH             = 0x5f,
+	TCC_PERF_SEL_ALL_TC_OP_WB_OR_INV_FINISH          = 0x60,
+	TCC_PERF_SEL_MDC_REQ                             = 0x61,
+	TCC_PERF_SEL_MDC_LEVEL                           = 0x62,
+	TCC_PERF_SEL_MDC_TAG_HIT                         = 0x63,
+	TCC_PERF_SEL_MDC_SECTOR_HIT                      = 0x64,
+	TCC_PERF_SEL_MDC_SECTOR_MISS                     = 0x65,
+	TCC_PERF_SEL_MDC_TAG_STALL                       = 0x66,
+	TCC_PERF_SEL_MDC_TAG_REPLACEMENT_LINE_IN_USE_STALL= 0x67,
+	TCC_PERF_SEL_MDC_TAG_DESECTORIZATION_FIFO_FULL_STALL= 0x68,
+	TCC_PERF_SEL_MDC_TAG_WAITING_FOR_INVALIDATE_COMPLETION_STALL= 0x69,
+	TCC_PERF_SEL_PROBE_FILTER_DISABLE_TRANSITION     = 0x6a,
+	TCC_PERF_SEL_PROBE_FILTER_DISABLED               = 0x6b,
+	TCC_PERF_SEL_CLIENT0_REQ                         = 0x80,
+	TCC_PERF_SEL_CLIENT1_REQ                         = 0x81,
+	TCC_PERF_SEL_CLIENT2_REQ                         = 0x82,
+	TCC_PERF_SEL_CLIENT3_REQ                         = 0x83,
+	TCC_PERF_SEL_CLIENT4_REQ                         = 0x84,
+	TCC_PERF_SEL_CLIENT5_REQ                         = 0x85,
+	TCC_PERF_SEL_CLIENT6_REQ                         = 0x86,
+	TCC_PERF_SEL_CLIENT7_REQ                         = 0x87,
+	TCC_PERF_SEL_CLIENT8_REQ                         = 0x88,
+	TCC_PERF_SEL_CLIENT9_REQ                         = 0x89,
+	TCC_PERF_SEL_CLIENT10_REQ                        = 0x8a,
+	TCC_PERF_SEL_CLIENT11_REQ                        = 0x8b,
+	TCC_PERF_SEL_CLIENT12_REQ                        = 0x8c,
+	TCC_PERF_SEL_CLIENT13_REQ                        = 0x8d,
+	TCC_PERF_SEL_CLIENT14_REQ                        = 0x8e,
+	TCC_PERF_SEL_CLIENT15_REQ                        = 0x8f,
+	TCC_PERF_SEL_CLIENT16_REQ                        = 0x90,
+	TCC_PERF_SEL_CLIENT17_REQ                        = 0x91,
+	TCC_PERF_SEL_CLIENT18_REQ                        = 0x92,
+	TCC_PERF_SEL_CLIENT19_REQ                        = 0x93,
+	TCC_PERF_SEL_CLIENT20_REQ                        = 0x94,
+	TCC_PERF_SEL_CLIENT21_REQ                        = 0x95,
+	TCC_PERF_SEL_CLIENT22_REQ                        = 0x96,
+	TCC_PERF_SEL_CLIENT23_REQ                        = 0x97,
+	TCC_PERF_SEL_CLIENT24_REQ                        = 0x98,
+	TCC_PERF_SEL_CLIENT25_REQ                        = 0x99,
+	TCC_PERF_SEL_CLIENT26_REQ                        = 0x9a,
+	TCC_PERF_SEL_CLIENT27_REQ                        = 0x9b,
+	TCC_PERF_SEL_CLIENT28_REQ                        = 0x9c,
+	TCC_PERF_SEL_CLIENT29_REQ                        = 0x9d,
+	TCC_PERF_SEL_CLIENT30_REQ                        = 0x9e,
+	TCC_PERF_SEL_CLIENT31_REQ                        = 0x9f,
+	TCC_PERF_SEL_CLIENT32_REQ                        = 0xa0,
+	TCC_PERF_SEL_CLIENT33_REQ                        = 0xa1,
+	TCC_PERF_SEL_CLIENT34_REQ                        = 0xa2,
+	TCC_PERF_SEL_CLIENT35_REQ                        = 0xa3,
+	TCC_PERF_SEL_CLIENT36_REQ                        = 0xa4,
+	TCC_PERF_SEL_CLIENT37_REQ                        = 0xa5,
+	TCC_PERF_SEL_CLIENT38_REQ                        = 0xa6,
+	TCC_PERF_SEL_CLIENT39_REQ                        = 0xa7,
+	TCC_PERF_SEL_CLIENT40_REQ                        = 0xa8,
+	TCC_PERF_SEL_CLIENT41_REQ                        = 0xa9,
+	TCC_PERF_SEL_CLIENT42_REQ                        = 0xaa,
+	TCC_PERF_SEL_CLIENT43_REQ                        = 0xab,
+	TCC_PERF_SEL_CLIENT44_REQ                        = 0xac,
+	TCC_PERF_SEL_CLIENT45_REQ                        = 0xad,
+	TCC_PERF_SEL_CLIENT46_REQ                        = 0xae,
+	TCC_PERF_SEL_CLIENT47_REQ                        = 0xaf,
+	TCC_PERF_SEL_CLIENT48_REQ                        = 0xb0,
+	TCC_PERF_SEL_CLIENT49_REQ                        = 0xb1,
+	TCC_PERF_SEL_CLIENT50_REQ                        = 0xb2,
+	TCC_PERF_SEL_CLIENT51_REQ                        = 0xb3,
+	TCC_PERF_SEL_CLIENT52_REQ                        = 0xb4,
+	TCC_PERF_SEL_CLIENT53_REQ                        = 0xb5,
+	TCC_PERF_SEL_CLIENT54_REQ                        = 0xb6,
+	TCC_PERF_SEL_CLIENT55_REQ                        = 0xb7,
+	TCC_PERF_SEL_CLIENT56_REQ                        = 0xb8,
+	TCC_PERF_SEL_CLIENT57_REQ                        = 0xb9,
+	TCC_PERF_SEL_CLIENT58_REQ                        = 0xba,
+	TCC_PERF_SEL_CLIENT59_REQ                        = 0xbb,
+	TCC_PERF_SEL_CLIENT60_REQ                        = 0xbc,
+	TCC_PERF_SEL_CLIENT61_REQ                        = 0xbd,
+	TCC_PERF_SEL_CLIENT62_REQ                        = 0xbe,
+	TCC_PERF_SEL_CLIENT63_REQ                        = 0xbf,
+	TCC_PERF_SEL_CLIENT64_REQ                        = 0xc0,
+	TCC_PERF_SEL_CLIENT65_REQ                        = 0xc1,
+	TCC_PERF_SEL_CLIENT66_REQ                        = 0xc2,
+	TCC_PERF_SEL_CLIENT67_REQ                        = 0xc3,
+	TCC_PERF_SEL_CLIENT68_REQ                        = 0xc4,
+	TCC_PERF_SEL_CLIENT69_REQ                        = 0xc5,
+	TCC_PERF_SEL_CLIENT70_REQ                        = 0xc6,
+	TCC_PERF_SEL_CLIENT71_REQ                        = 0xc7,
+	TCC_PERF_SEL_CLIENT72_REQ                        = 0xc8,
+	TCC_PERF_SEL_CLIENT73_REQ                        = 0xc9,
+	TCC_PERF_SEL_CLIENT74_REQ                        = 0xca,
+	TCC_PERF_SEL_CLIENT75_REQ                        = 0xcb,
+	TCC_PERF_SEL_CLIENT76_REQ                        = 0xcc,
+	TCC_PERF_SEL_CLIENT77_REQ                        = 0xcd,
+	TCC_PERF_SEL_CLIENT78_REQ                        = 0xce,
+	TCC_PERF_SEL_CLIENT79_REQ                        = 0xcf,
+	TCC_PERF_SEL_CLIENT80_REQ                        = 0xd0,
+	TCC_PERF_SEL_CLIENT81_REQ                        = 0xd1,
+	TCC_PERF_SEL_CLIENT82_REQ                        = 0xd2,
+	TCC_PERF_SEL_CLIENT83_REQ                        = 0xd3,
+	TCC_PERF_SEL_CLIENT84_REQ                        = 0xd4,
+	TCC_PERF_SEL_CLIENT85_REQ                        = 0xd5,
+	TCC_PERF_SEL_CLIENT86_REQ                        = 0xd6,
+	TCC_PERF_SEL_CLIENT87_REQ                        = 0xd7,
+	TCC_PERF_SEL_CLIENT88_REQ                        = 0xd8,
+	TCC_PERF_SEL_CLIENT89_REQ                        = 0xd9,
+	TCC_PERF_SEL_CLIENT90_REQ                        = 0xda,
+	TCC_PERF_SEL_CLIENT91_REQ                        = 0xdb,
+	TCC_PERF_SEL_CLIENT92_REQ                        = 0xdc,
+	TCC_PERF_SEL_CLIENT93_REQ                        = 0xdd,
+	TCC_PERF_SEL_CLIENT94_REQ                        = 0xde,
+	TCC_PERF_SEL_CLIENT95_REQ                        = 0xdf,
+	TCC_PERF_SEL_CLIENT96_REQ                        = 0xe0,
+	TCC_PERF_SEL_CLIENT97_REQ                        = 0xe1,
+	TCC_PERF_SEL_CLIENT98_REQ                        = 0xe2,
+	TCC_PERF_SEL_CLIENT99_REQ                        = 0xe3,
+	TCC_PERF_SEL_CLIENT100_REQ                       = 0xe4,
+	TCC_PERF_SEL_CLIENT101_REQ                       = 0xe5,
+	TCC_PERF_SEL_CLIENT102_REQ                       = 0xe6,
+	TCC_PERF_SEL_CLIENT103_REQ                       = 0xe7,
+	TCC_PERF_SEL_CLIENT104_REQ                       = 0xe8,
+	TCC_PERF_SEL_CLIENT105_REQ                       = 0xe9,
+	TCC_PERF_SEL_CLIENT106_REQ                       = 0xea,
+	TCC_PERF_SEL_CLIENT107_REQ                       = 0xeb,
+	TCC_PERF_SEL_CLIENT108_REQ                       = 0xec,
+	TCC_PERF_SEL_CLIENT109_REQ                       = 0xed,
+	TCC_PERF_SEL_CLIENT110_REQ                       = 0xee,
+	TCC_PERF_SEL_CLIENT111_REQ                       = 0xef,
+	TCC_PERF_SEL_CLIENT112_REQ                       = 0xf0,
+	TCC_PERF_SEL_CLIENT113_REQ                       = 0xf1,
+	TCC_PERF_SEL_CLIENT114_REQ                       = 0xf2,
+	TCC_PERF_SEL_CLIENT115_REQ                       = 0xf3,
+	TCC_PERF_SEL_CLIENT116_REQ                       = 0xf4,
+	TCC_PERF_SEL_CLIENT117_REQ                       = 0xf5,
+	TCC_PERF_SEL_CLIENT118_REQ                       = 0xf6,
+	TCC_PERF_SEL_CLIENT119_REQ                       = 0xf7,
+	TCC_PERF_SEL_CLIENT120_REQ                       = 0xf8,
+	TCC_PERF_SEL_CLIENT121_REQ                       = 0xf9,
+	TCC_PERF_SEL_CLIENT122_REQ                       = 0xfa,
+	TCC_PERF_SEL_CLIENT123_REQ                       = 0xfb,
+	TCC_PERF_SEL_CLIENT124_REQ                       = 0xfc,
+	TCC_PERF_SEL_CLIENT125_REQ                       = 0xfd,
+	TCC_PERF_SEL_CLIENT126_REQ                       = 0xfe,
+	TCC_PERF_SEL_CLIENT127_REQ                       = 0xff,
+} TCC_PERF_SEL;
+typedef enum TCA_PERF_SEL {
+	TCA_PERF_SEL_NONE                                = 0x0,
+	TCA_PERF_SEL_CYCLE                               = 0x1,
+	TCA_PERF_SEL_BUSY                                = 0x2,
+	TCA_PERF_SEL_FORCED_HOLE_TCC0                    = 0x3,
+	TCA_PERF_SEL_FORCED_HOLE_TCC1                    = 0x4,
+	TCA_PERF_SEL_FORCED_HOLE_TCC2                    = 0x5,
+	TCA_PERF_SEL_FORCED_HOLE_TCC3                    = 0x6,
+	TCA_PERF_SEL_FORCED_HOLE_TCC4                    = 0x7,
+	TCA_PERF_SEL_FORCED_HOLE_TCC5                    = 0x8,
+	TCA_PERF_SEL_FORCED_HOLE_TCC6                    = 0x9,
+	TCA_PERF_SEL_FORCED_HOLE_TCC7                    = 0xa,
+	TCA_PERF_SEL_REQ_TCC0                            = 0xb,
+	TCA_PERF_SEL_REQ_TCC1                            = 0xc,
+	TCA_PERF_SEL_REQ_TCC2                            = 0xd,
+	TCA_PERF_SEL_REQ_TCC3                            = 0xe,
+	TCA_PERF_SEL_REQ_TCC4                            = 0xf,
+	TCA_PERF_SEL_REQ_TCC5                            = 0x10,
+	TCA_PERF_SEL_REQ_TCC6                            = 0x11,
+	TCA_PERF_SEL_REQ_TCC7                            = 0x12,
+	TCA_PERF_SEL_CROSSBAR_DOUBLE_ARB_TCC0            = 0x13,
+	TCA_PERF_SEL_CROSSBAR_DOUBLE_ARB_TCC1            = 0x14,
+	TCA_PERF_SEL_CROSSBAR_DOUBLE_ARB_TCC2            = 0x15,
+	TCA_PERF_SEL_CROSSBAR_DOUBLE_ARB_TCC3            = 0x16,
+	TCA_PERF_SEL_CROSSBAR_DOUBLE_ARB_TCC4            = 0x17,
+	TCA_PERF_SEL_CROSSBAR_DOUBLE_ARB_TCC5            = 0x18,
+	TCA_PERF_SEL_CROSSBAR_DOUBLE_ARB_TCC6            = 0x19,
+	TCA_PERF_SEL_CROSSBAR_DOUBLE_ARB_TCC7            = 0x1a,
+	TCA_PERF_SEL_CROSSBAR_STALL_TCC0                 = 0x1b,
+	TCA_PERF_SEL_CROSSBAR_STALL_TCC1                 = 0x1c,
+	TCA_PERF_SEL_CROSSBAR_STALL_TCC2                 = 0x1d,
+	TCA_PERF_SEL_CROSSBAR_STALL_TCC3                 = 0x1e,
+	TCA_PERF_SEL_CROSSBAR_STALL_TCC4                 = 0x1f,
+	TCA_PERF_SEL_CROSSBAR_STALL_TCC5                 = 0x20,
+	TCA_PERF_SEL_CROSSBAR_STALL_TCC6                 = 0x21,
+	TCA_PERF_SEL_CROSSBAR_STALL_TCC7                 = 0x22,
+} TCA_PERF_SEL;
+typedef enum TA_TC_ADDR_MODES {
+	TA_TC_ADDR_MODE_DEFAULT                          = 0x0,
+	TA_TC_ADDR_MODE_COMP0                            = 0x1,
+	TA_TC_ADDR_MODE_COMP1                            = 0x2,
+	TA_TC_ADDR_MODE_COMP2                            = 0x3,
+	TA_TC_ADDR_MODE_COMP3                            = 0x4,
+	TA_TC_ADDR_MODE_UNALIGNED                        = 0x5,
+	TA_TC_ADDR_MODE_BORDER_COLOR                     = 0x6,
+} TA_TC_ADDR_MODES;
+typedef enum TA_PERFCOUNT_SEL {
+	TA_PERF_SEL_NULL                                 = 0x0,
+	TA_PERF_SEL_sh_fifo_busy                         = 0x1,
+	TA_PERF_SEL_sh_fifo_cmd_busy                     = 0x2,
+	TA_PERF_SEL_sh_fifo_addr_busy                    = 0x3,
+	TA_PERF_SEL_sh_fifo_data_busy                    = 0x4,
+	TA_PERF_SEL_sh_fifo_data_sfifo_busy              = 0x5,
+	TA_PERF_SEL_sh_fifo_data_tfifo_busy              = 0x6,
+	TA_PERF_SEL_gradient_busy                        = 0x7,
+	TA_PERF_SEL_gradient_fifo_busy                   = 0x8,
+	TA_PERF_SEL_lod_busy                             = 0x9,
+	TA_PERF_SEL_lod_fifo_busy                        = 0xa,
+	TA_PERF_SEL_addresser_busy                       = 0xb,
+	TA_PERF_SEL_addresser_fifo_busy                  = 0xc,
+	TA_PERF_SEL_aligner_busy                         = 0xd,
+	TA_PERF_SEL_write_path_busy                      = 0xe,
+	TA_PERF_SEL_ta_busy                              = 0xf,
+	TA_PERF_SEL_sq_ta_cmd_cycles                     = 0x10,
+	TA_PERF_SEL_sp_ta_addr_cycles                    = 0x11,
+	TA_PERF_SEL_sp_ta_data_cycles                    = 0x12,
+	TA_PERF_SEL_ta_fa_data_state_cycles              = 0x13,
+	TA_PERF_SEL_sh_fifo_addr_waiting_on_cmd_cycles   = 0x14,
+	TA_PERF_SEL_sh_fifo_cmd_waiting_on_addr_cycles   = 0x15,
+	TA_PERF_SEL_sh_fifo_addr_starved_while_busy_cycles= 0x16,
+	TA_PERF_SEL_sh_fifo_cmd_starved_while_busy_cycles= 0x17,
+	TA_PERF_SEL_sh_fifo_data_waiting_on_data_state_cycles= 0x18,
+	TA_PERF_SEL_sh_fifo_data_state_waiting_on_data_cycles= 0x19,
+	TA_PERF_SEL_sh_fifo_data_starved_while_busy_cycles= 0x1a,
+	TA_PERF_SEL_sh_fifo_data_state_starved_while_busy_cycles= 0x1b,
+	TA_PERF_SEL_RESERVED_28                          = 0x1c,
+	TA_PERF_SEL_RESERVED_29                          = 0x1d,
+	TA_PERF_SEL_sh_fifo_addr_cycles                  = 0x1e,
+	TA_PERF_SEL_sh_fifo_data_cycles                  = 0x1f,
+	TA_PERF_SEL_total_wavefronts                     = 0x20,
+	TA_PERF_SEL_gradient_cycles                      = 0x21,
+	TA_PERF_SEL_walker_cycles                        = 0x22,
+	TA_PERF_SEL_aligner_cycles                       = 0x23,
+	TA_PERF_SEL_image_wavefronts                     = 0x24,
+	TA_PERF_SEL_image_read_wavefronts                = 0x25,
+	TA_PERF_SEL_image_write_wavefronts               = 0x26,
+	TA_PERF_SEL_image_atomic_wavefronts              = 0x27,
+	TA_PERF_SEL_image_total_cycles                   = 0x28,
+	TA_PERF_SEL_RESERVED_41                          = 0x29,
+	TA_PERF_SEL_RESERVED_42                          = 0x2a,
+	TA_PERF_SEL_RESERVED_43                          = 0x2b,
+	TA_PERF_SEL_buffer_wavefronts                    = 0x2c,
+	TA_PERF_SEL_buffer_read_wavefronts               = 0x2d,
+	TA_PERF_SEL_buffer_write_wavefronts              = 0x2e,
+	TA_PERF_SEL_buffer_atomic_wavefronts             = 0x2f,
+	TA_PERF_SEL_buffer_coalescable_wavefronts        = 0x30,
+	TA_PERF_SEL_buffer_total_cycles                  = 0x31,
+	TA_PERF_SEL_buffer_coalescable_addr_multicycled_cycles= 0x32,
+	TA_PERF_SEL_buffer_coalescable_clamp_16kdword_multicycled_cycles= 0x33,
+	TA_PERF_SEL_buffer_coalesced_read_cycles         = 0x34,
+	TA_PERF_SEL_buffer_coalesced_write_cycles        = 0x35,
+	TA_PERF_SEL_addr_stalled_by_tc_cycles            = 0x36,
+	TA_PERF_SEL_addr_stalled_by_td_cycles            = 0x37,
+	TA_PERF_SEL_data_stalled_by_tc_cycles            = 0x38,
+	TA_PERF_SEL_addresser_stalled_by_aligner_only_cycles= 0x39,
+	TA_PERF_SEL_addresser_stalled_cycles             = 0x3a,
+	TA_PERF_SEL_aniso_stalled_by_addresser_only_cycles= 0x3b,
+	TA_PERF_SEL_aniso_stalled_cycles                 = 0x3c,
+	TA_PERF_SEL_deriv_stalled_by_aniso_only_cycles   = 0x3d,
+	TA_PERF_SEL_deriv_stalled_cycles                 = 0x3e,
+	TA_PERF_SEL_aniso_gt1_cycle_quads                = 0x3f,
+	TA_PERF_SEL_color_1_cycle_pixels                 = 0x40,
+	TA_PERF_SEL_color_2_cycle_pixels                 = 0x41,
+	TA_PERF_SEL_color_3_cycle_pixels                 = 0x42,
+	TA_PERF_SEL_color_4_cycle_pixels                 = 0x43,
+	TA_PERF_SEL_mip_1_cycle_pixels                   = 0x44,
+	TA_PERF_SEL_mip_2_cycle_pixels                   = 0x45,
+	TA_PERF_SEL_vol_1_cycle_pixels                   = 0x46,
+	TA_PERF_SEL_vol_2_cycle_pixels                   = 0x47,
+	TA_PERF_SEL_bilin_point_1_cycle_pixels           = 0x48,
+	TA_PERF_SEL_mipmap_lod_0_samples                 = 0x49,
+	TA_PERF_SEL_mipmap_lod_1_samples                 = 0x4a,
+	TA_PERF_SEL_mipmap_lod_2_samples                 = 0x4b,
+	TA_PERF_SEL_mipmap_lod_3_samples                 = 0x4c,
+	TA_PERF_SEL_mipmap_lod_4_samples                 = 0x4d,
+	TA_PERF_SEL_mipmap_lod_5_samples                 = 0x4e,
+	TA_PERF_SEL_mipmap_lod_6_samples                 = 0x4f,
+	TA_PERF_SEL_mipmap_lod_7_samples                 = 0x50,
+	TA_PERF_SEL_mipmap_lod_8_samples                 = 0x51,
+	TA_PERF_SEL_mipmap_lod_9_samples                 = 0x52,
+	TA_PERF_SEL_mipmap_lod_10_samples                = 0x53,
+	TA_PERF_SEL_mipmap_lod_11_samples                = 0x54,
+	TA_PERF_SEL_mipmap_lod_12_samples                = 0x55,
+	TA_PERF_SEL_mipmap_lod_13_samples                = 0x56,
+	TA_PERF_SEL_mipmap_lod_14_samples                = 0x57,
+	TA_PERF_SEL_mipmap_invalid_samples               = 0x58,
+	TA_PERF_SEL_aniso_1_cycle_quads                  = 0x59,
+	TA_PERF_SEL_aniso_2_cycle_quads                  = 0x5a,
+	TA_PERF_SEL_aniso_4_cycle_quads                  = 0x5b,
+	TA_PERF_SEL_aniso_6_cycle_quads                  = 0x5c,
+	TA_PERF_SEL_aniso_8_cycle_quads                  = 0x5d,
+	TA_PERF_SEL_aniso_10_cycle_quads                 = 0x5e,
+	TA_PERF_SEL_aniso_12_cycle_quads                 = 0x5f,
+	TA_PERF_SEL_aniso_14_cycle_quads                 = 0x60,
+	TA_PERF_SEL_aniso_16_cycle_quads                 = 0x61,
+	TA_PERF_SEL_write_path_input_cycles              = 0x62,
+	TA_PERF_SEL_write_path_output_cycles             = 0x63,
+	TA_PERF_SEL_flat_wavefronts                      = 0x64,
+	TA_PERF_SEL_flat_read_wavefronts                 = 0x65,
+	TA_PERF_SEL_flat_write_wavefronts                = 0x66,
+	TA_PERF_SEL_flat_atomic_wavefronts               = 0x67,
+	TA_PERF_SEL_flat_coalesceable_wavefronts         = 0x68,
+	TA_PERF_SEL_reg_sclk_vld                         = 0x69,
+	TA_PERF_SEL_local_cg_dyn_sclk_grp0_en            = 0x6a,
+	TA_PERF_SEL_local_cg_dyn_sclk_grp1_en            = 0x6b,
+	TA_PERF_SEL_local_cg_dyn_sclk_grp1_mems_en       = 0x6c,
+	TA_PERF_SEL_local_cg_dyn_sclk_grp4_en            = 0x6d,
+	TA_PERF_SEL_local_cg_dyn_sclk_grp5_en            = 0x6e,
+	TA_PERF_SEL_xnack_on_phase0                      = 0x6f,
+	TA_PERF_SEL_xnack_on_phase1                      = 0x70,
+	TA_PERF_SEL_xnack_on_phase2                      = 0x71,
+	TA_PERF_SEL_xnack_on_phase3                      = 0x72,
+	TA_PERF_SEL_first_xnack_on_phase0                = 0x73,
+	TA_PERF_SEL_first_xnack_on_phase1                = 0x74,
+	TA_PERF_SEL_first_xnack_on_phase2                = 0x75,
+	TA_PERF_SEL_first_xnack_on_phase3                = 0x76,
+} TA_PERFCOUNT_SEL;
+typedef enum TD_PERFCOUNT_SEL {
+	TD_PERF_SEL_none                                 = 0x0,
+	TD_PERF_SEL_td_busy                              = 0x1,
+	TD_PERF_SEL_input_busy                           = 0x2,
+	TD_PERF_SEL_output_busy                          = 0x3,
+	TD_PERF_SEL_lerp_busy                            = 0x4,
+	TD_PERF_SEL_reg_sclk_vld                         = 0x5,
+	TD_PERF_SEL_local_cg_dyn_sclk_grp0_en            = 0x6,
+	TD_PERF_SEL_local_cg_dyn_sclk_grp1_en            = 0x7,
+	TD_PERF_SEL_local_cg_dyn_sclk_grp4_en            = 0x8,
+	TD_PERF_SEL_local_cg_dyn_sclk_grp5_en            = 0x9,
+	TD_PERF_SEL_tc_td_fifo_full                      = 0xa,
+	TD_PERF_SEL_constant_state_full                  = 0xb,
+	TD_PERF_SEL_sample_state_full                    = 0xc,
+	TD_PERF_SEL_output_fifo_full                     = 0xd,
+	TD_PERF_SEL_RESERVED_14                          = 0xe,
+	TD_PERF_SEL_tc_stall                             = 0xf,
+	TD_PERF_SEL_pc_stall                             = 0x10,
+	TD_PERF_SEL_gds_stall                            = 0x11,
+	TD_PERF_SEL_RESERVED_18                          = 0x12,
+	TD_PERF_SEL_RESERVED_19                          = 0x13,
+	TD_PERF_SEL_gather4_wavefront                    = 0x14,
+	TD_PERF_SEL_sample_c_wavefront                   = 0x15,
+	TD_PERF_SEL_load_wavefront                       = 0x16,
+	TD_PERF_SEL_atomic_wavefront                     = 0x17,
+	TD_PERF_SEL_store_wavefront                      = 0x18,
+	TD_PERF_SEL_ldfptr_wavefront                     = 0x19,
+	TD_PERF_SEL_RESERVED_26                          = 0x1a,
+	TD_PERF_SEL_RESERVED_27                          = 0x1b,
+	TD_PERF_SEL_d16_en_wavefront                     = 0x1c,
+	TD_PERF_SEL_bicubic_filter_wavefront             = 0x1d,
+	TD_PERF_SEL_bypass_filter_wavefront              = 0x1e,
+	TD_PERF_SEL_min_max_filter_wavefront             = 0x1f,
+	TD_PERF_SEL_coalescable_wavefront                = 0x20,
+	TD_PERF_SEL_coalesced_phase                      = 0x21,
+	TD_PERF_SEL_four_phase_wavefront                 = 0x22,
+	TD_PERF_SEL_eight_phase_wavefront                = 0x23,
+	TD_PERF_SEL_sixteen_phase_wavefront              = 0x24,
+	TD_PERF_SEL_four_phase_forward_wavefront         = 0x25,
+	TD_PERF_SEL_write_ack_wavefront                  = 0x26,
+	TD_PERF_SEL_RESERVED_39                          = 0x27,
+	TD_PERF_SEL_user_defined_border                  = 0x28,
+	TD_PERF_SEL_white_border                         = 0x29,
+	TD_PERF_SEL_opaque_black_border                  = 0x2a,
+	TD_PERF_SEL_RESERVED_43                          = 0x2b,
+	TD_PERF_SEL_RESERVED_44                          = 0x2c,
+	TD_PERF_SEL_nack                                 = 0x2d,
+	TD_PERF_SEL_td_sp_traffic                        = 0x2e,
+	TD_PERF_SEL_consume_gds_traffic                  = 0x2f,
+	TD_PERF_SEL_addresscmd_poison                    = 0x30,
+	TD_PERF_SEL_data_poison                          = 0x31,
+	TD_PERF_SEL_start_cycle_0                        = 0x32,
+	TD_PERF_SEL_start_cycle_1                        = 0x33,
+	TD_PERF_SEL_start_cycle_2                        = 0x34,
+	TD_PERF_SEL_start_cycle_3                        = 0x35,
+	TD_PERF_SEL_null_cycle_output                    = 0x36,
+	TD_PERF_SEL_d16_data_packed                      = 0x37,
+} TD_PERFCOUNT_SEL;
+typedef enum TCP_PERFCOUNT_SELECT {
+	TCP_PERF_SEL_TA_TCP_ADDR_STARVE_CYCLES           = 0x0,
+	TCP_PERF_SEL_TA_TCP_DATA_STARVE_CYCLES           = 0x1,
+	TCP_PERF_SEL_TCP_TA_ADDR_STALL_CYCLES            = 0x2,
+	TCP_PERF_SEL_TCP_TA_DATA_STALL_CYCLES            = 0x3,
+	TCP_PERF_SEL_TD_TCP_STALL_CYCLES                 = 0x4,
+	TCP_PERF_SEL_TCR_TCP_STALL_CYCLES                = 0x5,
+	TCP_PERF_SEL_LOD_STALL_CYCLES                    = 0x6,
+	TCP_PERF_SEL_READ_TAGCONFLICT_STALL_CYCLES       = 0x7,
+	TCP_PERF_SEL_WRITE_TAGCONFLICT_STALL_CYCLES      = 0x8,
+	TCP_PERF_SEL_ATOMIC_TAGCONFLICT_STALL_CYCLES     = 0x9,
+	TCP_PERF_SEL_ALLOC_STALL_CYCLES                  = 0xa,
+	TCP_PERF_SEL_LFIFO_STALL_CYCLES                  = 0xb,
+	TCP_PERF_SEL_RFIFO_STALL_CYCLES                  = 0xc,
+	TCP_PERF_SEL_TCR_RDRET_STALL                     = 0xd,
+	TCP_PERF_SEL_WRITE_CONFLICT_STALL                = 0xe,
+	TCP_PERF_SEL_HOLE_READ_STALL                     = 0xf,
+	TCP_PERF_SEL_READCONFLICT_STALL_CYCLES           = 0x10,
+	TCP_PERF_SEL_PENDING_STALL_CYCLES                = 0x11,
+	TCP_PERF_SEL_READFIFO_STALL_CYCLES               = 0x12,
+	TCP_PERF_SEL_TCP_LATENCY                         = 0x13,
+	TCP_PERF_SEL_TCC_READ_REQ_LATENCY                = 0x14,
+	TCP_PERF_SEL_TCC_WRITE_REQ_LATENCY               = 0x15,
+	TCP_PERF_SEL_TCC_WRITE_REQ_HOLE_LATENCY          = 0x16,
+	TCP_PERF_SEL_TCC_READ_REQ                        = 0x17,
+	TCP_PERF_SEL_TCC_WRITE_REQ                       = 0x18,
+	TCP_PERF_SEL_TCC_ATOMIC_WITH_RET_REQ             = 0x19,
+	TCP_PERF_SEL_TCC_ATOMIC_WITHOUT_RET_REQ          = 0x1a,
+	TCP_PERF_SEL_TOTAL_LOCAL_READ                    = 0x1b,
+	TCP_PERF_SEL_TOTAL_GLOBAL_READ                   = 0x1c,
+	TCP_PERF_SEL_TOTAL_LOCAL_WRITE                   = 0x1d,
+	TCP_PERF_SEL_TOTAL_GLOBAL_WRITE                  = 0x1e,
+	TCP_PERF_SEL_TOTAL_ATOMIC_WITH_RET               = 0x1f,
+	TCP_PERF_SEL_TOTAL_ATOMIC_WITHOUT_RET            = 0x20,
+	TCP_PERF_SEL_TOTAL_WBINVL1                       = 0x21,
+	TCP_PERF_SEL_IMG_READ_FMT_1                      = 0x22,
+	TCP_PERF_SEL_IMG_READ_FMT_8                      = 0x23,
+	TCP_PERF_SEL_IMG_READ_FMT_16                     = 0x24,
+	TCP_PERF_SEL_IMG_READ_FMT_32                     = 0x25,
+	TCP_PERF_SEL_IMG_READ_FMT_32_AS_8                = 0x26,
+	TCP_PERF_SEL_IMG_READ_FMT_32_AS_16               = 0x27,
+	TCP_PERF_SEL_IMG_READ_FMT_32_AS_128              = 0x28,
+	TCP_PERF_SEL_IMG_READ_FMT_64_2_CYCLE             = 0x29,
+	TCP_PERF_SEL_IMG_READ_FMT_64_1_CYCLE             = 0x2a,
+	TCP_PERF_SEL_IMG_READ_FMT_96                     = 0x2b,
+	TCP_PERF_SEL_IMG_READ_FMT_128_4_CYCLE            = 0x2c,
+	TCP_PERF_SEL_IMG_READ_FMT_128_1_CYCLE            = 0x2d,
+	TCP_PERF_SEL_IMG_READ_FMT_BC1                    = 0x2e,
+	TCP_PERF_SEL_IMG_READ_FMT_BC2                    = 0x2f,
+	TCP_PERF_SEL_IMG_READ_FMT_BC3                    = 0x30,
+	TCP_PERF_SEL_IMG_READ_FMT_BC4                    = 0x31,
+	TCP_PERF_SEL_IMG_READ_FMT_BC5                    = 0x32,
+	TCP_PERF_SEL_IMG_READ_FMT_BC6                    = 0x33,
+	TCP_PERF_SEL_IMG_READ_FMT_BC7                    = 0x34,
+	TCP_PERF_SEL_IMG_READ_FMT_I8                     = 0x35,
+	TCP_PERF_SEL_IMG_READ_FMT_I16                    = 0x36,
+	TCP_PERF_SEL_IMG_READ_FMT_I32                    = 0x37,
+	TCP_PERF_SEL_IMG_READ_FMT_I32_AS_8               = 0x38,
+	TCP_PERF_SEL_IMG_READ_FMT_I32_AS_16              = 0x39,
+	TCP_PERF_SEL_IMG_READ_FMT_D8                     = 0x3a,
+	TCP_PERF_SEL_IMG_READ_FMT_D16                    = 0x3b,
+	TCP_PERF_SEL_IMG_READ_FMT_D32                    = 0x3c,
+	TCP_PERF_SEL_IMG_WRITE_FMT_8                     = 0x3d,
+	TCP_PERF_SEL_IMG_WRITE_FMT_16                    = 0x3e,
+	TCP_PERF_SEL_IMG_WRITE_FMT_32                    = 0x3f,
+	TCP_PERF_SEL_IMG_WRITE_FMT_64                    = 0x40,
+	TCP_PERF_SEL_IMG_WRITE_FMT_128                   = 0x41,
+	TCP_PERF_SEL_IMG_WRITE_FMT_D8                    = 0x42,
+	TCP_PERF_SEL_IMG_WRITE_FMT_D16                   = 0x43,
+	TCP_PERF_SEL_IMG_WRITE_FMT_D32                   = 0x44,
+	TCP_PERF_SEL_IMG_ATOMIC_WITH_RET_FMT_32          = 0x45,
+	TCP_PERF_SEL_IMG_ATOMIC_WITHOUT_RET_FMT_32       = 0x46,
+	TCP_PERF_SEL_IMG_ATOMIC_WITH_RET_FMT_64          = 0x47,
+	TCP_PERF_SEL_IMG_ATOMIC_WITHOUT_RET_FMT_64       = 0x48,
+	TCP_PERF_SEL_BUF_READ_FMT_8                      = 0x49,
+	TCP_PERF_SEL_BUF_READ_FMT_16                     = 0x4a,
+	TCP_PERF_SEL_BUF_READ_FMT_32                     = 0x4b,
+	TCP_PERF_SEL_BUF_WRITE_FMT_8                     = 0x4c,
+	TCP_PERF_SEL_BUF_WRITE_FMT_16                    = 0x4d,
+	TCP_PERF_SEL_BUF_WRITE_FMT_32                    = 0x4e,
+	TCP_PERF_SEL_BUF_ATOMIC_WITH_RET_FMT_32          = 0x4f,
+	TCP_PERF_SEL_BUF_ATOMIC_WITHOUT_RET_FMT_32       = 0x50,
+	TCP_PERF_SEL_BUF_ATOMIC_WITH_RET_FMT_64          = 0x51,
+	TCP_PERF_SEL_BUF_ATOMIC_WITHOUT_RET_FMT_64       = 0x52,
+	TCP_PERF_SEL_ARR_LINEAR_GENERAL                  = 0x53,
+	TCP_PERF_SEL_ARR_LINEAR_ALIGNED                  = 0x54,
+	TCP_PERF_SEL_ARR_1D_THIN1                        = 0x55,
+	TCP_PERF_SEL_ARR_1D_THICK                        = 0x56,
+	TCP_PERF_SEL_ARR_2D_THIN1                        = 0x57,
+	TCP_PERF_SEL_ARR_2D_THICK                        = 0x58,
+	TCP_PERF_SEL_ARR_2D_XTHICK                       = 0x59,
+	TCP_PERF_SEL_ARR_3D_THIN1                        = 0x5a,
+	TCP_PERF_SEL_ARR_3D_THICK                        = 0x5b,
+	TCP_PERF_SEL_ARR_3D_XTHICK                       = 0x5c,
+	TCP_PERF_SEL_DIM_1D                              = 0x5d,
+	TCP_PERF_SEL_DIM_2D                              = 0x5e,
+	TCP_PERF_SEL_DIM_3D                              = 0x5f,
+	TCP_PERF_SEL_DIM_1D_ARRAY                        = 0x60,
+	TCP_PERF_SEL_DIM_2D_ARRAY                        = 0x61,
+	TCP_PERF_SEL_DIM_2D_MSAA                         = 0x62,
+	TCP_PERF_SEL_DIM_2D_ARRAY_MSAA                   = 0x63,
+	TCP_PERF_SEL_DIM_CUBE_ARRAY                      = 0x64,
+	TCP_PERF_SEL_CP_TCP_INVALIDATE                   = 0x65,
+	TCP_PERF_SEL_TA_TCP_STATE_READ                   = 0x66,
+	TCP_PERF_SEL_TAGRAM0_REQ                         = 0x67,
+	TCP_PERF_SEL_TAGRAM1_REQ                         = 0x68,
+	TCP_PERF_SEL_TAGRAM2_REQ                         = 0x69,
+	TCP_PERF_SEL_TAGRAM3_REQ                         = 0x6a,
+	TCP_PERF_SEL_GATE_EN1                            = 0x6b,
+	TCP_PERF_SEL_GATE_EN2                            = 0x6c,
+	TCP_PERF_SEL_CORE_REG_SCLK_VLD                   = 0x6d,
+	TCP_PERF_SEL_TCC_REQ                             = 0x6e,
+	TCP_PERF_SEL_TCC_NON_READ_REQ                    = 0x6f,
+	TCP_PERF_SEL_TCC_BYPASS_READ_REQ                 = 0x70,
+	TCP_PERF_SEL_TCC_MISS_EVICT_READ_REQ             = 0x71,
+	TCP_PERF_SEL_TCC_VOLATILE_READ_REQ               = 0x72,
+	TCP_PERF_SEL_TCC_VOLATILE_BYPASS_READ_REQ        = 0x73,
+	TCP_PERF_SEL_TCC_VOLATILE_MISS_EVICT_READ_REQ    = 0x74,
+	TCP_PERF_SEL_TCC_BYPASS_WRITE_REQ                = 0x75,
+	TCP_PERF_SEL_TCC_MISS_EVICT_WRITE_REQ            = 0x76,
+	TCP_PERF_SEL_TCC_VOLATILE_BYPASS_WRITE_REQ       = 0x77,
+	TCP_PERF_SEL_TCC_VOLATILE_WRITE_REQ              = 0x78,
+	TCP_PERF_SEL_TCC_VOLATILE_MISS_EVICT_WRITE_REQ   = 0x79,
+	TCP_PERF_SEL_TCC_BYPASS_ATOMIC_REQ               = 0x7a,
+	TCP_PERF_SEL_TCC_ATOMIC_REQ                      = 0x7b,
+	TCP_PERF_SEL_TCC_VOLATILE_ATOMIC_REQ             = 0x7c,
+	TCP_PERF_SEL_TCC_DATA_BUS_BUSY                   = 0x7d,
+	TCP_PERF_SEL_TOTAL_ACCESSES                      = 0x7e,
+	TCP_PERF_SEL_TOTAL_READ                          = 0x7f,
+	TCP_PERF_SEL_TOTAL_HIT_LRU_READ                  = 0x80,
+	TCP_PERF_SEL_TOTAL_HIT_EVICT_READ                = 0x81,
+	TCP_PERF_SEL_TOTAL_MISS_LRU_READ                 = 0x82,
+	TCP_PERF_SEL_TOTAL_MISS_EVICT_READ               = 0x83,
+	TCP_PERF_SEL_TOTAL_NON_READ                      = 0x84,
+	TCP_PERF_SEL_TOTAL_WRITE                         = 0x85,
+	TCP_PERF_SEL_TOTAL_MISS_LRU_WRITE                = 0x86,
+	TCP_PERF_SEL_TOTAL_MISS_EVICT_WRITE              = 0x87,
+	TCP_PERF_SEL_TOTAL_WBINVL1_VOL                   = 0x88,
+	TCP_PERF_SEL_TOTAL_WRITEBACK_INVALIDATES         = 0x89,
+	TCP_PERF_SEL_DISPLAY_MICROTILING                 = 0x8a,
+	TCP_PERF_SEL_THIN_MICROTILING                    = 0x8b,
+	TCP_PERF_SEL_DEPTH_MICROTILING                   = 0x8c,
+	TCP_PERF_SEL_ARR_PRT_THIN1                       = 0x8d,
+	TCP_PERF_SEL_ARR_PRT_2D_THIN1                    = 0x8e,
+	TCP_PERF_SEL_ARR_PRT_3D_THIN1                    = 0x8f,
+	TCP_PERF_SEL_ARR_PRT_THICK                       = 0x90,
+	TCP_PERF_SEL_ARR_PRT_2D_THICK                    = 0x91,
+	TCP_PERF_SEL_ARR_PRT_3D_THICK                    = 0x92,
+	TCP_PERF_SEL_CP_TCP_INVALIDATE_VOL               = 0x93,
+	TCP_PERF_SEL_SQ_TCP_INVALIDATE_VOL               = 0x94,
+	TCP_PERF_SEL_UNALIGNED                           = 0x95,
+	TCP_PERF_SEL_ROTATED_MICROTILING                 = 0x96,
+	TCP_PERF_SEL_THICK_MICROTILING                   = 0x97,
+	TCP_PERF_SEL_ATC                                 = 0x98,
+	TCP_PERF_SEL_POWER_STALL                         = 0x99,
+	TCP_PERF_SEL_RESERVED_154                        = 0x9a,
+	TCP_PERF_SEL_TCC_LRU_REQ                         = 0x9b,
+	TCP_PERF_SEL_TCC_STREAM_REQ                      = 0x9c,
+	TCP_PERF_SEL_TCC_NC_READ_REQ                     = 0x9d,
+	TCP_PERF_SEL_TCC_NC_WRITE_REQ                    = 0x9e,
+	TCP_PERF_SEL_TCC_NC_ATOMIC_REQ                   = 0x9f,
+	TCP_PERF_SEL_TCC_UC_READ_REQ                     = 0xa0,
+	TCP_PERF_SEL_TCC_UC_WRITE_REQ                    = 0xa1,
+	TCP_PERF_SEL_TCC_UC_ATOMIC_REQ                   = 0xa2,
+	TCP_PERF_SEL_TCC_CC_READ_REQ                     = 0xa3,
+	TCP_PERF_SEL_TCC_CC_WRITE_REQ                    = 0xa4,
+	TCP_PERF_SEL_TCC_CC_ATOMIC_REQ                   = 0xa5,
+	TCP_PERF_SEL_TCC_DCC_REQ                         = 0xa6,
+	TCP_PERF_SEL_TCC_PHYSICAL_REQ                    = 0xa7,
+	TCP_PERF_SEL_UNORDERED_MTYPE_STALL               = 0xa8,
+	TCP_PERF_SEL_VOLATILE                            = 0xa9,
+	TCP_PERF_SEL_TC_TA_XNACK_STALL                   = 0xaa,
+	TCP_PERF_SEL_ATCL1_SERIALIZATION_STALL           = 0xab,
+	TCP_PERF_SEL_SHOOTDOWN                           = 0xac,
+	TCP_PERF_SEL_GATCL1_TRANSLATION_MISS             = 0xad,
+	TCP_PERF_SEL_GATCL1_PERMISSION_MISS              = 0xae,
+	TCP_PERF_SEL_GATCL1_REQUEST                      = 0xaf,
+	TCP_PERF_SEL_GATCL1_STALL_INFLIGHT_MAX           = 0xb0,
+	TCP_PERF_SEL_GATCL1_STALL_LRU_INFLIGHT           = 0xb1,
+	TCP_PERF_SEL_GATCL1_LFIFO_FULL                   = 0xb2,
+	TCP_PERF_SEL_GATCL1_STALL_LFIFO_NOT_RES          = 0xb3,
+	TCP_PERF_SEL_GATCL1_STALL_ATCL2_REQ_OUT_OF_CREDITS= 0xb4,
+	TCP_PERF_SEL_GATCL1_ATCL2_INFLIGHT               = 0xb5,
+	TCP_PERF_SEL_GATCL1_STALL_MISSFIFO_FULL          = 0xb6,
+	TCP_PERF_SEL_IMG_READ_FMT_ETC2_RGB               = 0xb7,
+	TCP_PERF_SEL_IMG_READ_FMT_ETC2_RGBA              = 0xb8,
+	TCP_PERF_SEL_IMG_READ_FMT_ETC2_RGBA1             = 0xb9,
+	TCP_PERF_SEL_IMG_READ_FMT_ETC2_R                 = 0xba,
+	TCP_PERF_SEL_IMG_READ_FMT_ETC2_RG                = 0xbb,
+	TCP_PERF_SEL_IMG_READ_FMT_8_AS_32                = 0xbc,
+	TCP_PERF_SEL_IMG_READ_FMT_8_AS_64                = 0xbd,
+	TCP_PERF_SEL_IMG_READ_FMT_16_AS_64               = 0xbe,
+	TCP_PERF_SEL_IMG_READ_FMT_16_AS_128              = 0xbf,
+	TCP_PERF_SEL_IMG_WRITE_FMT_8_AS_32               = 0xc0,
+	TCP_PERF_SEL_IMG_WRITE_FMT_8_AS_64               = 0xc1,
+	TCP_PERF_SEL_IMG_WRITE_FMT_16_AS_64              = 0xc2,
+	TCP_PERF_SEL_IMG_WRITE_FMT_16_AS_128             = 0xc3,
+} TCP_PERFCOUNT_SELECT;
+typedef enum TCP_CACHE_POLICIES {
+	TCP_CACHE_POLICY_MISS_LRU                        = 0x0,
+	TCP_CACHE_POLICY_MISS_EVICT                      = 0x1,
+	TCP_CACHE_POLICY_HIT_LRU                         = 0x2,
+	TCP_CACHE_POLICY_HIT_EVICT                       = 0x3,
+} TCP_CACHE_POLICIES;
+typedef enum TCP_CACHE_STORE_POLICIES {
+	TCP_CACHE_STORE_POLICY_WT_LRU                    = 0x0,
+	TCP_CACHE_STORE_POLICY_WT_EVICT                  = 0x1,
+} TCP_CACHE_STORE_POLICIES;
+typedef enum TCP_WATCH_MODES {
+	TCP_WATCH_MODE_READ                              = 0x0,
+	TCP_WATCH_MODE_NONREAD                           = 0x1,
+	TCP_WATCH_MODE_ATOMIC                            = 0x2,
+	TCP_WATCH_MODE_ALL                               = 0x3,
+} TCP_WATCH_MODES;
+typedef enum TCP_DSM_DATA_SEL {
+	TCP_DSM_DISABLE                                  = 0x0,
+	TCP_DSM_SEL0                                     = 0x1,
+	TCP_DSM_SEL1                                     = 0x2,
+	TCP_DSM_SEL_BOTH                                 = 0x3,
+} TCP_DSM_DATA_SEL;
+typedef enum TCP_DSM_SINGLE_WRITE {
+	TCP_DSM_SINGLE_WRITE_EN                          = 0x1,
+} TCP_DSM_SINGLE_WRITE;
+typedef enum VGT_OUT_PRIM_TYPE {
+	VGT_OUT_POINT                                    = 0x0,
+	VGT_OUT_LINE                                     = 0x1,
+	VGT_OUT_TRI                                      = 0x2,
+	VGT_OUT_RECT_V0                                  = 0x3,
+	VGT_OUT_RECT_V1                                  = 0x4,
+	VGT_OUT_RECT_V2                                  = 0x5,
+	VGT_OUT_RECT_V3                                  = 0x6,
+	VGT_OUT_RESERVED                                 = 0x7,
+	VGT_TE_QUAD                                      = 0x8,
+	VGT_TE_PRIM_INDEX_LINE                           = 0x9,
+	VGT_TE_PRIM_INDEX_TRI                            = 0xa,
+	VGT_TE_PRIM_INDEX_QUAD                           = 0xb,
+	VGT_OUT_LINE_ADJ                                 = 0xc,
+	VGT_OUT_TRI_ADJ                                  = 0xd,
+	VGT_OUT_PATCH                                    = 0xe,
+} VGT_OUT_PRIM_TYPE;
+typedef enum VGT_DI_PRIM_TYPE {
+	DI_PT_NONE                                       = 0x0,
+	DI_PT_POINTLIST                                  = 0x1,
+	DI_PT_LINELIST                                   = 0x2,
+	DI_PT_LINESTRIP                                  = 0x3,
+	DI_PT_TRILIST                                    = 0x4,
+	DI_PT_TRIFAN                                     = 0x5,
+	DI_PT_TRISTRIP                                   = 0x6,
+	DI_PT_UNUSED_0                                   = 0x7,
+	DI_PT_UNUSED_1                                   = 0x8,
+	DI_PT_PATCH                                      = 0x9,
+	DI_PT_LINELIST_ADJ                               = 0xa,
+	DI_PT_LINESTRIP_ADJ                              = 0xb,
+	DI_PT_TRILIST_ADJ                                = 0xc,
+	DI_PT_TRISTRIP_ADJ                               = 0xd,
+	DI_PT_UNUSED_3                                   = 0xe,
+	DI_PT_UNUSED_4                                   = 0xf,
+	DI_PT_TRI_WITH_WFLAGS                            = 0x10,
+	DI_PT_RECTLIST                                   = 0x11,
+	DI_PT_LINELOOP                                   = 0x12,
+	DI_PT_QUADLIST                                   = 0x13,
+	DI_PT_QUADSTRIP                                  = 0x14,
+	DI_PT_POLYGON                                    = 0x15,
+	DI_PT_2D_COPY_RECT_LIST_V0                       = 0x16,
+	DI_PT_2D_COPY_RECT_LIST_V1                       = 0x17,
+	DI_PT_2D_COPY_RECT_LIST_V2                       = 0x18,
+	DI_PT_2D_COPY_RECT_LIST_V3                       = 0x19,
+	DI_PT_2D_FILL_RECT_LIST                          = 0x1a,
+	DI_PT_2D_LINE_STRIP                              = 0x1b,
+	DI_PT_2D_TRI_STRIP                               = 0x1c,
+} VGT_DI_PRIM_TYPE;
+typedef enum VGT_DI_SOURCE_SELECT {
+	DI_SRC_SEL_DMA                                   = 0x0,
+	DI_SRC_SEL_IMMEDIATE                             = 0x1,
+	DI_SRC_SEL_AUTO_INDEX                            = 0x2,
+	DI_SRC_SEL_RESERVED                              = 0x3,
+} VGT_DI_SOURCE_SELECT;
+typedef enum VGT_DI_MAJOR_MODE_SELECT {
+	DI_MAJOR_MODE_0                                  = 0x0,
+	DI_MAJOR_MODE_1                                  = 0x1,
+} VGT_DI_MAJOR_MODE_SELECT;
+typedef enum VGT_DI_INDEX_SIZE {
+	DI_INDEX_SIZE_16_BIT                             = 0x0,
+	DI_INDEX_SIZE_32_BIT                             = 0x1,
+	DI_INDEX_SIZE_8_BIT                              = 0x2,
+} VGT_DI_INDEX_SIZE;
+typedef enum VGT_EVENT_TYPE {
+	Reserved_0x00                                    = 0x0,
+	SAMPLE_STREAMOUTSTATS1                           = 0x1,
+	SAMPLE_STREAMOUTSTATS2                           = 0x2,
+	SAMPLE_STREAMOUTSTATS3                           = 0x3,
+	CACHE_FLUSH_TS                                   = 0x4,
+	CONTEXT_DONE                                     = 0x5,
+	CACHE_FLUSH                                      = 0x6,
+	CS_PARTIAL_FLUSH                                 = 0x7,
+	VGT_STREAMOUT_SYNC                               = 0x8,
+	Reserved_0x09                                    = 0x9,
+	VGT_STREAMOUT_RESET                              = 0xa,
+	END_OF_PIPE_INCR_DE                              = 0xb,
+	END_OF_PIPE_IB_END                               = 0xc,
+	RST_PIX_CNT                                      = 0xd,
+	Reserved_0x0E                                    = 0xe,
+	VS_PARTIAL_FLUSH                                 = 0xf,
+	PS_PARTIAL_FLUSH                                 = 0x10,
+	FLUSH_HS_OUTPUT                                  = 0x11,
+	FLUSH_LS_OUTPUT                                  = 0x12,
+	Reserved_0x13                                    = 0x13,
+	CACHE_FLUSH_AND_INV_TS_EVENT                     = 0x14,
+	ZPASS_DONE                                       = 0x15,
+	CACHE_FLUSH_AND_INV_EVENT                        = 0x16,
+	PERFCOUNTER_START                                = 0x17,
+	PERFCOUNTER_STOP                                 = 0x18,
+	PIPELINESTAT_START                               = 0x19,
+	PIPELINESTAT_STOP                                = 0x1a,
+	PERFCOUNTER_SAMPLE                               = 0x1b,
+	FLUSH_ES_OUTPUT                                  = 0x1c,
+	FLUSH_GS_OUTPUT                                  = 0x1d,
+	SAMPLE_PIPELINESTAT                              = 0x1e,
+	SO_VGTSTREAMOUT_FLUSH                            = 0x1f,
+	SAMPLE_STREAMOUTSTATS                            = 0x20,
+	RESET_VTX_CNT                                    = 0x21,
+	BLOCK_CONTEXT_DONE                               = 0x22,
+	CS_CONTEXT_DONE                                  = 0x23,
+	VGT_FLUSH                                        = 0x24,
+	TGID_ROLLOVER                                    = 0x25,
+	SQ_NON_EVENT                                     = 0x26,
+	SC_SEND_DB_VPZ                                   = 0x27,
+	BOTTOM_OF_PIPE_TS                                = 0x28,
+	FLUSH_SX_TS                                      = 0x29,
+	DB_CACHE_FLUSH_AND_INV                           = 0x2a,
+	FLUSH_AND_INV_DB_DATA_TS                         = 0x2b,
+	FLUSH_AND_INV_DB_META                            = 0x2c,
+	FLUSH_AND_INV_CB_DATA_TS                         = 0x2d,
+	FLUSH_AND_INV_CB_META                            = 0x2e,
+	CS_DONE                                          = 0x2f,
+	PS_DONE                                          = 0x30,
+	FLUSH_AND_INV_CB_PIXEL_DATA                      = 0x31,
+	SX_CB_RAT_ACK_REQUEST                            = 0x32,
+	THREAD_TRACE_START                               = 0x33,
+	THREAD_TRACE_STOP                                = 0x34,
+	THREAD_TRACE_MARKER                              = 0x35,
+	THREAD_TRACE_FLUSH                               = 0x36,
+	THREAD_TRACE_FINISH                              = 0x37,
+	PIXEL_PIPE_STAT_CONTROL                          = 0x38,
+	PIXEL_PIPE_STAT_DUMP                             = 0x39,
+	PIXEL_PIPE_STAT_RESET                            = 0x3a,
+	CONTEXT_SUSPEND                                  = 0x3b,
+	OFFCHIP_HS_DEALLOC                               = 0x3c,
+} VGT_EVENT_TYPE;
+typedef enum VGT_DMA_SWAP_MODE {
+	VGT_DMA_SWAP_NONE                                = 0x0,
+	VGT_DMA_SWAP_16_BIT                              = 0x1,
+	VGT_DMA_SWAP_32_BIT                              = 0x2,
+	VGT_DMA_SWAP_WORD                                = 0x3,
+} VGT_DMA_SWAP_MODE;
+typedef enum VGT_INDEX_TYPE_MODE {
+	VGT_INDEX_16                                     = 0x0,
+	VGT_INDEX_32                                     = 0x1,
+	VGT_INDEX_8                                      = 0x2,
+} VGT_INDEX_TYPE_MODE;
+typedef enum VGT_DMA_BUF_TYPE {
+	VGT_DMA_BUF_MEM                                  = 0x0,
+	VGT_DMA_BUF_RING                                 = 0x1,
+	VGT_DMA_BUF_SETUP                                = 0x2,
+	VGT_DMA_PTR_UPDATE                               = 0x3,
+} VGT_DMA_BUF_TYPE;
+typedef enum VGT_OUTPATH_SELECT {
+	VGT_OUTPATH_VTX_REUSE                            = 0x0,
+	VGT_OUTPATH_TESS_EN                              = 0x1,
+	VGT_OUTPATH_PASSTHRU                             = 0x2,
+	VGT_OUTPATH_GS_BLOCK                             = 0x3,
+	VGT_OUTPATH_HS_BLOCK                             = 0x4,
+} VGT_OUTPATH_SELECT;
+typedef enum VGT_GRP_PRIM_TYPE {
+	VGT_GRP_3D_POINT                                 = 0x0,
+	VGT_GRP_3D_LINE                                  = 0x1,
+	VGT_GRP_3D_TRI                                   = 0x2,
+	VGT_GRP_3D_RECT                                  = 0x3,
+	VGT_GRP_3D_QUAD                                  = 0x4,
+	VGT_GRP_2D_COPY_RECT_V0                          = 0x5,
+	VGT_GRP_2D_COPY_RECT_V1                          = 0x6,
+	VGT_GRP_2D_COPY_RECT_V2                          = 0x7,
+	VGT_GRP_2D_COPY_RECT_V3                          = 0x8,
+	VGT_GRP_2D_FILL_RECT                             = 0x9,
+	VGT_GRP_2D_LINE                                  = 0xa,
+	VGT_GRP_2D_TRI                                   = 0xb,
+	VGT_GRP_PRIM_INDEX_LINE                          = 0xc,
+	VGT_GRP_PRIM_INDEX_TRI                           = 0xd,
+	VGT_GRP_PRIM_INDEX_QUAD                          = 0xe,
+	VGT_GRP_3D_LINE_ADJ                              = 0xf,
+	VGT_GRP_3D_TRI_ADJ                               = 0x10,
+	VGT_GRP_3D_PATCH                                 = 0x11,
+} VGT_GRP_PRIM_TYPE;
+typedef enum VGT_GRP_PRIM_ORDER {
+	VGT_GRP_LIST                                     = 0x0,
+	VGT_GRP_STRIP                                    = 0x1,
+	VGT_GRP_FAN                                      = 0x2,
+	VGT_GRP_LOOP                                     = 0x3,
+	VGT_GRP_POLYGON                                  = 0x4,
+} VGT_GRP_PRIM_ORDER;
+typedef enum VGT_GROUP_CONV_SEL {
+	VGT_GRP_INDEX_16                                 = 0x0,
+	VGT_GRP_INDEX_32                                 = 0x1,
+	VGT_GRP_UINT_16                                  = 0x2,
+	VGT_GRP_UINT_32                                  = 0x3,
+	VGT_GRP_SINT_16                                  = 0x4,
+	VGT_GRP_SINT_32                                  = 0x5,
+	VGT_GRP_FLOAT_32                                 = 0x6,
+	VGT_GRP_AUTO_PRIM                                = 0x7,
+	VGT_GRP_FIX_1_23_TO_FLOAT                        = 0x8,
+} VGT_GROUP_CONV_SEL;
+typedef enum VGT_GS_MODE_TYPE {
+	GS_OFF                                           = 0x0,
+	GS_SCENARIO_A                                    = 0x1,
+	GS_SCENARIO_B                                    = 0x2,
+	GS_SCENARIO_G                                    = 0x3,
+	GS_SCENARIO_C                                    = 0x4,
+	SPRITE_EN                                        = 0x5,
+} VGT_GS_MODE_TYPE;
+typedef enum VGT_GS_CUT_MODE {
+	GS_CUT_1024                                      = 0x0,
+	GS_CUT_512                                       = 0x1,
+	GS_CUT_256                                       = 0x2,
+	GS_CUT_128                                       = 0x3,
+} VGT_GS_CUT_MODE;
+typedef enum VGT_GS_OUTPRIM_TYPE {
+	POINTLIST                                        = 0x0,
+	LINESTRIP                                        = 0x1,
+	TRISTRIP                                         = 0x2,
+} VGT_GS_OUTPRIM_TYPE;
+typedef enum VGT_CACHE_INVALID_MODE {
+	VC_ONLY                                          = 0x0,
+	TC_ONLY                                          = 0x1,
+	VC_AND_TC                                        = 0x2,
+} VGT_CACHE_INVALID_MODE;
+typedef enum VGT_TESS_TYPE {
+	TESS_ISOLINE                                     = 0x0,
+	TESS_TRIANGLE                                    = 0x1,
+	TESS_QUAD                                        = 0x2,
+} VGT_TESS_TYPE;
+typedef enum VGT_TESS_PARTITION {
+	PART_INTEGER                                     = 0x0,
+	PART_POW2                                        = 0x1,
+	PART_FRAC_ODD                                    = 0x2,
+	PART_FRAC_EVEN                                   = 0x3,
+} VGT_TESS_PARTITION;
+typedef enum VGT_TESS_TOPOLOGY {
+	OUTPUT_POINT                                     = 0x0,
+	OUTPUT_LINE                                      = 0x1,
+	OUTPUT_TRIANGLE_CW                               = 0x2,
+	OUTPUT_TRIANGLE_CCW                              = 0x3,
+} VGT_TESS_TOPOLOGY;
+typedef enum VGT_RDREQ_POLICY {
+	VGT_POLICY_LRU                                   = 0x0,
+	VGT_POLICY_STREAM                                = 0x1,
+} VGT_RDREQ_POLICY;
+typedef enum VGT_DIST_MODE {
+	NO_DIST                                          = 0x0,
+	PATCHES                                          = 0x1,
+	DONUTS                                           = 0x2,
+} VGT_DIST_MODE;
+typedef enum VGT_STAGES_LS_EN {
+	LS_STAGE_OFF                                     = 0x0,
+	LS_STAGE_ON                                      = 0x1,
+	CS_STAGE_ON                                      = 0x2,
+	RESERVED_LS                                      = 0x3,
+} VGT_STAGES_LS_EN;
+typedef enum VGT_STAGES_HS_EN {
+	HS_STAGE_OFF                                     = 0x0,
+	HS_STAGE_ON                                      = 0x1,
+} VGT_STAGES_HS_EN;
+typedef enum VGT_STAGES_ES_EN {
+	ES_STAGE_OFF                                     = 0x0,
+	ES_STAGE_DS                                      = 0x1,
+	ES_STAGE_REAL                                    = 0x2,
+	RESERVED_ES                                      = 0x3,
+} VGT_STAGES_ES_EN;
+typedef enum VGT_STAGES_GS_EN {
+	GS_STAGE_OFF                                     = 0x0,
+	GS_STAGE_ON                                      = 0x1,
+} VGT_STAGES_GS_EN;
+typedef enum VGT_STAGES_VS_EN {
+	VS_STAGE_REAL                                    = 0x0,
+	VS_STAGE_DS                                      = 0x1,
+	VS_STAGE_COPY_SHADER                             = 0x2,
+	RESERVED_VS                                      = 0x3,
+} VGT_STAGES_VS_EN;
+typedef enum VGT_PERFCOUNT_SELECT {
+	vgt_perf_VGT_SPI_ESTHREAD_EVENT_WINDOW_ACTIVE    = 0x0,
+	vgt_perf_VGT_SPI_ESVERT_VALID                    = 0x1,
+	vgt_perf_VGT_SPI_ESVERT_EOV                      = 0x2,
+	vgt_perf_VGT_SPI_ESVERT_STALLED                  = 0x3,
+	vgt_perf_VGT_SPI_ESVERT_STARVED_BUSY             = 0x4,
+	vgt_perf_VGT_SPI_ESVERT_STARVED_IDLE             = 0x5,
+	vgt_perf_VGT_SPI_ESVERT_STATIC                   = 0x6,
+	vgt_perf_VGT_SPI_ESTHREAD_IS_EVENT               = 0x7,
+	vgt_perf_VGT_SPI_ESTHREAD_SEND                   = 0x8,
+	vgt_perf_VGT_SPI_GSPRIM_VALID                    = 0x9,
+	vgt_perf_VGT_SPI_GSPRIM_EOV                      = 0xa,
+	vgt_perf_VGT_SPI_GSPRIM_CONT                     = 0xb,
+	vgt_perf_VGT_SPI_GSPRIM_STALLED                  = 0xc,
+	vgt_perf_VGT_SPI_GSPRIM_STARVED_BUSY             = 0xd,
+	vgt_perf_VGT_SPI_GSPRIM_STARVED_IDLE             = 0xe,
+	vgt_perf_VGT_SPI_GSPRIM_STATIC                   = 0xf,
+	vgt_perf_VGT_SPI_GSTHREAD_EVENT_WINDOW_ACTIVE    = 0x10,
+	vgt_perf_VGT_SPI_GSTHREAD_IS_EVENT               = 0x11,
+	vgt_perf_VGT_SPI_GSTHREAD_SEND                   = 0x12,
+	vgt_perf_VGT_SPI_VSTHREAD_EVENT_WINDOW_ACTIVE    = 0x13,
+	vgt_perf_VGT_SPI_VSVERT_SEND                     = 0x14,
+	vgt_perf_VGT_SPI_VSVERT_EOV                      = 0x15,
+	vgt_perf_VGT_SPI_VSVERT_STALLED                  = 0x16,
+	vgt_perf_VGT_SPI_VSVERT_STARVED_BUSY             = 0x17,
+	vgt_perf_VGT_SPI_VSVERT_STARVED_IDLE             = 0x18,
+	vgt_perf_VGT_SPI_VSVERT_STATIC                   = 0x19,
+	vgt_perf_VGT_SPI_VSTHREAD_IS_EVENT               = 0x1a,
+	vgt_perf_VGT_SPI_VSTHREAD_SEND                   = 0x1b,
+	vgt_perf_VGT_PA_EVENT_WINDOW_ACTIVE              = 0x1c,
+	vgt_perf_VGT_PA_CLIPV_SEND                       = 0x1d,
+	vgt_perf_VGT_PA_CLIPV_FIRSTVERT                  = 0x1e,
+	vgt_perf_VGT_PA_CLIPV_STALLED                    = 0x1f,
+	vgt_perf_VGT_PA_CLIPV_STARVED_BUSY               = 0x20,
+	vgt_perf_VGT_PA_CLIPV_STARVED_IDLE               = 0x21,
+	vgt_perf_VGT_PA_CLIPV_STATIC                     = 0x22,
+	vgt_perf_VGT_PA_CLIPP_SEND                       = 0x23,
+	vgt_perf_VGT_PA_CLIPP_EOP                        = 0x24,
+	vgt_perf_VGT_PA_CLIPP_IS_EVENT                   = 0x25,
+	vgt_perf_VGT_PA_CLIPP_NULL_PRIM                  = 0x26,
+	vgt_perf_VGT_PA_CLIPP_NEW_VTX_VECT               = 0x27,
+	vgt_perf_VGT_PA_CLIPP_STALLED                    = 0x28,
+	vgt_perf_VGT_PA_CLIPP_STARVED_BUSY               = 0x29,
+	vgt_perf_VGT_PA_CLIPP_STARVED_IDLE               = 0x2a,
+	vgt_perf_VGT_PA_CLIPP_STATIC                     = 0x2b,
+	vgt_perf_VGT_PA_CLIPS_SEND                       = 0x2c,
+	vgt_perf_VGT_PA_CLIPS_STALLED                    = 0x2d,
+	vgt_perf_VGT_PA_CLIPS_STARVED_BUSY               = 0x2e,
+	vgt_perf_VGT_PA_CLIPS_STARVED_IDLE               = 0x2f,
+	vgt_perf_VGT_PA_CLIPS_STATIC                     = 0x30,
+	vgt_perf_vsvert_ds_send                          = 0x31,
+	vgt_perf_vsvert_api_send                         = 0x32,
+	vgt_perf_hs_tif_stall                            = 0x33,
+	vgt_perf_hs_input_stall                          = 0x34,
+	vgt_perf_hs_interface_stall                      = 0x35,
+	vgt_perf_hs_tfm_stall                            = 0x36,
+	vgt_perf_te11_starved                            = 0x37,
+	vgt_perf_gs_event_stall                          = 0x38,
+	vgt_perf_vgt_pa_clipp_send_not_event             = 0x39,
+	vgt_perf_vgt_pa_clipp_valid_prim                 = 0x3a,
+	vgt_perf_reused_es_indices                       = 0x3b,
+	vgt_perf_vs_cache_hits                           = 0x3c,
+	vgt_perf_gs_cache_hits                           = 0x3d,
+	vgt_perf_ds_cache_hits                           = 0x3e,
+	vgt_perf_total_cache_hits                        = 0x3f,
+	vgt_perf_vgt_busy                                = 0x40,
+	vgt_perf_vgt_gs_busy                             = 0x41,
+	vgt_perf_esvert_stalled_es_tbl                   = 0x42,
+	vgt_perf_esvert_stalled_gs_tbl                   = 0x43,
+	vgt_perf_esvert_stalled_gs_event                 = 0x44,
+	vgt_perf_esvert_stalled_gsprim                   = 0x45,
+	vgt_perf_gsprim_stalled_es_tbl                   = 0x46,
+	vgt_perf_gsprim_stalled_gs_tbl                   = 0x47,
+	vgt_perf_gsprim_stalled_gs_event                 = 0x48,
+	vgt_perf_gsprim_stalled_esvert                   = 0x49,
+	vgt_perf_esthread_stalled_es_rb_full             = 0x4a,
+	vgt_perf_esthread_stalled_spi_bp                 = 0x4b,
+	vgt_perf_counters_avail_stalled                  = 0x4c,
+	vgt_perf_gs_rb_space_avail_stalled               = 0x4d,
+	vgt_perf_gs_issue_rtr_stalled                    = 0x4e,
+	vgt_perf_gsthread_stalled                        = 0x4f,
+	vgt_perf_strmout_stalled                         = 0x50,
+	vgt_perf_wait_for_es_done_stalled                = 0x51,
+	vgt_perf_cm_stalled_by_gog                       = 0x52,
+	vgt_perf_cm_reading_stalled                      = 0x53,
+	vgt_perf_cm_stalled_by_gsfetch_done              = 0x54,
+	vgt_perf_gog_vs_tbl_stalled                      = 0x55,
+	vgt_perf_gog_out_indx_stalled                    = 0x56,
+	vgt_perf_gog_out_prim_stalled                    = 0x57,
+	vgt_perf_waveid_stalled                          = 0x58,
+	vgt_perf_gog_busy                                = 0x59,
+	vgt_perf_reused_vs_indices                       = 0x5a,
+	vgt_perf_sclk_reg_vld_event                      = 0x5b,
+	vgt_perf_vs_conflicting_indices                  = 0x5c,
+	vgt_perf_sclk_core_vld_event                     = 0x5d,
+	vgt_perf_hswave_stalled                          = 0x5e,
+	vgt_perf_sclk_gs_vld_event                       = 0x5f,
+	vgt_perf_VGT_SPI_LSVERT_VALID                    = 0x60,
+	vgt_perf_VGT_SPI_LSVERT_EOV                      = 0x61,
+	vgt_perf_VGT_SPI_LSVERT_STALLED                  = 0x62,
+	vgt_perf_VGT_SPI_LSVERT_STARVED_BUSY             = 0x63,
+	vgt_perf_VGT_SPI_LSVERT_STARVED_IDLE             = 0x64,
+	vgt_perf_VGT_SPI_LSVERT_STATIC                   = 0x65,
+	vgt_perf_VGT_SPI_LSWAVE_EVENT_WINDOW_ACTIVE      = 0x66,
+	vgt_perf_VGT_SPI_LSWAVE_IS_EVENT                 = 0x67,
+	vgt_perf_VGT_SPI_LSWAVE_SEND                     = 0x68,
+	vgt_perf_VGT_SPI_HSVERT_VALID                    = 0x69,
+	vgt_perf_VGT_SPI_HSVERT_EOV                      = 0x6a,
+	vgt_perf_VGT_SPI_HSVERT_STALLED                  = 0x6b,
+	vgt_perf_VGT_SPI_HSVERT_STARVED_BUSY             = 0x6c,
+	vgt_perf_VGT_SPI_HSVERT_STARVED_IDLE             = 0x6d,
+	vgt_perf_VGT_SPI_HSVERT_STATIC                   = 0x6e,
+	vgt_perf_VGT_SPI_HSWAVE_EVENT_WINDOW_ACTIVE      = 0x6f,
+	vgt_perf_VGT_SPI_HSWAVE_IS_EVENT                 = 0x70,
+	vgt_perf_VGT_SPI_HSWAVE_SEND                     = 0x71,
+	vgt_perf_ds_prims                                = 0x72,
+	vgt_perf_ls_thread_groups                        = 0x73,
+	vgt_perf_hs_thread_groups                        = 0x74,
+	vgt_perf_es_thread_groups                        = 0x75,
+	vgt_perf_vs_thread_groups                        = 0x76,
+	vgt_perf_ls_done_latency                         = 0x77,
+	vgt_perf_hs_done_latency                         = 0x78,
+	vgt_perf_es_done_latency                         = 0x79,
+	vgt_perf_gs_done_latency                         = 0x7a,
+	vgt_perf_vgt_hs_busy                             = 0x7b,
+	vgt_perf_vgt_te11_busy                           = 0x7c,
+	vgt_perf_ls_flush                                = 0x7d,
+	vgt_perf_hs_flush                                = 0x7e,
+	vgt_perf_es_flush                                = 0x7f,
+	vgt_perf_vgt_pa_clipp_eopg                       = 0x80,
+	vgt_perf_ls_done                                 = 0x81,
+	vgt_perf_hs_done                                 = 0x82,
+	vgt_perf_es_done                                 = 0x83,
+	vgt_perf_gs_done                                 = 0x84,
+	vgt_perf_vsfetch_done                            = 0x85,
+	vgt_perf_gs_done_received                        = 0x86,
+	vgt_perf_es_ring_high_water_mark                 = 0x87,
+	vgt_perf_gs_ring_high_water_mark                 = 0x88,
+	vgt_perf_vs_table_high_water_mark                = 0x89,
+	vgt_perf_hs_tgs_active_high_water_mark           = 0x8a,
+	vgt_perf_pa_clipp_dealloc                        = 0x8b,
+	vgt_perf_cut_mem_flush_stalled                   = 0x8c,
+	vgt_perf_vsvert_work_received                    = 0x8d,
+	vgt_perf_vgt_pa_clipp_starved_after_work         = 0x8e,
+	vgt_perf_te11_con_starved_after_work             = 0x8f,
+	vgt_perf_hs_waiting_on_ls_done_stall             = 0x90,
+	vgt_spi_vsvert_valid                             = 0x91,
+} VGT_PERFCOUNT_SELECT;
+typedef enum IA_PERFCOUNT_SELECT {
+	ia_perf_GRP_INPUT_EVENT_WINDOW_ACTIVE            = 0x0,
+	ia_perf_dma_data_fifo_full                       = 0x1,
+	ia_perf_RESERVED1                                = 0x2,
+	ia_perf_RESERVED2                                = 0x3,
+	ia_perf_RESERVED3                                = 0x4,
+	ia_perf_RESERVED4                                = 0x5,
+	ia_perf_RESERVED5                                = 0x6,
+	ia_perf_MC_LAT_BIN_0                             = 0x7,
+	ia_perf_MC_LAT_BIN_1                             = 0x8,
+	ia_perf_MC_LAT_BIN_2                             = 0x9,
+	ia_perf_MC_LAT_BIN_3                             = 0xa,
+	ia_perf_MC_LAT_BIN_4                             = 0xb,
+	ia_perf_MC_LAT_BIN_5                             = 0xc,
+	ia_perf_MC_LAT_BIN_6                             = 0xd,
+	ia_perf_MC_LAT_BIN_7                             = 0xe,
+	ia_perf_ia_busy                                  = 0xf,
+	ia_perf_ia_sclk_reg_vld_event                    = 0x10,
+	ia_perf_RESERVED6                                = 0x11,
+	ia_perf_ia_sclk_core_vld_event                   = 0x12,
+	ia_perf_RESERVED7                                = 0x13,
+	ia_perf_ia_dma_return                            = 0x14,
+	ia_perf_ia_stalled                               = 0x15,
+	ia_perf_shift_starved_pipe0_event                = 0x16,
+	ia_perf_shift_starved_pipe1_event                = 0x17,
+} IA_PERFCOUNT_SELECT;
+typedef enum WD_PERFCOUNT_SELECT {
+	wd_perf_RBIU_FIFOS_EVENT_WINDOW_ACTIVE           = 0x0,
+	wd_perf_RBIU_DR_FIFO_STARVED                     = 0x1,
+	wd_perf_RBIU_DR_FIFO_STALLED                     = 0x2,
+	wd_perf_RBIU_DI_FIFO_STARVED                     = 0x3,
+	wd_perf_RBIU_DI_FIFO_STALLED                     = 0x4,
+	wd_perf_wd_busy                                  = 0x5,
+	wd_perf_wd_sclk_reg_vld_event                    = 0x6,
+	wd_perf_wd_sclk_input_vld_event                  = 0x7,
+	wd_perf_wd_sclk_core_vld_event                   = 0x8,
+	wd_perf_wd_stalled                               = 0x9,
+	wd_perf_inside_tf_bin_0                          = 0xa,
+	wd_perf_inside_tf_bin_1                          = 0xb,
+	wd_perf_inside_tf_bin_2                          = 0xc,
+	wd_perf_inside_tf_bin_3                          = 0xd,
+	wd_perf_inside_tf_bin_4                          = 0xe,
+	wd_perf_inside_tf_bin_5                          = 0xf,
+	wd_perf_inside_tf_bin_6                          = 0x10,
+	wd_perf_inside_tf_bin_7                          = 0x11,
+	wd_perf_inside_tf_bin_8                          = 0x12,
+	wd_perf_tfreq_lat_bin_0                          = 0x13,
+	wd_perf_tfreq_lat_bin_1                          = 0x14,
+	wd_perf_tfreq_lat_bin_2                          = 0x15,
+	wd_perf_tfreq_lat_bin_3                          = 0x16,
+	wd_perf_tfreq_lat_bin_4                          = 0x17,
+	wd_perf_tfreq_lat_bin_5                          = 0x18,
+	wd_perf_tfreq_lat_bin_6                          = 0x19,
+	wd_perf_tfreq_lat_bin_7                          = 0x1a,
+	wd_starved_on_hs_done                            = 0x1b,
+	wd_perf_se0_hs_done_latency                      = 0x1c,
+	wd_perf_se1_hs_done_latency                      = 0x1d,
+	wd_perf_se2_hs_done_latency                      = 0x1e,
+	wd_perf_se3_hs_done_latency                      = 0x1f,
+	wd_perf_hs_done_se0                              = 0x20,
+	wd_perf_hs_done_se1                              = 0x21,
+	wd_perf_hs_done_se2                              = 0x22,
+	wd_perf_hs_done_se3                              = 0x23,
+	wd_perf_null_patches                             = 0x24,
+} WD_PERFCOUNT_SELECT;
+typedef enum WD_IA_DRAW_TYPE {
+	WD_IA_DRAW_TYPE_DI_MM0                           = 0x0,
+	WD_IA_DRAW_TYPE_DI_MM1                           = 0x1,
+	WD_IA_DRAW_TYPE_EVENT_INIT                       = 0x2,
+	WD_IA_DRAW_TYPE_EVENT_ADDR                       = 0x3,
+	WD_IA_DRAW_TYPE_MIN_INDX                         = 0x4,
+	WD_IA_DRAW_TYPE_MAX_INDX                         = 0x5,
+	WD_IA_DRAW_TYPE_INDX_OFF                         = 0x6,
+	WD_IA_DRAW_TYPE_IMM_DATA                         = 0x7,
+} WD_IA_DRAW_TYPE;
+typedef enum WD_IA_DRAW_SOURCE {
+	WD_IA_DRAW_SOURCE_DMA                            = 0x0,
+	WD_IA_DRAW_SOURCE_IMMD                           = 0x1,
+	WD_IA_DRAW_SOURCE_AUTO                           = 0x2,
+	WD_IA_DRAW_SOURCE_OPAQ                           = 0x3,
+} WD_IA_DRAW_SOURCE;
+#define GSTHREADID_SIZE                           0x2
+typedef enum DebugBlockId {
+	DBG_BLOCK_ID_RESERVED                            = 0x0,
+	DBG_BLOCK_ID_DBG                                 = 0x1,
+	DBG_BLOCK_ID_VMC                                 = 0x2,
+	DBG_BLOCK_ID_PDMA                                = 0x3,
+	DBG_BLOCK_ID_CG                                  = 0x4,
+	DBG_BLOCK_ID_SRBM                                = 0x5,
+	DBG_BLOCK_ID_GRBM                                = 0x6,
+	DBG_BLOCK_ID_RLC                                 = 0x7,
+	DBG_BLOCK_ID_CSC                                 = 0x8,
+	DBG_BLOCK_ID_SEM                                 = 0x9,
+	DBG_BLOCK_ID_IH                                  = 0xa,
+	DBG_BLOCK_ID_SC                                  = 0xb,
+	DBG_BLOCK_ID_SQ                                  = 0xc,
+	DBG_BLOCK_ID_UVDU                                = 0xd,
+	DBG_BLOCK_ID_SQA                                 = 0xe,
+	DBG_BLOCK_ID_SDMA0                               = 0xf,
+	DBG_BLOCK_ID_SDMA1                               = 0x10,
+	DBG_BLOCK_ID_SPIM                                = 0x11,
+	DBG_BLOCK_ID_GDS                                 = 0x12,
+	DBG_BLOCK_ID_VC0                                 = 0x13,
+	DBG_BLOCK_ID_VC1                                 = 0x14,
+	DBG_BLOCK_ID_PA0                                 = 0x15,
+	DBG_BLOCK_ID_PA1                                 = 0x16,
+	DBG_BLOCK_ID_CP0                                 = 0x17,
+	DBG_BLOCK_ID_CP1                                 = 0x18,
+	DBG_BLOCK_ID_CP2                                 = 0x19,
+	DBG_BLOCK_ID_XBR                                 = 0x1a,
+	DBG_BLOCK_ID_UVDM                                = 0x1b,
+	DBG_BLOCK_ID_VGT0                                = 0x1c,
+	DBG_BLOCK_ID_VGT1                                = 0x1d,
+	DBG_BLOCK_ID_IA                                  = 0x1e,
+	DBG_BLOCK_ID_SXM0                                = 0x1f,
+	DBG_BLOCK_ID_SXM1                                = 0x20,
+	DBG_BLOCK_ID_SCT0                                = 0x21,
+	DBG_BLOCK_ID_SCT1                                = 0x22,
+	DBG_BLOCK_ID_SPM0                                = 0x23,
+	DBG_BLOCK_ID_SPM1                                = 0x24,
+	DBG_BLOCK_ID_UNUSED0                             = 0x25,
+	DBG_BLOCK_ID_UNUSED1                             = 0x26,
+	DBG_BLOCK_ID_TCAA                                = 0x27,
+	DBG_BLOCK_ID_TCAB                                = 0x28,
+	DBG_BLOCK_ID_TCCA                                = 0x29,
+	DBG_BLOCK_ID_TCCB                                = 0x2a,
+	DBG_BLOCK_ID_MCC0                                = 0x2b,
+	DBG_BLOCK_ID_MCC1                                = 0x2c,
+	DBG_BLOCK_ID_MCC2                                = 0x2d,
+	DBG_BLOCK_ID_MCC3                                = 0x2e,
+	DBG_BLOCK_ID_SXS0                                = 0x2f,
+	DBG_BLOCK_ID_SXS1                                = 0x30,
+	DBG_BLOCK_ID_SXS2                                = 0x31,
+	DBG_BLOCK_ID_SXS3                                = 0x32,
+	DBG_BLOCK_ID_SXS4                                = 0x33,
+	DBG_BLOCK_ID_SXS5                                = 0x34,
+	DBG_BLOCK_ID_SXS6                                = 0x35,
+	DBG_BLOCK_ID_SXS7                                = 0x36,
+	DBG_BLOCK_ID_SXS8                                = 0x37,
+	DBG_BLOCK_ID_SXS9                                = 0x38,
+	DBG_BLOCK_ID_BCI0                                = 0x39,
+	DBG_BLOCK_ID_BCI1                                = 0x3a,
+	DBG_BLOCK_ID_BCI2                                = 0x3b,
+	DBG_BLOCK_ID_BCI3                                = 0x3c,
+	DBG_BLOCK_ID_MCB                                 = 0x3d,
+	DBG_BLOCK_ID_UNUSED6                             = 0x3e,
+	DBG_BLOCK_ID_SQA00                               = 0x3f,
+	DBG_BLOCK_ID_SQA01                               = 0x40,
+	DBG_BLOCK_ID_SQA02                               = 0x41,
+	DBG_BLOCK_ID_SQA10                               = 0x42,
+	DBG_BLOCK_ID_SQA11                               = 0x43,
+	DBG_BLOCK_ID_SQA12                               = 0x44,
+	DBG_BLOCK_ID_UNUSED7                             = 0x45,
+	DBG_BLOCK_ID_UNUSED8                             = 0x46,
+	DBG_BLOCK_ID_SQB00                               = 0x47,
+	DBG_BLOCK_ID_SQB01                               = 0x48,
+	DBG_BLOCK_ID_SQB10                               = 0x49,
+	DBG_BLOCK_ID_SQB11                               = 0x4a,
+	DBG_BLOCK_ID_SQ00                                = 0x4b,
+	DBG_BLOCK_ID_SQ01                                = 0x4c,
+	DBG_BLOCK_ID_SQ10                                = 0x4d,
+	DBG_BLOCK_ID_SQ11                                = 0x4e,
+	DBG_BLOCK_ID_CB00                                = 0x4f,
+	DBG_BLOCK_ID_CB01                                = 0x50,
+	DBG_BLOCK_ID_CB02                                = 0x51,
+	DBG_BLOCK_ID_CB03                                = 0x52,
+	DBG_BLOCK_ID_CB04                                = 0x53,
+	DBG_BLOCK_ID_UNUSED9                             = 0x54,
+	DBG_BLOCK_ID_UNUSED10                            = 0x55,
+	DBG_BLOCK_ID_UNUSED11                            = 0x56,
+	DBG_BLOCK_ID_CB10                                = 0x57,
+	DBG_BLOCK_ID_CB11                                = 0x58,
+	DBG_BLOCK_ID_CB12                                = 0x59,
+	DBG_BLOCK_ID_CB13                                = 0x5a,
+	DBG_BLOCK_ID_CB14                                = 0x5b,
+	DBG_BLOCK_ID_UNUSED12                            = 0x5c,
+	DBG_BLOCK_ID_UNUSED13                            = 0x5d,
+	DBG_BLOCK_ID_UNUSED14                            = 0x5e,
+	DBG_BLOCK_ID_TCP0                                = 0x5f,
+	DBG_BLOCK_ID_TCP1                                = 0x60,
+	DBG_BLOCK_ID_TCP2                                = 0x61,
+	DBG_BLOCK_ID_TCP3                                = 0x62,
+	DBG_BLOCK_ID_TCP4                                = 0x63,
+	DBG_BLOCK_ID_TCP5                                = 0x64,
+	DBG_BLOCK_ID_TCP6                                = 0x65,
+	DBG_BLOCK_ID_TCP7                                = 0x66,
+	DBG_BLOCK_ID_TCP8                                = 0x67,
+	DBG_BLOCK_ID_TCP9                                = 0x68,
+	DBG_BLOCK_ID_TCP10                               = 0x69,
+	DBG_BLOCK_ID_TCP11                               = 0x6a,
+	DBG_BLOCK_ID_TCP12                               = 0x6b,
+	DBG_BLOCK_ID_TCP13                               = 0x6c,
+	DBG_BLOCK_ID_TCP14                               = 0x6d,
+	DBG_BLOCK_ID_TCP15                               = 0x6e,
+	DBG_BLOCK_ID_TCP16                               = 0x6f,
+	DBG_BLOCK_ID_TCP17                               = 0x70,
+	DBG_BLOCK_ID_TCP18                               = 0x71,
+	DBG_BLOCK_ID_TCP19                               = 0x72,
+	DBG_BLOCK_ID_TCP20                               = 0x73,
+	DBG_BLOCK_ID_TCP21                               = 0x74,
+	DBG_BLOCK_ID_TCP22                               = 0x75,
+	DBG_BLOCK_ID_TCP23                               = 0x76,
+	DBG_BLOCK_ID_TCP_RESERVED0                       = 0x77,
+	DBG_BLOCK_ID_TCP_RESERVED1                       = 0x78,
+	DBG_BLOCK_ID_TCP_RESERVED2                       = 0x79,
+	DBG_BLOCK_ID_TCP_RESERVED3                       = 0x7a,
+	DBG_BLOCK_ID_TCP_RESERVED4                       = 0x7b,
+	DBG_BLOCK_ID_TCP_RESERVED5                       = 0x7c,
+	DBG_BLOCK_ID_TCP_RESERVED6                       = 0x7d,
+	DBG_BLOCK_ID_TCP_RESERVED7                       = 0x7e,
+	DBG_BLOCK_ID_DB00                                = 0x7f,
+	DBG_BLOCK_ID_DB01                                = 0x80,
+	DBG_BLOCK_ID_DB02                                = 0x81,
+	DBG_BLOCK_ID_DB03                                = 0x82,
+	DBG_BLOCK_ID_DB04                                = 0x83,
+	DBG_BLOCK_ID_UNUSED15                            = 0x84,
+	DBG_BLOCK_ID_UNUSED16                            = 0x85,
+	DBG_BLOCK_ID_UNUSED17                            = 0x86,
+	DBG_BLOCK_ID_DB10                                = 0x87,
+	DBG_BLOCK_ID_DB11                                = 0x88,
+	DBG_BLOCK_ID_DB12                                = 0x89,
+	DBG_BLOCK_ID_DB13                                = 0x8a,
+	DBG_BLOCK_ID_DB14                                = 0x8b,
+	DBG_BLOCK_ID_UNUSED18                            = 0x8c,
+	DBG_BLOCK_ID_UNUSED19                            = 0x8d,
+	DBG_BLOCK_ID_UNUSED20                            = 0x8e,
+	DBG_BLOCK_ID_TCC0                                = 0x8f,
+	DBG_BLOCK_ID_TCC1                                = 0x90,
+	DBG_BLOCK_ID_TCC2                                = 0x91,
+	DBG_BLOCK_ID_TCC3                                = 0x92,
+	DBG_BLOCK_ID_TCC4                                = 0x93,
+	DBG_BLOCK_ID_TCC5                                = 0x94,
+	DBG_BLOCK_ID_TCC6                                = 0x95,
+	DBG_BLOCK_ID_TCC7                                = 0x96,
+	DBG_BLOCK_ID_SPS00                               = 0x97,
+	DBG_BLOCK_ID_SPS01                               = 0x98,
+	DBG_BLOCK_ID_SPS02                               = 0x99,
+	DBG_BLOCK_ID_SPS10                               = 0x9a,
+	DBG_BLOCK_ID_SPS11                               = 0x9b,
+	DBG_BLOCK_ID_SPS12                               = 0x9c,
+	DBG_BLOCK_ID_UNUSED21                            = 0x9d,
+	DBG_BLOCK_ID_UNUSED22                            = 0x9e,
+	DBG_BLOCK_ID_TA00                                = 0x9f,
+	DBG_BLOCK_ID_TA01                                = 0xa0,
+	DBG_BLOCK_ID_TA02                                = 0xa1,
+	DBG_BLOCK_ID_TA03                                = 0xa2,
+	DBG_BLOCK_ID_TA04                                = 0xa3,
+	DBG_BLOCK_ID_TA05                                = 0xa4,
+	DBG_BLOCK_ID_TA06                                = 0xa5,
+	DBG_BLOCK_ID_TA07                                = 0xa6,
+	DBG_BLOCK_ID_TA08                                = 0xa7,
+	DBG_BLOCK_ID_TA09                                = 0xa8,
+	DBG_BLOCK_ID_TA0A                                = 0xa9,
+	DBG_BLOCK_ID_TA0B                                = 0xaa,
+	DBG_BLOCK_ID_UNUSED23                            = 0xab,
+	DBG_BLOCK_ID_UNUSED24                            = 0xac,
+	DBG_BLOCK_ID_UNUSED25                            = 0xad,
+	DBG_BLOCK_ID_UNUSED26                            = 0xae,
+	DBG_BLOCK_ID_TA10                                = 0xaf,
+	DBG_BLOCK_ID_TA11                                = 0xb0,
+	DBG_BLOCK_ID_TA12                                = 0xb1,
+	DBG_BLOCK_ID_TA13                                = 0xb2,
+	DBG_BLOCK_ID_TA14                                = 0xb3,
+	DBG_BLOCK_ID_TA15                                = 0xb4,
+	DBG_BLOCK_ID_TA16                                = 0xb5,
+	DBG_BLOCK_ID_TA17                                = 0xb6,
+	DBG_BLOCK_ID_TA18                                = 0xb7,
+	DBG_BLOCK_ID_TA19                                = 0xb8,
+	DBG_BLOCK_ID_TA1A                                = 0xb9,
+	DBG_BLOCK_ID_TA1B                                = 0xba,
+	DBG_BLOCK_ID_UNUSED27                            = 0xbb,
+	DBG_BLOCK_ID_UNUSED28                            = 0xbc,
+	DBG_BLOCK_ID_UNUSED29                            = 0xbd,
+	DBG_BLOCK_ID_UNUSED30                            = 0xbe,
+	DBG_BLOCK_ID_TD00                                = 0xbf,
+	DBG_BLOCK_ID_TD01                                = 0xc0,
+	DBG_BLOCK_ID_TD02                                = 0xc1,
+	DBG_BLOCK_ID_TD03                                = 0xc2,
+	DBG_BLOCK_ID_TD04                                = 0xc3,
+	DBG_BLOCK_ID_TD05                                = 0xc4,
+	DBG_BLOCK_ID_TD06                                = 0xc5,
+	DBG_BLOCK_ID_TD07                                = 0xc6,
+	DBG_BLOCK_ID_TD08                                = 0xc7,
+	DBG_BLOCK_ID_TD09                                = 0xc8,
+	DBG_BLOCK_ID_TD0A                                = 0xc9,
+	DBG_BLOCK_ID_TD0B                                = 0xca,
+	DBG_BLOCK_ID_UNUSED31                            = 0xcb,
+	DBG_BLOCK_ID_UNUSED32                            = 0xcc,
+	DBG_BLOCK_ID_UNUSED33                            = 0xcd,
+	DBG_BLOCK_ID_UNUSED34                            = 0xce,
+	DBG_BLOCK_ID_TD10                                = 0xcf,
+	DBG_BLOCK_ID_TD11                                = 0xd0,
+	DBG_BLOCK_ID_TD12                                = 0xd1,
+	DBG_BLOCK_ID_TD13                                = 0xd2,
+	DBG_BLOCK_ID_TD14                                = 0xd3,
+	DBG_BLOCK_ID_TD15                                = 0xd4,
+	DBG_BLOCK_ID_TD16                                = 0xd5,
+	DBG_BLOCK_ID_TD17                                = 0xd6,
+	DBG_BLOCK_ID_TD18                                = 0xd7,
+	DBG_BLOCK_ID_TD19                                = 0xd8,
+	DBG_BLOCK_ID_TD1A                                = 0xd9,
+	DBG_BLOCK_ID_TD1B                                = 0xda,
+	DBG_BLOCK_ID_UNUSED35                            = 0xdb,
+	DBG_BLOCK_ID_UNUSED36                            = 0xdc,
+	DBG_BLOCK_ID_UNUSED37                            = 0xdd,
+	DBG_BLOCK_ID_UNUSED38                            = 0xde,
+	DBG_BLOCK_ID_LDS00                               = 0xdf,
+	DBG_BLOCK_ID_LDS01                               = 0xe0,
+	DBG_BLOCK_ID_LDS02                               = 0xe1,
+	DBG_BLOCK_ID_LDS03                               = 0xe2,
+	DBG_BLOCK_ID_LDS04                               = 0xe3,
+	DBG_BLOCK_ID_LDS05                               = 0xe4,
+	DBG_BLOCK_ID_LDS06                               = 0xe5,
+	DBG_BLOCK_ID_LDS07                               = 0xe6,
+	DBG_BLOCK_ID_LDS08                               = 0xe7,
+	DBG_BLOCK_ID_LDS09                               = 0xe8,
+	DBG_BLOCK_ID_LDS0A                               = 0xe9,
+	DBG_BLOCK_ID_LDS0B                               = 0xea,
+	DBG_BLOCK_ID_UNUSED39                            = 0xeb,
+	DBG_BLOCK_ID_UNUSED40                            = 0xec,
+	DBG_BLOCK_ID_UNUSED41                            = 0xed,
+	DBG_BLOCK_ID_UNUSED42                            = 0xee,
+	DBG_BLOCK_ID_LDS10                               = 0xef,
+	DBG_BLOCK_ID_LDS11                               = 0xf0,
+	DBG_BLOCK_ID_LDS12                               = 0xf1,
+	DBG_BLOCK_ID_LDS13                               = 0xf2,
+	DBG_BLOCK_ID_LDS14                               = 0xf3,
+	DBG_BLOCK_ID_LDS15                               = 0xf4,
+	DBG_BLOCK_ID_LDS16                               = 0xf5,
+	DBG_BLOCK_ID_LDS17                               = 0xf6,
+	DBG_BLOCK_ID_LDS18                               = 0xf7,
+	DBG_BLOCK_ID_LDS19                               = 0xf8,
+	DBG_BLOCK_ID_LDS1A                               = 0xf9,
+	DBG_BLOCK_ID_LDS1B                               = 0xfa,
+	DBG_BLOCK_ID_UNUSED43                            = 0xfb,
+	DBG_BLOCK_ID_UNUSED44                            = 0xfc,
+	DBG_BLOCK_ID_UNUSED45                            = 0xfd,
+	DBG_BLOCK_ID_UNUSED46                            = 0xfe,
+} DebugBlockId;
+typedef enum DebugBlockId_BY2 {
+	DBG_BLOCK_ID_RESERVED_BY2                        = 0x0,
+	DBG_BLOCK_ID_VMC_BY2                             = 0x1,
+	DBG_BLOCK_ID_UNUSED0_BY2                         = 0x2,
+	DBG_BLOCK_ID_GRBM_BY2                            = 0x3,
+	DBG_BLOCK_ID_CSC_BY2                             = 0x4,
+	DBG_BLOCK_ID_IH_BY2                              = 0x5,
+	DBG_BLOCK_ID_SQ_BY2                              = 0x6,
+	DBG_BLOCK_ID_UVD_BY2                             = 0x7,
+	DBG_BLOCK_ID_SDMA0_BY2                           = 0x8,
+	DBG_BLOCK_ID_SPIM_BY2                            = 0x9,
+	DBG_BLOCK_ID_VC0_BY2                             = 0xa,
+	DBG_BLOCK_ID_PA_BY2                              = 0xb,
+	DBG_BLOCK_ID_CP0_BY2                             = 0xc,
+	DBG_BLOCK_ID_CP2_BY2                             = 0xd,
+	DBG_BLOCK_ID_PC0_BY2                             = 0xe,
+	DBG_BLOCK_ID_BCI0_BY2                            = 0xf,
+	DBG_BLOCK_ID_SXM0_BY2                            = 0x10,
+	DBG_BLOCK_ID_SCT0_BY2                            = 0x11,
+	DBG_BLOCK_ID_SPM0_BY2                            = 0x12,
+	DBG_BLOCK_ID_BCI2_BY2                            = 0x13,
+	DBG_BLOCK_ID_TCA_BY2                             = 0x14,
+	DBG_BLOCK_ID_TCCA_BY2                            = 0x15,
+	DBG_BLOCK_ID_MCC_BY2                             = 0x16,
+	DBG_BLOCK_ID_MCC2_BY2                            = 0x17,
+	DBG_BLOCK_ID_MCD_BY2                             = 0x18,
+	DBG_BLOCK_ID_MCD2_BY2                            = 0x19,
+	DBG_BLOCK_ID_MCD4_BY2                            = 0x1a,
+	DBG_BLOCK_ID_MCB_BY2                             = 0x1b,
+	DBG_BLOCK_ID_SQA_BY2                             = 0x1c,
+	DBG_BLOCK_ID_SQA02_BY2                           = 0x1d,
+	DBG_BLOCK_ID_SQA11_BY2                           = 0x1e,
+	DBG_BLOCK_ID_UNUSED8_BY2                         = 0x1f,
+	DBG_BLOCK_ID_SQB_BY2                             = 0x20,
+	DBG_BLOCK_ID_SQB10_BY2                           = 0x21,
+	DBG_BLOCK_ID_UNUSED10_BY2                        = 0x22,
+	DBG_BLOCK_ID_UNUSED12_BY2                        = 0x23,
+	DBG_BLOCK_ID_CB_BY2                              = 0x24,
+	DBG_BLOCK_ID_CB02_BY2                            = 0x25,
+	DBG_BLOCK_ID_CB10_BY2                            = 0x26,
+	DBG_BLOCK_ID_CB12_BY2                            = 0x27,
+	DBG_BLOCK_ID_SXS_BY2                             = 0x28,
+	DBG_BLOCK_ID_SXS2_BY2                            = 0x29,
+	DBG_BLOCK_ID_SXS4_BY2                            = 0x2a,
+	DBG_BLOCK_ID_SXS6_BY2                            = 0x2b,
+	DBG_BLOCK_ID_DB_BY2                              = 0x2c,
+	DBG_BLOCK_ID_DB02_BY2                            = 0x2d,
+	DBG_BLOCK_ID_DB10_BY2                            = 0x2e,
+	DBG_BLOCK_ID_DB12_BY2                            = 0x2f,
+	DBG_BLOCK_ID_TCP_BY2                             = 0x30,
+	DBG_BLOCK_ID_TCP2_BY2                            = 0x31,
+	DBG_BLOCK_ID_TCP4_BY2                            = 0x32,
+	DBG_BLOCK_ID_TCP6_BY2                            = 0x33,
+	DBG_BLOCK_ID_TCP8_BY2                            = 0x34,
+	DBG_BLOCK_ID_TCP10_BY2                           = 0x35,
+	DBG_BLOCK_ID_TCP12_BY2                           = 0x36,
+	DBG_BLOCK_ID_TCP14_BY2                           = 0x37,
+	DBG_BLOCK_ID_TCP16_BY2                           = 0x38,
+	DBG_BLOCK_ID_TCP18_BY2                           = 0x39,
+	DBG_BLOCK_ID_TCP20_BY2                           = 0x3a,
+	DBG_BLOCK_ID_TCP22_BY2                           = 0x3b,
+	DBG_BLOCK_ID_TCP_RESERVED0_BY2                   = 0x3c,
+	DBG_BLOCK_ID_TCP_RESERVED2_BY2                   = 0x3d,
+	DBG_BLOCK_ID_TCP_RESERVED4_BY2                   = 0x3e,
+	DBG_BLOCK_ID_TCP_RESERVED6_BY2                   = 0x3f,
+	DBG_BLOCK_ID_TCC_BY2                             = 0x40,
+	DBG_BLOCK_ID_TCC2_BY2                            = 0x41,
+	DBG_BLOCK_ID_TCC4_BY2                            = 0x42,
+	DBG_BLOCK_ID_TCC6_BY2                            = 0x43,
+	DBG_BLOCK_ID_SPS_BY2                             = 0x44,
+	DBG_BLOCK_ID_SPS02_BY2                           = 0x45,
+	DBG_BLOCK_ID_SPS11_BY2                           = 0x46,
+	DBG_BLOCK_ID_UNUSED14_BY2                        = 0x47,
+	DBG_BLOCK_ID_TA_BY2                              = 0x48,
+	DBG_BLOCK_ID_TA02_BY2                            = 0x49,
+	DBG_BLOCK_ID_TA04_BY2                            = 0x4a,
+	DBG_BLOCK_ID_TA06_BY2                            = 0x4b,
+	DBG_BLOCK_ID_TA08_BY2                            = 0x4c,
+	DBG_BLOCK_ID_TA0A_BY2                            = 0x4d,
+	DBG_BLOCK_ID_UNUSED20_BY2                        = 0x4e,
+	DBG_BLOCK_ID_UNUSED22_BY2                        = 0x4f,
+	DBG_BLOCK_ID_TA10_BY2                            = 0x50,
+	DBG_BLOCK_ID_TA12_BY2                            = 0x51,
+	DBG_BLOCK_ID_TA14_BY2                            = 0x52,
+	DBG_BLOCK_ID_TA16_BY2                            = 0x53,
+	DBG_BLOCK_ID_TA18_BY2                            = 0x54,
+	DBG_BLOCK_ID_TA1A_BY2                            = 0x55,
+	DBG_BLOCK_ID_UNUSED24_BY2                        = 0x56,
+	DBG_BLOCK_ID_UNUSED26_BY2                        = 0x57,
+	DBG_BLOCK_ID_TD_BY2                              = 0x58,
+	DBG_BLOCK_ID_TD02_BY2                            = 0x59,
+	DBG_BLOCK_ID_TD04_BY2                            = 0x5a,
+	DBG_BLOCK_ID_TD06_BY2                            = 0x5b,
+	DBG_BLOCK_ID_TD08_BY2                            = 0x5c,
+	DBG_BLOCK_ID_TD0A_BY2                            = 0x5d,
+	DBG_BLOCK_ID_UNUSED28_BY2                        = 0x5e,
+	DBG_BLOCK_ID_UNUSED30_BY2                        = 0x5f,
+	DBG_BLOCK_ID_TD10_BY2                            = 0x60,
+	DBG_BLOCK_ID_TD12_BY2                            = 0x61,
+	DBG_BLOCK_ID_TD14_BY2                            = 0x62,
+	DBG_BLOCK_ID_TD16_BY2                            = 0x63,
+	DBG_BLOCK_ID_TD18_BY2                            = 0x64,
+	DBG_BLOCK_ID_TD1A_BY2                            = 0x65,
+	DBG_BLOCK_ID_UNUSED32_BY2                        = 0x66,
+	DBG_BLOCK_ID_UNUSED34_BY2                        = 0x67,
+	DBG_BLOCK_ID_LDS_BY2                             = 0x68,
+	DBG_BLOCK_ID_LDS02_BY2                           = 0x69,
+	DBG_BLOCK_ID_LDS04_BY2                           = 0x6a,
+	DBG_BLOCK_ID_LDS06_BY2                           = 0x6b,
+	DBG_BLOCK_ID_LDS08_BY2                           = 0x6c,
+	DBG_BLOCK_ID_LDS0A_BY2                           = 0x6d,
+	DBG_BLOCK_ID_UNUSED36_BY2                        = 0x6e,
+	DBG_BLOCK_ID_UNUSED38_BY2                        = 0x6f,
+	DBG_BLOCK_ID_LDS10_BY2                           = 0x70,
+	DBG_BLOCK_ID_LDS12_BY2                           = 0x71,
+	DBG_BLOCK_ID_LDS14_BY2                           = 0x72,
+	DBG_BLOCK_ID_LDS16_BY2                           = 0x73,
+	DBG_BLOCK_ID_LDS18_BY2                           = 0x74,
+	DBG_BLOCK_ID_LDS1A_BY2                           = 0x75,
+	DBG_BLOCK_ID_UNUSED40_BY2                        = 0x76,
+	DBG_BLOCK_ID_UNUSED42_BY2                        = 0x77,
+} DebugBlockId_BY2;
+typedef enum DebugBlockId_BY4 {
+	DBG_BLOCK_ID_RESERVED_BY4                        = 0x0,
+	DBG_BLOCK_ID_UNUSED0_BY4                         = 0x1,
+	DBG_BLOCK_ID_CSC_BY4                             = 0x2,
+	DBG_BLOCK_ID_SQ_BY4                              = 0x3,
+	DBG_BLOCK_ID_SDMA0_BY4                           = 0x4,
+	DBG_BLOCK_ID_VC0_BY4                             = 0x5,
+	DBG_BLOCK_ID_CP0_BY4                             = 0x6,
+	DBG_BLOCK_ID_UNUSED1_BY4                         = 0x7,
+	DBG_BLOCK_ID_SXM0_BY4                            = 0x8,
+	DBG_BLOCK_ID_SPM0_BY4                            = 0x9,
+	DBG_BLOCK_ID_TCAA_BY4                            = 0xa,
+	DBG_BLOCK_ID_MCC_BY4                             = 0xb,
+	DBG_BLOCK_ID_MCD_BY4                             = 0xc,
+	DBG_BLOCK_ID_MCD4_BY4                            = 0xd,
+	DBG_BLOCK_ID_SQA_BY4                             = 0xe,
+	DBG_BLOCK_ID_SQA11_BY4                           = 0xf,
+	DBG_BLOCK_ID_SQB_BY4                             = 0x10,
+	DBG_BLOCK_ID_UNUSED10_BY4                        = 0x11,
+	DBG_BLOCK_ID_CB_BY4                              = 0x12,
+	DBG_BLOCK_ID_CB10_BY4                            = 0x13,
+	DBG_BLOCK_ID_SXS_BY4                             = 0x14,
+	DBG_BLOCK_ID_SXS4_BY4                            = 0x15,
+	DBG_BLOCK_ID_DB_BY4                              = 0x16,
+	DBG_BLOCK_ID_DB10_BY4                            = 0x17,
+	DBG_BLOCK_ID_TCP_BY4                             = 0x18,
+	DBG_BLOCK_ID_TCP4_BY4                            = 0x19,
+	DBG_BLOCK_ID_TCP8_BY4                            = 0x1a,
+	DBG_BLOCK_ID_TCP12_BY4                           = 0x1b,
+	DBG_BLOCK_ID_TCP16_BY4                           = 0x1c,
+	DBG_BLOCK_ID_TCP20_BY4                           = 0x1d,
+	DBG_BLOCK_ID_TCP_RESERVED0_BY4                   = 0x1e,
+	DBG_BLOCK_ID_TCP_RESERVED4_BY4                   = 0x1f,
+	DBG_BLOCK_ID_TCC_BY4                             = 0x20,
+	DBG_BLOCK_ID_TCC4_BY4                            = 0x21,
+	DBG_BLOCK_ID_SPS_BY4                             = 0x22,
+	DBG_BLOCK_ID_SPS11_BY4                           = 0x23,
+	DBG_BLOCK_ID_TA_BY4                              = 0x24,
+	DBG_BLOCK_ID_TA04_BY4                            = 0x25,
+	DBG_BLOCK_ID_TA08_BY4                            = 0x26,
+	DBG_BLOCK_ID_UNUSED20_BY4                        = 0x27,
+	DBG_BLOCK_ID_TA10_BY4                            = 0x28,
+	DBG_BLOCK_ID_TA14_BY4                            = 0x29,
+	DBG_BLOCK_ID_TA18_BY4                            = 0x2a,
+	DBG_BLOCK_ID_UNUSED24_BY4                        = 0x2b,
+	DBG_BLOCK_ID_TD_BY4                              = 0x2c,
+	DBG_BLOCK_ID_TD04_BY4                            = 0x2d,
+	DBG_BLOCK_ID_TD08_BY4                            = 0x2e,
+	DBG_BLOCK_ID_UNUSED28_BY4                        = 0x2f,
+	DBG_BLOCK_ID_TD10_BY4                            = 0x30,
+	DBG_BLOCK_ID_TD14_BY4                            = 0x31,
+	DBG_BLOCK_ID_TD18_BY4                            = 0x32,
+	DBG_BLOCK_ID_UNUSED32_BY4                        = 0x33,
+	DBG_BLOCK_ID_LDS_BY4                             = 0x34,
+	DBG_BLOCK_ID_LDS04_BY4                           = 0x35,
+	DBG_BLOCK_ID_LDS08_BY4                           = 0x36,
+	DBG_BLOCK_ID_UNUSED36_BY4                        = 0x37,
+	DBG_BLOCK_ID_LDS10_BY4                           = 0x38,
+	DBG_BLOCK_ID_LDS14_BY4                           = 0x39,
+	DBG_BLOCK_ID_LDS18_BY4                           = 0x3a,
+	DBG_BLOCK_ID_UNUSED40_BY4                        = 0x3b,
+} DebugBlockId_BY4;
+typedef enum DebugBlockId_BY8 {
+	DBG_BLOCK_ID_RESERVED_BY8                        = 0x0,
+	DBG_BLOCK_ID_CSC_BY8                             = 0x1,
+	DBG_BLOCK_ID_SDMA0_BY8                           = 0x2,
+	DBG_BLOCK_ID_CP0_BY8                             = 0x3,
+	DBG_BLOCK_ID_SXM0_BY8                            = 0x4,
+	DBG_BLOCK_ID_TCA_BY8                             = 0x5,
+	DBG_BLOCK_ID_MCD_BY8                             = 0x6,
+	DBG_BLOCK_ID_SQA_BY8                             = 0x7,
+	DBG_BLOCK_ID_SQB_BY8                             = 0x8,
+	DBG_BLOCK_ID_CB_BY8                              = 0x9,
+	DBG_BLOCK_ID_SXS_BY8                             = 0xa,
+	DBG_BLOCK_ID_DB_BY8                              = 0xb,
+	DBG_BLOCK_ID_TCP_BY8                             = 0xc,
+	DBG_BLOCK_ID_TCP8_BY8                            = 0xd,
+	DBG_BLOCK_ID_TCP16_BY8                           = 0xe,
+	DBG_BLOCK_ID_TCP_RESERVED0_BY8                   = 0xf,
+	DBG_BLOCK_ID_TCC_BY8                             = 0x10,
+	DBG_BLOCK_ID_SPS_BY8                             = 0x11,
+	DBG_BLOCK_ID_TA_BY8                              = 0x12,
+	DBG_BLOCK_ID_TA08_BY8                            = 0x13,
+	DBG_BLOCK_ID_TA10_BY8                            = 0x14,
+	DBG_BLOCK_ID_TA18_BY8                            = 0x15,
+	DBG_BLOCK_ID_TD_BY8                              = 0x16,
+	DBG_BLOCK_ID_TD08_BY8                            = 0x17,
+	DBG_BLOCK_ID_TD10_BY8                            = 0x18,
+	DBG_BLOCK_ID_TD18_BY8                            = 0x19,
+	DBG_BLOCK_ID_LDS_BY8                             = 0x1a,
+	DBG_BLOCK_ID_LDS08_BY8                           = 0x1b,
+	DBG_BLOCK_ID_LDS10_BY8                           = 0x1c,
+	DBG_BLOCK_ID_LDS18_BY8                           = 0x1d,
+} DebugBlockId_BY8;
+typedef enum DebugBlockId_BY16 {
+	DBG_BLOCK_ID_RESERVED_BY16                       = 0x0,
+	DBG_BLOCK_ID_SDMA0_BY16                          = 0x1,
+	DBG_BLOCK_ID_SXM_BY16                            = 0x2,
+	DBG_BLOCK_ID_MCD_BY16                            = 0x3,
+	DBG_BLOCK_ID_SQB_BY16                            = 0x4,
+	DBG_BLOCK_ID_SXS_BY16                            = 0x5,
+	DBG_BLOCK_ID_TCP_BY16                            = 0x6,
+	DBG_BLOCK_ID_TCP16_BY16                          = 0x7,
+	DBG_BLOCK_ID_TCC_BY16                            = 0x8,
+	DBG_BLOCK_ID_TA_BY16                             = 0x9,
+	DBG_BLOCK_ID_TA10_BY16                           = 0xa,
+	DBG_BLOCK_ID_TD_BY16                             = 0xb,
+	DBG_BLOCK_ID_TD10_BY16                           = 0xc,
+	DBG_BLOCK_ID_LDS_BY16                            = 0xd,
+	DBG_BLOCK_ID_LDS10_BY16                          = 0xe,
+} DebugBlockId_BY16;
+typedef enum SurfaceEndian {
+	ENDIAN_NONE                                      = 0x0,
+	ENDIAN_8IN16                                     = 0x1,
+	ENDIAN_8IN32                                     = 0x2,
+	ENDIAN_8IN64                                     = 0x3,
+} SurfaceEndian;
+typedef enum ArrayMode {
+	ARRAY_LINEAR_GENERAL                             = 0x0,
+	ARRAY_LINEAR_ALIGNED                             = 0x1,
+	ARRAY_1D_TILED_THIN1                             = 0x2,
+	ARRAY_1D_TILED_THICK                             = 0x3,
+	ARRAY_2D_TILED_THIN1                             = 0x4,
+	ARRAY_PRT_TILED_THIN1                            = 0x5,
+	ARRAY_PRT_2D_TILED_THIN1                         = 0x6,
+	ARRAY_2D_TILED_THICK                             = 0x7,
+	ARRAY_2D_TILED_XTHICK                            = 0x8,
+	ARRAY_PRT_TILED_THICK                            = 0x9,
+	ARRAY_PRT_2D_TILED_THICK                         = 0xa,
+	ARRAY_PRT_3D_TILED_THIN1                         = 0xb,
+	ARRAY_3D_TILED_THIN1                             = 0xc,
+	ARRAY_3D_TILED_THICK                             = 0xd,
+	ARRAY_3D_TILED_XTHICK                            = 0xe,
+	ARRAY_PRT_3D_TILED_THICK                         = 0xf,
+} ArrayMode;
+typedef enum PipeTiling {
+	CONFIG_1_PIPE                                    = 0x0,
+	CONFIG_2_PIPE                                    = 0x1,
+	CONFIG_4_PIPE                                    = 0x2,
+	CONFIG_8_PIPE                                    = 0x3,
+} PipeTiling;
+typedef enum BankTiling {
+	CONFIG_4_BANK                                    = 0x0,
+	CONFIG_8_BANK                                    = 0x1,
+} BankTiling;
+typedef enum GroupInterleave {
+	CONFIG_256B_GROUP                                = 0x0,
+	CONFIG_512B_GROUP                                = 0x1,
+} GroupInterleave;
+typedef enum RowTiling {
+	CONFIG_1KB_ROW                                   = 0x0,
+	CONFIG_2KB_ROW                                   = 0x1,
+	CONFIG_4KB_ROW                                   = 0x2,
+	CONFIG_8KB_ROW                                   = 0x3,
+	CONFIG_1KB_ROW_OPT                               = 0x4,
+	CONFIG_2KB_ROW_OPT                               = 0x5,
+	CONFIG_4KB_ROW_OPT                               = 0x6,
+	CONFIG_8KB_ROW_OPT                               = 0x7,
+} RowTiling;
+typedef enum BankSwapBytes {
+	CONFIG_128B_SWAPS                                = 0x0,
+	CONFIG_256B_SWAPS                                = 0x1,
+	CONFIG_512B_SWAPS                                = 0x2,
+	CONFIG_1KB_SWAPS                                 = 0x3,
+} BankSwapBytes;
+typedef enum SampleSplitBytes {
+	CONFIG_1KB_SPLIT                                 = 0x0,
+	CONFIG_2KB_SPLIT                                 = 0x1,
+	CONFIG_4KB_SPLIT                                 = 0x2,
+	CONFIG_8KB_SPLIT                                 = 0x3,
+} SampleSplitBytes;
+typedef enum NumPipes {
+	ADDR_CONFIG_1_PIPE                               = 0x0,
+	ADDR_CONFIG_2_PIPE                               = 0x1,
+	ADDR_CONFIG_4_PIPE                               = 0x2,
+	ADDR_CONFIG_8_PIPE                               = 0x3,
+} NumPipes;
+typedef enum PipeInterleaveSize {
+	ADDR_CONFIG_PIPE_INTERLEAVE_256B                 = 0x0,
+	ADDR_CONFIG_PIPE_INTERLEAVE_512B                 = 0x1,
+} PipeInterleaveSize;
+typedef enum BankInterleaveSize {
+	ADDR_CONFIG_BANK_INTERLEAVE_1                    = 0x0,
+	ADDR_CONFIG_BANK_INTERLEAVE_2                    = 0x1,
+	ADDR_CONFIG_BANK_INTERLEAVE_4                    = 0x2,
+	ADDR_CONFIG_BANK_INTERLEAVE_8                    = 0x3,
+} BankInterleaveSize;
+typedef enum NumShaderEngines {
+	ADDR_CONFIG_1_SHADER_ENGINE                      = 0x0,
+	ADDR_CONFIG_2_SHADER_ENGINE                      = 0x1,
+} NumShaderEngines;
+typedef enum ShaderEngineTileSize {
+	ADDR_CONFIG_SE_TILE_16                           = 0x0,
+	ADDR_CONFIG_SE_TILE_32                           = 0x1,
+} ShaderEngineTileSize;
+typedef enum NumGPUs {
+	ADDR_CONFIG_1_GPU                                = 0x0,
+	ADDR_CONFIG_2_GPU                                = 0x1,
+	ADDR_CONFIG_4_GPU                                = 0x2,
+} NumGPUs;
+typedef enum MultiGPUTileSize {
+	ADDR_CONFIG_GPU_TILE_16                          = 0x0,
+	ADDR_CONFIG_GPU_TILE_32                          = 0x1,
+	ADDR_CONFIG_GPU_TILE_64                          = 0x2,
+	ADDR_CONFIG_GPU_TILE_128                         = 0x3,
+} MultiGPUTileSize;
+typedef enum RowSize {
+	ADDR_CONFIG_1KB_ROW                              = 0x0,
+	ADDR_CONFIG_2KB_ROW                              = 0x1,
+	ADDR_CONFIG_4KB_ROW                              = 0x2,
+} RowSize;
+typedef enum NumLowerPipes {
+	ADDR_CONFIG_1_LOWER_PIPES                        = 0x0,
+	ADDR_CONFIG_2_LOWER_PIPES                        = 0x1,
+} NumLowerPipes;
+typedef enum ColorTransform {
+	DCC_CT_AUTO                                      = 0x0,
+	DCC_CT_NONE                                      = 0x1,
+	ABGR_TO_A_BG_G_RB                                = 0x2,
+	BGRA_TO_BG_G_RB_A                                = 0x3,
+} ColorTransform;
+typedef enum CompareRef {
+	REF_NEVER                                        = 0x0,
+	REF_LESS                                         = 0x1,
+	REF_EQUAL                                        = 0x2,
+	REF_LEQUAL                                       = 0x3,
+	REF_GREATER                                      = 0x4,
+	REF_NOTEQUAL                                     = 0x5,
+	REF_GEQUAL                                       = 0x6,
+	REF_ALWAYS                                       = 0x7,
+} CompareRef;
+typedef enum ReadSize {
+	READ_256_BITS                                    = 0x0,
+	READ_512_BITS                                    = 0x1,
+} ReadSize;
+typedef enum DepthFormat {
+	DEPTH_INVALID                                    = 0x0,
+	DEPTH_16                                         = 0x1,
+	DEPTH_X8_24                                      = 0x2,
+	DEPTH_8_24                                       = 0x3,
+	DEPTH_X8_24_FLOAT                                = 0x4,
+	DEPTH_8_24_FLOAT                                 = 0x5,
+	DEPTH_32_FLOAT                                   = 0x6,
+	DEPTH_X24_8_32_FLOAT                             = 0x7,
+} DepthFormat;
+typedef enum ZFormat {
+	Z_INVALID                                        = 0x0,
+	Z_16                                             = 0x1,
+	Z_24                                             = 0x2,
+	Z_32_FLOAT                                       = 0x3,
+} ZFormat;
+typedef enum StencilFormat {
+	STENCIL_INVALID                                  = 0x0,
+	STENCIL_8                                        = 0x1,
+} StencilFormat;
+typedef enum CmaskMode {
+	CMASK_CLEAR_NONE                                 = 0x0,
+	CMASK_CLEAR_ONE                                  = 0x1,
+	CMASK_CLEAR_ALL                                  = 0x2,
+	CMASK_ANY_EXPANDED                               = 0x3,
+	CMASK_ALPHA0_FRAG1                               = 0x4,
+	CMASK_ALPHA0_FRAG2                               = 0x5,
+	CMASK_ALPHA0_FRAG4                               = 0x6,
+	CMASK_ALPHA0_FRAGS                               = 0x7,
+	CMASK_ALPHA1_FRAG1                               = 0x8,
+	CMASK_ALPHA1_FRAG2                               = 0x9,
+	CMASK_ALPHA1_FRAG4                               = 0xa,
+	CMASK_ALPHA1_FRAGS                               = 0xb,
+	CMASK_ALPHAX_FRAG1                               = 0xc,
+	CMASK_ALPHAX_FRAG2                               = 0xd,
+	CMASK_ALPHAX_FRAG4                               = 0xe,
+	CMASK_ALPHAX_FRAGS                               = 0xf,
+} CmaskMode;
+typedef enum QuadExportFormat {
+	EXPORT_UNUSED                                    = 0x0,
+	EXPORT_32_R                                      = 0x1,
+	EXPORT_32_GR                                     = 0x2,
+	EXPORT_32_AR                                     = 0x3,
+	EXPORT_FP16_ABGR                                 = 0x4,
+	EXPORT_UNSIGNED16_ABGR                           = 0x5,
+	EXPORT_SIGNED16_ABGR                             = 0x6,
+	EXPORT_32_ABGR                                   = 0x7,
+	EXPORT_32BPP_8PIX                                = 0x8,
+	EXPORT_16_16_UNSIGNED_8PIX                       = 0x9,
+	EXPORT_16_16_SIGNED_8PIX                         = 0xa,
+	EXPORT_16_16_FLOAT_8PIX                          = 0xb,
+} QuadExportFormat;
+typedef enum QuadExportFormatOld {
+	EXPORT_4P_32BPC_ABGR                             = 0x0,
+	EXPORT_4P_16BPC_ABGR                             = 0x1,
+	EXPORT_4P_32BPC_GR                               = 0x2,
+	EXPORT_4P_32BPC_AR                               = 0x3,
+	EXPORT_2P_32BPC_ABGR                             = 0x4,
+	EXPORT_8P_32BPC_R                                = 0x5,
+} QuadExportFormatOld;
+typedef enum ColorFormat {
+	COLOR_INVALID                                    = 0x0,
+	COLOR_8                                          = 0x1,
+	COLOR_16                                         = 0x2,
+	COLOR_8_8                                        = 0x3,
+	COLOR_32                                         = 0x4,
+	COLOR_16_16                                      = 0x5,
+	COLOR_10_11_11                                   = 0x6,
+	COLOR_11_11_10                                   = 0x7,
+	COLOR_10_10_10_2                                 = 0x8,
+	COLOR_2_10_10_10                                 = 0x9,
+	COLOR_8_8_8_8                                    = 0xa,
+	COLOR_32_32                                      = 0xb,
+	COLOR_16_16_16_16                                = 0xc,
+	COLOR_RESERVED_13                                = 0xd,
+	COLOR_32_32_32_32                                = 0xe,
+	COLOR_RESERVED_15                                = 0xf,
+	COLOR_5_6_5                                      = 0x10,
+	COLOR_1_5_5_5                                    = 0x11,
+	COLOR_5_5_5_1                                    = 0x12,
+	COLOR_4_4_4_4                                    = 0x13,
+	COLOR_8_24                                       = 0x14,
+	COLOR_24_8                                       = 0x15,
+	COLOR_X24_8_32_FLOAT                             = 0x16,
+	COLOR_RESERVED_23                                = 0x17,
+	COLOR_RESERVED_24                                = 0x18,
+	COLOR_RESERVED_25                                = 0x19,
+	COLOR_RESERVED_26                                = 0x1a,
+	COLOR_RESERVED_27                                = 0x1b,
+	COLOR_RESERVED_28                                = 0x1c,
+	COLOR_RESERVED_29                                = 0x1d,
+	COLOR_RESERVED_30                                = 0x1e,
+} ColorFormat;
+typedef enum SurfaceFormat {
+	FMT_INVALID                                      = 0x0,
+	FMT_8                                            = 0x1,
+	FMT_16                                           = 0x2,
+	FMT_8_8                                          = 0x3,
+	FMT_32                                           = 0x4,
+	FMT_16_16                                        = 0x5,
+	FMT_10_11_11                                     = 0x6,
+	FMT_11_11_10                                     = 0x7,
+	FMT_10_10_10_2                                   = 0x8,
+	FMT_2_10_10_10                                   = 0x9,
+	FMT_8_8_8_8                                      = 0xa,
+	FMT_32_32                                        = 0xb,
+	FMT_16_16_16_16                                  = 0xc,
+	FMT_32_32_32                                     = 0xd,
+	FMT_32_32_32_32                                  = 0xe,
+	FMT_RESERVED_4                                   = 0xf,
+	FMT_5_6_5                                        = 0x10,
+	FMT_1_5_5_5                                      = 0x11,
+	FMT_5_5_5_1                                      = 0x12,
+	FMT_4_4_4_4                                      = 0x13,
+	FMT_8_24                                         = 0x14,
+	FMT_24_8                                         = 0x15,
+	FMT_X24_8_32_FLOAT                               = 0x16,
+	FMT_RESERVED_33                                  = 0x17,
+	FMT_11_11_10_FLOAT                               = 0x18,
+	FMT_16_FLOAT                                     = 0x19,
+	FMT_32_FLOAT                                     = 0x1a,
+	FMT_16_16_FLOAT                                  = 0x1b,
+	FMT_8_24_FLOAT                                   = 0x1c,
+	FMT_24_8_FLOAT                                   = 0x1d,
+	FMT_32_32_FLOAT                                  = 0x1e,
+	FMT_10_11_11_FLOAT                               = 0x1f,
+	FMT_16_16_16_16_FLOAT                            = 0x20,
+	FMT_3_3_2                                        = 0x21,
+	FMT_6_5_5                                        = 0x22,
+	FMT_32_32_32_32_FLOAT                            = 0x23,
+	FMT_RESERVED_36                                  = 0x24,
+	FMT_1                                            = 0x25,
+	FMT_1_REVERSED                                   = 0x26,
+	FMT_GB_GR                                        = 0x27,
+	FMT_BG_RG                                        = 0x28,
+	FMT_32_AS_8                                      = 0x29,
+	FMT_32_AS_8_8                                    = 0x2a,
+	FMT_5_9_9_9_SHAREDEXP                            = 0x2b,
+	FMT_8_8_8                                        = 0x2c,
+	FMT_16_16_16                                     = 0x2d,
+	FMT_16_16_16_FLOAT                               = 0x2e,
+	FMT_4_4                                          = 0x2f,
+	FMT_32_32_32_FLOAT                               = 0x30,
+	FMT_BC1                                          = 0x31,
+	FMT_BC2                                          = 0x32,
+	FMT_BC3                                          = 0x33,
+	FMT_BC4                                          = 0x34,
+	FMT_BC5                                          = 0x35,
+	FMT_BC6                                          = 0x36,
+	FMT_BC7                                          = 0x37,
+	FMT_32_AS_32_32_32_32                            = 0x38,
+	FMT_APC3                                         = 0x39,
+	FMT_APC4                                         = 0x3a,
+	FMT_APC5                                         = 0x3b,
+	FMT_APC6                                         = 0x3c,
+	FMT_APC7                                         = 0x3d,
+	FMT_CTX1                                         = 0x3e,
+	FMT_RESERVED_63                                  = 0x3f,
+} SurfaceFormat;
+typedef enum BUF_DATA_FORMAT {
+	BUF_DATA_FORMAT_INVALID                          = 0x0,
+	BUF_DATA_FORMAT_8                                = 0x1,
+	BUF_DATA_FORMAT_16                               = 0x2,
+	BUF_DATA_FORMAT_8_8                              = 0x3,
+	BUF_DATA_FORMAT_32                               = 0x4,
+	BUF_DATA_FORMAT_16_16                            = 0x5,
+	BUF_DATA_FORMAT_10_11_11                         = 0x6,
+	BUF_DATA_FORMAT_11_11_10                         = 0x7,
+	BUF_DATA_FORMAT_10_10_10_2                       = 0x8,
+	BUF_DATA_FORMAT_2_10_10_10                       = 0x9,
+	BUF_DATA_FORMAT_8_8_8_8                          = 0xa,
+	BUF_DATA_FORMAT_32_32                            = 0xb,
+	BUF_DATA_FORMAT_16_16_16_16                      = 0xc,
+	BUF_DATA_FORMAT_32_32_32                         = 0xd,
+	BUF_DATA_FORMAT_32_32_32_32                      = 0xe,
+	BUF_DATA_FORMAT_RESERVED_15                      = 0xf,
+} BUF_DATA_FORMAT;
+typedef enum IMG_DATA_FORMAT {
+	IMG_DATA_FORMAT_INVALID                          = 0x0,
+	IMG_DATA_FORMAT_8                                = 0x1,
+	IMG_DATA_FORMAT_16                               = 0x2,
+	IMG_DATA_FORMAT_8_8                              = 0x3,
+	IMG_DATA_FORMAT_32                               = 0x4,
+	IMG_DATA_FORMAT_16_16                            = 0x5,
+	IMG_DATA_FORMAT_10_11_11                         = 0x6,
+	IMG_DATA_FORMAT_11_11_10                         = 0x7,
+	IMG_DATA_FORMAT_10_10_10_2                       = 0x8,
+	IMG_DATA_FORMAT_2_10_10_10                       = 0x9,
+	IMG_DATA_FORMAT_8_8_8_8                          = 0xa,
+	IMG_DATA_FORMAT_32_32                            = 0xb,
+	IMG_DATA_FORMAT_16_16_16_16                      = 0xc,
+	IMG_DATA_FORMAT_32_32_32                         = 0xd,
+	IMG_DATA_FORMAT_32_32_32_32                      = 0xe,
+	IMG_DATA_FORMAT_16_AS_32_32                      = 0xf,
+	IMG_DATA_FORMAT_5_6_5                            = 0x10,
+	IMG_DATA_FORMAT_1_5_5_5                          = 0x11,
+	IMG_DATA_FORMAT_5_5_5_1                          = 0x12,
+	IMG_DATA_FORMAT_4_4_4_4                          = 0x13,
+	IMG_DATA_FORMAT_8_24                             = 0x14,
+	IMG_DATA_FORMAT_24_8                             = 0x15,
+	IMG_DATA_FORMAT_X24_8_32                         = 0x16,
+	IMG_DATA_FORMAT_8_AS_8_8_8_8                     = 0x17,
+	IMG_DATA_FORMAT_ETC2_RGB                         = 0x18,
+	IMG_DATA_FORMAT_ETC2_RGBA                        = 0x19,
+	IMG_DATA_FORMAT_ETC2_R                           = 0x1a,
+	IMG_DATA_FORMAT_ETC2_RG                          = 0x1b,
+	IMG_DATA_FORMAT_ETC2_RGBA1                       = 0x1c,
+	IMG_DATA_FORMAT_RESERVED_29                      = 0x1d,
+	IMG_DATA_FORMAT_RESERVED_30                      = 0x1e,
+	IMG_DATA_FORMAT_RESERVED_31                      = 0x1f,
+	IMG_DATA_FORMAT_GB_GR                            = 0x20,
+	IMG_DATA_FORMAT_BG_RG                            = 0x21,
+	IMG_DATA_FORMAT_5_9_9_9                          = 0x22,
+	IMG_DATA_FORMAT_BC1                              = 0x23,
+	IMG_DATA_FORMAT_BC2                              = 0x24,
+	IMG_DATA_FORMAT_BC3                              = 0x25,
+	IMG_DATA_FORMAT_BC4                              = 0x26,
+	IMG_DATA_FORMAT_BC5                              = 0x27,
+	IMG_DATA_FORMAT_BC6                              = 0x28,
+	IMG_DATA_FORMAT_BC7                              = 0x29,
+	IMG_DATA_FORMAT_16_AS_16_16_16_16                = 0x2a,
+	IMG_DATA_FORMAT_16_AS_32_32_32_32                = 0x2b,
+	IMG_DATA_FORMAT_FMASK8_S2_F1                     = 0x2c,
+	IMG_DATA_FORMAT_FMASK8_S4_F1                     = 0x2d,
+	IMG_DATA_FORMAT_FMASK8_S8_F1                     = 0x2e,
+	IMG_DATA_FORMAT_FMASK8_S2_F2                     = 0x2f,
+	IMG_DATA_FORMAT_FMASK8_S4_F2                     = 0x30,
+	IMG_DATA_FORMAT_FMASK8_S4_F4                     = 0x31,
+	IMG_DATA_FORMAT_FMASK16_S16_F1                   = 0x32,
+	IMG_DATA_FORMAT_FMASK16_S8_F2                    = 0x33,
+	IMG_DATA_FORMAT_FMASK32_S16_F2                   = 0x34,
+	IMG_DATA_FORMAT_FMASK32_S8_F4                    = 0x35,
+	IMG_DATA_FORMAT_FMASK32_S8_F8                    = 0x36,
+	IMG_DATA_FORMAT_FMASK64_S16_F4                   = 0x37,
+	IMG_DATA_FORMAT_FMASK64_S16_F8                   = 0x38,
+	IMG_DATA_FORMAT_4_4                              = 0x39,
+	IMG_DATA_FORMAT_6_5_5                            = 0x3a,
+	IMG_DATA_FORMAT_1                                = 0x3b,
+	IMG_DATA_FORMAT_1_REVERSED                       = 0x3c,
+	IMG_DATA_FORMAT_8_AS_32                          = 0x3d,
+	IMG_DATA_FORMAT_8_AS_32_32                       = 0x3e,
+	IMG_DATA_FORMAT_32_AS_32_32_32_32                = 0x3f,
+} IMG_DATA_FORMAT;
+typedef enum BUF_NUM_FORMAT {
+	BUF_NUM_FORMAT_UNORM                             = 0x0,
+	BUF_NUM_FORMAT_SNORM                             = 0x1,
+	BUF_NUM_FORMAT_USCALED                           = 0x2,
+	BUF_NUM_FORMAT_SSCALED                           = 0x3,
+	BUF_NUM_FORMAT_UINT                              = 0x4,
+	BUF_NUM_FORMAT_SINT                              = 0x5,
+	BUF_NUM_FORMAT_RESERVED_6                        = 0x6,
+	BUF_NUM_FORMAT_FLOAT                             = 0x7,
+} BUF_NUM_FORMAT;
+typedef enum IMG_NUM_FORMAT {
+	IMG_NUM_FORMAT_UNORM                             = 0x0,
+	IMG_NUM_FORMAT_SNORM                             = 0x1,
+	IMG_NUM_FORMAT_USCALED                           = 0x2,
+	IMG_NUM_FORMAT_SSCALED                           = 0x3,
+	IMG_NUM_FORMAT_UINT                              = 0x4,
+	IMG_NUM_FORMAT_SINT                              = 0x5,
+	IMG_NUM_FORMAT_RESERVED_6                        = 0x6,
+	IMG_NUM_FORMAT_FLOAT                             = 0x7,
+	IMG_NUM_FORMAT_RESERVED_8                        = 0x8,
+	IMG_NUM_FORMAT_SRGB                              = 0x9,
+	IMG_NUM_FORMAT_RESERVED_10                       = 0xa,
+	IMG_NUM_FORMAT_RESERVED_11                       = 0xb,
+	IMG_NUM_FORMAT_RESERVED_12                       = 0xc,
+	IMG_NUM_FORMAT_RESERVED_13                       = 0xd,
+	IMG_NUM_FORMAT_RESERVED_14                       = 0xe,
+	IMG_NUM_FORMAT_RESERVED_15                       = 0xf,
+} IMG_NUM_FORMAT;
+typedef enum TileType {
+	ARRAY_COLOR_TILE                                 = 0x0,
+	ARRAY_DEPTH_TILE                                 = 0x1,
+} TileType;
+typedef enum NonDispTilingOrder {
+	ADDR_SURF_MICRO_TILING_DISPLAY                   = 0x0,
+	ADDR_SURF_MICRO_TILING_NON_DISPLAY               = 0x1,
+} NonDispTilingOrder;
+typedef enum MicroTileMode {
+	ADDR_SURF_DISPLAY_MICRO_TILING                   = 0x0,
+	ADDR_SURF_THIN_MICRO_TILING                      = 0x1,
+	ADDR_SURF_DEPTH_MICRO_TILING                     = 0x2,
+	ADDR_SURF_ROTATED_MICRO_TILING                   = 0x3,
+	ADDR_SURF_THICK_MICRO_TILING                     = 0x4,
+} MicroTileMode;
+typedef enum TileSplit {
+	ADDR_SURF_TILE_SPLIT_64B                         = 0x0,
+	ADDR_SURF_TILE_SPLIT_128B                        = 0x1,
+	ADDR_SURF_TILE_SPLIT_256B                        = 0x2,
+	ADDR_SURF_TILE_SPLIT_512B                        = 0x3,
+	ADDR_SURF_TILE_SPLIT_1KB                         = 0x4,
+	ADDR_SURF_TILE_SPLIT_2KB                         = 0x5,
+	ADDR_SURF_TILE_SPLIT_4KB                         = 0x6,
+} TileSplit;
+typedef enum SampleSplit {
+	ADDR_SURF_SAMPLE_SPLIT_1                         = 0x0,
+	ADDR_SURF_SAMPLE_SPLIT_2                         = 0x1,
+	ADDR_SURF_SAMPLE_SPLIT_4                         = 0x2,
+	ADDR_SURF_SAMPLE_SPLIT_8                         = 0x3,
+} SampleSplit;
+typedef enum PipeConfig {
+	ADDR_SURF_P2                                     = 0x0,
+	ADDR_SURF_P2_RESERVED0                           = 0x1,
+	ADDR_SURF_P2_RESERVED1                           = 0x2,
+	ADDR_SURF_P2_RESERVED2                           = 0x3,
+	ADDR_SURF_P4_8x16                                = 0x4,
+	ADDR_SURF_P4_16x16                               = 0x5,
+	ADDR_SURF_P4_16x32                               = 0x6,
+	ADDR_SURF_P4_32x32                               = 0x7,
+	ADDR_SURF_P8_16x16_8x16                          = 0x8,
+	ADDR_SURF_P8_16x32_8x16                          = 0x9,
+	ADDR_SURF_P8_32x32_8x16                          = 0xa,
+	ADDR_SURF_P8_16x32_16x16                         = 0xb,
+	ADDR_SURF_P8_32x32_16x16                         = 0xc,
+	ADDR_SURF_P8_32x32_16x32                         = 0xd,
+	ADDR_SURF_P8_32x64_32x32                         = 0xe,
+	ADDR_SURF_P8_RESERVED0                           = 0xf,
+	ADDR_SURF_P16_32x32_8x16                         = 0x10,
+	ADDR_SURF_P16_32x32_16x16                        = 0x11,
+} PipeConfig;
+typedef enum NumBanks {
+	ADDR_SURF_2_BANK                                 = 0x0,
+	ADDR_SURF_4_BANK                                 = 0x1,
+	ADDR_SURF_8_BANK                                 = 0x2,
+	ADDR_SURF_16_BANK                                = 0x3,
+} NumBanks;
+typedef enum BankWidth {
+	ADDR_SURF_BANK_WIDTH_1                           = 0x0,
+	ADDR_SURF_BANK_WIDTH_2                           = 0x1,
+	ADDR_SURF_BANK_WIDTH_4                           = 0x2,
+	ADDR_SURF_BANK_WIDTH_8                           = 0x3,
+} BankWidth;
+typedef enum BankHeight {
+	ADDR_SURF_BANK_HEIGHT_1                          = 0x0,
+	ADDR_SURF_BANK_HEIGHT_2                          = 0x1,
+	ADDR_SURF_BANK_HEIGHT_4                          = 0x2,
+	ADDR_SURF_BANK_HEIGHT_8                          = 0x3,
+} BankHeight;
+typedef enum BankWidthHeight {
+	ADDR_SURF_BANK_WH_1                              = 0x0,
+	ADDR_SURF_BANK_WH_2                              = 0x1,
+	ADDR_SURF_BANK_WH_4                              = 0x2,
+	ADDR_SURF_BANK_WH_8                              = 0x3,
+} BankWidthHeight;
+typedef enum MacroTileAspect {
+	ADDR_SURF_MACRO_ASPECT_1                         = 0x0,
+	ADDR_SURF_MACRO_ASPECT_2                         = 0x1,
+	ADDR_SURF_MACRO_ASPECT_4                         = 0x2,
+	ADDR_SURF_MACRO_ASPECT_8                         = 0x3,
+} MacroTileAspect;
+typedef enum GATCL1RequestType {
+	GATCL1_TYPE_NORMAL                               = 0x0,
+	GATCL1_TYPE_SHOOTDOWN                            = 0x1,
+	GATCL1_TYPE_BYPASS                               = 0x2,
+} GATCL1RequestType;
+typedef enum TCC_CACHE_POLICIES {
+	TCC_CACHE_POLICY_LRU                             = 0x0,
+	TCC_CACHE_POLICY_STREAM                          = 0x1,
+} TCC_CACHE_POLICIES;
+typedef enum MTYPE {
+	MTYPE_NC_NV                                      = 0x0,
+	MTYPE_NC                                         = 0x1,
+	MTYPE_CC                                         = 0x2,
+	MTYPE_UC                                         = 0x3,
+} MTYPE;
+typedef enum PERFMON_COUNTER_MODE {
+	PERFMON_COUNTER_MODE_ACCUM                       = 0x0,
+	PERFMON_COUNTER_MODE_ACTIVE_CYCLES               = 0x1,
+	PERFMON_COUNTER_MODE_MAX                         = 0x2,
+	PERFMON_COUNTER_MODE_DIRTY                       = 0x3,
+	PERFMON_COUNTER_MODE_SAMPLE                      = 0x4,
+	PERFMON_COUNTER_MODE_CYCLES_SINCE_FIRST_EVENT    = 0x5,
+	PERFMON_COUNTER_MODE_CYCLES_SINCE_LAST_EVENT     = 0x6,
+	PERFMON_COUNTER_MODE_CYCLES_GE_HI                = 0x7,
+	PERFMON_COUNTER_MODE_CYCLES_EQ_HI                = 0x8,
+	PERFMON_COUNTER_MODE_INACTIVE_CYCLES             = 0x9,
+	PERFMON_COUNTER_MODE_RESERVED                    = 0xf,
+} PERFMON_COUNTER_MODE;
+typedef enum PERFMON_SPM_MODE {
+	PERFMON_SPM_MODE_OFF                             = 0x0,
+	PERFMON_SPM_MODE_16BIT_CLAMP                     = 0x1,
+	PERFMON_SPM_MODE_16BIT_NO_CLAMP                  = 0x2,
+	PERFMON_SPM_MODE_32BIT_CLAMP                     = 0x3,
+	PERFMON_SPM_MODE_32BIT_NO_CLAMP                  = 0x4,
+	PERFMON_SPM_MODE_RESERVED_5                      = 0x5,
+	PERFMON_SPM_MODE_RESERVED_6                      = 0x6,
+	PERFMON_SPM_MODE_RESERVED_7                      = 0x7,
+	PERFMON_SPM_MODE_TEST_MODE_0                     = 0x8,
+	PERFMON_SPM_MODE_TEST_MODE_1                     = 0x9,
+	PERFMON_SPM_MODE_TEST_MODE_2                     = 0xa,
+} PERFMON_SPM_MODE;
+typedef enum SurfaceTiling {
+	ARRAY_LINEAR                                     = 0x0,
+	ARRAY_TILED                                      = 0x1,
+} SurfaceTiling;
+typedef enum SurfaceArray {
+	ARRAY_1D                                         = 0x0,
+	ARRAY_2D                                         = 0x1,
+	ARRAY_3D                                         = 0x2,
+	ARRAY_3D_SLICE                                   = 0x3,
+} SurfaceArray;
+typedef enum ColorArray {
+	ARRAY_2D_ALT_COLOR                               = 0x0,
+	ARRAY_2D_COLOR                                   = 0x1,
+	ARRAY_3D_SLICE_COLOR                             = 0x3,
+} ColorArray;
+typedef enum DepthArray {
+	ARRAY_2D_ALT_DEPTH                               = 0x0,
+	ARRAY_2D_DEPTH                                   = 0x1,
+} DepthArray;
+typedef enum ENUM_NUM_SIMD_PER_CU {
+	NUM_SIMD_PER_CU                                  = 0x4,
+} ENUM_NUM_SIMD_PER_CU;
+typedef enum MEM_PWR_FORCE_CTRL {
+	NO_FORCE_REQUEST                                 = 0x0,
+	FORCE_LIGHT_SLEEP_REQUEST                        = 0x1,
+	FORCE_DEEP_SLEEP_REQUEST                         = 0x2,
+	FORCE_SHUT_DOWN_REQUEST                          = 0x3,
+} MEM_PWR_FORCE_CTRL;
+typedef enum MEM_PWR_FORCE_CTRL2 {
+	NO_FORCE_REQ                                     = 0x0,
+	FORCE_LIGHT_SLEEP_REQ                            = 0x1,
+} MEM_PWR_FORCE_CTRL2;
+typedef enum MEM_PWR_DIS_CTRL {
+	ENABLE_MEM_PWR_CTRL                              = 0x0,
+	DISABLE_MEM_PWR_CTRL                             = 0x1,
+} MEM_PWR_DIS_CTRL;
+typedef enum MEM_PWR_SEL_CTRL {
+	DYNAMIC_SHUT_DOWN_ENABLE                         = 0x0,
+	DYNAMIC_DEEP_SLEEP_ENABLE                        = 0x1,
+	DYNAMIC_LIGHT_SLEEP_ENABLE                       = 0x2,
+} MEM_PWR_SEL_CTRL;
+typedef enum MEM_PWR_SEL_CTRL2 {
+	DYNAMIC_DEEP_SLEEP_EN                            = 0x0,
+	DYNAMIC_LIGHT_SLEEP_EN                           = 0x1,
+} MEM_PWR_SEL_CTRL2;
+
+#endif /* GFX_8_1_ENUM_H */
diff --git a/drivers/gpu/drm/amd/include/asic_reg/gca/gfx_8_1_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/gca/gfx_8_1_sh_mask.h
new file mode 100644
index 0000000..397705a
--- /dev/null
+++ b/drivers/gpu/drm/amd/include/asic_reg/gca/gfx_8_1_sh_mask.h
@@ -0,0 +1,21368 @@
+/*
+ * GFX_8_1 Register documentation
+ *
+ * Copyright (C) 2014  Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included
+ * in all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
+ * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN
+ * AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+ */
+
+#ifndef GFX_8_1_SH_MASK_H
+#define GFX_8_1_SH_MASK_H
+
+#define CB_BLEND_RED__BLEND_RED_MASK 0xffffffff
+#define CB_BLEND_RED__BLEND_RED__SHIFT 0x0
+#define CB_BLEND_GREEN__BLEND_GREEN_MASK 0xffffffff
+#define CB_BLEND_GREEN__BLEND_GREEN__SHIFT 0x0
+#define CB_BLEND_BLUE__BLEND_BLUE_MASK 0xffffffff
+#define CB_BLEND_BLUE__BLEND_BLUE__SHIFT 0x0
+#define CB_BLEND_ALPHA__BLEND_ALPHA_MASK 0xffffffff
+#define CB_BLEND_ALPHA__BLEND_ALPHA__SHIFT 0x0
+#define CB_DCC_CONTROL__OVERWRITE_COMBINER_DISABLE_MASK 0x1
+#define CB_DCC_CONTROL__OVERWRITE_COMBINER_DISABLE__SHIFT 0x0
+#define CB_DCC_CONTROL__OVERWRITE_COMBINER_MRT_SHARING_DISABLE_MASK 0x2
+#define CB_DCC_CONTROL__OVERWRITE_COMBINER_MRT_SHARING_DISABLE__SHIFT 0x1
+#define CB_DCC_CONTROL__OVERWRITE_COMBINER_WATERMARK_MASK 0x7c
+#define CB_DCC_CONTROL__OVERWRITE_COMBINER_WATERMARK__SHIFT 0x2
+#define CB_COLOR_CONTROL__DISABLE_DUAL_QUAD_MASK 0x1
+#define CB_COLOR_CONTROL__DISABLE_DUAL_QUAD__SHIFT 0x0
+#define CB_COLOR_CONTROL__DEGAMMA_ENABLE_MASK 0x8
+#define CB_COLOR_CONTROL__DEGAMMA_ENABLE__SHIFT 0x3
+#define CB_COLOR_CONTROL__MODE_MASK 0x70
+#define CB_COLOR_CONTROL__MODE__SHIFT 0x4
+#define CB_COLOR_CONTROL__ROP3_MASK 0xff0000
+#define CB_COLOR_CONTROL__ROP3__SHIFT 0x10
+#define CB_BLEND0_CONTROL__COLOR_SRCBLEND_MASK 0x1f
+#define CB_BLEND0_CONTROL__COLOR_SRCBLEND__SHIFT 0x0
+#define CB_BLEND0_CONTROL__COLOR_COMB_FCN_MASK 0xe0
+#define CB_BLEND0_CONTROL__COLOR_COMB_FCN__SHIFT 0x5
+#define CB_BLEND0_CONTROL__COLOR_DESTBLEND_MASK 0x1f00
+#define CB_BLEND0_CONTROL__COLOR_DESTBLEND__SHIFT 0x8
+#define CB_BLEND0_CONTROL__ALPHA_SRCBLEND_MASK 0x1f0000
+#define CB_BLEND0_CONTROL__ALPHA_SRCBLEND__SHIFT 0x10
+#define CB_BLEND0_CONTROL__ALPHA_COMB_FCN_MASK 0xe00000
+#define CB_BLEND0_CONTROL__ALPHA_COMB_FCN__SHIFT 0x15
+#define CB_BLEND0_CONTROL__ALPHA_DESTBLEND_MASK 0x1f000000
+#define CB_BLEND0_CONTROL__ALPHA_DESTBLEND__SHIFT 0x18
+#define CB_BLEND0_CONTROL__SEPARATE_ALPHA_BLEND_MASK 0x20000000
+#define CB_BLEND0_CONTROL__SEPARATE_ALPHA_BLEND__SHIFT 0x1d
+#define CB_BLEND0_CONTROL__ENABLE_MASK 0x40000000
+#define CB_BLEND0_CONTROL__ENABLE__SHIFT 0x1e
+#define CB_BLEND0_CONTROL__DISABLE_ROP3_MASK 0x80000000
+#define CB_BLEND0_CONTROL__DISABLE_ROP3__SHIFT 0x1f
+#define CB_BLEND1_CONTROL__COLOR_SRCBLEND_MASK 0x1f
+#define CB_BLEND1_CONTROL__COLOR_SRCBLEND__SHIFT 0x0
+#define CB_BLEND1_CONTROL__COLOR_COMB_FCN_MASK 0xe0
+#define CB_BLEND1_CONTROL__COLOR_COMB_FCN__SHIFT 0x5
+#define CB_BLEND1_CONTROL__COLOR_DESTBLEND_MASK 0x1f00
+#define CB_BLEND1_CONTROL__COLOR_DESTBLEND__SHIFT 0x8
+#define CB_BLEND1_CONTROL__ALPHA_SRCBLEND_MASK 0x1f0000
+#define CB_BLEND1_CONTROL__ALPHA_SRCBLEND__SHIFT 0x10
+#define CB_BLEND1_CONTROL__ALPHA_COMB_FCN_MASK 0xe00000
+#define CB_BLEND1_CONTROL__ALPHA_COMB_FCN__SHIFT 0x15
+#define CB_BLEND1_CONTROL__ALPHA_DESTBLEND_MASK 0x1f000000
+#define CB_BLEND1_CONTROL__ALPHA_DESTBLEND__SHIFT 0x18
+#define CB_BLEND1_CONTROL__SEPARATE_ALPHA_BLEND_MASK 0x20000000
+#define CB_BLEND1_CONTROL__SEPARATE_ALPHA_BLEND__SHIFT 0x1d
+#define CB_BLEND1_CONTROL__ENABLE_MASK 0x40000000
+#define CB_BLEND1_CONTROL__ENABLE__SHIFT 0x1e
+#define CB_BLEND1_CONTROL__DISABLE_ROP3_MASK 0x80000000
+#define CB_BLEND1_CONTROL__DISABLE_ROP3__SHIFT 0x1f
+#define CB_BLEND2_CONTROL__COLOR_SRCBLEND_MASK 0x1f
+#define CB_BLEND2_CONTROL__COLOR_SRCBLEND__SHIFT 0x0
+#define CB_BLEND2_CONTROL__COLOR_COMB_FCN_MASK 0xe0
+#define CB_BLEND2_CONTROL__COLOR_COMB_FCN__SHIFT 0x5
+#define CB_BLEND2_CONTROL__COLOR_DESTBLEND_MASK 0x1f00
+#define CB_BLEND2_CONTROL__COLOR_DESTBLEND__SHIFT 0x8
+#define CB_BLEND2_CONTROL__ALPHA_SRCBLEND_MASK 0x1f0000
+#define CB_BLEND2_CONTROL__ALPHA_SRCBLEND__SHIFT 0x10
+#define CB_BLEND2_CONTROL__ALPHA_COMB_FCN_MASK 0xe00000
+#define CB_BLEND2_CONTROL__ALPHA_COMB_FCN__SHIFT 0x15
+#define CB_BLEND2_CONTROL__ALPHA_DESTBLEND_MASK 0x1f000000
+#define CB_BLEND2_CONTROL__ALPHA_DESTBLEND__SHIFT 0x18
+#define CB_BLEND2_CONTROL__SEPARATE_ALPHA_BLEND_MASK 0x20000000
+#define CB_BLEND2_CONTROL__SEPARATE_ALPHA_BLEND__SHIFT 0x1d
+#define CB_BLEND2_CONTROL__ENABLE_MASK 0x40000000
+#define CB_BLEND2_CONTROL__ENABLE__SHIFT 0x1e
+#define CB_BLEND2_CONTROL__DISABLE_ROP3_MASK 0x80000000
+#define CB_BLEND2_CONTROL__DISABLE_ROP3__SHIFT 0x1f
+#define CB_BLEND3_CONTROL__COLOR_SRCBLEND_MASK 0x1f
+#define CB_BLEND3_CONTROL__COLOR_SRCBLEND__SHIFT 0x0
+#define CB_BLEND3_CONTROL__COLOR_COMB_FCN_MASK 0xe0
+#define CB_BLEND3_CONTROL__COLOR_COMB_FCN__SHIFT 0x5
+#define CB_BLEND3_CONTROL__COLOR_DESTBLEND_MASK 0x1f00
+#define CB_BLEND3_CONTROL__COLOR_DESTBLEND__SHIFT 0x8
+#define CB_BLEND3_CONTROL__ALPHA_SRCBLEND_MASK 0x1f0000
+#define CB_BLEND3_CONTROL__ALPHA_SRCBLEND__SHIFT 0x10
+#define CB_BLEND3_CONTROL__ALPHA_COMB_FCN_MASK 0xe00000
+#define CB_BLEND3_CONTROL__ALPHA_COMB_FCN__SHIFT 0x15
+#define CB_BLEND3_CONTROL__ALPHA_DESTBLEND_MASK 0x1f000000
+#define CB_BLEND3_CONTROL__ALPHA_DESTBLEND__SHIFT 0x18
+#define CB_BLEND3_CONTROL__SEPARATE_ALPHA_BLEND_MASK 0x20000000
+#define CB_BLEND3_CONTROL__SEPARATE_ALPHA_BLEND__SHIFT 0x1d
+#define CB_BLEND3_CONTROL__ENABLE_MASK 0x40000000
+#define CB_BLEND3_CONTROL__ENABLE__SHIFT 0x1e
+#define CB_BLEND3_CONTROL__DISABLE_ROP3_MASK 0x80000000
+#define CB_BLEND3_CONTROL__DISABLE_ROP3__SHIFT 0x1f
+#define CB_BLEND4_CONTROL__COLOR_SRCBLEND_MASK 0x1f
+#define CB_BLEND4_CONTROL__COLOR_SRCBLEND__SHIFT 0x0
+#define CB_BLEND4_CONTROL__COLOR_COMB_FCN_MASK 0xe0
+#define CB_BLEND4_CONTROL__COLOR_COMB_FCN__SHIFT 0x5
+#define CB_BLEND4_CONTROL__COLOR_DESTBLEND_MASK 0x1f00
+#define CB_BLEND4_CONTROL__COLOR_DESTBLEND__SHIFT 0x8
+#define CB_BLEND4_CONTROL__ALPHA_SRCBLEND_MASK 0x1f0000
+#define CB_BLEND4_CONTROL__ALPHA_SRCBLEND__SHIFT 0x10
+#define CB_BLEND4_CONTROL__ALPHA_COMB_FCN_MASK 0xe00000
+#define CB_BLEND4_CONTROL__ALPHA_COMB_FCN__SHIFT 0x15
+#define CB_BLEND4_CONTROL__ALPHA_DESTBLEND_MASK 0x1f000000
+#define CB_BLEND4_CONTROL__ALPHA_DESTBLEND__SHIFT 0x18
+#define CB_BLEND4_CONTROL__SEPARATE_ALPHA_BLEND_MASK 0x20000000
+#define CB_BLEND4_CONTROL__SEPARATE_ALPHA_BLEND__SHIFT 0x1d
+#define CB_BLEND4_CONTROL__ENABLE_MASK 0x40000000
+#define CB_BLEND4_CONTROL__ENABLE__SHIFT 0x1e
+#define CB_BLEND4_CONTROL__DISABLE_ROP3_MASK 0x80000000
+#define CB_BLEND4_CONTROL__DISABLE_ROP3__SHIFT 0x1f
+#define CB_BLEND5_CONTROL__COLOR_SRCBLEND_MASK 0x1f
+#define CB_BLEND5_CONTROL__COLOR_SRCBLEND__SHIFT 0x0
+#define CB_BLEND5_CONTROL__COLOR_COMB_FCN_MASK 0xe0
+#define CB_BLEND5_CONTROL__COLOR_COMB_FCN__SHIFT 0x5
+#define CB_BLEND5_CONTROL__COLOR_DESTBLEND_MASK 0x1f00
+#define CB_BLEND5_CONTROL__COLOR_DESTBLEND__SHIFT 0x8
+#define CB_BLEND5_CONTROL__ALPHA_SRCBLEND_MASK 0x1f0000
+#define CB_BLEND5_CONTROL__ALPHA_SRCBLEND__SHIFT 0x10
+#define CB_BLEND5_CONTROL__ALPHA_COMB_FCN_MASK 0xe00000
+#define CB_BLEND5_CONTROL__ALPHA_COMB_FCN__SHIFT 0x15
+#define CB_BLEND5_CONTROL__ALPHA_DESTBLEND_MASK 0x1f000000
+#define CB_BLEND5_CONTROL__ALPHA_DESTBLEND__SHIFT 0x18
+#define CB_BLEND5_CONTROL__SEPARATE_ALPHA_BLEND_MASK 0x20000000
+#define CB_BLEND5_CONTROL__SEPARATE_ALPHA_BLEND__SHIFT 0x1d
+#define CB_BLEND5_CONTROL__ENABLE_MASK 0x40000000
+#define CB_BLEND5_CONTROL__ENABLE__SHIFT 0x1e
+#define CB_BLEND5_CONTROL__DISABLE_ROP3_MASK 0x80000000
+#define CB_BLEND5_CONTROL__DISABLE_ROP3__SHIFT 0x1f
+#define CB_BLEND6_CONTROL__COLOR_SRCBLEND_MASK 0x1f
+#define CB_BLEND6_CONTROL__COLOR_SRCBLEND__SHIFT 0x0
+#define CB_BLEND6_CONTROL__COLOR_COMB_FCN_MASK 0xe0
+#define CB_BLEND6_CONTROL__COLOR_COMB_FCN__SHIFT 0x5
+#define CB_BLEND6_CONTROL__COLOR_DESTBLEND_MASK 0x1f00
+#define CB_BLEND6_CONTROL__COLOR_DESTBLEND__SHIFT 0x8
+#define CB_BLEND6_CONTROL__ALPHA_SRCBLEND_MASK 0x1f0000
+#define CB_BLEND6_CONTROL__ALPHA_SRCBLEND__SHIFT 0x10
+#define CB_BLEND6_CONTROL__ALPHA_COMB_FCN_MASK 0xe00000
+#define CB_BLEND6_CONTROL__ALPHA_COMB_FCN__SHIFT 0x15
+#define CB_BLEND6_CONTROL__ALPHA_DESTBLEND_MASK 0x1f000000
+#define CB_BLEND6_CONTROL__ALPHA_DESTBLEND__SHIFT 0x18
+#define CB_BLEND6_CONTROL__SEPARATE_ALPHA_BLEND_MASK 0x20000000
+#define CB_BLEND6_CONTROL__SEPARATE_ALPHA_BLEND__SHIFT 0x1d
+#define CB_BLEND6_CONTROL__ENABLE_MASK 0x40000000
+#define CB_BLEND6_CONTROL__ENABLE__SHIFT 0x1e
+#define CB_BLEND6_CONTROL__DISABLE_ROP3_MASK 0x80000000
+#define CB_BLEND6_CONTROL__DISABLE_ROP3__SHIFT 0x1f
+#define CB_BLEND7_CONTROL__COLOR_SRCBLEND_MASK 0x1f
+#define CB_BLEND7_CONTROL__COLOR_SRCBLEND__SHIFT 0x0
+#define CB_BLEND7_CONTROL__COLOR_COMB_FCN_MASK 0xe0
+#define CB_BLEND7_CONTROL__COLOR_COMB_FCN__SHIFT 0x5
+#define CB_BLEND7_CONTROL__COLOR_DESTBLEND_MASK 0x1f00
+#define CB_BLEND7_CONTROL__COLOR_DESTBLEND__SHIFT 0x8
+#define CB_BLEND7_CONTROL__ALPHA_SRCBLEND_MASK 0x1f0000
+#define CB_BLEND7_CONTROL__ALPHA_SRCBLEND__SHIFT 0x10
+#define CB_BLEND7_CONTROL__ALPHA_COMB_FCN_MASK 0xe00000
+#define CB_BLEND7_CONTROL__ALPHA_COMB_FCN__SHIFT 0x15
+#define CB_BLEND7_CONTROL__ALPHA_DESTBLEND_MASK 0x1f000000
+#define CB_BLEND7_CONTROL__ALPHA_DESTBLEND__SHIFT 0x18
+#define CB_BLEND7_CONTROL__SEPARATE_ALPHA_BLEND_MASK 0x20000000
+#define CB_BLEND7_CONTROL__SEPARATE_ALPHA_BLEND__SHIFT 0x1d
+#define CB_BLEND7_CONTROL__ENABLE_MASK 0x40000000
+#define CB_BLEND7_CONTROL__ENABLE__SHIFT 0x1e
+#define CB_BLEND7_CONTROL__DISABLE_ROP3_MASK 0x80000000
+#define CB_BLEND7_CONTROL__DISABLE_ROP3__SHIFT 0x1f
+#define CB_COLOR0_BASE__BASE_256B_MASK 0xffffffff
+#define CB_COLOR0_BASE__BASE_256B__SHIFT 0x0
+#define CB_COLOR1_BASE__BASE_256B_MASK 0xffffffff
+#define CB_COLOR1_BASE__BASE_256B__SHIFT 0x0
+#define CB_COLOR2_BASE__BASE_256B_MASK 0xffffffff
+#define CB_COLOR2_BASE__BASE_256B__SHIFT 0x0
+#define CB_COLOR3_BASE__BASE_256B_MASK 0xffffffff
+#define CB_COLOR3_BASE__BASE_256B__SHIFT 0x0
+#define CB_COLOR4_BASE__BASE_256B_MASK 0xffffffff
+#define CB_COLOR4_BASE__BASE_256B__SHIFT 0x0
+#define CB_COLOR5_BASE__BASE_256B_MASK 0xffffffff
+#define CB_COLOR5_BASE__BASE_256B__SHIFT 0x0
+#define CB_COLOR6_BASE__BASE_256B_MASK 0xffffffff
+#define CB_COLOR6_BASE__BASE_256B__SHIFT 0x0
+#define CB_COLOR7_BASE__BASE_256B_MASK 0xffffffff
+#define CB_COLOR7_BASE__BASE_256B__SHIFT 0x0
+#define CB_COLOR0_PITCH__TILE_MAX_MASK 0x7ff
+#define CB_COLOR0_PITCH__TILE_MAX__SHIFT 0x0
+#define CB_COLOR0_PITCH__FMASK_TILE_MAX_MASK 0x7ff00000
+#define CB_COLOR0_PITCH__FMASK_TILE_MAX__SHIFT 0x14
+#define CB_COLOR1_PITCH__TILE_MAX_MASK 0x7ff
+#define CB_COLOR1_PITCH__TILE_MAX__SHIFT 0x0
+#define CB_COLOR1_PITCH__FMASK_TILE_MAX_MASK 0x7ff00000
+#define CB_COLOR1_PITCH__FMASK_TILE_MAX__SHIFT 0x14
+#define CB_COLOR2_PITCH__TILE_MAX_MASK 0x7ff
+#define CB_COLOR2_PITCH__TILE_MAX__SHIFT 0x0
+#define CB_COLOR2_PITCH__FMASK_TILE_MAX_MASK 0x7ff00000
+#define CB_COLOR2_PITCH__FMASK_TILE_MAX__SHIFT 0x14
+#define CB_COLOR3_PITCH__TILE_MAX_MASK 0x7ff
+#define CB_COLOR3_PITCH__TILE_MAX__SHIFT 0x0
+#define CB_COLOR3_PITCH__FMASK_TILE_MAX_MASK 0x7ff00000
+#define CB_COLOR3_PITCH__FMASK_TILE_MAX__SHIFT 0x14
+#define CB_COLOR4_PITCH__TILE_MAX_MASK 0x7ff
+#define CB_COLOR4_PITCH__TILE_MAX__SHIFT 0x0
+#define CB_COLOR4_PITCH__FMASK_TILE_MAX_MASK 0x7ff00000
+#define CB_COLOR4_PITCH__FMASK_TILE_MAX__SHIFT 0x14
+#define CB_COLOR5_PITCH__TILE_MAX_MASK 0x7ff
+#define CB_COLOR5_PITCH__TILE_MAX__SHIFT 0x0
+#define CB_COLOR5_PITCH__FMASK_TILE_MAX_MASK 0x7ff00000
+#define CB_COLOR5_PITCH__FMASK_TILE_MAX__SHIFT 0x14
+#define CB_COLOR6_PITCH__TILE_MAX_MASK 0x7ff
+#define CB_COLOR6_PITCH__TILE_MAX__SHIFT 0x0
+#define CB_COLOR6_PITCH__FMASK_TILE_MAX_MASK 0x7ff00000
+#define CB_COLOR6_PITCH__FMASK_TILE_MAX__SHIFT 0x14
+#define CB_COLOR7_PITCH__TILE_MAX_MASK 0x7ff
+#define CB_COLOR7_PITCH__TILE_MAX__SHIFT 0x0
+#define CB_COLOR7_PITCH__FMASK_TILE_MAX_MASK 0x7ff00000
+#define CB_COLOR7_PITCH__FMASK_TILE_MAX__SHIFT 0x14
+#define CB_COLOR0_SLICE__TILE_MAX_MASK 0x3fffff
+#define CB_COLOR0_SLICE__TILE_MAX__SHIFT 0x0
+#define CB_COLOR1_SLICE__TILE_MAX_MASK 0x3fffff
+#define CB_COLOR1_SLICE__TILE_MAX__SHIFT 0x0
+#define CB_COLOR2_SLICE__TILE_MAX_MASK 0x3fffff
+#define CB_COLOR2_SLICE__TILE_MAX__SHIFT 0x0
+#define CB_COLOR3_SLICE__TILE_MAX_MASK 0x3fffff
+#define CB_COLOR3_SLICE__TILE_MAX__SHIFT 0x0
+#define CB_COLOR4_SLICE__TILE_MAX_MASK 0x3fffff
+#define CB_COLOR4_SLICE__TILE_MAX__SHIFT 0x0
+#define CB_COLOR5_SLICE__TILE_MAX_MASK 0x3fffff
+#define CB_COLOR5_SLICE__TILE_MAX__SHIFT 0x0
+#define CB_COLOR6_SLICE__TILE_MAX_MASK 0x3fffff
+#define CB_COLOR6_SLICE__TILE_MAX__SHIFT 0x0
+#define CB_COLOR7_SLICE__TILE_MAX_MASK 0x3fffff
+#define CB_COLOR7_SLICE__TILE_MAX__SHIFT 0x0
+#define CB_COLOR0_VIEW__SLICE_START_MASK 0x7ff
+#define CB_COLOR0_VIEW__SLICE_START__SHIFT 0x0
+#define CB_COLOR0_VIEW__SLICE_MAX_MASK 0xffe000
+#define CB_COLOR0_VIEW__SLICE_MAX__SHIFT 0xd
+#define CB_COLOR1_VIEW__SLICE_START_MASK 0x7ff
+#define CB_COLOR1_VIEW__SLICE_START__SHIFT 0x0
+#define CB_COLOR1_VIEW__SLICE_MAX_MASK 0xffe000
+#define CB_COLOR1_VIEW__SLICE_MAX__SHIFT 0xd
+#define CB_COLOR2_VIEW__SLICE_START_MASK 0x7ff
+#define CB_COLOR2_VIEW__SLICE_START__SHIFT 0x0
+#define CB_COLOR2_VIEW__SLICE_MAX_MASK 0xffe000
+#define CB_COLOR2_VIEW__SLICE_MAX__SHIFT 0xd
+#define CB_COLOR3_VIEW__SLICE_START_MASK 0x7ff
+#define CB_COLOR3_VIEW__SLICE_START__SHIFT 0x0
+#define CB_COLOR3_VIEW__SLICE_MAX_MASK 0xffe000
+#define CB_COLOR3_VIEW__SLICE_MAX__SHIFT 0xd
+#define CB_COLOR4_VIEW__SLICE_START_MASK 0x7ff
+#define CB_COLOR4_VIEW__SLICE_START__SHIFT 0x0
+#define CB_COLOR4_VIEW__SLICE_MAX_MASK 0xffe000
+#define CB_COLOR4_VIEW__SLICE_MAX__SHIFT 0xd
+#define CB_COLOR5_VIEW__SLICE_START_MASK 0x7ff
+#define CB_COLOR5_VIEW__SLICE_START__SHIFT 0x0
+#define CB_COLOR5_VIEW__SLICE_MAX_MASK 0xffe000
+#define CB_COLOR5_VIEW__SLICE_MAX__SHIFT 0xd
+#define CB_COLOR6_VIEW__SLICE_START_MASK 0x7ff
+#define CB_COLOR6_VIEW__SLICE_START__SHIFT 0x0
+#define CB_COLOR6_VIEW__SLICE_MAX_MASK 0xffe000
+#define CB_COLOR6_VIEW__SLICE_MAX__SHIFT 0xd
+#define CB_COLOR7_VIEW__SLICE_START_MASK 0x7ff
+#define CB_COLOR7_VIEW__SLICE_START__SHIFT 0x0
+#define CB_COLOR7_VIEW__SLICE_MAX_MASK 0xffe000
+#define CB_COLOR7_VIEW__SLICE_MAX__SHIFT 0xd
+#define CB_COLOR0_INFO__ENDIAN_MASK 0x3
+#define CB_COLOR0_INFO__ENDIAN__SHIFT 0x0
+#define CB_COLOR0_INFO__FORMAT_MASK 0x7c
+#define CB_COLOR0_INFO__FORMAT__SHIFT 0x2
+#define CB_COLOR0_INFO__LINEAR_GENERAL_MASK 0x80
+#define CB_COLOR0_INFO__LINEAR_GENERAL__SHIFT 0x7
+#define CB_COLOR0_INFO__NUMBER_TYPE_MASK 0x700
+#define CB_COLOR0_INFO__NUMBER_TYPE__SHIFT 0x8
+#define CB_COLOR0_INFO__COMP_SWAP_MASK 0x1800
+#define CB_COLOR0_INFO__COMP_SWAP__SHIFT 0xb
+#define CB_COLOR0_INFO__FAST_CLEAR_MASK 0x2000
+#define CB_COLOR0_INFO__FAST_CLEAR__SHIFT 0xd
+#define CB_COLOR0_INFO__COMPRESSION_MASK 0x4000
+#define CB_COLOR0_INFO__COMPRESSION__SHIFT 0xe
+#define CB_COLOR0_INFO__BLEND_CLAMP_MASK 0x8000
+#define CB_COLOR0_INFO__BLEND_CLAMP__SHIFT 0xf
+#define CB_COLOR0_INFO__BLEND_BYPASS_MASK 0x10000
+#define CB_COLOR0_INFO__BLEND_BYPASS__SHIFT 0x10
+#define CB_COLOR0_INFO__SIMPLE_FLOAT_MASK 0x20000
+#define CB_COLOR0_INFO__SIMPLE_FLOAT__SHIFT 0x11
+#define CB_COLOR0_INFO__ROUND_MODE_MASK 0x40000
+#define CB_COLOR0_INFO__ROUND_MODE__SHIFT 0x12
+#define CB_COLOR0_INFO__CMASK_IS_LINEAR_MASK 0x80000
+#define CB_COLOR0_INFO__CMASK_IS_LINEAR__SHIFT 0x13
+#define CB_COLOR0_INFO__BLEND_OPT_DONT_RD_DST_MASK 0x700000
+#define CB_COLOR0_INFO__BLEND_OPT_DONT_RD_DST__SHIFT 0x14
+#define CB_COLOR0_INFO__BLEND_OPT_DISCARD_PIXEL_MASK 0x3800000
+#define CB_COLOR0_INFO__BLEND_OPT_DISCARD_PIXEL__SHIFT 0x17
+#define CB_COLOR0_INFO__FMASK_COMPRESSION_DISABLE_MASK 0x4000000
+#define CB_COLOR0_INFO__FMASK_COMPRESSION_DISABLE__SHIFT 0x1a
+#define CB_COLOR0_INFO__FMASK_COMPRESS_1FRAG_ONLY_MASK 0x8000000
+#define CB_COLOR0_INFO__FMASK_COMPRESS_1FRAG_ONLY__SHIFT 0x1b
+#define CB_COLOR0_INFO__DCC_ENABLE_MASK 0x10000000
+#define CB_COLOR0_INFO__DCC_ENABLE__SHIFT 0x1c
+#define CB_COLOR0_INFO__CMASK_ADDR_TYPE_MASK 0x60000000
+#define CB_COLOR0_INFO__CMASK_ADDR_TYPE__SHIFT 0x1d
+#define CB_COLOR1_INFO__ENDIAN_MASK 0x3
+#define CB_COLOR1_INFO__ENDIAN__SHIFT 0x0
+#define CB_COLOR1_INFO__FORMAT_MASK 0x7c
+#define CB_COLOR1_INFO__FORMAT__SHIFT 0x2
+#define CB_COLOR1_INFO__LINEAR_GENERAL_MASK 0x80
+#define CB_COLOR1_INFO__LINEAR_GENERAL__SHIFT 0x7
+#define CB_COLOR1_INFO__NUMBER_TYPE_MASK 0x700
+#define CB_COLOR1_INFO__NUMBER_TYPE__SHIFT 0x8
+#define CB_COLOR1_INFO__COMP_SWAP_MASK 0x1800
+#define CB_COLOR1_INFO__COMP_SWAP__SHIFT 0xb
+#define CB_COLOR1_INFO__FAST_CLEAR_MASK 0x2000
+#define CB_COLOR1_INFO__FAST_CLEAR__SHIFT 0xd
+#define CB_COLOR1_INFO__COMPRESSION_MASK 0x4000
+#define CB_COLOR1_INFO__COMPRESSION__SHIFT 0xe
+#define CB_COLOR1_INFO__BLEND_CLAMP_MASK 0x8000
+#define CB_COLOR1_INFO__BLEND_CLAMP__SHIFT 0xf
+#define CB_COLOR1_INFO__BLEND_BYPASS_MASK 0x10000
+#define CB_COLOR1_INFO__BLEND_BYPASS__SHIFT 0x10
+#define CB_COLOR1_INFO__SIMPLE_FLOAT_MASK 0x20000
+#define CB_COLOR1_INFO__SIMPLE_FLOAT__SHIFT 0x11
+#define CB_COLOR1_INFO__ROUND_MODE_MASK 0x40000
+#define CB_COLOR1_INFO__ROUND_MODE__SHIFT 0x12
+#define CB_COLOR1_INFO__CMASK_IS_LINEAR_MASK 0x80000
+#define CB_COLOR1_INFO__CMASK_IS_LINEAR__SHIFT 0x13
+#define CB_COLOR1_INFO__BLEND_OPT_DONT_RD_DST_MASK 0x700000
+#define CB_COLOR1_INFO__BLEND_OPT_DONT_RD_DST__SHIFT 0x14
+#define CB_COLOR1_INFO__BLEND_OPT_DISCARD_PIXEL_MASK 0x3800000
+#define CB_COLOR1_INFO__BLEND_OPT_DISCARD_PIXEL__SHIFT 0x17
+#define CB_COLOR1_INFO__FMASK_COMPRESSION_DISABLE_MASK 0x4000000
+#define CB_COLOR1_INFO__FMASK_COMPRESSION_DISABLE__SHIFT 0x1a
+#define CB_COLOR1_INFO__FMASK_COMPRESS_1FRAG_ONLY_MASK 0x8000000
+#define CB_COLOR1_INFO__FMASK_COMPRESS_1FRAG_ONLY__SHIFT 0x1b
+#define CB_COLOR1_INFO__DCC_ENABLE_MASK 0x10000000
+#define CB_COLOR1_INFO__DCC_ENABLE__SHIFT 0x1c
+#define CB_COLOR1_INFO__CMASK_ADDR_TYPE_MASK 0x60000000
+#define CB_COLOR1_INFO__CMASK_ADDR_TYPE__SHIFT 0x1d
+#define CB_COLOR2_INFO__ENDIAN_MASK 0x3
+#define CB_COLOR2_INFO__ENDIAN__SHIFT 0x0
+#define CB_COLOR2_INFO__FORMAT_MASK 0x7c
+#define CB_COLOR2_INFO__FORMAT__SHIFT 0x2
+#define CB_COLOR2_INFO__LINEAR_GENERAL_MASK 0x80
+#define CB_COLOR2_INFO__LINEAR_GENERAL__SHIFT 0x7
+#define CB_COLOR2_INFO__NUMBER_TYPE_MASK 0x700
+#define CB_COLOR2_INFO__NUMBER_TYPE__SHIFT 0x8
+#define CB_COLOR2_INFO__COMP_SWAP_MASK 0x1800
+#define CB_COLOR2_INFO__COMP_SWAP__SHIFT 0xb
+#define CB_COLOR2_INFO__FAST_CLEAR_MASK 0x2000
+#define CB_COLOR2_INFO__FAST_CLEAR__SHIFT 0xd
+#define CB_COLOR2_INFO__COMPRESSION_MASK 0x4000
+#define CB_COLOR2_INFO__COMPRESSION__SHIFT 0xe
+#define CB_COLOR2_INFO__BLEND_CLAMP_MASK 0x8000
+#define CB_COLOR2_INFO__BLEND_CLAMP__SHIFT 0xf
+#define CB_COLOR2_INFO__BLEND_BYPASS_MASK 0x10000
+#define CB_COLOR2_INFO__BLEND_BYPASS__SHIFT 0x10
+#define CB_COLOR2_INFO__SIMPLE_FLOAT_MASK 0x20000
+#define CB_COLOR2_INFO__SIMPLE_FLOAT__SHIFT 0x11
+#define CB_COLOR2_INFO__ROUND_MODE_MASK 0x40000
+#define CB_COLOR2_INFO__ROUND_MODE__SHIFT 0x12
+#define CB_COLOR2_INFO__CMASK_IS_LINEAR_MASK 0x80000
+#define CB_COLOR2_INFO__CMASK_IS_LINEAR__SHIFT 0x13
+#define CB_COLOR2_INFO__BLEND_OPT_DONT_RD_DST_MASK 0x700000
+#define CB_COLOR2_INFO__BLEND_OPT_DONT_RD_DST__SHIFT 0x14
+#define CB_COLOR2_INFO__BLEND_OPT_DISCARD_PIXEL_MASK 0x3800000
+#define CB_COLOR2_INFO__BLEND_OPT_DISCARD_PIXEL__SHIFT 0x17
+#define CB_COLOR2_INFO__FMASK_COMPRESSION_DISABLE_MASK 0x4000000
+#define CB_COLOR2_INFO__FMASK_COMPRESSION_DISABLE__SHIFT 0x1a
+#define CB_COLOR2_INFO__FMASK_COMPRESS_1FRAG_ONLY_MASK 0x8000000
+#define CB_COLOR2_INFO__FMASK_COMPRESS_1FRAG_ONLY__SHIFT 0x1b
+#define CB_COLOR2_INFO__DCC_ENABLE_MASK 0x10000000
+#define CB_COLOR2_INFO__DCC_ENABLE__SHIFT 0x1c
+#define CB_COLOR2_INFO__CMASK_ADDR_TYPE_MASK 0x60000000
+#define CB_COLOR2_INFO__CMASK_ADDR_TYPE__SHIFT 0x1d
+#define CB_COLOR3_INFO__ENDIAN_MASK 0x3
+#define CB_COLOR3_INFO__ENDIAN__SHIFT 0x0
+#define CB_COLOR3_INFO__FORMAT_MASK 0x7c
+#define CB_COLOR3_INFO__FORMAT__SHIFT 0x2
+#define CB_COLOR3_INFO__LINEAR_GENERAL_MASK 0x80
+#define CB_COLOR3_INFO__LINEAR_GENERAL__SHIFT 0x7
+#define CB_COLOR3_INFO__NUMBER_TYPE_MASK 0x700
+#define CB_COLOR3_INFO__NUMBER_TYPE__SHIFT 0x8
+#define CB_COLOR3_INFO__COMP_SWAP_MASK 0x1800
+#define CB_COLOR3_INFO__COMP_SWAP__SHIFT 0xb
+#define CB_COLOR3_INFO__FAST_CLEAR_MASK 0x2000
+#define CB_COLOR3_INFO__FAST_CLEAR__SHIFT 0xd
+#define CB_COLOR3_INFO__COMPRESSION_MASK 0x4000
+#define CB_COLOR3_INFO__COMPRESSION__SHIFT 0xe
+#define CB_COLOR3_INFO__BLEND_CLAMP_MASK 0x8000
+#define CB_COLOR3_INFO__BLEND_CLAMP__SHIFT 0xf
+#define CB_COLOR3_INFO__BLEND_BYPASS_MASK 0x10000
+#define CB_COLOR3_INFO__BLEND_BYPASS__SHIFT 0x10
+#define CB_COLOR3_INFO__SIMPLE_FLOAT_MASK 0x20000
+#define CB_COLOR3_INFO__SIMPLE_FLOAT__SHIFT 0x11
+#define CB_COLOR3_INFO__ROUND_MODE_MASK 0x40000
+#define CB_COLOR3_INFO__ROUND_MODE__SHIFT 0x12
+#define CB_COLOR3_INFO__CMASK_IS_LINEAR_MASK 0x80000
+#define CB_COLOR3_INFO__CMASK_IS_LINEAR__SHIFT 0x13
+#define CB_COLOR3_INFO__BLEND_OPT_DONT_RD_DST_MASK 0x700000
+#define CB_COLOR3_INFO__BLEND_OPT_DONT_RD_DST__SHIFT 0x14
+#define CB_COLOR3_INFO__BLEND_OPT_DISCARD_PIXEL_MASK 0x3800000
+#define CB_COLOR3_INFO__BLEND_OPT_DISCARD_PIXEL__SHIFT 0x17
+#define CB_COLOR3_INFO__FMASK_COMPRESSION_DISABLE_MASK 0x4000000
+#define CB_COLOR3_INFO__FMASK_COMPRESSION_DISABLE__SHIFT 0x1a
+#define CB_COLOR3_INFO__FMASK_COMPRESS_1FRAG_ONLY_MASK 0x8000000
+#define CB_COLOR3_INFO__FMASK_COMPRESS_1FRAG_ONLY__SHIFT 0x1b
+#define CB_COLOR3_INFO__DCC_ENABLE_MASK 0x10000000
+#define CB_COLOR3_INFO__DCC_ENABLE__SHIFT 0x1c
+#define CB_COLOR3_INFO__CMASK_ADDR_TYPE_MASK 0x60000000
+#define CB_COLOR3_INFO__CMASK_ADDR_TYPE__SHIFT 0x1d
+#define CB_COLOR4_INFO__ENDIAN_MASK 0x3
+#define CB_COLOR4_INFO__ENDIAN__SHIFT 0x0
+#define CB_COLOR4_INFO__FORMAT_MASK 0x7c
+#define CB_COLOR4_INFO__FORMAT__SHIFT 0x2
+#define CB_COLOR4_INFO__LINEAR_GENERAL_MASK 0x80
+#define CB_COLOR4_INFO__LINEAR_GENERAL__SHIFT 0x7
+#define CB_COLOR4_INFO__NUMBER_TYPE_MASK 0x700
+#define CB_COLOR4_INFO__NUMBER_TYPE__SHIFT 0x8
+#define CB_COLOR4_INFO__COMP_SWAP_MASK 0x1800
+#define CB_COLOR4_INFO__COMP_SWAP__SHIFT 0xb
+#define CB_COLOR4_INFO__FAST_CLEAR_MASK 0x2000
+#define CB_COLOR4_INFO__FAST_CLEAR__SHIFT 0xd
+#define CB_COLOR4_INFO__COMPRESSION_MASK 0x4000
+#define CB_COLOR4_INFO__COMPRESSION__SHIFT 0xe
+#define CB_COLOR4_INFO__BLEND_CLAMP_MASK 0x8000
+#define CB_COLOR4_INFO__BLEND_CLAMP__SHIFT 0xf
+#define CB_COLOR4_INFO__BLEND_BYPASS_MASK 0x10000
+#define CB_COLOR4_INFO__BLEND_BYPASS__SHIFT 0x10
+#define CB_COLOR4_INFO__SIMPLE_FLOAT_MASK 0x20000
+#define CB_COLOR4_INFO__SIMPLE_FLOAT__SHIFT 0x11
+#define CB_COLOR4_INFO__ROUND_MODE_MASK 0x40000
+#define CB_COLOR4_INFO__ROUND_MODE__SHIFT 0x12
+#define CB_COLOR4_INFO__CMASK_IS_LINEAR_MASK 0x80000
+#define CB_COLOR4_INFO__CMASK_IS_LINEAR__SHIFT 0x13
+#define CB_COLOR4_INFO__BLEND_OPT_DONT_RD_DST_MASK 0x700000
+#define CB_COLOR4_INFO__BLEND_OPT_DONT_RD_DST__SHIFT 0x14
+#define CB_COLOR4_INFO__BLEND_OPT_DISCARD_PIXEL_MASK 0x3800000
+#define CB_COLOR4_INFO__BLEND_OPT_DISCARD_PIXEL__SHIFT 0x17
+#define CB_COLOR4_INFO__FMASK_COMPRESSION_DISABLE_MASK 0x4000000
+#define CB_COLOR4_INFO__FMASK_COMPRESSION_DISABLE__SHIFT 0x1a
+#define CB_COLOR4_INFO__FMASK_COMPRESS_1FRAG_ONLY_MASK 0x8000000
+#define CB_COLOR4_INFO__FMASK_COMPRESS_1FRAG_ONLY__SHIFT 0x1b
+#define CB_COLOR4_INFO__DCC_ENABLE_MASK 0x10000000
+#define CB_COLOR4_INFO__DCC_ENABLE__SHIFT 0x1c
+#define CB_COLOR4_INFO__CMASK_ADDR_TYPE_MASK 0x60000000
+#define CB_COLOR4_INFO__CMASK_ADDR_TYPE__SHIFT 0x1d
+#define CB_COLOR5_INFO__ENDIAN_MASK 0x3
+#define CB_COLOR5_INFO__ENDIAN__SHIFT 0x0
+#define CB_COLOR5_INFO__FORMAT_MASK 0x7c
+#define CB_COLOR5_INFO__FORMAT__SHIFT 0x2
+#define CB_COLOR5_INFO__LINEAR_GENERAL_MASK 0x80
+#define CB_COLOR5_INFO__LINEAR_GENERAL__SHIFT 0x7
+#define CB_COLOR5_INFO__NUMBER_TYPE_MASK 0x700
+#define CB_COLOR5_INFO__NUMBER_TYPE__SHIFT 0x8
+#define CB_COLOR5_INFO__COMP_SWAP_MASK 0x1800
+#define CB_COLOR5_INFO__COMP_SWAP__SHIFT 0xb
+#define CB_COLOR5_INFO__FAST_CLEAR_MASK 0x2000
+#define CB_COLOR5_INFO__FAST_CLEAR__SHIFT 0xd
+#define CB_COLOR5_INFO__COMPRESSION_MASK 0x4000
+#define CB_COLOR5_INFO__COMPRESSION__SHIFT 0xe
+#define CB_COLOR5_INFO__BLEND_CLAMP_MASK 0x8000
+#define CB_COLOR5_INFO__BLEND_CLAMP__SHIFT 0xf
+#define CB_COLOR5_INFO__BLEND_BYPASS_MASK 0x10000
+#define CB_COLOR5_INFO__BLEND_BYPASS__SHIFT 0x10
+#define CB_COLOR5_INFO__SIMPLE_FLOAT_MASK 0x20000
+#define CB_COLOR5_INFO__SIMPLE_FLOAT__SHIFT 0x11
+#define CB_COLOR5_INFO__ROUND_MODE_MASK 0x40000
+#define CB_COLOR5_INFO__ROUND_MODE__SHIFT 0x12
+#define CB_COLOR5_INFO__CMASK_IS_LINEAR_MASK 0x80000
+#define CB_COLOR5_INFO__CMASK_IS_LINEAR__SHIFT 0x13
+#define CB_COLOR5_INFO__BLEND_OPT_DONT_RD_DST_MASK 0x700000
+#define CB_COLOR5_INFO__BLEND_OPT_DONT_RD_DST__SHIFT 0x14
+#define CB_COLOR5_INFO__BLEND_OPT_DISCARD_PIXEL_MASK 0x3800000
+#define CB_COLOR5_INFO__BLEND_OPT_DISCARD_PIXEL__SHIFT 0x17
+#define CB_COLOR5_INFO__FMASK_COMPRESSION_DISABLE_MASK 0x4000000
+#define CB_COLOR5_INFO__FMASK_COMPRESSION_DISABLE__SHIFT 0x1a
+#define CB_COLOR5_INFO__FMASK_COMPRESS_1FRAG_ONLY_MASK 0x8000000
+#define CB_COLOR5_INFO__FMASK_COMPRESS_1FRAG_ONLY__SHIFT 0x1b
+#define CB_COLOR5_INFO__DCC_ENABLE_MASK 0x10000000
+#define CB_COLOR5_INFO__DCC_ENABLE__SHIFT 0x1c
+#define CB_COLOR5_INFO__CMASK_ADDR_TYPE_MASK 0x60000000
+#define CB_COLOR5_INFO__CMASK_ADDR_TYPE__SHIFT 0x1d
+#define CB_COLOR6_INFO__ENDIAN_MASK 0x3
+#define CB_COLOR6_INFO__ENDIAN__SHIFT 0x0
+#define CB_COLOR6_INFO__FORMAT_MASK 0x7c
+#define CB_COLOR6_INFO__FORMAT__SHIFT 0x2
+#define CB_COLOR6_INFO__LINEAR_GENERAL_MASK 0x80
+#define CB_COLOR6_INFO__LINEAR_GENERAL__SHIFT 0x7
+#define CB_COLOR6_INFO__NUMBER_TYPE_MASK 0x700
+#define CB_COLOR6_INFO__NUMBER_TYPE__SHIFT 0x8
+#define CB_COLOR6_INFO__COMP_SWAP_MASK 0x1800
+#define CB_COLOR6_INFO__COMP_SWAP__SHIFT 0xb
+#define CB_COLOR6_INFO__FAST_CLEAR_MASK 0x2000
+#define CB_COLOR6_INFO__FAST_CLEAR__SHIFT 0xd
+#define CB_COLOR6_INFO__COMPRESSION_MASK 0x4000
+#define CB_COLOR6_INFO__COMPRESSION__SHIFT 0xe
+#define CB_COLOR6_INFO__BLEND_CLAMP_MASK 0x8000
+#define CB_COLOR6_INFO__BLEND_CLAMP__SHIFT 0xf
+#define CB_COLOR6_INFO__BLEND_BYPASS_MASK 0x10000
+#define CB_COLOR6_INFO__BLEND_BYPASS__SHIFT 0x10
+#define CB_COLOR6_INFO__SIMPLE_FLOAT_MASK 0x20000
+#define CB_COLOR6_INFO__SIMPLE_FLOAT__SHIFT 0x11
+#define CB_COLOR6_INFO__ROUND_MODE_MASK 0x40000
+#define CB_COLOR6_INFO__ROUND_MODE__SHIFT 0x12
+#define CB_COLOR6_INFO__CMASK_IS_LINEAR_MASK 0x80000
+#define CB_COLOR6_INFO__CMASK_IS_LINEAR__SHIFT 0x13
+#define CB_COLOR6_INFO__BLEND_OPT_DONT_RD_DST_MASK 0x700000
+#define CB_COLOR6_INFO__BLEND_OPT_DONT_RD_DST__SHIFT 0x14
+#define CB_COLOR6_INFO__BLEND_OPT_DISCARD_PIXEL_MASK 0x3800000
+#define CB_COLOR6_INFO__BLEND_OPT_DISCARD_PIXEL__SHIFT 0x17
+#define CB_COLOR6_INFO__FMASK_COMPRESSION_DISABLE_MASK 0x4000000
+#define CB_COLOR6_INFO__FMASK_COMPRESSION_DISABLE__SHIFT 0x1a
+#define CB_COLOR6_INFO__FMASK_COMPRESS_1FRAG_ONLY_MASK 0x8000000
+#define CB_COLOR6_INFO__FMASK_COMPRESS_1FRAG_ONLY__SHIFT 0x1b
+#define CB_COLOR6_INFO__DCC_ENABLE_MASK 0x10000000
+#define CB_COLOR6_INFO__DCC_ENABLE__SHIFT 0x1c
+#define CB_COLOR6_INFO__CMASK_ADDR_TYPE_MASK 0x60000000
+#define CB_COLOR6_INFO__CMASK_ADDR_TYPE__SHIFT 0x1d
+#define CB_COLOR7_INFO__ENDIAN_MASK 0x3
+#define CB_COLOR7_INFO__ENDIAN__SHIFT 0x0
+#define CB_COLOR7_INFO__FORMAT_MASK 0x7c
+#define CB_COLOR7_INFO__FORMAT__SHIFT 0x2
+#define CB_COLOR7_INFO__LINEAR_GENERAL_MASK 0x80
+#define CB_COLOR7_INFO__LINEAR_GENERAL__SHIFT 0x7
+#define CB_COLOR7_INFO__NUMBER_TYPE_MASK 0x700
+#define CB_COLOR7_INFO__NUMBER_TYPE__SHIFT 0x8
+#define CB_COLOR7_INFO__COMP_SWAP_MASK 0x1800
+#define CB_COLOR7_INFO__COMP_SWAP__SHIFT 0xb
+#define CB_COLOR7_INFO__FAST_CLEAR_MASK 0x2000
+#define CB_COLOR7_INFO__FAST_CLEAR__SHIFT 0xd
+#define CB_COLOR7_INFO__COMPRESSION_MASK 0x4000
+#define CB_COLOR7_INFO__COMPRESSION__SHIFT 0xe
+#define CB_COLOR7_INFO__BLEND_CLAMP_MASK 0x8000
+#define CB_COLOR7_INFO__BLEND_CLAMP__SHIFT 0xf
+#define CB_COLOR7_INFO__BLEND_BYPASS_MASK 0x10000
+#define CB_COLOR7_INFO__BLEND_BYPASS__SHIFT 0x10
+#define CB_COLOR7_INFO__SIMPLE_FLOAT_MASK 0x20000
+#define CB_COLOR7_INFO__SIMPLE_FLOAT__SHIFT 0x11
+#define CB_COLOR7_INFO__ROUND_MODE_MASK 0x40000
+#define CB_COLOR7_INFO__ROUND_MODE__SHIFT 0x12
+#define CB_COLOR7_INFO__CMASK_IS_LINEAR_MASK 0x80000
+#define CB_COLOR7_INFO__CMASK_IS_LINEAR__SHIFT 0x13
+#define CB_COLOR7_INFO__BLEND_OPT_DONT_RD_DST_MASK 0x700000
+#define CB_COLOR7_INFO__BLEND_OPT_DONT_RD_DST__SHIFT 0x14
+#define CB_COLOR7_INFO__BLEND_OPT_DISCARD_PIXEL_MASK 0x3800000
+#define CB_COLOR7_INFO__BLEND_OPT_DISCARD_PIXEL__SHIFT 0x17
+#define CB_COLOR7_INFO__FMASK_COMPRESSION_DISABLE_MASK 0x4000000
+#define CB_COLOR7_INFO__FMASK_COMPRESSION_DISABLE__SHIFT 0x1a
+#define CB_COLOR7_INFO__FMASK_COMPRESS_1FRAG_ONLY_MASK 0x8000000
+#define CB_COLOR7_INFO__FMASK_COMPRESS_1FRAG_ONLY__SHIFT 0x1b
+#define CB_COLOR7_INFO__DCC_ENABLE_MASK 0x10000000
+#define CB_COLOR7_INFO__DCC_ENABLE__SHIFT 0x1c
+#define CB_COLOR7_INFO__CMASK_ADDR_TYPE_MASK 0x60000000
+#define CB_COLOR7_INFO__CMASK_ADDR_TYPE__SHIFT 0x1d
+#define CB_COLOR0_ATTRIB__TILE_MODE_INDEX_MASK 0x1f
+#define CB_COLOR0_ATTRIB__TILE_MODE_INDEX__SHIFT 0x0
+#define CB_COLOR0_ATTRIB__FMASK_TILE_MODE_INDEX_MASK 0x3e0
+#define CB_COLOR0_ATTRIB__FMASK_TILE_MODE_INDEX__SHIFT 0x5
+#define CB_COLOR0_ATTRIB__FMASK_BANK_HEIGHT_MASK 0xc00
+#define CB_COLOR0_ATTRIB__FMASK_BANK_HEIGHT__SHIFT 0xa
+#define CB_COLOR0_ATTRIB__NUM_SAMPLES_MASK 0x7000
+#define CB_COLOR0_ATTRIB__NUM_SAMPLES__SHIFT 0xc
+#define CB_COLOR0_ATTRIB__NUM_FRAGMENTS_MASK 0x18000
+#define CB_COLOR0_ATTRIB__NUM_FRAGMENTS__SHIFT 0xf
+#define CB_COLOR0_ATTRIB__FORCE_DST_ALPHA_1_MASK 0x20000
+#define CB_COLOR0_ATTRIB__FORCE_DST_ALPHA_1__SHIFT 0x11
+#define CB_COLOR1_ATTRIB__TILE_MODE_INDEX_MASK 0x1f
+#define CB_COLOR1_ATTRIB__TILE_MODE_INDEX__SHIFT 0x0
+#define CB_COLOR1_ATTRIB__FMASK_TILE_MODE_INDEX_MASK 0x3e0
+#define CB_COLOR1_ATTRIB__FMASK_TILE_MODE_INDEX__SHIFT 0x5
+#define CB_COLOR1_ATTRIB__FMASK_BANK_HEIGHT_MASK 0xc00
+#define CB_COLOR1_ATTRIB__FMASK_BANK_HEIGHT__SHIFT 0xa
+#define CB_COLOR1_ATTRIB__NUM_SAMPLES_MASK 0x7000
+#define CB_COLOR1_ATTRIB__NUM_SAMPLES__SHIFT 0xc
+#define CB_COLOR1_ATTRIB__NUM_FRAGMENTS_MASK 0x18000
+#define CB_COLOR1_ATTRIB__NUM_FRAGMENTS__SHIFT 0xf
+#define CB_COLOR1_ATTRIB__FORCE_DST_ALPHA_1_MASK 0x20000
+#define CB_COLOR1_ATTRIB__FORCE_DST_ALPHA_1__SHIFT 0x11
+#define CB_COLOR2_ATTRIB__TILE_MODE_INDEX_MASK 0x1f
+#define CB_COLOR2_ATTRIB__TILE_MODE_INDEX__SHIFT 0x0
+#define CB_COLOR2_ATTRIB__FMASK_TILE_MODE_INDEX_MASK 0x3e0
+#define CB_COLOR2_ATTRIB__FMASK_TILE_MODE_INDEX__SHIFT 0x5
+#define CB_COLOR2_ATTRIB__FMASK_BANK_HEIGHT_MASK 0xc00
+#define CB_COLOR2_ATTRIB__FMASK_BANK_HEIGHT__SHIFT 0xa
+#define CB_COLOR2_ATTRIB__NUM_SAMPLES_MASK 0x7000
+#define CB_COLOR2_ATTRIB__NUM_SAMPLES__SHIFT 0xc
+#define CB_COLOR2_ATTRIB__NUM_FRAGMENTS_MASK 0x18000
+#define CB_COLOR2_ATTRIB__NUM_FRAGMENTS__SHIFT 0xf
+#define CB_COLOR2_ATTRIB__FORCE_DST_ALPHA_1_MASK 0x20000
+#define CB_COLOR2_ATTRIB__FORCE_DST_ALPHA_1__SHIFT 0x11
+#define CB_COLOR3_ATTRIB__TILE_MODE_INDEX_MASK 0x1f
+#define CB_COLOR3_ATTRIB__TILE_MODE_INDEX__SHIFT 0x0
+#define CB_COLOR3_ATTRIB__FMASK_TILE_MODE_INDEX_MASK 0x3e0
+#define CB_COLOR3_ATTRIB__FMASK_TILE_MODE_INDEX__SHIFT 0x5
+#define CB_COLOR3_ATTRIB__FMASK_BANK_HEIGHT_MASK 0xc00
+#define CB_COLOR3_ATTRIB__FMASK_BANK_HEIGHT__SHIFT 0xa
+#define CB_COLOR3_ATTRIB__NUM_SAMPLES_MASK 0x7000
+#define CB_COLOR3_ATTRIB__NUM_SAMPLES__SHIFT 0xc
+#define CB_COLOR3_ATTRIB__NUM_FRAGMENTS_MASK 0x18000
+#define CB_COLOR3_ATTRIB__NUM_FRAGMENTS__SHIFT 0xf
+#define CB_COLOR3_ATTRIB__FORCE_DST_ALPHA_1_MASK 0x20000
+#define CB_COLOR3_ATTRIB__FORCE_DST_ALPHA_1__SHIFT 0x11
+#define CB_COLOR4_ATTRIB__TILE_MODE_INDEX_MASK 0x1f
+#define CB_COLOR4_ATTRIB__TILE_MODE_INDEX__SHIFT 0x0
+#define CB_COLOR4_ATTRIB__FMASK_TILE_MODE_INDEX_MASK 0x3e0
+#define CB_COLOR4_ATTRIB__FMASK_TILE_MODE_INDEX__SHIFT 0x5
+#define CB_COLOR4_ATTRIB__FMASK_BANK_HEIGHT_MASK 0xc00
+#define CB_COLOR4_ATTRIB__FMASK_BANK_HEIGHT__SHIFT 0xa
+#define CB_COLOR4_ATTRIB__NUM_SAMPLES_MASK 0x7000
+#define CB_COLOR4_ATTRIB__NUM_SAMPLES__SHIFT 0xc
+#define CB_COLOR4_ATTRIB__NUM_FRAGMENTS_MASK 0x18000
+#define CB_COLOR4_ATTRIB__NUM_FRAGMENTS__SHIFT 0xf
+#define CB_COLOR4_ATTRIB__FORCE_DST_ALPHA_1_MASK 0x20000
+#define CB_COLOR4_ATTRIB__FORCE_DST_ALPHA_1__SHIFT 0x11
+#define CB_COLOR5_ATTRIB__TILE_MODE_INDEX_MASK 0x1f
+#define CB_COLOR5_ATTRIB__TILE_MODE_INDEX__SHIFT 0x0
+#define CB_COLOR5_ATTRIB__FMASK_TILE_MODE_INDEX_MASK 0x3e0
+#define CB_COLOR5_ATTRIB__FMASK_TILE_MODE_INDEX__SHIFT 0x5
+#define CB_COLOR5_ATTRIB__FMASK_BANK_HEIGHT_MASK 0xc00
+#define CB_COLOR5_ATTRIB__FMASK_BANK_HEIGHT__SHIFT 0xa
+#define CB_COLOR5_ATTRIB__NUM_SAMPLES_MASK 0x7000
+#define CB_COLOR5_ATTRIB__NUM_SAMPLES__SHIFT 0xc
+#define CB_COLOR5_ATTRIB__NUM_FRAGMENTS_MASK 0x18000
+#define CB_COLOR5_ATTRIB__NUM_FRAGMENTS__SHIFT 0xf
+#define CB_COLOR5_ATTRIB__FORCE_DST_ALPHA_1_MASK 0x20000
+#define CB_COLOR5_ATTRIB__FORCE_DST_ALPHA_1__SHIFT 0x11
+#define CB_COLOR6_ATTRIB__TILE_MODE_INDEX_MASK 0x1f
+#define CB_COLOR6_ATTRIB__TILE_MODE_INDEX__SHIFT 0x0
+#define CB_COLOR6_ATTRIB__FMASK_TILE_MODE_INDEX_MASK 0x3e0
+#define CB_COLOR6_ATTRIB__FMASK_TILE_MODE_INDEX__SHIFT 0x5
+#define CB_COLOR6_ATTRIB__FMASK_BANK_HEIGHT_MASK 0xc00
+#define CB_COLOR6_ATTRIB__FMASK_BANK_HEIGHT__SHIFT 0xa
+#define CB_COLOR6_ATTRIB__NUM_SAMPLES_MASK 0x7000
+#define CB_COLOR6_ATTRIB__NUM_SAMPLES__SHIFT 0xc
+#define CB_COLOR6_ATTRIB__NUM_FRAGMENTS_MASK 0x18000
+#define CB_COLOR6_ATTRIB__NUM_FRAGMENTS__SHIFT 0xf
+#define CB_COLOR6_ATTRIB__FORCE_DST_ALPHA_1_MASK 0x20000
+#define CB_COLOR6_ATTRIB__FORCE_DST_ALPHA_1__SHIFT 0x11
+#define CB_COLOR7_ATTRIB__TILE_MODE_INDEX_MASK 0x1f
+#define CB_COLOR7_ATTRIB__TILE_MODE_INDEX__SHIFT 0x0
+#define CB_COLOR7_ATTRIB__FMASK_TILE_MODE_INDEX_MASK 0x3e0
+#define CB_COLOR7_ATTRIB__FMASK_TILE_MODE_INDEX__SHIFT 0x5
+#define CB_COLOR7_ATTRIB__FMASK_BANK_HEIGHT_MASK 0xc00
+#define CB_COLOR7_ATTRIB__FMASK_BANK_HEIGHT__SHIFT 0xa
+#define CB_COLOR7_ATTRIB__NUM_SAMPLES_MASK 0x7000
+#define CB_COLOR7_ATTRIB__NUM_SAMPLES__SHIFT 0xc
+#define CB_COLOR7_ATTRIB__NUM_FRAGMENTS_MASK 0x18000
+#define CB_COLOR7_ATTRIB__NUM_FRAGMENTS__SHIFT 0xf
+#define CB_COLOR7_ATTRIB__FORCE_DST_ALPHA_1_MASK 0x20000
+#define CB_COLOR7_ATTRIB__FORCE_DST_ALPHA_1__SHIFT 0x11
+#define CB_COLOR0_DCC_CONTROL__OVERWRITE_COMBINER_DISABLE_MASK 0x1
+#define CB_COLOR0_DCC_CONTROL__OVERWRITE_COMBINER_DISABLE__SHIFT 0x0
+#define CB_COLOR0_DCC_CONTROL__KEY_CLEAR_ENABLE_MASK 0x2
+#define CB_COLOR0_DCC_CONTROL__KEY_CLEAR_ENABLE__SHIFT 0x1
+#define CB_COLOR0_DCC_CONTROL__MAX_UNCOMPRESSED_BLOCK_SIZE_MASK 0xc
+#define CB_COLOR0_DCC_CONTROL__MAX_UNCOMPRESSED_BLOCK_SIZE__SHIFT 0x2
+#define CB_COLOR0_DCC_CONTROL__MIN_COMPRESSED_BLOCK_SIZE_MASK 0x10
+#define CB_COLOR0_DCC_CONTROL__MIN_COMPRESSED_BLOCK_SIZE__SHIFT 0x4
+#define CB_COLOR0_DCC_CONTROL__MAX_COMPRESSED_BLOCK_SIZE_MASK 0x60
+#define CB_COLOR0_DCC_CONTROL__MAX_COMPRESSED_BLOCK_SIZE__SHIFT 0x5
+#define CB_COLOR0_DCC_CONTROL__COLOR_TRANSFORM_MASK 0x180
+#define CB_COLOR0_DCC_CONTROL__COLOR_TRANSFORM__SHIFT 0x7
+#define CB_COLOR0_DCC_CONTROL__INDEPENDENT_64B_BLOCKS_MASK 0x200
+#define CB_COLOR0_DCC_CONTROL__INDEPENDENT_64B_BLOCKS__SHIFT 0x9
+#define CB_COLOR0_DCC_CONTROL__LOSSY_RGB_PRECISION_MASK 0x3c00
+#define CB_COLOR0_DCC_CONTROL__LOSSY_RGB_PRECISION__SHIFT 0xa
+#define CB_COLOR0_DCC_CONTROL__LOSSY_ALPHA_PRECISION_MASK 0x3c000
+#define CB_COLOR0_DCC_CONTROL__LOSSY_ALPHA_PRECISION__SHIFT 0xe
+#define CB_COLOR1_DCC_CONTROL__OVERWRITE_COMBINER_DISABLE_MASK 0x1
+#define CB_COLOR1_DCC_CONTROL__OVERWRITE_COMBINER_DISABLE__SHIFT 0x0
+#define CB_COLOR1_DCC_CONTROL__KEY_CLEAR_ENABLE_MASK 0x2
+#define CB_COLOR1_DCC_CONTROL__KEY_CLEAR_ENABLE__SHIFT 0x1
+#define CB_COLOR1_DCC_CONTROL__MAX_UNCOMPRESSED_BLOCK_SIZE_MASK 0xc
+#define CB_COLOR1_DCC_CONTROL__MAX_UNCOMPRESSED_BLOCK_SIZE__SHIFT 0x2
+#define CB_COLOR1_DCC_CONTROL__MIN_COMPRESSED_BLOCK_SIZE_MASK 0x10
+#define CB_COLOR1_DCC_CONTROL__MIN_COMPRESSED_BLOCK_SIZE__SHIFT 0x4
+#define CB_COLOR1_DCC_CONTROL__MAX_COMPRESSED_BLOCK_SIZE_MASK 0x60
+#define CB_COLOR1_DCC_CONTROL__MAX_COMPRESSED_BLOCK_SIZE__SHIFT 0x5
+#define CB_COLOR1_DCC_CONTROL__COLOR_TRANSFORM_MASK 0x180
+#define CB_COLOR1_DCC_CONTROL__COLOR_TRANSFORM__SHIFT 0x7
+#define CB_COLOR1_DCC_CONTROL__INDEPENDENT_64B_BLOCKS_MASK 0x200
+#define CB_COLOR1_DCC_CONTROL__INDEPENDENT_64B_BLOCKS__SHIFT 0x9
+#define CB_COLOR1_DCC_CONTROL__LOSSY_RGB_PRECISION_MASK 0x3c00
+#define CB_COLOR1_DCC_CONTROL__LOSSY_RGB_PRECISION__SHIFT 0xa
+#define CB_COLOR1_DCC_CONTROL__LOSSY_ALPHA_PRECISION_MASK 0x3c000
+#define CB_COLOR1_DCC_CONTROL__LOSSY_ALPHA_PRECISION__SHIFT 0xe
+#define CB_COLOR2_DCC_CONTROL__OVERWRITE_COMBINER_DISABLE_MASK 0x1
+#define CB_COLOR2_DCC_CONTROL__OVERWRITE_COMBINER_DISABLE__SHIFT 0x0
+#define CB_COLOR2_DCC_CONTROL__KEY_CLEAR_ENABLE_MASK 0x2
+#define CB_COLOR2_DCC_CONTROL__KEY_CLEAR_ENABLE__SHIFT 0x1
+#define CB_COLOR2_DCC_CONTROL__MAX_UNCOMPRESSED_BLOCK_SIZE_MASK 0xc
+#define CB_COLOR2_DCC_CONTROL__MAX_UNCOMPRESSED_BLOCK_SIZE__SHIFT 0x2
+#define CB_COLOR2_DCC_CONTROL__MIN_COMPRESSED_BLOCK_SIZE_MASK 0x10
+#define CB_COLOR2_DCC_CONTROL__MIN_COMPRESSED_BLOCK_SIZE__SHIFT 0x4
+#define CB_COLOR2_DCC_CONTROL__MAX_COMPRESSED_BLOCK_SIZE_MASK 0x60
+#define CB_COLOR2_DCC_CONTROL__MAX_COMPRESSED_BLOCK_SIZE__SHIFT 0x5
+#define CB_COLOR2_DCC_CONTROL__COLOR_TRANSFORM_MASK 0x180
+#define CB_COLOR2_DCC_CONTROL__COLOR_TRANSFORM__SHIFT 0x7
+#define CB_COLOR2_DCC_CONTROL__INDEPENDENT_64B_BLOCKS_MASK 0x200
+#define CB_COLOR2_DCC_CONTROL__INDEPENDENT_64B_BLOCKS__SHIFT 0x9
+#define CB_COLOR2_DCC_CONTROL__LOSSY_RGB_PRECISION_MASK 0x3c00
+#define CB_COLOR2_DCC_CONTROL__LOSSY_RGB_PRECISION__SHIFT 0xa
+#define CB_COLOR2_DCC_CONTROL__LOSSY_ALPHA_PRECISION_MASK 0x3c000
+#define CB_COLOR2_DCC_CONTROL__LOSSY_ALPHA_PRECISION__SHIFT 0xe
+#define CB_COLOR3_DCC_CONTROL__OVERWRITE_COMBINER_DISABLE_MASK 0x1
+#define CB_COLOR3_DCC_CONTROL__OVERWRITE_COMBINER_DISABLE__SHIFT 0x0
+#define CB_COLOR3_DCC_CONTROL__KEY_CLEAR_ENABLE_MASK 0x2
+#define CB_COLOR3_DCC_CONTROL__KEY_CLEAR_ENABLE__SHIFT 0x1
+#define CB_COLOR3_DCC_CONTROL__MAX_UNCOMPRESSED_BLOCK_SIZE_MASK 0xc
+#define CB_COLOR3_DCC_CONTROL__MAX_UNCOMPRESSED_BLOCK_SIZE__SHIFT 0x2
+#define CB_COLOR3_DCC_CONTROL__MIN_COMPRESSED_BLOCK_SIZE_MASK 0x10
+#define CB_COLOR3_DCC_CONTROL__MIN_COMPRESSED_BLOCK_SIZE__SHIFT 0x4
+#define CB_COLOR3_DCC_CONTROL__MAX_COMPRESSED_BLOCK_SIZE_MASK 0x60
+#define CB_COLOR3_DCC_CONTROL__MAX_COMPRESSED_BLOCK_SIZE__SHIFT 0x5
+#define CB_COLOR3_DCC_CONTROL__COLOR_TRANSFORM_MASK 0x180
+#define CB_COLOR3_DCC_CONTROL__COLOR_TRANSFORM__SHIFT 0x7
+#define CB_COLOR3_DCC_CONTROL__INDEPENDENT_64B_BLOCKS_MASK 0x200
+#define CB_COLOR3_DCC_CONTROL__INDEPENDENT_64B_BLOCKS__SHIFT 0x9
+#define CB_COLOR3_DCC_CONTROL__LOSSY_RGB_PRECISION_MASK 0x3c00
+#define CB_COLOR3_DCC_CONTROL__LOSSY_RGB_PRECISION__SHIFT 0xa
+#define CB_COLOR3_DCC_CONTROL__LOSSY_ALPHA_PRECISION_MASK 0x3c000
+#define CB_COLOR3_DCC_CONTROL__LOSSY_ALPHA_PRECISION__SHIFT 0xe
+#define CB_COLOR4_DCC_CONTROL__OVERWRITE_COMBINER_DISABLE_MASK 0x1
+#define CB_COLOR4_DCC_CONTROL__OVERWRITE_COMBINER_DISABLE__SHIFT 0x0
+#define CB_COLOR4_DCC_CONTROL__KEY_CLEAR_ENABLE_MASK 0x2
+#define CB_COLOR4_DCC_CONTROL__KEY_CLEAR_ENABLE__SHIFT 0x1
+#define CB_COLOR4_DCC_CONTROL__MAX_UNCOMPRESSED_BLOCK_SIZE_MASK 0xc
+#define CB_COLOR4_DCC_CONTROL__MAX_UNCOMPRESSED_BLOCK_SIZE__SHIFT 0x2
+#define CB_COLOR4_DCC_CONTROL__MIN_COMPRESSED_BLOCK_SIZE_MASK 0x10
+#define CB_COLOR4_DCC_CONTROL__MIN_COMPRESSED_BLOCK_SIZE__SHIFT 0x4
+#define CB_COLOR4_DCC_CONTROL__MAX_COMPRESSED_BLOCK_SIZE_MASK 0x60
+#define CB_COLOR4_DCC_CONTROL__MAX_COMPRESSED_BLOCK_SIZE__SHIFT 0x5
+#define CB_COLOR4_DCC_CONTROL__COLOR_TRANSFORM_MASK 0x180
+#define CB_COLOR4_DCC_CONTROL__COLOR_TRANSFORM__SHIFT 0x7
+#define CB_COLOR4_DCC_CONTROL__INDEPENDENT_64B_BLOCKS_MASK 0x200
+#define CB_COLOR4_DCC_CONTROL__INDEPENDENT_64B_BLOCKS__SHIFT 0x9
+#define CB_COLOR4_DCC_CONTROL__LOSSY_RGB_PRECISION_MASK 0x3c00
+#define CB_COLOR4_DCC_CONTROL__LOSSY_RGB_PRECISION__SHIFT 0xa
+#define CB_COLOR4_DCC_CONTROL__LOSSY_ALPHA_PRECISION_MASK 0x3c000
+#define CB_COLOR4_DCC_CONTROL__LOSSY_ALPHA_PRECISION__SHIFT 0xe
+#define CB_COLOR5_DCC_CONTROL__OVERWRITE_COMBINER_DISABLE_MASK 0x1
+#define CB_COLOR5_DCC_CONTROL__OVERWRITE_COMBINER_DISABLE__SHIFT 0x0
+#define CB_COLOR5_DCC_CONTROL__KEY_CLEAR_ENABLE_MASK 0x2
+#define CB_COLOR5_DCC_CONTROL__KEY_CLEAR_ENABLE__SHIFT 0x1
+#define CB_COLOR5_DCC_CONTROL__MAX_UNCOMPRESSED_BLOCK_SIZE_MASK 0xc
+#define CB_COLOR5_DCC_CONTROL__MAX_UNCOMPRESSED_BLOCK_SIZE__SHIFT 0x2
+#define CB_COLOR5_DCC_CONTROL__MIN_COMPRESSED_BLOCK_SIZE_MASK 0x10
+#define CB_COLOR5_DCC_CONTROL__MIN_COMPRESSED_BLOCK_SIZE__SHIFT 0x4
+#define CB_COLOR5_DCC_CONTROL__MAX_COMPRESSED_BLOCK_SIZE_MASK 0x60
+#define CB_COLOR5_DCC_CONTROL__MAX_COMPRESSED_BLOCK_SIZE__SHIFT 0x5
+#define CB_COLOR5_DCC_CONTROL__COLOR_TRANSFORM_MASK 0x180
+#define CB_COLOR5_DCC_CONTROL__COLOR_TRANSFORM__SHIFT 0x7
+#define CB_COLOR5_DCC_CONTROL__INDEPENDENT_64B_BLOCKS_MASK 0x200
+#define CB_COLOR5_DCC_CONTROL__INDEPENDENT_64B_BLOCKS__SHIFT 0x9
+#define CB_COLOR5_DCC_CONTROL__LOSSY_RGB_PRECISION_MASK 0x3c00
+#define CB_COLOR5_DCC_CONTROL__LOSSY_RGB_PRECISION__SHIFT 0xa
+#define CB_COLOR5_DCC_CONTROL__LOSSY_ALPHA_PRECISION_MASK 0x3c000
+#define CB_COLOR5_DCC_CONTROL__LOSSY_ALPHA_PRECISION__SHIFT 0xe
+#define CB_COLOR6_DCC_CONTROL__OVERWRITE_COMBINER_DISABLE_MASK 0x1
+#define CB_COLOR6_DCC_CONTROL__OVERWRITE_COMBINER_DISABLE__SHIFT 0x0
+#define CB_COLOR6_DCC_CONTROL__KEY_CLEAR_ENABLE_MASK 0x2
+#define CB_COLOR6_DCC_CONTROL__KEY_CLEAR_ENABLE__SHIFT 0x1
+#define CB_COLOR6_DCC_CONTROL__MAX_UNCOMPRESSED_BLOCK_SIZE_MASK 0xc
+#define CB_COLOR6_DCC_CONTROL__MAX_UNCOMPRESSED_BLOCK_SIZE__SHIFT 0x2
+#define CB_COLOR6_DCC_CONTROL__MIN_COMPRESSED_BLOCK_SIZE_MASK 0x10
+#define CB_COLOR6_DCC_CONTROL__MIN_COMPRESSED_BLOCK_SIZE__SHIFT 0x4
+#define CB_COLOR6_DCC_CONTROL__MAX_COMPRESSED_BLOCK_SIZE_MASK 0x60
+#define CB_COLOR6_DCC_CONTROL__MAX_COMPRESSED_BLOCK_SIZE__SHIFT 0x5
+#define CB_COLOR6_DCC_CONTROL__COLOR_TRANSFORM_MASK 0x180
+#define CB_COLOR6_DCC_CONTROL__COLOR_TRANSFORM__SHIFT 0x7
+#define CB_COLOR6_DCC_CONTROL__INDEPENDENT_64B_BLOCKS_MASK 0x200
+#define CB_COLOR6_DCC_CONTROL__INDEPENDENT_64B_BLOCKS__SHIFT 0x9
+#define CB_COLOR6_DCC_CONTROL__LOSSY_RGB_PRECISION_MASK 0x3c00
+#define CB_COLOR6_DCC_CONTROL__LOSSY_RGB_PRECISION__SHIFT 0xa
+#define CB_COLOR6_DCC_CONTROL__LOSSY_ALPHA_PRECISION_MASK 0x3c000
+#define CB_COLOR6_DCC_CONTROL__LOSSY_ALPHA_PRECISION__SHIFT 0xe
+#define CB_COLOR7_DCC_CONTROL__OVERWRITE_COMBINER_DISABLE_MASK 0x1
+#define CB_COLOR7_DCC_CONTROL__OVERWRITE_COMBINER_DISABLE__SHIFT 0x0
+#define CB_COLOR7_DCC_CONTROL__KEY_CLEAR_ENABLE_MASK 0x2
+#define CB_COLOR7_DCC_CONTROL__KEY_CLEAR_ENABLE__SHIFT 0x1
+#define CB_COLOR7_DCC_CONTROL__MAX_UNCOMPRESSED_BLOCK_SIZE_MASK 0xc
+#define CB_COLOR7_DCC_CONTROL__MAX_UNCOMPRESSED_BLOCK_SIZE__SHIFT 0x2
+#define CB_COLOR7_DCC_CONTROL__MIN_COMPRESSED_BLOCK_SIZE_MASK 0x10
+#define CB_COLOR7_DCC_CONTROL__MIN_COMPRESSED_BLOCK_SIZE__SHIFT 0x4
+#define CB_COLOR7_DCC_CONTROL__MAX_COMPRESSED_BLOCK_SIZE_MASK 0x60
+#define CB_COLOR7_DCC_CONTROL__MAX_COMPRESSED_BLOCK_SIZE__SHIFT 0x5
+#define CB_COLOR7_DCC_CONTROL__COLOR_TRANSFORM_MASK 0x180
+#define CB_COLOR7_DCC_CONTROL__COLOR_TRANSFORM__SHIFT 0x7
+#define CB_COLOR7_DCC_CONTROL__INDEPENDENT_64B_BLOCKS_MASK 0x200
+#define CB_COLOR7_DCC_CONTROL__INDEPENDENT_64B_BLOCKS__SHIFT 0x9
+#define CB_COLOR7_DCC_CONTROL__LOSSY_RGB_PRECISION_MASK 0x3c00
+#define CB_COLOR7_DCC_CONTROL__LOSSY_RGB_PRECISION__SHIFT 0xa
+#define CB_COLOR7_DCC_CONTROL__LOSSY_ALPHA_PRECISION_MASK 0x3c000
+#define CB_COLOR7_DCC_CONTROL__LOSSY_ALPHA_PRECISION__SHIFT 0xe
+#define CB_COLOR0_CMASK__BASE_256B_MASK 0xffffffff
+#define CB_COLOR0_CMASK__BASE_256B__SHIFT 0x0
+#define CB_COLOR1_CMASK__BASE_256B_MASK 0xffffffff
+#define CB_COLOR1_CMASK__BASE_256B__SHIFT 0x0
+#define CB_COLOR2_CMASK__BASE_256B_MASK 0xffffffff
+#define CB_COLOR2_CMASK__BASE_256B__SHIFT 0x0
+#define CB_COLOR3_CMASK__BASE_256B_MASK 0xffffffff
+#define CB_COLOR3_CMASK__BASE_256B__SHIFT 0x0
+#define CB_COLOR4_CMASK__BASE_256B_MASK 0xffffffff
+#define CB_COLOR4_CMASK__BASE_256B__SHIFT 0x0
+#define CB_COLOR5_CMASK__BASE_256B_MASK 0xffffffff
+#define CB_COLOR5_CMASK__BASE_256B__SHIFT 0x0
+#define CB_COLOR6_CMASK__BASE_256B_MASK 0xffffffff
+#define CB_COLOR6_CMASK__BASE_256B__SHIFT 0x0
+#define CB_COLOR7_CMASK__BASE_256B_MASK 0xffffffff
+#define CB_COLOR7_CMASK__BASE_256B__SHIFT 0x0
+#define CB_COLOR0_CMASK_SLICE__TILE_MAX_MASK 0x3fff
+#define CB_COLOR0_CMASK_SLICE__TILE_MAX__SHIFT 0x0
+#define CB_COLOR1_CMASK_SLICE__TILE_MAX_MASK 0x3fff
+#define CB_COLOR1_CMASK_SLICE__TILE_MAX__SHIFT 0x0
+#define CB_COLOR2_CMASK_SLICE__TILE_MAX_MASK 0x3fff
+#define CB_COLOR2_CMASK_SLICE__TILE_MAX__SHIFT 0x0
+#define CB_COLOR3_CMASK_SLICE__TILE_MAX_MASK 0x3fff
+#define CB_COLOR3_CMASK_SLICE__TILE_MAX__SHIFT 0x0
+#define CB_COLOR4_CMASK_SLICE__TILE_MAX_MASK 0x3fff
+#define CB_COLOR4_CMASK_SLICE__TILE_MAX__SHIFT 0x0
+#define CB_COLOR5_CMASK_SLICE__TILE_MAX_MASK 0x3fff
+#define CB_COLOR5_CMASK_SLICE__TILE_MAX__SHIFT 0x0
+#define CB_COLOR6_CMASK_SLICE__TILE_MAX_MASK 0x3fff
+#define CB_COLOR6_CMASK_SLICE__TILE_MAX__SHIFT 0x0
+#define CB_COLOR7_CMASK_SLICE__TILE_MAX_MASK 0x3fff
+#define CB_COLOR7_CMASK_SLICE__TILE_MAX__SHIFT 0x0
+#define CB_COLOR0_FMASK__BASE_256B_MASK 0xffffffff
+#define CB_COLOR0_FMASK__BASE_256B__SHIFT 0x0
+#define CB_COLOR1_FMASK__BASE_256B_MASK 0xffffffff
+#define CB_COLOR1_FMASK__BASE_256B__SHIFT 0x0
+#define CB_COLOR2_FMASK__BASE_256B_MASK 0xffffffff
+#define CB_COLOR2_FMASK__BASE_256B__SHIFT 0x0
+#define CB_COLOR3_FMASK__BASE_256B_MASK 0xffffffff
+#define CB_COLOR3_FMASK__BASE_256B__SHIFT 0x0
+#define CB_COLOR4_FMASK__BASE_256B_MASK 0xffffffff
+#define CB_COLOR4_FMASK__BASE_256B__SHIFT 0x0
+#define CB_COLOR5_FMASK__BASE_256B_MASK 0xffffffff
+#define CB_COLOR5_FMASK__BASE_256B__SHIFT 0x0
+#define CB_COLOR6_FMASK__BASE_256B_MASK 0xffffffff
+#define CB_COLOR6_FMASK__BASE_256B__SHIFT 0x0
+#define CB_COLOR7_FMASK__BASE_256B_MASK 0xffffffff
+#define CB_COLOR7_FMASK__BASE_256B__SHIFT 0x0
+#define CB_COLOR0_FMASK_SLICE__TILE_MAX_MASK 0x3fffff
+#define CB_COLOR0_FMASK_SLICE__TILE_MAX__SHIFT 0x0
+#define CB_COLOR1_FMASK_SLICE__TILE_MAX_MASK 0x3fffff
+#define CB_COLOR1_FMASK_SLICE__TILE_MAX__SHIFT 0x0
+#define CB_COLOR2_FMASK_SLICE__TILE_MAX_MASK 0x3fffff
+#define CB_COLOR2_FMASK_SLICE__TILE_MAX__SHIFT 0x0
+#define CB_COLOR3_FMASK_SLICE__TILE_MAX_MASK 0x3fffff
+#define CB_COLOR3_FMASK_SLICE__TILE_MAX__SHIFT 0x0
+#define CB_COLOR4_FMASK_SLICE__TILE_MAX_MASK 0x3fffff
+#define CB_COLOR4_FMASK_SLICE__TILE_MAX__SHIFT 0x0
+#define CB_COLOR5_FMASK_SLICE__TILE_MAX_MASK 0x3fffff
+#define CB_COLOR5_FMASK_SLICE__TILE_MAX__SHIFT 0x0
+#define CB_COLOR6_FMASK_SLICE__TILE_MAX_MASK 0x3fffff
+#define CB_COLOR6_FMASK_SLICE__TILE_MAX__SHIFT 0x0
+#define CB_COLOR7_FMASK_SLICE__TILE_MAX_MASK 0x3fffff
+#define CB_COLOR7_FMASK_SLICE__TILE_MAX__SHIFT 0x0
+#define CB_COLOR0_CLEAR_WORD0__CLEAR_WORD0_MASK 0xffffffff
+#define CB_COLOR0_CLEAR_WORD0__CLEAR_WORD0__SHIFT 0x0
+#define CB_COLOR1_CLEAR_WORD0__CLEAR_WORD0_MASK 0xffffffff
+#define CB_COLOR1_CLEAR_WORD0__CLEAR_WORD0__SHIFT 0x0
+#define CB_COLOR2_CLEAR_WORD0__CLEAR_WORD0_MASK 0xffffffff
+#define CB_COLOR2_CLEAR_WORD0__CLEAR_WORD0__SHIFT 0x0
+#define CB_COLOR3_CLEAR_WORD0__CLEAR_WORD0_MASK 0xffffffff
+#define CB_COLOR3_CLEAR_WORD0__CLEAR_WORD0__SHIFT 0x0
+#define CB_COLOR4_CLEAR_WORD0__CLEAR_WORD0_MASK 0xffffffff
+#define CB_COLOR4_CLEAR_WORD0__CLEAR_WORD0__SHIFT 0x0
+#define CB_COLOR5_CLEAR_WORD0__CLEAR_WORD0_MASK 0xffffffff
+#define CB_COLOR5_CLEAR_WORD0__CLEAR_WORD0__SHIFT 0x0
+#define CB_COLOR6_CLEAR_WORD0__CLEAR_WORD0_MASK 0xffffffff
+#define CB_COLOR6_CLEAR_WORD0__CLEAR_WORD0__SHIFT 0x0
+#define CB_COLOR7_CLEAR_WORD0__CLEAR_WORD0_MASK 0xffffffff
+#define CB_COLOR7_CLEAR_WORD0__CLEAR_WORD0__SHIFT 0x0
+#define CB_COLOR0_CLEAR_WORD1__CLEAR_WORD1_MASK 0xffffffff
+#define CB_COLOR0_CLEAR_WORD1__CLEAR_WORD1__SHIFT 0x0
+#define CB_COLOR1_CLEAR_WORD1__CLEAR_WORD1_MASK 0xffffffff
+#define CB_COLOR1_CLEAR_WORD1__CLEAR_WORD1__SHIFT 0x0
+#define CB_COLOR2_CLEAR_WORD1__CLEAR_WORD1_MASK 0xffffffff
+#define CB_COLOR2_CLEAR_WORD1__CLEAR_WORD1__SHIFT 0x0
+#define CB_COLOR3_CLEAR_WORD1__CLEAR_WORD1_MASK 0xffffffff
+#define CB_COLOR3_CLEAR_WORD1__CLEAR_WORD1__SHIFT 0x0
+#define CB_COLOR4_CLEAR_WORD1__CLEAR_WORD1_MASK 0xffffffff
+#define CB_COLOR4_CLEAR_WORD1__CLEAR_WORD1__SHIFT 0x0
+#define CB_COLOR5_CLEAR_WORD1__CLEAR_WORD1_MASK 0xffffffff
+#define CB_COLOR5_CLEAR_WORD1__CLEAR_WORD1__SHIFT 0x0
+#define CB_COLOR6_CLEAR_WORD1__CLEAR_WORD1_MASK 0xffffffff
+#define CB_COLOR6_CLEAR_WORD1__CLEAR_WORD1__SHIFT 0x0
+#define CB_COLOR7_CLEAR_WORD1__CLEAR_WORD1_MASK 0xffffffff
+#define CB_COLOR7_CLEAR_WORD1__CLEAR_WORD1__SHIFT 0x0
+#define CB_COLOR0_DCC_BASE__BASE_256B_MASK 0xffffffff
+#define CB_COLOR0_DCC_BASE__BASE_256B__SHIFT 0x0
+#define CB_COLOR1_DCC_BASE__BASE_256B_MASK 0xffffffff
+#define CB_COLOR1_DCC_BASE__BASE_256B__SHIFT 0x0
+#define CB_COLOR2_DCC_BASE__BASE_256B_MASK 0xffffffff
+#define CB_COLOR2_DCC_BASE__BASE_256B__SHIFT 0x0
+#define CB_COLOR3_DCC_BASE__BASE_256B_MASK 0xffffffff
+#define CB_COLOR3_DCC_BASE__BASE_256B__SHIFT 0x0
+#define CB_COLOR4_DCC_BASE__BASE_256B_MASK 0xffffffff
+#define CB_COLOR4_DCC_BASE__BASE_256B__SHIFT 0x0
+#define CB_COLOR5_DCC_BASE__BASE_256B_MASK 0xffffffff
+#define CB_COLOR5_DCC_BASE__BASE_256B__SHIFT 0x0
+#define CB_COLOR6_DCC_BASE__BASE_256B_MASK 0xffffffff
+#define CB_COLOR6_DCC_BASE__BASE_256B__SHIFT 0x0
+#define CB_COLOR7_DCC_BASE__BASE_256B_MASK 0xffffffff
+#define CB_COLOR7_DCC_BASE__BASE_256B__SHIFT 0x0
+#define CB_TARGET_MASK__TARGET0_ENABLE_MASK 0xf
+#define CB_TARGET_MASK__TARGET0_ENABLE__SHIFT 0x0
+#define CB_TARGET_MASK__TARGET1_ENABLE_MASK 0xf0
+#define CB_TARGET_MASK__TARGET1_ENABLE__SHIFT 0x4
+#define CB_TARGET_MASK__TARGET2_ENABLE_MASK 0xf00
+#define CB_TARGET_MASK__TARGET2_ENABLE__SHIFT 0x8
+#define CB_TARGET_MASK__TARGET3_ENABLE_MASK 0xf000
+#define CB_TARGET_MASK__TARGET3_ENABLE__SHIFT 0xc
+#define CB_TARGET_MASK__TARGET4_ENABLE_MASK 0xf0000
+#define CB_TARGET_MASK__TARGET4_ENABLE__SHIFT 0x10
+#define CB_TARGET_MASK__TARGET5_ENABLE_MASK 0xf00000
+#define CB_TARGET_MASK__TARGET5_ENABLE__SHIFT 0x14
+#define CB_TARGET_MASK__TARGET6_ENABLE_MASK 0xf000000
+#define CB_TARGET_MASK__TARGET6_ENABLE__SHIFT 0x18
+#define CB_TARGET_MASK__TARGET7_ENABLE_MASK 0xf0000000
+#define CB_TARGET_MASK__TARGET7_ENABLE__SHIFT 0x1c
+#define CB_SHADER_MASK__OUTPUT0_ENABLE_MASK 0xf
+#define CB_SHADER_MASK__OUTPUT0_ENABLE__SHIFT 0x0
+#define CB_SHADER_MASK__OUTPUT1_ENABLE_MASK 0xf0
+#define CB_SHADER_MASK__OUTPUT1_ENABLE__SHIFT 0x4
+#define CB_SHADER_MASK__OUTPUT2_ENABLE_MASK 0xf00
+#define CB_SHADER_MASK__OUTPUT2_ENABLE__SHIFT 0x8
+#define CB_SHADER_MASK__OUTPUT3_ENABLE_MASK 0xf000
+#define CB_SHADER_MASK__OUTPUT3_ENABLE__SHIFT 0xc
+#define CB_SHADER_MASK__OUTPUT4_ENABLE_MASK 0xf0000
+#define CB_SHADER_MASK__OUTPUT4_ENABLE__SHIFT 0x10
+#define CB_SHADER_MASK__OUTPUT5_ENABLE_MASK 0xf00000
+#define CB_SHADER_MASK__OUTPUT5_ENABLE__SHIFT 0x14
+#define CB_SHADER_MASK__OUTPUT6_ENABLE_MASK 0xf000000
+#define CB_SHADER_MASK__OUTPUT6_ENABLE__SHIFT 0x18
+#define CB_SHADER_MASK__OUTPUT7_ENABLE_MASK 0xf0000000
+#define CB_SHADER_MASK__OUTPUT7_ENABLE__SHIFT 0x1c
+#define CB_HW_CONTROL__CM_CACHE_EVICT_POINT_MASK 0xf
+#define CB_HW_CONTROL__CM_CACHE_EVICT_POINT__SHIFT 0x0
+#define CB_HW_CONTROL__FC_CACHE_EVICT_POINT_MASK 0x3c0
+#define CB_HW_CONTROL__FC_CACHE_EVICT_POINT__SHIFT 0x6
+#define CB_HW_CONTROL__CC_CACHE_EVICT_POINT_MASK 0xf000
+#define CB_HW_CONTROL__CC_CACHE_EVICT_POINT__SHIFT 0xc
+#define CB_HW_CONTROL__ALLOW_MRT_WITH_DUAL_SOURCE_MASK 0x10000
+#define CB_HW_CONTROL__ALLOW_MRT_WITH_DUAL_SOURCE__SHIFT 0x10
+#define CB_HW_CONTROL__DISABLE_INTNORM_LE11BPC_CLAMPING_MASK 0x40000
+#define CB_HW_CONTROL__DISABLE_INTNORM_LE11BPC_CLAMPING__SHIFT 0x12
+#define CB_HW_CONTROL__FORCE_NEEDS_DST_MASK 0x80000
+#define CB_HW_CONTROL__FORCE_NEEDS_DST__SHIFT 0x13
+#define CB_HW_CONTROL__FORCE_ALWAYS_TOGGLE_MASK 0x100000
+#define CB_HW_CONTROL__FORCE_ALWAYS_TOGGLE__SHIFT 0x14
+#define CB_HW_CONTROL__DISABLE_BLEND_OPT_RESULT_EQ_DEST_MASK 0x200000
+#define CB_HW_CONTROL__DISABLE_BLEND_OPT_RESULT_EQ_DEST__SHIFT 0x15
+#define CB_HW_CONTROL__DISABLE_FULL_WRITE_MASK_MASK 0x400000
+#define CB_HW_CONTROL__DISABLE_FULL_WRITE_MASK__SHIFT 0x16
+#define CB_HW_CONTROL__DISABLE_RESOLVE_OPT_FOR_SINGLE_FRAG_MASK 0x800000
+#define CB_HW_CONTROL__DISABLE_RESOLVE_OPT_FOR_SINGLE_FRAG__SHIFT 0x17
+#define CB_HW_CONTROL__DISABLE_BLEND_OPT_DONT_RD_DST_MASK 0x1000000
+#define CB_HW_CONTROL__DISABLE_BLEND_OPT_DONT_RD_DST__SHIFT 0x18
+#define CB_HW_CONTROL__DISABLE_BLEND_OPT_BYPASS_MASK 0x2000000
+#define CB_HW_CONTROL__DISABLE_BLEND_OPT_BYPASS__SHIFT 0x19
+#define CB_HW_CONTROL__DISABLE_BLEND_OPT_DISCARD_PIXEL_MASK 0x4000000
+#define CB_HW_CONTROL__DISABLE_BLEND_OPT_DISCARD_PIXEL__SHIFT 0x1a
+#define CB_HW_CONTROL__DISABLE_BLEND_OPT_WHEN_DISABLED_SRCALPHA_IS_USED_MASK 0x8000000
+#define CB_HW_CONTROL__DISABLE_BLEND_OPT_WHEN_DISABLED_SRCALPHA_IS_USED__SHIFT 0x1b
+#define CB_HW_CONTROL__PRIORITIZE_FC_WR_OVER_FC_RD_ON_CMASK_CONFLICT_MASK 0x10000000
+#define CB_HW_CONTROL__PRIORITIZE_FC_WR_OVER_FC_RD_ON_CMASK_CONFLICT__SHIFT 0x1c
+#define CB_HW_CONTROL__PRIORITIZE_FC_EVICT_OVER_FOP_RD_ON_BANK_CONFLICT_MASK 0x20000000
+#define CB_HW_CONTROL__PRIORITIZE_FC_EVICT_OVER_FOP_RD_ON_BANK_CONFLICT__SHIFT 0x1d
+#define CB_HW_CONTROL__DISABLE_CC_IB_SERIALIZER_STATE_OPT_MASK 0x40000000
+#define CB_HW_CONTROL__DISABLE_CC_IB_SERIALIZER_STATE_OPT__SHIFT 0x1e
+#define CB_HW_CONTROL__DISABLE_PIXEL_IN_QUAD_FIX_FOR_LINEAR_SURFACE_MASK 0x80000000
+#define CB_HW_CONTROL__DISABLE_PIXEL_IN_QUAD_FIX_FOR_LINEAR_SURFACE__SHIFT 0x1f
+#define CB_HW_CONTROL_1__CM_CACHE_NUM_TAGS_MASK 0x1f
+#define CB_HW_CONTROL_1__CM_CACHE_NUM_TAGS__SHIFT 0x0
+#define CB_HW_CONTROL_1__FC_CACHE_NUM_TAGS_MASK 0x7e0
+#define CB_HW_CONTROL_1__FC_CACHE_NUM_TAGS__SHIFT 0x5
+#define CB_HW_CONTROL_1__CC_CACHE_NUM_TAGS_MASK 0x1f800
+#define CB_HW_CONTROL_1__CC_CACHE_NUM_TAGS__SHIFT 0xb
+#define CB_HW_CONTROL_1__CM_TILE_FIFO_DEPTH_MASK 0x3fe0000
+#define CB_HW_CONTROL_1__CM_TILE_FIFO_DEPTH__SHIFT 0x11
+#define CB_HW_CONTROL_1__CHICKEN_BITS_MASK 0xfc000000
+#define CB_HW_CONTROL_1__CHICKEN_BITS__SHIFT 0x1a
+#define CB_HW_CONTROL_2__CC_EVEN_ODD_FIFO_DEPTH_MASK 0xff
+#define CB_HW_CONTROL_2__CC_EVEN_ODD_FIFO_DEPTH__SHIFT 0x0
+#define CB_HW_CONTROL_2__FC_RDLAT_TILE_FIFO_DEPTH_MASK 0x7f00
+#define CB_HW_CONTROL_2__FC_RDLAT_TILE_FIFO_DEPTH__SHIFT 0x8
+#define CB_HW_CONTROL_2__FC_RDLAT_QUAD_FIFO_DEPTH_MASK 0x7f8000
+#define CB_HW_CONTROL_2__FC_RDLAT_QUAD_FIFO_DEPTH__SHIFT 0xf
+#define CB_HW_CONTROL_2__DRR_ASSUMED_FIFO_DEPTH_DIV8_MASK 0xf000000
+#define CB_HW_CONTROL_2__DRR_ASSUMED_FIFO_DEPTH_DIV8__SHIFT 0x18
+#define CB_HW_CONTROL_2__CHICKEN_BITS_MASK 0xf0000000
+#define CB_HW_CONTROL_2__CHICKEN_BITS__SHIFT 0x1c
+#define CB_HW_CONTROL_3__DISABLE_SLOW_MODE_EMPTY_HALF_QUAD_KILL_MASK 0x1
+#define CB_HW_CONTROL_3__DISABLE_SLOW_MODE_EMPTY_HALF_QUAD_KILL__SHIFT 0x0
+#define CB_HW_CONTROL_3__RAM_ADDRESS_CONFLICTS_DISALLOWED_MASK 0x2
+#define CB_HW_CONTROL_3__RAM_ADDRESS_CONFLICTS_DISALLOWED__SHIFT 0x1
+#define CB_HW_CONTROL_3__DISABLE_FAST_CLEAR_FETCH_OPT_MASK 0x4
+#define CB_HW_CONTROL_3__DISABLE_FAST_CLEAR_FETCH_OPT__SHIFT 0x2
+#define CB_HW_CONTROL_3__DISABLE_QUAD_MARKER_DROP_STOP_MASK 0x8
+#define CB_HW_CONTROL_3__DISABLE_QUAD_MARKER_DROP_STOP__SHIFT 0x3
+#define CB_HW_CONTROL_3__DISABLE_OVERWRITE_COMBINER_CAM_CLR_MASK 0x10
+#define CB_HW_CONTROL_3__DISABLE_OVERWRITE_COMBINER_CAM_CLR__SHIFT 0x4
+#define CB_HW_CONTROL_3__DISABLE_CC_CACHE_OVWR_STATUS_ACCUM_MASK 0x20
+#define CB_HW_CONTROL_3__DISABLE_CC_CACHE_OVWR_STATUS_ACCUM__SHIFT 0x5
+#define CB_HW_CONTROL_3__DISABLE_CC_CACHE_PANIC_GATING_MASK 0x80
+#define CB_HW_CONTROL_3__DISABLE_CC_CACHE_PANIC_GATING__SHIFT 0x7
+#define CB_HW_CONTROL_3__DISABLE_OVERWRITE_COMBINER_TARGET_MASK_VALIDATION_MASK 0x100
+#define CB_HW_CONTROL_3__DISABLE_OVERWRITE_COMBINER_TARGET_MASK_VALIDATION__SHIFT 0x8
+#define CB_HW_CONTROL_3__SPLIT_ALL_FAST_MODE_TRANSFERS_MASK 0x200
+#define CB_HW_CONTROL_3__SPLIT_ALL_FAST_MODE_TRANSFERS__SHIFT 0x9
+#define CB_HW_CONTROL_3__DISABLE_SHADER_BLEND_OPTS_MASK 0x400
+#define CB_HW_CONTROL_3__DISABLE_SHADER_BLEND_OPTS__SHIFT 0xa
+#define CB_HW_CONTROL_3__DISABLE_CMASK_LAST_QUAD_INSERTION_MASK 0x800
+#define CB_HW_CONTROL_3__DISABLE_CMASK_LAST_QUAD_INSERTION__SHIFT 0xb
+#define CB_HW_CONTROL_3__DISABLE_ROP3_FIXES_OF_BUG_511967_MASK 0x1000
+#define CB_HW_CONTROL_3__DISABLE_ROP3_FIXES_OF_BUG_511967__SHIFT 0xc
+#define CB_HW_CONTROL_3__DISABLE_ROP3_FIXES_OF_BUG_520657_MASK 0x2000
+#define CB_HW_CONTROL_3__DISABLE_ROP3_FIXES_OF_BUG_520657__SHIFT 0xd
+#define CB_DCC_CONFIG__OVERWRITE_COMBINER_DEPTH_MASK 0x1f
+#define CB_DCC_CONFIG__OVERWRITE_COMBINER_DEPTH__SHIFT 0x0
+#define CB_DCC_CONFIG__OVERWRITE_COMBINER_DISABLE_MASK 0x20
+#define CB_DCC_CONFIG__OVERWRITE_COMBINER_DISABLE__SHIFT 0x5
+#define CB_DCC_CONFIG__OVERWRITE_COMBINER_CC_POP_DISABLE_MASK 0x40
+#define CB_DCC_CONFIG__OVERWRITE_COMBINER_CC_POP_DISABLE__SHIFT 0x6
+#define CB_DCC_CONFIG__FC_RDLAT_KEYID_FIFO_DEPTH_MASK 0xff00
+#define CB_DCC_CONFIG__FC_RDLAT_KEYID_FIFO_DEPTH__SHIFT 0x8
+#define CB_DCC_CONFIG__READ_RETURN_SKID_FIFO_DEPTH_MASK 0x7f0000
+#define CB_DCC_CONFIG__READ_RETURN_SKID_FIFO_DEPTH__SHIFT 0x10
+#define CB_DCC_CONFIG__DCC_CACHE_EVICT_POINT_MASK 0xf000000
+#define CB_DCC_CONFIG__DCC_CACHE_EVICT_POINT__SHIFT 0x18
+#define CB_DCC_CONFIG__DCC_CACHE_NUM_TAGS_MASK 0xf0000000
+#define CB_DCC_CONFIG__DCC_CACHE_NUM_TAGS__SHIFT 0x1c
+#define CB_PERFCOUNTER_FILTER__OP_FILTER_ENABLE_MASK 0x1
+#define CB_PERFCOUNTER_FILTER__OP_FILTER_ENABLE__SHIFT 0x0
+#define CB_PERFCOUNTER_FILTER__OP_FILTER_SEL_MASK 0xe
+#define CB_PERFCOUNTER_FILTER__OP_FILTER_SEL__SHIFT 0x1
+#define CB_PERFCOUNTER_FILTER__FORMAT_FILTER_ENABLE_MASK 0x10
+#define CB_PERFCOUNTER_FILTER__FORMAT_FILTER_ENABLE__SHIFT 0x4
+#define CB_PERFCOUNTER_FILTER__FORMAT_FILTER_SEL_MASK 0x3e0
+#define CB_PERFCOUNTER_FILTER__FORMAT_FILTER_SEL__SHIFT 0x5
+#define CB_PERFCOUNTER_FILTER__CLEAR_FILTER_ENABLE_MASK 0x400
+#define CB_PERFCOUNTER_FILTER__CLEAR_FILTER_ENABLE__SHIFT 0xa
+#define CB_PERFCOUNTER_FILTER__CLEAR_FILTER_SEL_MASK 0x800
+#define CB_PERFCOUNTER_FILTER__CLEAR_FILTER_SEL__SHIFT 0xb
+#define CB_PERFCOUNTER_FILTER__MRT_FILTER_ENABLE_MASK 0x1000
+#define CB_PERFCOUNTER_FILTER__MRT_FILTER_ENABLE__SHIFT 0xc
+#define CB_PERFCOUNTER_FILTER__MRT_FILTER_SEL_MASK 0xe000
+#define CB_PERFCOUNTER_FILTER__MRT_FILTER_SEL__SHIFT 0xd
+#define CB_PERFCOUNTER_FILTER__NUM_SAMPLES_FILTER_ENABLE_MASK 0x20000
+#define CB_PERFCOUNTER_FILTER__NUM_SAMPLES_FILTER_ENABLE__SHIFT 0x11
+#define CB_PERFCOUNTER_FILTER__NUM_SAMPLES_FILTER_SEL_MASK 0x1c0000
+#define CB_PERFCOUNTER_FILTER__NUM_SAMPLES_FILTER_SEL__SHIFT 0x12
+#define CB_PERFCOUNTER_FILTER__NUM_FRAGMENTS_FILTER_ENABLE_MASK 0x200000
+#define CB_PERFCOUNTER_FILTER__NUM_FRAGMENTS_FILTER_ENABLE__SHIFT 0x15
+#define CB_PERFCOUNTER_FILTER__NUM_FRAGMENTS_FILTER_SEL_MASK 0xc00000
+#define CB_PERFCOUNTER_FILTER__NUM_FRAGMENTS_FILTER_SEL__SHIFT 0x16
+#define CB_PERFCOUNTER0_SELECT__PERF_SEL_MASK 0x1ff
+#define CB_PERFCOUNTER0_SELECT__PERF_SEL__SHIFT 0x0
+#define CB_PERFCOUNTER0_SELECT__PERF_SEL1_MASK 0x7fc00
+#define CB_PERFCOUNTER0_SELECT__PERF_SEL1__SHIFT 0xa
+#define CB_PERFCOUNTER0_SELECT__CNTR_MODE_MASK 0xf00000
+#define CB_PERFCOUNTER0_SELECT__CNTR_MODE__SHIFT 0x14
+#define CB_PERFCOUNTER0_SELECT__PERF_MODE1_MASK 0xf000000
+#define CB_PERFCOUNTER0_SELECT__PERF_MODE1__SHIFT 0x18
+#define CB_PERFCOUNTER0_SELECT__PERF_MODE_MASK 0xf0000000
+#define CB_PERFCOUNTER0_SELECT__PERF_MODE__SHIFT 0x1c
+#define CB_PERFCOUNTER0_SELECT1__PERF_SEL2_MASK 0x1ff
+#define CB_PERFCOUNTER0_SELECT1__PERF_SEL2__SHIFT 0x0
+#define CB_PERFCOUNTER0_SELECT1__PERF_SEL3_MASK 0x7fc00
+#define CB_PERFCOUNTER0_SELECT1__PERF_SEL3__SHIFT 0xa
+#define CB_PERFCOUNTER0_SELECT1__PERF_MODE3_MASK 0xf000000
+#define CB_PERFCOUNTER0_SELECT1__PERF_MODE3__SHIFT 0x18
+#define CB_PERFCOUNTER0_SELECT1__PERF_MODE2_MASK 0xf0000000
+#define CB_PERFCOUNTER0_SELECT1__PERF_MODE2__SHIFT 0x1c
+#define CB_PERFCOUNTER1_SELECT__PERF_SEL_MASK 0x1ff
+#define CB_PERFCOUNTER1_SELECT__PERF_SEL__SHIFT 0x0
+#define CB_PERFCOUNTER1_SELECT__PERF_MODE_MASK 0xf0000000
+#define CB_PERFCOUNTER1_SELECT__PERF_MODE__SHIFT 0x1c
+#define CB_PERFCOUNTER2_SELECT__PERF_SEL_MASK 0x1ff
+#define CB_PERFCOUNTER2_SELECT__PERF_SEL__SHIFT 0x0
+#define CB_PERFCOUNTER2_SELECT__PERF_MODE_MASK 0xf0000000
+#define CB_PERFCOUNTER2_SELECT__PERF_MODE__SHIFT 0x1c
+#define CB_PERFCOUNTER3_SELECT__PERF_SEL_MASK 0x1ff
+#define CB_PERFCOUNTER3_SELECT__PERF_SEL__SHIFT 0x0
+#define CB_PERFCOUNTER3_SELECT__PERF_MODE_MASK 0xf0000000
+#define CB_PERFCOUNTER3_SELECT__PERF_MODE__SHIFT 0x1c
+#define CB_PERFCOUNTER0_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define CB_PERFCOUNTER0_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define CB_PERFCOUNTER1_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define CB_PERFCOUNTER1_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define CB_PERFCOUNTER2_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define CB_PERFCOUNTER2_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define CB_PERFCOUNTER3_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define CB_PERFCOUNTER3_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define CB_PERFCOUNTER0_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define CB_PERFCOUNTER0_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define CB_PERFCOUNTER1_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define CB_PERFCOUNTER1_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define CB_PERFCOUNTER2_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define CB_PERFCOUNTER2_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define CB_PERFCOUNTER3_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define CB_PERFCOUNTER3_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define CB_CGTT_SCLK_CTRL__ON_DELAY_MASK 0xf
+#define CB_CGTT_SCLK_CTRL__ON_DELAY__SHIFT 0x0
+#define CB_CGTT_SCLK_CTRL__OFF_HYSTERESIS_MASK 0xff0
+#define CB_CGTT_SCLK_CTRL__OFF_HYSTERESIS__SHIFT 0x4
+#define CB_CGTT_SCLK_CTRL__SOFT_OVERRIDE7_MASK 0x1000000
+#define CB_CGTT_SCLK_CTRL__SOFT_OVERRIDE7__SHIFT 0x18
+#define CB_CGTT_SCLK_CTRL__SOFT_OVERRIDE6_MASK 0x2000000
+#define CB_CGTT_SCLK_CTRL__SOFT_OVERRIDE6__SHIFT 0x19
+#define CB_CGTT_SCLK_CTRL__SOFT_OVERRIDE5_MASK 0x4000000
+#define CB_CGTT_SCLK_CTRL__SOFT_OVERRIDE5__SHIFT 0x1a
+#define CB_CGTT_SCLK_CTRL__SOFT_OVERRIDE4_MASK 0x8000000
+#define CB_CGTT_SCLK_CTRL__SOFT_OVERRIDE4__SHIFT 0x1b
+#define CB_CGTT_SCLK_CTRL__SOFT_OVERRIDE3_MASK 0x10000000
+#define CB_CGTT_SCLK_CTRL__SOFT_OVERRIDE3__SHIFT 0x1c
+#define CB_CGTT_SCLK_CTRL__SOFT_OVERRIDE2_MASK 0x20000000
+#define CB_CGTT_SCLK_CTRL__SOFT_OVERRIDE2__SHIFT 0x1d
+#define CB_CGTT_SCLK_CTRL__SOFT_OVERRIDE1_MASK 0x40000000
+#define CB_CGTT_SCLK_CTRL__SOFT_OVERRIDE1__SHIFT 0x1e
+#define CB_CGTT_SCLK_CTRL__SOFT_OVERRIDE0_MASK 0x80000000
+#define CB_CGTT_SCLK_CTRL__SOFT_OVERRIDE0__SHIFT 0x1f
+#define CB_DEBUG_BUS_1__CB_BUSY_MASK 0x1
+#define CB_DEBUG_BUS_1__CB_BUSY__SHIFT 0x0
+#define CB_DEBUG_BUS_1__DB_CB_TILE_VALID_READY_MASK 0x2
+#define CB_DEBUG_BUS_1__DB_CB_TILE_VALID_READY__SHIFT 0x1
+#define CB_DEBUG_BUS_1__DB_CB_TILE_VALID_READYB_MASK 0x4
+#define CB_DEBUG_BUS_1__DB_CB_TILE_VALID_READYB__SHIFT 0x2
+#define CB_DEBUG_BUS_1__DB_CB_TILE_VALIDB_READY_MASK 0x8
+#define CB_DEBUG_BUS_1__DB_CB_TILE_VALIDB_READY__SHIFT 0x3
+#define CB_DEBUG_BUS_1__DB_CB_TILE_VALIDB_READYB_MASK 0x10
+#define CB_DEBUG_BUS_1__DB_CB_TILE_VALIDB_READYB__SHIFT 0x4
+#define CB_DEBUG_BUS_1__DB_CB_LQUAD_VALID_READY_MASK 0x20
+#define CB_DEBUG_BUS_1__DB_CB_LQUAD_VALID_READY__SHIFT 0x5
+#define CB_DEBUG_BUS_1__DB_CB_LQUAD_VALID_READYB_MASK 0x40
+#define CB_DEBUG_BUS_1__DB_CB_LQUAD_VALID_READYB__SHIFT 0x6
+#define CB_DEBUG_BUS_1__DB_CB_LQUAD_VALIDB_READY_MASK 0x80
+#define CB_DEBUG_BUS_1__DB_CB_LQUAD_VALIDB_READY__SHIFT 0x7
+#define CB_DEBUG_BUS_1__DB_CB_LQUAD_VALIDB_READYB_MASK 0x100
+#define CB_DEBUG_BUS_1__DB_CB_LQUAD_VALIDB_READYB__SHIFT 0x8
+#define CB_DEBUG_BUS_1__CB_TAP_WRREQ_VALID_READY_MASK 0x200
+#define CB_DEBUG_BUS_1__CB_TAP_WRREQ_VALID_READY__SHIFT 0x9
+#define CB_DEBUG_BUS_1__CB_TAP_WRREQ_VALID_READYB_MASK 0x400
+#define CB_DEBUG_BUS_1__CB_TAP_WRREQ_VALID_READYB__SHIFT 0xa
+#define CB_DEBUG_BUS_1__CB_TAP_WRREQ_VALIDB_READY_MASK 0x800
+#define CB_DEBUG_BUS_1__CB_TAP_WRREQ_VALIDB_READY__SHIFT 0xb
+#define CB_DEBUG_BUS_1__CB_TAP_WRREQ_VALIDB_READYB_MASK 0x1000
+#define CB_DEBUG_BUS_1__CB_TAP_WRREQ_VALIDB_READYB__SHIFT 0xc
+#define CB_DEBUG_BUS_1__CB_TAP_RDREQ_VALID_READY_MASK 0x2000
+#define CB_DEBUG_BUS_1__CB_TAP_RDREQ_VALID_READY__SHIFT 0xd
+#define CB_DEBUG_BUS_1__CB_TAP_RDREQ_VALID_READYB_MASK 0x4000
+#define CB_DEBUG_BUS_1__CB_TAP_RDREQ_VALID_READYB__SHIFT 0xe
+#define CB_DEBUG_BUS_1__CB_TAP_RDREQ_VALIDB_READY_MASK 0x8000
+#define CB_DEBUG_BUS_1__CB_TAP_RDREQ_VALIDB_READY__SHIFT 0xf
+#define CB_DEBUG_BUS_1__CB_TAP_RDREQ_VALIDB_READYB_MASK 0x10000
+#define CB_DEBUG_BUS_1__CB_TAP_RDREQ_VALIDB_READYB__SHIFT 0x10
+#define CB_DEBUG_BUS_1__CM_FC_TILE_VALID_READY_MASK 0x20000
+#define CB_DEBUG_BUS_1__CM_FC_TILE_VALID_READY__SHIFT 0x11
+#define CB_DEBUG_BUS_1__CM_FC_TILE_VALID_READYB_MASK 0x40000
+#define CB_DEBUG_BUS_1__CM_FC_TILE_VALID_READYB__SHIFT 0x12
+#define CB_DEBUG_BUS_1__CM_FC_TILE_VALIDB_READY_MASK 0x80000
+#define CB_DEBUG_BUS_1__CM_FC_TILE_VALIDB_READY__SHIFT 0x13
+#define CB_DEBUG_BUS_1__CM_FC_TILE_VALIDB_READYB_MASK 0x100000
+#define CB_DEBUG_BUS_1__CM_FC_TILE_VALIDB_READYB__SHIFT 0x14
+#define CB_DEBUG_BUS_1__FC_CLEAR_QUAD_VALID_READY_MASK 0x200000
+#define CB_DEBUG_BUS_1__FC_CLEAR_QUAD_VALID_READY__SHIFT 0x15
+#define CB_DEBUG_BUS_1__FC_CLEAR_QUAD_VALID_READYB_MASK 0x400000
+#define CB_DEBUG_BUS_1__FC_CLEAR_QUAD_VALID_READYB__SHIFT 0x16
+#define CB_DEBUG_BUS_1__FC_CLEAR_QUAD_VALIDB_READY_MASK 0x800000
+#define CB_DEBUG_BUS_1__FC_CLEAR_QUAD_VALIDB_READY__SHIFT 0x17
+#define CB_DEBUG_BUS_2__FC_CLEAR_QUAD_VALIDB_READYB_MASK 0x1
+#define CB_DEBUG_BUS_2__FC_CLEAR_QUAD_VALIDB_READYB__SHIFT 0x0
+#define CB_DEBUG_BUS_2__FC_QUAD_RESIDENCY_STALL_MASK 0x2
+#define CB_DEBUG_BUS_2__FC_QUAD_RESIDENCY_STALL__SHIFT 0x1
+#define CB_DEBUG_BUS_2__FC_CC_QUADFRAG_VALID_READY_MASK 0x4
+#define CB_DEBUG_BUS_2__FC_CC_QUADFRAG_VALID_READY__SHIFT 0x2
+#define CB_DEBUG_BUS_2__FC_CC_QUADFRAG_VALID_READYB_MASK 0x8
+#define CB_DEBUG_BUS_2__FC_CC_QUADFRAG_VALID_READYB__SHIFT 0x3
+#define CB_DEBUG_BUS_2__FC_CC_QUADFRAG_VALIDB_READY_MASK 0x10
+#define CB_DEBUG_BUS_2__FC_CC_QUADFRAG_VALIDB_READY__SHIFT 0x4
+#define CB_DEBUG_BUS_2__FC_CC_QUADFRAG_VALIDB_READYB_MASK 0x20
+#define CB_DEBUG_BUS_2__FC_CC_QUADFRAG_VALIDB_READYB__SHIFT 0x5
+#define CB_DEBUG_BUS_2__FOP_IN_VALID_READY_MASK 0x40
+#define CB_DEBUG_BUS_2__FOP_IN_VALID_READY__SHIFT 0x6
+#define CB_DEBUG_BUS_2__FOP_IN_VALID_READYB_MASK 0x80
+#define CB_DEBUG_BUS_2__FOP_IN_VALID_READYB__SHIFT 0x7
+#define CB_DEBUG_BUS_2__FOP_IN_VALIDB_READY_MASK 0x100
+#define CB_DEBUG_BUS_2__FOP_IN_VALIDB_READY__SHIFT 0x8
+#define CB_DEBUG_BUS_2__FOP_IN_VALIDB_READYB_MASK 0x200
+#define CB_DEBUG_BUS_2__FOP_IN_VALIDB_READYB__SHIFT 0x9
+#define CB_DEBUG_BUS_2__FOP_FMASK_RAW_STALL_MASK 0x400
+#define CB_DEBUG_BUS_2__FOP_FMASK_RAW_STALL__SHIFT 0xa
+#define CB_DEBUG_BUS_2__FOP_FMASK_BYPASS_STALL_MASK 0x800
+#define CB_DEBUG_BUS_2__FOP_FMASK_BYPASS_STALL__SHIFT 0xb
+#define CB_DEBUG_BUS_2__CC_IB_TB_FRAG_VALID_READY_MASK 0x1000
+#define CB_DEBUG_BUS_2__CC_IB_TB_FRAG_VALID_READY__SHIFT 0xc
+#define CB_DEBUG_BUS_2__CC_IB_TB_FRAG_VALID_READYB_MASK 0x2000
+#define CB_DEBUG_BUS_2__CC_IB_TB_FRAG_VALID_READYB__SHIFT 0xd
+#define CB_DEBUG_BUS_2__CC_IB_TB_FRAG_VALIDB_READY_MASK 0x4000
+#define CB_DEBUG_BUS_2__CC_IB_TB_FRAG_VALIDB_READY__SHIFT 0xe
+#define CB_DEBUG_BUS_2__CC_IB_TB_FRAG_VALIDB_READYB_MASK 0x8000
+#define CB_DEBUG_BUS_2__CC_IB_TB_FRAG_VALIDB_READYB__SHIFT 0xf
+#define CB_DEBUG_BUS_2__CC_IB_SR_FRAG_VALID_READY_MASK 0x10000
+#define CB_DEBUG_BUS_2__CC_IB_SR_FRAG_VALID_READY__SHIFT 0x10
+#define CB_DEBUG_BUS_2__CC_IB_SR_FRAG_VALID_READYB_MASK 0x20000
+#define CB_DEBUG_BUS_2__CC_IB_SR_FRAG_VALID_READYB__SHIFT 0x11
+#define CB_DEBUG_BUS_2__CC_IB_SR_FRAG_VALIDB_READY_MASK 0x40000
+#define CB_DEBUG_BUS_2__CC_IB_SR_FRAG_VALIDB_READY__SHIFT 0x12
+#define CB_DEBUG_BUS_2__CC_IB_SR_FRAG_VALIDB_READYB_MASK 0x80000
+#define CB_DEBUG_BUS_2__CC_IB_SR_FRAG_VALIDB_READYB__SHIFT 0x13
+#define CB_DEBUG_BUS_2__CC_RB_BC_EVENFRAG_VALID_READY_MASK 0x100000
+#define CB_DEBUG_BUS_2__CC_RB_BC_EVENFRAG_VALID_READY__SHIFT 0x14
+#define CB_DEBUG_BUS_2__CC_RB_BC_EVENFRAG_VALID_READYB_MASK 0x200000
+#define CB_DEBUG_BUS_2__CC_RB_BC_EVENFRAG_VALID_READYB__SHIFT 0x15
+#define CB_DEBUG_BUS_2__CC_RB_BC_EVENFRAG_VALIDB_READY_MASK 0x400000
+#define CB_DEBUG_BUS_2__CC_RB_BC_EVENFRAG_VALIDB_READY__SHIFT 0x16
+#define CB_DEBUG_BUS_2__CC_RB_BC_EVENFRAG_VALIDB_READYB_MASK 0x800000
+#define CB_DEBUG_BUS_2__CC_RB_BC_EVENFRAG_VALIDB_READYB__SHIFT 0x17
+#define CB_DEBUG_BUS_3__CC_RB_BC_ODDFRAG_VALID_READY_MASK 0x1
+#define CB_DEBUG_BUS_3__CC_RB_BC_ODDFRAG_VALID_READY__SHIFT 0x0
+#define CB_DEBUG_BUS_3__CC_RB_BC_ODDFRAG_VALID_READYB_MASK 0x2
+#define CB_DEBUG_BUS_3__CC_RB_BC_ODDFRAG_VALID_READYB__SHIFT 0x1
+#define CB_DEBUG_BUS_3__CC_RB_BC_ODDFRAG_VALIDB_READY_MASK 0x4
+#define CB_DEBUG_BUS_3__CC_RB_BC_ODDFRAG_VALIDB_READY__SHIFT 0x2
+#define CB_DEBUG_BUS_3__CC_RB_BC_ODDFRAG_VALIDB_READYB_MASK 0x8
+#define CB_DEBUG_BUS_3__CC_RB_BC_ODDFRAG_VALIDB_READYB__SHIFT 0x3
+#define CB_DEBUG_BUS_3__CC_BC_CS_FRAG_VALID_MASK 0x10
+#define CB_DEBUG_BUS_3__CC_BC_CS_FRAG_VALID__SHIFT 0x4
+#define CB_DEBUG_BUS_3__CC_SF_FULL_MASK 0x20
+#define CB_DEBUG_BUS_3__CC_SF_FULL__SHIFT 0x5
+#define CB_DEBUG_BUS_3__CC_RB_FULL_MASK 0x40
+#define CB_DEBUG_BUS_3__CC_RB_FULL__SHIFT 0x6
+#define CB_DEBUG_BUS_3__CC_EVENFIFO_QUAD_RESIDENCY_STALL_MASK 0x80
+#define CB_DEBUG_BUS_3__CC_EVENFIFO_QUAD_RESIDENCY_STALL__SHIFT 0x7
+#define CB_DEBUG_BUS_3__CC_ODDFIFO_QUAD_RESIDENCY_STALL_MASK 0x100
+#define CB_DEBUG_BUS_3__CC_ODDFIFO_QUAD_RESIDENCY_STALL__SHIFT 0x8
+#define CB_DEBUG_BUS_3__CM_TQ_FULL_MASK 0x200
+#define CB_DEBUG_BUS_3__CM_TQ_FULL__SHIFT 0x9
+#define CB_DEBUG_BUS_3__CM_TILE_RESIDENCY_STALL_MASK 0x400
+#define CB_DEBUG_BUS_3__CM_TILE_RESIDENCY_STALL__SHIFT 0xa
+#define CB_DEBUG_BUS_3__LQUAD_NO_TILE_MASK 0x800
+#define CB_DEBUG_BUS_3__LQUAD_NO_TILE__SHIFT 0xb
+#define CB_DEBUG_BUS_3__LQUAD_FORMAT_IS_EXPORT_32_R_MASK 0x1000
+#define CB_DEBUG_BUS_3__LQUAD_FORMAT_IS_EXPORT_32_R__SHIFT 0xc
+#define CB_DEBUG_BUS_3__LQUAD_FORMAT_IS_EXPORT_32_AR_MASK 0x2000
+#define CB_DEBUG_BUS_3__LQUAD_FORMAT_IS_EXPORT_32_AR__SHIFT 0xd
+#define CB_DEBUG_BUS_3__LQUAD_FORMAT_IS_EXPORT_32_GR_MASK 0x4000
+#define CB_DEBUG_BUS_3__LQUAD_FORMAT_IS_EXPORT_32_GR__SHIFT 0xe
+#define CB_DEBUG_BUS_3__LQUAD_FORMAT_IS_EXPORT_32_ABGR_MASK 0x8000
+#define CB_DEBUG_BUS_3__LQUAD_FORMAT_IS_EXPORT_32_ABGR__SHIFT 0xf
+#define CB_DEBUG_BUS_3__LQUAD_FORMAT_IS_EXPORT_FP16_ABGR_MASK 0x10000
+#define CB_DEBUG_BUS_3__LQUAD_FORMAT_IS_EXPORT_FP16_ABGR__SHIFT 0x10
+#define CB_DEBUG_BUS_3__LQUAD_FORMAT_IS_EXPORT_SIGNED16_ABGR_MASK 0x20000
+#define CB_DEBUG_BUS_3__LQUAD_FORMAT_IS_EXPORT_SIGNED16_ABGR__SHIFT 0x11
+#define CB_DEBUG_BUS_3__LQUAD_FORMAT_IS_EXPORT_UNSIGNED16_ABGR_MASK 0x40000
+#define CB_DEBUG_BUS_3__LQUAD_FORMAT_IS_EXPORT_UNSIGNED16_ABGR__SHIFT 0x12
+#define CB_DEBUG_BUS_3__CM_CACHE_HIT_MASK 0x80000
+#define CB_DEBUG_BUS_3__CM_CACHE_HIT__SHIFT 0x13
+#define CB_DEBUG_BUS_3__CM_CACHE_TAG_MISS_MASK 0x100000
+#define CB_DEBUG_BUS_3__CM_CACHE_TAG_MISS__SHIFT 0x14
+#define CB_DEBUG_BUS_3__CM_CACHE_SECTOR_MISS_MASK 0x200000
+#define CB_DEBUG_BUS_3__CM_CACHE_SECTOR_MISS__SHIFT 0x15
+#define CB_DEBUG_BUS_3__CM_CACHE_REEVICTION_STALL_MASK 0x400000
+#define CB_DEBUG_BUS_3__CM_CACHE_REEVICTION_STALL__SHIFT 0x16
+#define CB_DEBUG_BUS_3__CM_CACHE_EVICT_NONZERO_INFLIGHT_STALL_MASK 0x800000
+#define CB_DEBUG_BUS_3__CM_CACHE_EVICT_NONZERO_INFLIGHT_STALL__SHIFT 0x17
+#define CB_DEBUG_BUS_4__CM_CACHE_REPLACE_PENDING_EVICT_STALL_MASK 0x1
+#define CB_DEBUG_BUS_4__CM_CACHE_REPLACE_PENDING_EVICT_STALL__SHIFT 0x0
+#define CB_DEBUG_BUS_4__CM_CACHE_INFLIGHT_COUNTER_MAXIMUM_STALL_MASK 0x2
+#define CB_DEBUG_BUS_4__CM_CACHE_INFLIGHT_COUNTER_MAXIMUM_STALL__SHIFT 0x1
+#define CB_DEBUG_BUS_4__CM_CACHE_READ_OUTPUT_STALL_MASK 0x4
+#define CB_DEBUG_BUS_4__CM_CACHE_READ_OUTPUT_STALL__SHIFT 0x2
+#define CB_DEBUG_BUS_4__CM_CACHE_WRITE_OUTPUT_STALL_MASK 0x8
+#define CB_DEBUG_BUS_4__CM_CACHE_WRITE_OUTPUT_STALL__SHIFT 0x3
+#define CB_DEBUG_BUS_4__CM_CACHE_ACK_OUTPUT_STALL_MASK 0x10
+#define CB_DEBUG_BUS_4__CM_CACHE_ACK_OUTPUT_STALL__SHIFT 0x4
+#define CB_DEBUG_BUS_4__CM_CACHE_STALL_MASK 0x20
+#define CB_DEBUG_BUS_4__CM_CACHE_STALL__SHIFT 0x5
+#define CB_DEBUG_BUS_4__FC_CACHE_HIT_MASK 0x40
+#define CB_DEBUG_BUS_4__FC_CACHE_HIT__SHIFT 0x6
+#define CB_DEBUG_BUS_4__FC_CACHE_TAG_MISS_MASK 0x80
+#define CB_DEBUG_BUS_4__FC_CACHE_TAG_MISS__SHIFT 0x7
+#define CB_DEBUG_BUS_4__FC_CACHE_SECTOR_MISS_MASK 0x100
+#define CB_DEBUG_BUS_4__FC_CACHE_SECTOR_MISS__SHIFT 0x8
+#define CB_DEBUG_BUS_4__FC_CACHE_REEVICTION_STALL_MASK 0x200
+#define CB_DEBUG_BUS_4__FC_CACHE_REEVICTION_STALL__SHIFT 0x9
+#define CB_DEBUG_BUS_4__FC_CACHE_EVICT_NONZERO_INFLIGHT_STALL_MASK 0x400
+#define CB_DEBUG_BUS_4__FC_CACHE_EVICT_NONZERO_INFLIGHT_STALL__SHIFT 0xa
+#define CB_DEBUG_BUS_4__FC_CACHE_REPLACE_PENDING_EVICT_STALL_MASK 0x800
+#define CB_DEBUG_BUS_4__FC_CACHE_REPLACE_PENDING_EVICT_STALL__SHIFT 0xb
+#define CB_DEBUG_BUS_4__FC_CACHE_INFLIGHT_COUNTER_MAXIMUM_STALL_MASK 0x1000
+#define CB_DEBUG_BUS_4__FC_CACHE_INFLIGHT_COUNTER_MAXIMUM_STALL__SHIFT 0xc
+#define CB_DEBUG_BUS_4__FC_CACHE_READ_OUTPUT_STALL_MASK 0x2000
+#define CB_DEBUG_BUS_4__FC_CACHE_READ_OUTPUT_STALL__SHIFT 0xd
+#define CB_DEBUG_BUS_4__FC_CACHE_WRITE_OUTPUT_STALL_MASK 0x4000
+#define CB_DEBUG_BUS_4__FC_CACHE_WRITE_OUTPUT_STALL__SHIFT 0xe
+#define CB_DEBUG_BUS_4__FC_CACHE_ACK_OUTPUT_STALL_MASK 0x8000
+#define CB_DEBUG_BUS_4__FC_CACHE_ACK_OUTPUT_STALL__SHIFT 0xf
+#define CB_DEBUG_BUS_4__FC_CACHE_STALL_MASK 0x10000
+#define CB_DEBUG_BUS_4__FC_CACHE_STALL__SHIFT 0x10
+#define CB_DEBUG_BUS_4__CC_CACHE_HIT_MASK 0x20000
+#define CB_DEBUG_BUS_4__CC_CACHE_HIT__SHIFT 0x11
+#define CB_DEBUG_BUS_4__CC_CACHE_TAG_MISS_MASK 0x40000
+#define CB_DEBUG_BUS_4__CC_CACHE_TAG_MISS__SHIFT 0x12
+#define CB_DEBUG_BUS_4__CC_CACHE_SECTOR_MISS_MASK 0x80000
+#define CB_DEBUG_BUS_4__CC_CACHE_SECTOR_MISS__SHIFT 0x13
+#define CB_DEBUG_BUS_4__CC_CACHE_REEVICTION_STALL_MASK 0x100000
+#define CB_DEBUG_BUS_4__CC_CACHE_REEVICTION_STALL__SHIFT 0x14
+#define CB_DEBUG_BUS_4__CC_CACHE_EVICT_NONZERO_INFLIGHT_STALL_MASK 0x200000
+#define CB_DEBUG_BUS_4__CC_CACHE_EVICT_NONZERO_INFLIGHT_STALL__SHIFT 0x15
+#define CB_DEBUG_BUS_4__CC_CACHE_REPLACE_PENDING_EVICT_STALL_MASK 0x400000
+#define CB_DEBUG_BUS_4__CC_CACHE_REPLACE_PENDING_EVICT_STALL__SHIFT 0x16
+#define CB_DEBUG_BUS_4__CC_CACHE_INFLIGHT_COUNTER_MAXIMUM_STALL_MASK 0x800000
+#define CB_DEBUG_BUS_4__CC_CACHE_INFLIGHT_COUNTER_MAXIMUM_STALL__SHIFT 0x17
+#define CB_DEBUG_BUS_5__CC_CACHE_READ_OUTPUT_STALL_MASK 0x1
+#define CB_DEBUG_BUS_5__CC_CACHE_READ_OUTPUT_STALL__SHIFT 0x0
+#define CB_DEBUG_BUS_5__CC_CACHE_WRITE_OUTPUT_STALL_MASK 0x2
+#define CB_DEBUG_BUS_5__CC_CACHE_WRITE_OUTPUT_STALL__SHIFT 0x1
+#define CB_DEBUG_BUS_5__CC_CACHE_ACK_OUTPUT_STALL_MASK 0x4
+#define CB_DEBUG_BUS_5__CC_CACHE_ACK_OUTPUT_STALL__SHIFT 0x2
+#define CB_DEBUG_BUS_5__CC_CACHE_STALL_MASK 0x8
+#define CB_DEBUG_BUS_5__CC_CACHE_STALL__SHIFT 0x3
+#define CB_DEBUG_BUS_5__CC_CACHE_WA_TO_RMW_CONVERSION_MASK 0x10
+#define CB_DEBUG_BUS_5__CC_CACHE_WA_TO_RMW_CONVERSION__SHIFT 0x4
+#define CB_DEBUG_BUS_5__CM_CACHE_FLUSH_MASK 0x20
+#define CB_DEBUG_BUS_5__CM_CACHE_FLUSH__SHIFT 0x5
+#define CB_DEBUG_BUS_5__CM_CACHE_TAGS_FLUSHED_MASK 0x40
+#define CB_DEBUG_BUS_5__CM_CACHE_TAGS_FLUSHED__SHIFT 0x6
+#define CB_DEBUG_BUS_5__CM_CACHE_SECTORS_FLUSHED_MASK 0x80
+#define CB_DEBUG_BUS_5__CM_CACHE_SECTORS_FLUSHED__SHIFT 0x7
+#define CB_DEBUG_BUS_5__CM_CACHE_DIRTY_SECTORS_FLUSHED_MASK 0x100
+#define CB_DEBUG_BUS_5__CM_CACHE_DIRTY_SECTORS_FLUSHED__SHIFT 0x8
+#define CB_DEBUG_BUS_5__FC_CACHE_FLUSH_MASK 0x200
+#define CB_DEBUG_BUS_5__FC_CACHE_FLUSH__SHIFT 0x9
+#define CB_DEBUG_BUS_5__FC_CACHE_TAGS_FLUSHED_MASK 0x400
+#define CB_DEBUG_BUS_5__FC_CACHE_TAGS_FLUSHED__SHIFT 0xa
+#define CB_DEBUG_BUS_5__FC_CACHE_SECTORS_FLUSHED_MASK 0x3800
+#define CB_DEBUG_BUS_5__FC_CACHE_SECTORS_FLUSHED__SHIFT 0xb
+#define CB_DEBUG_BUS_5__FC_CACHE_DIRTY_SECTORS_FLUSHED_MASK 0x1c000
+#define CB_DEBUG_BUS_5__FC_CACHE_DIRTY_SECTORS_FLUSHED__SHIFT 0xe
+#define CB_DEBUG_BUS_5__CC_CACHE_FLUSH_MASK 0x20000
+#define CB_DEBUG_BUS_5__CC_CACHE_FLUSH__SHIFT 0x11
+#define CB_DEBUG_BUS_5__CC_CACHE_TAGS_FLUSHED_MASK 0x40000
+#define CB_DEBUG_BUS_5__CC_CACHE_TAGS_FLUSHED__SHIFT 0x12
+#define CB_DEBUG_BUS_5__CC_CACHE_SECTORS_FLUSHED_MASK 0x380000
+#define CB_DEBUG_BUS_5__CC_CACHE_SECTORS_FLUSHED__SHIFT 0x13
+#define CB_DEBUG_BUS_6__CC_CACHE_DIRTY_SECTORS_FLUSHED_MASK 0x7
+#define CB_DEBUG_BUS_6__CC_CACHE_DIRTY_SECTORS_FLUSHED__SHIFT 0x0
+#define CB_DEBUG_BUS_6__CM_MC_READ_REQUEST_MASK 0x8
+#define CB_DEBUG_BUS_6__CM_MC_READ_REQUEST__SHIFT 0x3
+#define CB_DEBUG_BUS_6__FC_MC_READ_REQUEST_MASK 0x10
+#define CB_DEBUG_BUS_6__FC_MC_READ_REQUEST__SHIFT 0x4
+#define CB_DEBUG_BUS_6__CC_MC_READ_REQUEST_MASK 0x20
+#define CB_DEBUG_BUS_6__CC_MC_READ_REQUEST__SHIFT 0x5
+#define CB_DEBUG_BUS_6__CM_MC_WRITE_REQUEST_MASK 0x40
+#define CB_DEBUG_BUS_6__CM_MC_WRITE_REQUEST__SHIFT 0x6
+#define CB_DEBUG_BUS_6__FC_MC_WRITE_REQUEST_MASK 0x80
+#define CB_DEBUG_BUS_6__FC_MC_WRITE_REQUEST__SHIFT 0x7
+#define CB_DEBUG_BUS_6__CC_MC_WRITE_REQUEST_MASK 0x100
+#define CB_DEBUG_BUS_6__CC_MC_WRITE_REQUEST__SHIFT 0x8
+#define CB_DEBUG_BUS_6__CM_MC_READ_REQUESTS_IN_FLIGHT_MASK 0x1fe00
+#define CB_DEBUG_BUS_6__CM_MC_READ_REQUESTS_IN_FLIGHT__SHIFT 0x9
+#define CB_DEBUG_BUS_7__FC_MC_READ_REQUESTS_IN_FLIGHT_MASK 0x7ff
+#define CB_DEBUG_BUS_7__FC_MC_READ_REQUESTS_IN_FLIGHT__SHIFT 0x0
+#define CB_DEBUG_BUS_7__CC_MC_READ_REQUESTS_IN_FLIGHT_MASK 0x1ff800
+#define CB_DEBUG_BUS_7__CC_MC_READ_REQUESTS_IN_FLIGHT__SHIFT 0xb
+#define CB_DEBUG_BUS_8__CM_MC_WRITE_REQUESTS_IN_FLIGHT_MASK 0xff
+#define CB_DEBUG_BUS_8__CM_MC_WRITE_REQUESTS_IN_FLIGHT__SHIFT 0x0
+#define CB_DEBUG_BUS_8__FC_MC_WRITE_REQUESTS_IN_FLIGHT_MASK 0x7ff00
+#define CB_DEBUG_BUS_8__FC_MC_WRITE_REQUESTS_IN_FLIGHT__SHIFT 0x8
+#define CB_DEBUG_BUS_8__FC_SEQUENCER_FMASK_COMPRESSION_DISABLE_MASK 0x80000
+#define CB_DEBUG_BUS_8__FC_SEQUENCER_FMASK_COMPRESSION_DISABLE__SHIFT 0x13
+#define CB_DEBUG_BUS_8__FC_SEQUENCER_FMASK_DECOMPRESS_MASK 0x100000
+#define CB_DEBUG_BUS_8__FC_SEQUENCER_FMASK_DECOMPRESS__SHIFT 0x14
+#define CB_DEBUG_BUS_8__FC_SEQUENCER_ELIMINATE_FAST_CLEAR_MASK 0x200000
+#define CB_DEBUG_BUS_8__FC_SEQUENCER_ELIMINATE_FAST_CLEAR__SHIFT 0x15
+#define CB_DEBUG_BUS_8__FC_SEQUENCER_CLEAR_MASK 0x400000
+#define CB_DEBUG_BUS_8__FC_SEQUENCER_CLEAR__SHIFT 0x16
+#define CB_DEBUG_BUS_9__CC_MC_WRITE_REQUESTS_IN_FLIGHT_MASK 0x3ff
+#define CB_DEBUG_BUS_9__CC_MC_WRITE_REQUESTS_IN_FLIGHT__SHIFT 0x0
+#define CB_DEBUG_BUS_9__CC_SURFACE_SYNC_MASK 0x400
+#define CB_DEBUG_BUS_9__CC_SURFACE_SYNC__SHIFT 0xa
+#define CB_DEBUG_BUS_9__TWO_PROBE_QUAD_FRAGMENT_MASK 0x800
+#define CB_DEBUG_BUS_9__TWO_PROBE_QUAD_FRAGMENT__SHIFT 0xb
+#define CB_DEBUG_BUS_9__EXPORT_32_ABGR_QUAD_FRAGMENT_MASK 0x1000
+#define CB_DEBUG_BUS_9__EXPORT_32_ABGR_QUAD_FRAGMENT__SHIFT 0xc
+#define CB_DEBUG_BUS_9__DUAL_SOURCE_COLOR_QUAD_FRAGMENT_MASK 0x2000
+#define CB_DEBUG_BUS_9__DUAL_SOURCE_COLOR_QUAD_FRAGMENT__SHIFT 0xd
+#define CB_DEBUG_BUS_9__DEBUG_BUS_DRAWN_QUAD_MASK 0x4000
+#define CB_DEBUG_BUS_9__DEBUG_BUS_DRAWN_QUAD__SHIFT 0xe
+#define CB_DEBUG_BUS_9__DEBUG_BUS_DRAWN_PIXEL_MASK 0x78000
+#define CB_DEBUG_BUS_9__DEBUG_BUS_DRAWN_PIXEL__SHIFT 0xf
+#define CB_DEBUG_BUS_9__DEBUG_BUS_DRAWN_QUAD_FRAGMENT_MASK 0x80000
+#define CB_DEBUG_BUS_9__DEBUG_BUS_DRAWN_QUAD_FRAGMENT__SHIFT 0x13
+#define CB_DEBUG_BUS_9__DEBUG_BUS_DRAWN_TILE_MASK 0x100000
+#define CB_DEBUG_BUS_9__DEBUG_BUS_DRAWN_TILE__SHIFT 0x14
+#define CB_DEBUG_BUS_9__EVENT_ALL_MASK 0x200000
+#define CB_DEBUG_BUS_9__EVENT_ALL__SHIFT 0x15
+#define CB_DEBUG_BUS_9__EVENT_CACHE_FLUSH_TS_MASK 0x400000
+#define CB_DEBUG_BUS_9__EVENT_CACHE_FLUSH_TS__SHIFT 0x16
+#define CB_DEBUG_BUS_9__EVENT_CONTEXT_DONE_MASK 0x800000
+#define CB_DEBUG_BUS_9__EVENT_CONTEXT_DONE__SHIFT 0x17
+#define CB_DEBUG_BUS_10__EVENT_CACHE_FLUSH_MASK 0x1
+#define CB_DEBUG_BUS_10__EVENT_CACHE_FLUSH__SHIFT 0x0
+#define CB_DEBUG_BUS_10__EVENT_CACHE_FLUSH_AND_INV_TS_EVENT_MASK 0x2
+#define CB_DEBUG_BUS_10__EVENT_CACHE_FLUSH_AND_INV_TS_EVENT__SHIFT 0x1
+#define CB_DEBUG_BUS_10__EVENT_CACHE_FLUSH_AND_INV_EVENT_MASK 0x4
+#define CB_DEBUG_BUS_10__EVENT_CACHE_FLUSH_AND_INV_EVENT__SHIFT 0x2
+#define CB_DEBUG_BUS_10__EVENT_FLUSH_AND_INV_CB_DATA_TS_MASK 0x8
+#define CB_DEBUG_BUS_10__EVENT_FLUSH_AND_INV_CB_DATA_TS__SHIFT 0x3
+#define CB_DEBUG_BUS_10__EVENT_FLUSH_AND_INV_CB_META_MASK 0x10
+#define CB_DEBUG_BUS_10__EVENT_FLUSH_AND_INV_CB_META__SHIFT 0x4
+#define CB_DEBUG_BUS_10__CMASK_READ_DATA_0XC_MASK 0x20
+#define CB_DEBUG_BUS_10__CMASK_READ_DATA_0XC__SHIFT 0x5
+#define CB_DEBUG_BUS_10__CMASK_READ_DATA_0XD_MASK 0x40
+#define CB_DEBUG_BUS_10__CMASK_READ_DATA_0XD__SHIFT 0x6
+#define CB_DEBUG_BUS_10__CMASK_READ_DATA_0XE_MASK 0x80
+#define CB_DEBUG_BUS_10__CMASK_READ_DATA_0XE__SHIFT 0x7
+#define CB_DEBUG_BUS_10__CMASK_READ_DATA_0XF_MASK 0x100
+#define CB_DEBUG_BUS_10__CMASK_READ_DATA_0XF__SHIFT 0x8
+#define CB_DEBUG_BUS_10__CMASK_WRITE_DATA_0XC_MASK 0x200
+#define CB_DEBUG_BUS_10__CMASK_WRITE_DATA_0XC__SHIFT 0x9
+#define CB_DEBUG_BUS_10__CMASK_WRITE_DATA_0XD_MASK 0x400
+#define CB_DEBUG_BUS_10__CMASK_WRITE_DATA_0XD__SHIFT 0xa
+#define CB_DEBUG_BUS_10__CMASK_WRITE_DATA_0XE_MASK 0x800
+#define CB_DEBUG_BUS_10__CMASK_WRITE_DATA_0XE__SHIFT 0xb
+#define CB_DEBUG_BUS_10__CMASK_WRITE_DATA_0XF_MASK 0x1000
+#define CB_DEBUG_BUS_10__CMASK_WRITE_DATA_0XF__SHIFT 0xc
+#define CB_DEBUG_BUS_10__CORE_SCLK_VLD_MASK 0x2000
+#define CB_DEBUG_BUS_10__CORE_SCLK_VLD__SHIFT 0xd
+#define CB_DEBUG_BUS_10__REG_SCLK0_VLD_MASK 0x4000
+#define CB_DEBUG_BUS_10__REG_SCLK0_VLD__SHIFT 0xe
+#define CB_DEBUG_BUS_10__REG_SCLK1_VLD_MASK 0x8000
+#define CB_DEBUG_BUS_10__REG_SCLK1_VLD__SHIFT 0xf
+#define CB_DEBUG_BUS_10__MERGE_TILE_ONLY_VALID_READY_MASK 0x10000
+#define CB_DEBUG_BUS_10__MERGE_TILE_ONLY_VALID_READY__SHIFT 0x10
+#define CB_DEBUG_BUS_10__MERGE_TILE_ONLY_VALID_READYB_MASK 0x20000
+#define CB_DEBUG_BUS_10__MERGE_TILE_ONLY_VALID_READYB__SHIFT 0x11
+#define CB_DEBUG_BUS_10__FC_QUAD_RDLAT_FIFO_FULL_MASK 0x40000
+#define CB_DEBUG_BUS_10__FC_QUAD_RDLAT_FIFO_FULL__SHIFT 0x12
+#define CB_DEBUG_BUS_10__FC_TILE_RDLAT_FIFO_FULL_MASK 0x80000
+#define CB_DEBUG_BUS_10__FC_TILE_RDLAT_FIFO_FULL__SHIFT 0x13
+#define CB_DEBUG_BUS_10__FOP_QUAD_HAS_1_FRAGMENT_BEFORE_UPDATE_MASK 0x100000
+#define CB_DEBUG_BUS_10__FOP_QUAD_HAS_1_FRAGMENT_BEFORE_UPDATE__SHIFT 0x14
+#define CB_DEBUG_BUS_10__FOP_QUAD_HAS_2_FRAGMENTS_BEFORE_UPDATE_MASK 0x200000
+#define CB_DEBUG_BUS_10__FOP_QUAD_HAS_2_FRAGMENTS_BEFORE_UPDATE__SHIFT 0x15
+#define CB_DEBUG_BUS_10__FOP_QUAD_HAS_3_FRAGMENTS_BEFORE_UPDATE_MASK 0x400000
+#define CB_DEBUG_BUS_10__FOP_QUAD_HAS_3_FRAGMENTS_BEFORE_UPDATE__SHIFT 0x16
+#define CB_DEBUG_BUS_10__FOP_QUAD_HAS_4_FRAGMENTS_BEFORE_UPDATE_MASK 0x800000
+#define CB_DEBUG_BUS_10__FOP_QUAD_HAS_4_FRAGMENTS_BEFORE_UPDATE__SHIFT 0x17
+#define CB_DEBUG_BUS_11__FOP_QUAD_HAS_5_FRAGMENTS_BEFORE_UPDATE_MASK 0x1
+#define CB_DEBUG_BUS_11__FOP_QUAD_HAS_5_FRAGMENTS_BEFORE_UPDATE__SHIFT 0x0
+#define CB_DEBUG_BUS_11__FOP_QUAD_HAS_6_FRAGMENTS_BEFORE_UPDATE_MASK 0x2
+#define CB_DEBUG_BUS_11__FOP_QUAD_HAS_6_FRAGMENTS_BEFORE_UPDATE__SHIFT 0x1
+#define CB_DEBUG_BUS_11__FOP_QUAD_HAS_7_FRAGMENTS_BEFORE_UPDATE_MASK 0x4
+#define CB_DEBUG_BUS_11__FOP_QUAD_HAS_7_FRAGMENTS_BEFORE_UPDATE__SHIFT 0x2
+#define CB_DEBUG_BUS_11__FOP_QUAD_HAS_8_FRAGMENTS_BEFORE_UPDATE_MASK 0x8
+#define CB_DEBUG_BUS_11__FOP_QUAD_HAS_8_FRAGMENTS_BEFORE_UPDATE__SHIFT 0x3
+#define CB_DEBUG_BUS_11__FOP_QUAD_HAS_1_FRAGMENT_AFTER_UPDATE_MASK 0x10
+#define CB_DEBUG_BUS_11__FOP_QUAD_HAS_1_FRAGMENT_AFTER_UPDATE__SHIFT 0x4
+#define CB_DEBUG_BUS_11__FOP_QUAD_HAS_2_FRAGMENTS_AFTER_UPDATE_MASK 0x20
+#define CB_DEBUG_BUS_11__FOP_QUAD_HAS_2_FRAGMENTS_AFTER_UPDATE__SHIFT 0x5
+#define CB_DEBUG_BUS_11__FOP_QUAD_HAS_3_FRAGMENTS_AFTER_UPDATE_MASK 0x40
+#define CB_DEBUG_BUS_11__FOP_QUAD_HAS_3_FRAGMENTS_AFTER_UPDATE__SHIFT 0x6
+#define CB_DEBUG_BUS_11__FOP_QUAD_HAS_4_FRAGMENTS_AFTER_UPDATE_MASK 0x80
+#define CB_DEBUG_BUS_11__FOP_QUAD_HAS_4_FRAGMENTS_AFTER_UPDATE__SHIFT 0x7
+#define CB_DEBUG_BUS_11__FOP_QUAD_HAS_5_FRAGMENTS_AFTER_UPDATE_MASK 0x100
+#define CB_DEBUG_BUS_11__FOP_QUAD_HAS_5_FRAGMENTS_AFTER_UPDATE__SHIFT 0x8
+#define CB_DEBUG_BUS_11__FOP_QUAD_HAS_6_FRAGMENTS_AFTER_UPDATE_MASK 0x200
+#define CB_DEBUG_BUS_11__FOP_QUAD_HAS_6_FRAGMENTS_AFTER_UPDATE__SHIFT 0x9
+#define CB_DEBUG_BUS_11__FOP_QUAD_HAS_7_FRAGMENTS_AFTER_UPDATE_MASK 0x400
+#define CB_DEBUG_BUS_11__FOP_QUAD_HAS_7_FRAGMENTS_AFTER_UPDATE__SHIFT 0xa
+#define CB_DEBUG_BUS_11__FOP_QUAD_HAS_8_FRAGMENTS_AFTER_UPDATE_MASK 0x800
+#define CB_DEBUG_BUS_11__FOP_QUAD_HAS_8_FRAGMENTS_AFTER_UPDATE__SHIFT 0xb
+#define CB_DEBUG_BUS_11__FOP_QUAD_ADDED_1_FRAGMENT_MASK 0x1000
+#define CB_DEBUG_BUS_11__FOP_QUAD_ADDED_1_FRAGMENT__SHIFT 0xc
+#define CB_DEBUG_BUS_11__FOP_QUAD_ADDED_2_FRAGMENTS_MASK 0x2000
+#define CB_DEBUG_BUS_11__FOP_QUAD_ADDED_2_FRAGMENTS__SHIFT 0xd
+#define CB_DEBUG_BUS_11__FOP_QUAD_ADDED_3_FRAGMENTS_MASK 0x4000
+#define CB_DEBUG_BUS_11__FOP_QUAD_ADDED_3_FRAGMENTS__SHIFT 0xe
+#define CB_DEBUG_BUS_11__FOP_QUAD_ADDED_4_FRAGMENTS_MASK 0x8000
+#define CB_DEBUG_BUS_11__FOP_QUAD_ADDED_4_FRAGMENTS__SHIFT 0xf
+#define CB_DEBUG_BUS_11__FOP_QUAD_ADDED_5_FRAGMENTS_MASK 0x10000
+#define CB_DEBUG_BUS_11__FOP_QUAD_ADDED_5_FRAGMENTS__SHIFT 0x10
+#define CB_DEBUG_BUS_11__FOP_QUAD_ADDED_6_FRAGMENTS_MASK 0x20000
+#define CB_DEBUG_BUS_11__FOP_QUAD_ADDED_6_FRAGMENTS__SHIFT 0x11
+#define CB_DEBUG_BUS_11__FOP_QUAD_ADDED_7_FRAGMENTS_MASK 0x40000
+#define CB_DEBUG_BUS_11__FOP_QUAD_ADDED_7_FRAGMENTS__SHIFT 0x12
+#define CB_DEBUG_BUS_11__FOP_QUAD_REMOVED_1_FRAGMENT_MASK 0x80000
+#define CB_DEBUG_BUS_11__FOP_QUAD_REMOVED_1_FRAGMENT__SHIFT 0x13
+#define CB_DEBUG_BUS_11__FOP_QUAD_REMOVED_2_FRAGMENTS_MASK 0x100000
+#define CB_DEBUG_BUS_11__FOP_QUAD_REMOVED_2_FRAGMENTS__SHIFT 0x14
+#define CB_DEBUG_BUS_11__FOP_QUAD_REMOVED_3_FRAGMENTS_MASK 0x200000
+#define CB_DEBUG_BUS_11__FOP_QUAD_REMOVED_3_FRAGMENTS__SHIFT 0x15
+#define CB_DEBUG_BUS_11__FOP_QUAD_REMOVED_4_FRAGMENTS_MASK 0x400000
+#define CB_DEBUG_BUS_11__FOP_QUAD_REMOVED_4_FRAGMENTS__SHIFT 0x16
+#define CB_DEBUG_BUS_11__FOP_QUAD_REMOVED_5_FRAGMENTS_MASK 0x800000
+#define CB_DEBUG_BUS_11__FOP_QUAD_REMOVED_5_FRAGMENTS__SHIFT 0x17
+#define CB_DEBUG_BUS_12__FOP_QUAD_REMOVED_6_FRAGMENTS_MASK 0x1
+#define CB_DEBUG_BUS_12__FOP_QUAD_REMOVED_6_FRAGMENTS__SHIFT 0x0
+#define CB_DEBUG_BUS_12__FOP_QUAD_REMOVED_7_FRAGMENTS_MASK 0x2
+#define CB_DEBUG_BUS_12__FOP_QUAD_REMOVED_7_FRAGMENTS__SHIFT 0x1
+#define CB_DEBUG_BUS_12__FC_CC_QUADFRAG_READS_FRAGMENT_0_MASK 0x4
+#define CB_DEBUG_BUS_12__FC_CC_QUADFRAG_READS_FRAGMENT_0__SHIFT 0x2
+#define CB_DEBUG_BUS_12__FC_CC_QUADFRAG_READS_FRAGMENT_1_MASK 0x8
+#define CB_DEBUG_BUS_12__FC_CC_QUADFRAG_READS_FRAGMENT_1__SHIFT 0x3
+#define CB_DEBUG_BUS_12__FC_CC_QUADFRAG_READS_FRAGMENT_2_MASK 0x10
+#define CB_DEBUG_BUS_12__FC_CC_QUADFRAG_READS_FRAGMENT_2__SHIFT 0x4
+#define CB_DEBUG_BUS_12__FC_CC_QUADFRAG_READS_FRAGMENT_3_MASK 0x20
+#define CB_DEBUG_BUS_12__FC_CC_QUADFRAG_READS_FRAGMENT_3__SHIFT 0x5
+#define CB_DEBUG_BUS_12__FC_CC_QUADFRAG_READS_FRAGMENT_4_MASK 0x40
+#define CB_DEBUG_BUS_12__FC_CC_QUADFRAG_READS_FRAGMENT_4__SHIFT 0x6
+#define CB_DEBUG_BUS_12__FC_CC_QUADFRAG_READS_FRAGMENT_5_MASK 0x80
+#define CB_DEBUG_BUS_12__FC_CC_QUADFRAG_READS_FRAGMENT_5__SHIFT 0x7
+#define CB_DEBUG_BUS_12__FC_CC_QUADFRAG_READS_FRAGMENT_6_MASK 0x100
+#define CB_DEBUG_BUS_12__FC_CC_QUADFRAG_READS_FRAGMENT_6__SHIFT 0x8
+#define CB_DEBUG_BUS_12__FC_CC_QUADFRAG_READS_FRAGMENT_7_MASK 0x200
+#define CB_DEBUG_BUS_12__FC_CC_QUADFRAG_READS_FRAGMENT_7__SHIFT 0x9
+#define CB_DEBUG_BUS_12__FC_CC_QUADFRAG_WRITES_FRAGMENT_0_MASK 0x400
+#define CB_DEBUG_BUS_12__FC_CC_QUADFRAG_WRITES_FRAGMENT_0__SHIFT 0xa
+#define CB_DEBUG_BUS_12__FC_CC_QUADFRAG_WRITES_FRAGMENT_1_MASK 0x800
+#define CB_DEBUG_BUS_12__FC_CC_QUADFRAG_WRITES_FRAGMENT_1__SHIFT 0xb
+#define CB_DEBUG_BUS_12__FC_CC_QUADFRAG_WRITES_FRAGMENT_2_MASK 0x1000
+#define CB_DEBUG_BUS_12__FC_CC_QUADFRAG_WRITES_FRAGMENT_2__SHIFT 0xc
+#define CB_DEBUG_BUS_12__FC_CC_QUADFRAG_WRITES_FRAGMENT_3_MASK 0x2000
+#define CB_DEBUG_BUS_12__FC_CC_QUADFRAG_WRITES_FRAGMENT_3__SHIFT 0xd
+#define CB_DEBUG_BUS_12__FC_CC_QUADFRAG_WRITES_FRAGMENT_4_MASK 0x4000
+#define CB_DEBUG_BUS_12__FC_CC_QUADFRAG_WRITES_FRAGMENT_4__SHIFT 0xe
+#define CB_DEBUG_BUS_12__FC_CC_QUADFRAG_WRITES_FRAGMENT_5_MASK 0x8000
+#define CB_DEBUG_BUS_12__FC_CC_QUADFRAG_WRITES_FRAGMENT_5__SHIFT 0xf
+#define CB_DEBUG_BUS_12__FC_CC_QUADFRAG_WRITES_FRAGMENT_6_MASK 0x10000
+#define CB_DEBUG_BUS_12__FC_CC_QUADFRAG_WRITES_FRAGMENT_6__SHIFT 0x10
+#define CB_DEBUG_BUS_12__FC_CC_QUADFRAG_WRITES_FRAGMENT_7_MASK 0x20000
+#define CB_DEBUG_BUS_12__FC_CC_QUADFRAG_WRITES_FRAGMENT_7__SHIFT 0x11
+#define CB_DEBUG_BUS_12__FC_QUAD_BLEND_OPT_DONT_READ_DST_MASK 0x40000
+#define CB_DEBUG_BUS_12__FC_QUAD_BLEND_OPT_DONT_READ_DST__SHIFT 0x12
+#define CB_DEBUG_BUS_12__FC_QUAD_BLEND_OPT_BLEND_BYPASS_MASK 0x80000
+#define CB_DEBUG_BUS_12__FC_QUAD_BLEND_OPT_BLEND_BYPASS__SHIFT 0x13
+#define CB_DEBUG_BUS_12__FC_QUAD_BLEND_OPT_DISCARD_PIXELS_MASK 0x100000
+#define CB_DEBUG_BUS_12__FC_QUAD_BLEND_OPT_DISCARD_PIXELS__SHIFT 0x14
+#define CB_DEBUG_BUS_12__FC_QUAD_KILLED_BY_EXTRA_PIXEL_EXPORT_MASK 0x200000
+#define CB_DEBUG_BUS_12__FC_QUAD_KILLED_BY_EXTRA_PIXEL_EXPORT__SHIFT 0x15
+#define CB_DEBUG_BUS_12__FC_QUAD_KILLED_BY_COLOR_INVALID_MASK 0x400000
+#define CB_DEBUG_BUS_12__FC_QUAD_KILLED_BY_COLOR_INVALID__SHIFT 0x16
+#define CB_DEBUG_BUS_12__FC_QUAD_KILLED_BY_NULL_TARGET_SHADER_MASK_MASK 0x800000
+#define CB_DEBUG_BUS_12__FC_QUAD_KILLED_BY_NULL_TARGET_SHADER_MASK__SHIFT 0x17
+#define CB_DEBUG_BUS_13__FC_PF_FC_KEYID_RDLAT_FIFO_FULL_MASK 0x1
+#define CB_DEBUG_BUS_13__FC_PF_FC_KEYID_RDLAT_FIFO_FULL__SHIFT 0x0
+#define CB_DEBUG_BUS_13__FC_DOC_QTILE_CAM_MISS_MASK 0x2
+#define CB_DEBUG_BUS_13__FC_DOC_QTILE_CAM_MISS__SHIFT 0x1
+#define CB_DEBUG_BUS_13__FC_DOC_QTILE_CAM_HIT_MASK 0x4
+#define CB_DEBUG_BUS_13__FC_DOC_QTILE_CAM_HIT__SHIFT 0x2
+#define CB_DEBUG_BUS_13__FC_DOC_CLINE_CAM_MISS_MASK 0x8
+#define CB_DEBUG_BUS_13__FC_DOC_CLINE_CAM_MISS__SHIFT 0x3
+#define CB_DEBUG_BUS_13__FC_DOC_CLINE_CAM_HIT_MASK 0x10
+#define CB_DEBUG_BUS_13__FC_DOC_CLINE_CAM_HIT__SHIFT 0x4
+#define CB_DEBUG_BUS_13__FC_DOC_OVERWROTE_1_SECTOR_MASK 0x20
+#define CB_DEBUG_BUS_13__FC_DOC_OVERWROTE_1_SECTOR__SHIFT 0x5
+#define CB_DEBUG_BUS_13__FC_DOC_OVERWROTE_2_SECTORS_MASK 0x40
+#define CB_DEBUG_BUS_13__FC_DOC_OVERWROTE_2_SECTORS__SHIFT 0x6
+#define CB_DEBUG_BUS_13__FC_DOC_OVERWROTE_3_SECTORS_MASK 0x80
+#define CB_DEBUG_BUS_13__FC_DOC_OVERWROTE_3_SECTORS__SHIFT 0x7
+#define CB_DEBUG_BUS_13__FC_DOC_OVERWROTE_4_SECTORS_MASK 0x100
+#define CB_DEBUG_BUS_13__FC_DOC_OVERWROTE_4_SECTORS__SHIFT 0x8
+#define CB_DEBUG_BUS_13__FC_PF_DCC_CACHE_HIT_MASK 0x200
+#define CB_DEBUG_BUS_13__FC_PF_DCC_CACHE_HIT__SHIFT 0x9
+#define CB_DEBUG_BUS_13__FC_PF_DCC_CACHE_TAG_MISS_MASK 0x400
+#define CB_DEBUG_BUS_13__FC_PF_DCC_CACHE_TAG_MISS__SHIFT 0xa
+#define CB_DEBUG_BUS_13__FC_PF_DCC_CACHE_SECTOR_MISS_MASK 0x800
+#define CB_DEBUG_BUS_13__FC_PF_DCC_CACHE_SECTOR_MISS__SHIFT 0xb
+#define CB_DEBUG_BUS_13__FC_PF_DCC_CACHE_REEVICTION_STALL_MASK 0x1000
+#define CB_DEBUG_BUS_13__FC_PF_DCC_CACHE_REEVICTION_STALL__SHIFT 0xc
+#define CB_DEBUG_BUS_13__FC_PF_DCC_CACHE_EVICT_NONZERO_INFLIGHT_STALL_MASK 0x2000
+#define CB_DEBUG_BUS_13__FC_PF_DCC_CACHE_EVICT_NONZERO_INFLIGHT_STALL__SHIFT 0xd
+#define CB_DEBUG_BUS_13__FC_PF_DCC_CACHE_REPLACE_PENDING_EVICT_STALL_MASK 0x4000
+#define CB_DEBUG_BUS_13__FC_PF_DCC_CACHE_REPLACE_PENDING_EVICT_STALL__SHIFT 0xe
+#define CB_DEBUG_BUS_13__FC_PF_DCC_CACHE_INFLIGHT_COUNTER_MAXIMUM_STALL_MASK 0x8000
+#define CB_DEBUG_BUS_13__FC_PF_DCC_CACHE_INFLIGHT_COUNTER_MAXIMUM_STALL__SHIFT 0xf
+#define CB_DEBUG_BUS_13__FC_PF_DCC_CACHE_READ_OUTPUT_STALL_MASK 0x10000
+#define CB_DEBUG_BUS_13__FC_PF_DCC_CACHE_READ_OUTPUT_STALL__SHIFT 0x10
+#define CB_DEBUG_BUS_13__FC_PF_DCC_CACHE_WRITE_OUTPUT_STALL_MASK 0x20000
+#define CB_DEBUG_BUS_13__FC_PF_DCC_CACHE_WRITE_OUTPUT_STALL__SHIFT 0x11
+#define CB_DEBUG_BUS_13__FC_PF_DCC_CACHE_ACK_OUTPUT_STALL_MASK 0x40000
+#define CB_DEBUG_BUS_13__FC_PF_DCC_CACHE_ACK_OUTPUT_STALL__SHIFT 0x12
+#define CB_DEBUG_BUS_13__FC_PF_DCC_CACHE_STALL_MASK 0x80000
+#define CB_DEBUG_BUS_13__FC_PF_DCC_CACHE_STALL__SHIFT 0x13
+#define CB_DEBUG_BUS_13__FC_PF_DCC_CACHE_FLUSH_MASK 0x100000
+#define CB_DEBUG_BUS_13__FC_PF_DCC_CACHE_FLUSH__SHIFT 0x14
+#define CB_DEBUG_BUS_13__FC_PF_DCC_CACHE_SECTORS_FLUSHED_MASK 0x200000
+#define CB_DEBUG_BUS_13__FC_PF_DCC_CACHE_SECTORS_FLUSHED__SHIFT 0x15
+#define CB_DEBUG_BUS_13__FC_PF_DCC_CACHE_DIRTY_SECTORS_FLUSHED_MASK 0x400000
+#define CB_DEBUG_BUS_13__FC_PF_DCC_CACHE_DIRTY_SECTORS_FLUSHED__SHIFT 0x16
+#define CB_DEBUG_BUS_13__FC_PF_DCC_CACHE_TAGS_FLUSHED_MASK 0x800000
+#define CB_DEBUG_BUS_13__FC_PF_DCC_CACHE_TAGS_FLUSHED__SHIFT 0x17
+#define CB_DEBUG_BUS_14__FC_MC_DCC_WRITE_REQUESTS_IN_FLIGHT_MASK 0x7ff
+#define CB_DEBUG_BUS_14__FC_MC_DCC_WRITE_REQUESTS_IN_FLIGHT__SHIFT 0x0
+#define CB_DEBUG_BUS_14__FC_MC_DCC_READ_REQUESTS_IN_FLIGHT_MASK 0x3ff800
+#define CB_DEBUG_BUS_14__FC_MC_DCC_READ_REQUESTS_IN_FLIGHT__SHIFT 0xb
+#define CB_DEBUG_BUS_14__CC_PF_DCC_BEYOND_TILE_SPLIT_MASK 0x400000
+#define CB_DEBUG_BUS_14__CC_PF_DCC_BEYOND_TILE_SPLIT__SHIFT 0x16
+#define CB_DEBUG_BUS_14__CC_PF_DCC_RDREQ_STALL_MASK 0x800000
+#define CB_DEBUG_BUS_14__CC_PF_DCC_RDREQ_STALL__SHIFT 0x17
+#define CB_DEBUG_BUS_15__CC_PF_DCC_COMPRESS_RATIO_2TO1_MASK 0x7
+#define CB_DEBUG_BUS_15__CC_PF_DCC_COMPRESS_RATIO_2TO1__SHIFT 0x0
+#define CB_DEBUG_BUS_15__CC_PF_DCC_COMPRESS_RATIO_4TO1_MASK 0x18
+#define CB_DEBUG_BUS_15__CC_PF_DCC_COMPRESS_RATIO_4TO1__SHIFT 0x3
+#define CB_DEBUG_BUS_15__CC_PF_DCC_COMPRESS_RATIO_4TO2_MASK 0x60
+#define CB_DEBUG_BUS_15__CC_PF_DCC_COMPRESS_RATIO_4TO2__SHIFT 0x5
+#define CB_DEBUG_BUS_15__CC_PF_DCC_COMPRESS_RATIO_4TO3_MASK 0x180
+#define CB_DEBUG_BUS_15__CC_PF_DCC_COMPRESS_RATIO_4TO3__SHIFT 0x7
+#define CB_DEBUG_BUS_15__CC_PF_DCC_COMPRESS_RATIO_6TO1_MASK 0x600
+#define CB_DEBUG_BUS_15__CC_PF_DCC_COMPRESS_RATIO_6TO1__SHIFT 0x9
+#define CB_DEBUG_BUS_15__CC_PF_DCC_COMPRESS_RATIO_6TO2_MASK 0x1800
+#define CB_DEBUG_BUS_15__CC_PF_DCC_COMPRESS_RATIO_6TO2__SHIFT 0xb
+#define CB_DEBUG_BUS_15__CC_PF_DCC_COMPRESS_RATIO_6TO3_MASK 0x6000
+#define CB_DEBUG_BUS_15__CC_PF_DCC_COMPRESS_RATIO_6TO3__SHIFT 0xd
+#define CB_DEBUG_BUS_15__CC_PF_DCC_COMPRESS_RATIO_6TO4_MASK 0x18000
+#define CB_DEBUG_BUS_15__CC_PF_DCC_COMPRESS_RATIO_6TO4__SHIFT 0xf
+#define CB_DEBUG_BUS_15__CC_PF_DCC_COMPRESS_RATIO_6TO5_MASK 0x60000
+#define CB_DEBUG_BUS_15__CC_PF_DCC_COMPRESS_RATIO_6TO5__SHIFT 0x11
+#define CB_DEBUG_BUS_16__CC_PF_DCC_COMPRESS_RATIO_8TO1_MASK 0x1
+#define CB_DEBUG_BUS_16__CC_PF_DCC_COMPRESS_RATIO_8TO1__SHIFT 0x0
+#define CB_DEBUG_BUS_16__CC_PF_DCC_COMPRESS_RATIO_8TO2_MASK 0x2
+#define CB_DEBUG_BUS_16__CC_PF_DCC_COMPRESS_RATIO_8TO2__SHIFT 0x1
+#define CB_DEBUG_BUS_16__CC_PF_DCC_COMPRESS_RATIO_8TO3_MASK 0x4
+#define CB_DEBUG_BUS_16__CC_PF_DCC_COMPRESS_RATIO_8TO3__SHIFT 0x2
+#define CB_DEBUG_BUS_16__CC_PF_DCC_COMPRESS_RATIO_8TO4_MASK 0x8
+#define CB_DEBUG_BUS_16__CC_PF_DCC_COMPRESS_RATIO_8TO4__SHIFT 0x3
+#define CB_DEBUG_BUS_16__CC_PF_DCC_COMPRESS_RATIO_8TO5_MASK 0x10
+#define CB_DEBUG_BUS_16__CC_PF_DCC_COMPRESS_RATIO_8TO5__SHIFT 0x4
+#define CB_DEBUG_BUS_16__CC_PF_DCC_COMPRESS_RATIO_8TO6_MASK 0x20
+#define CB_DEBUG_BUS_16__CC_PF_DCC_COMPRESS_RATIO_8TO6__SHIFT 0x5
+#define CB_DEBUG_BUS_16__CC_PF_DCC_COMPRESS_RATIO_8TO7_MASK 0x40
+#define CB_DEBUG_BUS_16__CC_PF_DCC_COMPRESS_RATIO_8TO7__SHIFT 0x6
+#define CB_DEBUG_BUS_17__TILE_INTFC_BUSY_MASK 0x1
+#define CB_DEBUG_BUS_17__TILE_INTFC_BUSY__SHIFT 0x0
+#define CB_DEBUG_BUS_17__MU_BUSY_MASK 0x2
+#define CB_DEBUG_BUS_17__MU_BUSY__SHIFT 0x1
+#define CB_DEBUG_BUS_17__TQ_BUSY_MASK 0x4
+#define CB_DEBUG_BUS_17__TQ_BUSY__SHIFT 0x2
+#define CB_DEBUG_BUS_17__AC_BUSY_MASK 0x8
+#define CB_DEBUG_BUS_17__AC_BUSY__SHIFT 0x3
+#define CB_DEBUG_BUS_17__CRW_BUSY_MASK 0x10
+#define CB_DEBUG_BUS_17__CRW_BUSY__SHIFT 0x4
+#define CB_DEBUG_BUS_17__CACHE_CTRL_BUSY_MASK 0x20
+#define CB_DEBUG_BUS_17__CACHE_CTRL_BUSY__SHIFT 0x5
+#define CB_DEBUG_BUS_17__MC_WR_PENDING_MASK 0x40
+#define CB_DEBUG_BUS_17__MC_WR_PENDING__SHIFT 0x6
+#define CB_DEBUG_BUS_17__FC_WR_PENDING_MASK 0x80
+#define CB_DEBUG_BUS_17__FC_WR_PENDING__SHIFT 0x7
+#define CB_DEBUG_BUS_17__FC_RD_PENDING_MASK 0x100
+#define CB_DEBUG_BUS_17__FC_RD_PENDING__SHIFT 0x8
+#define CB_DEBUG_BUS_17__EVICT_PENDING_MASK 0x200
+#define CB_DEBUG_BUS_17__EVICT_PENDING__SHIFT 0x9
+#define CB_DEBUG_BUS_17__LAST_RD_ARB_WINNER_MASK 0x400
+#define CB_DEBUG_BUS_17__LAST_RD_ARB_WINNER__SHIFT 0xa
+#define CB_DEBUG_BUS_17__MU_STATE_MASK 0x7f800
+#define CB_DEBUG_BUS_17__MU_STATE__SHIFT 0xb
+#define CB_DEBUG_BUS_18__TILE_RETIREMENT_BUSY_MASK 0x1
+#define CB_DEBUG_BUS_18__TILE_RETIREMENT_BUSY__SHIFT 0x0
+#define CB_DEBUG_BUS_18__FOP_BUSY_MASK 0x2
+#define CB_DEBUG_BUS_18__FOP_BUSY__SHIFT 0x1
+#define CB_DEBUG_BUS_18__CLEAR_BUSY_MASK 0x4
+#define CB_DEBUG_BUS_18__CLEAR_BUSY__SHIFT 0x2
+#define CB_DEBUG_BUS_18__LAT_BUSY_MASK 0x8
+#define CB_DEBUG_BUS_18__LAT_BUSY__SHIFT 0x3
+#define CB_DEBUG_BUS_18__CACHE_CTL_BUSY_MASK 0x10
+#define CB_DEBUG_BUS_18__CACHE_CTL_BUSY__SHIFT 0x4
+#define CB_DEBUG_BUS_18__ADDR_BUSY_MASK 0x20
+#define CB_DEBUG_BUS_18__ADDR_BUSY__SHIFT 0x5
+#define CB_DEBUG_BUS_18__MERGE_BUSY_MASK 0x40
+#define CB_DEBUG_BUS_18__MERGE_BUSY__SHIFT 0x6
+#define CB_DEBUG_BUS_18__QUAD_BUSY_MASK 0x80
+#define CB_DEBUG_BUS_18__QUAD_BUSY__SHIFT 0x7
+#define CB_DEBUG_BUS_18__TILE_BUSY_MASK 0x100
+#define CB_DEBUG_BUS_18__TILE_BUSY__SHIFT 0x8
+#define CB_DEBUG_BUS_18__DCC_BUSY_MASK 0x200
+#define CB_DEBUG_BUS_18__DCC_BUSY__SHIFT 0x9
+#define CB_DEBUG_BUS_18__DOC_BUSY_MASK 0x400
+#define CB_DEBUG_BUS_18__DOC_BUSY__SHIFT 0xa
+#define CB_DEBUG_BUS_18__DAG_BUSY_MASK 0x800
+#define CB_DEBUG_BUS_18__DAG_BUSY__SHIFT 0xb
+#define CB_DEBUG_BUS_18__DOC_STALL_MASK 0x1000
+#define CB_DEBUG_BUS_18__DOC_STALL__SHIFT 0xc
+#define CB_DEBUG_BUS_18__DOC_QT_CAM_FULL_MASK 0x2000
+#define CB_DEBUG_BUS_18__DOC_QT_CAM_FULL__SHIFT 0xd
+#define CB_DEBUG_BUS_18__DOC_CL_CAM_FULL_MASK 0x4000
+#define CB_DEBUG_BUS_18__DOC_CL_CAM_FULL__SHIFT 0xe
+#define CB_DEBUG_BUS_18__DOC_QUAD_PTR_FIFO_FULL_MASK 0x8000
+#define CB_DEBUG_BUS_18__DOC_QUAD_PTR_FIFO_FULL__SHIFT 0xf
+#define CB_DEBUG_BUS_18__DOC_SECTOR_MASK_FIFO_FULL_MASK 0x10000
+#define CB_DEBUG_BUS_18__DOC_SECTOR_MASK_FIFO_FULL__SHIFT 0x10
+#define CB_DEBUG_BUS_18__DCS_READ_WINNER_LAST_MASK 0x20000
+#define CB_DEBUG_BUS_18__DCS_READ_WINNER_LAST__SHIFT 0x11
+#define CB_DEBUG_BUS_18__DCS_READ_EV_PENDING_MASK 0x40000
+#define CB_DEBUG_BUS_18__DCS_READ_EV_PENDING__SHIFT 0x12
+#define CB_DEBUG_BUS_18__DCS_WRITE_CC_PENDING_MASK 0x80000
+#define CB_DEBUG_BUS_18__DCS_WRITE_CC_PENDING__SHIFT 0x13
+#define CB_DEBUG_BUS_18__DCS_READ_CC_PENDING_MASK 0x100000
+#define CB_DEBUG_BUS_18__DCS_READ_CC_PENDING__SHIFT 0x14
+#define CB_DEBUG_BUS_18__DCS_WRITE_MC_PENDING_MASK 0x200000
+#define CB_DEBUG_BUS_18__DCS_WRITE_MC_PENDING__SHIFT 0x15
+#define CB_DEBUG_BUS_19__SURF_SYNC_STATE_MASK 0x3
+#define CB_DEBUG_BUS_19__SURF_SYNC_STATE__SHIFT 0x0
+#define CB_DEBUG_BUS_19__SURF_SYNC_START_MASK 0x4
+#define CB_DEBUG_BUS_19__SURF_SYNC_START__SHIFT 0x2
+#define CB_DEBUG_BUS_19__SF_BUSY_MASK 0x8
+#define CB_DEBUG_BUS_19__SF_BUSY__SHIFT 0x3
+#define CB_DEBUG_BUS_19__CS_BUSY_MASK 0x10
+#define CB_DEBUG_BUS_19__CS_BUSY__SHIFT 0x4
+#define CB_DEBUG_BUS_19__RB_BUSY_MASK 0x20
+#define CB_DEBUG_BUS_19__RB_BUSY__SHIFT 0x5
+#define CB_DEBUG_BUS_19__DS_BUSY_MASK 0x40
+#define CB_DEBUG_BUS_19__DS_BUSY__SHIFT 0x6
+#define CB_DEBUG_BUS_19__TB_BUSY_MASK 0x80
+#define CB_DEBUG_BUS_19__TB_BUSY__SHIFT 0x7
+#define CB_DEBUG_BUS_19__IB_BUSY_MASK 0x100
+#define CB_DEBUG_BUS_19__IB_BUSY__SHIFT 0x8
+#define CB_DEBUG_BUS_19__DRR_BUSY_MASK 0x200
+#define CB_DEBUG_BUS_19__DRR_BUSY__SHIFT 0x9
+#define CB_DEBUG_BUS_19__DF_BUSY_MASK 0x400
+#define CB_DEBUG_BUS_19__DF_BUSY__SHIFT 0xa
+#define CB_DEBUG_BUS_19__DD_BUSY_MASK 0x800
+#define CB_DEBUG_BUS_19__DD_BUSY__SHIFT 0xb
+#define CB_DEBUG_BUS_19__DC_BUSY_MASK 0x1000
+#define CB_DEBUG_BUS_19__DC_BUSY__SHIFT 0xc
+#define CB_DEBUG_BUS_19__DK_BUSY_MASK 0x2000
+#define CB_DEBUG_BUS_19__DK_BUSY__SHIFT 0xd
+#define CB_DEBUG_BUS_19__DF_SKID_FIFO_EMPTY_MASK 0x4000
+#define CB_DEBUG_BUS_19__DF_SKID_FIFO_EMPTY__SHIFT 0xe
+#define CB_DEBUG_BUS_19__DF_CLEAR_FIFO_EMPTY_MASK 0x8000
+#define CB_DEBUG_BUS_19__DF_CLEAR_FIFO_EMPTY__SHIFT 0xf
+#define CB_DEBUG_BUS_19__DD_READY_MASK 0x10000
+#define CB_DEBUG_BUS_19__DD_READY__SHIFT 0x10
+#define CB_DEBUG_BUS_19__DC_FIFO_FULL_MASK 0x20000
+#define CB_DEBUG_BUS_19__DC_FIFO_FULL__SHIFT 0x11
+#define CB_DEBUG_BUS_19__DC_READY_MASK 0x40000
+#define CB_DEBUG_BUS_19__DC_READY__SHIFT 0x12
+#define CB_DEBUG_BUS_20__MC_RDREQ_CREDITS_MASK 0x3f
+#define CB_DEBUG_BUS_20__MC_RDREQ_CREDITS__SHIFT 0x0
+#define CB_DEBUG_BUS_20__MC_WRREQ_CREDITS_MASK 0xfc0
+#define CB_DEBUG_BUS_20__MC_WRREQ_CREDITS__SHIFT 0x6
+#define CB_DEBUG_BUS_20__CC_RDREQ_HAD_ITS_TURN_MASK 0x1000
+#define CB_DEBUG_BUS_20__CC_RDREQ_HAD_ITS_TURN__SHIFT 0xc
+#define CB_DEBUG_BUS_20__FC_RDREQ_HAD_ITS_TURN_MASK 0x2000
+#define CB_DEBUG_BUS_20__FC_RDREQ_HAD_ITS_TURN__SHIFT 0xd
+#define CB_DEBUG_BUS_20__CM_RDREQ_HAD_ITS_TURN_MASK 0x4000
+#define CB_DEBUG_BUS_20__CM_RDREQ_HAD_ITS_TURN__SHIFT 0xe
+#define CB_DEBUG_BUS_20__CC_WRREQ_HAD_ITS_TURN_MASK 0x10000
+#define CB_DEBUG_BUS_20__CC_WRREQ_HAD_ITS_TURN__SHIFT 0x10
+#define CB_DEBUG_BUS_20__FC_WRREQ_HAD_ITS_TURN_MASK 0x20000
+#define CB_DEBUG_BUS_20__FC_WRREQ_HAD_ITS_TURN__SHIFT 0x11
+#define CB_DEBUG_BUS_20__CM_WRREQ_HAD_ITS_TURN_MASK 0x40000
+#define CB_DEBUG_BUS_20__CM_WRREQ_HAD_ITS_TURN__SHIFT 0x12
+#define CB_DEBUG_BUS_20__CC_WRREQ_FIFO_EMPTY_MASK 0x100000
+#define CB_DEBUG_BUS_20__CC_WRREQ_FIFO_EMPTY__SHIFT 0x14
+#define CB_DEBUG_BUS_20__FC_WRREQ_FIFO_EMPTY_MASK 0x200000
+#define CB_DEBUG_BUS_20__FC_WRREQ_FIFO_EMPTY__SHIFT 0x15
+#define CB_DEBUG_BUS_20__CM_WRREQ_FIFO_EMPTY_MASK 0x400000
+#define CB_DEBUG_BUS_20__CM_WRREQ_FIFO_EMPTY__SHIFT 0x16
+#define CB_DEBUG_BUS_20__DCC_WRREQ_FIFO_EMPTY_MASK 0x800000
+#define CB_DEBUG_BUS_20__DCC_WRREQ_FIFO_EMPTY__SHIFT 0x17
+#define CB_DEBUG_BUS_21__CM_BUSY_MASK 0x1
+#define CB_DEBUG_BUS_21__CM_BUSY__SHIFT 0x0
+#define CB_DEBUG_BUS_21__FC_BUSY_MASK 0x2
+#define CB_DEBUG_BUS_21__FC_BUSY__SHIFT 0x1
+#define CB_DEBUG_BUS_21__CC_BUSY_MASK 0x4
+#define CB_DEBUG_BUS_21__CC_BUSY__SHIFT 0x2
+#define CB_DEBUG_BUS_21__BB_BUSY_MASK 0x8
+#define CB_DEBUG_BUS_21__BB_BUSY__SHIFT 0x3
+#define CB_DEBUG_BUS_21__MA_BUSY_MASK 0x10
+#define CB_DEBUG_BUS_21__MA_BUSY__SHIFT 0x4
+#define CB_DEBUG_BUS_21__CORE_SCLK_VLD_MASK 0x20
+#define CB_DEBUG_BUS_21__CORE_SCLK_VLD__SHIFT 0x5
+#define CB_DEBUG_BUS_21__REG_SCLK1_VLD_MASK 0x40
+#define CB_DEBUG_BUS_21__REG_SCLK1_VLD__SHIFT 0x6
+#define CB_DEBUG_BUS_21__REG_SCLK0_VLD_MASK 0x80
+#define CB_DEBUG_BUS_21__REG_SCLK0_VLD__SHIFT 0x7
+#define CB_DEBUG_BUS_22__OUTSTANDING_MC_READS_MASK 0xfff
+#define CB_DEBUG_BUS_22__OUTSTANDING_MC_READS__SHIFT 0x0
+#define CB_DEBUG_BUS_22__OUTSTANDING_MC_WRITES_MASK 0xfff000
+#define CB_DEBUG_BUS_22__OUTSTANDING_MC_WRITES__SHIFT 0xc
+#define CP_DFY_CNTL__POLICY_MASK 0x1
+#define CP_DFY_CNTL__POLICY__SHIFT 0x0
+#define CP_DFY_CNTL__MTYPE_MASK 0xc
+#define CP_DFY_CNTL__MTYPE__SHIFT 0x2
+#define CP_DFY_CNTL__LFSR_RESET_MASK 0x10000000
+#define CP_DFY_CNTL__LFSR_RESET__SHIFT 0x1c
+#define CP_DFY_CNTL__MODE_MASK 0x60000000
+#define CP_DFY_CNTL__MODE__SHIFT 0x1d
+#define CP_DFY_CNTL__ENABLE_MASK 0x80000000
+#define CP_DFY_CNTL__ENABLE__SHIFT 0x1f
+#define CP_DFY_STAT__BURST_COUNT_MASK 0xffff
+#define CP_DFY_STAT__BURST_COUNT__SHIFT 0x0
+#define CP_DFY_STAT__TAGS_PENDING_MASK 0x1ff0000
+#define CP_DFY_STAT__TAGS_PENDING__SHIFT 0x10
+#define CP_DFY_STAT__BUSY_MASK 0x80000000
+#define CP_DFY_STAT__BUSY__SHIFT 0x1f
+#define CP_DFY_ADDR_HI__ADDR_HI_MASK 0xffffffff
+#define CP_DFY_ADDR_HI__ADDR_HI__SHIFT 0x0
+#define CP_DFY_ADDR_LO__ADDR_LO_MASK 0xffffffe0
+#define CP_DFY_ADDR_LO__ADDR_LO__SHIFT 0x5
+#define CP_DFY_DATA_0__DATA_MASK 0xffffffff
+#define CP_DFY_DATA_0__DATA__SHIFT 0x0
+#define CP_DFY_DATA_1__DATA_MASK 0xffffffff
+#define CP_DFY_DATA_1__DATA__SHIFT 0x0
+#define CP_DFY_DATA_2__DATA_MASK 0xffffffff
+#define CP_DFY_DATA_2__DATA__SHIFT 0x0
+#define CP_DFY_DATA_3__DATA_MASK 0xffffffff
+#define CP_DFY_DATA_3__DATA__SHIFT 0x0
+#define CP_DFY_DATA_4__DATA_MASK 0xffffffff
+#define CP_DFY_DATA_4__DATA__SHIFT 0x0
+#define CP_DFY_DATA_5__DATA_MASK 0xffffffff
+#define CP_DFY_DATA_5__DATA__SHIFT 0x0
+#define CP_DFY_DATA_6__DATA_MASK 0xffffffff
+#define CP_DFY_DATA_6__DATA__SHIFT 0x0
+#define CP_DFY_DATA_7__DATA_MASK 0xffffffff
+#define CP_DFY_DATA_7__DATA__SHIFT 0x0
+#define CP_DFY_DATA_8__DATA_MASK 0xffffffff
+#define CP_DFY_DATA_8__DATA__SHIFT 0x0
+#define CP_DFY_DATA_9__DATA_MASK 0xffffffff
+#define CP_DFY_DATA_9__DATA__SHIFT 0x0
+#define CP_DFY_DATA_10__DATA_MASK 0xffffffff
+#define CP_DFY_DATA_10__DATA__SHIFT 0x0
+#define CP_DFY_DATA_11__DATA_MASK 0xffffffff
+#define CP_DFY_DATA_11__DATA__SHIFT 0x0
+#define CP_DFY_DATA_12__DATA_MASK 0xffffffff
+#define CP_DFY_DATA_12__DATA__SHIFT 0x0
+#define CP_DFY_DATA_13__DATA_MASK 0xffffffff
+#define CP_DFY_DATA_13__DATA__SHIFT 0x0
+#define CP_DFY_DATA_14__DATA_MASK 0xffffffff
+#define CP_DFY_DATA_14__DATA__SHIFT 0x0
+#define CP_DFY_DATA_15__DATA_MASK 0xffffffff
+#define CP_DFY_DATA_15__DATA__SHIFT 0x0
+#define CP_DFY_CMD__OFFSET_MASK 0x1ff
+#define CP_DFY_CMD__OFFSET__SHIFT 0x0
+#define CP_DFY_CMD__SIZE_MASK 0xffff0000
+#define CP_DFY_CMD__SIZE__SHIFT 0x10
+#define CP_CPC_MGCG_SYNC_CNTL__COOLDOWN_PERIOD_MASK 0xff
+#define CP_CPC_MGCG_SYNC_CNTL__COOLDOWN_PERIOD__SHIFT 0x0
+#define CP_CPC_MGCG_SYNC_CNTL__WARMUP_PERIOD_MASK 0xff00
+#define CP_CPC_MGCG_SYNC_CNTL__WARMUP_PERIOD__SHIFT 0x8
+#define CP_ATCL1_CNTL__XNACK_REDO_TIMER_CNT_MASK 0x3ff
+#define CP_ATCL1_CNTL__XNACK_REDO_TIMER_CNT__SHIFT 0x0
+#define CP_RB0_BASE__RB_BASE_MASK 0xffffffff
+#define CP_RB0_BASE__RB_BASE__SHIFT 0x0
+#define CP_RB0_BASE_HI__RB_BASE_HI_MASK 0xff
+#define CP_RB0_BASE_HI__RB_BASE_HI__SHIFT 0x0
+#define CP_RB_BASE__RB_BASE_MASK 0xffffffff
+#define CP_RB_BASE__RB_BASE__SHIFT 0x0
+#define CP_RB1_BASE__RB_BASE_MASK 0xffffffff
+#define CP_RB1_BASE__RB_BASE__SHIFT 0x0
+#define CP_RB1_BASE_HI__RB_BASE_HI_MASK 0xff
+#define CP_RB1_BASE_HI__RB_BASE_HI__SHIFT 0x0
+#define CP_RB2_BASE__RB_BASE_MASK 0xffffffff
+#define CP_RB2_BASE__RB_BASE__SHIFT 0x0
+#define CP_RB0_CNTL__RB_BUFSZ_MASK 0x3f
+#define CP_RB0_CNTL__RB_BUFSZ__SHIFT 0x0
+#define CP_RB0_CNTL__RB_BLKSZ_MASK 0x3f00
+#define CP_RB0_CNTL__RB_BLKSZ__SHIFT 0x8
+#define CP_RB0_CNTL__MTYPE_MASK 0x18000
+#define CP_RB0_CNTL__MTYPE__SHIFT 0xf
+#define CP_RB0_CNTL__BUF_SWAP_MASK 0x60000
+#define CP_RB0_CNTL__BUF_SWAP__SHIFT 0x11
+#define CP_RB0_CNTL__MIN_AVAILSZ_MASK 0x300000
+#define CP_RB0_CNTL__MIN_AVAILSZ__SHIFT 0x14
+#define CP_RB0_CNTL__MIN_IB_AVAILSZ_MASK 0xc00000
+#define CP_RB0_CNTL__MIN_IB_AVAILSZ__SHIFT 0x16
+#define CP_RB0_CNTL__CACHE_POLICY_MASK 0x1000000
+#define CP_RB0_CNTL__CACHE_POLICY__SHIFT 0x18
+#define CP_RB0_CNTL__RB_NO_UPDATE_MASK 0x8000000
+#define CP_RB0_CNTL__RB_NO_UPDATE__SHIFT 0x1b
+#define CP_RB0_CNTL__RB_RPTR_WR_ENA_MASK 0x80000000
+#define CP_RB0_CNTL__RB_RPTR_WR_ENA__SHIFT 0x1f
+#define CP_RB_CNTL__RB_BUFSZ_MASK 0x3f
+#define CP_RB_CNTL__RB_BUFSZ__SHIFT 0x0
+#define CP_RB_CNTL__RB_BLKSZ_MASK 0x3f00
+#define CP_RB_CNTL__RB_BLKSZ__SHIFT 0x8
+#define CP_RB_CNTL__MTYPE_MASK 0x18000
+#define CP_RB_CNTL__MTYPE__SHIFT 0xf
+#define CP_RB_CNTL__BUF_SWAP_MASK 0x60000
+#define CP_RB_CNTL__BUF_SWAP__SHIFT 0x11
+#define CP_RB_CNTL__MIN_AVAILSZ_MASK 0x300000
+#define CP_RB_CNTL__MIN_AVAILSZ__SHIFT 0x14
+#define CP_RB_CNTL__MIN_IB_AVAILSZ_MASK 0xc00000
+#define CP_RB_CNTL__MIN_IB_AVAILSZ__SHIFT 0x16
+#define CP_RB_CNTL__CACHE_POLICY_MASK 0x1000000
+#define CP_RB_CNTL__CACHE_POLICY__SHIFT 0x18
+#define CP_RB_CNTL__RB_NO_UPDATE_MASK 0x8000000
+#define CP_RB_CNTL__RB_NO_UPDATE__SHIFT 0x1b
+#define CP_RB_CNTL__RB_RPTR_WR_ENA_MASK 0x80000000
+#define CP_RB_CNTL__RB_RPTR_WR_ENA__SHIFT 0x1f
+#define CP_RB1_CNTL__RB_BUFSZ_MASK 0x3f
+#define CP_RB1_CNTL__RB_BUFSZ__SHIFT 0x0
+#define CP_RB1_CNTL__RB_BLKSZ_MASK 0x3f00
+#define CP_RB1_CNTL__RB_BLKSZ__SHIFT 0x8
+#define CP_RB1_CNTL__MTYPE_MASK 0x18000
+#define CP_RB1_CNTL__MTYPE__SHIFT 0xf
+#define CP_RB1_CNTL__MIN_AVAILSZ_MASK 0x300000
+#define CP_RB1_CNTL__MIN_AVAILSZ__SHIFT 0x14
+#define CP_RB1_CNTL__MIN_IB_AVAILSZ_MASK 0xc00000
+#define CP_RB1_CNTL__MIN_IB_AVAILSZ__SHIFT 0x16
+#define CP_RB1_CNTL__CACHE_POLICY_MASK 0x1000000
+#define CP_RB1_CNTL__CACHE_POLICY__SHIFT 0x18
+#define CP_RB1_CNTL__RB_NO_UPDATE_MASK 0x8000000
+#define CP_RB1_CNTL__RB_NO_UPDATE__SHIFT 0x1b
+#define CP_RB1_CNTL__RB_RPTR_WR_ENA_MASK 0x80000000
+#define CP_RB1_CNTL__RB_RPTR_WR_ENA__SHIFT 0x1f
+#define CP_RB2_CNTL__RB_BUFSZ_MASK 0x3f
+#define CP_RB2_CNTL__RB_BUFSZ__SHIFT 0x0
+#define CP_RB2_CNTL__RB_BLKSZ_MASK 0x3f00
+#define CP_RB2_CNTL__RB_BLKSZ__SHIFT 0x8
+#define CP_RB2_CNTL__MTYPE_MASK 0x18000
+#define CP_RB2_CNTL__MTYPE__SHIFT 0xf
+#define CP_RB2_CNTL__MIN_AVAILSZ_MASK 0x300000
+#define CP_RB2_CNTL__MIN_AVAILSZ__SHIFT 0x14
+#define CP_RB2_CNTL__MIN_IB_AVAILSZ_MASK 0xc00000
+#define CP_RB2_CNTL__MIN_IB_AVAILSZ__SHIFT 0x16
+#define CP_RB2_CNTL__CACHE_POLICY_MASK 0x1000000
+#define CP_RB2_CNTL__CACHE_POLICY__SHIFT 0x18
+#define CP_RB2_CNTL__RB_NO_UPDATE_MASK 0x8000000
+#define CP_RB2_CNTL__RB_NO_UPDATE__SHIFT 0x1b
+#define CP_RB2_CNTL__RB_RPTR_WR_ENA_MASK 0x80000000
+#define CP_RB2_CNTL__RB_RPTR_WR_ENA__SHIFT 0x1f
+#define CP_RB_RPTR_WR__RB_RPTR_WR_MASK 0xfffff
+#define CP_RB_RPTR_WR__RB_RPTR_WR__SHIFT 0x0
+#define CP_RB0_RPTR_ADDR__RB_RPTR_SWAP_MASK 0x3
+#define CP_RB0_RPTR_ADDR__RB_RPTR_SWAP__SHIFT 0x0
+#define CP_RB0_RPTR_ADDR__RB_RPTR_ADDR_MASK 0xfffffffc
+#define CP_RB0_RPTR_ADDR__RB_RPTR_ADDR__SHIFT 0x2
+#define CP_RB_RPTR_ADDR__RB_RPTR_SWAP_MASK 0x3
+#define CP_RB_RPTR_ADDR__RB_RPTR_SWAP__SHIFT 0x0
+#define CP_RB_RPTR_ADDR__RB_RPTR_ADDR_MASK 0xfffffffc
+#define CP_RB_RPTR_ADDR__RB_RPTR_ADDR__SHIFT 0x2
+#define CP_RB1_RPTR_ADDR__RB_RPTR_SWAP_MASK 0x3
+#define CP_RB1_RPTR_ADDR__RB_RPTR_SWAP__SHIFT 0x0
+#define CP_RB1_RPTR_ADDR__RB_RPTR_ADDR_MASK 0xfffffffc
+#define CP_RB1_RPTR_ADDR__RB_RPTR_ADDR__SHIFT 0x2
+#define CP_RB2_RPTR_ADDR__RB_RPTR_SWAP_MASK 0x3
+#define CP_RB2_RPTR_ADDR__RB_RPTR_SWAP__SHIFT 0x0
+#define CP_RB2_RPTR_ADDR__RB_RPTR_ADDR_MASK 0xfffffffc
+#define CP_RB2_RPTR_ADDR__RB_RPTR_ADDR__SHIFT 0x2
+#define CP_RB0_RPTR_ADDR_HI__RB_RPTR_ADDR_HI_MASK 0xffff
+#define CP_RB0_RPTR_ADDR_HI__RB_RPTR_ADDR_HI__SHIFT 0x0
+#define CP_RB_RPTR_ADDR_HI__RB_RPTR_ADDR_HI_MASK 0xffff
+#define CP_RB_RPTR_ADDR_HI__RB_RPTR_ADDR_HI__SHIFT 0x0
+#define CP_RB1_RPTR_ADDR_HI__RB_RPTR_ADDR_HI_MASK 0xffff
+#define CP_RB1_RPTR_ADDR_HI__RB_RPTR_ADDR_HI__SHIFT 0x0
+#define CP_RB2_RPTR_ADDR_HI__RB_RPTR_ADDR_HI_MASK 0xffff
+#define CP_RB2_RPTR_ADDR_HI__RB_RPTR_ADDR_HI__SHIFT 0x0
+#define CP_RB0_WPTR__RB_WPTR_MASK 0xfffff
+#define CP_RB0_WPTR__RB_WPTR__SHIFT 0x0
+#define CP_RB_WPTR__RB_WPTR_MASK 0xfffff
+#define CP_RB_WPTR__RB_WPTR__SHIFT 0x0
+#define CP_RB1_WPTR__RB_WPTR_MASK 0xfffff
+#define CP_RB1_WPTR__RB_WPTR__SHIFT 0x0
+#define CP_RB2_WPTR__RB_WPTR_MASK 0xfffff
+#define CP_RB2_WPTR__RB_WPTR__SHIFT 0x0
+#define CP_RB_WPTR_POLL_ADDR_LO__RB_WPTR_POLL_ADDR_LO_MASK 0xfffffffc
+#define CP_RB_WPTR_POLL_ADDR_LO__RB_WPTR_POLL_ADDR_LO__SHIFT 0x2
+#define CP_RB_WPTR_POLL_ADDR_HI__RB_WPTR_POLL_ADDR_HI_MASK 0xff
+#define CP_RB_WPTR_POLL_ADDR_HI__RB_WPTR_POLL_ADDR_HI__SHIFT 0x0
+#define CP_INT_CNTL__CP_VM_DOORBELL_WR_INT_ENABLE_MASK 0x800
+#define CP_INT_CNTL__CP_VM_DOORBELL_WR_INT_ENABLE__SHIFT 0xb
+#define CP_INT_CNTL__CP_ECC_ERROR_INT_ENABLE_MASK 0x4000
+#define CP_INT_CNTL__CP_ECC_ERROR_INT_ENABLE__SHIFT 0xe
+#define CP_INT_CNTL__WRM_POLL_TIMEOUT_INT_ENABLE_MASK 0x20000
+#define CP_INT_CNTL__WRM_POLL_TIMEOUT_INT_ENABLE__SHIFT 0x11
+#define CP_INT_CNTL__CMP_BUSY_INT_ENABLE_MASK 0x40000
+#define CP_INT_CNTL__CMP_BUSY_INT_ENABLE__SHIFT 0x12
+#define CP_INT_CNTL__CNTX_BUSY_INT_ENABLE_MASK 0x80000
+#define CP_INT_CNTL__CNTX_BUSY_INT_ENABLE__SHIFT 0x13
+#define CP_INT_CNTL__CNTX_EMPTY_INT_ENABLE_MASK 0x100000
+#define CP_INT_CNTL__CNTX_EMPTY_INT_ENABLE__SHIFT 0x14
+#define CP_INT_CNTL__GFX_IDLE_INT_ENABLE_MASK 0x200000
+#define CP_INT_CNTL__GFX_IDLE_INT_ENABLE__SHIFT 0x15
+#define CP_INT_CNTL__PRIV_INSTR_INT_ENABLE_MASK 0x400000
+#define CP_INT_CNTL__PRIV_INSTR_INT_ENABLE__SHIFT 0x16
+#define CP_INT_CNTL__PRIV_REG_INT_ENABLE_MASK 0x800000
+#define CP_INT_CNTL__PRIV_REG_INT_ENABLE__SHIFT 0x17
+#define CP_INT_CNTL__OPCODE_ERROR_INT_ENABLE_MASK 0x1000000
+#define CP_INT_CNTL__OPCODE_ERROR_INT_ENABLE__SHIFT 0x18
+#define CP_INT_CNTL__TIME_STAMP_INT_ENABLE_MASK 0x4000000
+#define CP_INT_CNTL__TIME_STAMP_INT_ENABLE__SHIFT 0x1a
+#define CP_INT_CNTL__RESERVED_BIT_ERROR_INT_ENABLE_MASK 0x8000000
+#define CP_INT_CNTL__RESERVED_BIT_ERROR_INT_ENABLE__SHIFT 0x1b
+#define CP_INT_CNTL__GENERIC2_INT_ENABLE_MASK 0x20000000
+#define CP_INT_CNTL__GENERIC2_INT_ENABLE__SHIFT 0x1d
+#define CP_INT_CNTL__GENERIC1_INT_ENABLE_MASK 0x40000000
+#define CP_INT_CNTL__GENERIC1_INT_ENABLE__SHIFT 0x1e
+#define CP_INT_CNTL__GENERIC0_INT_ENABLE_MASK 0x80000000
+#define CP_INT_CNTL__GENERIC0_INT_ENABLE__SHIFT 0x1f
+#define CP_INT_CNTL_RING0__CP_VM_DOORBELL_WR_INT_ENABLE_MASK 0x800
+#define CP_INT_CNTL_RING0__CP_VM_DOORBELL_WR_INT_ENABLE__SHIFT 0xb
+#define CP_INT_CNTL_RING0__CP_ECC_ERROR_INT_ENABLE_MASK 0x4000
+#define CP_INT_CNTL_RING0__CP_ECC_ERROR_INT_ENABLE__SHIFT 0xe
+#define CP_INT_CNTL_RING0__WRM_POLL_TIMEOUT_INT_ENABLE_MASK 0x20000
+#define CP_INT_CNTL_RING0__WRM_POLL_TIMEOUT_INT_ENABLE__SHIFT 0x11
+#define CP_INT_CNTL_RING0__CMP_BUSY_INT_ENABLE_MASK 0x40000
+#define CP_INT_CNTL_RING0__CMP_BUSY_INT_ENABLE__SHIFT 0x12
+#define CP_INT_CNTL_RING0__CNTX_BUSY_INT_ENABLE_MASK 0x80000
+#define CP_INT_CNTL_RING0__CNTX_BUSY_INT_ENABLE__SHIFT 0x13
+#define CP_INT_CNTL_RING0__CNTX_EMPTY_INT_ENABLE_MASK 0x100000
+#define CP_INT_CNTL_RING0__CNTX_EMPTY_INT_ENABLE__SHIFT 0x14
+#define CP_INT_CNTL_RING0__GFX_IDLE_INT_ENABLE_MASK 0x200000
+#define CP_INT_CNTL_RING0__GFX_IDLE_INT_ENABLE__SHIFT 0x15
+#define CP_INT_CNTL_RING0__PRIV_INSTR_INT_ENABLE_MASK 0x400000
+#define CP_INT_CNTL_RING0__PRIV_INSTR_INT_ENABLE__SHIFT 0x16
+#define CP_INT_CNTL_RING0__PRIV_REG_INT_ENABLE_MASK 0x800000
+#define CP_INT_CNTL_RING0__PRIV_REG_INT_ENABLE__SHIFT 0x17
+#define CP_INT_CNTL_RING0__OPCODE_ERROR_INT_ENABLE_MASK 0x1000000
+#define CP_INT_CNTL_RING0__OPCODE_ERROR_INT_ENABLE__SHIFT 0x18
+#define CP_INT_CNTL_RING0__TIME_STAMP_INT_ENABLE_MASK 0x4000000
+#define CP_INT_CNTL_RING0__TIME_STAMP_INT_ENABLE__SHIFT 0x1a
+#define CP_INT_CNTL_RING0__RESERVED_BIT_ERROR_INT_ENABLE_MASK 0x8000000
+#define CP_INT_CNTL_RING0__RESERVED_BIT_ERROR_INT_ENABLE__SHIFT 0x1b
+#define CP_INT_CNTL_RING0__GENERIC2_INT_ENABLE_MASK 0x20000000
+#define CP_INT_CNTL_RING0__GENERIC2_INT_ENABLE__SHIFT 0x1d
+#define CP_INT_CNTL_RING0__GENERIC1_INT_ENABLE_MASK 0x40000000
+#define CP_INT_CNTL_RING0__GENERIC1_INT_ENABLE__SHIFT 0x1e
+#define CP_INT_CNTL_RING0__GENERIC0_INT_ENABLE_MASK 0x80000000
+#define CP_INT_CNTL_RING0__GENERIC0_INT_ENABLE__SHIFT 0x1f
+#define CP_INT_CNTL_RING1__CP_VM_DOORBELL_WR_INT_ENABLE_MASK 0x800
+#define CP_INT_CNTL_RING1__CP_VM_DOORBELL_WR_INT_ENABLE__SHIFT 0xb
+#define CP_INT_CNTL_RING1__CP_ECC_ERROR_INT_ENABLE_MASK 0x4000
+#define CP_INT_CNTL_RING1__CP_ECC_ERROR_INT_ENABLE__SHIFT 0xe
+#define CP_INT_CNTL_RING1__WRM_POLL_TIMEOUT_INT_ENABLE_MASK 0x20000
+#define CP_INT_CNTL_RING1__WRM_POLL_TIMEOUT_INT_ENABLE__SHIFT 0x11
+#define CP_INT_CNTL_RING1__CMP_BUSY_INT_ENABLE_MASK 0x40000
+#define CP_INT_CNTL_RING1__CMP_BUSY_INT_ENABLE__SHIFT 0x12
+#define CP_INT_CNTL_RING1__CNTX_BUSY_INT_ENABLE_MASK 0x80000
+#define CP_INT_CNTL_RING1__CNTX_BUSY_INT_ENABLE__SHIFT 0x13
+#define CP_INT_CNTL_RING1__CNTX_EMPTY_INT_ENABLE_MASK 0x100000
+#define CP_INT_CNTL_RING1__CNTX_EMPTY_INT_ENABLE__SHIFT 0x14
+#define CP_INT_CNTL_RING1__GFX_IDLE_INT_ENABLE_MASK 0x200000
+#define CP_INT_CNTL_RING1__GFX_IDLE_INT_ENABLE__SHIFT 0x15
+#define CP_INT_CNTL_RING1__PRIV_INSTR_INT_ENABLE_MASK 0x400000
+#define CP_INT_CNTL_RING1__PRIV_INSTR_INT_ENABLE__SHIFT 0x16
+#define CP_INT_CNTL_RING1__PRIV_REG_INT_ENABLE_MASK 0x800000
+#define CP_INT_CNTL_RING1__PRIV_REG_INT_ENABLE__SHIFT 0x17
+#define CP_INT_CNTL_RING1__OPCODE_ERROR_INT_ENABLE_MASK 0x1000000
+#define CP_INT_CNTL_RING1__OPCODE_ERROR_INT_ENABLE__SHIFT 0x18
+#define CP_INT_CNTL_RING1__TIME_STAMP_INT_ENABLE_MASK 0x4000000
+#define CP_INT_CNTL_RING1__TIME_STAMP_INT_ENABLE__SHIFT 0x1a
+#define CP_INT_CNTL_RING1__RESERVED_BIT_ERROR_INT_ENABLE_MASK 0x8000000
+#define CP_INT_CNTL_RING1__RESERVED_BIT_ERROR_INT_ENABLE__SHIFT 0x1b
+#define CP_INT_CNTL_RING1__GENERIC2_INT_ENABLE_MASK 0x20000000
+#define CP_INT_CNTL_RING1__GENERIC2_INT_ENABLE__SHIFT 0x1d
+#define CP_INT_CNTL_RING1__GENERIC1_INT_ENABLE_MASK 0x40000000
+#define CP_INT_CNTL_RING1__GENERIC1_INT_ENABLE__SHIFT 0x1e
+#define CP_INT_CNTL_RING1__GENERIC0_INT_ENABLE_MASK 0x80000000
+#define CP_INT_CNTL_RING1__GENERIC0_INT_ENABLE__SHIFT 0x1f
+#define CP_INT_CNTL_RING2__CP_VM_DOORBELL_WR_INT_ENABLE_MASK 0x800
+#define CP_INT_CNTL_RING2__CP_VM_DOORBELL_WR_INT_ENABLE__SHIFT 0xb
+#define CP_INT_CNTL_RING2__CP_ECC_ERROR_INT_ENABLE_MASK 0x4000
+#define CP_INT_CNTL_RING2__CP_ECC_ERROR_INT_ENABLE__SHIFT 0xe
+#define CP_INT_CNTL_RING2__WRM_POLL_TIMEOUT_INT_ENABLE_MASK 0x20000
+#define CP_INT_CNTL_RING2__WRM_POLL_TIMEOUT_INT_ENABLE__SHIFT 0x11
+#define CP_INT_CNTL_RING2__CMP_BUSY_INT_ENABLE_MASK 0x40000
+#define CP_INT_CNTL_RING2__CMP_BUSY_INT_ENABLE__SHIFT 0x12
+#define CP_INT_CNTL_RING2__CNTX_BUSY_INT_ENABLE_MASK 0x80000
+#define CP_INT_CNTL_RING2__CNTX_BUSY_INT_ENABLE__SHIFT 0x13
+#define CP_INT_CNTL_RING2__CNTX_EMPTY_INT_ENABLE_MASK 0x100000
+#define CP_INT_CNTL_RING2__CNTX_EMPTY_INT_ENABLE__SHIFT 0x14
+#define CP_INT_CNTL_RING2__GFX_IDLE_INT_ENABLE_MASK 0x200000
+#define CP_INT_CNTL_RING2__GFX_IDLE_INT_ENABLE__SHIFT 0x15
+#define CP_INT_CNTL_RING2__PRIV_INSTR_INT_ENABLE_MASK 0x400000
+#define CP_INT_CNTL_RING2__PRIV_INSTR_INT_ENABLE__SHIFT 0x16
+#define CP_INT_CNTL_RING2__PRIV_REG_INT_ENABLE_MASK 0x800000
+#define CP_INT_CNTL_RING2__PRIV_REG_INT_ENABLE__SHIFT 0x17
+#define CP_INT_CNTL_RING2__OPCODE_ERROR_INT_ENABLE_MASK 0x1000000
+#define CP_INT_CNTL_RING2__OPCODE_ERROR_INT_ENABLE__SHIFT 0x18
+#define CP_INT_CNTL_RING2__TIME_STAMP_INT_ENABLE_MASK 0x4000000
+#define CP_INT_CNTL_RING2__TIME_STAMP_INT_ENABLE__SHIFT 0x1a
+#define CP_INT_CNTL_RING2__RESERVED_BIT_ERROR_INT_ENABLE_MASK 0x8000000
+#define CP_INT_CNTL_RING2__RESERVED_BIT_ERROR_INT_ENABLE__SHIFT 0x1b
+#define CP_INT_CNTL_RING2__GENERIC2_INT_ENABLE_MASK 0x20000000
+#define CP_INT_CNTL_RING2__GENERIC2_INT_ENABLE__SHIFT 0x1d
+#define CP_INT_CNTL_RING2__GENERIC1_INT_ENABLE_MASK 0x40000000
+#define CP_INT_CNTL_RING2__GENERIC1_INT_ENABLE__SHIFT 0x1e
+#define CP_INT_CNTL_RING2__GENERIC0_INT_ENABLE_MASK 0x80000000
+#define CP_INT_CNTL_RING2__GENERIC0_INT_ENABLE__SHIFT 0x1f
+#define CP_INT_STATUS__CP_VM_DOORBELL_WR_INT_STAT_MASK 0x800
+#define CP_INT_STATUS__CP_VM_DOORBELL_WR_INT_STAT__SHIFT 0xb
+#define CP_INT_STATUS__CP_ECC_ERROR_INT_STAT_MASK 0x4000
+#define CP_INT_STATUS__CP_ECC_ERROR_INT_STAT__SHIFT 0xe
+#define CP_INT_STATUS__WRM_POLL_TIMEOUT_INT_STAT_MASK 0x20000
+#define CP_INT_STATUS__WRM_POLL_TIMEOUT_INT_STAT__SHIFT 0x11
+#define CP_INT_STATUS__CMP_BUSY_INT_STAT_MASK 0x40000
+#define CP_INT_STATUS__CMP_BUSY_INT_STAT__SHIFT 0x12
+#define CP_INT_STATUS__CNTX_BUSY_INT_STAT_MASK 0x80000
+#define CP_INT_STATUS__CNTX_BUSY_INT_STAT__SHIFT 0x13
+#define CP_INT_STATUS__CNTX_EMPTY_INT_STAT_MASK 0x100000
+#define CP_INT_STATUS__CNTX_EMPTY_INT_STAT__SHIFT 0x14
+#define CP_INT_STATUS__GFX_IDLE_INT_STAT_MASK 0x200000
+#define CP_INT_STATUS__GFX_IDLE_INT_STAT__SHIFT 0x15
+#define CP_INT_STATUS__PRIV_INSTR_INT_STAT_MASK 0x400000
+#define CP_INT_STATUS__PRIV_INSTR_INT_STAT__SHIFT 0x16
+#define CP_INT_STATUS__PRIV_REG_INT_STAT_MASK 0x800000
+#define CP_INT_STATUS__PRIV_REG_INT_STAT__SHIFT 0x17
+#define CP_INT_STATUS__OPCODE_ERROR_INT_STAT_MASK 0x1000000
+#define CP_INT_STATUS__OPCODE_ERROR_INT_STAT__SHIFT 0x18
+#define CP_INT_STATUS__TIME_STAMP_INT_STAT_MASK 0x4000000
+#define CP_INT_STATUS__TIME_STAMP_INT_STAT__SHIFT 0x1a
+#define CP_INT_STATUS__RESERVED_BIT_ERROR_INT_STAT_MASK 0x8000000
+#define CP_INT_STATUS__RESERVED_BIT_ERROR_INT_STAT__SHIFT 0x1b
+#define CP_INT_STATUS__GENERIC2_INT_STAT_MASK 0x20000000
+#define CP_INT_STATUS__GENERIC2_INT_STAT__SHIFT 0x1d
+#define CP_INT_STATUS__GENERIC1_INT_STAT_MASK 0x40000000
+#define CP_INT_STATUS__GENERIC1_INT_STAT__SHIFT 0x1e
+#define CP_INT_STATUS__GENERIC0_INT_STAT_MASK 0x80000000
+#define CP_INT_STATUS__GENERIC0_INT_STAT__SHIFT 0x1f
+#define CP_INT_STATUS_RING0__CP_VM_DOORBELL_WR_INT_STAT_MASK 0x800
+#define CP_INT_STATUS_RING0__CP_VM_DOORBELL_WR_INT_STAT__SHIFT 0xb
+#define CP_INT_STATUS_RING0__CP_ECC_ERROR_INT_STAT_MASK 0x4000
+#define CP_INT_STATUS_RING0__CP_ECC_ERROR_INT_STAT__SHIFT 0xe
+#define CP_INT_STATUS_RING0__WRM_POLL_TIMEOUT_INT_STAT_MASK 0x20000
+#define CP_INT_STATUS_RING0__WRM_POLL_TIMEOUT_INT_STAT__SHIFT 0x11
+#define CP_INT_STATUS_RING0__CMP_BUSY_INT_STAT_MASK 0x40000
+#define CP_INT_STATUS_RING0__CMP_BUSY_INT_STAT__SHIFT 0x12
+#define CP_INT_STATUS_RING0__GCNTX_BUSY_INT_STAT_MASK 0x80000
+#define CP_INT_STATUS_RING0__GCNTX_BUSY_INT_STAT__SHIFT 0x13
+#define CP_INT_STATUS_RING0__CNTX_EMPTY_INT_STAT_MASK 0x100000
+#define CP_INT_STATUS_RING0__CNTX_EMPTY_INT_STAT__SHIFT 0x14
+#define CP_INT_STATUS_RING0__GFX_IDLE_INT_STAT_MASK 0x200000
+#define CP_INT_STATUS_RING0__GFX_IDLE_INT_STAT__SHIFT 0x15
+#define CP_INT_STATUS_RING0__PRIV_INSTR_INT_STAT_MASK 0x400000
+#define CP_INT_STATUS_RING0__PRIV_INSTR_INT_STAT__SHIFT 0x16
+#define CP_INT_STATUS_RING0__PRIV_REG_INT_STAT_MASK 0x800000
+#define CP_INT_STATUS_RING0__PRIV_REG_INT_STAT__SHIFT 0x17
+#define CP_INT_STATUS_RING0__OPCODE_ERROR_INT_STAT_MASK 0x1000000
+#define CP_INT_STATUS_RING0__OPCODE_ERROR_INT_STAT__SHIFT 0x18
+#define CP_INT_STATUS_RING0__TIME_STAMP_INT_STAT_MASK 0x4000000
+#define CP_INT_STATUS_RING0__TIME_STAMP_INT_STAT__SHIFT 0x1a
+#define CP_INT_STATUS_RING0__RESERVED_BIT_ERROR_INT_STAT_MASK 0x8000000
+#define CP_INT_STATUS_RING0__RESERVED_BIT_ERROR_INT_STAT__SHIFT 0x1b
+#define CP_INT_STATUS_RING0__GENERIC2_INT_STAT_MASK 0x20000000
+#define CP_INT_STATUS_RING0__GENERIC2_INT_STAT__SHIFT 0x1d
+#define CP_INT_STATUS_RING0__GENERIC1_INT_STAT_MASK 0x40000000
+#define CP_INT_STATUS_RING0__GENERIC1_INT_STAT__SHIFT 0x1e
+#define CP_INT_STATUS_RING0__GENERIC0_INT_STAT_MASK 0x80000000
+#define CP_INT_STATUS_RING0__GENERIC0_INT_STAT__SHIFT 0x1f
+#define CP_INT_STATUS_RING1__CP_VM_DOORBELL_WR_INT_STAT_MASK 0x800
+#define CP_INT_STATUS_RING1__CP_VM_DOORBELL_WR_INT_STAT__SHIFT 0xb
+#define CP_INT_STATUS_RING1__CP_ECC_ERROR_INT_STAT_MASK 0x4000
+#define CP_INT_STATUS_RING1__CP_ECC_ERROR_INT_STAT__SHIFT 0xe
+#define CP_INT_STATUS_RING1__WRM_POLL_TIMEOUT_INT_STAT_MASK 0x20000
+#define CP_INT_STATUS_RING1__WRM_POLL_TIMEOUT_INT_STAT__SHIFT 0x11
+#define CP_INT_STATUS_RING1__CMP_BUSY_INT_STAT_MASK 0x40000
+#define CP_INT_STATUS_RING1__CMP_BUSY_INT_STAT__SHIFT 0x12
+#define CP_INT_STATUS_RING1__CNTX_BUSY_INT_STAT_MASK 0x80000
+#define CP_INT_STATUS_RING1__CNTX_BUSY_INT_STAT__SHIFT 0x13
+#define CP_INT_STATUS_RING1__CNTX_EMPTY_INT_STAT_MASK 0x100000
+#define CP_INT_STATUS_RING1__CNTX_EMPTY_INT_STAT__SHIFT 0x14
+#define CP_INT_STATUS_RING1__GFX_IDLE_INT_STAT_MASK 0x200000
+#define CP_INT_STATUS_RING1__GFX_IDLE_INT_STAT__SHIFT 0x15
+#define CP_INT_STATUS_RING1__PRIV_INSTR_INT_STAT_MASK 0x400000
+#define CP_INT_STATUS_RING1__PRIV_INSTR_INT_STAT__SHIFT 0x16
+#define CP_INT_STATUS_RING1__PRIV_REG_INT_STAT_MASK 0x800000
+#define CP_INT_STATUS_RING1__PRIV_REG_INT_STAT__SHIFT 0x17
+#define CP_INT_STATUS_RING1__OPCODE_ERROR_INT_STAT_MASK 0x1000000
+#define CP_INT_STATUS_RING1__OPCODE_ERROR_INT_STAT__SHIFT 0x18
+#define CP_INT_STATUS_RING1__TIME_STAMP_INT_STAT_MASK 0x4000000
+#define CP_INT_STATUS_RING1__TIME_STAMP_INT_STAT__SHIFT 0x1a
+#define CP_INT_STATUS_RING1__RESERVED_BIT_ERROR_INT_STAT_MASK 0x8000000
+#define CP_INT_STATUS_RING1__RESERVED_BIT_ERROR_INT_STAT__SHIFT 0x1b
+#define CP_INT_STATUS_RING1__GENERIC2_INT_STAT_MASK 0x20000000
+#define CP_INT_STATUS_RING1__GENERIC2_INT_STAT__SHIFT 0x1d
+#define CP_INT_STATUS_RING1__GENERIC1_INT_STAT_MASK 0x40000000
+#define CP_INT_STATUS_RING1__GENERIC1_INT_STAT__SHIFT 0x1e
+#define CP_INT_STATUS_RING1__GENERIC0_INT_STAT_MASK 0x80000000
+#define CP_INT_STATUS_RING1__GENERIC0_INT_STAT__SHIFT 0x1f
+#define CP_INT_STATUS_RING2__CP_VM_DOORBELL_WR_INT_STAT_MASK 0x800
+#define CP_INT_STATUS_RING2__CP_VM_DOORBELL_WR_INT_STAT__SHIFT 0xb
+#define CP_INT_STATUS_RING2__CP_ECC_ERROR_INT_STAT_MASK 0x4000
+#define CP_INT_STATUS_RING2__CP_ECC_ERROR_INT_STAT__SHIFT 0xe
+#define CP_INT_STATUS_RING2__WRM_POLL_TIMEOUT_INT_STAT_MASK 0x20000
+#define CP_INT_STATUS_RING2__WRM_POLL_TIMEOUT_INT_STAT__SHIFT 0x11
+#define CP_INT_STATUS_RING2__CMP_BUSY_INT_STAT_MASK 0x40000
+#define CP_INT_STATUS_RING2__CMP_BUSY_INT_STAT__SHIFT 0x12
+#define CP_INT_STATUS_RING2__CNTX_BUSY_INT_STAT_MASK 0x80000
+#define CP_INT_STATUS_RING2__CNTX_BUSY_INT_STAT__SHIFT 0x13
+#define CP_INT_STATUS_RING2__CNTX_EMPTY_INT_STAT_MASK 0x100000
+#define CP_INT_STATUS_RING2__CNTX_EMPTY_INT_STAT__SHIFT 0x14
+#define CP_INT_STATUS_RING2__GFX_IDLE_INT_STAT_MASK 0x200000
+#define CP_INT_STATUS_RING2__GFX_IDLE_INT_STAT__SHIFT 0x15
+#define CP_INT_STATUS_RING2__PRIV_INSTR_INT_STAT_MASK 0x400000
+#define CP_INT_STATUS_RING2__PRIV_INSTR_INT_STAT__SHIFT 0x16
+#define CP_INT_STATUS_RING2__PRIV_REG_INT_STAT_MASK 0x800000
+#define CP_INT_STATUS_RING2__PRIV_REG_INT_STAT__SHIFT 0x17
+#define CP_INT_STATUS_RING2__OPCODE_ERROR_INT_STAT_MASK 0x1000000
+#define CP_INT_STATUS_RING2__OPCODE_ERROR_INT_STAT__SHIFT 0x18
+#define CP_INT_STATUS_RING2__TIME_STAMP_INT_STAT_MASK 0x4000000
+#define CP_INT_STATUS_RING2__TIME_STAMP_INT_STAT__SHIFT 0x1a
+#define CP_INT_STATUS_RING2__RESERVED_BIT_ERROR_INT_STAT_MASK 0x8000000
+#define CP_INT_STATUS_RING2__RESERVED_BIT_ERROR_INT_STAT__SHIFT 0x1b
+#define CP_INT_STATUS_RING2__GENERIC2_INT_STAT_MASK 0x20000000
+#define CP_INT_STATUS_RING2__GENERIC2_INT_STAT__SHIFT 0x1d
+#define CP_INT_STATUS_RING2__GENERIC1_INT_STAT_MASK 0x40000000
+#define CP_INT_STATUS_RING2__GENERIC1_INT_STAT__SHIFT 0x1e
+#define CP_INT_STATUS_RING2__GENERIC0_INT_STAT_MASK 0x80000000
+#define CP_INT_STATUS_RING2__GENERIC0_INT_STAT__SHIFT 0x1f
+#define CP_DEVICE_ID__DEVICE_ID_MASK 0xff
+#define CP_DEVICE_ID__DEVICE_ID__SHIFT 0x0
+#define CP_RING_PRIORITY_CNTS__PRIORITY1_CNT_MASK 0xff
+#define CP_RING_PRIORITY_CNTS__PRIORITY1_CNT__SHIFT 0x0
+#define CP_RING_PRIORITY_CNTS__PRIORITY2A_CNT_MASK 0xff00
+#define CP_RING_PRIORITY_CNTS__PRIORITY2A_CNT__SHIFT 0x8
+#define CP_RING_PRIORITY_CNTS__PRIORITY2B_CNT_MASK 0xff0000
+#define CP_RING_PRIORITY_CNTS__PRIORITY2B_CNT__SHIFT 0x10
+#define CP_RING_PRIORITY_CNTS__PRIORITY3_CNT_MASK 0xff000000
+#define CP_RING_PRIORITY_CNTS__PRIORITY3_CNT__SHIFT 0x18
+#define CP_ME0_PIPE_PRIORITY_CNTS__PRIORITY1_CNT_MASK 0xff
+#define CP_ME0_PIPE_PRIORITY_CNTS__PRIORITY1_CNT__SHIFT 0x0
+#define CP_ME0_PIPE_PRIORITY_CNTS__PRIORITY2A_CNT_MASK 0xff00
+#define CP_ME0_PIPE_PRIORITY_CNTS__PRIORITY2A_CNT__SHIFT 0x8
+#define CP_ME0_PIPE_PRIORITY_CNTS__PRIORITY2B_CNT_MASK 0xff0000
+#define CP_ME0_PIPE_PRIORITY_CNTS__PRIORITY2B_CNT__SHIFT 0x10
+#define CP_ME0_PIPE_PRIORITY_CNTS__PRIORITY3_CNT_MASK 0xff000000
+#define CP_ME0_PIPE_PRIORITY_CNTS__PRIORITY3_CNT__SHIFT 0x18
+#define CP_RING0_PRIORITY__PRIORITY_MASK 0x3
+#define CP_RING0_PRIORITY__PRIORITY__SHIFT 0x0
+#define CP_ME0_PIPE0_PRIORITY__PRIORITY_MASK 0x3
+#define CP_ME0_PIPE0_PRIORITY__PRIORITY__SHIFT 0x0
+#define CP_RING1_PRIORITY__PRIORITY_MASK 0x3
+#define CP_RING1_PRIORITY__PRIORITY__SHIFT 0x0
+#define CP_ME0_PIPE1_PRIORITY__PRIORITY_MASK 0x3
+#define CP_ME0_PIPE1_PRIORITY__PRIORITY__SHIFT 0x0
+#define CP_RING2_PRIORITY__PRIORITY_MASK 0x3
+#define CP_RING2_PRIORITY__PRIORITY__SHIFT 0x0
+#define CP_ME0_PIPE2_PRIORITY__PRIORITY_MASK 0x3
+#define CP_ME0_PIPE2_PRIORITY__PRIORITY__SHIFT 0x0
+#define CP_ENDIAN_SWAP__ENDIAN_SWAP_MASK 0x3
+#define CP_ENDIAN_SWAP__ENDIAN_SWAP__SHIFT 0x0
+#define CP_RB_VMID__RB0_VMID_MASK 0xf
+#define CP_RB_VMID__RB0_VMID__SHIFT 0x0
+#define CP_RB_VMID__RB1_VMID_MASK 0xf00
+#define CP_RB_VMID__RB1_VMID__SHIFT 0x8
+#define CP_RB_VMID__RB2_VMID_MASK 0xf0000
+#define CP_RB_VMID__RB2_VMID__SHIFT 0x10
+#define CP_ME0_PIPE0_VMID__VMID_MASK 0xf
+#define CP_ME0_PIPE0_VMID__VMID__SHIFT 0x0
+#define CP_ME0_PIPE1_VMID__VMID_MASK 0xf
+#define CP_ME0_PIPE1_VMID__VMID__SHIFT 0x0
+#define CP_RB_DOORBELL_CONTROL__DOORBELL_OFFSET_MASK 0x7ffffc
+#define CP_RB_DOORBELL_CONTROL__DOORBELL_OFFSET__SHIFT 0x2
+#define CP_RB_DOORBELL_CONTROL__DOORBELL_EN_MASK 0x40000000
+#define CP_RB_DOORBELL_CONTROL__DOORBELL_EN__SHIFT 0x1e
+#define CP_RB_DOORBELL_CONTROL__DOORBELL_HIT_MASK 0x80000000
+#define CP_RB_DOORBELL_CONTROL__DOORBELL_HIT__SHIFT 0x1f
+#define CP_RB_DOORBELL_RANGE_LOWER__DOORBELL_RANGE_LOWER_MASK 0x7ffffc
+#define CP_RB_DOORBELL_RANGE_LOWER__DOORBELL_RANGE_LOWER__SHIFT 0x2
+#define CP_RB_DOORBELL_RANGE_UPPER__DOORBELL_RANGE_UPPER_MASK 0x7ffffc
+#define CP_RB_DOORBELL_RANGE_UPPER__DOORBELL_RANGE_UPPER__SHIFT 0x2
+#define CP_MEC_DOORBELL_RANGE_LOWER__DOORBELL_RANGE_LOWER_MASK 0x7ffffc
+#define CP_MEC_DOORBELL_RANGE_LOWER__DOORBELL_RANGE_LOWER__SHIFT 0x2
+#define CP_MEC_DOORBELL_RANGE_UPPER__DOORBELL_RANGE_UPPER_MASK 0x7ffffc
+#define CP_MEC_DOORBELL_RANGE_UPPER__DOORBELL_RANGE_UPPER__SHIFT 0x2
+#define CP_PFP_UCODE_ADDR__UCODE_ADDR_MASK 0x1fff
+#define CP_PFP_UCODE_ADDR__UCODE_ADDR__SHIFT 0x0
+#define CP_PFP_UCODE_DATA__UCODE_DATA_MASK 0xffffffff
+#define CP_PFP_UCODE_DATA__UCODE_DATA__SHIFT 0x0
+#define CP_ME_RAM_RADDR__ME_RAM_RADDR_MASK 0x1fff
+#define CP_ME_RAM_RADDR__ME_RAM_RADDR__SHIFT 0x0
+#define CP_ME_RAM_WADDR__ME_RAM_WADDR_MASK 0x1fff
+#define CP_ME_RAM_WADDR__ME_RAM_WADDR__SHIFT 0x0
+#define CP_ME_RAM_DATA__ME_RAM_DATA_MASK 0xffffffff
+#define CP_ME_RAM_DATA__ME_RAM_DATA__SHIFT 0x0
+#define CGTT_CPC_CLK_CTRL__ON_DELAY_MASK 0xf
+#define CGTT_CPC_CLK_CTRL__ON_DELAY__SHIFT 0x0
+#define CGTT_CPC_CLK_CTRL__OFF_HYSTERESIS_MASK 0xff0
+#define CGTT_CPC_CLK_CTRL__OFF_HYSTERESIS__SHIFT 0x4
+#define CGTT_CPC_CLK_CTRL__SOFT_OVERRIDE_PERFMON_MASK 0x20000000
+#define CGTT_CPC_CLK_CTRL__SOFT_OVERRIDE_PERFMON__SHIFT 0x1d
+#define CGTT_CPC_CLK_CTRL__SOFT_OVERRIDE_DYN_MASK 0x40000000
+#define CGTT_CPC_CLK_CTRL__SOFT_OVERRIDE_DYN__SHIFT 0x1e
+#define CGTT_CPC_CLK_CTRL__SOFT_OVERRIDE_REG_MASK 0x80000000
+#define CGTT_CPC_CLK_CTRL__SOFT_OVERRIDE_REG__SHIFT 0x1f
+#define CGTT_CPF_CLK_CTRL__ON_DELAY_MASK 0xf
+#define CGTT_CPF_CLK_CTRL__ON_DELAY__SHIFT 0x0
+#define CGTT_CPF_CLK_CTRL__OFF_HYSTERESIS_MASK 0xff0
+#define CGTT_CPF_CLK_CTRL__OFF_HYSTERESIS__SHIFT 0x4
+#define CGTT_CPF_CLK_CTRL__SOFT_OVERRIDE_PERFMON_MASK 0x20000000
+#define CGTT_CPF_CLK_CTRL__SOFT_OVERRIDE_PERFMON__SHIFT 0x1d
+#define CGTT_CPF_CLK_CTRL__SOFT_OVERRIDE_DYN_MASK 0x40000000
+#define CGTT_CPF_CLK_CTRL__SOFT_OVERRIDE_DYN__SHIFT 0x1e
+#define CGTT_CPF_CLK_CTRL__SOFT_OVERRIDE_REG_MASK 0x80000000
+#define CGTT_CPF_CLK_CTRL__SOFT_OVERRIDE_REG__SHIFT 0x1f
+#define CGTT_CP_CLK_CTRL__ON_DELAY_MASK 0xf
+#define CGTT_CP_CLK_CTRL__ON_DELAY__SHIFT 0x0
+#define CGTT_CP_CLK_CTRL__OFF_HYSTERESIS_MASK 0xff0
+#define CGTT_CP_CLK_CTRL__OFF_HYSTERESIS__SHIFT 0x4
+#define CGTT_CP_CLK_CTRL__SOFT_OVERRIDE_PERFMON_MASK 0x20000000
+#define CGTT_CP_CLK_CTRL__SOFT_OVERRIDE_PERFMON__SHIFT 0x1d
+#define CGTT_CP_CLK_CTRL__SOFT_OVERRIDE_DYN_MASK 0x40000000
+#define CGTT_CP_CLK_CTRL__SOFT_OVERRIDE_DYN__SHIFT 0x1e
+#define CGTT_CP_CLK_CTRL__SOFT_OVERRIDE_REG_MASK 0x80000000
+#define CGTT_CP_CLK_CTRL__SOFT_OVERRIDE_REG__SHIFT 0x1f
+#define CP_CE_UCODE_ADDR__UCODE_ADDR_MASK 0xfff
+#define CP_CE_UCODE_ADDR__UCODE_ADDR__SHIFT 0x0
+#define CP_CE_UCODE_DATA__UCODE_DATA_MASK 0xffffffff
+#define CP_CE_UCODE_DATA__UCODE_DATA__SHIFT 0x0
+#define CP_MEC_ME1_UCODE_ADDR__UCODE_ADDR_MASK 0x1ffff
+#define CP_MEC_ME1_UCODE_ADDR__UCODE_ADDR__SHIFT 0x0
+#define CP_MEC_ME1_UCODE_DATA__UCODE_DATA_MASK 0xffffffff
+#define CP_MEC_ME1_UCODE_DATA__UCODE_DATA__SHIFT 0x0
+#define CP_MEC_ME2_UCODE_ADDR__UCODE_ADDR_MASK 0x1ffff
+#define CP_MEC_ME2_UCODE_ADDR__UCODE_ADDR__SHIFT 0x0
+#define CP_MEC_ME2_UCODE_DATA__UCODE_DATA_MASK 0xffffffff
+#define CP_MEC_ME2_UCODE_DATA__UCODE_DATA__SHIFT 0x0
+#define CP_PFP_F32_INTERRUPT__PRIV_REG_INT_MASK 0x2
+#define CP_PFP_F32_INTERRUPT__PRIV_REG_INT__SHIFT 0x1
+#define CP_MEC1_F32_INTERRUPT__PRIV_REG_INT_MASK 0x2
+#define CP_MEC1_F32_INTERRUPT__PRIV_REG_INT__SHIFT 0x1
+#define CP_MEC2_F32_INTERRUPT__PRIV_REG_INT_MASK 0x2
+#define CP_MEC2_F32_INTERRUPT__PRIV_REG_INT__SHIFT 0x1
+#define CP_MEC1_F32_INT_DIS__EDC_ROQ_FED_INT_MASK 0x1
+#define CP_MEC1_F32_INT_DIS__EDC_ROQ_FED_INT__SHIFT 0x0
+#define CP_MEC1_F32_INT_DIS__PRIV_REG_INT_MASK 0x2
+#define CP_MEC1_F32_INT_DIS__PRIV_REG_INT__SHIFT 0x1
+#define CP_MEC1_F32_INT_DIS__RESERVED_BIT_ERR_INT_MASK 0x4
+#define CP_MEC1_F32_INT_DIS__RESERVED_BIT_ERR_INT__SHIFT 0x2
+#define CP_MEC1_F32_INT_DIS__EDC_TC_FED_INT_MASK 0x8
+#define CP_MEC1_F32_INT_DIS__EDC_TC_FED_INT__SHIFT 0x3
+#define CP_MEC1_F32_INT_DIS__EDC_GDS_FED_INT_MASK 0x10
+#define CP_MEC1_F32_INT_DIS__EDC_GDS_FED_INT__SHIFT 0x4
+#define CP_MEC1_F32_INT_DIS__EDC_SCRATCH_FED_INT_MASK 0x20
+#define CP_MEC1_F32_INT_DIS__EDC_SCRATCH_FED_INT__SHIFT 0x5
+#define CP_MEC1_F32_INT_DIS__WAVE_RESTORE_INT_MASK 0x40
+#define CP_MEC1_F32_INT_DIS__WAVE_RESTORE_INT__SHIFT 0x6
+#define CP_MEC1_F32_INT_DIS__SUA_VIOLATION_INT_MASK 0x80
+#define CP_MEC1_F32_INT_DIS__SUA_VIOLATION_INT__SHIFT 0x7
+#define CP_MEC1_F32_INT_DIS__EDC_DMA_FED_INT_MASK 0x100
+#define CP_MEC1_F32_INT_DIS__EDC_DMA_FED_INT__SHIFT 0x8
+#define CP_MEC1_F32_INT_DIS__IQ_TIMER_INT_MASK 0x200
+#define CP_MEC1_F32_INT_DIS__IQ_TIMER_INT__SHIFT 0x9
+#define CP_MEC2_F32_INT_DIS__EDC_ROQ_FED_INT_MASK 0x1
+#define CP_MEC2_F32_INT_DIS__EDC_ROQ_FED_INT__SHIFT 0x0
+#define CP_MEC2_F32_INT_DIS__PRIV_REG_INT_MASK 0x2
+#define CP_MEC2_F32_INT_DIS__PRIV_REG_INT__SHIFT 0x1
+#define CP_MEC2_F32_INT_DIS__RESERVED_BIT_ERR_INT_MASK 0x4
+#define CP_MEC2_F32_INT_DIS__RESERVED_BIT_ERR_INT__SHIFT 0x2
+#define CP_MEC2_F32_INT_DIS__EDC_TC_FED_INT_MASK 0x8
+#define CP_MEC2_F32_INT_DIS__EDC_TC_FED_INT__SHIFT 0x3
+#define CP_MEC2_F32_INT_DIS__EDC_GDS_FED_INT_MASK 0x10
+#define CP_MEC2_F32_INT_DIS__EDC_GDS_FED_INT__SHIFT 0x4
+#define CP_MEC2_F32_INT_DIS__EDC_SCRATCH_FED_INT_MASK 0x20
+#define CP_MEC2_F32_INT_DIS__EDC_SCRATCH_FED_INT__SHIFT 0x5
+#define CP_MEC2_F32_INT_DIS__WAVE_RESTORE_INT_MASK 0x40
+#define CP_MEC2_F32_INT_DIS__WAVE_RESTORE_INT__SHIFT 0x6
+#define CP_MEC2_F32_INT_DIS__SUA_VIOLATION_INT_MASK 0x80
+#define CP_MEC2_F32_INT_DIS__SUA_VIOLATION_INT__SHIFT 0x7
+#define CP_MEC2_F32_INT_DIS__EDC_DMA_FED_INT_MASK 0x100
+#define CP_MEC2_F32_INT_DIS__EDC_DMA_FED_INT__SHIFT 0x8
+#define CP_MEC2_F32_INT_DIS__IQ_TIMER_INT_MASK 0x200
+#define CP_MEC2_F32_INT_DIS__IQ_TIMER_INT__SHIFT 0x9
+#define CP_PWR_CNTL__GFX_CLK_HALT_ME0_PIPE0_MASK 0x1
+#define CP_PWR_CNTL__GFX_CLK_HALT_ME0_PIPE0__SHIFT 0x0
+#define CP_PWR_CNTL__GFX_CLK_HALT_ME0_PIPE1_MASK 0x2
+#define CP_PWR_CNTL__GFX_CLK_HALT_ME0_PIPE1__SHIFT 0x1
+#define CP_PWR_CNTL__CMP_CLK_HALT_ME1_PIPE0_MASK 0x100
+#define CP_PWR_CNTL__CMP_CLK_HALT_ME1_PIPE0__SHIFT 0x8
+#define CP_PWR_CNTL__CMP_CLK_HALT_ME1_PIPE1_MASK 0x200
+#define CP_PWR_CNTL__CMP_CLK_HALT_ME1_PIPE1__SHIFT 0x9
+#define CP_PWR_CNTL__CMP_CLK_HALT_ME1_PIPE2_MASK 0x400
+#define CP_PWR_CNTL__CMP_CLK_HALT_ME1_PIPE2__SHIFT 0xa
+#define CP_PWR_CNTL__CMP_CLK_HALT_ME1_PIPE3_MASK 0x800
+#define CP_PWR_CNTL__CMP_CLK_HALT_ME1_PIPE3__SHIFT 0xb
+#define CP_PWR_CNTL__CMP_CLK_HALT_ME2_PIPE0_MASK 0x10000
+#define CP_PWR_CNTL__CMP_CLK_HALT_ME2_PIPE0__SHIFT 0x10
+#define CP_PWR_CNTL__CMP_CLK_HALT_ME2_PIPE1_MASK 0x20000
+#define CP_PWR_CNTL__CMP_CLK_HALT_ME2_PIPE1__SHIFT 0x11
+#define CP_PWR_CNTL__CMP_CLK_HALT_ME2_PIPE2_MASK 0x40000
+#define CP_PWR_CNTL__CMP_CLK_HALT_ME2_PIPE2__SHIFT 0x12
+#define CP_PWR_CNTL__CMP_CLK_HALT_ME2_PIPE3_MASK 0x80000
+#define CP_PWR_CNTL__CMP_CLK_HALT_ME2_PIPE3__SHIFT 0x13
+#define CP_MEM_SLP_CNTL__CP_MEM_LS_EN_MASK 0x1
+#define CP_MEM_SLP_CNTL__CP_MEM_LS_EN__SHIFT 0x0
+#define CP_MEM_SLP_CNTL__CP_MEM_DS_EN_MASK 0x2
+#define CP_MEM_SLP_CNTL__CP_MEM_DS_EN__SHIFT 0x1
+#define CP_MEM_SLP_CNTL__RESERVED_MASK 0x7c
+#define CP_MEM_SLP_CNTL__RESERVED__SHIFT 0x2
+#define CP_MEM_SLP_CNTL__CP_LS_DS_BUSY_OVERRIDE_MASK 0x80
+#define CP_MEM_SLP_CNTL__CP_LS_DS_BUSY_OVERRIDE__SHIFT 0x7
+#define CP_MEM_SLP_CNTL__CP_MEM_LS_ON_DELAY_MASK 0xff00
+#define CP_MEM_SLP_CNTL__CP_MEM_LS_ON_DELAY__SHIFT 0x8
+#define CP_MEM_SLP_CNTL__CP_MEM_LS_OFF_DELAY_MASK 0xff0000
+#define CP_MEM_SLP_CNTL__CP_MEM_LS_OFF_DELAY__SHIFT 0x10
+#define CP_MEM_SLP_CNTL__RESERVED1_MASK 0xff000000
+#define CP_MEM_SLP_CNTL__RESERVED1__SHIFT 0x18
+#define CP_ECC_FIRSTOCCURRENCE__INTERFACE_MASK 0x3
+#define CP_ECC_FIRSTOCCURRENCE__INTERFACE__SHIFT 0x0
+#define CP_ECC_FIRSTOCCURRENCE__CLIENT_MASK 0xf0
+#define CP_ECC_FIRSTOCCURRENCE__CLIENT__SHIFT 0x4
+#define CP_ECC_FIRSTOCCURRENCE__ME_MASK 0x300
+#define CP_ECC_FIRSTOCCURRENCE__ME__SHIFT 0x8
+#define CP_ECC_FIRSTOCCURRENCE__PIPE_MASK 0xc00
+#define CP_ECC_FIRSTOCCURRENCE__PIPE__SHIFT 0xa
+#define CP_ECC_FIRSTOCCURRENCE__QUEUE_MASK 0x7000
+#define CP_ECC_FIRSTOCCURRENCE__QUEUE__SHIFT 0xc
+#define CP_ECC_FIRSTOCCURRENCE__VMID_MASK 0xf0000
+#define CP_ECC_FIRSTOCCURRENCE__VMID__SHIFT 0x10
+#define CP_ECC_FIRSTOCCURRENCE_RING0__OBSOLETE_MASK 0xffffffff
+#define CP_ECC_FIRSTOCCURRENCE_RING0__OBSOLETE__SHIFT 0x0
+#define CP_ECC_FIRSTOCCURRENCE_RING1__OBSOLETE_MASK 0xffffffff
+#define CP_ECC_FIRSTOCCURRENCE_RING1__OBSOLETE__SHIFT 0x0
+#define CP_ECC_FIRSTOCCURRENCE_RING2__OBSOLETE_MASK 0xffffffff
+#define CP_ECC_FIRSTOCCURRENCE_RING2__OBSOLETE__SHIFT 0x0
+#define CP_PQ_WPTR_POLL_CNTL__PERIOD_MASK 0xff
+#define CP_PQ_WPTR_POLL_CNTL__PERIOD__SHIFT 0x0
+#define CP_PQ_WPTR_POLL_CNTL__POLL_ACTIVE_MASK 0x40000000
+#define CP_PQ_WPTR_POLL_CNTL__POLL_ACTIVE__SHIFT 0x1e
+#define CP_PQ_WPTR_POLL_CNTL__EN_MASK 0x80000000
+#define CP_PQ_WPTR_POLL_CNTL__EN__SHIFT 0x1f
+#define CP_PQ_WPTR_POLL_CNTL1__QUEUE_MASK_MASK 0xffffffff
+#define CP_PQ_WPTR_POLL_CNTL1__QUEUE_MASK__SHIFT 0x0
+#define CPC_INT_CNTL__CMP_QUERY_STATUS_INT_ENABLE_MASK 0x1000
+#define CPC_INT_CNTL__CMP_QUERY_STATUS_INT_ENABLE__SHIFT 0xc
+#define CPC_INT_CNTL__DEQUEUE_REQUEST_INT_ENABLE_MASK 0x2000
+#define CPC_INT_CNTL__DEQUEUE_REQUEST_INT_ENABLE__SHIFT 0xd
+#define CPC_INT_CNTL__CP_ECC_ERROR_INT_ENABLE_MASK 0x4000
+#define CPC_INT_CNTL__CP_ECC_ERROR_INT_ENABLE__SHIFT 0xe
+#define CPC_INT_CNTL__SUA_VIOLATION_INT_ENABLE_MASK 0x8000
+#define CPC_INT_CNTL__SUA_VIOLATION_INT_ENABLE__SHIFT 0xf
+#define CPC_INT_CNTL__WRM_POLL_TIMEOUT_INT_ENABLE_MASK 0x20000
+#define CPC_INT_CNTL__WRM_POLL_TIMEOUT_INT_ENABLE__SHIFT 0x11
+#define CPC_INT_CNTL__PRIV_REG_INT_ENABLE_MASK 0x800000
+#define CPC_INT_CNTL__PRIV_REG_INT_ENABLE__SHIFT 0x17
+#define CPC_INT_CNTL__OPCODE_ERROR_INT_ENABLE_MASK 0x1000000
+#define CPC_INT_CNTL__OPCODE_ERROR_INT_ENABLE__SHIFT 0x18
+#define CPC_INT_CNTL__TIME_STAMP_INT_ENABLE_MASK 0x4000000
+#define CPC_INT_CNTL__TIME_STAMP_INT_ENABLE__SHIFT 0x1a
+#define CPC_INT_CNTL__RESERVED_BIT_ERROR_INT_ENABLE_MASK 0x8000000
+#define CPC_INT_CNTL__RESERVED_BIT_ERROR_INT_ENABLE__SHIFT 0x1b
+#define CPC_INT_CNTL__GENERIC2_INT_ENABLE_MASK 0x20000000
+#define CPC_INT_CNTL__GENERIC2_INT_ENABLE__SHIFT 0x1d
+#define CPC_INT_CNTL__GENERIC1_INT_ENABLE_MASK 0x40000000
+#define CPC_INT_CNTL__GENERIC1_INT_ENABLE__SHIFT 0x1e
+#define CPC_INT_CNTL__GENERIC0_INT_ENABLE_MASK 0x80000000
+#define CPC_INT_CNTL__GENERIC0_INT_ENABLE__SHIFT 0x1f
+#define CP_ME1_PIPE0_INT_CNTL__CMP_QUERY_STATUS_INT_ENABLE_MASK 0x1000
+#define CP_ME1_PIPE0_INT_CNTL__CMP_QUERY_STATUS_INT_ENABLE__SHIFT 0xc
+#define CP_ME1_PIPE0_INT_CNTL__DEQUEUE_REQUEST_INT_ENABLE_MASK 0x2000
+#define CP_ME1_PIPE0_INT_CNTL__DEQUEUE_REQUEST_INT_ENABLE__SHIFT 0xd
+#define CP_ME1_PIPE0_INT_CNTL__CP_ECC_ERROR_INT_ENABLE_MASK 0x4000
+#define CP_ME1_PIPE0_INT_CNTL__CP_ECC_ERROR_INT_ENABLE__SHIFT 0xe
+#define CP_ME1_PIPE0_INT_CNTL__SUA_VIOLATION_INT_ENABLE_MASK 0x8000
+#define CP_ME1_PIPE0_INT_CNTL__SUA_VIOLATION_INT_ENABLE__SHIFT 0xf
+#define CP_ME1_PIPE0_INT_CNTL__WRM_POLL_TIMEOUT_INT_ENABLE_MASK 0x20000
+#define CP_ME1_PIPE0_INT_CNTL__WRM_POLL_TIMEOUT_INT_ENABLE__SHIFT 0x11
+#define CP_ME1_PIPE0_INT_CNTL__PRIV_REG_INT_ENABLE_MASK 0x800000
+#define CP_ME1_PIPE0_INT_CNTL__PRIV_REG_INT_ENABLE__SHIFT 0x17
+#define CP_ME1_PIPE0_INT_CNTL__OPCODE_ERROR_INT_ENABLE_MASK 0x1000000
+#define CP_ME1_PIPE0_INT_CNTL__OPCODE_ERROR_INT_ENABLE__SHIFT 0x18
+#define CP_ME1_PIPE0_INT_CNTL__TIME_STAMP_INT_ENABLE_MASK 0x4000000
+#define CP_ME1_PIPE0_INT_CNTL__TIME_STAMP_INT_ENABLE__SHIFT 0x1a
+#define CP_ME1_PIPE0_INT_CNTL__RESERVED_BIT_ERROR_INT_ENABLE_MASK 0x8000000
+#define CP_ME1_PIPE0_INT_CNTL__RESERVED_BIT_ERROR_INT_ENABLE__SHIFT 0x1b
+#define CP_ME1_PIPE0_INT_CNTL__GENERIC2_INT_ENABLE_MASK 0x20000000
+#define CP_ME1_PIPE0_INT_CNTL__GENERIC2_INT_ENABLE__SHIFT 0x1d
+#define CP_ME1_PIPE0_INT_CNTL__GENERIC1_INT_ENABLE_MASK 0x40000000
+#define CP_ME1_PIPE0_INT_CNTL__GENERIC1_INT_ENABLE__SHIFT 0x1e
+#define CP_ME1_PIPE0_INT_CNTL__GENERIC0_INT_ENABLE_MASK 0x80000000
+#define CP_ME1_PIPE0_INT_CNTL__GENERIC0_INT_ENABLE__SHIFT 0x1f
+#define CP_ME1_PIPE1_INT_CNTL__CMP_QUERY_STATUS_INT_ENABLE_MASK 0x1000
+#define CP_ME1_PIPE1_INT_CNTL__CMP_QUERY_STATUS_INT_ENABLE__SHIFT 0xc
+#define CP_ME1_PIPE1_INT_CNTL__DEQUEUE_REQUEST_INT_ENABLE_MASK 0x2000
+#define CP_ME1_PIPE1_INT_CNTL__DEQUEUE_REQUEST_INT_ENABLE__SHIFT 0xd
+#define CP_ME1_PIPE1_INT_CNTL__CP_ECC_ERROR_INT_ENABLE_MASK 0x4000
+#define CP_ME1_PIPE1_INT_CNTL__CP_ECC_ERROR_INT_ENABLE__SHIFT 0xe
+#define CP_ME1_PIPE1_INT_CNTL__SUA_VIOLATION_INT_ENABLE_MASK 0x8000
+#define CP_ME1_PIPE1_INT_CNTL__SUA_VIOLATION_INT_ENABLE__SHIFT 0xf
+#define CP_ME1_PIPE1_INT_CNTL__WRM_POLL_TIMEOUT_INT_ENABLE_MASK 0x20000
+#define CP_ME1_PIPE1_INT_CNTL__WRM_POLL_TIMEOUT_INT_ENABLE__SHIFT 0x11
+#define CP_ME1_PIPE1_INT_CNTL__PRIV_REG_INT_ENABLE_MASK 0x800000
+#define CP_ME1_PIPE1_INT_CNTL__PRIV_REG_INT_ENABLE__SHIFT 0x17
+#define CP_ME1_PIPE1_INT_CNTL__OPCODE_ERROR_INT_ENABLE_MASK 0x1000000
+#define CP_ME1_PIPE1_INT_CNTL__OPCODE_ERROR_INT_ENABLE__SHIFT 0x18
+#define CP_ME1_PIPE1_INT_CNTL__TIME_STAMP_INT_ENABLE_MASK 0x4000000
+#define CP_ME1_PIPE1_INT_CNTL__TIME_STAMP_INT_ENABLE__SHIFT 0x1a
+#define CP_ME1_PIPE1_INT_CNTL__RESERVED_BIT_ERROR_INT_ENABLE_MASK 0x8000000
+#define CP_ME1_PIPE1_INT_CNTL__RESERVED_BIT_ERROR_INT_ENABLE__SHIFT 0x1b
+#define CP_ME1_PIPE1_INT_CNTL__GENERIC2_INT_ENABLE_MASK 0x20000000
+#define CP_ME1_PIPE1_INT_CNTL__GENERIC2_INT_ENABLE__SHIFT 0x1d
+#define CP_ME1_PIPE1_INT_CNTL__GENERIC1_INT_ENABLE_MASK 0x40000000
+#define CP_ME1_PIPE1_INT_CNTL__GENERIC1_INT_ENABLE__SHIFT 0x1e
+#define CP_ME1_PIPE1_INT_CNTL__GENERIC0_INT_ENABLE_MASK 0x80000000
+#define CP_ME1_PIPE1_INT_CNTL__GENERIC0_INT_ENABLE__SHIFT 0x1f
+#define CP_ME1_PIPE2_INT_CNTL__CMP_QUERY_STATUS_INT_ENABLE_MASK 0x1000
+#define CP_ME1_PIPE2_INT_CNTL__CMP_QUERY_STATUS_INT_ENABLE__SHIFT 0xc
+#define CP_ME1_PIPE2_INT_CNTL__DEQUEUE_REQUEST_INT_ENABLE_MASK 0x2000
+#define CP_ME1_PIPE2_INT_CNTL__DEQUEUE_REQUEST_INT_ENABLE__SHIFT 0xd
+#define CP_ME1_PIPE2_INT_CNTL__CP_ECC_ERROR_INT_ENABLE_MASK 0x4000
+#define CP_ME1_PIPE2_INT_CNTL__CP_ECC_ERROR_INT_ENABLE__SHIFT 0xe
+#define CP_ME1_PIPE2_INT_CNTL__SUA_VIOLATION_INT_ENABLE_MASK 0x8000
+#define CP_ME1_PIPE2_INT_CNTL__SUA_VIOLATION_INT_ENABLE__SHIFT 0xf
+#define CP_ME1_PIPE2_INT_CNTL__WRM_POLL_TIMEOUT_INT_ENABLE_MASK 0x20000
+#define CP_ME1_PIPE2_INT_CNTL__WRM_POLL_TIMEOUT_INT_ENABLE__SHIFT 0x11
+#define CP_ME1_PIPE2_INT_CNTL__PRIV_REG_INT_ENABLE_MASK 0x800000
+#define CP_ME1_PIPE2_INT_CNTL__PRIV_REG_INT_ENABLE__SHIFT 0x17
+#define CP_ME1_PIPE2_INT_CNTL__OPCODE_ERROR_INT_ENABLE_MASK 0x1000000
+#define CP_ME1_PIPE2_INT_CNTL__OPCODE_ERROR_INT_ENABLE__SHIFT 0x18
+#define CP_ME1_PIPE2_INT_CNTL__TIME_STAMP_INT_ENABLE_MASK 0x4000000
+#define CP_ME1_PIPE2_INT_CNTL__TIME_STAMP_INT_ENABLE__SHIFT 0x1a
+#define CP_ME1_PIPE2_INT_CNTL__RESERVED_BIT_ERROR_INT_ENABLE_MASK 0x8000000
+#define CP_ME1_PIPE2_INT_CNTL__RESERVED_BIT_ERROR_INT_ENABLE__SHIFT 0x1b
+#define CP_ME1_PIPE2_INT_CNTL__GENERIC2_INT_ENABLE_MASK 0x20000000
+#define CP_ME1_PIPE2_INT_CNTL__GENERIC2_INT_ENABLE__SHIFT 0x1d
+#define CP_ME1_PIPE2_INT_CNTL__GENERIC1_INT_ENABLE_MASK 0x40000000
+#define CP_ME1_PIPE2_INT_CNTL__GENERIC1_INT_ENABLE__SHIFT 0x1e
+#define CP_ME1_PIPE2_INT_CNTL__GENERIC0_INT_ENABLE_MASK 0x80000000
+#define CP_ME1_PIPE2_INT_CNTL__GENERIC0_INT_ENABLE__SHIFT 0x1f
+#define CP_ME1_PIPE3_INT_CNTL__CMP_QUERY_STATUS_INT_ENABLE_MASK 0x1000
+#define CP_ME1_PIPE3_INT_CNTL__CMP_QUERY_STATUS_INT_ENABLE__SHIFT 0xc
+#define CP_ME1_PIPE3_INT_CNTL__DEQUEUE_REQUEST_INT_ENABLE_MASK 0x2000
+#define CP_ME1_PIPE3_INT_CNTL__DEQUEUE_REQUEST_INT_ENABLE__SHIFT 0xd
+#define CP_ME1_PIPE3_INT_CNTL__CP_ECC_ERROR_INT_ENABLE_MASK 0x4000
+#define CP_ME1_PIPE3_INT_CNTL__CP_ECC_ERROR_INT_ENABLE__SHIFT 0xe
+#define CP_ME1_PIPE3_INT_CNTL__SUA_VIOLATION_INT_ENABLE_MASK 0x8000
+#define CP_ME1_PIPE3_INT_CNTL__SUA_VIOLATION_INT_ENABLE__SHIFT 0xf
+#define CP_ME1_PIPE3_INT_CNTL__WRM_POLL_TIMEOUT_INT_ENABLE_MASK 0x20000
+#define CP_ME1_PIPE3_INT_CNTL__WRM_POLL_TIMEOUT_INT_ENABLE__SHIFT 0x11
+#define CP_ME1_PIPE3_INT_CNTL__PRIV_REG_INT_ENABLE_MASK 0x800000
+#define CP_ME1_PIPE3_INT_CNTL__PRIV_REG_INT_ENABLE__SHIFT 0x17
+#define CP_ME1_PIPE3_INT_CNTL__OPCODE_ERROR_INT_ENABLE_MASK 0x1000000
+#define CP_ME1_PIPE3_INT_CNTL__OPCODE_ERROR_INT_ENABLE__SHIFT 0x18
+#define CP_ME1_PIPE3_INT_CNTL__TIME_STAMP_INT_ENABLE_MASK 0x4000000
+#define CP_ME1_PIPE3_INT_CNTL__TIME_STAMP_INT_ENABLE__SHIFT 0x1a
+#define CP_ME1_PIPE3_INT_CNTL__RESERVED_BIT_ERROR_INT_ENABLE_MASK 0x8000000
+#define CP_ME1_PIPE3_INT_CNTL__RESERVED_BIT_ERROR_INT_ENABLE__SHIFT 0x1b
+#define CP_ME1_PIPE3_INT_CNTL__GENERIC2_INT_ENABLE_MASK 0x20000000
+#define CP_ME1_PIPE3_INT_CNTL__GENERIC2_INT_ENABLE__SHIFT 0x1d
+#define CP_ME1_PIPE3_INT_CNTL__GENERIC1_INT_ENABLE_MASK 0x40000000
+#define CP_ME1_PIPE3_INT_CNTL__GENERIC1_INT_ENABLE__SHIFT 0x1e
+#define CP_ME1_PIPE3_INT_CNTL__GENERIC0_INT_ENABLE_MASK 0x80000000
+#define CP_ME1_PIPE3_INT_CNTL__GENERIC0_INT_ENABLE__SHIFT 0x1f
+#define CP_ME2_PIPE0_INT_CNTL__CMP_QUERY_STATUS_INT_ENABLE_MASK 0x1000
+#define CP_ME2_PIPE0_INT_CNTL__CMP_QUERY_STATUS_INT_ENABLE__SHIFT 0xc
+#define CP_ME2_PIPE0_INT_CNTL__DEQUEUE_REQUEST_INT_ENABLE_MASK 0x2000
+#define CP_ME2_PIPE0_INT_CNTL__DEQUEUE_REQUEST_INT_ENABLE__SHIFT 0xd
+#define CP_ME2_PIPE0_INT_CNTL__CP_ECC_ERROR_INT_ENABLE_MASK 0x4000
+#define CP_ME2_PIPE0_INT_CNTL__CP_ECC_ERROR_INT_ENABLE__SHIFT 0xe
+#define CP_ME2_PIPE0_INT_CNTL__SUA_VIOLATION_INT_ENABLE_MASK 0x8000
+#define CP_ME2_PIPE0_INT_CNTL__SUA_VIOLATION_INT_ENABLE__SHIFT 0xf
+#define CP_ME2_PIPE0_INT_CNTL__WRM_POLL_TIMEOUT_INT_ENABLE_MASK 0x20000
+#define CP_ME2_PIPE0_INT_CNTL__WRM_POLL_TIMEOUT_INT_ENABLE__SHIFT 0x11
+#define CP_ME2_PIPE0_INT_CNTL__PRIV_REG_INT_ENABLE_MASK 0x800000
+#define CP_ME2_PIPE0_INT_CNTL__PRIV_REG_INT_ENABLE__SHIFT 0x17
+#define CP_ME2_PIPE0_INT_CNTL__OPCODE_ERROR_INT_ENABLE_MASK 0x1000000
+#define CP_ME2_PIPE0_INT_CNTL__OPCODE_ERROR_INT_ENABLE__SHIFT 0x18
+#define CP_ME2_PIPE0_INT_CNTL__TIME_STAMP_INT_ENABLE_MASK 0x4000000
+#define CP_ME2_PIPE0_INT_CNTL__TIME_STAMP_INT_ENABLE__SHIFT 0x1a
+#define CP_ME2_PIPE0_INT_CNTL__RESERVED_BIT_ERROR_INT_ENABLE_MASK 0x8000000
+#define CP_ME2_PIPE0_INT_CNTL__RESERVED_BIT_ERROR_INT_ENABLE__SHIFT 0x1b
+#define CP_ME2_PIPE0_INT_CNTL__GENERIC2_INT_ENABLE_MASK 0x20000000
+#define CP_ME2_PIPE0_INT_CNTL__GENERIC2_INT_ENABLE__SHIFT 0x1d
+#define CP_ME2_PIPE0_INT_CNTL__GENERIC1_INT_ENABLE_MASK 0x40000000
+#define CP_ME2_PIPE0_INT_CNTL__GENERIC1_INT_ENABLE__SHIFT 0x1e
+#define CP_ME2_PIPE0_INT_CNTL__GENERIC0_INT_ENABLE_MASK 0x80000000
+#define CP_ME2_PIPE0_INT_CNTL__GENERIC0_INT_ENABLE__SHIFT 0x1f
+#define CP_ME2_PIPE1_INT_CNTL__CMP_QUERY_STATUS_INT_ENABLE_MASK 0x1000
+#define CP_ME2_PIPE1_INT_CNTL__CMP_QUERY_STATUS_INT_ENABLE__SHIFT 0xc
+#define CP_ME2_PIPE1_INT_CNTL__DEQUEUE_REQUEST_INT_ENABLE_MASK 0x2000
+#define CP_ME2_PIPE1_INT_CNTL__DEQUEUE_REQUEST_INT_ENABLE__SHIFT 0xd
+#define CP_ME2_PIPE1_INT_CNTL__CP_ECC_ERROR_INT_ENABLE_MASK 0x4000
+#define CP_ME2_PIPE1_INT_CNTL__CP_ECC_ERROR_INT_ENABLE__SHIFT 0xe
+#define CP_ME2_PIPE1_INT_CNTL__SUA_VIOLATION_INT_ENABLE_MASK 0x8000
+#define CP_ME2_PIPE1_INT_CNTL__SUA_VIOLATION_INT_ENABLE__SHIFT 0xf
+#define CP_ME2_PIPE1_INT_CNTL__WRM_POLL_TIMEOUT_INT_ENABLE_MASK 0x20000
+#define CP_ME2_PIPE1_INT_CNTL__WRM_POLL_TIMEOUT_INT_ENABLE__SHIFT 0x11
+#define CP_ME2_PIPE1_INT_CNTL__PRIV_REG_INT_ENABLE_MASK 0x800000
+#define CP_ME2_PIPE1_INT_CNTL__PRIV_REG_INT_ENABLE__SHIFT 0x17
+#define CP_ME2_PIPE1_INT_CNTL__OPCODE_ERROR_INT_ENABLE_MASK 0x1000000
+#define CP_ME2_PIPE1_INT_CNTL__OPCODE_ERROR_INT_ENABLE__SHIFT 0x18
+#define CP_ME2_PIPE1_INT_CNTL__TIME_STAMP_INT_ENABLE_MASK 0x4000000
+#define CP_ME2_PIPE1_INT_CNTL__TIME_STAMP_INT_ENABLE__SHIFT 0x1a
+#define CP_ME2_PIPE1_INT_CNTL__RESERVED_BIT_ERROR_INT_ENABLE_MASK 0x8000000
+#define CP_ME2_PIPE1_INT_CNTL__RESERVED_BIT_ERROR_INT_ENABLE__SHIFT 0x1b
+#define CP_ME2_PIPE1_INT_CNTL__GENERIC2_INT_ENABLE_MASK 0x20000000
+#define CP_ME2_PIPE1_INT_CNTL__GENERIC2_INT_ENABLE__SHIFT 0x1d
+#define CP_ME2_PIPE1_INT_CNTL__GENERIC1_INT_ENABLE_MASK 0x40000000
+#define CP_ME2_PIPE1_INT_CNTL__GENERIC1_INT_ENABLE__SHIFT 0x1e
+#define CP_ME2_PIPE1_INT_CNTL__GENERIC0_INT_ENABLE_MASK 0x80000000
+#define CP_ME2_PIPE1_INT_CNTL__GENERIC0_INT_ENABLE__SHIFT 0x1f
+#define CP_ME2_PIPE2_INT_CNTL__CMP_QUERY_STATUS_INT_ENABLE_MASK 0x1000
+#define CP_ME2_PIPE2_INT_CNTL__CMP_QUERY_STATUS_INT_ENABLE__SHIFT 0xc
+#define CP_ME2_PIPE2_INT_CNTL__DEQUEUE_REQUEST_INT_ENABLE_MASK 0x2000
+#define CP_ME2_PIPE2_INT_CNTL__DEQUEUE_REQUEST_INT_ENABLE__SHIFT 0xd
+#define CP_ME2_PIPE2_INT_CNTL__CP_ECC_ERROR_INT_ENABLE_MASK 0x4000
+#define CP_ME2_PIPE2_INT_CNTL__CP_ECC_ERROR_INT_ENABLE__SHIFT 0xe
+#define CP_ME2_PIPE2_INT_CNTL__SUA_VIOLATION_INT_ENABLE_MASK 0x8000
+#define CP_ME2_PIPE2_INT_CNTL__SUA_VIOLATION_INT_ENABLE__SHIFT 0xf
+#define CP_ME2_PIPE2_INT_CNTL__WRM_POLL_TIMEOUT_INT_ENABLE_MASK 0x20000
+#define CP_ME2_PIPE2_INT_CNTL__WRM_POLL_TIMEOUT_INT_ENABLE__SHIFT 0x11
+#define CP_ME2_PIPE2_INT_CNTL__PRIV_REG_INT_ENABLE_MASK 0x800000
+#define CP_ME2_PIPE2_INT_CNTL__PRIV_REG_INT_ENABLE__SHIFT 0x17
+#define CP_ME2_PIPE2_INT_CNTL__OPCODE_ERROR_INT_ENABLE_MASK 0x1000000
+#define CP_ME2_PIPE2_INT_CNTL__OPCODE_ERROR_INT_ENABLE__SHIFT 0x18
+#define CP_ME2_PIPE2_INT_CNTL__TIME_STAMP_INT_ENABLE_MASK 0x4000000
+#define CP_ME2_PIPE2_INT_CNTL__TIME_STAMP_INT_ENABLE__SHIFT 0x1a
+#define CP_ME2_PIPE2_INT_CNTL__RESERVED_BIT_ERROR_INT_ENABLE_MASK 0x8000000
+#define CP_ME2_PIPE2_INT_CNTL__RESERVED_BIT_ERROR_INT_ENABLE__SHIFT 0x1b
+#define CP_ME2_PIPE2_INT_CNTL__GENERIC2_INT_ENABLE_MASK 0x20000000
+#define CP_ME2_PIPE2_INT_CNTL__GENERIC2_INT_ENABLE__SHIFT 0x1d
+#define CP_ME2_PIPE2_INT_CNTL__GENERIC1_INT_ENABLE_MASK 0x40000000
+#define CP_ME2_PIPE2_INT_CNTL__GENERIC1_INT_ENABLE__SHIFT 0x1e
+#define CP_ME2_PIPE2_INT_CNTL__GENERIC0_INT_ENABLE_MASK 0x80000000
+#define CP_ME2_PIPE2_INT_CNTL__GENERIC0_INT_ENABLE__SHIFT 0x1f
+#define CP_ME2_PIPE3_INT_CNTL__CMP_QUERY_STATUS_INT_ENABLE_MASK 0x1000
+#define CP_ME2_PIPE3_INT_CNTL__CMP_QUERY_STATUS_INT_ENABLE__SHIFT 0xc
+#define CP_ME2_PIPE3_INT_CNTL__DEQUEUE_REQUEST_INT_ENABLE_MASK 0x2000
+#define CP_ME2_PIPE3_INT_CNTL__DEQUEUE_REQUEST_INT_ENABLE__SHIFT 0xd
+#define CP_ME2_PIPE3_INT_CNTL__CP_ECC_ERROR_INT_ENABLE_MASK 0x4000
+#define CP_ME2_PIPE3_INT_CNTL__CP_ECC_ERROR_INT_ENABLE__SHIFT 0xe
+#define CP_ME2_PIPE3_INT_CNTL__SUA_VIOLATION_INT_ENABLE_MASK 0x8000
+#define CP_ME2_PIPE3_INT_CNTL__SUA_VIOLATION_INT_ENABLE__SHIFT 0xf
+#define CP_ME2_PIPE3_INT_CNTL__WRM_POLL_TIMEOUT_INT_ENABLE_MASK 0x20000
+#define CP_ME2_PIPE3_INT_CNTL__WRM_POLL_TIMEOUT_INT_ENABLE__SHIFT 0x11
+#define CP_ME2_PIPE3_INT_CNTL__PRIV_REG_INT_ENABLE_MASK 0x800000
+#define CP_ME2_PIPE3_INT_CNTL__PRIV_REG_INT_ENABLE__SHIFT 0x17
+#define CP_ME2_PIPE3_INT_CNTL__OPCODE_ERROR_INT_ENABLE_MASK 0x1000000
+#define CP_ME2_PIPE3_INT_CNTL__OPCODE_ERROR_INT_ENABLE__SHIFT 0x18
+#define CP_ME2_PIPE3_INT_CNTL__TIME_STAMP_INT_ENABLE_MASK 0x4000000
+#define CP_ME2_PIPE3_INT_CNTL__TIME_STAMP_INT_ENABLE__SHIFT 0x1a
+#define CP_ME2_PIPE3_INT_CNTL__RESERVED_BIT_ERROR_INT_ENABLE_MASK 0x8000000
+#define CP_ME2_PIPE3_INT_CNTL__RESERVED_BIT_ERROR_INT_ENABLE__SHIFT 0x1b
+#define CP_ME2_PIPE3_INT_CNTL__GENERIC2_INT_ENABLE_MASK 0x20000000
+#define CP_ME2_PIPE3_INT_CNTL__GENERIC2_INT_ENABLE__SHIFT 0x1d
+#define CP_ME2_PIPE3_INT_CNTL__GENERIC1_INT_ENABLE_MASK 0x40000000
+#define CP_ME2_PIPE3_INT_CNTL__GENERIC1_INT_ENABLE__SHIFT 0x1e
+#define CP_ME2_PIPE3_INT_CNTL__GENERIC0_INT_ENABLE_MASK 0x80000000
+#define CP_ME2_PIPE3_INT_CNTL__GENERIC0_INT_ENABLE__SHIFT 0x1f
+#define CPC_INT_STATUS__CMP_QUERY_STATUS_INT_STATUS_MASK 0x1000
+#define CPC_INT_STATUS__CMP_QUERY_STATUS_INT_STATUS__SHIFT 0xc
+#define CPC_INT_STATUS__DEQUEUE_REQUEST_INT_STATUS_MASK 0x2000
+#define CPC_INT_STATUS__DEQUEUE_REQUEST_INT_STATUS__SHIFT 0xd
+#define CPC_INT_STATUS__CP_ECC_ERROR_INT_STATUS_MASK 0x4000
+#define CPC_INT_STATUS__CP_ECC_ERROR_INT_STATUS__SHIFT 0xe
+#define CPC_INT_STATUS__SUA_VIOLATION_INT_STATUS_MASK 0x8000
+#define CPC_INT_STATUS__SUA_VIOLATION_INT_STATUS__SHIFT 0xf
+#define CPC_INT_STATUS__WRM_POLL_TIMEOUT_INT_STATUS_MASK 0x20000
+#define CPC_INT_STATUS__WRM_POLL_TIMEOUT_INT_STATUS__SHIFT 0x11
+#define CPC_INT_STATUS__PRIV_REG_INT_STATUS_MASK 0x800000
+#define CPC_INT_STATUS__PRIV_REG_INT_STATUS__SHIFT 0x17
+#define CPC_INT_STATUS__OPCODE_ERROR_INT_STATUS_MASK 0x1000000
+#define CPC_INT_STATUS__OPCODE_ERROR_INT_STATUS__SHIFT 0x18
+#define CPC_INT_STATUS__TIME_STAMP_INT_STATUS_MASK 0x4000000
+#define CPC_INT_STATUS__TIME_STAMP_INT_STATUS__SHIFT 0x1a
+#define CPC_INT_STATUS__RESERVED_BIT_ERROR_INT_STATUS_MASK 0x8000000
+#define CPC_INT_STATUS__RESERVED_BIT_ERROR_INT_STATUS__SHIFT 0x1b
+#define CPC_INT_STATUS__GENERIC2_INT_STATUS_MASK 0x20000000
+#define CPC_INT_STATUS__GENERIC2_INT_STATUS__SHIFT 0x1d
+#define CPC_INT_STATUS__GENERIC1_INT_STATUS_MASK 0x40000000
+#define CPC_INT_STATUS__GENERIC1_INT_STATUS__SHIFT 0x1e
+#define CPC_INT_STATUS__GENERIC0_INT_STATUS_MASK 0x80000000
+#define CPC_INT_STATUS__GENERIC0_INT_STATUS__SHIFT 0x1f
+#define CP_ME1_PIPE0_INT_STATUS__CMP_QUERY_STATUS_INT_STATUS_MASK 0x1000
+#define CP_ME1_PIPE0_INT_STATUS__CMP_QUERY_STATUS_INT_STATUS__SHIFT 0xc
+#define CP_ME1_PIPE0_INT_STATUS__DEQUEUE_REQUEST_INT_STATUS_MASK 0x2000
+#define CP_ME1_PIPE0_INT_STATUS__DEQUEUE_REQUEST_INT_STATUS__SHIFT 0xd
+#define CP_ME1_PIPE0_INT_STATUS__CP_ECC_ERROR_INT_STATUS_MASK 0x4000
+#define CP_ME1_PIPE0_INT_STATUS__CP_ECC_ERROR_INT_STATUS__SHIFT 0xe
+#define CP_ME1_PIPE0_INT_STATUS__SUA_VIOLATION_INT_STATUS_MASK 0x8000
+#define CP_ME1_PIPE0_INT_STATUS__SUA_VIOLATION_INT_STATUS__SHIFT 0xf
+#define CP_ME1_PIPE0_INT_STATUS__WRM_POLL_TIMEOUT_INT_STATUS_MASK 0x20000
+#define CP_ME1_PIPE0_INT_STATUS__WRM_POLL_TIMEOUT_INT_STATUS__SHIFT 0x11
+#define CP_ME1_PIPE0_INT_STATUS__PRIV_REG_INT_STATUS_MASK 0x800000
+#define CP_ME1_PIPE0_INT_STATUS__PRIV_REG_INT_STATUS__SHIFT 0x17
+#define CP_ME1_PIPE0_INT_STATUS__OPCODE_ERROR_INT_STATUS_MASK 0x1000000
+#define CP_ME1_PIPE0_INT_STATUS__OPCODE_ERROR_INT_STATUS__SHIFT 0x18
+#define CP_ME1_PIPE0_INT_STATUS__TIME_STAMP_INT_STATUS_MASK 0x4000000
+#define CP_ME1_PIPE0_INT_STATUS__TIME_STAMP_INT_STATUS__SHIFT 0x1a
+#define CP_ME1_PIPE0_INT_STATUS__RESERVED_BIT_ERROR_INT_STATUS_MASK 0x8000000
+#define CP_ME1_PIPE0_INT_STATUS__RESERVED_BIT_ERROR_INT_STATUS__SHIFT 0x1b
+#define CP_ME1_PIPE0_INT_STATUS__GENERIC2_INT_STATUS_MASK 0x20000000
+#define CP_ME1_PIPE0_INT_STATUS__GENERIC2_INT_STATUS__SHIFT 0x1d
+#define CP_ME1_PIPE0_INT_STATUS__GENERIC1_INT_STATUS_MASK 0x40000000
+#define CP_ME1_PIPE0_INT_STATUS__GENERIC1_INT_STATUS__SHIFT 0x1e
+#define CP_ME1_PIPE0_INT_STATUS__GENERIC0_INT_STATUS_MASK 0x80000000
+#define CP_ME1_PIPE0_INT_STATUS__GENERIC0_INT_STATUS__SHIFT 0x1f
+#define CP_ME1_PIPE1_INT_STATUS__CMP_QUERY_STATUS_INT_STATUS_MASK 0x1000
+#define CP_ME1_PIPE1_INT_STATUS__CMP_QUERY_STATUS_INT_STATUS__SHIFT 0xc
+#define CP_ME1_PIPE1_INT_STATUS__DEQUEUE_REQUEST_INT_STATUS_MASK 0x2000
+#define CP_ME1_PIPE1_INT_STATUS__DEQUEUE_REQUEST_INT_STATUS__SHIFT 0xd
+#define CP_ME1_PIPE1_INT_STATUS__CP_ECC_ERROR_INT_STATUS_MASK 0x4000
+#define CP_ME1_PIPE1_INT_STATUS__CP_ECC_ERROR_INT_STATUS__SHIFT 0xe
+#define CP_ME1_PIPE1_INT_STATUS__SUA_VIOLATION_INT_STATUS_MASK 0x8000
+#define CP_ME1_PIPE1_INT_STATUS__SUA_VIOLATION_INT_STATUS__SHIFT 0xf
+#define CP_ME1_PIPE1_INT_STATUS__WRM_POLL_TIMEOUT_INT_STATUS_MASK 0x20000
+#define CP_ME1_PIPE1_INT_STATUS__WRM_POLL_TIMEOUT_INT_STATUS__SHIFT 0x11
+#define CP_ME1_PIPE1_INT_STATUS__PRIV_REG_INT_STATUS_MASK 0x800000
+#define CP_ME1_PIPE1_INT_STATUS__PRIV_REG_INT_STATUS__SHIFT 0x17
+#define CP_ME1_PIPE1_INT_STATUS__OPCODE_ERROR_INT_STATUS_MASK 0x1000000
+#define CP_ME1_PIPE1_INT_STATUS__OPCODE_ERROR_INT_STATUS__SHIFT 0x18
+#define CP_ME1_PIPE1_INT_STATUS__TIME_STAMP_INT_STATUS_MASK 0x4000000
+#define CP_ME1_PIPE1_INT_STATUS__TIME_STAMP_INT_STATUS__SHIFT 0x1a
+#define CP_ME1_PIPE1_INT_STATUS__RESERVED_BIT_ERROR_INT_STATUS_MASK 0x8000000
+#define CP_ME1_PIPE1_INT_STATUS__RESERVED_BIT_ERROR_INT_STATUS__SHIFT 0x1b
+#define CP_ME1_PIPE1_INT_STATUS__GENERIC2_INT_STATUS_MASK 0x20000000
+#define CP_ME1_PIPE1_INT_STATUS__GENERIC2_INT_STATUS__SHIFT 0x1d
+#define CP_ME1_PIPE1_INT_STATUS__GENERIC1_INT_STATUS_MASK 0x40000000
+#define CP_ME1_PIPE1_INT_STATUS__GENERIC1_INT_STATUS__SHIFT 0x1e
+#define CP_ME1_PIPE1_INT_STATUS__GENERIC0_INT_STATUS_MASK 0x80000000
+#define CP_ME1_PIPE1_INT_STATUS__GENERIC0_INT_STATUS__SHIFT 0x1f
+#define CP_ME1_PIPE2_INT_STATUS__CMP_QUERY_STATUS_INT_STATUS_MASK 0x1000
+#define CP_ME1_PIPE2_INT_STATUS__CMP_QUERY_STATUS_INT_STATUS__SHIFT 0xc
+#define CP_ME1_PIPE2_INT_STATUS__DEQUEUE_REQUEST_INT_STATUS_MASK 0x2000
+#define CP_ME1_PIPE2_INT_STATUS__DEQUEUE_REQUEST_INT_STATUS__SHIFT 0xd
+#define CP_ME1_PIPE2_INT_STATUS__CP_ECC_ERROR_INT_STATUS_MASK 0x4000
+#define CP_ME1_PIPE2_INT_STATUS__CP_ECC_ERROR_INT_STATUS__SHIFT 0xe
+#define CP_ME1_PIPE2_INT_STATUS__SUA_VIOLATION_INT_STATUS_MASK 0x8000
+#define CP_ME1_PIPE2_INT_STATUS__SUA_VIOLATION_INT_STATUS__SHIFT 0xf
+#define CP_ME1_PIPE2_INT_STATUS__WRM_POLL_TIMEOUT_INT_STATUS_MASK 0x20000
+#define CP_ME1_PIPE2_INT_STATUS__WRM_POLL_TIMEOUT_INT_STATUS__SHIFT 0x11
+#define CP_ME1_PIPE2_INT_STATUS__PRIV_REG_INT_STATUS_MASK 0x800000
+#define CP_ME1_PIPE2_INT_STATUS__PRIV_REG_INT_STATUS__SHIFT 0x17
+#define CP_ME1_PIPE2_INT_STATUS__OPCODE_ERROR_INT_STATUS_MASK 0x1000000
+#define CP_ME1_PIPE2_INT_STATUS__OPCODE_ERROR_INT_STATUS__SHIFT 0x18
+#define CP_ME1_PIPE2_INT_STATUS__TIME_STAMP_INT_STATUS_MASK 0x4000000
+#define CP_ME1_PIPE2_INT_STATUS__TIME_STAMP_INT_STATUS__SHIFT 0x1a
+#define CP_ME1_PIPE2_INT_STATUS__RESERVED_BIT_ERROR_INT_STATUS_MASK 0x8000000
+#define CP_ME1_PIPE2_INT_STATUS__RESERVED_BIT_ERROR_INT_STATUS__SHIFT 0x1b
+#define CP_ME1_PIPE2_INT_STATUS__GENERIC2_INT_STATUS_MASK 0x20000000
+#define CP_ME1_PIPE2_INT_STATUS__GENERIC2_INT_STATUS__SHIFT 0x1d
+#define CP_ME1_PIPE2_INT_STATUS__GENERIC1_INT_STATUS_MASK 0x40000000
+#define CP_ME1_PIPE2_INT_STATUS__GENERIC1_INT_STATUS__SHIFT 0x1e
+#define CP_ME1_PIPE2_INT_STATUS__GENERIC0_INT_STATUS_MASK 0x80000000
+#define CP_ME1_PIPE2_INT_STATUS__GENERIC0_INT_STATUS__SHIFT 0x1f
+#define CP_ME1_PIPE3_INT_STATUS__CMP_QUERY_STATUS_INT_STATUS_MASK 0x1000
+#define CP_ME1_PIPE3_INT_STATUS__CMP_QUERY_STATUS_INT_STATUS__SHIFT 0xc
+#define CP_ME1_PIPE3_INT_STATUS__DEQUEUE_REQUEST_INT_STATUS_MASK 0x2000
+#define CP_ME1_PIPE3_INT_STATUS__DEQUEUE_REQUEST_INT_STATUS__SHIFT 0xd
+#define CP_ME1_PIPE3_INT_STATUS__CP_ECC_ERROR_INT_STATUS_MASK 0x4000
+#define CP_ME1_PIPE3_INT_STATUS__CP_ECC_ERROR_INT_STATUS__SHIFT 0xe
+#define CP_ME1_PIPE3_INT_STATUS__SUA_VIOLATION_INT_STATUS_MASK 0x8000
+#define CP_ME1_PIPE3_INT_STATUS__SUA_VIOLATION_INT_STATUS__SHIFT 0xf
+#define CP_ME1_PIPE3_INT_STATUS__WRM_POLL_TIMEOUT_INT_STATUS_MASK 0x20000
+#define CP_ME1_PIPE3_INT_STATUS__WRM_POLL_TIMEOUT_INT_STATUS__SHIFT 0x11
+#define CP_ME1_PIPE3_INT_STATUS__PRIV_REG_INT_STATUS_MASK 0x800000
+#define CP_ME1_PIPE3_INT_STATUS__PRIV_REG_INT_STATUS__SHIFT 0x17
+#define CP_ME1_PIPE3_INT_STATUS__OPCODE_ERROR_INT_STATUS_MASK 0x1000000
+#define CP_ME1_PIPE3_INT_STATUS__OPCODE_ERROR_INT_STATUS__SHIFT 0x18
+#define CP_ME1_PIPE3_INT_STATUS__TIME_STAMP_INT_STATUS_MASK 0x4000000
+#define CP_ME1_PIPE3_INT_STATUS__TIME_STAMP_INT_STATUS__SHIFT 0x1a
+#define CP_ME1_PIPE3_INT_STATUS__RESERVED_BIT_ERROR_INT_STATUS_MASK 0x8000000
+#define CP_ME1_PIPE3_INT_STATUS__RESERVED_BIT_ERROR_INT_STATUS__SHIFT 0x1b
+#define CP_ME1_PIPE3_INT_STATUS__GENERIC2_INT_STATUS_MASK 0x20000000
+#define CP_ME1_PIPE3_INT_STATUS__GENERIC2_INT_STATUS__SHIFT 0x1d
+#define CP_ME1_PIPE3_INT_STATUS__GENERIC1_INT_STATUS_MASK 0x40000000
+#define CP_ME1_PIPE3_INT_STATUS__GENERIC1_INT_STATUS__SHIFT 0x1e
+#define CP_ME1_PIPE3_INT_STATUS__GENERIC0_INT_STATUS_MASK 0x80000000
+#define CP_ME1_PIPE3_INT_STATUS__GENERIC0_INT_STATUS__SHIFT 0x1f
+#define CP_ME2_PIPE0_INT_STATUS__CMP_QUERY_STATUS_INT_STATUS_MASK 0x1000
+#define CP_ME2_PIPE0_INT_STATUS__CMP_QUERY_STATUS_INT_STATUS__SHIFT 0xc
+#define CP_ME2_PIPE0_INT_STATUS__DEQUEUE_REQUEST_INT_STATUS_MASK 0x2000
+#define CP_ME2_PIPE0_INT_STATUS__DEQUEUE_REQUEST_INT_STATUS__SHIFT 0xd
+#define CP_ME2_PIPE0_INT_STATUS__CP_ECC_ERROR_INT_STATUS_MASK 0x4000
+#define CP_ME2_PIPE0_INT_STATUS__CP_ECC_ERROR_INT_STATUS__SHIFT 0xe
+#define CP_ME2_PIPE0_INT_STATUS__SUA_VIOLATION_INT_STATUS_MASK 0x8000
+#define CP_ME2_PIPE0_INT_STATUS__SUA_VIOLATION_INT_STATUS__SHIFT 0xf
+#define CP_ME2_PIPE0_INT_STATUS__WRM_POLL_TIMEOUT_INT_STATUS_MASK 0x20000
+#define CP_ME2_PIPE0_INT_STATUS__WRM_POLL_TIMEOUT_INT_STATUS__SHIFT 0x11
+#define CP_ME2_PIPE0_INT_STATUS__PRIV_REG_INT_STATUS_MASK 0x800000
+#define CP_ME2_PIPE0_INT_STATUS__PRIV_REG_INT_STATUS__SHIFT 0x17
+#define CP_ME2_PIPE0_INT_STATUS__OPCODE_ERROR_INT_STATUS_MASK 0x1000000
+#define CP_ME2_PIPE0_INT_STATUS__OPCODE_ERROR_INT_STATUS__SHIFT 0x18
+#define CP_ME2_PIPE0_INT_STATUS__TIME_STAMP_INT_STATUS_MASK 0x4000000
+#define CP_ME2_PIPE0_INT_STATUS__TIME_STAMP_INT_STATUS__SHIFT 0x1a
+#define CP_ME2_PIPE0_INT_STATUS__RESERVED_BIT_ERROR_INT_STATUS_MASK 0x8000000
+#define CP_ME2_PIPE0_INT_STATUS__RESERVED_BIT_ERROR_INT_STATUS__SHIFT 0x1b
+#define CP_ME2_PIPE0_INT_STATUS__GENERIC2_INT_STATUS_MASK 0x20000000
+#define CP_ME2_PIPE0_INT_STATUS__GENERIC2_INT_STATUS__SHIFT 0x1d
+#define CP_ME2_PIPE0_INT_STATUS__GENERIC1_INT_STATUS_MASK 0x40000000
+#define CP_ME2_PIPE0_INT_STATUS__GENERIC1_INT_STATUS__SHIFT 0x1e
+#define CP_ME2_PIPE0_INT_STATUS__GENERIC0_INT_STATUS_MASK 0x80000000
+#define CP_ME2_PIPE0_INT_STATUS__GENERIC0_INT_STATUS__SHIFT 0x1f
+#define CP_ME2_PIPE1_INT_STATUS__CMP_QUERY_STATUS_INT_STATUS_MASK 0x1000
+#define CP_ME2_PIPE1_INT_STATUS__CMP_QUERY_STATUS_INT_STATUS__SHIFT 0xc
+#define CP_ME2_PIPE1_INT_STATUS__DEQUEUE_REQUEST_INT_STATUS_MASK 0x2000
+#define CP_ME2_PIPE1_INT_STATUS__DEQUEUE_REQUEST_INT_STATUS__SHIFT 0xd
+#define CP_ME2_PIPE1_INT_STATUS__CP_ECC_ERROR_INT_STATUS_MASK 0x4000
+#define CP_ME2_PIPE1_INT_STATUS__CP_ECC_ERROR_INT_STATUS__SHIFT 0xe
+#define CP_ME2_PIPE1_INT_STATUS__SUA_VIOLATION_INT_STATUS_MASK 0x8000
+#define CP_ME2_PIPE1_INT_STATUS__SUA_VIOLATION_INT_STATUS__SHIFT 0xf
+#define CP_ME2_PIPE1_INT_STATUS__WRM_POLL_TIMEOUT_INT_STATUS_MASK 0x20000
+#define CP_ME2_PIPE1_INT_STATUS__WRM_POLL_TIMEOUT_INT_STATUS__SHIFT 0x11
+#define CP_ME2_PIPE1_INT_STATUS__PRIV_REG_INT_STATUS_MASK 0x800000
+#define CP_ME2_PIPE1_INT_STATUS__PRIV_REG_INT_STATUS__SHIFT 0x17
+#define CP_ME2_PIPE1_INT_STATUS__OPCODE_ERROR_INT_STATUS_MASK 0x1000000
+#define CP_ME2_PIPE1_INT_STATUS__OPCODE_ERROR_INT_STATUS__SHIFT 0x18
+#define CP_ME2_PIPE1_INT_STATUS__TIME_STAMP_INT_STATUS_MASK 0x4000000
+#define CP_ME2_PIPE1_INT_STATUS__TIME_STAMP_INT_STATUS__SHIFT 0x1a
+#define CP_ME2_PIPE1_INT_STATUS__RESERVED_BIT_ERROR_INT_STATUS_MASK 0x8000000
+#define CP_ME2_PIPE1_INT_STATUS__RESERVED_BIT_ERROR_INT_STATUS__SHIFT 0x1b
+#define CP_ME2_PIPE1_INT_STATUS__GENERIC2_INT_STATUS_MASK 0x20000000
+#define CP_ME2_PIPE1_INT_STATUS__GENERIC2_INT_STATUS__SHIFT 0x1d
+#define CP_ME2_PIPE1_INT_STATUS__GENERIC1_INT_STATUS_MASK 0x40000000
+#define CP_ME2_PIPE1_INT_STATUS__GENERIC1_INT_STATUS__SHIFT 0x1e
+#define CP_ME2_PIPE1_INT_STATUS__GENERIC0_INT_STATUS_MASK 0x80000000
+#define CP_ME2_PIPE1_INT_STATUS__GENERIC0_INT_STATUS__SHIFT 0x1f
+#define CP_ME2_PIPE2_INT_STATUS__CMP_QUERY_STATUS_INT_STATUS_MASK 0x1000
+#define CP_ME2_PIPE2_INT_STATUS__CMP_QUERY_STATUS_INT_STATUS__SHIFT 0xc
+#define CP_ME2_PIPE2_INT_STATUS__DEQUEUE_REQUEST_INT_STATUS_MASK 0x2000
+#define CP_ME2_PIPE2_INT_STATUS__DEQUEUE_REQUEST_INT_STATUS__SHIFT 0xd
+#define CP_ME2_PIPE2_INT_STATUS__CP_ECC_ERROR_INT_STATUS_MASK 0x4000
+#define CP_ME2_PIPE2_INT_STATUS__CP_ECC_ERROR_INT_STATUS__SHIFT 0xe
+#define CP_ME2_PIPE2_INT_STATUS__SUA_VIOLATION_INT_STATUS_MASK 0x8000
+#define CP_ME2_PIPE2_INT_STATUS__SUA_VIOLATION_INT_STATUS__SHIFT 0xf
+#define CP_ME2_PIPE2_INT_STATUS__WRM_POLL_TIMEOUT_INT_STATUS_MASK 0x20000
+#define CP_ME2_PIPE2_INT_STATUS__WRM_POLL_TIMEOUT_INT_STATUS__SHIFT 0x11
+#define CP_ME2_PIPE2_INT_STATUS__PRIV_REG_INT_STATUS_MASK 0x800000
+#define CP_ME2_PIPE2_INT_STATUS__PRIV_REG_INT_STATUS__SHIFT 0x17
+#define CP_ME2_PIPE2_INT_STATUS__OPCODE_ERROR_INT_STATUS_MASK 0x1000000
+#define CP_ME2_PIPE2_INT_STATUS__OPCODE_ERROR_INT_STATUS__SHIFT 0x18
+#define CP_ME2_PIPE2_INT_STATUS__TIME_STAMP_INT_STATUS_MASK 0x4000000
+#define CP_ME2_PIPE2_INT_STATUS__TIME_STAMP_INT_STATUS__SHIFT 0x1a
+#define CP_ME2_PIPE2_INT_STATUS__RESERVED_BIT_ERROR_INT_STATUS_MASK 0x8000000
+#define CP_ME2_PIPE2_INT_STATUS__RESERVED_BIT_ERROR_INT_STATUS__SHIFT 0x1b
+#define CP_ME2_PIPE2_INT_STATUS__GENERIC2_INT_STATUS_MASK 0x20000000
+#define CP_ME2_PIPE2_INT_STATUS__GENERIC2_INT_STATUS__SHIFT 0x1d
+#define CP_ME2_PIPE2_INT_STATUS__GENERIC1_INT_STATUS_MASK 0x40000000
+#define CP_ME2_PIPE2_INT_STATUS__GENERIC1_INT_STATUS__SHIFT 0x1e
+#define CP_ME2_PIPE2_INT_STATUS__GENERIC0_INT_STATUS_MASK 0x80000000
+#define CP_ME2_PIPE2_INT_STATUS__GENERIC0_INT_STATUS__SHIFT 0x1f
+#define CP_ME2_PIPE3_INT_STATUS__CMP_QUERY_STATUS_INT_STATUS_MASK 0x1000
+#define CP_ME2_PIPE3_INT_STATUS__CMP_QUERY_STATUS_INT_STATUS__SHIFT 0xc
+#define CP_ME2_PIPE3_INT_STATUS__DEQUEUE_REQUEST_INT_STATUS_MASK 0x2000
+#define CP_ME2_PIPE3_INT_STATUS__DEQUEUE_REQUEST_INT_STATUS__SHIFT 0xd
+#define CP_ME2_PIPE3_INT_STATUS__CP_ECC_ERROR_INT_STATUS_MASK 0x4000
+#define CP_ME2_PIPE3_INT_STATUS__CP_ECC_ERROR_INT_STATUS__SHIFT 0xe
+#define CP_ME2_PIPE3_INT_STATUS__SUA_VIOLATION_INT_STATUS_MASK 0x8000
+#define CP_ME2_PIPE3_INT_STATUS__SUA_VIOLATION_INT_STATUS__SHIFT 0xf
+#define CP_ME2_PIPE3_INT_STATUS__WRM_POLL_TIMEOUT_INT_STATUS_MASK 0x20000
+#define CP_ME2_PIPE3_INT_STATUS__WRM_POLL_TIMEOUT_INT_STATUS__SHIFT 0x11
+#define CP_ME2_PIPE3_INT_STATUS__PRIV_REG_INT_STATUS_MASK 0x800000
+#define CP_ME2_PIPE3_INT_STATUS__PRIV_REG_INT_STATUS__SHIFT 0x17
+#define CP_ME2_PIPE3_INT_STATUS__OPCODE_ERROR_INT_STATUS_MASK 0x1000000
+#define CP_ME2_PIPE3_INT_STATUS__OPCODE_ERROR_INT_STATUS__SHIFT 0x18
+#define CP_ME2_PIPE3_INT_STATUS__TIME_STAMP_INT_STATUS_MASK 0x4000000
+#define CP_ME2_PIPE3_INT_STATUS__TIME_STAMP_INT_STATUS__SHIFT 0x1a
+#define CP_ME2_PIPE3_INT_STATUS__RESERVED_BIT_ERROR_INT_STATUS_MASK 0x8000000
+#define CP_ME2_PIPE3_INT_STATUS__RESERVED_BIT_ERROR_INT_STATUS__SHIFT 0x1b
+#define CP_ME2_PIPE3_INT_STATUS__GENERIC2_INT_STATUS_MASK 0x20000000
+#define CP_ME2_PIPE3_INT_STATUS__GENERIC2_INT_STATUS__SHIFT 0x1d
+#define CP_ME2_PIPE3_INT_STATUS__GENERIC1_INT_STATUS_MASK 0x40000000
+#define CP_ME2_PIPE3_INT_STATUS__GENERIC1_INT_STATUS__SHIFT 0x1e
+#define CP_ME2_PIPE3_INT_STATUS__GENERIC0_INT_STATUS_MASK 0x80000000
+#define CP_ME2_PIPE3_INT_STATUS__GENERIC0_INT_STATUS__SHIFT 0x1f
+#define CP_ME1_INT_STAT_DEBUG__CMP_QUERY_STATUS_INT_ASSERTED_MASK 0x1000
+#define CP_ME1_INT_STAT_DEBUG__CMP_QUERY_STATUS_INT_ASSERTED__SHIFT 0xc
+#define CP_ME1_INT_STAT_DEBUG__DEQUEUE_REQUEST_INT_ASSERTED_MASK 0x2000
+#define CP_ME1_INT_STAT_DEBUG__DEQUEUE_REQUEST_INT_ASSERTED__SHIFT 0xd
+#define CP_ME1_INT_STAT_DEBUG__CP_ECC_ERROR_INT_ASSERTED_MASK 0x4000
+#define CP_ME1_INT_STAT_DEBUG__CP_ECC_ERROR_INT_ASSERTED__SHIFT 0xe
+#define CP_ME1_INT_STAT_DEBUG__SUA_VIOLATION_INT_STATUS_MASK 0x8000
+#define CP_ME1_INT_STAT_DEBUG__SUA_VIOLATION_INT_STATUS__SHIFT 0xf
+#define CP_ME1_INT_STAT_DEBUG__WRM_POLL_TIMEOUT_INT_ASSERTED_MASK 0x20000
+#define CP_ME1_INT_STAT_DEBUG__WRM_POLL_TIMEOUT_INT_ASSERTED__SHIFT 0x11
+#define CP_ME1_INT_STAT_DEBUG__PRIV_REG_INT_ASSERTED_MASK 0x800000
+#define CP_ME1_INT_STAT_DEBUG__PRIV_REG_INT_ASSERTED__SHIFT 0x17
+#define CP_ME1_INT_STAT_DEBUG__OPCODE_ERROR_INT_ASSERTED_MASK 0x1000000
+#define CP_ME1_INT_STAT_DEBUG__OPCODE_ERROR_INT_ASSERTED__SHIFT 0x18
+#define CP_ME1_INT_STAT_DEBUG__TIME_STAMP_INT_ASSERTED_MASK 0x4000000
+#define CP_ME1_INT_STAT_DEBUG__TIME_STAMP_INT_ASSERTED__SHIFT 0x1a
+#define CP_ME1_INT_STAT_DEBUG__RESERVED_BIT_ERROR_INT_ASSERTED_MASK 0x8000000
+#define CP_ME1_INT_STAT_DEBUG__RESERVED_BIT_ERROR_INT_ASSERTED__SHIFT 0x1b
+#define CP_ME1_INT_STAT_DEBUG__GENERIC2_INT_ASSERTED_MASK 0x20000000
+#define CP_ME1_INT_STAT_DEBUG__GENERIC2_INT_ASSERTED__SHIFT 0x1d
+#define CP_ME1_INT_STAT_DEBUG__GENERIC1_INT_ASSERTED_MASK 0x40000000
+#define CP_ME1_INT_STAT_DEBUG__GENERIC1_INT_ASSERTED__SHIFT 0x1e
+#define CP_ME1_INT_STAT_DEBUG__GENERIC0_INT_ASSERTED_MASK 0x80000000
+#define CP_ME1_INT_STAT_DEBUG__GENERIC0_INT_ASSERTED__SHIFT 0x1f
+#define CP_ME2_INT_STAT_DEBUG__CMP_QUERY_STATUS_INT_ASSERTED_MASK 0x1000
+#define CP_ME2_INT_STAT_DEBUG__CMP_QUERY_STATUS_INT_ASSERTED__SHIFT 0xc
+#define CP_ME2_INT_STAT_DEBUG__DEQUEUE_REQUEST_INT_ASSERTED_MASK 0x2000
+#define CP_ME2_INT_STAT_DEBUG__DEQUEUE_REQUEST_INT_ASSERTED__SHIFT 0xd
+#define CP_ME2_INT_STAT_DEBUG__CP_ECC_ERROR_INT_ASSERTED_MASK 0x4000
+#define CP_ME2_INT_STAT_DEBUG__CP_ECC_ERROR_INT_ASSERTED__SHIFT 0xe
+#define CP_ME2_INT_STAT_DEBUG__SUA_VIOLATION_INT_STATUS_MASK 0x8000
+#define CP_ME2_INT_STAT_DEBUG__SUA_VIOLATION_INT_STATUS__SHIFT 0xf
+#define CP_ME2_INT_STAT_DEBUG__WRM_POLL_TIMEOUT_INT_ASSERTED_MASK 0x20000
+#define CP_ME2_INT_STAT_DEBUG__WRM_POLL_TIMEOUT_INT_ASSERTED__SHIFT 0x11
+#define CP_ME2_INT_STAT_DEBUG__PRIV_REG_INT_ASSERTED_MASK 0x800000
+#define CP_ME2_INT_STAT_DEBUG__PRIV_REG_INT_ASSERTED__SHIFT 0x17
+#define CP_ME2_INT_STAT_DEBUG__OPCODE_ERROR_INT_ASSERTED_MASK 0x1000000
+#define CP_ME2_INT_STAT_DEBUG__OPCODE_ERROR_INT_ASSERTED__SHIFT 0x18
+#define CP_ME2_INT_STAT_DEBUG__TIME_STAMP_INT_ASSERTED_MASK 0x4000000
+#define CP_ME2_INT_STAT_DEBUG__TIME_STAMP_INT_ASSERTED__SHIFT 0x1a
+#define CP_ME2_INT_STAT_DEBUG__RESERVED_BIT_ERROR_INT_ASSERTED_MASK 0x8000000
+#define CP_ME2_INT_STAT_DEBUG__RESERVED_BIT_ERROR_INT_ASSERTED__SHIFT 0x1b
+#define CP_ME2_INT_STAT_DEBUG__GENERIC2_INT_ASSERTED_MASK 0x20000000
+#define CP_ME2_INT_STAT_DEBUG__GENERIC2_INT_ASSERTED__SHIFT 0x1d
+#define CP_ME2_INT_STAT_DEBUG__GENERIC1_INT_ASSERTED_MASK 0x40000000
+#define CP_ME2_INT_STAT_DEBUG__GENERIC1_INT_ASSERTED__SHIFT 0x1e
+#define CP_ME2_INT_STAT_DEBUG__GENERIC0_INT_ASSERTED_MASK 0x80000000
+#define CP_ME2_INT_STAT_DEBUG__GENERIC0_INT_ASSERTED__SHIFT 0x1f
+#define CP_ME1_PIPE_PRIORITY_CNTS__PRIORITY1_CNT_MASK 0xff
+#define CP_ME1_PIPE_PRIORITY_CNTS__PRIORITY1_CNT__SHIFT 0x0
+#define CP_ME1_PIPE_PRIORITY_CNTS__PRIORITY2A_CNT_MASK 0xff00
+#define CP_ME1_PIPE_PRIORITY_CNTS__PRIORITY2A_CNT__SHIFT 0x8
+#define CP_ME1_PIPE_PRIORITY_CNTS__PRIORITY2B_CNT_MASK 0xff0000
+#define CP_ME1_PIPE_PRIORITY_CNTS__PRIORITY2B_CNT__SHIFT 0x10
+#define CP_ME1_PIPE_PRIORITY_CNTS__PRIORITY3_CNT_MASK 0xff000000
+#define CP_ME1_PIPE_PRIORITY_CNTS__PRIORITY3_CNT__SHIFT 0x18
+#define CP_ME1_PIPE0_PRIORITY__PRIORITY_MASK 0x3
+#define CP_ME1_PIPE0_PRIORITY__PRIORITY__SHIFT 0x0
+#define CP_ME1_PIPE1_PRIORITY__PRIORITY_MASK 0x3
+#define CP_ME1_PIPE1_PRIORITY__PRIORITY__SHIFT 0x0
+#define CP_ME1_PIPE2_PRIORITY__PRIORITY_MASK 0x3
+#define CP_ME1_PIPE2_PRIORITY__PRIORITY__SHIFT 0x0
+#define CP_ME1_PIPE3_PRIORITY__PRIORITY_MASK 0x3
+#define CP_ME1_PIPE3_PRIORITY__PRIORITY__SHIFT 0x0
+#define CP_ME2_PIPE_PRIORITY_CNTS__PRIORITY1_CNT_MASK 0xff
+#define CP_ME2_PIPE_PRIORITY_CNTS__PRIORITY1_CNT__SHIFT 0x0
+#define CP_ME2_PIPE_PRIORITY_CNTS__PRIORITY2A_CNT_MASK 0xff00
+#define CP_ME2_PIPE_PRIORITY_CNTS__PRIORITY2A_CNT__SHIFT 0x8
+#define CP_ME2_PIPE_PRIORITY_CNTS__PRIORITY2B_CNT_MASK 0xff0000
+#define CP_ME2_PIPE_PRIORITY_CNTS__PRIORITY2B_CNT__SHIFT 0x10
+#define CP_ME2_PIPE_PRIORITY_CNTS__PRIORITY3_CNT_MASK 0xff000000
+#define CP_ME2_PIPE_PRIORITY_CNTS__PRIORITY3_CNT__SHIFT 0x18
+#define CP_ME2_PIPE0_PRIORITY__PRIORITY_MASK 0x3
+#define CP_ME2_PIPE0_PRIORITY__PRIORITY__SHIFT 0x0
+#define CP_ME2_PIPE1_PRIORITY__PRIORITY_MASK 0x3
+#define CP_ME2_PIPE1_PRIORITY__PRIORITY__SHIFT 0x0
+#define CP_ME2_PIPE2_PRIORITY__PRIORITY_MASK 0x3
+#define CP_ME2_PIPE2_PRIORITY__PRIORITY__SHIFT 0x0
+#define CP_ME2_PIPE3_PRIORITY__PRIORITY_MASK 0x3
+#define CP_ME2_PIPE3_PRIORITY__PRIORITY__SHIFT 0x0
+#define CP_CE_PRGRM_CNTR_START__IP_START_MASK 0x7ff
+#define CP_CE_PRGRM_CNTR_START__IP_START__SHIFT 0x0
+#define CP_PFP_PRGRM_CNTR_START__IP_START_MASK 0xfff
+#define CP_PFP_PRGRM_CNTR_START__IP_START__SHIFT 0x0
+#define CP_ME_PRGRM_CNTR_START__IP_START_MASK 0xfff
+#define CP_ME_PRGRM_CNTR_START__IP_START__SHIFT 0x0
+#define CP_MEC1_PRGRM_CNTR_START__IP_START_MASK 0xffff
+#define CP_MEC1_PRGRM_CNTR_START__IP_START__SHIFT 0x0
+#define CP_MEC2_PRGRM_CNTR_START__IP_START_MASK 0xffff
+#define CP_MEC2_PRGRM_CNTR_START__IP_START__SHIFT 0x0
+#define CP_CE_INTR_ROUTINE_START__IR_START_MASK 0x7ff
+#define CP_CE_INTR_ROUTINE_START__IR_START__SHIFT 0x0
+#define CP_PFP_INTR_ROUTINE_START__IR_START_MASK 0xfff
+#define CP_PFP_INTR_ROUTINE_START__IR_START__SHIFT 0x0
+#define CP_ME_INTR_ROUTINE_START__IR_START_MASK 0xfff
+#define CP_ME_INTR_ROUTINE_START__IR_START__SHIFT 0x0
+#define CP_MEC1_INTR_ROUTINE_START__IR_START_MASK 0xffff
+#define CP_MEC1_INTR_ROUTINE_START__IR_START__SHIFT 0x0
+#define CP_MEC2_INTR_ROUTINE_START__IR_START_MASK 0xffff
+#define CP_MEC2_INTR_ROUTINE_START__IR_START__SHIFT 0x0
+#define CP_CONTEXT_CNTL__ME0PIPE0_MAX_WD_CNTX_MASK 0x7
+#define CP_CONTEXT_CNTL__ME0PIPE0_MAX_WD_CNTX__SHIFT 0x0
+#define CP_CONTEXT_CNTL__ME0PIPE0_MAX_PIPE_CNTX_MASK 0x70
+#define CP_CONTEXT_CNTL__ME0PIPE0_MAX_PIPE_CNTX__SHIFT 0x4
+#define CP_CONTEXT_CNTL__ME0PIPE1_MAX_WD_CNTX_MASK 0x70000
+#define CP_CONTEXT_CNTL__ME0PIPE1_MAX_WD_CNTX__SHIFT 0x10
+#define CP_CONTEXT_CNTL__ME0PIPE1_MAX_PIPE_CNTX_MASK 0x700000
+#define CP_CONTEXT_CNTL__ME0PIPE1_MAX_PIPE_CNTX__SHIFT 0x14
+#define CP_MAX_CONTEXT__MAX_CONTEXT_MASK 0x7
+#define CP_MAX_CONTEXT__MAX_CONTEXT__SHIFT 0x0
+#define CP_IQ_WAIT_TIME1__IB_OFFLOAD_MASK 0xff
+#define CP_IQ_WAIT_TIME1__IB_OFFLOAD__SHIFT 0x0
+#define CP_IQ_WAIT_TIME1__ATOMIC_OFFLOAD_MASK 0xff00
+#define CP_IQ_WAIT_TIME1__ATOMIC_OFFLOAD__SHIFT 0x8
+#define CP_IQ_WAIT_TIME1__WRM_OFFLOAD_MASK 0xff0000
+#define CP_IQ_WAIT_TIME1__WRM_OFFLOAD__SHIFT 0x10
+#define CP_IQ_WAIT_TIME1__GWS_MASK 0xff000000
+#define CP_IQ_WAIT_TIME1__GWS__SHIFT 0x18
+#define CP_IQ_WAIT_TIME2__QUE_SLEEP_MASK 0xff
+#define CP_IQ_WAIT_TIME2__QUE_SLEEP__SHIFT 0x0
+#define CP_IQ_WAIT_TIME2__SCH_WAVE_MASK 0xff00
+#define CP_IQ_WAIT_TIME2__SCH_WAVE__SHIFT 0x8
+#define CP_IQ_WAIT_TIME2__SEM_REARM_MASK 0xff0000
+#define CP_IQ_WAIT_TIME2__SEM_REARM__SHIFT 0x10
+#define CP_IQ_WAIT_TIME2__DEQ_RETRY_MASK 0xff000000
+#define CP_IQ_WAIT_TIME2__DEQ_RETRY__SHIFT 0x18
+#define CP_VMID_RESET__RESET_REQUEST_MASK 0xffff
+#define CP_VMID_RESET__RESET_REQUEST__SHIFT 0x0
+#define CP_VMID_RESET__RESET_STATUS_MASK 0xffff0000
+#define CP_VMID_RESET__RESET_STATUS__SHIFT 0x10
+#define CP_VMID_PREEMPT__PREEMPT_REQUEST_MASK 0xffff
+#define CP_VMID_PREEMPT__PREEMPT_REQUEST__SHIFT 0x0
+#define CP_VMID_PREEMPT__VIRT_COMMAND_MASK 0xf0000
+#define CP_VMID_PREEMPT__VIRT_COMMAND__SHIFT 0x10
+#define CP_VMID_STATUS__PREEMPT_DE_STATUS_MASK 0xffff
+#define CP_VMID_STATUS__PREEMPT_DE_STATUS__SHIFT 0x0
+#define CP_VMID_STATUS__PREEMPT_CE_STATUS_MASK 0xffff0000
+#define CP_VMID_STATUS__PREEMPT_CE_STATUS__SHIFT 0x10
+#define CPC_INT_CNTX_ID__CNTX_ID_MASK 0xfffffff
+#define CPC_INT_CNTX_ID__CNTX_ID__SHIFT 0x0
+#define CPC_INT_CNTX_ID__QUEUE_ID_MASK 0x70000000
+#define CPC_INT_CNTX_ID__QUEUE_ID__SHIFT 0x1c
+#define CP_PQ_STATUS__DOORBELL_UPDATED_MASK 0x1
+#define CP_PQ_STATUS__DOORBELL_UPDATED__SHIFT 0x0
+#define CP_PQ_STATUS__DOORBELL_ENABLE_MASK 0x2
+#define CP_PQ_STATUS__DOORBELL_ENABLE__SHIFT 0x1
+#define CP_CPC_IC_BASE_LO__IC_BASE_LO_MASK 0xfffff000
+#define CP_CPC_IC_BASE_LO__IC_BASE_LO__SHIFT 0xc
+#define CP_CPC_IC_BASE_HI__IC_BASE_HI_MASK 0xffff
+#define CP_CPC_IC_BASE_HI__IC_BASE_HI__SHIFT 0x0
+#define CP_CPC_IC_BASE_CNTL__VMID_MASK 0xf
+#define CP_CPC_IC_BASE_CNTL__VMID__SHIFT 0x0
+#define CP_CPC_IC_BASE_CNTL__ATC_MASK 0x800000
+#define CP_CPC_IC_BASE_CNTL__ATC__SHIFT 0x17
+#define CP_CPC_IC_BASE_CNTL__CACHE_POLICY_MASK 0x1000000
+#define CP_CPC_IC_BASE_CNTL__CACHE_POLICY__SHIFT 0x18
+#define CP_CPC_IC_BASE_CNTL__MTYPE_MASK 0x18000000
+#define CP_CPC_IC_BASE_CNTL__MTYPE__SHIFT 0x1b
+#define CP_CPC_IC_OP_CNTL__INVALIDATE_CACHE_MASK 0x1
+#define CP_CPC_IC_OP_CNTL__INVALIDATE_CACHE__SHIFT 0x0
+#define CP_CPC_IC_OP_CNTL__PRIME_ICACHE_MASK 0x10
+#define CP_CPC_IC_OP_CNTL__PRIME_ICACHE__SHIFT 0x4
+#define CP_CPC_IC_OP_CNTL__ICACHE_PRIMED_MASK 0x20
+#define CP_CPC_IC_OP_CNTL__ICACHE_PRIMED__SHIFT 0x5
+#define CP_CPC_STATUS__MEC1_BUSY_MASK 0x1
+#define CP_CPC_STATUS__MEC1_BUSY__SHIFT 0x0
+#define CP_CPC_STATUS__MEC2_BUSY_MASK 0x2
+#define CP_CPC_STATUS__MEC2_BUSY__SHIFT 0x1
+#define CP_CPC_STATUS__DC0_BUSY_MASK 0x4
+#define CP_CPC_STATUS__DC0_BUSY__SHIFT 0x2
+#define CP_CPC_STATUS__DC1_BUSY_MASK 0x8
+#define CP_CPC_STATUS__DC1_BUSY__SHIFT 0x3
+#define CP_CPC_STATUS__RCIU1_BUSY_MASK 0x10
+#define CP_CPC_STATUS__RCIU1_BUSY__SHIFT 0x4
+#define CP_CPC_STATUS__RCIU2_BUSY_MASK 0x20
+#define CP_CPC_STATUS__RCIU2_BUSY__SHIFT 0x5
+#define CP_CPC_STATUS__ROQ1_BUSY_MASK 0x40
+#define CP_CPC_STATUS__ROQ1_BUSY__SHIFT 0x6
+#define CP_CPC_STATUS__ROQ2_BUSY_MASK 0x80
+#define CP_CPC_STATUS__ROQ2_BUSY__SHIFT 0x7
+#define CP_CPC_STATUS__TCIU_BUSY_MASK 0x400
+#define CP_CPC_STATUS__TCIU_BUSY__SHIFT 0xa
+#define CP_CPC_STATUS__SCRATCH_RAM_BUSY_MASK 0x800
+#define CP_CPC_STATUS__SCRATCH_RAM_BUSY__SHIFT 0xb
+#define CP_CPC_STATUS__QU_BUSY_MASK 0x1000
+#define CP_CPC_STATUS__QU_BUSY__SHIFT 0xc
+#define CP_CPC_STATUS__ATCL2IU_BUSY_MASK 0x2000
+#define CP_CPC_STATUS__ATCL2IU_BUSY__SHIFT 0xd
+#define CP_CPC_STATUS__CPG_CPC_BUSY_MASK 0x20000000
+#define CP_CPC_STATUS__CPG_CPC_BUSY__SHIFT 0x1d
+#define CP_CPC_STATUS__CPF_CPC_BUSY_MASK 0x40000000
+#define CP_CPC_STATUS__CPF_CPC_BUSY__SHIFT 0x1e
+#define CP_CPC_STATUS__CPC_BUSY_MASK 0x80000000
+#define CP_CPC_STATUS__CPC_BUSY__SHIFT 0x1f
+#define CP_CPC_BUSY_STAT__MEC1_LOAD_BUSY_MASK 0x1
+#define CP_CPC_BUSY_STAT__MEC1_LOAD_BUSY__SHIFT 0x0
+#define CP_CPC_BUSY_STAT__MEC1_SEMAPOHRE_BUSY_MASK 0x2
+#define CP_CPC_BUSY_STAT__MEC1_SEMAPOHRE_BUSY__SHIFT 0x1
+#define CP_CPC_BUSY_STAT__MEC1_MUTEX_BUSY_MASK 0x4
+#define CP_CPC_BUSY_STAT__MEC1_MUTEX_BUSY__SHIFT 0x2
+#define CP_CPC_BUSY_STAT__MEC1_MESSAGE_BUSY_MASK 0x8
+#define CP_CPC_BUSY_STAT__MEC1_MESSAGE_BUSY__SHIFT 0x3
+#define CP_CPC_BUSY_STAT__MEC1_EOP_QUEUE_BUSY_MASK 0x10
+#define CP_CPC_BUSY_STAT__MEC1_EOP_QUEUE_BUSY__SHIFT 0x4
+#define CP_CPC_BUSY_STAT__MEC1_IQ_QUEUE_BUSY_MASK 0x20
+#define CP_CPC_BUSY_STAT__MEC1_IQ_QUEUE_BUSY__SHIFT 0x5
+#define CP_CPC_BUSY_STAT__MEC1_IB_QUEUE_BUSY_MASK 0x40
+#define CP_CPC_BUSY_STAT__MEC1_IB_QUEUE_BUSY__SHIFT 0x6
+#define CP_CPC_BUSY_STAT__MEC1_TC_BUSY_MASK 0x80
+#define CP_CPC_BUSY_STAT__MEC1_TC_BUSY__SHIFT 0x7
+#define CP_CPC_BUSY_STAT__MEC1_DMA_BUSY_MASK 0x100
+#define CP_CPC_BUSY_STAT__MEC1_DMA_BUSY__SHIFT 0x8
+#define CP_CPC_BUSY_STAT__MEC1_PARTIAL_FLUSH_BUSY_MASK 0x200
+#define CP_CPC_BUSY_STAT__MEC1_PARTIAL_FLUSH_BUSY__SHIFT 0x9
+#define CP_CPC_BUSY_STAT__MEC1_PIPE0_BUSY_MASK 0x400
+#define CP_CPC_BUSY_STAT__MEC1_PIPE0_BUSY__SHIFT 0xa
+#define CP_CPC_BUSY_STAT__MEC1_PIPE1_BUSY_MASK 0x800
+#define CP_CPC_BUSY_STAT__MEC1_PIPE1_BUSY__SHIFT 0xb
+#define CP_CPC_BUSY_STAT__MEC1_PIPE2_BUSY_MASK 0x1000
+#define CP_CPC_BUSY_STAT__MEC1_PIPE2_BUSY__SHIFT 0xc
+#define CP_CPC_BUSY_STAT__MEC1_PIPE3_BUSY_MASK 0x2000
+#define CP_CPC_BUSY_STAT__MEC1_PIPE3_BUSY__SHIFT 0xd
+#define CP_CPC_BUSY_STAT__MEC2_LOAD_BUSY_MASK 0x10000
+#define CP_CPC_BUSY_STAT__MEC2_LOAD_BUSY__SHIFT 0x10
+#define CP_CPC_BUSY_STAT__MEC2_SEMAPOHRE_BUSY_MASK 0x20000
+#define CP_CPC_BUSY_STAT__MEC2_SEMAPOHRE_BUSY__SHIFT 0x11
+#define CP_CPC_BUSY_STAT__MEC2_MUTEX_BUSY_MASK 0x40000
+#define CP_CPC_BUSY_STAT__MEC2_MUTEX_BUSY__SHIFT 0x12
+#define CP_CPC_BUSY_STAT__MEC2_MESSAGE_BUSY_MASK 0x80000
+#define CP_CPC_BUSY_STAT__MEC2_MESSAGE_BUSY__SHIFT 0x13
+#define CP_CPC_BUSY_STAT__MEC2_EOP_QUEUE_BUSY_MASK 0x100000
+#define CP_CPC_BUSY_STAT__MEC2_EOP_QUEUE_BUSY__SHIFT 0x14
+#define CP_CPC_BUSY_STAT__MEC2_IQ_QUEUE_BUSY_MASK 0x200000
+#define CP_CPC_BUSY_STAT__MEC2_IQ_QUEUE_BUSY__SHIFT 0x15
+#define CP_CPC_BUSY_STAT__MEC2_IB_QUEUE_BUSY_MASK 0x400000
+#define CP_CPC_BUSY_STAT__MEC2_IB_QUEUE_BUSY__SHIFT 0x16
+#define CP_CPC_BUSY_STAT__MEC2_TC_BUSY_MASK 0x800000
+#define CP_CPC_BUSY_STAT__MEC2_TC_BUSY__SHIFT 0x17
+#define CP_CPC_BUSY_STAT__MEC2_DMA_BUSY_MASK 0x1000000
+#define CP_CPC_BUSY_STAT__MEC2_DMA_BUSY__SHIFT 0x18
+#define CP_CPC_BUSY_STAT__MEC2_PARTIAL_FLUSH_BUSY_MASK 0x2000000
+#define CP_CPC_BUSY_STAT__MEC2_PARTIAL_FLUSH_BUSY__SHIFT 0x19
+#define CP_CPC_BUSY_STAT__MEC2_PIPE0_BUSY_MASK 0x4000000
+#define CP_CPC_BUSY_STAT__MEC2_PIPE0_BUSY__SHIFT 0x1a
+#define CP_CPC_BUSY_STAT__MEC2_PIPE1_BUSY_MASK 0x8000000
+#define CP_CPC_BUSY_STAT__MEC2_PIPE1_BUSY__SHIFT 0x1b
+#define CP_CPC_BUSY_STAT__MEC2_PIPE2_BUSY_MASK 0x10000000
+#define CP_CPC_BUSY_STAT__MEC2_PIPE2_BUSY__SHIFT 0x1c
+#define CP_CPC_BUSY_STAT__MEC2_PIPE3_BUSY_MASK 0x20000000
+#define CP_CPC_BUSY_STAT__MEC2_PIPE3_BUSY__SHIFT 0x1d
+#define CP_CPC_STALLED_STAT1__RCIU_TX_FREE_STALL_MASK 0x8
+#define CP_CPC_STALLED_STAT1__RCIU_TX_FREE_STALL__SHIFT 0x3
+#define CP_CPC_STALLED_STAT1__RCIU_PRIV_VIOLATION_MASK 0x10
+#define CP_CPC_STALLED_STAT1__RCIU_PRIV_VIOLATION__SHIFT 0x4
+#define CP_CPC_STALLED_STAT1__TCIU_TX_FREE_STALL_MASK 0x40
+#define CP_CPC_STALLED_STAT1__TCIU_TX_FREE_STALL__SHIFT 0x6
+#define CP_CPC_STALLED_STAT1__MEC1_DECODING_PACKET_MASK 0x100
+#define CP_CPC_STALLED_STAT1__MEC1_DECODING_PACKET__SHIFT 0x8
+#define CP_CPC_STALLED_STAT1__MEC1_WAIT_ON_RCIU_MASK 0x200
+#define CP_CPC_STALLED_STAT1__MEC1_WAIT_ON_RCIU__SHIFT 0x9
+#define CP_CPC_STALLED_STAT1__MEC1_WAIT_ON_RCIU_READ_MASK 0x400
+#define CP_CPC_STALLED_STAT1__MEC1_WAIT_ON_RCIU_READ__SHIFT 0xa
+#define CP_CPC_STALLED_STAT1__MEC1_WAIT_ON_ROQ_DATA_MASK 0x2000
+#define CP_CPC_STALLED_STAT1__MEC1_WAIT_ON_ROQ_DATA__SHIFT 0xd
+#define CP_CPC_STALLED_STAT1__MEC2_DECODING_PACKET_MASK 0x10000
+#define CP_CPC_STALLED_STAT1__MEC2_DECODING_PACKET__SHIFT 0x10
+#define CP_CPC_STALLED_STAT1__MEC2_WAIT_ON_RCIU_MASK 0x20000
+#define CP_CPC_STALLED_STAT1__MEC2_WAIT_ON_RCIU__SHIFT 0x11
+#define CP_CPC_STALLED_STAT1__MEC2_WAIT_ON_RCIU_READ_MASK 0x40000
+#define CP_CPC_STALLED_STAT1__MEC2_WAIT_ON_RCIU_READ__SHIFT 0x12
+#define CP_CPC_STALLED_STAT1__MEC2_WAIT_ON_ROQ_DATA_MASK 0x200000
+#define CP_CPC_STALLED_STAT1__MEC2_WAIT_ON_ROQ_DATA__SHIFT 0x15
+#define CP_CPC_STALLED_STAT1__ATCL2IU_WAITING_ON_FREE_MASK 0x400000
+#define CP_CPC_STALLED_STAT1__ATCL2IU_WAITING_ON_FREE__SHIFT 0x16
+#define CP_CPC_STALLED_STAT1__ATCL2IU_WAITING_ON_TAGS_MASK 0x800000
+#define CP_CPC_STALLED_STAT1__ATCL2IU_WAITING_ON_TAGS__SHIFT 0x17
+#define CP_CPC_STALLED_STAT1__ATCL1_WAITING_ON_TRANS_MASK 0x1000000
+#define CP_CPC_STALLED_STAT1__ATCL1_WAITING_ON_TRANS__SHIFT 0x18
+#define CP_CPF_STATUS__POST_WPTR_GFX_BUSY_MASK 0x1
+#define CP_CPF_STATUS__POST_WPTR_GFX_BUSY__SHIFT 0x0
+#define CP_CPF_STATUS__CSF_BUSY_MASK 0x2
+#define CP_CPF_STATUS__CSF_BUSY__SHIFT 0x1
+#define CP_CPF_STATUS__ROQ_ALIGN_BUSY_MASK 0x10
+#define CP_CPF_STATUS__ROQ_ALIGN_BUSY__SHIFT 0x4
+#define CP_CPF_STATUS__ROQ_RING_BUSY_MASK 0x20
+#define CP_CPF_STATUS__ROQ_RING_BUSY__SHIFT 0x5
+#define CP_CPF_STATUS__ROQ_INDIRECT1_BUSY_MASK 0x40
+#define CP_CPF_STATUS__ROQ_INDIRECT1_BUSY__SHIFT 0x6
+#define CP_CPF_STATUS__ROQ_INDIRECT2_BUSY_MASK 0x80
+#define CP_CPF_STATUS__ROQ_INDIRECT2_BUSY__SHIFT 0x7
+#define CP_CPF_STATUS__ROQ_STATE_BUSY_MASK 0x100
+#define CP_CPF_STATUS__ROQ_STATE_BUSY__SHIFT 0x8
+#define CP_CPF_STATUS__ROQ_CE_RING_BUSY_MASK 0x200
+#define CP_CPF_STATUS__ROQ_CE_RING_BUSY__SHIFT 0x9
+#define CP_CPF_STATUS__ROQ_CE_INDIRECT1_BUSY_MASK 0x400
+#define CP_CPF_STATUS__ROQ_CE_INDIRECT1_BUSY__SHIFT 0xa
+#define CP_CPF_STATUS__ROQ_CE_INDIRECT2_BUSY_MASK 0x800
+#define CP_CPF_STATUS__ROQ_CE_INDIRECT2_BUSY__SHIFT 0xb
+#define CP_CPF_STATUS__SEMAPHORE_BUSY_MASK 0x1000
+#define CP_CPF_STATUS__SEMAPHORE_BUSY__SHIFT 0xc
+#define CP_CPF_STATUS__INTERRUPT_BUSY_MASK 0x2000
+#define CP_CPF_STATUS__INTERRUPT_BUSY__SHIFT 0xd
+#define CP_CPF_STATUS__TCIU_BUSY_MASK 0x4000
+#define CP_CPF_STATUS__TCIU_BUSY__SHIFT 0xe
+#define CP_CPF_STATUS__HQD_BUSY_MASK 0x8000
+#define CP_CPF_STATUS__HQD_BUSY__SHIFT 0xf
+#define CP_CPF_STATUS__PRT_BUSY_MASK 0x10000
+#define CP_CPF_STATUS__PRT_BUSY__SHIFT 0x10
+#define CP_CPF_STATUS__ATCL2IU_BUSY_MASK 0x20000
+#define CP_CPF_STATUS__ATCL2IU_BUSY__SHIFT 0x11
+#define CP_CPF_STATUS__CPF_GFX_BUSY_MASK 0x4000000
+#define CP_CPF_STATUS__CPF_GFX_BUSY__SHIFT 0x1a
+#define CP_CPF_STATUS__CPF_CMP_BUSY_MASK 0x8000000
+#define CP_CPF_STATUS__CPF_CMP_BUSY__SHIFT 0x1b
+#define CP_CPF_STATUS__GRBM_CPF_STAT_BUSY_MASK 0x30000000
+#define CP_CPF_STATUS__GRBM_CPF_STAT_BUSY__SHIFT 0x1c
+#define CP_CPF_STATUS__CPC_CPF_BUSY_MASK 0x40000000
+#define CP_CPF_STATUS__CPC_CPF_BUSY__SHIFT 0x1e
+#define CP_CPF_STATUS__CPF_BUSY_MASK 0x80000000
+#define CP_CPF_STATUS__CPF_BUSY__SHIFT 0x1f
+#define CP_CPF_BUSY_STAT__REG_BUS_FIFO_BUSY_MASK 0x1
+#define CP_CPF_BUSY_STAT__REG_BUS_FIFO_BUSY__SHIFT 0x0
+#define CP_CPF_BUSY_STAT__CSF_RING_BUSY_MASK 0x2
+#define CP_CPF_BUSY_STAT__CSF_RING_BUSY__SHIFT 0x1
+#define CP_CPF_BUSY_STAT__CSF_INDIRECT1_BUSY_MASK 0x4
+#define CP_CPF_BUSY_STAT__CSF_INDIRECT1_BUSY__SHIFT 0x2
+#define CP_CPF_BUSY_STAT__CSF_INDIRECT2_BUSY_MASK 0x8
+#define CP_CPF_BUSY_STAT__CSF_INDIRECT2_BUSY__SHIFT 0x3
+#define CP_CPF_BUSY_STAT__CSF_STATE_BUSY_MASK 0x10
+#define CP_CPF_BUSY_STAT__CSF_STATE_BUSY__SHIFT 0x4
+#define CP_CPF_BUSY_STAT__CSF_CE_INDR1_BUSY_MASK 0x20
+#define CP_CPF_BUSY_STAT__CSF_CE_INDR1_BUSY__SHIFT 0x5
+#define CP_CPF_BUSY_STAT__CSF_CE_INDR2_BUSY_MASK 0x40
+#define CP_CPF_BUSY_STAT__CSF_CE_INDR2_BUSY__SHIFT 0x6
+#define CP_CPF_BUSY_STAT__CSF_ARBITER_BUSY_MASK 0x80
+#define CP_CPF_BUSY_STAT__CSF_ARBITER_BUSY__SHIFT 0x7
+#define CP_CPF_BUSY_STAT__CSF_INPUT_BUSY_MASK 0x100
+#define CP_CPF_BUSY_STAT__CSF_INPUT_BUSY__SHIFT 0x8
+#define CP_CPF_BUSY_STAT__OUTSTANDING_READ_TAGS_MASK 0x200
+#define CP_CPF_BUSY_STAT__OUTSTANDING_READ_TAGS__SHIFT 0x9
+#define CP_CPF_BUSY_STAT__HPD_PROCESSING_EOP_BUSY_MASK 0x800
+#define CP_CPF_BUSY_STAT__HPD_PROCESSING_EOP_BUSY__SHIFT 0xb
+#define CP_CPF_BUSY_STAT__HQD_DISPATCH_BUSY_MASK 0x1000
+#define CP_CPF_BUSY_STAT__HQD_DISPATCH_BUSY__SHIFT 0xc
+#define CP_CPF_BUSY_STAT__HQD_IQ_TIMER_BUSY_MASK 0x2000
+#define CP_CPF_BUSY_STAT__HQD_IQ_TIMER_BUSY__SHIFT 0xd
+#define CP_CPF_BUSY_STAT__HQD_DMA_OFFLOAD_BUSY_MASK 0x4000
+#define CP_CPF_BUSY_STAT__HQD_DMA_OFFLOAD_BUSY__SHIFT 0xe
+#define CP_CPF_BUSY_STAT__HQD_WAIT_SEMAPHORE_BUSY_MASK 0x8000
+#define CP_CPF_BUSY_STAT__HQD_WAIT_SEMAPHORE_BUSY__SHIFT 0xf
+#define CP_CPF_BUSY_STAT__HQD_SIGNAL_SEMAPHORE_BUSY_MASK 0x10000
+#define CP_CPF_BUSY_STAT__HQD_SIGNAL_SEMAPHORE_BUSY__SHIFT 0x10
+#define CP_CPF_BUSY_STAT__HQD_MESSAGE_BUSY_MASK 0x20000
+#define CP_CPF_BUSY_STAT__HQD_MESSAGE_BUSY__SHIFT 0x11
+#define CP_CPF_BUSY_STAT__HQD_PQ_FETCHER_BUSY_MASK 0x40000
+#define CP_CPF_BUSY_STAT__HQD_PQ_FETCHER_BUSY__SHIFT 0x12
+#define CP_CPF_BUSY_STAT__HQD_IB_FETCHER_BUSY_MASK 0x80000
+#define CP_CPF_BUSY_STAT__HQD_IB_FETCHER_BUSY__SHIFT 0x13
+#define CP_CPF_BUSY_STAT__HQD_IQ_FETCHER_BUSY_MASK 0x100000
+#define CP_CPF_BUSY_STAT__HQD_IQ_FETCHER_BUSY__SHIFT 0x14
+#define CP_CPF_BUSY_STAT__HQD_EOP_FETCHER_BUSY_MASK 0x200000
+#define CP_CPF_BUSY_STAT__HQD_EOP_FETCHER_BUSY__SHIFT 0x15
+#define CP_CPF_BUSY_STAT__HQD_CONSUMED_RPTR_BUSY_MASK 0x400000
+#define CP_CPF_BUSY_STAT__HQD_CONSUMED_RPTR_BUSY__SHIFT 0x16
+#define CP_CPF_BUSY_STAT__HQD_FETCHER_ARB_BUSY_MASK 0x800000
+#define CP_CPF_BUSY_STAT__HQD_FETCHER_ARB_BUSY__SHIFT 0x17
+#define CP_CPF_BUSY_STAT__HQD_ROQ_ALIGN_BUSY_MASK 0x1000000
+#define CP_CPF_BUSY_STAT__HQD_ROQ_ALIGN_BUSY__SHIFT 0x18
+#define CP_CPF_BUSY_STAT__HQD_ROQ_EOP_BUSY_MASK 0x2000000
+#define CP_CPF_BUSY_STAT__HQD_ROQ_EOP_BUSY__SHIFT 0x19
+#define CP_CPF_BUSY_STAT__HQD_ROQ_IQ_BUSY_MASK 0x4000000
+#define CP_CPF_BUSY_STAT__HQD_ROQ_IQ_BUSY__SHIFT 0x1a
+#define CP_CPF_BUSY_STAT__HQD_ROQ_PQ_BUSY_MASK 0x8000000
+#define CP_CPF_BUSY_STAT__HQD_ROQ_PQ_BUSY__SHIFT 0x1b
+#define CP_CPF_BUSY_STAT__HQD_ROQ_IB_BUSY_MASK 0x10000000
+#define CP_CPF_BUSY_STAT__HQD_ROQ_IB_BUSY__SHIFT 0x1c
+#define CP_CPF_BUSY_STAT__HQD_WPTR_POLL_BUSY_MASK 0x20000000
+#define CP_CPF_BUSY_STAT__HQD_WPTR_POLL_BUSY__SHIFT 0x1d
+#define CP_CPF_BUSY_STAT__HQD_PQ_BUSY_MASK 0x40000000
+#define CP_CPF_BUSY_STAT__HQD_PQ_BUSY__SHIFT 0x1e
+#define CP_CPF_BUSY_STAT__HQD_IB_BUSY_MASK 0x80000000
+#define CP_CPF_BUSY_STAT__HQD_IB_BUSY__SHIFT 0x1f
+#define CP_CPF_STALLED_STAT1__RING_FETCHING_DATA_MASK 0x1
+#define CP_CPF_STALLED_STAT1__RING_FETCHING_DATA__SHIFT 0x0
+#define CP_CPF_STALLED_STAT1__INDR1_FETCHING_DATA_MASK 0x2
+#define CP_CPF_STALLED_STAT1__INDR1_FETCHING_DATA__SHIFT 0x1
+#define CP_CPF_STALLED_STAT1__INDR2_FETCHING_DATA_MASK 0x4
+#define CP_CPF_STALLED_STAT1__INDR2_FETCHING_DATA__SHIFT 0x2
+#define CP_CPF_STALLED_STAT1__STATE_FETCHING_DATA_MASK 0x8
+#define CP_CPF_STALLED_STAT1__STATE_FETCHING_DATA__SHIFT 0x3
+#define CP_CPF_STALLED_STAT1__TCIU_WAITING_ON_FREE_MASK 0x20
+#define CP_CPF_STALLED_STAT1__TCIU_WAITING_ON_FREE__SHIFT 0x5
+#define CP_CPF_STALLED_STAT1__TCIU_WAITING_ON_TAGS_MASK 0x40
+#define CP_CPF_STALLED_STAT1__TCIU_WAITING_ON_TAGS__SHIFT 0x6
+#define CP_CPF_STALLED_STAT1__ATCL2IU_WAITING_ON_FREE_MASK 0x80
+#define CP_CPF_STALLED_STAT1__ATCL2IU_WAITING_ON_FREE__SHIFT 0x7
+#define CP_CPF_STALLED_STAT1__ATCL2IU_WAITING_ON_TAGS_MASK 0x100
+#define CP_CPF_STALLED_STAT1__ATCL2IU_WAITING_ON_TAGS__SHIFT 0x8
+#define CP_CPF_STALLED_STAT1__ATCL1_WAITING_ON_TRANS_MASK 0x200
+#define CP_CPF_STALLED_STAT1__ATCL1_WAITING_ON_TRANS__SHIFT 0x9
+#define CP_CPC_GRBM_FREE_COUNT__FREE_COUNT_MASK 0x3f
+#define CP_CPC_GRBM_FREE_COUNT__FREE_COUNT__SHIFT 0x0
+#define CP_MEC_CNTL__MEC_INVALIDATE_ICACHE_MASK 0x10
+#define CP_MEC_CNTL__MEC_INVALIDATE_ICACHE__SHIFT 0x4
+#define CP_MEC_CNTL__MEC_ME1_PIPE0_RESET_MASK 0x10000
+#define CP_MEC_CNTL__MEC_ME1_PIPE0_RESET__SHIFT 0x10
+#define CP_MEC_CNTL__MEC_ME1_PIPE1_RESET_MASK 0x20000
+#define CP_MEC_CNTL__MEC_ME1_PIPE1_RESET__SHIFT 0x11
+#define CP_MEC_CNTL__MEC_ME1_PIPE2_RESET_MASK 0x40000
+#define CP_MEC_CNTL__MEC_ME1_PIPE2_RESET__SHIFT 0x12
+#define CP_MEC_CNTL__MEC_ME1_PIPE3_RESET_MASK 0x80000
+#define CP_MEC_CNTL__MEC_ME1_PIPE3_RESET__SHIFT 0x13
+#define CP_MEC_CNTL__MEC_ME2_PIPE0_RESET_MASK 0x100000
+#define CP_MEC_CNTL__MEC_ME2_PIPE0_RESET__SHIFT 0x14
+#define CP_MEC_CNTL__MEC_ME2_PIPE1_RESET_MASK 0x200000
+#define CP_MEC_CNTL__MEC_ME2_PIPE1_RESET__SHIFT 0x15
+#define CP_MEC_CNTL__MEC_ME2_HALT_MASK 0x10000000
+#define CP_MEC_CNTL__MEC_ME2_HALT__SHIFT 0x1c
+#define CP_MEC_CNTL__MEC_ME2_STEP_MASK 0x20000000
+#define CP_MEC_CNTL__MEC_ME2_STEP__SHIFT 0x1d
+#define CP_MEC_CNTL__MEC_ME1_HALT_MASK 0x40000000
+#define CP_MEC_CNTL__MEC_ME1_HALT__SHIFT 0x1e
+#define CP_MEC_CNTL__MEC_ME1_STEP_MASK 0x80000000
+#define CP_MEC_CNTL__MEC_ME1_STEP__SHIFT 0x1f
+#define CP_MEC_ME1_HEADER_DUMP__HEADER_DUMP_MASK 0xffffffff
+#define CP_MEC_ME1_HEADER_DUMP__HEADER_DUMP__SHIFT 0x0
+#define CP_MEC_ME2_HEADER_DUMP__HEADER_DUMP_MASK 0xffffffff
+#define CP_MEC_ME2_HEADER_DUMP__HEADER_DUMP__SHIFT 0x0
+#define CP_CPC_SCRATCH_INDEX__SCRATCH_INDEX_MASK 0x1ff
+#define CP_CPC_SCRATCH_INDEX__SCRATCH_INDEX__SHIFT 0x0
+#define CP_CPC_SCRATCH_DATA__SCRATCH_DATA_MASK 0xffffffff
+#define CP_CPC_SCRATCH_DATA__SCRATCH_DATA__SHIFT 0x0
+#define CPG_PERFCOUNTER1_SELECT__PERF_SEL_MASK 0x3f
+#define CPG_PERFCOUNTER1_SELECT__PERF_SEL__SHIFT 0x0
+#define CPG_PERFCOUNTER1_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define CPG_PERFCOUNTER1_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define CPG_PERFCOUNTER1_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define CPG_PERFCOUNTER1_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define CPG_PERFCOUNTER0_SELECT1__PERF_SEL2_MASK 0x3f
+#define CPG_PERFCOUNTER0_SELECT1__PERF_SEL2__SHIFT 0x0
+#define CPG_PERFCOUNTER0_SELECT1__PERF_SEL3_MASK 0xfc00
+#define CPG_PERFCOUNTER0_SELECT1__PERF_SEL3__SHIFT 0xa
+#define CPG_PERFCOUNTER0_SELECT__PERF_SEL_MASK 0x3f
+#define CPG_PERFCOUNTER0_SELECT__PERF_SEL__SHIFT 0x0
+#define CPG_PERFCOUNTER0_SELECT__PERF_SEL1_MASK 0xfc00
+#define CPG_PERFCOUNTER0_SELECT__PERF_SEL1__SHIFT 0xa
+#define CPG_PERFCOUNTER0_SELECT__CNTR_MODE_MASK 0xf00000
+#define CPG_PERFCOUNTER0_SELECT__CNTR_MODE__SHIFT 0x14
+#define CPG_PERFCOUNTER0_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define CPG_PERFCOUNTER0_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define CPG_PERFCOUNTER0_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define CPG_PERFCOUNTER0_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define CPC_PERFCOUNTER1_SELECT__PERF_SEL_MASK 0x3f
+#define CPC_PERFCOUNTER1_SELECT__PERF_SEL__SHIFT 0x0
+#define CPC_PERFCOUNTER1_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define CPC_PERFCOUNTER1_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define CPC_PERFCOUNTER1_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define CPC_PERFCOUNTER1_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define CPC_PERFCOUNTER0_SELECT1__PERF_SEL2_MASK 0x3f
+#define CPC_PERFCOUNTER0_SELECT1__PERF_SEL2__SHIFT 0x0
+#define CPC_PERFCOUNTER0_SELECT1__PERF_SEL3_MASK 0xfc00
+#define CPC_PERFCOUNTER0_SELECT1__PERF_SEL3__SHIFT 0xa
+#define CPC_PERFCOUNTER0_SELECT__PERF_SEL_MASK 0x3f
+#define CPC_PERFCOUNTER0_SELECT__PERF_SEL__SHIFT 0x0
+#define CPC_PERFCOUNTER0_SELECT__PERF_SEL1_MASK 0xfc00
+#define CPC_PERFCOUNTER0_SELECT__PERF_SEL1__SHIFT 0xa
+#define CPC_PERFCOUNTER0_SELECT__CNTR_MODE_MASK 0xf00000
+#define CPC_PERFCOUNTER0_SELECT__CNTR_MODE__SHIFT 0x14
+#define CPC_PERFCOUNTER0_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define CPC_PERFCOUNTER0_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define CPC_PERFCOUNTER0_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define CPC_PERFCOUNTER0_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define CPF_PERFCOUNTER1_SELECT__PERF_SEL_MASK 0x3f
+#define CPF_PERFCOUNTER1_SELECT__PERF_SEL__SHIFT 0x0
+#define CPF_PERFCOUNTER1_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define CPF_PERFCOUNTER1_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define CPF_PERFCOUNTER1_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define CPF_PERFCOUNTER1_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define CPF_PERFCOUNTER0_SELECT1__PERF_SEL2_MASK 0x3f
+#define CPF_PERFCOUNTER0_SELECT1__PERF_SEL2__SHIFT 0x0
+#define CPF_PERFCOUNTER0_SELECT1__PERF_SEL3_MASK 0xfc00
+#define CPF_PERFCOUNTER0_SELECT1__PERF_SEL3__SHIFT 0xa
+#define CPF_PERFCOUNTER0_SELECT__PERF_SEL_MASK 0x3f
+#define CPF_PERFCOUNTER0_SELECT__PERF_SEL__SHIFT 0x0
+#define CPF_PERFCOUNTER0_SELECT__PERF_SEL1_MASK 0xfc00
+#define CPF_PERFCOUNTER0_SELECT__PERF_SEL1__SHIFT 0xa
+#define CPF_PERFCOUNTER0_SELECT__CNTR_MODE_MASK 0xf00000
+#define CPF_PERFCOUNTER0_SELECT__CNTR_MODE__SHIFT 0x14
+#define CPF_PERFCOUNTER0_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define CPF_PERFCOUNTER0_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define CPF_PERFCOUNTER0_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define CPF_PERFCOUNTER0_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define CP_CPC_HALT_HYST_COUNT__COUNT_MASK 0xf
+#define CP_CPC_HALT_HYST_COUNT__COUNT__SHIFT 0x0
+#define CP_DRAW_OBJECT__OBJECT_MASK 0xffffffff
+#define CP_DRAW_OBJECT__OBJECT__SHIFT 0x0
+#define CP_DRAW_OBJECT_COUNTER__COUNT_MASK 0xffff
+#define CP_DRAW_OBJECT_COUNTER__COUNT__SHIFT 0x0
+#define CP_DRAW_WINDOW_MASK_HI__WINDOW_MASK_HI_MASK 0xffffffff
+#define CP_DRAW_WINDOW_MASK_HI__WINDOW_MASK_HI__SHIFT 0x0
+#define CP_DRAW_WINDOW_HI__WINDOW_HI_MASK 0xffffffff
+#define CP_DRAW_WINDOW_HI__WINDOW_HI__SHIFT 0x0
+#define CP_DRAW_WINDOW_LO__MIN_MASK 0xffff
+#define CP_DRAW_WINDOW_LO__MIN__SHIFT 0x0
+#define CP_DRAW_WINDOW_LO__MAX_MASK 0xffff0000
+#define CP_DRAW_WINDOW_LO__MAX__SHIFT 0x10
+#define CP_DRAW_WINDOW_CNTL__DISABLE_DRAW_WINDOW_LO_MAX_MASK 0x1
+#define CP_DRAW_WINDOW_CNTL__DISABLE_DRAW_WINDOW_LO_MAX__SHIFT 0x0
+#define CP_DRAW_WINDOW_CNTL__DISABLE_DRAW_WINDOW_LO_MIN_MASK 0x2
+#define CP_DRAW_WINDOW_CNTL__DISABLE_DRAW_WINDOW_LO_MIN__SHIFT 0x1
+#define CP_DRAW_WINDOW_CNTL__DISABLE_DRAW_WINDOW_HI_MASK 0x4
+#define CP_DRAW_WINDOW_CNTL__DISABLE_DRAW_WINDOW_HI__SHIFT 0x2
+#define CP_DRAW_WINDOW_CNTL__MODE_MASK 0x100
+#define CP_DRAW_WINDOW_CNTL__MODE__SHIFT 0x8
+#define CP_PRT_LOD_STATS_CNTL0__BU_SIZE_MASK 0xffffffff
+#define CP_PRT_LOD_STATS_CNTL0__BU_SIZE__SHIFT 0x0
+#define CP_PRT_LOD_STATS_CNTL1__BASE_LO_MASK 0xffffffff
+#define CP_PRT_LOD_STATS_CNTL1__BASE_LO__SHIFT 0x0
+#define CP_PRT_LOD_STATS_CNTL2__BASE_HI_MASK 0x3
+#define CP_PRT_LOD_STATS_CNTL2__BASE_HI__SHIFT 0x0
+#define CP_PRT_LOD_STATS_CNTL2__INTERVAL_MASK 0x3fc
+#define CP_PRT_LOD_STATS_CNTL2__INTERVAL__SHIFT 0x2
+#define CP_PRT_LOD_STATS_CNTL2__RESET_CNT_MASK 0x3fc00
+#define CP_PRT_LOD_STATS_CNTL2__RESET_CNT__SHIFT 0xa
+#define CP_PRT_LOD_STATS_CNTL2__RESET_FORCE_MASK 0x40000
+#define CP_PRT_LOD_STATS_CNTL2__RESET_FORCE__SHIFT 0x12
+#define CP_PRT_LOD_STATS_CNTL2__REPORT_AND_RESET_MASK 0x80000
+#define CP_PRT_LOD_STATS_CNTL2__REPORT_AND_RESET__SHIFT 0x13
+#define CP_PRT_LOD_STATS_CNTL2__MC_VMID_MASK 0x7800000
+#define CP_PRT_LOD_STATS_CNTL2__MC_VMID__SHIFT 0x17
+#define CP_PRT_LOD_STATS_CNTL2__CACHE_POLICY_MASK 0x10000000
+#define CP_PRT_LOD_STATS_CNTL2__CACHE_POLICY__SHIFT 0x1c
+#define CP_PRT_LOD_STATS_CNTL2__MTYPE_MASK 0xc0000000
+#define CP_PRT_LOD_STATS_CNTL2__MTYPE__SHIFT 0x1e
+#define CP_CE_COMPARE_COUNT__COMPARE_COUNT_MASK 0xffffffff
+#define CP_CE_COMPARE_COUNT__COMPARE_COUNT__SHIFT 0x0
+#define CP_CE_DE_COUNT__DRAW_ENGINE_COUNT_MASK 0xffffffff
+#define CP_CE_DE_COUNT__DRAW_ENGINE_COUNT__SHIFT 0x0
+#define CP_DE_CE_COUNT__CONST_ENGINE_COUNT_MASK 0xffffffff
+#define CP_DE_CE_COUNT__CONST_ENGINE_COUNT__SHIFT 0x0
+#define CP_DE_LAST_INVAL_COUNT__LAST_INVAL_COUNT_MASK 0xffffffff
+#define CP_DE_LAST_INVAL_COUNT__LAST_INVAL_COUNT__SHIFT 0x0
+#define CP_DE_DE_COUNT__DRAW_ENGINE_COUNT_MASK 0xffffffff
+#define CP_DE_DE_COUNT__DRAW_ENGINE_COUNT__SHIFT 0x0
+#define CP_EOP_DONE_EVENT_CNTL__WBINV_TC_OP_MASK 0x7f
+#define CP_EOP_DONE_EVENT_CNTL__WBINV_TC_OP__SHIFT 0x0
+#define CP_EOP_DONE_EVENT_CNTL__WBINV_ACTION_ENA_MASK 0x3f000
+#define CP_EOP_DONE_EVENT_CNTL__WBINV_ACTION_ENA__SHIFT 0xc
+#define CP_EOP_DONE_EVENT_CNTL__CACHE_CONTROL_MASK 0x2000000
+#define CP_EOP_DONE_EVENT_CNTL__CACHE_CONTROL__SHIFT 0x19
+#define CP_EOP_DONE_EVENT_CNTL__MTYPE_MASK 0x18000000
+#define CP_EOP_DONE_EVENT_CNTL__MTYPE__SHIFT 0x1b
+#define CP_EOP_DONE_DATA_CNTL__CNTX_ID_MASK 0xffff
+#define CP_EOP_DONE_DATA_CNTL__CNTX_ID__SHIFT 0x0
+#define CP_EOP_DONE_DATA_CNTL__DST_SEL_MASK 0x30000
+#define CP_EOP_DONE_DATA_CNTL__DST_SEL__SHIFT 0x10
+#define CP_EOP_DONE_DATA_CNTL__INT_SEL_MASK 0x7000000
+#define CP_EOP_DONE_DATA_CNTL__INT_SEL__SHIFT 0x18
+#define CP_EOP_DONE_DATA_CNTL__DATA_SEL_MASK 0xe0000000
+#define CP_EOP_DONE_DATA_CNTL__DATA_SEL__SHIFT 0x1d
+#define CP_EOP_DONE_CNTX_ID__CNTX_ID_MASK 0xfffffff
+#define CP_EOP_DONE_CNTX_ID__CNTX_ID__SHIFT 0x0
+#define CP_EOP_DONE_ADDR_LO__ADDR_LO_MASK 0xfffffffc
+#define CP_EOP_DONE_ADDR_LO__ADDR_LO__SHIFT 0x2
+#define CP_EOP_DONE_ADDR_HI__ADDR_HI_MASK 0xffff
+#define CP_EOP_DONE_ADDR_HI__ADDR_HI__SHIFT 0x0
+#define CP_EOP_DONE_DATA_LO__DATA_LO_MASK 0xffffffff
+#define CP_EOP_DONE_DATA_LO__DATA_LO__SHIFT 0x0
+#define CP_EOP_DONE_DATA_HI__DATA_HI_MASK 0xffffffff
+#define CP_EOP_DONE_DATA_HI__DATA_HI__SHIFT 0x0
+#define CP_EOP_LAST_FENCE_LO__LAST_FENCE_LO_MASK 0xffffffff
+#define CP_EOP_LAST_FENCE_LO__LAST_FENCE_LO__SHIFT 0x0
+#define CP_EOP_LAST_FENCE_HI__LAST_FENCE_HI_MASK 0xffffffff
+#define CP_EOP_LAST_FENCE_HI__LAST_FENCE_HI__SHIFT 0x0
+#define CP_STREAM_OUT_ADDR_LO__STREAM_OUT_ADDR_LO_MASK 0xfffffffc
+#define CP_STREAM_OUT_ADDR_LO__STREAM_OUT_ADDR_LO__SHIFT 0x2
+#define CP_STREAM_OUT_ADDR_HI__STREAM_OUT_ADDR_HI_MASK 0xffff
+#define CP_STREAM_OUT_ADDR_HI__STREAM_OUT_ADDR_HI__SHIFT 0x0
+#define CP_NUM_PRIM_WRITTEN_COUNT0_LO__NUM_PRIM_WRITTEN_CNT0_LO_MASK 0xffffffff
+#define CP_NUM_PRIM_WRITTEN_COUNT0_LO__NUM_PRIM_WRITTEN_CNT0_LO__SHIFT 0x0
+#define CP_NUM_PRIM_WRITTEN_COUNT0_HI__NUM_PRIM_WRITTEN_CNT0_HI_MASK 0xffffffff
+#define CP_NUM_PRIM_WRITTEN_COUNT0_HI__NUM_PRIM_WRITTEN_CNT0_HI__SHIFT 0x0
+#define CP_NUM_PRIM_NEEDED_COUNT0_LO__NUM_PRIM_NEEDED_CNT0_LO_MASK 0xffffffff
+#define CP_NUM_PRIM_NEEDED_COUNT0_LO__NUM_PRIM_NEEDED_CNT0_LO__SHIFT 0x0
+#define CP_NUM_PRIM_NEEDED_COUNT0_HI__NUM_PRIM_NEEDED_CNT0_HI_MASK 0xffffffff
+#define CP_NUM_PRIM_NEEDED_COUNT0_HI__NUM_PRIM_NEEDED_CNT0_HI__SHIFT 0x0
+#define CP_NUM_PRIM_WRITTEN_COUNT1_LO__NUM_PRIM_WRITTEN_CNT1_LO_MASK 0xffffffff
+#define CP_NUM_PRIM_WRITTEN_COUNT1_LO__NUM_PRIM_WRITTEN_CNT1_LO__SHIFT 0x0
+#define CP_NUM_PRIM_WRITTEN_COUNT1_HI__NUM_PRIM_WRITTEN_CNT1_HI_MASK 0xffffffff
+#define CP_NUM_PRIM_WRITTEN_COUNT1_HI__NUM_PRIM_WRITTEN_CNT1_HI__SHIFT 0x0
+#define CP_NUM_PRIM_NEEDED_COUNT1_LO__NUM_PRIM_NEEDED_CNT1_LO_MASK 0xffffffff
+#define CP_NUM_PRIM_NEEDED_COUNT1_LO__NUM_PRIM_NEEDED_CNT1_LO__SHIFT 0x0
+#define CP_NUM_PRIM_NEEDED_COUNT1_HI__NUM_PRIM_NEEDED_CNT1_HI_MASK 0xffffffff
+#define CP_NUM_PRIM_NEEDED_COUNT1_HI__NUM_PRIM_NEEDED_CNT1_HI__SHIFT 0x0
+#define CP_NUM_PRIM_WRITTEN_COUNT2_LO__NUM_PRIM_WRITTEN_CNT2_LO_MASK 0xffffffff
+#define CP_NUM_PRIM_WRITTEN_COUNT2_LO__NUM_PRIM_WRITTEN_CNT2_LO__SHIFT 0x0
+#define CP_NUM_PRIM_WRITTEN_COUNT2_HI__NUM_PRIM_WRITTEN_CNT2_HI_MASK 0xffffffff
+#define CP_NUM_PRIM_WRITTEN_COUNT2_HI__NUM_PRIM_WRITTEN_CNT2_HI__SHIFT 0x0
+#define CP_NUM_PRIM_NEEDED_COUNT2_LO__NUM_PRIM_NEEDED_CNT2_LO_MASK 0xffffffff
+#define CP_NUM_PRIM_NEEDED_COUNT2_LO__NUM_PRIM_NEEDED_CNT2_LO__SHIFT 0x0
+#define CP_NUM_PRIM_NEEDED_COUNT2_HI__NUM_PRIM_NEEDED_CNT2_HI_MASK 0xffffffff
+#define CP_NUM_PRIM_NEEDED_COUNT2_HI__NUM_PRIM_NEEDED_CNT2_HI__SHIFT 0x0
+#define CP_NUM_PRIM_WRITTEN_COUNT3_LO__NUM_PRIM_WRITTEN_CNT3_LO_MASK 0xffffffff
+#define CP_NUM_PRIM_WRITTEN_COUNT3_LO__NUM_PRIM_WRITTEN_CNT3_LO__SHIFT 0x0
+#define CP_NUM_PRIM_WRITTEN_COUNT3_HI__NUM_PRIM_WRITTEN_CNT3_HI_MASK 0xffffffff
+#define CP_NUM_PRIM_WRITTEN_COUNT3_HI__NUM_PRIM_WRITTEN_CNT3_HI__SHIFT 0x0
+#define CP_NUM_PRIM_NEEDED_COUNT3_LO__NUM_PRIM_NEEDED_CNT3_LO_MASK 0xffffffff
+#define CP_NUM_PRIM_NEEDED_COUNT3_LO__NUM_PRIM_NEEDED_CNT3_LO__SHIFT 0x0
+#define CP_NUM_PRIM_NEEDED_COUNT3_HI__NUM_PRIM_NEEDED_CNT3_HI_MASK 0xffffffff
+#define CP_NUM_PRIM_NEEDED_COUNT3_HI__NUM_PRIM_NEEDED_CNT3_HI__SHIFT 0x0
+#define CP_PIPE_STATS_ADDR_LO__PIPE_STATS_ADDR_LO_MASK 0xfffffffc
+#define CP_PIPE_STATS_ADDR_LO__PIPE_STATS_ADDR_LO__SHIFT 0x2
+#define CP_PIPE_STATS_ADDR_HI__PIPE_STATS_ADDR_HI_MASK 0xffff
+#define CP_PIPE_STATS_ADDR_HI__PIPE_STATS_ADDR_HI__SHIFT 0x0
+#define CP_VGT_IAVERT_COUNT_LO__IAVERT_COUNT_LO_MASK 0xffffffff
+#define CP_VGT_IAVERT_COUNT_LO__IAVERT_COUNT_LO__SHIFT 0x0
+#define CP_VGT_IAVERT_COUNT_HI__IAVERT_COUNT_HI_MASK 0xffffffff
+#define CP_VGT_IAVERT_COUNT_HI__IAVERT_COUNT_HI__SHIFT 0x0
+#define CP_VGT_IAPRIM_COUNT_LO__IAPRIM_COUNT_LO_MASK 0xffffffff
+#define CP_VGT_IAPRIM_COUNT_LO__IAPRIM_COUNT_LO__SHIFT 0x0
+#define CP_VGT_IAPRIM_COUNT_HI__IAPRIM_COUNT_HI_MASK 0xffffffff
+#define CP_VGT_IAPRIM_COUNT_HI__IAPRIM_COUNT_HI__SHIFT 0x0
+#define CP_VGT_GSPRIM_COUNT_LO__GSPRIM_COUNT_LO_MASK 0xffffffff
+#define CP_VGT_GSPRIM_COUNT_LO__GSPRIM_COUNT_LO__SHIFT 0x0
+#define CP_VGT_GSPRIM_COUNT_HI__GSPRIM_COUNT_HI_MASK 0xffffffff
+#define CP_VGT_GSPRIM_COUNT_HI__GSPRIM_COUNT_HI__SHIFT 0x0
+#define CP_VGT_VSINVOC_COUNT_LO__VSINVOC_COUNT_LO_MASK 0xffffffff
+#define CP_VGT_VSINVOC_COUNT_LO__VSINVOC_COUNT_LO__SHIFT 0x0
+#define CP_VGT_VSINVOC_COUNT_HI__VSINVOC_COUNT_HI_MASK 0xffffffff
+#define CP_VGT_VSINVOC_COUNT_HI__VSINVOC_COUNT_HI__SHIFT 0x0
+#define CP_VGT_GSINVOC_COUNT_LO__GSINVOC_COUNT_LO_MASK 0xffffffff
+#define CP_VGT_GSINVOC_COUNT_LO__GSINVOC_COUNT_LO__SHIFT 0x0
+#define CP_VGT_GSINVOC_COUNT_HI__GSINVOC_COUNT_HI_MASK 0xffffffff
+#define CP_VGT_GSINVOC_COUNT_HI__GSINVOC_COUNT_HI__SHIFT 0x0
+#define CP_VGT_HSINVOC_COUNT_LO__HSINVOC_COUNT_LO_MASK 0xffffffff
+#define CP_VGT_HSINVOC_COUNT_LO__HSINVOC_COUNT_LO__SHIFT 0x0
+#define CP_VGT_HSINVOC_COUNT_HI__HSINVOC_COUNT_HI_MASK 0xffffffff
+#define CP_VGT_HSINVOC_COUNT_HI__HSINVOC_COUNT_HI__SHIFT 0x0
+#define CP_VGT_DSINVOC_COUNT_LO__DSINVOC_COUNT_LO_MASK 0xffffffff
+#define CP_VGT_DSINVOC_COUNT_LO__DSINVOC_COUNT_LO__SHIFT 0x0
+#define CP_VGT_DSINVOC_COUNT_HI__DSINVOC_COUNT_HI_MASK 0xffffffff
+#define CP_VGT_DSINVOC_COUNT_HI__DSINVOC_COUNT_HI__SHIFT 0x0
+#define CP_PA_CINVOC_COUNT_LO__CINVOC_COUNT_LO_MASK 0xffffffff
+#define CP_PA_CINVOC_COUNT_LO__CINVOC_COUNT_LO__SHIFT 0x0
+#define CP_PA_CINVOC_COUNT_HI__CINVOC_COUNT_HI_MASK 0xffffffff
+#define CP_PA_CINVOC_COUNT_HI__CINVOC_COUNT_HI__SHIFT 0x0
+#define CP_PA_CPRIM_COUNT_LO__CPRIM_COUNT_LO_MASK 0xffffffff
+#define CP_PA_CPRIM_COUNT_LO__CPRIM_COUNT_LO__SHIFT 0x0
+#define CP_PA_CPRIM_COUNT_HI__CPRIM_COUNT_HI_MASK 0xffffffff
+#define CP_PA_CPRIM_COUNT_HI__CPRIM_COUNT_HI__SHIFT 0x0
+#define CP_SC_PSINVOC_COUNT0_LO__PSINVOC_COUNT0_LO_MASK 0xffffffff
+#define CP_SC_PSINVOC_COUNT0_LO__PSINVOC_COUNT0_LO__SHIFT 0x0
+#define CP_SC_PSINVOC_COUNT0_HI__PSINVOC_COUNT0_HI_MASK 0xffffffff
+#define CP_SC_PSINVOC_COUNT0_HI__PSINVOC_COUNT0_HI__SHIFT 0x0
+#define CP_SC_PSINVOC_COUNT1_LO__OBSOLETE_MASK 0xffffffff
+#define CP_SC_PSINVOC_COUNT1_LO__OBSOLETE__SHIFT 0x0
+#define CP_SC_PSINVOC_COUNT1_HI__OBSOLETE_MASK 0xffffffff
+#define CP_SC_PSINVOC_COUNT1_HI__OBSOLETE__SHIFT 0x0
+#define CP_VGT_CSINVOC_COUNT_LO__CSINVOC_COUNT_LO_MASK 0xffffffff
+#define CP_VGT_CSINVOC_COUNT_LO__CSINVOC_COUNT_LO__SHIFT 0x0
+#define CP_VGT_CSINVOC_COUNT_HI__CSINVOC_COUNT_HI_MASK 0xffffffff
+#define CP_VGT_CSINVOC_COUNT_HI__CSINVOC_COUNT_HI__SHIFT 0x0
+#define CP_PIPE_STATS_CONTROL__CACHE_CONTROL_MASK 0x2000000
+#define CP_PIPE_STATS_CONTROL__CACHE_CONTROL__SHIFT 0x19
+#define CP_PIPE_STATS_CONTROL__MTYPE_MASK 0x18000000
+#define CP_PIPE_STATS_CONTROL__MTYPE__SHIFT 0x1b
+#define CP_STREAM_OUT_CONTROL__CACHE_CONTROL_MASK 0x2000000
+#define CP_STREAM_OUT_CONTROL__CACHE_CONTROL__SHIFT 0x19
+#define CP_STREAM_OUT_CONTROL__MTYPE_MASK 0x18000000
+#define CP_STREAM_OUT_CONTROL__MTYPE__SHIFT 0x1b
+#define CP_STRMOUT_CNTL__OFFSET_UPDATE_DONE_MASK 0x1
+#define CP_STRMOUT_CNTL__OFFSET_UPDATE_DONE__SHIFT 0x0
+#define SCRATCH_REG0__SCRATCH_REG0_MASK 0xffffffff
+#define SCRATCH_REG0__SCRATCH_REG0__SHIFT 0x0
+#define SCRATCH_REG1__SCRATCH_REG1_MASK 0xffffffff
+#define SCRATCH_REG1__SCRATCH_REG1__SHIFT 0x0
+#define SCRATCH_REG2__SCRATCH_REG2_MASK 0xffffffff
+#define SCRATCH_REG2__SCRATCH_REG2__SHIFT 0x0
+#define SCRATCH_REG3__SCRATCH_REG3_MASK 0xffffffff
+#define SCRATCH_REG3__SCRATCH_REG3__SHIFT 0x0
+#define SCRATCH_REG4__SCRATCH_REG4_MASK 0xffffffff
+#define SCRATCH_REG4__SCRATCH_REG4__SHIFT 0x0
+#define SCRATCH_REG5__SCRATCH_REG5_MASK 0xffffffff
+#define SCRATCH_REG5__SCRATCH_REG5__SHIFT 0x0
+#define SCRATCH_REG6__SCRATCH_REG6_MASK 0xffffffff
+#define SCRATCH_REG6__SCRATCH_REG6__SHIFT 0x0
+#define SCRATCH_REG7__SCRATCH_REG7_MASK 0xffffffff
+#define SCRATCH_REG7__SCRATCH_REG7__SHIFT 0x0
+#define SCRATCH_UMSK__OBSOLETE_UMSK_MASK 0xff
+#define SCRATCH_UMSK__OBSOLETE_UMSK__SHIFT 0x0
+#define SCRATCH_UMSK__OBSOLETE_SWAP_MASK 0x30000
+#define SCRATCH_UMSK__OBSOLETE_SWAP__SHIFT 0x10
+#define SCRATCH_ADDR__OBSOLETE_ADDR_MASK 0xffffffff
+#define SCRATCH_ADDR__OBSOLETE_ADDR__SHIFT 0x0
+#define CP_PFP_ATOMIC_PREOP_LO__ATOMIC_PREOP_LO_MASK 0xffffffff
+#define CP_PFP_ATOMIC_PREOP_LO__ATOMIC_PREOP_LO__SHIFT 0x0
+#define CP_PFP_ATOMIC_PREOP_HI__ATOMIC_PREOP_HI_MASK 0xffffffff
+#define CP_PFP_ATOMIC_PREOP_HI__ATOMIC_PREOP_HI__SHIFT 0x0
+#define CP_PFP_GDS_ATOMIC0_PREOP_LO__GDS_ATOMIC0_PREOP_LO_MASK 0xffffffff
+#define CP_PFP_GDS_ATOMIC0_PREOP_LO__GDS_ATOMIC0_PREOP_LO__SHIFT 0x0
+#define CP_PFP_GDS_ATOMIC0_PREOP_HI__GDS_ATOMIC0_PREOP_HI_MASK 0xffffffff
+#define CP_PFP_GDS_ATOMIC0_PREOP_HI__GDS_ATOMIC0_PREOP_HI__SHIFT 0x0
+#define CP_PFP_GDS_ATOMIC1_PREOP_LO__GDS_ATOMIC1_PREOP_LO_MASK 0xffffffff
+#define CP_PFP_GDS_ATOMIC1_PREOP_LO__GDS_ATOMIC1_PREOP_LO__SHIFT 0x0
+#define CP_PFP_GDS_ATOMIC1_PREOP_HI__GDS_ATOMIC1_PREOP_HI_MASK 0xffffffff
+#define CP_PFP_GDS_ATOMIC1_PREOP_HI__GDS_ATOMIC1_PREOP_HI__SHIFT 0x0
+#define CP_APPEND_ADDR_LO__MEM_ADDR_LO_MASK 0xfffffffc
+#define CP_APPEND_ADDR_LO__MEM_ADDR_LO__SHIFT 0x2
+#define CP_APPEND_ADDR_HI__MEM_ADDR_HI_MASK 0xffff
+#define CP_APPEND_ADDR_HI__MEM_ADDR_HI__SHIFT 0x0
+#define CP_APPEND_ADDR_HI__CS_PS_SEL_MASK 0x10000
+#define CP_APPEND_ADDR_HI__CS_PS_SEL__SHIFT 0x10
+#define CP_APPEND_ADDR_HI__CACHE_POLICY_MASK 0x2000000
+#define CP_APPEND_ADDR_HI__CACHE_POLICY__SHIFT 0x19
+#define CP_APPEND_ADDR_HI__MTYPE_MASK 0x18000000
+#define CP_APPEND_ADDR_HI__MTYPE__SHIFT 0x1b
+#define CP_APPEND_ADDR_HI__COMMAND_MASK 0xe0000000
+#define CP_APPEND_ADDR_HI__COMMAND__SHIFT 0x1d
+#define CP_APPEND_DATA__DATA_MASK 0xffffffff
+#define CP_APPEND_DATA__DATA__SHIFT 0x0
+#define CP_APPEND_LAST_CS_FENCE__LAST_FENCE_MASK 0xffffffff
+#define CP_APPEND_LAST_CS_FENCE__LAST_FENCE__SHIFT 0x0
+#define CP_APPEND_LAST_PS_FENCE__LAST_FENCE_MASK 0xffffffff
+#define CP_APPEND_LAST_PS_FENCE__LAST_FENCE__SHIFT 0x0
+#define CP_ATOMIC_PREOP_LO__ATOMIC_PREOP_LO_MASK 0xffffffff
+#define CP_ATOMIC_PREOP_LO__ATOMIC_PREOP_LO__SHIFT 0x0
+#define CP_ME_ATOMIC_PREOP_LO__ATOMIC_PREOP_LO_MASK 0xffffffff
+#define CP_ME_ATOMIC_PREOP_LO__ATOMIC_PREOP_LO__SHIFT 0x0
+#define CP_ATOMIC_PREOP_HI__ATOMIC_PREOP_HI_MASK 0xffffffff
+#define CP_ATOMIC_PREOP_HI__ATOMIC_PREOP_HI__SHIFT 0x0
+#define CP_ME_ATOMIC_PREOP_HI__ATOMIC_PREOP_HI_MASK 0xffffffff
+#define CP_ME_ATOMIC_PREOP_HI__ATOMIC_PREOP_HI__SHIFT 0x0
+#define CP_GDS_ATOMIC0_PREOP_LO__GDS_ATOMIC0_PREOP_LO_MASK 0xffffffff
+#define CP_GDS_ATOMIC0_PREOP_LO__GDS_ATOMIC0_PREOP_LO__SHIFT 0x0
+#define CP_ME_GDS_ATOMIC0_PREOP_LO__GDS_ATOMIC0_PREOP_LO_MASK 0xffffffff
+#define CP_ME_GDS_ATOMIC0_PREOP_LO__GDS_ATOMIC0_PREOP_LO__SHIFT 0x0
+#define CP_GDS_ATOMIC0_PREOP_HI__GDS_ATOMIC0_PREOP_HI_MASK 0xffffffff
+#define CP_GDS_ATOMIC0_PREOP_HI__GDS_ATOMIC0_PREOP_HI__SHIFT 0x0
+#define CP_ME_GDS_ATOMIC0_PREOP_HI__GDS_ATOMIC0_PREOP_HI_MASK 0xffffffff
+#define CP_ME_GDS_ATOMIC0_PREOP_HI__GDS_ATOMIC0_PREOP_HI__SHIFT 0x0
+#define CP_GDS_ATOMIC1_PREOP_LO__GDS_ATOMIC1_PREOP_LO_MASK 0xffffffff
+#define CP_GDS_ATOMIC1_PREOP_LO__GDS_ATOMIC1_PREOP_LO__SHIFT 0x0
+#define CP_ME_GDS_ATOMIC1_PREOP_LO__GDS_ATOMIC1_PREOP_LO_MASK 0xffffffff
+#define CP_ME_GDS_ATOMIC1_PREOP_LO__GDS_ATOMIC1_PREOP_LO__SHIFT 0x0
+#define CP_GDS_ATOMIC1_PREOP_HI__GDS_ATOMIC1_PREOP_HI_MASK 0xffffffff
+#define CP_GDS_ATOMIC1_PREOP_HI__GDS_ATOMIC1_PREOP_HI__SHIFT 0x0
+#define CP_ME_GDS_ATOMIC1_PREOP_HI__GDS_ATOMIC1_PREOP_HI_MASK 0xffffffff
+#define CP_ME_GDS_ATOMIC1_PREOP_HI__GDS_ATOMIC1_PREOP_HI__SHIFT 0x0
+#define CP_ME_MC_WADDR_LO__ME_MC_WADDR_SWAP_MASK 0x3
+#define CP_ME_MC_WADDR_LO__ME_MC_WADDR_SWAP__SHIFT 0x0
+#define CP_ME_MC_WADDR_LO__ME_MC_WADDR_LO_MASK 0xfffffffc
+#define CP_ME_MC_WADDR_LO__ME_MC_WADDR_LO__SHIFT 0x2
+#define CP_ME_MC_WADDR_HI__ME_MC_WADDR_HI_MASK 0xffff
+#define CP_ME_MC_WADDR_HI__ME_MC_WADDR_HI__SHIFT 0x0
+#define CP_ME_MC_WADDR_HI__MTYPE_MASK 0x300000
+#define CP_ME_MC_WADDR_HI__MTYPE__SHIFT 0x14
+#define CP_ME_MC_WADDR_HI__CACHE_POLICY_MASK 0x400000
+#define CP_ME_MC_WADDR_HI__CACHE_POLICY__SHIFT 0x16
+#define CP_ME_MC_WDATA_LO__ME_MC_WDATA_LO_MASK 0xffffffff
+#define CP_ME_MC_WDATA_LO__ME_MC_WDATA_LO__SHIFT 0x0
+#define CP_ME_MC_WDATA_HI__ME_MC_WDATA_HI_MASK 0xffffffff
+#define CP_ME_MC_WDATA_HI__ME_MC_WDATA_HI__SHIFT 0x0
+#define CP_ME_MC_RADDR_LO__ME_MC_RADDR_SWAP_MASK 0x3
+#define CP_ME_MC_RADDR_LO__ME_MC_RADDR_SWAP__SHIFT 0x0
+#define CP_ME_MC_RADDR_LO__ME_MC_RADDR_LO_MASK 0xfffffffc
+#define CP_ME_MC_RADDR_LO__ME_MC_RADDR_LO__SHIFT 0x2
+#define CP_ME_MC_RADDR_HI__ME_MC_RADDR_HI_MASK 0xffff
+#define CP_ME_MC_RADDR_HI__ME_MC_RADDR_HI__SHIFT 0x0
+#define CP_ME_MC_RADDR_HI__MTYPE_MASK 0x300000
+#define CP_ME_MC_RADDR_HI__MTYPE__SHIFT 0x14
+#define CP_ME_MC_RADDR_HI__CACHE_POLICY_MASK 0x400000
+#define CP_ME_MC_RADDR_HI__CACHE_POLICY__SHIFT 0x16
+#define CP_SEM_WAIT_TIMER__SEM_WAIT_TIMER_MASK 0xffffffff
+#define CP_SEM_WAIT_TIMER__SEM_WAIT_TIMER__SHIFT 0x0
+#define CP_SIG_SEM_ADDR_LO__SEM_ADDR_SWAP_MASK 0x3
+#define CP_SIG_SEM_ADDR_LO__SEM_ADDR_SWAP__SHIFT 0x0
+#define CP_SIG_SEM_ADDR_LO__SEM_ADDR_LO_MASK 0xfffffff8
+#define CP_SIG_SEM_ADDR_LO__SEM_ADDR_LO__SHIFT 0x3
+#define CP_SIG_SEM_ADDR_HI__SEM_ADDR_HI_MASK 0xffff
+#define CP_SIG_SEM_ADDR_HI__SEM_ADDR_HI__SHIFT 0x0
+#define CP_SIG_SEM_ADDR_HI__SEM_USE_MAILBOX_MASK 0x10000
+#define CP_SIG_SEM_ADDR_HI__SEM_USE_MAILBOX__SHIFT 0x10
+#define CP_SIG_SEM_ADDR_HI__SEM_SIGNAL_TYPE_MASK 0x100000
+#define CP_SIG_SEM_ADDR_HI__SEM_SIGNAL_TYPE__SHIFT 0x14
+#define CP_SIG_SEM_ADDR_HI__SEM_CLIENT_CODE_MASK 0x3000000
+#define CP_SIG_SEM_ADDR_HI__SEM_CLIENT_CODE__SHIFT 0x18
+#define CP_SIG_SEM_ADDR_HI__SEM_SELECT_MASK 0xe0000000
+#define CP_SIG_SEM_ADDR_HI__SEM_SELECT__SHIFT 0x1d
+#define CP_WAIT_SEM_ADDR_LO__SEM_ADDR_SWAP_MASK 0x3
+#define CP_WAIT_SEM_ADDR_LO__SEM_ADDR_SWAP__SHIFT 0x0
+#define CP_WAIT_SEM_ADDR_LO__SEM_ADDR_LO_MASK 0xfffffff8
+#define CP_WAIT_SEM_ADDR_LO__SEM_ADDR_LO__SHIFT 0x3
+#define CP_WAIT_SEM_ADDR_HI__SEM_ADDR_HI_MASK 0xffff
+#define CP_WAIT_SEM_ADDR_HI__SEM_ADDR_HI__SHIFT 0x0
+#define CP_WAIT_SEM_ADDR_HI__SEM_USE_MAILBOX_MASK 0x10000
+#define CP_WAIT_SEM_ADDR_HI__SEM_USE_MAILBOX__SHIFT 0x10
+#define CP_WAIT_SEM_ADDR_HI__SEM_SIGNAL_TYPE_MASK 0x100000
+#define CP_WAIT_SEM_ADDR_HI__SEM_SIGNAL_TYPE__SHIFT 0x14
+#define CP_WAIT_SEM_ADDR_HI__SEM_CLIENT_CODE_MASK 0x3000000
+#define CP_WAIT_SEM_ADDR_HI__SEM_CLIENT_CODE__SHIFT 0x18
+#define CP_WAIT_SEM_ADDR_HI__SEM_SELECT_MASK 0xe0000000
+#define CP_WAIT_SEM_ADDR_HI__SEM_SELECT__SHIFT 0x1d
+#define CP_WAIT_REG_MEM_TIMEOUT__WAIT_REG_MEM_TIMEOUT_MASK 0xffffffff
+#define CP_WAIT_REG_MEM_TIMEOUT__WAIT_REG_MEM_TIMEOUT__SHIFT 0x0
+#define CP_COHER_START_DELAY__START_DELAY_COUNT_MASK 0x3f
+#define CP_COHER_START_DELAY__START_DELAY_COUNT__SHIFT 0x0
+#define CP_COHER_CNTL__DEST_BASE_0_ENA_MASK 0x1
+#define CP_COHER_CNTL__DEST_BASE_0_ENA__SHIFT 0x0
+#define CP_COHER_CNTL__DEST_BASE_1_ENA_MASK 0x2
+#define CP_COHER_CNTL__DEST_BASE_1_ENA__SHIFT 0x1
+#define CP_COHER_CNTL__TC_SD_ACTION_ENA_MASK 0x4
+#define CP_COHER_CNTL__TC_SD_ACTION_ENA__SHIFT 0x2
+#define CP_COHER_CNTL__TC_NC_ACTION_ENA_MASK 0x8
+#define CP_COHER_CNTL__TC_NC_ACTION_ENA__SHIFT 0x3
+#define CP_COHER_CNTL__CB0_DEST_BASE_ENA_MASK 0x40
+#define CP_COHER_CNTL__CB0_DEST_BASE_ENA__SHIFT 0x6
+#define CP_COHER_CNTL__CB1_DEST_BASE_ENA_MASK 0x80
+#define CP_COHER_CNTL__CB1_DEST_BASE_ENA__SHIFT 0x7
+#define CP_COHER_CNTL__CB2_DEST_BASE_ENA_MASK 0x100
+#define CP_COHER_CNTL__CB2_DEST_BASE_ENA__SHIFT 0x8
+#define CP_COHER_CNTL__CB3_DEST_BASE_ENA_MASK 0x200
+#define CP_COHER_CNTL__CB3_DEST_BASE_ENA__SHIFT 0x9
+#define CP_COHER_CNTL__CB4_DEST_BASE_ENA_MASK 0x400
+#define CP_COHER_CNTL__CB4_DEST_BASE_ENA__SHIFT 0xa
+#define CP_COHER_CNTL__CB5_DEST_BASE_ENA_MASK 0x800
+#define CP_COHER_CNTL__CB5_DEST_BASE_ENA__SHIFT 0xb
+#define CP_COHER_CNTL__CB6_DEST_BASE_ENA_MASK 0x1000
+#define CP_COHER_CNTL__CB6_DEST_BASE_ENA__SHIFT 0xc
+#define CP_COHER_CNTL__CB7_DEST_BASE_ENA_MASK 0x2000
+#define CP_COHER_CNTL__CB7_DEST_BASE_ENA__SHIFT 0xd
+#define CP_COHER_CNTL__DB_DEST_BASE_ENA_MASK 0x4000
+#define CP_COHER_CNTL__DB_DEST_BASE_ENA__SHIFT 0xe
+#define CP_COHER_CNTL__TCL1_VOL_ACTION_ENA_MASK 0x8000
+#define CP_COHER_CNTL__TCL1_VOL_ACTION_ENA__SHIFT 0xf
+#define CP_COHER_CNTL__TC_WB_ACTION_ENA_MASK 0x40000
+#define CP_COHER_CNTL__TC_WB_ACTION_ENA__SHIFT 0x12
+#define CP_COHER_CNTL__DEST_BASE_2_ENA_MASK 0x80000
+#define CP_COHER_CNTL__DEST_BASE_2_ENA__SHIFT 0x13
+#define CP_COHER_CNTL__DEST_BASE_3_ENA_MASK 0x200000
+#define CP_COHER_CNTL__DEST_BASE_3_ENA__SHIFT 0x15
+#define CP_COHER_CNTL__TCL1_ACTION_ENA_MASK 0x400000
+#define CP_COHER_CNTL__TCL1_ACTION_ENA__SHIFT 0x16
+#define CP_COHER_CNTL__TC_ACTION_ENA_MASK 0x800000
+#define CP_COHER_CNTL__TC_ACTION_ENA__SHIFT 0x17
+#define CP_COHER_CNTL__CB_ACTION_ENA_MASK 0x2000000
+#define CP_COHER_CNTL__CB_ACTION_ENA__SHIFT 0x19
+#define CP_COHER_CNTL__DB_ACTION_ENA_MASK 0x4000000
+#define CP_COHER_CNTL__DB_ACTION_ENA__SHIFT 0x1a
+#define CP_COHER_CNTL__SH_KCACHE_ACTION_ENA_MASK 0x8000000
+#define CP_COHER_CNTL__SH_KCACHE_ACTION_ENA__SHIFT 0x1b
+#define CP_COHER_CNTL__SH_KCACHE_VOL_ACTION_ENA_MASK 0x10000000
+#define CP_COHER_CNTL__SH_KCACHE_VOL_ACTION_ENA__SHIFT 0x1c
+#define CP_COHER_CNTL__SH_ICACHE_ACTION_ENA_MASK 0x20000000
+#define CP_COHER_CNTL__SH_ICACHE_ACTION_ENA__SHIFT 0x1d
+#define CP_COHER_CNTL__SH_KCACHE_WB_ACTION_ENA_MASK 0x40000000
+#define CP_COHER_CNTL__SH_KCACHE_WB_ACTION_ENA__SHIFT 0x1e
+#define CP_COHER_CNTL__SH_SD_ACTION_ENA_MASK 0x80000000
+#define CP_COHER_CNTL__SH_SD_ACTION_ENA__SHIFT 0x1f
+#define CP_COHER_SIZE__COHER_SIZE_256B_MASK 0xffffffff
+#define CP_COHER_SIZE__COHER_SIZE_256B__SHIFT 0x0
+#define CP_COHER_SIZE_HI__COHER_SIZE_HI_256B_MASK 0xff
+#define CP_COHER_SIZE_HI__COHER_SIZE_HI_256B__SHIFT 0x0
+#define CP_COHER_BASE__COHER_BASE_256B_MASK 0xffffffff
+#define CP_COHER_BASE__COHER_BASE_256B__SHIFT 0x0
+#define CP_COHER_BASE_HI__COHER_BASE_HI_256B_MASK 0xff
+#define CP_COHER_BASE_HI__COHER_BASE_HI_256B__SHIFT 0x0
+#define CP_COHER_STATUS__MATCHING_GFX_CNTX_MASK 0xff
+#define CP_COHER_STATUS__MATCHING_GFX_CNTX__SHIFT 0x0
+#define CP_COHER_STATUS__MEID_MASK 0x3000000
+#define CP_COHER_STATUS__MEID__SHIFT 0x18
+#define CP_COHER_STATUS__PHASE1_STATUS_MASK 0x40000000
+#define CP_COHER_STATUS__PHASE1_STATUS__SHIFT 0x1e
+#define CP_COHER_STATUS__STATUS_MASK 0x80000000
+#define CP_COHER_STATUS__STATUS__SHIFT 0x1f
+#define COHER_DEST_BASE_0__DEST_BASE_256B_MASK 0xffffffff
+#define COHER_DEST_BASE_0__DEST_BASE_256B__SHIFT 0x0
+#define COHER_DEST_BASE_1__DEST_BASE_256B_MASK 0xffffffff
+#define COHER_DEST_BASE_1__DEST_BASE_256B__SHIFT 0x0
+#define COHER_DEST_BASE_2__DEST_BASE_256B_MASK 0xffffffff
+#define COHER_DEST_BASE_2__DEST_BASE_256B__SHIFT 0x0
+#define COHER_DEST_BASE_3__DEST_BASE_256B_MASK 0xffffffff
+#define COHER_DEST_BASE_3__DEST_BASE_256B__SHIFT 0x0
+#define COHER_DEST_BASE_HI_0__DEST_BASE_HI_256B_MASK 0xffffffff
+#define COHER_DEST_BASE_HI_0__DEST_BASE_HI_256B__SHIFT 0x0
+#define COHER_DEST_BASE_HI_1__DEST_BASE_HI_256B_MASK 0xffffffff
+#define COHER_DEST_BASE_HI_1__DEST_BASE_HI_256B__SHIFT 0x0
+#define COHER_DEST_BASE_HI_2__DEST_BASE_HI_256B_MASK 0xffffffff
+#define COHER_DEST_BASE_HI_2__DEST_BASE_HI_256B__SHIFT 0x0
+#define COHER_DEST_BASE_HI_3__DEST_BASE_HI_256B_MASK 0xffffffff
+#define COHER_DEST_BASE_HI_3__DEST_BASE_HI_256B__SHIFT 0x0
+#define CP_DMA_ME_SRC_ADDR__SRC_ADDR_MASK 0xffffffff
+#define CP_DMA_ME_SRC_ADDR__SRC_ADDR__SHIFT 0x0
+#define CP_DMA_ME_SRC_ADDR_HI__SRC_ADDR_HI_MASK 0xffff
+#define CP_DMA_ME_SRC_ADDR_HI__SRC_ADDR_HI__SHIFT 0x0
+#define CP_DMA_ME_DST_ADDR__DST_ADDR_MASK 0xffffffff
+#define CP_DMA_ME_DST_ADDR__DST_ADDR__SHIFT 0x0
+#define CP_DMA_ME_DST_ADDR_HI__DST_ADDR_HI_MASK 0xffff
+#define CP_DMA_ME_DST_ADDR_HI__DST_ADDR_HI__SHIFT 0x0
+#define CP_DMA_ME_CONTROL__SRC_MTYPE_MASK 0xc00
+#define CP_DMA_ME_CONTROL__SRC_MTYPE__SHIFT 0xa
+#define CP_DMA_ME_CONTROL__SRC_ATC_MASK 0x1000
+#define CP_DMA_ME_CONTROL__SRC_ATC__SHIFT 0xc
+#define CP_DMA_ME_CONTROL__SRC_CACHE_POLICY_MASK 0x2000
+#define CP_DMA_ME_CONTROL__SRC_CACHE_POLICY__SHIFT 0xd
+#define CP_DMA_ME_CONTROL__DST_SELECT_MASK 0x300000
+#define CP_DMA_ME_CONTROL__DST_SELECT__SHIFT 0x14
+#define CP_DMA_ME_CONTROL__DST_MTYPE_MASK 0xc00000
+#define CP_DMA_ME_CONTROL__DST_MTYPE__SHIFT 0x16
+#define CP_DMA_ME_CONTROL__DST_ATC_MASK 0x1000000
+#define CP_DMA_ME_CONTROL__DST_ATC__SHIFT 0x18
+#define CP_DMA_ME_CONTROL__DST_CACHE_POLICY_MASK 0x2000000
+#define CP_DMA_ME_CONTROL__DST_CACHE_POLICY__SHIFT 0x19
+#define CP_DMA_ME_CONTROL__SRC_SELECT_MASK 0x60000000
+#define CP_DMA_ME_CONTROL__SRC_SELECT__SHIFT 0x1d
+#define CP_DMA_ME_COMMAND__BYTE_COUNT_MASK 0x1fffff
+#define CP_DMA_ME_COMMAND__BYTE_COUNT__SHIFT 0x0
+#define CP_DMA_ME_COMMAND__DIS_WC_MASK 0x200000
+#define CP_DMA_ME_COMMAND__DIS_WC__SHIFT 0x15
+#define CP_DMA_ME_COMMAND__SRC_SWAP_MASK 0xc00000
+#define CP_DMA_ME_COMMAND__SRC_SWAP__SHIFT 0x16
+#define CP_DMA_ME_COMMAND__DST_SWAP_MASK 0x3000000
+#define CP_DMA_ME_COMMAND__DST_SWAP__SHIFT 0x18
+#define CP_DMA_ME_COMMAND__SAS_MASK 0x4000000
+#define CP_DMA_ME_COMMAND__SAS__SHIFT 0x1a
+#define CP_DMA_ME_COMMAND__DAS_MASK 0x8000000
+#define CP_DMA_ME_COMMAND__DAS__SHIFT 0x1b
+#define CP_DMA_ME_COMMAND__SAIC_MASK 0x10000000
+#define CP_DMA_ME_COMMAND__SAIC__SHIFT 0x1c
+#define CP_DMA_ME_COMMAND__DAIC_MASK 0x20000000
+#define CP_DMA_ME_COMMAND__DAIC__SHIFT 0x1d
+#define CP_DMA_ME_COMMAND__RAW_WAIT_MASK 0x40000000
+#define CP_DMA_ME_COMMAND__RAW_WAIT__SHIFT 0x1e
+#define CP_DMA_PFP_SRC_ADDR__SRC_ADDR_MASK 0xffffffff
+#define CP_DMA_PFP_SRC_ADDR__SRC_ADDR__SHIFT 0x0
+#define CP_DMA_PFP_SRC_ADDR_HI__SRC_ADDR_HI_MASK 0xffff
+#define CP_DMA_PFP_SRC_ADDR_HI__SRC_ADDR_HI__SHIFT 0x0
+#define CP_DMA_PFP_DST_ADDR__DST_ADDR_MASK 0xffffffff
+#define CP_DMA_PFP_DST_ADDR__DST_ADDR__SHIFT 0x0
+#define CP_DMA_PFP_DST_ADDR_HI__DST_ADDR_HI_MASK 0xffff
+#define CP_DMA_PFP_DST_ADDR_HI__DST_ADDR_HI__SHIFT 0x0
+#define CP_DMA_PFP_CONTROL__SRC_MTYPE_MASK 0xc00
+#define CP_DMA_PFP_CONTROL__SRC_MTYPE__SHIFT 0xa
+#define CP_DMA_PFP_CONTROL__SRC_ATC_MASK 0x1000
+#define CP_DMA_PFP_CONTROL__SRC_ATC__SHIFT 0xc
+#define CP_DMA_PFP_CONTROL__SRC_CACHE_POLICY_MASK 0x2000
+#define CP_DMA_PFP_CONTROL__SRC_CACHE_POLICY__SHIFT 0xd
+#define CP_DMA_PFP_CONTROL__DST_SELECT_MASK 0x300000
+#define CP_DMA_PFP_CONTROL__DST_SELECT__SHIFT 0x14
+#define CP_DMA_PFP_CONTROL__DST_MTYPE_MASK 0xc00000
+#define CP_DMA_PFP_CONTROL__DST_MTYPE__SHIFT 0x16
+#define CP_DMA_PFP_CONTROL__DST_ATC_MASK 0x1000000
+#define CP_DMA_PFP_CONTROL__DST_ATC__SHIFT 0x18
+#define CP_DMA_PFP_CONTROL__DST_CACHE_POLICY_MASK 0x2000000
+#define CP_DMA_PFP_CONTROL__DST_CACHE_POLICY__SHIFT 0x19
+#define CP_DMA_PFP_CONTROL__SRC_SELECT_MASK 0x60000000
+#define CP_DMA_PFP_CONTROL__SRC_SELECT__SHIFT 0x1d
+#define CP_DMA_PFP_COMMAND__BYTE_COUNT_MASK 0x1fffff
+#define CP_DMA_PFP_COMMAND__BYTE_COUNT__SHIFT 0x0
+#define CP_DMA_PFP_COMMAND__DIS_WC_MASK 0x200000
+#define CP_DMA_PFP_COMMAND__DIS_WC__SHIFT 0x15
+#define CP_DMA_PFP_COMMAND__SRC_SWAP_MASK 0xc00000
+#define CP_DMA_PFP_COMMAND__SRC_SWAP__SHIFT 0x16
+#define CP_DMA_PFP_COMMAND__DST_SWAP_MASK 0x3000000
+#define CP_DMA_PFP_COMMAND__DST_SWAP__SHIFT 0x18
+#define CP_DMA_PFP_COMMAND__SAS_MASK 0x4000000
+#define CP_DMA_PFP_COMMAND__SAS__SHIFT 0x1a
+#define CP_DMA_PFP_COMMAND__DAS_MASK 0x8000000
+#define CP_DMA_PFP_COMMAND__DAS__SHIFT 0x1b
+#define CP_DMA_PFP_COMMAND__SAIC_MASK 0x10000000
+#define CP_DMA_PFP_COMMAND__SAIC__SHIFT 0x1c
+#define CP_DMA_PFP_COMMAND__DAIC_MASK 0x20000000
+#define CP_DMA_PFP_COMMAND__DAIC__SHIFT 0x1d
+#define CP_DMA_PFP_COMMAND__RAW_WAIT_MASK 0x40000000
+#define CP_DMA_PFP_COMMAND__RAW_WAIT__SHIFT 0x1e
+#define CP_DMA_CNTL__MIN_AVAILSZ_MASK 0x30
+#define CP_DMA_CNTL__MIN_AVAILSZ__SHIFT 0x4
+#define CP_DMA_CNTL__BUFFER_DEPTH_MASK 0xf0000
+#define CP_DMA_CNTL__BUFFER_DEPTH__SHIFT 0x10
+#define CP_DMA_CNTL__PIO_FIFO_EMPTY_MASK 0x10000000
+#define CP_DMA_CNTL__PIO_FIFO_EMPTY__SHIFT 0x1c
+#define CP_DMA_CNTL__PIO_FIFO_FULL_MASK 0x20000000
+#define CP_DMA_CNTL__PIO_FIFO_FULL__SHIFT 0x1d
+#define CP_DMA_CNTL__PIO_COUNT_MASK 0xc0000000
+#define CP_DMA_CNTL__PIO_COUNT__SHIFT 0x1e
+#define CP_DMA_READ_TAGS__DMA_READ_TAG_MASK 0x3ffffff
+#define CP_DMA_READ_TAGS__DMA_READ_TAG__SHIFT 0x0
+#define CP_DMA_READ_TAGS__DMA_READ_TAG_VALID_MASK 0x10000000
+#define CP_DMA_READ_TAGS__DMA_READ_TAG_VALID__SHIFT 0x1c
+#define CP_PFP_IB_CONTROL__IB_EN_MASK 0xff
+#define CP_PFP_IB_CONTROL__IB_EN__SHIFT 0x0
+#define CP_PFP_LOAD_CONTROL__CONFIG_REG_EN_MASK 0x1
+#define CP_PFP_LOAD_CONTROL__CONFIG_REG_EN__SHIFT 0x0
+#define CP_PFP_LOAD_CONTROL__CNTX_REG_EN_MASK 0x2
+#define CP_PFP_LOAD_CONTROL__CNTX_REG_EN__SHIFT 0x1
+#define CP_PFP_LOAD_CONTROL__SH_GFX_REG_EN_MASK 0x10000
+#define CP_PFP_LOAD_CONTROL__SH_GFX_REG_EN__SHIFT 0x10
+#define CP_PFP_LOAD_CONTROL__SH_CS_REG_EN_MASK 0x1000000
+#define CP_PFP_LOAD_CONTROL__SH_CS_REG_EN__SHIFT 0x18
+#define CP_SCRATCH_INDEX__SCRATCH_INDEX_MASK 0xff
+#define CP_SCRATCH_INDEX__SCRATCH_INDEX__SHIFT 0x0
+#define CP_SCRATCH_DATA__SCRATCH_DATA_MASK 0xffffffff
+#define CP_SCRATCH_DATA__SCRATCH_DATA__SHIFT 0x0
+#define CP_RB_OFFSET__RB_OFFSET_MASK 0xfffff
+#define CP_RB_OFFSET__RB_OFFSET__SHIFT 0x0
+#define CP_IB1_OFFSET__IB1_OFFSET_MASK 0xfffff
+#define CP_IB1_OFFSET__IB1_OFFSET__SHIFT 0x0
+#define CP_IB2_OFFSET__IB2_OFFSET_MASK 0xfffff
+#define CP_IB2_OFFSET__IB2_OFFSET__SHIFT 0x0
+#define CP_IB1_PREAMBLE_BEGIN__IB1_PREAMBLE_BEGIN_MASK 0xfffff
+#define CP_IB1_PREAMBLE_BEGIN__IB1_PREAMBLE_BEGIN__SHIFT 0x0
+#define CP_IB1_PREAMBLE_END__IB1_PREAMBLE_END_MASK 0xfffff
+#define CP_IB1_PREAMBLE_END__IB1_PREAMBLE_END__SHIFT 0x0
+#define CP_IB2_PREAMBLE_BEGIN__IB2_PREAMBLE_BEGIN_MASK 0xfffff
+#define CP_IB2_PREAMBLE_BEGIN__IB2_PREAMBLE_BEGIN__SHIFT 0x0
+#define CP_IB2_PREAMBLE_END__IB2_PREAMBLE_END_MASK 0xfffff
+#define CP_IB2_PREAMBLE_END__IB2_PREAMBLE_END__SHIFT 0x0
+#define CP_CE_IB1_OFFSET__IB1_OFFSET_MASK 0xfffff
+#define CP_CE_IB1_OFFSET__IB1_OFFSET__SHIFT 0x0
+#define CP_CE_IB2_OFFSET__IB2_OFFSET_MASK 0xfffff
+#define CP_CE_IB2_OFFSET__IB2_OFFSET__SHIFT 0x0
+#define CP_CE_COUNTER__CONST_ENGINE_COUNT_MASK 0xffffffff
+#define CP_CE_COUNTER__CONST_ENGINE_COUNT__SHIFT 0x0
+#define CP_CE_RB_OFFSET__RB_OFFSET_MASK 0xfffff
+#define CP_CE_RB_OFFSET__RB_OFFSET__SHIFT 0x0
+#define CP_PFP_COMPLETION_STATUS__STATUS_MASK 0x3
+#define CP_PFP_COMPLETION_STATUS__STATUS__SHIFT 0x0
+#define CP_CE_COMPLETION_STATUS__STATUS_MASK 0x3
+#define CP_CE_COMPLETION_STATUS__STATUS__SHIFT 0x0
+#define CP_PRED_NOT_VISIBLE__NOT_VISIBLE_MASK 0x1
+#define CP_PRED_NOT_VISIBLE__NOT_VISIBLE__SHIFT 0x0
+#define CP_PFP_METADATA_BASE_ADDR__ADDR_LO_MASK 0xffffffff
+#define CP_PFP_METADATA_BASE_ADDR__ADDR_LO__SHIFT 0x0
+#define CP_PFP_METADATA_BASE_ADDR_HI__ADDR_HI_MASK 0xffff
+#define CP_PFP_METADATA_BASE_ADDR_HI__ADDR_HI__SHIFT 0x0
+#define CP_CE_METADATA_BASE_ADDR__ADDR_LO_MASK 0xffffffff
+#define CP_CE_METADATA_BASE_ADDR__ADDR_LO__SHIFT 0x0
+#define CP_CE_METADATA_BASE_ADDR_HI__ADDR_HI_MASK 0xffff
+#define CP_CE_METADATA_BASE_ADDR_HI__ADDR_HI__SHIFT 0x0
+#define CP_DRAW_INDX_INDR_ADDR__ADDR_LO_MASK 0xffffffff
+#define CP_DRAW_INDX_INDR_ADDR__ADDR_LO__SHIFT 0x0
+#define CP_DRAW_INDX_INDR_ADDR_HI__ADDR_HI_MASK 0xffff
+#define CP_DRAW_INDX_INDR_ADDR_HI__ADDR_HI__SHIFT 0x0
+#define CP_DISPATCH_INDR_ADDR__ADDR_LO_MASK 0xffffffff
+#define CP_DISPATCH_INDR_ADDR__ADDR_LO__SHIFT 0x0
+#define CP_DISPATCH_INDR_ADDR_HI__ADDR_HI_MASK 0xffff
+#define CP_DISPATCH_INDR_ADDR_HI__ADDR_HI__SHIFT 0x0
+#define CP_INDEX_BASE_ADDR__ADDR_LO_MASK 0xffffffff
+#define CP_INDEX_BASE_ADDR__ADDR_LO__SHIFT 0x0
+#define CP_INDEX_BASE_ADDR_HI__ADDR_HI_MASK 0xffff
+#define CP_INDEX_BASE_ADDR_HI__ADDR_HI__SHIFT 0x0
+#define CP_INDEX_TYPE__INDEX_TYPE_MASK 0x3
+#define CP_INDEX_TYPE__INDEX_TYPE__SHIFT 0x0
+#define CP_GDS_BKUP_ADDR__ADDR_LO_MASK 0xffffffff
+#define CP_GDS_BKUP_ADDR__ADDR_LO__SHIFT 0x0
+#define CP_GDS_BKUP_ADDR_HI__ADDR_HI_MASK 0xffff
+#define CP_GDS_BKUP_ADDR_HI__ADDR_HI__SHIFT 0x0
+#define CP_SAMPLE_STATUS__Z_PASS_ACITVE_MASK 0x1
+#define CP_SAMPLE_STATUS__Z_PASS_ACITVE__SHIFT 0x0
+#define CP_SAMPLE_STATUS__STREAMOUT_ACTIVE_MASK 0x2
+#define CP_SAMPLE_STATUS__STREAMOUT_ACTIVE__SHIFT 0x1
+#define CP_SAMPLE_STATUS__PIPELINE_ACTIVE_MASK 0x4
+#define CP_SAMPLE_STATUS__PIPELINE_ACTIVE__SHIFT 0x2
+#define CP_SAMPLE_STATUS__STIPPLE_ACTIVE_MASK 0x8
+#define CP_SAMPLE_STATUS__STIPPLE_ACTIVE__SHIFT 0x3
+#define CP_SAMPLE_STATUS__VGT_BUFFERS_ACTIVE_MASK 0x10
+#define CP_SAMPLE_STATUS__VGT_BUFFERS_ACTIVE__SHIFT 0x4
+#define CP_SAMPLE_STATUS__SCREEN_EXT_ACTIVE_MASK 0x20
+#define CP_SAMPLE_STATUS__SCREEN_EXT_ACTIVE__SHIFT 0x5
+#define CP_SAMPLE_STATUS__DRAW_INDIRECT_ACTIVE_MASK 0x40
+#define CP_SAMPLE_STATUS__DRAW_INDIRECT_ACTIVE__SHIFT 0x6
+#define CP_SAMPLE_STATUS__DISP_INDIRECT_ACTIVE_MASK 0x80
+#define CP_SAMPLE_STATUS__DISP_INDIRECT_ACTIVE__SHIFT 0x7
+#define CP_STALLED_STAT1__RBIU_TO_DMA_NOT_RDY_TO_RCV_MASK 0x1
+#define CP_STALLED_STAT1__RBIU_TO_DMA_NOT_RDY_TO_RCV__SHIFT 0x0
+#define CP_STALLED_STAT1__RBIU_TO_SEM_NOT_RDY_TO_RCV_MASK 0x4
+#define CP_STALLED_STAT1__RBIU_TO_SEM_NOT_RDY_TO_RCV__SHIFT 0x2
+#define CP_STALLED_STAT1__RBIU_TO_MEMWR_NOT_RDY_TO_RCV_MASK 0x10
+#define CP_STALLED_STAT1__RBIU_TO_MEMWR_NOT_RDY_TO_RCV__SHIFT 0x4
+#define CP_STALLED_STAT1__ME_HAS_ACTIVE_CE_BUFFER_FLAG_MASK 0x400
+#define CP_STALLED_STAT1__ME_HAS_ACTIVE_CE_BUFFER_FLAG__SHIFT 0xa
+#define CP_STALLED_STAT1__ME_HAS_ACTIVE_DE_BUFFER_FLAG_MASK 0x800
+#define CP_STALLED_STAT1__ME_HAS_ACTIVE_DE_BUFFER_FLAG__SHIFT 0xb
+#define CP_STALLED_STAT1__ME_STALLED_ON_TC_WR_CONFIRM_MASK 0x1000
+#define CP_STALLED_STAT1__ME_STALLED_ON_TC_WR_CONFIRM__SHIFT 0xc
+#define CP_STALLED_STAT1__ME_STALLED_ON_ATOMIC_RTN_DATA_MASK 0x2000
+#define CP_STALLED_STAT1__ME_STALLED_ON_ATOMIC_RTN_DATA__SHIFT 0xd
+#define CP_STALLED_STAT1__ME_WAITING_ON_TC_READ_DATA_MASK 0x4000
+#define CP_STALLED_STAT1__ME_WAITING_ON_TC_READ_DATA__SHIFT 0xe
+#define CP_STALLED_STAT1__ME_WAITING_ON_REG_READ_DATA_MASK 0x8000
+#define CP_STALLED_STAT1__ME_WAITING_ON_REG_READ_DATA__SHIFT 0xf
+#define CP_STALLED_STAT1__RCIU_WAITING_ON_GDS_FREE_MASK 0x800000
+#define CP_STALLED_STAT1__RCIU_WAITING_ON_GDS_FREE__SHIFT 0x17
+#define CP_STALLED_STAT1__RCIU_WAITING_ON_GRBM_FREE_MASK 0x1000000
+#define CP_STALLED_STAT1__RCIU_WAITING_ON_GRBM_FREE__SHIFT 0x18
+#define CP_STALLED_STAT1__RCIU_WAITING_ON_VGT_FREE_MASK 0x2000000
+#define CP_STALLED_STAT1__RCIU_WAITING_ON_VGT_FREE__SHIFT 0x19
+#define CP_STALLED_STAT1__RCIU_STALLED_ON_ME_READ_MASK 0x4000000
+#define CP_STALLED_STAT1__RCIU_STALLED_ON_ME_READ__SHIFT 0x1a
+#define CP_STALLED_STAT1__RCIU_STALLED_ON_DMA_READ_MASK 0x8000000
+#define CP_STALLED_STAT1__RCIU_STALLED_ON_DMA_READ__SHIFT 0x1b
+#define CP_STALLED_STAT1__RCIU_STALLED_ON_APPEND_READ_MASK 0x10000000
+#define CP_STALLED_STAT1__RCIU_STALLED_ON_APPEND_READ__SHIFT 0x1c
+#define CP_STALLED_STAT1__RCIU_HALTED_BY_REG_VIOLATION_MASK 0x20000000
+#define CP_STALLED_STAT1__RCIU_HALTED_BY_REG_VIOLATION__SHIFT 0x1d
+#define CP_STALLED_STAT2__PFP_TO_CSF_NOT_RDY_TO_RCV_MASK 0x1
+#define CP_STALLED_STAT2__PFP_TO_CSF_NOT_RDY_TO_RCV__SHIFT 0x0
+#define CP_STALLED_STAT2__PFP_TO_MEQ_NOT_RDY_TO_RCV_MASK 0x2
+#define CP_STALLED_STAT2__PFP_TO_MEQ_NOT_RDY_TO_RCV__SHIFT 0x1
+#define CP_STALLED_STAT2__PFP_TO_RCIU_NOT_RDY_TO_RCV_MASK 0x4
+#define CP_STALLED_STAT2__PFP_TO_RCIU_NOT_RDY_TO_RCV__SHIFT 0x2
+#define CP_STALLED_STAT2__PFP_TO_VGT_WRITES_PENDING_MASK 0x10
+#define CP_STALLED_STAT2__PFP_TO_VGT_WRITES_PENDING__SHIFT 0x4
+#define CP_STALLED_STAT2__PFP_RCIU_READ_PENDING_MASK 0x20
+#define CP_STALLED_STAT2__PFP_RCIU_READ_PENDING__SHIFT 0x5
+#define CP_STALLED_STAT2__PFP_WAITING_ON_BUFFER_DATA_MASK 0x100
+#define CP_STALLED_STAT2__PFP_WAITING_ON_BUFFER_DATA__SHIFT 0x8
+#define CP_STALLED_STAT2__ME_WAIT_ON_CE_COUNTER_MASK 0x200
+#define CP_STALLED_STAT2__ME_WAIT_ON_CE_COUNTER__SHIFT 0x9
+#define CP_STALLED_STAT2__ME_WAIT_ON_AVAIL_BUFFER_MASK 0x400
+#define CP_STALLED_STAT2__ME_WAIT_ON_AVAIL_BUFFER__SHIFT 0xa
+#define CP_STALLED_STAT2__GFX_CNTX_NOT_AVAIL_TO_ME_MASK 0x800
+#define CP_STALLED_STAT2__GFX_CNTX_NOT_AVAIL_TO_ME__SHIFT 0xb
+#define CP_STALLED_STAT2__ME_RCIU_NOT_RDY_TO_RCV_MASK 0x1000
+#define CP_STALLED_STAT2__ME_RCIU_NOT_RDY_TO_RCV__SHIFT 0xc
+#define CP_STALLED_STAT2__ME_TO_CONST_NOT_RDY_TO_RCV_MASK 0x2000
+#define CP_STALLED_STAT2__ME_TO_CONST_NOT_RDY_TO_RCV__SHIFT 0xd
+#define CP_STALLED_STAT2__ME_WAITING_DATA_FROM_PFP_MASK 0x4000
+#define CP_STALLED_STAT2__ME_WAITING_DATA_FROM_PFP__SHIFT 0xe
+#define CP_STALLED_STAT2__ME_WAITING_ON_PARTIAL_FLUSH_MASK 0x8000
+#define CP_STALLED_STAT2__ME_WAITING_ON_PARTIAL_FLUSH__SHIFT 0xf
+#define CP_STALLED_STAT2__MEQ_TO_ME_NOT_RDY_TO_RCV_MASK 0x10000
+#define CP_STALLED_STAT2__MEQ_TO_ME_NOT_RDY_TO_RCV__SHIFT 0x10
+#define CP_STALLED_STAT2__STQ_TO_ME_NOT_RDY_TO_RCV_MASK 0x20000
+#define CP_STALLED_STAT2__STQ_TO_ME_NOT_RDY_TO_RCV__SHIFT 0x11
+#define CP_STALLED_STAT2__ME_WAITING_DATA_FROM_STQ_MASK 0x40000
+#define CP_STALLED_STAT2__ME_WAITING_DATA_FROM_STQ__SHIFT 0x12
+#define CP_STALLED_STAT2__PFP_STALLED_ON_TC_WR_CONFIRM_MASK 0x80000
+#define CP_STALLED_STAT2__PFP_STALLED_ON_TC_WR_CONFIRM__SHIFT 0x13
+#define CP_STALLED_STAT2__PFP_STALLED_ON_ATOMIC_RTN_DATA_MASK 0x100000
+#define CP_STALLED_STAT2__PFP_STALLED_ON_ATOMIC_RTN_DATA__SHIFT 0x14
+#define CP_STALLED_STAT2__EOPD_FIFO_NEEDS_SC_EOP_DONE_MASK 0x200000
+#define CP_STALLED_STAT2__EOPD_FIFO_NEEDS_SC_EOP_DONE__SHIFT 0x15
+#define CP_STALLED_STAT2__EOPD_FIFO_NEEDS_WR_CONFIRM_MASK 0x400000
+#define CP_STALLED_STAT2__EOPD_FIFO_NEEDS_WR_CONFIRM__SHIFT 0x16
+#define CP_STALLED_STAT2__STRMO_WR_OF_PRIM_DATA_PENDING_MASK 0x800000
+#define CP_STALLED_STAT2__STRMO_WR_OF_PRIM_DATA_PENDING__SHIFT 0x17
+#define CP_STALLED_STAT2__PIPE_STATS_WR_DATA_PENDING_MASK 0x1000000
+#define CP_STALLED_STAT2__PIPE_STATS_WR_DATA_PENDING__SHIFT 0x18
+#define CP_STALLED_STAT2__APPEND_RDY_WAIT_ON_CS_DONE_MASK 0x2000000
+#define CP_STALLED_STAT2__APPEND_RDY_WAIT_ON_CS_DONE__SHIFT 0x19
+#define CP_STALLED_STAT2__APPEND_RDY_WAIT_ON_PS_DONE_MASK 0x4000000
+#define CP_STALLED_STAT2__APPEND_RDY_WAIT_ON_PS_DONE__SHIFT 0x1a
+#define CP_STALLED_STAT2__APPEND_WAIT_ON_WR_CONFIRM_MASK 0x8000000
+#define CP_STALLED_STAT2__APPEND_WAIT_ON_WR_CONFIRM__SHIFT 0x1b
+#define CP_STALLED_STAT2__APPEND_ACTIVE_PARTITION_MASK 0x10000000
+#define CP_STALLED_STAT2__APPEND_ACTIVE_PARTITION__SHIFT 0x1c
+#define CP_STALLED_STAT2__APPEND_WAITING_TO_SEND_MEMWRITE_MASK 0x20000000
+#define CP_STALLED_STAT2__APPEND_WAITING_TO_SEND_MEMWRITE__SHIFT 0x1d
+#define CP_STALLED_STAT2__SURF_SYNC_NEEDS_IDLE_CNTXS_MASK 0x40000000
+#define CP_STALLED_STAT2__SURF_SYNC_NEEDS_IDLE_CNTXS__SHIFT 0x1e
+#define CP_STALLED_STAT2__SURF_SYNC_NEEDS_ALL_CLEAN_MASK 0x80000000
+#define CP_STALLED_STAT2__SURF_SYNC_NEEDS_ALL_CLEAN__SHIFT 0x1f
+#define CP_STALLED_STAT3__CE_TO_CSF_NOT_RDY_TO_RCV_MASK 0x1
+#define CP_STALLED_STAT3__CE_TO_CSF_NOT_RDY_TO_RCV__SHIFT 0x0
+#define CP_STALLED_STAT3__CE_TO_RAM_INIT_FETCHER_NOT_RDY_TO_RCV_MASK 0x2
+#define CP_STALLED_STAT3__CE_TO_RAM_INIT_FETCHER_NOT_RDY_TO_RCV__SHIFT 0x1
+#define CP_STALLED_STAT3__CE_WAITING_ON_DATA_FROM_RAM_INIT_FETCHER_MASK 0x4
+#define CP_STALLED_STAT3__CE_WAITING_ON_DATA_FROM_RAM_INIT_FETCHER__SHIFT 0x2
+#define CP_STALLED_STAT3__CE_TO_RAM_INIT_NOT_RDY_MASK 0x8
+#define CP_STALLED_STAT3__CE_TO_RAM_INIT_NOT_RDY__SHIFT 0x3
+#define CP_STALLED_STAT3__CE_TO_RAM_DUMP_NOT_RDY_MASK 0x10
+#define CP_STALLED_STAT3__CE_TO_RAM_DUMP_NOT_RDY__SHIFT 0x4
+#define CP_STALLED_STAT3__CE_TO_RAM_WRITE_NOT_RDY_MASK 0x20
+#define CP_STALLED_STAT3__CE_TO_RAM_WRITE_NOT_RDY__SHIFT 0x5
+#define CP_STALLED_STAT3__CE_TO_INC_FIFO_NOT_RDY_TO_RCV_MASK 0x40
+#define CP_STALLED_STAT3__CE_TO_INC_FIFO_NOT_RDY_TO_RCV__SHIFT 0x6
+#define CP_STALLED_STAT3__CE_TO_WR_FIFO_NOT_RDY_TO_RCV_MASK 0x80
+#define CP_STALLED_STAT3__CE_TO_WR_FIFO_NOT_RDY_TO_RCV__SHIFT 0x7
+#define CP_STALLED_STAT3__CE_WAITING_ON_BUFFER_DATA_MASK 0x400
+#define CP_STALLED_STAT3__CE_WAITING_ON_BUFFER_DATA__SHIFT 0xa
+#define CP_STALLED_STAT3__CE_WAITING_ON_CE_BUFFER_FLAG_MASK 0x800
+#define CP_STALLED_STAT3__CE_WAITING_ON_CE_BUFFER_FLAG__SHIFT 0xb
+#define CP_STALLED_STAT3__CE_WAITING_ON_DE_COUNTER_MASK 0x1000
+#define CP_STALLED_STAT3__CE_WAITING_ON_DE_COUNTER__SHIFT 0xc
+#define CP_STALLED_STAT3__CE_WAITING_ON_DE_COUNTER_UNDERFLOW_MASK 0x2000
+#define CP_STALLED_STAT3__CE_WAITING_ON_DE_COUNTER_UNDERFLOW__SHIFT 0xd
+#define CP_STALLED_STAT3__TCIU_WAITING_ON_FREE_MASK 0x4000
+#define CP_STALLED_STAT3__TCIU_WAITING_ON_FREE__SHIFT 0xe
+#define CP_STALLED_STAT3__TCIU_WAITING_ON_TAGS_MASK 0x8000
+#define CP_STALLED_STAT3__TCIU_WAITING_ON_TAGS__SHIFT 0xf
+#define CP_STALLED_STAT3__CE_STALLED_ON_TC_WR_CONFIRM_MASK 0x10000
+#define CP_STALLED_STAT3__CE_STALLED_ON_TC_WR_CONFIRM__SHIFT 0x10
+#define CP_STALLED_STAT3__CE_STALLED_ON_ATOMIC_RTN_DATA_MASK 0x20000
+#define CP_STALLED_STAT3__CE_STALLED_ON_ATOMIC_RTN_DATA__SHIFT 0x11
+#define CP_STALLED_STAT3__ATCL2IU_WAITING_ON_FREE_MASK 0x40000
+#define CP_STALLED_STAT3__ATCL2IU_WAITING_ON_FREE__SHIFT 0x12
+#define CP_STALLED_STAT3__ATCL2IU_WAITING_ON_TAGS_MASK 0x80000
+#define CP_STALLED_STAT3__ATCL2IU_WAITING_ON_TAGS__SHIFT 0x13
+#define CP_STALLED_STAT3__ATCL1_WAITING_ON_TRANS_MASK 0x100000
+#define CP_STALLED_STAT3__ATCL1_WAITING_ON_TRANS__SHIFT 0x14
+#define CP_BUSY_STAT__REG_BUS_FIFO_BUSY_MASK 0x1
+#define CP_BUSY_STAT__REG_BUS_FIFO_BUSY__SHIFT 0x0
+#define CP_BUSY_STAT__COHER_CNT_NEQ_ZERO_MASK 0x40
+#define CP_BUSY_STAT__COHER_CNT_NEQ_ZERO__SHIFT 0x6
+#define CP_BUSY_STAT__PFP_PARSING_PACKETS_MASK 0x80
+#define CP_BUSY_STAT__PFP_PARSING_PACKETS__SHIFT 0x7
+#define CP_BUSY_STAT__ME_PARSING_PACKETS_MASK 0x100
+#define CP_BUSY_STAT__ME_PARSING_PACKETS__SHIFT 0x8
+#define CP_BUSY_STAT__RCIU_PFP_BUSY_MASK 0x200
+#define CP_BUSY_STAT__RCIU_PFP_BUSY__SHIFT 0x9
+#define CP_BUSY_STAT__RCIU_ME_BUSY_MASK 0x400
+#define CP_BUSY_STAT__RCIU_ME_BUSY__SHIFT 0xa
+#define CP_BUSY_STAT__SEM_CMDFIFO_NOT_EMPTY_MASK 0x1000
+#define CP_BUSY_STAT__SEM_CMDFIFO_NOT_EMPTY__SHIFT 0xc
+#define CP_BUSY_STAT__SEM_FAILED_AND_HOLDING_MASK 0x2000
+#define CP_BUSY_STAT__SEM_FAILED_AND_HOLDING__SHIFT 0xd
+#define CP_BUSY_STAT__SEM_POLLING_FOR_PASS_MASK 0x4000
+#define CP_BUSY_STAT__SEM_POLLING_FOR_PASS__SHIFT 0xe
+#define CP_BUSY_STAT__GFX_CONTEXT_BUSY_MASK 0x8000
+#define CP_BUSY_STAT__GFX_CONTEXT_BUSY__SHIFT 0xf
+#define CP_BUSY_STAT__ME_PARSER_BUSY_MASK 0x20000
+#define CP_BUSY_STAT__ME_PARSER_BUSY__SHIFT 0x11
+#define CP_BUSY_STAT__EOP_DONE_BUSY_MASK 0x40000
+#define CP_BUSY_STAT__EOP_DONE_BUSY__SHIFT 0x12
+#define CP_BUSY_STAT__STRM_OUT_BUSY_MASK 0x80000
+#define CP_BUSY_STAT__STRM_OUT_BUSY__SHIFT 0x13
+#define CP_BUSY_STAT__PIPE_STATS_BUSY_MASK 0x100000
+#define CP_BUSY_STAT__PIPE_STATS_BUSY__SHIFT 0x14
+#define CP_BUSY_STAT__RCIU_CE_BUSY_MASK 0x200000
+#define CP_BUSY_STAT__RCIU_CE_BUSY__SHIFT 0x15
+#define CP_BUSY_STAT__CE_PARSING_PACKETS_MASK 0x400000
+#define CP_BUSY_STAT__CE_PARSING_PACKETS__SHIFT 0x16
+#define CP_STAT__ROQ_RING_BUSY_MASK 0x200
+#define CP_STAT__ROQ_RING_BUSY__SHIFT 0x9
+#define CP_STAT__ROQ_INDIRECT1_BUSY_MASK 0x400
+#define CP_STAT__ROQ_INDIRECT1_BUSY__SHIFT 0xa
+#define CP_STAT__ROQ_INDIRECT2_BUSY_MASK 0x800
+#define CP_STAT__ROQ_INDIRECT2_BUSY__SHIFT 0xb
+#define CP_STAT__ROQ_STATE_BUSY_MASK 0x1000
+#define CP_STAT__ROQ_STATE_BUSY__SHIFT 0xc
+#define CP_STAT__DC_BUSY_MASK 0x2000
+#define CP_STAT__DC_BUSY__SHIFT 0xd
+#define CP_STAT__ATCL2IU_BUSY_MASK 0x4000
+#define CP_STAT__ATCL2IU_BUSY__SHIFT 0xe
+#define CP_STAT__PFP_BUSY_MASK 0x8000
+#define CP_STAT__PFP_BUSY__SHIFT 0xf
+#define CP_STAT__MEQ_BUSY_MASK 0x10000
+#define CP_STAT__MEQ_BUSY__SHIFT 0x10
+#define CP_STAT__ME_BUSY_MASK 0x20000
+#define CP_STAT__ME_BUSY__SHIFT 0x11
+#define CP_STAT__QUERY_BUSY_MASK 0x40000
+#define CP_STAT__QUERY_BUSY__SHIFT 0x12
+#define CP_STAT__SEMAPHORE_BUSY_MASK 0x80000
+#define CP_STAT__SEMAPHORE_BUSY__SHIFT 0x13
+#define CP_STAT__INTERRUPT_BUSY_MASK 0x100000
+#define CP_STAT__INTERRUPT_BUSY__SHIFT 0x14
+#define CP_STAT__SURFACE_SYNC_BUSY_MASK 0x200000
+#define CP_STAT__SURFACE_SYNC_BUSY__SHIFT 0x15
+#define CP_STAT__DMA_BUSY_MASK 0x400000
+#define CP_STAT__DMA_BUSY__SHIFT 0x16
+#define CP_STAT__RCIU_BUSY_MASK 0x800000
+#define CP_STAT__RCIU_BUSY__SHIFT 0x17
+#define CP_STAT__SCRATCH_RAM_BUSY_MASK 0x1000000
+#define CP_STAT__SCRATCH_RAM_BUSY__SHIFT 0x18
+#define CP_STAT__CPC_CPG_BUSY_MASK 0x2000000
+#define CP_STAT__CPC_CPG_BUSY__SHIFT 0x19
+#define CP_STAT__CE_BUSY_MASK 0x4000000
+#define CP_STAT__CE_BUSY__SHIFT 0x1a
+#define CP_STAT__TCIU_BUSY_MASK 0x8000000
+#define CP_STAT__TCIU_BUSY__SHIFT 0x1b
+#define CP_STAT__ROQ_CE_RING_BUSY_MASK 0x10000000
+#define CP_STAT__ROQ_CE_RING_BUSY__SHIFT 0x1c
+#define CP_STAT__ROQ_CE_INDIRECT1_BUSY_MASK 0x20000000
+#define CP_STAT__ROQ_CE_INDIRECT1_BUSY__SHIFT 0x1d
+#define CP_STAT__ROQ_CE_INDIRECT2_BUSY_MASK 0x40000000
+#define CP_STAT__ROQ_CE_INDIRECT2_BUSY__SHIFT 0x1e
+#define CP_STAT__CP_BUSY_MASK 0x80000000
+#define CP_STAT__CP_BUSY__SHIFT 0x1f
+#define CP_ME_HEADER_DUMP__ME_HEADER_DUMP_MASK 0xffffffff
+#define CP_ME_HEADER_DUMP__ME_HEADER_DUMP__SHIFT 0x0
+#define CP_PFP_HEADER_DUMP__PFP_HEADER_DUMP_MASK 0xffffffff
+#define CP_PFP_HEADER_DUMP__PFP_HEADER_DUMP__SHIFT 0x0
+#define CP_GRBM_FREE_COUNT__FREE_COUNT_MASK 0x3f
+#define CP_GRBM_FREE_COUNT__FREE_COUNT__SHIFT 0x0
+#define CP_GRBM_FREE_COUNT__FREE_COUNT_GDS_MASK 0x3f00
+#define CP_GRBM_FREE_COUNT__FREE_COUNT_GDS__SHIFT 0x8
+#define CP_GRBM_FREE_COUNT__FREE_COUNT_PFP_MASK 0x3f0000
+#define CP_GRBM_FREE_COUNT__FREE_COUNT_PFP__SHIFT 0x10
+#define CP_CE_HEADER_DUMP__CE_HEADER_DUMP_MASK 0xffffffff
+#define CP_CE_HEADER_DUMP__CE_HEADER_DUMP__SHIFT 0x0
+#define CP_CSF_STAT__BUFFER_SLOTS_ALLOCATED_MASK 0xf
+#define CP_CSF_STAT__BUFFER_SLOTS_ALLOCATED__SHIFT 0x0
+#define CP_CSF_STAT__BUFFER_REQUEST_COUNT_MASK 0x1ff00
+#define CP_CSF_STAT__BUFFER_REQUEST_COUNT__SHIFT 0x8
+#define CP_CSF_CNTL__FETCH_BUFFER_DEPTH_MASK 0xf
+#define CP_CSF_CNTL__FETCH_BUFFER_DEPTH__SHIFT 0x0
+#define CP_ME_CNTL__CE_INVALIDATE_ICACHE_MASK 0x10
+#define CP_ME_CNTL__CE_INVALIDATE_ICACHE__SHIFT 0x4
+#define CP_ME_CNTL__PFP_INVALIDATE_ICACHE_MASK 0x40
+#define CP_ME_CNTL__PFP_INVALIDATE_ICACHE__SHIFT 0x6
+#define CP_ME_CNTL__ME_INVALIDATE_ICACHE_MASK 0x100
+#define CP_ME_CNTL__ME_INVALIDATE_ICACHE__SHIFT 0x8
+#define CP_ME_CNTL__CE_PIPE0_RESET_MASK 0x10000
+#define CP_ME_CNTL__CE_PIPE0_RESET__SHIFT 0x10
+#define CP_ME_CNTL__PFP_PIPE0_RESET_MASK 0x40000
+#define CP_ME_CNTL__PFP_PIPE0_RESET__SHIFT 0x12
+#define CP_ME_CNTL__ME_PIPE0_RESET_MASK 0x100000
+#define CP_ME_CNTL__ME_PIPE0_RESET__SHIFT 0x14
+#define CP_ME_CNTL__CE_HALT_MASK 0x1000000
+#define CP_ME_CNTL__CE_HALT__SHIFT 0x18
+#define CP_ME_CNTL__CE_STEP_MASK 0x2000000
+#define CP_ME_CNTL__CE_STEP__SHIFT 0x19
+#define CP_ME_CNTL__PFP_HALT_MASK 0x4000000
+#define CP_ME_CNTL__PFP_HALT__SHIFT 0x1a
+#define CP_ME_CNTL__PFP_STEP_MASK 0x8000000
+#define CP_ME_CNTL__PFP_STEP__SHIFT 0x1b
+#define CP_ME_CNTL__ME_HALT_MASK 0x10000000
+#define CP_ME_CNTL__ME_HALT__SHIFT 0x1c
+#define CP_ME_CNTL__ME_STEP_MASK 0x20000000
+#define CP_ME_CNTL__ME_STEP__SHIFT 0x1d
+#define CP_CNTX_STAT__ACTIVE_HP3D_CONTEXTS_MASK 0xff
+#define CP_CNTX_STAT__ACTIVE_HP3D_CONTEXTS__SHIFT 0x0
+#define CP_CNTX_STAT__CURRENT_HP3D_CONTEXT_MASK 0x700
+#define CP_CNTX_STAT__CURRENT_HP3D_CONTEXT__SHIFT 0x8
+#define CP_CNTX_STAT__ACTIVE_GFX_CONTEXTS_MASK 0xff00000
+#define CP_CNTX_STAT__ACTIVE_GFX_CONTEXTS__SHIFT 0x14
+#define CP_CNTX_STAT__CURRENT_GFX_CONTEXT_MASK 0x70000000
+#define CP_CNTX_STAT__CURRENT_GFX_CONTEXT__SHIFT 0x1c
+#define CP_ME_PREEMPTION__OBSOLETE_MASK 0x1
+#define CP_ME_PREEMPTION__OBSOLETE__SHIFT 0x0
+#define CP_RB0_RPTR__RB_RPTR_MASK 0xfffff
+#define CP_RB0_RPTR__RB_RPTR__SHIFT 0x0
+#define CP_RB_RPTR__RB_RPTR_MASK 0xfffff
+#define CP_RB_RPTR__RB_RPTR__SHIFT 0x0
+#define CP_RB1_RPTR__RB_RPTR_MASK 0xfffff
+#define CP_RB1_RPTR__RB_RPTR__SHIFT 0x0
+#define CP_RB2_RPTR__RB_RPTR_MASK 0xfffff
+#define CP_RB2_RPTR__RB_RPTR__SHIFT 0x0
+#define CP_RB_WPTR_DELAY__PRE_WRITE_TIMER_MASK 0xfffffff
+#define CP_RB_WPTR_DELAY__PRE_WRITE_TIMER__SHIFT 0x0
+#define CP_RB_WPTR_DELAY__PRE_WRITE_LIMIT_MASK 0xf0000000
+#define CP_RB_WPTR_DELAY__PRE_WRITE_LIMIT__SHIFT 0x1c
+#define CP_RB_WPTR_POLL_CNTL__POLL_FREQUENCY_MASK 0xffff
+#define CP_RB_WPTR_POLL_CNTL__POLL_FREQUENCY__SHIFT 0x0
+#define CP_RB_WPTR_POLL_CNTL__IDLE_POLL_COUNT_MASK 0xffff0000
+#define CP_RB_WPTR_POLL_CNTL__IDLE_POLL_COUNT__SHIFT 0x10
+#define CP_CE_INIT_BASE_LO__INIT_BASE_LO_MASK 0xffffffe0
+#define CP_CE_INIT_BASE_LO__INIT_BASE_LO__SHIFT 0x5
+#define CP_CE_INIT_BASE_HI__INIT_BASE_HI_MASK 0xffff
+#define CP_CE_INIT_BASE_HI__INIT_BASE_HI__SHIFT 0x0
+#define CP_CE_INIT_BUFSZ__INIT_BUFSZ_MASK 0xfff
+#define CP_CE_INIT_BUFSZ__INIT_BUFSZ__SHIFT 0x0
+#define CP_CE_IB1_BASE_LO__IB1_BASE_LO_MASK 0xfffffffc
+#define CP_CE_IB1_BASE_LO__IB1_BASE_LO__SHIFT 0x2
+#define CP_CE_IB1_BASE_HI__IB1_BASE_HI_MASK 0xffff
+#define CP_CE_IB1_BASE_HI__IB1_BASE_HI__SHIFT 0x0
+#define CP_CE_IB1_BUFSZ__IB1_BUFSZ_MASK 0xfffff
+#define CP_CE_IB1_BUFSZ__IB1_BUFSZ__SHIFT 0x0
+#define CP_CE_IB2_BASE_LO__IB2_BASE_LO_MASK 0xfffffffc
+#define CP_CE_IB2_BASE_LO__IB2_BASE_LO__SHIFT 0x2
+#define CP_CE_IB2_BASE_HI__IB2_BASE_HI_MASK 0xffff
+#define CP_CE_IB2_BASE_HI__IB2_BASE_HI__SHIFT 0x0
+#define CP_CE_IB2_BUFSZ__IB2_BUFSZ_MASK 0xfffff
+#define CP_CE_IB2_BUFSZ__IB2_BUFSZ__SHIFT 0x0
+#define CP_IB1_BASE_LO__IB1_BASE_LO_MASK 0xfffffffc
+#define CP_IB1_BASE_LO__IB1_BASE_LO__SHIFT 0x2
+#define CP_IB1_BASE_HI__IB1_BASE_HI_MASK 0xffff
+#define CP_IB1_BASE_HI__IB1_BASE_HI__SHIFT 0x0
+#define CP_IB1_BUFSZ__IB1_BUFSZ_MASK 0xfffff
+#define CP_IB1_BUFSZ__IB1_BUFSZ__SHIFT 0x0
+#define CP_IB2_BASE_LO__IB2_BASE_LO_MASK 0xfffffffc
+#define CP_IB2_BASE_LO__IB2_BASE_LO__SHIFT 0x2
+#define CP_IB2_BASE_HI__IB2_BASE_HI_MASK 0xffff
+#define CP_IB2_BASE_HI__IB2_BASE_HI__SHIFT 0x0
+#define CP_IB2_BUFSZ__IB2_BUFSZ_MASK 0xfffff
+#define CP_IB2_BUFSZ__IB2_BUFSZ__SHIFT 0x0
+#define CP_ST_BASE_LO__ST_BASE_LO_MASK 0xfffffffc
+#define CP_ST_BASE_LO__ST_BASE_LO__SHIFT 0x2
+#define CP_ST_BASE_HI__ST_BASE_HI_MASK 0xffff
+#define CP_ST_BASE_HI__ST_BASE_HI__SHIFT 0x0
+#define CP_ST_BUFSZ__ST_BUFSZ_MASK 0xfffff
+#define CP_ST_BUFSZ__ST_BUFSZ__SHIFT 0x0
+#define CP_ROQ_THRESHOLDS__IB1_START_MASK 0xff
+#define CP_ROQ_THRESHOLDS__IB1_START__SHIFT 0x0
+#define CP_ROQ_THRESHOLDS__IB2_START_MASK 0xff00
+#define CP_ROQ_THRESHOLDS__IB2_START__SHIFT 0x8
+#define CP_MEQ_STQ_THRESHOLD__STQ_START_MASK 0xff
+#define CP_MEQ_STQ_THRESHOLD__STQ_START__SHIFT 0x0
+#define CP_ROQ1_THRESHOLDS__RB1_START_MASK 0xff
+#define CP_ROQ1_THRESHOLDS__RB1_START__SHIFT 0x0
+#define CP_ROQ1_THRESHOLDS__RB2_START_MASK 0xff00
+#define CP_ROQ1_THRESHOLDS__RB2_START__SHIFT 0x8
+#define CP_ROQ1_THRESHOLDS__R0_IB1_START_MASK 0xff0000
+#define CP_ROQ1_THRESHOLDS__R0_IB1_START__SHIFT 0x10
+#define CP_ROQ1_THRESHOLDS__R1_IB1_START_MASK 0xff000000
+#define CP_ROQ1_THRESHOLDS__R1_IB1_START__SHIFT 0x18
+#define CP_ROQ2_THRESHOLDS__R2_IB1_START_MASK 0xff
+#define CP_ROQ2_THRESHOLDS__R2_IB1_START__SHIFT 0x0
+#define CP_ROQ2_THRESHOLDS__R0_IB2_START_MASK 0xff00
+#define CP_ROQ2_THRESHOLDS__R0_IB2_START__SHIFT 0x8
+#define CP_ROQ2_THRESHOLDS__R1_IB2_START_MASK 0xff0000
+#define CP_ROQ2_THRESHOLDS__R1_IB2_START__SHIFT 0x10
+#define CP_ROQ2_THRESHOLDS__R2_IB2_START_MASK 0xff000000
+#define CP_ROQ2_THRESHOLDS__R2_IB2_START__SHIFT 0x18
+#define CP_STQ_THRESHOLDS__STQ0_START_MASK 0xff
+#define CP_STQ_THRESHOLDS__STQ0_START__SHIFT 0x0
+#define CP_STQ_THRESHOLDS__STQ1_START_MASK 0xff00
+#define CP_STQ_THRESHOLDS__STQ1_START__SHIFT 0x8
+#define CP_STQ_THRESHOLDS__STQ2_START_MASK 0xff0000
+#define CP_STQ_THRESHOLDS__STQ2_START__SHIFT 0x10
+#define CP_QUEUE_THRESHOLDS__ROQ_IB1_START_MASK 0x3f
+#define CP_QUEUE_THRESHOLDS__ROQ_IB1_START__SHIFT 0x0
+#define CP_QUEUE_THRESHOLDS__ROQ_IB2_START_MASK 0x3f00
+#define CP_QUEUE_THRESHOLDS__ROQ_IB2_START__SHIFT 0x8
+#define CP_MEQ_THRESHOLDS__MEQ1_START_MASK 0xff
+#define CP_MEQ_THRESHOLDS__MEQ1_START__SHIFT 0x0
+#define CP_MEQ_THRESHOLDS__MEQ2_START_MASK 0xff00
+#define CP_MEQ_THRESHOLDS__MEQ2_START__SHIFT 0x8
+#define CP_ROQ_AVAIL__ROQ_CNT_RING_MASK 0x7ff
+#define CP_ROQ_AVAIL__ROQ_CNT_RING__SHIFT 0x0
+#define CP_ROQ_AVAIL__ROQ_CNT_IB1_MASK 0x7ff0000
+#define CP_ROQ_AVAIL__ROQ_CNT_IB1__SHIFT 0x10
+#define CP_STQ_AVAIL__STQ_CNT_MASK 0x1ff
+#define CP_STQ_AVAIL__STQ_CNT__SHIFT 0x0
+#define CP_ROQ2_AVAIL__ROQ_CNT_IB2_MASK 0x7ff
+#define CP_ROQ2_AVAIL__ROQ_CNT_IB2__SHIFT 0x0
+#define CP_MEQ_AVAIL__MEQ_CNT_MASK 0x3ff
+#define CP_MEQ_AVAIL__MEQ_CNT__SHIFT 0x0
+#define CP_CMD_INDEX__CMD_INDEX_MASK 0x7ff
+#define CP_CMD_INDEX__CMD_INDEX__SHIFT 0x0
+#define CP_CMD_INDEX__CMD_ME_SEL_MASK 0x3000
+#define CP_CMD_INDEX__CMD_ME_SEL__SHIFT 0xc
+#define CP_CMD_INDEX__CMD_QUEUE_SEL_MASK 0x70000
+#define CP_CMD_INDEX__CMD_QUEUE_SEL__SHIFT 0x10
+#define CP_CMD_DATA__CMD_DATA_MASK 0xffffffff
+#define CP_CMD_DATA__CMD_DATA__SHIFT 0x0
+#define CP_ROQ_RB_STAT__ROQ_RPTR_PRIMARY_MASK 0x3ff
+#define CP_ROQ_RB_STAT__ROQ_RPTR_PRIMARY__SHIFT 0x0
+#define CP_ROQ_RB_STAT__ROQ_WPTR_PRIMARY_MASK 0x3ff0000
+#define CP_ROQ_RB_STAT__ROQ_WPTR_PRIMARY__SHIFT 0x10
+#define CP_ROQ_IB1_STAT__ROQ_RPTR_INDIRECT1_MASK 0x3ff
+#define CP_ROQ_IB1_STAT__ROQ_RPTR_INDIRECT1__SHIFT 0x0
+#define CP_ROQ_IB1_STAT__ROQ_WPTR_INDIRECT1_MASK 0x3ff0000
+#define CP_ROQ_IB1_STAT__ROQ_WPTR_INDIRECT1__SHIFT 0x10
+#define CP_ROQ_IB2_STAT__ROQ_RPTR_INDIRECT2_MASK 0x3ff
+#define CP_ROQ_IB2_STAT__ROQ_RPTR_INDIRECT2__SHIFT 0x0
+#define CP_ROQ_IB2_STAT__ROQ_WPTR_INDIRECT2_MASK 0x3ff0000
+#define CP_ROQ_IB2_STAT__ROQ_WPTR_INDIRECT2__SHIFT 0x10
+#define CP_STQ_STAT__STQ_RPTR_MASK 0x3ff
+#define CP_STQ_STAT__STQ_RPTR__SHIFT 0x0
+#define CP_STQ_WR_STAT__STQ_WPTR_MASK 0x3ff
+#define CP_STQ_WR_STAT__STQ_WPTR__SHIFT 0x0
+#define CP_MEQ_STAT__MEQ_RPTR_MASK 0x3ff
+#define CP_MEQ_STAT__MEQ_RPTR__SHIFT 0x0
+#define CP_MEQ_STAT__MEQ_WPTR_MASK 0x3ff0000
+#define CP_MEQ_STAT__MEQ_WPTR__SHIFT 0x10
+#define CP_CEQ1_AVAIL__CEQ_CNT_RING_MASK 0x7ff
+#define CP_CEQ1_AVAIL__CEQ_CNT_RING__SHIFT 0x0
+#define CP_CEQ1_AVAIL__CEQ_CNT_IB1_MASK 0x7ff0000
+#define CP_CEQ1_AVAIL__CEQ_CNT_IB1__SHIFT 0x10
+#define CP_CEQ2_AVAIL__CEQ_CNT_IB2_MASK 0x7ff
+#define CP_CEQ2_AVAIL__CEQ_CNT_IB2__SHIFT 0x0
+#define CP_CE_ROQ_RB_STAT__CEQ_RPTR_PRIMARY_MASK 0x3ff
+#define CP_CE_ROQ_RB_STAT__CEQ_RPTR_PRIMARY__SHIFT 0x0
+#define CP_CE_ROQ_RB_STAT__CEQ_WPTR_PRIMARY_MASK 0x3ff0000
+#define CP_CE_ROQ_RB_STAT__CEQ_WPTR_PRIMARY__SHIFT 0x10
+#define CP_CE_ROQ_IB1_STAT__CEQ_RPTR_INDIRECT1_MASK 0x3ff
+#define CP_CE_ROQ_IB1_STAT__CEQ_RPTR_INDIRECT1__SHIFT 0x0
+#define CP_CE_ROQ_IB1_STAT__CEQ_WPTR_INDIRECT1_MASK 0x3ff0000
+#define CP_CE_ROQ_IB1_STAT__CEQ_WPTR_INDIRECT1__SHIFT 0x10
+#define CP_CE_ROQ_IB2_STAT__CEQ_RPTR_INDIRECT2_MASK 0x3ff
+#define CP_CE_ROQ_IB2_STAT__CEQ_RPTR_INDIRECT2__SHIFT 0x0
+#define CP_CE_ROQ_IB2_STAT__CEQ_WPTR_INDIRECT2_MASK 0x3ff0000
+#define CP_CE_ROQ_IB2_STAT__CEQ_WPTR_INDIRECT2__SHIFT 0x10
+#define CP_INT_STAT_DEBUG__CP_VM_DOORBELL_WR_INT_ASSERTED_MASK 0x800
+#define CP_INT_STAT_DEBUG__CP_VM_DOORBELL_WR_INT_ASSERTED__SHIFT 0xb
+#define CP_INT_STAT_DEBUG__CP_ECC_ERROR_INT_ASSERTED_MASK 0x4000
+#define CP_INT_STAT_DEBUG__CP_ECC_ERROR_INT_ASSERTED__SHIFT 0xe
+#define CP_INT_STAT_DEBUG__WRM_POLL_TIMEOUT_INT_ASSERTED_MASK 0x20000
+#define CP_INT_STAT_DEBUG__WRM_POLL_TIMEOUT_INT_ASSERTED__SHIFT 0x11
+#define CP_INT_STAT_DEBUG__CMP_BUSY_INT_ASSERTED_MASK 0x40000
+#define CP_INT_STAT_DEBUG__CMP_BUSY_INT_ASSERTED__SHIFT 0x12
+#define CP_INT_STAT_DEBUG__CNTX_BUSY_INT_ASSERTED_MASK 0x80000
+#define CP_INT_STAT_DEBUG__CNTX_BUSY_INT_ASSERTED__SHIFT 0x13
+#define CP_INT_STAT_DEBUG__CNTX_EMPTY_INT_ASSERTED_MASK 0x100000
+#define CP_INT_STAT_DEBUG__CNTX_EMPTY_INT_ASSERTED__SHIFT 0x14
+#define CP_INT_STAT_DEBUG__GFX_IDLE_INT_ASSERTED_MASK 0x200000
+#define CP_INT_STAT_DEBUG__GFX_IDLE_INT_ASSERTED__SHIFT 0x15
+#define CP_INT_STAT_DEBUG__PRIV_INSTR_INT_ASSERTED_MASK 0x400000
+#define CP_INT_STAT_DEBUG__PRIV_INSTR_INT_ASSERTED__SHIFT 0x16
+#define CP_INT_STAT_DEBUG__PRIV_REG_INT_ASSERTED_MASK 0x800000
+#define CP_INT_STAT_DEBUG__PRIV_REG_INT_ASSERTED__SHIFT 0x17
+#define CP_INT_STAT_DEBUG__OPCODE_ERROR_INT_ASSERTED_MASK 0x1000000
+#define CP_INT_STAT_DEBUG__OPCODE_ERROR_INT_ASSERTED__SHIFT 0x18
+#define CP_INT_STAT_DEBUG__TIME_STAMP_INT_ASSERTED_MASK 0x4000000
+#define CP_INT_STAT_DEBUG__TIME_STAMP_INT_ASSERTED__SHIFT 0x1a
+#define CP_INT_STAT_DEBUG__RESERVED_BIT_ERROR_INT_ASSERTED_MASK 0x8000000
+#define CP_INT_STAT_DEBUG__RESERVED_BIT_ERROR_INT_ASSERTED__SHIFT 0x1b
+#define CP_INT_STAT_DEBUG__GENERIC2_INT_ASSERTED_MASK 0x20000000
+#define CP_INT_STAT_DEBUG__GENERIC2_INT_ASSERTED__SHIFT 0x1d
+#define CP_INT_STAT_DEBUG__GENERIC1_INT_ASSERTED_MASK 0x40000000
+#define CP_INT_STAT_DEBUG__GENERIC1_INT_ASSERTED__SHIFT 0x1e
+#define CP_INT_STAT_DEBUG__GENERIC0_INT_ASSERTED_MASK 0x80000000
+#define CP_INT_STAT_DEBUG__GENERIC0_INT_ASSERTED__SHIFT 0x1f
+#define CP_PERFMON_CNTL__PERFMON_STATE_MASK 0xf
+#define CP_PERFMON_CNTL__PERFMON_STATE__SHIFT 0x0
+#define CP_PERFMON_CNTL__SPM_PERFMON_STATE_MASK 0xf0
+#define CP_PERFMON_CNTL__SPM_PERFMON_STATE__SHIFT 0x4
+#define CP_PERFMON_CNTL__PERFMON_ENABLE_MODE_MASK 0x300
+#define CP_PERFMON_CNTL__PERFMON_ENABLE_MODE__SHIFT 0x8
+#define CP_PERFMON_CNTL__PERFMON_SAMPLE_ENABLE_MASK 0x400
+#define CP_PERFMON_CNTL__PERFMON_SAMPLE_ENABLE__SHIFT 0xa
+#define CP_PERFMON_CNTX_CNTL__PERFMON_ENABLE_MASK 0x80000000
+#define CP_PERFMON_CNTX_CNTL__PERFMON_ENABLE__SHIFT 0x1f
+#define CP_RINGID__RINGID_MASK 0x3
+#define CP_RINGID__RINGID__SHIFT 0x0
+#define CP_PIPEID__PIPE_ID_MASK 0x3
+#define CP_PIPEID__PIPE_ID__SHIFT 0x0
+#define CP_VMID__VMID_MASK 0xf
+#define CP_VMID__VMID__SHIFT 0x0
+#define CP_HPD_ROQ_OFFSETS__IQ_OFFSET_MASK 0x7
+#define CP_HPD_ROQ_OFFSETS__IQ_OFFSET__SHIFT 0x0
+#define CP_HPD_ROQ_OFFSETS__PQ_OFFSET_MASK 0x3f00
+#define CP_HPD_ROQ_OFFSETS__PQ_OFFSET__SHIFT 0x8
+#define CP_HPD_ROQ_OFFSETS__IB_OFFSET_MASK 0x3f0000
+#define CP_HPD_ROQ_OFFSETS__IB_OFFSET__SHIFT 0x10
+#define CP_HPD_STATUS0__QUEUE_STATE_MASK 0x1f
+#define CP_HPD_STATUS0__QUEUE_STATE__SHIFT 0x0
+#define CP_HPD_STATUS0__MAPPED_QUEUE_MASK 0xe0
+#define CP_HPD_STATUS0__MAPPED_QUEUE__SHIFT 0x5
+#define CP_HPD_STATUS0__QUEUE_AVAILABLE_MASK 0xff00
+#define CP_HPD_STATUS0__QUEUE_AVAILABLE__SHIFT 0x8
+#define CP_MQD_BASE_ADDR__BASE_ADDR_MASK 0xfffffffc
+#define CP_MQD_BASE_ADDR__BASE_ADDR__SHIFT 0x2
+#define CP_MQD_BASE_ADDR_HI__BASE_ADDR_HI_MASK 0xffff
+#define CP_MQD_BASE_ADDR_HI__BASE_ADDR_HI__SHIFT 0x0
+#define CP_HQD_ACTIVE__ACTIVE_MASK 0x1
+#define CP_HQD_ACTIVE__ACTIVE__SHIFT 0x0
+#define CP_HQD_ACTIVE__BUSY_GATE_MASK 0x2
+#define CP_HQD_ACTIVE__BUSY_GATE__SHIFT 0x1
+#define CP_HQD_VMID__VMID_MASK 0xf
+#define CP_HQD_VMID__VMID__SHIFT 0x0
+#define CP_HQD_VMID__IB_VMID_MASK 0xf00
+#define CP_HQD_VMID__IB_VMID__SHIFT 0x8
+#define CP_HQD_VMID__VQID_MASK 0x3ff0000
+#define CP_HQD_VMID__VQID__SHIFT 0x10
+#define CP_HQD_PERSISTENT_STATE__PRELOAD_REQ_MASK 0x1
+#define CP_HQD_PERSISTENT_STATE__PRELOAD_REQ__SHIFT 0x0
+#define CP_HQD_PERSISTENT_STATE__PRELOAD_SIZE_MASK 0x3ff00
+#define CP_HQD_PERSISTENT_STATE__PRELOAD_SIZE__SHIFT 0x8
+#define CP_HQD_PERSISTENT_STATE__RESTORE_ACTIVE_MASK 0x10000000
+#define CP_HQD_PERSISTENT_STATE__RESTORE_ACTIVE__SHIFT 0x1c
+#define CP_HQD_PERSISTENT_STATE__RELAUNCH_WAVES_MASK 0x20000000
+#define CP_HQD_PERSISTENT_STATE__RELAUNCH_WAVES__SHIFT 0x1d
+#define CP_HQD_PERSISTENT_STATE__QSWITCH_MODE_MASK 0x40000000
+#define CP_HQD_PERSISTENT_STATE__QSWITCH_MODE__SHIFT 0x1e
+#define CP_HQD_PERSISTENT_STATE__DISP_ACTIVE_MASK 0x80000000
+#define CP_HQD_PERSISTENT_STATE__DISP_ACTIVE__SHIFT 0x1f
+#define CP_HQD_PIPE_PRIORITY__PIPE_PRIORITY_MASK 0x3
+#define CP_HQD_PIPE_PRIORITY__PIPE_PRIORITY__SHIFT 0x0
+#define CP_HQD_QUEUE_PRIORITY__PRIORITY_LEVEL_MASK 0xf
+#define CP_HQD_QUEUE_PRIORITY__PRIORITY_LEVEL__SHIFT 0x0
+#define CP_HQD_QUANTUM__QUANTUM_EN_MASK 0x1
+#define CP_HQD_QUANTUM__QUANTUM_EN__SHIFT 0x0
+#define CP_HQD_QUANTUM__QUANTUM_SCALE_MASK 0x10
+#define CP_HQD_QUANTUM__QUANTUM_SCALE__SHIFT 0x4
+#define CP_HQD_QUANTUM__QUANTUM_DURATION_MASK 0x3f00
+#define CP_HQD_QUANTUM__QUANTUM_DURATION__SHIFT 0x8
+#define CP_HQD_QUANTUM__QUANTUM_ACTIVE_MASK 0x80000000
+#define CP_HQD_QUANTUM__QUANTUM_ACTIVE__SHIFT 0x1f
+#define CP_HQD_PQ_BASE__ADDR_MASK 0xffffffff
+#define CP_HQD_PQ_BASE__ADDR__SHIFT 0x0
+#define CP_HQD_PQ_BASE_HI__ADDR_HI_MASK 0xff
+#define CP_HQD_PQ_BASE_HI__ADDR_HI__SHIFT 0x0
+#define CP_HQD_PQ_RPTR__CONSUMED_OFFSET_MASK 0xffffffff
+#define CP_HQD_PQ_RPTR__CONSUMED_OFFSET__SHIFT 0x0
+#define CP_HQD_PQ_RPTR_REPORT_ADDR__RPTR_REPORT_ADDR_MASK 0xfffffffc
+#define CP_HQD_PQ_RPTR_REPORT_ADDR__RPTR_REPORT_ADDR__SHIFT 0x2
+#define CP_HQD_PQ_RPTR_REPORT_ADDR_HI__RPTR_REPORT_ADDR_HI_MASK 0xffff
+#define CP_HQD_PQ_RPTR_REPORT_ADDR_HI__RPTR_REPORT_ADDR_HI__SHIFT 0x0
+#define CP_HQD_PQ_WPTR_POLL_ADDR__WPTR_ADDR_MASK 0xfffffffc
+#define CP_HQD_PQ_WPTR_POLL_ADDR__WPTR_ADDR__SHIFT 0x2
+#define CP_HQD_PQ_WPTR_POLL_ADDR_HI__WPTR_ADDR_HI_MASK 0xffff
+#define CP_HQD_PQ_WPTR_POLL_ADDR_HI__WPTR_ADDR_HI__SHIFT 0x0
+#define CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_MODE_MASK 0x1
+#define CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_MODE__SHIFT 0x0
+#define CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_BIF_DROP_MASK 0x2
+#define CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_BIF_DROP__SHIFT 0x1
+#define CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_OFFSET_MASK 0x7ffffc
+#define CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_OFFSET__SHIFT 0x2
+#define CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_CARRY_BITS_MASK 0x3800000
+#define CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_CARRY_BITS__SHIFT 0x17
+#define CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_SOURCE_MASK 0x10000000
+#define CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_SOURCE__SHIFT 0x1c
+#define CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_SCHD_HIT_MASK 0x20000000
+#define CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_SCHD_HIT__SHIFT 0x1d
+#define CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_EN_MASK 0x40000000
+#define CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_EN__SHIFT 0x1e
+#define CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_HIT_MASK 0x80000000
+#define CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_HIT__SHIFT 0x1f
+#define CP_HQD_PQ_WPTR__OFFSET_MASK 0xffffffff
+#define CP_HQD_PQ_WPTR__OFFSET__SHIFT 0x0
+#define CP_HQD_PQ_CONTROL__QUEUE_SIZE_MASK 0x3f
+#define CP_HQD_PQ_CONTROL__QUEUE_SIZE__SHIFT 0x0
+#define CP_HQD_PQ_CONTROL__RPTR_BLOCK_SIZE_MASK 0x3f00
+#define CP_HQD_PQ_CONTROL__RPTR_BLOCK_SIZE__SHIFT 0x8
+#define CP_HQD_PQ_CONTROL__MTYPE_MASK 0x18000
+#define CP_HQD_PQ_CONTROL__MTYPE__SHIFT 0xf
+#define CP_HQD_PQ_CONTROL__ENDIAN_SWAP_MASK 0x60000
+#define CP_HQD_PQ_CONTROL__ENDIAN_SWAP__SHIFT 0x11
+#define CP_HQD_PQ_CONTROL__MIN_AVAIL_SIZE_MASK 0x300000
+#define CP_HQD_PQ_CONTROL__MIN_AVAIL_SIZE__SHIFT 0x14
+#define CP_HQD_PQ_CONTROL__PQ_ATC_MASK 0x800000
+#define CP_HQD_PQ_CONTROL__PQ_ATC__SHIFT 0x17
+#define CP_HQD_PQ_CONTROL__CACHE_POLICY_MASK 0x1000000
+#define CP_HQD_PQ_CONTROL__CACHE_POLICY__SHIFT 0x18
+#define CP_HQD_PQ_CONTROL__SLOT_BASED_WPTR_MASK 0x6000000
+#define CP_HQD_PQ_CONTROL__SLOT_BASED_WPTR__SHIFT 0x19
+#define CP_HQD_PQ_CONTROL__NO_UPDATE_RPTR_MASK 0x8000000
+#define CP_HQD_PQ_CONTROL__NO_UPDATE_RPTR__SHIFT 0x1b
+#define CP_HQD_PQ_CONTROL__UNORD_DISPATCH_MASK 0x10000000
+#define CP_HQD_PQ_CONTROL__UNORD_DISPATCH__SHIFT 0x1c
+#define CP_HQD_PQ_CONTROL__ROQ_PQ_IB_FLIP_MASK 0x20000000
+#define CP_HQD_PQ_CONTROL__ROQ_PQ_IB_FLIP__SHIFT 0x1d
+#define CP_HQD_PQ_CONTROL__PRIV_STATE_MASK 0x40000000
+#define CP_HQD_PQ_CONTROL__PRIV_STATE__SHIFT 0x1e
+#define CP_HQD_PQ_CONTROL__KMD_QUEUE_MASK 0x80000000
+#define CP_HQD_PQ_CONTROL__KMD_QUEUE__SHIFT 0x1f
+#define CP_HQD_IB_BASE_ADDR__IB_BASE_ADDR_MASK 0xfffffffc
+#define CP_HQD_IB_BASE_ADDR__IB_BASE_ADDR__SHIFT 0x2
+#define CP_HQD_IB_BASE_ADDR_HI__IB_BASE_ADDR_HI_MASK 0xffff
+#define CP_HQD_IB_BASE_ADDR_HI__IB_BASE_ADDR_HI__SHIFT 0x0
+#define CP_HQD_IB_RPTR__CONSUMED_OFFSET_MASK 0xfffff
+#define CP_HQD_IB_RPTR__CONSUMED_OFFSET__SHIFT 0x0
+#define CP_HQD_IB_CONTROL__IB_SIZE_MASK 0xfffff
+#define CP_HQD_IB_CONTROL__IB_SIZE__SHIFT 0x0
+#define CP_HQD_IB_CONTROL__MIN_IB_AVAIL_SIZE_MASK 0x300000
+#define CP_HQD_IB_CONTROL__MIN_IB_AVAIL_SIZE__SHIFT 0x14
+#define CP_HQD_IB_CONTROL__IB_ATC_MASK 0x800000
+#define CP_HQD_IB_CONTROL__IB_ATC__SHIFT 0x17
+#define CP_HQD_IB_CONTROL__IB_CACHE_POLICY_MASK 0x1000000
+#define CP_HQD_IB_CONTROL__IB_CACHE_POLICY__SHIFT 0x18
+#define CP_HQD_IB_CONTROL__MTYPE_MASK 0x18000000
+#define CP_HQD_IB_CONTROL__MTYPE__SHIFT 0x1b
+#define CP_HQD_IB_CONTROL__PROCESSING_IB_MASK 0x80000000
+#define CP_HQD_IB_CONTROL__PROCESSING_IB__SHIFT 0x1f
+#define CP_HQD_IQ_TIMER__WAIT_TIME_MASK 0xff
+#define CP_HQD_IQ_TIMER__WAIT_TIME__SHIFT 0x0
+#define CP_HQD_IQ_TIMER__RETRY_TYPE_MASK 0x700
+#define CP_HQD_IQ_TIMER__RETRY_TYPE__SHIFT 0x8
+#define CP_HQD_IQ_TIMER__IMMEDIATE_EXPIRE_MASK 0x800
+#define CP_HQD_IQ_TIMER__IMMEDIATE_EXPIRE__SHIFT 0xb
+#define CP_HQD_IQ_TIMER__INTERRUPT_TYPE_MASK 0x3000
+#define CP_HQD_IQ_TIMER__INTERRUPT_TYPE__SHIFT 0xc
+#define CP_HQD_IQ_TIMER__CLOCK_COUNT_MASK 0xc000
+#define CP_HQD_IQ_TIMER__CLOCK_COUNT__SHIFT 0xe
+#define CP_HQD_IQ_TIMER__INTERRUPT_SIZE_MASK 0x3f0000
+#define CP_HQD_IQ_TIMER__INTERRUPT_SIZE__SHIFT 0x10
+#define CP_HQD_IQ_TIMER__QUANTUM_TIMER_MASK 0x400000
+#define CP_HQD_IQ_TIMER__QUANTUM_TIMER__SHIFT 0x16
+#define CP_HQD_IQ_TIMER__IQ_ATC_MASK 0x800000
+#define CP_HQD_IQ_TIMER__IQ_ATC__SHIFT 0x17
+#define CP_HQD_IQ_TIMER__CACHE_POLICY_MASK 0x1000000
+#define CP_HQD_IQ_TIMER__CACHE_POLICY__SHIFT 0x18
+#define CP_HQD_IQ_TIMER__MTYPE_MASK 0x18000000
+#define CP_HQD_IQ_TIMER__MTYPE__SHIFT 0x1b
+#define CP_HQD_IQ_TIMER__PROCESS_IQ_EN_MASK 0x20000000
+#define CP_HQD_IQ_TIMER__PROCESS_IQ_EN__SHIFT 0x1d
+#define CP_HQD_IQ_TIMER__PROCESSING_IQ_MASK 0x40000000
+#define CP_HQD_IQ_TIMER__PROCESSING_IQ__SHIFT 0x1e
+#define CP_HQD_IQ_TIMER__ACTIVE_MASK 0x80000000
+#define CP_HQD_IQ_TIMER__ACTIVE__SHIFT 0x1f
+#define CP_HQD_IQ_RPTR__OFFSET_MASK 0x3f
+#define CP_HQD_IQ_RPTR__OFFSET__SHIFT 0x0
+#define CP_HQD_DEQUEUE_REQUEST__DEQUEUE_REQ_MASK 0x7
+#define CP_HQD_DEQUEUE_REQUEST__DEQUEUE_REQ__SHIFT 0x0
+#define CP_HQD_DEQUEUE_REQUEST__IQ_REQ_PEND_MASK 0x10
+#define CP_HQD_DEQUEUE_REQUEST__IQ_REQ_PEND__SHIFT 0x4
+#define CP_HQD_DEQUEUE_REQUEST__DEQUEUE_INT_MASK 0x100
+#define CP_HQD_DEQUEUE_REQUEST__DEQUEUE_INT__SHIFT 0x8
+#define CP_HQD_DEQUEUE_REQUEST__IQ_REQ_PEND_EN_MASK 0x200
+#define CP_HQD_DEQUEUE_REQUEST__IQ_REQ_PEND_EN__SHIFT 0x9
+#define CP_HQD_DEQUEUE_REQUEST__DEQUEUE_REQ_EN_MASK 0x400
+#define CP_HQD_DEQUEUE_REQUEST__DEQUEUE_REQ_EN__SHIFT 0xa
+#define CP_HQD_DMA_OFFLOAD__DMA_OFFLOAD_MASK 0x1
+#define CP_HQD_DMA_OFFLOAD__DMA_OFFLOAD__SHIFT 0x0
+#define CP_HQD_OFFLOAD__DMA_OFFLOAD_MASK 0x1
+#define CP_HQD_OFFLOAD__DMA_OFFLOAD__SHIFT 0x0
+#define CP_HQD_OFFLOAD__DMA_OFFLOAD_EN_MASK 0x2
+#define CP_HQD_OFFLOAD__DMA_OFFLOAD_EN__SHIFT 0x1
+#define CP_HQD_OFFLOAD__EOP_OFFLOAD_MASK 0x10
+#define CP_HQD_OFFLOAD__EOP_OFFLOAD__SHIFT 0x4
+#define CP_HQD_OFFLOAD__EOP_OFFLOAD_EN_MASK 0x20
+#define CP_HQD_OFFLOAD__EOP_OFFLOAD_EN__SHIFT 0x5
+#define CP_HQD_SEMA_CMD__RETRY_MASK 0x1
+#define CP_HQD_SEMA_CMD__RETRY__SHIFT 0x0
+#define CP_HQD_SEMA_CMD__RESULT_MASK 0x6
+#define CP_HQD_SEMA_CMD__RESULT__SHIFT 0x1
+#define CP_HQD_MSG_TYPE__ACTION_MASK 0x7
+#define CP_HQD_MSG_TYPE__ACTION__SHIFT 0x0
+#define CP_HQD_MSG_TYPE__SAVE_STATE_MASK 0x70
+#define CP_HQD_MSG_TYPE__SAVE_STATE__SHIFT 0x4
+#define CP_HQD_ATOMIC0_PREOP_LO__ATOMIC0_PREOP_LO_MASK 0xffffffff
+#define CP_HQD_ATOMIC0_PREOP_LO__ATOMIC0_PREOP_LO__SHIFT 0x0
+#define CP_HQD_ATOMIC0_PREOP_HI__ATOMIC0_PREOP_HI_MASK 0xffffffff
+#define CP_HQD_ATOMIC0_PREOP_HI__ATOMIC0_PREOP_HI__SHIFT 0x0
+#define CP_HQD_ATOMIC1_PREOP_LO__ATOMIC1_PREOP_LO_MASK 0xffffffff
+#define CP_HQD_ATOMIC1_PREOP_LO__ATOMIC1_PREOP_LO__SHIFT 0x0
+#define CP_HQD_ATOMIC1_PREOP_HI__ATOMIC1_PREOP_HI_MASK 0xffffffff
+#define CP_HQD_ATOMIC1_PREOP_HI__ATOMIC1_PREOP_HI__SHIFT 0x0
+#define CP_HQD_HQ_SCHEDULER0__SCHEDULER_MASK 0xffffffff
+#define CP_HQD_HQ_SCHEDULER0__SCHEDULER__SHIFT 0x0
+#define CP_HQD_HQ_STATUS0__DEQUEUE_STATUS_MASK 0x3
+#define CP_HQD_HQ_STATUS0__DEQUEUE_STATUS__SHIFT 0x0
+#define CP_HQD_HQ_STATUS0__DEQUEUE_RETRY_CNT_MASK 0xc
+#define CP_HQD_HQ_STATUS0__DEQUEUE_RETRY_CNT__SHIFT 0x2
+#define CP_HQD_HQ_STATUS0__RSV_6_4_MASK 0x70
+#define CP_HQD_HQ_STATUS0__RSV_6_4__SHIFT 0x4
+#define CP_HQD_HQ_STATUS0__SCRATCH_RAM_INIT_MASK 0x80
+#define CP_HQD_HQ_STATUS0__SCRATCH_RAM_INIT__SHIFT 0x7
+#define CP_HQD_HQ_STATUS0__TCL2_DIRTY_MASK 0x100
+#define CP_HQD_HQ_STATUS0__TCL2_DIRTY__SHIFT 0x8
+#define CP_HQD_HQ_STATUS0__PG_ACTIVATED_MASK 0x200
+#define CP_HQD_HQ_STATUS0__PG_ACTIVATED__SHIFT 0x9
+#define CP_HQD_HQ_STATUS0__RSVR_31_10_MASK 0xfffffc00
+#define CP_HQD_HQ_STATUS0__RSVR_31_10__SHIFT 0xa
+#define CP_HQD_HQ_SCHEDULER1__SCHEDULER_MASK 0xffffffff
+#define CP_HQD_HQ_SCHEDULER1__SCHEDULER__SHIFT 0x0
+#define CP_HQD_HQ_CONTROL0__CONTROL_MASK 0xffffffff
+#define CP_HQD_HQ_CONTROL0__CONTROL__SHIFT 0x0
+#define CP_MQD_CONTROL__VMID_MASK 0xf
+#define CP_MQD_CONTROL__VMID__SHIFT 0x0
+#define CP_MQD_CONTROL__PROCESSING_MQD_MASK 0x1000
+#define CP_MQD_CONTROL__PROCESSING_MQD__SHIFT 0xc
+#define CP_MQD_CONTROL__PROCESSING_MQD_EN_MASK 0x2000
+#define CP_MQD_CONTROL__PROCESSING_MQD_EN__SHIFT 0xd
+#define CP_MQD_CONTROL__MQD_ATC_MASK 0x800000
+#define CP_MQD_CONTROL__MQD_ATC__SHIFT 0x17
+#define CP_MQD_CONTROL__CACHE_POLICY_MASK 0x1000000
+#define CP_MQD_CONTROL__CACHE_POLICY__SHIFT 0x18
+#define CP_MQD_CONTROL__MTYPE_MASK 0x18000000
+#define CP_MQD_CONTROL__MTYPE__SHIFT 0x1b
+#define CP_HQD_HQ_STATUS1__STATUS_MASK 0xffffffff
+#define CP_HQD_HQ_STATUS1__STATUS__SHIFT 0x0
+#define CP_HQD_HQ_CONTROL1__CONTROL_MASK 0xffffffff
+#define CP_HQD_HQ_CONTROL1__CONTROL__SHIFT 0x0
+#define CP_HQD_EOP_BASE_ADDR__BASE_ADDR_MASK 0xffffffff
+#define CP_HQD_EOP_BASE_ADDR__BASE_ADDR__SHIFT 0x0
+#define CP_HQD_EOP_BASE_ADDR_HI__BASE_ADDR_HI_MASK 0xff
+#define CP_HQD_EOP_BASE_ADDR_HI__BASE_ADDR_HI__SHIFT 0x0
+#define CP_HQD_EOP_CONTROL__EOP_SIZE_MASK 0x3f
+#define CP_HQD_EOP_CONTROL__EOP_SIZE__SHIFT 0x0
+#define CP_HQD_EOP_CONTROL__PROCESSING_EOP_MASK 0x100
+#define CP_HQD_EOP_CONTROL__PROCESSING_EOP__SHIFT 0x8
+#define CP_HQD_EOP_CONTROL__PROCESS_EOP_EN_MASK 0x1000
+#define CP_HQD_EOP_CONTROL__PROCESS_EOP_EN__SHIFT 0xc
+#define CP_HQD_EOP_CONTROL__PROCESSING_EOPIB_MASK 0x2000
+#define CP_HQD_EOP_CONTROL__PROCESSING_EOPIB__SHIFT 0xd
+#define CP_HQD_EOP_CONTROL__PROCESS_EOPIB_EN_MASK 0x4000
+#define CP_HQD_EOP_CONTROL__PROCESS_EOPIB_EN__SHIFT 0xe
+#define CP_HQD_EOP_CONTROL__MTYPE_MASK 0x18000
+#define CP_HQD_EOP_CONTROL__MTYPE__SHIFT 0xf
+#define CP_HQD_EOP_CONTROL__EOP_ATC_MASK 0x800000
+#define CP_HQD_EOP_CONTROL__EOP_ATC__SHIFT 0x17
+#define CP_HQD_EOP_CONTROL__CACHE_POLICY_MASK 0x1000000
+#define CP_HQD_EOP_CONTROL__CACHE_POLICY__SHIFT 0x18
+#define CP_HQD_EOP_CONTROL__SIG_SEM_RESULT_MASK 0x60000000
+#define CP_HQD_EOP_CONTROL__SIG_SEM_RESULT__SHIFT 0x1d
+#define CP_HQD_EOP_CONTROL__PEND_SIG_SEM_MASK 0x80000000
+#define CP_HQD_EOP_CONTROL__PEND_SIG_SEM__SHIFT 0x1f
+#define CP_HQD_EOP_RPTR__RPTR_MASK 0x1fff
+#define CP_HQD_EOP_RPTR__RPTR__SHIFT 0x0
+#define CP_HQD_EOP_RPTR__RPTR_EQ_CSMD_WPTR_MASK 0x40000000
+#define CP_HQD_EOP_RPTR__RPTR_EQ_CSMD_WPTR__SHIFT 0x1e
+#define CP_HQD_EOP_RPTR__INIT_FETCHER_MASK 0x80000000
+#define CP_HQD_EOP_RPTR__INIT_FETCHER__SHIFT 0x1f
+#define CP_HQD_EOP_WPTR__WPTR_MASK 0x1fff
+#define CP_HQD_EOP_WPTR__WPTR__SHIFT 0x0
+#define CP_HQD_EOP_WPTR__EOP_AVAIL_MASK 0x1fff0000
+#define CP_HQD_EOP_WPTR__EOP_AVAIL__SHIFT 0x10
+#define CP_HQD_EOP_EVENTS__EVENT_COUNT_MASK 0xfff
+#define CP_HQD_EOP_EVENTS__EVENT_COUNT__SHIFT 0x0
+#define CP_HQD_EOP_EVENTS__CS_PARTIAL_FLUSH_PEND_MASK 0x10000
+#define CP_HQD_EOP_EVENTS__CS_PARTIAL_FLUSH_PEND__SHIFT 0x10
+#define CP_HQD_CTX_SAVE_BASE_ADDR_LO__ADDR_MASK 0xfffff000
+#define CP_HQD_CTX_SAVE_BASE_ADDR_LO__ADDR__SHIFT 0xc
+#define CP_HQD_CTX_SAVE_BASE_ADDR_HI__ADDR_HI_MASK 0xffff
+#define CP_HQD_CTX_SAVE_BASE_ADDR_HI__ADDR_HI__SHIFT 0x0
+#define CP_HQD_CTX_SAVE_CONTROL__ATC_MASK 0x1
+#define CP_HQD_CTX_SAVE_CONTROL__ATC__SHIFT 0x0
+#define CP_HQD_CTX_SAVE_CONTROL__MTYPE_MASK 0x6
+#define CP_HQD_CTX_SAVE_CONTROL__MTYPE__SHIFT 0x1
+#define CP_HQD_CTX_SAVE_CONTROL__POLICY_MASK 0x8
+#define CP_HQD_CTX_SAVE_CONTROL__POLICY__SHIFT 0x3
+#define CP_HQD_CNTL_STACK_OFFSET__OFFSET_MASK 0x7ffc
+#define CP_HQD_CNTL_STACK_OFFSET__OFFSET__SHIFT 0x2
+#define CP_HQD_CNTL_STACK_SIZE__SIZE_MASK 0x7000
+#define CP_HQD_CNTL_STACK_SIZE__SIZE__SHIFT 0xc
+#define CP_HQD_WG_STATE_OFFSET__OFFSET_MASK 0x1fffffc
+#define CP_HQD_WG_STATE_OFFSET__OFFSET__SHIFT 0x2
+#define CP_HQD_CTX_SAVE_SIZE__SIZE_MASK 0x1fff000
+#define CP_HQD_CTX_SAVE_SIZE__SIZE__SHIFT 0xc
+#define CP_HQD_GDS_RESOURCE_STATE__OA_REQUIRED_MASK 0x1
+#define CP_HQD_GDS_RESOURCE_STATE__OA_REQUIRED__SHIFT 0x0
+#define CP_HQD_GDS_RESOURCE_STATE__OA_ACQUIRED_MASK 0x2
+#define CP_HQD_GDS_RESOURCE_STATE__OA_ACQUIRED__SHIFT 0x1
+#define CP_HQD_GDS_RESOURCE_STATE__GWS_SIZE_MASK 0x3f0
+#define CP_HQD_GDS_RESOURCE_STATE__GWS_SIZE__SHIFT 0x4
+#define CP_HQD_GDS_RESOURCE_STATE__GWS_PNTR_MASK 0x3f000
+#define CP_HQD_GDS_RESOURCE_STATE__GWS_PNTR__SHIFT 0xc
+#define CP_HQD_ERROR__EDC_ERROR_ID_MASK 0xf
+#define CP_HQD_ERROR__EDC_ERROR_ID__SHIFT 0x0
+#define CP_HQD_ERROR__SUA_ERROR_MASK 0x10
+#define CP_HQD_ERROR__SUA_ERROR__SHIFT 0x4
+#define CP_HQD_EOP_WPTR_MEM__WPTR_MASK 0x1fff
+#define CP_HQD_EOP_WPTR_MEM__WPTR__SHIFT 0x0
+#define CP_HQD_EOP_DONES__DONE_COUNT_MASK 0xffffffff
+#define CP_HQD_EOP_DONES__DONE_COUNT__SHIFT 0x0
+#define DB_Z_READ_BASE__BASE_256B_MASK 0xffffffff
+#define DB_Z_READ_BASE__BASE_256B__SHIFT 0x0
+#define DB_STENCIL_READ_BASE__BASE_256B_MASK 0xffffffff
+#define DB_STENCIL_READ_BASE__BASE_256B__SHIFT 0x0
+#define DB_Z_WRITE_BASE__BASE_256B_MASK 0xffffffff
+#define DB_Z_WRITE_BASE__BASE_256B__SHIFT 0x0
+#define DB_STENCIL_WRITE_BASE__BASE_256B_MASK 0xffffffff
+#define DB_STENCIL_WRITE_BASE__BASE_256B__SHIFT 0x0
+#define DB_DEPTH_INFO__ADDR5_SWIZZLE_MASK_MASK 0xf
+#define DB_DEPTH_INFO__ADDR5_SWIZZLE_MASK__SHIFT 0x0
+#define DB_DEPTH_INFO__ARRAY_MODE_MASK 0xf0
+#define DB_DEPTH_INFO__ARRAY_MODE__SHIFT 0x4
+#define DB_DEPTH_INFO__PIPE_CONFIG_MASK 0x1f00
+#define DB_DEPTH_INFO__PIPE_CONFIG__SHIFT 0x8
+#define DB_DEPTH_INFO__BANK_WIDTH_MASK 0x6000
+#define DB_DEPTH_INFO__BANK_WIDTH__SHIFT 0xd
+#define DB_DEPTH_INFO__BANK_HEIGHT_MASK 0x18000
+#define DB_DEPTH_INFO__BANK_HEIGHT__SHIFT 0xf
+#define DB_DEPTH_INFO__MACRO_TILE_ASPECT_MASK 0x60000
+#define DB_DEPTH_INFO__MACRO_TILE_ASPECT__SHIFT 0x11
+#define DB_DEPTH_INFO__NUM_BANKS_MASK 0x180000
+#define DB_DEPTH_INFO__NUM_BANKS__SHIFT 0x13
+#define DB_Z_INFO__FORMAT_MASK 0x3
+#define DB_Z_INFO__FORMAT__SHIFT 0x0
+#define DB_Z_INFO__NUM_SAMPLES_MASK 0xc
+#define DB_Z_INFO__NUM_SAMPLES__SHIFT 0x2
+#define DB_Z_INFO__TILE_SPLIT_MASK 0xe000
+#define DB_Z_INFO__TILE_SPLIT__SHIFT 0xd
+#define DB_Z_INFO__TILE_MODE_INDEX_MASK 0x700000
+#define DB_Z_INFO__TILE_MODE_INDEX__SHIFT 0x14
+#define DB_Z_INFO__DECOMPRESS_ON_N_ZPLANES_MASK 0x7800000
+#define DB_Z_INFO__DECOMPRESS_ON_N_ZPLANES__SHIFT 0x17
+#define DB_Z_INFO__ALLOW_EXPCLEAR_MASK 0x8000000
+#define DB_Z_INFO__ALLOW_EXPCLEAR__SHIFT 0x1b
+#define DB_Z_INFO__READ_SIZE_MASK 0x10000000
+#define DB_Z_INFO__READ_SIZE__SHIFT 0x1c
+#define DB_Z_INFO__TILE_SURFACE_ENABLE_MASK 0x20000000
+#define DB_Z_INFO__TILE_SURFACE_ENABLE__SHIFT 0x1d
+#define DB_Z_INFO__CLEAR_DISALLOWED_MASK 0x40000000
+#define DB_Z_INFO__CLEAR_DISALLOWED__SHIFT 0x1e
+#define DB_Z_INFO__ZRANGE_PRECISION_MASK 0x80000000
+#define DB_Z_INFO__ZRANGE_PRECISION__SHIFT 0x1f
+#define DB_STENCIL_INFO__FORMAT_MASK 0x1
+#define DB_STENCIL_INFO__FORMAT__SHIFT 0x0
+#define DB_STENCIL_INFO__TILE_SPLIT_MASK 0xe000
+#define DB_STENCIL_INFO__TILE_SPLIT__SHIFT 0xd
+#define DB_STENCIL_INFO__TILE_MODE_INDEX_MASK 0x700000
+#define DB_STENCIL_INFO__TILE_MODE_INDEX__SHIFT 0x14
+#define DB_STENCIL_INFO__ALLOW_EXPCLEAR_MASK 0x8000000
+#define DB_STENCIL_INFO__ALLOW_EXPCLEAR__SHIFT 0x1b
+#define DB_STENCIL_INFO__TILE_STENCIL_DISABLE_MASK 0x20000000
+#define DB_STENCIL_INFO__TILE_STENCIL_DISABLE__SHIFT 0x1d
+#define DB_STENCIL_INFO__CLEAR_DISALLOWED_MASK 0x40000000
+#define DB_STENCIL_INFO__CLEAR_DISALLOWED__SHIFT 0x1e
+#define DB_DEPTH_SIZE__PITCH_TILE_MAX_MASK 0x7ff
+#define DB_DEPTH_SIZE__PITCH_TILE_MAX__SHIFT 0x0
+#define DB_DEPTH_SIZE__HEIGHT_TILE_MAX_MASK 0x3ff800
+#define DB_DEPTH_SIZE__HEIGHT_TILE_MAX__SHIFT 0xb
+#define DB_DEPTH_SLICE__SLICE_TILE_MAX_MASK 0x3fffff
+#define DB_DEPTH_SLICE__SLICE_TILE_MAX__SHIFT 0x0
+#define DB_DEPTH_VIEW__SLICE_START_MASK 0x7ff
+#define DB_DEPTH_VIEW__SLICE_START__SHIFT 0x0
+#define DB_DEPTH_VIEW__SLICE_MAX_MASK 0xffe000
+#define DB_DEPTH_VIEW__SLICE_MAX__SHIFT 0xd
+#define DB_DEPTH_VIEW__Z_READ_ONLY_MASK 0x1000000
+#define DB_DEPTH_VIEW__Z_READ_ONLY__SHIFT 0x18
+#define DB_DEPTH_VIEW__STENCIL_READ_ONLY_MASK 0x2000000
+#define DB_DEPTH_VIEW__STENCIL_READ_ONLY__SHIFT 0x19
+#define DB_RENDER_CONTROL__DEPTH_CLEAR_ENABLE_MASK 0x1
+#define DB_RENDER_CONTROL__DEPTH_CLEAR_ENABLE__SHIFT 0x0
+#define DB_RENDER_CONTROL__STENCIL_CLEAR_ENABLE_MASK 0x2
+#define DB_RENDER_CONTROL__STENCIL_CLEAR_ENABLE__SHIFT 0x1
+#define DB_RENDER_CONTROL__DEPTH_COPY_MASK 0x4
+#define DB_RENDER_CONTROL__DEPTH_COPY__SHIFT 0x2
+#define DB_RENDER_CONTROL__STENCIL_COPY_MASK 0x8
+#define DB_RENDER_CONTROL__STENCIL_COPY__SHIFT 0x3
+#define DB_RENDER_CONTROL__RESUMMARIZE_ENABLE_MASK 0x10
+#define DB_RENDER_CONTROL__RESUMMARIZE_ENABLE__SHIFT 0x4
+#define DB_RENDER_CONTROL__STENCIL_COMPRESS_DISABLE_MASK 0x20
+#define DB_RENDER_CONTROL__STENCIL_COMPRESS_DISABLE__SHIFT 0x5
+#define DB_RENDER_CONTROL__DEPTH_COMPRESS_DISABLE_MASK 0x40
+#define DB_RENDER_CONTROL__DEPTH_COMPRESS_DISABLE__SHIFT 0x6
+#define DB_RENDER_CONTROL__COPY_CENTROID_MASK 0x80
+#define DB_RENDER_CONTROL__COPY_CENTROID__SHIFT 0x7
+#define DB_RENDER_CONTROL__COPY_SAMPLE_MASK 0xf00
+#define DB_RENDER_CONTROL__COPY_SAMPLE__SHIFT 0x8
+#define DB_RENDER_CONTROL__DECOMPRESS_ENABLE_MASK 0x1000
+#define DB_RENDER_CONTROL__DECOMPRESS_ENABLE__SHIFT 0xc
+#define DB_COUNT_CONTROL__ZPASS_INCREMENT_DISABLE_MASK 0x1
+#define DB_COUNT_CONTROL__ZPASS_INCREMENT_DISABLE__SHIFT 0x0
+#define DB_COUNT_CONTROL__PERFECT_ZPASS_COUNTS_MASK 0x2
+#define DB_COUNT_CONTROL__PERFECT_ZPASS_COUNTS__SHIFT 0x1
+#define DB_COUNT_CONTROL__SAMPLE_RATE_MASK 0x70
+#define DB_COUNT_CONTROL__SAMPLE_RATE__SHIFT 0x4
+#define DB_COUNT_CONTROL__ZPASS_ENABLE_MASK 0xf00
+#define DB_COUNT_CONTROL__ZPASS_ENABLE__SHIFT 0x8
+#define DB_COUNT_CONTROL__ZFAIL_ENABLE_MASK 0xf000
+#define DB_COUNT_CONTROL__ZFAIL_ENABLE__SHIFT 0xc
+#define DB_COUNT_CONTROL__SFAIL_ENABLE_MASK 0xf0000
+#define DB_COUNT_CONTROL__SFAIL_ENABLE__SHIFT 0x10
+#define DB_COUNT_CONTROL__DBFAIL_ENABLE_MASK 0xf00000
+#define DB_COUNT_CONTROL__DBFAIL_ENABLE__SHIFT 0x14
+#define DB_COUNT_CONTROL__SLICE_EVEN_ENABLE_MASK 0xf000000
+#define DB_COUNT_CONTROL__SLICE_EVEN_ENABLE__SHIFT 0x18
+#define DB_COUNT_CONTROL__SLICE_ODD_ENABLE_MASK 0xf0000000
+#define DB_COUNT_CONTROL__SLICE_ODD_ENABLE__SHIFT 0x1c
+#define DB_RENDER_OVERRIDE__FORCE_HIZ_ENABLE_MASK 0x3
+#define DB_RENDER_OVERRIDE__FORCE_HIZ_ENABLE__SHIFT 0x0
+#define DB_RENDER_OVERRIDE__FORCE_HIS_ENABLE0_MASK 0xc
+#define DB_RENDER_OVERRIDE__FORCE_HIS_ENABLE0__SHIFT 0x2
+#define DB_RENDER_OVERRIDE__FORCE_HIS_ENABLE1_MASK 0x30
+#define DB_RENDER_OVERRIDE__FORCE_HIS_ENABLE1__SHIFT 0x4
+#define DB_RENDER_OVERRIDE__FORCE_SHADER_Z_ORDER_MASK 0x40
+#define DB_RENDER_OVERRIDE__FORCE_SHADER_Z_ORDER__SHIFT 0x6
+#define DB_RENDER_OVERRIDE__FAST_Z_DISABLE_MASK 0x80
+#define DB_RENDER_OVERRIDE__FAST_Z_DISABLE__SHIFT 0x7
+#define DB_RENDER_OVERRIDE__FAST_STENCIL_DISABLE_MASK 0x100
+#define DB_RENDER_OVERRIDE__FAST_STENCIL_DISABLE__SHIFT 0x8
+#define DB_RENDER_OVERRIDE__NOOP_CULL_DISABLE_MASK 0x200
+#define DB_RENDER_OVERRIDE__NOOP_CULL_DISABLE__SHIFT 0x9
+#define DB_RENDER_OVERRIDE__FORCE_COLOR_KILL_MASK 0x400
+#define DB_RENDER_OVERRIDE__FORCE_COLOR_KILL__SHIFT 0xa
+#define DB_RENDER_OVERRIDE__FORCE_Z_READ_MASK 0x800
+#define DB_RENDER_OVERRIDE__FORCE_Z_READ__SHIFT 0xb
+#define DB_RENDER_OVERRIDE__FORCE_STENCIL_READ_MASK 0x1000
+#define DB_RENDER_OVERRIDE__FORCE_STENCIL_READ__SHIFT 0xc
+#define DB_RENDER_OVERRIDE__FORCE_FULL_Z_RANGE_MASK 0x6000
+#define DB_RENDER_OVERRIDE__FORCE_FULL_Z_RANGE__SHIFT 0xd
+#define DB_RENDER_OVERRIDE__FORCE_QC_SMASK_CONFLICT_MASK 0x8000
+#define DB_RENDER_OVERRIDE__FORCE_QC_SMASK_CONFLICT__SHIFT 0xf
+#define DB_RENDER_OVERRIDE__DISABLE_VIEWPORT_CLAMP_MASK 0x10000
+#define DB_RENDER_OVERRIDE__DISABLE_VIEWPORT_CLAMP__SHIFT 0x10
+#define DB_RENDER_OVERRIDE__IGNORE_SC_ZRANGE_MASK 0x20000
+#define DB_RENDER_OVERRIDE__IGNORE_SC_ZRANGE__SHIFT 0x11
+#define DB_RENDER_OVERRIDE__DISABLE_FULLY_COVERED_MASK 0x40000
+#define DB_RENDER_OVERRIDE__DISABLE_FULLY_COVERED__SHIFT 0x12
+#define DB_RENDER_OVERRIDE__FORCE_Z_LIMIT_SUMM_MASK 0x180000
+#define DB_RENDER_OVERRIDE__FORCE_Z_LIMIT_SUMM__SHIFT 0x13
+#define DB_RENDER_OVERRIDE__MAX_TILES_IN_DTT_MASK 0x3e00000
+#define DB_RENDER_OVERRIDE__MAX_TILES_IN_DTT__SHIFT 0x15
+#define DB_RENDER_OVERRIDE__DISABLE_TILE_RATE_TILES_MASK 0x4000000
+#define DB_RENDER_OVERRIDE__DISABLE_TILE_RATE_TILES__SHIFT 0x1a
+#define DB_RENDER_OVERRIDE__FORCE_Z_DIRTY_MASK 0x8000000
+#define DB_RENDER_OVERRIDE__FORCE_Z_DIRTY__SHIFT 0x1b
+#define DB_RENDER_OVERRIDE__FORCE_STENCIL_DIRTY_MASK 0x10000000
+#define DB_RENDER_OVERRIDE__FORCE_STENCIL_DIRTY__SHIFT 0x1c
+#define DB_RENDER_OVERRIDE__FORCE_Z_VALID_MASK 0x20000000
+#define DB_RENDER_OVERRIDE__FORCE_Z_VALID__SHIFT 0x1d
+#define DB_RENDER_OVERRIDE__FORCE_STENCIL_VALID_MASK 0x40000000
+#define DB_RENDER_OVERRIDE__FORCE_STENCIL_VALID__SHIFT 0x1e
+#define DB_RENDER_OVERRIDE__PRESERVE_COMPRESSION_MASK 0x80000000
+#define DB_RENDER_OVERRIDE__PRESERVE_COMPRESSION__SHIFT 0x1f
+#define DB_RENDER_OVERRIDE2__PARTIAL_SQUAD_LAUNCH_CONTROL_MASK 0x3
+#define DB_RENDER_OVERRIDE2__PARTIAL_SQUAD_LAUNCH_CONTROL__SHIFT 0x0
+#define DB_RENDER_OVERRIDE2__PARTIAL_SQUAD_LAUNCH_COUNTDOWN_MASK 0x1c
+#define DB_RENDER_OVERRIDE2__PARTIAL_SQUAD_LAUNCH_COUNTDOWN__SHIFT 0x2
+#define DB_RENDER_OVERRIDE2__DISABLE_ZMASK_EXPCLEAR_OPTIMIZATION_MASK 0x20
+#define DB_RENDER_OVERRIDE2__DISABLE_ZMASK_EXPCLEAR_OPTIMIZATION__SHIFT 0x5
+#define DB_RENDER_OVERRIDE2__DISABLE_SMEM_EXPCLEAR_OPTIMIZATION_MASK 0x40
+#define DB_RENDER_OVERRIDE2__DISABLE_SMEM_EXPCLEAR_OPTIMIZATION__SHIFT 0x6
+#define DB_RENDER_OVERRIDE2__DISABLE_COLOR_ON_VALIDATION_MASK 0x80
+#define DB_RENDER_OVERRIDE2__DISABLE_COLOR_ON_VALIDATION__SHIFT 0x7
+#define DB_RENDER_OVERRIDE2__DECOMPRESS_Z_ON_FLUSH_MASK 0x100
+#define DB_RENDER_OVERRIDE2__DECOMPRESS_Z_ON_FLUSH__SHIFT 0x8
+#define DB_RENDER_OVERRIDE2__DISABLE_REG_SNOOP_MASK 0x200
+#define DB_RENDER_OVERRIDE2__DISABLE_REG_SNOOP__SHIFT 0x9
+#define DB_RENDER_OVERRIDE2__DEPTH_BOUNDS_HIER_DEPTH_DISABLE_MASK 0x400
+#define DB_RENDER_OVERRIDE2__DEPTH_BOUNDS_HIER_DEPTH_DISABLE__SHIFT 0xa
+#define DB_RENDER_OVERRIDE2__SEPARATE_HIZS_FUNC_ENABLE_MASK 0x800
+#define DB_RENDER_OVERRIDE2__SEPARATE_HIZS_FUNC_ENABLE__SHIFT 0xb
+#define DB_RENDER_OVERRIDE2__HIZ_ZFUNC_MASK 0x7000
+#define DB_RENDER_OVERRIDE2__HIZ_ZFUNC__SHIFT 0xc
+#define DB_RENDER_OVERRIDE2__HIS_SFUNC_FF_MASK 0x38000
+#define DB_RENDER_OVERRIDE2__HIS_SFUNC_FF__SHIFT 0xf
+#define DB_RENDER_OVERRIDE2__HIS_SFUNC_BF_MASK 0x1c0000
+#define DB_RENDER_OVERRIDE2__HIS_SFUNC_BF__SHIFT 0x12
+#define DB_RENDER_OVERRIDE2__PRESERVE_ZRANGE_MASK 0x200000
+#define DB_RENDER_OVERRIDE2__PRESERVE_ZRANGE__SHIFT 0x15
+#define DB_RENDER_OVERRIDE2__PRESERVE_SRESULTS_MASK 0x400000
+#define DB_RENDER_OVERRIDE2__PRESERVE_SRESULTS__SHIFT 0x16
+#define DB_RENDER_OVERRIDE2__DISABLE_FAST_PASS_MASK 0x800000
+#define DB_RENDER_OVERRIDE2__DISABLE_FAST_PASS__SHIFT 0x17
+#define DB_EQAA__MAX_ANCHOR_SAMPLES_MASK 0x7
+#define DB_EQAA__MAX_ANCHOR_SAMPLES__SHIFT 0x0
+#define DB_EQAA__PS_ITER_SAMPLES_MASK 0x70
+#define DB_EQAA__PS_ITER_SAMPLES__SHIFT 0x4
+#define DB_EQAA__MASK_EXPORT_NUM_SAMPLES_MASK 0x700
+#define DB_EQAA__MASK_EXPORT_NUM_SAMPLES__SHIFT 0x8
+#define DB_EQAA__ALPHA_TO_MASK_NUM_SAMPLES_MASK 0x7000
+#define DB_EQAA__ALPHA_TO_MASK_NUM_SAMPLES__SHIFT 0xc
+#define DB_EQAA__HIGH_QUALITY_INTERSECTIONS_MASK 0x10000
+#define DB_EQAA__HIGH_QUALITY_INTERSECTIONS__SHIFT 0x10
+#define DB_EQAA__INCOHERENT_EQAA_READS_MASK 0x20000
+#define DB_EQAA__INCOHERENT_EQAA_READS__SHIFT 0x11
+#define DB_EQAA__INTERPOLATE_COMP_Z_MASK 0x40000
+#define DB_EQAA__INTERPOLATE_COMP_Z__SHIFT 0x12
+#define DB_EQAA__INTERPOLATE_SRC_Z_MASK 0x80000
+#define DB_EQAA__INTERPOLATE_SRC_Z__SHIFT 0x13
+#define DB_EQAA__STATIC_ANCHOR_ASSOCIATIONS_MASK 0x100000
+#define DB_EQAA__STATIC_ANCHOR_ASSOCIATIONS__SHIFT 0x14
+#define DB_EQAA__ALPHA_TO_MASK_EQAA_DISABLE_MASK 0x200000
+#define DB_EQAA__ALPHA_TO_MASK_EQAA_DISABLE__SHIFT 0x15
+#define DB_EQAA__OVERRASTERIZATION_AMOUNT_MASK 0x7000000
+#define DB_EQAA__OVERRASTERIZATION_AMOUNT__SHIFT 0x18
+#define DB_EQAA__ENABLE_POSTZ_OVERRASTERIZATION_MASK 0x8000000
+#define DB_EQAA__ENABLE_POSTZ_OVERRASTERIZATION__SHIFT 0x1b
+#define DB_SHADER_CONTROL__Z_EXPORT_ENABLE_MASK 0x1
+#define DB_SHADER_CONTROL__Z_EXPORT_ENABLE__SHIFT 0x0
+#define DB_SHADER_CONTROL__STENCIL_TEST_VAL_EXPORT_ENABLE_MASK 0x2
+#define DB_SHADER_CONTROL__STENCIL_TEST_VAL_EXPORT_ENABLE__SHIFT 0x1
+#define DB_SHADER_CONTROL__STENCIL_OP_VAL_EXPORT_ENABLE_MASK 0x4
+#define DB_SHADER_CONTROL__STENCIL_OP_VAL_EXPORT_ENABLE__SHIFT 0x2
+#define DB_SHADER_CONTROL__Z_ORDER_MASK 0x30
+#define DB_SHADER_CONTROL__Z_ORDER__SHIFT 0x4
+#define DB_SHADER_CONTROL__KILL_ENABLE_MASK 0x40
+#define DB_SHADER_CONTROL__KILL_ENABLE__SHIFT 0x6
+#define DB_SHADER_CONTROL__COVERAGE_TO_MASK_ENABLE_MASK 0x80
+#define DB_SHADER_CONTROL__COVERAGE_TO_MASK_ENABLE__SHIFT 0x7
+#define DB_SHADER_CONTROL__MASK_EXPORT_ENABLE_MASK 0x100
+#define DB_SHADER_CONTROL__MASK_EXPORT_ENABLE__SHIFT 0x8
+#define DB_SHADER_CONTROL__EXEC_ON_HIER_FAIL_MASK 0x200
+#define DB_SHADER_CONTROL__EXEC_ON_HIER_FAIL__SHIFT 0x9
+#define DB_SHADER_CONTROL__EXEC_ON_NOOP_MASK 0x400
+#define DB_SHADER_CONTROL__EXEC_ON_NOOP__SHIFT 0xa
+#define DB_SHADER_CONTROL__ALPHA_TO_MASK_DISABLE_MASK 0x800
+#define DB_SHADER_CONTROL__ALPHA_TO_MASK_DISABLE__SHIFT 0xb
+#define DB_SHADER_CONTROL__DEPTH_BEFORE_SHADER_MASK 0x1000
+#define DB_SHADER_CONTROL__DEPTH_BEFORE_SHADER__SHIFT 0xc
+#define DB_SHADER_CONTROL__CONSERVATIVE_Z_EXPORT_MASK 0x6000
+#define DB_SHADER_CONTROL__CONSERVATIVE_Z_EXPORT__SHIFT 0xd
+#define DB_SHADER_CONTROL__DUAL_QUAD_DISABLE_MASK 0x8000
+#define DB_SHADER_CONTROL__DUAL_QUAD_DISABLE__SHIFT 0xf
+#define DB_DEPTH_BOUNDS_MIN__MIN_MASK 0xffffffff
+#define DB_DEPTH_BOUNDS_MIN__MIN__SHIFT 0x0
+#define DB_DEPTH_BOUNDS_MAX__MAX_MASK 0xffffffff
+#define DB_DEPTH_BOUNDS_MAX__MAX__SHIFT 0x0
+#define DB_STENCIL_CLEAR__CLEAR_MASK 0xff
+#define DB_STENCIL_CLEAR__CLEAR__SHIFT 0x0
+#define DB_DEPTH_CLEAR__DEPTH_CLEAR_MASK 0xffffffff
+#define DB_DEPTH_CLEAR__DEPTH_CLEAR__SHIFT 0x0
+#define DB_HTILE_DATA_BASE__BASE_256B_MASK 0xffffffff
+#define DB_HTILE_DATA_BASE__BASE_256B__SHIFT 0x0
+#define DB_HTILE_SURFACE__LINEAR_MASK 0x1
+#define DB_HTILE_SURFACE__LINEAR__SHIFT 0x0
+#define DB_HTILE_SURFACE__FULL_CACHE_MASK 0x2
+#define DB_HTILE_SURFACE__FULL_CACHE__SHIFT 0x1
+#define DB_HTILE_SURFACE__HTILE_USES_PRELOAD_WIN_MASK 0x4
+#define DB_HTILE_SURFACE__HTILE_USES_PRELOAD_WIN__SHIFT 0x2
+#define DB_HTILE_SURFACE__PRELOAD_MASK 0x8
+#define DB_HTILE_SURFACE__PRELOAD__SHIFT 0x3
+#define DB_HTILE_SURFACE__PREFETCH_WIDTH_MASK 0x3f0
+#define DB_HTILE_SURFACE__PREFETCH_WIDTH__SHIFT 0x4
+#define DB_HTILE_SURFACE__PREFETCH_HEIGHT_MASK 0xfc00
+#define DB_HTILE_SURFACE__PREFETCH_HEIGHT__SHIFT 0xa
+#define DB_HTILE_SURFACE__DST_OUTSIDE_ZERO_TO_ONE_MASK 0x10000
+#define DB_HTILE_SURFACE__DST_OUTSIDE_ZERO_TO_ONE__SHIFT 0x10
+#define DB_HTILE_SURFACE__TC_COMPATIBLE_MASK 0x20000
+#define DB_HTILE_SURFACE__TC_COMPATIBLE__SHIFT 0x11
+#define DB_PRELOAD_CONTROL__START_X_MASK 0xff
+#define DB_PRELOAD_CONTROL__START_X__SHIFT 0x0
+#define DB_PRELOAD_CONTROL__START_Y_MASK 0xff00
+#define DB_PRELOAD_CONTROL__START_Y__SHIFT 0x8
+#define DB_PRELOAD_CONTROL__MAX_X_MASK 0xff0000
+#define DB_PRELOAD_CONTROL__MAX_X__SHIFT 0x10
+#define DB_PRELOAD_CONTROL__MAX_Y_MASK 0xff000000
+#define DB_PRELOAD_CONTROL__MAX_Y__SHIFT 0x18
+#define DB_STENCILREFMASK__STENCILTESTVAL_MASK 0xff
+#define DB_STENCILREFMASK__STENCILTESTVAL__SHIFT 0x0
+#define DB_STENCILREFMASK__STENCILMASK_MASK 0xff00
+#define DB_STENCILREFMASK__STENCILMASK__SHIFT 0x8
+#define DB_STENCILREFMASK__STENCILWRITEMASK_MASK 0xff0000
+#define DB_STENCILREFMASK__STENCILWRITEMASK__SHIFT 0x10
+#define DB_STENCILREFMASK__STENCILOPVAL_MASK 0xff000000
+#define DB_STENCILREFMASK__STENCILOPVAL__SHIFT 0x18
+#define DB_STENCILREFMASK_BF__STENCILTESTVAL_BF_MASK 0xff
+#define DB_STENCILREFMASK_BF__STENCILTESTVAL_BF__SHIFT 0x0
+#define DB_STENCILREFMASK_BF__STENCILMASK_BF_MASK 0xff00
+#define DB_STENCILREFMASK_BF__STENCILMASK_BF__SHIFT 0x8
+#define DB_STENCILREFMASK_BF__STENCILWRITEMASK_BF_MASK 0xff0000
+#define DB_STENCILREFMASK_BF__STENCILWRITEMASK_BF__SHIFT 0x10
+#define DB_STENCILREFMASK_BF__STENCILOPVAL_BF_MASK 0xff000000
+#define DB_STENCILREFMASK_BF__STENCILOPVAL_BF__SHIFT 0x18
+#define DB_SRESULTS_COMPARE_STATE0__COMPAREFUNC0_MASK 0x7
+#define DB_SRESULTS_COMPARE_STATE0__COMPAREFUNC0__SHIFT 0x0
+#define DB_SRESULTS_COMPARE_STATE0__COMPAREVALUE0_MASK 0xff0
+#define DB_SRESULTS_COMPARE_STATE0__COMPAREVALUE0__SHIFT 0x4
+#define DB_SRESULTS_COMPARE_STATE0__COMPAREMASK0_MASK 0xff000
+#define DB_SRESULTS_COMPARE_STATE0__COMPAREMASK0__SHIFT 0xc
+#define DB_SRESULTS_COMPARE_STATE0__ENABLE0_MASK 0x1000000
+#define DB_SRESULTS_COMPARE_STATE0__ENABLE0__SHIFT 0x18
+#define DB_SRESULTS_COMPARE_STATE1__COMPAREFUNC1_MASK 0x7
+#define DB_SRESULTS_COMPARE_STATE1__COMPAREFUNC1__SHIFT 0x0
+#define DB_SRESULTS_COMPARE_STATE1__COMPAREVALUE1_MASK 0xff0
+#define DB_SRESULTS_COMPARE_STATE1__COMPAREVALUE1__SHIFT 0x4
+#define DB_SRESULTS_COMPARE_STATE1__COMPAREMASK1_MASK 0xff000
+#define DB_SRESULTS_COMPARE_STATE1__COMPAREMASK1__SHIFT 0xc
+#define DB_SRESULTS_COMPARE_STATE1__ENABLE1_MASK 0x1000000
+#define DB_SRESULTS_COMPARE_STATE1__ENABLE1__SHIFT 0x18
+#define DB_DEPTH_CONTROL__STENCIL_ENABLE_MASK 0x1
+#define DB_DEPTH_CONTROL__STENCIL_ENABLE__SHIFT 0x0
+#define DB_DEPTH_CONTROL__Z_ENABLE_MASK 0x2
+#define DB_DEPTH_CONTROL__Z_ENABLE__SHIFT 0x1
+#define DB_DEPTH_CONTROL__Z_WRITE_ENABLE_MASK 0x4
+#define DB_DEPTH_CONTROL__Z_WRITE_ENABLE__SHIFT 0x2
+#define DB_DEPTH_CONTROL__DEPTH_BOUNDS_ENABLE_MASK 0x8
+#define DB_DEPTH_CONTROL__DEPTH_BOUNDS_ENABLE__SHIFT 0x3
+#define DB_DEPTH_CONTROL__ZFUNC_MASK 0x70
+#define DB_DEPTH_CONTROL__ZFUNC__SHIFT 0x4
+#define DB_DEPTH_CONTROL__BACKFACE_ENABLE_MASK 0x80
+#define DB_DEPTH_CONTROL__BACKFACE_ENABLE__SHIFT 0x7
+#define DB_DEPTH_CONTROL__STENCILFUNC_MASK 0x700
+#define DB_DEPTH_CONTROL__STENCILFUNC__SHIFT 0x8
+#define DB_DEPTH_CONTROL__STENCILFUNC_BF_MASK 0x700000
+#define DB_DEPTH_CONTROL__STENCILFUNC_BF__SHIFT 0x14
+#define DB_DEPTH_CONTROL__ENABLE_COLOR_WRITES_ON_DEPTH_FAIL_MASK 0x40000000
+#define DB_DEPTH_CONTROL__ENABLE_COLOR_WRITES_ON_DEPTH_FAIL__SHIFT 0x1e
+#define DB_DEPTH_CONTROL__DISABLE_COLOR_WRITES_ON_DEPTH_PASS_MASK 0x80000000
+#define DB_DEPTH_CONTROL__DISABLE_COLOR_WRITES_ON_DEPTH_PASS__SHIFT 0x1f
+#define DB_STENCIL_CONTROL__STENCILFAIL_MASK 0xf
+#define DB_STENCIL_CONTROL__STENCILFAIL__SHIFT 0x0
+#define DB_STENCIL_CONTROL__STENCILZPASS_MASK 0xf0
+#define DB_STENCIL_CONTROL__STENCILZPASS__SHIFT 0x4
+#define DB_STENCIL_CONTROL__STENCILZFAIL_MASK 0xf00
+#define DB_STENCIL_CONTROL__STENCILZFAIL__SHIFT 0x8
+#define DB_STENCIL_CONTROL__STENCILFAIL_BF_MASK 0xf000
+#define DB_STENCIL_CONTROL__STENCILFAIL_BF__SHIFT 0xc
+#define DB_STENCIL_CONTROL__STENCILZPASS_BF_MASK 0xf0000
+#define DB_STENCIL_CONTROL__STENCILZPASS_BF__SHIFT 0x10
+#define DB_STENCIL_CONTROL__STENCILZFAIL_BF_MASK 0xf00000
+#define DB_STENCIL_CONTROL__STENCILZFAIL_BF__SHIFT 0x14
+#define DB_ALPHA_TO_MASK__ALPHA_TO_MASK_ENABLE_MASK 0x1
+#define DB_ALPHA_TO_MASK__ALPHA_TO_MASK_ENABLE__SHIFT 0x0
+#define DB_ALPHA_TO_MASK__ALPHA_TO_MASK_OFFSET0_MASK 0x300
+#define DB_ALPHA_TO_MASK__ALPHA_TO_MASK_OFFSET0__SHIFT 0x8
+#define DB_ALPHA_TO_MASK__ALPHA_TO_MASK_OFFSET1_MASK 0xc00
+#define DB_ALPHA_TO_MASK__ALPHA_TO_MASK_OFFSET1__SHIFT 0xa
+#define DB_ALPHA_TO_MASK__ALPHA_TO_MASK_OFFSET2_MASK 0x3000
+#define DB_ALPHA_TO_MASK__ALPHA_TO_MASK_OFFSET2__SHIFT 0xc
+#define DB_ALPHA_TO_MASK__ALPHA_TO_MASK_OFFSET3_MASK 0xc000
+#define DB_ALPHA_TO_MASK__ALPHA_TO_MASK_OFFSET3__SHIFT 0xe
+#define DB_ALPHA_TO_MASK__OFFSET_ROUND_MASK 0x10000
+#define DB_ALPHA_TO_MASK__OFFSET_ROUND__SHIFT 0x10
+#define DB_PERFCOUNTER0_SELECT__PERF_SEL_MASK 0x3ff
+#define DB_PERFCOUNTER0_SELECT__PERF_SEL__SHIFT 0x0
+#define DB_PERFCOUNTER0_SELECT__PERF_SEL1_MASK 0xffc00
+#define DB_PERFCOUNTER0_SELECT__PERF_SEL1__SHIFT 0xa
+#define DB_PERFCOUNTER0_SELECT__CNTR_MODE_MASK 0xf00000
+#define DB_PERFCOUNTER0_SELECT__CNTR_MODE__SHIFT 0x14
+#define DB_PERFCOUNTER0_SELECT__PERF_MODE1_MASK 0xf000000
+#define DB_PERFCOUNTER0_SELECT__PERF_MODE1__SHIFT 0x18
+#define DB_PERFCOUNTER0_SELECT__PERF_MODE_MASK 0xf0000000
+#define DB_PERFCOUNTER0_SELECT__PERF_MODE__SHIFT 0x1c
+#define DB_PERFCOUNTER1_SELECT__PERF_SEL_MASK 0x3ff
+#define DB_PERFCOUNTER1_SELECT__PERF_SEL__SHIFT 0x0
+#define DB_PERFCOUNTER1_SELECT__PERF_SEL1_MASK 0xffc00
+#define DB_PERFCOUNTER1_SELECT__PERF_SEL1__SHIFT 0xa
+#define DB_PERFCOUNTER1_SELECT__CNTR_MODE_MASK 0xf00000
+#define DB_PERFCOUNTER1_SELECT__CNTR_MODE__SHIFT 0x14
+#define DB_PERFCOUNTER1_SELECT__PERF_MODE1_MASK 0xf000000
+#define DB_PERFCOUNTER1_SELECT__PERF_MODE1__SHIFT 0x18
+#define DB_PERFCOUNTER1_SELECT__PERF_MODE_MASK 0xf0000000
+#define DB_PERFCOUNTER1_SELECT__PERF_MODE__SHIFT 0x1c
+#define DB_PERFCOUNTER2_SELECT__PERF_SEL_MASK 0x3ff
+#define DB_PERFCOUNTER2_SELECT__PERF_SEL__SHIFT 0x0
+#define DB_PERFCOUNTER2_SELECT__PERF_SEL1_MASK 0xffc00
+#define DB_PERFCOUNTER2_SELECT__PERF_SEL1__SHIFT 0xa
+#define DB_PERFCOUNTER2_SELECT__CNTR_MODE_MASK 0xf00000
+#define DB_PERFCOUNTER2_SELECT__CNTR_MODE__SHIFT 0x14
+#define DB_PERFCOUNTER2_SELECT__PERF_MODE1_MASK 0xf000000
+#define DB_PERFCOUNTER2_SELECT__PERF_MODE1__SHIFT 0x18
+#define DB_PERFCOUNTER2_SELECT__PERF_MODE_MASK 0xf0000000
+#define DB_PERFCOUNTER2_SELECT__PERF_MODE__SHIFT 0x1c
+#define DB_PERFCOUNTER3_SELECT__PERF_SEL_MASK 0x3ff
+#define DB_PERFCOUNTER3_SELECT__PERF_SEL__SHIFT 0x0
+#define DB_PERFCOUNTER3_SELECT__PERF_SEL1_MASK 0xffc00
+#define DB_PERFCOUNTER3_SELECT__PERF_SEL1__SHIFT 0xa
+#define DB_PERFCOUNTER3_SELECT__CNTR_MODE_MASK 0xf00000
+#define DB_PERFCOUNTER3_SELECT__CNTR_MODE__SHIFT 0x14
+#define DB_PERFCOUNTER3_SELECT__PERF_MODE1_MASK 0xf000000
+#define DB_PERFCOUNTER3_SELECT__PERF_MODE1__SHIFT 0x18
+#define DB_PERFCOUNTER3_SELECT__PERF_MODE_MASK 0xf0000000
+#define DB_PERFCOUNTER3_SELECT__PERF_MODE__SHIFT 0x1c
+#define DB_PERFCOUNTER0_SELECT1__PERF_SEL2_MASK 0x3ff
+#define DB_PERFCOUNTER0_SELECT1__PERF_SEL2__SHIFT 0x0
+#define DB_PERFCOUNTER0_SELECT1__PERF_SEL3_MASK 0xffc00
+#define DB_PERFCOUNTER0_SELECT1__PERF_SEL3__SHIFT 0xa
+#define DB_PERFCOUNTER0_SELECT1__PERF_MODE3_MASK 0xf000000
+#define DB_PERFCOUNTER0_SELECT1__PERF_MODE3__SHIFT 0x18
+#define DB_PERFCOUNTER0_SELECT1__PERF_MODE2_MASK 0xf0000000
+#define DB_PERFCOUNTER0_SELECT1__PERF_MODE2__SHIFT 0x1c
+#define DB_PERFCOUNTER1_SELECT1__PERF_SEL2_MASK 0x3ff
+#define DB_PERFCOUNTER1_SELECT1__PERF_SEL2__SHIFT 0x0
+#define DB_PERFCOUNTER1_SELECT1__PERF_SEL3_MASK 0xffc00
+#define DB_PERFCOUNTER1_SELECT1__PERF_SEL3__SHIFT 0xa
+#define DB_PERFCOUNTER1_SELECT1__PERF_MODE3_MASK 0xf000000
+#define DB_PERFCOUNTER1_SELECT1__PERF_MODE3__SHIFT 0x18
+#define DB_PERFCOUNTER1_SELECT1__PERF_MODE2_MASK 0xf0000000
+#define DB_PERFCOUNTER1_SELECT1__PERF_MODE2__SHIFT 0x1c
+#define DB_PERFCOUNTER0_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define DB_PERFCOUNTER0_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define DB_PERFCOUNTER1_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define DB_PERFCOUNTER1_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define DB_PERFCOUNTER2_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define DB_PERFCOUNTER2_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define DB_PERFCOUNTER3_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define DB_PERFCOUNTER3_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define DB_PERFCOUNTER0_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define DB_PERFCOUNTER0_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define DB_PERFCOUNTER1_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define DB_PERFCOUNTER1_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define DB_PERFCOUNTER2_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define DB_PERFCOUNTER2_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define DB_PERFCOUNTER3_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define DB_PERFCOUNTER3_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define DB_DEBUG__DEBUG_STENCIL_COMPRESS_DISABLE_MASK 0x1
+#define DB_DEBUG__DEBUG_STENCIL_COMPRESS_DISABLE__SHIFT 0x0
+#define DB_DEBUG__DEBUG_DEPTH_COMPRESS_DISABLE_MASK 0x2
+#define DB_DEBUG__DEBUG_DEPTH_COMPRESS_DISABLE__SHIFT 0x1
+#define DB_DEBUG__FETCH_FULL_Z_TILE_MASK 0x4
+#define DB_DEBUG__FETCH_FULL_Z_TILE__SHIFT 0x2
+#define DB_DEBUG__FETCH_FULL_STENCIL_TILE_MASK 0x8
+#define DB_DEBUG__FETCH_FULL_STENCIL_TILE__SHIFT 0x3
+#define DB_DEBUG__FORCE_Z_MODE_MASK 0x30
+#define DB_DEBUG__FORCE_Z_MODE__SHIFT 0x4
+#define DB_DEBUG__DEBUG_FORCE_DEPTH_READ_MASK 0x40
+#define DB_DEBUG__DEBUG_FORCE_DEPTH_READ__SHIFT 0x6
+#define DB_DEBUG__DEBUG_FORCE_STENCIL_READ_MASK 0x80
+#define DB_DEBUG__DEBUG_FORCE_STENCIL_READ__SHIFT 0x7
+#define DB_DEBUG__DEBUG_FORCE_HIZ_ENABLE_MASK 0x300
+#define DB_DEBUG__DEBUG_FORCE_HIZ_ENABLE__SHIFT 0x8
+#define DB_DEBUG__DEBUG_FORCE_HIS_ENABLE0_MASK 0xc00
+#define DB_DEBUG__DEBUG_FORCE_HIS_ENABLE0__SHIFT 0xa
+#define DB_DEBUG__DEBUG_FORCE_HIS_ENABLE1_MASK 0x3000
+#define DB_DEBUG__DEBUG_FORCE_HIS_ENABLE1__SHIFT 0xc
+#define DB_DEBUG__DEBUG_FAST_Z_DISABLE_MASK 0x4000
+#define DB_DEBUG__DEBUG_FAST_Z_DISABLE__SHIFT 0xe
+#define DB_DEBUG__DEBUG_FAST_STENCIL_DISABLE_MASK 0x8000
+#define DB_DEBUG__DEBUG_FAST_STENCIL_DISABLE__SHIFT 0xf
+#define DB_DEBUG__DEBUG_NOOP_CULL_DISABLE_MASK 0x10000
+#define DB_DEBUG__DEBUG_NOOP_CULL_DISABLE__SHIFT 0x10
+#define DB_DEBUG__DISABLE_SUMM_SQUADS_MASK 0x20000
+#define DB_DEBUG__DISABLE_SUMM_SQUADS__SHIFT 0x11
+#define DB_DEBUG__DEPTH_CACHE_FORCE_MISS_MASK 0x40000
+#define DB_DEBUG__DEPTH_CACHE_FORCE_MISS__SHIFT 0x12
+#define DB_DEBUG__DEBUG_FORCE_FULL_Z_RANGE_MASK 0x180000
+#define DB_DEBUG__DEBUG_FORCE_FULL_Z_RANGE__SHIFT 0x13
+#define DB_DEBUG__NEVER_FREE_Z_ONLY_MASK 0x200000
+#define DB_DEBUG__NEVER_FREE_Z_ONLY__SHIFT 0x15
+#define DB_DEBUG__ZPASS_COUNTS_LOOK_AT_PIPE_STAT_EVENTS_MASK 0x400000
+#define DB_DEBUG__ZPASS_COUNTS_LOOK_AT_PIPE_STAT_EVENTS__SHIFT 0x16
+#define DB_DEBUG__DISABLE_VPORT_ZPLANE_OPTIMIZATION_MASK 0x800000
+#define DB_DEBUG__DISABLE_VPORT_ZPLANE_OPTIMIZATION__SHIFT 0x17
+#define DB_DEBUG__DECOMPRESS_AFTER_N_ZPLANES_MASK 0xf000000
+#define DB_DEBUG__DECOMPRESS_AFTER_N_ZPLANES__SHIFT 0x18
+#define DB_DEBUG__ONE_FREE_IN_FLIGHT_MASK 0x10000000
+#define DB_DEBUG__ONE_FREE_IN_FLIGHT__SHIFT 0x1c
+#define DB_DEBUG__FORCE_MISS_IF_NOT_INFLIGHT_MASK 0x20000000
+#define DB_DEBUG__FORCE_MISS_IF_NOT_INFLIGHT__SHIFT 0x1d
+#define DB_DEBUG__DISABLE_DEPTH_SURFACE_SYNC_MASK 0x40000000
+#define DB_DEBUG__DISABLE_DEPTH_SURFACE_SYNC__SHIFT 0x1e
+#define DB_DEBUG__DISABLE_HTILE_SURFACE_SYNC_MASK 0x80000000
+#define DB_DEBUG__DISABLE_HTILE_SURFACE_SYNC__SHIFT 0x1f
+#define DB_DEBUG2__ALLOW_COMPZ_BYTE_MASKING_MASK 0x1
+#define DB_DEBUG2__ALLOW_COMPZ_BYTE_MASKING__SHIFT 0x0
+#define DB_DEBUG2__DISABLE_TC_ZRANGE_L0_CACHE_MASK 0x2
+#define DB_DEBUG2__DISABLE_TC_ZRANGE_L0_CACHE__SHIFT 0x1
+#define DB_DEBUG2__DISABLE_TC_MASK_L0_CACHE_MASK 0x4
+#define DB_DEBUG2__DISABLE_TC_MASK_L0_CACHE__SHIFT 0x2
+#define DB_DEBUG2__DTR_ROUND_ROBIN_ARB_MASK 0x8
+#define DB_DEBUG2__DTR_ROUND_ROBIN_ARB__SHIFT 0x3
+#define DB_DEBUG2__DTR_PREZ_STALLS_FOR_ETF_ROOM_MASK 0x10
+#define DB_DEBUG2__DTR_PREZ_STALLS_FOR_ETF_ROOM__SHIFT 0x4
+#define DB_DEBUG2__DISABLE_PREZL_LPF_STALL_MASK 0x20
+#define DB_DEBUG2__DISABLE_PREZL_LPF_STALL__SHIFT 0x5
+#define DB_DEBUG2__ENABLE_PREZL_CB_STALL_MASK 0x40
+#define DB_DEBUG2__ENABLE_PREZL_CB_STALL__SHIFT 0x6
+#define DB_DEBUG2__DISABLE_PREZL_LPF_STALL_REZ_MASK 0x80
+#define DB_DEBUG2__DISABLE_PREZL_LPF_STALL_REZ__SHIFT 0x7
+#define DB_DEBUG2__DISABLE_PREZL_CB_STALL_REZ_MASK 0x100
+#define DB_DEBUG2__DISABLE_PREZL_CB_STALL_REZ__SHIFT 0x8
+#define DB_DEBUG2__CLK_OFF_DELAY_MASK 0x3e00
+#define DB_DEBUG2__CLK_OFF_DELAY__SHIFT 0x9
+#define DB_DEBUG2__DISABLE_TILE_COVERED_FOR_PS_ITER_MASK 0x4000
+#define DB_DEBUG2__DISABLE_TILE_COVERED_FOR_PS_ITER__SHIFT 0xe
+#define DB_DEBUG2__ENABLE_SUBTILE_GROUPING_MASK 0x8000
+#define DB_DEBUG2__ENABLE_SUBTILE_GROUPING__SHIFT 0xf
+#define DB_DEBUG2__DISABLE_HTILE_PAIRED_PIPES_MASK 0x10000
+#define DB_DEBUG2__DISABLE_HTILE_PAIRED_PIPES__SHIFT 0x10
+#define DB_DEBUG2__DISABLE_NULL_EOT_FORWARDING_MASK 0x20000
+#define DB_DEBUG2__DISABLE_NULL_EOT_FORWARDING__SHIFT 0x11
+#define DB_DEBUG2__DISABLE_DTT_DATA_FORWARDING_MASK 0x40000
+#define DB_DEBUG2__DISABLE_DTT_DATA_FORWARDING__SHIFT 0x12
+#define DB_DEBUG2__DISABLE_QUAD_COHERENCY_STALL_MASK 0x80000
+#define DB_DEBUG2__DISABLE_QUAD_COHERENCY_STALL__SHIFT 0x13
+#define DB_DEBUG2__ENABLE_PREZ_OF_REZ_SUMM_MASK 0x10000000
+#define DB_DEBUG2__ENABLE_PREZ_OF_REZ_SUMM__SHIFT 0x1c
+#define DB_DEBUG2__DISABLE_PREZL_VIEWPORT_STALL_MASK 0x20000000
+#define DB_DEBUG2__DISABLE_PREZL_VIEWPORT_STALL__SHIFT 0x1d
+#define DB_DEBUG2__DISABLE_SINGLE_STENCIL_QUAD_SUMM_MASK 0x40000000
+#define DB_DEBUG2__DISABLE_SINGLE_STENCIL_QUAD_SUMM__SHIFT 0x1e
+#define DB_DEBUG2__DISABLE_WRITE_STALL_ON_RDWR_CONFLICT_MASK 0x80000000
+#define DB_DEBUG2__DISABLE_WRITE_STALL_ON_RDWR_CONFLICT__SHIFT 0x1f
+#define DB_DEBUG3__FORCE_DB_IS_GOOD_MASK 0x4
+#define DB_DEBUG3__FORCE_DB_IS_GOOD__SHIFT 0x2
+#define DB_DEBUG3__DISABLE_TL_SSO_NULL_SUPPRESSION_MASK 0x8
+#define DB_DEBUG3__DISABLE_TL_SSO_NULL_SUPPRESSION__SHIFT 0x3
+#define DB_DEBUG3__DISABLE_HIZ_ON_VPORT_CLAMP_MASK 0x10
+#define DB_DEBUG3__DISABLE_HIZ_ON_VPORT_CLAMP__SHIFT 0x4
+#define DB_DEBUG3__EQAA_INTERPOLATE_COMP_Z_MASK 0x20
+#define DB_DEBUG3__EQAA_INTERPOLATE_COMP_Z__SHIFT 0x5
+#define DB_DEBUG3__EQAA_INTERPOLATE_SRC_Z_MASK 0x40
+#define DB_DEBUG3__EQAA_INTERPOLATE_SRC_Z__SHIFT 0x6
+#define DB_DEBUG3__DISABLE_TCP_CAM_BYPASS_MASK 0x80
+#define DB_DEBUG3__DISABLE_TCP_CAM_BYPASS__SHIFT 0x7
+#define DB_DEBUG3__DISABLE_ZCMP_DIRTY_SUPPRESSION_MASK 0x100
+#define DB_DEBUG3__DISABLE_ZCMP_DIRTY_SUPPRESSION__SHIFT 0x8
+#define DB_DEBUG3__DISABLE_REDUNDANT_PLANE_FLUSHES_OPT_MASK 0x200
+#define DB_DEBUG3__DISABLE_REDUNDANT_PLANE_FLUSHES_OPT__SHIFT 0x9
+#define DB_DEBUG3__DISABLE_RECOMP_TO_1ZPLANE_WITHOUT_FASTOP_MASK 0x400
+#define DB_DEBUG3__DISABLE_RECOMP_TO_1ZPLANE_WITHOUT_FASTOP__SHIFT 0xa
+#define DB_DEBUG3__ENABLE_INCOHERENT_EQAA_READS_MASK 0x800
+#define DB_DEBUG3__ENABLE_INCOHERENT_EQAA_READS__SHIFT 0xb
+#define DB_DEBUG3__DISABLE_OP_Z_DATA_FORWARDING_MASK 0x1000
+#define DB_DEBUG3__DISABLE_OP_Z_DATA_FORWARDING__SHIFT 0xc
+#define DB_DEBUG3__DISABLE_OP_DF_BYPASS_MASK 0x2000
+#define DB_DEBUG3__DISABLE_OP_DF_BYPASS__SHIFT 0xd
+#define DB_DEBUG3__DISABLE_OP_DF_WRITE_COMBINE_MASK 0x4000
+#define DB_DEBUG3__DISABLE_OP_DF_WRITE_COMBINE__SHIFT 0xe
+#define DB_DEBUG3__DISABLE_OP_DF_DIRECT_FEEDBACK_MASK 0x8000
+#define DB_DEBUG3__DISABLE_OP_DF_DIRECT_FEEDBACK__SHIFT 0xf
+#define DB_DEBUG3__ALLOW_RF2P_RW_COLLISION_MASK 0x10000
+#define DB_DEBUG3__ALLOW_RF2P_RW_COLLISION__SHIFT 0x10
+#define DB_DEBUG3__SLOW_PREZ_TO_A2M_OMASK_RATE_MASK 0x20000
+#define DB_DEBUG3__SLOW_PREZ_TO_A2M_OMASK_RATE__SHIFT 0x11
+#define DB_DEBUG3__DISABLE_OP_S_DATA_FORWARDING_MASK 0x40000
+#define DB_DEBUG3__DISABLE_OP_S_DATA_FORWARDING__SHIFT 0x12
+#define DB_DEBUG3__DISABLE_TC_UPDATE_WRITE_COMBINE_MASK 0x80000
+#define DB_DEBUG3__DISABLE_TC_UPDATE_WRITE_COMBINE__SHIFT 0x13
+#define DB_DEBUG3__DISABLE_HZ_TC_WRITE_COMBINE_MASK 0x100000
+#define DB_DEBUG3__DISABLE_HZ_TC_WRITE_COMBINE__SHIFT 0x14
+#define DB_DEBUG3__ENABLE_RECOMP_ZDIRTY_SUPPRESSION_OPT_MASK 0x200000
+#define DB_DEBUG3__ENABLE_RECOMP_ZDIRTY_SUPPRESSION_OPT__SHIFT 0x15
+#define DB_DEBUG3__ENABLE_TC_MA_ROUND_ROBIN_ARB_MASK 0x400000
+#define DB_DEBUG3__ENABLE_TC_MA_ROUND_ROBIN_ARB__SHIFT 0x16
+#define DB_DEBUG3__DISABLE_RAM_READ_SUPPRESION_ON_FWD_MASK 0x800000
+#define DB_DEBUG3__DISABLE_RAM_READ_SUPPRESION_ON_FWD__SHIFT 0x17
+#define DB_DEBUG3__DISABLE_EQAA_A2M_PERF_OPT_MASK 0x1000000
+#define DB_DEBUG3__DISABLE_EQAA_A2M_PERF_OPT__SHIFT 0x18
+#define DB_DEBUG3__DISABLE_DI_DT_STALL_MASK 0x2000000
+#define DB_DEBUG3__DISABLE_DI_DT_STALL__SHIFT 0x19
+#define DB_DEBUG3__ENABLE_DB_PROCESS_RESET_MASK 0x4000000
+#define DB_DEBUG3__ENABLE_DB_PROCESS_RESET__SHIFT 0x1a
+#define DB_DEBUG3__DISABLE_OVERRASTERIZATION_FIX_MASK 0x8000000
+#define DB_DEBUG3__DISABLE_OVERRASTERIZATION_FIX__SHIFT 0x1b
+#define DB_DEBUG3__DONT_INSERT_CONTEXT_SUSPEND_MASK 0x10000000
+#define DB_DEBUG3__DONT_INSERT_CONTEXT_SUSPEND__SHIFT 0x1c
+#define DB_DEBUG3__DONT_DELETE_CONTEXT_SUSPEND_MASK 0x20000000
+#define DB_DEBUG3__DONT_DELETE_CONTEXT_SUSPEND__SHIFT 0x1d
+#define DB_DEBUG3__DISABLE_4XAA_2P_DELAYED_WRITE_MASK 0x40000000
+#define DB_DEBUG3__DISABLE_4XAA_2P_DELAYED_WRITE__SHIFT 0x1e
+#define DB_DEBUG3__DISABLE_4XAA_2P_INTERLEAVED_PMASK_MASK 0x80000000
+#define DB_DEBUG3__DISABLE_4XAA_2P_INTERLEAVED_PMASK__SHIFT 0x1f
+#define DB_DEBUG4__DISABLE_QC_Z_MASK_SUMMATION_MASK 0x1
+#define DB_DEBUG4__DISABLE_QC_Z_MASK_SUMMATION__SHIFT 0x0
+#define DB_DEBUG4__DISABLE_QC_STENCIL_MASK_SUMMATION_MASK 0x2
+#define DB_DEBUG4__DISABLE_QC_STENCIL_MASK_SUMMATION__SHIFT 0x1
+#define DB_DEBUG4__DISABLE_RESUMM_TO_SINGLE_STENCIL_MASK 0x4
+#define DB_DEBUG4__DISABLE_RESUMM_TO_SINGLE_STENCIL__SHIFT 0x2
+#define DB_DEBUG4__DISABLE_PREZ_POSTZ_DTILE_CONFLICT_STALL_MASK 0x8
+#define DB_DEBUG4__DISABLE_PREZ_POSTZ_DTILE_CONFLICT_STALL__SHIFT 0x3
+#define DB_DEBUG4__DISABLE_4XAA_2P_ZD_HOLDOFF_MASK 0x10
+#define DB_DEBUG4__DISABLE_4XAA_2P_ZD_HOLDOFF__SHIFT 0x4
+#define DB_DEBUG4__ENABLE_A2M_DQUAD_OPTIMIZATION_MASK 0x20
+#define DB_DEBUG4__ENABLE_A2M_DQUAD_OPTIMIZATION__SHIFT 0x5
+#define DB_DEBUG4__ENABLE_DBCB_SLOW_FORMAT_COLLAPSE_MASK 0x40
+#define DB_DEBUG4__ENABLE_DBCB_SLOW_FORMAT_COLLAPSE__SHIFT 0x6
+#define DB_DEBUG4__DB_EXTRA_DEBUG4_MASK 0xffffff80
+#define DB_DEBUG4__DB_EXTRA_DEBUG4__SHIFT 0x7
+#define DB_CREDIT_LIMIT__DB_SC_TILE_CREDITS_MASK 0x1f
+#define DB_CREDIT_LIMIT__DB_SC_TILE_CREDITS__SHIFT 0x0
+#define DB_CREDIT_LIMIT__DB_SC_QUAD_CREDITS_MASK 0x3e0
+#define DB_CREDIT_LIMIT__DB_SC_QUAD_CREDITS__SHIFT 0x5
+#define DB_CREDIT_LIMIT__DB_CB_LQUAD_CREDITS_MASK 0x1c00
+#define DB_CREDIT_LIMIT__DB_CB_LQUAD_CREDITS__SHIFT 0xa
+#define DB_CREDIT_LIMIT__DB_CB_TILE_CREDITS_MASK 0x7f000000
+#define DB_CREDIT_LIMIT__DB_CB_TILE_CREDITS__SHIFT 0x18
+#define DB_WATERMARKS__DEPTH_FREE_MASK 0x1f
+#define DB_WATERMARKS__DEPTH_FREE__SHIFT 0x0
+#define DB_WATERMARKS__DEPTH_FLUSH_MASK 0x7e0
+#define DB_WATERMARKS__DEPTH_FLUSH__SHIFT 0x5
+#define DB_WATERMARKS__FORCE_SUMMARIZE_MASK 0x7800
+#define DB_WATERMARKS__FORCE_SUMMARIZE__SHIFT 0xb
+#define DB_WATERMARKS__DEPTH_PENDING_FREE_MASK 0xf8000
+#define DB_WATERMARKS__DEPTH_PENDING_FREE__SHIFT 0xf
+#define DB_WATERMARKS__DEPTH_CACHELINE_FREE_MASK 0x7f00000
+#define DB_WATERMARKS__DEPTH_CACHELINE_FREE__SHIFT 0x14
+#define DB_WATERMARKS__EARLY_Z_PANIC_DISABLE_MASK 0x8000000
+#define DB_WATERMARKS__EARLY_Z_PANIC_DISABLE__SHIFT 0x1b
+#define DB_WATERMARKS__LATE_Z_PANIC_DISABLE_MASK 0x10000000
+#define DB_WATERMARKS__LATE_Z_PANIC_DISABLE__SHIFT 0x1c
+#define DB_WATERMARKS__RE_Z_PANIC_DISABLE_MASK 0x20000000
+#define DB_WATERMARKS__RE_Z_PANIC_DISABLE__SHIFT 0x1d
+#define DB_WATERMARKS__AUTO_FLUSH_HTILE_MASK 0x40000000
+#define DB_WATERMARKS__AUTO_FLUSH_HTILE__SHIFT 0x1e
+#define DB_WATERMARKS__AUTO_FLUSH_QUAD_MASK 0x80000000
+#define DB_WATERMARKS__AUTO_FLUSH_QUAD__SHIFT 0x1f
+#define DB_SUBTILE_CONTROL__MSAA1_X_MASK 0x3
+#define DB_SUBTILE_CONTROL__MSAA1_X__SHIFT 0x0
+#define DB_SUBTILE_CONTROL__MSAA1_Y_MASK 0xc
+#define DB_SUBTILE_CONTROL__MSAA1_Y__SHIFT 0x2
+#define DB_SUBTILE_CONTROL__MSAA2_X_MASK 0x30
+#define DB_SUBTILE_CONTROL__MSAA2_X__SHIFT 0x4
+#define DB_SUBTILE_CONTROL__MSAA2_Y_MASK 0xc0
+#define DB_SUBTILE_CONTROL__MSAA2_Y__SHIFT 0x6
+#define DB_SUBTILE_CONTROL__MSAA4_X_MASK 0x300
+#define DB_SUBTILE_CONTROL__MSAA4_X__SHIFT 0x8
+#define DB_SUBTILE_CONTROL__MSAA4_Y_MASK 0xc00
+#define DB_SUBTILE_CONTROL__MSAA4_Y__SHIFT 0xa
+#define DB_SUBTILE_CONTROL__MSAA8_X_MASK 0x3000
+#define DB_SUBTILE_CONTROL__MSAA8_X__SHIFT 0xc
+#define DB_SUBTILE_CONTROL__MSAA8_Y_MASK 0xc000
+#define DB_SUBTILE_CONTROL__MSAA8_Y__SHIFT 0xe
+#define DB_SUBTILE_CONTROL__MSAA16_X_MASK 0x30000
+#define DB_SUBTILE_CONTROL__MSAA16_X__SHIFT 0x10
+#define DB_SUBTILE_CONTROL__MSAA16_Y_MASK 0xc0000
+#define DB_SUBTILE_CONTROL__MSAA16_Y__SHIFT 0x12
+#define DB_FREE_CACHELINES__FREE_DTILE_DEPTH_MASK 0x7f
+#define DB_FREE_CACHELINES__FREE_DTILE_DEPTH__SHIFT 0x0
+#define DB_FREE_CACHELINES__FREE_PLANE_DEPTH_MASK 0x3f80
+#define DB_FREE_CACHELINES__FREE_PLANE_DEPTH__SHIFT 0x7
+#define DB_FREE_CACHELINES__FREE_Z_DEPTH_MASK 0x1fc000
+#define DB_FREE_CACHELINES__FREE_Z_DEPTH__SHIFT 0xe
+#define DB_FREE_CACHELINES__FREE_HTILE_DEPTH_MASK 0x1e00000
+#define DB_FREE_CACHELINES__FREE_HTILE_DEPTH__SHIFT 0x15
+#define DB_FREE_CACHELINES__QUAD_READ_REQS_MASK 0xfe000000
+#define DB_FREE_CACHELINES__QUAD_READ_REQS__SHIFT 0x19
+#define DB_FIFO_DEPTH1__MI_RDREQ_FIFO_DEPTH_MASK 0x1f
+#define DB_FIFO_DEPTH1__MI_RDREQ_FIFO_DEPTH__SHIFT 0x0
+#define DB_FIFO_DEPTH1__MI_WRREQ_FIFO_DEPTH_MASK 0x3e0
+#define DB_FIFO_DEPTH1__MI_WRREQ_FIFO_DEPTH__SHIFT 0x5
+#define DB_FIFO_DEPTH1__MCC_DEPTH_MASK 0xfc00
+#define DB_FIFO_DEPTH1__MCC_DEPTH__SHIFT 0xa
+#define DB_FIFO_DEPTH1__QC_DEPTH_MASK 0x1f0000
+#define DB_FIFO_DEPTH1__QC_DEPTH__SHIFT 0x10
+#define DB_FIFO_DEPTH1__LTILE_PROBE_FIFO_DEPTH_MASK 0x1fe00000
+#define DB_FIFO_DEPTH1__LTILE_PROBE_FIFO_DEPTH__SHIFT 0x15
+#define DB_FIFO_DEPTH2__EQUAD_FIFO_DEPTH_MASK 0xff
+#define DB_FIFO_DEPTH2__EQUAD_FIFO_DEPTH__SHIFT 0x0
+#define DB_FIFO_DEPTH2__ETILE_OP_FIFO_DEPTH_MASK 0x7f00
+#define DB_FIFO_DEPTH2__ETILE_OP_FIFO_DEPTH__SHIFT 0x8
+#define DB_FIFO_DEPTH2__LQUAD_FIFO_DEPTH_MASK 0x1ff8000
+#define DB_FIFO_DEPTH2__LQUAD_FIFO_DEPTH__SHIFT 0xf
+#define DB_FIFO_DEPTH2__LTILE_OP_FIFO_DEPTH_MASK 0xfe000000
+#define DB_FIFO_DEPTH2__LTILE_OP_FIFO_DEPTH__SHIFT 0x19
+#define DB_CGTT_CLK_CTRL_0__ON_DELAY_MASK 0xf
+#define DB_CGTT_CLK_CTRL_0__ON_DELAY__SHIFT 0x0
+#define DB_CGTT_CLK_CTRL_0__OFF_HYSTERESIS_MASK 0xff0
+#define DB_CGTT_CLK_CTRL_0__OFF_HYSTERESIS__SHIFT 0x4
+#define DB_CGTT_CLK_CTRL_0__RESERVED_MASK 0xfff000
+#define DB_CGTT_CLK_CTRL_0__RESERVED__SHIFT 0xc
+#define DB_CGTT_CLK_CTRL_0__SOFT_OVERRIDE7_MASK 0x1000000
+#define DB_CGTT_CLK_CTRL_0__SOFT_OVERRIDE7__SHIFT 0x18
+#define DB_CGTT_CLK_CTRL_0__SOFT_OVERRIDE6_MASK 0x2000000
+#define DB_CGTT_CLK_CTRL_0__SOFT_OVERRIDE6__SHIFT 0x19
+#define DB_CGTT_CLK_CTRL_0__SOFT_OVERRIDE5_MASK 0x4000000
+#define DB_CGTT_CLK_CTRL_0__SOFT_OVERRIDE5__SHIFT 0x1a
+#define DB_CGTT_CLK_CTRL_0__SOFT_OVERRIDE4_MASK 0x8000000
+#define DB_CGTT_CLK_CTRL_0__SOFT_OVERRIDE4__SHIFT 0x1b
+#define DB_CGTT_CLK_CTRL_0__SOFT_OVERRIDE3_MASK 0x10000000
+#define DB_CGTT_CLK_CTRL_0__SOFT_OVERRIDE3__SHIFT 0x1c
+#define DB_CGTT_CLK_CTRL_0__SOFT_OVERRIDE2_MASK 0x20000000
+#define DB_CGTT_CLK_CTRL_0__SOFT_OVERRIDE2__SHIFT 0x1d
+#define DB_CGTT_CLK_CTRL_0__SOFT_OVERRIDE1_MASK 0x40000000
+#define DB_CGTT_CLK_CTRL_0__SOFT_OVERRIDE1__SHIFT 0x1e
+#define DB_CGTT_CLK_CTRL_0__SOFT_OVERRIDE0_MASK 0x80000000
+#define DB_CGTT_CLK_CTRL_0__SOFT_OVERRIDE0__SHIFT 0x1f
+#define DB_ZPASS_COUNT_LOW__COUNT_LOW_MASK 0xffffffff
+#define DB_ZPASS_COUNT_LOW__COUNT_LOW__SHIFT 0x0
+#define DB_ZPASS_COUNT_HI__COUNT_HI_MASK 0x7fffffff
+#define DB_ZPASS_COUNT_HI__COUNT_HI__SHIFT 0x0
+#define DB_RING_CONTROL__COUNTER_CONTROL_MASK 0x3
+#define DB_RING_CONTROL__COUNTER_CONTROL__SHIFT 0x0
+#define DB_READ_DEBUG_0__BUSY_DATA0_MASK 0xffffffff
+#define DB_READ_DEBUG_0__BUSY_DATA0__SHIFT 0x0
+#define DB_READ_DEBUG_1__BUSY_DATA1_MASK 0xffffffff
+#define DB_READ_DEBUG_1__BUSY_DATA1__SHIFT 0x0
+#define DB_READ_DEBUG_2__BUSY_DATA2_MASK 0xffffffff
+#define DB_READ_DEBUG_2__BUSY_DATA2__SHIFT 0x0
+#define DB_READ_DEBUG_3__DEBUG_DATA_MASK 0xffffffff
+#define DB_READ_DEBUG_3__DEBUG_DATA__SHIFT 0x0
+#define DB_READ_DEBUG_4__DEBUG_DATA_MASK 0xffffffff
+#define DB_READ_DEBUG_4__DEBUG_DATA__SHIFT 0x0
+#define DB_READ_DEBUG_5__DEBUG_DATA_MASK 0xffffffff
+#define DB_READ_DEBUG_5__DEBUG_DATA__SHIFT 0x0
+#define DB_READ_DEBUG_6__DEBUG_DATA_MASK 0xffffffff
+#define DB_READ_DEBUG_6__DEBUG_DATA__SHIFT 0x0
+#define DB_READ_DEBUG_7__DEBUG_DATA_MASK 0xffffffff
+#define DB_READ_DEBUG_7__DEBUG_DATA__SHIFT 0x0
+#define DB_READ_DEBUG_8__DEBUG_DATA_MASK 0xffffffff
+#define DB_READ_DEBUG_8__DEBUG_DATA__SHIFT 0x0
+#define DB_READ_DEBUG_9__DEBUG_DATA_MASK 0xffffffff
+#define DB_READ_DEBUG_9__DEBUG_DATA__SHIFT 0x0
+#define DB_READ_DEBUG_A__DEBUG_DATA_MASK 0xffffffff
+#define DB_READ_DEBUG_A__DEBUG_DATA__SHIFT 0x0
+#define DB_READ_DEBUG_B__DEBUG_DATA_MASK 0xffffffff
+#define DB_READ_DEBUG_B__DEBUG_DATA__SHIFT 0x0
+#define DB_READ_DEBUG_C__DEBUG_DATA_MASK 0xffffffff
+#define DB_READ_DEBUG_C__DEBUG_DATA__SHIFT 0x0
+#define DB_READ_DEBUG_D__DEBUG_DATA_MASK 0xffffffff
+#define DB_READ_DEBUG_D__DEBUG_DATA__SHIFT 0x0
+#define DB_READ_DEBUG_E__DEBUG_DATA_MASK 0xffffffff
+#define DB_READ_DEBUG_E__DEBUG_DATA__SHIFT 0x0
+#define DB_READ_DEBUG_F__DEBUG_DATA_MASK 0xffffffff
+#define DB_READ_DEBUG_F__DEBUG_DATA__SHIFT 0x0
+#define DB_OCCLUSION_COUNT0_LOW__COUNT_LOW_MASK 0xffffffff
+#define DB_OCCLUSION_COUNT0_LOW__COUNT_LOW__SHIFT 0x0
+#define DB_OCCLUSION_COUNT0_HI__COUNT_HI_MASK 0x7fffffff
+#define DB_OCCLUSION_COUNT0_HI__COUNT_HI__SHIFT 0x0
+#define DB_OCCLUSION_COUNT1_LOW__COUNT_LOW_MASK 0xffffffff
+#define DB_OCCLUSION_COUNT1_LOW__COUNT_LOW__SHIFT 0x0
+#define DB_OCCLUSION_COUNT1_HI__COUNT_HI_MASK 0x7fffffff
+#define DB_OCCLUSION_COUNT1_HI__COUNT_HI__SHIFT 0x0
+#define DB_OCCLUSION_COUNT2_LOW__COUNT_LOW_MASK 0xffffffff
+#define DB_OCCLUSION_COUNT2_LOW__COUNT_LOW__SHIFT 0x0
+#define DB_OCCLUSION_COUNT2_HI__COUNT_HI_MASK 0x7fffffff
+#define DB_OCCLUSION_COUNT2_HI__COUNT_HI__SHIFT 0x0
+#define DB_OCCLUSION_COUNT3_LOW__COUNT_LOW_MASK 0xffffffff
+#define DB_OCCLUSION_COUNT3_LOW__COUNT_LOW__SHIFT 0x0
+#define DB_OCCLUSION_COUNT3_HI__COUNT_HI_MASK 0x7fffffff
+#define DB_OCCLUSION_COUNT3_HI__COUNT_HI__SHIFT 0x0
+#define CC_RB_REDUNDANCY__FAILED_RB0_MASK 0xf00
+#define CC_RB_REDUNDANCY__FAILED_RB0__SHIFT 0x8
+#define CC_RB_REDUNDANCY__EN_REDUNDANCY0_MASK 0x1000
+#define CC_RB_REDUNDANCY__EN_REDUNDANCY0__SHIFT 0xc
+#define CC_RB_REDUNDANCY__FAILED_RB1_MASK 0xf0000
+#define CC_RB_REDUNDANCY__FAILED_RB1__SHIFT 0x10
+#define CC_RB_REDUNDANCY__EN_REDUNDANCY1_MASK 0x100000
+#define CC_RB_REDUNDANCY__EN_REDUNDANCY1__SHIFT 0x14
+#define CC_RB_BACKEND_DISABLE__BACKEND_DISABLE_MASK 0xff0000
+#define CC_RB_BACKEND_DISABLE__BACKEND_DISABLE__SHIFT 0x10
+#define GC_USER_RB_REDUNDANCY__FAILED_RB0_MASK 0xf00
+#define GC_USER_RB_REDUNDANCY__FAILED_RB0__SHIFT 0x8
+#define GC_USER_RB_REDUNDANCY__EN_REDUNDANCY0_MASK 0x1000
+#define GC_USER_RB_REDUNDANCY__EN_REDUNDANCY0__SHIFT 0xc
+#define GC_USER_RB_REDUNDANCY__FAILED_RB1_MASK 0xf0000
+#define GC_USER_RB_REDUNDANCY__FAILED_RB1__SHIFT 0x10
+#define GC_USER_RB_REDUNDANCY__EN_REDUNDANCY1_MASK 0x100000
+#define GC_USER_RB_REDUNDANCY__EN_REDUNDANCY1__SHIFT 0x14
+#define GC_USER_RB_BACKEND_DISABLE__BACKEND_DISABLE_MASK 0xff0000
+#define GC_USER_RB_BACKEND_DISABLE__BACKEND_DISABLE__SHIFT 0x10
+#define GB_ADDR_CONFIG__NUM_PIPES_MASK 0x7
+#define GB_ADDR_CONFIG__NUM_PIPES__SHIFT 0x0
+#define GB_ADDR_CONFIG__PIPE_INTERLEAVE_SIZE_MASK 0x70
+#define GB_ADDR_CONFIG__PIPE_INTERLEAVE_SIZE__SHIFT 0x4
+#define GB_ADDR_CONFIG__BANK_INTERLEAVE_SIZE_MASK 0x700
+#define GB_ADDR_CONFIG__BANK_INTERLEAVE_SIZE__SHIFT 0x8
+#define GB_ADDR_CONFIG__NUM_SHADER_ENGINES_MASK 0x3000
+#define GB_ADDR_CONFIG__NUM_SHADER_ENGINES__SHIFT 0xc
+#define GB_ADDR_CONFIG__SHADER_ENGINE_TILE_SIZE_MASK 0x70000
+#define GB_ADDR_CONFIG__SHADER_ENGINE_TILE_SIZE__SHIFT 0x10
+#define GB_ADDR_CONFIG__NUM_GPUS_MASK 0x700000
+#define GB_ADDR_CONFIG__NUM_GPUS__SHIFT 0x14
+#define GB_ADDR_CONFIG__MULTI_GPU_TILE_SIZE_MASK 0x3000000
+#define GB_ADDR_CONFIG__MULTI_GPU_TILE_SIZE__SHIFT 0x18
+#define GB_ADDR_CONFIG__ROW_SIZE_MASK 0x30000000
+#define GB_ADDR_CONFIG__ROW_SIZE__SHIFT 0x1c
+#define GB_ADDR_CONFIG__NUM_LOWER_PIPES_MASK 0x40000000
+#define GB_ADDR_CONFIG__NUM_LOWER_PIPES__SHIFT 0x1e
+#define GB_BACKEND_MAP__BACKEND_MAP_MASK 0xffffffff
+#define GB_BACKEND_MAP__BACKEND_MAP__SHIFT 0x0
+#define GB_GPU_ID__GPU_ID_MASK 0xf
+#define GB_GPU_ID__GPU_ID__SHIFT 0x0
+#define CC_RB_DAISY_CHAIN__RB_0_MASK 0xf
+#define CC_RB_DAISY_CHAIN__RB_0__SHIFT 0x0
+#define CC_RB_DAISY_CHAIN__RB_1_MASK 0xf0
+#define CC_RB_DAISY_CHAIN__RB_1__SHIFT 0x4
+#define CC_RB_DAISY_CHAIN__RB_2_MASK 0xf00
+#define CC_RB_DAISY_CHAIN__RB_2__SHIFT 0x8
+#define CC_RB_DAISY_CHAIN__RB_3_MASK 0xf000
+#define CC_RB_DAISY_CHAIN__RB_3__SHIFT 0xc
+#define CC_RB_DAISY_CHAIN__RB_4_MASK 0xf0000
+#define CC_RB_DAISY_CHAIN__RB_4__SHIFT 0x10
+#define CC_RB_DAISY_CHAIN__RB_5_MASK 0xf00000
+#define CC_RB_DAISY_CHAIN__RB_5__SHIFT 0x14
+#define CC_RB_DAISY_CHAIN__RB_6_MASK 0xf000000
+#define CC_RB_DAISY_CHAIN__RB_6__SHIFT 0x18
+#define CC_RB_DAISY_CHAIN__RB_7_MASK 0xf0000000
+#define CC_RB_DAISY_CHAIN__RB_7__SHIFT 0x1c
+#define GB_TILE_MODE0__ARRAY_MODE_MASK 0x3c
+#define GB_TILE_MODE0__ARRAY_MODE__SHIFT 0x2
+#define GB_TILE_MODE0__PIPE_CONFIG_MASK 0x7c0
+#define GB_TILE_MODE0__PIPE_CONFIG__SHIFT 0x6
+#define GB_TILE_MODE0__TILE_SPLIT_MASK 0x3800
+#define GB_TILE_MODE0__TILE_SPLIT__SHIFT 0xb
+#define GB_TILE_MODE0__MICRO_TILE_MODE_NEW_MASK 0x1c00000
+#define GB_TILE_MODE0__MICRO_TILE_MODE_NEW__SHIFT 0x16
+#define GB_TILE_MODE0__SAMPLE_SPLIT_MASK 0x6000000
+#define GB_TILE_MODE0__SAMPLE_SPLIT__SHIFT 0x19
+#define GB_TILE_MODE1__ARRAY_MODE_MASK 0x3c
+#define GB_TILE_MODE1__ARRAY_MODE__SHIFT 0x2
+#define GB_TILE_MODE1__PIPE_CONFIG_MASK 0x7c0
+#define GB_TILE_MODE1__PIPE_CONFIG__SHIFT 0x6
+#define GB_TILE_MODE1__TILE_SPLIT_MASK 0x3800
+#define GB_TILE_MODE1__TILE_SPLIT__SHIFT 0xb
+#define GB_TILE_MODE1__MICRO_TILE_MODE_NEW_MASK 0x1c00000
+#define GB_TILE_MODE1__MICRO_TILE_MODE_NEW__SHIFT 0x16
+#define GB_TILE_MODE1__SAMPLE_SPLIT_MASK 0x6000000
+#define GB_TILE_MODE1__SAMPLE_SPLIT__SHIFT 0x19
+#define GB_TILE_MODE2__ARRAY_MODE_MASK 0x3c
+#define GB_TILE_MODE2__ARRAY_MODE__SHIFT 0x2
+#define GB_TILE_MODE2__PIPE_CONFIG_MASK 0x7c0
+#define GB_TILE_MODE2__PIPE_CONFIG__SHIFT 0x6
+#define GB_TILE_MODE2__TILE_SPLIT_MASK 0x3800
+#define GB_TILE_MODE2__TILE_SPLIT__SHIFT 0xb
+#define GB_TILE_MODE2__MICRO_TILE_MODE_NEW_MASK 0x1c00000
+#define GB_TILE_MODE2__MICRO_TILE_MODE_NEW__SHIFT 0x16
+#define GB_TILE_MODE2__SAMPLE_SPLIT_MASK 0x6000000
+#define GB_TILE_MODE2__SAMPLE_SPLIT__SHIFT 0x19
+#define GB_TILE_MODE3__ARRAY_MODE_MASK 0x3c
+#define GB_TILE_MODE3__ARRAY_MODE__SHIFT 0x2
+#define GB_TILE_MODE3__PIPE_CONFIG_MASK 0x7c0
+#define GB_TILE_MODE3__PIPE_CONFIG__SHIFT 0x6
+#define GB_TILE_MODE3__TILE_SPLIT_MASK 0x3800
+#define GB_TILE_MODE3__TILE_SPLIT__SHIFT 0xb
+#define GB_TILE_MODE3__MICRO_TILE_MODE_NEW_MASK 0x1c00000
+#define GB_TILE_MODE3__MICRO_TILE_MODE_NEW__SHIFT 0x16
+#define GB_TILE_MODE3__SAMPLE_SPLIT_MASK 0x6000000
+#define GB_TILE_MODE3__SAMPLE_SPLIT__SHIFT 0x19
+#define GB_TILE_MODE4__ARRAY_MODE_MASK 0x3c
+#define GB_TILE_MODE4__ARRAY_MODE__SHIFT 0x2
+#define GB_TILE_MODE4__PIPE_CONFIG_MASK 0x7c0
+#define GB_TILE_MODE4__PIPE_CONFIG__SHIFT 0x6
+#define GB_TILE_MODE4__TILE_SPLIT_MASK 0x3800
+#define GB_TILE_MODE4__TILE_SPLIT__SHIFT 0xb
+#define GB_TILE_MODE4__MICRO_TILE_MODE_NEW_MASK 0x1c00000
+#define GB_TILE_MODE4__MICRO_TILE_MODE_NEW__SHIFT 0x16
+#define GB_TILE_MODE4__SAMPLE_SPLIT_MASK 0x6000000
+#define GB_TILE_MODE4__SAMPLE_SPLIT__SHIFT 0x19
+#define GB_TILE_MODE5__ARRAY_MODE_MASK 0x3c
+#define GB_TILE_MODE5__ARRAY_MODE__SHIFT 0x2
+#define GB_TILE_MODE5__PIPE_CONFIG_MASK 0x7c0
+#define GB_TILE_MODE5__PIPE_CONFIG__SHIFT 0x6
+#define GB_TILE_MODE5__TILE_SPLIT_MASK 0x3800
+#define GB_TILE_MODE5__TILE_SPLIT__SHIFT 0xb
+#define GB_TILE_MODE5__MICRO_TILE_MODE_NEW_MASK 0x1c00000
+#define GB_TILE_MODE5__MICRO_TILE_MODE_NEW__SHIFT 0x16
+#define GB_TILE_MODE5__SAMPLE_SPLIT_MASK 0x6000000
+#define GB_TILE_MODE5__SAMPLE_SPLIT__SHIFT 0x19
+#define GB_TILE_MODE6__ARRAY_MODE_MASK 0x3c
+#define GB_TILE_MODE6__ARRAY_MODE__SHIFT 0x2
+#define GB_TILE_MODE6__PIPE_CONFIG_MASK 0x7c0
+#define GB_TILE_MODE6__PIPE_CONFIG__SHIFT 0x6
+#define GB_TILE_MODE6__TILE_SPLIT_MASK 0x3800
+#define GB_TILE_MODE6__TILE_SPLIT__SHIFT 0xb
+#define GB_TILE_MODE6__MICRO_TILE_MODE_NEW_MASK 0x1c00000
+#define GB_TILE_MODE6__MICRO_TILE_MODE_NEW__SHIFT 0x16
+#define GB_TILE_MODE6__SAMPLE_SPLIT_MASK 0x6000000
+#define GB_TILE_MODE6__SAMPLE_SPLIT__SHIFT 0x19
+#define GB_TILE_MODE7__ARRAY_MODE_MASK 0x3c
+#define GB_TILE_MODE7__ARRAY_MODE__SHIFT 0x2
+#define GB_TILE_MODE7__PIPE_CONFIG_MASK 0x7c0
+#define GB_TILE_MODE7__PIPE_CONFIG__SHIFT 0x6
+#define GB_TILE_MODE7__TILE_SPLIT_MASK 0x3800
+#define GB_TILE_MODE7__TILE_SPLIT__SHIFT 0xb
+#define GB_TILE_MODE7__MICRO_TILE_MODE_NEW_MASK 0x1c00000
+#define GB_TILE_MODE7__MICRO_TILE_MODE_NEW__SHIFT 0x16
+#define GB_TILE_MODE7__SAMPLE_SPLIT_MASK 0x6000000
+#define GB_TILE_MODE7__SAMPLE_SPLIT__SHIFT 0x19
+#define GB_TILE_MODE8__ARRAY_MODE_MASK 0x3c
+#define GB_TILE_MODE8__ARRAY_MODE__SHIFT 0x2
+#define GB_TILE_MODE8__PIPE_CONFIG_MASK 0x7c0
+#define GB_TILE_MODE8__PIPE_CONFIG__SHIFT 0x6
+#define GB_TILE_MODE8__TILE_SPLIT_MASK 0x3800
+#define GB_TILE_MODE8__TILE_SPLIT__SHIFT 0xb
+#define GB_TILE_MODE8__MICRO_TILE_MODE_NEW_MASK 0x1c00000
+#define GB_TILE_MODE8__MICRO_TILE_MODE_NEW__SHIFT 0x16
+#define GB_TILE_MODE8__SAMPLE_SPLIT_MASK 0x6000000
+#define GB_TILE_MODE8__SAMPLE_SPLIT__SHIFT 0x19
+#define GB_TILE_MODE9__ARRAY_MODE_MASK 0x3c
+#define GB_TILE_MODE9__ARRAY_MODE__SHIFT 0x2
+#define GB_TILE_MODE9__PIPE_CONFIG_MASK 0x7c0
+#define GB_TILE_MODE9__PIPE_CONFIG__SHIFT 0x6
+#define GB_TILE_MODE9__TILE_SPLIT_MASK 0x3800
+#define GB_TILE_MODE9__TILE_SPLIT__SHIFT 0xb
+#define GB_TILE_MODE9__MICRO_TILE_MODE_NEW_MASK 0x1c00000
+#define GB_TILE_MODE9__MICRO_TILE_MODE_NEW__SHIFT 0x16
+#define GB_TILE_MODE9__SAMPLE_SPLIT_MASK 0x6000000
+#define GB_TILE_MODE9__SAMPLE_SPLIT__SHIFT 0x19
+#define GB_TILE_MODE10__ARRAY_MODE_MASK 0x3c
+#define GB_TILE_MODE10__ARRAY_MODE__SHIFT 0x2
+#define GB_TILE_MODE10__PIPE_CONFIG_MASK 0x7c0
+#define GB_TILE_MODE10__PIPE_CONFIG__SHIFT 0x6
+#define GB_TILE_MODE10__TILE_SPLIT_MASK 0x3800
+#define GB_TILE_MODE10__TILE_SPLIT__SHIFT 0xb
+#define GB_TILE_MODE10__MICRO_TILE_MODE_NEW_MASK 0x1c00000
+#define GB_TILE_MODE10__MICRO_TILE_MODE_NEW__SHIFT 0x16
+#define GB_TILE_MODE10__SAMPLE_SPLIT_MASK 0x6000000
+#define GB_TILE_MODE10__SAMPLE_SPLIT__SHIFT 0x19
+#define GB_TILE_MODE11__ARRAY_MODE_MASK 0x3c
+#define GB_TILE_MODE11__ARRAY_MODE__SHIFT 0x2
+#define GB_TILE_MODE11__PIPE_CONFIG_MASK 0x7c0
+#define GB_TILE_MODE11__PIPE_CONFIG__SHIFT 0x6
+#define GB_TILE_MODE11__TILE_SPLIT_MASK 0x3800
+#define GB_TILE_MODE11__TILE_SPLIT__SHIFT 0xb
+#define GB_TILE_MODE11__MICRO_TILE_MODE_NEW_MASK 0x1c00000
+#define GB_TILE_MODE11__MICRO_TILE_MODE_NEW__SHIFT 0x16
+#define GB_TILE_MODE11__SAMPLE_SPLIT_MASK 0x6000000
+#define GB_TILE_MODE11__SAMPLE_SPLIT__SHIFT 0x19
+#define GB_TILE_MODE12__ARRAY_MODE_MASK 0x3c
+#define GB_TILE_MODE12__ARRAY_MODE__SHIFT 0x2
+#define GB_TILE_MODE12__PIPE_CONFIG_MASK 0x7c0
+#define GB_TILE_MODE12__PIPE_CONFIG__SHIFT 0x6
+#define GB_TILE_MODE12__TILE_SPLIT_MASK 0x3800
+#define GB_TILE_MODE12__TILE_SPLIT__SHIFT 0xb
+#define GB_TILE_MODE12__MICRO_TILE_MODE_NEW_MASK 0x1c00000
+#define GB_TILE_MODE12__MICRO_TILE_MODE_NEW__SHIFT 0x16
+#define GB_TILE_MODE12__SAMPLE_SPLIT_MASK 0x6000000
+#define GB_TILE_MODE12__SAMPLE_SPLIT__SHIFT 0x19
+#define GB_TILE_MODE13__ARRAY_MODE_MASK 0x3c
+#define GB_TILE_MODE13__ARRAY_MODE__SHIFT 0x2
+#define GB_TILE_MODE13__PIPE_CONFIG_MASK 0x7c0
+#define GB_TILE_MODE13__PIPE_CONFIG__SHIFT 0x6
+#define GB_TILE_MODE13__TILE_SPLIT_MASK 0x3800
+#define GB_TILE_MODE13__TILE_SPLIT__SHIFT 0xb
+#define GB_TILE_MODE13__MICRO_TILE_MODE_NEW_MASK 0x1c00000
+#define GB_TILE_MODE13__MICRO_TILE_MODE_NEW__SHIFT 0x16
+#define GB_TILE_MODE13__SAMPLE_SPLIT_MASK 0x6000000
+#define GB_TILE_MODE13__SAMPLE_SPLIT__SHIFT 0x19
+#define GB_TILE_MODE14__ARRAY_MODE_MASK 0x3c
+#define GB_TILE_MODE14__ARRAY_MODE__SHIFT 0x2
+#define GB_TILE_MODE14__PIPE_CONFIG_MASK 0x7c0
+#define GB_TILE_MODE14__PIPE_CONFIG__SHIFT 0x6
+#define GB_TILE_MODE14__TILE_SPLIT_MASK 0x3800
+#define GB_TILE_MODE14__TILE_SPLIT__SHIFT 0xb
+#define GB_TILE_MODE14__MICRO_TILE_MODE_NEW_MASK 0x1c00000
+#define GB_TILE_MODE14__MICRO_TILE_MODE_NEW__SHIFT 0x16
+#define GB_TILE_MODE14__SAMPLE_SPLIT_MASK 0x6000000
+#define GB_TILE_MODE14__SAMPLE_SPLIT__SHIFT 0x19
+#define GB_TILE_MODE15__ARRAY_MODE_MASK 0x3c
+#define GB_TILE_MODE15__ARRAY_MODE__SHIFT 0x2
+#define GB_TILE_MODE15__PIPE_CONFIG_MASK 0x7c0
+#define GB_TILE_MODE15__PIPE_CONFIG__SHIFT 0x6
+#define GB_TILE_MODE15__TILE_SPLIT_MASK 0x3800
+#define GB_TILE_MODE15__TILE_SPLIT__SHIFT 0xb
+#define GB_TILE_MODE15__MICRO_TILE_MODE_NEW_MASK 0x1c00000
+#define GB_TILE_MODE15__MICRO_TILE_MODE_NEW__SHIFT 0x16
+#define GB_TILE_MODE15__SAMPLE_SPLIT_MASK 0x6000000
+#define GB_TILE_MODE15__SAMPLE_SPLIT__SHIFT 0x19
+#define GB_TILE_MODE16__ARRAY_MODE_MASK 0x3c
+#define GB_TILE_MODE16__ARRAY_MODE__SHIFT 0x2
+#define GB_TILE_MODE16__PIPE_CONFIG_MASK 0x7c0
+#define GB_TILE_MODE16__PIPE_CONFIG__SHIFT 0x6
+#define GB_TILE_MODE16__TILE_SPLIT_MASK 0x3800
+#define GB_TILE_MODE16__TILE_SPLIT__SHIFT 0xb
+#define GB_TILE_MODE16__MICRO_TILE_MODE_NEW_MASK 0x1c00000
+#define GB_TILE_MODE16__MICRO_TILE_MODE_NEW__SHIFT 0x16
+#define GB_TILE_MODE16__SAMPLE_SPLIT_MASK 0x6000000
+#define GB_TILE_MODE16__SAMPLE_SPLIT__SHIFT 0x19
+#define GB_TILE_MODE17__ARRAY_MODE_MASK 0x3c
+#define GB_TILE_MODE17__ARRAY_MODE__SHIFT 0x2
+#define GB_TILE_MODE17__PIPE_CONFIG_MASK 0x7c0
+#define GB_TILE_MODE17__PIPE_CONFIG__SHIFT 0x6
+#define GB_TILE_MODE17__TILE_SPLIT_MASK 0x3800
+#define GB_TILE_MODE17__TILE_SPLIT__SHIFT 0xb
+#define GB_TILE_MODE17__MICRO_TILE_MODE_NEW_MASK 0x1c00000
+#define GB_TILE_MODE17__MICRO_TILE_MODE_NEW__SHIFT 0x16
+#define GB_TILE_MODE17__SAMPLE_SPLIT_MASK 0x6000000
+#define GB_TILE_MODE17__SAMPLE_SPLIT__SHIFT 0x19
+#define GB_TILE_MODE18__ARRAY_MODE_MASK 0x3c
+#define GB_TILE_MODE18__ARRAY_MODE__SHIFT 0x2
+#define GB_TILE_MODE18__PIPE_CONFIG_MASK 0x7c0
+#define GB_TILE_MODE18__PIPE_CONFIG__SHIFT 0x6
+#define GB_TILE_MODE18__TILE_SPLIT_MASK 0x3800
+#define GB_TILE_MODE18__TILE_SPLIT__SHIFT 0xb
+#define GB_TILE_MODE18__MICRO_TILE_MODE_NEW_MASK 0x1c00000
+#define GB_TILE_MODE18__MICRO_TILE_MODE_NEW__SHIFT 0x16
+#define GB_TILE_MODE18__SAMPLE_SPLIT_MASK 0x6000000
+#define GB_TILE_MODE18__SAMPLE_SPLIT__SHIFT 0x19
+#define GB_TILE_MODE19__ARRAY_MODE_MASK 0x3c
+#define GB_TILE_MODE19__ARRAY_MODE__SHIFT 0x2
+#define GB_TILE_MODE19__PIPE_CONFIG_MASK 0x7c0
+#define GB_TILE_MODE19__PIPE_CONFIG__SHIFT 0x6
+#define GB_TILE_MODE19__TILE_SPLIT_MASK 0x3800
+#define GB_TILE_MODE19__TILE_SPLIT__SHIFT 0xb
+#define GB_TILE_MODE19__MICRO_TILE_MODE_NEW_MASK 0x1c00000
+#define GB_TILE_MODE19__MICRO_TILE_MODE_NEW__SHIFT 0x16
+#define GB_TILE_MODE19__SAMPLE_SPLIT_MASK 0x6000000
+#define GB_TILE_MODE19__SAMPLE_SPLIT__SHIFT 0x19
+#define GB_TILE_MODE20__ARRAY_MODE_MASK 0x3c
+#define GB_TILE_MODE20__ARRAY_MODE__SHIFT 0x2
+#define GB_TILE_MODE20__PIPE_CONFIG_MASK 0x7c0
+#define GB_TILE_MODE20__PIPE_CONFIG__SHIFT 0x6
+#define GB_TILE_MODE20__TILE_SPLIT_MASK 0x3800
+#define GB_TILE_MODE20__TILE_SPLIT__SHIFT 0xb
+#define GB_TILE_MODE20__MICRO_TILE_MODE_NEW_MASK 0x1c00000
+#define GB_TILE_MODE20__MICRO_TILE_MODE_NEW__SHIFT 0x16
+#define GB_TILE_MODE20__SAMPLE_SPLIT_MASK 0x6000000
+#define GB_TILE_MODE20__SAMPLE_SPLIT__SHIFT 0x19
+#define GB_TILE_MODE21__ARRAY_MODE_MASK 0x3c
+#define GB_TILE_MODE21__ARRAY_MODE__SHIFT 0x2
+#define GB_TILE_MODE21__PIPE_CONFIG_MASK 0x7c0
+#define GB_TILE_MODE21__PIPE_CONFIG__SHIFT 0x6
+#define GB_TILE_MODE21__TILE_SPLIT_MASK 0x3800
+#define GB_TILE_MODE21__TILE_SPLIT__SHIFT 0xb
+#define GB_TILE_MODE21__MICRO_TILE_MODE_NEW_MASK 0x1c00000
+#define GB_TILE_MODE21__MICRO_TILE_MODE_NEW__SHIFT 0x16
+#define GB_TILE_MODE21__SAMPLE_SPLIT_MASK 0x6000000
+#define GB_TILE_MODE21__SAMPLE_SPLIT__SHIFT 0x19
+#define GB_TILE_MODE22__ARRAY_MODE_MASK 0x3c
+#define GB_TILE_MODE22__ARRAY_MODE__SHIFT 0x2
+#define GB_TILE_MODE22__PIPE_CONFIG_MASK 0x7c0
+#define GB_TILE_MODE22__PIPE_CONFIG__SHIFT 0x6
+#define GB_TILE_MODE22__TILE_SPLIT_MASK 0x3800
+#define GB_TILE_MODE22__TILE_SPLIT__SHIFT 0xb
+#define GB_TILE_MODE22__MICRO_TILE_MODE_NEW_MASK 0x1c00000
+#define GB_TILE_MODE22__MICRO_TILE_MODE_NEW__SHIFT 0x16
+#define GB_TILE_MODE22__SAMPLE_SPLIT_MASK 0x6000000
+#define GB_TILE_MODE22__SAMPLE_SPLIT__SHIFT 0x19
+#define GB_TILE_MODE23__ARRAY_MODE_MASK 0x3c
+#define GB_TILE_MODE23__ARRAY_MODE__SHIFT 0x2
+#define GB_TILE_MODE23__PIPE_CONFIG_MASK 0x7c0
+#define GB_TILE_MODE23__PIPE_CONFIG__SHIFT 0x6
+#define GB_TILE_MODE23__TILE_SPLIT_MASK 0x3800
+#define GB_TILE_MODE23__TILE_SPLIT__SHIFT 0xb
+#define GB_TILE_MODE23__MICRO_TILE_MODE_NEW_MASK 0x1c00000
+#define GB_TILE_MODE23__MICRO_TILE_MODE_NEW__SHIFT 0x16
+#define GB_TILE_MODE23__SAMPLE_SPLIT_MASK 0x6000000
+#define GB_TILE_MODE23__SAMPLE_SPLIT__SHIFT 0x19
+#define GB_TILE_MODE24__ARRAY_MODE_MASK 0x3c
+#define GB_TILE_MODE24__ARRAY_MODE__SHIFT 0x2
+#define GB_TILE_MODE24__PIPE_CONFIG_MASK 0x7c0
+#define GB_TILE_MODE24__PIPE_CONFIG__SHIFT 0x6
+#define GB_TILE_MODE24__TILE_SPLIT_MASK 0x3800
+#define GB_TILE_MODE24__TILE_SPLIT__SHIFT 0xb
+#define GB_TILE_MODE24__MICRO_TILE_MODE_NEW_MASK 0x1c00000
+#define GB_TILE_MODE24__MICRO_TILE_MODE_NEW__SHIFT 0x16
+#define GB_TILE_MODE24__SAMPLE_SPLIT_MASK 0x6000000
+#define GB_TILE_MODE24__SAMPLE_SPLIT__SHIFT 0x19
+#define GB_TILE_MODE25__ARRAY_MODE_MASK 0x3c
+#define GB_TILE_MODE25__ARRAY_MODE__SHIFT 0x2
+#define GB_TILE_MODE25__PIPE_CONFIG_MASK 0x7c0
+#define GB_TILE_MODE25__PIPE_CONFIG__SHIFT 0x6
+#define GB_TILE_MODE25__TILE_SPLIT_MASK 0x3800
+#define GB_TILE_MODE25__TILE_SPLIT__SHIFT 0xb
+#define GB_TILE_MODE25__MICRO_TILE_MODE_NEW_MASK 0x1c00000
+#define GB_TILE_MODE25__MICRO_TILE_MODE_NEW__SHIFT 0x16
+#define GB_TILE_MODE25__SAMPLE_SPLIT_MASK 0x6000000
+#define GB_TILE_MODE25__SAMPLE_SPLIT__SHIFT 0x19
+#define GB_TILE_MODE26__ARRAY_MODE_MASK 0x3c
+#define GB_TILE_MODE26__ARRAY_MODE__SHIFT 0x2
+#define GB_TILE_MODE26__PIPE_CONFIG_MASK 0x7c0
+#define GB_TILE_MODE26__PIPE_CONFIG__SHIFT 0x6
+#define GB_TILE_MODE26__TILE_SPLIT_MASK 0x3800
+#define GB_TILE_MODE26__TILE_SPLIT__SHIFT 0xb
+#define GB_TILE_MODE26__MICRO_TILE_MODE_NEW_MASK 0x1c00000
+#define GB_TILE_MODE26__MICRO_TILE_MODE_NEW__SHIFT 0x16
+#define GB_TILE_MODE26__SAMPLE_SPLIT_MASK 0x6000000
+#define GB_TILE_MODE26__SAMPLE_SPLIT__SHIFT 0x19
+#define GB_TILE_MODE27__ARRAY_MODE_MASK 0x3c
+#define GB_TILE_MODE27__ARRAY_MODE__SHIFT 0x2
+#define GB_TILE_MODE27__PIPE_CONFIG_MASK 0x7c0
+#define GB_TILE_MODE27__PIPE_CONFIG__SHIFT 0x6
+#define GB_TILE_MODE27__TILE_SPLIT_MASK 0x3800
+#define GB_TILE_MODE27__TILE_SPLIT__SHIFT 0xb
+#define GB_TILE_MODE27__MICRO_TILE_MODE_NEW_MASK 0x1c00000
+#define GB_TILE_MODE27__MICRO_TILE_MODE_NEW__SHIFT 0x16
+#define GB_TILE_MODE27__SAMPLE_SPLIT_MASK 0x6000000
+#define GB_TILE_MODE27__SAMPLE_SPLIT__SHIFT 0x19
+#define GB_TILE_MODE28__ARRAY_MODE_MASK 0x3c
+#define GB_TILE_MODE28__ARRAY_MODE__SHIFT 0x2
+#define GB_TILE_MODE28__PIPE_CONFIG_MASK 0x7c0
+#define GB_TILE_MODE28__PIPE_CONFIG__SHIFT 0x6
+#define GB_TILE_MODE28__TILE_SPLIT_MASK 0x3800
+#define GB_TILE_MODE28__TILE_SPLIT__SHIFT 0xb
+#define GB_TILE_MODE28__MICRO_TILE_MODE_NEW_MASK 0x1c00000
+#define GB_TILE_MODE28__MICRO_TILE_MODE_NEW__SHIFT 0x16
+#define GB_TILE_MODE28__SAMPLE_SPLIT_MASK 0x6000000
+#define GB_TILE_MODE28__SAMPLE_SPLIT__SHIFT 0x19
+#define GB_TILE_MODE29__ARRAY_MODE_MASK 0x3c
+#define GB_TILE_MODE29__ARRAY_MODE__SHIFT 0x2
+#define GB_TILE_MODE29__PIPE_CONFIG_MASK 0x7c0
+#define GB_TILE_MODE29__PIPE_CONFIG__SHIFT 0x6
+#define GB_TILE_MODE29__TILE_SPLIT_MASK 0x3800
+#define GB_TILE_MODE29__TILE_SPLIT__SHIFT 0xb
+#define GB_TILE_MODE29__MICRO_TILE_MODE_NEW_MASK 0x1c00000
+#define GB_TILE_MODE29__MICRO_TILE_MODE_NEW__SHIFT 0x16
+#define GB_TILE_MODE29__SAMPLE_SPLIT_MASK 0x6000000
+#define GB_TILE_MODE29__SAMPLE_SPLIT__SHIFT 0x19
+#define GB_TILE_MODE30__ARRAY_MODE_MASK 0x3c
+#define GB_TILE_MODE30__ARRAY_MODE__SHIFT 0x2
+#define GB_TILE_MODE30__PIPE_CONFIG_MASK 0x7c0
+#define GB_TILE_MODE30__PIPE_CONFIG__SHIFT 0x6
+#define GB_TILE_MODE30__TILE_SPLIT_MASK 0x3800
+#define GB_TILE_MODE30__TILE_SPLIT__SHIFT 0xb
+#define GB_TILE_MODE30__MICRO_TILE_MODE_NEW_MASK 0x1c00000
+#define GB_TILE_MODE30__MICRO_TILE_MODE_NEW__SHIFT 0x16
+#define GB_TILE_MODE30__SAMPLE_SPLIT_MASK 0x6000000
+#define GB_TILE_MODE30__SAMPLE_SPLIT__SHIFT 0x19
+#define GB_TILE_MODE31__ARRAY_MODE_MASK 0x3c
+#define GB_TILE_MODE31__ARRAY_MODE__SHIFT 0x2
+#define GB_TILE_MODE31__PIPE_CONFIG_MASK 0x7c0
+#define GB_TILE_MODE31__PIPE_CONFIG__SHIFT 0x6
+#define GB_TILE_MODE31__TILE_SPLIT_MASK 0x3800
+#define GB_TILE_MODE31__TILE_SPLIT__SHIFT 0xb
+#define GB_TILE_MODE31__MICRO_TILE_MODE_NEW_MASK 0x1c00000
+#define GB_TILE_MODE31__MICRO_TILE_MODE_NEW__SHIFT 0x16
+#define GB_TILE_MODE31__SAMPLE_SPLIT_MASK 0x6000000
+#define GB_TILE_MODE31__SAMPLE_SPLIT__SHIFT 0x19
+#define GB_MACROTILE_MODE0__BANK_WIDTH_MASK 0x3
+#define GB_MACROTILE_MODE0__BANK_WIDTH__SHIFT 0x0
+#define GB_MACROTILE_MODE0__BANK_HEIGHT_MASK 0xc
+#define GB_MACROTILE_MODE0__BANK_HEIGHT__SHIFT 0x2
+#define GB_MACROTILE_MODE0__MACRO_TILE_ASPECT_MASK 0x30
+#define GB_MACROTILE_MODE0__MACRO_TILE_ASPECT__SHIFT 0x4
+#define GB_MACROTILE_MODE0__NUM_BANKS_MASK 0xc0
+#define GB_MACROTILE_MODE0__NUM_BANKS__SHIFT 0x6
+#define GB_MACROTILE_MODE1__BANK_WIDTH_MASK 0x3
+#define GB_MACROTILE_MODE1__BANK_WIDTH__SHIFT 0x0
+#define GB_MACROTILE_MODE1__BANK_HEIGHT_MASK 0xc
+#define GB_MACROTILE_MODE1__BANK_HEIGHT__SHIFT 0x2
+#define GB_MACROTILE_MODE1__MACRO_TILE_ASPECT_MASK 0x30
+#define GB_MACROTILE_MODE1__MACRO_TILE_ASPECT__SHIFT 0x4
+#define GB_MACROTILE_MODE1__NUM_BANKS_MASK 0xc0
+#define GB_MACROTILE_MODE1__NUM_BANKS__SHIFT 0x6
+#define GB_MACROTILE_MODE2__BANK_WIDTH_MASK 0x3
+#define GB_MACROTILE_MODE2__BANK_WIDTH__SHIFT 0x0
+#define GB_MACROTILE_MODE2__BANK_HEIGHT_MASK 0xc
+#define GB_MACROTILE_MODE2__BANK_HEIGHT__SHIFT 0x2
+#define GB_MACROTILE_MODE2__MACRO_TILE_ASPECT_MASK 0x30
+#define GB_MACROTILE_MODE2__MACRO_TILE_ASPECT__SHIFT 0x4
+#define GB_MACROTILE_MODE2__NUM_BANKS_MASK 0xc0
+#define GB_MACROTILE_MODE2__NUM_BANKS__SHIFT 0x6
+#define GB_MACROTILE_MODE3__BANK_WIDTH_MASK 0x3
+#define GB_MACROTILE_MODE3__BANK_WIDTH__SHIFT 0x0
+#define GB_MACROTILE_MODE3__BANK_HEIGHT_MASK 0xc
+#define GB_MACROTILE_MODE3__BANK_HEIGHT__SHIFT 0x2
+#define GB_MACROTILE_MODE3__MACRO_TILE_ASPECT_MASK 0x30
+#define GB_MACROTILE_MODE3__MACRO_TILE_ASPECT__SHIFT 0x4
+#define GB_MACROTILE_MODE3__NUM_BANKS_MASK 0xc0
+#define GB_MACROTILE_MODE3__NUM_BANKS__SHIFT 0x6
+#define GB_MACROTILE_MODE4__BANK_WIDTH_MASK 0x3
+#define GB_MACROTILE_MODE4__BANK_WIDTH__SHIFT 0x0
+#define GB_MACROTILE_MODE4__BANK_HEIGHT_MASK 0xc
+#define GB_MACROTILE_MODE4__BANK_HEIGHT__SHIFT 0x2
+#define GB_MACROTILE_MODE4__MACRO_TILE_ASPECT_MASK 0x30
+#define GB_MACROTILE_MODE4__MACRO_TILE_ASPECT__SHIFT 0x4
+#define GB_MACROTILE_MODE4__NUM_BANKS_MASK 0xc0
+#define GB_MACROTILE_MODE4__NUM_BANKS__SHIFT 0x6
+#define GB_MACROTILE_MODE5__BANK_WIDTH_MASK 0x3
+#define GB_MACROTILE_MODE5__BANK_WIDTH__SHIFT 0x0
+#define GB_MACROTILE_MODE5__BANK_HEIGHT_MASK 0xc
+#define GB_MACROTILE_MODE5__BANK_HEIGHT__SHIFT 0x2
+#define GB_MACROTILE_MODE5__MACRO_TILE_ASPECT_MASK 0x30
+#define GB_MACROTILE_MODE5__MACRO_TILE_ASPECT__SHIFT 0x4
+#define GB_MACROTILE_MODE5__NUM_BANKS_MASK 0xc0
+#define GB_MACROTILE_MODE5__NUM_BANKS__SHIFT 0x6
+#define GB_MACROTILE_MODE6__BANK_WIDTH_MASK 0x3
+#define GB_MACROTILE_MODE6__BANK_WIDTH__SHIFT 0x0
+#define GB_MACROTILE_MODE6__BANK_HEIGHT_MASK 0xc
+#define GB_MACROTILE_MODE6__BANK_HEIGHT__SHIFT 0x2
+#define GB_MACROTILE_MODE6__MACRO_TILE_ASPECT_MASK 0x30
+#define GB_MACROTILE_MODE6__MACRO_TILE_ASPECT__SHIFT 0x4
+#define GB_MACROTILE_MODE6__NUM_BANKS_MASK 0xc0
+#define GB_MACROTILE_MODE6__NUM_BANKS__SHIFT 0x6
+#define GB_MACROTILE_MODE7__BANK_WIDTH_MASK 0x3
+#define GB_MACROTILE_MODE7__BANK_WIDTH__SHIFT 0x0
+#define GB_MACROTILE_MODE7__BANK_HEIGHT_MASK 0xc
+#define GB_MACROTILE_MODE7__BANK_HEIGHT__SHIFT 0x2
+#define GB_MACROTILE_MODE7__MACRO_TILE_ASPECT_MASK 0x30
+#define GB_MACROTILE_MODE7__MACRO_TILE_ASPECT__SHIFT 0x4
+#define GB_MACROTILE_MODE7__NUM_BANKS_MASK 0xc0
+#define GB_MACROTILE_MODE7__NUM_BANKS__SHIFT 0x6
+#define GB_MACROTILE_MODE8__BANK_WIDTH_MASK 0x3
+#define GB_MACROTILE_MODE8__BANK_WIDTH__SHIFT 0x0
+#define GB_MACROTILE_MODE8__BANK_HEIGHT_MASK 0xc
+#define GB_MACROTILE_MODE8__BANK_HEIGHT__SHIFT 0x2
+#define GB_MACROTILE_MODE8__MACRO_TILE_ASPECT_MASK 0x30
+#define GB_MACROTILE_MODE8__MACRO_TILE_ASPECT__SHIFT 0x4
+#define GB_MACROTILE_MODE8__NUM_BANKS_MASK 0xc0
+#define GB_MACROTILE_MODE8__NUM_BANKS__SHIFT 0x6
+#define GB_MACROTILE_MODE9__BANK_WIDTH_MASK 0x3
+#define GB_MACROTILE_MODE9__BANK_WIDTH__SHIFT 0x0
+#define GB_MACROTILE_MODE9__BANK_HEIGHT_MASK 0xc
+#define GB_MACROTILE_MODE9__BANK_HEIGHT__SHIFT 0x2
+#define GB_MACROTILE_MODE9__MACRO_TILE_ASPECT_MASK 0x30
+#define GB_MACROTILE_MODE9__MACRO_TILE_ASPECT__SHIFT 0x4
+#define GB_MACROTILE_MODE9__NUM_BANKS_MASK 0xc0
+#define GB_MACROTILE_MODE9__NUM_BANKS__SHIFT 0x6
+#define GB_MACROTILE_MODE10__BANK_WIDTH_MASK 0x3
+#define GB_MACROTILE_MODE10__BANK_WIDTH__SHIFT 0x0
+#define GB_MACROTILE_MODE10__BANK_HEIGHT_MASK 0xc
+#define GB_MACROTILE_MODE10__BANK_HEIGHT__SHIFT 0x2
+#define GB_MACROTILE_MODE10__MACRO_TILE_ASPECT_MASK 0x30
+#define GB_MACROTILE_MODE10__MACRO_TILE_ASPECT__SHIFT 0x4
+#define GB_MACROTILE_MODE10__NUM_BANKS_MASK 0xc0
+#define GB_MACROTILE_MODE10__NUM_BANKS__SHIFT 0x6
+#define GB_MACROTILE_MODE11__BANK_WIDTH_MASK 0x3
+#define GB_MACROTILE_MODE11__BANK_WIDTH__SHIFT 0x0
+#define GB_MACROTILE_MODE11__BANK_HEIGHT_MASK 0xc
+#define GB_MACROTILE_MODE11__BANK_HEIGHT__SHIFT 0x2
+#define GB_MACROTILE_MODE11__MACRO_TILE_ASPECT_MASK 0x30
+#define GB_MACROTILE_MODE11__MACRO_TILE_ASPECT__SHIFT 0x4
+#define GB_MACROTILE_MODE11__NUM_BANKS_MASK 0xc0
+#define GB_MACROTILE_MODE11__NUM_BANKS__SHIFT 0x6
+#define GB_MACROTILE_MODE12__BANK_WIDTH_MASK 0x3
+#define GB_MACROTILE_MODE12__BANK_WIDTH__SHIFT 0x0
+#define GB_MACROTILE_MODE12__BANK_HEIGHT_MASK 0xc
+#define GB_MACROTILE_MODE12__BANK_HEIGHT__SHIFT 0x2
+#define GB_MACROTILE_MODE12__MACRO_TILE_ASPECT_MASK 0x30
+#define GB_MACROTILE_MODE12__MACRO_TILE_ASPECT__SHIFT 0x4
+#define GB_MACROTILE_MODE12__NUM_BANKS_MASK 0xc0
+#define GB_MACROTILE_MODE12__NUM_BANKS__SHIFT 0x6
+#define GB_MACROTILE_MODE13__BANK_WIDTH_MASK 0x3
+#define GB_MACROTILE_MODE13__BANK_WIDTH__SHIFT 0x0
+#define GB_MACROTILE_MODE13__BANK_HEIGHT_MASK 0xc
+#define GB_MACROTILE_MODE13__BANK_HEIGHT__SHIFT 0x2
+#define GB_MACROTILE_MODE13__MACRO_TILE_ASPECT_MASK 0x30
+#define GB_MACROTILE_MODE13__MACRO_TILE_ASPECT__SHIFT 0x4
+#define GB_MACROTILE_MODE13__NUM_BANKS_MASK 0xc0
+#define GB_MACROTILE_MODE13__NUM_BANKS__SHIFT 0x6
+#define GB_MACROTILE_MODE14__BANK_WIDTH_MASK 0x3
+#define GB_MACROTILE_MODE14__BANK_WIDTH__SHIFT 0x0
+#define GB_MACROTILE_MODE14__BANK_HEIGHT_MASK 0xc
+#define GB_MACROTILE_MODE14__BANK_HEIGHT__SHIFT 0x2
+#define GB_MACROTILE_MODE14__MACRO_TILE_ASPECT_MASK 0x30
+#define GB_MACROTILE_MODE14__MACRO_TILE_ASPECT__SHIFT 0x4
+#define GB_MACROTILE_MODE14__NUM_BANKS_MASK 0xc0
+#define GB_MACROTILE_MODE14__NUM_BANKS__SHIFT 0x6
+#define GB_MACROTILE_MODE15__BANK_WIDTH_MASK 0x3
+#define GB_MACROTILE_MODE15__BANK_WIDTH__SHIFT 0x0
+#define GB_MACROTILE_MODE15__BANK_HEIGHT_MASK 0xc
+#define GB_MACROTILE_MODE15__BANK_HEIGHT__SHIFT 0x2
+#define GB_MACROTILE_MODE15__MACRO_TILE_ASPECT_MASK 0x30
+#define GB_MACROTILE_MODE15__MACRO_TILE_ASPECT__SHIFT 0x4
+#define GB_MACROTILE_MODE15__NUM_BANKS_MASK 0xc0
+#define GB_MACROTILE_MODE15__NUM_BANKS__SHIFT 0x6
+#define GB_EDC_MODE__FORCE_SEC_ON_DED_MASK 0x10000
+#define GB_EDC_MODE__FORCE_SEC_ON_DED__SHIFT 0x10
+#define GB_EDC_MODE__DED_MODE_MASK 0x300000
+#define GB_EDC_MODE__DED_MODE__SHIFT 0x14
+#define GB_EDC_MODE__PROP_FED_MASK 0x20000000
+#define GB_EDC_MODE__PROP_FED__SHIFT 0x1d
+#define GB_EDC_MODE__BYPASS_MASK 0x80000000
+#define GB_EDC_MODE__BYPASS__SHIFT 0x1f
+#define CC_GC_EDC_CONFIG__DIS_EDC_MASK 0x2
+#define CC_GC_EDC_CONFIG__DIS_EDC__SHIFT 0x1
+#define RAS_SIGNATURE_CONTROL__ENABLE_MASK 0x1
+#define RAS_SIGNATURE_CONTROL__ENABLE__SHIFT 0x0
+#define RAS_SIGNATURE_MASK__INPUT_BUS_MASK_MASK 0xffffffff
+#define RAS_SIGNATURE_MASK__INPUT_BUS_MASK__SHIFT 0x0
+#define RAS_SX_SIGNATURE0__SIGNATURE_MASK 0xffffffff
+#define RAS_SX_SIGNATURE0__SIGNATURE__SHIFT 0x0
+#define RAS_SX_SIGNATURE1__SIGNATURE_MASK 0xffffffff
+#define RAS_SX_SIGNATURE1__SIGNATURE__SHIFT 0x0
+#define RAS_SX_SIGNATURE2__SIGNATURE_MASK 0xffffffff
+#define RAS_SX_SIGNATURE2__SIGNATURE__SHIFT 0x0
+#define RAS_SX_SIGNATURE3__SIGNATURE_MASK 0xffffffff
+#define RAS_SX_SIGNATURE3__SIGNATURE__SHIFT 0x0
+#define RAS_DB_SIGNATURE0__SIGNATURE_MASK 0xffffffff
+#define RAS_DB_SIGNATURE0__SIGNATURE__SHIFT 0x0
+#define RAS_PA_SIGNATURE0__SIGNATURE_MASK 0xffffffff
+#define RAS_PA_SIGNATURE0__SIGNATURE__SHIFT 0x0
+#define RAS_VGT_SIGNATURE0__SIGNATURE_MASK 0xffffffff
+#define RAS_VGT_SIGNATURE0__SIGNATURE__SHIFT 0x0
+#define RAS_SC_SIGNATURE0__SIGNATURE_MASK 0xffffffff
+#define RAS_SC_SIGNATURE0__SIGNATURE__SHIFT 0x0
+#define RAS_SC_SIGNATURE1__SIGNATURE_MASK 0xffffffff
+#define RAS_SC_SIGNATURE1__SIGNATURE__SHIFT 0x0
+#define RAS_SC_SIGNATURE2__SIGNATURE_MASK 0xffffffff
+#define RAS_SC_SIGNATURE2__SIGNATURE__SHIFT 0x0
+#define RAS_SC_SIGNATURE3__SIGNATURE_MASK 0xffffffff
+#define RAS_SC_SIGNATURE3__SIGNATURE__SHIFT 0x0
+#define RAS_SC_SIGNATURE4__SIGNATURE_MASK 0xffffffff
+#define RAS_SC_SIGNATURE4__SIGNATURE__SHIFT 0x0
+#define RAS_SC_SIGNATURE5__SIGNATURE_MASK 0xffffffff
+#define RAS_SC_SIGNATURE5__SIGNATURE__SHIFT 0x0
+#define RAS_SC_SIGNATURE6__SIGNATURE_MASK 0xffffffff
+#define RAS_SC_SIGNATURE6__SIGNATURE__SHIFT 0x0
+#define RAS_SC_SIGNATURE7__SIGNATURE_MASK 0xffffffff
+#define RAS_SC_SIGNATURE7__SIGNATURE__SHIFT 0x0
+#define RAS_IA_SIGNATURE0__SIGNATURE_MASK 0xffffffff
+#define RAS_IA_SIGNATURE0__SIGNATURE__SHIFT 0x0
+#define RAS_IA_SIGNATURE1__SIGNATURE_MASK 0xffffffff
+#define RAS_IA_SIGNATURE1__SIGNATURE__SHIFT 0x0
+#define RAS_SPI_SIGNATURE0__SIGNATURE_MASK 0xffffffff
+#define RAS_SPI_SIGNATURE0__SIGNATURE__SHIFT 0x0
+#define RAS_SPI_SIGNATURE1__SIGNATURE_MASK 0xffffffff
+#define RAS_SPI_SIGNATURE1__SIGNATURE__SHIFT 0x0
+#define RAS_TA_SIGNATURE0__SIGNATURE_MASK 0xffffffff
+#define RAS_TA_SIGNATURE0__SIGNATURE__SHIFT 0x0
+#define RAS_TD_SIGNATURE0__SIGNATURE_MASK 0xffffffff
+#define RAS_TD_SIGNATURE0__SIGNATURE__SHIFT 0x0
+#define RAS_CB_SIGNATURE0__SIGNATURE_MASK 0xffffffff
+#define RAS_CB_SIGNATURE0__SIGNATURE__SHIFT 0x0
+#define RAS_BCI_SIGNATURE0__SIGNATURE_MASK 0xffffffff
+#define RAS_BCI_SIGNATURE0__SIGNATURE__SHIFT 0x0
+#define RAS_BCI_SIGNATURE1__SIGNATURE_MASK 0xffffffff
+#define RAS_BCI_SIGNATURE1__SIGNATURE__SHIFT 0x0
+#define RAS_TA_SIGNATURE1__SIGNATURE_MASK 0xffffffff
+#define RAS_TA_SIGNATURE1__SIGNATURE__SHIFT 0x0
+#define GRBM_HYP_CAM_INDEX__CAM_INDEX_MASK 0x7
+#define GRBM_HYP_CAM_INDEX__CAM_INDEX__SHIFT 0x0
+#define GRBM_CAM_INDEX__CAM_INDEX_MASK 0x7
+#define GRBM_CAM_INDEX__CAM_INDEX__SHIFT 0x0
+#define GRBM_HYP_CAM_DATA__CAM_ADDR_MASK 0xffff
+#define GRBM_HYP_CAM_DATA__CAM_ADDR__SHIFT 0x0
+#define GRBM_HYP_CAM_DATA__CAM_REMAPADDR_MASK 0xffff0000
+#define GRBM_HYP_CAM_DATA__CAM_REMAPADDR__SHIFT 0x10
+#define GRBM_CAM_DATA__CAM_ADDR_MASK 0xffff
+#define GRBM_CAM_DATA__CAM_ADDR__SHIFT 0x0
+#define GRBM_CAM_DATA__CAM_REMAPADDR_MASK 0xffff0000
+#define GRBM_CAM_DATA__CAM_REMAPADDR__SHIFT 0x10
+#define GRBM_CNTL__READ_TIMEOUT_MASK 0xff
+#define GRBM_CNTL__READ_TIMEOUT__SHIFT 0x0
+#define GRBM_CNTL__REPORT_LAST_RDERR_MASK 0x80000000
+#define GRBM_CNTL__REPORT_LAST_RDERR__SHIFT 0x1f
+#define GRBM_SKEW_CNTL__SKEW_TOP_THRESHOLD_MASK 0x3f
+#define GRBM_SKEW_CNTL__SKEW_TOP_THRESHOLD__SHIFT 0x0
+#define GRBM_SKEW_CNTL__SKEW_COUNT_MASK 0xfc0
+#define GRBM_SKEW_CNTL__SKEW_COUNT__SHIFT 0x6
+#define GRBM_PWR_CNTL__ALL_REQ_TYPE_MASK 0x3
+#define GRBM_PWR_CNTL__ALL_REQ_TYPE__SHIFT 0x0
+#define GRBM_PWR_CNTL__GFX_REQ_TYPE_MASK 0xc
+#define GRBM_PWR_CNTL__GFX_REQ_TYPE__SHIFT 0x2
+#define GRBM_PWR_CNTL__ALL_RSP_TYPE_MASK 0x30
+#define GRBM_PWR_CNTL__ALL_RSP_TYPE__SHIFT 0x4
+#define GRBM_PWR_CNTL__GFX_RSP_TYPE_MASK 0xc0
+#define GRBM_PWR_CNTL__GFX_RSP_TYPE__SHIFT 0x6
+#define GRBM_PWR_CNTL__GFX_REQ_EN_MASK 0x4000
+#define GRBM_PWR_CNTL__GFX_REQ_EN__SHIFT 0xe
+#define GRBM_PWR_CNTL__ALL_REQ_EN_MASK 0x8000
+#define GRBM_PWR_CNTL__ALL_REQ_EN__SHIFT 0xf
+#define GRBM_STATUS__ME0PIPE0_CMDFIFO_AVAIL_MASK 0xf
+#define GRBM_STATUS__ME0PIPE0_CMDFIFO_AVAIL__SHIFT 0x0
+#define GRBM_STATUS__SRBM_RQ_PENDING_MASK 0x20
+#define GRBM_STATUS__SRBM_RQ_PENDING__SHIFT 0x5
+#define GRBM_STATUS__ME0PIPE0_CF_RQ_PENDING_MASK 0x80
+#define GRBM_STATUS__ME0PIPE0_CF_RQ_PENDING__SHIFT 0x7
+#define GRBM_STATUS__ME0PIPE0_PF_RQ_PENDING_MASK 0x100
+#define GRBM_STATUS__ME0PIPE0_PF_RQ_PENDING__SHIFT 0x8
+#define GRBM_STATUS__GDS_DMA_RQ_PENDING_MASK 0x200
+#define GRBM_STATUS__GDS_DMA_RQ_PENDING__SHIFT 0x9
+#define GRBM_STATUS__DB_CLEAN_MASK 0x1000
+#define GRBM_STATUS__DB_CLEAN__SHIFT 0xc
+#define GRBM_STATUS__CB_CLEAN_MASK 0x2000
+#define GRBM_STATUS__CB_CLEAN__SHIFT 0xd
+#define GRBM_STATUS__TA_BUSY_MASK 0x4000
+#define GRBM_STATUS__TA_BUSY__SHIFT 0xe
+#define GRBM_STATUS__GDS_BUSY_MASK 0x8000
+#define GRBM_STATUS__GDS_BUSY__SHIFT 0xf
+#define GRBM_STATUS__WD_BUSY_NO_DMA_MASK 0x10000
+#define GRBM_STATUS__WD_BUSY_NO_DMA__SHIFT 0x10
+#define GRBM_STATUS__VGT_BUSY_MASK 0x20000
+#define GRBM_STATUS__VGT_BUSY__SHIFT 0x11
+#define GRBM_STATUS__IA_BUSY_NO_DMA_MASK 0x40000
+#define GRBM_STATUS__IA_BUSY_NO_DMA__SHIFT 0x12
+#define GRBM_STATUS__IA_BUSY_MASK 0x80000
+#define GRBM_STATUS__IA_BUSY__SHIFT 0x13
+#define GRBM_STATUS__SX_BUSY_MASK 0x100000
+#define GRBM_STATUS__SX_BUSY__SHIFT 0x14
+#define GRBM_STATUS__WD_BUSY_MASK 0x200000
+#define GRBM_STATUS__WD_BUSY__SHIFT 0x15
+#define GRBM_STATUS__SPI_BUSY_MASK 0x400000
+#define GRBM_STATUS__SPI_BUSY__SHIFT 0x16
+#define GRBM_STATUS__BCI_BUSY_MASK 0x800000
+#define GRBM_STATUS__BCI_BUSY__SHIFT 0x17
+#define GRBM_STATUS__SC_BUSY_MASK 0x1000000
+#define GRBM_STATUS__SC_BUSY__SHIFT 0x18
+#define GRBM_STATUS__PA_BUSY_MASK 0x2000000
+#define GRBM_STATUS__PA_BUSY__SHIFT 0x19
+#define GRBM_STATUS__DB_BUSY_MASK 0x4000000
+#define GRBM_STATUS__DB_BUSY__SHIFT 0x1a
+#define GRBM_STATUS__CP_COHERENCY_BUSY_MASK 0x10000000
+#define GRBM_STATUS__CP_COHERENCY_BUSY__SHIFT 0x1c
+#define GRBM_STATUS__CP_BUSY_MASK 0x20000000
+#define GRBM_STATUS__CP_BUSY__SHIFT 0x1d
+#define GRBM_STATUS__CB_BUSY_MASK 0x40000000
+#define GRBM_STATUS__CB_BUSY__SHIFT 0x1e
+#define GRBM_STATUS__GUI_ACTIVE_MASK 0x80000000
+#define GRBM_STATUS__GUI_ACTIVE__SHIFT 0x1f
+#define GRBM_STATUS2__ME0PIPE1_CMDFIFO_AVAIL_MASK 0xf
+#define GRBM_STATUS2__ME0PIPE1_CMDFIFO_AVAIL__SHIFT 0x0
+#define GRBM_STATUS2__ME0PIPE1_CF_RQ_PENDING_MASK 0x10
+#define GRBM_STATUS2__ME0PIPE1_CF_RQ_PENDING__SHIFT 0x4
+#define GRBM_STATUS2__ME0PIPE1_PF_RQ_PENDING_MASK 0x20
+#define GRBM_STATUS2__ME0PIPE1_PF_RQ_PENDING__SHIFT 0x5
+#define GRBM_STATUS2__ME1PIPE0_RQ_PENDING_MASK 0x40
+#define GRBM_STATUS2__ME1PIPE0_RQ_PENDING__SHIFT 0x6
+#define GRBM_STATUS2__ME1PIPE1_RQ_PENDING_MASK 0x80
+#define GRBM_STATUS2__ME1PIPE1_RQ_PENDING__SHIFT 0x7
+#define GRBM_STATUS2__ME1PIPE2_RQ_PENDING_MASK 0x100
+#define GRBM_STATUS2__ME1PIPE2_RQ_PENDING__SHIFT 0x8
+#define GRBM_STATUS2__ME1PIPE3_RQ_PENDING_MASK 0x200
+#define GRBM_STATUS2__ME1PIPE3_RQ_PENDING__SHIFT 0x9
+#define GRBM_STATUS2__ME2PIPE0_RQ_PENDING_MASK 0x400
+#define GRBM_STATUS2__ME2PIPE0_RQ_PENDING__SHIFT 0xa
+#define GRBM_STATUS2__ME2PIPE1_RQ_PENDING_MASK 0x800
+#define GRBM_STATUS2__ME2PIPE1_RQ_PENDING__SHIFT 0xb
+#define GRBM_STATUS2__ME2PIPE2_RQ_PENDING_MASK 0x1000
+#define GRBM_STATUS2__ME2PIPE2_RQ_PENDING__SHIFT 0xc
+#define GRBM_STATUS2__ME2PIPE3_RQ_PENDING_MASK 0x2000
+#define GRBM_STATUS2__ME2PIPE3_RQ_PENDING__SHIFT 0xd
+#define GRBM_STATUS2__RLC_RQ_PENDING_MASK 0x4000
+#define GRBM_STATUS2__RLC_RQ_PENDING__SHIFT 0xe
+#define GRBM_STATUS2__RLC_BUSY_MASK 0x1000000
+#define GRBM_STATUS2__RLC_BUSY__SHIFT 0x18
+#define GRBM_STATUS2__TC_BUSY_MASK 0x2000000
+#define GRBM_STATUS2__TC_BUSY__SHIFT 0x19
+#define GRBM_STATUS2__TCC_CC_RESIDENT_MASK 0x4000000
+#define GRBM_STATUS2__TCC_CC_RESIDENT__SHIFT 0x1a
+#define GRBM_STATUS2__CPF_BUSY_MASK 0x10000000
+#define GRBM_STATUS2__CPF_BUSY__SHIFT 0x1c
+#define GRBM_STATUS2__CPC_BUSY_MASK 0x20000000
+#define GRBM_STATUS2__CPC_BUSY__SHIFT 0x1d
+#define GRBM_STATUS2__CPG_BUSY_MASK 0x40000000
+#define GRBM_STATUS2__CPG_BUSY__SHIFT 0x1e
+#define GRBM_STATUS_SE0__DB_CLEAN_MASK 0x2
+#define GRBM_STATUS_SE0__DB_CLEAN__SHIFT 0x1
+#define GRBM_STATUS_SE0__CB_CLEAN_MASK 0x4
+#define GRBM_STATUS_SE0__CB_CLEAN__SHIFT 0x2
+#define GRBM_STATUS_SE0__BCI_BUSY_MASK 0x400000
+#define GRBM_STATUS_SE0__BCI_BUSY__SHIFT 0x16
+#define GRBM_STATUS_SE0__VGT_BUSY_MASK 0x800000
+#define GRBM_STATUS_SE0__VGT_BUSY__SHIFT 0x17
+#define GRBM_STATUS_SE0__PA_BUSY_MASK 0x1000000
+#define GRBM_STATUS_SE0__PA_BUSY__SHIFT 0x18
+#define GRBM_STATUS_SE0__TA_BUSY_MASK 0x2000000
+#define GRBM_STATUS_SE0__TA_BUSY__SHIFT 0x19
+#define GRBM_STATUS_SE0__SX_BUSY_MASK 0x4000000
+#define GRBM_STATUS_SE0__SX_BUSY__SHIFT 0x1a
+#define GRBM_STATUS_SE0__SPI_BUSY_MASK 0x8000000
+#define GRBM_STATUS_SE0__SPI_BUSY__SHIFT 0x1b
+#define GRBM_STATUS_SE0__SC_BUSY_MASK 0x20000000
+#define GRBM_STATUS_SE0__SC_BUSY__SHIFT 0x1d
+#define GRBM_STATUS_SE0__DB_BUSY_MASK 0x40000000
+#define GRBM_STATUS_SE0__DB_BUSY__SHIFT 0x1e
+#define GRBM_STATUS_SE0__CB_BUSY_MASK 0x80000000
+#define GRBM_STATUS_SE0__CB_BUSY__SHIFT 0x1f
+#define GRBM_STATUS_SE1__DB_CLEAN_MASK 0x2
+#define GRBM_STATUS_SE1__DB_CLEAN__SHIFT 0x1
+#define GRBM_STATUS_SE1__CB_CLEAN_MASK 0x4
+#define GRBM_STATUS_SE1__CB_CLEAN__SHIFT 0x2
+#define GRBM_STATUS_SE1__BCI_BUSY_MASK 0x400000
+#define GRBM_STATUS_SE1__BCI_BUSY__SHIFT 0x16
+#define GRBM_STATUS_SE1__VGT_BUSY_MASK 0x800000
+#define GRBM_STATUS_SE1__VGT_BUSY__SHIFT 0x17
+#define GRBM_STATUS_SE1__PA_BUSY_MASK 0x1000000
+#define GRBM_STATUS_SE1__PA_BUSY__SHIFT 0x18
+#define GRBM_STATUS_SE1__TA_BUSY_MASK 0x2000000
+#define GRBM_STATUS_SE1__TA_BUSY__SHIFT 0x19
+#define GRBM_STATUS_SE1__SX_BUSY_MASK 0x4000000
+#define GRBM_STATUS_SE1__SX_BUSY__SHIFT 0x1a
+#define GRBM_STATUS_SE1__SPI_BUSY_MASK 0x8000000
+#define GRBM_STATUS_SE1__SPI_BUSY__SHIFT 0x1b
+#define GRBM_STATUS_SE1__SC_BUSY_MASK 0x20000000
+#define GRBM_STATUS_SE1__SC_BUSY__SHIFT 0x1d
+#define GRBM_STATUS_SE1__DB_BUSY_MASK 0x40000000
+#define GRBM_STATUS_SE1__DB_BUSY__SHIFT 0x1e
+#define GRBM_STATUS_SE1__CB_BUSY_MASK 0x80000000
+#define GRBM_STATUS_SE1__CB_BUSY__SHIFT 0x1f
+#define GRBM_STATUS_SE2__DB_CLEAN_MASK 0x2
+#define GRBM_STATUS_SE2__DB_CLEAN__SHIFT 0x1
+#define GRBM_STATUS_SE2__CB_CLEAN_MASK 0x4
+#define GRBM_STATUS_SE2__CB_CLEAN__SHIFT 0x2
+#define GRBM_STATUS_SE2__BCI_BUSY_MASK 0x400000
+#define GRBM_STATUS_SE2__BCI_BUSY__SHIFT 0x16
+#define GRBM_STATUS_SE2__VGT_BUSY_MASK 0x800000
+#define GRBM_STATUS_SE2__VGT_BUSY__SHIFT 0x17
+#define GRBM_STATUS_SE2__PA_BUSY_MASK 0x1000000
+#define GRBM_STATUS_SE2__PA_BUSY__SHIFT 0x18
+#define GRBM_STATUS_SE2__TA_BUSY_MASK 0x2000000
+#define GRBM_STATUS_SE2__TA_BUSY__SHIFT 0x19
+#define GRBM_STATUS_SE2__SX_BUSY_MASK 0x4000000
+#define GRBM_STATUS_SE2__SX_BUSY__SHIFT 0x1a
+#define GRBM_STATUS_SE2__SPI_BUSY_MASK 0x8000000
+#define GRBM_STATUS_SE2__SPI_BUSY__SHIFT 0x1b
+#define GRBM_STATUS_SE2__SC_BUSY_MASK 0x20000000
+#define GRBM_STATUS_SE2__SC_BUSY__SHIFT 0x1d
+#define GRBM_STATUS_SE2__DB_BUSY_MASK 0x40000000
+#define GRBM_STATUS_SE2__DB_BUSY__SHIFT 0x1e
+#define GRBM_STATUS_SE2__CB_BUSY_MASK 0x80000000
+#define GRBM_STATUS_SE2__CB_BUSY__SHIFT 0x1f
+#define GRBM_STATUS_SE3__DB_CLEAN_MASK 0x2
+#define GRBM_STATUS_SE3__DB_CLEAN__SHIFT 0x1
+#define GRBM_STATUS_SE3__CB_CLEAN_MASK 0x4
+#define GRBM_STATUS_SE3__CB_CLEAN__SHIFT 0x2
+#define GRBM_STATUS_SE3__BCI_BUSY_MASK 0x400000
+#define GRBM_STATUS_SE3__BCI_BUSY__SHIFT 0x16
+#define GRBM_STATUS_SE3__VGT_BUSY_MASK 0x800000
+#define GRBM_STATUS_SE3__VGT_BUSY__SHIFT 0x17
+#define GRBM_STATUS_SE3__PA_BUSY_MASK 0x1000000
+#define GRBM_STATUS_SE3__PA_BUSY__SHIFT 0x18
+#define GRBM_STATUS_SE3__TA_BUSY_MASK 0x2000000
+#define GRBM_STATUS_SE3__TA_BUSY__SHIFT 0x19
+#define GRBM_STATUS_SE3__SX_BUSY_MASK 0x4000000
+#define GRBM_STATUS_SE3__SX_BUSY__SHIFT 0x1a
+#define GRBM_STATUS_SE3__SPI_BUSY_MASK 0x8000000
+#define GRBM_STATUS_SE3__SPI_BUSY__SHIFT 0x1b
+#define GRBM_STATUS_SE3__SC_BUSY_MASK 0x20000000
+#define GRBM_STATUS_SE3__SC_BUSY__SHIFT 0x1d
+#define GRBM_STATUS_SE3__DB_BUSY_MASK 0x40000000
+#define GRBM_STATUS_SE3__DB_BUSY__SHIFT 0x1e
+#define GRBM_STATUS_SE3__CB_BUSY_MASK 0x80000000
+#define GRBM_STATUS_SE3__CB_BUSY__SHIFT 0x1f
+#define GRBM_SOFT_RESET__SOFT_RESET_CP_MASK 0x1
+#define GRBM_SOFT_RESET__SOFT_RESET_CP__SHIFT 0x0
+#define GRBM_SOFT_RESET__SOFT_RESET_RLC_MASK 0x4
+#define GRBM_SOFT_RESET__SOFT_RESET_RLC__SHIFT 0x2
+#define GRBM_SOFT_RESET__SOFT_RESET_GFX_MASK 0x10000
+#define GRBM_SOFT_RESET__SOFT_RESET_GFX__SHIFT 0x10
+#define GRBM_SOFT_RESET__SOFT_RESET_CPF_MASK 0x20000
+#define GRBM_SOFT_RESET__SOFT_RESET_CPF__SHIFT 0x11
+#define GRBM_SOFT_RESET__SOFT_RESET_CPC_MASK 0x40000
+#define GRBM_SOFT_RESET__SOFT_RESET_CPC__SHIFT 0x12
+#define GRBM_SOFT_RESET__SOFT_RESET_CPG_MASK 0x80000
+#define GRBM_SOFT_RESET__SOFT_RESET_CPG__SHIFT 0x13
+#define GRBM_SOFT_RESET__SOFT_RESET_CAC_MASK 0x100000
+#define GRBM_SOFT_RESET__SOFT_RESET_CAC__SHIFT 0x14
+#define GRBM_DEBUG_CNTL__GRBM_DEBUG_INDEX_MASK 0x3f
+#define GRBM_DEBUG_CNTL__GRBM_DEBUG_INDEX__SHIFT 0x0
+#define GRBM_DEBUG_DATA__DATA_MASK 0xffffffff
+#define GRBM_DEBUG_DATA__DATA__SHIFT 0x0
+#define GRBM_CGTT_CLK_CNTL__ON_DELAY_MASK 0xf
+#define GRBM_CGTT_CLK_CNTL__ON_DELAY__SHIFT 0x0
+#define GRBM_CGTT_CLK_CNTL__OFF_HYSTERESIS_MASK 0xff0
+#define GRBM_CGTT_CLK_CNTL__OFF_HYSTERESIS__SHIFT 0x4
+#define GRBM_CGTT_CLK_CNTL__SOFT_OVERRIDE_DYN_MASK 0x40000000
+#define GRBM_CGTT_CLK_CNTL__SOFT_OVERRIDE_DYN__SHIFT 0x1e
+#define GRBM_GFX_INDEX__INSTANCE_INDEX_MASK 0xff
+#define GRBM_GFX_INDEX__INSTANCE_INDEX__SHIFT 0x0
+#define GRBM_GFX_INDEX__SH_INDEX_MASK 0xff00
+#define GRBM_GFX_INDEX__SH_INDEX__SHIFT 0x8
+#define GRBM_GFX_INDEX__SE_INDEX_MASK 0xff0000
+#define GRBM_GFX_INDEX__SE_INDEX__SHIFT 0x10
+#define GRBM_GFX_INDEX__SH_BROADCAST_WRITES_MASK 0x20000000
+#define GRBM_GFX_INDEX__SH_BROADCAST_WRITES__SHIFT 0x1d
+#define GRBM_GFX_INDEX__INSTANCE_BROADCAST_WRITES_MASK 0x40000000
+#define GRBM_GFX_INDEX__INSTANCE_BROADCAST_WRITES__SHIFT 0x1e
+#define GRBM_GFX_INDEX__SE_BROADCAST_WRITES_MASK 0x80000000
+#define GRBM_GFX_INDEX__SE_BROADCAST_WRITES__SHIFT 0x1f
+#define GRBM_GFX_CLKEN_CNTL__PREFIX_DELAY_CNT_MASK 0xf
+#define GRBM_GFX_CLKEN_CNTL__PREFIX_DELAY_CNT__SHIFT 0x0
+#define GRBM_GFX_CLKEN_CNTL__POST_DELAY_CNT_MASK 0x1f00
+#define GRBM_GFX_CLKEN_CNTL__POST_DELAY_CNT__SHIFT 0x8
+#define GRBM_WAIT_IDLE_CLOCKS__WAIT_IDLE_CLOCKS_MASK 0xff
+#define GRBM_WAIT_IDLE_CLOCKS__WAIT_IDLE_CLOCKS__SHIFT 0x0
+#define GRBM_DEBUG__IGNORE_RDY_MASK 0x2
+#define GRBM_DEBUG__IGNORE_RDY__SHIFT 0x1
+#define GRBM_DEBUG__IGNORE_FAO_MASK 0x20
+#define GRBM_DEBUG__IGNORE_FAO__SHIFT 0x5
+#define GRBM_DEBUG__DISABLE_READ_TIMEOUT_MASK 0x40
+#define GRBM_DEBUG__DISABLE_READ_TIMEOUT__SHIFT 0x6
+#define GRBM_DEBUG__SNAPSHOT_FREE_CNTRS_MASK 0x80
+#define GRBM_DEBUG__SNAPSHOT_FREE_CNTRS__SHIFT 0x7
+#define GRBM_DEBUG__HYSTERESIS_GUI_ACTIVE_MASK 0xf00
+#define GRBM_DEBUG__HYSTERESIS_GUI_ACTIVE__SHIFT 0x8
+#define GRBM_DEBUG__GFX_CLOCK_DOMAIN_OVERRIDE_MASK 0x1000
+#define GRBM_DEBUG__GFX_CLOCK_DOMAIN_OVERRIDE__SHIFT 0xc
+#define GRBM_DEBUG__GRBM_TRAP_ENABLE_MASK 0x2000
+#define GRBM_DEBUG__GRBM_TRAP_ENABLE__SHIFT 0xd
+#define GRBM_DEBUG__DEBUG_BUS_FGCG_EN_MASK 0x80000000
+#define GRBM_DEBUG__DEBUG_BUS_FGCG_EN__SHIFT 0x1f
+#define GRBM_DEBUG_SNAPSHOT__CPF_RDY_MASK 0x1
+#define GRBM_DEBUG_SNAPSHOT__CPF_RDY__SHIFT 0x0
+#define GRBM_DEBUG_SNAPSHOT__CPG_RDY_MASK 0x2
+#define GRBM_DEBUG_SNAPSHOT__CPG_RDY__SHIFT 0x1
+#define GRBM_DEBUG_SNAPSHOT__SRBM_RDY_MASK 0x4
+#define GRBM_DEBUG_SNAPSHOT__SRBM_RDY__SHIFT 0x2
+#define GRBM_DEBUG_SNAPSHOT__WD_ME0PIPE0_RDY_MASK 0x8
+#define GRBM_DEBUG_SNAPSHOT__WD_ME0PIPE0_RDY__SHIFT 0x3
+#define GRBM_DEBUG_SNAPSHOT__WD_ME0PIPE1_RDY_MASK 0x10
+#define GRBM_DEBUG_SNAPSHOT__WD_ME0PIPE1_RDY__SHIFT 0x4
+#define GRBM_DEBUG_SNAPSHOT__GDS_RDY_MASK 0x20
+#define GRBM_DEBUG_SNAPSHOT__GDS_RDY__SHIFT 0x5
+#define GRBM_DEBUG_SNAPSHOT__SE0SPI_ME0PIPE0_RDY0_MASK 0x40
+#define GRBM_DEBUG_SNAPSHOT__SE0SPI_ME0PIPE0_RDY0__SHIFT 0x6
+#define GRBM_DEBUG_SNAPSHOT__SE0SPI_ME0PIPE1_RDY0_MASK 0x80
+#define GRBM_DEBUG_SNAPSHOT__SE0SPI_ME0PIPE1_RDY0__SHIFT 0x7
+#define GRBM_DEBUG_SNAPSHOT__SE1SPI_ME0PIPE0_RDY0_MASK 0x100
+#define GRBM_DEBUG_SNAPSHOT__SE1SPI_ME0PIPE0_RDY0__SHIFT 0x8
+#define GRBM_DEBUG_SNAPSHOT__SE1SPI_ME0PIPE1_RDY0_MASK 0x200
+#define GRBM_DEBUG_SNAPSHOT__SE1SPI_ME0PIPE1_RDY0__SHIFT 0x9
+#define GRBM_DEBUG_SNAPSHOT__SE2SPI_ME0PIPE0_RDY0_MASK 0x400
+#define GRBM_DEBUG_SNAPSHOT__SE2SPI_ME0PIPE0_RDY0__SHIFT 0xa
+#define GRBM_DEBUG_SNAPSHOT__SE2SPI_ME0PIPE1_RDY0_MASK 0x800
+#define GRBM_DEBUG_SNAPSHOT__SE2SPI_ME0PIPE1_RDY0__SHIFT 0xb
+#define GRBM_DEBUG_SNAPSHOT__SE3SPI_ME0PIPE0_RDY0_MASK 0x1000
+#define GRBM_DEBUG_SNAPSHOT__SE3SPI_ME0PIPE0_RDY0__SHIFT 0xc
+#define GRBM_DEBUG_SNAPSHOT__SE3SPI_ME0PIPE1_RDY0_MASK 0x2000
+#define GRBM_DEBUG_SNAPSHOT__SE3SPI_ME0PIPE1_RDY0__SHIFT 0xd
+#define GRBM_DEBUG_SNAPSHOT__SE0SPI_ME0PIPE0_RDY1_MASK 0x4000
+#define GRBM_DEBUG_SNAPSHOT__SE0SPI_ME0PIPE0_RDY1__SHIFT 0xe
+#define GRBM_DEBUG_SNAPSHOT__SE0SPI_ME0PIPE1_RDY1_MASK 0x8000
+#define GRBM_DEBUG_SNAPSHOT__SE0SPI_ME0PIPE1_RDY1__SHIFT 0xf
+#define GRBM_DEBUG_SNAPSHOT__SE1SPI_ME0PIPE0_RDY1_MASK 0x10000
+#define GRBM_DEBUG_SNAPSHOT__SE1SPI_ME0PIPE0_RDY1__SHIFT 0x10
+#define GRBM_DEBUG_SNAPSHOT__SE1SPI_ME0PIPE1_RDY1_MASK 0x20000
+#define GRBM_DEBUG_SNAPSHOT__SE1SPI_ME0PIPE1_RDY1__SHIFT 0x11
+#define GRBM_DEBUG_SNAPSHOT__SE2SPI_ME0PIPE0_RDY1_MASK 0x40000
+#define GRBM_DEBUG_SNAPSHOT__SE2SPI_ME0PIPE0_RDY1__SHIFT 0x12
+#define GRBM_DEBUG_SNAPSHOT__SE2SPI_ME0PIPE1_RDY1_MASK 0x80000
+#define GRBM_DEBUG_SNAPSHOT__SE2SPI_ME0PIPE1_RDY1__SHIFT 0x13
+#define GRBM_DEBUG_SNAPSHOT__SE3SPI_ME0PIPE0_RDY1_MASK 0x100000
+#define GRBM_DEBUG_SNAPSHOT__SE3SPI_ME0PIPE0_RDY1__SHIFT 0x14
+#define GRBM_DEBUG_SNAPSHOT__SE3SPI_ME0PIPE1_RDY1_MASK 0x200000
+#define GRBM_DEBUG_SNAPSHOT__SE3SPI_ME0PIPE1_RDY1__SHIFT 0x15
+#define GRBM_READ_ERROR__READ_ADDRESS_MASK 0x3fffc
+#define GRBM_READ_ERROR__READ_ADDRESS__SHIFT 0x2
+#define GRBM_READ_ERROR__READ_PIPEID_MASK 0x300000
+#define GRBM_READ_ERROR__READ_PIPEID__SHIFT 0x14
+#define GRBM_READ_ERROR__READ_MEID_MASK 0xc00000
+#define GRBM_READ_ERROR__READ_MEID__SHIFT 0x16
+#define GRBM_READ_ERROR__READ_ERROR_MASK 0x80000000
+#define GRBM_READ_ERROR__READ_ERROR__SHIFT 0x1f
+#define GRBM_READ_ERROR2__READ_REQUESTER_SRBM_MASK 0x20000
+#define GRBM_READ_ERROR2__READ_REQUESTER_SRBM__SHIFT 0x11
+#define GRBM_READ_ERROR2__READ_REQUESTER_RLC_MASK 0x40000
+#define GRBM_READ_ERROR2__READ_REQUESTER_RLC__SHIFT 0x12
+#define GRBM_READ_ERROR2__READ_REQUESTER_GDS_DMA_MASK 0x80000
+#define GRBM_READ_ERROR2__READ_REQUESTER_GDS_DMA__SHIFT 0x13
+#define GRBM_READ_ERROR2__READ_REQUESTER_ME0PIPE0_CF_MASK 0x100000
+#define GRBM_READ_ERROR2__READ_REQUESTER_ME0PIPE0_CF__SHIFT 0x14
+#define GRBM_READ_ERROR2__READ_REQUESTER_ME0PIPE0_PF_MASK 0x200000
+#define GRBM_READ_ERROR2__READ_REQUESTER_ME0PIPE0_PF__SHIFT 0x15
+#define GRBM_READ_ERROR2__READ_REQUESTER_ME0PIPE1_CF_MASK 0x400000
+#define GRBM_READ_ERROR2__READ_REQUESTER_ME0PIPE1_CF__SHIFT 0x16
+#define GRBM_READ_ERROR2__READ_REQUESTER_ME0PIPE1_PF_MASK 0x800000
+#define GRBM_READ_ERROR2__READ_REQUESTER_ME0PIPE1_PF__SHIFT 0x17
+#define GRBM_READ_ERROR2__READ_REQUESTER_ME1PIPE0_MASK 0x1000000
+#define GRBM_READ_ERROR2__READ_REQUESTER_ME1PIPE0__SHIFT 0x18
+#define GRBM_READ_ERROR2__READ_REQUESTER_ME1PIPE1_MASK 0x2000000
+#define GRBM_READ_ERROR2__READ_REQUESTER_ME1PIPE1__SHIFT 0x19
+#define GRBM_READ_ERROR2__READ_REQUESTER_ME1PIPE2_MASK 0x4000000
+#define GRBM_READ_ERROR2__READ_REQUESTER_ME1PIPE2__SHIFT 0x1a
+#define GRBM_READ_ERROR2__READ_REQUESTER_ME1PIPE3_MASK 0x8000000
+#define GRBM_READ_ERROR2__READ_REQUESTER_ME1PIPE3__SHIFT 0x1b
+#define GRBM_READ_ERROR2__READ_REQUESTER_ME2PIPE0_MASK 0x10000000
+#define GRBM_READ_ERROR2__READ_REQUESTER_ME2PIPE0__SHIFT 0x1c
+#define GRBM_READ_ERROR2__READ_REQUESTER_ME2PIPE1_MASK 0x20000000
+#define GRBM_READ_ERROR2__READ_REQUESTER_ME2PIPE1__SHIFT 0x1d
+#define GRBM_READ_ERROR2__READ_REQUESTER_ME2PIPE2_MASK 0x40000000
+#define GRBM_READ_ERROR2__READ_REQUESTER_ME2PIPE2__SHIFT 0x1e
+#define GRBM_READ_ERROR2__READ_REQUESTER_ME2PIPE3_MASK 0x80000000
+#define GRBM_READ_ERROR2__READ_REQUESTER_ME2PIPE3__SHIFT 0x1f
+#define GRBM_INT_CNTL__RDERR_INT_ENABLE_MASK 0x1
+#define GRBM_INT_CNTL__RDERR_INT_ENABLE__SHIFT 0x0
+#define GRBM_INT_CNTL__GUI_IDLE_INT_ENABLE_MASK 0x80000
+#define GRBM_INT_CNTL__GUI_IDLE_INT_ENABLE__SHIFT 0x13
+#define GRBM_TRAP_OP__RW_MASK 0x1
+#define GRBM_TRAP_OP__RW__SHIFT 0x0
+#define GRBM_TRAP_ADDR__DATA_MASK 0xffff
+#define GRBM_TRAP_ADDR__DATA__SHIFT 0x0
+#define GRBM_TRAP_ADDR_MSK__DATA_MASK 0xffff
+#define GRBM_TRAP_ADDR_MSK__DATA__SHIFT 0x0
+#define GRBM_TRAP_WD__DATA_MASK 0xffffffff
+#define GRBM_TRAP_WD__DATA__SHIFT 0x0
+#define GRBM_TRAP_WD_MSK__DATA_MASK 0xffffffff
+#define GRBM_TRAP_WD_MSK__DATA__SHIFT 0x0
+#define GRBM_DSM_BYPASS__BYPASS_BITS_MASK 0x3
+#define GRBM_DSM_BYPASS__BYPASS_BITS__SHIFT 0x0
+#define GRBM_DSM_BYPASS__BYPASS_EN_MASK 0x4
+#define GRBM_DSM_BYPASS__BYPASS_EN__SHIFT 0x2
+#define GRBM_WRITE_ERROR__WRITE_REQUESTER_RLC_MASK 0x1
+#define GRBM_WRITE_ERROR__WRITE_REQUESTER_RLC__SHIFT 0x0
+#define GRBM_WRITE_ERROR__WRITE_REQUESTER_SRBM_MASK 0x2
+#define GRBM_WRITE_ERROR__WRITE_REQUESTER_SRBM__SHIFT 0x1
+#define GRBM_WRITE_ERROR__WRITE_SSRCID_MASK 0x1c
+#define GRBM_WRITE_ERROR__WRITE_SSRCID__SHIFT 0x2
+#define GRBM_WRITE_ERROR__WRITE_VFID_MASK 0x1e0
+#define GRBM_WRITE_ERROR__WRITE_VFID__SHIFT 0x5
+#define GRBM_WRITE_ERROR__WRITE_VF_MASK 0x1000
+#define GRBM_WRITE_ERROR__WRITE_VF__SHIFT 0xc
+#define GRBM_WRITE_ERROR__WRITE_VMID_MASK 0x1e000
+#define GRBM_WRITE_ERROR__WRITE_VMID__SHIFT 0xd
+#define GRBM_WRITE_ERROR__WRITE_PIPEID_MASK 0x300000
+#define GRBM_WRITE_ERROR__WRITE_PIPEID__SHIFT 0x14
+#define GRBM_WRITE_ERROR__WRITE_MEID_MASK 0xc00000
+#define GRBM_WRITE_ERROR__WRITE_MEID__SHIFT 0x16
+#define GRBM_WRITE_ERROR__WRITE_ERROR_MASK 0x80000000
+#define GRBM_WRITE_ERROR__WRITE_ERROR__SHIFT 0x1f
+#define GRBM_PERFCOUNTER0_SELECT__PERF_SEL_MASK 0x3f
+#define GRBM_PERFCOUNTER0_SELECT__PERF_SEL__SHIFT 0x0
+#define GRBM_PERFCOUNTER0_SELECT__DB_CLEAN_USER_DEFINED_MASK_MASK 0x400
+#define GRBM_PERFCOUNTER0_SELECT__DB_CLEAN_USER_DEFINED_MASK__SHIFT 0xa
+#define GRBM_PERFCOUNTER0_SELECT__CB_CLEAN_USER_DEFINED_MASK_MASK 0x800
+#define GRBM_PERFCOUNTER0_SELECT__CB_CLEAN_USER_DEFINED_MASK__SHIFT 0xb
+#define GRBM_PERFCOUNTER0_SELECT__VGT_BUSY_USER_DEFINED_MASK_MASK 0x1000
+#define GRBM_PERFCOUNTER0_SELECT__VGT_BUSY_USER_DEFINED_MASK__SHIFT 0xc
+#define GRBM_PERFCOUNTER0_SELECT__TA_BUSY_USER_DEFINED_MASK_MASK 0x2000
+#define GRBM_PERFCOUNTER0_SELECT__TA_BUSY_USER_DEFINED_MASK__SHIFT 0xd
+#define GRBM_PERFCOUNTER0_SELECT__SX_BUSY_USER_DEFINED_MASK_MASK 0x4000
+#define GRBM_PERFCOUNTER0_SELECT__SX_BUSY_USER_DEFINED_MASK__SHIFT 0xe
+#define GRBM_PERFCOUNTER0_SELECT__SPI_BUSY_USER_DEFINED_MASK_MASK 0x10000
+#define GRBM_PERFCOUNTER0_SELECT__SPI_BUSY_USER_DEFINED_MASK__SHIFT 0x10
+#define GRBM_PERFCOUNTER0_SELECT__SC_BUSY_USER_DEFINED_MASK_MASK 0x20000
+#define GRBM_PERFCOUNTER0_SELECT__SC_BUSY_USER_DEFINED_MASK__SHIFT 0x11
+#define GRBM_PERFCOUNTER0_SELECT__PA_BUSY_USER_DEFINED_MASK_MASK 0x40000
+#define GRBM_PERFCOUNTER0_SELECT__PA_BUSY_USER_DEFINED_MASK__SHIFT 0x12
+#define GRBM_PERFCOUNTER0_SELECT__GRBM_BUSY_USER_DEFINED_MASK_MASK 0x80000
+#define GRBM_PERFCOUNTER0_SELECT__GRBM_BUSY_USER_DEFINED_MASK__SHIFT 0x13
+#define GRBM_PERFCOUNTER0_SELECT__DB_BUSY_USER_DEFINED_MASK_MASK 0x100000
+#define GRBM_PERFCOUNTER0_SELECT__DB_BUSY_USER_DEFINED_MASK__SHIFT 0x14
+#define GRBM_PERFCOUNTER0_SELECT__CB_BUSY_USER_DEFINED_MASK_MASK 0x200000
+#define GRBM_PERFCOUNTER0_SELECT__CB_BUSY_USER_DEFINED_MASK__SHIFT 0x15
+#define GRBM_PERFCOUNTER0_SELECT__CP_BUSY_USER_DEFINED_MASK_MASK 0x400000
+#define GRBM_PERFCOUNTER0_SELECT__CP_BUSY_USER_DEFINED_MASK__SHIFT 0x16
+#define GRBM_PERFCOUNTER0_SELECT__IA_BUSY_USER_DEFINED_MASK_MASK 0x800000
+#define GRBM_PERFCOUNTER0_SELECT__IA_BUSY_USER_DEFINED_MASK__SHIFT 0x17
+#define GRBM_PERFCOUNTER0_SELECT__GDS_BUSY_USER_DEFINED_MASK_MASK 0x1000000
+#define GRBM_PERFCOUNTER0_SELECT__GDS_BUSY_USER_DEFINED_MASK__SHIFT 0x18
+#define GRBM_PERFCOUNTER0_SELECT__BCI_BUSY_USER_DEFINED_MASK_MASK 0x2000000
+#define GRBM_PERFCOUNTER0_SELECT__BCI_BUSY_USER_DEFINED_MASK__SHIFT 0x19
+#define GRBM_PERFCOUNTER0_SELECT__RLC_BUSY_USER_DEFINED_MASK_MASK 0x4000000
+#define GRBM_PERFCOUNTER0_SELECT__RLC_BUSY_USER_DEFINED_MASK__SHIFT 0x1a
+#define GRBM_PERFCOUNTER0_SELECT__TC_BUSY_USER_DEFINED_MASK_MASK 0x8000000
+#define GRBM_PERFCOUNTER0_SELECT__TC_BUSY_USER_DEFINED_MASK__SHIFT 0x1b
+#define GRBM_PERFCOUNTER0_SELECT__WD_BUSY_USER_DEFINED_MASK_MASK 0x10000000
+#define GRBM_PERFCOUNTER0_SELECT__WD_BUSY_USER_DEFINED_MASK__SHIFT 0x1c
+#define GRBM_PERFCOUNTER1_SELECT__PERF_SEL_MASK 0x3f
+#define GRBM_PERFCOUNTER1_SELECT__PERF_SEL__SHIFT 0x0
+#define GRBM_PERFCOUNTER1_SELECT__DB_CLEAN_USER_DEFINED_MASK_MASK 0x400
+#define GRBM_PERFCOUNTER1_SELECT__DB_CLEAN_USER_DEFINED_MASK__SHIFT 0xa
+#define GRBM_PERFCOUNTER1_SELECT__CB_CLEAN_USER_DEFINED_MASK_MASK 0x800
+#define GRBM_PERFCOUNTER1_SELECT__CB_CLEAN_USER_DEFINED_MASK__SHIFT 0xb
+#define GRBM_PERFCOUNTER1_SELECT__VGT_BUSY_USER_DEFINED_MASK_MASK 0x1000
+#define GRBM_PERFCOUNTER1_SELECT__VGT_BUSY_USER_DEFINED_MASK__SHIFT 0xc
+#define GRBM_PERFCOUNTER1_SELECT__TA_BUSY_USER_DEFINED_MASK_MASK 0x2000
+#define GRBM_PERFCOUNTER1_SELECT__TA_BUSY_USER_DEFINED_MASK__SHIFT 0xd
+#define GRBM_PERFCOUNTER1_SELECT__SX_BUSY_USER_DEFINED_MASK_MASK 0x4000
+#define GRBM_PERFCOUNTER1_SELECT__SX_BUSY_USER_DEFINED_MASK__SHIFT 0xe
+#define GRBM_PERFCOUNTER1_SELECT__SPI_BUSY_USER_DEFINED_MASK_MASK 0x10000
+#define GRBM_PERFCOUNTER1_SELECT__SPI_BUSY_USER_DEFINED_MASK__SHIFT 0x10
+#define GRBM_PERFCOUNTER1_SELECT__SC_BUSY_USER_DEFINED_MASK_MASK 0x20000
+#define GRBM_PERFCOUNTER1_SELECT__SC_BUSY_USER_DEFINED_MASK__SHIFT 0x11
+#define GRBM_PERFCOUNTER1_SELECT__PA_BUSY_USER_DEFINED_MASK_MASK 0x40000
+#define GRBM_PERFCOUNTER1_SELECT__PA_BUSY_USER_DEFINED_MASK__SHIFT 0x12
+#define GRBM_PERFCOUNTER1_SELECT__GRBM_BUSY_USER_DEFINED_MASK_MASK 0x80000
+#define GRBM_PERFCOUNTER1_SELECT__GRBM_BUSY_USER_DEFINED_MASK__SHIFT 0x13
+#define GRBM_PERFCOUNTER1_SELECT__DB_BUSY_USER_DEFINED_MASK_MASK 0x100000
+#define GRBM_PERFCOUNTER1_SELECT__DB_BUSY_USER_DEFINED_MASK__SHIFT 0x14
+#define GRBM_PERFCOUNTER1_SELECT__CB_BUSY_USER_DEFINED_MASK_MASK 0x200000
+#define GRBM_PERFCOUNTER1_SELECT__CB_BUSY_USER_DEFINED_MASK__SHIFT 0x15
+#define GRBM_PERFCOUNTER1_SELECT__CP_BUSY_USER_DEFINED_MASK_MASK 0x400000
+#define GRBM_PERFCOUNTER1_SELECT__CP_BUSY_USER_DEFINED_MASK__SHIFT 0x16
+#define GRBM_PERFCOUNTER1_SELECT__IA_BUSY_USER_DEFINED_MASK_MASK 0x800000
+#define GRBM_PERFCOUNTER1_SELECT__IA_BUSY_USER_DEFINED_MASK__SHIFT 0x17
+#define GRBM_PERFCOUNTER1_SELECT__GDS_BUSY_USER_DEFINED_MASK_MASK 0x1000000
+#define GRBM_PERFCOUNTER1_SELECT__GDS_BUSY_USER_DEFINED_MASK__SHIFT 0x18
+#define GRBM_PERFCOUNTER1_SELECT__BCI_BUSY_USER_DEFINED_MASK_MASK 0x2000000
+#define GRBM_PERFCOUNTER1_SELECT__BCI_BUSY_USER_DEFINED_MASK__SHIFT 0x19
+#define GRBM_PERFCOUNTER1_SELECT__RLC_BUSY_USER_DEFINED_MASK_MASK 0x4000000
+#define GRBM_PERFCOUNTER1_SELECT__RLC_BUSY_USER_DEFINED_MASK__SHIFT 0x1a
+#define GRBM_PERFCOUNTER1_SELECT__TC_BUSY_USER_DEFINED_MASK_MASK 0x8000000
+#define GRBM_PERFCOUNTER1_SELECT__TC_BUSY_USER_DEFINED_MASK__SHIFT 0x1b
+#define GRBM_PERFCOUNTER1_SELECT__WD_BUSY_USER_DEFINED_MASK_MASK 0x10000000
+#define GRBM_PERFCOUNTER1_SELECT__WD_BUSY_USER_DEFINED_MASK__SHIFT 0x1c
+#define GRBM_SE0_PERFCOUNTER_SELECT__PERF_SEL_MASK 0x3f
+#define GRBM_SE0_PERFCOUNTER_SELECT__PERF_SEL__SHIFT 0x0
+#define GRBM_SE0_PERFCOUNTER_SELECT__DB_CLEAN_USER_DEFINED_MASK_MASK 0x400
+#define GRBM_SE0_PERFCOUNTER_SELECT__DB_CLEAN_USER_DEFINED_MASK__SHIFT 0xa
+#define GRBM_SE0_PERFCOUNTER_SELECT__CB_CLEAN_USER_DEFINED_MASK_MASK 0x800
+#define GRBM_SE0_PERFCOUNTER_SELECT__CB_CLEAN_USER_DEFINED_MASK__SHIFT 0xb
+#define GRBM_SE0_PERFCOUNTER_SELECT__TA_BUSY_USER_DEFINED_MASK_MASK 0x1000
+#define GRBM_SE0_PERFCOUNTER_SELECT__TA_BUSY_USER_DEFINED_MASK__SHIFT 0xc
+#define GRBM_SE0_PERFCOUNTER_SELECT__SX_BUSY_USER_DEFINED_MASK_MASK 0x2000
+#define GRBM_SE0_PERFCOUNTER_SELECT__SX_BUSY_USER_DEFINED_MASK__SHIFT 0xd
+#define GRBM_SE0_PERFCOUNTER_SELECT__SPI_BUSY_USER_DEFINED_MASK_MASK 0x8000
+#define GRBM_SE0_PERFCOUNTER_SELECT__SPI_BUSY_USER_DEFINED_MASK__SHIFT 0xf
+#define GRBM_SE0_PERFCOUNTER_SELECT__SC_BUSY_USER_DEFINED_MASK_MASK 0x10000
+#define GRBM_SE0_PERFCOUNTER_SELECT__SC_BUSY_USER_DEFINED_MASK__SHIFT 0x10
+#define GRBM_SE0_PERFCOUNTER_SELECT__DB_BUSY_USER_DEFINED_MASK_MASK 0x20000
+#define GRBM_SE0_PERFCOUNTER_SELECT__DB_BUSY_USER_DEFINED_MASK__SHIFT 0x11
+#define GRBM_SE0_PERFCOUNTER_SELECT__CB_BUSY_USER_DEFINED_MASK_MASK 0x40000
+#define GRBM_SE0_PERFCOUNTER_SELECT__CB_BUSY_USER_DEFINED_MASK__SHIFT 0x12
+#define GRBM_SE0_PERFCOUNTER_SELECT__VGT_BUSY_USER_DEFINED_MASK_MASK 0x80000
+#define GRBM_SE0_PERFCOUNTER_SELECT__VGT_BUSY_USER_DEFINED_MASK__SHIFT 0x13
+#define GRBM_SE0_PERFCOUNTER_SELECT__PA_BUSY_USER_DEFINED_MASK_MASK 0x100000
+#define GRBM_SE0_PERFCOUNTER_SELECT__PA_BUSY_USER_DEFINED_MASK__SHIFT 0x14
+#define GRBM_SE0_PERFCOUNTER_SELECT__BCI_BUSY_USER_DEFINED_MASK_MASK 0x200000
+#define GRBM_SE0_PERFCOUNTER_SELECT__BCI_BUSY_USER_DEFINED_MASK__SHIFT 0x15
+#define GRBM_SE1_PERFCOUNTER_SELECT__PERF_SEL_MASK 0x3f
+#define GRBM_SE1_PERFCOUNTER_SELECT__PERF_SEL__SHIFT 0x0
+#define GRBM_SE1_PERFCOUNTER_SELECT__DB_CLEAN_USER_DEFINED_MASK_MASK 0x400
+#define GRBM_SE1_PERFCOUNTER_SELECT__DB_CLEAN_USER_DEFINED_MASK__SHIFT 0xa
+#define GRBM_SE1_PERFCOUNTER_SELECT__CB_CLEAN_USER_DEFINED_MASK_MASK 0x800
+#define GRBM_SE1_PERFCOUNTER_SELECT__CB_CLEAN_USER_DEFINED_MASK__SHIFT 0xb
+#define GRBM_SE1_PERFCOUNTER_SELECT__TA_BUSY_USER_DEFINED_MASK_MASK 0x1000
+#define GRBM_SE1_PERFCOUNTER_SELECT__TA_BUSY_USER_DEFINED_MASK__SHIFT 0xc
+#define GRBM_SE1_PERFCOUNTER_SELECT__SX_BUSY_USER_DEFINED_MASK_MASK 0x2000
+#define GRBM_SE1_PERFCOUNTER_SELECT__SX_BUSY_USER_DEFINED_MASK__SHIFT 0xd
+#define GRBM_SE1_PERFCOUNTER_SELECT__SPI_BUSY_USER_DEFINED_MASK_MASK 0x8000
+#define GRBM_SE1_PERFCOUNTER_SELECT__SPI_BUSY_USER_DEFINED_MASK__SHIFT 0xf
+#define GRBM_SE1_PERFCOUNTER_SELECT__SC_BUSY_USER_DEFINED_MASK_MASK 0x10000
+#define GRBM_SE1_PERFCOUNTER_SELECT__SC_BUSY_USER_DEFINED_MASK__SHIFT 0x10
+#define GRBM_SE1_PERFCOUNTER_SELECT__DB_BUSY_USER_DEFINED_MASK_MASK 0x20000
+#define GRBM_SE1_PERFCOUNTER_SELECT__DB_BUSY_USER_DEFINED_MASK__SHIFT 0x11
+#define GRBM_SE1_PERFCOUNTER_SELECT__CB_BUSY_USER_DEFINED_MASK_MASK 0x40000
+#define GRBM_SE1_PERFCOUNTER_SELECT__CB_BUSY_USER_DEFINED_MASK__SHIFT 0x12
+#define GRBM_SE1_PERFCOUNTER_SELECT__VGT_BUSY_USER_DEFINED_MASK_MASK 0x80000
+#define GRBM_SE1_PERFCOUNTER_SELECT__VGT_BUSY_USER_DEFINED_MASK__SHIFT 0x13
+#define GRBM_SE1_PERFCOUNTER_SELECT__PA_BUSY_USER_DEFINED_MASK_MASK 0x100000
+#define GRBM_SE1_PERFCOUNTER_SELECT__PA_BUSY_USER_DEFINED_MASK__SHIFT 0x14
+#define GRBM_SE1_PERFCOUNTER_SELECT__BCI_BUSY_USER_DEFINED_MASK_MASK 0x200000
+#define GRBM_SE1_PERFCOUNTER_SELECT__BCI_BUSY_USER_DEFINED_MASK__SHIFT 0x15
+#define GRBM_SE2_PERFCOUNTER_SELECT__PERF_SEL_MASK 0x3f
+#define GRBM_SE2_PERFCOUNTER_SELECT__PERF_SEL__SHIFT 0x0
+#define GRBM_SE2_PERFCOUNTER_SELECT__DB_CLEAN_USER_DEFINED_MASK_MASK 0x400
+#define GRBM_SE2_PERFCOUNTER_SELECT__DB_CLEAN_USER_DEFINED_MASK__SHIFT 0xa
+#define GRBM_SE2_PERFCOUNTER_SELECT__CB_CLEAN_USER_DEFINED_MASK_MASK 0x800
+#define GRBM_SE2_PERFCOUNTER_SELECT__CB_CLEAN_USER_DEFINED_MASK__SHIFT 0xb
+#define GRBM_SE2_PERFCOUNTER_SELECT__TA_BUSY_USER_DEFINED_MASK_MASK 0x1000
+#define GRBM_SE2_PERFCOUNTER_SELECT__TA_BUSY_USER_DEFINED_MASK__SHIFT 0xc
+#define GRBM_SE2_PERFCOUNTER_SELECT__SX_BUSY_USER_DEFINED_MASK_MASK 0x2000
+#define GRBM_SE2_PERFCOUNTER_SELECT__SX_BUSY_USER_DEFINED_MASK__SHIFT 0xd
+#define GRBM_SE2_PERFCOUNTER_SELECT__SPI_BUSY_USER_DEFINED_MASK_MASK 0x8000
+#define GRBM_SE2_PERFCOUNTER_SELECT__SPI_BUSY_USER_DEFINED_MASK__SHIFT 0xf
+#define GRBM_SE2_PERFCOUNTER_SELECT__SC_BUSY_USER_DEFINED_MASK_MASK 0x10000
+#define GRBM_SE2_PERFCOUNTER_SELECT__SC_BUSY_USER_DEFINED_MASK__SHIFT 0x10
+#define GRBM_SE2_PERFCOUNTER_SELECT__DB_BUSY_USER_DEFINED_MASK_MASK 0x20000
+#define GRBM_SE2_PERFCOUNTER_SELECT__DB_BUSY_USER_DEFINED_MASK__SHIFT 0x11
+#define GRBM_SE2_PERFCOUNTER_SELECT__CB_BUSY_USER_DEFINED_MASK_MASK 0x40000
+#define GRBM_SE2_PERFCOUNTER_SELECT__CB_BUSY_USER_DEFINED_MASK__SHIFT 0x12
+#define GRBM_SE2_PERFCOUNTER_SELECT__VGT_BUSY_USER_DEFINED_MASK_MASK 0x80000
+#define GRBM_SE2_PERFCOUNTER_SELECT__VGT_BUSY_USER_DEFINED_MASK__SHIFT 0x13
+#define GRBM_SE2_PERFCOUNTER_SELECT__PA_BUSY_USER_DEFINED_MASK_MASK 0x100000
+#define GRBM_SE2_PERFCOUNTER_SELECT__PA_BUSY_USER_DEFINED_MASK__SHIFT 0x14
+#define GRBM_SE2_PERFCOUNTER_SELECT__BCI_BUSY_USER_DEFINED_MASK_MASK 0x200000
+#define GRBM_SE2_PERFCOUNTER_SELECT__BCI_BUSY_USER_DEFINED_MASK__SHIFT 0x15
+#define GRBM_SE3_PERFCOUNTER_SELECT__PERF_SEL_MASK 0x3f
+#define GRBM_SE3_PERFCOUNTER_SELECT__PERF_SEL__SHIFT 0x0
+#define GRBM_SE3_PERFCOUNTER_SELECT__DB_CLEAN_USER_DEFINED_MASK_MASK 0x400
+#define GRBM_SE3_PERFCOUNTER_SELECT__DB_CLEAN_USER_DEFINED_MASK__SHIFT 0xa
+#define GRBM_SE3_PERFCOUNTER_SELECT__CB_CLEAN_USER_DEFINED_MASK_MASK 0x800
+#define GRBM_SE3_PERFCOUNTER_SELECT__CB_CLEAN_USER_DEFINED_MASK__SHIFT 0xb
+#define GRBM_SE3_PERFCOUNTER_SELECT__TA_BUSY_USER_DEFINED_MASK_MASK 0x1000
+#define GRBM_SE3_PERFCOUNTER_SELECT__TA_BUSY_USER_DEFINED_MASK__SHIFT 0xc
+#define GRBM_SE3_PERFCOUNTER_SELECT__SX_BUSY_USER_DEFINED_MASK_MASK 0x2000
+#define GRBM_SE3_PERFCOUNTER_SELECT__SX_BUSY_USER_DEFINED_MASK__SHIFT 0xd
+#define GRBM_SE3_PERFCOUNTER_SELECT__SPI_BUSY_USER_DEFINED_MASK_MASK 0x8000
+#define GRBM_SE3_PERFCOUNTER_SELECT__SPI_BUSY_USER_DEFINED_MASK__SHIFT 0xf
+#define GRBM_SE3_PERFCOUNTER_SELECT__SC_BUSY_USER_DEFINED_MASK_MASK 0x10000
+#define GRBM_SE3_PERFCOUNTER_SELECT__SC_BUSY_USER_DEFINED_MASK__SHIFT 0x10
+#define GRBM_SE3_PERFCOUNTER_SELECT__DB_BUSY_USER_DEFINED_MASK_MASK 0x20000
+#define GRBM_SE3_PERFCOUNTER_SELECT__DB_BUSY_USER_DEFINED_MASK__SHIFT 0x11
+#define GRBM_SE3_PERFCOUNTER_SELECT__CB_BUSY_USER_DEFINED_MASK_MASK 0x40000
+#define GRBM_SE3_PERFCOUNTER_SELECT__CB_BUSY_USER_DEFINED_MASK__SHIFT 0x12
+#define GRBM_SE3_PERFCOUNTER_SELECT__VGT_BUSY_USER_DEFINED_MASK_MASK 0x80000
+#define GRBM_SE3_PERFCOUNTER_SELECT__VGT_BUSY_USER_DEFINED_MASK__SHIFT 0x13
+#define GRBM_SE3_PERFCOUNTER_SELECT__PA_BUSY_USER_DEFINED_MASK_MASK 0x100000
+#define GRBM_SE3_PERFCOUNTER_SELECT__PA_BUSY_USER_DEFINED_MASK__SHIFT 0x14
+#define GRBM_SE3_PERFCOUNTER_SELECT__BCI_BUSY_USER_DEFINED_MASK_MASK 0x200000
+#define GRBM_SE3_PERFCOUNTER_SELECT__BCI_BUSY_USER_DEFINED_MASK__SHIFT 0x15
+#define GRBM_PERFCOUNTER0_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define GRBM_PERFCOUNTER0_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define GRBM_PERFCOUNTER0_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define GRBM_PERFCOUNTER0_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define GRBM_PERFCOUNTER1_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define GRBM_PERFCOUNTER1_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define GRBM_PERFCOUNTER1_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define GRBM_PERFCOUNTER1_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define GRBM_SE0_PERFCOUNTER_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define GRBM_SE0_PERFCOUNTER_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define GRBM_SE0_PERFCOUNTER_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define GRBM_SE0_PERFCOUNTER_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define GRBM_SE1_PERFCOUNTER_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define GRBM_SE1_PERFCOUNTER_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define GRBM_SE1_PERFCOUNTER_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define GRBM_SE1_PERFCOUNTER_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define GRBM_SE2_PERFCOUNTER_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define GRBM_SE2_PERFCOUNTER_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define GRBM_SE2_PERFCOUNTER_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define GRBM_SE2_PERFCOUNTER_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define GRBM_SE3_PERFCOUNTER_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define GRBM_SE3_PERFCOUNTER_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define GRBM_SE3_PERFCOUNTER_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define GRBM_SE3_PERFCOUNTER_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define GRBM_SCRATCH_REG0__SCRATCH_REG0_MASK 0xffffffff
+#define GRBM_SCRATCH_REG0__SCRATCH_REG0__SHIFT 0x0
+#define GRBM_SCRATCH_REG1__SCRATCH_REG1_MASK 0xffffffff
+#define GRBM_SCRATCH_REG1__SCRATCH_REG1__SHIFT 0x0
+#define GRBM_SCRATCH_REG2__SCRATCH_REG2_MASK 0xffffffff
+#define GRBM_SCRATCH_REG2__SCRATCH_REG2__SHIFT 0x0
+#define GRBM_SCRATCH_REG3__SCRATCH_REG3_MASK 0xffffffff
+#define GRBM_SCRATCH_REG3__SCRATCH_REG3__SHIFT 0x0
+#define GRBM_SCRATCH_REG4__SCRATCH_REG4_MASK 0xffffffff
+#define GRBM_SCRATCH_REG4__SCRATCH_REG4__SHIFT 0x0
+#define GRBM_SCRATCH_REG5__SCRATCH_REG5_MASK 0xffffffff
+#define GRBM_SCRATCH_REG5__SCRATCH_REG5__SHIFT 0x0
+#define GRBM_SCRATCH_REG6__SCRATCH_REG6_MASK 0xffffffff
+#define GRBM_SCRATCH_REG6__SCRATCH_REG6__SHIFT 0x0
+#define GRBM_SCRATCH_REG7__SCRATCH_REG7_MASK 0xffffffff
+#define GRBM_SCRATCH_REG7__SCRATCH_REG7__SHIFT 0x0
+#define DEBUG_INDEX__DEBUG_INDEX_MASK 0x3ffff
+#define DEBUG_INDEX__DEBUG_INDEX__SHIFT 0x0
+#define DEBUG_DATA__DEBUG_DATA_MASK 0xffffffff
+#define DEBUG_DATA__DEBUG_DATA__SHIFT 0x0
+#define GRBM_NOWHERE__DATA_MASK 0xffffffff
+#define GRBM_NOWHERE__DATA__SHIFT 0x0
+#define PA_CL_VPORT_XSCALE__VPORT_XSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_XSCALE__VPORT_XSCALE__SHIFT 0x0
+#define PA_CL_VPORT_XOFFSET__VPORT_XOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_XOFFSET__VPORT_XOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_YSCALE__VPORT_YSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_YSCALE__VPORT_YSCALE__SHIFT 0x0
+#define PA_CL_VPORT_YOFFSET__VPORT_YOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_YOFFSET__VPORT_YOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_ZSCALE__VPORT_ZSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_ZSCALE__VPORT_ZSCALE__SHIFT 0x0
+#define PA_CL_VPORT_ZOFFSET__VPORT_ZOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_ZOFFSET__VPORT_ZOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_XSCALE_1__VPORT_XSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_XSCALE_1__VPORT_XSCALE__SHIFT 0x0
+#define PA_CL_VPORT_XSCALE_2__VPORT_XSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_XSCALE_2__VPORT_XSCALE__SHIFT 0x0
+#define PA_CL_VPORT_XSCALE_3__VPORT_XSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_XSCALE_3__VPORT_XSCALE__SHIFT 0x0
+#define PA_CL_VPORT_XSCALE_4__VPORT_XSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_XSCALE_4__VPORT_XSCALE__SHIFT 0x0
+#define PA_CL_VPORT_XSCALE_5__VPORT_XSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_XSCALE_5__VPORT_XSCALE__SHIFT 0x0
+#define PA_CL_VPORT_XSCALE_6__VPORT_XSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_XSCALE_6__VPORT_XSCALE__SHIFT 0x0
+#define PA_CL_VPORT_XSCALE_7__VPORT_XSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_XSCALE_7__VPORT_XSCALE__SHIFT 0x0
+#define PA_CL_VPORT_XSCALE_8__VPORT_XSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_XSCALE_8__VPORT_XSCALE__SHIFT 0x0
+#define PA_CL_VPORT_XSCALE_9__VPORT_XSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_XSCALE_9__VPORT_XSCALE__SHIFT 0x0
+#define PA_CL_VPORT_XSCALE_10__VPORT_XSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_XSCALE_10__VPORT_XSCALE__SHIFT 0x0
+#define PA_CL_VPORT_XSCALE_11__VPORT_XSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_XSCALE_11__VPORT_XSCALE__SHIFT 0x0
+#define PA_CL_VPORT_XSCALE_12__VPORT_XSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_XSCALE_12__VPORT_XSCALE__SHIFT 0x0
+#define PA_CL_VPORT_XSCALE_13__VPORT_XSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_XSCALE_13__VPORT_XSCALE__SHIFT 0x0
+#define PA_CL_VPORT_XSCALE_14__VPORT_XSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_XSCALE_14__VPORT_XSCALE__SHIFT 0x0
+#define PA_CL_VPORT_XSCALE_15__VPORT_XSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_XSCALE_15__VPORT_XSCALE__SHIFT 0x0
+#define PA_CL_VPORT_XOFFSET_1__VPORT_XOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_XOFFSET_1__VPORT_XOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_XOFFSET_2__VPORT_XOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_XOFFSET_2__VPORT_XOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_XOFFSET_3__VPORT_XOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_XOFFSET_3__VPORT_XOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_XOFFSET_4__VPORT_XOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_XOFFSET_4__VPORT_XOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_XOFFSET_5__VPORT_XOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_XOFFSET_5__VPORT_XOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_XOFFSET_6__VPORT_XOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_XOFFSET_6__VPORT_XOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_XOFFSET_7__VPORT_XOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_XOFFSET_7__VPORT_XOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_XOFFSET_8__VPORT_XOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_XOFFSET_8__VPORT_XOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_XOFFSET_9__VPORT_XOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_XOFFSET_9__VPORT_XOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_XOFFSET_10__VPORT_XOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_XOFFSET_10__VPORT_XOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_XOFFSET_11__VPORT_XOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_XOFFSET_11__VPORT_XOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_XOFFSET_12__VPORT_XOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_XOFFSET_12__VPORT_XOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_XOFFSET_13__VPORT_XOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_XOFFSET_13__VPORT_XOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_XOFFSET_14__VPORT_XOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_XOFFSET_14__VPORT_XOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_XOFFSET_15__VPORT_XOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_XOFFSET_15__VPORT_XOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_YSCALE_1__VPORT_YSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_YSCALE_1__VPORT_YSCALE__SHIFT 0x0
+#define PA_CL_VPORT_YSCALE_2__VPORT_YSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_YSCALE_2__VPORT_YSCALE__SHIFT 0x0
+#define PA_CL_VPORT_YSCALE_3__VPORT_YSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_YSCALE_3__VPORT_YSCALE__SHIFT 0x0
+#define PA_CL_VPORT_YSCALE_4__VPORT_YSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_YSCALE_4__VPORT_YSCALE__SHIFT 0x0
+#define PA_CL_VPORT_YSCALE_5__VPORT_YSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_YSCALE_5__VPORT_YSCALE__SHIFT 0x0
+#define PA_CL_VPORT_YSCALE_6__VPORT_YSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_YSCALE_6__VPORT_YSCALE__SHIFT 0x0
+#define PA_CL_VPORT_YSCALE_7__VPORT_YSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_YSCALE_7__VPORT_YSCALE__SHIFT 0x0
+#define PA_CL_VPORT_YSCALE_8__VPORT_YSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_YSCALE_8__VPORT_YSCALE__SHIFT 0x0
+#define PA_CL_VPORT_YSCALE_9__VPORT_YSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_YSCALE_9__VPORT_YSCALE__SHIFT 0x0
+#define PA_CL_VPORT_YSCALE_10__VPORT_YSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_YSCALE_10__VPORT_YSCALE__SHIFT 0x0
+#define PA_CL_VPORT_YSCALE_11__VPORT_YSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_YSCALE_11__VPORT_YSCALE__SHIFT 0x0
+#define PA_CL_VPORT_YSCALE_12__VPORT_YSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_YSCALE_12__VPORT_YSCALE__SHIFT 0x0
+#define PA_CL_VPORT_YSCALE_13__VPORT_YSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_YSCALE_13__VPORT_YSCALE__SHIFT 0x0
+#define PA_CL_VPORT_YSCALE_14__VPORT_YSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_YSCALE_14__VPORT_YSCALE__SHIFT 0x0
+#define PA_CL_VPORT_YSCALE_15__VPORT_YSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_YSCALE_15__VPORT_YSCALE__SHIFT 0x0
+#define PA_CL_VPORT_YOFFSET_1__VPORT_YOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_YOFFSET_1__VPORT_YOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_YOFFSET_2__VPORT_YOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_YOFFSET_2__VPORT_YOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_YOFFSET_3__VPORT_YOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_YOFFSET_3__VPORT_YOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_YOFFSET_4__VPORT_YOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_YOFFSET_4__VPORT_YOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_YOFFSET_5__VPORT_YOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_YOFFSET_5__VPORT_YOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_YOFFSET_6__VPORT_YOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_YOFFSET_6__VPORT_YOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_YOFFSET_7__VPORT_YOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_YOFFSET_7__VPORT_YOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_YOFFSET_8__VPORT_YOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_YOFFSET_8__VPORT_YOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_YOFFSET_9__VPORT_YOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_YOFFSET_9__VPORT_YOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_YOFFSET_10__VPORT_YOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_YOFFSET_10__VPORT_YOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_YOFFSET_11__VPORT_YOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_YOFFSET_11__VPORT_YOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_YOFFSET_12__VPORT_YOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_YOFFSET_12__VPORT_YOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_YOFFSET_13__VPORT_YOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_YOFFSET_13__VPORT_YOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_YOFFSET_14__VPORT_YOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_YOFFSET_14__VPORT_YOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_YOFFSET_15__VPORT_YOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_YOFFSET_15__VPORT_YOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_ZSCALE_1__VPORT_ZSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_ZSCALE_1__VPORT_ZSCALE__SHIFT 0x0
+#define PA_CL_VPORT_ZSCALE_2__VPORT_ZSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_ZSCALE_2__VPORT_ZSCALE__SHIFT 0x0
+#define PA_CL_VPORT_ZSCALE_3__VPORT_ZSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_ZSCALE_3__VPORT_ZSCALE__SHIFT 0x0
+#define PA_CL_VPORT_ZSCALE_4__VPORT_ZSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_ZSCALE_4__VPORT_ZSCALE__SHIFT 0x0
+#define PA_CL_VPORT_ZSCALE_5__VPORT_ZSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_ZSCALE_5__VPORT_ZSCALE__SHIFT 0x0
+#define PA_CL_VPORT_ZSCALE_6__VPORT_ZSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_ZSCALE_6__VPORT_ZSCALE__SHIFT 0x0
+#define PA_CL_VPORT_ZSCALE_7__VPORT_ZSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_ZSCALE_7__VPORT_ZSCALE__SHIFT 0x0
+#define PA_CL_VPORT_ZSCALE_8__VPORT_ZSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_ZSCALE_8__VPORT_ZSCALE__SHIFT 0x0
+#define PA_CL_VPORT_ZSCALE_9__VPORT_ZSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_ZSCALE_9__VPORT_ZSCALE__SHIFT 0x0
+#define PA_CL_VPORT_ZSCALE_10__VPORT_ZSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_ZSCALE_10__VPORT_ZSCALE__SHIFT 0x0
+#define PA_CL_VPORT_ZSCALE_11__VPORT_ZSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_ZSCALE_11__VPORT_ZSCALE__SHIFT 0x0
+#define PA_CL_VPORT_ZSCALE_12__VPORT_ZSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_ZSCALE_12__VPORT_ZSCALE__SHIFT 0x0
+#define PA_CL_VPORT_ZSCALE_13__VPORT_ZSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_ZSCALE_13__VPORT_ZSCALE__SHIFT 0x0
+#define PA_CL_VPORT_ZSCALE_14__VPORT_ZSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_ZSCALE_14__VPORT_ZSCALE__SHIFT 0x0
+#define PA_CL_VPORT_ZSCALE_15__VPORT_ZSCALE_MASK 0xffffffff
+#define PA_CL_VPORT_ZSCALE_15__VPORT_ZSCALE__SHIFT 0x0
+#define PA_CL_VPORT_ZOFFSET_1__VPORT_ZOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_ZOFFSET_1__VPORT_ZOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_ZOFFSET_2__VPORT_ZOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_ZOFFSET_2__VPORT_ZOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_ZOFFSET_3__VPORT_ZOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_ZOFFSET_3__VPORT_ZOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_ZOFFSET_4__VPORT_ZOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_ZOFFSET_4__VPORT_ZOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_ZOFFSET_5__VPORT_ZOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_ZOFFSET_5__VPORT_ZOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_ZOFFSET_6__VPORT_ZOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_ZOFFSET_6__VPORT_ZOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_ZOFFSET_7__VPORT_ZOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_ZOFFSET_7__VPORT_ZOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_ZOFFSET_8__VPORT_ZOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_ZOFFSET_8__VPORT_ZOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_ZOFFSET_9__VPORT_ZOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_ZOFFSET_9__VPORT_ZOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_ZOFFSET_10__VPORT_ZOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_ZOFFSET_10__VPORT_ZOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_ZOFFSET_11__VPORT_ZOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_ZOFFSET_11__VPORT_ZOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_ZOFFSET_12__VPORT_ZOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_ZOFFSET_12__VPORT_ZOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_ZOFFSET_13__VPORT_ZOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_ZOFFSET_13__VPORT_ZOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_ZOFFSET_14__VPORT_ZOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_ZOFFSET_14__VPORT_ZOFFSET__SHIFT 0x0
+#define PA_CL_VPORT_ZOFFSET_15__VPORT_ZOFFSET_MASK 0xffffffff
+#define PA_CL_VPORT_ZOFFSET_15__VPORT_ZOFFSET__SHIFT 0x0
+#define PA_CL_VTE_CNTL__VPORT_X_SCALE_ENA_MASK 0x1
+#define PA_CL_VTE_CNTL__VPORT_X_SCALE_ENA__SHIFT 0x0
+#define PA_CL_VTE_CNTL__VPORT_X_OFFSET_ENA_MASK 0x2
+#define PA_CL_VTE_CNTL__VPORT_X_OFFSET_ENA__SHIFT 0x1
+#define PA_CL_VTE_CNTL__VPORT_Y_SCALE_ENA_MASK 0x4
+#define PA_CL_VTE_CNTL__VPORT_Y_SCALE_ENA__SHIFT 0x2
+#define PA_CL_VTE_CNTL__VPORT_Y_OFFSET_ENA_MASK 0x8
+#define PA_CL_VTE_CNTL__VPORT_Y_OFFSET_ENA__SHIFT 0x3
+#define PA_CL_VTE_CNTL__VPORT_Z_SCALE_ENA_MASK 0x10
+#define PA_CL_VTE_CNTL__VPORT_Z_SCALE_ENA__SHIFT 0x4
+#define PA_CL_VTE_CNTL__VPORT_Z_OFFSET_ENA_MASK 0x20
+#define PA_CL_VTE_CNTL__VPORT_Z_OFFSET_ENA__SHIFT 0x5
+#define PA_CL_VTE_CNTL__VTX_XY_FMT_MASK 0x100
+#define PA_CL_VTE_CNTL__VTX_XY_FMT__SHIFT 0x8
+#define PA_CL_VTE_CNTL__VTX_Z_FMT_MASK 0x200
+#define PA_CL_VTE_CNTL__VTX_Z_FMT__SHIFT 0x9
+#define PA_CL_VTE_CNTL__VTX_W0_FMT_MASK 0x400
+#define PA_CL_VTE_CNTL__VTX_W0_FMT__SHIFT 0xa
+#define PA_CL_VTE_CNTL__PERFCOUNTER_REF_MASK 0x800
+#define PA_CL_VTE_CNTL__PERFCOUNTER_REF__SHIFT 0xb
+#define PA_CL_VS_OUT_CNTL__CLIP_DIST_ENA_0_MASK 0x1
+#define PA_CL_VS_OUT_CNTL__CLIP_DIST_ENA_0__SHIFT 0x0
+#define PA_CL_VS_OUT_CNTL__CLIP_DIST_ENA_1_MASK 0x2
+#define PA_CL_VS_OUT_CNTL__CLIP_DIST_ENA_1__SHIFT 0x1
+#define PA_CL_VS_OUT_CNTL__CLIP_DIST_ENA_2_MASK 0x4
+#define PA_CL_VS_OUT_CNTL__CLIP_DIST_ENA_2__SHIFT 0x2
+#define PA_CL_VS_OUT_CNTL__CLIP_DIST_ENA_3_MASK 0x8
+#define PA_CL_VS_OUT_CNTL__CLIP_DIST_ENA_3__SHIFT 0x3
+#define PA_CL_VS_OUT_CNTL__CLIP_DIST_ENA_4_MASK 0x10
+#define PA_CL_VS_OUT_CNTL__CLIP_DIST_ENA_4__SHIFT 0x4
+#define PA_CL_VS_OUT_CNTL__CLIP_DIST_ENA_5_MASK 0x20
+#define PA_CL_VS_OUT_CNTL__CLIP_DIST_ENA_5__SHIFT 0x5
+#define PA_CL_VS_OUT_CNTL__CLIP_DIST_ENA_6_MASK 0x40
+#define PA_CL_VS_OUT_CNTL__CLIP_DIST_ENA_6__SHIFT 0x6
+#define PA_CL_VS_OUT_CNTL__CLIP_DIST_ENA_7_MASK 0x80
+#define PA_CL_VS_OUT_CNTL__CLIP_DIST_ENA_7__SHIFT 0x7
+#define PA_CL_VS_OUT_CNTL__CULL_DIST_ENA_0_MASK 0x100
+#define PA_CL_VS_OUT_CNTL__CULL_DIST_ENA_0__SHIFT 0x8
+#define PA_CL_VS_OUT_CNTL__CULL_DIST_ENA_1_MASK 0x200
+#define PA_CL_VS_OUT_CNTL__CULL_DIST_ENA_1__SHIFT 0x9
+#define PA_CL_VS_OUT_CNTL__CULL_DIST_ENA_2_MASK 0x400
+#define PA_CL_VS_OUT_CNTL__CULL_DIST_ENA_2__SHIFT 0xa
+#define PA_CL_VS_OUT_CNTL__CULL_DIST_ENA_3_MASK 0x800
+#define PA_CL_VS_OUT_CNTL__CULL_DIST_ENA_3__SHIFT 0xb
+#define PA_CL_VS_OUT_CNTL__CULL_DIST_ENA_4_MASK 0x1000
+#define PA_CL_VS_OUT_CNTL__CULL_DIST_ENA_4__SHIFT 0xc
+#define PA_CL_VS_OUT_CNTL__CULL_DIST_ENA_5_MASK 0x2000
+#define PA_CL_VS_OUT_CNTL__CULL_DIST_ENA_5__SHIFT 0xd
+#define PA_CL_VS_OUT_CNTL__CULL_DIST_ENA_6_MASK 0x4000
+#define PA_CL_VS_OUT_CNTL__CULL_DIST_ENA_6__SHIFT 0xe
+#define PA_CL_VS_OUT_CNTL__CULL_DIST_ENA_7_MASK 0x8000
+#define PA_CL_VS_OUT_CNTL__CULL_DIST_ENA_7__SHIFT 0xf
+#define PA_CL_VS_OUT_CNTL__USE_VTX_POINT_SIZE_MASK 0x10000
+#define PA_CL_VS_OUT_CNTL__USE_VTX_POINT_SIZE__SHIFT 0x10
+#define PA_CL_VS_OUT_CNTL__USE_VTX_EDGE_FLAG_MASK 0x20000
+#define PA_CL_VS_OUT_CNTL__USE_VTX_EDGE_FLAG__SHIFT 0x11
+#define PA_CL_VS_OUT_CNTL__USE_VTX_RENDER_TARGET_INDX_MASK 0x40000
+#define PA_CL_VS_OUT_CNTL__USE_VTX_RENDER_TARGET_INDX__SHIFT 0x12
+#define PA_CL_VS_OUT_CNTL__USE_VTX_VIEWPORT_INDX_MASK 0x80000
+#define PA_CL_VS_OUT_CNTL__USE_VTX_VIEWPORT_INDX__SHIFT 0x13
+#define PA_CL_VS_OUT_CNTL__USE_VTX_KILL_FLAG_MASK 0x100000
+#define PA_CL_VS_OUT_CNTL__USE_VTX_KILL_FLAG__SHIFT 0x14
+#define PA_CL_VS_OUT_CNTL__VS_OUT_MISC_VEC_ENA_MASK 0x200000
+#define PA_CL_VS_OUT_CNTL__VS_OUT_MISC_VEC_ENA__SHIFT 0x15
+#define PA_CL_VS_OUT_CNTL__VS_OUT_CCDIST0_VEC_ENA_MASK 0x400000
+#define PA_CL_VS_OUT_CNTL__VS_OUT_CCDIST0_VEC_ENA__SHIFT 0x16
+#define PA_CL_VS_OUT_CNTL__VS_OUT_CCDIST1_VEC_ENA_MASK 0x800000
+#define PA_CL_VS_OUT_CNTL__VS_OUT_CCDIST1_VEC_ENA__SHIFT 0x17
+#define PA_CL_VS_OUT_CNTL__VS_OUT_MISC_SIDE_BUS_ENA_MASK 0x1000000
+#define PA_CL_VS_OUT_CNTL__VS_OUT_MISC_SIDE_BUS_ENA__SHIFT 0x18
+#define PA_CL_VS_OUT_CNTL__USE_VTX_GS_CUT_FLAG_MASK 0x2000000
+#define PA_CL_VS_OUT_CNTL__USE_VTX_GS_CUT_FLAG__SHIFT 0x19
+#define PA_CL_VS_OUT_CNTL__USE_VTX_LINE_WIDTH_MASK 0x4000000
+#define PA_CL_VS_OUT_CNTL__USE_VTX_LINE_WIDTH__SHIFT 0x1a
+#define PA_CL_NANINF_CNTL__VTE_XY_INF_DISCARD_MASK 0x1
+#define PA_CL_NANINF_CNTL__VTE_XY_INF_DISCARD__SHIFT 0x0
+#define PA_CL_NANINF_CNTL__VTE_Z_INF_DISCARD_MASK 0x2
+#define PA_CL_NANINF_CNTL__VTE_Z_INF_DISCARD__SHIFT 0x1
+#define PA_CL_NANINF_CNTL__VTE_W_INF_DISCARD_MASK 0x4
+#define PA_CL_NANINF_CNTL__VTE_W_INF_DISCARD__SHIFT 0x2
+#define PA_CL_NANINF_CNTL__VTE_0XNANINF_IS_0_MASK 0x8
+#define PA_CL_NANINF_CNTL__VTE_0XNANINF_IS_0__SHIFT 0x3
+#define PA_CL_NANINF_CNTL__VTE_XY_NAN_RETAIN_MASK 0x10
+#define PA_CL_NANINF_CNTL__VTE_XY_NAN_RETAIN__SHIFT 0x4
+#define PA_CL_NANINF_CNTL__VTE_Z_NAN_RETAIN_MASK 0x20
+#define PA_CL_NANINF_CNTL__VTE_Z_NAN_RETAIN__SHIFT 0x5
+#define PA_CL_NANINF_CNTL__VTE_W_NAN_RETAIN_MASK 0x40
+#define PA_CL_NANINF_CNTL__VTE_W_NAN_RETAIN__SHIFT 0x6
+#define PA_CL_NANINF_CNTL__VTE_W_RECIP_NAN_IS_0_MASK 0x80
+#define PA_CL_NANINF_CNTL__VTE_W_RECIP_NAN_IS_0__SHIFT 0x7
+#define PA_CL_NANINF_CNTL__VS_XY_NAN_TO_INF_MASK 0x100
+#define PA_CL_NANINF_CNTL__VS_XY_NAN_TO_INF__SHIFT 0x8
+#define PA_CL_NANINF_CNTL__VS_XY_INF_RETAIN_MASK 0x200
+#define PA_CL_NANINF_CNTL__VS_XY_INF_RETAIN__SHIFT 0x9
+#define PA_CL_NANINF_CNTL__VS_Z_NAN_TO_INF_MASK 0x400
+#define PA_CL_NANINF_CNTL__VS_Z_NAN_TO_INF__SHIFT 0xa
+#define PA_CL_NANINF_CNTL__VS_Z_INF_RETAIN_MASK 0x800
+#define PA_CL_NANINF_CNTL__VS_Z_INF_RETAIN__SHIFT 0xb
+#define PA_CL_NANINF_CNTL__VS_W_NAN_TO_INF_MASK 0x1000
+#define PA_CL_NANINF_CNTL__VS_W_NAN_TO_INF__SHIFT 0xc
+#define PA_CL_NANINF_CNTL__VS_W_INF_RETAIN_MASK 0x2000
+#define PA_CL_NANINF_CNTL__VS_W_INF_RETAIN__SHIFT 0xd
+#define PA_CL_NANINF_CNTL__VS_CLIP_DIST_INF_DISCARD_MASK 0x4000
+#define PA_CL_NANINF_CNTL__VS_CLIP_DIST_INF_DISCARD__SHIFT 0xe
+#define PA_CL_NANINF_CNTL__VTE_NO_OUTPUT_NEG_0_MASK 0x100000
+#define PA_CL_NANINF_CNTL__VTE_NO_OUTPUT_NEG_0__SHIFT 0x14
+#define PA_CL_CLIP_CNTL__UCP_ENA_0_MASK 0x1
+#define PA_CL_CLIP_CNTL__UCP_ENA_0__SHIFT 0x0
+#define PA_CL_CLIP_CNTL__UCP_ENA_1_MASK 0x2
+#define PA_CL_CLIP_CNTL__UCP_ENA_1__SHIFT 0x1
+#define PA_CL_CLIP_CNTL__UCP_ENA_2_MASK 0x4
+#define PA_CL_CLIP_CNTL__UCP_ENA_2__SHIFT 0x2
+#define PA_CL_CLIP_CNTL__UCP_ENA_3_MASK 0x8
+#define PA_CL_CLIP_CNTL__UCP_ENA_3__SHIFT 0x3
+#define PA_CL_CLIP_CNTL__UCP_ENA_4_MASK 0x10
+#define PA_CL_CLIP_CNTL__UCP_ENA_4__SHIFT 0x4
+#define PA_CL_CLIP_CNTL__UCP_ENA_5_MASK 0x20
+#define PA_CL_CLIP_CNTL__UCP_ENA_5__SHIFT 0x5
+#define PA_CL_CLIP_CNTL__PS_UCP_Y_SCALE_NEG_MASK 0x2000
+#define PA_CL_CLIP_CNTL__PS_UCP_Y_SCALE_NEG__SHIFT 0xd
+#define PA_CL_CLIP_CNTL__PS_UCP_MODE_MASK 0xc000
+#define PA_CL_CLIP_CNTL__PS_UCP_MODE__SHIFT 0xe
+#define PA_CL_CLIP_CNTL__CLIP_DISABLE_MASK 0x10000
+#define PA_CL_CLIP_CNTL__CLIP_DISABLE__SHIFT 0x10
+#define PA_CL_CLIP_CNTL__UCP_CULL_ONLY_ENA_MASK 0x20000
+#define PA_CL_CLIP_CNTL__UCP_CULL_ONLY_ENA__SHIFT 0x11
+#define PA_CL_CLIP_CNTL__BOUNDARY_EDGE_FLAG_ENA_MASK 0x40000
+#define PA_CL_CLIP_CNTL__BOUNDARY_EDGE_FLAG_ENA__SHIFT 0x12
+#define PA_CL_CLIP_CNTL__DX_CLIP_SPACE_DEF_MASK 0x80000
+#define PA_CL_CLIP_CNTL__DX_CLIP_SPACE_DEF__SHIFT 0x13
+#define PA_CL_CLIP_CNTL__DIS_CLIP_ERR_DETECT_MASK 0x100000
+#define PA_CL_CLIP_CNTL__DIS_CLIP_ERR_DETECT__SHIFT 0x14
+#define PA_CL_CLIP_CNTL__VTX_KILL_OR_MASK 0x200000
+#define PA_CL_CLIP_CNTL__VTX_KILL_OR__SHIFT 0x15
+#define PA_CL_CLIP_CNTL__DX_RASTERIZATION_KILL_MASK 0x400000
+#define PA_CL_CLIP_CNTL__DX_RASTERIZATION_KILL__SHIFT 0x16
+#define PA_CL_CLIP_CNTL__DX_LINEAR_ATTR_CLIP_ENA_MASK 0x1000000
+#define PA_CL_CLIP_CNTL__DX_LINEAR_ATTR_CLIP_ENA__SHIFT 0x18
+#define PA_CL_CLIP_CNTL__VTE_VPORT_PROVOKE_DISABLE_MASK 0x2000000
+#define PA_CL_CLIP_CNTL__VTE_VPORT_PROVOKE_DISABLE__SHIFT 0x19
+#define PA_CL_CLIP_CNTL__ZCLIP_NEAR_DISABLE_MASK 0x4000000
+#define PA_CL_CLIP_CNTL__ZCLIP_NEAR_DISABLE__SHIFT 0x1a
+#define PA_CL_CLIP_CNTL__ZCLIP_FAR_DISABLE_MASK 0x8000000
+#define PA_CL_CLIP_CNTL__ZCLIP_FAR_DISABLE__SHIFT 0x1b
+#define PA_CL_GB_VERT_CLIP_ADJ__DATA_REGISTER_MASK 0xffffffff
+#define PA_CL_GB_VERT_CLIP_ADJ__DATA_REGISTER__SHIFT 0x0
+#define PA_CL_GB_VERT_DISC_ADJ__DATA_REGISTER_MASK 0xffffffff
+#define PA_CL_GB_VERT_DISC_ADJ__DATA_REGISTER__SHIFT 0x0
+#define PA_CL_GB_HORZ_CLIP_ADJ__DATA_REGISTER_MASK 0xffffffff
+#define PA_CL_GB_HORZ_CLIP_ADJ__DATA_REGISTER__SHIFT 0x0
+#define PA_CL_GB_HORZ_DISC_ADJ__DATA_REGISTER_MASK 0xffffffff
+#define PA_CL_GB_HORZ_DISC_ADJ__DATA_REGISTER__SHIFT 0x0
+#define PA_CL_UCP_0_X__DATA_REGISTER_MASK 0xffffffff
+#define PA_CL_UCP_0_X__DATA_REGISTER__SHIFT 0x0
+#define PA_CL_UCP_0_Y__DATA_REGISTER_MASK 0xffffffff
+#define PA_CL_UCP_0_Y__DATA_REGISTER__SHIFT 0x0
+#define PA_CL_UCP_0_Z__DATA_REGISTER_MASK 0xffffffff
+#define PA_CL_UCP_0_Z__DATA_REGISTER__SHIFT 0x0
+#define PA_CL_UCP_0_W__DATA_REGISTER_MASK 0xffffffff
+#define PA_CL_UCP_0_W__DATA_REGISTER__SHIFT 0x0
+#define PA_CL_UCP_1_X__DATA_REGISTER_MASK 0xffffffff
+#define PA_CL_UCP_1_X__DATA_REGISTER__SHIFT 0x0
+#define PA_CL_UCP_1_Y__DATA_REGISTER_MASK 0xffffffff
+#define PA_CL_UCP_1_Y__DATA_REGISTER__SHIFT 0x0
+#define PA_CL_UCP_1_Z__DATA_REGISTER_MASK 0xffffffff
+#define PA_CL_UCP_1_Z__DATA_REGISTER__SHIFT 0x0
+#define PA_CL_UCP_1_W__DATA_REGISTER_MASK 0xffffffff
+#define PA_CL_UCP_1_W__DATA_REGISTER__SHIFT 0x0
+#define PA_CL_UCP_2_X__DATA_REGISTER_MASK 0xffffffff
+#define PA_CL_UCP_2_X__DATA_REGISTER__SHIFT 0x0
+#define PA_CL_UCP_2_Y__DATA_REGISTER_MASK 0xffffffff
+#define PA_CL_UCP_2_Y__DATA_REGISTER__SHIFT 0x0
+#define PA_CL_UCP_2_Z__DATA_REGISTER_MASK 0xffffffff
+#define PA_CL_UCP_2_Z__DATA_REGISTER__SHIFT 0x0
+#define PA_CL_UCP_2_W__DATA_REGISTER_MASK 0xffffffff
+#define PA_CL_UCP_2_W__DATA_REGISTER__SHIFT 0x0
+#define PA_CL_UCP_3_X__DATA_REGISTER_MASK 0xffffffff
+#define PA_CL_UCP_3_X__DATA_REGISTER__SHIFT 0x0
+#define PA_CL_UCP_3_Y__DATA_REGISTER_MASK 0xffffffff
+#define PA_CL_UCP_3_Y__DATA_REGISTER__SHIFT 0x0
+#define PA_CL_UCP_3_Z__DATA_REGISTER_MASK 0xffffffff
+#define PA_CL_UCP_3_Z__DATA_REGISTER__SHIFT 0x0
+#define PA_CL_UCP_3_W__DATA_REGISTER_MASK 0xffffffff
+#define PA_CL_UCP_3_W__DATA_REGISTER__SHIFT 0x0
+#define PA_CL_UCP_4_X__DATA_REGISTER_MASK 0xffffffff
+#define PA_CL_UCP_4_X__DATA_REGISTER__SHIFT 0x0
+#define PA_CL_UCP_4_Y__DATA_REGISTER_MASK 0xffffffff
+#define PA_CL_UCP_4_Y__DATA_REGISTER__SHIFT 0x0
+#define PA_CL_UCP_4_Z__DATA_REGISTER_MASK 0xffffffff
+#define PA_CL_UCP_4_Z__DATA_REGISTER__SHIFT 0x0
+#define PA_CL_UCP_4_W__DATA_REGISTER_MASK 0xffffffff
+#define PA_CL_UCP_4_W__DATA_REGISTER__SHIFT 0x0
+#define PA_CL_UCP_5_X__DATA_REGISTER_MASK 0xffffffff
+#define PA_CL_UCP_5_X__DATA_REGISTER__SHIFT 0x0
+#define PA_CL_UCP_5_Y__DATA_REGISTER_MASK 0xffffffff
+#define PA_CL_UCP_5_Y__DATA_REGISTER__SHIFT 0x0
+#define PA_CL_UCP_5_Z__DATA_REGISTER_MASK 0xffffffff
+#define PA_CL_UCP_5_Z__DATA_REGISTER__SHIFT 0x0
+#define PA_CL_UCP_5_W__DATA_REGISTER_MASK 0xffffffff
+#define PA_CL_UCP_5_W__DATA_REGISTER__SHIFT 0x0
+#define PA_CL_POINT_X_RAD__DATA_REGISTER_MASK 0xffffffff
+#define PA_CL_POINT_X_RAD__DATA_REGISTER__SHIFT 0x0
+#define PA_CL_POINT_Y_RAD__DATA_REGISTER_MASK 0xffffffff
+#define PA_CL_POINT_Y_RAD__DATA_REGISTER__SHIFT 0x0
+#define PA_CL_POINT_SIZE__DATA_REGISTER_MASK 0xffffffff
+#define PA_CL_POINT_SIZE__DATA_REGISTER__SHIFT 0x0
+#define PA_CL_POINT_CULL_RAD__DATA_REGISTER_MASK 0xffffffff
+#define PA_CL_POINT_CULL_RAD__DATA_REGISTER__SHIFT 0x0
+#define PA_CL_ENHANCE__CLIP_VTX_REORDER_ENA_MASK 0x1
+#define PA_CL_ENHANCE__CLIP_VTX_REORDER_ENA__SHIFT 0x0
+#define PA_CL_ENHANCE__NUM_CLIP_SEQ_MASK 0x6
+#define PA_CL_ENHANCE__NUM_CLIP_SEQ__SHIFT 0x1
+#define PA_CL_ENHANCE__CLIPPED_PRIM_SEQ_STALL_MASK 0x8
+#define PA_CL_ENHANCE__CLIPPED_PRIM_SEQ_STALL__SHIFT 0x3
+#define PA_CL_ENHANCE__VE_NAN_PROC_DISABLE_MASK 0x10
+#define PA_CL_ENHANCE__VE_NAN_PROC_DISABLE__SHIFT 0x4
+#define PA_CL_ENHANCE__XTRA_DEBUG_REG_SEL_MASK 0x20
+#define PA_CL_ENHANCE__XTRA_DEBUG_REG_SEL__SHIFT 0x5
+#define PA_CL_ENHANCE__ECO_SPARE3_MASK 0x10000000
+#define PA_CL_ENHANCE__ECO_SPARE3__SHIFT 0x1c
+#define PA_CL_ENHANCE__ECO_SPARE2_MASK 0x20000000
+#define PA_CL_ENHANCE__ECO_SPARE2__SHIFT 0x1d
+#define PA_CL_ENHANCE__ECO_SPARE1_MASK 0x40000000
+#define PA_CL_ENHANCE__ECO_SPARE1__SHIFT 0x1e
+#define PA_CL_ENHANCE__ECO_SPARE0_MASK 0x80000000
+#define PA_CL_ENHANCE__ECO_SPARE0__SHIFT 0x1f
+#define PA_CL_RESET_DEBUG__CL_TRIV_DISC_DISABLE_MASK 0x1
+#define PA_CL_RESET_DEBUG__CL_TRIV_DISC_DISABLE__SHIFT 0x0
+#define PA_SU_VTX_CNTL__PIX_CENTER_MASK 0x1
+#define PA_SU_VTX_CNTL__PIX_CENTER__SHIFT 0x0
+#define PA_SU_VTX_CNTL__ROUND_MODE_MASK 0x6
+#define PA_SU_VTX_CNTL__ROUND_MODE__SHIFT 0x1
+#define PA_SU_VTX_CNTL__QUANT_MODE_MASK 0x38
+#define PA_SU_VTX_CNTL__QUANT_MODE__SHIFT 0x3
+#define PA_SU_POINT_SIZE__HEIGHT_MASK 0xffff
+#define PA_SU_POINT_SIZE__HEIGHT__SHIFT 0x0
+#define PA_SU_POINT_SIZE__WIDTH_MASK 0xffff0000
+#define PA_SU_POINT_SIZE__WIDTH__SHIFT 0x10
+#define PA_SU_POINT_MINMAX__MIN_SIZE_MASK 0xffff
+#define PA_SU_POINT_MINMAX__MIN_SIZE__SHIFT 0x0
+#define PA_SU_POINT_MINMAX__MAX_SIZE_MASK 0xffff0000
+#define PA_SU_POINT_MINMAX__MAX_SIZE__SHIFT 0x10
+#define PA_SU_LINE_CNTL__WIDTH_MASK 0xffff
+#define PA_SU_LINE_CNTL__WIDTH__SHIFT 0x0
+#define PA_SU_LINE_STIPPLE_CNTL__LINE_STIPPLE_RESET_MASK 0x3
+#define PA_SU_LINE_STIPPLE_CNTL__LINE_STIPPLE_RESET__SHIFT 0x0
+#define PA_SU_LINE_STIPPLE_CNTL__EXPAND_FULL_LENGTH_MASK 0x4
+#define PA_SU_LINE_STIPPLE_CNTL__EXPAND_FULL_LENGTH__SHIFT 0x2
+#define PA_SU_LINE_STIPPLE_CNTL__FRACTIONAL_ACCUM_MASK 0x8
+#define PA_SU_LINE_STIPPLE_CNTL__FRACTIONAL_ACCUM__SHIFT 0x3
+#define PA_SU_LINE_STIPPLE_CNTL__DIAMOND_ADJUST_MASK 0x10
+#define PA_SU_LINE_STIPPLE_CNTL__DIAMOND_ADJUST__SHIFT 0x4
+#define PA_SU_LINE_STIPPLE_SCALE__LINE_STIPPLE_SCALE_MASK 0xffffffff
+#define PA_SU_LINE_STIPPLE_SCALE__LINE_STIPPLE_SCALE__SHIFT 0x0
+#define PA_SU_PRIM_FILTER_CNTL__TRIANGLE_FILTER_DISABLE_MASK 0x1
+#define PA_SU_PRIM_FILTER_CNTL__TRIANGLE_FILTER_DISABLE__SHIFT 0x0
+#define PA_SU_PRIM_FILTER_CNTL__LINE_FILTER_DISABLE_MASK 0x2
+#define PA_SU_PRIM_FILTER_CNTL__LINE_FILTER_DISABLE__SHIFT 0x1
+#define PA_SU_PRIM_FILTER_CNTL__POINT_FILTER_DISABLE_MASK 0x4
+#define PA_SU_PRIM_FILTER_CNTL__POINT_FILTER_DISABLE__SHIFT 0x2
+#define PA_SU_PRIM_FILTER_CNTL__RECTANGLE_FILTER_DISABLE_MASK 0x8
+#define PA_SU_PRIM_FILTER_CNTL__RECTANGLE_FILTER_DISABLE__SHIFT 0x3
+#define PA_SU_PRIM_FILTER_CNTL__TRIANGLE_EXPAND_ENA_MASK 0x10
+#define PA_SU_PRIM_FILTER_CNTL__TRIANGLE_EXPAND_ENA__SHIFT 0x4
+#define PA_SU_PRIM_FILTER_CNTL__LINE_EXPAND_ENA_MASK 0x20
+#define PA_SU_PRIM_FILTER_CNTL__LINE_EXPAND_ENA__SHIFT 0x5
+#define PA_SU_PRIM_FILTER_CNTL__POINT_EXPAND_ENA_MASK 0x40
+#define PA_SU_PRIM_FILTER_CNTL__POINT_EXPAND_ENA__SHIFT 0x6
+#define PA_SU_PRIM_FILTER_CNTL__RECTANGLE_EXPAND_ENA_MASK 0x80
+#define PA_SU_PRIM_FILTER_CNTL__RECTANGLE_EXPAND_ENA__SHIFT 0x7
+#define PA_SU_PRIM_FILTER_CNTL__PRIM_EXPAND_CONSTANT_MASK 0xff00
+#define PA_SU_PRIM_FILTER_CNTL__PRIM_EXPAND_CONSTANT__SHIFT 0x8
+#define PA_SU_PRIM_FILTER_CNTL__XMAX_RIGHT_EXCLUSION_MASK 0x40000000
+#define PA_SU_PRIM_FILTER_CNTL__XMAX_RIGHT_EXCLUSION__SHIFT 0x1e
+#define PA_SU_PRIM_FILTER_CNTL__YMAX_BOTTOM_EXCLUSION_MASK 0x80000000
+#define PA_SU_PRIM_FILTER_CNTL__YMAX_BOTTOM_EXCLUSION__SHIFT 0x1f
+#define PA_SU_SC_MODE_CNTL__CULL_FRONT_MASK 0x1
+#define PA_SU_SC_MODE_CNTL__CULL_FRONT__SHIFT 0x0
+#define PA_SU_SC_MODE_CNTL__CULL_BACK_MASK 0x2
+#define PA_SU_SC_MODE_CNTL__CULL_BACK__SHIFT 0x1
+#define PA_SU_SC_MODE_CNTL__FACE_MASK 0x4
+#define PA_SU_SC_MODE_CNTL__FACE__SHIFT 0x2
+#define PA_SU_SC_MODE_CNTL__POLY_MODE_MASK 0x18
+#define PA_SU_SC_MODE_CNTL__POLY_MODE__SHIFT 0x3
+#define PA_SU_SC_MODE_CNTL__POLYMODE_FRONT_PTYPE_MASK 0xe0
+#define PA_SU_SC_MODE_CNTL__POLYMODE_FRONT_PTYPE__SHIFT 0x5
+#define PA_SU_SC_MODE_CNTL__POLYMODE_BACK_PTYPE_MASK 0x700
+#define PA_SU_SC_MODE_CNTL__POLYMODE_BACK_PTYPE__SHIFT 0x8
+#define PA_SU_SC_MODE_CNTL__POLY_OFFSET_FRONT_ENABLE_MASK 0x800
+#define PA_SU_SC_MODE_CNTL__POLY_OFFSET_FRONT_ENABLE__SHIFT 0xb
+#define PA_SU_SC_MODE_CNTL__POLY_OFFSET_BACK_ENABLE_MASK 0x1000
+#define PA_SU_SC_MODE_CNTL__POLY_OFFSET_BACK_ENABLE__SHIFT 0xc
+#define PA_SU_SC_MODE_CNTL__POLY_OFFSET_PARA_ENABLE_MASK 0x2000
+#define PA_SU_SC_MODE_CNTL__POLY_OFFSET_PARA_ENABLE__SHIFT 0xd
+#define PA_SU_SC_MODE_CNTL__VTX_WINDOW_OFFSET_ENABLE_MASK 0x10000
+#define PA_SU_SC_MODE_CNTL__VTX_WINDOW_OFFSET_ENABLE__SHIFT 0x10
+#define PA_SU_SC_MODE_CNTL__PROVOKING_VTX_LAST_MASK 0x80000
+#define PA_SU_SC_MODE_CNTL__PROVOKING_VTX_LAST__SHIFT 0x13
+#define PA_SU_SC_MODE_CNTL__PERSP_CORR_DIS_MASK 0x100000
+#define PA_SU_SC_MODE_CNTL__PERSP_CORR_DIS__SHIFT 0x14
+#define PA_SU_SC_MODE_CNTL__MULTI_PRIM_IB_ENA_MASK 0x200000
+#define PA_SU_SC_MODE_CNTL__MULTI_PRIM_IB_ENA__SHIFT 0x15
+#define PA_SU_POLY_OFFSET_DB_FMT_CNTL__POLY_OFFSET_NEG_NUM_DB_BITS_MASK 0xff
+#define PA_SU_POLY_OFFSET_DB_FMT_CNTL__POLY_OFFSET_NEG_NUM_DB_BITS__SHIFT 0x0
+#define PA_SU_POLY_OFFSET_DB_FMT_CNTL__POLY_OFFSET_DB_IS_FLOAT_FMT_MASK 0x100
+#define PA_SU_POLY_OFFSET_DB_FMT_CNTL__POLY_OFFSET_DB_IS_FLOAT_FMT__SHIFT 0x8
+#define PA_SU_POLY_OFFSET_CLAMP__CLAMP_MASK 0xffffffff
+#define PA_SU_POLY_OFFSET_CLAMP__CLAMP__SHIFT 0x0
+#define PA_SU_POLY_OFFSET_FRONT_SCALE__SCALE_MASK 0xffffffff
+#define PA_SU_POLY_OFFSET_FRONT_SCALE__SCALE__SHIFT 0x0
+#define PA_SU_POLY_OFFSET_FRONT_OFFSET__OFFSET_MASK 0xffffffff
+#define PA_SU_POLY_OFFSET_FRONT_OFFSET__OFFSET__SHIFT 0x0
+#define PA_SU_POLY_OFFSET_BACK_SCALE__SCALE_MASK 0xffffffff
+#define PA_SU_POLY_OFFSET_BACK_SCALE__SCALE__SHIFT 0x0
+#define PA_SU_POLY_OFFSET_BACK_OFFSET__OFFSET_MASK 0xffffffff
+#define PA_SU_POLY_OFFSET_BACK_OFFSET__OFFSET__SHIFT 0x0
+#define PA_SU_HARDWARE_SCREEN_OFFSET__HW_SCREEN_OFFSET_X_MASK 0x1ff
+#define PA_SU_HARDWARE_SCREEN_OFFSET__HW_SCREEN_OFFSET_X__SHIFT 0x0
+#define PA_SU_HARDWARE_SCREEN_OFFSET__HW_SCREEN_OFFSET_Y_MASK 0x1ff0000
+#define PA_SU_HARDWARE_SCREEN_OFFSET__HW_SCREEN_OFFSET_Y__SHIFT 0x10
+#define PA_SU_LINE_STIPPLE_VALUE__LINE_STIPPLE_VALUE_MASK 0xffffff
+#define PA_SU_LINE_STIPPLE_VALUE__LINE_STIPPLE_VALUE__SHIFT 0x0
+#define PA_SU_PERFCOUNTER0_SELECT__PERF_SEL_MASK 0x3ff
+#define PA_SU_PERFCOUNTER0_SELECT__PERF_SEL__SHIFT 0x0
+#define PA_SU_PERFCOUNTER0_SELECT__PERF_SEL1_MASK 0xffc00
+#define PA_SU_PERFCOUNTER0_SELECT__PERF_SEL1__SHIFT 0xa
+#define PA_SU_PERFCOUNTER0_SELECT__CNTR_MODE_MASK 0xf00000
+#define PA_SU_PERFCOUNTER0_SELECT__CNTR_MODE__SHIFT 0x14
+#define PA_SU_PERFCOUNTER0_SELECT1__PERF_SEL2_MASK 0x3ff
+#define PA_SU_PERFCOUNTER0_SELECT1__PERF_SEL2__SHIFT 0x0
+#define PA_SU_PERFCOUNTER0_SELECT1__PERF_SEL3_MASK 0xffc00
+#define PA_SU_PERFCOUNTER0_SELECT1__PERF_SEL3__SHIFT 0xa
+#define PA_SU_PERFCOUNTER1_SELECT__PERF_SEL_MASK 0x3ff
+#define PA_SU_PERFCOUNTER1_SELECT__PERF_SEL__SHIFT 0x0
+#define PA_SU_PERFCOUNTER1_SELECT__PERF_SEL1_MASK 0xffc00
+#define PA_SU_PERFCOUNTER1_SELECT__PERF_SEL1__SHIFT 0xa
+#define PA_SU_PERFCOUNTER1_SELECT__CNTR_MODE_MASK 0xf00000
+#define PA_SU_PERFCOUNTER1_SELECT__CNTR_MODE__SHIFT 0x14
+#define PA_SU_PERFCOUNTER1_SELECT1__PERF_SEL2_MASK 0x3ff
+#define PA_SU_PERFCOUNTER1_SELECT1__PERF_SEL2__SHIFT 0x0
+#define PA_SU_PERFCOUNTER1_SELECT1__PERF_SEL3_MASK 0xffc00
+#define PA_SU_PERFCOUNTER1_SELECT1__PERF_SEL3__SHIFT 0xa
+#define PA_SU_PERFCOUNTER2_SELECT__PERF_SEL_MASK 0x3ff
+#define PA_SU_PERFCOUNTER2_SELECT__PERF_SEL__SHIFT 0x0
+#define PA_SU_PERFCOUNTER2_SELECT__CNTR_MODE_MASK 0xf00000
+#define PA_SU_PERFCOUNTER2_SELECT__CNTR_MODE__SHIFT 0x14
+#define PA_SU_PERFCOUNTER3_SELECT__PERF_SEL_MASK 0x3ff
+#define PA_SU_PERFCOUNTER3_SELECT__PERF_SEL__SHIFT 0x0
+#define PA_SU_PERFCOUNTER3_SELECT__CNTR_MODE_MASK 0xf00000
+#define PA_SU_PERFCOUNTER3_SELECT__CNTR_MODE__SHIFT 0x14
+#define PA_SU_PERFCOUNTER0_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define PA_SU_PERFCOUNTER0_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define PA_SU_PERFCOUNTER0_HI__PERFCOUNTER_HI_MASK 0xffff
+#define PA_SU_PERFCOUNTER0_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define PA_SU_PERFCOUNTER1_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define PA_SU_PERFCOUNTER1_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define PA_SU_PERFCOUNTER1_HI__PERFCOUNTER_HI_MASK 0xffff
+#define PA_SU_PERFCOUNTER1_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define PA_SU_PERFCOUNTER2_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define PA_SU_PERFCOUNTER2_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define PA_SU_PERFCOUNTER2_HI__PERFCOUNTER_HI_MASK 0xffff
+#define PA_SU_PERFCOUNTER2_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define PA_SU_PERFCOUNTER3_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define PA_SU_PERFCOUNTER3_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define PA_SU_PERFCOUNTER3_HI__PERFCOUNTER_HI_MASK 0xffff
+#define PA_SU_PERFCOUNTER3_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define PA_SC_AA_CONFIG__MSAA_NUM_SAMPLES_MASK 0x7
+#define PA_SC_AA_CONFIG__MSAA_NUM_SAMPLES__SHIFT 0x0
+#define PA_SC_AA_CONFIG__AA_MASK_CENTROID_DTMN_MASK 0x10
+#define PA_SC_AA_CONFIG__AA_MASK_CENTROID_DTMN__SHIFT 0x4
+#define PA_SC_AA_CONFIG__MAX_SAMPLE_DIST_MASK 0x1e000
+#define PA_SC_AA_CONFIG__MAX_SAMPLE_DIST__SHIFT 0xd
+#define PA_SC_AA_CONFIG__MSAA_EXPOSED_SAMPLES_MASK 0x700000
+#define PA_SC_AA_CONFIG__MSAA_EXPOSED_SAMPLES__SHIFT 0x14
+#define PA_SC_AA_CONFIG__DETAIL_TO_EXPOSED_MODE_MASK 0x3000000
+#define PA_SC_AA_CONFIG__DETAIL_TO_EXPOSED_MODE__SHIFT 0x18
+#define PA_SC_AA_MASK_X0Y0_X1Y0__AA_MASK_X0Y0_MASK 0xffff
+#define PA_SC_AA_MASK_X0Y0_X1Y0__AA_MASK_X0Y0__SHIFT 0x0
+#define PA_SC_AA_MASK_X0Y0_X1Y0__AA_MASK_X1Y0_MASK 0xffff0000
+#define PA_SC_AA_MASK_X0Y0_X1Y0__AA_MASK_X1Y0__SHIFT 0x10
+#define PA_SC_AA_MASK_X0Y1_X1Y1__AA_MASK_X0Y1_MASK 0xffff
+#define PA_SC_AA_MASK_X0Y1_X1Y1__AA_MASK_X0Y1__SHIFT 0x0
+#define PA_SC_AA_MASK_X0Y1_X1Y1__AA_MASK_X1Y1_MASK 0xffff0000
+#define PA_SC_AA_MASK_X0Y1_X1Y1__AA_MASK_X1Y1__SHIFT 0x10
+#define PA_SC_SHADER_CONTROL__REALIGN_DQUADS_AFTER_N_WAVES_MASK 0x3
+#define PA_SC_SHADER_CONTROL__REALIGN_DQUADS_AFTER_N_WAVES__SHIFT 0x0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_0__S0_X_MASK 0xf
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_0__S0_X__SHIFT 0x0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_0__S0_Y_MASK 0xf0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_0__S0_Y__SHIFT 0x4
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_0__S1_X_MASK 0xf00
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_0__S1_X__SHIFT 0x8
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_0__S1_Y_MASK 0xf000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_0__S1_Y__SHIFT 0xc
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_0__S2_X_MASK 0xf0000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_0__S2_X__SHIFT 0x10
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_0__S2_Y_MASK 0xf00000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_0__S2_Y__SHIFT 0x14
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_0__S3_X_MASK 0xf000000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_0__S3_X__SHIFT 0x18
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_0__S3_Y_MASK 0xf0000000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_0__S3_Y__SHIFT 0x1c
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_1__S4_X_MASK 0xf
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_1__S4_X__SHIFT 0x0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_1__S4_Y_MASK 0xf0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_1__S4_Y__SHIFT 0x4
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_1__S5_X_MASK 0xf00
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_1__S5_X__SHIFT 0x8
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_1__S5_Y_MASK 0xf000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_1__S5_Y__SHIFT 0xc
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_1__S6_X_MASK 0xf0000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_1__S6_X__SHIFT 0x10
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_1__S6_Y_MASK 0xf00000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_1__S6_Y__SHIFT 0x14
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_1__S7_X_MASK 0xf000000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_1__S7_X__SHIFT 0x18
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_1__S7_Y_MASK 0xf0000000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_1__S7_Y__SHIFT 0x1c
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_2__S8_X_MASK 0xf
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_2__S8_X__SHIFT 0x0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_2__S8_Y_MASK 0xf0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_2__S8_Y__SHIFT 0x4
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_2__S9_X_MASK 0xf00
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_2__S9_X__SHIFT 0x8
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_2__S9_Y_MASK 0xf000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_2__S9_Y__SHIFT 0xc
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_2__S10_X_MASK 0xf0000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_2__S10_X__SHIFT 0x10
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_2__S10_Y_MASK 0xf00000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_2__S10_Y__SHIFT 0x14
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_2__S11_X_MASK 0xf000000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_2__S11_X__SHIFT 0x18
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_2__S11_Y_MASK 0xf0000000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_2__S11_Y__SHIFT 0x1c
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_3__S12_X_MASK 0xf
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_3__S12_X__SHIFT 0x0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_3__S12_Y_MASK 0xf0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_3__S12_Y__SHIFT 0x4
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_3__S13_X_MASK 0xf00
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_3__S13_X__SHIFT 0x8
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_3__S13_Y_MASK 0xf000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_3__S13_Y__SHIFT 0xc
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_3__S14_X_MASK 0xf0000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_3__S14_X__SHIFT 0x10
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_3__S14_Y_MASK 0xf00000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_3__S14_Y__SHIFT 0x14
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_3__S15_X_MASK 0xf000000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_3__S15_X__SHIFT 0x18
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_3__S15_Y_MASK 0xf0000000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y0_3__S15_Y__SHIFT 0x1c
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_0__S0_X_MASK 0xf
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_0__S0_X__SHIFT 0x0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_0__S0_Y_MASK 0xf0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_0__S0_Y__SHIFT 0x4
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_0__S1_X_MASK 0xf00
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_0__S1_X__SHIFT 0x8
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_0__S1_Y_MASK 0xf000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_0__S1_Y__SHIFT 0xc
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_0__S2_X_MASK 0xf0000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_0__S2_X__SHIFT 0x10
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_0__S2_Y_MASK 0xf00000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_0__S2_Y__SHIFT 0x14
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_0__S3_X_MASK 0xf000000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_0__S3_X__SHIFT 0x18
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_0__S3_Y_MASK 0xf0000000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_0__S3_Y__SHIFT 0x1c
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_1__S4_X_MASK 0xf
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_1__S4_X__SHIFT 0x0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_1__S4_Y_MASK 0xf0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_1__S4_Y__SHIFT 0x4
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_1__S5_X_MASK 0xf00
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_1__S5_X__SHIFT 0x8
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_1__S5_Y_MASK 0xf000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_1__S5_Y__SHIFT 0xc
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_1__S6_X_MASK 0xf0000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_1__S6_X__SHIFT 0x10
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_1__S6_Y_MASK 0xf00000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_1__S6_Y__SHIFT 0x14
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_1__S7_X_MASK 0xf000000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_1__S7_X__SHIFT 0x18
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_1__S7_Y_MASK 0xf0000000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_1__S7_Y__SHIFT 0x1c
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_2__S8_X_MASK 0xf
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_2__S8_X__SHIFT 0x0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_2__S8_Y_MASK 0xf0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_2__S8_Y__SHIFT 0x4
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_2__S9_X_MASK 0xf00
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_2__S9_X__SHIFT 0x8
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_2__S9_Y_MASK 0xf000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_2__S9_Y__SHIFT 0xc
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_2__S10_X_MASK 0xf0000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_2__S10_X__SHIFT 0x10
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_2__S10_Y_MASK 0xf00000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_2__S10_Y__SHIFT 0x14
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_2__S11_X_MASK 0xf000000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_2__S11_X__SHIFT 0x18
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_2__S11_Y_MASK 0xf0000000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_2__S11_Y__SHIFT 0x1c
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_3__S12_X_MASK 0xf
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_3__S12_X__SHIFT 0x0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_3__S12_Y_MASK 0xf0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_3__S12_Y__SHIFT 0x4
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_3__S13_X_MASK 0xf00
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_3__S13_X__SHIFT 0x8
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_3__S13_Y_MASK 0xf000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_3__S13_Y__SHIFT 0xc
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_3__S14_X_MASK 0xf0000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_3__S14_X__SHIFT 0x10
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_3__S14_Y_MASK 0xf00000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_3__S14_Y__SHIFT 0x14
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_3__S15_X_MASK 0xf000000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_3__S15_X__SHIFT 0x18
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_3__S15_Y_MASK 0xf0000000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y0_3__S15_Y__SHIFT 0x1c
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_0__S0_X_MASK 0xf
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_0__S0_X__SHIFT 0x0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_0__S0_Y_MASK 0xf0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_0__S0_Y__SHIFT 0x4
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_0__S1_X_MASK 0xf00
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_0__S1_X__SHIFT 0x8
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_0__S1_Y_MASK 0xf000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_0__S1_Y__SHIFT 0xc
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_0__S2_X_MASK 0xf0000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_0__S2_X__SHIFT 0x10
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_0__S2_Y_MASK 0xf00000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_0__S2_Y__SHIFT 0x14
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_0__S3_X_MASK 0xf000000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_0__S3_X__SHIFT 0x18
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_0__S3_Y_MASK 0xf0000000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_0__S3_Y__SHIFT 0x1c
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_1__S4_X_MASK 0xf
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_1__S4_X__SHIFT 0x0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_1__S4_Y_MASK 0xf0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_1__S4_Y__SHIFT 0x4
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_1__S5_X_MASK 0xf00
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_1__S5_X__SHIFT 0x8
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_1__S5_Y_MASK 0xf000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_1__S5_Y__SHIFT 0xc
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_1__S6_X_MASK 0xf0000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_1__S6_X__SHIFT 0x10
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_1__S6_Y_MASK 0xf00000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_1__S6_Y__SHIFT 0x14
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_1__S7_X_MASK 0xf000000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_1__S7_X__SHIFT 0x18
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_1__S7_Y_MASK 0xf0000000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_1__S7_Y__SHIFT 0x1c
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_2__S8_X_MASK 0xf
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_2__S8_X__SHIFT 0x0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_2__S8_Y_MASK 0xf0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_2__S8_Y__SHIFT 0x4
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_2__S9_X_MASK 0xf00
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_2__S9_X__SHIFT 0x8
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_2__S9_Y_MASK 0xf000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_2__S9_Y__SHIFT 0xc
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_2__S10_X_MASK 0xf0000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_2__S10_X__SHIFT 0x10
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_2__S10_Y_MASK 0xf00000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_2__S10_Y__SHIFT 0x14
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_2__S11_X_MASK 0xf000000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_2__S11_X__SHIFT 0x18
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_2__S11_Y_MASK 0xf0000000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_2__S11_Y__SHIFT 0x1c
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_3__S12_X_MASK 0xf
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_3__S12_X__SHIFT 0x0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_3__S12_Y_MASK 0xf0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_3__S12_Y__SHIFT 0x4
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_3__S13_X_MASK 0xf00
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_3__S13_X__SHIFT 0x8
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_3__S13_Y_MASK 0xf000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_3__S13_Y__SHIFT 0xc
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_3__S14_X_MASK 0xf0000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_3__S14_X__SHIFT 0x10
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_3__S14_Y_MASK 0xf00000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_3__S14_Y__SHIFT 0x14
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_3__S15_X_MASK 0xf000000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_3__S15_X__SHIFT 0x18
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_3__S15_Y_MASK 0xf0000000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X0Y1_3__S15_Y__SHIFT 0x1c
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_0__S0_X_MASK 0xf
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_0__S0_X__SHIFT 0x0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_0__S0_Y_MASK 0xf0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_0__S0_Y__SHIFT 0x4
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_0__S1_X_MASK 0xf00
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_0__S1_X__SHIFT 0x8
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_0__S1_Y_MASK 0xf000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_0__S1_Y__SHIFT 0xc
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_0__S2_X_MASK 0xf0000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_0__S2_X__SHIFT 0x10
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_0__S2_Y_MASK 0xf00000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_0__S2_Y__SHIFT 0x14
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_0__S3_X_MASK 0xf000000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_0__S3_X__SHIFT 0x18
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_0__S3_Y_MASK 0xf0000000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_0__S3_Y__SHIFT 0x1c
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_1__S4_X_MASK 0xf
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_1__S4_X__SHIFT 0x0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_1__S4_Y_MASK 0xf0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_1__S4_Y__SHIFT 0x4
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_1__S5_X_MASK 0xf00
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_1__S5_X__SHIFT 0x8
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_1__S5_Y_MASK 0xf000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_1__S5_Y__SHIFT 0xc
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_1__S6_X_MASK 0xf0000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_1__S6_X__SHIFT 0x10
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_1__S6_Y_MASK 0xf00000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_1__S6_Y__SHIFT 0x14
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_1__S7_X_MASK 0xf000000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_1__S7_X__SHIFT 0x18
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_1__S7_Y_MASK 0xf0000000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_1__S7_Y__SHIFT 0x1c
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_2__S8_X_MASK 0xf
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_2__S8_X__SHIFT 0x0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_2__S8_Y_MASK 0xf0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_2__S8_Y__SHIFT 0x4
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_2__S9_X_MASK 0xf00
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_2__S9_X__SHIFT 0x8
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_2__S9_Y_MASK 0xf000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_2__S9_Y__SHIFT 0xc
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_2__S10_X_MASK 0xf0000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_2__S10_X__SHIFT 0x10
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_2__S10_Y_MASK 0xf00000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_2__S10_Y__SHIFT 0x14
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_2__S11_X_MASK 0xf000000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_2__S11_X__SHIFT 0x18
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_2__S11_Y_MASK 0xf0000000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_2__S11_Y__SHIFT 0x1c
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_3__S12_X_MASK 0xf
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_3__S12_X__SHIFT 0x0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_3__S12_Y_MASK 0xf0
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_3__S12_Y__SHIFT 0x4
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_3__S13_X_MASK 0xf00
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_3__S13_X__SHIFT 0x8
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_3__S13_Y_MASK 0xf000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_3__S13_Y__SHIFT 0xc
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_3__S14_X_MASK 0xf0000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_3__S14_X__SHIFT 0x10
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_3__S14_Y_MASK 0xf00000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_3__S14_Y__SHIFT 0x14
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_3__S15_X_MASK 0xf000000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_3__S15_X__SHIFT 0x18
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_3__S15_Y_MASK 0xf0000000
+#define PA_SC_AA_SAMPLE_LOCS_PIXEL_X1Y1_3__S15_Y__SHIFT 0x1c
+#define PA_SC_CENTROID_PRIORITY_0__DISTANCE_0_MASK 0xf
+#define PA_SC_CENTROID_PRIORITY_0__DISTANCE_0__SHIFT 0x0
+#define PA_SC_CENTROID_PRIORITY_0__DISTANCE_1_MASK 0xf0
+#define PA_SC_CENTROID_PRIORITY_0__DISTANCE_1__SHIFT 0x4
+#define PA_SC_CENTROID_PRIORITY_0__DISTANCE_2_MASK 0xf00
+#define PA_SC_CENTROID_PRIORITY_0__DISTANCE_2__SHIFT 0x8
+#define PA_SC_CENTROID_PRIORITY_0__DISTANCE_3_MASK 0xf000
+#define PA_SC_CENTROID_PRIORITY_0__DISTANCE_3__SHIFT 0xc
+#define PA_SC_CENTROID_PRIORITY_0__DISTANCE_4_MASK 0xf0000
+#define PA_SC_CENTROID_PRIORITY_0__DISTANCE_4__SHIFT 0x10
+#define PA_SC_CENTROID_PRIORITY_0__DISTANCE_5_MASK 0xf00000
+#define PA_SC_CENTROID_PRIORITY_0__DISTANCE_5__SHIFT 0x14
+#define PA_SC_CENTROID_PRIORITY_0__DISTANCE_6_MASK 0xf000000
+#define PA_SC_CENTROID_PRIORITY_0__DISTANCE_6__SHIFT 0x18
+#define PA_SC_CENTROID_PRIORITY_0__DISTANCE_7_MASK 0xf0000000
+#define PA_SC_CENTROID_PRIORITY_0__DISTANCE_7__SHIFT 0x1c
+#define PA_SC_CENTROID_PRIORITY_1__DISTANCE_8_MASK 0xf
+#define PA_SC_CENTROID_PRIORITY_1__DISTANCE_8__SHIFT 0x0
+#define PA_SC_CENTROID_PRIORITY_1__DISTANCE_9_MASK 0xf0
+#define PA_SC_CENTROID_PRIORITY_1__DISTANCE_9__SHIFT 0x4
+#define PA_SC_CENTROID_PRIORITY_1__DISTANCE_10_MASK 0xf00
+#define PA_SC_CENTROID_PRIORITY_1__DISTANCE_10__SHIFT 0x8
+#define PA_SC_CENTROID_PRIORITY_1__DISTANCE_11_MASK 0xf000
+#define PA_SC_CENTROID_PRIORITY_1__DISTANCE_11__SHIFT 0xc
+#define PA_SC_CENTROID_PRIORITY_1__DISTANCE_12_MASK 0xf0000
+#define PA_SC_CENTROID_PRIORITY_1__DISTANCE_12__SHIFT 0x10
+#define PA_SC_CENTROID_PRIORITY_1__DISTANCE_13_MASK 0xf00000
+#define PA_SC_CENTROID_PRIORITY_1__DISTANCE_13__SHIFT 0x14
+#define PA_SC_CENTROID_PRIORITY_1__DISTANCE_14_MASK 0xf000000
+#define PA_SC_CENTROID_PRIORITY_1__DISTANCE_14__SHIFT 0x18
+#define PA_SC_CENTROID_PRIORITY_1__DISTANCE_15_MASK 0xf0000000
+#define PA_SC_CENTROID_PRIORITY_1__DISTANCE_15__SHIFT 0x1c
+#define PA_SC_CLIPRECT_0_TL__TL_X_MASK 0x7fff
+#define PA_SC_CLIPRECT_0_TL__TL_X__SHIFT 0x0
+#define PA_SC_CLIPRECT_0_TL__TL_Y_MASK 0x7fff0000
+#define PA_SC_CLIPRECT_0_TL__TL_Y__SHIFT 0x10
+#define PA_SC_CLIPRECT_0_BR__BR_X_MASK 0x7fff
+#define PA_SC_CLIPRECT_0_BR__BR_X__SHIFT 0x0
+#define PA_SC_CLIPRECT_0_BR__BR_Y_MASK 0x7fff0000
+#define PA_SC_CLIPRECT_0_BR__BR_Y__SHIFT 0x10
+#define PA_SC_CLIPRECT_1_TL__TL_X_MASK 0x7fff
+#define PA_SC_CLIPRECT_1_TL__TL_X__SHIFT 0x0
+#define PA_SC_CLIPRECT_1_TL__TL_Y_MASK 0x7fff0000
+#define PA_SC_CLIPRECT_1_TL__TL_Y__SHIFT 0x10
+#define PA_SC_CLIPRECT_1_BR__BR_X_MASK 0x7fff
+#define PA_SC_CLIPRECT_1_BR__BR_X__SHIFT 0x0
+#define PA_SC_CLIPRECT_1_BR__BR_Y_MASK 0x7fff0000
+#define PA_SC_CLIPRECT_1_BR__BR_Y__SHIFT 0x10
+#define PA_SC_CLIPRECT_2_TL__TL_X_MASK 0x7fff
+#define PA_SC_CLIPRECT_2_TL__TL_X__SHIFT 0x0
+#define PA_SC_CLIPRECT_2_TL__TL_Y_MASK 0x7fff0000
+#define PA_SC_CLIPRECT_2_TL__TL_Y__SHIFT 0x10
+#define PA_SC_CLIPRECT_2_BR__BR_X_MASK 0x7fff
+#define PA_SC_CLIPRECT_2_BR__BR_X__SHIFT 0x0
+#define PA_SC_CLIPRECT_2_BR__BR_Y_MASK 0x7fff0000
+#define PA_SC_CLIPRECT_2_BR__BR_Y__SHIFT 0x10
+#define PA_SC_CLIPRECT_3_TL__TL_X_MASK 0x7fff
+#define PA_SC_CLIPRECT_3_TL__TL_X__SHIFT 0x0
+#define PA_SC_CLIPRECT_3_TL__TL_Y_MASK 0x7fff0000
+#define PA_SC_CLIPRECT_3_TL__TL_Y__SHIFT 0x10
+#define PA_SC_CLIPRECT_3_BR__BR_X_MASK 0x7fff
+#define PA_SC_CLIPRECT_3_BR__BR_X__SHIFT 0x0
+#define PA_SC_CLIPRECT_3_BR__BR_Y_MASK 0x7fff0000
+#define PA_SC_CLIPRECT_3_BR__BR_Y__SHIFT 0x10
+#define PA_SC_CLIPRECT_RULE__CLIP_RULE_MASK 0xffff
+#define PA_SC_CLIPRECT_RULE__CLIP_RULE__SHIFT 0x0
+#define PA_SC_EDGERULE__ER_TRI_MASK 0xf
+#define PA_SC_EDGERULE__ER_TRI__SHIFT 0x0
+#define PA_SC_EDGERULE__ER_POINT_MASK 0xf0
+#define PA_SC_EDGERULE__ER_POINT__SHIFT 0x4
+#define PA_SC_EDGERULE__ER_RECT_MASK 0xf00
+#define PA_SC_EDGERULE__ER_RECT__SHIFT 0x8
+#define PA_SC_EDGERULE__ER_LINE_LR_MASK 0x3f000
+#define PA_SC_EDGERULE__ER_LINE_LR__SHIFT 0xc
+#define PA_SC_EDGERULE__ER_LINE_RL_MASK 0xfc0000
+#define PA_SC_EDGERULE__ER_LINE_RL__SHIFT 0x12
+#define PA_SC_EDGERULE__ER_LINE_TB_MASK 0xf000000
+#define PA_SC_EDGERULE__ER_LINE_TB__SHIFT 0x18
+#define PA_SC_EDGERULE__ER_LINE_BT_MASK 0xf0000000
+#define PA_SC_EDGERULE__ER_LINE_BT__SHIFT 0x1c
+#define PA_SC_LINE_CNTL__EXPAND_LINE_WIDTH_MASK 0x200
+#define PA_SC_LINE_CNTL__EXPAND_LINE_WIDTH__SHIFT 0x9
+#define PA_SC_LINE_CNTL__LAST_PIXEL_MASK 0x400
+#define PA_SC_LINE_CNTL__LAST_PIXEL__SHIFT 0xa
+#define PA_SC_LINE_CNTL__PERPENDICULAR_ENDCAP_ENA_MASK 0x800
+#define PA_SC_LINE_CNTL__PERPENDICULAR_ENDCAP_ENA__SHIFT 0xb
+#define PA_SC_LINE_CNTL__DX10_DIAMOND_TEST_ENA_MASK 0x1000
+#define PA_SC_LINE_CNTL__DX10_DIAMOND_TEST_ENA__SHIFT 0xc
+#define PA_SC_LINE_STIPPLE__LINE_PATTERN_MASK 0xffff
+#define PA_SC_LINE_STIPPLE__LINE_PATTERN__SHIFT 0x0
+#define PA_SC_LINE_STIPPLE__REPEAT_COUNT_MASK 0xff0000
+#define PA_SC_LINE_STIPPLE__REPEAT_COUNT__SHIFT 0x10
+#define PA_SC_LINE_STIPPLE__PATTERN_BIT_ORDER_MASK 0x10000000
+#define PA_SC_LINE_STIPPLE__PATTERN_BIT_ORDER__SHIFT 0x1c
+#define PA_SC_LINE_STIPPLE__AUTO_RESET_CNTL_MASK 0x60000000
+#define PA_SC_LINE_STIPPLE__AUTO_RESET_CNTL__SHIFT 0x1d
+#define PA_SC_MODE_CNTL_0__MSAA_ENABLE_MASK 0x1
+#define PA_SC_MODE_CNTL_0__MSAA_ENABLE__SHIFT 0x0
+#define PA_SC_MODE_CNTL_0__VPORT_SCISSOR_ENABLE_MASK 0x2
+#define PA_SC_MODE_CNTL_0__VPORT_SCISSOR_ENABLE__SHIFT 0x1
+#define PA_SC_MODE_CNTL_0__LINE_STIPPLE_ENABLE_MASK 0x4
+#define PA_SC_MODE_CNTL_0__LINE_STIPPLE_ENABLE__SHIFT 0x2
+#define PA_SC_MODE_CNTL_0__SEND_UNLIT_STILES_TO_PKR_MASK 0x8
+#define PA_SC_MODE_CNTL_0__SEND_UNLIT_STILES_TO_PKR__SHIFT 0x3
+#define PA_SC_MODE_CNTL_1__WALK_SIZE_MASK 0x1
+#define PA_SC_MODE_CNTL_1__WALK_SIZE__SHIFT 0x0
+#define PA_SC_MODE_CNTL_1__WALK_ALIGNMENT_MASK 0x2
+#define PA_SC_MODE_CNTL_1__WALK_ALIGNMENT__SHIFT 0x1
+#define PA_SC_MODE_CNTL_1__WALK_ALIGN8_PRIM_FITS_ST_MASK 0x4
+#define PA_SC_MODE_CNTL_1__WALK_ALIGN8_PRIM_FITS_ST__SHIFT 0x2
+#define PA_SC_MODE_CNTL_1__WALK_FENCE_ENABLE_MASK 0x8
+#define PA_SC_MODE_CNTL_1__WALK_FENCE_ENABLE__SHIFT 0x3
+#define PA_SC_MODE_CNTL_1__WALK_FENCE_SIZE_MASK 0x70
+#define PA_SC_MODE_CNTL_1__WALK_FENCE_SIZE__SHIFT 0x4
+#define PA_SC_MODE_CNTL_1__SUPERTILE_WALK_ORDER_ENABLE_MASK 0x80
+#define PA_SC_MODE_CNTL_1__SUPERTILE_WALK_ORDER_ENABLE__SHIFT 0x7
+#define PA_SC_MODE_CNTL_1__TILE_WALK_ORDER_ENABLE_MASK 0x100
+#define PA_SC_MODE_CNTL_1__TILE_WALK_ORDER_ENABLE__SHIFT 0x8
+#define PA_SC_MODE_CNTL_1__TILE_COVER_DISABLE_MASK 0x200
+#define PA_SC_MODE_CNTL_1__TILE_COVER_DISABLE__SHIFT 0x9
+#define PA_SC_MODE_CNTL_1__TILE_COVER_NO_SCISSOR_MASK 0x400
+#define PA_SC_MODE_CNTL_1__TILE_COVER_NO_SCISSOR__SHIFT 0xa
+#define PA_SC_MODE_CNTL_1__ZMM_LINE_EXTENT_MASK 0x800
+#define PA_SC_MODE_CNTL_1__ZMM_LINE_EXTENT__SHIFT 0xb
+#define PA_SC_MODE_CNTL_1__ZMM_LINE_OFFSET_MASK 0x1000
+#define PA_SC_MODE_CNTL_1__ZMM_LINE_OFFSET__SHIFT 0xc
+#define PA_SC_MODE_CNTL_1__ZMM_RECT_EXTENT_MASK 0x2000
+#define PA_SC_MODE_CNTL_1__ZMM_RECT_EXTENT__SHIFT 0xd
+#define PA_SC_MODE_CNTL_1__KILL_PIX_POST_HI_Z_MASK 0x4000
+#define PA_SC_MODE_CNTL_1__KILL_PIX_POST_HI_Z__SHIFT 0xe
+#define PA_SC_MODE_CNTL_1__KILL_PIX_POST_DETAIL_MASK_MASK 0x8000
+#define PA_SC_MODE_CNTL_1__KILL_PIX_POST_DETAIL_MASK__SHIFT 0xf
+#define PA_SC_MODE_CNTL_1__PS_ITER_SAMPLE_MASK 0x10000
+#define PA_SC_MODE_CNTL_1__PS_ITER_SAMPLE__SHIFT 0x10
+#define PA_SC_MODE_CNTL_1__MULTI_SHADER_ENGINE_PRIM_DISCARD_ENABLE_MASK 0x20000
+#define PA_SC_MODE_CNTL_1__MULTI_SHADER_ENGINE_PRIM_DISCARD_ENABLE__SHIFT 0x11
+#define PA_SC_MODE_CNTL_1__MULTI_GPU_SUPERTILE_ENABLE_MASK 0x40000
+#define PA_SC_MODE_CNTL_1__MULTI_GPU_SUPERTILE_ENABLE__SHIFT 0x12
+#define PA_SC_MODE_CNTL_1__GPU_ID_OVERRIDE_ENABLE_MASK 0x80000
+#define PA_SC_MODE_CNTL_1__GPU_ID_OVERRIDE_ENABLE__SHIFT 0x13
+#define PA_SC_MODE_CNTL_1__GPU_ID_OVERRIDE_MASK 0xf00000
+#define PA_SC_MODE_CNTL_1__GPU_ID_OVERRIDE__SHIFT 0x14
+#define PA_SC_MODE_CNTL_1__MULTI_GPU_PRIM_DISCARD_ENABLE_MASK 0x1000000
+#define PA_SC_MODE_CNTL_1__MULTI_GPU_PRIM_DISCARD_ENABLE__SHIFT 0x18
+#define PA_SC_MODE_CNTL_1__FORCE_EOV_CNTDWN_ENABLE_MASK 0x2000000
+#define PA_SC_MODE_CNTL_1__FORCE_EOV_CNTDWN_ENABLE__SHIFT 0x19
+#define PA_SC_MODE_CNTL_1__FORCE_EOV_REZ_ENABLE_MASK 0x4000000
+#define PA_SC_MODE_CNTL_1__FORCE_EOV_REZ_ENABLE__SHIFT 0x1a
+#define PA_SC_MODE_CNTL_1__OUT_OF_ORDER_PRIMITIVE_ENABLE_MASK 0x8000000
+#define PA_SC_MODE_CNTL_1__OUT_OF_ORDER_PRIMITIVE_ENABLE__SHIFT 0x1b
+#define PA_SC_MODE_CNTL_1__OUT_OF_ORDER_WATER_MARK_MASK 0x70000000
+#define PA_SC_MODE_CNTL_1__OUT_OF_ORDER_WATER_MARK__SHIFT 0x1c
+#define PA_SC_RASTER_CONFIG__RB_MAP_PKR0_MASK 0x3
+#define PA_SC_RASTER_CONFIG__RB_MAP_PKR0__SHIFT 0x0
+#define PA_SC_RASTER_CONFIG__RB_MAP_PKR1_MASK 0xc
+#define PA_SC_RASTER_CONFIG__RB_MAP_PKR1__SHIFT 0x2
+#define PA_SC_RASTER_CONFIG__RB_XSEL2_MASK 0x30
+#define PA_SC_RASTER_CONFIG__RB_XSEL2__SHIFT 0x4
+#define PA_SC_RASTER_CONFIG__RB_XSEL_MASK 0x40
+#define PA_SC_RASTER_CONFIG__RB_XSEL__SHIFT 0x6
+#define PA_SC_RASTER_CONFIG__RB_YSEL_MASK 0x80
+#define PA_SC_RASTER_CONFIG__RB_YSEL__SHIFT 0x7
+#define PA_SC_RASTER_CONFIG__PKR_MAP_MASK 0x300
+#define PA_SC_RASTER_CONFIG__PKR_MAP__SHIFT 0x8
+#define PA_SC_RASTER_CONFIG__PKR_XSEL_MASK 0xc00
+#define PA_SC_RASTER_CONFIG__PKR_XSEL__SHIFT 0xa
+#define PA_SC_RASTER_CONFIG__PKR_YSEL_MASK 0x3000
+#define PA_SC_RASTER_CONFIG__PKR_YSEL__SHIFT 0xc
+#define PA_SC_RASTER_CONFIG__PKR_XSEL2_MASK 0xc000
+#define PA_SC_RASTER_CONFIG__PKR_XSEL2__SHIFT 0xe
+#define PA_SC_RASTER_CONFIG__SC_MAP_MASK 0x30000
+#define PA_SC_RASTER_CONFIG__SC_MAP__SHIFT 0x10
+#define PA_SC_RASTER_CONFIG__SC_XSEL_MASK 0xc0000
+#define PA_SC_RASTER_CONFIG__SC_XSEL__SHIFT 0x12
+#define PA_SC_RASTER_CONFIG__SC_YSEL_MASK 0x300000
+#define PA_SC_RASTER_CONFIG__SC_YSEL__SHIFT 0x14
+#define PA_SC_RASTER_CONFIG__SE_MAP_MASK 0x3000000
+#define PA_SC_RASTER_CONFIG__SE_MAP__SHIFT 0x18
+#define PA_SC_RASTER_CONFIG__SE_XSEL_MASK 0xc000000
+#define PA_SC_RASTER_CONFIG__SE_XSEL__SHIFT 0x1a
+#define PA_SC_RASTER_CONFIG__SE_YSEL_MASK 0x30000000
+#define PA_SC_RASTER_CONFIG__SE_YSEL__SHIFT 0x1c
+#define PA_SC_RASTER_CONFIG_1__SE_PAIR_MAP_MASK 0x3
+#define PA_SC_RASTER_CONFIG_1__SE_PAIR_MAP__SHIFT 0x0
+#define PA_SC_RASTER_CONFIG_1__SE_PAIR_XSEL_MASK 0xc
+#define PA_SC_RASTER_CONFIG_1__SE_PAIR_XSEL__SHIFT 0x2
+#define PA_SC_RASTER_CONFIG_1__SE_PAIR_YSEL_MASK 0x30
+#define PA_SC_RASTER_CONFIG_1__SE_PAIR_YSEL__SHIFT 0x4
+#define PA_SC_SCREEN_EXTENT_CONTROL__SLICE_EVEN_ENABLE_MASK 0x3
+#define PA_SC_SCREEN_EXTENT_CONTROL__SLICE_EVEN_ENABLE__SHIFT 0x0
+#define PA_SC_SCREEN_EXTENT_CONTROL__SLICE_ODD_ENABLE_MASK 0xc
+#define PA_SC_SCREEN_EXTENT_CONTROL__SLICE_ODD_ENABLE__SHIFT 0x2
+#define PA_SC_GENERIC_SCISSOR_TL__TL_X_MASK 0x7fff
+#define PA_SC_GENERIC_SCISSOR_TL__TL_X__SHIFT 0x0
+#define PA_SC_GENERIC_SCISSOR_TL__TL_Y_MASK 0x7fff0000
+#define PA_SC_GENERIC_SCISSOR_TL__TL_Y__SHIFT 0x10
+#define PA_SC_GENERIC_SCISSOR_TL__WINDOW_OFFSET_DISABLE_MASK 0x80000000
+#define PA_SC_GENERIC_SCISSOR_TL__WINDOW_OFFSET_DISABLE__SHIFT 0x1f
+#define PA_SC_GENERIC_SCISSOR_BR__BR_X_MASK 0x7fff
+#define PA_SC_GENERIC_SCISSOR_BR__BR_X__SHIFT 0x0
+#define PA_SC_GENERIC_SCISSOR_BR__BR_Y_MASK 0x7fff0000
+#define PA_SC_GENERIC_SCISSOR_BR__BR_Y__SHIFT 0x10
+#define PA_SC_SCREEN_SCISSOR_TL__TL_X_MASK 0xffff
+#define PA_SC_SCREEN_SCISSOR_TL__TL_X__SHIFT 0x0
+#define PA_SC_SCREEN_SCISSOR_TL__TL_Y_MASK 0xffff0000
+#define PA_SC_SCREEN_SCISSOR_TL__TL_Y__SHIFT 0x10
+#define PA_SC_SCREEN_SCISSOR_BR__BR_X_MASK 0xffff
+#define PA_SC_SCREEN_SCISSOR_BR__BR_X__SHIFT 0x0
+#define PA_SC_SCREEN_SCISSOR_BR__BR_Y_MASK 0xffff0000
+#define PA_SC_SCREEN_SCISSOR_BR__BR_Y__SHIFT 0x10
+#define PA_SC_WINDOW_OFFSET__WINDOW_X_OFFSET_MASK 0xffff
+#define PA_SC_WINDOW_OFFSET__WINDOW_X_OFFSET__SHIFT 0x0
+#define PA_SC_WINDOW_OFFSET__WINDOW_Y_OFFSET_MASK 0xffff0000
+#define PA_SC_WINDOW_OFFSET__WINDOW_Y_OFFSET__SHIFT 0x10
+#define PA_SC_WINDOW_SCISSOR_TL__TL_X_MASK 0x7fff
+#define PA_SC_WINDOW_SCISSOR_TL__TL_X__SHIFT 0x0
+#define PA_SC_WINDOW_SCISSOR_TL__TL_Y_MASK 0x7fff0000
+#define PA_SC_WINDOW_SCISSOR_TL__TL_Y__SHIFT 0x10
+#define PA_SC_WINDOW_SCISSOR_TL__WINDOW_OFFSET_DISABLE_MASK 0x80000000
+#define PA_SC_WINDOW_SCISSOR_TL__WINDOW_OFFSET_DISABLE__SHIFT 0x1f
+#define PA_SC_WINDOW_SCISSOR_BR__BR_X_MASK 0x7fff
+#define PA_SC_WINDOW_SCISSOR_BR__BR_X__SHIFT 0x0
+#define PA_SC_WINDOW_SCISSOR_BR__BR_Y_MASK 0x7fff0000
+#define PA_SC_WINDOW_SCISSOR_BR__BR_Y__SHIFT 0x10
+#define PA_SC_VPORT_SCISSOR_0_TL__TL_X_MASK 0x7fff
+#define PA_SC_VPORT_SCISSOR_0_TL__TL_X__SHIFT 0x0
+#define PA_SC_VPORT_SCISSOR_0_TL__TL_Y_MASK 0x7fff0000
+#define PA_SC_VPORT_SCISSOR_0_TL__TL_Y__SHIFT 0x10
+#define PA_SC_VPORT_SCISSOR_0_TL__WINDOW_OFFSET_DISABLE_MASK 0x80000000
+#define PA_SC_VPORT_SCISSOR_0_TL__WINDOW_OFFSET_DISABLE__SHIFT 0x1f
+#define PA_SC_VPORT_SCISSOR_1_TL__TL_X_MASK 0x7fff
+#define PA_SC_VPORT_SCISSOR_1_TL__TL_X__SHIFT 0x0
+#define PA_SC_VPORT_SCISSOR_1_TL__TL_Y_MASK 0x7fff0000
+#define PA_SC_VPORT_SCISSOR_1_TL__TL_Y__SHIFT 0x10
+#define PA_SC_VPORT_SCISSOR_1_TL__WINDOW_OFFSET_DISABLE_MASK 0x80000000
+#define PA_SC_VPORT_SCISSOR_1_TL__WINDOW_OFFSET_DISABLE__SHIFT 0x1f
+#define PA_SC_VPORT_SCISSOR_2_TL__TL_X_MASK 0x7fff
+#define PA_SC_VPORT_SCISSOR_2_TL__TL_X__SHIFT 0x0
+#define PA_SC_VPORT_SCISSOR_2_TL__TL_Y_MASK 0x7fff0000
+#define PA_SC_VPORT_SCISSOR_2_TL__TL_Y__SHIFT 0x10
+#define PA_SC_VPORT_SCISSOR_2_TL__WINDOW_OFFSET_DISABLE_MASK 0x80000000
+#define PA_SC_VPORT_SCISSOR_2_TL__WINDOW_OFFSET_DISABLE__SHIFT 0x1f
+#define PA_SC_VPORT_SCISSOR_3_TL__TL_X_MASK 0x7fff
+#define PA_SC_VPORT_SCISSOR_3_TL__TL_X__SHIFT 0x0
+#define PA_SC_VPORT_SCISSOR_3_TL__TL_Y_MASK 0x7fff0000
+#define PA_SC_VPORT_SCISSOR_3_TL__TL_Y__SHIFT 0x10
+#define PA_SC_VPORT_SCISSOR_3_TL__WINDOW_OFFSET_DISABLE_MASK 0x80000000
+#define PA_SC_VPORT_SCISSOR_3_TL__WINDOW_OFFSET_DISABLE__SHIFT 0x1f
+#define PA_SC_VPORT_SCISSOR_4_TL__TL_X_MASK 0x7fff
+#define PA_SC_VPORT_SCISSOR_4_TL__TL_X__SHIFT 0x0
+#define PA_SC_VPORT_SCISSOR_4_TL__TL_Y_MASK 0x7fff0000
+#define PA_SC_VPORT_SCISSOR_4_TL__TL_Y__SHIFT 0x10
+#define PA_SC_VPORT_SCISSOR_4_TL__WINDOW_OFFSET_DISABLE_MASK 0x80000000
+#define PA_SC_VPORT_SCISSOR_4_TL__WINDOW_OFFSET_DISABLE__SHIFT 0x1f
+#define PA_SC_VPORT_SCISSOR_5_TL__TL_X_MASK 0x7fff
+#define PA_SC_VPORT_SCISSOR_5_TL__TL_X__SHIFT 0x0
+#define PA_SC_VPORT_SCISSOR_5_TL__TL_Y_MASK 0x7fff0000
+#define PA_SC_VPORT_SCISSOR_5_TL__TL_Y__SHIFT 0x10
+#define PA_SC_VPORT_SCISSOR_5_TL__WINDOW_OFFSET_DISABLE_MASK 0x80000000
+#define PA_SC_VPORT_SCISSOR_5_TL__WINDOW_OFFSET_DISABLE__SHIFT 0x1f
+#define PA_SC_VPORT_SCISSOR_6_TL__TL_X_MASK 0x7fff
+#define PA_SC_VPORT_SCISSOR_6_TL__TL_X__SHIFT 0x0
+#define PA_SC_VPORT_SCISSOR_6_TL__TL_Y_MASK 0x7fff0000
+#define PA_SC_VPORT_SCISSOR_6_TL__TL_Y__SHIFT 0x10
+#define PA_SC_VPORT_SCISSOR_6_TL__WINDOW_OFFSET_DISABLE_MASK 0x80000000
+#define PA_SC_VPORT_SCISSOR_6_TL__WINDOW_OFFSET_DISABLE__SHIFT 0x1f
+#define PA_SC_VPORT_SCISSOR_7_TL__TL_X_MASK 0x7fff
+#define PA_SC_VPORT_SCISSOR_7_TL__TL_X__SHIFT 0x0
+#define PA_SC_VPORT_SCISSOR_7_TL__TL_Y_MASK 0x7fff0000
+#define PA_SC_VPORT_SCISSOR_7_TL__TL_Y__SHIFT 0x10
+#define PA_SC_VPORT_SCISSOR_7_TL__WINDOW_OFFSET_DISABLE_MASK 0x80000000
+#define PA_SC_VPORT_SCISSOR_7_TL__WINDOW_OFFSET_DISABLE__SHIFT 0x1f
+#define PA_SC_VPORT_SCISSOR_8_TL__TL_X_MASK 0x7fff
+#define PA_SC_VPORT_SCISSOR_8_TL__TL_X__SHIFT 0x0
+#define PA_SC_VPORT_SCISSOR_8_TL__TL_Y_MASK 0x7fff0000
+#define PA_SC_VPORT_SCISSOR_8_TL__TL_Y__SHIFT 0x10
+#define PA_SC_VPORT_SCISSOR_8_TL__WINDOW_OFFSET_DISABLE_MASK 0x80000000
+#define PA_SC_VPORT_SCISSOR_8_TL__WINDOW_OFFSET_DISABLE__SHIFT 0x1f
+#define PA_SC_VPORT_SCISSOR_9_TL__TL_X_MASK 0x7fff
+#define PA_SC_VPORT_SCISSOR_9_TL__TL_X__SHIFT 0x0
+#define PA_SC_VPORT_SCISSOR_9_TL__TL_Y_MASK 0x7fff0000
+#define PA_SC_VPORT_SCISSOR_9_TL__TL_Y__SHIFT 0x10
+#define PA_SC_VPORT_SCISSOR_9_TL__WINDOW_OFFSET_DISABLE_MASK 0x80000000
+#define PA_SC_VPORT_SCISSOR_9_TL__WINDOW_OFFSET_DISABLE__SHIFT 0x1f
+#define PA_SC_VPORT_SCISSOR_10_TL__TL_X_MASK 0x7fff
+#define PA_SC_VPORT_SCISSOR_10_TL__TL_X__SHIFT 0x0
+#define PA_SC_VPORT_SCISSOR_10_TL__TL_Y_MASK 0x7fff0000
+#define PA_SC_VPORT_SCISSOR_10_TL__TL_Y__SHIFT 0x10
+#define PA_SC_VPORT_SCISSOR_10_TL__WINDOW_OFFSET_DISABLE_MASK 0x80000000
+#define PA_SC_VPORT_SCISSOR_10_TL__WINDOW_OFFSET_DISABLE__SHIFT 0x1f
+#define PA_SC_VPORT_SCISSOR_11_TL__TL_X_MASK 0x7fff
+#define PA_SC_VPORT_SCISSOR_11_TL__TL_X__SHIFT 0x0
+#define PA_SC_VPORT_SCISSOR_11_TL__TL_Y_MASK 0x7fff0000
+#define PA_SC_VPORT_SCISSOR_11_TL__TL_Y__SHIFT 0x10
+#define PA_SC_VPORT_SCISSOR_11_TL__WINDOW_OFFSET_DISABLE_MASK 0x80000000
+#define PA_SC_VPORT_SCISSOR_11_TL__WINDOW_OFFSET_DISABLE__SHIFT 0x1f
+#define PA_SC_VPORT_SCISSOR_12_TL__TL_X_MASK 0x7fff
+#define PA_SC_VPORT_SCISSOR_12_TL__TL_X__SHIFT 0x0
+#define PA_SC_VPORT_SCISSOR_12_TL__TL_Y_MASK 0x7fff0000
+#define PA_SC_VPORT_SCISSOR_12_TL__TL_Y__SHIFT 0x10
+#define PA_SC_VPORT_SCISSOR_12_TL__WINDOW_OFFSET_DISABLE_MASK 0x80000000
+#define PA_SC_VPORT_SCISSOR_12_TL__WINDOW_OFFSET_DISABLE__SHIFT 0x1f
+#define PA_SC_VPORT_SCISSOR_13_TL__TL_X_MASK 0x7fff
+#define PA_SC_VPORT_SCISSOR_13_TL__TL_X__SHIFT 0x0
+#define PA_SC_VPORT_SCISSOR_13_TL__TL_Y_MASK 0x7fff0000
+#define PA_SC_VPORT_SCISSOR_13_TL__TL_Y__SHIFT 0x10
+#define PA_SC_VPORT_SCISSOR_13_TL__WINDOW_OFFSET_DISABLE_MASK 0x80000000
+#define PA_SC_VPORT_SCISSOR_13_TL__WINDOW_OFFSET_DISABLE__SHIFT 0x1f
+#define PA_SC_VPORT_SCISSOR_14_TL__TL_X_MASK 0x7fff
+#define PA_SC_VPORT_SCISSOR_14_TL__TL_X__SHIFT 0x0
+#define PA_SC_VPORT_SCISSOR_14_TL__TL_Y_MASK 0x7fff0000
+#define PA_SC_VPORT_SCISSOR_14_TL__TL_Y__SHIFT 0x10
+#define PA_SC_VPORT_SCISSOR_14_TL__WINDOW_OFFSET_DISABLE_MASK 0x80000000
+#define PA_SC_VPORT_SCISSOR_14_TL__WINDOW_OFFSET_DISABLE__SHIFT 0x1f
+#define PA_SC_VPORT_SCISSOR_15_TL__TL_X_MASK 0x7fff
+#define PA_SC_VPORT_SCISSOR_15_TL__TL_X__SHIFT 0x0
+#define PA_SC_VPORT_SCISSOR_15_TL__TL_Y_MASK 0x7fff0000
+#define PA_SC_VPORT_SCISSOR_15_TL__TL_Y__SHIFT 0x10
+#define PA_SC_VPORT_SCISSOR_15_TL__WINDOW_OFFSET_DISABLE_MASK 0x80000000
+#define PA_SC_VPORT_SCISSOR_15_TL__WINDOW_OFFSET_DISABLE__SHIFT 0x1f
+#define PA_SC_VPORT_SCISSOR_0_BR__BR_X_MASK 0x7fff
+#define PA_SC_VPORT_SCISSOR_0_BR__BR_X__SHIFT 0x0
+#define PA_SC_VPORT_SCISSOR_0_BR__BR_Y_MASK 0x7fff0000
+#define PA_SC_VPORT_SCISSOR_0_BR__BR_Y__SHIFT 0x10
+#define PA_SC_VPORT_SCISSOR_1_BR__BR_X_MASK 0x7fff
+#define PA_SC_VPORT_SCISSOR_1_BR__BR_X__SHIFT 0x0
+#define PA_SC_VPORT_SCISSOR_1_BR__BR_Y_MASK 0x7fff0000
+#define PA_SC_VPORT_SCISSOR_1_BR__BR_Y__SHIFT 0x10
+#define PA_SC_VPORT_SCISSOR_2_BR__BR_X_MASK 0x7fff
+#define PA_SC_VPORT_SCISSOR_2_BR__BR_X__SHIFT 0x0
+#define PA_SC_VPORT_SCISSOR_2_BR__BR_Y_MASK 0x7fff0000
+#define PA_SC_VPORT_SCISSOR_2_BR__BR_Y__SHIFT 0x10
+#define PA_SC_VPORT_SCISSOR_3_BR__BR_X_MASK 0x7fff
+#define PA_SC_VPORT_SCISSOR_3_BR__BR_X__SHIFT 0x0
+#define PA_SC_VPORT_SCISSOR_3_BR__BR_Y_MASK 0x7fff0000
+#define PA_SC_VPORT_SCISSOR_3_BR__BR_Y__SHIFT 0x10
+#define PA_SC_VPORT_SCISSOR_4_BR__BR_X_MASK 0x7fff
+#define PA_SC_VPORT_SCISSOR_4_BR__BR_X__SHIFT 0x0
+#define PA_SC_VPORT_SCISSOR_4_BR__BR_Y_MASK 0x7fff0000
+#define PA_SC_VPORT_SCISSOR_4_BR__BR_Y__SHIFT 0x10
+#define PA_SC_VPORT_SCISSOR_5_BR__BR_X_MASK 0x7fff
+#define PA_SC_VPORT_SCISSOR_5_BR__BR_X__SHIFT 0x0
+#define PA_SC_VPORT_SCISSOR_5_BR__BR_Y_MASK 0x7fff0000
+#define PA_SC_VPORT_SCISSOR_5_BR__BR_Y__SHIFT 0x10
+#define PA_SC_VPORT_SCISSOR_6_BR__BR_X_MASK 0x7fff
+#define PA_SC_VPORT_SCISSOR_6_BR__BR_X__SHIFT 0x0
+#define PA_SC_VPORT_SCISSOR_6_BR__BR_Y_MASK 0x7fff0000
+#define PA_SC_VPORT_SCISSOR_6_BR__BR_Y__SHIFT 0x10
+#define PA_SC_VPORT_SCISSOR_7_BR__BR_X_MASK 0x7fff
+#define PA_SC_VPORT_SCISSOR_7_BR__BR_X__SHIFT 0x0
+#define PA_SC_VPORT_SCISSOR_7_BR__BR_Y_MASK 0x7fff0000
+#define PA_SC_VPORT_SCISSOR_7_BR__BR_Y__SHIFT 0x10
+#define PA_SC_VPORT_SCISSOR_8_BR__BR_X_MASK 0x7fff
+#define PA_SC_VPORT_SCISSOR_8_BR__BR_X__SHIFT 0x0
+#define PA_SC_VPORT_SCISSOR_8_BR__BR_Y_MASK 0x7fff0000
+#define PA_SC_VPORT_SCISSOR_8_BR__BR_Y__SHIFT 0x10
+#define PA_SC_VPORT_SCISSOR_9_BR__BR_X_MASK 0x7fff
+#define PA_SC_VPORT_SCISSOR_9_BR__BR_X__SHIFT 0x0
+#define PA_SC_VPORT_SCISSOR_9_BR__BR_Y_MASK 0x7fff0000
+#define PA_SC_VPORT_SCISSOR_9_BR__BR_Y__SHIFT 0x10
+#define PA_SC_VPORT_SCISSOR_10_BR__BR_X_MASK 0x7fff
+#define PA_SC_VPORT_SCISSOR_10_BR__BR_X__SHIFT 0x0
+#define PA_SC_VPORT_SCISSOR_10_BR__BR_Y_MASK 0x7fff0000
+#define PA_SC_VPORT_SCISSOR_10_BR__BR_Y__SHIFT 0x10
+#define PA_SC_VPORT_SCISSOR_11_BR__BR_X_MASK 0x7fff
+#define PA_SC_VPORT_SCISSOR_11_BR__BR_X__SHIFT 0x0
+#define PA_SC_VPORT_SCISSOR_11_BR__BR_Y_MASK 0x7fff0000
+#define PA_SC_VPORT_SCISSOR_11_BR__BR_Y__SHIFT 0x10
+#define PA_SC_VPORT_SCISSOR_12_BR__BR_X_MASK 0x7fff
+#define PA_SC_VPORT_SCISSOR_12_BR__BR_X__SHIFT 0x0
+#define PA_SC_VPORT_SCISSOR_12_BR__BR_Y_MASK 0x7fff0000
+#define PA_SC_VPORT_SCISSOR_12_BR__BR_Y__SHIFT 0x10
+#define PA_SC_VPORT_SCISSOR_13_BR__BR_X_MASK 0x7fff
+#define PA_SC_VPORT_SCISSOR_13_BR__BR_X__SHIFT 0x0
+#define PA_SC_VPORT_SCISSOR_13_BR__BR_Y_MASK 0x7fff0000
+#define PA_SC_VPORT_SCISSOR_13_BR__BR_Y__SHIFT 0x10
+#define PA_SC_VPORT_SCISSOR_14_BR__BR_X_MASK 0x7fff
+#define PA_SC_VPORT_SCISSOR_14_BR__BR_X__SHIFT 0x0
+#define PA_SC_VPORT_SCISSOR_14_BR__BR_Y_MASK 0x7fff0000
+#define PA_SC_VPORT_SCISSOR_14_BR__BR_Y__SHIFT 0x10
+#define PA_SC_VPORT_SCISSOR_15_BR__BR_X_MASK 0x7fff
+#define PA_SC_VPORT_SCISSOR_15_BR__BR_X__SHIFT 0x0
+#define PA_SC_VPORT_SCISSOR_15_BR__BR_Y_MASK 0x7fff0000
+#define PA_SC_VPORT_SCISSOR_15_BR__BR_Y__SHIFT 0x10
+#define PA_SC_VPORT_ZMIN_0__VPORT_ZMIN_MASK 0xffffffff
+#define PA_SC_VPORT_ZMIN_0__VPORT_ZMIN__SHIFT 0x0
+#define PA_SC_VPORT_ZMIN_1__VPORT_ZMIN_MASK 0xffffffff
+#define PA_SC_VPORT_ZMIN_1__VPORT_ZMIN__SHIFT 0x0
+#define PA_SC_VPORT_ZMIN_2__VPORT_ZMIN_MASK 0xffffffff
+#define PA_SC_VPORT_ZMIN_2__VPORT_ZMIN__SHIFT 0x0
+#define PA_SC_VPORT_ZMIN_3__VPORT_ZMIN_MASK 0xffffffff
+#define PA_SC_VPORT_ZMIN_3__VPORT_ZMIN__SHIFT 0x0
+#define PA_SC_VPORT_ZMIN_4__VPORT_ZMIN_MASK 0xffffffff
+#define PA_SC_VPORT_ZMIN_4__VPORT_ZMIN__SHIFT 0x0
+#define PA_SC_VPORT_ZMIN_5__VPORT_ZMIN_MASK 0xffffffff
+#define PA_SC_VPORT_ZMIN_5__VPORT_ZMIN__SHIFT 0x0
+#define PA_SC_VPORT_ZMIN_6__VPORT_ZMIN_MASK 0xffffffff
+#define PA_SC_VPORT_ZMIN_6__VPORT_ZMIN__SHIFT 0x0
+#define PA_SC_VPORT_ZMIN_7__VPORT_ZMIN_MASK 0xffffffff
+#define PA_SC_VPORT_ZMIN_7__VPORT_ZMIN__SHIFT 0x0
+#define PA_SC_VPORT_ZMIN_8__VPORT_ZMIN_MASK 0xffffffff
+#define PA_SC_VPORT_ZMIN_8__VPORT_ZMIN__SHIFT 0x0
+#define PA_SC_VPORT_ZMIN_9__VPORT_ZMIN_MASK 0xffffffff
+#define PA_SC_VPORT_ZMIN_9__VPORT_ZMIN__SHIFT 0x0
+#define PA_SC_VPORT_ZMIN_10__VPORT_ZMIN_MASK 0xffffffff
+#define PA_SC_VPORT_ZMIN_10__VPORT_ZMIN__SHIFT 0x0
+#define PA_SC_VPORT_ZMIN_11__VPORT_ZMIN_MASK 0xffffffff
+#define PA_SC_VPORT_ZMIN_11__VPORT_ZMIN__SHIFT 0x0
+#define PA_SC_VPORT_ZMIN_12__VPORT_ZMIN_MASK 0xffffffff
+#define PA_SC_VPORT_ZMIN_12__VPORT_ZMIN__SHIFT 0x0
+#define PA_SC_VPORT_ZMIN_13__VPORT_ZMIN_MASK 0xffffffff
+#define PA_SC_VPORT_ZMIN_13__VPORT_ZMIN__SHIFT 0x0
+#define PA_SC_VPORT_ZMIN_14__VPORT_ZMIN_MASK 0xffffffff
+#define PA_SC_VPORT_ZMIN_14__VPORT_ZMIN__SHIFT 0x0
+#define PA_SC_VPORT_ZMIN_15__VPORT_ZMIN_MASK 0xffffffff
+#define PA_SC_VPORT_ZMIN_15__VPORT_ZMIN__SHIFT 0x0
+#define PA_SC_VPORT_ZMAX_0__VPORT_ZMAX_MASK 0xffffffff
+#define PA_SC_VPORT_ZMAX_0__VPORT_ZMAX__SHIFT 0x0
+#define PA_SC_VPORT_ZMAX_1__VPORT_ZMAX_MASK 0xffffffff
+#define PA_SC_VPORT_ZMAX_1__VPORT_ZMAX__SHIFT 0x0
+#define PA_SC_VPORT_ZMAX_2__VPORT_ZMAX_MASK 0xffffffff
+#define PA_SC_VPORT_ZMAX_2__VPORT_ZMAX__SHIFT 0x0
+#define PA_SC_VPORT_ZMAX_3__VPORT_ZMAX_MASK 0xffffffff
+#define PA_SC_VPORT_ZMAX_3__VPORT_ZMAX__SHIFT 0x0
+#define PA_SC_VPORT_ZMAX_4__VPORT_ZMAX_MASK 0xffffffff
+#define PA_SC_VPORT_ZMAX_4__VPORT_ZMAX__SHIFT 0x0
+#define PA_SC_VPORT_ZMAX_5__VPORT_ZMAX_MASK 0xffffffff
+#define PA_SC_VPORT_ZMAX_5__VPORT_ZMAX__SHIFT 0x0
+#define PA_SC_VPORT_ZMAX_6__VPORT_ZMAX_MASK 0xffffffff
+#define PA_SC_VPORT_ZMAX_6__VPORT_ZMAX__SHIFT 0x0
+#define PA_SC_VPORT_ZMAX_7__VPORT_ZMAX_MASK 0xffffffff
+#define PA_SC_VPORT_ZMAX_7__VPORT_ZMAX__SHIFT 0x0
+#define PA_SC_VPORT_ZMAX_8__VPORT_ZMAX_MASK 0xffffffff
+#define PA_SC_VPORT_ZMAX_8__VPORT_ZMAX__SHIFT 0x0
+#define PA_SC_VPORT_ZMAX_9__VPORT_ZMAX_MASK 0xffffffff
+#define PA_SC_VPORT_ZMAX_9__VPORT_ZMAX__SHIFT 0x0
+#define PA_SC_VPORT_ZMAX_10__VPORT_ZMAX_MASK 0xffffffff
+#define PA_SC_VPORT_ZMAX_10__VPORT_ZMAX__SHIFT 0x0
+#define PA_SC_VPORT_ZMAX_11__VPORT_ZMAX_MASK 0xffffffff
+#define PA_SC_VPORT_ZMAX_11__VPORT_ZMAX__SHIFT 0x0
+#define PA_SC_VPORT_ZMAX_12__VPORT_ZMAX_MASK 0xffffffff
+#define PA_SC_VPORT_ZMAX_12__VPORT_ZMAX__SHIFT 0x0
+#define PA_SC_VPORT_ZMAX_13__VPORT_ZMAX_MASK 0xffffffff
+#define PA_SC_VPORT_ZMAX_13__VPORT_ZMAX__SHIFT 0x0
+#define PA_SC_VPORT_ZMAX_14__VPORT_ZMAX_MASK 0xffffffff
+#define PA_SC_VPORT_ZMAX_14__VPORT_ZMAX__SHIFT 0x0
+#define PA_SC_VPORT_ZMAX_15__VPORT_ZMAX_MASK 0xffffffff
+#define PA_SC_VPORT_ZMAX_15__VPORT_ZMAX__SHIFT 0x0
+#define PA_SC_ENHANCE__ENABLE_PA_SC_OUT_OF_ORDER_MASK 0x1
+#define PA_SC_ENHANCE__ENABLE_PA_SC_OUT_OF_ORDER__SHIFT 0x0
+#define PA_SC_ENHANCE__DISABLE_SC_DB_TILE_FIX_MASK 0x2
+#define PA_SC_ENHANCE__DISABLE_SC_DB_TILE_FIX__SHIFT 0x1
+#define PA_SC_ENHANCE__DISABLE_AA_MASK_FULL_FIX_MASK 0x4
+#define PA_SC_ENHANCE__DISABLE_AA_MASK_FULL_FIX__SHIFT 0x2
+#define PA_SC_ENHANCE__ENABLE_1XMSAA_SAMPLE_LOCATIONS_MASK 0x8
+#define PA_SC_ENHANCE__ENABLE_1XMSAA_SAMPLE_LOCATIONS__SHIFT 0x3
+#define PA_SC_ENHANCE__ENABLE_1XMSAA_SAMPLE_LOC_CENTROID_MASK 0x10
+#define PA_SC_ENHANCE__ENABLE_1XMSAA_SAMPLE_LOC_CENTROID__SHIFT 0x4
+#define PA_SC_ENHANCE__DISABLE_SCISSOR_FIX_MASK 0x20
+#define PA_SC_ENHANCE__DISABLE_SCISSOR_FIX__SHIFT 0x5
+#define PA_SC_ENHANCE__DISABLE_PW_BUBBLE_COLLAPSE_MASK 0xc0
+#define PA_SC_ENHANCE__DISABLE_PW_BUBBLE_COLLAPSE__SHIFT 0x6
+#define PA_SC_ENHANCE__SEND_UNLIT_STILES_TO_PACKER_MASK 0x100
+#define PA_SC_ENHANCE__SEND_UNLIT_STILES_TO_PACKER__SHIFT 0x8
+#define PA_SC_ENHANCE__DISABLE_DUALGRAD_PERF_OPTIMIZATION_MASK 0x200
+#define PA_SC_ENHANCE__DISABLE_DUALGRAD_PERF_OPTIMIZATION__SHIFT 0x9
+#define PA_SC_ENHANCE__DISABLE_SC_PROCESS_RESET_PRIM_MASK 0x400
+#define PA_SC_ENHANCE__DISABLE_SC_PROCESS_RESET_PRIM__SHIFT 0xa
+#define PA_SC_ENHANCE__DISABLE_SC_PROCESS_RESET_SUPERTILE_MASK 0x800
+#define PA_SC_ENHANCE__DISABLE_SC_PROCESS_RESET_SUPERTILE__SHIFT 0xb
+#define PA_SC_ENHANCE__DISABLE_SC_PROCESS_RESET_TILE_MASK 0x1000
+#define PA_SC_ENHANCE__DISABLE_SC_PROCESS_RESET_TILE__SHIFT 0xc
+#define PA_SC_ENHANCE__DISABLE_PA_SC_GUIDANCE_MASK 0x2000
+#define PA_SC_ENHANCE__DISABLE_PA_SC_GUIDANCE__SHIFT 0xd
+#define PA_SC_ENHANCE__DISABLE_EOV_ALL_CTRL_ONLY_COMBINATIONS_MASK 0x4000
+#define PA_SC_ENHANCE__DISABLE_EOV_ALL_CTRL_ONLY_COMBINATIONS__SHIFT 0xe
+#define PA_SC_ENHANCE__ENABLE_MULTICYCLE_BUBBLE_FREEZE_MASK 0x8000
+#define PA_SC_ENHANCE__ENABLE_MULTICYCLE_BUBBLE_FREEZE__SHIFT 0xf
+#define PA_SC_ENHANCE__DISABLE_OUT_OF_ORDER_PA_SC_GUIDANCE_MASK 0x10000
+#define PA_SC_ENHANCE__DISABLE_OUT_OF_ORDER_PA_SC_GUIDANCE__SHIFT 0x10
+#define PA_SC_ENHANCE__ENABLE_OUT_OF_ORDER_POLY_MODE_MASK 0x20000
+#define PA_SC_ENHANCE__ENABLE_OUT_OF_ORDER_POLY_MODE__SHIFT 0x11
+#define PA_SC_ENHANCE__DISABLE_OUT_OF_ORDER_EOP_SYNC_NULL_PRIMS_LAST_MASK 0x40000
+#define PA_SC_ENHANCE__DISABLE_OUT_OF_ORDER_EOP_SYNC_NULL_PRIMS_LAST__SHIFT 0x12
+#define PA_SC_ENHANCE__DISABLE_OUT_OF_ORDER_THRESHOLD_SWITCHING_MASK 0x80000
+#define PA_SC_ENHANCE__DISABLE_OUT_OF_ORDER_THRESHOLD_SWITCHING__SHIFT 0x13
+#define PA_SC_ENHANCE__ENABLE_OUT_OF_ORDER_THRESHOLD_SWITCH_AT_EOPG_ONLY_MASK 0x100000
+#define PA_SC_ENHANCE__ENABLE_OUT_OF_ORDER_THRESHOLD_SWITCH_AT_EOPG_ONLY__SHIFT 0x14
+#define PA_SC_ENHANCE__DISABLE_OUT_OF_ORDER_DESIRED_FIFO_EMPTY_SWITCHING_MASK 0x200000
+#define PA_SC_ENHANCE__DISABLE_OUT_OF_ORDER_DESIRED_FIFO_EMPTY_SWITCHING__SHIFT 0x15
+#define PA_SC_ENHANCE__DISABLE_OUT_OF_ORDER_SELECTED_FIFO_EMPTY_SWITCHING_MASK 0x400000
+#define PA_SC_ENHANCE__DISABLE_OUT_OF_ORDER_SELECTED_FIFO_EMPTY_SWITCHING__SHIFT 0x16
+#define PA_SC_ENHANCE__DISABLE_OUT_OF_ORDER_EMPTY_SWITCHING_HYSTERYSIS_MASK 0x800000
+#define PA_SC_ENHANCE__DISABLE_OUT_OF_ORDER_EMPTY_SWITCHING_HYSTERYSIS__SHIFT 0x17
+#define PA_SC_ENHANCE__ENABLE_OUT_OF_ORDER_DESIRED_FIFO_IS_NEXT_FEID_MASK 0x1000000
+#define PA_SC_ENHANCE__ENABLE_OUT_OF_ORDER_DESIRED_FIFO_IS_NEXT_FEID__SHIFT 0x18
+#define PA_SC_ENHANCE__DISABLE_OOO_NO_EOPG_SKEW_DESIRED_FIFO_IS_CURRENT_FIFO_MASK 0x2000000
+#define PA_SC_ENHANCE__DISABLE_OOO_NO_EOPG_SKEW_DESIRED_FIFO_IS_CURRENT_FIFO__SHIFT 0x19
+#define PA_SC_ENHANCE__OOO_DISABLE_EOP_ON_FIRST_LIVE_PRIM_HIT_MASK 0x4000000
+#define PA_SC_ENHANCE__OOO_DISABLE_EOP_ON_FIRST_LIVE_PRIM_HIT__SHIFT 0x1a
+#define PA_SC_ENHANCE__OOO_DISABLE_EOPG_SKEW_THRESHOLD_SWITCHING_MASK 0x8000000
+#define PA_SC_ENHANCE__OOO_DISABLE_EOPG_SKEW_THRESHOLD_SWITCHING__SHIFT 0x1b
+#define PA_SC_ENHANCE__DISABLE_EOP_LINE_STIPPLE_RESET_MASK 0x10000000
+#define PA_SC_ENHANCE__DISABLE_EOP_LINE_STIPPLE_RESET__SHIFT 0x1c
+#define PA_SC_ENHANCE__DISABLE_VPZ_EOP_LINE_STIPPLE_RESET_MASK 0x20000000
+#define PA_SC_ENHANCE__DISABLE_VPZ_EOP_LINE_STIPPLE_RESET__SHIFT 0x1d
+#define PA_SC_ENHANCE__ECO_SPARE1_MASK 0x40000000
+#define PA_SC_ENHANCE__ECO_SPARE1__SHIFT 0x1e
+#define PA_SC_ENHANCE__ECO_SPARE0_MASK 0x80000000
+#define PA_SC_ENHANCE__ECO_SPARE0__SHIFT 0x1f
+#define PA_SC_ENHANCE_1__REALIGN_DQUADS_OVERRIDE_ENABLE_MASK 0x1
+#define PA_SC_ENHANCE_1__REALIGN_DQUADS_OVERRIDE_ENABLE__SHIFT 0x0
+#define PA_SC_ENHANCE_1__REALIGN_DQUADS_OVERRIDE_MASK 0x6
+#define PA_SC_ENHANCE_1__REALIGN_DQUADS_OVERRIDE__SHIFT 0x1
+#define PA_SC_ENHANCE_1__ENABLE_SC_BINNING_MASK 0x8
+#define PA_SC_ENHANCE_1__ENABLE_SC_BINNING__SHIFT 0x3
+#define PA_SC_ENHANCE_1__ECO_SPARE0_MASK 0x10
+#define PA_SC_ENHANCE_1__ECO_SPARE0__SHIFT 0x4
+#define PA_SC_ENHANCE_1__ECO_SPARE1_MASK 0x20
+#define PA_SC_ENHANCE_1__ECO_SPARE1__SHIFT 0x5
+#define PA_SC_ENHANCE_1__ECO_SPARE2_MASK 0x40
+#define PA_SC_ENHANCE_1__ECO_SPARE2__SHIFT 0x6
+#define PA_SC_ENHANCE_1__ECO_SPARE3_MASK 0x80
+#define PA_SC_ENHANCE_1__ECO_SPARE3__SHIFT 0x7
+#define PA_SC_DSM_CNTL__FORCE_EOV_REZ_0_MASK 0x1
+#define PA_SC_DSM_CNTL__FORCE_EOV_REZ_0__SHIFT 0x0
+#define PA_SC_DSM_CNTL__FORCE_EOV_REZ_1_MASK 0x2
+#define PA_SC_DSM_CNTL__FORCE_EOV_REZ_1__SHIFT 0x1
+#define PA_SC_FIFO_SIZE__SC_FRONTEND_PRIM_FIFO_SIZE_MASK 0x3f
+#define PA_SC_FIFO_SIZE__SC_FRONTEND_PRIM_FIFO_SIZE__SHIFT 0x0
+#define PA_SC_FIFO_SIZE__SC_BACKEND_PRIM_FIFO_SIZE_MASK 0x7fc0
+#define PA_SC_FIFO_SIZE__SC_BACKEND_PRIM_FIFO_SIZE__SHIFT 0x6
+#define PA_SC_FIFO_SIZE__SC_HIZ_TILE_FIFO_SIZE_MASK 0x1f8000
+#define PA_SC_FIFO_SIZE__SC_HIZ_TILE_FIFO_SIZE__SHIFT 0xf
+#define PA_SC_FIFO_SIZE__SC_EARLYZ_TILE_FIFO_SIZE_MASK 0xff800000
+#define PA_SC_FIFO_SIZE__SC_EARLYZ_TILE_FIFO_SIZE__SHIFT 0x17
+#define PA_SC_IF_FIFO_SIZE__SC_DB_TILE_IF_FIFO_SIZE_MASK 0x3f
+#define PA_SC_IF_FIFO_SIZE__SC_DB_TILE_IF_FIFO_SIZE__SHIFT 0x0
+#define PA_SC_IF_FIFO_SIZE__SC_DB_QUAD_IF_FIFO_SIZE_MASK 0xfc0
+#define PA_SC_IF_FIFO_SIZE__SC_DB_QUAD_IF_FIFO_SIZE__SHIFT 0x6
+#define PA_SC_IF_FIFO_SIZE__SC_SPI_IF_FIFO_SIZE_MASK 0x3f000
+#define PA_SC_IF_FIFO_SIZE__SC_SPI_IF_FIFO_SIZE__SHIFT 0xc
+#define PA_SC_IF_FIFO_SIZE__SC_BCI_IF_FIFO_SIZE_MASK 0xfc0000
+#define PA_SC_IF_FIFO_SIZE__SC_BCI_IF_FIFO_SIZE__SHIFT 0x12
+#define PA_SC_FORCE_EOV_MAX_CNTS__FORCE_EOV_MAX_CLK_CNT_MASK 0xffff
+#define PA_SC_FORCE_EOV_MAX_CNTS__FORCE_EOV_MAX_CLK_CNT__SHIFT 0x0
+#define PA_SC_FORCE_EOV_MAX_CNTS__FORCE_EOV_MAX_REZ_CNT_MASK 0xffff0000
+#define PA_SC_FORCE_EOV_MAX_CNTS__FORCE_EOV_MAX_REZ_CNT__SHIFT 0x10
+#define PA_SC_LINE_STIPPLE_STATE__CURRENT_PTR_MASK 0xf
+#define PA_SC_LINE_STIPPLE_STATE__CURRENT_PTR__SHIFT 0x0
+#define PA_SC_LINE_STIPPLE_STATE__CURRENT_COUNT_MASK 0xff00
+#define PA_SC_LINE_STIPPLE_STATE__CURRENT_COUNT__SHIFT 0x8
+#define PA_SC_SCREEN_EXTENT_MIN_0__X_MASK 0xffff
+#define PA_SC_SCREEN_EXTENT_MIN_0__X__SHIFT 0x0
+#define PA_SC_SCREEN_EXTENT_MIN_0__Y_MASK 0xffff0000
+#define PA_SC_SCREEN_EXTENT_MIN_0__Y__SHIFT 0x10
+#define PA_SC_SCREEN_EXTENT_MAX_0__X_MASK 0xffff
+#define PA_SC_SCREEN_EXTENT_MAX_0__X__SHIFT 0x0
+#define PA_SC_SCREEN_EXTENT_MAX_0__Y_MASK 0xffff0000
+#define PA_SC_SCREEN_EXTENT_MAX_0__Y__SHIFT 0x10
+#define PA_SC_SCREEN_EXTENT_MIN_1__X_MASK 0xffff
+#define PA_SC_SCREEN_EXTENT_MIN_1__X__SHIFT 0x0
+#define PA_SC_SCREEN_EXTENT_MIN_1__Y_MASK 0xffff0000
+#define PA_SC_SCREEN_EXTENT_MIN_1__Y__SHIFT 0x10
+#define PA_SC_SCREEN_EXTENT_MAX_1__X_MASK 0xffff
+#define PA_SC_SCREEN_EXTENT_MAX_1__X__SHIFT 0x0
+#define PA_SC_SCREEN_EXTENT_MAX_1__Y_MASK 0xffff0000
+#define PA_SC_SCREEN_EXTENT_MAX_1__Y__SHIFT 0x10
+#define PA_SC_PERFCOUNTER0_SELECT__PERF_SEL_MASK 0x3ff
+#define PA_SC_PERFCOUNTER0_SELECT__PERF_SEL__SHIFT 0x0
+#define PA_SC_PERFCOUNTER0_SELECT__PERF_SEL1_MASK 0xffc00
+#define PA_SC_PERFCOUNTER0_SELECT__PERF_SEL1__SHIFT 0xa
+#define PA_SC_PERFCOUNTER0_SELECT__CNTR_MODE_MASK 0xf00000
+#define PA_SC_PERFCOUNTER0_SELECT__CNTR_MODE__SHIFT 0x14
+#define PA_SC_PERFCOUNTER0_SELECT1__PERF_SEL2_MASK 0x3ff
+#define PA_SC_PERFCOUNTER0_SELECT1__PERF_SEL2__SHIFT 0x0
+#define PA_SC_PERFCOUNTER0_SELECT1__PERF_SEL3_MASK 0xffc00
+#define PA_SC_PERFCOUNTER0_SELECT1__PERF_SEL3__SHIFT 0xa
+#define PA_SC_PERFCOUNTER1_SELECT__PERF_SEL_MASK 0x3ff
+#define PA_SC_PERFCOUNTER1_SELECT__PERF_SEL__SHIFT 0x0
+#define PA_SC_PERFCOUNTER2_SELECT__PERF_SEL_MASK 0x3ff
+#define PA_SC_PERFCOUNTER2_SELECT__PERF_SEL__SHIFT 0x0
+#define PA_SC_PERFCOUNTER3_SELECT__PERF_SEL_MASK 0x3ff
+#define PA_SC_PERFCOUNTER3_SELECT__PERF_SEL__SHIFT 0x0
+#define PA_SC_PERFCOUNTER4_SELECT__PERF_SEL_MASK 0x3ff
+#define PA_SC_PERFCOUNTER4_SELECT__PERF_SEL__SHIFT 0x0
+#define PA_SC_PERFCOUNTER5_SELECT__PERF_SEL_MASK 0x3ff
+#define PA_SC_PERFCOUNTER5_SELECT__PERF_SEL__SHIFT 0x0
+#define PA_SC_PERFCOUNTER6_SELECT__PERF_SEL_MASK 0x3ff
+#define PA_SC_PERFCOUNTER6_SELECT__PERF_SEL__SHIFT 0x0
+#define PA_SC_PERFCOUNTER7_SELECT__PERF_SEL_MASK 0x3ff
+#define PA_SC_PERFCOUNTER7_SELECT__PERF_SEL__SHIFT 0x0
+#define PA_SC_PERFCOUNTER0_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define PA_SC_PERFCOUNTER0_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define PA_SC_PERFCOUNTER0_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define PA_SC_PERFCOUNTER0_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define PA_SC_PERFCOUNTER1_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define PA_SC_PERFCOUNTER1_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define PA_SC_PERFCOUNTER1_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define PA_SC_PERFCOUNTER1_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define PA_SC_PERFCOUNTER2_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define PA_SC_PERFCOUNTER2_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define PA_SC_PERFCOUNTER2_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define PA_SC_PERFCOUNTER2_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define PA_SC_PERFCOUNTER3_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define PA_SC_PERFCOUNTER3_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define PA_SC_PERFCOUNTER3_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define PA_SC_PERFCOUNTER3_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define PA_SC_PERFCOUNTER4_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define PA_SC_PERFCOUNTER4_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define PA_SC_PERFCOUNTER4_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define PA_SC_PERFCOUNTER4_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define PA_SC_PERFCOUNTER5_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define PA_SC_PERFCOUNTER5_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define PA_SC_PERFCOUNTER5_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define PA_SC_PERFCOUNTER5_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define PA_SC_PERFCOUNTER6_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define PA_SC_PERFCOUNTER6_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define PA_SC_PERFCOUNTER6_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define PA_SC_PERFCOUNTER6_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define PA_SC_PERFCOUNTER7_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define PA_SC_PERFCOUNTER7_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define PA_SC_PERFCOUNTER7_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define PA_SC_PERFCOUNTER7_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define PA_SC_P3D_TRAP_SCREEN_HV_EN__ENABLE_HV_PRE_SHADER_MASK 0x1
+#define PA_SC_P3D_TRAP_SCREEN_HV_EN__ENABLE_HV_PRE_SHADER__SHIFT 0x0
+#define PA_SC_P3D_TRAP_SCREEN_HV_EN__FORCE_PRE_SHADER_ALL_PIXELS_MASK 0x2
+#define PA_SC_P3D_TRAP_SCREEN_HV_EN__FORCE_PRE_SHADER_ALL_PIXELS__SHIFT 0x1
+#define PA_SC_P3D_TRAP_SCREEN_H__X_COORD_MASK 0x3fff
+#define PA_SC_P3D_TRAP_SCREEN_H__X_COORD__SHIFT 0x0
+#define PA_SC_P3D_TRAP_SCREEN_V__Y_COORD_MASK 0x3fff
+#define PA_SC_P3D_TRAP_SCREEN_V__Y_COORD__SHIFT 0x0
+#define PA_SC_P3D_TRAP_SCREEN_OCCURRENCE__COUNT_MASK 0xffff
+#define PA_SC_P3D_TRAP_SCREEN_OCCURRENCE__COUNT__SHIFT 0x0
+#define PA_SC_P3D_TRAP_SCREEN_COUNT__COUNT_MASK 0xffff
+#define PA_SC_P3D_TRAP_SCREEN_COUNT__COUNT__SHIFT 0x0
+#define PA_SC_HP3D_TRAP_SCREEN_HV_EN__ENABLE_HV_PRE_SHADER_MASK 0x1
+#define PA_SC_HP3D_TRAP_SCREEN_HV_EN__ENABLE_HV_PRE_SHADER__SHIFT 0x0
+#define PA_SC_HP3D_TRAP_SCREEN_HV_EN__FORCE_PRE_SHADER_ALL_PIXELS_MASK 0x2
+#define PA_SC_HP3D_TRAP_SCREEN_HV_EN__FORCE_PRE_SHADER_ALL_PIXELS__SHIFT 0x1
+#define PA_SC_HP3D_TRAP_SCREEN_H__X_COORD_MASK 0x3fff
+#define PA_SC_HP3D_TRAP_SCREEN_H__X_COORD__SHIFT 0x0
+#define PA_SC_HP3D_TRAP_SCREEN_V__Y_COORD_MASK 0x3fff
+#define PA_SC_HP3D_TRAP_SCREEN_V__Y_COORD__SHIFT 0x0
+#define PA_SC_HP3D_TRAP_SCREEN_OCCURRENCE__COUNT_MASK 0xffff
+#define PA_SC_HP3D_TRAP_SCREEN_OCCURRENCE__COUNT__SHIFT 0x0
+#define PA_SC_HP3D_TRAP_SCREEN_COUNT__COUNT_MASK 0xffff
+#define PA_SC_HP3D_TRAP_SCREEN_COUNT__COUNT__SHIFT 0x0
+#define PA_SC_TRAP_SCREEN_HV_EN__ENABLE_HV_PRE_SHADER_MASK 0x1
+#define PA_SC_TRAP_SCREEN_HV_EN__ENABLE_HV_PRE_SHADER__SHIFT 0x0
+#define PA_SC_TRAP_SCREEN_HV_EN__FORCE_PRE_SHADER_ALL_PIXELS_MASK 0x2
+#define PA_SC_TRAP_SCREEN_HV_EN__FORCE_PRE_SHADER_ALL_PIXELS__SHIFT 0x1
+#define PA_SC_TRAP_SCREEN_H__X_COORD_MASK 0x3fff
+#define PA_SC_TRAP_SCREEN_H__X_COORD__SHIFT 0x0
+#define PA_SC_TRAP_SCREEN_V__Y_COORD_MASK 0x3fff
+#define PA_SC_TRAP_SCREEN_V__Y_COORD__SHIFT 0x0
+#define PA_SC_TRAP_SCREEN_OCCURRENCE__COUNT_MASK 0xffff
+#define PA_SC_TRAP_SCREEN_OCCURRENCE__COUNT__SHIFT 0x0
+#define PA_SC_TRAP_SCREEN_COUNT__COUNT_MASK 0xffff
+#define PA_SC_TRAP_SCREEN_COUNT__COUNT__SHIFT 0x0
+#define PA_SC_P3D_TRAP_SCREEN_HV_LOCK__DISABLE_NON_PRIV_WRITES_MASK 0x1
+#define PA_SC_P3D_TRAP_SCREEN_HV_LOCK__DISABLE_NON_PRIV_WRITES__SHIFT 0x0
+#define PA_SC_HP3D_TRAP_SCREEN_HV_LOCK__DISABLE_NON_PRIV_WRITES_MASK 0x1
+#define PA_SC_HP3D_TRAP_SCREEN_HV_LOCK__DISABLE_NON_PRIV_WRITES__SHIFT 0x0
+#define PA_SC_TRAP_SCREEN_HV_LOCK__DISABLE_NON_PRIV_WRITES_MASK 0x1
+#define PA_SC_TRAP_SCREEN_HV_LOCK__DISABLE_NON_PRIV_WRITES__SHIFT 0x0
+#define PA_CL_CNTL_STATUS__CL_BUSY_MASK 0x80000000
+#define PA_CL_CNTL_STATUS__CL_BUSY__SHIFT 0x1f
+#define PA_SU_CNTL_STATUS__SU_BUSY_MASK 0x80000000
+#define PA_SU_CNTL_STATUS__SU_BUSY__SHIFT 0x1f
+#define PA_SC_FIFO_DEPTH_CNTL__DEPTH_MASK 0x3ff
+#define PA_SC_FIFO_DEPTH_CNTL__DEPTH__SHIFT 0x0
+#define CGTT_PA_CLK_CTRL__ON_DELAY_MASK 0xf
+#define CGTT_PA_CLK_CTRL__ON_DELAY__SHIFT 0x0
+#define CGTT_PA_CLK_CTRL__OFF_HYSTERESIS_MASK 0xff0
+#define CGTT_PA_CLK_CTRL__OFF_HYSTERESIS__SHIFT 0x4
+#define CGTT_PA_CLK_CTRL__SOFT_OVERRIDE7_MASK 0x1000000
+#define CGTT_PA_CLK_CTRL__SOFT_OVERRIDE7__SHIFT 0x18
+#define CGTT_PA_CLK_CTRL__SOFT_OVERRIDE6_MASK 0x2000000
+#define CGTT_PA_CLK_CTRL__SOFT_OVERRIDE6__SHIFT 0x19
+#define CGTT_PA_CLK_CTRL__SOFT_OVERRIDE5_MASK 0x4000000
+#define CGTT_PA_CLK_CTRL__SOFT_OVERRIDE5__SHIFT 0x1a
+#define CGTT_PA_CLK_CTRL__SOFT_OVERRIDE4_MASK 0x8000000
+#define CGTT_PA_CLK_CTRL__SOFT_OVERRIDE4__SHIFT 0x1b
+#define CGTT_PA_CLK_CTRL__SOFT_OVERRIDE3_MASK 0x10000000
+#define CGTT_PA_CLK_CTRL__SOFT_OVERRIDE3__SHIFT 0x1c
+#define CGTT_PA_CLK_CTRL__SU_CLK_OVERRIDE_MASK 0x20000000
+#define CGTT_PA_CLK_CTRL__SU_CLK_OVERRIDE__SHIFT 0x1d
+#define CGTT_PA_CLK_CTRL__CL_CLK_OVERRIDE_MASK 0x40000000
+#define CGTT_PA_CLK_CTRL__CL_CLK_OVERRIDE__SHIFT 0x1e
+#define CGTT_PA_CLK_CTRL__REG_CLK_OVERRIDE_MASK 0x80000000
+#define CGTT_PA_CLK_CTRL__REG_CLK_OVERRIDE__SHIFT 0x1f
+#define CGTT_SC_CLK_CTRL__ON_DELAY_MASK 0xf
+#define CGTT_SC_CLK_CTRL__ON_DELAY__SHIFT 0x0
+#define CGTT_SC_CLK_CTRL__OFF_HYSTERESIS_MASK 0xff0
+#define CGTT_SC_CLK_CTRL__OFF_HYSTERESIS__SHIFT 0x4
+#define CGTT_SC_CLK_CTRL__SOFT_OVERRIDE7_MASK 0x1000000
+#define CGTT_SC_CLK_CTRL__SOFT_OVERRIDE7__SHIFT 0x18
+#define CGTT_SC_CLK_CTRL__SOFT_OVERRIDE6_MASK 0x2000000
+#define CGTT_SC_CLK_CTRL__SOFT_OVERRIDE6__SHIFT 0x19
+#define CGTT_SC_CLK_CTRL__SOFT_OVERRIDE5_MASK 0x4000000
+#define CGTT_SC_CLK_CTRL__SOFT_OVERRIDE5__SHIFT 0x1a
+#define CGTT_SC_CLK_CTRL__SOFT_OVERRIDE4_MASK 0x8000000
+#define CGTT_SC_CLK_CTRL__SOFT_OVERRIDE4__SHIFT 0x1b
+#define CGTT_SC_CLK_CTRL__SOFT_OVERRIDE3_MASK 0x10000000
+#define CGTT_SC_CLK_CTRL__SOFT_OVERRIDE3__SHIFT 0x1c
+#define CGTT_SC_CLK_CTRL__SOFT_OVERRIDE2_MASK 0x20000000
+#define CGTT_SC_CLK_CTRL__SOFT_OVERRIDE2__SHIFT 0x1d
+#define CGTT_SC_CLK_CTRL__SOFT_OVERRIDE1_MASK 0x40000000
+#define CGTT_SC_CLK_CTRL__SOFT_OVERRIDE1__SHIFT 0x1e
+#define CGTT_SC_CLK_CTRL__SOFT_OVERRIDE0_MASK 0x80000000
+#define CGTT_SC_CLK_CTRL__SOFT_OVERRIDE0__SHIFT 0x1f
+#define PA_SU_DEBUG_CNTL__SU_DEBUG_INDX_MASK 0x1f
+#define PA_SU_DEBUG_CNTL__SU_DEBUG_INDX__SHIFT 0x0
+#define PA_SU_DEBUG_DATA__DATA_MASK 0xffffffff
+#define PA_SU_DEBUG_DATA__DATA__SHIFT 0x0
+#define PA_SC_DEBUG_CNTL__SC_DEBUG_INDX_MASK 0x3f
+#define PA_SC_DEBUG_CNTL__SC_DEBUG_INDX__SHIFT 0x0
+#define PA_SC_DEBUG_DATA__DATA_MASK 0xffffffff
+#define PA_SC_DEBUG_DATA__DATA__SHIFT 0x0
+#define CLIPPER_DEBUG_REG00__ALWAYS_ZERO_MASK 0xff
+#define CLIPPER_DEBUG_REG00__ALWAYS_ZERO__SHIFT 0x0
+#define CLIPPER_DEBUG_REG00__clip_ga_bc_fifo_write_MASK 0x100
+#define CLIPPER_DEBUG_REG00__clip_ga_bc_fifo_write__SHIFT 0x8
+#define CLIPPER_DEBUG_REG00__su_clip_baryc_free_MASK 0x600
+#define CLIPPER_DEBUG_REG00__su_clip_baryc_free__SHIFT 0x9
+#define CLIPPER_DEBUG_REG00__clip_to_ga_fifo_write_MASK 0x800
+#define CLIPPER_DEBUG_REG00__clip_to_ga_fifo_write__SHIFT 0xb
+#define CLIPPER_DEBUG_REG00__clip_to_ga_fifo_full_MASK 0x1000
+#define CLIPPER_DEBUG_REG00__clip_to_ga_fifo_full__SHIFT 0xc
+#define CLIPPER_DEBUG_REG00__primic_to_clprim_fifo_empty_MASK 0x2000
+#define CLIPPER_DEBUG_REG00__primic_to_clprim_fifo_empty__SHIFT 0xd
+#define CLIPPER_DEBUG_REG00__primic_to_clprim_fifo_full_MASK 0x4000
+#define CLIPPER_DEBUG_REG00__primic_to_clprim_fifo_full__SHIFT 0xe
+#define CLIPPER_DEBUG_REG00__clip_to_outsm_fifo_empty_MASK 0x8000
+#define CLIPPER_DEBUG_REG00__clip_to_outsm_fifo_empty__SHIFT 0xf
+#define CLIPPER_DEBUG_REG00__clip_to_outsm_fifo_full_MASK 0x10000
+#define CLIPPER_DEBUG_REG00__clip_to_outsm_fifo_full__SHIFT 0x10
+#define CLIPPER_DEBUG_REG00__vgt_to_clipp_fifo_empty_MASK 0x20000
+#define CLIPPER_DEBUG_REG00__vgt_to_clipp_fifo_empty__SHIFT 0x11
+#define CLIPPER_DEBUG_REG00__vgt_to_clipp_fifo_full_MASK 0x40000
+#define CLIPPER_DEBUG_REG00__vgt_to_clipp_fifo_full__SHIFT 0x12
+#define CLIPPER_DEBUG_REG00__vgt_to_clips_fifo_empty_MASK 0x80000
+#define CLIPPER_DEBUG_REG00__vgt_to_clips_fifo_empty__SHIFT 0x13
+#define CLIPPER_DEBUG_REG00__vgt_to_clips_fifo_full_MASK 0x100000
+#define CLIPPER_DEBUG_REG00__vgt_to_clips_fifo_full__SHIFT 0x14
+#define CLIPPER_DEBUG_REG00__clipcode_fifo_fifo_empty_MASK 0x200000
+#define CLIPPER_DEBUG_REG00__clipcode_fifo_fifo_empty__SHIFT 0x15
+#define CLIPPER_DEBUG_REG00__clipcode_fifo_full_MASK 0x400000
+#define CLIPPER_DEBUG_REG00__clipcode_fifo_full__SHIFT 0x16
+#define CLIPPER_DEBUG_REG00__vte_out_clip_fifo_fifo_empty_MASK 0x800000
+#define CLIPPER_DEBUG_REG00__vte_out_clip_fifo_fifo_empty__SHIFT 0x17
+#define CLIPPER_DEBUG_REG00__vte_out_clip_fifo_fifo_full_MASK 0x1000000
+#define CLIPPER_DEBUG_REG00__vte_out_clip_fifo_fifo_full__SHIFT 0x18
+#define CLIPPER_DEBUG_REG00__vte_out_orig_fifo_fifo_empty_MASK 0x2000000
+#define CLIPPER_DEBUG_REG00__vte_out_orig_fifo_fifo_empty__SHIFT 0x19
+#define CLIPPER_DEBUG_REG00__vte_out_orig_fifo_fifo_full_MASK 0x4000000
+#define CLIPPER_DEBUG_REG00__vte_out_orig_fifo_fifo_full__SHIFT 0x1a
+#define CLIPPER_DEBUG_REG00__ccgen_to_clipcc_fifo_empty_MASK 0x8000000
+#define CLIPPER_DEBUG_REG00__ccgen_to_clipcc_fifo_empty__SHIFT 0x1b
+#define CLIPPER_DEBUG_REG00__ccgen_to_clipcc_fifo_full_MASK 0x10000000
+#define CLIPPER_DEBUG_REG00__ccgen_to_clipcc_fifo_full__SHIFT 0x1c
+#define CLIPPER_DEBUG_REG00__clip_to_outsm_fifo_write_MASK 0x20000000
+#define CLIPPER_DEBUG_REG00__clip_to_outsm_fifo_write__SHIFT 0x1d
+#define CLIPPER_DEBUG_REG00__vte_out_orig_fifo_fifo_write_MASK 0x40000000
+#define CLIPPER_DEBUG_REG00__vte_out_orig_fifo_fifo_write__SHIFT 0x1e
+#define CLIPPER_DEBUG_REG00__vgt_to_clipp_fifo_write_MASK 0x80000000
+#define CLIPPER_DEBUG_REG00__vgt_to_clipp_fifo_write__SHIFT 0x1f
+#define CLIPPER_DEBUG_REG01__ALWAYS_ZERO_MASK 0xff
+#define CLIPPER_DEBUG_REG01__ALWAYS_ZERO__SHIFT 0x0
+#define CLIPPER_DEBUG_REG01__clip_extra_bc_valid_MASK 0x700
+#define CLIPPER_DEBUG_REG01__clip_extra_bc_valid__SHIFT 0x8
+#define CLIPPER_DEBUG_REG01__clip_vert_vte_valid_MASK 0x3800
+#define CLIPPER_DEBUG_REG01__clip_vert_vte_valid__SHIFT 0xb
+#define CLIPPER_DEBUG_REG01__clip_to_outsm_vertex_deallocate_MASK 0x1c000
+#define CLIPPER_DEBUG_REG01__clip_to_outsm_vertex_deallocate__SHIFT 0xe
+#define CLIPPER_DEBUG_REG01__clip_to_outsm_deallocate_slot_MASK 0xe0000
+#define CLIPPER_DEBUG_REG01__clip_to_outsm_deallocate_slot__SHIFT 0x11
+#define CLIPPER_DEBUG_REG01__clip_to_outsm_null_primitive_MASK 0x100000
+#define CLIPPER_DEBUG_REG01__clip_to_outsm_null_primitive__SHIFT 0x14
+#define CLIPPER_DEBUG_REG01__vte_positions_vte_clip_vte_naninf_kill_2_MASK 0x200000
+#define CLIPPER_DEBUG_REG01__vte_positions_vte_clip_vte_naninf_kill_2__SHIFT 0x15
+#define CLIPPER_DEBUG_REG01__vte_positions_vte_clip_vte_naninf_kill_1_MASK 0x400000
+#define CLIPPER_DEBUG_REG01__vte_positions_vte_clip_vte_naninf_kill_1__SHIFT 0x16
+#define CLIPPER_DEBUG_REG01__vte_positions_vte_clip_vte_naninf_kill_0_MASK 0x800000
+#define CLIPPER_DEBUG_REG01__vte_positions_vte_clip_vte_naninf_kill_0__SHIFT 0x17
+#define CLIPPER_DEBUG_REG01__vte_out_clip_rd_extra_bc_valid_MASK 0x1000000
+#define CLIPPER_DEBUG_REG01__vte_out_clip_rd_extra_bc_valid__SHIFT 0x18
+#define CLIPPER_DEBUG_REG01__vte_out_clip_rd_vte_naninf_kill_MASK 0x2000000
+#define CLIPPER_DEBUG_REG01__vte_out_clip_rd_vte_naninf_kill__SHIFT 0x19
+#define CLIPPER_DEBUG_REG01__vte_out_clip_rd_vertex_store_indx_MASK 0xc000000
+#define CLIPPER_DEBUG_REG01__vte_out_clip_rd_vertex_store_indx__SHIFT 0x1a
+#define CLIPPER_DEBUG_REG01__clip_ga_bc_fifo_write_MASK 0x10000000
+#define CLIPPER_DEBUG_REG01__clip_ga_bc_fifo_write__SHIFT 0x1c
+#define CLIPPER_DEBUG_REG01__clip_to_ga_fifo_write_MASK 0x20000000
+#define CLIPPER_DEBUG_REG01__clip_to_ga_fifo_write__SHIFT 0x1d
+#define CLIPPER_DEBUG_REG01__vte_out_clip_fifo_fifo_advanceread_MASK 0x40000000
+#define CLIPPER_DEBUG_REG01__vte_out_clip_fifo_fifo_advanceread__SHIFT 0x1e
+#define CLIPPER_DEBUG_REG01__vte_out_clip_fifo_fifo_empty_MASK 0x80000000
+#define CLIPPER_DEBUG_REG01__vte_out_clip_fifo_fifo_empty__SHIFT 0x1f
+#define CLIPPER_DEBUG_REG02__clip_extra_bc_valid_MASK 0x7
+#define CLIPPER_DEBUG_REG02__clip_extra_bc_valid__SHIFT 0x0
+#define CLIPPER_DEBUG_REG02__clip_vert_vte_valid_MASK 0x38
+#define CLIPPER_DEBUG_REG02__clip_vert_vte_valid__SHIFT 0x3
+#define CLIPPER_DEBUG_REG02__clip_to_outsm_clip_seq_indx_MASK 0xc0
+#define CLIPPER_DEBUG_REG02__clip_to_outsm_clip_seq_indx__SHIFT 0x6
+#define CLIPPER_DEBUG_REG02__clip_to_outsm_vertex_store_indx_2_MASK 0xf00
+#define CLIPPER_DEBUG_REG02__clip_to_outsm_vertex_store_indx_2__SHIFT 0x8
+#define CLIPPER_DEBUG_REG02__clip_to_outsm_vertex_store_indx_1_MASK 0xf000
+#define CLIPPER_DEBUG_REG02__clip_to_outsm_vertex_store_indx_1__SHIFT 0xc
+#define CLIPPER_DEBUG_REG02__clip_to_outsm_vertex_store_indx_0_MASK 0xf0000
+#define CLIPPER_DEBUG_REG02__clip_to_outsm_vertex_store_indx_0__SHIFT 0x10
+#define CLIPPER_DEBUG_REG02__clip_to_clipga_extra_bc_coords_MASK 0x100000
+#define CLIPPER_DEBUG_REG02__clip_to_clipga_extra_bc_coords__SHIFT 0x14
+#define CLIPPER_DEBUG_REG02__clip_to_clipga_vte_naninf_kill_MASK 0x200000
+#define CLIPPER_DEBUG_REG02__clip_to_clipga_vte_naninf_kill__SHIFT 0x15
+#define CLIPPER_DEBUG_REG02__clip_to_outsm_end_of_packet_MASK 0x400000
+#define CLIPPER_DEBUG_REG02__clip_to_outsm_end_of_packet__SHIFT 0x16
+#define CLIPPER_DEBUG_REG02__clip_to_outsm_first_prim_of_slot_MASK 0x800000
+#define CLIPPER_DEBUG_REG02__clip_to_outsm_first_prim_of_slot__SHIFT 0x17
+#define CLIPPER_DEBUG_REG02__clip_to_outsm_clipped_prim_MASK 0x1000000
+#define CLIPPER_DEBUG_REG02__clip_to_outsm_clipped_prim__SHIFT 0x18
+#define CLIPPER_DEBUG_REG02__clip_to_outsm_null_primitive_MASK 0x2000000
+#define CLIPPER_DEBUG_REG02__clip_to_outsm_null_primitive__SHIFT 0x19
+#define CLIPPER_DEBUG_REG02__clip_ga_bc_fifo_full_MASK 0x4000000
+#define CLIPPER_DEBUG_REG02__clip_ga_bc_fifo_full__SHIFT 0x1a
+#define CLIPPER_DEBUG_REG02__clip_to_ga_fifo_full_MASK 0x8000000
+#define CLIPPER_DEBUG_REG02__clip_to_ga_fifo_full__SHIFT 0x1b
+#define CLIPPER_DEBUG_REG02__clip_ga_bc_fifo_write_MASK 0x10000000
+#define CLIPPER_DEBUG_REG02__clip_ga_bc_fifo_write__SHIFT 0x1c
+#define CLIPPER_DEBUG_REG02__clip_to_ga_fifo_write_MASK 0x20000000
+#define CLIPPER_DEBUG_REG02__clip_to_ga_fifo_write__SHIFT 0x1d
+#define CLIPPER_DEBUG_REG02__clip_to_outsm_fifo_advanceread_MASK 0x40000000
+#define CLIPPER_DEBUG_REG02__clip_to_outsm_fifo_advanceread__SHIFT 0x1e
+#define CLIPPER_DEBUG_REG02__clip_to_outsm_fifo_empty_MASK 0x80000000
+#define CLIPPER_DEBUG_REG02__clip_to_outsm_fifo_empty__SHIFT 0x1f
+#define CLIPPER_DEBUG_REG03__clipsm0_clprim_to_clip_clip_code_or_MASK 0x3fff
+#define CLIPPER_DEBUG_REG03__clipsm0_clprim_to_clip_clip_code_or__SHIFT 0x0
+#define CLIPPER_DEBUG_REG03__clipsm0_clprim_to_clip_event_id_MASK 0xfc000
+#define CLIPPER_DEBUG_REG03__clipsm0_clprim_to_clip_event_id__SHIFT 0xe
+#define CLIPPER_DEBUG_REG03__clipsm0_clprim_to_clip_state_var_indx_MASK 0x700000
+#define CLIPPER_DEBUG_REG03__clipsm0_clprim_to_clip_state_var_indx__SHIFT 0x14
+#define CLIPPER_DEBUG_REG03__clipsm0_clprim_to_clip_clip_primitive_MASK 0x800000
+#define CLIPPER_DEBUG_REG03__clipsm0_clprim_to_clip_clip_primitive__SHIFT 0x17
+#define CLIPPER_DEBUG_REG03__clipsm0_clprim_to_clip_deallocate_slot_MASK 0x7000000
+#define CLIPPER_DEBUG_REG03__clipsm0_clprim_to_clip_deallocate_slot__SHIFT 0x18
+#define CLIPPER_DEBUG_REG03__clipsm0_clprim_to_clip_first_prim_of_slot_MASK 0x8000000
+#define CLIPPER_DEBUG_REG03__clipsm0_clprim_to_clip_first_prim_of_slot__SHIFT 0x1b
+#define CLIPPER_DEBUG_REG03__clipsm0_clprim_to_clip_end_of_packet_MASK 0x10000000
+#define CLIPPER_DEBUG_REG03__clipsm0_clprim_to_clip_end_of_packet__SHIFT 0x1c
+#define CLIPPER_DEBUG_REG03__clipsm0_clprim_to_clip_event_MASK 0x20000000
+#define CLIPPER_DEBUG_REG03__clipsm0_clprim_to_clip_event__SHIFT 0x1d
+#define CLIPPER_DEBUG_REG03__clipsm0_clprim_to_clip_null_primitive_MASK 0x40000000
+#define CLIPPER_DEBUG_REG03__clipsm0_clprim_to_clip_null_primitive__SHIFT 0x1e
+#define CLIPPER_DEBUG_REG03__clipsm0_clprim_to_clip_prim_valid_MASK 0x80000000
+#define CLIPPER_DEBUG_REG03__clipsm0_clprim_to_clip_prim_valid__SHIFT 0x1f
+#define CLIPPER_DEBUG_REG04__clipsm0_clprim_to_clip_param_cache_indx_0_MASK 0x7fe
+#define CLIPPER_DEBUG_REG04__clipsm0_clprim_to_clip_param_cache_indx_0__SHIFT 0x1
+#define CLIPPER_DEBUG_REG04__clipsm0_clprim_to_clip_vertex_store_indx_2_MASK 0x1f800
+#define CLIPPER_DEBUG_REG04__clipsm0_clprim_to_clip_vertex_store_indx_2__SHIFT 0xb
+#define CLIPPER_DEBUG_REG04__clipsm0_clprim_to_clip_vertex_store_indx_1_MASK 0x7e0000
+#define CLIPPER_DEBUG_REG04__clipsm0_clprim_to_clip_vertex_store_indx_1__SHIFT 0x11
+#define CLIPPER_DEBUG_REG04__clipsm0_clprim_to_clip_vertex_store_indx_0_MASK 0x1f800000
+#define CLIPPER_DEBUG_REG04__clipsm0_clprim_to_clip_vertex_store_indx_0__SHIFT 0x17
+#define CLIPPER_DEBUG_REG04__clipsm0_clprim_to_clip_event_MASK 0x20000000
+#define CLIPPER_DEBUG_REG04__clipsm0_clprim_to_clip_event__SHIFT 0x1d
+#define CLIPPER_DEBUG_REG04__clipsm0_clprim_to_clip_null_primitive_MASK 0x40000000
+#define CLIPPER_DEBUG_REG04__clipsm0_clprim_to_clip_null_primitive__SHIFT 0x1e
+#define CLIPPER_DEBUG_REG04__clipsm0_clprim_to_clip_prim_valid_MASK 0x80000000
+#define CLIPPER_DEBUG_REG04__clipsm0_clprim_to_clip_prim_valid__SHIFT 0x1f
+#define CLIPPER_DEBUG_REG05__clipsm1_clprim_to_clip_clip_code_or_MASK 0x3fff
+#define CLIPPER_DEBUG_REG05__clipsm1_clprim_to_clip_clip_code_or__SHIFT 0x0
+#define CLIPPER_DEBUG_REG05__clipsm1_clprim_to_clip_event_id_MASK 0xfc000
+#define CLIPPER_DEBUG_REG05__clipsm1_clprim_to_clip_event_id__SHIFT 0xe
+#define CLIPPER_DEBUG_REG05__clipsm1_clprim_to_clip_state_var_indx_MASK 0x700000
+#define CLIPPER_DEBUG_REG05__clipsm1_clprim_to_clip_state_var_indx__SHIFT 0x14
+#define CLIPPER_DEBUG_REG05__clipsm1_clprim_to_clip_clip_primitive_MASK 0x800000
+#define CLIPPER_DEBUG_REG05__clipsm1_clprim_to_clip_clip_primitive__SHIFT 0x17
+#define CLIPPER_DEBUG_REG05__clipsm1_clprim_to_clip_deallocate_slot_MASK 0x7000000
+#define CLIPPER_DEBUG_REG05__clipsm1_clprim_to_clip_deallocate_slot__SHIFT 0x18
+#define CLIPPER_DEBUG_REG05__clipsm1_clprim_to_clip_first_prim_of_slot_MASK 0x8000000
+#define CLIPPER_DEBUG_REG05__clipsm1_clprim_to_clip_first_prim_of_slot__SHIFT 0x1b
+#define CLIPPER_DEBUG_REG05__clipsm1_clprim_to_clip_end_of_packet_MASK 0x10000000
+#define CLIPPER_DEBUG_REG05__clipsm1_clprim_to_clip_end_of_packet__SHIFT 0x1c
+#define CLIPPER_DEBUG_REG05__clipsm1_clprim_to_clip_event_MASK 0x20000000
+#define CLIPPER_DEBUG_REG05__clipsm1_clprim_to_clip_event__SHIFT 0x1d
+#define CLIPPER_DEBUG_REG05__clipsm1_clprim_to_clip_null_primitive_MASK 0x40000000
+#define CLIPPER_DEBUG_REG05__clipsm1_clprim_to_clip_null_primitive__SHIFT 0x1e
+#define CLIPPER_DEBUG_REG05__clipsm1_clprim_to_clip_prim_valid_MASK 0x80000000
+#define CLIPPER_DEBUG_REG05__clipsm1_clprim_to_clip_prim_valid__SHIFT 0x1f
+#define CLIPPER_DEBUG_REG06__clipsm1_clprim_to_clip_param_cache_indx_0_MASK 0x7fe
+#define CLIPPER_DEBUG_REG06__clipsm1_clprim_to_clip_param_cache_indx_0__SHIFT 0x1
+#define CLIPPER_DEBUG_REG06__clipsm1_clprim_to_clip_vertex_store_indx_2_MASK 0x1f800
+#define CLIPPER_DEBUG_REG06__clipsm1_clprim_to_clip_vertex_store_indx_2__SHIFT 0xb
+#define CLIPPER_DEBUG_REG06__clipsm1_clprim_to_clip_vertex_store_indx_1_MASK 0x7e0000
+#define CLIPPER_DEBUG_REG06__clipsm1_clprim_to_clip_vertex_store_indx_1__SHIFT 0x11
+#define CLIPPER_DEBUG_REG06__clipsm1_clprim_to_clip_vertex_store_indx_0_MASK 0x1f800000
+#define CLIPPER_DEBUG_REG06__clipsm1_clprim_to_clip_vertex_store_indx_0__SHIFT 0x17
+#define CLIPPER_DEBUG_REG06__clipsm1_clprim_to_clip_event_MASK 0x20000000
+#define CLIPPER_DEBUG_REG06__clipsm1_clprim_to_clip_event__SHIFT 0x1d
+#define CLIPPER_DEBUG_REG06__clipsm1_clprim_to_clip_null_primitive_MASK 0x40000000
+#define CLIPPER_DEBUG_REG06__clipsm1_clprim_to_clip_null_primitive__SHIFT 0x1e
+#define CLIPPER_DEBUG_REG06__clipsm1_clprim_to_clip_prim_valid_MASK 0x80000000
+#define CLIPPER_DEBUG_REG06__clipsm1_clprim_to_clip_prim_valid__SHIFT 0x1f
+#define CLIPPER_DEBUG_REG07__clipsm2_clprim_to_clip_clip_code_or_MASK 0x3fff
+#define CLIPPER_DEBUG_REG07__clipsm2_clprim_to_clip_clip_code_or__SHIFT 0x0
+#define CLIPPER_DEBUG_REG07__clipsm2_clprim_to_clip_event_id_MASK 0xfc000
+#define CLIPPER_DEBUG_REG07__clipsm2_clprim_to_clip_event_id__SHIFT 0xe
+#define CLIPPER_DEBUG_REG07__clipsm2_clprim_to_clip_state_var_indx_MASK 0x700000
+#define CLIPPER_DEBUG_REG07__clipsm2_clprim_to_clip_state_var_indx__SHIFT 0x14
+#define CLIPPER_DEBUG_REG07__clipsm2_clprim_to_clip_clip_primitive_MASK 0x800000
+#define CLIPPER_DEBUG_REG07__clipsm2_clprim_to_clip_clip_primitive__SHIFT 0x17
+#define CLIPPER_DEBUG_REG07__clipsm2_clprim_to_clip_deallocate_slot_MASK 0x7000000
+#define CLIPPER_DEBUG_REG07__clipsm2_clprim_to_clip_deallocate_slot__SHIFT 0x18
+#define CLIPPER_DEBUG_REG07__clipsm2_clprim_to_clip_first_prim_of_slot_MASK 0x8000000
+#define CLIPPER_DEBUG_REG07__clipsm2_clprim_to_clip_first_prim_of_slot__SHIFT 0x1b
+#define CLIPPER_DEBUG_REG07__clipsm2_clprim_to_clip_end_of_packet_MASK 0x10000000
+#define CLIPPER_DEBUG_REG07__clipsm2_clprim_to_clip_end_of_packet__SHIFT 0x1c
+#define CLIPPER_DEBUG_REG07__clipsm2_clprim_to_clip_event_MASK 0x20000000
+#define CLIPPER_DEBUG_REG07__clipsm2_clprim_to_clip_event__SHIFT 0x1d
+#define CLIPPER_DEBUG_REG07__clipsm2_clprim_to_clip_null_primitive_MASK 0x40000000
+#define CLIPPER_DEBUG_REG07__clipsm2_clprim_to_clip_null_primitive__SHIFT 0x1e
+#define CLIPPER_DEBUG_REG07__clipsm2_clprim_to_clip_prim_valid_MASK 0x80000000
+#define CLIPPER_DEBUG_REG07__clipsm2_clprim_to_clip_prim_valid__SHIFT 0x1f
+#define CLIPPER_DEBUG_REG08__clipsm2_clprim_to_clip_param_cache_indx_0_MASK 0x7fe
+#define CLIPPER_DEBUG_REG08__clipsm2_clprim_to_clip_param_cache_indx_0__SHIFT 0x1
+#define CLIPPER_DEBUG_REG08__clipsm2_clprim_to_clip_vertex_store_indx_2_MASK 0x1f800
+#define CLIPPER_DEBUG_REG08__clipsm2_clprim_to_clip_vertex_store_indx_2__SHIFT 0xb
+#define CLIPPER_DEBUG_REG08__clipsm2_clprim_to_clip_vertex_store_indx_1_MASK 0x7e0000
+#define CLIPPER_DEBUG_REG08__clipsm2_clprim_to_clip_vertex_store_indx_1__SHIFT 0x11
+#define CLIPPER_DEBUG_REG08__clipsm2_clprim_to_clip_vertex_store_indx_0_MASK 0x1f800000
+#define CLIPPER_DEBUG_REG08__clipsm2_clprim_to_clip_vertex_store_indx_0__SHIFT 0x17
+#define CLIPPER_DEBUG_REG08__clipsm2_clprim_to_clip_event_MASK 0x20000000
+#define CLIPPER_DEBUG_REG08__clipsm2_clprim_to_clip_event__SHIFT 0x1d
+#define CLIPPER_DEBUG_REG08__clipsm2_clprim_to_clip_null_primitive_MASK 0x40000000
+#define CLIPPER_DEBUG_REG08__clipsm2_clprim_to_clip_null_primitive__SHIFT 0x1e
+#define CLIPPER_DEBUG_REG08__clipsm2_clprim_to_clip_prim_valid_MASK 0x80000000
+#define CLIPPER_DEBUG_REG08__clipsm2_clprim_to_clip_prim_valid__SHIFT 0x1f
+#define CLIPPER_DEBUG_REG09__clipsm3_clprim_to_clip_clip_code_or_MASK 0x3fff
+#define CLIPPER_DEBUG_REG09__clipsm3_clprim_to_clip_clip_code_or__SHIFT 0x0
+#define CLIPPER_DEBUG_REG09__clipsm3_clprim_to_clip_event_id_MASK 0xfc000
+#define CLIPPER_DEBUG_REG09__clipsm3_clprim_to_clip_event_id__SHIFT 0xe
+#define CLIPPER_DEBUG_REG09__clipsm3_clprim_to_clip_state_var_indx_MASK 0x700000
+#define CLIPPER_DEBUG_REG09__clipsm3_clprim_to_clip_state_var_indx__SHIFT 0x14
+#define CLIPPER_DEBUG_REG09__clipsm3_clprim_to_clip_clip_primitive_MASK 0x800000
+#define CLIPPER_DEBUG_REG09__clipsm3_clprim_to_clip_clip_primitive__SHIFT 0x17
+#define CLIPPER_DEBUG_REG09__clipsm3_clprim_to_clip_deallocate_slot_MASK 0x7000000
+#define CLIPPER_DEBUG_REG09__clipsm3_clprim_to_clip_deallocate_slot__SHIFT 0x18
+#define CLIPPER_DEBUG_REG09__clipsm3_clprim_to_clip_first_prim_of_slot_MASK 0x8000000
+#define CLIPPER_DEBUG_REG09__clipsm3_clprim_to_clip_first_prim_of_slot__SHIFT 0x1b
+#define CLIPPER_DEBUG_REG09__clipsm3_clprim_to_clip_end_of_packet_MASK 0x10000000
+#define CLIPPER_DEBUG_REG09__clipsm3_clprim_to_clip_end_of_packet__SHIFT 0x1c
+#define CLIPPER_DEBUG_REG09__clipsm3_clprim_to_clip_event_MASK 0x20000000
+#define CLIPPER_DEBUG_REG09__clipsm3_clprim_to_clip_event__SHIFT 0x1d
+#define CLIPPER_DEBUG_REG09__clipsm3_clprim_to_clip_null_primitive_MASK 0x40000000
+#define CLIPPER_DEBUG_REG09__clipsm3_clprim_to_clip_null_primitive__SHIFT 0x1e
+#define CLIPPER_DEBUG_REG09__clipsm3_clprim_to_clip_prim_valid_MASK 0x80000000
+#define CLIPPER_DEBUG_REG09__clipsm3_clprim_to_clip_prim_valid__SHIFT 0x1f
+#define CLIPPER_DEBUG_REG10__clipsm3_clprim_to_clip_param_cache_indx_0_MASK 0x7fe
+#define CLIPPER_DEBUG_REG10__clipsm3_clprim_to_clip_param_cache_indx_0__SHIFT 0x1
+#define CLIPPER_DEBUG_REG10__clipsm3_clprim_to_clip_vertex_store_indx_2_MASK 0x1f800
+#define CLIPPER_DEBUG_REG10__clipsm3_clprim_to_clip_vertex_store_indx_2__SHIFT 0xb
+#define CLIPPER_DEBUG_REG10__clipsm3_clprim_to_clip_vertex_store_indx_1_MASK 0x7e0000
+#define CLIPPER_DEBUG_REG10__clipsm3_clprim_to_clip_vertex_store_indx_1__SHIFT 0x11
+#define CLIPPER_DEBUG_REG10__clipsm3_clprim_to_clip_vertex_store_indx_0_MASK 0x1f800000
+#define CLIPPER_DEBUG_REG10__clipsm3_clprim_to_clip_vertex_store_indx_0__SHIFT 0x17
+#define CLIPPER_DEBUG_REG10__clipsm3_clprim_to_clip_event_MASK 0x20000000
+#define CLIPPER_DEBUG_REG10__clipsm3_clprim_to_clip_event__SHIFT 0x1d
+#define CLIPPER_DEBUG_REG10__clipsm3_clprim_to_clip_null_primitive_MASK 0x40000000
+#define CLIPPER_DEBUG_REG10__clipsm3_clprim_to_clip_null_primitive__SHIFT 0x1e
+#define CLIPPER_DEBUG_REG10__clipsm3_clprim_to_clip_prim_valid_MASK 0x80000000
+#define CLIPPER_DEBUG_REG10__clipsm3_clprim_to_clip_prim_valid__SHIFT 0x1f
+#define CLIPPER_DEBUG_REG11__clipsm3_clip_to_clipga_event_MASK 0x1
+#define CLIPPER_DEBUG_REG11__clipsm3_clip_to_clipga_event__SHIFT 0x0
+#define CLIPPER_DEBUG_REG11__clipsm2_clip_to_clipga_event_MASK 0x2
+#define CLIPPER_DEBUG_REG11__clipsm2_clip_to_clipga_event__SHIFT 0x1
+#define CLIPPER_DEBUG_REG11__clipsm1_clip_to_clipga_event_MASK 0x4
+#define CLIPPER_DEBUG_REG11__clipsm1_clip_to_clipga_event__SHIFT 0x2
+#define CLIPPER_DEBUG_REG11__clipsm0_clip_to_clipga_event_MASK 0x8
+#define CLIPPER_DEBUG_REG11__clipsm0_clip_to_clipga_event__SHIFT 0x3
+#define CLIPPER_DEBUG_REG11__clipsm3_clip_to_clipga_clip_primitive_MASK 0x10
+#define CLIPPER_DEBUG_REG11__clipsm3_clip_to_clipga_clip_primitive__SHIFT 0x4
+#define CLIPPER_DEBUG_REG11__clipsm2_clip_to_clipga_clip_primitive_MASK 0x20
+#define CLIPPER_DEBUG_REG11__clipsm2_clip_to_clipga_clip_primitive__SHIFT 0x5
+#define CLIPPER_DEBUG_REG11__clipsm1_clip_to_clipga_clip_primitive_MASK 0x40
+#define CLIPPER_DEBUG_REG11__clipsm1_clip_to_clipga_clip_primitive__SHIFT 0x6
+#define CLIPPER_DEBUG_REG11__clipsm0_clip_to_clipga_clip_primitive_MASK 0x80
+#define CLIPPER_DEBUG_REG11__clipsm0_clip_to_clipga_clip_primitive__SHIFT 0x7
+#define CLIPPER_DEBUG_REG11__clipsm3_clip_to_clipga_clip_to_outsm_cnt_MASK 0xf00
+#define CLIPPER_DEBUG_REG11__clipsm3_clip_to_clipga_clip_to_outsm_cnt__SHIFT 0x8
+#define CLIPPER_DEBUG_REG11__clipsm2_clip_to_clipga_clip_to_outsm_cnt_MASK 0xf000
+#define CLIPPER_DEBUG_REG11__clipsm2_clip_to_clipga_clip_to_outsm_cnt__SHIFT 0xc
+#define CLIPPER_DEBUG_REG11__clipsm1_clip_to_clipga_clip_to_outsm_cnt_MASK 0xf0000
+#define CLIPPER_DEBUG_REG11__clipsm1_clip_to_clipga_clip_to_outsm_cnt__SHIFT 0x10
+#define CLIPPER_DEBUG_REG11__clipsm0_clip_to_clipga_clip_to_outsm_cnt_MASK 0xf00000
+#define CLIPPER_DEBUG_REG11__clipsm0_clip_to_clipga_clip_to_outsm_cnt__SHIFT 0x14
+#define CLIPPER_DEBUG_REG11__clipsm3_clip_to_clipga_prim_valid_MASK 0x1000000
+#define CLIPPER_DEBUG_REG11__clipsm3_clip_to_clipga_prim_valid__SHIFT 0x18
+#define CLIPPER_DEBUG_REG11__clipsm2_clip_to_clipga_prim_valid_MASK 0x2000000
+#define CLIPPER_DEBUG_REG11__clipsm2_clip_to_clipga_prim_valid__SHIFT 0x19
+#define CLIPPER_DEBUG_REG11__clipsm1_clip_to_clipga_prim_valid_MASK 0x4000000
+#define CLIPPER_DEBUG_REG11__clipsm1_clip_to_clipga_prim_valid__SHIFT 0x1a
+#define CLIPPER_DEBUG_REG11__clipsm0_clip_to_clipga_prim_valid_MASK 0x8000000
+#define CLIPPER_DEBUG_REG11__clipsm0_clip_to_clipga_prim_valid__SHIFT 0x1b
+#define CLIPPER_DEBUG_REG11__clipsm3_inc_clip_to_clipga_clip_to_outsm_cnt_MASK 0x10000000
+#define CLIPPER_DEBUG_REG11__clipsm3_inc_clip_to_clipga_clip_to_outsm_cnt__SHIFT 0x1c
+#define CLIPPER_DEBUG_REG11__clipsm2_inc_clip_to_clipga_clip_to_outsm_cnt_MASK 0x20000000
+#define CLIPPER_DEBUG_REG11__clipsm2_inc_clip_to_clipga_clip_to_outsm_cnt__SHIFT 0x1d
+#define CLIPPER_DEBUG_REG11__clipsm1_inc_clip_to_clipga_clip_to_outsm_cnt_MASK 0x40000000
+#define CLIPPER_DEBUG_REG11__clipsm1_inc_clip_to_clipga_clip_to_outsm_cnt__SHIFT 0x1e
+#define CLIPPER_DEBUG_REG11__clipsm0_inc_clip_to_clipga_clip_to_outsm_cnt_MASK 0x80000000
+#define CLIPPER_DEBUG_REG11__clipsm0_inc_clip_to_clipga_clip_to_outsm_cnt__SHIFT 0x1f
+#define CLIPPER_DEBUG_REG12__ALWAYS_ZERO_MASK 0xff
+#define CLIPPER_DEBUG_REG12__ALWAYS_ZERO__SHIFT 0x0
+#define CLIPPER_DEBUG_REG12__clip_priority_available_vte_out_clip_MASK 0x1f00
+#define CLIPPER_DEBUG_REG12__clip_priority_available_vte_out_clip__SHIFT 0x8
+#define CLIPPER_DEBUG_REG12__clip_priority_available_clip_verts_MASK 0x3e000
+#define CLIPPER_DEBUG_REG12__clip_priority_available_clip_verts__SHIFT 0xd
+#define CLIPPER_DEBUG_REG12__clip_priority_seq_indx_out_MASK 0xc0000
+#define CLIPPER_DEBUG_REG12__clip_priority_seq_indx_out__SHIFT 0x12
+#define CLIPPER_DEBUG_REG12__clip_priority_seq_indx_vert_MASK 0x300000
+#define CLIPPER_DEBUG_REG12__clip_priority_seq_indx_vert__SHIFT 0x14
+#define CLIPPER_DEBUG_REG12__clip_priority_seq_indx_load_MASK 0xc00000
+#define CLIPPER_DEBUG_REG12__clip_priority_seq_indx_load__SHIFT 0x16
+#define CLIPPER_DEBUG_REG12__clipsm3_clprim_to_clip_clip_primitive_MASK 0x1000000
+#define CLIPPER_DEBUG_REG12__clipsm3_clprim_to_clip_clip_primitive__SHIFT 0x18
+#define CLIPPER_DEBUG_REG12__clipsm3_clprim_to_clip_prim_valid_MASK 0x2000000
+#define CLIPPER_DEBUG_REG12__clipsm3_clprim_to_clip_prim_valid__SHIFT 0x19
+#define CLIPPER_DEBUG_REG12__clipsm2_clprim_to_clip_clip_primitive_MASK 0x4000000
+#define CLIPPER_DEBUG_REG12__clipsm2_clprim_to_clip_clip_primitive__SHIFT 0x1a
+#define CLIPPER_DEBUG_REG12__clipsm2_clprim_to_clip_prim_valid_MASK 0x8000000
+#define CLIPPER_DEBUG_REG12__clipsm2_clprim_to_clip_prim_valid__SHIFT 0x1b
+#define CLIPPER_DEBUG_REG12__clipsm1_clprim_to_clip_clip_primitive_MASK 0x10000000
+#define CLIPPER_DEBUG_REG12__clipsm1_clprim_to_clip_clip_primitive__SHIFT 0x1c
+#define CLIPPER_DEBUG_REG12__clipsm1_clprim_to_clip_prim_valid_MASK 0x20000000
+#define CLIPPER_DEBUG_REG12__clipsm1_clprim_to_clip_prim_valid__SHIFT 0x1d
+#define CLIPPER_DEBUG_REG12__clipsm0_clprim_to_clip_clip_primitive_MASK 0x40000000
+#define CLIPPER_DEBUG_REG12__clipsm0_clprim_to_clip_clip_primitive__SHIFT 0x1e
+#define CLIPPER_DEBUG_REG12__clipsm0_clprim_to_clip_prim_valid_MASK 0x80000000
+#define CLIPPER_DEBUG_REG12__clipsm0_clprim_to_clip_prim_valid__SHIFT 0x1f
+#define CLIPPER_DEBUG_REG13__clprim_in_back_state_var_indx_MASK 0x7
+#define CLIPPER_DEBUG_REG13__clprim_in_back_state_var_indx__SHIFT 0x0
+#define CLIPPER_DEBUG_REG13__point_clip_candidate_MASK 0x8
+#define CLIPPER_DEBUG_REG13__point_clip_candidate__SHIFT 0x3
+#define CLIPPER_DEBUG_REG13__prim_nan_kill_MASK 0x10
+#define CLIPPER_DEBUG_REG13__prim_nan_kill__SHIFT 0x4
+#define CLIPPER_DEBUG_REG13__clprim_clip_primitive_MASK 0x20
+#define CLIPPER_DEBUG_REG13__clprim_clip_primitive__SHIFT 0x5
+#define CLIPPER_DEBUG_REG13__clprim_cull_primitive_MASK 0x40
+#define CLIPPER_DEBUG_REG13__clprim_cull_primitive__SHIFT 0x6
+#define CLIPPER_DEBUG_REG13__prim_back_valid_MASK 0x80
+#define CLIPPER_DEBUG_REG13__prim_back_valid__SHIFT 0x7
+#define CLIPPER_DEBUG_REG13__vertval_bits_vertex_cc_next_valid_MASK 0xf00
+#define CLIPPER_DEBUG_REG13__vertval_bits_vertex_cc_next_valid__SHIFT 0x8
+#define CLIPPER_DEBUG_REG13__clipcc_vertex_store_indx_MASK 0x3000
+#define CLIPPER_DEBUG_REG13__clipcc_vertex_store_indx__SHIFT 0xc
+#define CLIPPER_DEBUG_REG13__vte_out_orig_fifo_fifo_empty_MASK 0x4000
+#define CLIPPER_DEBUG_REG13__vte_out_orig_fifo_fifo_empty__SHIFT 0xe
+#define CLIPPER_DEBUG_REG13__clipcode_fifo_fifo_empty_MASK 0x8000
+#define CLIPPER_DEBUG_REG13__clipcode_fifo_fifo_empty__SHIFT 0xf
+#define CLIPPER_DEBUG_REG13__ccgen_to_clipcc_fifo_empty_MASK 0x10000
+#define CLIPPER_DEBUG_REG13__ccgen_to_clipcc_fifo_empty__SHIFT 0x10
+#define CLIPPER_DEBUG_REG13__clip_priority_seq_indx_out_cnt_MASK 0x1e0000
+#define CLIPPER_DEBUG_REG13__clip_priority_seq_indx_out_cnt__SHIFT 0x11
+#define CLIPPER_DEBUG_REG13__outsm_clr_rd_orig_vertices_MASK 0x600000
+#define CLIPPER_DEBUG_REG13__outsm_clr_rd_orig_vertices__SHIFT 0x15
+#define CLIPPER_DEBUG_REG13__outsm_clr_rd_clipsm_wait_MASK 0x800000
+#define CLIPPER_DEBUG_REG13__outsm_clr_rd_clipsm_wait__SHIFT 0x17
+#define CLIPPER_DEBUG_REG13__outsm_clr_fifo_contents_MASK 0x1f000000
+#define CLIPPER_DEBUG_REG13__outsm_clr_fifo_contents__SHIFT 0x18
+#define CLIPPER_DEBUG_REG13__outsm_clr_fifo_full_MASK 0x20000000
+#define CLIPPER_DEBUG_REG13__outsm_clr_fifo_full__SHIFT 0x1d
+#define CLIPPER_DEBUG_REG13__outsm_clr_fifo_advanceread_MASK 0x40000000
+#define CLIPPER_DEBUG_REG13__outsm_clr_fifo_advanceread__SHIFT 0x1e
+#define CLIPPER_DEBUG_REG13__outsm_clr_fifo_write_MASK 0x80000000
+#define CLIPPER_DEBUG_REG13__outsm_clr_fifo_write__SHIFT 0x1f
+#define CLIPPER_DEBUG_REG14__clprim_in_back_vertex_store_indx_2_MASK 0x3f
+#define CLIPPER_DEBUG_REG14__clprim_in_back_vertex_store_indx_2__SHIFT 0x0
+#define CLIPPER_DEBUG_REG14__clprim_in_back_vertex_store_indx_1_MASK 0xfc0
+#define CLIPPER_DEBUG_REG14__clprim_in_back_vertex_store_indx_1__SHIFT 0x6
+#define CLIPPER_DEBUG_REG14__clprim_in_back_vertex_store_indx_0_MASK 0x3f000
+#define CLIPPER_DEBUG_REG14__clprim_in_back_vertex_store_indx_0__SHIFT 0xc
+#define CLIPPER_DEBUG_REG14__outputclprimtoclip_null_primitive_MASK 0x40000
+#define CLIPPER_DEBUG_REG14__outputclprimtoclip_null_primitive__SHIFT 0x12
+#define CLIPPER_DEBUG_REG14__clprim_in_back_end_of_packet_MASK 0x80000
+#define CLIPPER_DEBUG_REG14__clprim_in_back_end_of_packet__SHIFT 0x13
+#define CLIPPER_DEBUG_REG14__clprim_in_back_first_prim_of_slot_MASK 0x100000
+#define CLIPPER_DEBUG_REG14__clprim_in_back_first_prim_of_slot__SHIFT 0x14
+#define CLIPPER_DEBUG_REG14__clprim_in_back_deallocate_slot_MASK 0xe00000
+#define CLIPPER_DEBUG_REG14__clprim_in_back_deallocate_slot__SHIFT 0x15
+#define CLIPPER_DEBUG_REG14__clprim_in_back_event_id_MASK 0x3f000000
+#define CLIPPER_DEBUG_REG14__clprim_in_back_event_id__SHIFT 0x18
+#define CLIPPER_DEBUG_REG14__clprim_in_back_event_MASK 0x40000000
+#define CLIPPER_DEBUG_REG14__clprim_in_back_event__SHIFT 0x1e
+#define CLIPPER_DEBUG_REG14__prim_back_valid_MASK 0x80000000
+#define CLIPPER_DEBUG_REG14__prim_back_valid__SHIFT 0x1f
+#define CLIPPER_DEBUG_REG15__vertval_bits_vertex_vertex_store_msb_MASK 0xffff
+#define CLIPPER_DEBUG_REG15__vertval_bits_vertex_vertex_store_msb__SHIFT 0x0
+#define CLIPPER_DEBUG_REG15__primic_to_clprim_fifo_vertex_store_indx_2_MASK 0x1f0000
+#define CLIPPER_DEBUG_REG15__primic_to_clprim_fifo_vertex_store_indx_2__SHIFT 0x10
+#define CLIPPER_DEBUG_REG15__primic_to_clprim_fifo_vertex_store_indx_1_MASK 0x3e00000
+#define CLIPPER_DEBUG_REG15__primic_to_clprim_fifo_vertex_store_indx_1__SHIFT 0x15
+#define CLIPPER_DEBUG_REG15__primic_to_clprim_fifo_vertex_store_indx_0_MASK 0x7c000000
+#define CLIPPER_DEBUG_REG15__primic_to_clprim_fifo_vertex_store_indx_0__SHIFT 0x1a
+#define CLIPPER_DEBUG_REG15__primic_to_clprim_valid_MASK 0x80000000
+#define CLIPPER_DEBUG_REG15__primic_to_clprim_valid__SHIFT 0x1f
+#define CLIPPER_DEBUG_REG16__sm0_prim_end_state_MASK 0x7f
+#define CLIPPER_DEBUG_REG16__sm0_prim_end_state__SHIFT 0x0
+#define CLIPPER_DEBUG_REG16__sm0_ps_expand_MASK 0x80
+#define CLIPPER_DEBUG_REG16__sm0_ps_expand__SHIFT 0x7
+#define CLIPPER_DEBUG_REG16__sm0_clip_vert_cnt_MASK 0x1f00
+#define CLIPPER_DEBUG_REG16__sm0_clip_vert_cnt__SHIFT 0x8
+#define CLIPPER_DEBUG_REG16__sm0_vertex_clip_cnt_MASK 0x3e000
+#define CLIPPER_DEBUG_REG16__sm0_vertex_clip_cnt__SHIFT 0xd
+#define CLIPPER_DEBUG_REG16__sm0_inv_to_clip_data_valid_1_MASK 0x40000
+#define CLIPPER_DEBUG_REG16__sm0_inv_to_clip_data_valid_1__SHIFT 0x12
+#define CLIPPER_DEBUG_REG16__sm0_inv_to_clip_data_valid_0_MASK 0x80000
+#define CLIPPER_DEBUG_REG16__sm0_inv_to_clip_data_valid_0__SHIFT 0x13
+#define CLIPPER_DEBUG_REG16__sm0_current_state_MASK 0x7f00000
+#define CLIPPER_DEBUG_REG16__sm0_current_state__SHIFT 0x14
+#define CLIPPER_DEBUG_REG16__sm0_clip_to_clipga_clip_to_outsm_cnt_eq0_MASK 0x8000000
+#define CLIPPER_DEBUG_REG16__sm0_clip_to_clipga_clip_to_outsm_cnt_eq0__SHIFT 0x1b
+#define CLIPPER_DEBUG_REG16__sm0_clip_to_outsm_fifo_full_MASK 0x10000000
+#define CLIPPER_DEBUG_REG16__sm0_clip_to_outsm_fifo_full__SHIFT 0x1c
+#define CLIPPER_DEBUG_REG16__sm0_highest_priority_seq_MASK 0x20000000
+#define CLIPPER_DEBUG_REG16__sm0_highest_priority_seq__SHIFT 0x1d
+#define CLIPPER_DEBUG_REG16__sm0_outputcliptoclipga_0_MASK 0x40000000
+#define CLIPPER_DEBUG_REG16__sm0_outputcliptoclipga_0__SHIFT 0x1e
+#define CLIPPER_DEBUG_REG16__sm0_clprim_to_clip_prim_valid_MASK 0x80000000
+#define CLIPPER_DEBUG_REG16__sm0_clprim_to_clip_prim_valid__SHIFT 0x1f
+#define CLIPPER_DEBUG_REG17__sm1_prim_end_state_MASK 0x7f
+#define CLIPPER_DEBUG_REG17__sm1_prim_end_state__SHIFT 0x0
+#define CLIPPER_DEBUG_REG17__sm1_ps_expand_MASK 0x80
+#define CLIPPER_DEBUG_REG17__sm1_ps_expand__SHIFT 0x7
+#define CLIPPER_DEBUG_REG17__sm1_clip_vert_cnt_MASK 0x1f00
+#define CLIPPER_DEBUG_REG17__sm1_clip_vert_cnt__SHIFT 0x8
+#define CLIPPER_DEBUG_REG17__sm1_vertex_clip_cnt_MASK 0x3e000
+#define CLIPPER_DEBUG_REG17__sm1_vertex_clip_cnt__SHIFT 0xd
+#define CLIPPER_DEBUG_REG17__sm1_inv_to_clip_data_valid_1_MASK 0x40000
+#define CLIPPER_DEBUG_REG17__sm1_inv_to_clip_data_valid_1__SHIFT 0x12
+#define CLIPPER_DEBUG_REG17__sm1_inv_to_clip_data_valid_0_MASK 0x80000
+#define CLIPPER_DEBUG_REG17__sm1_inv_to_clip_data_valid_0__SHIFT 0x13
+#define CLIPPER_DEBUG_REG17__sm1_current_state_MASK 0x7f00000
+#define CLIPPER_DEBUG_REG17__sm1_current_state__SHIFT 0x14
+#define CLIPPER_DEBUG_REG17__sm1_clip_to_clipga_clip_to_outsm_cnt_eq0_MASK 0x8000000
+#define CLIPPER_DEBUG_REG17__sm1_clip_to_clipga_clip_to_outsm_cnt_eq0__SHIFT 0x1b
+#define CLIPPER_DEBUG_REG17__sm1_clip_to_outsm_fifo_full_MASK 0x10000000
+#define CLIPPER_DEBUG_REG17__sm1_clip_to_outsm_fifo_full__SHIFT 0x1c
+#define CLIPPER_DEBUG_REG17__sm1_highest_priority_seq_MASK 0x20000000
+#define CLIPPER_DEBUG_REG17__sm1_highest_priority_seq__SHIFT 0x1d
+#define CLIPPER_DEBUG_REG17__sm1_outputcliptoclipga_0_MASK 0x40000000
+#define CLIPPER_DEBUG_REG17__sm1_outputcliptoclipga_0__SHIFT 0x1e
+#define CLIPPER_DEBUG_REG17__sm1_clprim_to_clip_prim_valid_MASK 0x80000000
+#define CLIPPER_DEBUG_REG17__sm1_clprim_to_clip_prim_valid__SHIFT 0x1f
+#define CLIPPER_DEBUG_REG18__sm2_prim_end_state_MASK 0x7f
+#define CLIPPER_DEBUG_REG18__sm2_prim_end_state__SHIFT 0x0
+#define CLIPPER_DEBUG_REG18__sm2_ps_expand_MASK 0x80
+#define CLIPPER_DEBUG_REG18__sm2_ps_expand__SHIFT 0x7
+#define CLIPPER_DEBUG_REG18__sm2_clip_vert_cnt_MASK 0x1f00
+#define CLIPPER_DEBUG_REG18__sm2_clip_vert_cnt__SHIFT 0x8
+#define CLIPPER_DEBUG_REG18__sm2_vertex_clip_cnt_MASK 0x3e000
+#define CLIPPER_DEBUG_REG18__sm2_vertex_clip_cnt__SHIFT 0xd
+#define CLIPPER_DEBUG_REG18__sm2_inv_to_clip_data_valid_1_MASK 0x40000
+#define CLIPPER_DEBUG_REG18__sm2_inv_to_clip_data_valid_1__SHIFT 0x12
+#define CLIPPER_DEBUG_REG18__sm2_inv_to_clip_data_valid_0_MASK 0x80000
+#define CLIPPER_DEBUG_REG18__sm2_inv_to_clip_data_valid_0__SHIFT 0x13
+#define CLIPPER_DEBUG_REG18__sm2_current_state_MASK 0x7f00000
+#define CLIPPER_DEBUG_REG18__sm2_current_state__SHIFT 0x14
+#define CLIPPER_DEBUG_REG18__sm2_clip_to_clipga_clip_to_outsm_cnt_eq0_MASK 0x8000000
+#define CLIPPER_DEBUG_REG18__sm2_clip_to_clipga_clip_to_outsm_cnt_eq0__SHIFT 0x1b
+#define CLIPPER_DEBUG_REG18__sm2_clip_to_outsm_fifo_full_MASK 0x10000000
+#define CLIPPER_DEBUG_REG18__sm2_clip_to_outsm_fifo_full__SHIFT 0x1c
+#define CLIPPER_DEBUG_REG18__sm2_highest_priority_seq_MASK 0x20000000
+#define CLIPPER_DEBUG_REG18__sm2_highest_priority_seq__SHIFT 0x1d
+#define CLIPPER_DEBUG_REG18__sm2_outputcliptoclipga_0_MASK 0x40000000
+#define CLIPPER_DEBUG_REG18__sm2_outputcliptoclipga_0__SHIFT 0x1e
+#define CLIPPER_DEBUG_REG18__sm2_clprim_to_clip_prim_valid_MASK 0x80000000
+#define CLIPPER_DEBUG_REG18__sm2_clprim_to_clip_prim_valid__SHIFT 0x1f
+#define CLIPPER_DEBUG_REG19__sm3_prim_end_state_MASK 0x7f
+#define CLIPPER_DEBUG_REG19__sm3_prim_end_state__SHIFT 0x0
+#define CLIPPER_DEBUG_REG19__sm3_ps_expand_MASK 0x80
+#define CLIPPER_DEBUG_REG19__sm3_ps_expand__SHIFT 0x7
+#define CLIPPER_DEBUG_REG19__sm3_clip_vert_cnt_MASK 0x1f00
+#define CLIPPER_DEBUG_REG19__sm3_clip_vert_cnt__SHIFT 0x8
+#define CLIPPER_DEBUG_REG19__sm3_vertex_clip_cnt_MASK 0x3e000
+#define CLIPPER_DEBUG_REG19__sm3_vertex_clip_cnt__SHIFT 0xd
+#define CLIPPER_DEBUG_REG19__sm3_inv_to_clip_data_valid_1_MASK 0x40000
+#define CLIPPER_DEBUG_REG19__sm3_inv_to_clip_data_valid_1__SHIFT 0x12
+#define CLIPPER_DEBUG_REG19__sm3_inv_to_clip_data_valid_0_MASK 0x80000
+#define CLIPPER_DEBUG_REG19__sm3_inv_to_clip_data_valid_0__SHIFT 0x13
+#define CLIPPER_DEBUG_REG19__sm3_current_state_MASK 0x7f00000
+#define CLIPPER_DEBUG_REG19__sm3_current_state__SHIFT 0x14
+#define CLIPPER_DEBUG_REG19__sm3_clip_to_clipga_clip_to_outsm_cnt_eq0_MASK 0x8000000
+#define CLIPPER_DEBUG_REG19__sm3_clip_to_clipga_clip_to_outsm_cnt_eq0__SHIFT 0x1b
+#define CLIPPER_DEBUG_REG19__sm3_clip_to_outsm_fifo_full_MASK 0x10000000
+#define CLIPPER_DEBUG_REG19__sm3_clip_to_outsm_fifo_full__SHIFT 0x1c
+#define CLIPPER_DEBUG_REG19__sm3_highest_priority_seq_MASK 0x20000000
+#define CLIPPER_DEBUG_REG19__sm3_highest_priority_seq__SHIFT 0x1d
+#define CLIPPER_DEBUG_REG19__sm3_outputcliptoclipga_0_MASK 0x40000000
+#define CLIPPER_DEBUG_REG19__sm3_outputcliptoclipga_0__SHIFT 0x1e
+#define CLIPPER_DEBUG_REG19__sm3_clprim_to_clip_prim_valid_MASK 0x80000000
+#define CLIPPER_DEBUG_REG19__sm3_clprim_to_clip_prim_valid__SHIFT 0x1f
+#define SXIFCCG_DEBUG_REG0__position_address_MASK 0x3f
+#define SXIFCCG_DEBUG_REG0__position_address__SHIFT 0x0
+#define SXIFCCG_DEBUG_REG0__point_address_MASK 0x1c0
+#define SXIFCCG_DEBUG_REG0__point_address__SHIFT 0x6
+#define SXIFCCG_DEBUG_REG0__sx_pending_rd_state_var_indx_MASK 0xe00
+#define SXIFCCG_DEBUG_REG0__sx_pending_rd_state_var_indx__SHIFT 0x9
+#define SXIFCCG_DEBUG_REG0__sx_pending_rd_req_mask_MASK 0xf000
+#define SXIFCCG_DEBUG_REG0__sx_pending_rd_req_mask__SHIFT 0xc
+#define SXIFCCG_DEBUG_REG0__sx_pending_rd_pci_MASK 0x3ff0000
+#define SXIFCCG_DEBUG_REG0__sx_pending_rd_pci__SHIFT 0x10
+#define SXIFCCG_DEBUG_REG0__sx_pending_rd_aux_sel_MASK 0xc000000
+#define SXIFCCG_DEBUG_REG0__sx_pending_rd_aux_sel__SHIFT 0x1a
+#define SXIFCCG_DEBUG_REG0__sx_pending_rd_sp_id_MASK 0x30000000
+#define SXIFCCG_DEBUG_REG0__sx_pending_rd_sp_id__SHIFT 0x1c
+#define SXIFCCG_DEBUG_REG0__sx_pending_rd_aux_inc_MASK 0x40000000
+#define SXIFCCG_DEBUG_REG0__sx_pending_rd_aux_inc__SHIFT 0x1e
+#define SXIFCCG_DEBUG_REG0__sx_pending_rd_advance_MASK 0x80000000
+#define SXIFCCG_DEBUG_REG0__sx_pending_rd_advance__SHIFT 0x1f
+#define SXIFCCG_DEBUG_REG1__available_positions_MASK 0x7f
+#define SXIFCCG_DEBUG_REG1__available_positions__SHIFT 0x0
+#define SXIFCCG_DEBUG_REG1__sx_receive_indx_MASK 0x380
+#define SXIFCCG_DEBUG_REG1__sx_receive_indx__SHIFT 0x7
+#define SXIFCCG_DEBUG_REG1__sx_pending_fifo_contents_MASK 0x7c00
+#define SXIFCCG_DEBUG_REG1__sx_pending_fifo_contents__SHIFT 0xa
+#define SXIFCCG_DEBUG_REG1__statevar_bits_vs_out_misc_vec_ena_MASK 0x8000
+#define SXIFCCG_DEBUG_REG1__statevar_bits_vs_out_misc_vec_ena__SHIFT 0xf
+#define SXIFCCG_DEBUG_REG1__statevar_bits_disable_sp_MASK 0xf0000
+#define SXIFCCG_DEBUG_REG1__statevar_bits_disable_sp__SHIFT 0x10
+#define SXIFCCG_DEBUG_REG1__aux_sel_MASK 0x300000
+#define SXIFCCG_DEBUG_REG1__aux_sel__SHIFT 0x14
+#define SXIFCCG_DEBUG_REG1__sx_to_pa_empty_1_MASK 0x400000
+#define SXIFCCG_DEBUG_REG1__sx_to_pa_empty_1__SHIFT 0x16
+#define SXIFCCG_DEBUG_REG1__sx_to_pa_empty_0_MASK 0x800000
+#define SXIFCCG_DEBUG_REG1__sx_to_pa_empty_0__SHIFT 0x17
+#define SXIFCCG_DEBUG_REG1__pasx_req_cnt_1_MASK 0xf000000
+#define SXIFCCG_DEBUG_REG1__pasx_req_cnt_1__SHIFT 0x18
+#define SXIFCCG_DEBUG_REG1__pasx_req_cnt_0_MASK 0xf0000000
+#define SXIFCCG_DEBUG_REG1__pasx_req_cnt_0__SHIFT 0x1c
+#define SXIFCCG_DEBUG_REG2__param_cache_base_MASK 0x7f
+#define SXIFCCG_DEBUG_REG2__param_cache_base__SHIFT 0x0
+#define SXIFCCG_DEBUG_REG2__sx_aux_MASK 0x180
+#define SXIFCCG_DEBUG_REG2__sx_aux__SHIFT 0x7
+#define SXIFCCG_DEBUG_REG2__sx_request_indx_MASK 0x7e00
+#define SXIFCCG_DEBUG_REG2__sx_request_indx__SHIFT 0x9
+#define SXIFCCG_DEBUG_REG2__req_active_verts_loaded_MASK 0x8000
+#define SXIFCCG_DEBUG_REG2__req_active_verts_loaded__SHIFT 0xf
+#define SXIFCCG_DEBUG_REG2__req_active_verts_MASK 0x7f0000
+#define SXIFCCG_DEBUG_REG2__req_active_verts__SHIFT 0x10
+#define SXIFCCG_DEBUG_REG2__vgt_to_ccgen_state_var_indx_MASK 0x3800000
+#define SXIFCCG_DEBUG_REG2__vgt_to_ccgen_state_var_indx__SHIFT 0x17
+#define SXIFCCG_DEBUG_REG2__vgt_to_ccgen_active_verts_MASK 0xfc000000
+#define SXIFCCG_DEBUG_REG2__vgt_to_ccgen_active_verts__SHIFT 0x1a
+#define SXIFCCG_DEBUG_REG3__ALWAYS_ZERO_MASK 0xff
+#define SXIFCCG_DEBUG_REG3__ALWAYS_ZERO__SHIFT 0x0
+#define SXIFCCG_DEBUG_REG3__vertex_fifo_entriesavailable_MASK 0xf00
+#define SXIFCCG_DEBUG_REG3__vertex_fifo_entriesavailable__SHIFT 0x8
+#define SXIFCCG_DEBUG_REG3__statevar_bits_vs_out_ccdist1_vec_ena_MASK 0x1000
+#define SXIFCCG_DEBUG_REG3__statevar_bits_vs_out_ccdist1_vec_ena__SHIFT 0xc
+#define SXIFCCG_DEBUG_REG3__statevar_bits_vs_out_ccdist0_vec_ena_MASK 0x2000
+#define SXIFCCG_DEBUG_REG3__statevar_bits_vs_out_ccdist0_vec_ena__SHIFT 0xd
+#define SXIFCCG_DEBUG_REG3__available_positions_MASK 0x1fc000
+#define SXIFCCG_DEBUG_REG3__available_positions__SHIFT 0xe
+#define SXIFCCG_DEBUG_REG3__current_state_MASK 0x600000
+#define SXIFCCG_DEBUG_REG3__current_state__SHIFT 0x15
+#define SXIFCCG_DEBUG_REG3__vertex_fifo_empty_MASK 0x800000
+#define SXIFCCG_DEBUG_REG3__vertex_fifo_empty__SHIFT 0x17
+#define SXIFCCG_DEBUG_REG3__vertex_fifo_full_MASK 0x1000000
+#define SXIFCCG_DEBUG_REG3__vertex_fifo_full__SHIFT 0x18
+#define SXIFCCG_DEBUG_REG3__sx0_receive_fifo_empty_MASK 0x2000000
+#define SXIFCCG_DEBUG_REG3__sx0_receive_fifo_empty__SHIFT 0x19
+#define SXIFCCG_DEBUG_REG3__sx0_receive_fifo_full_MASK 0x4000000
+#define SXIFCCG_DEBUG_REG3__sx0_receive_fifo_full__SHIFT 0x1a
+#define SXIFCCG_DEBUG_REG3__vgt_to_ccgen_fifo_empty_MASK 0x8000000
+#define SXIFCCG_DEBUG_REG3__vgt_to_ccgen_fifo_empty__SHIFT 0x1b
+#define SXIFCCG_DEBUG_REG3__vgt_to_ccgen_fifo_full_MASK 0x10000000
+#define SXIFCCG_DEBUG_REG3__vgt_to_ccgen_fifo_full__SHIFT 0x1c
+#define SXIFCCG_DEBUG_REG3__ccgen_to_clipcc_fifo_full_MASK 0x20000000
+#define SXIFCCG_DEBUG_REG3__ccgen_to_clipcc_fifo_full__SHIFT 0x1d
+#define SXIFCCG_DEBUG_REG3__sx0_receive_fifo_write_MASK 0x40000000
+#define SXIFCCG_DEBUG_REG3__sx0_receive_fifo_write__SHIFT 0x1e
+#define SXIFCCG_DEBUG_REG3__ccgen_to_clipcc_write_MASK 0x80000000
+#define SXIFCCG_DEBUG_REG3__ccgen_to_clipcc_write__SHIFT 0x1f
+#define SETUP_DEBUG_REG0__su_baryc_cntl_state_MASK 0x3
+#define SETUP_DEBUG_REG0__su_baryc_cntl_state__SHIFT 0x0
+#define SETUP_DEBUG_REG0__su_cntl_state_MASK 0x3c
+#define SETUP_DEBUG_REG0__su_cntl_state__SHIFT 0x2
+#define SETUP_DEBUG_REG0__pmode_state_MASK 0x3f00
+#define SETUP_DEBUG_REG0__pmode_state__SHIFT 0x8
+#define SETUP_DEBUG_REG0__ge_stallb_MASK 0x4000
+#define SETUP_DEBUG_REG0__ge_stallb__SHIFT 0xe
+#define SETUP_DEBUG_REG0__geom_enable_MASK 0x8000
+#define SETUP_DEBUG_REG0__geom_enable__SHIFT 0xf
+#define SETUP_DEBUG_REG0__su_clip_baryc_free_MASK 0x30000
+#define SETUP_DEBUG_REG0__su_clip_baryc_free__SHIFT 0x10
+#define SETUP_DEBUG_REG0__su_clip_rtr_MASK 0x40000
+#define SETUP_DEBUG_REG0__su_clip_rtr__SHIFT 0x12
+#define SETUP_DEBUG_REG0__pfifo_busy_MASK 0x80000
+#define SETUP_DEBUG_REG0__pfifo_busy__SHIFT 0x13
+#define SETUP_DEBUG_REG0__su_cntl_busy_MASK 0x100000
+#define SETUP_DEBUG_REG0__su_cntl_busy__SHIFT 0x14
+#define SETUP_DEBUG_REG0__geom_busy_MASK 0x200000
+#define SETUP_DEBUG_REG0__geom_busy__SHIFT 0x15
+#define SETUP_DEBUG_REG0__event_id_gated_MASK 0xfc00000
+#define SETUP_DEBUG_REG0__event_id_gated__SHIFT 0x16
+#define SETUP_DEBUG_REG0__event_gated_MASK 0x10000000
+#define SETUP_DEBUG_REG0__event_gated__SHIFT 0x1c
+#define SETUP_DEBUG_REG0__pmode_prim_gated_MASK 0x20000000
+#define SETUP_DEBUG_REG0__pmode_prim_gated__SHIFT 0x1d
+#define SETUP_DEBUG_REG0__su_dyn_sclk_vld_MASK 0x40000000
+#define SETUP_DEBUG_REG0__su_dyn_sclk_vld__SHIFT 0x1e
+#define SETUP_DEBUG_REG0__cl_dyn_sclk_vld_MASK 0x80000000
+#define SETUP_DEBUG_REG0__cl_dyn_sclk_vld__SHIFT 0x1f
+#define SETUP_DEBUG_REG1__y_sort0_gated_23_8_MASK 0xffff
+#define SETUP_DEBUG_REG1__y_sort0_gated_23_8__SHIFT 0x0
+#define SETUP_DEBUG_REG1__x_sort0_gated_23_8_MASK 0xffff0000
+#define SETUP_DEBUG_REG1__x_sort0_gated_23_8__SHIFT 0x10
+#define SETUP_DEBUG_REG2__y_sort1_gated_23_8_MASK 0xffff
+#define SETUP_DEBUG_REG2__y_sort1_gated_23_8__SHIFT 0x0
+#define SETUP_DEBUG_REG2__x_sort1_gated_23_8_MASK 0xffff0000
+#define SETUP_DEBUG_REG2__x_sort1_gated_23_8__SHIFT 0x10
+#define SETUP_DEBUG_REG3__y_sort2_gated_23_8_MASK 0xffff
+#define SETUP_DEBUG_REG3__y_sort2_gated_23_8__SHIFT 0x0
+#define SETUP_DEBUG_REG3__x_sort2_gated_23_8_MASK 0xffff0000
+#define SETUP_DEBUG_REG3__x_sort2_gated_23_8__SHIFT 0x10
+#define SETUP_DEBUG_REG4__attr_indx_sort0_gated_MASK 0x3fff
+#define SETUP_DEBUG_REG4__attr_indx_sort0_gated__SHIFT 0x0
+#define SETUP_DEBUG_REG4__null_prim_gated_MASK 0x4000
+#define SETUP_DEBUG_REG4__null_prim_gated__SHIFT 0xe
+#define SETUP_DEBUG_REG4__backfacing_gated_MASK 0x8000
+#define SETUP_DEBUG_REG4__backfacing_gated__SHIFT 0xf
+#define SETUP_DEBUG_REG4__st_indx_gated_MASK 0x70000
+#define SETUP_DEBUG_REG4__st_indx_gated__SHIFT 0x10
+#define SETUP_DEBUG_REG4__clipped_gated_MASK 0x80000
+#define SETUP_DEBUG_REG4__clipped_gated__SHIFT 0x13
+#define SETUP_DEBUG_REG4__dealloc_slot_gated_MASK 0x700000
+#define SETUP_DEBUG_REG4__dealloc_slot_gated__SHIFT 0x14
+#define SETUP_DEBUG_REG4__xmajor_gated_MASK 0x800000
+#define SETUP_DEBUG_REG4__xmajor_gated__SHIFT 0x17
+#define SETUP_DEBUG_REG4__diamond_rule_gated_MASK 0x3000000
+#define SETUP_DEBUG_REG4__diamond_rule_gated__SHIFT 0x18
+#define SETUP_DEBUG_REG4__type_gated_MASK 0x1c000000
+#define SETUP_DEBUG_REG4__type_gated__SHIFT 0x1a
+#define SETUP_DEBUG_REG4__fpov_gated_MASK 0x60000000
+#define SETUP_DEBUG_REG4__fpov_gated__SHIFT 0x1d
+#define SETUP_DEBUG_REG4__eop_gated_MASK 0x80000000
+#define SETUP_DEBUG_REG4__eop_gated__SHIFT 0x1f
+#define SETUP_DEBUG_REG5__attr_indx_sort2_gated_MASK 0x3fff
+#define SETUP_DEBUG_REG5__attr_indx_sort2_gated__SHIFT 0x0
+#define SETUP_DEBUG_REG5__attr_indx_sort1_gated_MASK 0xfffc000
+#define SETUP_DEBUG_REG5__attr_indx_sort1_gated__SHIFT 0xe
+#define SETUP_DEBUG_REG5__provoking_vtx_gated_MASK 0x30000000
+#define SETUP_DEBUG_REG5__provoking_vtx_gated__SHIFT 0x1c
+#define SETUP_DEBUG_REG5__valid_prim_gated_MASK 0x40000000
+#define SETUP_DEBUG_REG5__valid_prim_gated__SHIFT 0x1e
+#define SETUP_DEBUG_REG5__pa_reg_sclk_vld_MASK 0x80000000
+#define SETUP_DEBUG_REG5__pa_reg_sclk_vld__SHIFT 0x1f
+#define PA_SC_DEBUG_REG0__REG0_FIELD0_MASK 0x3
+#define PA_SC_DEBUG_REG0__REG0_FIELD0__SHIFT 0x0
+#define PA_SC_DEBUG_REG0__REG0_FIELD1_MASK 0xc
+#define PA_SC_DEBUG_REG0__REG0_FIELD1__SHIFT 0x2
+#define PA_SC_DEBUG_REG1__REG1_FIELD0_MASK 0x3
+#define PA_SC_DEBUG_REG1__REG1_FIELD0__SHIFT 0x0
+#define PA_SC_DEBUG_REG1__REG1_FIELD1_MASK 0xc
+#define PA_SC_DEBUG_REG1__REG1_FIELD1__SHIFT 0x2
+#define COMPUTE_DISPATCH_INITIATOR__COMPUTE_SHADER_EN_MASK 0x1
+#define COMPUTE_DISPATCH_INITIATOR__COMPUTE_SHADER_EN__SHIFT 0x0
+#define COMPUTE_DISPATCH_INITIATOR__PARTIAL_TG_EN_MASK 0x2
+#define COMPUTE_DISPATCH_INITIATOR__PARTIAL_TG_EN__SHIFT 0x1
+#define COMPUTE_DISPATCH_INITIATOR__FORCE_START_AT_000_MASK 0x4
+#define COMPUTE_DISPATCH_INITIATOR__FORCE_START_AT_000__SHIFT 0x2
+#define COMPUTE_DISPATCH_INITIATOR__ORDERED_APPEND_ENBL_MASK 0x8
+#define COMPUTE_DISPATCH_INITIATOR__ORDERED_APPEND_ENBL__SHIFT 0x3
+#define COMPUTE_DISPATCH_INITIATOR__ORDERED_APPEND_MODE_MASK 0x10
+#define COMPUTE_DISPATCH_INITIATOR__ORDERED_APPEND_MODE__SHIFT 0x4
+#define COMPUTE_DISPATCH_INITIATOR__USE_THREAD_DIMENSIONS_MASK 0x20
+#define COMPUTE_DISPATCH_INITIATOR__USE_THREAD_DIMENSIONS__SHIFT 0x5
+#define COMPUTE_DISPATCH_INITIATOR__ORDER_MODE_MASK 0x40
+#define COMPUTE_DISPATCH_INITIATOR__ORDER_MODE__SHIFT 0x6
+#define COMPUTE_DISPATCH_INITIATOR__DISPATCH_CACHE_CNTL_MASK 0x380
+#define COMPUTE_DISPATCH_INITIATOR__DISPATCH_CACHE_CNTL__SHIFT 0x7
+#define COMPUTE_DISPATCH_INITIATOR__SCALAR_L1_INV_VOL_MASK 0x400
+#define COMPUTE_DISPATCH_INITIATOR__SCALAR_L1_INV_VOL__SHIFT 0xa
+#define COMPUTE_DISPATCH_INITIATOR__VECTOR_L1_INV_VOL_MASK 0x800
+#define COMPUTE_DISPATCH_INITIATOR__VECTOR_L1_INV_VOL__SHIFT 0xb
+#define COMPUTE_DISPATCH_INITIATOR__DATA_ATC_MASK 0x1000
+#define COMPUTE_DISPATCH_INITIATOR__DATA_ATC__SHIFT 0xc
+#define COMPUTE_DISPATCH_INITIATOR__RESTORE_MASK 0x4000
+#define COMPUTE_DISPATCH_INITIATOR__RESTORE__SHIFT 0xe
+#define COMPUTE_DIM_X__SIZE_MASK 0xffffffff
+#define COMPUTE_DIM_X__SIZE__SHIFT 0x0
+#define COMPUTE_DIM_Y__SIZE_MASK 0xffffffff
+#define COMPUTE_DIM_Y__SIZE__SHIFT 0x0
+#define COMPUTE_DIM_Z__SIZE_MASK 0xffffffff
+#define COMPUTE_DIM_Z__SIZE__SHIFT 0x0
+#define COMPUTE_START_X__START_MASK 0xffffffff
+#define COMPUTE_START_X__START__SHIFT 0x0
+#define COMPUTE_START_Y__START_MASK 0xffffffff
+#define COMPUTE_START_Y__START__SHIFT 0x0
+#define COMPUTE_START_Z__START_MASK 0xffffffff
+#define COMPUTE_START_Z__START__SHIFT 0x0
+#define COMPUTE_NUM_THREAD_X__NUM_THREAD_FULL_MASK 0xffff
+#define COMPUTE_NUM_THREAD_X__NUM_THREAD_FULL__SHIFT 0x0
+#define COMPUTE_NUM_THREAD_X__NUM_THREAD_PARTIAL_MASK 0xffff0000
+#define COMPUTE_NUM_THREAD_X__NUM_THREAD_PARTIAL__SHIFT 0x10
+#define COMPUTE_NUM_THREAD_Y__NUM_THREAD_FULL_MASK 0xffff
+#define COMPUTE_NUM_THREAD_Y__NUM_THREAD_FULL__SHIFT 0x0
+#define COMPUTE_NUM_THREAD_Y__NUM_THREAD_PARTIAL_MASK 0xffff0000
+#define COMPUTE_NUM_THREAD_Y__NUM_THREAD_PARTIAL__SHIFT 0x10
+#define COMPUTE_NUM_THREAD_Z__NUM_THREAD_FULL_MASK 0xffff
+#define COMPUTE_NUM_THREAD_Z__NUM_THREAD_FULL__SHIFT 0x0
+#define COMPUTE_NUM_THREAD_Z__NUM_THREAD_PARTIAL_MASK 0xffff0000
+#define COMPUTE_NUM_THREAD_Z__NUM_THREAD_PARTIAL__SHIFT 0x10
+#define COMPUTE_PIPELINESTAT_ENABLE__PIPELINESTAT_ENABLE_MASK 0x1
+#define COMPUTE_PIPELINESTAT_ENABLE__PIPELINESTAT_ENABLE__SHIFT 0x0
+#define COMPUTE_PERFCOUNT_ENABLE__PERFCOUNT_ENABLE_MASK 0x1
+#define COMPUTE_PERFCOUNT_ENABLE__PERFCOUNT_ENABLE__SHIFT 0x0
+#define COMPUTE_PGM_LO__DATA_MASK 0xffffffff
+#define COMPUTE_PGM_LO__DATA__SHIFT 0x0
+#define COMPUTE_PGM_HI__DATA_MASK 0xff
+#define COMPUTE_PGM_HI__DATA__SHIFT 0x0
+#define COMPUTE_PGM_HI__INST_ATC_MASK 0x100
+#define COMPUTE_PGM_HI__INST_ATC__SHIFT 0x8
+#define COMPUTE_TBA_LO__DATA_MASK 0xffffffff
+#define COMPUTE_TBA_LO__DATA__SHIFT 0x0
+#define COMPUTE_TBA_HI__DATA_MASK 0xff
+#define COMPUTE_TBA_HI__DATA__SHIFT 0x0
+#define COMPUTE_TMA_LO__DATA_MASK 0xffffffff
+#define COMPUTE_TMA_LO__DATA__SHIFT 0x0
+#define COMPUTE_TMA_HI__DATA_MASK 0xff
+#define COMPUTE_TMA_HI__DATA__SHIFT 0x0
+#define COMPUTE_PGM_RSRC1__VGPRS_MASK 0x3f
+#define COMPUTE_PGM_RSRC1__VGPRS__SHIFT 0x0
+#define COMPUTE_PGM_RSRC1__SGPRS_MASK 0x3c0
+#define COMPUTE_PGM_RSRC1__SGPRS__SHIFT 0x6
+#define COMPUTE_PGM_RSRC1__PRIORITY_MASK 0xc00
+#define COMPUTE_PGM_RSRC1__PRIORITY__SHIFT 0xa
+#define COMPUTE_PGM_RSRC1__FLOAT_MODE_MASK 0xff000
+#define COMPUTE_PGM_RSRC1__FLOAT_MODE__SHIFT 0xc
+#define COMPUTE_PGM_RSRC1__PRIV_MASK 0x100000
+#define COMPUTE_PGM_RSRC1__PRIV__SHIFT 0x14
+#define COMPUTE_PGM_RSRC1__DX10_CLAMP_MASK 0x200000
+#define COMPUTE_PGM_RSRC1__DX10_CLAMP__SHIFT 0x15
+#define COMPUTE_PGM_RSRC1__DEBUG_MODE_MASK 0x400000
+#define COMPUTE_PGM_RSRC1__DEBUG_MODE__SHIFT 0x16
+#define COMPUTE_PGM_RSRC1__IEEE_MODE_MASK 0x800000
+#define COMPUTE_PGM_RSRC1__IEEE_MODE__SHIFT 0x17
+#define COMPUTE_PGM_RSRC1__BULKY_MASK 0x1000000
+#define COMPUTE_PGM_RSRC1__BULKY__SHIFT 0x18
+#define COMPUTE_PGM_RSRC1__CDBG_USER_MASK 0x2000000
+#define COMPUTE_PGM_RSRC1__CDBG_USER__SHIFT 0x19
+#define COMPUTE_PGM_RSRC2__SCRATCH_EN_MASK 0x1
+#define COMPUTE_PGM_RSRC2__SCRATCH_EN__SHIFT 0x0
+#define COMPUTE_PGM_RSRC2__USER_SGPR_MASK 0x3e
+#define COMPUTE_PGM_RSRC2__USER_SGPR__SHIFT 0x1
+#define COMPUTE_PGM_RSRC2__TRAP_PRESENT_MASK 0x40
+#define COMPUTE_PGM_RSRC2__TRAP_PRESENT__SHIFT 0x6
+#define COMPUTE_PGM_RSRC2__TGID_X_EN_MASK 0x80
+#define COMPUTE_PGM_RSRC2__TGID_X_EN__SHIFT 0x7
+#define COMPUTE_PGM_RSRC2__TGID_Y_EN_MASK 0x100
+#define COMPUTE_PGM_RSRC2__TGID_Y_EN__SHIFT 0x8
+#define COMPUTE_PGM_RSRC2__TGID_Z_EN_MASK 0x200
+#define COMPUTE_PGM_RSRC2__TGID_Z_EN__SHIFT 0x9
+#define COMPUTE_PGM_RSRC2__TG_SIZE_EN_MASK 0x400
+#define COMPUTE_PGM_RSRC2__TG_SIZE_EN__SHIFT 0xa
+#define COMPUTE_PGM_RSRC2__TIDIG_COMP_CNT_MASK 0x1800
+#define COMPUTE_PGM_RSRC2__TIDIG_COMP_CNT__SHIFT 0xb
+#define COMPUTE_PGM_RSRC2__EXCP_EN_MSB_MASK 0x6000
+#define COMPUTE_PGM_RSRC2__EXCP_EN_MSB__SHIFT 0xd
+#define COMPUTE_PGM_RSRC2__LDS_SIZE_MASK 0xff8000
+#define COMPUTE_PGM_RSRC2__LDS_SIZE__SHIFT 0xf
+#define COMPUTE_PGM_RSRC2__EXCP_EN_MASK 0x7f000000
+#define COMPUTE_PGM_RSRC2__EXCP_EN__SHIFT 0x18
+#define COMPUTE_VMID__DATA_MASK 0xf
+#define COMPUTE_VMID__DATA__SHIFT 0x0
+#define COMPUTE_RESOURCE_LIMITS__WAVES_PER_SH_MASK 0x3ff
+#define COMPUTE_RESOURCE_LIMITS__WAVES_PER_SH__SHIFT 0x0
+#define COMPUTE_RESOURCE_LIMITS__TG_PER_CU_MASK 0xf000
+#define COMPUTE_RESOURCE_LIMITS__TG_PER_CU__SHIFT 0xc
+#define COMPUTE_RESOURCE_LIMITS__LOCK_THRESHOLD_MASK 0x3f0000
+#define COMPUTE_RESOURCE_LIMITS__LOCK_THRESHOLD__SHIFT 0x10
+#define COMPUTE_RESOURCE_LIMITS__SIMD_DEST_CNTL_MASK 0x400000
+#define COMPUTE_RESOURCE_LIMITS__SIMD_DEST_CNTL__SHIFT 0x16
+#define COMPUTE_RESOURCE_LIMITS__FORCE_SIMD_DIST_MASK 0x800000
+#define COMPUTE_RESOURCE_LIMITS__FORCE_SIMD_DIST__SHIFT 0x17
+#define COMPUTE_RESOURCE_LIMITS__CU_GROUP_COUNT_MASK 0x7000000
+#define COMPUTE_RESOURCE_LIMITS__CU_GROUP_COUNT__SHIFT 0x18
+#define COMPUTE_STATIC_THREAD_MGMT_SE0__SH0_CU_EN_MASK 0xffff
+#define COMPUTE_STATIC_THREAD_MGMT_SE0__SH0_CU_EN__SHIFT 0x0
+#define COMPUTE_STATIC_THREAD_MGMT_SE0__SH1_CU_EN_MASK 0xffff0000
+#define COMPUTE_STATIC_THREAD_MGMT_SE0__SH1_CU_EN__SHIFT 0x10
+#define COMPUTE_STATIC_THREAD_MGMT_SE1__SH0_CU_EN_MASK 0xffff
+#define COMPUTE_STATIC_THREAD_MGMT_SE1__SH0_CU_EN__SHIFT 0x0
+#define COMPUTE_STATIC_THREAD_MGMT_SE1__SH1_CU_EN_MASK 0xffff0000
+#define COMPUTE_STATIC_THREAD_MGMT_SE1__SH1_CU_EN__SHIFT 0x10
+#define COMPUTE_TMPRING_SIZE__WAVES_MASK 0xfff
+#define COMPUTE_TMPRING_SIZE__WAVES__SHIFT 0x0
+#define COMPUTE_TMPRING_SIZE__WAVESIZE_MASK 0x1fff000
+#define COMPUTE_TMPRING_SIZE__WAVESIZE__SHIFT 0xc
+#define COMPUTE_STATIC_THREAD_MGMT_SE2__SH0_CU_EN_MASK 0xffff
+#define COMPUTE_STATIC_THREAD_MGMT_SE2__SH0_CU_EN__SHIFT 0x0
+#define COMPUTE_STATIC_THREAD_MGMT_SE2__SH1_CU_EN_MASK 0xffff0000
+#define COMPUTE_STATIC_THREAD_MGMT_SE2__SH1_CU_EN__SHIFT 0x10
+#define COMPUTE_STATIC_THREAD_MGMT_SE3__SH0_CU_EN_MASK 0xffff
+#define COMPUTE_STATIC_THREAD_MGMT_SE3__SH0_CU_EN__SHIFT 0x0
+#define COMPUTE_STATIC_THREAD_MGMT_SE3__SH1_CU_EN_MASK 0xffff0000
+#define COMPUTE_STATIC_THREAD_MGMT_SE3__SH1_CU_EN__SHIFT 0x10
+#define COMPUTE_RESTART_X__RESTART_MASK 0xffffffff
+#define COMPUTE_RESTART_X__RESTART__SHIFT 0x0
+#define COMPUTE_RESTART_Y__RESTART_MASK 0xffffffff
+#define COMPUTE_RESTART_Y__RESTART__SHIFT 0x0
+#define COMPUTE_RESTART_Z__RESTART_MASK 0xffffffff
+#define COMPUTE_RESTART_Z__RESTART__SHIFT 0x0
+#define COMPUTE_THREAD_TRACE_ENABLE__THREAD_TRACE_ENABLE_MASK 0x1
+#define COMPUTE_THREAD_TRACE_ENABLE__THREAD_TRACE_ENABLE__SHIFT 0x0
+#define COMPUTE_MISC_RESERVED__SEND_SEID_MASK 0x3
+#define COMPUTE_MISC_RESERVED__SEND_SEID__SHIFT 0x0
+#define COMPUTE_MISC_RESERVED__RESERVED2_MASK 0x4
+#define COMPUTE_MISC_RESERVED__RESERVED2__SHIFT 0x2
+#define COMPUTE_MISC_RESERVED__RESERVED3_MASK 0x8
+#define COMPUTE_MISC_RESERVED__RESERVED3__SHIFT 0x3
+#define COMPUTE_MISC_RESERVED__RESERVED4_MASK 0x10
+#define COMPUTE_MISC_RESERVED__RESERVED4__SHIFT 0x4
+#define COMPUTE_MISC_RESERVED__WAVE_ID_BASE_MASK 0x1ffe0
+#define COMPUTE_MISC_RESERVED__WAVE_ID_BASE__SHIFT 0x5
+#define COMPUTE_DISPATCH_ID__DISPATCH_ID_MASK 0xffffffff
+#define COMPUTE_DISPATCH_ID__DISPATCH_ID__SHIFT 0x0
+#define COMPUTE_THREADGROUP_ID__THREADGROUP_ID_MASK 0xffffffff
+#define COMPUTE_THREADGROUP_ID__THREADGROUP_ID__SHIFT 0x0
+#define COMPUTE_RELAUNCH__PAYLOAD_MASK 0x3fffffff
+#define COMPUTE_RELAUNCH__PAYLOAD__SHIFT 0x0
+#define COMPUTE_RELAUNCH__IS_EVENT_MASK 0x40000000
+#define COMPUTE_RELAUNCH__IS_EVENT__SHIFT 0x1e
+#define COMPUTE_RELAUNCH__IS_STATE_MASK 0x80000000
+#define COMPUTE_RELAUNCH__IS_STATE__SHIFT 0x1f
+#define COMPUTE_WAVE_RESTORE_ADDR_LO__ADDR_MASK 0xffffffff
+#define COMPUTE_WAVE_RESTORE_ADDR_LO__ADDR__SHIFT 0x0
+#define COMPUTE_WAVE_RESTORE_ADDR_HI__ADDR_MASK 0xffff
+#define COMPUTE_WAVE_RESTORE_ADDR_HI__ADDR__SHIFT 0x0
+#define COMPUTE_WAVE_RESTORE_CONTROL__ATC_MASK 0x1
+#define COMPUTE_WAVE_RESTORE_CONTROL__ATC__SHIFT 0x0
+#define COMPUTE_WAVE_RESTORE_CONTROL__MTYPE_MASK 0x6
+#define COMPUTE_WAVE_RESTORE_CONTROL__MTYPE__SHIFT 0x1
+#define COMPUTE_USER_DATA_0__DATA_MASK 0xffffffff
+#define COMPUTE_USER_DATA_0__DATA__SHIFT 0x0
+#define COMPUTE_USER_DATA_1__DATA_MASK 0xffffffff
+#define COMPUTE_USER_DATA_1__DATA__SHIFT 0x0
+#define COMPUTE_USER_DATA_2__DATA_MASK 0xffffffff
+#define COMPUTE_USER_DATA_2__DATA__SHIFT 0x0
+#define COMPUTE_USER_DATA_3__DATA_MASK 0xffffffff
+#define COMPUTE_USER_DATA_3__DATA__SHIFT 0x0
+#define COMPUTE_USER_DATA_4__DATA_MASK 0xffffffff
+#define COMPUTE_USER_DATA_4__DATA__SHIFT 0x0
+#define COMPUTE_USER_DATA_5__DATA_MASK 0xffffffff
+#define COMPUTE_USER_DATA_5__DATA__SHIFT 0x0
+#define COMPUTE_USER_DATA_6__DATA_MASK 0xffffffff
+#define COMPUTE_USER_DATA_6__DATA__SHIFT 0x0
+#define COMPUTE_USER_DATA_7__DATA_MASK 0xffffffff
+#define COMPUTE_USER_DATA_7__DATA__SHIFT 0x0
+#define COMPUTE_USER_DATA_8__DATA_MASK 0xffffffff
+#define COMPUTE_USER_DATA_8__DATA__SHIFT 0x0
+#define COMPUTE_USER_DATA_9__DATA_MASK 0xffffffff
+#define COMPUTE_USER_DATA_9__DATA__SHIFT 0x0
+#define COMPUTE_USER_DATA_10__DATA_MASK 0xffffffff
+#define COMPUTE_USER_DATA_10__DATA__SHIFT 0x0
+#define COMPUTE_USER_DATA_11__DATA_MASK 0xffffffff
+#define COMPUTE_USER_DATA_11__DATA__SHIFT 0x0
+#define COMPUTE_USER_DATA_12__DATA_MASK 0xffffffff
+#define COMPUTE_USER_DATA_12__DATA__SHIFT 0x0
+#define COMPUTE_USER_DATA_13__DATA_MASK 0xffffffff
+#define COMPUTE_USER_DATA_13__DATA__SHIFT 0x0
+#define COMPUTE_USER_DATA_14__DATA_MASK 0xffffffff
+#define COMPUTE_USER_DATA_14__DATA__SHIFT 0x0
+#define COMPUTE_USER_DATA_15__DATA_MASK 0xffffffff
+#define COMPUTE_USER_DATA_15__DATA__SHIFT 0x0
+#define COMPUTE_NOWHERE__DATA_MASK 0xffffffff
+#define COMPUTE_NOWHERE__DATA__SHIFT 0x0
+#define CSPRIV_CONNECT__DOORBELL_OFFSET_MASK 0x1fffff
+#define CSPRIV_CONNECT__DOORBELL_OFFSET__SHIFT 0x0
+#define CSPRIV_CONNECT__QUEUE_ID_MASK 0xe00000
+#define CSPRIV_CONNECT__QUEUE_ID__SHIFT 0x15
+#define CSPRIV_CONNECT__VMID_MASK 0x3c000000
+#define CSPRIV_CONNECT__VMID__SHIFT 0x1a
+#define CSPRIV_CONNECT__UNORD_DISP_MASK 0x80000000
+#define CSPRIV_CONNECT__UNORD_DISP__SHIFT 0x1f
+#define CSPRIV_THREAD_TRACE_TG0__TGID_X_MASK 0xffffffff
+#define CSPRIV_THREAD_TRACE_TG0__TGID_X__SHIFT 0x0
+#define CSPRIV_THREAD_TRACE_TG1__TGID_Y_MASK 0xffffffff
+#define CSPRIV_THREAD_TRACE_TG1__TGID_Y__SHIFT 0x0
+#define CSPRIV_THREAD_TRACE_TG2__TGID_Z_MASK 0xffffffff
+#define CSPRIV_THREAD_TRACE_TG2__TGID_Z__SHIFT 0x0
+#define CSPRIV_THREAD_TRACE_TG3__WAVE_ID_BASE_MASK 0xfff
+#define CSPRIV_THREAD_TRACE_TG3__WAVE_ID_BASE__SHIFT 0x0
+#define CSPRIV_THREAD_TRACE_TG3__THREADS_IN_GROUP_MASK 0xfff000
+#define CSPRIV_THREAD_TRACE_TG3__THREADS_IN_GROUP__SHIFT 0xc
+#define CSPRIV_THREAD_TRACE_TG3__PARTIAL_X_FLAG_MASK 0x1000000
+#define CSPRIV_THREAD_TRACE_TG3__PARTIAL_X_FLAG__SHIFT 0x18
+#define CSPRIV_THREAD_TRACE_TG3__PARTIAL_Y_FLAG_MASK 0x2000000
+#define CSPRIV_THREAD_TRACE_TG3__PARTIAL_Y_FLAG__SHIFT 0x19
+#define CSPRIV_THREAD_TRACE_TG3__PARTIAL_Z_FLAG_MASK 0x4000000
+#define CSPRIV_THREAD_TRACE_TG3__PARTIAL_Z_FLAG__SHIFT 0x1a
+#define CSPRIV_THREAD_TRACE_TG3__LAST_TG_MASK 0x8000000
+#define CSPRIV_THREAD_TRACE_TG3__LAST_TG__SHIFT 0x1b
+#define CSPRIV_THREAD_TRACE_TG3__FIRST_TG_MASK 0x10000000
+#define CSPRIV_THREAD_TRACE_TG3__FIRST_TG__SHIFT 0x1c
+#define CSPRIV_THREAD_TRACE_EVENT__EVENT_ID_MASK 0x1f
+#define CSPRIV_THREAD_TRACE_EVENT__EVENT_ID__SHIFT 0x0
+#define RLC_CNTL__RLC_ENABLE_F32_MASK 0x1
+#define RLC_CNTL__RLC_ENABLE_F32__SHIFT 0x0
+#define RLC_CNTL__FORCE_RETRY_MASK 0x2
+#define RLC_CNTL__FORCE_RETRY__SHIFT 0x1
+#define RLC_CNTL__READ_CACHE_DISABLE_MASK 0x4
+#define RLC_CNTL__READ_CACHE_DISABLE__SHIFT 0x2
+#define RLC_CNTL__RLC_STEP_F32_MASK 0x8
+#define RLC_CNTL__RLC_STEP_F32__SHIFT 0x3
+#define RLC_CNTL__SOFT_RESET_DEBUG_MODE_MASK 0x10
+#define RLC_CNTL__SOFT_RESET_DEBUG_MODE__SHIFT 0x4
+#define RLC_CNTL__RESERVED_MASK 0xffffff00
+#define RLC_CNTL__RESERVED__SHIFT 0x8
+#define RLC_DEBUG_SELECT__SELECT_MASK 0xff
+#define RLC_DEBUG_SELECT__SELECT__SHIFT 0x0
+#define RLC_DEBUG_SELECT__RESERVED_MASK 0xffffff00
+#define RLC_DEBUG_SELECT__RESERVED__SHIFT 0x8
+#define RLC_DEBUG__DATA_MASK 0xffffffff
+#define RLC_DEBUG__DATA__SHIFT 0x0
+#define RLC_MC_CNTL__WRREQ_SWAP_MASK 0x3
+#define RLC_MC_CNTL__WRREQ_SWAP__SHIFT 0x0
+#define RLC_MC_CNTL__WRREQ_TRAN_MASK 0x4
+#define RLC_MC_CNTL__WRREQ_TRAN__SHIFT 0x2
+#define RLC_MC_CNTL__WRREQ_PRIV_MASK 0x8
+#define RLC_MC_CNTL__WRREQ_PRIV__SHIFT 0x3
+#define RLC_MC_CNTL__WRNFO_STALL_MASK 0x10
+#define RLC_MC_CNTL__WRNFO_STALL__SHIFT 0x4
+#define RLC_MC_CNTL__WRNFO_URG_MASK 0x1e0
+#define RLC_MC_CNTL__WRNFO_URG__SHIFT 0x5
+#define RLC_MC_CNTL__WRREQ_DW_IMASK_MASK 0x1e00
+#define RLC_MC_CNTL__WRREQ_DW_IMASK__SHIFT 0x9
+#define RLC_MC_CNTL__RESERVED_B_MASK 0xfe000
+#define RLC_MC_CNTL__RESERVED_B__SHIFT 0xd
+#define RLC_MC_CNTL__RDNFO_URG_MASK 0xf00000
+#define RLC_MC_CNTL__RDNFO_URG__SHIFT 0x14
+#define RLC_MC_CNTL__RDREQ_SWAP_MASK 0x3000000
+#define RLC_MC_CNTL__RDREQ_SWAP__SHIFT 0x18
+#define RLC_MC_CNTL__RDREQ_TRAN_MASK 0x4000000
+#define RLC_MC_CNTL__RDREQ_TRAN__SHIFT 0x1a
+#define RLC_MC_CNTL__RDREQ_PRIV_MASK 0x8000000
+#define RLC_MC_CNTL__RDREQ_PRIV__SHIFT 0x1b
+#define RLC_MC_CNTL__RDNFO_STALL_MASK 0x10000000
+#define RLC_MC_CNTL__RDNFO_STALL__SHIFT 0x1c
+#define RLC_MC_CNTL__RESERVED_MASK 0xe0000000
+#define RLC_MC_CNTL__RESERVED__SHIFT 0x1d
+#define RLC_STAT__RLC_BUSY_MASK 0x1
+#define RLC_STAT__RLC_BUSY__SHIFT 0x0
+#define RLC_STAT__RLC_GPM_BUSY_MASK 0x2
+#define RLC_STAT__RLC_GPM_BUSY__SHIFT 0x1
+#define RLC_STAT__RLC_SPM_BUSY_MASK 0x4
+#define RLC_STAT__RLC_SPM_BUSY__SHIFT 0x2
+#define RLC_STAT__RLC_SRM_BUSY_MASK 0x8
+#define RLC_STAT__RLC_SRM_BUSY__SHIFT 0x3
+#define RLC_STAT__RESERVED_MASK 0xfffffff0
+#define RLC_STAT__RESERVED__SHIFT 0x4
+#define RLC_SAFE_MODE__CMD_MASK 0x1
+#define RLC_SAFE_MODE__CMD__SHIFT 0x0
+#define RLC_SAFE_MODE__MESSAGE_MASK 0x1e
+#define RLC_SAFE_MODE__MESSAGE__SHIFT 0x1
+#define RLC_SAFE_MODE__RESERVED1_MASK 0xe0
+#define RLC_SAFE_MODE__RESERVED1__SHIFT 0x5
+#define RLC_SAFE_MODE__RESPONSE_MASK 0xf00
+#define RLC_SAFE_MODE__RESPONSE__SHIFT 0x8
+#define RLC_SAFE_MODE__RESERVED_MASK 0xfffff000
+#define RLC_SAFE_MODE__RESERVED__SHIFT 0xc
+#define RLC_MEM_SLP_CNTL__RLC_MEM_LS_EN_MASK 0x1
+#define RLC_MEM_SLP_CNTL__RLC_MEM_LS_EN__SHIFT 0x0
+#define RLC_MEM_SLP_CNTL__RLC_MEM_DS_EN_MASK 0x2
+#define RLC_MEM_SLP_CNTL__RLC_MEM_DS_EN__SHIFT 0x1
+#define RLC_MEM_SLP_CNTL__RESERVED_MASK 0x7c
+#define RLC_MEM_SLP_CNTL__RESERVED__SHIFT 0x2
+#define RLC_MEM_SLP_CNTL__RLC_LS_DS_BUSY_OVERRIDE_MASK 0x80
+#define RLC_MEM_SLP_CNTL__RLC_LS_DS_BUSY_OVERRIDE__SHIFT 0x7
+#define RLC_MEM_SLP_CNTL__RLC_MEM_LS_ON_DELAY_MASK 0xff00
+#define RLC_MEM_SLP_CNTL__RLC_MEM_LS_ON_DELAY__SHIFT 0x8
+#define RLC_MEM_SLP_CNTL__RLC_MEM_LS_OFF_DELAY_MASK 0xff0000
+#define RLC_MEM_SLP_CNTL__RLC_MEM_LS_OFF_DELAY__SHIFT 0x10
+#define RLC_MEM_SLP_CNTL__RESERVED1_MASK 0xff000000
+#define RLC_MEM_SLP_CNTL__RESERVED1__SHIFT 0x18
+#define SMU_RLC_RESPONSE__RESP_MASK 0xffffffff
+#define SMU_RLC_RESPONSE__RESP__SHIFT 0x0
+#define RLC_RLCV_SAFE_MODE__CMD_MASK 0x1
+#define RLC_RLCV_SAFE_MODE__CMD__SHIFT 0x0
+#define RLC_RLCV_SAFE_MODE__MESSAGE_MASK 0x1e
+#define RLC_RLCV_SAFE_MODE__MESSAGE__SHIFT 0x1
+#define RLC_RLCV_SAFE_MODE__RESERVED1_MASK 0xe0
+#define RLC_RLCV_SAFE_MODE__RESERVED1__SHIFT 0x5
+#define RLC_RLCV_SAFE_MODE__RESPONSE_MASK 0xf00
+#define RLC_RLCV_SAFE_MODE__RESPONSE__SHIFT 0x8
+#define RLC_RLCV_SAFE_MODE__RESERVED_MASK 0xfffff000
+#define RLC_RLCV_SAFE_MODE__RESERVED__SHIFT 0xc
+#define RLC_SMU_SAFE_MODE__CMD_MASK 0x1
+#define RLC_SMU_SAFE_MODE__CMD__SHIFT 0x0
+#define RLC_SMU_SAFE_MODE__MESSAGE_MASK 0x1e
+#define RLC_SMU_SAFE_MODE__MESSAGE__SHIFT 0x1
+#define RLC_SMU_SAFE_MODE__RESERVED1_MASK 0xe0
+#define RLC_SMU_SAFE_MODE__RESERVED1__SHIFT 0x5
+#define RLC_SMU_SAFE_MODE__RESPONSE_MASK 0xf00
+#define RLC_SMU_SAFE_MODE__RESPONSE__SHIFT 0x8
+#define RLC_SMU_SAFE_MODE__RESERVED_MASK 0xfffff000
+#define RLC_SMU_SAFE_MODE__RESERVED__SHIFT 0xc
+#define RLC_RLCV_COMMAND__CMD_MASK 0xf
+#define RLC_RLCV_COMMAND__CMD__SHIFT 0x0
+#define RLC_RLCV_COMMAND__RESERVED_MASK 0xfffffff0
+#define RLC_RLCV_COMMAND__RESERVED__SHIFT 0x4
+#define RLC_CLK_CNTL__RLC_SRM_CLK_CNTL_MASK 0x1
+#define RLC_CLK_CNTL__RLC_SRM_CLK_CNTL__SHIFT 0x0
+#define RLC_CLK_CNTL__RLC_SPM_CLK_CNTL_MASK 0x2
+#define RLC_CLK_CNTL__RLC_SPM_CLK_CNTL__SHIFT 0x1
+#define RLC_CLK_CNTL__RESERVED_MASK 0xfffffffc
+#define RLC_CLK_CNTL__RESERVED__SHIFT 0x2
+#define RLC_PERFMON_CLK_CNTL__PERFMON_CLOCK_STATE_MASK 0x1
+#define RLC_PERFMON_CLK_CNTL__PERFMON_CLOCK_STATE__SHIFT 0x0
+#define RLC_PERFMON_CNTL__PERFMON_STATE_MASK 0x7
+#define RLC_PERFMON_CNTL__PERFMON_STATE__SHIFT 0x0
+#define RLC_PERFMON_CNTL__PERFMON_SAMPLE_ENABLE_MASK 0x400
+#define RLC_PERFMON_CNTL__PERFMON_SAMPLE_ENABLE__SHIFT 0xa
+#define RLC_PERFCOUNTER0_SELECT__PERFCOUNTER_SELECT_MASK 0xff
+#define RLC_PERFCOUNTER0_SELECT__PERFCOUNTER_SELECT__SHIFT 0x0
+#define RLC_PERFCOUNTER1_SELECT__PERFCOUNTER_SELECT_MASK 0xff
+#define RLC_PERFCOUNTER1_SELECT__PERFCOUNTER_SELECT__SHIFT 0x0
+#define RLC_PERFCOUNTER0_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define RLC_PERFCOUNTER0_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define RLC_PERFCOUNTER1_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define RLC_PERFCOUNTER1_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define RLC_PERFCOUNTER0_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define RLC_PERFCOUNTER0_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define RLC_PERFCOUNTER1_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define RLC_PERFCOUNTER1_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define CGTT_RLC_CLK_CTRL__ON_DELAY_MASK 0xf
+#define CGTT_RLC_CLK_CTRL__ON_DELAY__SHIFT 0x0
+#define CGTT_RLC_CLK_CTRL__OFF_HYSTERESIS_MASK 0xff0
+#define CGTT_RLC_CLK_CTRL__OFF_HYSTERESIS__SHIFT 0x4
+#define CGTT_RLC_CLK_CTRL__SOFT_OVERRIDE_DYN_MASK 0x40000000
+#define CGTT_RLC_CLK_CTRL__SOFT_OVERRIDE_DYN__SHIFT 0x1e
+#define CGTT_RLC_CLK_CTRL__SOFT_OVERRIDE_REG_MASK 0x80000000
+#define CGTT_RLC_CLK_CTRL__SOFT_OVERRIDE_REG__SHIFT 0x1f
+#define RLC_LB_CNTL__LOAD_BALANCE_ENABLE_MASK 0x1
+#define RLC_LB_CNTL__LOAD_BALANCE_ENABLE__SHIFT 0x0
+#define RLC_LB_CNTL__LB_CNT_CP_BUSY_MASK 0x2
+#define RLC_LB_CNTL__LB_CNT_CP_BUSY__SHIFT 0x1
+#define RLC_LB_CNTL__LB_CNT_SPIM_ACTIVE_MASK 0x4
+#define RLC_LB_CNTL__LB_CNT_SPIM_ACTIVE__SHIFT 0x2
+#define RLC_LB_CNTL__LB_CNT_REG_INC_MASK 0x8
+#define RLC_LB_CNTL__LB_CNT_REG_INC__SHIFT 0x3
+#define RLC_LB_CNTL__CU_MASK_USED_OFF_HYST_MASK 0xff0
+#define RLC_LB_CNTL__CU_MASK_USED_OFF_HYST__SHIFT 0x4
+#define RLC_LB_CNTL__RESERVED_MASK 0xfffff000
+#define RLC_LB_CNTL__RESERVED__SHIFT 0xc
+#define RLC_LB_CNTR_MAX__LB_CNTR_MAX_MASK 0xffffffff
+#define RLC_LB_CNTR_MAX__LB_CNTR_MAX__SHIFT 0x0
+#define RLC_LB_CNTR_INIT__LB_CNTR_INIT_MASK 0xffffffff
+#define RLC_LB_CNTR_INIT__LB_CNTR_INIT__SHIFT 0x0
+#define RLC_LOAD_BALANCE_CNTR__RLC_LOAD_BALANCE_CNTR_MASK 0xffffffff
+#define RLC_LOAD_BALANCE_CNTR__RLC_LOAD_BALANCE_CNTR__SHIFT 0x0
+#define RLC_JUMP_TABLE_RESTORE__ADDR_MASK 0xffffffff
+#define RLC_JUMP_TABLE_RESTORE__ADDR__SHIFT 0x0
+#define RLC_PG_DELAY_2__SERDES_TIMEOUT_VALUE_MASK 0xff
+#define RLC_PG_DELAY_2__SERDES_TIMEOUT_VALUE__SHIFT 0x0
+#define RLC_PG_DELAY_2__SERDES_CMD_DELAY_MASK 0xff00
+#define RLC_PG_DELAY_2__SERDES_CMD_DELAY__SHIFT 0x8
+#define RLC_PG_DELAY_2__PERCU_TIMEOUT_VALUE_MASK 0xffff0000
+#define RLC_PG_DELAY_2__PERCU_TIMEOUT_VALUE__SHIFT 0x10
+#define RLC_GPM_DEBUG_SELECT__SELECT_MASK 0xff
+#define RLC_GPM_DEBUG_SELECT__SELECT__SHIFT 0x0
+#define RLC_GPM_DEBUG_SELECT__F32_DEBUG_SELECT_MASK 0x300
+#define RLC_GPM_DEBUG_SELECT__F32_DEBUG_SELECT__SHIFT 0x8
+#define RLC_GPM_DEBUG_SELECT__RESERVED_MASK 0xfffffc00
+#define RLC_GPM_DEBUG_SELECT__RESERVED__SHIFT 0xa
+#define RLC_GPM_DEBUG__DATA_MASK 0xffffffff
+#define RLC_GPM_DEBUG__DATA__SHIFT 0x0
+#define RLC_GPM_DEBUG_INST_A__INST_A_MASK 0xffffffff
+#define RLC_GPM_DEBUG_INST_A__INST_A__SHIFT 0x0
+#define RLC_GPM_DEBUG_INST_B__INST_B_MASK 0xffffffff
+#define RLC_GPM_DEBUG_INST_B__INST_B__SHIFT 0x0
+#define RLC_GPM_DEBUG_INST_ADDR__ADRR_A_MASK 0xffff
+#define RLC_GPM_DEBUG_INST_ADDR__ADRR_A__SHIFT 0x0
+#define RLC_GPM_DEBUG_INST_ADDR__ADDR_B_MASK 0xffff0000
+#define RLC_GPM_DEBUG_INST_ADDR__ADDR_B__SHIFT 0x10
+#define RLC_GPM_UCODE_ADDR__UCODE_ADDR_MASK 0xfff
+#define RLC_GPM_UCODE_ADDR__UCODE_ADDR__SHIFT 0x0
+#define RLC_GPM_UCODE_ADDR__RESERVED_MASK 0xfffff000
+#define RLC_GPM_UCODE_ADDR__RESERVED__SHIFT 0xc
+#define RLC_GPM_UCODE_DATA__UCODE_DATA_MASK 0xffffffff
+#define RLC_GPM_UCODE_DATA__UCODE_DATA__SHIFT 0x0
+#define GPU_BIST_CONTROL__STOP_ON_FAIL_HW_MASK 0x1
+#define GPU_BIST_CONTROL__STOP_ON_FAIL_HW__SHIFT 0x0
+#define GPU_BIST_CONTROL__STOP_ON_FAIL_CU_HARV_MASK 0x2
+#define GPU_BIST_CONTROL__STOP_ON_FAIL_CU_HARV__SHIFT 0x1
+#define GPU_BIST_CONTROL__CU_HARV_LOOP_COUNT_MASK 0x3c
+#define GPU_BIST_CONTROL__CU_HARV_LOOP_COUNT__SHIFT 0x2
+#define GPU_BIST_CONTROL__RESERVED_MASK 0xffff80
+#define GPU_BIST_CONTROL__RESERVED__SHIFT 0x7
+#define GPU_BIST_CONTROL__GLOBAL_LOOP_COUNT_MASK 0xff000000
+#define GPU_BIST_CONTROL__GLOBAL_LOOP_COUNT__SHIFT 0x18
+#define RLC_ROM_CNTL__USE_ROM_MASK 0x1
+#define RLC_ROM_CNTL__USE_ROM__SHIFT 0x0
+#define RLC_ROM_CNTL__SLP_MODE_EN_MASK 0x2
+#define RLC_ROM_CNTL__SLP_MODE_EN__SHIFT 0x1
+#define RLC_ROM_CNTL__EFUSE_DISTRIB_EN_MASK 0x4
+#define RLC_ROM_CNTL__EFUSE_DISTRIB_EN__SHIFT 0x2
+#define RLC_ROM_CNTL__HELLOWORLD_EN_MASK 0x8
+#define RLC_ROM_CNTL__HELLOWORLD_EN__SHIFT 0x3
+#define RLC_ROM_CNTL__CU_HARVEST_EN_MASK 0x10
+#define RLC_ROM_CNTL__CU_HARVEST_EN__SHIFT 0x4
+#define RLC_ROM_CNTL__RESERVED_MASK 0xffffffe0
+#define RLC_ROM_CNTL__RESERVED__SHIFT 0x5
+#define RLC_GPU_CLOCK_COUNT_LSB__GPU_CLOCKS_LSB_MASK 0xffffffff
+#define RLC_GPU_CLOCK_COUNT_LSB__GPU_CLOCKS_LSB__SHIFT 0x0
+#define RLC_GPU_CLOCK_COUNT_MSB__GPU_CLOCKS_MSB_MASK 0xffffffff
+#define RLC_GPU_CLOCK_COUNT_MSB__GPU_CLOCKS_MSB__SHIFT 0x0
+#define RLC_CAPTURE_GPU_CLOCK_COUNT__CAPTURE_MASK 0x1
+#define RLC_CAPTURE_GPU_CLOCK_COUNT__CAPTURE__SHIFT 0x0
+#define RLC_CAPTURE_GPU_CLOCK_COUNT__RESERVED_MASK 0xfffffffe
+#define RLC_CAPTURE_GPU_CLOCK_COUNT__RESERVED__SHIFT 0x1
+#define RLC_UCODE_CNTL__RLC_UCODE_FLAGS_MASK 0xffffffff
+#define RLC_UCODE_CNTL__RLC_UCODE_FLAGS__SHIFT 0x0
+#define RLC_GPM_STAT__RLC_BUSY_MASK 0x1
+#define RLC_GPM_STAT__RLC_BUSY__SHIFT 0x0
+#define RLC_GPM_STAT__GFX_POWER_STATUS_MASK 0x2
+#define RLC_GPM_STAT__GFX_POWER_STATUS__SHIFT 0x1
+#define RLC_GPM_STAT__GFX_CLOCK_STATUS_MASK 0x4
+#define RLC_GPM_STAT__GFX_CLOCK_STATUS__SHIFT 0x2
+#define RLC_GPM_STAT__GFX_LS_STATUS_MASK 0x8
+#define RLC_GPM_STAT__GFX_LS_STATUS__SHIFT 0x3
+#define RLC_GPM_STAT__GFX_PIPELINE_POWER_STATUS_MASK 0x10
+#define RLC_GPM_STAT__GFX_PIPELINE_POWER_STATUS__SHIFT 0x4
+#define RLC_GPM_STAT__CNTX_IDLE_BEING_PROCESSED_MASK 0x20
+#define RLC_GPM_STAT__CNTX_IDLE_BEING_PROCESSED__SHIFT 0x5
+#define RLC_GPM_STAT__CNTX_BUSY_BEING_PROCESSED_MASK 0x40
+#define RLC_GPM_STAT__CNTX_BUSY_BEING_PROCESSED__SHIFT 0x6
+#define RLC_GPM_STAT__GFX_IDLE_BEING_PROCESSED_MASK 0x80
+#define RLC_GPM_STAT__GFX_IDLE_BEING_PROCESSED__SHIFT 0x7
+#define RLC_GPM_STAT__CMP_BUSY_BEING_PROCESSED_MASK 0x100
+#define RLC_GPM_STAT__CMP_BUSY_BEING_PROCESSED__SHIFT 0x8
+#define RLC_GPM_STAT__SAVING_REGISTERS_MASK 0x200
+#define RLC_GPM_STAT__SAVING_REGISTERS__SHIFT 0x9
+#define RLC_GPM_STAT__RESTORING_REGISTERS_MASK 0x400
+#define RLC_GPM_STAT__RESTORING_REGISTERS__SHIFT 0xa
+#define RLC_GPM_STAT__GFX3D_BLOCKS_CHANGING_POWER_STATE_MASK 0x800
+#define RLC_GPM_STAT__GFX3D_BLOCKS_CHANGING_POWER_STATE__SHIFT 0xb
+#define RLC_GPM_STAT__CMP_BLOCKS_CHANGING_POWER_STATE_MASK 0x1000
+#define RLC_GPM_STAT__CMP_BLOCKS_CHANGING_POWER_STATE__SHIFT 0xc
+#define RLC_GPM_STAT__STATIC_CU_POWERING_UP_MASK 0x2000
+#define RLC_GPM_STAT__STATIC_CU_POWERING_UP__SHIFT 0xd
+#define RLC_GPM_STAT__STATIC_CU_POWERING_DOWN_MASK 0x4000
+#define RLC_GPM_STAT__STATIC_CU_POWERING_DOWN__SHIFT 0xe
+#define RLC_GPM_STAT__DYN_CU_POWERING_UP_MASK 0x8000
+#define RLC_GPM_STAT__DYN_CU_POWERING_UP__SHIFT 0xf
+#define RLC_GPM_STAT__DYN_CU_POWERING_DOWN_MASK 0x10000
+#define RLC_GPM_STAT__DYN_CU_POWERING_DOWN__SHIFT 0x10
+#define RLC_GPM_STAT__ABORTED_PD_SEQUENCE_MASK 0x20000
+#define RLC_GPM_STAT__ABORTED_PD_SEQUENCE__SHIFT 0x11
+#define RLC_GPM_STAT__RESERVED_MASK 0xfc0000
+#define RLC_GPM_STAT__RESERVED__SHIFT 0x12
+#define RLC_GPM_STAT__PG_ERROR_STATUS_MASK 0xff000000
+#define RLC_GPM_STAT__PG_ERROR_STATUS__SHIFT 0x18
+#define RLC_GPU_CLOCK_32_RES_SEL__RES_SEL_MASK 0x3f
+#define RLC_GPU_CLOCK_32_RES_SEL__RES_SEL__SHIFT 0x0
+#define RLC_GPU_CLOCK_32_RES_SEL__RESERVED_MASK 0xffffffc0
+#define RLC_GPU_CLOCK_32_RES_SEL__RESERVED__SHIFT 0x6
+#define RLC_GPU_CLOCK_32__GPU_CLOCK_32_MASK 0xffffffff
+#define RLC_GPU_CLOCK_32__GPU_CLOCK_32__SHIFT 0x0
+#define RLC_PG_CNTL__GFX_POWER_GATING_ENABLE_MASK 0x1
+#define RLC_PG_CNTL__GFX_POWER_GATING_ENABLE__SHIFT 0x0
+#define RLC_PG_CNTL__GFX_POWER_GATING_SRC_MASK 0x2
+#define RLC_PG_CNTL__GFX_POWER_GATING_SRC__SHIFT 0x1
+#define RLC_PG_CNTL__DYN_PER_CU_PG_ENABLE_MASK 0x4
+#define RLC_PG_CNTL__DYN_PER_CU_PG_ENABLE__SHIFT 0x2
+#define RLC_PG_CNTL__STATIC_PER_CU_PG_ENABLE_MASK 0x8
+#define RLC_PG_CNTL__STATIC_PER_CU_PG_ENABLE__SHIFT 0x3
+#define RLC_PG_CNTL__GFX_PIPELINE_PG_ENABLE_MASK 0x10
+#define RLC_PG_CNTL__GFX_PIPELINE_PG_ENABLE__SHIFT 0x4
+#define RLC_PG_CNTL__RESERVED_MASK 0x3fe0
+#define RLC_PG_CNTL__RESERVED__SHIFT 0x5
+#define RLC_PG_CNTL__PG_OVERRIDE_MASK 0x4000
+#define RLC_PG_CNTL__PG_OVERRIDE__SHIFT 0xe
+#define RLC_PG_CNTL__CP_PG_DISABLE_MASK 0x8000
+#define RLC_PG_CNTL__CP_PG_DISABLE__SHIFT 0xf
+#define RLC_PG_CNTL__CHUB_HANDSHAKE_ENABLE_MASK 0x10000
+#define RLC_PG_CNTL__CHUB_HANDSHAKE_ENABLE__SHIFT 0x10
+#define RLC_PG_CNTL__SMU_CLK_SLOWDOWN_ON_PU_ENABLE_MASK 0x20000
+#define RLC_PG_CNTL__SMU_CLK_SLOWDOWN_ON_PU_ENABLE__SHIFT 0x11
+#define RLC_PG_CNTL__SMU_CLK_SLOWDOWN_ON_PD_ENABLE_MASK 0x40000
+#define RLC_PG_CNTL__SMU_CLK_SLOWDOWN_ON_PD_ENABLE__SHIFT 0x12
+#define RLC_PG_CNTL__SMU_HANDSHAKE_ENABLE_MASK 0x80000
+#define RLC_PG_CNTL__SMU_HANDSHAKE_ENABLE__SHIFT 0x13
+#define RLC_PG_CNTL__RESERVED1_MASK 0xf00000
+#define RLC_PG_CNTL__RESERVED1__SHIFT 0x14
+#define RLC_GPM_THREAD_PRIORITY__THREAD0_PRIORITY_MASK 0xff
+#define RLC_GPM_THREAD_PRIORITY__THREAD0_PRIORITY__SHIFT 0x0
+#define RLC_GPM_THREAD_PRIORITY__THREAD1_PRIORITY_MASK 0xff00
+#define RLC_GPM_THREAD_PRIORITY__THREAD1_PRIORITY__SHIFT 0x8
+#define RLC_GPM_THREAD_PRIORITY__THREAD2_PRIORITY_MASK 0xff0000
+#define RLC_GPM_THREAD_PRIORITY__THREAD2_PRIORITY__SHIFT 0x10
+#define RLC_GPM_THREAD_PRIORITY__THREAD3_PRIORITY_MASK 0xff000000
+#define RLC_GPM_THREAD_PRIORITY__THREAD3_PRIORITY__SHIFT 0x18
+#define RLC_GPM_THREAD_ENABLE__THREAD0_ENABLE_MASK 0x1
+#define RLC_GPM_THREAD_ENABLE__THREAD0_ENABLE__SHIFT 0x0
+#define RLC_GPM_THREAD_ENABLE__THREAD1_ENABLE_MASK 0x2
+#define RLC_GPM_THREAD_ENABLE__THREAD1_ENABLE__SHIFT 0x1
+#define RLC_GPM_THREAD_ENABLE__THREAD2_ENABLE_MASK 0x4
+#define RLC_GPM_THREAD_ENABLE__THREAD2_ENABLE__SHIFT 0x2
+#define RLC_GPM_THREAD_ENABLE__THREAD3_ENABLE_MASK 0x8
+#define RLC_GPM_THREAD_ENABLE__THREAD3_ENABLE__SHIFT 0x3
+#define RLC_GPM_THREAD_ENABLE__RESERVED_MASK 0xfffffff0
+#define RLC_GPM_THREAD_ENABLE__RESERVED__SHIFT 0x4
+#define RLC_GPM_VMID_THREAD0__RLC_VMID_MASK 0xf
+#define RLC_GPM_VMID_THREAD0__RLC_VMID__SHIFT 0x0
+#define RLC_GPM_VMID_THREAD0__RESERVED0_MASK 0xf0
+#define RLC_GPM_VMID_THREAD0__RESERVED0__SHIFT 0x4
+#define RLC_GPM_VMID_THREAD0__RLC_QUEUEID_MASK 0x700
+#define RLC_GPM_VMID_THREAD0__RLC_QUEUEID__SHIFT 0x8
+#define RLC_GPM_VMID_THREAD0__RESERVED1_MASK 0xfffff800
+#define RLC_GPM_VMID_THREAD0__RESERVED1__SHIFT 0xb
+#define RLC_GPM_VMID_THREAD1__RLC_VMID_MASK 0xf
+#define RLC_GPM_VMID_THREAD1__RLC_VMID__SHIFT 0x0
+#define RLC_GPM_VMID_THREAD1__RESERVED0_MASK 0xf0
+#define RLC_GPM_VMID_THREAD1__RESERVED0__SHIFT 0x4
+#define RLC_GPM_VMID_THREAD1__RLC_QUEUEID_MASK 0x700
+#define RLC_GPM_VMID_THREAD1__RLC_QUEUEID__SHIFT 0x8
+#define RLC_GPM_VMID_THREAD1__RESERVED1_MASK 0xfffff800
+#define RLC_GPM_VMID_THREAD1__RESERVED1__SHIFT 0xb
+#define RLC_CGTT_MGCG_OVERRIDE__OVERRIDE_MASK 0xffffffff
+#define RLC_CGTT_MGCG_OVERRIDE__OVERRIDE__SHIFT 0x0
+#define RLC_CGCG_CGLS_CTRL__CGCG_EN_MASK 0x1
+#define RLC_CGCG_CGLS_CTRL__CGCG_EN__SHIFT 0x0
+#define RLC_CGCG_CGLS_CTRL__CGLS_EN_MASK 0x2
+#define RLC_CGCG_CGLS_CTRL__CGLS_EN__SHIFT 0x1
+#define RLC_CGCG_CGLS_CTRL__CGLS_REP_COMPANSAT_DELAY_MASK 0xfc
+#define RLC_CGCG_CGLS_CTRL__CGLS_REP_COMPANSAT_DELAY__SHIFT 0x2
+#define RLC_CGCG_CGLS_CTRL__CGCG_GFX_IDLE_THRESHOLD_MASK 0x7ffff00
+#define RLC_CGCG_CGLS_CTRL__CGCG_GFX_IDLE_THRESHOLD__SHIFT 0x8
+#define RLC_CGCG_CGLS_CTRL__CGCG_CONTROLLER_MASK 0x8000000
+#define RLC_CGCG_CGLS_CTRL__CGCG_CONTROLLER__SHIFT 0x1b
+#define RLC_CGCG_CGLS_CTRL__CGCG_REG_CTRL_MASK 0x10000000
+#define RLC_CGCG_CGLS_CTRL__CGCG_REG_CTRL__SHIFT 0x1c
+#define RLC_CGCG_CGLS_CTRL__SLEEP_MODE_MASK 0x60000000
+#define RLC_CGCG_CGLS_CTRL__SLEEP_MODE__SHIFT 0x1d
+#define RLC_CGCG_CGLS_CTRL__SIM_SILICON_EN_MASK 0x80000000
+#define RLC_CGCG_CGLS_CTRL__SIM_SILICON_EN__SHIFT 0x1f
+#define RLC_CGCG_RAMP_CTRL__DOWN_DIV_START_UNIT_MASK 0xf
+#define RLC_CGCG_RAMP_CTRL__DOWN_DIV_START_UNIT__SHIFT 0x0
+#define RLC_CGCG_RAMP_CTRL__DOWN_DIV_STEP_UNIT_MASK 0xf0
+#define RLC_CGCG_RAMP_CTRL__DOWN_DIV_STEP_UNIT__SHIFT 0x4
+#define RLC_CGCG_RAMP_CTRL__UP_DIV_START_UNIT_MASK 0xf00
+#define RLC_CGCG_RAMP_CTRL__UP_DIV_START_UNIT__SHIFT 0x8
+#define RLC_CGCG_RAMP_CTRL__UP_DIV_STEP_UNIT_MASK 0xf000
+#define RLC_CGCG_RAMP_CTRL__UP_DIV_STEP_UNIT__SHIFT 0xc
+#define RLC_CGCG_RAMP_CTRL__STEP_DELAY_CNT_MASK 0xfff0000
+#define RLC_CGCG_RAMP_CTRL__STEP_DELAY_CNT__SHIFT 0x10
+#define RLC_CGCG_RAMP_CTRL__STEP_DELAY_UNIT_MASK 0xf0000000
+#define RLC_CGCG_RAMP_CTRL__STEP_DELAY_UNIT__SHIFT 0x1c
+#define RLC_DYN_PG_STATUS__PG_STATUS_CU_MASK_MASK 0xffffffff
+#define RLC_DYN_PG_STATUS__PG_STATUS_CU_MASK__SHIFT 0x0
+#define RLC_DYN_PG_REQUEST__PG_REQUEST_CU_MASK_MASK 0xffffffff
+#define RLC_DYN_PG_REQUEST__PG_REQUEST_CU_MASK__SHIFT 0x0
+#define RLC_PG_DELAY__POWER_UP_DELAY_MASK 0xff
+#define RLC_PG_DELAY__POWER_UP_DELAY__SHIFT 0x0
+#define RLC_PG_DELAY__POWER_DOWN_DELAY_MASK 0xff00
+#define RLC_PG_DELAY__POWER_DOWN_DELAY__SHIFT 0x8
+#define RLC_PG_DELAY__CMD_PROPAGATE_DELAY_MASK 0xff0000
+#define RLC_PG_DELAY__CMD_PROPAGATE_DELAY__SHIFT 0x10
+#define RLC_PG_DELAY__MEM_SLEEP_DELAY_MASK 0xff000000
+#define RLC_PG_DELAY__MEM_SLEEP_DELAY__SHIFT 0x18
+#define RLC_CU_STATUS__WORK_PENDING_MASK 0xffffffff
+#define RLC_CU_STATUS__WORK_PENDING__SHIFT 0x0
+#define RLC_LB_INIT_CU_MASK__INIT_CU_MASK_MASK 0xffffffff
+#define RLC_LB_INIT_CU_MASK__INIT_CU_MASK__SHIFT 0x0
+#define RLC_LB_ALWAYS_ACTIVE_CU_MASK__ALWAYS_ACTIVE_CU_MASK_MASK 0xffffffff
+#define RLC_LB_ALWAYS_ACTIVE_CU_MASK__ALWAYS_ACTIVE_CU_MASK__SHIFT 0x0
+#define RLC_LB_PARAMS__SKIP_L2_CHECK_MASK 0x1
+#define RLC_LB_PARAMS__SKIP_L2_CHECK__SHIFT 0x0
+#define RLC_LB_PARAMS__FIFO_SAMPLES_MASK 0xfe
+#define RLC_LB_PARAMS__FIFO_SAMPLES__SHIFT 0x1
+#define RLC_LB_PARAMS__PG_IDLE_SAMPLES_MASK 0xff00
+#define RLC_LB_PARAMS__PG_IDLE_SAMPLES__SHIFT 0x8
+#define RLC_LB_PARAMS__PG_IDLE_SAMPLE_INTERVAL_MASK 0xffff0000
+#define RLC_LB_PARAMS__PG_IDLE_SAMPLE_INTERVAL__SHIFT 0x10
+#define RLC_THREAD1_DELAY__CU_IDEL_DELAY_MASK 0xff
+#define RLC_THREAD1_DELAY__CU_IDEL_DELAY__SHIFT 0x0
+#define RLC_THREAD1_DELAY__LBPW_INNER_LOOP_DELAY_MASK 0xff00
+#define RLC_THREAD1_DELAY__LBPW_INNER_LOOP_DELAY__SHIFT 0x8
+#define RLC_THREAD1_DELAY__LBPW_OUTER_LOOP_DELAY_MASK 0xff0000
+#define RLC_THREAD1_DELAY__LBPW_OUTER_LOOP_DELAY__SHIFT 0x10
+#define RLC_THREAD1_DELAY__SPARE_MASK 0xff000000
+#define RLC_THREAD1_DELAY__SPARE__SHIFT 0x18
+#define RLC_PG_ALWAYS_ON_CU_MASK__AON_CU_MASK_MASK 0xffffffff
+#define RLC_PG_ALWAYS_ON_CU_MASK__AON_CU_MASK__SHIFT 0x0
+#define RLC_MAX_PG_CU__MAX_POWERED_UP_CU_MASK 0xff
+#define RLC_MAX_PG_CU__MAX_POWERED_UP_CU__SHIFT 0x0
+#define RLC_MAX_PG_CU__SPARE_MASK 0xffffff00
+#define RLC_MAX_PG_CU__SPARE__SHIFT 0x8
+#define RLC_AUTO_PG_CTRL__AUTO_PG_EN_MASK 0x1
+#define RLC_AUTO_PG_CTRL__AUTO_PG_EN__SHIFT 0x0
+#define RLC_AUTO_PG_CTRL__AUTO_GRBM_REG_SAVE_ON_IDLE_EN_MASK 0x2
+#define RLC_AUTO_PG_CTRL__AUTO_GRBM_REG_SAVE_ON_IDLE_EN__SHIFT 0x1
+#define RLC_AUTO_PG_CTRL__AUTO_WAKE_UP_EN_MASK 0x4
+#define RLC_AUTO_PG_CTRL__AUTO_WAKE_UP_EN__SHIFT 0x2
+#define RLC_AUTO_PG_CTRL__GRBM_REG_SAVE_GFX_IDLE_THRESHOLD_MASK 0x7fff8
+#define RLC_AUTO_PG_CTRL__GRBM_REG_SAVE_GFX_IDLE_THRESHOLD__SHIFT 0x3
+#define RLC_AUTO_PG_CTRL__PG_AFTER_GRBM_REG_SAVE_THRESHOLD_MASK 0xfff80000
+#define RLC_AUTO_PG_CTRL__PG_AFTER_GRBM_REG_SAVE_THRESHOLD__SHIFT 0x13
+#define RLC_SMU_GRBM_REG_SAVE_CTRL__START_GRBM_REG_SAVE_MASK 0x1
+#define RLC_SMU_GRBM_REG_SAVE_CTRL__START_GRBM_REG_SAVE__SHIFT 0x0
+#define RLC_SMU_GRBM_REG_SAVE_CTRL__SPARE_MASK 0xfffffffe
+#define RLC_SMU_GRBM_REG_SAVE_CTRL__SPARE__SHIFT 0x1
+#define RLC_SERDES_RD_MASTER_INDEX__CU_ID_MASK 0xf
+#define RLC_SERDES_RD_MASTER_INDEX__CU_ID__SHIFT 0x0
+#define RLC_SERDES_RD_MASTER_INDEX__SH_ID_MASK 0x30
+#define RLC_SERDES_RD_MASTER_INDEX__SH_ID__SHIFT 0x4
+#define RLC_SERDES_RD_MASTER_INDEX__SE_ID_MASK 0x1c0
+#define RLC_SERDES_RD_MASTER_INDEX__SE_ID__SHIFT 0x6
+#define RLC_SERDES_RD_MASTER_INDEX__SE_NONCU_ID_MASK 0x200
+#define RLC_SERDES_RD_MASTER_INDEX__SE_NONCU_ID__SHIFT 0x9
+#define RLC_SERDES_RD_MASTER_INDEX__SE_NONCU_MASK 0x400
+#define RLC_SERDES_RD_MASTER_INDEX__SE_NONCU__SHIFT 0xa
+#define RLC_SERDES_RD_MASTER_INDEX__NON_SE_MASK 0x7800
+#define RLC_SERDES_RD_MASTER_INDEX__NON_SE__SHIFT 0xb
+#define RLC_SERDES_RD_MASTER_INDEX__DATA_REG_ID_MASK 0x18000
+#define RLC_SERDES_RD_MASTER_INDEX__DATA_REG_ID__SHIFT 0xf
+#define RLC_SERDES_RD_MASTER_INDEX__SPARE_MASK 0xfffe0000
+#define RLC_SERDES_RD_MASTER_INDEX__SPARE__SHIFT 0x11
+#define RLC_SERDES_RD_DATA_0__DATA_MASK 0xffffffff
+#define RLC_SERDES_RD_DATA_0__DATA__SHIFT 0x0
+#define RLC_SERDES_RD_DATA_1__DATA_MASK 0xffffffff
+#define RLC_SERDES_RD_DATA_1__DATA__SHIFT 0x0
+#define RLC_SERDES_RD_DATA_2__DATA_MASK 0xffffffff
+#define RLC_SERDES_RD_DATA_2__DATA__SHIFT 0x0
+#define RLC_SERDES_WR_CU_MASTER_MASK__MASTER_MASK_MASK 0xffffffff
+#define RLC_SERDES_WR_CU_MASTER_MASK__MASTER_MASK__SHIFT 0x0
+#define RLC_SERDES_WR_NONCU_MASTER_MASK__SE_MASTER_MASK_MASK 0xffff
+#define RLC_SERDES_WR_NONCU_MASTER_MASK__SE_MASTER_MASK__SHIFT 0x0
+#define RLC_SERDES_WR_NONCU_MASTER_MASK__GC_MASTER_MASK_MASK 0x10000
+#define RLC_SERDES_WR_NONCU_MASTER_MASK__GC_MASTER_MASK__SHIFT 0x10
+#define RLC_SERDES_WR_NONCU_MASTER_MASK__GC_GFX_MASTER_MASK_MASK 0x20000
+#define RLC_SERDES_WR_NONCU_MASTER_MASK__GC_GFX_MASTER_MASK__SHIFT 0x11
+#define RLC_SERDES_WR_NONCU_MASTER_MASK__TC0_MASTER_MASK_MASK 0x40000
+#define RLC_SERDES_WR_NONCU_MASTER_MASK__TC0_MASTER_MASK__SHIFT 0x12
+#define RLC_SERDES_WR_NONCU_MASTER_MASK__TC1_MASTER_MASK_MASK 0x80000
+#define RLC_SERDES_WR_NONCU_MASTER_MASK__TC1_MASTER_MASK__SHIFT 0x13
+#define RLC_SERDES_WR_NONCU_MASTER_MASK__SPARE0_MASTER_MASK_MASK 0x100000
+#define RLC_SERDES_WR_NONCU_MASTER_MASK__SPARE0_MASTER_MASK__SHIFT 0x14
+#define RLC_SERDES_WR_NONCU_MASTER_MASK__SPARE1_MASTER_MASK_MASK 0x200000
+#define RLC_SERDES_WR_NONCU_MASTER_MASK__SPARE1_MASTER_MASK__SHIFT 0x15
+#define RLC_SERDES_WR_NONCU_MASTER_MASK__SPARE2_MASTER_MASK_MASK 0x400000
+#define RLC_SERDES_WR_NONCU_MASTER_MASK__SPARE2_MASTER_MASK__SHIFT 0x16
+#define RLC_SERDES_WR_NONCU_MASTER_MASK__SPARE3_MASTER_MASK_MASK 0x800000
+#define RLC_SERDES_WR_NONCU_MASTER_MASK__SPARE3_MASTER_MASK__SHIFT 0x17
+#define RLC_SERDES_WR_NONCU_MASTER_MASK__RESERVED_MASK 0xff000000
+#define RLC_SERDES_WR_NONCU_MASTER_MASK__RESERVED__SHIFT 0x18
+#define RLC_SERDES_WR_CTRL__BPM_ADDR_MASK 0xff
+#define RLC_SERDES_WR_CTRL__BPM_ADDR__SHIFT 0x0
+#define RLC_SERDES_WR_CTRL__POWER_DOWN_MASK 0x100
+#define RLC_SERDES_WR_CTRL__POWER_DOWN__SHIFT 0x8
+#define RLC_SERDES_WR_CTRL__POWER_UP_MASK 0x200
+#define RLC_SERDES_WR_CTRL__POWER_UP__SHIFT 0x9
+#define RLC_SERDES_WR_CTRL__P1_SELECT_MASK 0x400
+#define RLC_SERDES_WR_CTRL__P1_SELECT__SHIFT 0xa
+#define RLC_SERDES_WR_CTRL__P2_SELECT_MASK 0x800
+#define RLC_SERDES_WR_CTRL__P2_SELECT__SHIFT 0xb
+#define RLC_SERDES_WR_CTRL__WRITE_COMMAND_MASK 0x1000
+#define RLC_SERDES_WR_CTRL__WRITE_COMMAND__SHIFT 0xc
+#define RLC_SERDES_WR_CTRL__READ_COMMAND_MASK 0x2000
+#define RLC_SERDES_WR_CTRL__READ_COMMAND__SHIFT 0xd
+#define RLC_SERDES_WR_CTRL__RDDATA_RESET_MASK 0x4000
+#define RLC_SERDES_WR_CTRL__RDDATA_RESET__SHIFT 0xe
+#define RLC_SERDES_WR_CTRL__SHORT_FORMAT_MASK 0x8000
+#define RLC_SERDES_WR_CTRL__SHORT_FORMAT__SHIFT 0xf
+#define RLC_SERDES_WR_CTRL__BPM_DATA_MASK 0x3ff0000
+#define RLC_SERDES_WR_CTRL__BPM_DATA__SHIFT 0x10
+#define RLC_SERDES_WR_CTRL__SRBM_OVERRIDE_MASK 0x4000000
+#define RLC_SERDES_WR_CTRL__SRBM_OVERRIDE__SHIFT 0x1a
+#define RLC_SERDES_WR_CTRL__RSVD_BPM_ADDR_MASK 0x8000000
+#define RLC_SERDES_WR_CTRL__RSVD_BPM_ADDR__SHIFT 0x1b
+#define RLC_SERDES_WR_CTRL__REG_ADDR_MASK 0xf0000000
+#define RLC_SERDES_WR_CTRL__REG_ADDR__SHIFT 0x1c
+#define RLC_SERDES_WR_DATA__DATA_MASK 0xffffffff
+#define RLC_SERDES_WR_DATA__DATA__SHIFT 0x0
+#define RLC_SERDES_CU_MASTER_BUSY__BUSY_BUSY_MASK 0xffffffff
+#define RLC_SERDES_CU_MASTER_BUSY__BUSY_BUSY__SHIFT 0x0
+#define RLC_SERDES_NONCU_MASTER_BUSY__SE_MASTER_BUSY_MASK 0xffff
+#define RLC_SERDES_NONCU_MASTER_BUSY__SE_MASTER_BUSY__SHIFT 0x0
+#define RLC_SERDES_NONCU_MASTER_BUSY__GC_MASTER_BUSY_MASK 0x10000
+#define RLC_SERDES_NONCU_MASTER_BUSY__GC_MASTER_BUSY__SHIFT 0x10
+#define RLC_SERDES_NONCU_MASTER_BUSY__GC_GFX_MASTER_BUSY_MASK 0x20000
+#define RLC_SERDES_NONCU_MASTER_BUSY__GC_GFX_MASTER_BUSY__SHIFT 0x11
+#define RLC_SERDES_NONCU_MASTER_BUSY__TC0_MASTER_BUSY_MASK 0x40000
+#define RLC_SERDES_NONCU_MASTER_BUSY__TC0_MASTER_BUSY__SHIFT 0x12
+#define RLC_SERDES_NONCU_MASTER_BUSY__TC1_MASTER_BUSY_MASK 0x80000
+#define RLC_SERDES_NONCU_MASTER_BUSY__TC1_MASTER_BUSY__SHIFT 0x13
+#define RLC_SERDES_NONCU_MASTER_BUSY__SPARE0_MASTER_BUSY_MASK 0x100000
+#define RLC_SERDES_NONCU_MASTER_BUSY__SPARE0_MASTER_BUSY__SHIFT 0x14
+#define RLC_SERDES_NONCU_MASTER_BUSY__SPARE1_MASTER_BUSY_MASK 0x200000
+#define RLC_SERDES_NONCU_MASTER_BUSY__SPARE1_MASTER_BUSY__SHIFT 0x15
+#define RLC_SERDES_NONCU_MASTER_BUSY__SPARE2_MASTER_BUSY_MASK 0x400000
+#define RLC_SERDES_NONCU_MASTER_BUSY__SPARE2_MASTER_BUSY__SHIFT 0x16
+#define RLC_SERDES_NONCU_MASTER_BUSY__SPARE3_MASTER_BUSY_MASK 0x800000
+#define RLC_SERDES_NONCU_MASTER_BUSY__SPARE3_MASTER_BUSY__SHIFT 0x17
+#define RLC_SERDES_NONCU_MASTER_BUSY__RESERVED_MASK 0xff000000
+#define RLC_SERDES_NONCU_MASTER_BUSY__RESERVED__SHIFT 0x18
+#define RLC_GPM_GENERAL_0__DATA_MASK 0xffffffff
+#define RLC_GPM_GENERAL_0__DATA__SHIFT 0x0
+#define RLC_GPM_GENERAL_1__DATA_MASK 0xffffffff
+#define RLC_GPM_GENERAL_1__DATA__SHIFT 0x0
+#define RLC_GPM_GENERAL_2__DATA_MASK 0xffffffff
+#define RLC_GPM_GENERAL_2__DATA__SHIFT 0x0
+#define RLC_GPM_GENERAL_3__DATA_MASK 0xffffffff
+#define RLC_GPM_GENERAL_3__DATA__SHIFT 0x0
+#define RLC_GPM_GENERAL_4__DATA_MASK 0xffffffff
+#define RLC_GPM_GENERAL_4__DATA__SHIFT 0x0
+#define RLC_GPM_GENERAL_5__DATA_MASK 0xffffffff
+#define RLC_GPM_GENERAL_5__DATA__SHIFT 0x0
+#define RLC_GPM_GENERAL_6__DATA_MASK 0xffffffff
+#define RLC_GPM_GENERAL_6__DATA__SHIFT 0x0
+#define RLC_GPM_GENERAL_7__DATA_MASK 0xffffffff
+#define RLC_GPM_GENERAL_7__DATA__SHIFT 0x0
+#define RLC_GPM_SCRATCH_ADDR__ADDR_MASK 0x1ff
+#define RLC_GPM_SCRATCH_ADDR__ADDR__SHIFT 0x0
+#define RLC_GPM_SCRATCH_ADDR__RESERVED_MASK 0xfffffe00
+#define RLC_GPM_SCRATCH_ADDR__RESERVED__SHIFT 0x9
+#define RLC_GPM_SCRATCH_DATA__DATA_MASK 0xffffffff
+#define RLC_GPM_SCRATCH_DATA__DATA__SHIFT 0x0
+#define RLC_STATIC_PG_STATUS__PG_STATUS_CU_MASK_MASK 0xffffffff
+#define RLC_STATIC_PG_STATUS__PG_STATUS_CU_MASK__SHIFT 0x0
+#define RLC_GPM_PERF_COUNT_0__FEATURE_SEL_MASK 0xf
+#define RLC_GPM_PERF_COUNT_0__FEATURE_SEL__SHIFT 0x0
+#define RLC_GPM_PERF_COUNT_0__SE_INDEX_MASK 0xf0
+#define RLC_GPM_PERF_COUNT_0__SE_INDEX__SHIFT 0x4
+#define RLC_GPM_PERF_COUNT_0__SH_INDEX_MASK 0xf00
+#define RLC_GPM_PERF_COUNT_0__SH_INDEX__SHIFT 0x8
+#define RLC_GPM_PERF_COUNT_0__CU_INDEX_MASK 0xf000
+#define RLC_GPM_PERF_COUNT_0__CU_INDEX__SHIFT 0xc
+#define RLC_GPM_PERF_COUNT_0__EVENT_SEL_MASK 0x30000
+#define RLC_GPM_PERF_COUNT_0__EVENT_SEL__SHIFT 0x10
+#define RLC_GPM_PERF_COUNT_0__UNUSED_MASK 0xc0000
+#define RLC_GPM_PERF_COUNT_0__UNUSED__SHIFT 0x12
+#define RLC_GPM_PERF_COUNT_0__ENABLE_MASK 0x100000
+#define RLC_GPM_PERF_COUNT_0__ENABLE__SHIFT 0x14
+#define RLC_GPM_PERF_COUNT_0__RESERVED_MASK 0xffe00000
+#define RLC_GPM_PERF_COUNT_0__RESERVED__SHIFT 0x15
+#define RLC_GPM_PERF_COUNT_1__FEATURE_SEL_MASK 0xf
+#define RLC_GPM_PERF_COUNT_1__FEATURE_SEL__SHIFT 0x0
+#define RLC_GPM_PERF_COUNT_1__SE_INDEX_MASK 0xf0
+#define RLC_GPM_PERF_COUNT_1__SE_INDEX__SHIFT 0x4
+#define RLC_GPM_PERF_COUNT_1__SH_INDEX_MASK 0xf00
+#define RLC_GPM_PERF_COUNT_1__SH_INDEX__SHIFT 0x8
+#define RLC_GPM_PERF_COUNT_1__CU_INDEX_MASK 0xf000
+#define RLC_GPM_PERF_COUNT_1__CU_INDEX__SHIFT 0xc
+#define RLC_GPM_PERF_COUNT_1__EVENT_SEL_MASK 0x30000
+#define RLC_GPM_PERF_COUNT_1__EVENT_SEL__SHIFT 0x10
+#define RLC_GPM_PERF_COUNT_1__UNUSED_MASK 0xc0000
+#define RLC_GPM_PERF_COUNT_1__UNUSED__SHIFT 0x12
+#define RLC_GPM_PERF_COUNT_1__ENABLE_MASK 0x100000
+#define RLC_GPM_PERF_COUNT_1__ENABLE__SHIFT 0x14
+#define RLC_GPM_PERF_COUNT_1__RESERVED_MASK 0xffe00000
+#define RLC_GPM_PERF_COUNT_1__RESERVED__SHIFT 0x15
+#define RLC_GPR_REG1__DATA_MASK 0xffffffff
+#define RLC_GPR_REG1__DATA__SHIFT 0x0
+#define RLC_GPR_REG2__DATA_MASK 0xffffffff
+#define RLC_GPR_REG2__DATA__SHIFT 0x0
+#define RLC_MGCG_CTRL__MGCG_EN_MASK 0x1
+#define RLC_MGCG_CTRL__MGCG_EN__SHIFT 0x0
+#define RLC_MGCG_CTRL__SILICON_EN_MASK 0x2
+#define RLC_MGCG_CTRL__SILICON_EN__SHIFT 0x1
+#define RLC_MGCG_CTRL__SIMULATION_EN_MASK 0x4
+#define RLC_MGCG_CTRL__SIMULATION_EN__SHIFT 0x2
+#define RLC_MGCG_CTRL__ON_DELAY_MASK 0x78
+#define RLC_MGCG_CTRL__ON_DELAY__SHIFT 0x3
+#define RLC_MGCG_CTRL__OFF_HYSTERESIS_MASK 0x7f80
+#define RLC_MGCG_CTRL__OFF_HYSTERESIS__SHIFT 0x7
+#define RLC_MGCG_CTRL__GC_CAC_MGCG_CLK_CNTL_MASK 0x8000
+#define RLC_MGCG_CTRL__GC_CAC_MGCG_CLK_CNTL__SHIFT 0xf
+#define RLC_MGCG_CTRL__SE_CAC_MGCG_CLK_CNTL_MASK 0x10000
+#define RLC_MGCG_CTRL__SE_CAC_MGCG_CLK_CNTL__SHIFT 0x10
+#define RLC_MGCG_CTRL__SPARE_MASK 0xfffe0000
+#define RLC_MGCG_CTRL__SPARE__SHIFT 0x11
+#define RLC_GPM_THREAD_RESET__THREAD0_RESET_MASK 0x1
+#define RLC_GPM_THREAD_RESET__THREAD0_RESET__SHIFT 0x0
+#define RLC_GPM_THREAD_RESET__THREAD1_RESET_MASK 0x2
+#define RLC_GPM_THREAD_RESET__THREAD1_RESET__SHIFT 0x1
+#define RLC_GPM_THREAD_RESET__THREAD2_RESET_MASK 0x4
+#define RLC_GPM_THREAD_RESET__THREAD2_RESET__SHIFT 0x2
+#define RLC_GPM_THREAD_RESET__THREAD3_RESET_MASK 0x8
+#define RLC_GPM_THREAD_RESET__THREAD3_RESET__SHIFT 0x3
+#define RLC_GPM_THREAD_RESET__RESERVED_MASK 0xfffffff0
+#define RLC_GPM_THREAD_RESET__RESERVED__SHIFT 0x4
+#define RLC_SPM_VMID__RLC_SPM_VMID_MASK 0xf
+#define RLC_SPM_VMID__RLC_SPM_VMID__SHIFT 0x0
+#define RLC_SPM_VMID__RESERVED_MASK 0xfffffff0
+#define RLC_SPM_VMID__RESERVED__SHIFT 0x4
+#define RLC_SPM_INT_CNTL__RLC_SPM_INT_CNTL_MASK 0x1
+#define RLC_SPM_INT_CNTL__RLC_SPM_INT_CNTL__SHIFT 0x0
+#define RLC_SPM_INT_CNTL__RESERVED_MASK 0xfffffffe
+#define RLC_SPM_INT_CNTL__RESERVED__SHIFT 0x1
+#define RLC_SPM_INT_STATUS__RLC_SPM_INT_STATUS_MASK 0x1
+#define RLC_SPM_INT_STATUS__RLC_SPM_INT_STATUS__SHIFT 0x0
+#define RLC_SPM_INT_STATUS__RESERVED_MASK 0xfffffffe
+#define RLC_SPM_INT_STATUS__RESERVED__SHIFT 0x1
+#define RLC_SPM_DEBUG_SELECT__SELECT_MASK 0xff
+#define RLC_SPM_DEBUG_SELECT__SELECT__SHIFT 0x0
+#define RLC_SPM_DEBUG_SELECT__RESERVED_MASK 0x7f00
+#define RLC_SPM_DEBUG_SELECT__RESERVED__SHIFT 0x8
+#define RLC_SPM_DEBUG_SELECT__RLC_SPM_DEBUG_MODE_MASK 0x8000
+#define RLC_SPM_DEBUG_SELECT__RLC_SPM_DEBUG_MODE__SHIFT 0xf
+#define RLC_SPM_DEBUG_SELECT__RLC_SPM_NUM_SAMPLE_MASK 0xffff0000
+#define RLC_SPM_DEBUG_SELECT__RLC_SPM_NUM_SAMPLE__SHIFT 0x10
+#define RLC_SPM_DEBUG__DATA_MASK 0xffffffff
+#define RLC_SPM_DEBUG__DATA__SHIFT 0x0
+#define RLC_SMU_MESSAGE__CMD_MASK 0xffffffff
+#define RLC_SMU_MESSAGE__CMD__SHIFT 0x0
+#define RLC_GPM_LOG_SIZE__SIZE_MASK 0xffffffff
+#define RLC_GPM_LOG_SIZE__SIZE__SHIFT 0x0
+#define RLC_GPM_LOG_CONT__CONT_MASK 0xffffffff
+#define RLC_GPM_LOG_CONT__CONT__SHIFT 0x0
+#define RLC_PG_DELAY_3__CGCG_ACTIVE_BEFORE_CGPG_MASK 0xff
+#define RLC_PG_DELAY_3__CGCG_ACTIVE_BEFORE_CGPG__SHIFT 0x0
+#define RLC_PG_DELAY_3__RESERVED_MASK 0xffffff00
+#define RLC_PG_DELAY_3__RESERVED__SHIFT 0x8
+#define RLC_GPM_INT_DISABLE_TH0__DISABLE_MASK 0xffffffff
+#define RLC_GPM_INT_DISABLE_TH0__DISABLE__SHIFT 0x0
+#define RLC_GPM_INT_DISABLE_TH1__DISABLE_MASK 0xffffffff
+#define RLC_GPM_INT_DISABLE_TH1__DISABLE__SHIFT 0x0
+#define RLC_GPM_INT_FORCE_TH0__FORCE_MASK 0xffffffff
+#define RLC_GPM_INT_FORCE_TH0__FORCE__SHIFT 0x0
+#define RLC_GPM_INT_FORCE_TH1__FORCE_MASK 0xffffffff
+#define RLC_GPM_INT_FORCE_TH1__FORCE__SHIFT 0x0
+#define RLC_SRM_CNTL__SRM_ENABLE_MASK 0x1
+#define RLC_SRM_CNTL__SRM_ENABLE__SHIFT 0x0
+#define RLC_SRM_CNTL__AUTO_INCR_ADDR_MASK 0x2
+#define RLC_SRM_CNTL__AUTO_INCR_ADDR__SHIFT 0x1
+#define RLC_SRM_CNTL__RESERVED_MASK 0xfffffffc
+#define RLC_SRM_CNTL__RESERVED__SHIFT 0x2
+#define RLC_SRM_DEBUG_SELECT__SELECT_MASK 0xff
+#define RLC_SRM_DEBUG_SELECT__SELECT__SHIFT 0x0
+#define RLC_SRM_DEBUG_SELECT__RESERVED_MASK 0xffffff00
+#define RLC_SRM_DEBUG_SELECT__RESERVED__SHIFT 0x8
+#define RLC_SRM_DEBUG__DATA_MASK 0xffffffff
+#define RLC_SRM_DEBUG__DATA__SHIFT 0x0
+#define RLC_SRM_ARAM_ADDR__ADDR_MASK 0x3ff
+#define RLC_SRM_ARAM_ADDR__ADDR__SHIFT 0x0
+#define RLC_SRM_ARAM_ADDR__RESERVED_MASK 0xfffffc00
+#define RLC_SRM_ARAM_ADDR__RESERVED__SHIFT 0xa
+#define RLC_SRM_ARAM_DATA__DATA_MASK 0xffffffff
+#define RLC_SRM_ARAM_DATA__DATA__SHIFT 0x0
+#define RLC_SRM_DRAM_ADDR__ADDR_MASK 0x3ff
+#define RLC_SRM_DRAM_ADDR__ADDR__SHIFT 0x0
+#define RLC_SRM_DRAM_ADDR__RESERVED_MASK 0xfffffc00
+#define RLC_SRM_DRAM_ADDR__RESERVED__SHIFT 0xa
+#define RLC_SRM_DRAM_DATA__DATA_MASK 0xffffffff
+#define RLC_SRM_DRAM_DATA__DATA__SHIFT 0x0
+#define RLC_SRM_GPM_COMMAND__OP_MASK 0x1
+#define RLC_SRM_GPM_COMMAND__OP__SHIFT 0x0
+#define RLC_SRM_GPM_COMMAND__INDEX_CNTL_MASK 0x2
+#define RLC_SRM_GPM_COMMAND__INDEX_CNTL__SHIFT 0x1
+#define RLC_SRM_GPM_COMMAND__INDEX_CNTL_NUM_MASK 0x1c
+#define RLC_SRM_GPM_COMMAND__INDEX_CNTL_NUM__SHIFT 0x2
+#define RLC_SRM_GPM_COMMAND__SIZE_MASK 0x1ffe0
+#define RLC_SRM_GPM_COMMAND__SIZE__SHIFT 0x5
+#define RLC_SRM_GPM_COMMAND__START_OFFSET_MASK 0x1ffe0000
+#define RLC_SRM_GPM_COMMAND__START_OFFSET__SHIFT 0x11
+#define RLC_SRM_GPM_COMMAND__RESERVED1_MASK 0x60000000
+#define RLC_SRM_GPM_COMMAND__RESERVED1__SHIFT 0x1d
+#define RLC_SRM_GPM_COMMAND__DEST_MEMORY_MASK 0x80000000
+#define RLC_SRM_GPM_COMMAND__DEST_MEMORY__SHIFT 0x1f
+#define RLC_SRM_GPM_COMMAND_STATUS__FIFO_EMPTY_MASK 0x1
+#define RLC_SRM_GPM_COMMAND_STATUS__FIFO_EMPTY__SHIFT 0x0
+#define RLC_SRM_GPM_COMMAND_STATUS__FIFO_FULL_MASK 0x2
+#define RLC_SRM_GPM_COMMAND_STATUS__FIFO_FULL__SHIFT 0x1
+#define RLC_SRM_GPM_COMMAND_STATUS__RESERVED_MASK 0xfffffffc
+#define RLC_SRM_GPM_COMMAND_STATUS__RESERVED__SHIFT 0x2
+#define RLC_SRM_RLCV_COMMAND__OP_MASK 0x1
+#define RLC_SRM_RLCV_COMMAND__OP__SHIFT 0x0
+#define RLC_SRM_RLCV_COMMAND__RESERVED_MASK 0xe
+#define RLC_SRM_RLCV_COMMAND__RESERVED__SHIFT 0x1
+#define RLC_SRM_RLCV_COMMAND__SIZE_MASK 0xfff0
+#define RLC_SRM_RLCV_COMMAND__SIZE__SHIFT 0x4
+#define RLC_SRM_RLCV_COMMAND__START_OFFSET_MASK 0xfff0000
+#define RLC_SRM_RLCV_COMMAND__START_OFFSET__SHIFT 0x10
+#define RLC_SRM_RLCV_COMMAND__RESERVED1_MASK 0x70000000
+#define RLC_SRM_RLCV_COMMAND__RESERVED1__SHIFT 0x1c
+#define RLC_SRM_RLCV_COMMAND__DEST_MEMORY_MASK 0x80000000
+#define RLC_SRM_RLCV_COMMAND__DEST_MEMORY__SHIFT 0x1f
+#define RLC_SRM_RLCV_COMMAND_STATUS__FIFO_EMPTY_MASK 0x1
+#define RLC_SRM_RLCV_COMMAND_STATUS__FIFO_EMPTY__SHIFT 0x0
+#define RLC_SRM_RLCV_COMMAND_STATUS__FIFO_FULL_MASK 0x2
+#define RLC_SRM_RLCV_COMMAND_STATUS__FIFO_FULL__SHIFT 0x1
+#define RLC_SRM_RLCV_COMMAND_STATUS__RESERVED_MASK 0xfffffffc
+#define RLC_SRM_RLCV_COMMAND_STATUS__RESERVED__SHIFT 0x2
+#define RLC_SRM_INDEX_CNTL_ADDR_0__ADDRESS_MASK 0xffff
+#define RLC_SRM_INDEX_CNTL_ADDR_0__ADDRESS__SHIFT 0x0
+#define RLC_SRM_INDEX_CNTL_ADDR_0__RESERVED_MASK 0xffff0000
+#define RLC_SRM_INDEX_CNTL_ADDR_0__RESERVED__SHIFT 0x10
+#define RLC_SRM_INDEX_CNTL_ADDR_1__ADDRESS_MASK 0xffff
+#define RLC_SRM_INDEX_CNTL_ADDR_1__ADDRESS__SHIFT 0x0
+#define RLC_SRM_INDEX_CNTL_ADDR_1__RESERVED_MASK 0xffff0000
+#define RLC_SRM_INDEX_CNTL_ADDR_1__RESERVED__SHIFT 0x10
+#define RLC_SRM_INDEX_CNTL_ADDR_2__ADDRESS_MASK 0xffff
+#define RLC_SRM_INDEX_CNTL_ADDR_2__ADDRESS__SHIFT 0x0
+#define RLC_SRM_INDEX_CNTL_ADDR_2__RESERVED_MASK 0xffff0000
+#define RLC_SRM_INDEX_CNTL_ADDR_2__RESERVED__SHIFT 0x10
+#define RLC_SRM_INDEX_CNTL_ADDR_3__ADDRESS_MASK 0xffff
+#define RLC_SRM_INDEX_CNTL_ADDR_3__ADDRESS__SHIFT 0x0
+#define RLC_SRM_INDEX_CNTL_ADDR_3__RESERVED_MASK 0xffff0000
+#define RLC_SRM_INDEX_CNTL_ADDR_3__RESERVED__SHIFT 0x10
+#define RLC_SRM_INDEX_CNTL_ADDR_4__ADDRESS_MASK 0xffff
+#define RLC_SRM_INDEX_CNTL_ADDR_4__ADDRESS__SHIFT 0x0
+#define RLC_SRM_INDEX_CNTL_ADDR_4__RESERVED_MASK 0xffff0000
+#define RLC_SRM_INDEX_CNTL_ADDR_4__RESERVED__SHIFT 0x10
+#define RLC_SRM_INDEX_CNTL_ADDR_5__ADDRESS_MASK 0xffff
+#define RLC_SRM_INDEX_CNTL_ADDR_5__ADDRESS__SHIFT 0x0
+#define RLC_SRM_INDEX_CNTL_ADDR_5__RESERVED_MASK 0xffff0000
+#define RLC_SRM_INDEX_CNTL_ADDR_5__RESERVED__SHIFT 0x10
+#define RLC_SRM_INDEX_CNTL_ADDR_6__ADDRESS_MASK 0xffff
+#define RLC_SRM_INDEX_CNTL_ADDR_6__ADDRESS__SHIFT 0x0
+#define RLC_SRM_INDEX_CNTL_ADDR_6__RESERVED_MASK 0xffff0000
+#define RLC_SRM_INDEX_CNTL_ADDR_6__RESERVED__SHIFT 0x10
+#define RLC_SRM_INDEX_CNTL_ADDR_7__ADDRESS_MASK 0xffff
+#define RLC_SRM_INDEX_CNTL_ADDR_7__ADDRESS__SHIFT 0x0
+#define RLC_SRM_INDEX_CNTL_ADDR_7__RESERVED_MASK 0xffff0000
+#define RLC_SRM_INDEX_CNTL_ADDR_7__RESERVED__SHIFT 0x10
+#define RLC_SRM_INDEX_CNTL_DATA_0__DATA_MASK 0xffffffff
+#define RLC_SRM_INDEX_CNTL_DATA_0__DATA__SHIFT 0x0
+#define RLC_SRM_INDEX_CNTL_DATA_1__DATA_MASK 0xffffffff
+#define RLC_SRM_INDEX_CNTL_DATA_1__DATA__SHIFT 0x0
+#define RLC_SRM_INDEX_CNTL_DATA_2__DATA_MASK 0xffffffff
+#define RLC_SRM_INDEX_CNTL_DATA_2__DATA__SHIFT 0x0
+#define RLC_SRM_INDEX_CNTL_DATA_3__DATA_MASK 0xffffffff
+#define RLC_SRM_INDEX_CNTL_DATA_3__DATA__SHIFT 0x0
+#define RLC_SRM_INDEX_CNTL_DATA_4__DATA_MASK 0xffffffff
+#define RLC_SRM_INDEX_CNTL_DATA_4__DATA__SHIFT 0x0
+#define RLC_SRM_INDEX_CNTL_DATA_5__DATA_MASK 0xffffffff
+#define RLC_SRM_INDEX_CNTL_DATA_5__DATA__SHIFT 0x0
+#define RLC_SRM_INDEX_CNTL_DATA_6__DATA_MASK 0xffffffff
+#define RLC_SRM_INDEX_CNTL_DATA_6__DATA__SHIFT 0x0
+#define RLC_SRM_INDEX_CNTL_DATA_7__DATA_MASK 0xffffffff
+#define RLC_SRM_INDEX_CNTL_DATA_7__DATA__SHIFT 0x0
+#define RLC_SRM_STAT__SRM_STATUS_MASK 0x1
+#define RLC_SRM_STAT__SRM_STATUS__SHIFT 0x0
+#define RLC_SRM_STAT__RESERVED_MASK 0xfffffffe
+#define RLC_SRM_STAT__RESERVED__SHIFT 0x1
+#define RLC_SRM_GPM_ABORT__ABORT_MASK 0x1
+#define RLC_SRM_GPM_ABORT__ABORT__SHIFT 0x0
+#define RLC_SRM_GPM_ABORT__RESERVED_MASK 0xfffffffe
+#define RLC_SRM_GPM_ABORT__RESERVED__SHIFT 0x1
+#define RLC_CSIB_ADDR_LO__ADDRESS_MASK 0xffffffff
+#define RLC_CSIB_ADDR_LO__ADDRESS__SHIFT 0x0
+#define RLC_CSIB_ADDR_HI__ADDRESS_MASK 0xffff
+#define RLC_CSIB_ADDR_HI__ADDRESS__SHIFT 0x0
+#define RLC_CSIB_LENGTH__LENGTH_MASK 0xffffffff
+#define RLC_CSIB_LENGTH__LENGTH__SHIFT 0x0
+#define RLC_CP_RESPONSE0__RESPONSE_MASK 0xffffffff
+#define RLC_CP_RESPONSE0__RESPONSE__SHIFT 0x0
+#define RLC_CP_RESPONSE1__RESPONSE_MASK 0xffffffff
+#define RLC_CP_RESPONSE1__RESPONSE__SHIFT 0x0
+#define RLC_CP_RESPONSE2__RESPONSE_MASK 0xffffffff
+#define RLC_CP_RESPONSE2__RESPONSE__SHIFT 0x0
+#define RLC_CP_RESPONSE3__RESPONSE_MASK 0xffffffff
+#define RLC_CP_RESPONSE3__RESPONSE__SHIFT 0x0
+#define RLC_SMU_COMMAND__CMD_MASK 0xffffffff
+#define RLC_SMU_COMMAND__CMD__SHIFT 0x0
+#define RLC_CP_SCHEDULERS__scheduler0_MASK 0xff
+#define RLC_CP_SCHEDULERS__scheduler0__SHIFT 0x0
+#define RLC_CP_SCHEDULERS__scheduler1_MASK 0xff00
+#define RLC_CP_SCHEDULERS__scheduler1__SHIFT 0x8
+#define RLC_CP_SCHEDULERS__scheduler2_MASK 0xff0000
+#define RLC_CP_SCHEDULERS__scheduler2__SHIFT 0x10
+#define RLC_CP_SCHEDULERS__scheduler3_MASK 0xff000000
+#define RLC_CP_SCHEDULERS__scheduler3__SHIFT 0x18
+#define RLC_SMU_ARGUMENT_1__ARG_MASK 0xffffffff
+#define RLC_SMU_ARGUMENT_1__ARG__SHIFT 0x0
+#define RLC_SMU_ARGUMENT_2__ARG_MASK 0xffffffff
+#define RLC_SMU_ARGUMENT_2__ARG__SHIFT 0x0
+#define RLC_GPM_GENERAL_8__DATA_MASK 0xffffffff
+#define RLC_GPM_GENERAL_8__DATA__SHIFT 0x0
+#define RLC_GPM_GENERAL_9__DATA_MASK 0xffffffff
+#define RLC_GPM_GENERAL_9__DATA__SHIFT 0x0
+#define RLC_GPM_GENERAL_10__DATA_MASK 0xffffffff
+#define RLC_GPM_GENERAL_10__DATA__SHIFT 0x0
+#define RLC_GPM_GENERAL_11__DATA_MASK 0xffffffff
+#define RLC_GPM_GENERAL_11__DATA__SHIFT 0x0
+#define RLC_GPM_GENERAL_12__DATA_MASK 0xffffffff
+#define RLC_GPM_GENERAL_12__DATA__SHIFT 0x0
+#define RLC_SPM_PERFMON_CNTL__RESERVED1_MASK 0xfff
+#define RLC_SPM_PERFMON_CNTL__RESERVED1__SHIFT 0x0
+#define RLC_SPM_PERFMON_CNTL__PERFMON_RING_MODE_MASK 0x3000
+#define RLC_SPM_PERFMON_CNTL__PERFMON_RING_MODE__SHIFT 0xc
+#define RLC_SPM_PERFMON_CNTL__RESERVED_MASK 0xc000
+#define RLC_SPM_PERFMON_CNTL__RESERVED__SHIFT 0xe
+#define RLC_SPM_PERFMON_CNTL__PERFMON_SAMPLE_INTERVAL_MASK 0xffff0000
+#define RLC_SPM_PERFMON_CNTL__PERFMON_SAMPLE_INTERVAL__SHIFT 0x10
+#define RLC_SPM_PERFMON_RING_BASE_LO__RING_BASE_LO_MASK 0xffffffff
+#define RLC_SPM_PERFMON_RING_BASE_LO__RING_BASE_LO__SHIFT 0x0
+#define RLC_SPM_PERFMON_RING_BASE_HI__RING_BASE_HI_MASK 0xffff
+#define RLC_SPM_PERFMON_RING_BASE_HI__RING_BASE_HI__SHIFT 0x0
+#define RLC_SPM_PERFMON_RING_BASE_HI__RESERVED_MASK 0xffff0000
+#define RLC_SPM_PERFMON_RING_BASE_HI__RESERVED__SHIFT 0x10
+#define RLC_SPM_PERFMON_RING_SIZE__RING_BASE_SIZE_MASK 0xffffffff
+#define RLC_SPM_PERFMON_RING_SIZE__RING_BASE_SIZE__SHIFT 0x0
+#define RLC_SPM_PERFMON_SEGMENT_SIZE__PERFMON_SEGMENT_SIZE_MASK 0xff
+#define RLC_SPM_PERFMON_SEGMENT_SIZE__PERFMON_SEGMENT_SIZE__SHIFT 0x0
+#define RLC_SPM_PERFMON_SEGMENT_SIZE__RESERVED1_MASK 0x700
+#define RLC_SPM_PERFMON_SEGMENT_SIZE__RESERVED1__SHIFT 0x8
+#define RLC_SPM_PERFMON_SEGMENT_SIZE__GLOBAL_NUM_LINE_MASK 0xf800
+#define RLC_SPM_PERFMON_SEGMENT_SIZE__GLOBAL_NUM_LINE__SHIFT 0xb
+#define RLC_SPM_PERFMON_SEGMENT_SIZE__SE0_NUM_LINE_MASK 0x1f0000
+#define RLC_SPM_PERFMON_SEGMENT_SIZE__SE0_NUM_LINE__SHIFT 0x10
+#define RLC_SPM_PERFMON_SEGMENT_SIZE__SE1_NUM_LINE_MASK 0x3e00000
+#define RLC_SPM_PERFMON_SEGMENT_SIZE__SE1_NUM_LINE__SHIFT 0x15
+#define RLC_SPM_PERFMON_SEGMENT_SIZE__SE2_NUM_LINE_MASK 0x7c000000
+#define RLC_SPM_PERFMON_SEGMENT_SIZE__SE2_NUM_LINE__SHIFT 0x1a
+#define RLC_SPM_PERFMON_SEGMENT_SIZE__RESERVED_MASK 0x80000000
+#define RLC_SPM_PERFMON_SEGMENT_SIZE__RESERVED__SHIFT 0x1f
+#define RLC_SPM_SE_MUXSEL_ADDR__PERFMON_SEL_ADDR_MASK 0xffffffff
+#define RLC_SPM_SE_MUXSEL_ADDR__PERFMON_SEL_ADDR__SHIFT 0x0
+#define RLC_SPM_SE_MUXSEL_DATA__PERFMON_SEL_DATA_MASK 0xffffffff
+#define RLC_SPM_SE_MUXSEL_DATA__PERFMON_SEL_DATA__SHIFT 0x0
+#define RLC_SPM_CPG_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY_MASK 0xff
+#define RLC_SPM_CPG_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY__SHIFT 0x0
+#define RLC_SPM_CPG_PERFMON_SAMPLE_DELAY__RESERVED_MASK 0xffffff00
+#define RLC_SPM_CPG_PERFMON_SAMPLE_DELAY__RESERVED__SHIFT 0x8
+#define RLC_SPM_CPC_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY_MASK 0xff
+#define RLC_SPM_CPC_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY__SHIFT 0x0
+#define RLC_SPM_CPC_PERFMON_SAMPLE_DELAY__RESERVED_MASK 0xffffff00
+#define RLC_SPM_CPC_PERFMON_SAMPLE_DELAY__RESERVED__SHIFT 0x8
+#define RLC_SPM_CPF_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY_MASK 0xff
+#define RLC_SPM_CPF_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY__SHIFT 0x0
+#define RLC_SPM_CPF_PERFMON_SAMPLE_DELAY__RESERVED_MASK 0xffffff00
+#define RLC_SPM_CPF_PERFMON_SAMPLE_DELAY__RESERVED__SHIFT 0x8
+#define RLC_SPM_CB_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY_MASK 0xff
+#define RLC_SPM_CB_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY__SHIFT 0x0
+#define RLC_SPM_CB_PERFMON_SAMPLE_DELAY__RESERVED_MASK 0xffffff00
+#define RLC_SPM_CB_PERFMON_SAMPLE_DELAY__RESERVED__SHIFT 0x8
+#define RLC_SPM_DB_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY_MASK 0xff
+#define RLC_SPM_DB_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY__SHIFT 0x0
+#define RLC_SPM_DB_PERFMON_SAMPLE_DELAY__RESERVED_MASK 0xffffff00
+#define RLC_SPM_DB_PERFMON_SAMPLE_DELAY__RESERVED__SHIFT 0x8
+#define RLC_SPM_PA_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY_MASK 0xff
+#define RLC_SPM_PA_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY__SHIFT 0x0
+#define RLC_SPM_PA_PERFMON_SAMPLE_DELAY__RESERVED_MASK 0xffffff00
+#define RLC_SPM_PA_PERFMON_SAMPLE_DELAY__RESERVED__SHIFT 0x8
+#define RLC_SPM_GDS_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY_MASK 0xff
+#define RLC_SPM_GDS_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY__SHIFT 0x0
+#define RLC_SPM_GDS_PERFMON_SAMPLE_DELAY__RESERVED_MASK 0xffffff00
+#define RLC_SPM_GDS_PERFMON_SAMPLE_DELAY__RESERVED__SHIFT 0x8
+#define RLC_SPM_IA_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY_MASK 0xff
+#define RLC_SPM_IA_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY__SHIFT 0x0
+#define RLC_SPM_IA_PERFMON_SAMPLE_DELAY__RESERVED_MASK 0xffffff00
+#define RLC_SPM_IA_PERFMON_SAMPLE_DELAY__RESERVED__SHIFT 0x8
+#define RLC_SPM_SC_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY_MASK 0xff
+#define RLC_SPM_SC_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY__SHIFT 0x0
+#define RLC_SPM_SC_PERFMON_SAMPLE_DELAY__RESERVED_MASK 0xffffff00
+#define RLC_SPM_SC_PERFMON_SAMPLE_DELAY__RESERVED__SHIFT 0x8
+#define RLC_SPM_TCC_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY_MASK 0xff
+#define RLC_SPM_TCC_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY__SHIFT 0x0
+#define RLC_SPM_TCC_PERFMON_SAMPLE_DELAY__RESERVED_MASK 0xffffff00
+#define RLC_SPM_TCC_PERFMON_SAMPLE_DELAY__RESERVED__SHIFT 0x8
+#define RLC_SPM_TCA_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY_MASK 0xff
+#define RLC_SPM_TCA_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY__SHIFT 0x0
+#define RLC_SPM_TCA_PERFMON_SAMPLE_DELAY__RESERVED_MASK 0xffffff00
+#define RLC_SPM_TCA_PERFMON_SAMPLE_DELAY__RESERVED__SHIFT 0x8
+#define RLC_SPM_TCP_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY_MASK 0xff
+#define RLC_SPM_TCP_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY__SHIFT 0x0
+#define RLC_SPM_TCP_PERFMON_SAMPLE_DELAY__RESERVED_MASK 0xffffff00
+#define RLC_SPM_TCP_PERFMON_SAMPLE_DELAY__RESERVED__SHIFT 0x8
+#define RLC_SPM_TA_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY_MASK 0xff
+#define RLC_SPM_TA_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY__SHIFT 0x0
+#define RLC_SPM_TA_PERFMON_SAMPLE_DELAY__RESERVED_MASK 0xffffff00
+#define RLC_SPM_TA_PERFMON_SAMPLE_DELAY__RESERVED__SHIFT 0x8
+#define RLC_SPM_TD_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY_MASK 0xff
+#define RLC_SPM_TD_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY__SHIFT 0x0
+#define RLC_SPM_TD_PERFMON_SAMPLE_DELAY__RESERVED_MASK 0xffffff00
+#define RLC_SPM_TD_PERFMON_SAMPLE_DELAY__RESERVED__SHIFT 0x8
+#define RLC_SPM_VGT_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY_MASK 0xff
+#define RLC_SPM_VGT_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY__SHIFT 0x0
+#define RLC_SPM_VGT_PERFMON_SAMPLE_DELAY__RESERVED_MASK 0xffffff00
+#define RLC_SPM_VGT_PERFMON_SAMPLE_DELAY__RESERVED__SHIFT 0x8
+#define RLC_SPM_SPI_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY_MASK 0xff
+#define RLC_SPM_SPI_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY__SHIFT 0x0
+#define RLC_SPM_SPI_PERFMON_SAMPLE_DELAY__RESERVED_MASK 0xffffff00
+#define RLC_SPM_SPI_PERFMON_SAMPLE_DELAY__RESERVED__SHIFT 0x8
+#define RLC_SPM_SQG_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY_MASK 0xff
+#define RLC_SPM_SQG_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY__SHIFT 0x0
+#define RLC_SPM_SQG_PERFMON_SAMPLE_DELAY__RESERVED_MASK 0xffffff00
+#define RLC_SPM_SQG_PERFMON_SAMPLE_DELAY__RESERVED__SHIFT 0x8
+#define RLC_SPM_SX_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY_MASK 0xff
+#define RLC_SPM_SX_PERFMON_SAMPLE_DELAY__PERFMON_SAMPLE_DELAY__SHIFT 0x0
+#define RLC_SPM_SX_PERFMON_SAMPLE_DELAY__RESERVED_MASK 0xffffff00
+#define RLC_SPM_SX_PERFMON_SAMPLE_DELAY__RESERVED__SHIFT 0x8
+#define RLC_SPM_GLOBAL_MUXSEL_ADDR__PERFMON_SEL_ADDR_MASK 0xffffffff
+#define RLC_SPM_GLOBAL_MUXSEL_ADDR__PERFMON_SEL_ADDR__SHIFT 0x0
+#define RLC_SPM_GLOBAL_MUXSEL_DATA__PERFMON_SEL_DATA_MASK 0xffffffff
+#define RLC_SPM_GLOBAL_MUXSEL_DATA__PERFMON_SEL_DATA__SHIFT 0x0
+#define RLC_SPM_RING_RDPTR__PERFMON_RING_RDPTR_MASK 0xffffffff
+#define RLC_SPM_RING_RDPTR__PERFMON_RING_RDPTR__SHIFT 0x0
+#define RLC_SPM_SEGMENT_THRESHOLD__NUM_SEGMENT_THRESHOLD_MASK 0xffffffff
+#define RLC_SPM_SEGMENT_THRESHOLD__NUM_SEGMENT_THRESHOLD__SHIFT 0x0
+#define RLC_GPU_IOV_VF_ENABLE__VF_ENABLE_MASK 0x1
+#define RLC_GPU_IOV_VF_ENABLE__VF_ENABLE__SHIFT 0x0
+#define RLC_GPU_IOV_VF_ENABLE__RESERVED_MASK 0xfffe
+#define RLC_GPU_IOV_VF_ENABLE__RESERVED__SHIFT 0x1
+#define RLC_GPU_IOV_VF_ENABLE__VF_NUM_MASK 0xffff0000
+#define RLC_GPU_IOV_VF_ENABLE__VF_NUM__SHIFT 0x10
+#define RLC_GPU_IOV_RLC_RESPONSE__RESP_MASK 0xffffffff
+#define RLC_GPU_IOV_RLC_RESPONSE__RESP__SHIFT 0x0
+#define RLC_GPU_IOV_ACTIVE_FCN_ID__VF_ID_MASK 0xf
+#define RLC_GPU_IOV_ACTIVE_FCN_ID__VF_ID__SHIFT 0x0
+#define RLC_GPU_IOV_ACTIVE_FCN_ID__RESERVED_MASK 0x7ffffff0
+#define RLC_GPU_IOV_ACTIVE_FCN_ID__RESERVED__SHIFT 0x4
+#define RLC_GPU_IOV_ACTIVE_FCN_ID__PF_VF_MASK 0x80000000
+#define RLC_GPU_IOV_ACTIVE_FCN_ID__PF_VF__SHIFT 0x1f
+#define SPI_PS_INPUT_CNTL_0__OFFSET_MASK 0x3f
+#define SPI_PS_INPUT_CNTL_0__OFFSET__SHIFT 0x0
+#define SPI_PS_INPUT_CNTL_0__DEFAULT_VAL_MASK 0x300
+#define SPI_PS_INPUT_CNTL_0__DEFAULT_VAL__SHIFT 0x8
+#define SPI_PS_INPUT_CNTL_0__FLAT_SHADE_MASK 0x400
+#define SPI_PS_INPUT_CNTL_0__FLAT_SHADE__SHIFT 0xa
+#define SPI_PS_INPUT_CNTL_0__CYL_WRAP_MASK 0x1e000
+#define SPI_PS_INPUT_CNTL_0__CYL_WRAP__SHIFT 0xd
+#define SPI_PS_INPUT_CNTL_0__PT_SPRITE_TEX_MASK 0x20000
+#define SPI_PS_INPUT_CNTL_0__PT_SPRITE_TEX__SHIFT 0x11
+#define SPI_PS_INPUT_CNTL_0__DUP_MASK 0x40000
+#define SPI_PS_INPUT_CNTL_0__DUP__SHIFT 0x12
+#define SPI_PS_INPUT_CNTL_0__FP16_INTERP_MODE_MASK 0x80000
+#define SPI_PS_INPUT_CNTL_0__FP16_INTERP_MODE__SHIFT 0x13
+#define SPI_PS_INPUT_CNTL_0__USE_DEFAULT_ATTR1_MASK 0x100000
+#define SPI_PS_INPUT_CNTL_0__USE_DEFAULT_ATTR1__SHIFT 0x14
+#define SPI_PS_INPUT_CNTL_0__DEFAULT_VAL_ATTR1_MASK 0x600000
+#define SPI_PS_INPUT_CNTL_0__DEFAULT_VAL_ATTR1__SHIFT 0x15
+#define SPI_PS_INPUT_CNTL_0__PT_SPRITE_TEX_ATTR1_MASK 0x800000
+#define SPI_PS_INPUT_CNTL_0__PT_SPRITE_TEX_ATTR1__SHIFT 0x17
+#define SPI_PS_INPUT_CNTL_0__ATTR0_VALID_MASK 0x1000000
+#define SPI_PS_INPUT_CNTL_0__ATTR0_VALID__SHIFT 0x18
+#define SPI_PS_INPUT_CNTL_0__ATTR1_VALID_MASK 0x2000000
+#define SPI_PS_INPUT_CNTL_0__ATTR1_VALID__SHIFT 0x19
+#define SPI_PS_INPUT_CNTL_1__OFFSET_MASK 0x3f
+#define SPI_PS_INPUT_CNTL_1__OFFSET__SHIFT 0x0
+#define SPI_PS_INPUT_CNTL_1__DEFAULT_VAL_MASK 0x300
+#define SPI_PS_INPUT_CNTL_1__DEFAULT_VAL__SHIFT 0x8
+#define SPI_PS_INPUT_CNTL_1__FLAT_SHADE_MASK 0x400
+#define SPI_PS_INPUT_CNTL_1__FLAT_SHADE__SHIFT 0xa
+#define SPI_PS_INPUT_CNTL_1__CYL_WRAP_MASK 0x1e000
+#define SPI_PS_INPUT_CNTL_1__CYL_WRAP__SHIFT 0xd
+#define SPI_PS_INPUT_CNTL_1__PT_SPRITE_TEX_MASK 0x20000
+#define SPI_PS_INPUT_CNTL_1__PT_SPRITE_TEX__SHIFT 0x11
+#define SPI_PS_INPUT_CNTL_1__DUP_MASK 0x40000
+#define SPI_PS_INPUT_CNTL_1__DUP__SHIFT 0x12
+#define SPI_PS_INPUT_CNTL_1__FP16_INTERP_MODE_MASK 0x80000
+#define SPI_PS_INPUT_CNTL_1__FP16_INTERP_MODE__SHIFT 0x13
+#define SPI_PS_INPUT_CNTL_1__USE_DEFAULT_ATTR1_MASK 0x100000
+#define SPI_PS_INPUT_CNTL_1__USE_DEFAULT_ATTR1__SHIFT 0x14
+#define SPI_PS_INPUT_CNTL_1__DEFAULT_VAL_ATTR1_MASK 0x600000
+#define SPI_PS_INPUT_CNTL_1__DEFAULT_VAL_ATTR1__SHIFT 0x15
+#define SPI_PS_INPUT_CNTL_1__PT_SPRITE_TEX_ATTR1_MASK 0x800000
+#define SPI_PS_INPUT_CNTL_1__PT_SPRITE_TEX_ATTR1__SHIFT 0x17
+#define SPI_PS_INPUT_CNTL_1__ATTR0_VALID_MASK 0x1000000
+#define SPI_PS_INPUT_CNTL_1__ATTR0_VALID__SHIFT 0x18
+#define SPI_PS_INPUT_CNTL_1__ATTR1_VALID_MASK 0x2000000
+#define SPI_PS_INPUT_CNTL_1__ATTR1_VALID__SHIFT 0x19
+#define SPI_PS_INPUT_CNTL_2__OFFSET_MASK 0x3f
+#define SPI_PS_INPUT_CNTL_2__OFFSET__SHIFT 0x0
+#define SPI_PS_INPUT_CNTL_2__DEFAULT_VAL_MASK 0x300
+#define SPI_PS_INPUT_CNTL_2__DEFAULT_VAL__SHIFT 0x8
+#define SPI_PS_INPUT_CNTL_2__FLAT_SHADE_MASK 0x400
+#define SPI_PS_INPUT_CNTL_2__FLAT_SHADE__SHIFT 0xa
+#define SPI_PS_INPUT_CNTL_2__CYL_WRAP_MASK 0x1e000
+#define SPI_PS_INPUT_CNTL_2__CYL_WRAP__SHIFT 0xd
+#define SPI_PS_INPUT_CNTL_2__PT_SPRITE_TEX_MASK 0x20000
+#define SPI_PS_INPUT_CNTL_2__PT_SPRITE_TEX__SHIFT 0x11
+#define SPI_PS_INPUT_CNTL_2__DUP_MASK 0x40000
+#define SPI_PS_INPUT_CNTL_2__DUP__SHIFT 0x12
+#define SPI_PS_INPUT_CNTL_2__FP16_INTERP_MODE_MASK 0x80000
+#define SPI_PS_INPUT_CNTL_2__FP16_INTERP_MODE__SHIFT 0x13
+#define SPI_PS_INPUT_CNTL_2__USE_DEFAULT_ATTR1_MASK 0x100000
+#define SPI_PS_INPUT_CNTL_2__USE_DEFAULT_ATTR1__SHIFT 0x14
+#define SPI_PS_INPUT_CNTL_2__DEFAULT_VAL_ATTR1_MASK 0x600000
+#define SPI_PS_INPUT_CNTL_2__DEFAULT_VAL_ATTR1__SHIFT 0x15
+#define SPI_PS_INPUT_CNTL_2__PT_SPRITE_TEX_ATTR1_MASK 0x800000
+#define SPI_PS_INPUT_CNTL_2__PT_SPRITE_TEX_ATTR1__SHIFT 0x17
+#define SPI_PS_INPUT_CNTL_2__ATTR0_VALID_MASK 0x1000000
+#define SPI_PS_INPUT_CNTL_2__ATTR0_VALID__SHIFT 0x18
+#define SPI_PS_INPUT_CNTL_2__ATTR1_VALID_MASK 0x2000000
+#define SPI_PS_INPUT_CNTL_2__ATTR1_VALID__SHIFT 0x19
+#define SPI_PS_INPUT_CNTL_3__OFFSET_MASK 0x3f
+#define SPI_PS_INPUT_CNTL_3__OFFSET__SHIFT 0x0
+#define SPI_PS_INPUT_CNTL_3__DEFAULT_VAL_MASK 0x300
+#define SPI_PS_INPUT_CNTL_3__DEFAULT_VAL__SHIFT 0x8
+#define SPI_PS_INPUT_CNTL_3__FLAT_SHADE_MASK 0x400
+#define SPI_PS_INPUT_CNTL_3__FLAT_SHADE__SHIFT 0xa
+#define SPI_PS_INPUT_CNTL_3__CYL_WRAP_MASK 0x1e000
+#define SPI_PS_INPUT_CNTL_3__CYL_WRAP__SHIFT 0xd
+#define SPI_PS_INPUT_CNTL_3__PT_SPRITE_TEX_MASK 0x20000
+#define SPI_PS_INPUT_CNTL_3__PT_SPRITE_TEX__SHIFT 0x11
+#define SPI_PS_INPUT_CNTL_3__DUP_MASK 0x40000
+#define SPI_PS_INPUT_CNTL_3__DUP__SHIFT 0x12
+#define SPI_PS_INPUT_CNTL_3__FP16_INTERP_MODE_MASK 0x80000
+#define SPI_PS_INPUT_CNTL_3__FP16_INTERP_MODE__SHIFT 0x13
+#define SPI_PS_INPUT_CNTL_3__USE_DEFAULT_ATTR1_MASK 0x100000
+#define SPI_PS_INPUT_CNTL_3__USE_DEFAULT_ATTR1__SHIFT 0x14
+#define SPI_PS_INPUT_CNTL_3__DEFAULT_VAL_ATTR1_MASK 0x600000
+#define SPI_PS_INPUT_CNTL_3__DEFAULT_VAL_ATTR1__SHIFT 0x15
+#define SPI_PS_INPUT_CNTL_3__PT_SPRITE_TEX_ATTR1_MASK 0x800000
+#define SPI_PS_INPUT_CNTL_3__PT_SPRITE_TEX_ATTR1__SHIFT 0x17
+#define SPI_PS_INPUT_CNTL_3__ATTR0_VALID_MASK 0x1000000
+#define SPI_PS_INPUT_CNTL_3__ATTR0_VALID__SHIFT 0x18
+#define SPI_PS_INPUT_CNTL_3__ATTR1_VALID_MASK 0x2000000
+#define SPI_PS_INPUT_CNTL_3__ATTR1_VALID__SHIFT 0x19
+#define SPI_PS_INPUT_CNTL_4__OFFSET_MASK 0x3f
+#define SPI_PS_INPUT_CNTL_4__OFFSET__SHIFT 0x0
+#define SPI_PS_INPUT_CNTL_4__DEFAULT_VAL_MASK 0x300
+#define SPI_PS_INPUT_CNTL_4__DEFAULT_VAL__SHIFT 0x8
+#define SPI_PS_INPUT_CNTL_4__FLAT_SHADE_MASK 0x400
+#define SPI_PS_INPUT_CNTL_4__FLAT_SHADE__SHIFT 0xa
+#define SPI_PS_INPUT_CNTL_4__CYL_WRAP_MASK 0x1e000
+#define SPI_PS_INPUT_CNTL_4__CYL_WRAP__SHIFT 0xd
+#define SPI_PS_INPUT_CNTL_4__PT_SPRITE_TEX_MASK 0x20000
+#define SPI_PS_INPUT_CNTL_4__PT_SPRITE_TEX__SHIFT 0x11
+#define SPI_PS_INPUT_CNTL_4__DUP_MASK 0x40000
+#define SPI_PS_INPUT_CNTL_4__DUP__SHIFT 0x12
+#define SPI_PS_INPUT_CNTL_4__FP16_INTERP_MODE_MASK 0x80000
+#define SPI_PS_INPUT_CNTL_4__FP16_INTERP_MODE__SHIFT 0x13
+#define SPI_PS_INPUT_CNTL_4__USE_DEFAULT_ATTR1_MASK 0x100000
+#define SPI_PS_INPUT_CNTL_4__USE_DEFAULT_ATTR1__SHIFT 0x14
+#define SPI_PS_INPUT_CNTL_4__DEFAULT_VAL_ATTR1_MASK 0x600000
+#define SPI_PS_INPUT_CNTL_4__DEFAULT_VAL_ATTR1__SHIFT 0x15
+#define SPI_PS_INPUT_CNTL_4__PT_SPRITE_TEX_ATTR1_MASK 0x800000
+#define SPI_PS_INPUT_CNTL_4__PT_SPRITE_TEX_ATTR1__SHIFT 0x17
+#define SPI_PS_INPUT_CNTL_4__ATTR0_VALID_MASK 0x1000000
+#define SPI_PS_INPUT_CNTL_4__ATTR0_VALID__SHIFT 0x18
+#define SPI_PS_INPUT_CNTL_4__ATTR1_VALID_MASK 0x2000000
+#define SPI_PS_INPUT_CNTL_4__ATTR1_VALID__SHIFT 0x19
+#define SPI_PS_INPUT_CNTL_5__OFFSET_MASK 0x3f
+#define SPI_PS_INPUT_CNTL_5__OFFSET__SHIFT 0x0
+#define SPI_PS_INPUT_CNTL_5__DEFAULT_VAL_MASK 0x300
+#define SPI_PS_INPUT_CNTL_5__DEFAULT_VAL__SHIFT 0x8
+#define SPI_PS_INPUT_CNTL_5__FLAT_SHADE_MASK 0x400
+#define SPI_PS_INPUT_CNTL_5__FLAT_SHADE__SHIFT 0xa
+#define SPI_PS_INPUT_CNTL_5__CYL_WRAP_MASK 0x1e000
+#define SPI_PS_INPUT_CNTL_5__CYL_WRAP__SHIFT 0xd
+#define SPI_PS_INPUT_CNTL_5__PT_SPRITE_TEX_MASK 0x20000
+#define SPI_PS_INPUT_CNTL_5__PT_SPRITE_TEX__SHIFT 0x11
+#define SPI_PS_INPUT_CNTL_5__DUP_MASK 0x40000
+#define SPI_PS_INPUT_CNTL_5__DUP__SHIFT 0x12
+#define SPI_PS_INPUT_CNTL_5__FP16_INTERP_MODE_MASK 0x80000
+#define SPI_PS_INPUT_CNTL_5__FP16_INTERP_MODE__SHIFT 0x13
+#define SPI_PS_INPUT_CNTL_5__USE_DEFAULT_ATTR1_MASK 0x100000
+#define SPI_PS_INPUT_CNTL_5__USE_DEFAULT_ATTR1__SHIFT 0x14
+#define SPI_PS_INPUT_CNTL_5__DEFAULT_VAL_ATTR1_MASK 0x600000
+#define SPI_PS_INPUT_CNTL_5__DEFAULT_VAL_ATTR1__SHIFT 0x15
+#define SPI_PS_INPUT_CNTL_5__PT_SPRITE_TEX_ATTR1_MASK 0x800000
+#define SPI_PS_INPUT_CNTL_5__PT_SPRITE_TEX_ATTR1__SHIFT 0x17
+#define SPI_PS_INPUT_CNTL_5__ATTR0_VALID_MASK 0x1000000
+#define SPI_PS_INPUT_CNTL_5__ATTR0_VALID__SHIFT 0x18
+#define SPI_PS_INPUT_CNTL_5__ATTR1_VALID_MASK 0x2000000
+#define SPI_PS_INPUT_CNTL_5__ATTR1_VALID__SHIFT 0x19
+#define SPI_PS_INPUT_CNTL_6__OFFSET_MASK 0x3f
+#define SPI_PS_INPUT_CNTL_6__OFFSET__SHIFT 0x0
+#define SPI_PS_INPUT_CNTL_6__DEFAULT_VAL_MASK 0x300
+#define SPI_PS_INPUT_CNTL_6__DEFAULT_VAL__SHIFT 0x8
+#define SPI_PS_INPUT_CNTL_6__FLAT_SHADE_MASK 0x400
+#define SPI_PS_INPUT_CNTL_6__FLAT_SHADE__SHIFT 0xa
+#define SPI_PS_INPUT_CNTL_6__CYL_WRAP_MASK 0x1e000
+#define SPI_PS_INPUT_CNTL_6__CYL_WRAP__SHIFT 0xd
+#define SPI_PS_INPUT_CNTL_6__PT_SPRITE_TEX_MASK 0x20000
+#define SPI_PS_INPUT_CNTL_6__PT_SPRITE_TEX__SHIFT 0x11
+#define SPI_PS_INPUT_CNTL_6__DUP_MASK 0x40000
+#define SPI_PS_INPUT_CNTL_6__DUP__SHIFT 0x12
+#define SPI_PS_INPUT_CNTL_6__FP16_INTERP_MODE_MASK 0x80000
+#define SPI_PS_INPUT_CNTL_6__FP16_INTERP_MODE__SHIFT 0x13
+#define SPI_PS_INPUT_CNTL_6__USE_DEFAULT_ATTR1_MASK 0x100000
+#define SPI_PS_INPUT_CNTL_6__USE_DEFAULT_ATTR1__SHIFT 0x14
+#define SPI_PS_INPUT_CNTL_6__DEFAULT_VAL_ATTR1_MASK 0x600000
+#define SPI_PS_INPUT_CNTL_6__DEFAULT_VAL_ATTR1__SHIFT 0x15
+#define SPI_PS_INPUT_CNTL_6__PT_SPRITE_TEX_ATTR1_MASK 0x800000
+#define SPI_PS_INPUT_CNTL_6__PT_SPRITE_TEX_ATTR1__SHIFT 0x17
+#define SPI_PS_INPUT_CNTL_6__ATTR0_VALID_MASK 0x1000000
+#define SPI_PS_INPUT_CNTL_6__ATTR0_VALID__SHIFT 0x18
+#define SPI_PS_INPUT_CNTL_6__ATTR1_VALID_MASK 0x2000000
+#define SPI_PS_INPUT_CNTL_6__ATTR1_VALID__SHIFT 0x19
+#define SPI_PS_INPUT_CNTL_7__OFFSET_MASK 0x3f
+#define SPI_PS_INPUT_CNTL_7__OFFSET__SHIFT 0x0
+#define SPI_PS_INPUT_CNTL_7__DEFAULT_VAL_MASK 0x300
+#define SPI_PS_INPUT_CNTL_7__DEFAULT_VAL__SHIFT 0x8
+#define SPI_PS_INPUT_CNTL_7__FLAT_SHADE_MASK 0x400
+#define SPI_PS_INPUT_CNTL_7__FLAT_SHADE__SHIFT 0xa
+#define SPI_PS_INPUT_CNTL_7__CYL_WRAP_MASK 0x1e000
+#define SPI_PS_INPUT_CNTL_7__CYL_WRAP__SHIFT 0xd
+#define SPI_PS_INPUT_CNTL_7__PT_SPRITE_TEX_MASK 0x20000
+#define SPI_PS_INPUT_CNTL_7__PT_SPRITE_TEX__SHIFT 0x11
+#define SPI_PS_INPUT_CNTL_7__DUP_MASK 0x40000
+#define SPI_PS_INPUT_CNTL_7__DUP__SHIFT 0x12
+#define SPI_PS_INPUT_CNTL_7__FP16_INTERP_MODE_MASK 0x80000
+#define SPI_PS_INPUT_CNTL_7__FP16_INTERP_MODE__SHIFT 0x13
+#define SPI_PS_INPUT_CNTL_7__USE_DEFAULT_ATTR1_MASK 0x100000
+#define SPI_PS_INPUT_CNTL_7__USE_DEFAULT_ATTR1__SHIFT 0x14
+#define SPI_PS_INPUT_CNTL_7__DEFAULT_VAL_ATTR1_MASK 0x600000
+#define SPI_PS_INPUT_CNTL_7__DEFAULT_VAL_ATTR1__SHIFT 0x15
+#define SPI_PS_INPUT_CNTL_7__PT_SPRITE_TEX_ATTR1_MASK 0x800000
+#define SPI_PS_INPUT_CNTL_7__PT_SPRITE_TEX_ATTR1__SHIFT 0x17
+#define SPI_PS_INPUT_CNTL_7__ATTR0_VALID_MASK 0x1000000
+#define SPI_PS_INPUT_CNTL_7__ATTR0_VALID__SHIFT 0x18
+#define SPI_PS_INPUT_CNTL_7__ATTR1_VALID_MASK 0x2000000
+#define SPI_PS_INPUT_CNTL_7__ATTR1_VALID__SHIFT 0x19
+#define SPI_PS_INPUT_CNTL_8__OFFSET_MASK 0x3f
+#define SPI_PS_INPUT_CNTL_8__OFFSET__SHIFT 0x0
+#define SPI_PS_INPUT_CNTL_8__DEFAULT_VAL_MASK 0x300
+#define SPI_PS_INPUT_CNTL_8__DEFAULT_VAL__SHIFT 0x8
+#define SPI_PS_INPUT_CNTL_8__FLAT_SHADE_MASK 0x400
+#define SPI_PS_INPUT_CNTL_8__FLAT_SHADE__SHIFT 0xa
+#define SPI_PS_INPUT_CNTL_8__CYL_WRAP_MASK 0x1e000
+#define SPI_PS_INPUT_CNTL_8__CYL_WRAP__SHIFT 0xd
+#define SPI_PS_INPUT_CNTL_8__PT_SPRITE_TEX_MASK 0x20000
+#define SPI_PS_INPUT_CNTL_8__PT_SPRITE_TEX__SHIFT 0x11
+#define SPI_PS_INPUT_CNTL_8__DUP_MASK 0x40000
+#define SPI_PS_INPUT_CNTL_8__DUP__SHIFT 0x12
+#define SPI_PS_INPUT_CNTL_8__FP16_INTERP_MODE_MASK 0x80000
+#define SPI_PS_INPUT_CNTL_8__FP16_INTERP_MODE__SHIFT 0x13
+#define SPI_PS_INPUT_CNTL_8__USE_DEFAULT_ATTR1_MASK 0x100000
+#define SPI_PS_INPUT_CNTL_8__USE_DEFAULT_ATTR1__SHIFT 0x14
+#define SPI_PS_INPUT_CNTL_8__DEFAULT_VAL_ATTR1_MASK 0x600000
+#define SPI_PS_INPUT_CNTL_8__DEFAULT_VAL_ATTR1__SHIFT 0x15
+#define SPI_PS_INPUT_CNTL_8__PT_SPRITE_TEX_ATTR1_MASK 0x800000
+#define SPI_PS_INPUT_CNTL_8__PT_SPRITE_TEX_ATTR1__SHIFT 0x17
+#define SPI_PS_INPUT_CNTL_8__ATTR0_VALID_MASK 0x1000000
+#define SPI_PS_INPUT_CNTL_8__ATTR0_VALID__SHIFT 0x18
+#define SPI_PS_INPUT_CNTL_8__ATTR1_VALID_MASK 0x2000000
+#define SPI_PS_INPUT_CNTL_8__ATTR1_VALID__SHIFT 0x19
+#define SPI_PS_INPUT_CNTL_9__OFFSET_MASK 0x3f
+#define SPI_PS_INPUT_CNTL_9__OFFSET__SHIFT 0x0
+#define SPI_PS_INPUT_CNTL_9__DEFAULT_VAL_MASK 0x300
+#define SPI_PS_INPUT_CNTL_9__DEFAULT_VAL__SHIFT 0x8
+#define SPI_PS_INPUT_CNTL_9__FLAT_SHADE_MASK 0x400
+#define SPI_PS_INPUT_CNTL_9__FLAT_SHADE__SHIFT 0xa
+#define SPI_PS_INPUT_CNTL_9__CYL_WRAP_MASK 0x1e000
+#define SPI_PS_INPUT_CNTL_9__CYL_WRAP__SHIFT 0xd
+#define SPI_PS_INPUT_CNTL_9__PT_SPRITE_TEX_MASK 0x20000
+#define SPI_PS_INPUT_CNTL_9__PT_SPRITE_TEX__SHIFT 0x11
+#define SPI_PS_INPUT_CNTL_9__DUP_MASK 0x40000
+#define SPI_PS_INPUT_CNTL_9__DUP__SHIFT 0x12
+#define SPI_PS_INPUT_CNTL_9__FP16_INTERP_MODE_MASK 0x80000
+#define SPI_PS_INPUT_CNTL_9__FP16_INTERP_MODE__SHIFT 0x13
+#define SPI_PS_INPUT_CNTL_9__USE_DEFAULT_ATTR1_MASK 0x100000
+#define SPI_PS_INPUT_CNTL_9__USE_DEFAULT_ATTR1__SHIFT 0x14
+#define SPI_PS_INPUT_CNTL_9__DEFAULT_VAL_ATTR1_MASK 0x600000
+#define SPI_PS_INPUT_CNTL_9__DEFAULT_VAL_ATTR1__SHIFT 0x15
+#define SPI_PS_INPUT_CNTL_9__PT_SPRITE_TEX_ATTR1_MASK 0x800000
+#define SPI_PS_INPUT_CNTL_9__PT_SPRITE_TEX_ATTR1__SHIFT 0x17
+#define SPI_PS_INPUT_CNTL_9__ATTR0_VALID_MASK 0x1000000
+#define SPI_PS_INPUT_CNTL_9__ATTR0_VALID__SHIFT 0x18
+#define SPI_PS_INPUT_CNTL_9__ATTR1_VALID_MASK 0x2000000
+#define SPI_PS_INPUT_CNTL_9__ATTR1_VALID__SHIFT 0x19
+#define SPI_PS_INPUT_CNTL_10__OFFSET_MASK 0x3f
+#define SPI_PS_INPUT_CNTL_10__OFFSET__SHIFT 0x0
+#define SPI_PS_INPUT_CNTL_10__DEFAULT_VAL_MASK 0x300
+#define SPI_PS_INPUT_CNTL_10__DEFAULT_VAL__SHIFT 0x8
+#define SPI_PS_INPUT_CNTL_10__FLAT_SHADE_MASK 0x400
+#define SPI_PS_INPUT_CNTL_10__FLAT_SHADE__SHIFT 0xa
+#define SPI_PS_INPUT_CNTL_10__CYL_WRAP_MASK 0x1e000
+#define SPI_PS_INPUT_CNTL_10__CYL_WRAP__SHIFT 0xd
+#define SPI_PS_INPUT_CNTL_10__PT_SPRITE_TEX_MASK 0x20000
+#define SPI_PS_INPUT_CNTL_10__PT_SPRITE_TEX__SHIFT 0x11
+#define SPI_PS_INPUT_CNTL_10__DUP_MASK 0x40000
+#define SPI_PS_INPUT_CNTL_10__DUP__SHIFT 0x12
+#define SPI_PS_INPUT_CNTL_10__FP16_INTERP_MODE_MASK 0x80000
+#define SPI_PS_INPUT_CNTL_10__FP16_INTERP_MODE__SHIFT 0x13
+#define SPI_PS_INPUT_CNTL_10__USE_DEFAULT_ATTR1_MASK 0x100000
+#define SPI_PS_INPUT_CNTL_10__USE_DEFAULT_ATTR1__SHIFT 0x14
+#define SPI_PS_INPUT_CNTL_10__DEFAULT_VAL_ATTR1_MASK 0x600000
+#define SPI_PS_INPUT_CNTL_10__DEFAULT_VAL_ATTR1__SHIFT 0x15
+#define SPI_PS_INPUT_CNTL_10__PT_SPRITE_TEX_ATTR1_MASK 0x800000
+#define SPI_PS_INPUT_CNTL_10__PT_SPRITE_TEX_ATTR1__SHIFT 0x17
+#define SPI_PS_INPUT_CNTL_10__ATTR0_VALID_MASK 0x1000000
+#define SPI_PS_INPUT_CNTL_10__ATTR0_VALID__SHIFT 0x18
+#define SPI_PS_INPUT_CNTL_10__ATTR1_VALID_MASK 0x2000000
+#define SPI_PS_INPUT_CNTL_10__ATTR1_VALID__SHIFT 0x19
+#define SPI_PS_INPUT_CNTL_11__OFFSET_MASK 0x3f
+#define SPI_PS_INPUT_CNTL_11__OFFSET__SHIFT 0x0
+#define SPI_PS_INPUT_CNTL_11__DEFAULT_VAL_MASK 0x300
+#define SPI_PS_INPUT_CNTL_11__DEFAULT_VAL__SHIFT 0x8
+#define SPI_PS_INPUT_CNTL_11__FLAT_SHADE_MASK 0x400
+#define SPI_PS_INPUT_CNTL_11__FLAT_SHADE__SHIFT 0xa
+#define SPI_PS_INPUT_CNTL_11__CYL_WRAP_MASK 0x1e000
+#define SPI_PS_INPUT_CNTL_11__CYL_WRAP__SHIFT 0xd
+#define SPI_PS_INPUT_CNTL_11__PT_SPRITE_TEX_MASK 0x20000
+#define SPI_PS_INPUT_CNTL_11__PT_SPRITE_TEX__SHIFT 0x11
+#define SPI_PS_INPUT_CNTL_11__DUP_MASK 0x40000
+#define SPI_PS_INPUT_CNTL_11__DUP__SHIFT 0x12
+#define SPI_PS_INPUT_CNTL_11__FP16_INTERP_MODE_MASK 0x80000
+#define SPI_PS_INPUT_CNTL_11__FP16_INTERP_MODE__SHIFT 0x13
+#define SPI_PS_INPUT_CNTL_11__USE_DEFAULT_ATTR1_MASK 0x100000
+#define SPI_PS_INPUT_CNTL_11__USE_DEFAULT_ATTR1__SHIFT 0x14
+#define SPI_PS_INPUT_CNTL_11__DEFAULT_VAL_ATTR1_MASK 0x600000
+#define SPI_PS_INPUT_CNTL_11__DEFAULT_VAL_ATTR1__SHIFT 0x15
+#define SPI_PS_INPUT_CNTL_11__PT_SPRITE_TEX_ATTR1_MASK 0x800000
+#define SPI_PS_INPUT_CNTL_11__PT_SPRITE_TEX_ATTR1__SHIFT 0x17
+#define SPI_PS_INPUT_CNTL_11__ATTR0_VALID_MASK 0x1000000
+#define SPI_PS_INPUT_CNTL_11__ATTR0_VALID__SHIFT 0x18
+#define SPI_PS_INPUT_CNTL_11__ATTR1_VALID_MASK 0x2000000
+#define SPI_PS_INPUT_CNTL_11__ATTR1_VALID__SHIFT 0x19
+#define SPI_PS_INPUT_CNTL_12__OFFSET_MASK 0x3f
+#define SPI_PS_INPUT_CNTL_12__OFFSET__SHIFT 0x0
+#define SPI_PS_INPUT_CNTL_12__DEFAULT_VAL_MASK 0x300
+#define SPI_PS_INPUT_CNTL_12__DEFAULT_VAL__SHIFT 0x8
+#define SPI_PS_INPUT_CNTL_12__FLAT_SHADE_MASK 0x400
+#define SPI_PS_INPUT_CNTL_12__FLAT_SHADE__SHIFT 0xa
+#define SPI_PS_INPUT_CNTL_12__CYL_WRAP_MASK 0x1e000
+#define SPI_PS_INPUT_CNTL_12__CYL_WRAP__SHIFT 0xd
+#define SPI_PS_INPUT_CNTL_12__PT_SPRITE_TEX_MASK 0x20000
+#define SPI_PS_INPUT_CNTL_12__PT_SPRITE_TEX__SHIFT 0x11
+#define SPI_PS_INPUT_CNTL_12__DUP_MASK 0x40000
+#define SPI_PS_INPUT_CNTL_12__DUP__SHIFT 0x12
+#define SPI_PS_INPUT_CNTL_12__FP16_INTERP_MODE_MASK 0x80000
+#define SPI_PS_INPUT_CNTL_12__FP16_INTERP_MODE__SHIFT 0x13
+#define SPI_PS_INPUT_CNTL_12__USE_DEFAULT_ATTR1_MASK 0x100000
+#define SPI_PS_INPUT_CNTL_12__USE_DEFAULT_ATTR1__SHIFT 0x14
+#define SPI_PS_INPUT_CNTL_12__DEFAULT_VAL_ATTR1_MASK 0x600000
+#define SPI_PS_INPUT_CNTL_12__DEFAULT_VAL_ATTR1__SHIFT 0x15
+#define SPI_PS_INPUT_CNTL_12__PT_SPRITE_TEX_ATTR1_MASK 0x800000
+#define SPI_PS_INPUT_CNTL_12__PT_SPRITE_TEX_ATTR1__SHIFT 0x17
+#define SPI_PS_INPUT_CNTL_12__ATTR0_VALID_MASK 0x1000000
+#define SPI_PS_INPUT_CNTL_12__ATTR0_VALID__SHIFT 0x18
+#define SPI_PS_INPUT_CNTL_12__ATTR1_VALID_MASK 0x2000000
+#define SPI_PS_INPUT_CNTL_12__ATTR1_VALID__SHIFT 0x19
+#define SPI_PS_INPUT_CNTL_13__OFFSET_MASK 0x3f
+#define SPI_PS_INPUT_CNTL_13__OFFSET__SHIFT 0x0
+#define SPI_PS_INPUT_CNTL_13__DEFAULT_VAL_MASK 0x300
+#define SPI_PS_INPUT_CNTL_13__DEFAULT_VAL__SHIFT 0x8
+#define SPI_PS_INPUT_CNTL_13__FLAT_SHADE_MASK 0x400
+#define SPI_PS_INPUT_CNTL_13__FLAT_SHADE__SHIFT 0xa
+#define SPI_PS_INPUT_CNTL_13__CYL_WRAP_MASK 0x1e000
+#define SPI_PS_INPUT_CNTL_13__CYL_WRAP__SHIFT 0xd
+#define SPI_PS_INPUT_CNTL_13__PT_SPRITE_TEX_MASK 0x20000
+#define SPI_PS_INPUT_CNTL_13__PT_SPRITE_TEX__SHIFT 0x11
+#define SPI_PS_INPUT_CNTL_13__DUP_MASK 0x40000
+#define SPI_PS_INPUT_CNTL_13__DUP__SHIFT 0x12
+#define SPI_PS_INPUT_CNTL_13__FP16_INTERP_MODE_MASK 0x80000
+#define SPI_PS_INPUT_CNTL_13__FP16_INTERP_MODE__SHIFT 0x13
+#define SPI_PS_INPUT_CNTL_13__USE_DEFAULT_ATTR1_MASK 0x100000
+#define SPI_PS_INPUT_CNTL_13__USE_DEFAULT_ATTR1__SHIFT 0x14
+#define SPI_PS_INPUT_CNTL_13__DEFAULT_VAL_ATTR1_MASK 0x600000
+#define SPI_PS_INPUT_CNTL_13__DEFAULT_VAL_ATTR1__SHIFT 0x15
+#define SPI_PS_INPUT_CNTL_13__PT_SPRITE_TEX_ATTR1_MASK 0x800000
+#define SPI_PS_INPUT_CNTL_13__PT_SPRITE_TEX_ATTR1__SHIFT 0x17
+#define SPI_PS_INPUT_CNTL_13__ATTR0_VALID_MASK 0x1000000
+#define SPI_PS_INPUT_CNTL_13__ATTR0_VALID__SHIFT 0x18
+#define SPI_PS_INPUT_CNTL_13__ATTR1_VALID_MASK 0x2000000
+#define SPI_PS_INPUT_CNTL_13__ATTR1_VALID__SHIFT 0x19
+#define SPI_PS_INPUT_CNTL_14__OFFSET_MASK 0x3f
+#define SPI_PS_INPUT_CNTL_14__OFFSET__SHIFT 0x0
+#define SPI_PS_INPUT_CNTL_14__DEFAULT_VAL_MASK 0x300
+#define SPI_PS_INPUT_CNTL_14__DEFAULT_VAL__SHIFT 0x8
+#define SPI_PS_INPUT_CNTL_14__FLAT_SHADE_MASK 0x400
+#define SPI_PS_INPUT_CNTL_14__FLAT_SHADE__SHIFT 0xa
+#define SPI_PS_INPUT_CNTL_14__CYL_WRAP_MASK 0x1e000
+#define SPI_PS_INPUT_CNTL_14__CYL_WRAP__SHIFT 0xd
+#define SPI_PS_INPUT_CNTL_14__PT_SPRITE_TEX_MASK 0x20000
+#define SPI_PS_INPUT_CNTL_14__PT_SPRITE_TEX__SHIFT 0x11
+#define SPI_PS_INPUT_CNTL_14__DUP_MASK 0x40000
+#define SPI_PS_INPUT_CNTL_14__DUP__SHIFT 0x12
+#define SPI_PS_INPUT_CNTL_14__FP16_INTERP_MODE_MASK 0x80000
+#define SPI_PS_INPUT_CNTL_14__FP16_INTERP_MODE__SHIFT 0x13
+#define SPI_PS_INPUT_CNTL_14__USE_DEFAULT_ATTR1_MASK 0x100000
+#define SPI_PS_INPUT_CNTL_14__USE_DEFAULT_ATTR1__SHIFT 0x14
+#define SPI_PS_INPUT_CNTL_14__DEFAULT_VAL_ATTR1_MASK 0x600000
+#define SPI_PS_INPUT_CNTL_14__DEFAULT_VAL_ATTR1__SHIFT 0x15
+#define SPI_PS_INPUT_CNTL_14__PT_SPRITE_TEX_ATTR1_MASK 0x800000
+#define SPI_PS_INPUT_CNTL_14__PT_SPRITE_TEX_ATTR1__SHIFT 0x17
+#define SPI_PS_INPUT_CNTL_14__ATTR0_VALID_MASK 0x1000000
+#define SPI_PS_INPUT_CNTL_14__ATTR0_VALID__SHIFT 0x18
+#define SPI_PS_INPUT_CNTL_14__ATTR1_VALID_MASK 0x2000000
+#define SPI_PS_INPUT_CNTL_14__ATTR1_VALID__SHIFT 0x19
+#define SPI_PS_INPUT_CNTL_15__OFFSET_MASK 0x3f
+#define SPI_PS_INPUT_CNTL_15__OFFSET__SHIFT 0x0
+#define SPI_PS_INPUT_CNTL_15__DEFAULT_VAL_MASK 0x300
+#define SPI_PS_INPUT_CNTL_15__DEFAULT_VAL__SHIFT 0x8
+#define SPI_PS_INPUT_CNTL_15__FLAT_SHADE_MASK 0x400
+#define SPI_PS_INPUT_CNTL_15__FLAT_SHADE__SHIFT 0xa
+#define SPI_PS_INPUT_CNTL_15__CYL_WRAP_MASK 0x1e000
+#define SPI_PS_INPUT_CNTL_15__CYL_WRAP__SHIFT 0xd
+#define SPI_PS_INPUT_CNTL_15__PT_SPRITE_TEX_MASK 0x20000
+#define SPI_PS_INPUT_CNTL_15__PT_SPRITE_TEX__SHIFT 0x11
+#define SPI_PS_INPUT_CNTL_15__DUP_MASK 0x40000
+#define SPI_PS_INPUT_CNTL_15__DUP__SHIFT 0x12
+#define SPI_PS_INPUT_CNTL_15__FP16_INTERP_MODE_MASK 0x80000
+#define SPI_PS_INPUT_CNTL_15__FP16_INTERP_MODE__SHIFT 0x13
+#define SPI_PS_INPUT_CNTL_15__USE_DEFAULT_ATTR1_MASK 0x100000
+#define SPI_PS_INPUT_CNTL_15__USE_DEFAULT_ATTR1__SHIFT 0x14
+#define SPI_PS_INPUT_CNTL_15__DEFAULT_VAL_ATTR1_MASK 0x600000
+#define SPI_PS_INPUT_CNTL_15__DEFAULT_VAL_ATTR1__SHIFT 0x15
+#define SPI_PS_INPUT_CNTL_15__PT_SPRITE_TEX_ATTR1_MASK 0x800000
+#define SPI_PS_INPUT_CNTL_15__PT_SPRITE_TEX_ATTR1__SHIFT 0x17
+#define SPI_PS_INPUT_CNTL_15__ATTR0_VALID_MASK 0x1000000
+#define SPI_PS_INPUT_CNTL_15__ATTR0_VALID__SHIFT 0x18
+#define SPI_PS_INPUT_CNTL_15__ATTR1_VALID_MASK 0x2000000
+#define SPI_PS_INPUT_CNTL_15__ATTR1_VALID__SHIFT 0x19
+#define SPI_PS_INPUT_CNTL_16__OFFSET_MASK 0x3f
+#define SPI_PS_INPUT_CNTL_16__OFFSET__SHIFT 0x0
+#define SPI_PS_INPUT_CNTL_16__DEFAULT_VAL_MASK 0x300
+#define SPI_PS_INPUT_CNTL_16__DEFAULT_VAL__SHIFT 0x8
+#define SPI_PS_INPUT_CNTL_16__FLAT_SHADE_MASK 0x400
+#define SPI_PS_INPUT_CNTL_16__FLAT_SHADE__SHIFT 0xa
+#define SPI_PS_INPUT_CNTL_16__CYL_WRAP_MASK 0x1e000
+#define SPI_PS_INPUT_CNTL_16__CYL_WRAP__SHIFT 0xd
+#define SPI_PS_INPUT_CNTL_16__PT_SPRITE_TEX_MASK 0x20000
+#define SPI_PS_INPUT_CNTL_16__PT_SPRITE_TEX__SHIFT 0x11
+#define SPI_PS_INPUT_CNTL_16__DUP_MASK 0x40000
+#define SPI_PS_INPUT_CNTL_16__DUP__SHIFT 0x12
+#define SPI_PS_INPUT_CNTL_16__FP16_INTERP_MODE_MASK 0x80000
+#define SPI_PS_INPUT_CNTL_16__FP16_INTERP_MODE__SHIFT 0x13
+#define SPI_PS_INPUT_CNTL_16__USE_DEFAULT_ATTR1_MASK 0x100000
+#define SPI_PS_INPUT_CNTL_16__USE_DEFAULT_ATTR1__SHIFT 0x14
+#define SPI_PS_INPUT_CNTL_16__DEFAULT_VAL_ATTR1_MASK 0x600000
+#define SPI_PS_INPUT_CNTL_16__DEFAULT_VAL_ATTR1__SHIFT 0x15
+#define SPI_PS_INPUT_CNTL_16__PT_SPRITE_TEX_ATTR1_MASK 0x800000
+#define SPI_PS_INPUT_CNTL_16__PT_SPRITE_TEX_ATTR1__SHIFT 0x17
+#define SPI_PS_INPUT_CNTL_16__ATTR0_VALID_MASK 0x1000000
+#define SPI_PS_INPUT_CNTL_16__ATTR0_VALID__SHIFT 0x18
+#define SPI_PS_INPUT_CNTL_16__ATTR1_VALID_MASK 0x2000000
+#define SPI_PS_INPUT_CNTL_16__ATTR1_VALID__SHIFT 0x19
+#define SPI_PS_INPUT_CNTL_17__OFFSET_MASK 0x3f
+#define SPI_PS_INPUT_CNTL_17__OFFSET__SHIFT 0x0
+#define SPI_PS_INPUT_CNTL_17__DEFAULT_VAL_MASK 0x300
+#define SPI_PS_INPUT_CNTL_17__DEFAULT_VAL__SHIFT 0x8
+#define SPI_PS_INPUT_CNTL_17__FLAT_SHADE_MASK 0x400
+#define SPI_PS_INPUT_CNTL_17__FLAT_SHADE__SHIFT 0xa
+#define SPI_PS_INPUT_CNTL_17__CYL_WRAP_MASK 0x1e000
+#define SPI_PS_INPUT_CNTL_17__CYL_WRAP__SHIFT 0xd
+#define SPI_PS_INPUT_CNTL_17__PT_SPRITE_TEX_MASK 0x20000
+#define SPI_PS_INPUT_CNTL_17__PT_SPRITE_TEX__SHIFT 0x11
+#define SPI_PS_INPUT_CNTL_17__DUP_MASK 0x40000
+#define SPI_PS_INPUT_CNTL_17__DUP__SHIFT 0x12
+#define SPI_PS_INPUT_CNTL_17__FP16_INTERP_MODE_MASK 0x80000
+#define SPI_PS_INPUT_CNTL_17__FP16_INTERP_MODE__SHIFT 0x13
+#define SPI_PS_INPUT_CNTL_17__USE_DEFAULT_ATTR1_MASK 0x100000
+#define SPI_PS_INPUT_CNTL_17__USE_DEFAULT_ATTR1__SHIFT 0x14
+#define SPI_PS_INPUT_CNTL_17__DEFAULT_VAL_ATTR1_MASK 0x600000
+#define SPI_PS_INPUT_CNTL_17__DEFAULT_VAL_ATTR1__SHIFT 0x15
+#define SPI_PS_INPUT_CNTL_17__PT_SPRITE_TEX_ATTR1_MASK 0x800000
+#define SPI_PS_INPUT_CNTL_17__PT_SPRITE_TEX_ATTR1__SHIFT 0x17
+#define SPI_PS_INPUT_CNTL_17__ATTR0_VALID_MASK 0x1000000
+#define SPI_PS_INPUT_CNTL_17__ATTR0_VALID__SHIFT 0x18
+#define SPI_PS_INPUT_CNTL_17__ATTR1_VALID_MASK 0x2000000
+#define SPI_PS_INPUT_CNTL_17__ATTR1_VALID__SHIFT 0x19
+#define SPI_PS_INPUT_CNTL_18__OFFSET_MASK 0x3f
+#define SPI_PS_INPUT_CNTL_18__OFFSET__SHIFT 0x0
+#define SPI_PS_INPUT_CNTL_18__DEFAULT_VAL_MASK 0x300
+#define SPI_PS_INPUT_CNTL_18__DEFAULT_VAL__SHIFT 0x8
+#define SPI_PS_INPUT_CNTL_18__FLAT_SHADE_MASK 0x400
+#define SPI_PS_INPUT_CNTL_18__FLAT_SHADE__SHIFT 0xa
+#define SPI_PS_INPUT_CNTL_18__CYL_WRAP_MASK 0x1e000
+#define SPI_PS_INPUT_CNTL_18__CYL_WRAP__SHIFT 0xd
+#define SPI_PS_INPUT_CNTL_18__PT_SPRITE_TEX_MASK 0x20000
+#define SPI_PS_INPUT_CNTL_18__PT_SPRITE_TEX__SHIFT 0x11
+#define SPI_PS_INPUT_CNTL_18__DUP_MASK 0x40000
+#define SPI_PS_INPUT_CNTL_18__DUP__SHIFT 0x12
+#define SPI_PS_INPUT_CNTL_18__FP16_INTERP_MODE_MASK 0x80000
+#define SPI_PS_INPUT_CNTL_18__FP16_INTERP_MODE__SHIFT 0x13
+#define SPI_PS_INPUT_CNTL_18__USE_DEFAULT_ATTR1_MASK 0x100000
+#define SPI_PS_INPUT_CNTL_18__USE_DEFAULT_ATTR1__SHIFT 0x14
+#define SPI_PS_INPUT_CNTL_18__DEFAULT_VAL_ATTR1_MASK 0x600000
+#define SPI_PS_INPUT_CNTL_18__DEFAULT_VAL_ATTR1__SHIFT 0x15
+#define SPI_PS_INPUT_CNTL_18__PT_SPRITE_TEX_ATTR1_MASK 0x800000
+#define SPI_PS_INPUT_CNTL_18__PT_SPRITE_TEX_ATTR1__SHIFT 0x17
+#define SPI_PS_INPUT_CNTL_18__ATTR0_VALID_MASK 0x1000000
+#define SPI_PS_INPUT_CNTL_18__ATTR0_VALID__SHIFT 0x18
+#define SPI_PS_INPUT_CNTL_18__ATTR1_VALID_MASK 0x2000000
+#define SPI_PS_INPUT_CNTL_18__ATTR1_VALID__SHIFT 0x19
+#define SPI_PS_INPUT_CNTL_19__OFFSET_MASK 0x3f
+#define SPI_PS_INPUT_CNTL_19__OFFSET__SHIFT 0x0
+#define SPI_PS_INPUT_CNTL_19__DEFAULT_VAL_MASK 0x300
+#define SPI_PS_INPUT_CNTL_19__DEFAULT_VAL__SHIFT 0x8
+#define SPI_PS_INPUT_CNTL_19__FLAT_SHADE_MASK 0x400
+#define SPI_PS_INPUT_CNTL_19__FLAT_SHADE__SHIFT 0xa
+#define SPI_PS_INPUT_CNTL_19__CYL_WRAP_MASK 0x1e000
+#define SPI_PS_INPUT_CNTL_19__CYL_WRAP__SHIFT 0xd
+#define SPI_PS_INPUT_CNTL_19__PT_SPRITE_TEX_MASK 0x20000
+#define SPI_PS_INPUT_CNTL_19__PT_SPRITE_TEX__SHIFT 0x11
+#define SPI_PS_INPUT_CNTL_19__DUP_MASK 0x40000
+#define SPI_PS_INPUT_CNTL_19__DUP__SHIFT 0x12
+#define SPI_PS_INPUT_CNTL_19__FP16_INTERP_MODE_MASK 0x80000
+#define SPI_PS_INPUT_CNTL_19__FP16_INTERP_MODE__SHIFT 0x13
+#define SPI_PS_INPUT_CNTL_19__USE_DEFAULT_ATTR1_MASK 0x100000
+#define SPI_PS_INPUT_CNTL_19__USE_DEFAULT_ATTR1__SHIFT 0x14
+#define SPI_PS_INPUT_CNTL_19__DEFAULT_VAL_ATTR1_MASK 0x600000
+#define SPI_PS_INPUT_CNTL_19__DEFAULT_VAL_ATTR1__SHIFT 0x15
+#define SPI_PS_INPUT_CNTL_19__PT_SPRITE_TEX_ATTR1_MASK 0x800000
+#define SPI_PS_INPUT_CNTL_19__PT_SPRITE_TEX_ATTR1__SHIFT 0x17
+#define SPI_PS_INPUT_CNTL_19__ATTR0_VALID_MASK 0x1000000
+#define SPI_PS_INPUT_CNTL_19__ATTR0_VALID__SHIFT 0x18
+#define SPI_PS_INPUT_CNTL_19__ATTR1_VALID_MASK 0x2000000
+#define SPI_PS_INPUT_CNTL_19__ATTR1_VALID__SHIFT 0x19
+#define SPI_PS_INPUT_CNTL_20__OFFSET_MASK 0x3f
+#define SPI_PS_INPUT_CNTL_20__OFFSET__SHIFT 0x0
+#define SPI_PS_INPUT_CNTL_20__DEFAULT_VAL_MASK 0x300
+#define SPI_PS_INPUT_CNTL_20__DEFAULT_VAL__SHIFT 0x8
+#define SPI_PS_INPUT_CNTL_20__FLAT_SHADE_MASK 0x400
+#define SPI_PS_INPUT_CNTL_20__FLAT_SHADE__SHIFT 0xa
+#define SPI_PS_INPUT_CNTL_20__DUP_MASK 0x40000
+#define SPI_PS_INPUT_CNTL_20__DUP__SHIFT 0x12
+#define SPI_PS_INPUT_CNTL_20__FP16_INTERP_MODE_MASK 0x80000
+#define SPI_PS_INPUT_CNTL_20__FP16_INTERP_MODE__SHIFT 0x13
+#define SPI_PS_INPUT_CNTL_20__USE_DEFAULT_ATTR1_MASK 0x100000
+#define SPI_PS_INPUT_CNTL_20__USE_DEFAULT_ATTR1__SHIFT 0x14
+#define SPI_PS_INPUT_CNTL_20__DEFAULT_VAL_ATTR1_MASK 0x600000
+#define SPI_PS_INPUT_CNTL_20__DEFAULT_VAL_ATTR1__SHIFT 0x15
+#define SPI_PS_INPUT_CNTL_20__ATTR0_VALID_MASK 0x1000000
+#define SPI_PS_INPUT_CNTL_20__ATTR0_VALID__SHIFT 0x18
+#define SPI_PS_INPUT_CNTL_20__ATTR1_VALID_MASK 0x2000000
+#define SPI_PS_INPUT_CNTL_20__ATTR1_VALID__SHIFT 0x19
+#define SPI_PS_INPUT_CNTL_21__OFFSET_MASK 0x3f
+#define SPI_PS_INPUT_CNTL_21__OFFSET__SHIFT 0x0
+#define SPI_PS_INPUT_CNTL_21__DEFAULT_VAL_MASK 0x300
+#define SPI_PS_INPUT_CNTL_21__DEFAULT_VAL__SHIFT 0x8
+#define SPI_PS_INPUT_CNTL_21__FLAT_SHADE_MASK 0x400
+#define SPI_PS_INPUT_CNTL_21__FLAT_SHADE__SHIFT 0xa
+#define SPI_PS_INPUT_CNTL_21__DUP_MASK 0x40000
+#define SPI_PS_INPUT_CNTL_21__DUP__SHIFT 0x12
+#define SPI_PS_INPUT_CNTL_21__FP16_INTERP_MODE_MASK 0x80000
+#define SPI_PS_INPUT_CNTL_21__FP16_INTERP_MODE__SHIFT 0x13
+#define SPI_PS_INPUT_CNTL_21__USE_DEFAULT_ATTR1_MASK 0x100000
+#define SPI_PS_INPUT_CNTL_21__USE_DEFAULT_ATTR1__SHIFT 0x14
+#define SPI_PS_INPUT_CNTL_21__DEFAULT_VAL_ATTR1_MASK 0x600000
+#define SPI_PS_INPUT_CNTL_21__DEFAULT_VAL_ATTR1__SHIFT 0x15
+#define SPI_PS_INPUT_CNTL_21__ATTR0_VALID_MASK 0x1000000
+#define SPI_PS_INPUT_CNTL_21__ATTR0_VALID__SHIFT 0x18
+#define SPI_PS_INPUT_CNTL_21__ATTR1_VALID_MASK 0x2000000
+#define SPI_PS_INPUT_CNTL_21__ATTR1_VALID__SHIFT 0x19
+#define SPI_PS_INPUT_CNTL_22__OFFSET_MASK 0x3f
+#define SPI_PS_INPUT_CNTL_22__OFFSET__SHIFT 0x0
+#define SPI_PS_INPUT_CNTL_22__DEFAULT_VAL_MASK 0x300
+#define SPI_PS_INPUT_CNTL_22__DEFAULT_VAL__SHIFT 0x8
+#define SPI_PS_INPUT_CNTL_22__FLAT_SHADE_MASK 0x400
+#define SPI_PS_INPUT_CNTL_22__FLAT_SHADE__SHIFT 0xa
+#define SPI_PS_INPUT_CNTL_22__DUP_MASK 0x40000
+#define SPI_PS_INPUT_CNTL_22__DUP__SHIFT 0x12
+#define SPI_PS_INPUT_CNTL_22__FP16_INTERP_MODE_MASK 0x80000
+#define SPI_PS_INPUT_CNTL_22__FP16_INTERP_MODE__SHIFT 0x13
+#define SPI_PS_INPUT_CNTL_22__USE_DEFAULT_ATTR1_MASK 0x100000
+#define SPI_PS_INPUT_CNTL_22__USE_DEFAULT_ATTR1__SHIFT 0x14
+#define SPI_PS_INPUT_CNTL_22__DEFAULT_VAL_ATTR1_MASK 0x600000
+#define SPI_PS_INPUT_CNTL_22__DEFAULT_VAL_ATTR1__SHIFT 0x15
+#define SPI_PS_INPUT_CNTL_22__ATTR0_VALID_MASK 0x1000000
+#define SPI_PS_INPUT_CNTL_22__ATTR0_VALID__SHIFT 0x18
+#define SPI_PS_INPUT_CNTL_22__ATTR1_VALID_MASK 0x2000000
+#define SPI_PS_INPUT_CNTL_22__ATTR1_VALID__SHIFT 0x19
+#define SPI_PS_INPUT_CNTL_23__OFFSET_MASK 0x3f
+#define SPI_PS_INPUT_CNTL_23__OFFSET__SHIFT 0x0
+#define SPI_PS_INPUT_CNTL_23__DEFAULT_VAL_MASK 0x300
+#define SPI_PS_INPUT_CNTL_23__DEFAULT_VAL__SHIFT 0x8
+#define SPI_PS_INPUT_CNTL_23__FLAT_SHADE_MASK 0x400
+#define SPI_PS_INPUT_CNTL_23__FLAT_SHADE__SHIFT 0xa
+#define SPI_PS_INPUT_CNTL_23__DUP_MASK 0x40000
+#define SPI_PS_INPUT_CNTL_23__DUP__SHIFT 0x12
+#define SPI_PS_INPUT_CNTL_23__FP16_INTERP_MODE_MASK 0x80000
+#define SPI_PS_INPUT_CNTL_23__FP16_INTERP_MODE__SHIFT 0x13
+#define SPI_PS_INPUT_CNTL_23__USE_DEFAULT_ATTR1_MASK 0x100000
+#define SPI_PS_INPUT_CNTL_23__USE_DEFAULT_ATTR1__SHIFT 0x14
+#define SPI_PS_INPUT_CNTL_23__DEFAULT_VAL_ATTR1_MASK 0x600000
+#define SPI_PS_INPUT_CNTL_23__DEFAULT_VAL_ATTR1__SHIFT 0x15
+#define SPI_PS_INPUT_CNTL_23__ATTR0_VALID_MASK 0x1000000
+#define SPI_PS_INPUT_CNTL_23__ATTR0_VALID__SHIFT 0x18
+#define SPI_PS_INPUT_CNTL_23__ATTR1_VALID_MASK 0x2000000
+#define SPI_PS_INPUT_CNTL_23__ATTR1_VALID__SHIFT 0x19
+#define SPI_PS_INPUT_CNTL_24__OFFSET_MASK 0x3f
+#define SPI_PS_INPUT_CNTL_24__OFFSET__SHIFT 0x0
+#define SPI_PS_INPUT_CNTL_24__DEFAULT_VAL_MASK 0x300
+#define SPI_PS_INPUT_CNTL_24__DEFAULT_VAL__SHIFT 0x8
+#define SPI_PS_INPUT_CNTL_24__FLAT_SHADE_MASK 0x400
+#define SPI_PS_INPUT_CNTL_24__FLAT_SHADE__SHIFT 0xa
+#define SPI_PS_INPUT_CNTL_24__DUP_MASK 0x40000
+#define SPI_PS_INPUT_CNTL_24__DUP__SHIFT 0x12
+#define SPI_PS_INPUT_CNTL_24__FP16_INTERP_MODE_MASK 0x80000
+#define SPI_PS_INPUT_CNTL_24__FP16_INTERP_MODE__SHIFT 0x13
+#define SPI_PS_INPUT_CNTL_24__USE_DEFAULT_ATTR1_MASK 0x100000
+#define SPI_PS_INPUT_CNTL_24__USE_DEFAULT_ATTR1__SHIFT 0x14
+#define SPI_PS_INPUT_CNTL_24__DEFAULT_VAL_ATTR1_MASK 0x600000
+#define SPI_PS_INPUT_CNTL_24__DEFAULT_VAL_ATTR1__SHIFT 0x15
+#define SPI_PS_INPUT_CNTL_24__ATTR0_VALID_MASK 0x1000000
+#define SPI_PS_INPUT_CNTL_24__ATTR0_VALID__SHIFT 0x18
+#define SPI_PS_INPUT_CNTL_24__ATTR1_VALID_MASK 0x2000000
+#define SPI_PS_INPUT_CNTL_24__ATTR1_VALID__SHIFT 0x19
+#define SPI_PS_INPUT_CNTL_25__OFFSET_MASK 0x3f
+#define SPI_PS_INPUT_CNTL_25__OFFSET__SHIFT 0x0
+#define SPI_PS_INPUT_CNTL_25__DEFAULT_VAL_MASK 0x300
+#define SPI_PS_INPUT_CNTL_25__DEFAULT_VAL__SHIFT 0x8
+#define SPI_PS_INPUT_CNTL_25__FLAT_SHADE_MASK 0x400
+#define SPI_PS_INPUT_CNTL_25__FLAT_SHADE__SHIFT 0xa
+#define SPI_PS_INPUT_CNTL_25__DUP_MASK 0x40000
+#define SPI_PS_INPUT_CNTL_25__DUP__SHIFT 0x12
+#define SPI_PS_INPUT_CNTL_25__FP16_INTERP_MODE_MASK 0x80000
+#define SPI_PS_INPUT_CNTL_25__FP16_INTERP_MODE__SHIFT 0x13
+#define SPI_PS_INPUT_CNTL_25__USE_DEFAULT_ATTR1_MASK 0x100000
+#define SPI_PS_INPUT_CNTL_25__USE_DEFAULT_ATTR1__SHIFT 0x14
+#define SPI_PS_INPUT_CNTL_25__DEFAULT_VAL_ATTR1_MASK 0x600000
+#define SPI_PS_INPUT_CNTL_25__DEFAULT_VAL_ATTR1__SHIFT 0x15
+#define SPI_PS_INPUT_CNTL_25__ATTR0_VALID_MASK 0x1000000
+#define SPI_PS_INPUT_CNTL_25__ATTR0_VALID__SHIFT 0x18
+#define SPI_PS_INPUT_CNTL_25__ATTR1_VALID_MASK 0x2000000
+#define SPI_PS_INPUT_CNTL_25__ATTR1_VALID__SHIFT 0x19
+#define SPI_PS_INPUT_CNTL_26__OFFSET_MASK 0x3f
+#define SPI_PS_INPUT_CNTL_26__OFFSET__SHIFT 0x0
+#define SPI_PS_INPUT_CNTL_26__DEFAULT_VAL_MASK 0x300
+#define SPI_PS_INPUT_CNTL_26__DEFAULT_VAL__SHIFT 0x8
+#define SPI_PS_INPUT_CNTL_26__FLAT_SHADE_MASK 0x400
+#define SPI_PS_INPUT_CNTL_26__FLAT_SHADE__SHIFT 0xa
+#define SPI_PS_INPUT_CNTL_26__DUP_MASK 0x40000
+#define SPI_PS_INPUT_CNTL_26__DUP__SHIFT 0x12
+#define SPI_PS_INPUT_CNTL_26__FP16_INTERP_MODE_MASK 0x80000
+#define SPI_PS_INPUT_CNTL_26__FP16_INTERP_MODE__SHIFT 0x13
+#define SPI_PS_INPUT_CNTL_26__USE_DEFAULT_ATTR1_MASK 0x100000
+#define SPI_PS_INPUT_CNTL_26__USE_DEFAULT_ATTR1__SHIFT 0x14
+#define SPI_PS_INPUT_CNTL_26__DEFAULT_VAL_ATTR1_MASK 0x600000
+#define SPI_PS_INPUT_CNTL_26__DEFAULT_VAL_ATTR1__SHIFT 0x15
+#define SPI_PS_INPUT_CNTL_26__ATTR0_VALID_MASK 0x1000000
+#define SPI_PS_INPUT_CNTL_26__ATTR0_VALID__SHIFT 0x18
+#define SPI_PS_INPUT_CNTL_26__ATTR1_VALID_MASK 0x2000000
+#define SPI_PS_INPUT_CNTL_26__ATTR1_VALID__SHIFT 0x19
+#define SPI_PS_INPUT_CNTL_27__OFFSET_MASK 0x3f
+#define SPI_PS_INPUT_CNTL_27__OFFSET__SHIFT 0x0
+#define SPI_PS_INPUT_CNTL_27__DEFAULT_VAL_MASK 0x300
+#define SPI_PS_INPUT_CNTL_27__DEFAULT_VAL__SHIFT 0x8
+#define SPI_PS_INPUT_CNTL_27__FLAT_SHADE_MASK 0x400
+#define SPI_PS_INPUT_CNTL_27__FLAT_SHADE__SHIFT 0xa
+#define SPI_PS_INPUT_CNTL_27__DUP_MASK 0x40000
+#define SPI_PS_INPUT_CNTL_27__DUP__SHIFT 0x12
+#define SPI_PS_INPUT_CNTL_27__FP16_INTERP_MODE_MASK 0x80000
+#define SPI_PS_INPUT_CNTL_27__FP16_INTERP_MODE__SHIFT 0x13
+#define SPI_PS_INPUT_CNTL_27__USE_DEFAULT_ATTR1_MASK 0x100000
+#define SPI_PS_INPUT_CNTL_27__USE_DEFAULT_ATTR1__SHIFT 0x14
+#define SPI_PS_INPUT_CNTL_27__DEFAULT_VAL_ATTR1_MASK 0x600000
+#define SPI_PS_INPUT_CNTL_27__DEFAULT_VAL_ATTR1__SHIFT 0x15
+#define SPI_PS_INPUT_CNTL_27__ATTR0_VALID_MASK 0x1000000
+#define SPI_PS_INPUT_CNTL_27__ATTR0_VALID__SHIFT 0x18
+#define SPI_PS_INPUT_CNTL_27__ATTR1_VALID_MASK 0x2000000
+#define SPI_PS_INPUT_CNTL_27__ATTR1_VALID__SHIFT 0x19
+#define SPI_PS_INPUT_CNTL_28__OFFSET_MASK 0x3f
+#define SPI_PS_INPUT_CNTL_28__OFFSET__SHIFT 0x0
+#define SPI_PS_INPUT_CNTL_28__DEFAULT_VAL_MASK 0x300
+#define SPI_PS_INPUT_CNTL_28__DEFAULT_VAL__SHIFT 0x8
+#define SPI_PS_INPUT_CNTL_28__FLAT_SHADE_MASK 0x400
+#define SPI_PS_INPUT_CNTL_28__FLAT_SHADE__SHIFT 0xa
+#define SPI_PS_INPUT_CNTL_28__DUP_MASK 0x40000
+#define SPI_PS_INPUT_CNTL_28__DUP__SHIFT 0x12
+#define SPI_PS_INPUT_CNTL_28__FP16_INTERP_MODE_MASK 0x80000
+#define SPI_PS_INPUT_CNTL_28__FP16_INTERP_MODE__SHIFT 0x13
+#define SPI_PS_INPUT_CNTL_28__USE_DEFAULT_ATTR1_MASK 0x100000
+#define SPI_PS_INPUT_CNTL_28__USE_DEFAULT_ATTR1__SHIFT 0x14
+#define SPI_PS_INPUT_CNTL_28__DEFAULT_VAL_ATTR1_MASK 0x600000
+#define SPI_PS_INPUT_CNTL_28__DEFAULT_VAL_ATTR1__SHIFT 0x15
+#define SPI_PS_INPUT_CNTL_28__ATTR0_VALID_MASK 0x1000000
+#define SPI_PS_INPUT_CNTL_28__ATTR0_VALID__SHIFT 0x18
+#define SPI_PS_INPUT_CNTL_28__ATTR1_VALID_MASK 0x2000000
+#define SPI_PS_INPUT_CNTL_28__ATTR1_VALID__SHIFT 0x19
+#define SPI_PS_INPUT_CNTL_29__OFFSET_MASK 0x3f
+#define SPI_PS_INPUT_CNTL_29__OFFSET__SHIFT 0x0
+#define SPI_PS_INPUT_CNTL_29__DEFAULT_VAL_MASK 0x300
+#define SPI_PS_INPUT_CNTL_29__DEFAULT_VAL__SHIFT 0x8
+#define SPI_PS_INPUT_CNTL_29__FLAT_SHADE_MASK 0x400
+#define SPI_PS_INPUT_CNTL_29__FLAT_SHADE__SHIFT 0xa
+#define SPI_PS_INPUT_CNTL_29__DUP_MASK 0x40000
+#define SPI_PS_INPUT_CNTL_29__DUP__SHIFT 0x12
+#define SPI_PS_INPUT_CNTL_29__FP16_INTERP_MODE_MASK 0x80000
+#define SPI_PS_INPUT_CNTL_29__FP16_INTERP_MODE__SHIFT 0x13
+#define SPI_PS_INPUT_CNTL_29__USE_DEFAULT_ATTR1_MASK 0x100000
+#define SPI_PS_INPUT_CNTL_29__USE_DEFAULT_ATTR1__SHIFT 0x14
+#define SPI_PS_INPUT_CNTL_29__DEFAULT_VAL_ATTR1_MASK 0x600000
+#define SPI_PS_INPUT_CNTL_29__DEFAULT_VAL_ATTR1__SHIFT 0x15
+#define SPI_PS_INPUT_CNTL_29__ATTR0_VALID_MASK 0x1000000
+#define SPI_PS_INPUT_CNTL_29__ATTR0_VALID__SHIFT 0x18
+#define SPI_PS_INPUT_CNTL_29__ATTR1_VALID_MASK 0x2000000
+#define SPI_PS_INPUT_CNTL_29__ATTR1_VALID__SHIFT 0x19
+#define SPI_PS_INPUT_CNTL_30__OFFSET_MASK 0x3f
+#define SPI_PS_INPUT_CNTL_30__OFFSET__SHIFT 0x0
+#define SPI_PS_INPUT_CNTL_30__DEFAULT_VAL_MASK 0x300
+#define SPI_PS_INPUT_CNTL_30__DEFAULT_VAL__SHIFT 0x8
+#define SPI_PS_INPUT_CNTL_30__FLAT_SHADE_MASK 0x400
+#define SPI_PS_INPUT_CNTL_30__FLAT_SHADE__SHIFT 0xa
+#define SPI_PS_INPUT_CNTL_30__DUP_MASK 0x40000
+#define SPI_PS_INPUT_CNTL_30__DUP__SHIFT 0x12
+#define SPI_PS_INPUT_CNTL_30__FP16_INTERP_MODE_MASK 0x80000
+#define SPI_PS_INPUT_CNTL_30__FP16_INTERP_MODE__SHIFT 0x13
+#define SPI_PS_INPUT_CNTL_30__USE_DEFAULT_ATTR1_MASK 0x100000
+#define SPI_PS_INPUT_CNTL_30__USE_DEFAULT_ATTR1__SHIFT 0x14
+#define SPI_PS_INPUT_CNTL_30__DEFAULT_VAL_ATTR1_MASK 0x600000
+#define SPI_PS_INPUT_CNTL_30__DEFAULT_VAL_ATTR1__SHIFT 0x15
+#define SPI_PS_INPUT_CNTL_30__ATTR0_VALID_MASK 0x1000000
+#define SPI_PS_INPUT_CNTL_30__ATTR0_VALID__SHIFT 0x18
+#define SPI_PS_INPUT_CNTL_30__ATTR1_VALID_MASK 0x2000000
+#define SPI_PS_INPUT_CNTL_30__ATTR1_VALID__SHIFT 0x19
+#define SPI_PS_INPUT_CNTL_31__OFFSET_MASK 0x3f
+#define SPI_PS_INPUT_CNTL_31__OFFSET__SHIFT 0x0
+#define SPI_PS_INPUT_CNTL_31__DEFAULT_VAL_MASK 0x300
+#define SPI_PS_INPUT_CNTL_31__DEFAULT_VAL__SHIFT 0x8
+#define SPI_PS_INPUT_CNTL_31__FLAT_SHADE_MASK 0x400
+#define SPI_PS_INPUT_CNTL_31__FLAT_SHADE__SHIFT 0xa
+#define SPI_PS_INPUT_CNTL_31__DUP_MASK 0x40000
+#define SPI_PS_INPUT_CNTL_31__DUP__SHIFT 0x12
+#define SPI_PS_INPUT_CNTL_31__FP16_INTERP_MODE_MASK 0x80000
+#define SPI_PS_INPUT_CNTL_31__FP16_INTERP_MODE__SHIFT 0x13
+#define SPI_PS_INPUT_CNTL_31__USE_DEFAULT_ATTR1_MASK 0x100000
+#define SPI_PS_INPUT_CNTL_31__USE_DEFAULT_ATTR1__SHIFT 0x14
+#define SPI_PS_INPUT_CNTL_31__DEFAULT_VAL_ATTR1_MASK 0x600000
+#define SPI_PS_INPUT_CNTL_31__DEFAULT_VAL_ATTR1__SHIFT 0x15
+#define SPI_PS_INPUT_CNTL_31__ATTR0_VALID_MASK 0x1000000
+#define SPI_PS_INPUT_CNTL_31__ATTR0_VALID__SHIFT 0x18
+#define SPI_PS_INPUT_CNTL_31__ATTR1_VALID_MASK 0x2000000
+#define SPI_PS_INPUT_CNTL_31__ATTR1_VALID__SHIFT 0x19
+#define SPI_VS_OUT_CONFIG__VS_EXPORT_COUNT_MASK 0x3e
+#define SPI_VS_OUT_CONFIG__VS_EXPORT_COUNT__SHIFT 0x1
+#define SPI_VS_OUT_CONFIG__VS_HALF_PACK_MASK 0x40
+#define SPI_VS_OUT_CONFIG__VS_HALF_PACK__SHIFT 0x6
+#define SPI_PS_INPUT_ENA__PERSP_SAMPLE_ENA_MASK 0x1
+#define SPI_PS_INPUT_ENA__PERSP_SAMPLE_ENA__SHIFT 0x0
+#define SPI_PS_INPUT_ENA__PERSP_CENTER_ENA_MASK 0x2
+#define SPI_PS_INPUT_ENA__PERSP_CENTER_ENA__SHIFT 0x1
+#define SPI_PS_INPUT_ENA__PERSP_CENTROID_ENA_MASK 0x4
+#define SPI_PS_INPUT_ENA__PERSP_CENTROID_ENA__SHIFT 0x2
+#define SPI_PS_INPUT_ENA__PERSP_PULL_MODEL_ENA_MASK 0x8
+#define SPI_PS_INPUT_ENA__PERSP_PULL_MODEL_ENA__SHIFT 0x3
+#define SPI_PS_INPUT_ENA__LINEAR_SAMPLE_ENA_MASK 0x10
+#define SPI_PS_INPUT_ENA__LINEAR_SAMPLE_ENA__SHIFT 0x4
+#define SPI_PS_INPUT_ENA__LINEAR_CENTER_ENA_MASK 0x20
+#define SPI_PS_INPUT_ENA__LINEAR_CENTER_ENA__SHIFT 0x5
+#define SPI_PS_INPUT_ENA__LINEAR_CENTROID_ENA_MASK 0x40
+#define SPI_PS_INPUT_ENA__LINEAR_CENTROID_ENA__SHIFT 0x6
+#define SPI_PS_INPUT_ENA__LINE_STIPPLE_TEX_ENA_MASK 0x80
+#define SPI_PS_INPUT_ENA__LINE_STIPPLE_TEX_ENA__SHIFT 0x7
+#define SPI_PS_INPUT_ENA__POS_X_FLOAT_ENA_MASK 0x100
+#define SPI_PS_INPUT_ENA__POS_X_FLOAT_ENA__SHIFT 0x8
+#define SPI_PS_INPUT_ENA__POS_Y_FLOAT_ENA_MASK 0x200
+#define SPI_PS_INPUT_ENA__POS_Y_FLOAT_ENA__SHIFT 0x9
+#define SPI_PS_INPUT_ENA__POS_Z_FLOAT_ENA_MASK 0x400
+#define SPI_PS_INPUT_ENA__POS_Z_FLOAT_ENA__SHIFT 0xa
+#define SPI_PS_INPUT_ENA__POS_W_FLOAT_ENA_MASK 0x800
+#define SPI_PS_INPUT_ENA__POS_W_FLOAT_ENA__SHIFT 0xb
+#define SPI_PS_INPUT_ENA__FRONT_FACE_ENA_MASK 0x1000
+#define SPI_PS_INPUT_ENA__FRONT_FACE_ENA__SHIFT 0xc
+#define SPI_PS_INPUT_ENA__ANCILLARY_ENA_MASK 0x2000
+#define SPI_PS_INPUT_ENA__ANCILLARY_ENA__SHIFT 0xd
+#define SPI_PS_INPUT_ENA__SAMPLE_COVERAGE_ENA_MASK 0x4000
+#define SPI_PS_INPUT_ENA__SAMPLE_COVERAGE_ENA__SHIFT 0xe
+#define SPI_PS_INPUT_ENA__POS_FIXED_PT_ENA_MASK 0x8000
+#define SPI_PS_INPUT_ENA__POS_FIXED_PT_ENA__SHIFT 0xf
+#define SPI_PS_INPUT_ADDR__PERSP_SAMPLE_ENA_MASK 0x1
+#define SPI_PS_INPUT_ADDR__PERSP_SAMPLE_ENA__SHIFT 0x0
+#define SPI_PS_INPUT_ADDR__PERSP_CENTER_ENA_MASK 0x2
+#define SPI_PS_INPUT_ADDR__PERSP_CENTER_ENA__SHIFT 0x1
+#define SPI_PS_INPUT_ADDR__PERSP_CENTROID_ENA_MASK 0x4
+#define SPI_PS_INPUT_ADDR__PERSP_CENTROID_ENA__SHIFT 0x2
+#define SPI_PS_INPUT_ADDR__PERSP_PULL_MODEL_ENA_MASK 0x8
+#define SPI_PS_INPUT_ADDR__PERSP_PULL_MODEL_ENA__SHIFT 0x3
+#define SPI_PS_INPUT_ADDR__LINEAR_SAMPLE_ENA_MASK 0x10
+#define SPI_PS_INPUT_ADDR__LINEAR_SAMPLE_ENA__SHIFT 0x4
+#define SPI_PS_INPUT_ADDR__LINEAR_CENTER_ENA_MASK 0x20
+#define SPI_PS_INPUT_ADDR__LINEAR_CENTER_ENA__SHIFT 0x5
+#define SPI_PS_INPUT_ADDR__LINEAR_CENTROID_ENA_MASK 0x40
+#define SPI_PS_INPUT_ADDR__LINEAR_CENTROID_ENA__SHIFT 0x6
+#define SPI_PS_INPUT_ADDR__LINE_STIPPLE_TEX_ENA_MASK 0x80
+#define SPI_PS_INPUT_ADDR__LINE_STIPPLE_TEX_ENA__SHIFT 0x7
+#define SPI_PS_INPUT_ADDR__POS_X_FLOAT_ENA_MASK 0x100
+#define SPI_PS_INPUT_ADDR__POS_X_FLOAT_ENA__SHIFT 0x8
+#define SPI_PS_INPUT_ADDR__POS_Y_FLOAT_ENA_MASK 0x200
+#define SPI_PS_INPUT_ADDR__POS_Y_FLOAT_ENA__SHIFT 0x9
+#define SPI_PS_INPUT_ADDR__POS_Z_FLOAT_ENA_MASK 0x400
+#define SPI_PS_INPUT_ADDR__POS_Z_FLOAT_ENA__SHIFT 0xa
+#define SPI_PS_INPUT_ADDR__POS_W_FLOAT_ENA_MASK 0x800
+#define SPI_PS_INPUT_ADDR__POS_W_FLOAT_ENA__SHIFT 0xb
+#define SPI_PS_INPUT_ADDR__FRONT_FACE_ENA_MASK 0x1000
+#define SPI_PS_INPUT_ADDR__FRONT_FACE_ENA__SHIFT 0xc
+#define SPI_PS_INPUT_ADDR__ANCILLARY_ENA_MASK 0x2000
+#define SPI_PS_INPUT_ADDR__ANCILLARY_ENA__SHIFT 0xd
+#define SPI_PS_INPUT_ADDR__SAMPLE_COVERAGE_ENA_MASK 0x4000
+#define SPI_PS_INPUT_ADDR__SAMPLE_COVERAGE_ENA__SHIFT 0xe
+#define SPI_PS_INPUT_ADDR__POS_FIXED_PT_ENA_MASK 0x8000
+#define SPI_PS_INPUT_ADDR__POS_FIXED_PT_ENA__SHIFT 0xf
+#define SPI_INTERP_CONTROL_0__FLAT_SHADE_ENA_MASK 0x1
+#define SPI_INTERP_CONTROL_0__FLAT_SHADE_ENA__SHIFT 0x0
+#define SPI_INTERP_CONTROL_0__PNT_SPRITE_ENA_MASK 0x2
+#define SPI_INTERP_CONTROL_0__PNT_SPRITE_ENA__SHIFT 0x1
+#define SPI_INTERP_CONTROL_0__PNT_SPRITE_OVRD_X_MASK 0x1c
+#define SPI_INTERP_CONTROL_0__PNT_SPRITE_OVRD_X__SHIFT 0x2
+#define SPI_INTERP_CONTROL_0__PNT_SPRITE_OVRD_Y_MASK 0xe0
+#define SPI_INTERP_CONTROL_0__PNT_SPRITE_OVRD_Y__SHIFT 0x5
+#define SPI_INTERP_CONTROL_0__PNT_SPRITE_OVRD_Z_MASK 0x700
+#define SPI_INTERP_CONTROL_0__PNT_SPRITE_OVRD_Z__SHIFT 0x8
+#define SPI_INTERP_CONTROL_0__PNT_SPRITE_OVRD_W_MASK 0x3800
+#define SPI_INTERP_CONTROL_0__PNT_SPRITE_OVRD_W__SHIFT 0xb
+#define SPI_INTERP_CONTROL_0__PNT_SPRITE_TOP_1_MASK 0x4000
+#define SPI_INTERP_CONTROL_0__PNT_SPRITE_TOP_1__SHIFT 0xe
+#define SPI_PS_IN_CONTROL__NUM_INTERP_MASK 0x3f
+#define SPI_PS_IN_CONTROL__NUM_INTERP__SHIFT 0x0
+#define SPI_PS_IN_CONTROL__PARAM_GEN_MASK 0x40
+#define SPI_PS_IN_CONTROL__PARAM_GEN__SHIFT 0x6
+#define SPI_PS_IN_CONTROL__BC_OPTIMIZE_DISABLE_MASK 0x4000
+#define SPI_PS_IN_CONTROL__BC_OPTIMIZE_DISABLE__SHIFT 0xe
+#define SPI_BARYC_CNTL__PERSP_CENTER_CNTL_MASK 0x1
+#define SPI_BARYC_CNTL__PERSP_CENTER_CNTL__SHIFT 0x0
+#define SPI_BARYC_CNTL__PERSP_CENTROID_CNTL_MASK 0x10
+#define SPI_BARYC_CNTL__PERSP_CENTROID_CNTL__SHIFT 0x4
+#define SPI_BARYC_CNTL__LINEAR_CENTER_CNTL_MASK 0x100
+#define SPI_BARYC_CNTL__LINEAR_CENTER_CNTL__SHIFT 0x8
+#define SPI_BARYC_CNTL__LINEAR_CENTROID_CNTL_MASK 0x1000
+#define SPI_BARYC_CNTL__LINEAR_CENTROID_CNTL__SHIFT 0xc
+#define SPI_BARYC_CNTL__POS_FLOAT_LOCATION_MASK 0x30000
+#define SPI_BARYC_CNTL__POS_FLOAT_LOCATION__SHIFT 0x10
+#define SPI_BARYC_CNTL__POS_FLOAT_ULC_MASK 0x100000
+#define SPI_BARYC_CNTL__POS_FLOAT_ULC__SHIFT 0x14
+#define SPI_BARYC_CNTL__FRONT_FACE_ALL_BITS_MASK 0x1000000
+#define SPI_BARYC_CNTL__FRONT_FACE_ALL_BITS__SHIFT 0x18
+#define SPI_TMPRING_SIZE__WAVES_MASK 0xfff
+#define SPI_TMPRING_SIZE__WAVES__SHIFT 0x0
+#define SPI_TMPRING_SIZE__WAVESIZE_MASK 0x1fff000
+#define SPI_TMPRING_SIZE__WAVESIZE__SHIFT 0xc
+#define SPI_SHADER_POS_FORMAT__POS0_EXPORT_FORMAT_MASK 0xf
+#define SPI_SHADER_POS_FORMAT__POS0_EXPORT_FORMAT__SHIFT 0x0
+#define SPI_SHADER_POS_FORMAT__POS1_EXPORT_FORMAT_MASK 0xf0
+#define SPI_SHADER_POS_FORMAT__POS1_EXPORT_FORMAT__SHIFT 0x4
+#define SPI_SHADER_POS_FORMAT__POS2_EXPORT_FORMAT_MASK 0xf00
+#define SPI_SHADER_POS_FORMAT__POS2_EXPORT_FORMAT__SHIFT 0x8
+#define SPI_SHADER_POS_FORMAT__POS3_EXPORT_FORMAT_MASK 0xf000
+#define SPI_SHADER_POS_FORMAT__POS3_EXPORT_FORMAT__SHIFT 0xc
+#define SPI_SHADER_Z_FORMAT__Z_EXPORT_FORMAT_MASK 0xf
+#define SPI_SHADER_Z_FORMAT__Z_EXPORT_FORMAT__SHIFT 0x0
+#define SPI_SHADER_COL_FORMAT__COL0_EXPORT_FORMAT_MASK 0xf
+#define SPI_SHADER_COL_FORMAT__COL0_EXPORT_FORMAT__SHIFT 0x0
+#define SPI_SHADER_COL_FORMAT__COL1_EXPORT_FORMAT_MASK 0xf0
+#define SPI_SHADER_COL_FORMAT__COL1_EXPORT_FORMAT__SHIFT 0x4
+#define SPI_SHADER_COL_FORMAT__COL2_EXPORT_FORMAT_MASK 0xf00
+#define SPI_SHADER_COL_FORMAT__COL2_EXPORT_FORMAT__SHIFT 0x8
+#define SPI_SHADER_COL_FORMAT__COL3_EXPORT_FORMAT_MASK 0xf000
+#define SPI_SHADER_COL_FORMAT__COL3_EXPORT_FORMAT__SHIFT 0xc
+#define SPI_SHADER_COL_FORMAT__COL4_EXPORT_FORMAT_MASK 0xf0000
+#define SPI_SHADER_COL_FORMAT__COL4_EXPORT_FORMAT__SHIFT 0x10
+#define SPI_SHADER_COL_FORMAT__COL5_EXPORT_FORMAT_MASK 0xf00000
+#define SPI_SHADER_COL_FORMAT__COL5_EXPORT_FORMAT__SHIFT 0x14
+#define SPI_SHADER_COL_FORMAT__COL6_EXPORT_FORMAT_MASK 0xf000000
+#define SPI_SHADER_COL_FORMAT__COL6_EXPORT_FORMAT__SHIFT 0x18
+#define SPI_SHADER_COL_FORMAT__COL7_EXPORT_FORMAT_MASK 0xf0000000
+#define SPI_SHADER_COL_FORMAT__COL7_EXPORT_FORMAT__SHIFT 0x1c
+#define SPI_ARB_PRIORITY__PIPE_ORDER_TS0_MASK 0x7
+#define SPI_ARB_PRIORITY__PIPE_ORDER_TS0__SHIFT 0x0
+#define SPI_ARB_PRIORITY__PIPE_ORDER_TS1_MASK 0x38
+#define SPI_ARB_PRIORITY__PIPE_ORDER_TS1__SHIFT 0x3
+#define SPI_ARB_PRIORITY__PIPE_ORDER_TS2_MASK 0x1c0
+#define SPI_ARB_PRIORITY__PIPE_ORDER_TS2__SHIFT 0x6
+#define SPI_ARB_PRIORITY__PIPE_ORDER_TS3_MASK 0xe00
+#define SPI_ARB_PRIORITY__PIPE_ORDER_TS3__SHIFT 0x9
+#define SPI_ARB_PRIORITY__TS0_DUR_MULT_MASK 0x3000
+#define SPI_ARB_PRIORITY__TS0_DUR_MULT__SHIFT 0xc
+#define SPI_ARB_PRIORITY__TS1_DUR_MULT_MASK 0xc000
+#define SPI_ARB_PRIORITY__TS1_DUR_MULT__SHIFT 0xe
+#define SPI_ARB_PRIORITY__TS2_DUR_MULT_MASK 0x30000
+#define SPI_ARB_PRIORITY__TS2_DUR_MULT__SHIFT 0x10
+#define SPI_ARB_PRIORITY__TS3_DUR_MULT_MASK 0xc0000
+#define SPI_ARB_PRIORITY__TS3_DUR_MULT__SHIFT 0x12
+#define SPI_ARB_CYCLES_0__TS0_DURATION_MASK 0xffff
+#define SPI_ARB_CYCLES_0__TS0_DURATION__SHIFT 0x0
+#define SPI_ARB_CYCLES_0__TS1_DURATION_MASK 0xffff0000
+#define SPI_ARB_CYCLES_0__TS1_DURATION__SHIFT 0x10
+#define SPI_ARB_CYCLES_1__TS2_DURATION_MASK 0xffff
+#define SPI_ARB_CYCLES_1__TS2_DURATION__SHIFT 0x0
+#define SPI_ARB_CYCLES_1__TS3_DURATION_MASK 0xffff0000
+#define SPI_ARB_CYCLES_1__TS3_DURATION__SHIFT 0x10
+#define SPI_CDBG_SYS_GFX__PS_EN_MASK 0x1
+#define SPI_CDBG_SYS_GFX__PS_EN__SHIFT 0x0
+#define SPI_CDBG_SYS_GFX__VS_EN_MASK 0x2
+#define SPI_CDBG_SYS_GFX__VS_EN__SHIFT 0x1
+#define SPI_CDBG_SYS_GFX__GS_EN_MASK 0x4
+#define SPI_CDBG_SYS_GFX__GS_EN__SHIFT 0x2
+#define SPI_CDBG_SYS_GFX__ES_EN_MASK 0x8
+#define SPI_CDBG_SYS_GFX__ES_EN__SHIFT 0x3
+#define SPI_CDBG_SYS_GFX__HS_EN_MASK 0x10
+#define SPI_CDBG_SYS_GFX__HS_EN__SHIFT 0x4
+#define SPI_CDBG_SYS_GFX__LS_EN_MASK 0x20
+#define SPI_CDBG_SYS_GFX__LS_EN__SHIFT 0x5
+#define SPI_CDBG_SYS_GFX__CS_EN_MASK 0x40
+#define SPI_CDBG_SYS_GFX__CS_EN__SHIFT 0x6
+#define SPI_CDBG_SYS_HP3D__PS_EN_MASK 0x1
+#define SPI_CDBG_SYS_HP3D__PS_EN__SHIFT 0x0
+#define SPI_CDBG_SYS_HP3D__VS_EN_MASK 0x2
+#define SPI_CDBG_SYS_HP3D__VS_EN__SHIFT 0x1
+#define SPI_CDBG_SYS_HP3D__GS_EN_MASK 0x4
+#define SPI_CDBG_SYS_HP3D__GS_EN__SHIFT 0x2
+#define SPI_CDBG_SYS_HP3D__ES_EN_MASK 0x8
+#define SPI_CDBG_SYS_HP3D__ES_EN__SHIFT 0x3
+#define SPI_CDBG_SYS_HP3D__HS_EN_MASK 0x10
+#define SPI_CDBG_SYS_HP3D__HS_EN__SHIFT 0x4
+#define SPI_CDBG_SYS_HP3D__LS_EN_MASK 0x20
+#define SPI_CDBG_SYS_HP3D__LS_EN__SHIFT 0x5
+#define SPI_CDBG_SYS_CS0__PIPE0_MASK 0xff
+#define SPI_CDBG_SYS_CS0__PIPE0__SHIFT 0x0
+#define SPI_CDBG_SYS_CS0__PIPE1_MASK 0xff00
+#define SPI_CDBG_SYS_CS0__PIPE1__SHIFT 0x8
+#define SPI_CDBG_SYS_CS0__PIPE2_MASK 0xff0000
+#define SPI_CDBG_SYS_CS0__PIPE2__SHIFT 0x10
+#define SPI_CDBG_SYS_CS0__PIPE3_MASK 0xff000000
+#define SPI_CDBG_SYS_CS0__PIPE3__SHIFT 0x18
+#define SPI_CDBG_SYS_CS1__PIPE0_MASK 0xff
+#define SPI_CDBG_SYS_CS1__PIPE0__SHIFT 0x0
+#define SPI_CDBG_SYS_CS1__PIPE1_MASK 0xff00
+#define SPI_CDBG_SYS_CS1__PIPE1__SHIFT 0x8
+#define SPI_CDBG_SYS_CS1__PIPE2_MASK 0xff0000
+#define SPI_CDBG_SYS_CS1__PIPE2__SHIFT 0x10
+#define SPI_CDBG_SYS_CS1__PIPE3_MASK 0xff000000
+#define SPI_CDBG_SYS_CS1__PIPE3__SHIFT 0x18
+#define SPI_WCL_PIPE_PERCENT_GFX__VALUE_MASK 0x7f
+#define SPI_WCL_PIPE_PERCENT_GFX__VALUE__SHIFT 0x0
+#define SPI_WCL_PIPE_PERCENT_GFX__LS_GRP_VALUE_MASK 0xf80
+#define SPI_WCL_PIPE_PERCENT_GFX__LS_GRP_VALUE__SHIFT 0x7
+#define SPI_WCL_PIPE_PERCENT_GFX__HS_GRP_VALUE_MASK 0x1f000
+#define SPI_WCL_PIPE_PERCENT_GFX__HS_GRP_VALUE__SHIFT 0xc
+#define SPI_WCL_PIPE_PERCENT_GFX__ES_GRP_VALUE_MASK 0x3e0000
+#define SPI_WCL_PIPE_PERCENT_GFX__ES_GRP_VALUE__SHIFT 0x11
+#define SPI_WCL_PIPE_PERCENT_GFX__GS_GRP_VALUE_MASK 0x7c00000
+#define SPI_WCL_PIPE_PERCENT_GFX__GS_GRP_VALUE__SHIFT 0x16
+#define SPI_WCL_PIPE_PERCENT_HP3D__VALUE_MASK 0x7f
+#define SPI_WCL_PIPE_PERCENT_HP3D__VALUE__SHIFT 0x0
+#define SPI_WCL_PIPE_PERCENT_HP3D__LS_GRP_VALUE_MASK 0xf80
+#define SPI_WCL_PIPE_PERCENT_HP3D__LS_GRP_VALUE__SHIFT 0x7
+#define SPI_WCL_PIPE_PERCENT_HP3D__HS_GRP_VALUE_MASK 0x1f000
+#define SPI_WCL_PIPE_PERCENT_HP3D__HS_GRP_VALUE__SHIFT 0xc
+#define SPI_WCL_PIPE_PERCENT_HP3D__ES_GRP_VALUE_MASK 0x3e0000
+#define SPI_WCL_PIPE_PERCENT_HP3D__ES_GRP_VALUE__SHIFT 0x11
+#define SPI_WCL_PIPE_PERCENT_HP3D__GS_GRP_VALUE_MASK 0x7c00000
+#define SPI_WCL_PIPE_PERCENT_HP3D__GS_GRP_VALUE__SHIFT 0x16
+#define SPI_WCL_PIPE_PERCENT_CS0__VALUE_MASK 0x7f
+#define SPI_WCL_PIPE_PERCENT_CS0__VALUE__SHIFT 0x0
+#define SPI_WCL_PIPE_PERCENT_CS1__VALUE_MASK 0x7f
+#define SPI_WCL_PIPE_PERCENT_CS1__VALUE__SHIFT 0x0
+#define SPI_WCL_PIPE_PERCENT_CS2__VALUE_MASK 0x7f
+#define SPI_WCL_PIPE_PERCENT_CS2__VALUE__SHIFT 0x0
+#define SPI_WCL_PIPE_PERCENT_CS3__VALUE_MASK 0x7f
+#define SPI_WCL_PIPE_PERCENT_CS3__VALUE__SHIFT 0x0
+#define SPI_WCL_PIPE_PERCENT_CS4__VALUE_MASK 0x7f
+#define SPI_WCL_PIPE_PERCENT_CS4__VALUE__SHIFT 0x0
+#define SPI_WCL_PIPE_PERCENT_CS5__VALUE_MASK 0x7f
+#define SPI_WCL_PIPE_PERCENT_CS5__VALUE__SHIFT 0x0
+#define SPI_WCL_PIPE_PERCENT_CS6__VALUE_MASK 0x7f
+#define SPI_WCL_PIPE_PERCENT_CS6__VALUE__SHIFT 0x0
+#define SPI_WCL_PIPE_PERCENT_CS7__VALUE_MASK 0x7f
+#define SPI_WCL_PIPE_PERCENT_CS7__VALUE__SHIFT 0x0
+#define SPI_GDBG_WAVE_CNTL__STALL_RA_MASK 0x1
+#define SPI_GDBG_WAVE_CNTL__STALL_RA__SHIFT 0x0
+#define SPI_GDBG_WAVE_CNTL__STALL_VMID_MASK 0x1fffe
+#define SPI_GDBG_WAVE_CNTL__STALL_VMID__SHIFT 0x1
+#define SPI_GDBG_TRAP_CONFIG__ME_SEL_MASK 0x3
+#define SPI_GDBG_TRAP_CONFIG__ME_SEL__SHIFT 0x0
+#define SPI_GDBG_TRAP_CONFIG__PIPE_SEL_MASK 0xc
+#define SPI_GDBG_TRAP_CONFIG__PIPE_SEL__SHIFT 0x2
+#define SPI_GDBG_TRAP_CONFIG__QUEUE_SEL_MASK 0x70
+#define SPI_GDBG_TRAP_CONFIG__QUEUE_SEL__SHIFT 0x4
+#define SPI_GDBG_TRAP_CONFIG__ME_MATCH_MASK 0x80
+#define SPI_GDBG_TRAP_CONFIG__ME_MATCH__SHIFT 0x7
+#define SPI_GDBG_TRAP_CONFIG__PIPE_MATCH_MASK 0x100
+#define SPI_GDBG_TRAP_CONFIG__PIPE_MATCH__SHIFT 0x8
+#define SPI_GDBG_TRAP_CONFIG__QUEUE_MATCH_MASK 0x200
+#define SPI_GDBG_TRAP_CONFIG__QUEUE_MATCH__SHIFT 0x9
+#define SPI_GDBG_TRAP_CONFIG__TRAP_EN_MASK 0x8000
+#define SPI_GDBG_TRAP_CONFIG__TRAP_EN__SHIFT 0xf
+#define SPI_GDBG_TRAP_CONFIG__VMID_SEL_MASK 0xffff0000
+#define SPI_GDBG_TRAP_CONFIG__VMID_SEL__SHIFT 0x10
+#define SPI_GDBG_TRAP_MASK__EXCP_EN_MASK 0x1ff
+#define SPI_GDBG_TRAP_MASK__EXCP_EN__SHIFT 0x0
+#define SPI_GDBG_TRAP_MASK__REPLACE_MASK 0x200
+#define SPI_GDBG_TRAP_MASK__REPLACE__SHIFT 0x9
+#define SPI_GDBG_TBA_LO__MEM_BASE_MASK 0xffffffff
+#define SPI_GDBG_TBA_LO__MEM_BASE__SHIFT 0x0
+#define SPI_GDBG_TBA_HI__MEM_BASE_MASK 0xff
+#define SPI_GDBG_TBA_HI__MEM_BASE__SHIFT 0x0
+#define SPI_GDBG_TMA_LO__MEM_BASE_MASK 0xffffffff
+#define SPI_GDBG_TMA_LO__MEM_BASE__SHIFT 0x0
+#define SPI_GDBG_TMA_HI__MEM_BASE_MASK 0xff
+#define SPI_GDBG_TMA_HI__MEM_BASE__SHIFT 0x0
+#define SPI_GDBG_TRAP_DATA0__DATA_MASK 0xffffffff
+#define SPI_GDBG_TRAP_DATA0__DATA__SHIFT 0x0
+#define SPI_GDBG_TRAP_DATA1__DATA_MASK 0xffffffff
+#define SPI_GDBG_TRAP_DATA1__DATA__SHIFT 0x0
+#define SPI_RESET_DEBUG__DISABLE_GFX_RESET_MASK 0x1
+#define SPI_RESET_DEBUG__DISABLE_GFX_RESET__SHIFT 0x0
+#define SPI_RESET_DEBUG__DISABLE_GFX_RESET_PER_VMID_MASK 0x2
+#define SPI_RESET_DEBUG__DISABLE_GFX_RESET_PER_VMID__SHIFT 0x1
+#define SPI_RESET_DEBUG__DISABLE_GFX_RESET_ALL_VMID_MASK 0x4
+#define SPI_RESET_DEBUG__DISABLE_GFX_RESET_ALL_VMID__SHIFT 0x2
+#define SPI_RESET_DEBUG__DISABLE_GFX_RESET_RESOURCE_MASK 0x8
+#define SPI_RESET_DEBUG__DISABLE_GFX_RESET_RESOURCE__SHIFT 0x3
+#define SPI_RESET_DEBUG__DISABLE_GFX_RESET_PRIORITY_MASK 0x10
+#define SPI_RESET_DEBUG__DISABLE_GFX_RESET_PRIORITY__SHIFT 0x4
+#define SPI_COMPUTE_QUEUE_RESET__RESET_MASK 0x1
+#define SPI_COMPUTE_QUEUE_RESET__RESET__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_CU_0__VGPR_MASK 0xf
+#define SPI_RESOURCE_RESERVE_CU_0__VGPR__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_CU_0__SGPR_MASK 0xf0
+#define SPI_RESOURCE_RESERVE_CU_0__SGPR__SHIFT 0x4
+#define SPI_RESOURCE_RESERVE_CU_0__LDS_MASK 0xf00
+#define SPI_RESOURCE_RESERVE_CU_0__LDS__SHIFT 0x8
+#define SPI_RESOURCE_RESERVE_CU_0__WAVES_MASK 0x7000
+#define SPI_RESOURCE_RESERVE_CU_0__WAVES__SHIFT 0xc
+#define SPI_RESOURCE_RESERVE_CU_0__BARRIERS_MASK 0x78000
+#define SPI_RESOURCE_RESERVE_CU_0__BARRIERS__SHIFT 0xf
+#define SPI_RESOURCE_RESERVE_CU_1__VGPR_MASK 0xf
+#define SPI_RESOURCE_RESERVE_CU_1__VGPR__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_CU_1__SGPR_MASK 0xf0
+#define SPI_RESOURCE_RESERVE_CU_1__SGPR__SHIFT 0x4
+#define SPI_RESOURCE_RESERVE_CU_1__LDS_MASK 0xf00
+#define SPI_RESOURCE_RESERVE_CU_1__LDS__SHIFT 0x8
+#define SPI_RESOURCE_RESERVE_CU_1__WAVES_MASK 0x7000
+#define SPI_RESOURCE_RESERVE_CU_1__WAVES__SHIFT 0xc
+#define SPI_RESOURCE_RESERVE_CU_1__BARRIERS_MASK 0x78000
+#define SPI_RESOURCE_RESERVE_CU_1__BARRIERS__SHIFT 0xf
+#define SPI_RESOURCE_RESERVE_CU_2__VGPR_MASK 0xf
+#define SPI_RESOURCE_RESERVE_CU_2__VGPR__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_CU_2__SGPR_MASK 0xf0
+#define SPI_RESOURCE_RESERVE_CU_2__SGPR__SHIFT 0x4
+#define SPI_RESOURCE_RESERVE_CU_2__LDS_MASK 0xf00
+#define SPI_RESOURCE_RESERVE_CU_2__LDS__SHIFT 0x8
+#define SPI_RESOURCE_RESERVE_CU_2__WAVES_MASK 0x7000
+#define SPI_RESOURCE_RESERVE_CU_2__WAVES__SHIFT 0xc
+#define SPI_RESOURCE_RESERVE_CU_2__BARRIERS_MASK 0x78000
+#define SPI_RESOURCE_RESERVE_CU_2__BARRIERS__SHIFT 0xf
+#define SPI_RESOURCE_RESERVE_CU_3__VGPR_MASK 0xf
+#define SPI_RESOURCE_RESERVE_CU_3__VGPR__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_CU_3__SGPR_MASK 0xf0
+#define SPI_RESOURCE_RESERVE_CU_3__SGPR__SHIFT 0x4
+#define SPI_RESOURCE_RESERVE_CU_3__LDS_MASK 0xf00
+#define SPI_RESOURCE_RESERVE_CU_3__LDS__SHIFT 0x8
+#define SPI_RESOURCE_RESERVE_CU_3__WAVES_MASK 0x7000
+#define SPI_RESOURCE_RESERVE_CU_3__WAVES__SHIFT 0xc
+#define SPI_RESOURCE_RESERVE_CU_3__BARRIERS_MASK 0x78000
+#define SPI_RESOURCE_RESERVE_CU_3__BARRIERS__SHIFT 0xf
+#define SPI_RESOURCE_RESERVE_CU_4__VGPR_MASK 0xf
+#define SPI_RESOURCE_RESERVE_CU_4__VGPR__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_CU_4__SGPR_MASK 0xf0
+#define SPI_RESOURCE_RESERVE_CU_4__SGPR__SHIFT 0x4
+#define SPI_RESOURCE_RESERVE_CU_4__LDS_MASK 0xf00
+#define SPI_RESOURCE_RESERVE_CU_4__LDS__SHIFT 0x8
+#define SPI_RESOURCE_RESERVE_CU_4__WAVES_MASK 0x7000
+#define SPI_RESOURCE_RESERVE_CU_4__WAVES__SHIFT 0xc
+#define SPI_RESOURCE_RESERVE_CU_4__BARRIERS_MASK 0x78000
+#define SPI_RESOURCE_RESERVE_CU_4__BARRIERS__SHIFT 0xf
+#define SPI_RESOURCE_RESERVE_CU_5__VGPR_MASK 0xf
+#define SPI_RESOURCE_RESERVE_CU_5__VGPR__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_CU_5__SGPR_MASK 0xf0
+#define SPI_RESOURCE_RESERVE_CU_5__SGPR__SHIFT 0x4
+#define SPI_RESOURCE_RESERVE_CU_5__LDS_MASK 0xf00
+#define SPI_RESOURCE_RESERVE_CU_5__LDS__SHIFT 0x8
+#define SPI_RESOURCE_RESERVE_CU_5__WAVES_MASK 0x7000
+#define SPI_RESOURCE_RESERVE_CU_5__WAVES__SHIFT 0xc
+#define SPI_RESOURCE_RESERVE_CU_5__BARRIERS_MASK 0x78000
+#define SPI_RESOURCE_RESERVE_CU_5__BARRIERS__SHIFT 0xf
+#define SPI_RESOURCE_RESERVE_CU_6__VGPR_MASK 0xf
+#define SPI_RESOURCE_RESERVE_CU_6__VGPR__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_CU_6__SGPR_MASK 0xf0
+#define SPI_RESOURCE_RESERVE_CU_6__SGPR__SHIFT 0x4
+#define SPI_RESOURCE_RESERVE_CU_6__LDS_MASK 0xf00
+#define SPI_RESOURCE_RESERVE_CU_6__LDS__SHIFT 0x8
+#define SPI_RESOURCE_RESERVE_CU_6__WAVES_MASK 0x7000
+#define SPI_RESOURCE_RESERVE_CU_6__WAVES__SHIFT 0xc
+#define SPI_RESOURCE_RESERVE_CU_6__BARRIERS_MASK 0x78000
+#define SPI_RESOURCE_RESERVE_CU_6__BARRIERS__SHIFT 0xf
+#define SPI_RESOURCE_RESERVE_CU_7__VGPR_MASK 0xf
+#define SPI_RESOURCE_RESERVE_CU_7__VGPR__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_CU_7__SGPR_MASK 0xf0
+#define SPI_RESOURCE_RESERVE_CU_7__SGPR__SHIFT 0x4
+#define SPI_RESOURCE_RESERVE_CU_7__LDS_MASK 0xf00
+#define SPI_RESOURCE_RESERVE_CU_7__LDS__SHIFT 0x8
+#define SPI_RESOURCE_RESERVE_CU_7__WAVES_MASK 0x7000
+#define SPI_RESOURCE_RESERVE_CU_7__WAVES__SHIFT 0xc
+#define SPI_RESOURCE_RESERVE_CU_7__BARRIERS_MASK 0x78000
+#define SPI_RESOURCE_RESERVE_CU_7__BARRIERS__SHIFT 0xf
+#define SPI_RESOURCE_RESERVE_CU_8__VGPR_MASK 0xf
+#define SPI_RESOURCE_RESERVE_CU_8__VGPR__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_CU_8__SGPR_MASK 0xf0
+#define SPI_RESOURCE_RESERVE_CU_8__SGPR__SHIFT 0x4
+#define SPI_RESOURCE_RESERVE_CU_8__LDS_MASK 0xf00
+#define SPI_RESOURCE_RESERVE_CU_8__LDS__SHIFT 0x8
+#define SPI_RESOURCE_RESERVE_CU_8__WAVES_MASK 0x7000
+#define SPI_RESOURCE_RESERVE_CU_8__WAVES__SHIFT 0xc
+#define SPI_RESOURCE_RESERVE_CU_8__BARRIERS_MASK 0x78000
+#define SPI_RESOURCE_RESERVE_CU_8__BARRIERS__SHIFT 0xf
+#define SPI_RESOURCE_RESERVE_CU_9__VGPR_MASK 0xf
+#define SPI_RESOURCE_RESERVE_CU_9__VGPR__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_CU_9__SGPR_MASK 0xf0
+#define SPI_RESOURCE_RESERVE_CU_9__SGPR__SHIFT 0x4
+#define SPI_RESOURCE_RESERVE_CU_9__LDS_MASK 0xf00
+#define SPI_RESOURCE_RESERVE_CU_9__LDS__SHIFT 0x8
+#define SPI_RESOURCE_RESERVE_CU_9__WAVES_MASK 0x7000
+#define SPI_RESOURCE_RESERVE_CU_9__WAVES__SHIFT 0xc
+#define SPI_RESOURCE_RESERVE_CU_9__BARRIERS_MASK 0x78000
+#define SPI_RESOURCE_RESERVE_CU_9__BARRIERS__SHIFT 0xf
+#define SPI_RESOURCE_RESERVE_CU_10__VGPR_MASK 0xf
+#define SPI_RESOURCE_RESERVE_CU_10__VGPR__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_CU_10__SGPR_MASK 0xf0
+#define SPI_RESOURCE_RESERVE_CU_10__SGPR__SHIFT 0x4
+#define SPI_RESOURCE_RESERVE_CU_10__LDS_MASK 0xf00
+#define SPI_RESOURCE_RESERVE_CU_10__LDS__SHIFT 0x8
+#define SPI_RESOURCE_RESERVE_CU_10__WAVES_MASK 0x7000
+#define SPI_RESOURCE_RESERVE_CU_10__WAVES__SHIFT 0xc
+#define SPI_RESOURCE_RESERVE_CU_10__BARRIERS_MASK 0x78000
+#define SPI_RESOURCE_RESERVE_CU_10__BARRIERS__SHIFT 0xf
+#define SPI_RESOURCE_RESERVE_CU_11__VGPR_MASK 0xf
+#define SPI_RESOURCE_RESERVE_CU_11__VGPR__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_CU_11__SGPR_MASK 0xf0
+#define SPI_RESOURCE_RESERVE_CU_11__SGPR__SHIFT 0x4
+#define SPI_RESOURCE_RESERVE_CU_11__LDS_MASK 0xf00
+#define SPI_RESOURCE_RESERVE_CU_11__LDS__SHIFT 0x8
+#define SPI_RESOURCE_RESERVE_CU_11__WAVES_MASK 0x7000
+#define SPI_RESOURCE_RESERVE_CU_11__WAVES__SHIFT 0xc
+#define SPI_RESOURCE_RESERVE_CU_11__BARRIERS_MASK 0x78000
+#define SPI_RESOURCE_RESERVE_CU_11__BARRIERS__SHIFT 0xf
+#define SPI_RESOURCE_RESERVE_CU_12__VGPR_MASK 0xf
+#define SPI_RESOURCE_RESERVE_CU_12__VGPR__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_CU_12__SGPR_MASK 0xf0
+#define SPI_RESOURCE_RESERVE_CU_12__SGPR__SHIFT 0x4
+#define SPI_RESOURCE_RESERVE_CU_12__LDS_MASK 0xf00
+#define SPI_RESOURCE_RESERVE_CU_12__LDS__SHIFT 0x8
+#define SPI_RESOURCE_RESERVE_CU_12__WAVES_MASK 0x7000
+#define SPI_RESOURCE_RESERVE_CU_12__WAVES__SHIFT 0xc
+#define SPI_RESOURCE_RESERVE_CU_12__BARRIERS_MASK 0x78000
+#define SPI_RESOURCE_RESERVE_CU_12__BARRIERS__SHIFT 0xf
+#define SPI_RESOURCE_RESERVE_CU_13__VGPR_MASK 0xf
+#define SPI_RESOURCE_RESERVE_CU_13__VGPR__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_CU_13__SGPR_MASK 0xf0
+#define SPI_RESOURCE_RESERVE_CU_13__SGPR__SHIFT 0x4
+#define SPI_RESOURCE_RESERVE_CU_13__LDS_MASK 0xf00
+#define SPI_RESOURCE_RESERVE_CU_13__LDS__SHIFT 0x8
+#define SPI_RESOURCE_RESERVE_CU_13__WAVES_MASK 0x7000
+#define SPI_RESOURCE_RESERVE_CU_13__WAVES__SHIFT 0xc
+#define SPI_RESOURCE_RESERVE_CU_13__BARRIERS_MASK 0x78000
+#define SPI_RESOURCE_RESERVE_CU_13__BARRIERS__SHIFT 0xf
+#define SPI_RESOURCE_RESERVE_CU_14__VGPR_MASK 0xf
+#define SPI_RESOURCE_RESERVE_CU_14__VGPR__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_CU_14__SGPR_MASK 0xf0
+#define SPI_RESOURCE_RESERVE_CU_14__SGPR__SHIFT 0x4
+#define SPI_RESOURCE_RESERVE_CU_14__LDS_MASK 0xf00
+#define SPI_RESOURCE_RESERVE_CU_14__LDS__SHIFT 0x8
+#define SPI_RESOURCE_RESERVE_CU_14__WAVES_MASK 0x7000
+#define SPI_RESOURCE_RESERVE_CU_14__WAVES__SHIFT 0xc
+#define SPI_RESOURCE_RESERVE_CU_14__BARRIERS_MASK 0x78000
+#define SPI_RESOURCE_RESERVE_CU_14__BARRIERS__SHIFT 0xf
+#define SPI_RESOURCE_RESERVE_CU_15__VGPR_MASK 0xf
+#define SPI_RESOURCE_RESERVE_CU_15__VGPR__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_CU_15__SGPR_MASK 0xf0
+#define SPI_RESOURCE_RESERVE_CU_15__SGPR__SHIFT 0x4
+#define SPI_RESOURCE_RESERVE_CU_15__LDS_MASK 0xf00
+#define SPI_RESOURCE_RESERVE_CU_15__LDS__SHIFT 0x8
+#define SPI_RESOURCE_RESERVE_CU_15__WAVES_MASK 0x7000
+#define SPI_RESOURCE_RESERVE_CU_15__WAVES__SHIFT 0xc
+#define SPI_RESOURCE_RESERVE_CU_15__BARRIERS_MASK 0x78000
+#define SPI_RESOURCE_RESERVE_CU_15__BARRIERS__SHIFT 0xf
+#define SPI_RESOURCE_RESERVE_EN_CU_0__EN_MASK 0x1
+#define SPI_RESOURCE_RESERVE_EN_CU_0__EN__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_EN_CU_0__TYPE_MASK_MASK 0xfffe
+#define SPI_RESOURCE_RESERVE_EN_CU_0__TYPE_MASK__SHIFT 0x1
+#define SPI_RESOURCE_RESERVE_EN_CU_0__QUEUE_MASK_MASK 0xff0000
+#define SPI_RESOURCE_RESERVE_EN_CU_0__QUEUE_MASK__SHIFT 0x10
+#define SPI_RESOURCE_RESERVE_EN_CU_0__RESERVE_SPACE_ONLY_MASK 0x1000000
+#define SPI_RESOURCE_RESERVE_EN_CU_0__RESERVE_SPACE_ONLY__SHIFT 0x18
+#define SPI_RESOURCE_RESERVE_EN_CU_1__EN_MASK 0x1
+#define SPI_RESOURCE_RESERVE_EN_CU_1__EN__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_EN_CU_1__TYPE_MASK_MASK 0xfffe
+#define SPI_RESOURCE_RESERVE_EN_CU_1__TYPE_MASK__SHIFT 0x1
+#define SPI_RESOURCE_RESERVE_EN_CU_1__QUEUE_MASK_MASK 0xff0000
+#define SPI_RESOURCE_RESERVE_EN_CU_1__QUEUE_MASK__SHIFT 0x10
+#define SPI_RESOURCE_RESERVE_EN_CU_1__RESERVE_SPACE_ONLY_MASK 0x1000000
+#define SPI_RESOURCE_RESERVE_EN_CU_1__RESERVE_SPACE_ONLY__SHIFT 0x18
+#define SPI_RESOURCE_RESERVE_EN_CU_2__EN_MASK 0x1
+#define SPI_RESOURCE_RESERVE_EN_CU_2__EN__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_EN_CU_2__TYPE_MASK_MASK 0xfffe
+#define SPI_RESOURCE_RESERVE_EN_CU_2__TYPE_MASK__SHIFT 0x1
+#define SPI_RESOURCE_RESERVE_EN_CU_2__QUEUE_MASK_MASK 0xff0000
+#define SPI_RESOURCE_RESERVE_EN_CU_2__QUEUE_MASK__SHIFT 0x10
+#define SPI_RESOURCE_RESERVE_EN_CU_2__RESERVE_SPACE_ONLY_MASK 0x1000000
+#define SPI_RESOURCE_RESERVE_EN_CU_2__RESERVE_SPACE_ONLY__SHIFT 0x18
+#define SPI_RESOURCE_RESERVE_EN_CU_3__EN_MASK 0x1
+#define SPI_RESOURCE_RESERVE_EN_CU_3__EN__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_EN_CU_3__TYPE_MASK_MASK 0xfffe
+#define SPI_RESOURCE_RESERVE_EN_CU_3__TYPE_MASK__SHIFT 0x1
+#define SPI_RESOURCE_RESERVE_EN_CU_3__QUEUE_MASK_MASK 0xff0000
+#define SPI_RESOURCE_RESERVE_EN_CU_3__QUEUE_MASK__SHIFT 0x10
+#define SPI_RESOURCE_RESERVE_EN_CU_3__RESERVE_SPACE_ONLY_MASK 0x1000000
+#define SPI_RESOURCE_RESERVE_EN_CU_3__RESERVE_SPACE_ONLY__SHIFT 0x18
+#define SPI_RESOURCE_RESERVE_EN_CU_4__EN_MASK 0x1
+#define SPI_RESOURCE_RESERVE_EN_CU_4__EN__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_EN_CU_4__TYPE_MASK_MASK 0xfffe
+#define SPI_RESOURCE_RESERVE_EN_CU_4__TYPE_MASK__SHIFT 0x1
+#define SPI_RESOURCE_RESERVE_EN_CU_4__QUEUE_MASK_MASK 0xff0000
+#define SPI_RESOURCE_RESERVE_EN_CU_4__QUEUE_MASK__SHIFT 0x10
+#define SPI_RESOURCE_RESERVE_EN_CU_4__RESERVE_SPACE_ONLY_MASK 0x1000000
+#define SPI_RESOURCE_RESERVE_EN_CU_4__RESERVE_SPACE_ONLY__SHIFT 0x18
+#define SPI_RESOURCE_RESERVE_EN_CU_5__EN_MASK 0x1
+#define SPI_RESOURCE_RESERVE_EN_CU_5__EN__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_EN_CU_5__TYPE_MASK_MASK 0xfffe
+#define SPI_RESOURCE_RESERVE_EN_CU_5__TYPE_MASK__SHIFT 0x1
+#define SPI_RESOURCE_RESERVE_EN_CU_5__QUEUE_MASK_MASK 0xff0000
+#define SPI_RESOURCE_RESERVE_EN_CU_5__QUEUE_MASK__SHIFT 0x10
+#define SPI_RESOURCE_RESERVE_EN_CU_5__RESERVE_SPACE_ONLY_MASK 0x1000000
+#define SPI_RESOURCE_RESERVE_EN_CU_5__RESERVE_SPACE_ONLY__SHIFT 0x18
+#define SPI_RESOURCE_RESERVE_EN_CU_6__EN_MASK 0x1
+#define SPI_RESOURCE_RESERVE_EN_CU_6__EN__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_EN_CU_6__TYPE_MASK_MASK 0xfffe
+#define SPI_RESOURCE_RESERVE_EN_CU_6__TYPE_MASK__SHIFT 0x1
+#define SPI_RESOURCE_RESERVE_EN_CU_6__QUEUE_MASK_MASK 0xff0000
+#define SPI_RESOURCE_RESERVE_EN_CU_6__QUEUE_MASK__SHIFT 0x10
+#define SPI_RESOURCE_RESERVE_EN_CU_6__RESERVE_SPACE_ONLY_MASK 0x1000000
+#define SPI_RESOURCE_RESERVE_EN_CU_6__RESERVE_SPACE_ONLY__SHIFT 0x18
+#define SPI_RESOURCE_RESERVE_EN_CU_7__EN_MASK 0x1
+#define SPI_RESOURCE_RESERVE_EN_CU_7__EN__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_EN_CU_7__TYPE_MASK_MASK 0xfffe
+#define SPI_RESOURCE_RESERVE_EN_CU_7__TYPE_MASK__SHIFT 0x1
+#define SPI_RESOURCE_RESERVE_EN_CU_7__QUEUE_MASK_MASK 0xff0000
+#define SPI_RESOURCE_RESERVE_EN_CU_7__QUEUE_MASK__SHIFT 0x10
+#define SPI_RESOURCE_RESERVE_EN_CU_7__RESERVE_SPACE_ONLY_MASK 0x1000000
+#define SPI_RESOURCE_RESERVE_EN_CU_7__RESERVE_SPACE_ONLY__SHIFT 0x18
+#define SPI_RESOURCE_RESERVE_EN_CU_8__EN_MASK 0x1
+#define SPI_RESOURCE_RESERVE_EN_CU_8__EN__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_EN_CU_8__TYPE_MASK_MASK 0xfffe
+#define SPI_RESOURCE_RESERVE_EN_CU_8__TYPE_MASK__SHIFT 0x1
+#define SPI_RESOURCE_RESERVE_EN_CU_8__QUEUE_MASK_MASK 0xff0000
+#define SPI_RESOURCE_RESERVE_EN_CU_8__QUEUE_MASK__SHIFT 0x10
+#define SPI_RESOURCE_RESERVE_EN_CU_8__RESERVE_SPACE_ONLY_MASK 0x1000000
+#define SPI_RESOURCE_RESERVE_EN_CU_8__RESERVE_SPACE_ONLY__SHIFT 0x18
+#define SPI_RESOURCE_RESERVE_EN_CU_9__EN_MASK 0x1
+#define SPI_RESOURCE_RESERVE_EN_CU_9__EN__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_EN_CU_9__TYPE_MASK_MASK 0xfffe
+#define SPI_RESOURCE_RESERVE_EN_CU_9__TYPE_MASK__SHIFT 0x1
+#define SPI_RESOURCE_RESERVE_EN_CU_9__QUEUE_MASK_MASK 0xff0000
+#define SPI_RESOURCE_RESERVE_EN_CU_9__QUEUE_MASK__SHIFT 0x10
+#define SPI_RESOURCE_RESERVE_EN_CU_9__RESERVE_SPACE_ONLY_MASK 0x1000000
+#define SPI_RESOURCE_RESERVE_EN_CU_9__RESERVE_SPACE_ONLY__SHIFT 0x18
+#define SPI_RESOURCE_RESERVE_EN_CU_10__EN_MASK 0x1
+#define SPI_RESOURCE_RESERVE_EN_CU_10__EN__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_EN_CU_10__TYPE_MASK_MASK 0xfffe
+#define SPI_RESOURCE_RESERVE_EN_CU_10__TYPE_MASK__SHIFT 0x1
+#define SPI_RESOURCE_RESERVE_EN_CU_10__QUEUE_MASK_MASK 0xff0000
+#define SPI_RESOURCE_RESERVE_EN_CU_10__QUEUE_MASK__SHIFT 0x10
+#define SPI_RESOURCE_RESERVE_EN_CU_10__RESERVE_SPACE_ONLY_MASK 0x1000000
+#define SPI_RESOURCE_RESERVE_EN_CU_10__RESERVE_SPACE_ONLY__SHIFT 0x18
+#define SPI_RESOURCE_RESERVE_EN_CU_11__EN_MASK 0x1
+#define SPI_RESOURCE_RESERVE_EN_CU_11__EN__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_EN_CU_11__TYPE_MASK_MASK 0xfffe
+#define SPI_RESOURCE_RESERVE_EN_CU_11__TYPE_MASK__SHIFT 0x1
+#define SPI_RESOURCE_RESERVE_EN_CU_11__QUEUE_MASK_MASK 0xff0000
+#define SPI_RESOURCE_RESERVE_EN_CU_11__QUEUE_MASK__SHIFT 0x10
+#define SPI_RESOURCE_RESERVE_EN_CU_11__RESERVE_SPACE_ONLY_MASK 0x1000000
+#define SPI_RESOURCE_RESERVE_EN_CU_11__RESERVE_SPACE_ONLY__SHIFT 0x18
+#define SPI_RESOURCE_RESERVE_EN_CU_12__EN_MASK 0x1
+#define SPI_RESOURCE_RESERVE_EN_CU_12__EN__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_EN_CU_12__TYPE_MASK_MASK 0xfffe
+#define SPI_RESOURCE_RESERVE_EN_CU_12__TYPE_MASK__SHIFT 0x1
+#define SPI_RESOURCE_RESERVE_EN_CU_12__QUEUE_MASK_MASK 0xff0000
+#define SPI_RESOURCE_RESERVE_EN_CU_12__QUEUE_MASK__SHIFT 0x10
+#define SPI_RESOURCE_RESERVE_EN_CU_12__RESERVE_SPACE_ONLY_MASK 0x1000000
+#define SPI_RESOURCE_RESERVE_EN_CU_12__RESERVE_SPACE_ONLY__SHIFT 0x18
+#define SPI_RESOURCE_RESERVE_EN_CU_13__EN_MASK 0x1
+#define SPI_RESOURCE_RESERVE_EN_CU_13__EN__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_EN_CU_13__TYPE_MASK_MASK 0xfffe
+#define SPI_RESOURCE_RESERVE_EN_CU_13__TYPE_MASK__SHIFT 0x1
+#define SPI_RESOURCE_RESERVE_EN_CU_13__QUEUE_MASK_MASK 0xff0000
+#define SPI_RESOURCE_RESERVE_EN_CU_13__QUEUE_MASK__SHIFT 0x10
+#define SPI_RESOURCE_RESERVE_EN_CU_13__RESERVE_SPACE_ONLY_MASK 0x1000000
+#define SPI_RESOURCE_RESERVE_EN_CU_13__RESERVE_SPACE_ONLY__SHIFT 0x18
+#define SPI_RESOURCE_RESERVE_EN_CU_14__EN_MASK 0x1
+#define SPI_RESOURCE_RESERVE_EN_CU_14__EN__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_EN_CU_14__TYPE_MASK_MASK 0xfffe
+#define SPI_RESOURCE_RESERVE_EN_CU_14__TYPE_MASK__SHIFT 0x1
+#define SPI_RESOURCE_RESERVE_EN_CU_14__QUEUE_MASK_MASK 0xff0000
+#define SPI_RESOURCE_RESERVE_EN_CU_14__QUEUE_MASK__SHIFT 0x10
+#define SPI_RESOURCE_RESERVE_EN_CU_14__RESERVE_SPACE_ONLY_MASK 0x1000000
+#define SPI_RESOURCE_RESERVE_EN_CU_14__RESERVE_SPACE_ONLY__SHIFT 0x18
+#define SPI_RESOURCE_RESERVE_EN_CU_15__EN_MASK 0x1
+#define SPI_RESOURCE_RESERVE_EN_CU_15__EN__SHIFT 0x0
+#define SPI_RESOURCE_RESERVE_EN_CU_15__TYPE_MASK_MASK 0xfffe
+#define SPI_RESOURCE_RESERVE_EN_CU_15__TYPE_MASK__SHIFT 0x1
+#define SPI_RESOURCE_RESERVE_EN_CU_15__QUEUE_MASK_MASK 0xff0000
+#define SPI_RESOURCE_RESERVE_EN_CU_15__QUEUE_MASK__SHIFT 0x10
+#define SPI_RESOURCE_RESERVE_EN_CU_15__RESERVE_SPACE_ONLY_MASK 0x1000000
+#define SPI_RESOURCE_RESERVE_EN_CU_15__RESERVE_SPACE_ONLY__SHIFT 0x18
+#define SPI_COMPUTE_WF_CTX_SAVE__INITIATE_MASK 0x1
+#define SPI_COMPUTE_WF_CTX_SAVE__INITIATE__SHIFT 0x0
+#define SPI_COMPUTE_WF_CTX_SAVE__GDS_INTERRUPT_EN_MASK 0x2
+#define SPI_COMPUTE_WF_CTX_SAVE__GDS_INTERRUPT_EN__SHIFT 0x1
+#define SPI_COMPUTE_WF_CTX_SAVE__DONE_INTERRUPT_EN_MASK 0x4
+#define SPI_COMPUTE_WF_CTX_SAVE__DONE_INTERRUPT_EN__SHIFT 0x2
+#define SPI_COMPUTE_WF_CTX_SAVE__GDS_REQ_BUSY_MASK 0x40000000
+#define SPI_COMPUTE_WF_CTX_SAVE__GDS_REQ_BUSY__SHIFT 0x1e
+#define SPI_COMPUTE_WF_CTX_SAVE__SAVE_BUSY_MASK 0x80000000
+#define SPI_COMPUTE_WF_CTX_SAVE__SAVE_BUSY__SHIFT 0x1f
+#define SPI_PS_MAX_WAVE_ID__MAX_WAVE_ID_MASK 0xfff
+#define SPI_PS_MAX_WAVE_ID__MAX_WAVE_ID__SHIFT 0x0
+#define SPI_START_PHASE__VGPR_START_PHASE_MASK 0x3
+#define SPI_START_PHASE__VGPR_START_PHASE__SHIFT 0x0
+#define SPI_START_PHASE__SGPR_START_PHASE_MASK 0xc
+#define SPI_START_PHASE__SGPR_START_PHASE__SHIFT 0x2
+#define SPI_START_PHASE__WAVE_START_PHASE_MASK 0x30
+#define SPI_START_PHASE__WAVE_START_PHASE__SHIFT 0x4
+#define SPI_GFX_CNTL__RESET_COUNTS_MASK 0x1
+#define SPI_GFX_CNTL__RESET_COUNTS__SHIFT 0x0
+#define SPI_CONFIG_CNTL__GPR_WRITE_PRIORITY_MASK 0x1fffff
+#define SPI_CONFIG_CNTL__GPR_WRITE_PRIORITY__SHIFT 0x0
+#define SPI_CONFIG_CNTL__EXP_PRIORITY_ORDER_MASK 0xe00000
+#define SPI_CONFIG_CNTL__EXP_PRIORITY_ORDER__SHIFT 0x15
+#define SPI_CONFIG_CNTL__ENABLE_SQG_TOP_EVENTS_MASK 0x1000000
+#define SPI_CONFIG_CNTL__ENABLE_SQG_TOP_EVENTS__SHIFT 0x18
+#define SPI_CONFIG_CNTL__ENABLE_SQG_BOP_EVENTS_MASK 0x2000000
+#define SPI_CONFIG_CNTL__ENABLE_SQG_BOP_EVENTS__SHIFT 0x19
+#define SPI_CONFIG_CNTL__RSRC_MGMT_RESET_MASK 0x4000000
+#define SPI_CONFIG_CNTL__RSRC_MGMT_RESET__SHIFT 0x1a
+#define SPI_CONFIG_CNTL__TTRACE_STALL_ALL_MASK 0x8000000
+#define SPI_CONFIG_CNTL__TTRACE_STALL_ALL__SHIFT 0x1b
+#define SPI_DEBUG_CNTL__DEBUG_GRBM_OVERRIDE_MASK 0x1
+#define SPI_DEBUG_CNTL__DEBUG_GRBM_OVERRIDE__SHIFT 0x0
+#define SPI_DEBUG_CNTL__DEBUG_THREAD_TYPE_SEL_MASK 0xe
+#define SPI_DEBUG_CNTL__DEBUG_THREAD_TYPE_SEL__SHIFT 0x1
+#define SPI_DEBUG_CNTL__DEBUG_GROUP_SEL_MASK 0x3f0
+#define SPI_DEBUG_CNTL__DEBUG_GROUP_SEL__SHIFT 0x4
+#define SPI_DEBUG_CNTL__DEBUG_SIMD_SEL_MASK 0xfc00
+#define SPI_DEBUG_CNTL__DEBUG_SIMD_SEL__SHIFT 0xa
+#define SPI_DEBUG_CNTL__DEBUG_SH_SEL_MASK 0x10000
+#define SPI_DEBUG_CNTL__DEBUG_SH_SEL__SHIFT 0x10
+#define SPI_DEBUG_CNTL__SPI_ECO_SPARE_0_MASK 0x20000
+#define SPI_DEBUG_CNTL__SPI_ECO_SPARE_0__SHIFT 0x11
+#define SPI_DEBUG_CNTL__SPI_ECO_SPARE_1_MASK 0x40000
+#define SPI_DEBUG_CNTL__SPI_ECO_SPARE_1__SHIFT 0x12
+#define SPI_DEBUG_CNTL__SPI_ECO_SPARE_2_MASK 0x80000
+#define SPI_DEBUG_CNTL__SPI_ECO_SPARE_2__SHIFT 0x13
+#define SPI_DEBUG_CNTL__SPI_ECO_SPARE_3_MASK 0x100000
+#define SPI_DEBUG_CNTL__SPI_ECO_SPARE_3__SHIFT 0x14
+#define SPI_DEBUG_CNTL__SPI_ECO_SPARE_4_MASK 0x200000
+#define SPI_DEBUG_CNTL__SPI_ECO_SPARE_4__SHIFT 0x15
+#define SPI_DEBUG_CNTL__SPI_ECO_SPARE_5_MASK 0x400000
+#define SPI_DEBUG_CNTL__SPI_ECO_SPARE_5__SHIFT 0x16
+#define SPI_DEBUG_CNTL__SPI_ECO_SPARE_6_MASK 0x800000
+#define SPI_DEBUG_CNTL__SPI_ECO_SPARE_6__SHIFT 0x17
+#define SPI_DEBUG_CNTL__SPI_ECO_SPARE_7_MASK 0x1000000
+#define SPI_DEBUG_CNTL__SPI_ECO_SPARE_7__SHIFT 0x18
+#define SPI_DEBUG_CNTL__DEBUG_PIPE_SEL_MASK 0xe000000
+#define SPI_DEBUG_CNTL__DEBUG_PIPE_SEL__SHIFT 0x19
+#define SPI_DEBUG_CNTL__DEBUG_REG_EN_MASK 0x80000000
+#define SPI_DEBUG_CNTL__DEBUG_REG_EN__SHIFT 0x1f
+#define SPI_DEBUG_READ__DATA_MASK 0xffffff
+#define SPI_DEBUG_READ__DATA__SHIFT 0x0
+#define SPI_DSM_CNTL__Sel_DSM_SPI_Irritator_data0_MASK 0x1
+#define SPI_DSM_CNTL__Sel_DSM_SPI_Irritator_data0__SHIFT 0x0
+#define SPI_DSM_CNTL__Sel_DSM_SPI_Irritator_data1_MASK 0x2
+#define SPI_DSM_CNTL__Sel_DSM_SPI_Irritator_data1__SHIFT 0x1
+#define SPI_DSM_CNTL__SPI_Enable_Single_Write_MASK 0x4
+#define SPI_DSM_CNTL__SPI_Enable_Single_Write__SHIFT 0x2
+#define SPI_DSM_CNTL__UNUSED_MASK 0xfffffff8
+#define SPI_DSM_CNTL__UNUSED__SHIFT 0x3
+#define SPI_EDC_CNT__SED_MASK 0xff
+#define SPI_EDC_CNT__SED__SHIFT 0x0
+#define SPI_PERFCOUNTER0_SELECT__PERF_SEL_MASK 0x3ff
+#define SPI_PERFCOUNTER0_SELECT__PERF_SEL__SHIFT 0x0
+#define SPI_PERFCOUNTER0_SELECT__PERF_SEL1_MASK 0xffc00
+#define SPI_PERFCOUNTER0_SELECT__PERF_SEL1__SHIFT 0xa
+#define SPI_PERFCOUNTER0_SELECT__CNTR_MODE_MASK 0xf00000
+#define SPI_PERFCOUNTER0_SELECT__CNTR_MODE__SHIFT 0x14
+#define SPI_PERFCOUNTER1_SELECT__PERF_SEL_MASK 0x3ff
+#define SPI_PERFCOUNTER1_SELECT__PERF_SEL__SHIFT 0x0
+#define SPI_PERFCOUNTER1_SELECT__PERF_SEL1_MASK 0xffc00
+#define SPI_PERFCOUNTER1_SELECT__PERF_SEL1__SHIFT 0xa
+#define SPI_PERFCOUNTER1_SELECT__CNTR_MODE_MASK 0xf00000
+#define SPI_PERFCOUNTER1_SELECT__CNTR_MODE__SHIFT 0x14
+#define SPI_PERFCOUNTER2_SELECT__PERF_SEL_MASK 0x3ff
+#define SPI_PERFCOUNTER2_SELECT__PERF_SEL__SHIFT 0x0
+#define SPI_PERFCOUNTER2_SELECT__PERF_SEL1_MASK 0xffc00
+#define SPI_PERFCOUNTER2_SELECT__PERF_SEL1__SHIFT 0xa
+#define SPI_PERFCOUNTER2_SELECT__CNTR_MODE_MASK 0xf00000
+#define SPI_PERFCOUNTER2_SELECT__CNTR_MODE__SHIFT 0x14
+#define SPI_PERFCOUNTER3_SELECT__PERF_SEL_MASK 0x3ff
+#define SPI_PERFCOUNTER3_SELECT__PERF_SEL__SHIFT 0x0
+#define SPI_PERFCOUNTER3_SELECT__PERF_SEL1_MASK 0xffc00
+#define SPI_PERFCOUNTER3_SELECT__PERF_SEL1__SHIFT 0xa
+#define SPI_PERFCOUNTER3_SELECT__CNTR_MODE_MASK 0xf00000
+#define SPI_PERFCOUNTER3_SELECT__CNTR_MODE__SHIFT 0x14
+#define SPI_PERFCOUNTER0_SELECT1__PERF_SEL2_MASK 0x3ff
+#define SPI_PERFCOUNTER0_SELECT1__PERF_SEL2__SHIFT 0x0
+#define SPI_PERFCOUNTER0_SELECT1__PERF_SEL3_MASK 0xffc00
+#define SPI_PERFCOUNTER0_SELECT1__PERF_SEL3__SHIFT 0xa
+#define SPI_PERFCOUNTER1_SELECT1__PERF_SEL2_MASK 0x3ff
+#define SPI_PERFCOUNTER1_SELECT1__PERF_SEL2__SHIFT 0x0
+#define SPI_PERFCOUNTER1_SELECT1__PERF_SEL3_MASK 0xffc00
+#define SPI_PERFCOUNTER1_SELECT1__PERF_SEL3__SHIFT 0xa
+#define SPI_PERFCOUNTER2_SELECT1__PERF_SEL2_MASK 0x3ff
+#define SPI_PERFCOUNTER2_SELECT1__PERF_SEL2__SHIFT 0x0
+#define SPI_PERFCOUNTER2_SELECT1__PERF_SEL3_MASK 0xffc00
+#define SPI_PERFCOUNTER2_SELECT1__PERF_SEL3__SHIFT 0xa
+#define SPI_PERFCOUNTER3_SELECT1__PERF_SEL2_MASK 0x3ff
+#define SPI_PERFCOUNTER3_SELECT1__PERF_SEL2__SHIFT 0x0
+#define SPI_PERFCOUNTER3_SELECT1__PERF_SEL3_MASK 0xffc00
+#define SPI_PERFCOUNTER3_SELECT1__PERF_SEL3__SHIFT 0xa
+#define SPI_PERFCOUNTER4_SELECT__PERF_SEL_MASK 0xff
+#define SPI_PERFCOUNTER4_SELECT__PERF_SEL__SHIFT 0x0
+#define SPI_PERFCOUNTER5_SELECT__PERF_SEL_MASK 0xff
+#define SPI_PERFCOUNTER5_SELECT__PERF_SEL__SHIFT 0x0
+#define SPI_PERFCOUNTER_BINS__BIN0_MIN_MASK 0xf
+#define SPI_PERFCOUNTER_BINS__BIN0_MIN__SHIFT 0x0
+#define SPI_PERFCOUNTER_BINS__BIN0_MAX_MASK 0xf0
+#define SPI_PERFCOUNTER_BINS__BIN0_MAX__SHIFT 0x4
+#define SPI_PERFCOUNTER_BINS__BIN1_MIN_MASK 0xf00
+#define SPI_PERFCOUNTER_BINS__BIN1_MIN__SHIFT 0x8
+#define SPI_PERFCOUNTER_BINS__BIN1_MAX_MASK 0xf000
+#define SPI_PERFCOUNTER_BINS__BIN1_MAX__SHIFT 0xc
+#define SPI_PERFCOUNTER_BINS__BIN2_MIN_MASK 0xf0000
+#define SPI_PERFCOUNTER_BINS__BIN2_MIN__SHIFT 0x10
+#define SPI_PERFCOUNTER_BINS__BIN2_MAX_MASK 0xf00000
+#define SPI_PERFCOUNTER_BINS__BIN2_MAX__SHIFT 0x14
+#define SPI_PERFCOUNTER_BINS__BIN3_MIN_MASK 0xf000000
+#define SPI_PERFCOUNTER_BINS__BIN3_MIN__SHIFT 0x18
+#define SPI_PERFCOUNTER_BINS__BIN3_MAX_MASK 0xf0000000
+#define SPI_PERFCOUNTER_BINS__BIN3_MAX__SHIFT 0x1c
+#define SPI_PERFCOUNTER0_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define SPI_PERFCOUNTER0_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define SPI_PERFCOUNTER0_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define SPI_PERFCOUNTER0_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define SPI_PERFCOUNTER1_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define SPI_PERFCOUNTER1_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define SPI_PERFCOUNTER1_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define SPI_PERFCOUNTER1_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define SPI_PERFCOUNTER2_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define SPI_PERFCOUNTER2_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define SPI_PERFCOUNTER2_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define SPI_PERFCOUNTER2_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define SPI_PERFCOUNTER3_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define SPI_PERFCOUNTER3_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define SPI_PERFCOUNTER3_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define SPI_PERFCOUNTER3_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define SPI_PERFCOUNTER4_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define SPI_PERFCOUNTER4_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define SPI_PERFCOUNTER4_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define SPI_PERFCOUNTER4_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define SPI_PERFCOUNTER5_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define SPI_PERFCOUNTER5_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define SPI_PERFCOUNTER5_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define SPI_PERFCOUNTER5_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define SPI_CONFIG_CNTL_1__VTX_DONE_DELAY_MASK 0xf
+#define SPI_CONFIG_CNTL_1__VTX_DONE_DELAY__SHIFT 0x0
+#define SPI_CONFIG_CNTL_1__INTERP_ONE_PRIM_PER_ROW_MASK 0x10
+#define SPI_CONFIG_CNTL_1__INTERP_ONE_PRIM_PER_ROW__SHIFT 0x4
+#define SPI_CONFIG_CNTL_1__PC_LIMIT_ENABLE_MASK 0x40
+#define SPI_CONFIG_CNTL_1__PC_LIMIT_ENABLE__SHIFT 0x6
+#define SPI_CONFIG_CNTL_1__PC_LIMIT_STRICT_MASK 0x80
+#define SPI_CONFIG_CNTL_1__PC_LIMIT_STRICT__SHIFT 0x7
+#define SPI_CONFIG_CNTL_1__CRC_SIMD_ID_WADDR_DISABLE_MASK 0x100
+#define SPI_CONFIG_CNTL_1__CRC_SIMD_ID_WADDR_DISABLE__SHIFT 0x8
+#define SPI_CONFIG_CNTL_1__LBPW_CU_CHK_MODE_MASK 0x200
+#define SPI_CONFIG_CNTL_1__LBPW_CU_CHK_MODE__SHIFT 0x9
+#define SPI_CONFIG_CNTL_1__LBPW_CU_CHK_CNT_MASK 0x3c00
+#define SPI_CONFIG_CNTL_1__LBPW_CU_CHK_CNT__SHIFT 0xa
+#define SPI_CONFIG_CNTL_1__PC_LIMIT_SIZE_MASK 0xffff0000
+#define SPI_CONFIG_CNTL_1__PC_LIMIT_SIZE__SHIFT 0x10
+#define SPI_DEBUG_BUSY__LS_BUSY_MASK 0x1
+#define SPI_DEBUG_BUSY__LS_BUSY__SHIFT 0x0
+#define SPI_DEBUG_BUSY__HS_BUSY_MASK 0x2
+#define SPI_DEBUG_BUSY__HS_BUSY__SHIFT 0x1
+#define SPI_DEBUG_BUSY__ES_BUSY_MASK 0x4
+#define SPI_DEBUG_BUSY__ES_BUSY__SHIFT 0x2
+#define SPI_DEBUG_BUSY__GS_BUSY_MASK 0x8
+#define SPI_DEBUG_BUSY__GS_BUSY__SHIFT 0x3
+#define SPI_DEBUG_BUSY__VS_BUSY_MASK 0x10
+#define SPI_DEBUG_BUSY__VS_BUSY__SHIFT 0x4
+#define SPI_DEBUG_BUSY__PS0_BUSY_MASK 0x20
+#define SPI_DEBUG_BUSY__PS0_BUSY__SHIFT 0x5
+#define SPI_DEBUG_BUSY__PS1_BUSY_MASK 0x40
+#define SPI_DEBUG_BUSY__PS1_BUSY__SHIFT 0x6
+#define SPI_DEBUG_BUSY__CSG_BUSY_MASK 0x80
+#define SPI_DEBUG_BUSY__CSG_BUSY__SHIFT 0x7
+#define SPI_DEBUG_BUSY__CS0_BUSY_MASK 0x100
+#define SPI_DEBUG_BUSY__CS0_BUSY__SHIFT 0x8
+#define SPI_DEBUG_BUSY__CS1_BUSY_MASK 0x200
+#define SPI_DEBUG_BUSY__CS1_BUSY__SHIFT 0x9
+#define SPI_DEBUG_BUSY__CS2_BUSY_MASK 0x400
+#define SPI_DEBUG_BUSY__CS2_BUSY__SHIFT 0xa
+#define SPI_DEBUG_BUSY__CS3_BUSY_MASK 0x800
+#define SPI_DEBUG_BUSY__CS3_BUSY__SHIFT 0xb
+#define SPI_DEBUG_BUSY__CS4_BUSY_MASK 0x1000
+#define SPI_DEBUG_BUSY__CS4_BUSY__SHIFT 0xc
+#define SPI_DEBUG_BUSY__CS5_BUSY_MASK 0x2000
+#define SPI_DEBUG_BUSY__CS5_BUSY__SHIFT 0xd
+#define SPI_DEBUG_BUSY__CS6_BUSY_MASK 0x4000
+#define SPI_DEBUG_BUSY__CS6_BUSY__SHIFT 0xe
+#define SPI_DEBUG_BUSY__CS7_BUSY_MASK 0x8000
+#define SPI_DEBUG_BUSY__CS7_BUSY__SHIFT 0xf
+#define SPI_DEBUG_BUSY__LDS_WR_CTL0_BUSY_MASK 0x10000
+#define SPI_DEBUG_BUSY__LDS_WR_CTL0_BUSY__SHIFT 0x10
+#define SPI_DEBUG_BUSY__LDS_WR_CTL1_BUSY_MASK 0x20000
+#define SPI_DEBUG_BUSY__LDS_WR_CTL1_BUSY__SHIFT 0x11
+#define SPI_DEBUG_BUSY__RSRC_ALLOC0_BUSY_MASK 0x40000
+#define SPI_DEBUG_BUSY__RSRC_ALLOC0_BUSY__SHIFT 0x12
+#define SPI_DEBUG_BUSY__RSRC_ALLOC1_BUSY_MASK 0x80000
+#define SPI_DEBUG_BUSY__RSRC_ALLOC1_BUSY__SHIFT 0x13
+#define SPI_DEBUG_BUSY__PC_DEALLOC_BUSY_MASK 0x100000
+#define SPI_DEBUG_BUSY__PC_DEALLOC_BUSY__SHIFT 0x14
+#define SPI_DEBUG_BUSY__EVENT_CLCTR_BUSY_MASK 0x200000
+#define SPI_DEBUG_BUSY__EVENT_CLCTR_BUSY__SHIFT 0x15
+#define SPI_DEBUG_BUSY__GRBM_BUSY_MASK 0x400000
+#define SPI_DEBUG_BUSY__GRBM_BUSY__SHIFT 0x16
+#define SPI_DEBUG_BUSY__SPIS_BUSY_MASK 0x800000
+#define SPI_DEBUG_BUSY__SPIS_BUSY__SHIFT 0x17
+#define SPI_CONFIG_CNTL_2__CONTEXT_SAVE_WAIT_GDS_REQUEST_CYCLE_OVHD_MASK 0xf
+#define SPI_CONFIG_CNTL_2__CONTEXT_SAVE_WAIT_GDS_REQUEST_CYCLE_OVHD__SHIFT 0x0
+#define SPI_CONFIG_CNTL_2__CONTEXT_SAVE_WAIT_GDS_GRANT_CYCLE_OVHD_MASK 0xf0
+#define SPI_CONFIG_CNTL_2__CONTEXT_SAVE_WAIT_GDS_GRANT_CYCLE_OVHD__SHIFT 0x4
+#define CGTS_SM_CTRL_REG__ON_SEQ_DELAY_MASK 0xf
+#define CGTS_SM_CTRL_REG__ON_SEQ_DELAY__SHIFT 0x0
+#define CGTS_SM_CTRL_REG__OFF_SEQ_DELAY_MASK 0xff0
+#define CGTS_SM_CTRL_REG__OFF_SEQ_DELAY__SHIFT 0x4
+#define CGTS_SM_CTRL_REG__MGCG_ENABLED_MASK 0x1000
+#define CGTS_SM_CTRL_REG__MGCG_ENABLED__SHIFT 0xc
+#define CGTS_SM_CTRL_REG__BASE_MODE_MASK 0x10000
+#define CGTS_SM_CTRL_REG__BASE_MODE__SHIFT 0x10
+#define CGTS_SM_CTRL_REG__SM_MODE_MASK 0xe0000
+#define CGTS_SM_CTRL_REG__SM_MODE__SHIFT 0x11
+#define CGTS_SM_CTRL_REG__SM_MODE_ENABLE_MASK 0x100000
+#define CGTS_SM_CTRL_REG__SM_MODE_ENABLE__SHIFT 0x14
+#define CGTS_SM_CTRL_REG__OVERRIDE_MASK 0x200000
+#define CGTS_SM_CTRL_REG__OVERRIDE__SHIFT 0x15
+#define CGTS_SM_CTRL_REG__LS_OVERRIDE_MASK 0x400000
+#define CGTS_SM_CTRL_REG__LS_OVERRIDE__SHIFT 0x16
+#define CGTS_SM_CTRL_REG__ON_MONITOR_ADD_EN_MASK 0x800000
+#define CGTS_SM_CTRL_REG__ON_MONITOR_ADD_EN__SHIFT 0x17
+#define CGTS_SM_CTRL_REG__ON_MONITOR_ADD_MASK 0xff000000
+#define CGTS_SM_CTRL_REG__ON_MONITOR_ADD__SHIFT 0x18
+#define CGTS_RD_CTRL_REG__ROW_MUX_SEL_MASK 0x1f
+#define CGTS_RD_CTRL_REG__ROW_MUX_SEL__SHIFT 0x0
+#define CGTS_RD_CTRL_REG__REG_MUX_SEL_MASK 0x1f00
+#define CGTS_RD_CTRL_REG__REG_MUX_SEL__SHIFT 0x8
+#define CGTS_RD_REG__READ_DATA_MASK 0x3fff
+#define CGTS_RD_REG__READ_DATA__SHIFT 0x0
+#define CGTS_TCC_DISABLE__TCC_DISABLE_MASK 0xffff0000
+#define CGTS_TCC_DISABLE__TCC_DISABLE__SHIFT 0x10
+#define CGTS_USER_TCC_DISABLE__TCC_DISABLE_MASK 0xffff0000
+#define CGTS_USER_TCC_DISABLE__TCC_DISABLE__SHIFT 0x10
+#define CGTS_CU0_SP0_CTRL_REG__SP00_MASK 0x7f
+#define CGTS_CU0_SP0_CTRL_REG__SP00__SHIFT 0x0
+#define CGTS_CU0_SP0_CTRL_REG__SP00_OVERRIDE_MASK 0x80
+#define CGTS_CU0_SP0_CTRL_REG__SP00_OVERRIDE__SHIFT 0x7
+#define CGTS_CU0_SP0_CTRL_REG__SP00_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU0_SP0_CTRL_REG__SP00_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU0_SP0_CTRL_REG__SP00_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU0_SP0_CTRL_REG__SP00_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU0_SP0_CTRL_REG__SP00_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU0_SP0_CTRL_REG__SP00_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU0_SP0_CTRL_REG__SP01_MASK 0x7f0000
+#define CGTS_CU0_SP0_CTRL_REG__SP01__SHIFT 0x10
+#define CGTS_CU0_SP0_CTRL_REG__SP01_OVERRIDE_MASK 0x800000
+#define CGTS_CU0_SP0_CTRL_REG__SP01_OVERRIDE__SHIFT 0x17
+#define CGTS_CU0_SP0_CTRL_REG__SP01_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU0_SP0_CTRL_REG__SP01_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU0_SP0_CTRL_REG__SP01_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU0_SP0_CTRL_REG__SP01_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU0_SP0_CTRL_REG__SP01_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU0_SP0_CTRL_REG__SP01_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU0_LDS_SQ_CTRL_REG__LDS_MASK 0x7f
+#define CGTS_CU0_LDS_SQ_CTRL_REG__LDS__SHIFT 0x0
+#define CGTS_CU0_LDS_SQ_CTRL_REG__LDS_OVERRIDE_MASK 0x80
+#define CGTS_CU0_LDS_SQ_CTRL_REG__LDS_OVERRIDE__SHIFT 0x7
+#define CGTS_CU0_LDS_SQ_CTRL_REG__LDS_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU0_LDS_SQ_CTRL_REG__LDS_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU0_LDS_SQ_CTRL_REG__LDS_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU0_LDS_SQ_CTRL_REG__LDS_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU0_LDS_SQ_CTRL_REG__LDS_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU0_LDS_SQ_CTRL_REG__LDS_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU0_LDS_SQ_CTRL_REG__SQ_MASK 0x7f0000
+#define CGTS_CU0_LDS_SQ_CTRL_REG__SQ__SHIFT 0x10
+#define CGTS_CU0_LDS_SQ_CTRL_REG__SQ_OVERRIDE_MASK 0x800000
+#define CGTS_CU0_LDS_SQ_CTRL_REG__SQ_OVERRIDE__SHIFT 0x17
+#define CGTS_CU0_LDS_SQ_CTRL_REG__SQ_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU0_LDS_SQ_CTRL_REG__SQ_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU0_LDS_SQ_CTRL_REG__SQ_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU0_LDS_SQ_CTRL_REG__SQ_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU0_LDS_SQ_CTRL_REG__SQ_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU0_LDS_SQ_CTRL_REG__SQ_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU0_TA_SQC_CTRL_REG__TA_MASK 0x7f
+#define CGTS_CU0_TA_SQC_CTRL_REG__TA__SHIFT 0x0
+#define CGTS_CU0_TA_SQC_CTRL_REG__TA_OVERRIDE_MASK 0x80
+#define CGTS_CU0_TA_SQC_CTRL_REG__TA_OVERRIDE__SHIFT 0x7
+#define CGTS_CU0_TA_SQC_CTRL_REG__TA_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU0_TA_SQC_CTRL_REG__TA_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU0_TA_SQC_CTRL_REG__TA_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU0_TA_SQC_CTRL_REG__TA_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU0_TA_SQC_CTRL_REG__TA_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU0_TA_SQC_CTRL_REG__TA_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU0_TA_SQC_CTRL_REG__SQC_MASK 0x7f0000
+#define CGTS_CU0_TA_SQC_CTRL_REG__SQC__SHIFT 0x10
+#define CGTS_CU0_TA_SQC_CTRL_REG__SQC_OVERRIDE_MASK 0x800000
+#define CGTS_CU0_TA_SQC_CTRL_REG__SQC_OVERRIDE__SHIFT 0x17
+#define CGTS_CU0_TA_SQC_CTRL_REG__SQC_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU0_TA_SQC_CTRL_REG__SQC_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU0_TA_SQC_CTRL_REG__SQC_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU0_TA_SQC_CTRL_REG__SQC_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU0_TA_SQC_CTRL_REG__SQC_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU0_TA_SQC_CTRL_REG__SQC_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU0_SP1_CTRL_REG__SP10_MASK 0x7f
+#define CGTS_CU0_SP1_CTRL_REG__SP10__SHIFT 0x0
+#define CGTS_CU0_SP1_CTRL_REG__SP10_OVERRIDE_MASK 0x80
+#define CGTS_CU0_SP1_CTRL_REG__SP10_OVERRIDE__SHIFT 0x7
+#define CGTS_CU0_SP1_CTRL_REG__SP10_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU0_SP1_CTRL_REG__SP10_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU0_SP1_CTRL_REG__SP10_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU0_SP1_CTRL_REG__SP10_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU0_SP1_CTRL_REG__SP10_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU0_SP1_CTRL_REG__SP10_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU0_SP1_CTRL_REG__SP11_MASK 0x7f0000
+#define CGTS_CU0_SP1_CTRL_REG__SP11__SHIFT 0x10
+#define CGTS_CU0_SP1_CTRL_REG__SP11_OVERRIDE_MASK 0x800000
+#define CGTS_CU0_SP1_CTRL_REG__SP11_OVERRIDE__SHIFT 0x17
+#define CGTS_CU0_SP1_CTRL_REG__SP11_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU0_SP1_CTRL_REG__SP11_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU0_SP1_CTRL_REG__SP11_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU0_SP1_CTRL_REG__SP11_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU0_SP1_CTRL_REG__SP11_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU0_SP1_CTRL_REG__SP11_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU0_TD_TCP_CTRL_REG__TD_MASK 0x7f
+#define CGTS_CU0_TD_TCP_CTRL_REG__TD__SHIFT 0x0
+#define CGTS_CU0_TD_TCP_CTRL_REG__TD_OVERRIDE_MASK 0x80
+#define CGTS_CU0_TD_TCP_CTRL_REG__TD_OVERRIDE__SHIFT 0x7
+#define CGTS_CU0_TD_TCP_CTRL_REG__TD_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU0_TD_TCP_CTRL_REG__TD_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU0_TD_TCP_CTRL_REG__TD_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU0_TD_TCP_CTRL_REG__TD_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU0_TD_TCP_CTRL_REG__TD_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU0_TD_TCP_CTRL_REG__TD_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU0_TD_TCP_CTRL_REG__TCP_MASK 0x7f0000
+#define CGTS_CU0_TD_TCP_CTRL_REG__TCP__SHIFT 0x10
+#define CGTS_CU0_TD_TCP_CTRL_REG__TCP_OVERRIDE_MASK 0x800000
+#define CGTS_CU0_TD_TCP_CTRL_REG__TCP_OVERRIDE__SHIFT 0x17
+#define CGTS_CU0_TD_TCP_CTRL_REG__TCP_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU0_TD_TCP_CTRL_REG__TCP_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU0_TD_TCP_CTRL_REG__TCP_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU0_TD_TCP_CTRL_REG__TCP_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU0_TD_TCP_CTRL_REG__TCP_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU0_TD_TCP_CTRL_REG__TCP_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU1_SP0_CTRL_REG__SP00_MASK 0x7f
+#define CGTS_CU1_SP0_CTRL_REG__SP00__SHIFT 0x0
+#define CGTS_CU1_SP0_CTRL_REG__SP00_OVERRIDE_MASK 0x80
+#define CGTS_CU1_SP0_CTRL_REG__SP00_OVERRIDE__SHIFT 0x7
+#define CGTS_CU1_SP0_CTRL_REG__SP00_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU1_SP0_CTRL_REG__SP00_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU1_SP0_CTRL_REG__SP00_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU1_SP0_CTRL_REG__SP00_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU1_SP0_CTRL_REG__SP00_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU1_SP0_CTRL_REG__SP00_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU1_SP0_CTRL_REG__SP01_MASK 0x7f0000
+#define CGTS_CU1_SP0_CTRL_REG__SP01__SHIFT 0x10
+#define CGTS_CU1_SP0_CTRL_REG__SP01_OVERRIDE_MASK 0x800000
+#define CGTS_CU1_SP0_CTRL_REG__SP01_OVERRIDE__SHIFT 0x17
+#define CGTS_CU1_SP0_CTRL_REG__SP01_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU1_SP0_CTRL_REG__SP01_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU1_SP0_CTRL_REG__SP01_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU1_SP0_CTRL_REG__SP01_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU1_SP0_CTRL_REG__SP01_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU1_SP0_CTRL_REG__SP01_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU1_LDS_SQ_CTRL_REG__LDS_MASK 0x7f
+#define CGTS_CU1_LDS_SQ_CTRL_REG__LDS__SHIFT 0x0
+#define CGTS_CU1_LDS_SQ_CTRL_REG__LDS_OVERRIDE_MASK 0x80
+#define CGTS_CU1_LDS_SQ_CTRL_REG__LDS_OVERRIDE__SHIFT 0x7
+#define CGTS_CU1_LDS_SQ_CTRL_REG__LDS_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU1_LDS_SQ_CTRL_REG__LDS_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU1_LDS_SQ_CTRL_REG__LDS_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU1_LDS_SQ_CTRL_REG__LDS_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU1_LDS_SQ_CTRL_REG__LDS_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU1_LDS_SQ_CTRL_REG__LDS_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU1_LDS_SQ_CTRL_REG__SQ_MASK 0x7f0000
+#define CGTS_CU1_LDS_SQ_CTRL_REG__SQ__SHIFT 0x10
+#define CGTS_CU1_LDS_SQ_CTRL_REG__SQ_OVERRIDE_MASK 0x800000
+#define CGTS_CU1_LDS_SQ_CTRL_REG__SQ_OVERRIDE__SHIFT 0x17
+#define CGTS_CU1_LDS_SQ_CTRL_REG__SQ_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU1_LDS_SQ_CTRL_REG__SQ_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU1_LDS_SQ_CTRL_REG__SQ_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU1_LDS_SQ_CTRL_REG__SQ_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU1_LDS_SQ_CTRL_REG__SQ_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU1_LDS_SQ_CTRL_REG__SQ_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU1_TA_CTRL_REG__TA_MASK 0x7f
+#define CGTS_CU1_TA_CTRL_REG__TA__SHIFT 0x0
+#define CGTS_CU1_TA_CTRL_REG__TA_OVERRIDE_MASK 0x80
+#define CGTS_CU1_TA_CTRL_REG__TA_OVERRIDE__SHIFT 0x7
+#define CGTS_CU1_TA_CTRL_REG__TA_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU1_TA_CTRL_REG__TA_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU1_TA_CTRL_REG__TA_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU1_TA_CTRL_REG__TA_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU1_TA_CTRL_REG__TA_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU1_TA_CTRL_REG__TA_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU1_SP1_CTRL_REG__SP10_MASK 0x7f
+#define CGTS_CU1_SP1_CTRL_REG__SP10__SHIFT 0x0
+#define CGTS_CU1_SP1_CTRL_REG__SP10_OVERRIDE_MASK 0x80
+#define CGTS_CU1_SP1_CTRL_REG__SP10_OVERRIDE__SHIFT 0x7
+#define CGTS_CU1_SP1_CTRL_REG__SP10_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU1_SP1_CTRL_REG__SP10_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU1_SP1_CTRL_REG__SP10_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU1_SP1_CTRL_REG__SP10_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU1_SP1_CTRL_REG__SP10_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU1_SP1_CTRL_REG__SP10_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU1_SP1_CTRL_REG__SP11_MASK 0x7f0000
+#define CGTS_CU1_SP1_CTRL_REG__SP11__SHIFT 0x10
+#define CGTS_CU1_SP1_CTRL_REG__SP11_OVERRIDE_MASK 0x800000
+#define CGTS_CU1_SP1_CTRL_REG__SP11_OVERRIDE__SHIFT 0x17
+#define CGTS_CU1_SP1_CTRL_REG__SP11_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU1_SP1_CTRL_REG__SP11_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU1_SP1_CTRL_REG__SP11_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU1_SP1_CTRL_REG__SP11_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU1_SP1_CTRL_REG__SP11_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU1_SP1_CTRL_REG__SP11_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU1_TD_TCP_CTRL_REG__TD_MASK 0x7f
+#define CGTS_CU1_TD_TCP_CTRL_REG__TD__SHIFT 0x0
+#define CGTS_CU1_TD_TCP_CTRL_REG__TD_OVERRIDE_MASK 0x80
+#define CGTS_CU1_TD_TCP_CTRL_REG__TD_OVERRIDE__SHIFT 0x7
+#define CGTS_CU1_TD_TCP_CTRL_REG__TD_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU1_TD_TCP_CTRL_REG__TD_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU1_TD_TCP_CTRL_REG__TD_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU1_TD_TCP_CTRL_REG__TD_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU1_TD_TCP_CTRL_REG__TD_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU1_TD_TCP_CTRL_REG__TD_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU1_TD_TCP_CTRL_REG__TCP_MASK 0x7f0000
+#define CGTS_CU1_TD_TCP_CTRL_REG__TCP__SHIFT 0x10
+#define CGTS_CU1_TD_TCP_CTRL_REG__TCP_OVERRIDE_MASK 0x800000
+#define CGTS_CU1_TD_TCP_CTRL_REG__TCP_OVERRIDE__SHIFT 0x17
+#define CGTS_CU1_TD_TCP_CTRL_REG__TCP_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU1_TD_TCP_CTRL_REG__TCP_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU1_TD_TCP_CTRL_REG__TCP_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU1_TD_TCP_CTRL_REG__TCP_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU1_TD_TCP_CTRL_REG__TCP_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU1_TD_TCP_CTRL_REG__TCP_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU2_SP0_CTRL_REG__SP00_MASK 0x7f
+#define CGTS_CU2_SP0_CTRL_REG__SP00__SHIFT 0x0
+#define CGTS_CU2_SP0_CTRL_REG__SP00_OVERRIDE_MASK 0x80
+#define CGTS_CU2_SP0_CTRL_REG__SP00_OVERRIDE__SHIFT 0x7
+#define CGTS_CU2_SP0_CTRL_REG__SP00_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU2_SP0_CTRL_REG__SP00_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU2_SP0_CTRL_REG__SP00_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU2_SP0_CTRL_REG__SP00_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU2_SP0_CTRL_REG__SP00_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU2_SP0_CTRL_REG__SP00_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU2_SP0_CTRL_REG__SP01_MASK 0x7f0000
+#define CGTS_CU2_SP0_CTRL_REG__SP01__SHIFT 0x10
+#define CGTS_CU2_SP0_CTRL_REG__SP01_OVERRIDE_MASK 0x800000
+#define CGTS_CU2_SP0_CTRL_REG__SP01_OVERRIDE__SHIFT 0x17
+#define CGTS_CU2_SP0_CTRL_REG__SP01_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU2_SP0_CTRL_REG__SP01_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU2_SP0_CTRL_REG__SP01_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU2_SP0_CTRL_REG__SP01_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU2_SP0_CTRL_REG__SP01_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU2_SP0_CTRL_REG__SP01_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU2_LDS_SQ_CTRL_REG__LDS_MASK 0x7f
+#define CGTS_CU2_LDS_SQ_CTRL_REG__LDS__SHIFT 0x0
+#define CGTS_CU2_LDS_SQ_CTRL_REG__LDS_OVERRIDE_MASK 0x80
+#define CGTS_CU2_LDS_SQ_CTRL_REG__LDS_OVERRIDE__SHIFT 0x7
+#define CGTS_CU2_LDS_SQ_CTRL_REG__LDS_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU2_LDS_SQ_CTRL_REG__LDS_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU2_LDS_SQ_CTRL_REG__LDS_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU2_LDS_SQ_CTRL_REG__LDS_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU2_LDS_SQ_CTRL_REG__LDS_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU2_LDS_SQ_CTRL_REG__LDS_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU2_LDS_SQ_CTRL_REG__SQ_MASK 0x7f0000
+#define CGTS_CU2_LDS_SQ_CTRL_REG__SQ__SHIFT 0x10
+#define CGTS_CU2_LDS_SQ_CTRL_REG__SQ_OVERRIDE_MASK 0x800000
+#define CGTS_CU2_LDS_SQ_CTRL_REG__SQ_OVERRIDE__SHIFT 0x17
+#define CGTS_CU2_LDS_SQ_CTRL_REG__SQ_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU2_LDS_SQ_CTRL_REG__SQ_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU2_LDS_SQ_CTRL_REG__SQ_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU2_LDS_SQ_CTRL_REG__SQ_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU2_LDS_SQ_CTRL_REG__SQ_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU2_LDS_SQ_CTRL_REG__SQ_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU2_TA_CTRL_REG__TA_MASK 0x7f
+#define CGTS_CU2_TA_CTRL_REG__TA__SHIFT 0x0
+#define CGTS_CU2_TA_CTRL_REG__TA_OVERRIDE_MASK 0x80
+#define CGTS_CU2_TA_CTRL_REG__TA_OVERRIDE__SHIFT 0x7
+#define CGTS_CU2_TA_CTRL_REG__TA_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU2_TA_CTRL_REG__TA_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU2_TA_CTRL_REG__TA_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU2_TA_CTRL_REG__TA_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU2_TA_CTRL_REG__TA_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU2_TA_CTRL_REG__TA_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU2_SP1_CTRL_REG__SP10_MASK 0x7f
+#define CGTS_CU2_SP1_CTRL_REG__SP10__SHIFT 0x0
+#define CGTS_CU2_SP1_CTRL_REG__SP10_OVERRIDE_MASK 0x80
+#define CGTS_CU2_SP1_CTRL_REG__SP10_OVERRIDE__SHIFT 0x7
+#define CGTS_CU2_SP1_CTRL_REG__SP10_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU2_SP1_CTRL_REG__SP10_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU2_SP1_CTRL_REG__SP10_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU2_SP1_CTRL_REG__SP10_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU2_SP1_CTRL_REG__SP10_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU2_SP1_CTRL_REG__SP10_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU2_SP1_CTRL_REG__SP11_MASK 0x7f0000
+#define CGTS_CU2_SP1_CTRL_REG__SP11__SHIFT 0x10
+#define CGTS_CU2_SP1_CTRL_REG__SP11_OVERRIDE_MASK 0x800000
+#define CGTS_CU2_SP1_CTRL_REG__SP11_OVERRIDE__SHIFT 0x17
+#define CGTS_CU2_SP1_CTRL_REG__SP11_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU2_SP1_CTRL_REG__SP11_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU2_SP1_CTRL_REG__SP11_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU2_SP1_CTRL_REG__SP11_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU2_SP1_CTRL_REG__SP11_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU2_SP1_CTRL_REG__SP11_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU2_TD_TCP_CTRL_REG__TD_MASK 0x7f
+#define CGTS_CU2_TD_TCP_CTRL_REG__TD__SHIFT 0x0
+#define CGTS_CU2_TD_TCP_CTRL_REG__TD_OVERRIDE_MASK 0x80
+#define CGTS_CU2_TD_TCP_CTRL_REG__TD_OVERRIDE__SHIFT 0x7
+#define CGTS_CU2_TD_TCP_CTRL_REG__TD_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU2_TD_TCP_CTRL_REG__TD_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU2_TD_TCP_CTRL_REG__TD_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU2_TD_TCP_CTRL_REG__TD_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU2_TD_TCP_CTRL_REG__TD_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU2_TD_TCP_CTRL_REG__TD_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU2_TD_TCP_CTRL_REG__TCP_MASK 0x7f0000
+#define CGTS_CU2_TD_TCP_CTRL_REG__TCP__SHIFT 0x10
+#define CGTS_CU2_TD_TCP_CTRL_REG__TCP_OVERRIDE_MASK 0x800000
+#define CGTS_CU2_TD_TCP_CTRL_REG__TCP_OVERRIDE__SHIFT 0x17
+#define CGTS_CU2_TD_TCP_CTRL_REG__TCP_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU2_TD_TCP_CTRL_REG__TCP_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU2_TD_TCP_CTRL_REG__TCP_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU2_TD_TCP_CTRL_REG__TCP_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU2_TD_TCP_CTRL_REG__TCP_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU2_TD_TCP_CTRL_REG__TCP_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU3_SP0_CTRL_REG__SP00_MASK 0x7f
+#define CGTS_CU3_SP0_CTRL_REG__SP00__SHIFT 0x0
+#define CGTS_CU3_SP0_CTRL_REG__SP00_OVERRIDE_MASK 0x80
+#define CGTS_CU3_SP0_CTRL_REG__SP00_OVERRIDE__SHIFT 0x7
+#define CGTS_CU3_SP0_CTRL_REG__SP00_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU3_SP0_CTRL_REG__SP00_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU3_SP0_CTRL_REG__SP00_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU3_SP0_CTRL_REG__SP00_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU3_SP0_CTRL_REG__SP00_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU3_SP0_CTRL_REG__SP00_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU3_SP0_CTRL_REG__SP01_MASK 0x7f0000
+#define CGTS_CU3_SP0_CTRL_REG__SP01__SHIFT 0x10
+#define CGTS_CU3_SP0_CTRL_REG__SP01_OVERRIDE_MASK 0x800000
+#define CGTS_CU3_SP0_CTRL_REG__SP01_OVERRIDE__SHIFT 0x17
+#define CGTS_CU3_SP0_CTRL_REG__SP01_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU3_SP0_CTRL_REG__SP01_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU3_SP0_CTRL_REG__SP01_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU3_SP0_CTRL_REG__SP01_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU3_SP0_CTRL_REG__SP01_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU3_SP0_CTRL_REG__SP01_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU3_LDS_SQ_CTRL_REG__LDS_MASK 0x7f
+#define CGTS_CU3_LDS_SQ_CTRL_REG__LDS__SHIFT 0x0
+#define CGTS_CU3_LDS_SQ_CTRL_REG__LDS_OVERRIDE_MASK 0x80
+#define CGTS_CU3_LDS_SQ_CTRL_REG__LDS_OVERRIDE__SHIFT 0x7
+#define CGTS_CU3_LDS_SQ_CTRL_REG__LDS_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU3_LDS_SQ_CTRL_REG__LDS_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU3_LDS_SQ_CTRL_REG__LDS_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU3_LDS_SQ_CTRL_REG__LDS_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU3_LDS_SQ_CTRL_REG__LDS_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU3_LDS_SQ_CTRL_REG__LDS_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU3_LDS_SQ_CTRL_REG__SQ_MASK 0x7f0000
+#define CGTS_CU3_LDS_SQ_CTRL_REG__SQ__SHIFT 0x10
+#define CGTS_CU3_LDS_SQ_CTRL_REG__SQ_OVERRIDE_MASK 0x800000
+#define CGTS_CU3_LDS_SQ_CTRL_REG__SQ_OVERRIDE__SHIFT 0x17
+#define CGTS_CU3_LDS_SQ_CTRL_REG__SQ_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU3_LDS_SQ_CTRL_REG__SQ_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU3_LDS_SQ_CTRL_REG__SQ_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU3_LDS_SQ_CTRL_REG__SQ_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU3_LDS_SQ_CTRL_REG__SQ_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU3_LDS_SQ_CTRL_REG__SQ_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU3_TA_CTRL_REG__TA_MASK 0x7f
+#define CGTS_CU3_TA_CTRL_REG__TA__SHIFT 0x0
+#define CGTS_CU3_TA_CTRL_REG__TA_OVERRIDE_MASK 0x80
+#define CGTS_CU3_TA_CTRL_REG__TA_OVERRIDE__SHIFT 0x7
+#define CGTS_CU3_TA_CTRL_REG__TA_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU3_TA_CTRL_REG__TA_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU3_TA_CTRL_REG__TA_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU3_TA_CTRL_REG__TA_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU3_TA_CTRL_REG__TA_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU3_TA_CTRL_REG__TA_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU3_SP1_CTRL_REG__SP10_MASK 0x7f
+#define CGTS_CU3_SP1_CTRL_REG__SP10__SHIFT 0x0
+#define CGTS_CU3_SP1_CTRL_REG__SP10_OVERRIDE_MASK 0x80
+#define CGTS_CU3_SP1_CTRL_REG__SP10_OVERRIDE__SHIFT 0x7
+#define CGTS_CU3_SP1_CTRL_REG__SP10_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU3_SP1_CTRL_REG__SP10_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU3_SP1_CTRL_REG__SP10_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU3_SP1_CTRL_REG__SP10_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU3_SP1_CTRL_REG__SP10_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU3_SP1_CTRL_REG__SP10_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU3_SP1_CTRL_REG__SP11_MASK 0x7f0000
+#define CGTS_CU3_SP1_CTRL_REG__SP11__SHIFT 0x10
+#define CGTS_CU3_SP1_CTRL_REG__SP11_OVERRIDE_MASK 0x800000
+#define CGTS_CU3_SP1_CTRL_REG__SP11_OVERRIDE__SHIFT 0x17
+#define CGTS_CU3_SP1_CTRL_REG__SP11_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU3_SP1_CTRL_REG__SP11_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU3_SP1_CTRL_REG__SP11_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU3_SP1_CTRL_REG__SP11_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU3_SP1_CTRL_REG__SP11_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU3_SP1_CTRL_REG__SP11_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU3_TD_TCP_CTRL_REG__TD_MASK 0x7f
+#define CGTS_CU3_TD_TCP_CTRL_REG__TD__SHIFT 0x0
+#define CGTS_CU3_TD_TCP_CTRL_REG__TD_OVERRIDE_MASK 0x80
+#define CGTS_CU3_TD_TCP_CTRL_REG__TD_OVERRIDE__SHIFT 0x7
+#define CGTS_CU3_TD_TCP_CTRL_REG__TD_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU3_TD_TCP_CTRL_REG__TD_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU3_TD_TCP_CTRL_REG__TD_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU3_TD_TCP_CTRL_REG__TD_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU3_TD_TCP_CTRL_REG__TD_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU3_TD_TCP_CTRL_REG__TD_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU3_TD_TCP_CTRL_REG__TCP_MASK 0x7f0000
+#define CGTS_CU3_TD_TCP_CTRL_REG__TCP__SHIFT 0x10
+#define CGTS_CU3_TD_TCP_CTRL_REG__TCP_OVERRIDE_MASK 0x800000
+#define CGTS_CU3_TD_TCP_CTRL_REG__TCP_OVERRIDE__SHIFT 0x17
+#define CGTS_CU3_TD_TCP_CTRL_REG__TCP_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU3_TD_TCP_CTRL_REG__TCP_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU3_TD_TCP_CTRL_REG__TCP_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU3_TD_TCP_CTRL_REG__TCP_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU3_TD_TCP_CTRL_REG__TCP_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU3_TD_TCP_CTRL_REG__TCP_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU4_SP0_CTRL_REG__SP00_MASK 0x7f
+#define CGTS_CU4_SP0_CTRL_REG__SP00__SHIFT 0x0
+#define CGTS_CU4_SP0_CTRL_REG__SP00_OVERRIDE_MASK 0x80
+#define CGTS_CU4_SP0_CTRL_REG__SP00_OVERRIDE__SHIFT 0x7
+#define CGTS_CU4_SP0_CTRL_REG__SP00_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU4_SP0_CTRL_REG__SP00_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU4_SP0_CTRL_REG__SP00_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU4_SP0_CTRL_REG__SP00_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU4_SP0_CTRL_REG__SP00_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU4_SP0_CTRL_REG__SP00_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU4_SP0_CTRL_REG__SP01_MASK 0x7f0000
+#define CGTS_CU4_SP0_CTRL_REG__SP01__SHIFT 0x10
+#define CGTS_CU4_SP0_CTRL_REG__SP01_OVERRIDE_MASK 0x800000
+#define CGTS_CU4_SP0_CTRL_REG__SP01_OVERRIDE__SHIFT 0x17
+#define CGTS_CU4_SP0_CTRL_REG__SP01_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU4_SP0_CTRL_REG__SP01_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU4_SP0_CTRL_REG__SP01_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU4_SP0_CTRL_REG__SP01_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU4_SP0_CTRL_REG__SP01_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU4_SP0_CTRL_REG__SP01_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU4_LDS_SQ_CTRL_REG__LDS_MASK 0x7f
+#define CGTS_CU4_LDS_SQ_CTRL_REG__LDS__SHIFT 0x0
+#define CGTS_CU4_LDS_SQ_CTRL_REG__LDS_OVERRIDE_MASK 0x80
+#define CGTS_CU4_LDS_SQ_CTRL_REG__LDS_OVERRIDE__SHIFT 0x7
+#define CGTS_CU4_LDS_SQ_CTRL_REG__LDS_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU4_LDS_SQ_CTRL_REG__LDS_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU4_LDS_SQ_CTRL_REG__LDS_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU4_LDS_SQ_CTRL_REG__LDS_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU4_LDS_SQ_CTRL_REG__LDS_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU4_LDS_SQ_CTRL_REG__LDS_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU4_LDS_SQ_CTRL_REG__SQ_MASK 0x7f0000
+#define CGTS_CU4_LDS_SQ_CTRL_REG__SQ__SHIFT 0x10
+#define CGTS_CU4_LDS_SQ_CTRL_REG__SQ_OVERRIDE_MASK 0x800000
+#define CGTS_CU4_LDS_SQ_CTRL_REG__SQ_OVERRIDE__SHIFT 0x17
+#define CGTS_CU4_LDS_SQ_CTRL_REG__SQ_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU4_LDS_SQ_CTRL_REG__SQ_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU4_LDS_SQ_CTRL_REG__SQ_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU4_LDS_SQ_CTRL_REG__SQ_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU4_LDS_SQ_CTRL_REG__SQ_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU4_LDS_SQ_CTRL_REG__SQ_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU4_TA_SQC_CTRL_REG__TA_MASK 0x7f
+#define CGTS_CU4_TA_SQC_CTRL_REG__TA__SHIFT 0x0
+#define CGTS_CU4_TA_SQC_CTRL_REG__TA_OVERRIDE_MASK 0x80
+#define CGTS_CU4_TA_SQC_CTRL_REG__TA_OVERRIDE__SHIFT 0x7
+#define CGTS_CU4_TA_SQC_CTRL_REG__TA_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU4_TA_SQC_CTRL_REG__TA_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU4_TA_SQC_CTRL_REG__TA_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU4_TA_SQC_CTRL_REG__TA_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU4_TA_SQC_CTRL_REG__TA_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU4_TA_SQC_CTRL_REG__TA_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU4_TA_SQC_CTRL_REG__SQC_MASK 0x7f0000
+#define CGTS_CU4_TA_SQC_CTRL_REG__SQC__SHIFT 0x10
+#define CGTS_CU4_TA_SQC_CTRL_REG__SQC_OVERRIDE_MASK 0x800000
+#define CGTS_CU4_TA_SQC_CTRL_REG__SQC_OVERRIDE__SHIFT 0x17
+#define CGTS_CU4_TA_SQC_CTRL_REG__SQC_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU4_TA_SQC_CTRL_REG__SQC_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU4_TA_SQC_CTRL_REG__SQC_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU4_TA_SQC_CTRL_REG__SQC_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU4_TA_SQC_CTRL_REG__SQC_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU4_TA_SQC_CTRL_REG__SQC_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU4_SP1_CTRL_REG__SP10_MASK 0x7f
+#define CGTS_CU4_SP1_CTRL_REG__SP10__SHIFT 0x0
+#define CGTS_CU4_SP1_CTRL_REG__SP10_OVERRIDE_MASK 0x80
+#define CGTS_CU4_SP1_CTRL_REG__SP10_OVERRIDE__SHIFT 0x7
+#define CGTS_CU4_SP1_CTRL_REG__SP10_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU4_SP1_CTRL_REG__SP10_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU4_SP1_CTRL_REG__SP10_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU4_SP1_CTRL_REG__SP10_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU4_SP1_CTRL_REG__SP10_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU4_SP1_CTRL_REG__SP10_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU4_SP1_CTRL_REG__SP11_MASK 0x7f0000
+#define CGTS_CU4_SP1_CTRL_REG__SP11__SHIFT 0x10
+#define CGTS_CU4_SP1_CTRL_REG__SP11_OVERRIDE_MASK 0x800000
+#define CGTS_CU4_SP1_CTRL_REG__SP11_OVERRIDE__SHIFT 0x17
+#define CGTS_CU4_SP1_CTRL_REG__SP11_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU4_SP1_CTRL_REG__SP11_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU4_SP1_CTRL_REG__SP11_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU4_SP1_CTRL_REG__SP11_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU4_SP1_CTRL_REG__SP11_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU4_SP1_CTRL_REG__SP11_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU4_TD_TCP_CTRL_REG__TD_MASK 0x7f
+#define CGTS_CU4_TD_TCP_CTRL_REG__TD__SHIFT 0x0
+#define CGTS_CU4_TD_TCP_CTRL_REG__TD_OVERRIDE_MASK 0x80
+#define CGTS_CU4_TD_TCP_CTRL_REG__TD_OVERRIDE__SHIFT 0x7
+#define CGTS_CU4_TD_TCP_CTRL_REG__TD_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU4_TD_TCP_CTRL_REG__TD_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU4_TD_TCP_CTRL_REG__TD_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU4_TD_TCP_CTRL_REG__TD_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU4_TD_TCP_CTRL_REG__TD_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU4_TD_TCP_CTRL_REG__TD_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU4_TD_TCP_CTRL_REG__TCP_MASK 0x7f0000
+#define CGTS_CU4_TD_TCP_CTRL_REG__TCP__SHIFT 0x10
+#define CGTS_CU4_TD_TCP_CTRL_REG__TCP_OVERRIDE_MASK 0x800000
+#define CGTS_CU4_TD_TCP_CTRL_REG__TCP_OVERRIDE__SHIFT 0x17
+#define CGTS_CU4_TD_TCP_CTRL_REG__TCP_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU4_TD_TCP_CTRL_REG__TCP_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU4_TD_TCP_CTRL_REG__TCP_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU4_TD_TCP_CTRL_REG__TCP_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU4_TD_TCP_CTRL_REG__TCP_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU4_TD_TCP_CTRL_REG__TCP_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU5_SP0_CTRL_REG__SP00_MASK 0x7f
+#define CGTS_CU5_SP0_CTRL_REG__SP00__SHIFT 0x0
+#define CGTS_CU5_SP0_CTRL_REG__SP00_OVERRIDE_MASK 0x80
+#define CGTS_CU5_SP0_CTRL_REG__SP00_OVERRIDE__SHIFT 0x7
+#define CGTS_CU5_SP0_CTRL_REG__SP00_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU5_SP0_CTRL_REG__SP00_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU5_SP0_CTRL_REG__SP00_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU5_SP0_CTRL_REG__SP00_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU5_SP0_CTRL_REG__SP00_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU5_SP0_CTRL_REG__SP00_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU5_SP0_CTRL_REG__SP01_MASK 0x7f0000
+#define CGTS_CU5_SP0_CTRL_REG__SP01__SHIFT 0x10
+#define CGTS_CU5_SP0_CTRL_REG__SP01_OVERRIDE_MASK 0x800000
+#define CGTS_CU5_SP0_CTRL_REG__SP01_OVERRIDE__SHIFT 0x17
+#define CGTS_CU5_SP0_CTRL_REG__SP01_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU5_SP0_CTRL_REG__SP01_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU5_SP0_CTRL_REG__SP01_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU5_SP0_CTRL_REG__SP01_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU5_SP0_CTRL_REG__SP01_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU5_SP0_CTRL_REG__SP01_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU5_LDS_SQ_CTRL_REG__LDS_MASK 0x7f
+#define CGTS_CU5_LDS_SQ_CTRL_REG__LDS__SHIFT 0x0
+#define CGTS_CU5_LDS_SQ_CTRL_REG__LDS_OVERRIDE_MASK 0x80
+#define CGTS_CU5_LDS_SQ_CTRL_REG__LDS_OVERRIDE__SHIFT 0x7
+#define CGTS_CU5_LDS_SQ_CTRL_REG__LDS_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU5_LDS_SQ_CTRL_REG__LDS_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU5_LDS_SQ_CTRL_REG__LDS_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU5_LDS_SQ_CTRL_REG__LDS_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU5_LDS_SQ_CTRL_REG__LDS_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU5_LDS_SQ_CTRL_REG__LDS_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU5_LDS_SQ_CTRL_REG__SQ_MASK 0x7f0000
+#define CGTS_CU5_LDS_SQ_CTRL_REG__SQ__SHIFT 0x10
+#define CGTS_CU5_LDS_SQ_CTRL_REG__SQ_OVERRIDE_MASK 0x800000
+#define CGTS_CU5_LDS_SQ_CTRL_REG__SQ_OVERRIDE__SHIFT 0x17
+#define CGTS_CU5_LDS_SQ_CTRL_REG__SQ_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU5_LDS_SQ_CTRL_REG__SQ_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU5_LDS_SQ_CTRL_REG__SQ_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU5_LDS_SQ_CTRL_REG__SQ_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU5_LDS_SQ_CTRL_REG__SQ_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU5_LDS_SQ_CTRL_REG__SQ_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU5_TA_CTRL_REG__TA_MASK 0x7f
+#define CGTS_CU5_TA_CTRL_REG__TA__SHIFT 0x0
+#define CGTS_CU5_TA_CTRL_REG__TA_OVERRIDE_MASK 0x80
+#define CGTS_CU5_TA_CTRL_REG__TA_OVERRIDE__SHIFT 0x7
+#define CGTS_CU5_TA_CTRL_REG__TA_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU5_TA_CTRL_REG__TA_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU5_TA_CTRL_REG__TA_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU5_TA_CTRL_REG__TA_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU5_TA_CTRL_REG__TA_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU5_TA_CTRL_REG__TA_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU5_SP1_CTRL_REG__SP10_MASK 0x7f
+#define CGTS_CU5_SP1_CTRL_REG__SP10__SHIFT 0x0
+#define CGTS_CU5_SP1_CTRL_REG__SP10_OVERRIDE_MASK 0x80
+#define CGTS_CU5_SP1_CTRL_REG__SP10_OVERRIDE__SHIFT 0x7
+#define CGTS_CU5_SP1_CTRL_REG__SP10_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU5_SP1_CTRL_REG__SP10_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU5_SP1_CTRL_REG__SP10_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU5_SP1_CTRL_REG__SP10_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU5_SP1_CTRL_REG__SP10_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU5_SP1_CTRL_REG__SP10_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU5_SP1_CTRL_REG__SP11_MASK 0x7f0000
+#define CGTS_CU5_SP1_CTRL_REG__SP11__SHIFT 0x10
+#define CGTS_CU5_SP1_CTRL_REG__SP11_OVERRIDE_MASK 0x800000
+#define CGTS_CU5_SP1_CTRL_REG__SP11_OVERRIDE__SHIFT 0x17
+#define CGTS_CU5_SP1_CTRL_REG__SP11_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU5_SP1_CTRL_REG__SP11_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU5_SP1_CTRL_REG__SP11_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU5_SP1_CTRL_REG__SP11_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU5_SP1_CTRL_REG__SP11_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU5_SP1_CTRL_REG__SP11_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU5_TD_TCP_CTRL_REG__TD_MASK 0x7f
+#define CGTS_CU5_TD_TCP_CTRL_REG__TD__SHIFT 0x0
+#define CGTS_CU5_TD_TCP_CTRL_REG__TD_OVERRIDE_MASK 0x80
+#define CGTS_CU5_TD_TCP_CTRL_REG__TD_OVERRIDE__SHIFT 0x7
+#define CGTS_CU5_TD_TCP_CTRL_REG__TD_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU5_TD_TCP_CTRL_REG__TD_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU5_TD_TCP_CTRL_REG__TD_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU5_TD_TCP_CTRL_REG__TD_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU5_TD_TCP_CTRL_REG__TD_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU5_TD_TCP_CTRL_REG__TD_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU5_TD_TCP_CTRL_REG__TCP_MASK 0x7f0000
+#define CGTS_CU5_TD_TCP_CTRL_REG__TCP__SHIFT 0x10
+#define CGTS_CU5_TD_TCP_CTRL_REG__TCP_OVERRIDE_MASK 0x800000
+#define CGTS_CU5_TD_TCP_CTRL_REG__TCP_OVERRIDE__SHIFT 0x17
+#define CGTS_CU5_TD_TCP_CTRL_REG__TCP_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU5_TD_TCP_CTRL_REG__TCP_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU5_TD_TCP_CTRL_REG__TCP_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU5_TD_TCP_CTRL_REG__TCP_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU5_TD_TCP_CTRL_REG__TCP_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU5_TD_TCP_CTRL_REG__TCP_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU6_SP0_CTRL_REG__SP00_MASK 0x7f
+#define CGTS_CU6_SP0_CTRL_REG__SP00__SHIFT 0x0
+#define CGTS_CU6_SP0_CTRL_REG__SP00_OVERRIDE_MASK 0x80
+#define CGTS_CU6_SP0_CTRL_REG__SP00_OVERRIDE__SHIFT 0x7
+#define CGTS_CU6_SP0_CTRL_REG__SP00_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU6_SP0_CTRL_REG__SP00_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU6_SP0_CTRL_REG__SP00_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU6_SP0_CTRL_REG__SP00_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU6_SP0_CTRL_REG__SP00_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU6_SP0_CTRL_REG__SP00_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU6_SP0_CTRL_REG__SP01_MASK 0x7f0000
+#define CGTS_CU6_SP0_CTRL_REG__SP01__SHIFT 0x10
+#define CGTS_CU6_SP0_CTRL_REG__SP01_OVERRIDE_MASK 0x800000
+#define CGTS_CU6_SP0_CTRL_REG__SP01_OVERRIDE__SHIFT 0x17
+#define CGTS_CU6_SP0_CTRL_REG__SP01_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU6_SP0_CTRL_REG__SP01_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU6_SP0_CTRL_REG__SP01_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU6_SP0_CTRL_REG__SP01_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU6_SP0_CTRL_REG__SP01_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU6_SP0_CTRL_REG__SP01_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU6_LDS_SQ_CTRL_REG__LDS_MASK 0x7f
+#define CGTS_CU6_LDS_SQ_CTRL_REG__LDS__SHIFT 0x0
+#define CGTS_CU6_LDS_SQ_CTRL_REG__LDS_OVERRIDE_MASK 0x80
+#define CGTS_CU6_LDS_SQ_CTRL_REG__LDS_OVERRIDE__SHIFT 0x7
+#define CGTS_CU6_LDS_SQ_CTRL_REG__LDS_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU6_LDS_SQ_CTRL_REG__LDS_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU6_LDS_SQ_CTRL_REG__LDS_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU6_LDS_SQ_CTRL_REG__LDS_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU6_LDS_SQ_CTRL_REG__LDS_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU6_LDS_SQ_CTRL_REG__LDS_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU6_LDS_SQ_CTRL_REG__SQ_MASK 0x7f0000
+#define CGTS_CU6_LDS_SQ_CTRL_REG__SQ__SHIFT 0x10
+#define CGTS_CU6_LDS_SQ_CTRL_REG__SQ_OVERRIDE_MASK 0x800000
+#define CGTS_CU6_LDS_SQ_CTRL_REG__SQ_OVERRIDE__SHIFT 0x17
+#define CGTS_CU6_LDS_SQ_CTRL_REG__SQ_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU6_LDS_SQ_CTRL_REG__SQ_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU6_LDS_SQ_CTRL_REG__SQ_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU6_LDS_SQ_CTRL_REG__SQ_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU6_LDS_SQ_CTRL_REG__SQ_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU6_LDS_SQ_CTRL_REG__SQ_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU6_TA_CTRL_REG__TA_MASK 0x7f
+#define CGTS_CU6_TA_CTRL_REG__TA__SHIFT 0x0
+#define CGTS_CU6_TA_CTRL_REG__TA_OVERRIDE_MASK 0x80
+#define CGTS_CU6_TA_CTRL_REG__TA_OVERRIDE__SHIFT 0x7
+#define CGTS_CU6_TA_CTRL_REG__TA_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU6_TA_CTRL_REG__TA_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU6_TA_CTRL_REG__TA_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU6_TA_CTRL_REG__TA_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU6_TA_CTRL_REG__TA_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU6_TA_CTRL_REG__TA_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU6_SP1_CTRL_REG__SP10_MASK 0x7f
+#define CGTS_CU6_SP1_CTRL_REG__SP10__SHIFT 0x0
+#define CGTS_CU6_SP1_CTRL_REG__SP10_OVERRIDE_MASK 0x80
+#define CGTS_CU6_SP1_CTRL_REG__SP10_OVERRIDE__SHIFT 0x7
+#define CGTS_CU6_SP1_CTRL_REG__SP10_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU6_SP1_CTRL_REG__SP10_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU6_SP1_CTRL_REG__SP10_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU6_SP1_CTRL_REG__SP10_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU6_SP1_CTRL_REG__SP10_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU6_SP1_CTRL_REG__SP10_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU6_SP1_CTRL_REG__SP11_MASK 0x7f0000
+#define CGTS_CU6_SP1_CTRL_REG__SP11__SHIFT 0x10
+#define CGTS_CU6_SP1_CTRL_REG__SP11_OVERRIDE_MASK 0x800000
+#define CGTS_CU6_SP1_CTRL_REG__SP11_OVERRIDE__SHIFT 0x17
+#define CGTS_CU6_SP1_CTRL_REG__SP11_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU6_SP1_CTRL_REG__SP11_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU6_SP1_CTRL_REG__SP11_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU6_SP1_CTRL_REG__SP11_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU6_SP1_CTRL_REG__SP11_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU6_SP1_CTRL_REG__SP11_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU6_TD_TCP_CTRL_REG__TD_MASK 0x7f
+#define CGTS_CU6_TD_TCP_CTRL_REG__TD__SHIFT 0x0
+#define CGTS_CU6_TD_TCP_CTRL_REG__TD_OVERRIDE_MASK 0x80
+#define CGTS_CU6_TD_TCP_CTRL_REG__TD_OVERRIDE__SHIFT 0x7
+#define CGTS_CU6_TD_TCP_CTRL_REG__TD_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU6_TD_TCP_CTRL_REG__TD_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU6_TD_TCP_CTRL_REG__TD_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU6_TD_TCP_CTRL_REG__TD_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU6_TD_TCP_CTRL_REG__TD_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU6_TD_TCP_CTRL_REG__TD_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU6_TD_TCP_CTRL_REG__TCP_MASK 0x7f0000
+#define CGTS_CU6_TD_TCP_CTRL_REG__TCP__SHIFT 0x10
+#define CGTS_CU6_TD_TCP_CTRL_REG__TCP_OVERRIDE_MASK 0x800000
+#define CGTS_CU6_TD_TCP_CTRL_REG__TCP_OVERRIDE__SHIFT 0x17
+#define CGTS_CU6_TD_TCP_CTRL_REG__TCP_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU6_TD_TCP_CTRL_REG__TCP_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU6_TD_TCP_CTRL_REG__TCP_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU6_TD_TCP_CTRL_REG__TCP_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU6_TD_TCP_CTRL_REG__TCP_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU6_TD_TCP_CTRL_REG__TCP_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU7_SP0_CTRL_REG__SP00_MASK 0x7f
+#define CGTS_CU7_SP0_CTRL_REG__SP00__SHIFT 0x0
+#define CGTS_CU7_SP0_CTRL_REG__SP00_OVERRIDE_MASK 0x80
+#define CGTS_CU7_SP0_CTRL_REG__SP00_OVERRIDE__SHIFT 0x7
+#define CGTS_CU7_SP0_CTRL_REG__SP00_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU7_SP0_CTRL_REG__SP00_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU7_SP0_CTRL_REG__SP00_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU7_SP0_CTRL_REG__SP00_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU7_SP0_CTRL_REG__SP00_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU7_SP0_CTRL_REG__SP00_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU7_SP0_CTRL_REG__SP01_MASK 0x7f0000
+#define CGTS_CU7_SP0_CTRL_REG__SP01__SHIFT 0x10
+#define CGTS_CU7_SP0_CTRL_REG__SP01_OVERRIDE_MASK 0x800000
+#define CGTS_CU7_SP0_CTRL_REG__SP01_OVERRIDE__SHIFT 0x17
+#define CGTS_CU7_SP0_CTRL_REG__SP01_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU7_SP0_CTRL_REG__SP01_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU7_SP0_CTRL_REG__SP01_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU7_SP0_CTRL_REG__SP01_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU7_SP0_CTRL_REG__SP01_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU7_SP0_CTRL_REG__SP01_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU7_LDS_SQ_CTRL_REG__LDS_MASK 0x7f
+#define CGTS_CU7_LDS_SQ_CTRL_REG__LDS__SHIFT 0x0
+#define CGTS_CU7_LDS_SQ_CTRL_REG__LDS_OVERRIDE_MASK 0x80
+#define CGTS_CU7_LDS_SQ_CTRL_REG__LDS_OVERRIDE__SHIFT 0x7
+#define CGTS_CU7_LDS_SQ_CTRL_REG__LDS_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU7_LDS_SQ_CTRL_REG__LDS_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU7_LDS_SQ_CTRL_REG__LDS_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU7_LDS_SQ_CTRL_REG__LDS_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU7_LDS_SQ_CTRL_REG__LDS_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU7_LDS_SQ_CTRL_REG__LDS_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU7_LDS_SQ_CTRL_REG__SQ_MASK 0x7f0000
+#define CGTS_CU7_LDS_SQ_CTRL_REG__SQ__SHIFT 0x10
+#define CGTS_CU7_LDS_SQ_CTRL_REG__SQ_OVERRIDE_MASK 0x800000
+#define CGTS_CU7_LDS_SQ_CTRL_REG__SQ_OVERRIDE__SHIFT 0x17
+#define CGTS_CU7_LDS_SQ_CTRL_REG__SQ_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU7_LDS_SQ_CTRL_REG__SQ_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU7_LDS_SQ_CTRL_REG__SQ_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU7_LDS_SQ_CTRL_REG__SQ_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU7_LDS_SQ_CTRL_REG__SQ_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU7_LDS_SQ_CTRL_REG__SQ_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU7_TA_CTRL_REG__TA_MASK 0x7f
+#define CGTS_CU7_TA_CTRL_REG__TA__SHIFT 0x0
+#define CGTS_CU7_TA_CTRL_REG__TA_OVERRIDE_MASK 0x80
+#define CGTS_CU7_TA_CTRL_REG__TA_OVERRIDE__SHIFT 0x7
+#define CGTS_CU7_TA_CTRL_REG__TA_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU7_TA_CTRL_REG__TA_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU7_TA_CTRL_REG__TA_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU7_TA_CTRL_REG__TA_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU7_TA_CTRL_REG__TA_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU7_TA_CTRL_REG__TA_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU7_SP1_CTRL_REG__SP10_MASK 0x7f
+#define CGTS_CU7_SP1_CTRL_REG__SP10__SHIFT 0x0
+#define CGTS_CU7_SP1_CTRL_REG__SP10_OVERRIDE_MASK 0x80
+#define CGTS_CU7_SP1_CTRL_REG__SP10_OVERRIDE__SHIFT 0x7
+#define CGTS_CU7_SP1_CTRL_REG__SP10_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU7_SP1_CTRL_REG__SP10_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU7_SP1_CTRL_REG__SP10_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU7_SP1_CTRL_REG__SP10_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU7_SP1_CTRL_REG__SP10_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU7_SP1_CTRL_REG__SP10_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU7_SP1_CTRL_REG__SP11_MASK 0x7f0000
+#define CGTS_CU7_SP1_CTRL_REG__SP11__SHIFT 0x10
+#define CGTS_CU7_SP1_CTRL_REG__SP11_OVERRIDE_MASK 0x800000
+#define CGTS_CU7_SP1_CTRL_REG__SP11_OVERRIDE__SHIFT 0x17
+#define CGTS_CU7_SP1_CTRL_REG__SP11_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU7_SP1_CTRL_REG__SP11_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU7_SP1_CTRL_REG__SP11_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU7_SP1_CTRL_REG__SP11_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU7_SP1_CTRL_REG__SP11_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU7_SP1_CTRL_REG__SP11_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU7_TD_TCP_CTRL_REG__TD_MASK 0x7f
+#define CGTS_CU7_TD_TCP_CTRL_REG__TD__SHIFT 0x0
+#define CGTS_CU7_TD_TCP_CTRL_REG__TD_OVERRIDE_MASK 0x80
+#define CGTS_CU7_TD_TCP_CTRL_REG__TD_OVERRIDE__SHIFT 0x7
+#define CGTS_CU7_TD_TCP_CTRL_REG__TD_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU7_TD_TCP_CTRL_REG__TD_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU7_TD_TCP_CTRL_REG__TD_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU7_TD_TCP_CTRL_REG__TD_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU7_TD_TCP_CTRL_REG__TD_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU7_TD_TCP_CTRL_REG__TD_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU7_TD_TCP_CTRL_REG__TCP_MASK 0x7f0000
+#define CGTS_CU7_TD_TCP_CTRL_REG__TCP__SHIFT 0x10
+#define CGTS_CU7_TD_TCP_CTRL_REG__TCP_OVERRIDE_MASK 0x800000
+#define CGTS_CU7_TD_TCP_CTRL_REG__TCP_OVERRIDE__SHIFT 0x17
+#define CGTS_CU7_TD_TCP_CTRL_REG__TCP_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU7_TD_TCP_CTRL_REG__TCP_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU7_TD_TCP_CTRL_REG__TCP_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU7_TD_TCP_CTRL_REG__TCP_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU7_TD_TCP_CTRL_REG__TCP_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU7_TD_TCP_CTRL_REG__TCP_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU8_SP0_CTRL_REG__SP00_MASK 0x7f
+#define CGTS_CU8_SP0_CTRL_REG__SP00__SHIFT 0x0
+#define CGTS_CU8_SP0_CTRL_REG__SP00_OVERRIDE_MASK 0x80
+#define CGTS_CU8_SP0_CTRL_REG__SP00_OVERRIDE__SHIFT 0x7
+#define CGTS_CU8_SP0_CTRL_REG__SP00_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU8_SP0_CTRL_REG__SP00_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU8_SP0_CTRL_REG__SP00_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU8_SP0_CTRL_REG__SP00_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU8_SP0_CTRL_REG__SP00_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU8_SP0_CTRL_REG__SP00_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU8_SP0_CTRL_REG__SP01_MASK 0x7f0000
+#define CGTS_CU8_SP0_CTRL_REG__SP01__SHIFT 0x10
+#define CGTS_CU8_SP0_CTRL_REG__SP01_OVERRIDE_MASK 0x800000
+#define CGTS_CU8_SP0_CTRL_REG__SP01_OVERRIDE__SHIFT 0x17
+#define CGTS_CU8_SP0_CTRL_REG__SP01_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU8_SP0_CTRL_REG__SP01_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU8_SP0_CTRL_REG__SP01_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU8_SP0_CTRL_REG__SP01_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU8_SP0_CTRL_REG__SP01_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU8_SP0_CTRL_REG__SP01_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU8_LDS_SQ_CTRL_REG__LDS_MASK 0x7f
+#define CGTS_CU8_LDS_SQ_CTRL_REG__LDS__SHIFT 0x0
+#define CGTS_CU8_LDS_SQ_CTRL_REG__LDS_OVERRIDE_MASK 0x80
+#define CGTS_CU8_LDS_SQ_CTRL_REG__LDS_OVERRIDE__SHIFT 0x7
+#define CGTS_CU8_LDS_SQ_CTRL_REG__LDS_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU8_LDS_SQ_CTRL_REG__LDS_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU8_LDS_SQ_CTRL_REG__LDS_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU8_LDS_SQ_CTRL_REG__LDS_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU8_LDS_SQ_CTRL_REG__LDS_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU8_LDS_SQ_CTRL_REG__LDS_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU8_LDS_SQ_CTRL_REG__SQ_MASK 0x7f0000
+#define CGTS_CU8_LDS_SQ_CTRL_REG__SQ__SHIFT 0x10
+#define CGTS_CU8_LDS_SQ_CTRL_REG__SQ_OVERRIDE_MASK 0x800000
+#define CGTS_CU8_LDS_SQ_CTRL_REG__SQ_OVERRIDE__SHIFT 0x17
+#define CGTS_CU8_LDS_SQ_CTRL_REG__SQ_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU8_LDS_SQ_CTRL_REG__SQ_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU8_LDS_SQ_CTRL_REG__SQ_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU8_LDS_SQ_CTRL_REG__SQ_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU8_LDS_SQ_CTRL_REG__SQ_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU8_LDS_SQ_CTRL_REG__SQ_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU8_TA_SQC_CTRL_REG__TA_MASK 0x7f
+#define CGTS_CU8_TA_SQC_CTRL_REG__TA__SHIFT 0x0
+#define CGTS_CU8_TA_SQC_CTRL_REG__TA_OVERRIDE_MASK 0x80
+#define CGTS_CU8_TA_SQC_CTRL_REG__TA_OVERRIDE__SHIFT 0x7
+#define CGTS_CU8_TA_SQC_CTRL_REG__TA_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU8_TA_SQC_CTRL_REG__TA_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU8_TA_SQC_CTRL_REG__TA_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU8_TA_SQC_CTRL_REG__TA_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU8_TA_SQC_CTRL_REG__TA_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU8_TA_SQC_CTRL_REG__TA_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU8_TA_SQC_CTRL_REG__SQC_MASK 0x7f0000
+#define CGTS_CU8_TA_SQC_CTRL_REG__SQC__SHIFT 0x10
+#define CGTS_CU8_TA_SQC_CTRL_REG__SQC_OVERRIDE_MASK 0x800000
+#define CGTS_CU8_TA_SQC_CTRL_REG__SQC_OVERRIDE__SHIFT 0x17
+#define CGTS_CU8_TA_SQC_CTRL_REG__SQC_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU8_TA_SQC_CTRL_REG__SQC_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU8_TA_SQC_CTRL_REG__SQC_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU8_TA_SQC_CTRL_REG__SQC_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU8_TA_SQC_CTRL_REG__SQC_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU8_TA_SQC_CTRL_REG__SQC_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU8_SP1_CTRL_REG__SP10_MASK 0x7f
+#define CGTS_CU8_SP1_CTRL_REG__SP10__SHIFT 0x0
+#define CGTS_CU8_SP1_CTRL_REG__SP10_OVERRIDE_MASK 0x80
+#define CGTS_CU8_SP1_CTRL_REG__SP10_OVERRIDE__SHIFT 0x7
+#define CGTS_CU8_SP1_CTRL_REG__SP10_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU8_SP1_CTRL_REG__SP10_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU8_SP1_CTRL_REG__SP10_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU8_SP1_CTRL_REG__SP10_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU8_SP1_CTRL_REG__SP10_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU8_SP1_CTRL_REG__SP10_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU8_SP1_CTRL_REG__SP11_MASK 0x7f0000
+#define CGTS_CU8_SP1_CTRL_REG__SP11__SHIFT 0x10
+#define CGTS_CU8_SP1_CTRL_REG__SP11_OVERRIDE_MASK 0x800000
+#define CGTS_CU8_SP1_CTRL_REG__SP11_OVERRIDE__SHIFT 0x17
+#define CGTS_CU8_SP1_CTRL_REG__SP11_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU8_SP1_CTRL_REG__SP11_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU8_SP1_CTRL_REG__SP11_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU8_SP1_CTRL_REG__SP11_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU8_SP1_CTRL_REG__SP11_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU8_SP1_CTRL_REG__SP11_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU8_TD_TCP_CTRL_REG__TD_MASK 0x7f
+#define CGTS_CU8_TD_TCP_CTRL_REG__TD__SHIFT 0x0
+#define CGTS_CU8_TD_TCP_CTRL_REG__TD_OVERRIDE_MASK 0x80
+#define CGTS_CU8_TD_TCP_CTRL_REG__TD_OVERRIDE__SHIFT 0x7
+#define CGTS_CU8_TD_TCP_CTRL_REG__TD_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU8_TD_TCP_CTRL_REG__TD_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU8_TD_TCP_CTRL_REG__TD_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU8_TD_TCP_CTRL_REG__TD_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU8_TD_TCP_CTRL_REG__TD_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU8_TD_TCP_CTRL_REG__TD_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU8_TD_TCP_CTRL_REG__TCP_MASK 0x7f0000
+#define CGTS_CU8_TD_TCP_CTRL_REG__TCP__SHIFT 0x10
+#define CGTS_CU8_TD_TCP_CTRL_REG__TCP_OVERRIDE_MASK 0x800000
+#define CGTS_CU8_TD_TCP_CTRL_REG__TCP_OVERRIDE__SHIFT 0x17
+#define CGTS_CU8_TD_TCP_CTRL_REG__TCP_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU8_TD_TCP_CTRL_REG__TCP_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU8_TD_TCP_CTRL_REG__TCP_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU8_TD_TCP_CTRL_REG__TCP_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU8_TD_TCP_CTRL_REG__TCP_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU8_TD_TCP_CTRL_REG__TCP_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU9_SP0_CTRL_REG__SP00_MASK 0x7f
+#define CGTS_CU9_SP0_CTRL_REG__SP00__SHIFT 0x0
+#define CGTS_CU9_SP0_CTRL_REG__SP00_OVERRIDE_MASK 0x80
+#define CGTS_CU9_SP0_CTRL_REG__SP00_OVERRIDE__SHIFT 0x7
+#define CGTS_CU9_SP0_CTRL_REG__SP00_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU9_SP0_CTRL_REG__SP00_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU9_SP0_CTRL_REG__SP00_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU9_SP0_CTRL_REG__SP00_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU9_SP0_CTRL_REG__SP00_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU9_SP0_CTRL_REG__SP00_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU9_SP0_CTRL_REG__SP01_MASK 0x7f0000
+#define CGTS_CU9_SP0_CTRL_REG__SP01__SHIFT 0x10
+#define CGTS_CU9_SP0_CTRL_REG__SP01_OVERRIDE_MASK 0x800000
+#define CGTS_CU9_SP0_CTRL_REG__SP01_OVERRIDE__SHIFT 0x17
+#define CGTS_CU9_SP0_CTRL_REG__SP01_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU9_SP0_CTRL_REG__SP01_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU9_SP0_CTRL_REG__SP01_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU9_SP0_CTRL_REG__SP01_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU9_SP0_CTRL_REG__SP01_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU9_SP0_CTRL_REG__SP01_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU9_LDS_SQ_CTRL_REG__LDS_MASK 0x7f
+#define CGTS_CU9_LDS_SQ_CTRL_REG__LDS__SHIFT 0x0
+#define CGTS_CU9_LDS_SQ_CTRL_REG__LDS_OVERRIDE_MASK 0x80
+#define CGTS_CU9_LDS_SQ_CTRL_REG__LDS_OVERRIDE__SHIFT 0x7
+#define CGTS_CU9_LDS_SQ_CTRL_REG__LDS_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU9_LDS_SQ_CTRL_REG__LDS_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU9_LDS_SQ_CTRL_REG__LDS_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU9_LDS_SQ_CTRL_REG__LDS_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU9_LDS_SQ_CTRL_REG__LDS_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU9_LDS_SQ_CTRL_REG__LDS_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU9_LDS_SQ_CTRL_REG__SQ_MASK 0x7f0000
+#define CGTS_CU9_LDS_SQ_CTRL_REG__SQ__SHIFT 0x10
+#define CGTS_CU9_LDS_SQ_CTRL_REG__SQ_OVERRIDE_MASK 0x800000
+#define CGTS_CU9_LDS_SQ_CTRL_REG__SQ_OVERRIDE__SHIFT 0x17
+#define CGTS_CU9_LDS_SQ_CTRL_REG__SQ_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU9_LDS_SQ_CTRL_REG__SQ_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU9_LDS_SQ_CTRL_REG__SQ_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU9_LDS_SQ_CTRL_REG__SQ_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU9_LDS_SQ_CTRL_REG__SQ_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU9_LDS_SQ_CTRL_REG__SQ_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU9_TA_CTRL_REG__TA_MASK 0x7f
+#define CGTS_CU9_TA_CTRL_REG__TA__SHIFT 0x0
+#define CGTS_CU9_TA_CTRL_REG__TA_OVERRIDE_MASK 0x80
+#define CGTS_CU9_TA_CTRL_REG__TA_OVERRIDE__SHIFT 0x7
+#define CGTS_CU9_TA_CTRL_REG__TA_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU9_TA_CTRL_REG__TA_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU9_TA_CTRL_REG__TA_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU9_TA_CTRL_REG__TA_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU9_TA_CTRL_REG__TA_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU9_TA_CTRL_REG__TA_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU9_SP1_CTRL_REG__SP10_MASK 0x7f
+#define CGTS_CU9_SP1_CTRL_REG__SP10__SHIFT 0x0
+#define CGTS_CU9_SP1_CTRL_REG__SP10_OVERRIDE_MASK 0x80
+#define CGTS_CU9_SP1_CTRL_REG__SP10_OVERRIDE__SHIFT 0x7
+#define CGTS_CU9_SP1_CTRL_REG__SP10_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU9_SP1_CTRL_REG__SP10_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU9_SP1_CTRL_REG__SP10_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU9_SP1_CTRL_REG__SP10_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU9_SP1_CTRL_REG__SP10_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU9_SP1_CTRL_REG__SP10_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU9_SP1_CTRL_REG__SP11_MASK 0x7f0000
+#define CGTS_CU9_SP1_CTRL_REG__SP11__SHIFT 0x10
+#define CGTS_CU9_SP1_CTRL_REG__SP11_OVERRIDE_MASK 0x800000
+#define CGTS_CU9_SP1_CTRL_REG__SP11_OVERRIDE__SHIFT 0x17
+#define CGTS_CU9_SP1_CTRL_REG__SP11_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU9_SP1_CTRL_REG__SP11_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU9_SP1_CTRL_REG__SP11_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU9_SP1_CTRL_REG__SP11_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU9_SP1_CTRL_REG__SP11_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU9_SP1_CTRL_REG__SP11_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU9_TD_TCP_CTRL_REG__TD_MASK 0x7f
+#define CGTS_CU9_TD_TCP_CTRL_REG__TD__SHIFT 0x0
+#define CGTS_CU9_TD_TCP_CTRL_REG__TD_OVERRIDE_MASK 0x80
+#define CGTS_CU9_TD_TCP_CTRL_REG__TD_OVERRIDE__SHIFT 0x7
+#define CGTS_CU9_TD_TCP_CTRL_REG__TD_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU9_TD_TCP_CTRL_REG__TD_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU9_TD_TCP_CTRL_REG__TD_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU9_TD_TCP_CTRL_REG__TD_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU9_TD_TCP_CTRL_REG__TD_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU9_TD_TCP_CTRL_REG__TD_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU9_TD_TCP_CTRL_REG__TCP_MASK 0x7f0000
+#define CGTS_CU9_TD_TCP_CTRL_REG__TCP__SHIFT 0x10
+#define CGTS_CU9_TD_TCP_CTRL_REG__TCP_OVERRIDE_MASK 0x800000
+#define CGTS_CU9_TD_TCP_CTRL_REG__TCP_OVERRIDE__SHIFT 0x17
+#define CGTS_CU9_TD_TCP_CTRL_REG__TCP_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU9_TD_TCP_CTRL_REG__TCP_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU9_TD_TCP_CTRL_REG__TCP_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU9_TD_TCP_CTRL_REG__TCP_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU9_TD_TCP_CTRL_REG__TCP_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU9_TD_TCP_CTRL_REG__TCP_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU10_SP0_CTRL_REG__SP00_MASK 0x7f
+#define CGTS_CU10_SP0_CTRL_REG__SP00__SHIFT 0x0
+#define CGTS_CU10_SP0_CTRL_REG__SP00_OVERRIDE_MASK 0x80
+#define CGTS_CU10_SP0_CTRL_REG__SP00_OVERRIDE__SHIFT 0x7
+#define CGTS_CU10_SP0_CTRL_REG__SP00_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU10_SP0_CTRL_REG__SP00_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU10_SP0_CTRL_REG__SP00_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU10_SP0_CTRL_REG__SP00_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU10_SP0_CTRL_REG__SP00_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU10_SP0_CTRL_REG__SP00_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU10_SP0_CTRL_REG__SP01_MASK 0x7f0000
+#define CGTS_CU10_SP0_CTRL_REG__SP01__SHIFT 0x10
+#define CGTS_CU10_SP0_CTRL_REG__SP01_OVERRIDE_MASK 0x800000
+#define CGTS_CU10_SP0_CTRL_REG__SP01_OVERRIDE__SHIFT 0x17
+#define CGTS_CU10_SP0_CTRL_REG__SP01_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU10_SP0_CTRL_REG__SP01_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU10_SP0_CTRL_REG__SP01_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU10_SP0_CTRL_REG__SP01_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU10_SP0_CTRL_REG__SP01_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU10_SP0_CTRL_REG__SP01_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU10_LDS_SQ_CTRL_REG__LDS_MASK 0x7f
+#define CGTS_CU10_LDS_SQ_CTRL_REG__LDS__SHIFT 0x0
+#define CGTS_CU10_LDS_SQ_CTRL_REG__LDS_OVERRIDE_MASK 0x80
+#define CGTS_CU10_LDS_SQ_CTRL_REG__LDS_OVERRIDE__SHIFT 0x7
+#define CGTS_CU10_LDS_SQ_CTRL_REG__LDS_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU10_LDS_SQ_CTRL_REG__LDS_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU10_LDS_SQ_CTRL_REG__LDS_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU10_LDS_SQ_CTRL_REG__LDS_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU10_LDS_SQ_CTRL_REG__LDS_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU10_LDS_SQ_CTRL_REG__LDS_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU10_LDS_SQ_CTRL_REG__SQ_MASK 0x7f0000
+#define CGTS_CU10_LDS_SQ_CTRL_REG__SQ__SHIFT 0x10
+#define CGTS_CU10_LDS_SQ_CTRL_REG__SQ_OVERRIDE_MASK 0x800000
+#define CGTS_CU10_LDS_SQ_CTRL_REG__SQ_OVERRIDE__SHIFT 0x17
+#define CGTS_CU10_LDS_SQ_CTRL_REG__SQ_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU10_LDS_SQ_CTRL_REG__SQ_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU10_LDS_SQ_CTRL_REG__SQ_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU10_LDS_SQ_CTRL_REG__SQ_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU10_LDS_SQ_CTRL_REG__SQ_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU10_LDS_SQ_CTRL_REG__SQ_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU10_TA_CTRL_REG__TA_MASK 0x7f
+#define CGTS_CU10_TA_CTRL_REG__TA__SHIFT 0x0
+#define CGTS_CU10_TA_CTRL_REG__TA_OVERRIDE_MASK 0x80
+#define CGTS_CU10_TA_CTRL_REG__TA_OVERRIDE__SHIFT 0x7
+#define CGTS_CU10_TA_CTRL_REG__TA_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU10_TA_CTRL_REG__TA_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU10_TA_CTRL_REG__TA_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU10_TA_CTRL_REG__TA_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU10_TA_CTRL_REG__TA_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU10_TA_CTRL_REG__TA_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU10_SP1_CTRL_REG__SP10_MASK 0x7f
+#define CGTS_CU10_SP1_CTRL_REG__SP10__SHIFT 0x0
+#define CGTS_CU10_SP1_CTRL_REG__SP10_OVERRIDE_MASK 0x80
+#define CGTS_CU10_SP1_CTRL_REG__SP10_OVERRIDE__SHIFT 0x7
+#define CGTS_CU10_SP1_CTRL_REG__SP10_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU10_SP1_CTRL_REG__SP10_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU10_SP1_CTRL_REG__SP10_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU10_SP1_CTRL_REG__SP10_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU10_SP1_CTRL_REG__SP10_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU10_SP1_CTRL_REG__SP10_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU10_SP1_CTRL_REG__SP11_MASK 0x7f0000
+#define CGTS_CU10_SP1_CTRL_REG__SP11__SHIFT 0x10
+#define CGTS_CU10_SP1_CTRL_REG__SP11_OVERRIDE_MASK 0x800000
+#define CGTS_CU10_SP1_CTRL_REG__SP11_OVERRIDE__SHIFT 0x17
+#define CGTS_CU10_SP1_CTRL_REG__SP11_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU10_SP1_CTRL_REG__SP11_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU10_SP1_CTRL_REG__SP11_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU10_SP1_CTRL_REG__SP11_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU10_SP1_CTRL_REG__SP11_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU10_SP1_CTRL_REG__SP11_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU10_TD_TCP_CTRL_REG__TD_MASK 0x7f
+#define CGTS_CU10_TD_TCP_CTRL_REG__TD__SHIFT 0x0
+#define CGTS_CU10_TD_TCP_CTRL_REG__TD_OVERRIDE_MASK 0x80
+#define CGTS_CU10_TD_TCP_CTRL_REG__TD_OVERRIDE__SHIFT 0x7
+#define CGTS_CU10_TD_TCP_CTRL_REG__TD_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU10_TD_TCP_CTRL_REG__TD_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU10_TD_TCP_CTRL_REG__TD_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU10_TD_TCP_CTRL_REG__TD_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU10_TD_TCP_CTRL_REG__TD_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU10_TD_TCP_CTRL_REG__TD_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU10_TD_TCP_CTRL_REG__TCP_MASK 0x7f0000
+#define CGTS_CU10_TD_TCP_CTRL_REG__TCP__SHIFT 0x10
+#define CGTS_CU10_TD_TCP_CTRL_REG__TCP_OVERRIDE_MASK 0x800000
+#define CGTS_CU10_TD_TCP_CTRL_REG__TCP_OVERRIDE__SHIFT 0x17
+#define CGTS_CU10_TD_TCP_CTRL_REG__TCP_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU10_TD_TCP_CTRL_REG__TCP_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU10_TD_TCP_CTRL_REG__TCP_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU10_TD_TCP_CTRL_REG__TCP_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU10_TD_TCP_CTRL_REG__TCP_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU10_TD_TCP_CTRL_REG__TCP_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU11_SP0_CTRL_REG__SP00_MASK 0x7f
+#define CGTS_CU11_SP0_CTRL_REG__SP00__SHIFT 0x0
+#define CGTS_CU11_SP0_CTRL_REG__SP00_OVERRIDE_MASK 0x80
+#define CGTS_CU11_SP0_CTRL_REG__SP00_OVERRIDE__SHIFT 0x7
+#define CGTS_CU11_SP0_CTRL_REG__SP00_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU11_SP0_CTRL_REG__SP00_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU11_SP0_CTRL_REG__SP00_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU11_SP0_CTRL_REG__SP00_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU11_SP0_CTRL_REG__SP00_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU11_SP0_CTRL_REG__SP00_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU11_SP0_CTRL_REG__SP01_MASK 0x7f0000
+#define CGTS_CU11_SP0_CTRL_REG__SP01__SHIFT 0x10
+#define CGTS_CU11_SP0_CTRL_REG__SP01_OVERRIDE_MASK 0x800000
+#define CGTS_CU11_SP0_CTRL_REG__SP01_OVERRIDE__SHIFT 0x17
+#define CGTS_CU11_SP0_CTRL_REG__SP01_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU11_SP0_CTRL_REG__SP01_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU11_SP0_CTRL_REG__SP01_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU11_SP0_CTRL_REG__SP01_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU11_SP0_CTRL_REG__SP01_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU11_SP0_CTRL_REG__SP01_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU11_LDS_SQ_CTRL_REG__LDS_MASK 0x7f
+#define CGTS_CU11_LDS_SQ_CTRL_REG__LDS__SHIFT 0x0
+#define CGTS_CU11_LDS_SQ_CTRL_REG__LDS_OVERRIDE_MASK 0x80
+#define CGTS_CU11_LDS_SQ_CTRL_REG__LDS_OVERRIDE__SHIFT 0x7
+#define CGTS_CU11_LDS_SQ_CTRL_REG__LDS_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU11_LDS_SQ_CTRL_REG__LDS_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU11_LDS_SQ_CTRL_REG__LDS_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU11_LDS_SQ_CTRL_REG__LDS_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU11_LDS_SQ_CTRL_REG__LDS_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU11_LDS_SQ_CTRL_REG__LDS_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU11_LDS_SQ_CTRL_REG__SQ_MASK 0x7f0000
+#define CGTS_CU11_LDS_SQ_CTRL_REG__SQ__SHIFT 0x10
+#define CGTS_CU11_LDS_SQ_CTRL_REG__SQ_OVERRIDE_MASK 0x800000
+#define CGTS_CU11_LDS_SQ_CTRL_REG__SQ_OVERRIDE__SHIFT 0x17
+#define CGTS_CU11_LDS_SQ_CTRL_REG__SQ_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU11_LDS_SQ_CTRL_REG__SQ_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU11_LDS_SQ_CTRL_REG__SQ_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU11_LDS_SQ_CTRL_REG__SQ_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU11_LDS_SQ_CTRL_REG__SQ_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU11_LDS_SQ_CTRL_REG__SQ_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU11_TA_CTRL_REG__TA_MASK 0x7f
+#define CGTS_CU11_TA_CTRL_REG__TA__SHIFT 0x0
+#define CGTS_CU11_TA_CTRL_REG__TA_OVERRIDE_MASK 0x80
+#define CGTS_CU11_TA_CTRL_REG__TA_OVERRIDE__SHIFT 0x7
+#define CGTS_CU11_TA_CTRL_REG__TA_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU11_TA_CTRL_REG__TA_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU11_TA_CTRL_REG__TA_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU11_TA_CTRL_REG__TA_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU11_TA_CTRL_REG__TA_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU11_TA_CTRL_REG__TA_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU11_SP1_CTRL_REG__SP10_MASK 0x7f
+#define CGTS_CU11_SP1_CTRL_REG__SP10__SHIFT 0x0
+#define CGTS_CU11_SP1_CTRL_REG__SP10_OVERRIDE_MASK 0x80
+#define CGTS_CU11_SP1_CTRL_REG__SP10_OVERRIDE__SHIFT 0x7
+#define CGTS_CU11_SP1_CTRL_REG__SP10_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU11_SP1_CTRL_REG__SP10_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU11_SP1_CTRL_REG__SP10_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU11_SP1_CTRL_REG__SP10_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU11_SP1_CTRL_REG__SP10_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU11_SP1_CTRL_REG__SP10_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU11_SP1_CTRL_REG__SP11_MASK 0x7f0000
+#define CGTS_CU11_SP1_CTRL_REG__SP11__SHIFT 0x10
+#define CGTS_CU11_SP1_CTRL_REG__SP11_OVERRIDE_MASK 0x800000
+#define CGTS_CU11_SP1_CTRL_REG__SP11_OVERRIDE__SHIFT 0x17
+#define CGTS_CU11_SP1_CTRL_REG__SP11_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU11_SP1_CTRL_REG__SP11_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU11_SP1_CTRL_REG__SP11_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU11_SP1_CTRL_REG__SP11_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU11_SP1_CTRL_REG__SP11_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU11_SP1_CTRL_REG__SP11_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU11_TD_TCP_CTRL_REG__TD_MASK 0x7f
+#define CGTS_CU11_TD_TCP_CTRL_REG__TD__SHIFT 0x0
+#define CGTS_CU11_TD_TCP_CTRL_REG__TD_OVERRIDE_MASK 0x80
+#define CGTS_CU11_TD_TCP_CTRL_REG__TD_OVERRIDE__SHIFT 0x7
+#define CGTS_CU11_TD_TCP_CTRL_REG__TD_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU11_TD_TCP_CTRL_REG__TD_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU11_TD_TCP_CTRL_REG__TD_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU11_TD_TCP_CTRL_REG__TD_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU11_TD_TCP_CTRL_REG__TD_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU11_TD_TCP_CTRL_REG__TD_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU11_TD_TCP_CTRL_REG__TCP_MASK 0x7f0000
+#define CGTS_CU11_TD_TCP_CTRL_REG__TCP__SHIFT 0x10
+#define CGTS_CU11_TD_TCP_CTRL_REG__TCP_OVERRIDE_MASK 0x800000
+#define CGTS_CU11_TD_TCP_CTRL_REG__TCP_OVERRIDE__SHIFT 0x17
+#define CGTS_CU11_TD_TCP_CTRL_REG__TCP_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU11_TD_TCP_CTRL_REG__TCP_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU11_TD_TCP_CTRL_REG__TCP_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU11_TD_TCP_CTRL_REG__TCP_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU11_TD_TCP_CTRL_REG__TCP_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU11_TD_TCP_CTRL_REG__TCP_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU12_SP0_CTRL_REG__SP00_MASK 0x7f
+#define CGTS_CU12_SP0_CTRL_REG__SP00__SHIFT 0x0
+#define CGTS_CU12_SP0_CTRL_REG__SP00_OVERRIDE_MASK 0x80
+#define CGTS_CU12_SP0_CTRL_REG__SP00_OVERRIDE__SHIFT 0x7
+#define CGTS_CU12_SP0_CTRL_REG__SP00_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU12_SP0_CTRL_REG__SP00_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU12_SP0_CTRL_REG__SP00_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU12_SP0_CTRL_REG__SP00_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU12_SP0_CTRL_REG__SP00_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU12_SP0_CTRL_REG__SP00_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU12_SP0_CTRL_REG__SP01_MASK 0x7f0000
+#define CGTS_CU12_SP0_CTRL_REG__SP01__SHIFT 0x10
+#define CGTS_CU12_SP0_CTRL_REG__SP01_OVERRIDE_MASK 0x800000
+#define CGTS_CU12_SP0_CTRL_REG__SP01_OVERRIDE__SHIFT 0x17
+#define CGTS_CU12_SP0_CTRL_REG__SP01_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU12_SP0_CTRL_REG__SP01_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU12_SP0_CTRL_REG__SP01_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU12_SP0_CTRL_REG__SP01_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU12_SP0_CTRL_REG__SP01_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU12_SP0_CTRL_REG__SP01_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU12_LDS_SQ_CTRL_REG__LDS_MASK 0x7f
+#define CGTS_CU12_LDS_SQ_CTRL_REG__LDS__SHIFT 0x0
+#define CGTS_CU12_LDS_SQ_CTRL_REG__LDS_OVERRIDE_MASK 0x80
+#define CGTS_CU12_LDS_SQ_CTRL_REG__LDS_OVERRIDE__SHIFT 0x7
+#define CGTS_CU12_LDS_SQ_CTRL_REG__LDS_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU12_LDS_SQ_CTRL_REG__LDS_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU12_LDS_SQ_CTRL_REG__LDS_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU12_LDS_SQ_CTRL_REG__LDS_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU12_LDS_SQ_CTRL_REG__LDS_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU12_LDS_SQ_CTRL_REG__LDS_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU12_LDS_SQ_CTRL_REG__SQ_MASK 0x7f0000
+#define CGTS_CU12_LDS_SQ_CTRL_REG__SQ__SHIFT 0x10
+#define CGTS_CU12_LDS_SQ_CTRL_REG__SQ_OVERRIDE_MASK 0x800000
+#define CGTS_CU12_LDS_SQ_CTRL_REG__SQ_OVERRIDE__SHIFT 0x17
+#define CGTS_CU12_LDS_SQ_CTRL_REG__SQ_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU12_LDS_SQ_CTRL_REG__SQ_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU12_LDS_SQ_CTRL_REG__SQ_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU12_LDS_SQ_CTRL_REG__SQ_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU12_LDS_SQ_CTRL_REG__SQ_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU12_LDS_SQ_CTRL_REG__SQ_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU12_TA_SQC_CTRL_REG__TA_MASK 0x7f
+#define CGTS_CU12_TA_SQC_CTRL_REG__TA__SHIFT 0x0
+#define CGTS_CU12_TA_SQC_CTRL_REG__TA_OVERRIDE_MASK 0x80
+#define CGTS_CU12_TA_SQC_CTRL_REG__TA_OVERRIDE__SHIFT 0x7
+#define CGTS_CU12_TA_SQC_CTRL_REG__TA_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU12_TA_SQC_CTRL_REG__TA_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU12_TA_SQC_CTRL_REG__TA_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU12_TA_SQC_CTRL_REG__TA_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU12_TA_SQC_CTRL_REG__TA_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU12_TA_SQC_CTRL_REG__TA_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU12_TA_SQC_CTRL_REG__SQC_MASK 0x7f0000
+#define CGTS_CU12_TA_SQC_CTRL_REG__SQC__SHIFT 0x10
+#define CGTS_CU12_TA_SQC_CTRL_REG__SQC_OVERRIDE_MASK 0x800000
+#define CGTS_CU12_TA_SQC_CTRL_REG__SQC_OVERRIDE__SHIFT 0x17
+#define CGTS_CU12_TA_SQC_CTRL_REG__SQC_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU12_TA_SQC_CTRL_REG__SQC_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU12_TA_SQC_CTRL_REG__SQC_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU12_TA_SQC_CTRL_REG__SQC_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU12_TA_SQC_CTRL_REG__SQC_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU12_TA_SQC_CTRL_REG__SQC_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU12_SP1_CTRL_REG__SP10_MASK 0x7f
+#define CGTS_CU12_SP1_CTRL_REG__SP10__SHIFT 0x0
+#define CGTS_CU12_SP1_CTRL_REG__SP10_OVERRIDE_MASK 0x80
+#define CGTS_CU12_SP1_CTRL_REG__SP10_OVERRIDE__SHIFT 0x7
+#define CGTS_CU12_SP1_CTRL_REG__SP10_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU12_SP1_CTRL_REG__SP10_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU12_SP1_CTRL_REG__SP10_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU12_SP1_CTRL_REG__SP10_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU12_SP1_CTRL_REG__SP10_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU12_SP1_CTRL_REG__SP10_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU12_SP1_CTRL_REG__SP11_MASK 0x7f0000
+#define CGTS_CU12_SP1_CTRL_REG__SP11__SHIFT 0x10
+#define CGTS_CU12_SP1_CTRL_REG__SP11_OVERRIDE_MASK 0x800000
+#define CGTS_CU12_SP1_CTRL_REG__SP11_OVERRIDE__SHIFT 0x17
+#define CGTS_CU12_SP1_CTRL_REG__SP11_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU12_SP1_CTRL_REG__SP11_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU12_SP1_CTRL_REG__SP11_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU12_SP1_CTRL_REG__SP11_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU12_SP1_CTRL_REG__SP11_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU12_SP1_CTRL_REG__SP11_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU12_TD_TCP_CTRL_REG__TD_MASK 0x7f
+#define CGTS_CU12_TD_TCP_CTRL_REG__TD__SHIFT 0x0
+#define CGTS_CU12_TD_TCP_CTRL_REG__TD_OVERRIDE_MASK 0x80
+#define CGTS_CU12_TD_TCP_CTRL_REG__TD_OVERRIDE__SHIFT 0x7
+#define CGTS_CU12_TD_TCP_CTRL_REG__TD_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU12_TD_TCP_CTRL_REG__TD_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU12_TD_TCP_CTRL_REG__TD_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU12_TD_TCP_CTRL_REG__TD_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU12_TD_TCP_CTRL_REG__TD_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU12_TD_TCP_CTRL_REG__TD_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU12_TD_TCP_CTRL_REG__TCP_MASK 0x7f0000
+#define CGTS_CU12_TD_TCP_CTRL_REG__TCP__SHIFT 0x10
+#define CGTS_CU12_TD_TCP_CTRL_REG__TCP_OVERRIDE_MASK 0x800000
+#define CGTS_CU12_TD_TCP_CTRL_REG__TCP_OVERRIDE__SHIFT 0x17
+#define CGTS_CU12_TD_TCP_CTRL_REG__TCP_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU12_TD_TCP_CTRL_REG__TCP_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU12_TD_TCP_CTRL_REG__TCP_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU12_TD_TCP_CTRL_REG__TCP_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU12_TD_TCP_CTRL_REG__TCP_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU12_TD_TCP_CTRL_REG__TCP_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU13_SP0_CTRL_REG__SP00_MASK 0x7f
+#define CGTS_CU13_SP0_CTRL_REG__SP00__SHIFT 0x0
+#define CGTS_CU13_SP0_CTRL_REG__SP00_OVERRIDE_MASK 0x80
+#define CGTS_CU13_SP0_CTRL_REG__SP00_OVERRIDE__SHIFT 0x7
+#define CGTS_CU13_SP0_CTRL_REG__SP00_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU13_SP0_CTRL_REG__SP00_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU13_SP0_CTRL_REG__SP00_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU13_SP0_CTRL_REG__SP00_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU13_SP0_CTRL_REG__SP00_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU13_SP0_CTRL_REG__SP00_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU13_SP0_CTRL_REG__SP01_MASK 0x7f0000
+#define CGTS_CU13_SP0_CTRL_REG__SP01__SHIFT 0x10
+#define CGTS_CU13_SP0_CTRL_REG__SP01_OVERRIDE_MASK 0x800000
+#define CGTS_CU13_SP0_CTRL_REG__SP01_OVERRIDE__SHIFT 0x17
+#define CGTS_CU13_SP0_CTRL_REG__SP01_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU13_SP0_CTRL_REG__SP01_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU13_SP0_CTRL_REG__SP01_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU13_SP0_CTRL_REG__SP01_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU13_SP0_CTRL_REG__SP01_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU13_SP0_CTRL_REG__SP01_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU13_LDS_SQ_CTRL_REG__LDS_MASK 0x7f
+#define CGTS_CU13_LDS_SQ_CTRL_REG__LDS__SHIFT 0x0
+#define CGTS_CU13_LDS_SQ_CTRL_REG__LDS_OVERRIDE_MASK 0x80
+#define CGTS_CU13_LDS_SQ_CTRL_REG__LDS_OVERRIDE__SHIFT 0x7
+#define CGTS_CU13_LDS_SQ_CTRL_REG__LDS_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU13_LDS_SQ_CTRL_REG__LDS_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU13_LDS_SQ_CTRL_REG__LDS_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU13_LDS_SQ_CTRL_REG__LDS_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU13_LDS_SQ_CTRL_REG__LDS_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU13_LDS_SQ_CTRL_REG__LDS_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU13_LDS_SQ_CTRL_REG__SQ_MASK 0x7f0000
+#define CGTS_CU13_LDS_SQ_CTRL_REG__SQ__SHIFT 0x10
+#define CGTS_CU13_LDS_SQ_CTRL_REG__SQ_OVERRIDE_MASK 0x800000
+#define CGTS_CU13_LDS_SQ_CTRL_REG__SQ_OVERRIDE__SHIFT 0x17
+#define CGTS_CU13_LDS_SQ_CTRL_REG__SQ_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU13_LDS_SQ_CTRL_REG__SQ_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU13_LDS_SQ_CTRL_REG__SQ_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU13_LDS_SQ_CTRL_REG__SQ_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU13_LDS_SQ_CTRL_REG__SQ_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU13_LDS_SQ_CTRL_REG__SQ_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU13_TA_CTRL_REG__TA_MASK 0x7f
+#define CGTS_CU13_TA_CTRL_REG__TA__SHIFT 0x0
+#define CGTS_CU13_TA_CTRL_REG__TA_OVERRIDE_MASK 0x80
+#define CGTS_CU13_TA_CTRL_REG__TA_OVERRIDE__SHIFT 0x7
+#define CGTS_CU13_TA_CTRL_REG__TA_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU13_TA_CTRL_REG__TA_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU13_TA_CTRL_REG__TA_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU13_TA_CTRL_REG__TA_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU13_TA_CTRL_REG__TA_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU13_TA_CTRL_REG__TA_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU13_SP1_CTRL_REG__SP10_MASK 0x7f
+#define CGTS_CU13_SP1_CTRL_REG__SP10__SHIFT 0x0
+#define CGTS_CU13_SP1_CTRL_REG__SP10_OVERRIDE_MASK 0x80
+#define CGTS_CU13_SP1_CTRL_REG__SP10_OVERRIDE__SHIFT 0x7
+#define CGTS_CU13_SP1_CTRL_REG__SP10_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU13_SP1_CTRL_REG__SP10_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU13_SP1_CTRL_REG__SP10_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU13_SP1_CTRL_REG__SP10_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU13_SP1_CTRL_REG__SP10_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU13_SP1_CTRL_REG__SP10_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU13_SP1_CTRL_REG__SP11_MASK 0x7f0000
+#define CGTS_CU13_SP1_CTRL_REG__SP11__SHIFT 0x10
+#define CGTS_CU13_SP1_CTRL_REG__SP11_OVERRIDE_MASK 0x800000
+#define CGTS_CU13_SP1_CTRL_REG__SP11_OVERRIDE__SHIFT 0x17
+#define CGTS_CU13_SP1_CTRL_REG__SP11_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU13_SP1_CTRL_REG__SP11_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU13_SP1_CTRL_REG__SP11_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU13_SP1_CTRL_REG__SP11_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU13_SP1_CTRL_REG__SP11_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU13_SP1_CTRL_REG__SP11_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU13_TD_TCP_CTRL_REG__TD_MASK 0x7f
+#define CGTS_CU13_TD_TCP_CTRL_REG__TD__SHIFT 0x0
+#define CGTS_CU13_TD_TCP_CTRL_REG__TD_OVERRIDE_MASK 0x80
+#define CGTS_CU13_TD_TCP_CTRL_REG__TD_OVERRIDE__SHIFT 0x7
+#define CGTS_CU13_TD_TCP_CTRL_REG__TD_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU13_TD_TCP_CTRL_REG__TD_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU13_TD_TCP_CTRL_REG__TD_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU13_TD_TCP_CTRL_REG__TD_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU13_TD_TCP_CTRL_REG__TD_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU13_TD_TCP_CTRL_REG__TD_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU13_TD_TCP_CTRL_REG__TCP_MASK 0x7f0000
+#define CGTS_CU13_TD_TCP_CTRL_REG__TCP__SHIFT 0x10
+#define CGTS_CU13_TD_TCP_CTRL_REG__TCP_OVERRIDE_MASK 0x800000
+#define CGTS_CU13_TD_TCP_CTRL_REG__TCP_OVERRIDE__SHIFT 0x17
+#define CGTS_CU13_TD_TCP_CTRL_REG__TCP_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU13_TD_TCP_CTRL_REG__TCP_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU13_TD_TCP_CTRL_REG__TCP_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU13_TD_TCP_CTRL_REG__TCP_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU13_TD_TCP_CTRL_REG__TCP_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU13_TD_TCP_CTRL_REG__TCP_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU14_SP0_CTRL_REG__SP00_MASK 0x7f
+#define CGTS_CU14_SP0_CTRL_REG__SP00__SHIFT 0x0
+#define CGTS_CU14_SP0_CTRL_REG__SP00_OVERRIDE_MASK 0x80
+#define CGTS_CU14_SP0_CTRL_REG__SP00_OVERRIDE__SHIFT 0x7
+#define CGTS_CU14_SP0_CTRL_REG__SP00_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU14_SP0_CTRL_REG__SP00_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU14_SP0_CTRL_REG__SP00_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU14_SP0_CTRL_REG__SP00_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU14_SP0_CTRL_REG__SP00_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU14_SP0_CTRL_REG__SP00_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU14_SP0_CTRL_REG__SP01_MASK 0x7f0000
+#define CGTS_CU14_SP0_CTRL_REG__SP01__SHIFT 0x10
+#define CGTS_CU14_SP0_CTRL_REG__SP01_OVERRIDE_MASK 0x800000
+#define CGTS_CU14_SP0_CTRL_REG__SP01_OVERRIDE__SHIFT 0x17
+#define CGTS_CU14_SP0_CTRL_REG__SP01_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU14_SP0_CTRL_REG__SP01_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU14_SP0_CTRL_REG__SP01_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU14_SP0_CTRL_REG__SP01_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU14_SP0_CTRL_REG__SP01_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU14_SP0_CTRL_REG__SP01_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU14_LDS_SQ_CTRL_REG__LDS_MASK 0x7f
+#define CGTS_CU14_LDS_SQ_CTRL_REG__LDS__SHIFT 0x0
+#define CGTS_CU14_LDS_SQ_CTRL_REG__LDS_OVERRIDE_MASK 0x80
+#define CGTS_CU14_LDS_SQ_CTRL_REG__LDS_OVERRIDE__SHIFT 0x7
+#define CGTS_CU14_LDS_SQ_CTRL_REG__LDS_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU14_LDS_SQ_CTRL_REG__LDS_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU14_LDS_SQ_CTRL_REG__LDS_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU14_LDS_SQ_CTRL_REG__LDS_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU14_LDS_SQ_CTRL_REG__LDS_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU14_LDS_SQ_CTRL_REG__LDS_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU14_LDS_SQ_CTRL_REG__SQ_MASK 0x7f0000
+#define CGTS_CU14_LDS_SQ_CTRL_REG__SQ__SHIFT 0x10
+#define CGTS_CU14_LDS_SQ_CTRL_REG__SQ_OVERRIDE_MASK 0x800000
+#define CGTS_CU14_LDS_SQ_CTRL_REG__SQ_OVERRIDE__SHIFT 0x17
+#define CGTS_CU14_LDS_SQ_CTRL_REG__SQ_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU14_LDS_SQ_CTRL_REG__SQ_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU14_LDS_SQ_CTRL_REG__SQ_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU14_LDS_SQ_CTRL_REG__SQ_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU14_LDS_SQ_CTRL_REG__SQ_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU14_LDS_SQ_CTRL_REG__SQ_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU14_TA_CTRL_REG__TA_MASK 0x7f
+#define CGTS_CU14_TA_CTRL_REG__TA__SHIFT 0x0
+#define CGTS_CU14_TA_CTRL_REG__TA_OVERRIDE_MASK 0x80
+#define CGTS_CU14_TA_CTRL_REG__TA_OVERRIDE__SHIFT 0x7
+#define CGTS_CU14_TA_CTRL_REG__TA_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU14_TA_CTRL_REG__TA_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU14_TA_CTRL_REG__TA_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU14_TA_CTRL_REG__TA_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU14_TA_CTRL_REG__TA_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU14_TA_CTRL_REG__TA_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU14_SP1_CTRL_REG__SP10_MASK 0x7f
+#define CGTS_CU14_SP1_CTRL_REG__SP10__SHIFT 0x0
+#define CGTS_CU14_SP1_CTRL_REG__SP10_OVERRIDE_MASK 0x80
+#define CGTS_CU14_SP1_CTRL_REG__SP10_OVERRIDE__SHIFT 0x7
+#define CGTS_CU14_SP1_CTRL_REG__SP10_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU14_SP1_CTRL_REG__SP10_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU14_SP1_CTRL_REG__SP10_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU14_SP1_CTRL_REG__SP10_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU14_SP1_CTRL_REG__SP10_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU14_SP1_CTRL_REG__SP10_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU14_SP1_CTRL_REG__SP11_MASK 0x7f0000
+#define CGTS_CU14_SP1_CTRL_REG__SP11__SHIFT 0x10
+#define CGTS_CU14_SP1_CTRL_REG__SP11_OVERRIDE_MASK 0x800000
+#define CGTS_CU14_SP1_CTRL_REG__SP11_OVERRIDE__SHIFT 0x17
+#define CGTS_CU14_SP1_CTRL_REG__SP11_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU14_SP1_CTRL_REG__SP11_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU14_SP1_CTRL_REG__SP11_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU14_SP1_CTRL_REG__SP11_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU14_SP1_CTRL_REG__SP11_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU14_SP1_CTRL_REG__SP11_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU14_TD_TCP_CTRL_REG__TD_MASK 0x7f
+#define CGTS_CU14_TD_TCP_CTRL_REG__TD__SHIFT 0x0
+#define CGTS_CU14_TD_TCP_CTRL_REG__TD_OVERRIDE_MASK 0x80
+#define CGTS_CU14_TD_TCP_CTRL_REG__TD_OVERRIDE__SHIFT 0x7
+#define CGTS_CU14_TD_TCP_CTRL_REG__TD_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU14_TD_TCP_CTRL_REG__TD_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU14_TD_TCP_CTRL_REG__TD_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU14_TD_TCP_CTRL_REG__TD_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU14_TD_TCP_CTRL_REG__TD_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU14_TD_TCP_CTRL_REG__TD_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU14_TD_TCP_CTRL_REG__TCP_MASK 0x7f0000
+#define CGTS_CU14_TD_TCP_CTRL_REG__TCP__SHIFT 0x10
+#define CGTS_CU14_TD_TCP_CTRL_REG__TCP_OVERRIDE_MASK 0x800000
+#define CGTS_CU14_TD_TCP_CTRL_REG__TCP_OVERRIDE__SHIFT 0x17
+#define CGTS_CU14_TD_TCP_CTRL_REG__TCP_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU14_TD_TCP_CTRL_REG__TCP_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU14_TD_TCP_CTRL_REG__TCP_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU14_TD_TCP_CTRL_REG__TCP_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU14_TD_TCP_CTRL_REG__TCP_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU14_TD_TCP_CTRL_REG__TCP_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU15_SP0_CTRL_REG__SP00_MASK 0x7f
+#define CGTS_CU15_SP0_CTRL_REG__SP00__SHIFT 0x0
+#define CGTS_CU15_SP0_CTRL_REG__SP00_OVERRIDE_MASK 0x80
+#define CGTS_CU15_SP0_CTRL_REG__SP00_OVERRIDE__SHIFT 0x7
+#define CGTS_CU15_SP0_CTRL_REG__SP00_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU15_SP0_CTRL_REG__SP00_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU15_SP0_CTRL_REG__SP00_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU15_SP0_CTRL_REG__SP00_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU15_SP0_CTRL_REG__SP00_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU15_SP0_CTRL_REG__SP00_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU15_SP0_CTRL_REG__SP01_MASK 0x7f0000
+#define CGTS_CU15_SP0_CTRL_REG__SP01__SHIFT 0x10
+#define CGTS_CU15_SP0_CTRL_REG__SP01_OVERRIDE_MASK 0x800000
+#define CGTS_CU15_SP0_CTRL_REG__SP01_OVERRIDE__SHIFT 0x17
+#define CGTS_CU15_SP0_CTRL_REG__SP01_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU15_SP0_CTRL_REG__SP01_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU15_SP0_CTRL_REG__SP01_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU15_SP0_CTRL_REG__SP01_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU15_SP0_CTRL_REG__SP01_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU15_SP0_CTRL_REG__SP01_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU15_LDS_SQ_CTRL_REG__LDS_MASK 0x7f
+#define CGTS_CU15_LDS_SQ_CTRL_REG__LDS__SHIFT 0x0
+#define CGTS_CU15_LDS_SQ_CTRL_REG__LDS_OVERRIDE_MASK 0x80
+#define CGTS_CU15_LDS_SQ_CTRL_REG__LDS_OVERRIDE__SHIFT 0x7
+#define CGTS_CU15_LDS_SQ_CTRL_REG__LDS_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU15_LDS_SQ_CTRL_REG__LDS_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU15_LDS_SQ_CTRL_REG__LDS_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU15_LDS_SQ_CTRL_REG__LDS_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU15_LDS_SQ_CTRL_REG__LDS_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU15_LDS_SQ_CTRL_REG__LDS_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU15_LDS_SQ_CTRL_REG__SQ_MASK 0x7f0000
+#define CGTS_CU15_LDS_SQ_CTRL_REG__SQ__SHIFT 0x10
+#define CGTS_CU15_LDS_SQ_CTRL_REG__SQ_OVERRIDE_MASK 0x800000
+#define CGTS_CU15_LDS_SQ_CTRL_REG__SQ_OVERRIDE__SHIFT 0x17
+#define CGTS_CU15_LDS_SQ_CTRL_REG__SQ_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU15_LDS_SQ_CTRL_REG__SQ_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU15_LDS_SQ_CTRL_REG__SQ_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU15_LDS_SQ_CTRL_REG__SQ_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU15_LDS_SQ_CTRL_REG__SQ_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU15_LDS_SQ_CTRL_REG__SQ_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU15_TA_CTRL_REG__TA_MASK 0x7f
+#define CGTS_CU15_TA_CTRL_REG__TA__SHIFT 0x0
+#define CGTS_CU15_TA_CTRL_REG__TA_OVERRIDE_MASK 0x80
+#define CGTS_CU15_TA_CTRL_REG__TA_OVERRIDE__SHIFT 0x7
+#define CGTS_CU15_TA_CTRL_REG__TA_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU15_TA_CTRL_REG__TA_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU15_TA_CTRL_REG__TA_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU15_TA_CTRL_REG__TA_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU15_TA_CTRL_REG__TA_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU15_TA_CTRL_REG__TA_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU15_SP1_CTRL_REG__SP10_MASK 0x7f
+#define CGTS_CU15_SP1_CTRL_REG__SP10__SHIFT 0x0
+#define CGTS_CU15_SP1_CTRL_REG__SP10_OVERRIDE_MASK 0x80
+#define CGTS_CU15_SP1_CTRL_REG__SP10_OVERRIDE__SHIFT 0x7
+#define CGTS_CU15_SP1_CTRL_REG__SP10_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU15_SP1_CTRL_REG__SP10_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU15_SP1_CTRL_REG__SP10_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU15_SP1_CTRL_REG__SP10_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU15_SP1_CTRL_REG__SP10_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU15_SP1_CTRL_REG__SP10_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU15_SP1_CTRL_REG__SP11_MASK 0x7f0000
+#define CGTS_CU15_SP1_CTRL_REG__SP11__SHIFT 0x10
+#define CGTS_CU15_SP1_CTRL_REG__SP11_OVERRIDE_MASK 0x800000
+#define CGTS_CU15_SP1_CTRL_REG__SP11_OVERRIDE__SHIFT 0x17
+#define CGTS_CU15_SP1_CTRL_REG__SP11_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU15_SP1_CTRL_REG__SP11_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU15_SP1_CTRL_REG__SP11_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU15_SP1_CTRL_REG__SP11_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU15_SP1_CTRL_REG__SP11_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU15_SP1_CTRL_REG__SP11_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTS_CU15_TD_TCP_CTRL_REG__TD_MASK 0x7f
+#define CGTS_CU15_TD_TCP_CTRL_REG__TD__SHIFT 0x0
+#define CGTS_CU15_TD_TCP_CTRL_REG__TD_OVERRIDE_MASK 0x80
+#define CGTS_CU15_TD_TCP_CTRL_REG__TD_OVERRIDE__SHIFT 0x7
+#define CGTS_CU15_TD_TCP_CTRL_REG__TD_BUSY_OVERRIDE_MASK 0x300
+#define CGTS_CU15_TD_TCP_CTRL_REG__TD_BUSY_OVERRIDE__SHIFT 0x8
+#define CGTS_CU15_TD_TCP_CTRL_REG__TD_LS_OVERRIDE_MASK 0x400
+#define CGTS_CU15_TD_TCP_CTRL_REG__TD_LS_OVERRIDE__SHIFT 0xa
+#define CGTS_CU15_TD_TCP_CTRL_REG__TD_SIMDBUSY_OVERRIDE_MASK 0x800
+#define CGTS_CU15_TD_TCP_CTRL_REG__TD_SIMDBUSY_OVERRIDE__SHIFT 0xb
+#define CGTS_CU15_TD_TCP_CTRL_REG__TCP_MASK 0x7f0000
+#define CGTS_CU15_TD_TCP_CTRL_REG__TCP__SHIFT 0x10
+#define CGTS_CU15_TD_TCP_CTRL_REG__TCP_OVERRIDE_MASK 0x800000
+#define CGTS_CU15_TD_TCP_CTRL_REG__TCP_OVERRIDE__SHIFT 0x17
+#define CGTS_CU15_TD_TCP_CTRL_REG__TCP_BUSY_OVERRIDE_MASK 0x3000000
+#define CGTS_CU15_TD_TCP_CTRL_REG__TCP_BUSY_OVERRIDE__SHIFT 0x18
+#define CGTS_CU15_TD_TCP_CTRL_REG__TCP_LS_OVERRIDE_MASK 0x4000000
+#define CGTS_CU15_TD_TCP_CTRL_REG__TCP_LS_OVERRIDE__SHIFT 0x1a
+#define CGTS_CU15_TD_TCP_CTRL_REG__TCP_SIMDBUSY_OVERRIDE_MASK 0x8000000
+#define CGTS_CU15_TD_TCP_CTRL_REG__TCP_SIMDBUSY_OVERRIDE__SHIFT 0x1b
+#define CGTT_SPI_CLK_CTRL__ON_DELAY_MASK 0xf
+#define CGTT_SPI_CLK_CTRL__ON_DELAY__SHIFT 0x0
+#define CGTT_SPI_CLK_CTRL__OFF_HYSTERESIS_MASK 0xff0
+#define CGTT_SPI_CLK_CTRL__OFF_HYSTERESIS__SHIFT 0x4
+#define CGTT_SPI_CLK_CTRL__GRP5_CG_OFF_HYST_MASK 0xfc0000
+#define CGTT_SPI_CLK_CTRL__GRP5_CG_OFF_HYST__SHIFT 0x12
+#define CGTT_SPI_CLK_CTRL__GRP5_CG_OVERRIDE_MASK 0x1000000
+#define CGTT_SPI_CLK_CTRL__GRP5_CG_OVERRIDE__SHIFT 0x18
+#define CGTT_SPI_CLK_CTRL__ALL_CLK_ON_OVERRIDE_MASK 0x4000000
+#define CGTT_SPI_CLK_CTRL__ALL_CLK_ON_OVERRIDE__SHIFT 0x1a
+#define CGTT_SPI_CLK_CTRL__GRP3_OVERRIDE_MASK 0x8000000
+#define CGTT_SPI_CLK_CTRL__GRP3_OVERRIDE__SHIFT 0x1b
+#define CGTT_SPI_CLK_CTRL__GRP2_OVERRIDE_MASK 0x10000000
+#define CGTT_SPI_CLK_CTRL__GRP2_OVERRIDE__SHIFT 0x1c
+#define CGTT_SPI_CLK_CTRL__GRP1_OVERRIDE_MASK 0x20000000
+#define CGTT_SPI_CLK_CTRL__GRP1_OVERRIDE__SHIFT 0x1d
+#define CGTT_SPI_CLK_CTRL__GRP0_OVERRIDE_MASK 0x40000000
+#define CGTT_SPI_CLK_CTRL__GRP0_OVERRIDE__SHIFT 0x1e
+#define CGTT_SPI_CLK_CTRL__REG_OVERRIDE_MASK 0x80000000
+#define CGTT_SPI_CLK_CTRL__REG_OVERRIDE__SHIFT 0x1f
+#define CGTT_PC_CLK_CTRL__ON_DELAY_MASK 0xf
+#define CGTT_PC_CLK_CTRL__ON_DELAY__SHIFT 0x0
+#define CGTT_PC_CLK_CTRL__OFF_HYSTERESIS_MASK 0xff0
+#define CGTT_PC_CLK_CTRL__OFF_HYSTERESIS__SHIFT 0x4
+#define CGTT_PC_CLK_CTRL__GRP5_CG_OFF_HYST_MASK 0xfc0000
+#define CGTT_PC_CLK_CTRL__GRP5_CG_OFF_HYST__SHIFT 0x12
+#define CGTT_PC_CLK_CTRL__GRP5_CG_OVERRIDE_MASK 0x1000000
+#define CGTT_PC_CLK_CTRL__GRP5_CG_OVERRIDE__SHIFT 0x18
+#define CGTT_PC_CLK_CTRL__BACK_CLK_ON_OVERRIDE_MASK 0x2000000
+#define CGTT_PC_CLK_CTRL__BACK_CLK_ON_OVERRIDE__SHIFT 0x19
+#define CGTT_PC_CLK_CTRL__FRONT_CLK_ON_OVERRIDE_MASK 0x4000000
+#define CGTT_PC_CLK_CTRL__FRONT_CLK_ON_OVERRIDE__SHIFT 0x1a
+#define CGTT_PC_CLK_CTRL__CORE3_OVERRIDE_MASK 0x8000000
+#define CGTT_PC_CLK_CTRL__CORE3_OVERRIDE__SHIFT 0x1b
+#define CGTT_PC_CLK_CTRL__CORE2_OVERRIDE_MASK 0x10000000
+#define CGTT_PC_CLK_CTRL__CORE2_OVERRIDE__SHIFT 0x1c
+#define CGTT_PC_CLK_CTRL__CORE1_OVERRIDE_MASK 0x20000000
+#define CGTT_PC_CLK_CTRL__CORE1_OVERRIDE__SHIFT 0x1d
+#define CGTT_PC_CLK_CTRL__CORE0_OVERRIDE_MASK 0x40000000
+#define CGTT_PC_CLK_CTRL__CORE0_OVERRIDE__SHIFT 0x1e
+#define CGTT_PC_CLK_CTRL__REG_OVERRIDE_MASK 0x80000000
+#define CGTT_PC_CLK_CTRL__REG_OVERRIDE__SHIFT 0x1f
+#define CGTT_BCI_CLK_CTRL__ON_DELAY_MASK 0xf
+#define CGTT_BCI_CLK_CTRL__ON_DELAY__SHIFT 0x0
+#define CGTT_BCI_CLK_CTRL__OFF_HYSTERESIS_MASK 0xff0
+#define CGTT_BCI_CLK_CTRL__OFF_HYSTERESIS__SHIFT 0x4
+#define CGTT_BCI_CLK_CTRL__RESERVED_MASK 0xfff000
+#define CGTT_BCI_CLK_CTRL__RESERVED__SHIFT 0xc
+#define CGTT_BCI_CLK_CTRL__CORE6_OVERRIDE_MASK 0x1000000
+#define CGTT_BCI_CLK_CTRL__CORE6_OVERRIDE__SHIFT 0x18
+#define CGTT_BCI_CLK_CTRL__CORE5_OVERRIDE_MASK 0x2000000
+#define CGTT_BCI_CLK_CTRL__CORE5_OVERRIDE__SHIFT 0x19
+#define CGTT_BCI_CLK_CTRL__CORE4_OVERRIDE_MASK 0x4000000
+#define CGTT_BCI_CLK_CTRL__CORE4_OVERRIDE__SHIFT 0x1a
+#define CGTT_BCI_CLK_CTRL__CORE3_OVERRIDE_MASK 0x8000000
+#define CGTT_BCI_CLK_CTRL__CORE3_OVERRIDE__SHIFT 0x1b
+#define CGTT_BCI_CLK_CTRL__CORE2_OVERRIDE_MASK 0x10000000
+#define CGTT_BCI_CLK_CTRL__CORE2_OVERRIDE__SHIFT 0x1c
+#define CGTT_BCI_CLK_CTRL__CORE1_OVERRIDE_MASK 0x20000000
+#define CGTT_BCI_CLK_CTRL__CORE1_OVERRIDE__SHIFT 0x1d
+#define CGTT_BCI_CLK_CTRL__CORE0_OVERRIDE_MASK 0x40000000
+#define CGTT_BCI_CLK_CTRL__CORE0_OVERRIDE__SHIFT 0x1e
+#define CGTT_BCI_CLK_CTRL__REG_OVERRIDE_MASK 0x80000000
+#define CGTT_BCI_CLK_CTRL__REG_OVERRIDE__SHIFT 0x1f
+#define SPI_WF_LIFETIME_CNTL__SAMPLE_PERIOD_MASK 0xf
+#define SPI_WF_LIFETIME_CNTL__SAMPLE_PERIOD__SHIFT 0x0
+#define SPI_WF_LIFETIME_CNTL__EN_MASK 0x10
+#define SPI_WF_LIFETIME_CNTL__EN__SHIFT 0x4
+#define SPI_WF_LIFETIME_LIMIT_0__MAX_CNT_MASK 0x7fffffff
+#define SPI_WF_LIFETIME_LIMIT_0__MAX_CNT__SHIFT 0x0
+#define SPI_WF_LIFETIME_LIMIT_0__EN_WARN_MASK 0x80000000
+#define SPI_WF_LIFETIME_LIMIT_0__EN_WARN__SHIFT 0x1f
+#define SPI_WF_LIFETIME_LIMIT_1__MAX_CNT_MASK 0x7fffffff
+#define SPI_WF_LIFETIME_LIMIT_1__MAX_CNT__SHIFT 0x0
+#define SPI_WF_LIFETIME_LIMIT_1__EN_WARN_MASK 0x80000000
+#define SPI_WF_LIFETIME_LIMIT_1__EN_WARN__SHIFT 0x1f
+#define SPI_WF_LIFETIME_LIMIT_2__MAX_CNT_MASK 0x7fffffff
+#define SPI_WF_LIFETIME_LIMIT_2__MAX_CNT__SHIFT 0x0
+#define SPI_WF_LIFETIME_LIMIT_2__EN_WARN_MASK 0x80000000
+#define SPI_WF_LIFETIME_LIMIT_2__EN_WARN__SHIFT 0x1f
+#define SPI_WF_LIFETIME_LIMIT_3__MAX_CNT_MASK 0x7fffffff
+#define SPI_WF_LIFETIME_LIMIT_3__MAX_CNT__SHIFT 0x0
+#define SPI_WF_LIFETIME_LIMIT_3__EN_WARN_MASK 0x80000000
+#define SPI_WF_LIFETIME_LIMIT_3__EN_WARN__SHIFT 0x1f
+#define SPI_WF_LIFETIME_LIMIT_4__MAX_CNT_MASK 0x7fffffff
+#define SPI_WF_LIFETIME_LIMIT_4__MAX_CNT__SHIFT 0x0
+#define SPI_WF_LIFETIME_LIMIT_4__EN_WARN_MASK 0x80000000
+#define SPI_WF_LIFETIME_LIMIT_4__EN_WARN__SHIFT 0x1f
+#define SPI_WF_LIFETIME_LIMIT_5__MAX_CNT_MASK 0x7fffffff
+#define SPI_WF_LIFETIME_LIMIT_5__MAX_CNT__SHIFT 0x0
+#define SPI_WF_LIFETIME_LIMIT_5__EN_WARN_MASK 0x80000000
+#define SPI_WF_LIFETIME_LIMIT_5__EN_WARN__SHIFT 0x1f
+#define SPI_WF_LIFETIME_LIMIT_6__MAX_CNT_MASK 0x7fffffff
+#define SPI_WF_LIFETIME_LIMIT_6__MAX_CNT__SHIFT 0x0
+#define SPI_WF_LIFETIME_LIMIT_6__EN_WARN_MASK 0x80000000
+#define SPI_WF_LIFETIME_LIMIT_6__EN_WARN__SHIFT 0x1f
+#define SPI_WF_LIFETIME_LIMIT_7__MAX_CNT_MASK 0x7fffffff
+#define SPI_WF_LIFETIME_LIMIT_7__MAX_CNT__SHIFT 0x0
+#define SPI_WF_LIFETIME_LIMIT_7__EN_WARN_MASK 0x80000000
+#define SPI_WF_LIFETIME_LIMIT_7__EN_WARN__SHIFT 0x1f
+#define SPI_WF_LIFETIME_LIMIT_8__MAX_CNT_MASK 0x7fffffff
+#define SPI_WF_LIFETIME_LIMIT_8__MAX_CNT__SHIFT 0x0
+#define SPI_WF_LIFETIME_LIMIT_8__EN_WARN_MASK 0x80000000
+#define SPI_WF_LIFETIME_LIMIT_8__EN_WARN__SHIFT 0x1f
+#define SPI_WF_LIFETIME_LIMIT_9__MAX_CNT_MASK 0x7fffffff
+#define SPI_WF_LIFETIME_LIMIT_9__MAX_CNT__SHIFT 0x0
+#define SPI_WF_LIFETIME_LIMIT_9__EN_WARN_MASK 0x80000000
+#define SPI_WF_LIFETIME_LIMIT_9__EN_WARN__SHIFT 0x1f
+#define SPI_WF_LIFETIME_STATUS_0__MAX_CNT_MASK 0x7fffffff
+#define SPI_WF_LIFETIME_STATUS_0__MAX_CNT__SHIFT 0x0
+#define SPI_WF_LIFETIME_STATUS_0__INT_SENT_MASK 0x80000000
+#define SPI_WF_LIFETIME_STATUS_0__INT_SENT__SHIFT 0x1f
+#define SPI_WF_LIFETIME_STATUS_1__MAX_CNT_MASK 0x7fffffff
+#define SPI_WF_LIFETIME_STATUS_1__MAX_CNT__SHIFT 0x0
+#define SPI_WF_LIFETIME_STATUS_1__INT_SENT_MASK 0x80000000
+#define SPI_WF_LIFETIME_STATUS_1__INT_SENT__SHIFT 0x1f
+#define SPI_WF_LIFETIME_STATUS_2__MAX_CNT_MASK 0x7fffffff
+#define SPI_WF_LIFETIME_STATUS_2__MAX_CNT__SHIFT 0x0
+#define SPI_WF_LIFETIME_STATUS_2__INT_SENT_MASK 0x80000000
+#define SPI_WF_LIFETIME_STATUS_2__INT_SENT__SHIFT 0x1f
+#define SPI_WF_LIFETIME_STATUS_3__MAX_CNT_MASK 0x7fffffff
+#define SPI_WF_LIFETIME_STATUS_3__MAX_CNT__SHIFT 0x0
+#define SPI_WF_LIFETIME_STATUS_3__INT_SENT_MASK 0x80000000
+#define SPI_WF_LIFETIME_STATUS_3__INT_SENT__SHIFT 0x1f
+#define SPI_WF_LIFETIME_STATUS_4__MAX_CNT_MASK 0x7fffffff
+#define SPI_WF_LIFETIME_STATUS_4__MAX_CNT__SHIFT 0x0
+#define SPI_WF_LIFETIME_STATUS_4__INT_SENT_MASK 0x80000000
+#define SPI_WF_LIFETIME_STATUS_4__INT_SENT__SHIFT 0x1f
+#define SPI_WF_LIFETIME_STATUS_5__MAX_CNT_MASK 0x7fffffff
+#define SPI_WF_LIFETIME_STATUS_5__MAX_CNT__SHIFT 0x0
+#define SPI_WF_LIFETIME_STATUS_5__INT_SENT_MASK 0x80000000
+#define SPI_WF_LIFETIME_STATUS_5__INT_SENT__SHIFT 0x1f
+#define SPI_WF_LIFETIME_STATUS_6__MAX_CNT_MASK 0x7fffffff
+#define SPI_WF_LIFETIME_STATUS_6__MAX_CNT__SHIFT 0x0
+#define SPI_WF_LIFETIME_STATUS_6__INT_SENT_MASK 0x80000000
+#define SPI_WF_LIFETIME_STATUS_6__INT_SENT__SHIFT 0x1f
+#define SPI_WF_LIFETIME_STATUS_7__MAX_CNT_MASK 0x7fffffff
+#define SPI_WF_LIFETIME_STATUS_7__MAX_CNT__SHIFT 0x0
+#define SPI_WF_LIFETIME_STATUS_7__INT_SENT_MASK 0x80000000
+#define SPI_WF_LIFETIME_STATUS_7__INT_SENT__SHIFT 0x1f
+#define SPI_WF_LIFETIME_STATUS_8__MAX_CNT_MASK 0x7fffffff
+#define SPI_WF_LIFETIME_STATUS_8__MAX_CNT__SHIFT 0x0
+#define SPI_WF_LIFETIME_STATUS_8__INT_SENT_MASK 0x80000000
+#define SPI_WF_LIFETIME_STATUS_8__INT_SENT__SHIFT 0x1f
+#define SPI_WF_LIFETIME_STATUS_9__MAX_CNT_MASK 0x7fffffff
+#define SPI_WF_LIFETIME_STATUS_9__MAX_CNT__SHIFT 0x0
+#define SPI_WF_LIFETIME_STATUS_9__INT_SENT_MASK 0x80000000
+#define SPI_WF_LIFETIME_STATUS_9__INT_SENT__SHIFT 0x1f
+#define SPI_WF_LIFETIME_STATUS_10__MAX_CNT_MASK 0x7fffffff
+#define SPI_WF_LIFETIME_STATUS_10__MAX_CNT__SHIFT 0x0
+#define SPI_WF_LIFETIME_STATUS_10__INT_SENT_MASK 0x80000000
+#define SPI_WF_LIFETIME_STATUS_10__INT_SENT__SHIFT 0x1f
+#define SPI_WF_LIFETIME_STATUS_11__MAX_CNT_MASK 0x7fffffff
+#define SPI_WF_LIFETIME_STATUS_11__MAX_CNT__SHIFT 0x0
+#define SPI_WF_LIFETIME_STATUS_11__INT_SENT_MASK 0x80000000
+#define SPI_WF_LIFETIME_STATUS_11__INT_SENT__SHIFT 0x1f
+#define SPI_WF_LIFETIME_STATUS_12__MAX_CNT_MASK 0x7fffffff
+#define SPI_WF_LIFETIME_STATUS_12__MAX_CNT__SHIFT 0x0
+#define SPI_WF_LIFETIME_STATUS_12__INT_SENT_MASK 0x80000000
+#define SPI_WF_LIFETIME_STATUS_12__INT_SENT__SHIFT 0x1f
+#define SPI_WF_LIFETIME_STATUS_13__MAX_CNT_MASK 0x7fffffff
+#define SPI_WF_LIFETIME_STATUS_13__MAX_CNT__SHIFT 0x0
+#define SPI_WF_LIFETIME_STATUS_13__INT_SENT_MASK 0x80000000
+#define SPI_WF_LIFETIME_STATUS_13__INT_SENT__SHIFT 0x1f
+#define SPI_WF_LIFETIME_STATUS_14__MAX_CNT_MASK 0x7fffffff
+#define SPI_WF_LIFETIME_STATUS_14__MAX_CNT__SHIFT 0x0
+#define SPI_WF_LIFETIME_STATUS_14__INT_SENT_MASK 0x80000000
+#define SPI_WF_LIFETIME_STATUS_14__INT_SENT__SHIFT 0x1f
+#define SPI_WF_LIFETIME_STATUS_15__MAX_CNT_MASK 0x7fffffff
+#define SPI_WF_LIFETIME_STATUS_15__MAX_CNT__SHIFT 0x0
+#define SPI_WF_LIFETIME_STATUS_15__INT_SENT_MASK 0x80000000
+#define SPI_WF_LIFETIME_STATUS_15__INT_SENT__SHIFT 0x1f
+#define SPI_WF_LIFETIME_STATUS_16__MAX_CNT_MASK 0x7fffffff
+#define SPI_WF_LIFETIME_STATUS_16__MAX_CNT__SHIFT 0x0
+#define SPI_WF_LIFETIME_STATUS_16__INT_SENT_MASK 0x80000000
+#define SPI_WF_LIFETIME_STATUS_16__INT_SENT__SHIFT 0x1f
+#define SPI_WF_LIFETIME_STATUS_17__MAX_CNT_MASK 0x7fffffff
+#define SPI_WF_LIFETIME_STATUS_17__MAX_CNT__SHIFT 0x0
+#define SPI_WF_LIFETIME_STATUS_17__INT_SENT_MASK 0x80000000
+#define SPI_WF_LIFETIME_STATUS_17__INT_SENT__SHIFT 0x1f
+#define SPI_WF_LIFETIME_STATUS_18__MAX_CNT_MASK 0x7fffffff
+#define SPI_WF_LIFETIME_STATUS_18__MAX_CNT__SHIFT 0x0
+#define SPI_WF_LIFETIME_STATUS_18__INT_SENT_MASK 0x80000000
+#define SPI_WF_LIFETIME_STATUS_18__INT_SENT__SHIFT 0x1f
+#define SPI_WF_LIFETIME_STATUS_19__MAX_CNT_MASK 0x7fffffff
+#define SPI_WF_LIFETIME_STATUS_19__MAX_CNT__SHIFT 0x0
+#define SPI_WF_LIFETIME_STATUS_19__INT_SENT_MASK 0x80000000
+#define SPI_WF_LIFETIME_STATUS_19__INT_SENT__SHIFT 0x1f
+#define SPI_WF_LIFETIME_STATUS_20__MAX_CNT_MASK 0x7fffffff
+#define SPI_WF_LIFETIME_STATUS_20__MAX_CNT__SHIFT 0x0
+#define SPI_WF_LIFETIME_STATUS_20__INT_SENT_MASK 0x80000000
+#define SPI_WF_LIFETIME_STATUS_20__INT_SENT__SHIFT 0x1f
+#define SPI_WF_LIFETIME_DEBUG__START_VALUE_MASK 0x7fffffff
+#define SPI_WF_LIFETIME_DEBUG__START_VALUE__SHIFT 0x0
+#define SPI_WF_LIFETIME_DEBUG__OVERRIDE_EN_MASK 0x80000000
+#define SPI_WF_LIFETIME_DEBUG__OVERRIDE_EN__SHIFT 0x1f
+#define SPI_SLAVE_DEBUG_BUSY__LS_VTX_BUSY_MASK 0x1
+#define SPI_SLAVE_DEBUG_BUSY__LS_VTX_BUSY__SHIFT 0x0
+#define SPI_SLAVE_DEBUG_BUSY__HS_VTX_BUSY_MASK 0x2
+#define SPI_SLAVE_DEBUG_BUSY__HS_VTX_BUSY__SHIFT 0x1
+#define SPI_SLAVE_DEBUG_BUSY__ES_VTX_BUSY_MASK 0x4
+#define SPI_SLAVE_DEBUG_BUSY__ES_VTX_BUSY__SHIFT 0x2
+#define SPI_SLAVE_DEBUG_BUSY__GS_VTX_BUSY_MASK 0x8
+#define SPI_SLAVE_DEBUG_BUSY__GS_VTX_BUSY__SHIFT 0x3
+#define SPI_SLAVE_DEBUG_BUSY__VS_VTX_BUSY_MASK 0x10
+#define SPI_SLAVE_DEBUG_BUSY__VS_VTX_BUSY__SHIFT 0x4
+#define SPI_SLAVE_DEBUG_BUSY__VGPR_WC00_BUSY_MASK 0x20
+#define SPI_SLAVE_DEBUG_BUSY__VGPR_WC00_BUSY__SHIFT 0x5
+#define SPI_SLAVE_DEBUG_BUSY__VGPR_WC01_BUSY_MASK 0x40
+#define SPI_SLAVE_DEBUG_BUSY__VGPR_WC01_BUSY__SHIFT 0x6
+#define SPI_SLAVE_DEBUG_BUSY__VGPR_WC10_BUSY_MASK 0x80
+#define SPI_SLAVE_DEBUG_BUSY__VGPR_WC10_BUSY__SHIFT 0x7
+#define SPI_SLAVE_DEBUG_BUSY__VGPR_WC11_BUSY_MASK 0x100
+#define SPI_SLAVE_DEBUG_BUSY__VGPR_WC11_BUSY__SHIFT 0x8
+#define SPI_SLAVE_DEBUG_BUSY__SGPR_WC00_BUSY_MASK 0x200
+#define SPI_SLAVE_DEBUG_BUSY__SGPR_WC00_BUSY__SHIFT 0x9
+#define SPI_SLAVE_DEBUG_BUSY__SGPR_WC01_BUSY_MASK 0x400
+#define SPI_SLAVE_DEBUG_BUSY__SGPR_WC01_BUSY__SHIFT 0xa
+#define SPI_SLAVE_DEBUG_BUSY__SGPR_WC02_BUSY_MASK 0x800
+#define SPI_SLAVE_DEBUG_BUSY__SGPR_WC02_BUSY__SHIFT 0xb
+#define SPI_SLAVE_DEBUG_BUSY__SGPR_WC03_BUSY_MASK 0x1000
+#define SPI_SLAVE_DEBUG_BUSY__SGPR_WC03_BUSY__SHIFT 0xc
+#define SPI_SLAVE_DEBUG_BUSY__SGPR_WC10_BUSY_MASK 0x2000
+#define SPI_SLAVE_DEBUG_BUSY__SGPR_WC10_BUSY__SHIFT 0xd
+#define SPI_SLAVE_DEBUG_BUSY__SGPR_WC11_BUSY_MASK 0x4000
+#define SPI_SLAVE_DEBUG_BUSY__SGPR_WC11_BUSY__SHIFT 0xe
+#define SPI_SLAVE_DEBUG_BUSY__SGPR_WC12_BUSY_MASK 0x8000
+#define SPI_SLAVE_DEBUG_BUSY__SGPR_WC12_BUSY__SHIFT 0xf
+#define SPI_SLAVE_DEBUG_BUSY__SGPR_WC13_BUSY_MASK 0x10000
+#define SPI_SLAVE_DEBUG_BUSY__SGPR_WC13_BUSY__SHIFT 0x10
+#define SPI_SLAVE_DEBUG_BUSY__WAVEBUFFER0_BUSY_MASK 0x20000
+#define SPI_SLAVE_DEBUG_BUSY__WAVEBUFFER0_BUSY__SHIFT 0x11
+#define SPI_SLAVE_DEBUG_BUSY__WAVEBUFFER1_BUSY_MASK 0x40000
+#define SPI_SLAVE_DEBUG_BUSY__WAVEBUFFER1_BUSY__SHIFT 0x12
+#define SPI_SLAVE_DEBUG_BUSY__WAVE_WC0_BUSY_MASK 0x80000
+#define SPI_SLAVE_DEBUG_BUSY__WAVE_WC0_BUSY__SHIFT 0x13
+#define SPI_SLAVE_DEBUG_BUSY__WAVE_WC1_BUSY_MASK 0x100000
+#define SPI_SLAVE_DEBUG_BUSY__WAVE_WC1_BUSY__SHIFT 0x14
+#define SPI_SLAVE_DEBUG_BUSY__EVENT_CNTL_BUSY_MASK 0x200000
+#define SPI_SLAVE_DEBUG_BUSY__EVENT_CNTL_BUSY__SHIFT 0x15
+#define SPI_SLAVE_DEBUG_BUSY__SAVE_CTX_BUSY_MASK 0x400000
+#define SPI_SLAVE_DEBUG_BUSY__SAVE_CTX_BUSY__SHIFT 0x16
+#define SPI_LB_CTR_CTRL__LOAD_MASK 0x1
+#define SPI_LB_CTR_CTRL__LOAD__SHIFT 0x0
+#define SPI_LB_CU_MASK__CU_MASK_MASK 0xffff
+#define SPI_LB_CU_MASK__CU_MASK__SHIFT 0x0
+#define SPI_LB_DATA_REG__CNT_DATA_MASK 0xffffffff
+#define SPI_LB_DATA_REG__CNT_DATA__SHIFT 0x0
+#define SPI_PG_ENABLE_STATIC_CU_MASK__CU_MASK_MASK 0xffff
+#define SPI_PG_ENABLE_STATIC_CU_MASK__CU_MASK__SHIFT 0x0
+#define SPI_GDS_CREDITS__DS_DATA_CREDITS_MASK 0xff
+#define SPI_GDS_CREDITS__DS_DATA_CREDITS__SHIFT 0x0
+#define SPI_GDS_CREDITS__DS_CMD_CREDITS_MASK 0xff00
+#define SPI_GDS_CREDITS__DS_CMD_CREDITS__SHIFT 0x8
+#define SPI_GDS_CREDITS__UNUSED_MASK 0xffff0000
+#define SPI_GDS_CREDITS__UNUSED__SHIFT 0x10
+#define SPI_SX_EXPORT_BUFFER_SIZES__COLOR_BUFFER_SIZE_MASK 0xffff
+#define SPI_SX_EXPORT_BUFFER_SIZES__COLOR_BUFFER_SIZE__SHIFT 0x0
+#define SPI_SX_EXPORT_BUFFER_SIZES__POSITION_BUFFER_SIZE_MASK 0xffff0000
+#define SPI_SX_EXPORT_BUFFER_SIZES__POSITION_BUFFER_SIZE__SHIFT 0x10
+#define SPI_SX_SCOREBOARD_BUFFER_SIZES__COLOR_SCOREBOARD_SIZE_MASK 0xffff
+#define SPI_SX_SCOREBOARD_BUFFER_SIZES__COLOR_SCOREBOARD_SIZE__SHIFT 0x0
+#define SPI_SX_SCOREBOARD_BUFFER_SIZES__POSITION_SCOREBOARD_SIZE_MASK 0xffff0000
+#define SPI_SX_SCOREBOARD_BUFFER_SIZES__POSITION_SCOREBOARD_SIZE__SHIFT 0x10
+#define SPI_CSQ_WF_ACTIVE_STATUS__ACTIVE_MASK 0xffffffff
+#define SPI_CSQ_WF_ACTIVE_STATUS__ACTIVE__SHIFT 0x0
+#define SPI_CSQ_WF_ACTIVE_COUNT_0__COUNT_MASK 0x7ff
+#define SPI_CSQ_WF_ACTIVE_COUNT_0__COUNT__SHIFT 0x0
+#define SPI_CSQ_WF_ACTIVE_COUNT_1__COUNT_MASK 0x7ff
+#define SPI_CSQ_WF_ACTIVE_COUNT_1__COUNT__SHIFT 0x0
+#define SPI_CSQ_WF_ACTIVE_COUNT_2__COUNT_MASK 0x7ff
+#define SPI_CSQ_WF_ACTIVE_COUNT_2__COUNT__SHIFT 0x0
+#define SPI_CSQ_WF_ACTIVE_COUNT_3__COUNT_MASK 0x7ff
+#define SPI_CSQ_WF_ACTIVE_COUNT_3__COUNT__SHIFT 0x0
+#define SPI_CSQ_WF_ACTIVE_COUNT_4__COUNT_MASK 0x7ff
+#define SPI_CSQ_WF_ACTIVE_COUNT_4__COUNT__SHIFT 0x0
+#define SPI_CSQ_WF_ACTIVE_COUNT_5__COUNT_MASK 0x7ff
+#define SPI_CSQ_WF_ACTIVE_COUNT_5__COUNT__SHIFT 0x0
+#define SPI_CSQ_WF_ACTIVE_COUNT_6__COUNT_MASK 0x7ff
+#define SPI_CSQ_WF_ACTIVE_COUNT_6__COUNT__SHIFT 0x0
+#define SPI_CSQ_WF_ACTIVE_COUNT_7__COUNT_MASK 0x7ff
+#define SPI_CSQ_WF_ACTIVE_COUNT_7__COUNT__SHIFT 0x0
+#define BCI_DEBUG_READ__DATA_MASK 0xffffff
+#define BCI_DEBUG_READ__DATA__SHIFT 0x0
+#define SPI_P0_TRAP_SCREEN_PSBA_LO__MEM_BASE_MASK 0xffffffff
+#define SPI_P0_TRAP_SCREEN_PSBA_LO__MEM_BASE__SHIFT 0x0
+#define SPI_P0_TRAP_SCREEN_PSBA_HI__MEM_BASE_MASK 0xff
+#define SPI_P0_TRAP_SCREEN_PSBA_HI__MEM_BASE__SHIFT 0x0
+#define SPI_P0_TRAP_SCREEN_PSMA_LO__MEM_BASE_MASK 0xffffffff
+#define SPI_P0_TRAP_SCREEN_PSMA_LO__MEM_BASE__SHIFT 0x0
+#define SPI_P0_TRAP_SCREEN_PSMA_HI__MEM_BASE_MASK 0xff
+#define SPI_P0_TRAP_SCREEN_PSMA_HI__MEM_BASE__SHIFT 0x0
+#define SPI_P0_TRAP_SCREEN_GPR_MIN__VGPR_MIN_MASK 0x3f
+#define SPI_P0_TRAP_SCREEN_GPR_MIN__VGPR_MIN__SHIFT 0x0
+#define SPI_P0_TRAP_SCREEN_GPR_MIN__SGPR_MIN_MASK 0x3c0
+#define SPI_P0_TRAP_SCREEN_GPR_MIN__SGPR_MIN__SHIFT 0x6
+#define SPI_P1_TRAP_SCREEN_PSBA_LO__MEM_BASE_MASK 0xffffffff
+#define SPI_P1_TRAP_SCREEN_PSBA_LO__MEM_BASE__SHIFT 0x0
+#define SPI_P1_TRAP_SCREEN_PSBA_HI__MEM_BASE_MASK 0xff
+#define SPI_P1_TRAP_SCREEN_PSBA_HI__MEM_BASE__SHIFT 0x0
+#define SPI_P1_TRAP_SCREEN_PSMA_LO__MEM_BASE_MASK 0xffffffff
+#define SPI_P1_TRAP_SCREEN_PSMA_LO__MEM_BASE__SHIFT 0x0
+#define SPI_P1_TRAP_SCREEN_PSMA_HI__MEM_BASE_MASK 0xff
+#define SPI_P1_TRAP_SCREEN_PSMA_HI__MEM_BASE__SHIFT 0x0
+#define SPI_P1_TRAP_SCREEN_GPR_MIN__VGPR_MIN_MASK 0x3f
+#define SPI_P1_TRAP_SCREEN_GPR_MIN__VGPR_MIN__SHIFT 0x0
+#define SPI_P1_TRAP_SCREEN_GPR_MIN__SGPR_MIN_MASK 0x3c0
+#define SPI_P1_TRAP_SCREEN_GPR_MIN__SGPR_MIN__SHIFT 0x6
+#define SPI_SHADER_TBA_LO_PS__MEM_BASE_MASK 0xffffffff
+#define SPI_SHADER_TBA_LO_PS__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_TBA_HI_PS__MEM_BASE_MASK 0xff
+#define SPI_SHADER_TBA_HI_PS__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_TMA_LO_PS__MEM_BASE_MASK 0xffffffff
+#define SPI_SHADER_TMA_LO_PS__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_TMA_HI_PS__MEM_BASE_MASK 0xff
+#define SPI_SHADER_TMA_HI_PS__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_PGM_LO_PS__MEM_BASE_MASK 0xffffffff
+#define SPI_SHADER_PGM_LO_PS__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_PGM_HI_PS__MEM_BASE_MASK 0xff
+#define SPI_SHADER_PGM_HI_PS__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC1_PS__VGPRS_MASK 0x3f
+#define SPI_SHADER_PGM_RSRC1_PS__VGPRS__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC1_PS__SGPRS_MASK 0x3c0
+#define SPI_SHADER_PGM_RSRC1_PS__SGPRS__SHIFT 0x6
+#define SPI_SHADER_PGM_RSRC1_PS__PRIORITY_MASK 0xc00
+#define SPI_SHADER_PGM_RSRC1_PS__PRIORITY__SHIFT 0xa
+#define SPI_SHADER_PGM_RSRC1_PS__FLOAT_MODE_MASK 0xff000
+#define SPI_SHADER_PGM_RSRC1_PS__FLOAT_MODE__SHIFT 0xc
+#define SPI_SHADER_PGM_RSRC1_PS__PRIV_MASK 0x100000
+#define SPI_SHADER_PGM_RSRC1_PS__PRIV__SHIFT 0x14
+#define SPI_SHADER_PGM_RSRC1_PS__DX10_CLAMP_MASK 0x200000
+#define SPI_SHADER_PGM_RSRC1_PS__DX10_CLAMP__SHIFT 0x15
+#define SPI_SHADER_PGM_RSRC1_PS__DEBUG_MODE_MASK 0x400000
+#define SPI_SHADER_PGM_RSRC1_PS__DEBUG_MODE__SHIFT 0x16
+#define SPI_SHADER_PGM_RSRC1_PS__IEEE_MODE_MASK 0x800000
+#define SPI_SHADER_PGM_RSRC1_PS__IEEE_MODE__SHIFT 0x17
+#define SPI_SHADER_PGM_RSRC1_PS__CU_GROUP_DISABLE_MASK 0x1000000
+#define SPI_SHADER_PGM_RSRC1_PS__CU_GROUP_DISABLE__SHIFT 0x18
+#define SPI_SHADER_PGM_RSRC1_PS__CACHE_CTL_MASK 0xe000000
+#define SPI_SHADER_PGM_RSRC1_PS__CACHE_CTL__SHIFT 0x19
+#define SPI_SHADER_PGM_RSRC1_PS__CDBG_USER_MASK 0x10000000
+#define SPI_SHADER_PGM_RSRC1_PS__CDBG_USER__SHIFT 0x1c
+#define SPI_SHADER_PGM_RSRC2_PS__SCRATCH_EN_MASK 0x1
+#define SPI_SHADER_PGM_RSRC2_PS__SCRATCH_EN__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC2_PS__USER_SGPR_MASK 0x3e
+#define SPI_SHADER_PGM_RSRC2_PS__USER_SGPR__SHIFT 0x1
+#define SPI_SHADER_PGM_RSRC2_PS__TRAP_PRESENT_MASK 0x40
+#define SPI_SHADER_PGM_RSRC2_PS__TRAP_PRESENT__SHIFT 0x6
+#define SPI_SHADER_PGM_RSRC2_PS__WAVE_CNT_EN_MASK 0x80
+#define SPI_SHADER_PGM_RSRC2_PS__WAVE_CNT_EN__SHIFT 0x7
+#define SPI_SHADER_PGM_RSRC2_PS__EXTRA_LDS_SIZE_MASK 0xff00
+#define SPI_SHADER_PGM_RSRC2_PS__EXTRA_LDS_SIZE__SHIFT 0x8
+#define SPI_SHADER_PGM_RSRC2_PS__EXCP_EN_MASK 0x1ff0000
+#define SPI_SHADER_PGM_RSRC2_PS__EXCP_EN__SHIFT 0x10
+#define SPI_SHADER_PGM_RSRC3_PS__CU_EN_MASK 0xffff
+#define SPI_SHADER_PGM_RSRC3_PS__CU_EN__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC3_PS__WAVE_LIMIT_MASK 0x3f0000
+#define SPI_SHADER_PGM_RSRC3_PS__WAVE_LIMIT__SHIFT 0x10
+#define SPI_SHADER_PGM_RSRC3_PS__LOCK_LOW_THRESHOLD_MASK 0x3c00000
+#define SPI_SHADER_PGM_RSRC3_PS__LOCK_LOW_THRESHOLD__SHIFT 0x16
+#define SPI_SHADER_USER_DATA_PS_0__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_PS_0__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_PS_1__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_PS_1__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_PS_2__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_PS_2__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_PS_3__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_PS_3__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_PS_4__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_PS_4__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_PS_5__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_PS_5__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_PS_6__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_PS_6__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_PS_7__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_PS_7__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_PS_8__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_PS_8__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_PS_9__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_PS_9__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_PS_10__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_PS_10__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_PS_11__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_PS_11__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_PS_12__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_PS_12__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_PS_13__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_PS_13__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_PS_14__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_PS_14__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_PS_15__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_PS_15__DATA__SHIFT 0x0
+#define SPI_SHADER_TBA_LO_VS__MEM_BASE_MASK 0xffffffff
+#define SPI_SHADER_TBA_LO_VS__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_TBA_HI_VS__MEM_BASE_MASK 0xff
+#define SPI_SHADER_TBA_HI_VS__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_TMA_LO_VS__MEM_BASE_MASK 0xffffffff
+#define SPI_SHADER_TMA_LO_VS__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_TMA_HI_VS__MEM_BASE_MASK 0xff
+#define SPI_SHADER_TMA_HI_VS__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_PGM_LO_VS__MEM_BASE_MASK 0xffffffff
+#define SPI_SHADER_PGM_LO_VS__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_PGM_HI_VS__MEM_BASE_MASK 0xff
+#define SPI_SHADER_PGM_HI_VS__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC1_VS__VGPRS_MASK 0x3f
+#define SPI_SHADER_PGM_RSRC1_VS__VGPRS__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC1_VS__SGPRS_MASK 0x3c0
+#define SPI_SHADER_PGM_RSRC1_VS__SGPRS__SHIFT 0x6
+#define SPI_SHADER_PGM_RSRC1_VS__PRIORITY_MASK 0xc00
+#define SPI_SHADER_PGM_RSRC1_VS__PRIORITY__SHIFT 0xa
+#define SPI_SHADER_PGM_RSRC1_VS__FLOAT_MODE_MASK 0xff000
+#define SPI_SHADER_PGM_RSRC1_VS__FLOAT_MODE__SHIFT 0xc
+#define SPI_SHADER_PGM_RSRC1_VS__PRIV_MASK 0x100000
+#define SPI_SHADER_PGM_RSRC1_VS__PRIV__SHIFT 0x14
+#define SPI_SHADER_PGM_RSRC1_VS__DX10_CLAMP_MASK 0x200000
+#define SPI_SHADER_PGM_RSRC1_VS__DX10_CLAMP__SHIFT 0x15
+#define SPI_SHADER_PGM_RSRC1_VS__DEBUG_MODE_MASK 0x400000
+#define SPI_SHADER_PGM_RSRC1_VS__DEBUG_MODE__SHIFT 0x16
+#define SPI_SHADER_PGM_RSRC1_VS__IEEE_MODE_MASK 0x800000
+#define SPI_SHADER_PGM_RSRC1_VS__IEEE_MODE__SHIFT 0x17
+#define SPI_SHADER_PGM_RSRC1_VS__VGPR_COMP_CNT_MASK 0x3000000
+#define SPI_SHADER_PGM_RSRC1_VS__VGPR_COMP_CNT__SHIFT 0x18
+#define SPI_SHADER_PGM_RSRC1_VS__CU_GROUP_ENABLE_MASK 0x4000000
+#define SPI_SHADER_PGM_RSRC1_VS__CU_GROUP_ENABLE__SHIFT 0x1a
+#define SPI_SHADER_PGM_RSRC1_VS__CACHE_CTL_MASK 0x38000000
+#define SPI_SHADER_PGM_RSRC1_VS__CACHE_CTL__SHIFT 0x1b
+#define SPI_SHADER_PGM_RSRC1_VS__CDBG_USER_MASK 0x40000000
+#define SPI_SHADER_PGM_RSRC1_VS__CDBG_USER__SHIFT 0x1e
+#define SPI_SHADER_PGM_RSRC2_VS__SCRATCH_EN_MASK 0x1
+#define SPI_SHADER_PGM_RSRC2_VS__SCRATCH_EN__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC2_VS__USER_SGPR_MASK 0x3e
+#define SPI_SHADER_PGM_RSRC2_VS__USER_SGPR__SHIFT 0x1
+#define SPI_SHADER_PGM_RSRC2_VS__TRAP_PRESENT_MASK 0x40
+#define SPI_SHADER_PGM_RSRC2_VS__TRAP_PRESENT__SHIFT 0x6
+#define SPI_SHADER_PGM_RSRC2_VS__OC_LDS_EN_MASK 0x80
+#define SPI_SHADER_PGM_RSRC2_VS__OC_LDS_EN__SHIFT 0x7
+#define SPI_SHADER_PGM_RSRC2_VS__SO_BASE0_EN_MASK 0x100
+#define SPI_SHADER_PGM_RSRC2_VS__SO_BASE0_EN__SHIFT 0x8
+#define SPI_SHADER_PGM_RSRC2_VS__SO_BASE1_EN_MASK 0x200
+#define SPI_SHADER_PGM_RSRC2_VS__SO_BASE1_EN__SHIFT 0x9
+#define SPI_SHADER_PGM_RSRC2_VS__SO_BASE2_EN_MASK 0x400
+#define SPI_SHADER_PGM_RSRC2_VS__SO_BASE2_EN__SHIFT 0xa
+#define SPI_SHADER_PGM_RSRC2_VS__SO_BASE3_EN_MASK 0x800
+#define SPI_SHADER_PGM_RSRC2_VS__SO_BASE3_EN__SHIFT 0xb
+#define SPI_SHADER_PGM_RSRC2_VS__SO_EN_MASK 0x1000
+#define SPI_SHADER_PGM_RSRC2_VS__SO_EN__SHIFT 0xc
+#define SPI_SHADER_PGM_RSRC2_VS__EXCP_EN_MASK 0x3fe000
+#define SPI_SHADER_PGM_RSRC2_VS__EXCP_EN__SHIFT 0xd
+#define SPI_SHADER_PGM_RSRC2_VS__DISPATCH_DRAW_EN_MASK 0x1000000
+#define SPI_SHADER_PGM_RSRC2_VS__DISPATCH_DRAW_EN__SHIFT 0x18
+#define SPI_SHADER_PGM_RSRC3_VS__CU_EN_MASK 0xffff
+#define SPI_SHADER_PGM_RSRC3_VS__CU_EN__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC3_VS__WAVE_LIMIT_MASK 0x3f0000
+#define SPI_SHADER_PGM_RSRC3_VS__WAVE_LIMIT__SHIFT 0x10
+#define SPI_SHADER_PGM_RSRC3_VS__LOCK_LOW_THRESHOLD_MASK 0x3c00000
+#define SPI_SHADER_PGM_RSRC3_VS__LOCK_LOW_THRESHOLD__SHIFT 0x16
+#define SPI_SHADER_LATE_ALLOC_VS__LIMIT_MASK 0x3f
+#define SPI_SHADER_LATE_ALLOC_VS__LIMIT__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_VS_0__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_VS_0__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_VS_1__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_VS_1__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_VS_2__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_VS_2__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_VS_3__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_VS_3__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_VS_4__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_VS_4__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_VS_5__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_VS_5__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_VS_6__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_VS_6__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_VS_7__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_VS_7__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_VS_8__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_VS_8__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_VS_9__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_VS_9__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_VS_10__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_VS_10__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_VS_11__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_VS_11__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_VS_12__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_VS_12__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_VS_13__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_VS_13__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_VS_14__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_VS_14__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_VS_15__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_VS_15__DATA__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC2_ES_VS__SCRATCH_EN_MASK 0x1
+#define SPI_SHADER_PGM_RSRC2_ES_VS__SCRATCH_EN__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC2_ES_VS__USER_SGPR_MASK 0x3e
+#define SPI_SHADER_PGM_RSRC2_ES_VS__USER_SGPR__SHIFT 0x1
+#define SPI_SHADER_PGM_RSRC2_ES_VS__TRAP_PRESENT_MASK 0x40
+#define SPI_SHADER_PGM_RSRC2_ES_VS__TRAP_PRESENT__SHIFT 0x6
+#define SPI_SHADER_PGM_RSRC2_ES_VS__OC_LDS_EN_MASK 0x80
+#define SPI_SHADER_PGM_RSRC2_ES_VS__OC_LDS_EN__SHIFT 0x7
+#define SPI_SHADER_PGM_RSRC2_ES_VS__EXCP_EN_MASK 0x1ff00
+#define SPI_SHADER_PGM_RSRC2_ES_VS__EXCP_EN__SHIFT 0x8
+#define SPI_SHADER_PGM_RSRC2_ES_VS__LDS_SIZE_MASK 0x1ff00000
+#define SPI_SHADER_PGM_RSRC2_ES_VS__LDS_SIZE__SHIFT 0x14
+#define SPI_SHADER_PGM_RSRC2_LS_VS__SCRATCH_EN_MASK 0x1
+#define SPI_SHADER_PGM_RSRC2_LS_VS__SCRATCH_EN__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC2_LS_VS__USER_SGPR_MASK 0x3e
+#define SPI_SHADER_PGM_RSRC2_LS_VS__USER_SGPR__SHIFT 0x1
+#define SPI_SHADER_PGM_RSRC2_LS_VS__TRAP_PRESENT_MASK 0x40
+#define SPI_SHADER_PGM_RSRC2_LS_VS__TRAP_PRESENT__SHIFT 0x6
+#define SPI_SHADER_PGM_RSRC2_LS_VS__LDS_SIZE_MASK 0xff80
+#define SPI_SHADER_PGM_RSRC2_LS_VS__LDS_SIZE__SHIFT 0x7
+#define SPI_SHADER_PGM_RSRC2_LS_VS__EXCP_EN_MASK 0x1ff0000
+#define SPI_SHADER_PGM_RSRC2_LS_VS__EXCP_EN__SHIFT 0x10
+#define SPI_SHADER_TBA_LO_GS__MEM_BASE_MASK 0xffffffff
+#define SPI_SHADER_TBA_LO_GS__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_TBA_HI_GS__MEM_BASE_MASK 0xff
+#define SPI_SHADER_TBA_HI_GS__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_TMA_LO_GS__MEM_BASE_MASK 0xffffffff
+#define SPI_SHADER_TMA_LO_GS__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_TMA_HI_GS__MEM_BASE_MASK 0xff
+#define SPI_SHADER_TMA_HI_GS__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_PGM_LO_GS__MEM_BASE_MASK 0xffffffff
+#define SPI_SHADER_PGM_LO_GS__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_PGM_HI_GS__MEM_BASE_MASK 0xff
+#define SPI_SHADER_PGM_HI_GS__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC1_GS__VGPRS_MASK 0x3f
+#define SPI_SHADER_PGM_RSRC1_GS__VGPRS__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC1_GS__SGPRS_MASK 0x3c0
+#define SPI_SHADER_PGM_RSRC1_GS__SGPRS__SHIFT 0x6
+#define SPI_SHADER_PGM_RSRC1_GS__PRIORITY_MASK 0xc00
+#define SPI_SHADER_PGM_RSRC1_GS__PRIORITY__SHIFT 0xa
+#define SPI_SHADER_PGM_RSRC1_GS__FLOAT_MODE_MASK 0xff000
+#define SPI_SHADER_PGM_RSRC1_GS__FLOAT_MODE__SHIFT 0xc
+#define SPI_SHADER_PGM_RSRC1_GS__PRIV_MASK 0x100000
+#define SPI_SHADER_PGM_RSRC1_GS__PRIV__SHIFT 0x14
+#define SPI_SHADER_PGM_RSRC1_GS__DX10_CLAMP_MASK 0x200000
+#define SPI_SHADER_PGM_RSRC1_GS__DX10_CLAMP__SHIFT 0x15
+#define SPI_SHADER_PGM_RSRC1_GS__DEBUG_MODE_MASK 0x400000
+#define SPI_SHADER_PGM_RSRC1_GS__DEBUG_MODE__SHIFT 0x16
+#define SPI_SHADER_PGM_RSRC1_GS__IEEE_MODE_MASK 0x800000
+#define SPI_SHADER_PGM_RSRC1_GS__IEEE_MODE__SHIFT 0x17
+#define SPI_SHADER_PGM_RSRC1_GS__CU_GROUP_ENABLE_MASK 0x1000000
+#define SPI_SHADER_PGM_RSRC1_GS__CU_GROUP_ENABLE__SHIFT 0x18
+#define SPI_SHADER_PGM_RSRC1_GS__CACHE_CTL_MASK 0xe000000
+#define SPI_SHADER_PGM_RSRC1_GS__CACHE_CTL__SHIFT 0x19
+#define SPI_SHADER_PGM_RSRC1_GS__CDBG_USER_MASK 0x10000000
+#define SPI_SHADER_PGM_RSRC1_GS__CDBG_USER__SHIFT 0x1c
+#define SPI_SHADER_PGM_RSRC2_GS__SCRATCH_EN_MASK 0x1
+#define SPI_SHADER_PGM_RSRC2_GS__SCRATCH_EN__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC2_GS__USER_SGPR_MASK 0x3e
+#define SPI_SHADER_PGM_RSRC2_GS__USER_SGPR__SHIFT 0x1
+#define SPI_SHADER_PGM_RSRC2_GS__TRAP_PRESENT_MASK 0x40
+#define SPI_SHADER_PGM_RSRC2_GS__TRAP_PRESENT__SHIFT 0x6
+#define SPI_SHADER_PGM_RSRC2_GS__EXCP_EN_MASK 0xff80
+#define SPI_SHADER_PGM_RSRC2_GS__EXCP_EN__SHIFT 0x7
+#define SPI_SHADER_PGM_RSRC3_GS__CU_EN_MASK 0xffff
+#define SPI_SHADER_PGM_RSRC3_GS__CU_EN__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC3_GS__WAVE_LIMIT_MASK 0x3f0000
+#define SPI_SHADER_PGM_RSRC3_GS__WAVE_LIMIT__SHIFT 0x10
+#define SPI_SHADER_PGM_RSRC3_GS__LOCK_LOW_THRESHOLD_MASK 0x3c00000
+#define SPI_SHADER_PGM_RSRC3_GS__LOCK_LOW_THRESHOLD__SHIFT 0x16
+#define SPI_SHADER_PGM_RSRC3_GS__GROUP_FIFO_DEPTH_MASK 0xfc000000
+#define SPI_SHADER_PGM_RSRC3_GS__GROUP_FIFO_DEPTH__SHIFT 0x1a
+#define SPI_SHADER_USER_DATA_GS_0__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_GS_0__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_GS_1__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_GS_1__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_GS_2__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_GS_2__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_GS_3__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_GS_3__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_GS_4__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_GS_4__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_GS_5__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_GS_5__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_GS_6__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_GS_6__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_GS_7__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_GS_7__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_GS_8__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_GS_8__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_GS_9__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_GS_9__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_GS_10__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_GS_10__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_GS_11__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_GS_11__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_GS_12__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_GS_12__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_GS_13__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_GS_13__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_GS_14__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_GS_14__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_GS_15__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_GS_15__DATA__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC2_ES_GS__SCRATCH_EN_MASK 0x1
+#define SPI_SHADER_PGM_RSRC2_ES_GS__SCRATCH_EN__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC2_ES_GS__USER_SGPR_MASK 0x3e
+#define SPI_SHADER_PGM_RSRC2_ES_GS__USER_SGPR__SHIFT 0x1
+#define SPI_SHADER_PGM_RSRC2_ES_GS__TRAP_PRESENT_MASK 0x40
+#define SPI_SHADER_PGM_RSRC2_ES_GS__TRAP_PRESENT__SHIFT 0x6
+#define SPI_SHADER_PGM_RSRC2_ES_GS__OC_LDS_EN_MASK 0x80
+#define SPI_SHADER_PGM_RSRC2_ES_GS__OC_LDS_EN__SHIFT 0x7
+#define SPI_SHADER_PGM_RSRC2_ES_GS__EXCP_EN_MASK 0x1ff00
+#define SPI_SHADER_PGM_RSRC2_ES_GS__EXCP_EN__SHIFT 0x8
+#define SPI_SHADER_PGM_RSRC2_ES_GS__LDS_SIZE_MASK 0x1ff00000
+#define SPI_SHADER_PGM_RSRC2_ES_GS__LDS_SIZE__SHIFT 0x14
+#define SPI_SHADER_TBA_LO_ES__MEM_BASE_MASK 0xffffffff
+#define SPI_SHADER_TBA_LO_ES__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_TBA_HI_ES__MEM_BASE_MASK 0xff
+#define SPI_SHADER_TBA_HI_ES__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_TMA_LO_ES__MEM_BASE_MASK 0xffffffff
+#define SPI_SHADER_TMA_LO_ES__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_TMA_HI_ES__MEM_BASE_MASK 0xff
+#define SPI_SHADER_TMA_HI_ES__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_PGM_LO_ES__MEM_BASE_MASK 0xffffffff
+#define SPI_SHADER_PGM_LO_ES__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_PGM_HI_ES__MEM_BASE_MASK 0xff
+#define SPI_SHADER_PGM_HI_ES__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC1_ES__VGPRS_MASK 0x3f
+#define SPI_SHADER_PGM_RSRC1_ES__VGPRS__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC1_ES__SGPRS_MASK 0x3c0
+#define SPI_SHADER_PGM_RSRC1_ES__SGPRS__SHIFT 0x6
+#define SPI_SHADER_PGM_RSRC1_ES__PRIORITY_MASK 0xc00
+#define SPI_SHADER_PGM_RSRC1_ES__PRIORITY__SHIFT 0xa
+#define SPI_SHADER_PGM_RSRC1_ES__FLOAT_MODE_MASK 0xff000
+#define SPI_SHADER_PGM_RSRC1_ES__FLOAT_MODE__SHIFT 0xc
+#define SPI_SHADER_PGM_RSRC1_ES__PRIV_MASK 0x100000
+#define SPI_SHADER_PGM_RSRC1_ES__PRIV__SHIFT 0x14
+#define SPI_SHADER_PGM_RSRC1_ES__DX10_CLAMP_MASK 0x200000
+#define SPI_SHADER_PGM_RSRC1_ES__DX10_CLAMP__SHIFT 0x15
+#define SPI_SHADER_PGM_RSRC1_ES__DEBUG_MODE_MASK 0x400000
+#define SPI_SHADER_PGM_RSRC1_ES__DEBUG_MODE__SHIFT 0x16
+#define SPI_SHADER_PGM_RSRC1_ES__IEEE_MODE_MASK 0x800000
+#define SPI_SHADER_PGM_RSRC1_ES__IEEE_MODE__SHIFT 0x17
+#define SPI_SHADER_PGM_RSRC1_ES__VGPR_COMP_CNT_MASK 0x3000000
+#define SPI_SHADER_PGM_RSRC1_ES__VGPR_COMP_CNT__SHIFT 0x18
+#define SPI_SHADER_PGM_RSRC1_ES__CU_GROUP_ENABLE_MASK 0x4000000
+#define SPI_SHADER_PGM_RSRC1_ES__CU_GROUP_ENABLE__SHIFT 0x1a
+#define SPI_SHADER_PGM_RSRC1_ES__CACHE_CTL_MASK 0x38000000
+#define SPI_SHADER_PGM_RSRC1_ES__CACHE_CTL__SHIFT 0x1b
+#define SPI_SHADER_PGM_RSRC1_ES__CDBG_USER_MASK 0x40000000
+#define SPI_SHADER_PGM_RSRC1_ES__CDBG_USER__SHIFT 0x1e
+#define SPI_SHADER_PGM_RSRC2_ES__SCRATCH_EN_MASK 0x1
+#define SPI_SHADER_PGM_RSRC2_ES__SCRATCH_EN__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC2_ES__USER_SGPR_MASK 0x3e
+#define SPI_SHADER_PGM_RSRC2_ES__USER_SGPR__SHIFT 0x1
+#define SPI_SHADER_PGM_RSRC2_ES__TRAP_PRESENT_MASK 0x40
+#define SPI_SHADER_PGM_RSRC2_ES__TRAP_PRESENT__SHIFT 0x6
+#define SPI_SHADER_PGM_RSRC2_ES__OC_LDS_EN_MASK 0x80
+#define SPI_SHADER_PGM_RSRC2_ES__OC_LDS_EN__SHIFT 0x7
+#define SPI_SHADER_PGM_RSRC2_ES__EXCP_EN_MASK 0x1ff00
+#define SPI_SHADER_PGM_RSRC2_ES__EXCP_EN__SHIFT 0x8
+#define SPI_SHADER_PGM_RSRC2_ES__LDS_SIZE_MASK 0x1ff00000
+#define SPI_SHADER_PGM_RSRC2_ES__LDS_SIZE__SHIFT 0x14
+#define SPI_SHADER_PGM_RSRC3_ES__CU_EN_MASK 0xffff
+#define SPI_SHADER_PGM_RSRC3_ES__CU_EN__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC3_ES__WAVE_LIMIT_MASK 0x3f0000
+#define SPI_SHADER_PGM_RSRC3_ES__WAVE_LIMIT__SHIFT 0x10
+#define SPI_SHADER_PGM_RSRC3_ES__LOCK_LOW_THRESHOLD_MASK 0x3c00000
+#define SPI_SHADER_PGM_RSRC3_ES__LOCK_LOW_THRESHOLD__SHIFT 0x16
+#define SPI_SHADER_PGM_RSRC3_ES__GROUP_FIFO_DEPTH_MASK 0xfc000000
+#define SPI_SHADER_PGM_RSRC3_ES__GROUP_FIFO_DEPTH__SHIFT 0x1a
+#define SPI_SHADER_USER_DATA_ES_0__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_ES_0__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_ES_1__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_ES_1__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_ES_2__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_ES_2__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_ES_3__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_ES_3__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_ES_4__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_ES_4__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_ES_5__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_ES_5__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_ES_6__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_ES_6__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_ES_7__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_ES_7__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_ES_8__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_ES_8__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_ES_9__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_ES_9__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_ES_10__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_ES_10__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_ES_11__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_ES_11__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_ES_12__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_ES_12__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_ES_13__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_ES_13__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_ES_14__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_ES_14__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_ES_15__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_ES_15__DATA__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC2_LS_ES__SCRATCH_EN_MASK 0x1
+#define SPI_SHADER_PGM_RSRC2_LS_ES__SCRATCH_EN__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC2_LS_ES__USER_SGPR_MASK 0x3e
+#define SPI_SHADER_PGM_RSRC2_LS_ES__USER_SGPR__SHIFT 0x1
+#define SPI_SHADER_PGM_RSRC2_LS_ES__TRAP_PRESENT_MASK 0x40
+#define SPI_SHADER_PGM_RSRC2_LS_ES__TRAP_PRESENT__SHIFT 0x6
+#define SPI_SHADER_PGM_RSRC2_LS_ES__LDS_SIZE_MASK 0xff80
+#define SPI_SHADER_PGM_RSRC2_LS_ES__LDS_SIZE__SHIFT 0x7
+#define SPI_SHADER_PGM_RSRC2_LS_ES__EXCP_EN_MASK 0x1ff0000
+#define SPI_SHADER_PGM_RSRC2_LS_ES__EXCP_EN__SHIFT 0x10
+#define SPI_SHADER_TBA_LO_HS__MEM_BASE_MASK 0xffffffff
+#define SPI_SHADER_TBA_LO_HS__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_TBA_HI_HS__MEM_BASE_MASK 0xff
+#define SPI_SHADER_TBA_HI_HS__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_TMA_LO_HS__MEM_BASE_MASK 0xffffffff
+#define SPI_SHADER_TMA_LO_HS__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_TMA_HI_HS__MEM_BASE_MASK 0xff
+#define SPI_SHADER_TMA_HI_HS__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_PGM_LO_HS__MEM_BASE_MASK 0xffffffff
+#define SPI_SHADER_PGM_LO_HS__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_PGM_HI_HS__MEM_BASE_MASK 0xff
+#define SPI_SHADER_PGM_HI_HS__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC1_HS__VGPRS_MASK 0x3f
+#define SPI_SHADER_PGM_RSRC1_HS__VGPRS__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC1_HS__SGPRS_MASK 0x3c0
+#define SPI_SHADER_PGM_RSRC1_HS__SGPRS__SHIFT 0x6
+#define SPI_SHADER_PGM_RSRC1_HS__PRIORITY_MASK 0xc00
+#define SPI_SHADER_PGM_RSRC1_HS__PRIORITY__SHIFT 0xa
+#define SPI_SHADER_PGM_RSRC1_HS__FLOAT_MODE_MASK 0xff000
+#define SPI_SHADER_PGM_RSRC1_HS__FLOAT_MODE__SHIFT 0xc
+#define SPI_SHADER_PGM_RSRC1_HS__PRIV_MASK 0x100000
+#define SPI_SHADER_PGM_RSRC1_HS__PRIV__SHIFT 0x14
+#define SPI_SHADER_PGM_RSRC1_HS__DX10_CLAMP_MASK 0x200000
+#define SPI_SHADER_PGM_RSRC1_HS__DX10_CLAMP__SHIFT 0x15
+#define SPI_SHADER_PGM_RSRC1_HS__DEBUG_MODE_MASK 0x400000
+#define SPI_SHADER_PGM_RSRC1_HS__DEBUG_MODE__SHIFT 0x16
+#define SPI_SHADER_PGM_RSRC1_HS__IEEE_MODE_MASK 0x800000
+#define SPI_SHADER_PGM_RSRC1_HS__IEEE_MODE__SHIFT 0x17
+#define SPI_SHADER_PGM_RSRC1_HS__CACHE_CTL_MASK 0x7000000
+#define SPI_SHADER_PGM_RSRC1_HS__CACHE_CTL__SHIFT 0x18
+#define SPI_SHADER_PGM_RSRC1_HS__CDBG_USER_MASK 0x8000000
+#define SPI_SHADER_PGM_RSRC1_HS__CDBG_USER__SHIFT 0x1b
+#define SPI_SHADER_PGM_RSRC2_HS__SCRATCH_EN_MASK 0x1
+#define SPI_SHADER_PGM_RSRC2_HS__SCRATCH_EN__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC2_HS__USER_SGPR_MASK 0x3e
+#define SPI_SHADER_PGM_RSRC2_HS__USER_SGPR__SHIFT 0x1
+#define SPI_SHADER_PGM_RSRC2_HS__TRAP_PRESENT_MASK 0x40
+#define SPI_SHADER_PGM_RSRC2_HS__TRAP_PRESENT__SHIFT 0x6
+#define SPI_SHADER_PGM_RSRC2_HS__OC_LDS_EN_MASK 0x80
+#define SPI_SHADER_PGM_RSRC2_HS__OC_LDS_EN__SHIFT 0x7
+#define SPI_SHADER_PGM_RSRC2_HS__TG_SIZE_EN_MASK 0x100
+#define SPI_SHADER_PGM_RSRC2_HS__TG_SIZE_EN__SHIFT 0x8
+#define SPI_SHADER_PGM_RSRC2_HS__EXCP_EN_MASK 0x3fe00
+#define SPI_SHADER_PGM_RSRC2_HS__EXCP_EN__SHIFT 0x9
+#define SPI_SHADER_PGM_RSRC3_HS__WAVE_LIMIT_MASK 0x3f
+#define SPI_SHADER_PGM_RSRC3_HS__WAVE_LIMIT__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC3_HS__LOCK_LOW_THRESHOLD_MASK 0x3c0
+#define SPI_SHADER_PGM_RSRC3_HS__LOCK_LOW_THRESHOLD__SHIFT 0x6
+#define SPI_SHADER_PGM_RSRC3_HS__GROUP_FIFO_DEPTH_MASK 0xfc00
+#define SPI_SHADER_PGM_RSRC3_HS__GROUP_FIFO_DEPTH__SHIFT 0xa
+#define SPI_SHADER_USER_DATA_HS_0__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_HS_0__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_HS_1__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_HS_1__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_HS_2__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_HS_2__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_HS_3__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_HS_3__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_HS_4__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_HS_4__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_HS_5__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_HS_5__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_HS_6__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_HS_6__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_HS_7__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_HS_7__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_HS_8__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_HS_8__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_HS_9__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_HS_9__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_HS_10__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_HS_10__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_HS_11__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_HS_11__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_HS_12__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_HS_12__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_HS_13__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_HS_13__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_HS_14__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_HS_14__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_HS_15__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_HS_15__DATA__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC2_LS_HS__SCRATCH_EN_MASK 0x1
+#define SPI_SHADER_PGM_RSRC2_LS_HS__SCRATCH_EN__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC2_LS_HS__USER_SGPR_MASK 0x3e
+#define SPI_SHADER_PGM_RSRC2_LS_HS__USER_SGPR__SHIFT 0x1
+#define SPI_SHADER_PGM_RSRC2_LS_HS__TRAP_PRESENT_MASK 0x40
+#define SPI_SHADER_PGM_RSRC2_LS_HS__TRAP_PRESENT__SHIFT 0x6
+#define SPI_SHADER_PGM_RSRC2_LS_HS__LDS_SIZE_MASK 0xff80
+#define SPI_SHADER_PGM_RSRC2_LS_HS__LDS_SIZE__SHIFT 0x7
+#define SPI_SHADER_PGM_RSRC2_LS_HS__EXCP_EN_MASK 0x1ff0000
+#define SPI_SHADER_PGM_RSRC2_LS_HS__EXCP_EN__SHIFT 0x10
+#define SPI_SHADER_TBA_LO_LS__MEM_BASE_MASK 0xffffffff
+#define SPI_SHADER_TBA_LO_LS__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_TBA_HI_LS__MEM_BASE_MASK 0xff
+#define SPI_SHADER_TBA_HI_LS__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_TMA_LO_LS__MEM_BASE_MASK 0xffffffff
+#define SPI_SHADER_TMA_LO_LS__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_TMA_HI_LS__MEM_BASE_MASK 0xff
+#define SPI_SHADER_TMA_HI_LS__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_PGM_LO_LS__MEM_BASE_MASK 0xffffffff
+#define SPI_SHADER_PGM_LO_LS__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_PGM_HI_LS__MEM_BASE_MASK 0xff
+#define SPI_SHADER_PGM_HI_LS__MEM_BASE__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC1_LS__VGPRS_MASK 0x3f
+#define SPI_SHADER_PGM_RSRC1_LS__VGPRS__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC1_LS__SGPRS_MASK 0x3c0
+#define SPI_SHADER_PGM_RSRC1_LS__SGPRS__SHIFT 0x6
+#define SPI_SHADER_PGM_RSRC1_LS__PRIORITY_MASK 0xc00
+#define SPI_SHADER_PGM_RSRC1_LS__PRIORITY__SHIFT 0xa
+#define SPI_SHADER_PGM_RSRC1_LS__FLOAT_MODE_MASK 0xff000
+#define SPI_SHADER_PGM_RSRC1_LS__FLOAT_MODE__SHIFT 0xc
+#define SPI_SHADER_PGM_RSRC1_LS__PRIV_MASK 0x100000
+#define SPI_SHADER_PGM_RSRC1_LS__PRIV__SHIFT 0x14
+#define SPI_SHADER_PGM_RSRC1_LS__DX10_CLAMP_MASK 0x200000
+#define SPI_SHADER_PGM_RSRC1_LS__DX10_CLAMP__SHIFT 0x15
+#define SPI_SHADER_PGM_RSRC1_LS__DEBUG_MODE_MASK 0x400000
+#define SPI_SHADER_PGM_RSRC1_LS__DEBUG_MODE__SHIFT 0x16
+#define SPI_SHADER_PGM_RSRC1_LS__IEEE_MODE_MASK 0x800000
+#define SPI_SHADER_PGM_RSRC1_LS__IEEE_MODE__SHIFT 0x17
+#define SPI_SHADER_PGM_RSRC1_LS__VGPR_COMP_CNT_MASK 0x3000000
+#define SPI_SHADER_PGM_RSRC1_LS__VGPR_COMP_CNT__SHIFT 0x18
+#define SPI_SHADER_PGM_RSRC1_LS__CACHE_CTL_MASK 0x1c000000
+#define SPI_SHADER_PGM_RSRC1_LS__CACHE_CTL__SHIFT 0x1a
+#define SPI_SHADER_PGM_RSRC1_LS__CDBG_USER_MASK 0x20000000
+#define SPI_SHADER_PGM_RSRC1_LS__CDBG_USER__SHIFT 0x1d
+#define SPI_SHADER_PGM_RSRC2_LS__SCRATCH_EN_MASK 0x1
+#define SPI_SHADER_PGM_RSRC2_LS__SCRATCH_EN__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC2_LS__USER_SGPR_MASK 0x3e
+#define SPI_SHADER_PGM_RSRC2_LS__USER_SGPR__SHIFT 0x1
+#define SPI_SHADER_PGM_RSRC2_LS__TRAP_PRESENT_MASK 0x40
+#define SPI_SHADER_PGM_RSRC2_LS__TRAP_PRESENT__SHIFT 0x6
+#define SPI_SHADER_PGM_RSRC2_LS__LDS_SIZE_MASK 0xff80
+#define SPI_SHADER_PGM_RSRC2_LS__LDS_SIZE__SHIFT 0x7
+#define SPI_SHADER_PGM_RSRC2_LS__EXCP_EN_MASK 0x1ff0000
+#define SPI_SHADER_PGM_RSRC2_LS__EXCP_EN__SHIFT 0x10
+#define SPI_SHADER_PGM_RSRC3_LS__CU_EN_MASK 0xffff
+#define SPI_SHADER_PGM_RSRC3_LS__CU_EN__SHIFT 0x0
+#define SPI_SHADER_PGM_RSRC3_LS__WAVE_LIMIT_MASK 0x3f0000
+#define SPI_SHADER_PGM_RSRC3_LS__WAVE_LIMIT__SHIFT 0x10
+#define SPI_SHADER_PGM_RSRC3_LS__LOCK_LOW_THRESHOLD_MASK 0x3c00000
+#define SPI_SHADER_PGM_RSRC3_LS__LOCK_LOW_THRESHOLD__SHIFT 0x16
+#define SPI_SHADER_PGM_RSRC3_LS__GROUP_FIFO_DEPTH_MASK 0xfc000000
+#define SPI_SHADER_PGM_RSRC3_LS__GROUP_FIFO_DEPTH__SHIFT 0x1a
+#define SPI_SHADER_USER_DATA_LS_0__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_LS_0__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_LS_1__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_LS_1__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_LS_2__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_LS_2__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_LS_3__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_LS_3__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_LS_4__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_LS_4__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_LS_5__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_LS_5__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_LS_6__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_LS_6__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_LS_7__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_LS_7__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_LS_8__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_LS_8__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_LS_9__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_LS_9__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_LS_10__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_LS_10__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_LS_11__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_LS_11__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_LS_12__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_LS_12__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_LS_13__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_LS_13__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_LS_14__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_LS_14__DATA__SHIFT 0x0
+#define SPI_SHADER_USER_DATA_LS_15__DATA_MASK 0xffffffff
+#define SPI_SHADER_USER_DATA_LS_15__DATA__SHIFT 0x0
+#define SQ_CONFIG__UNUSED_MASK 0xff
+#define SQ_CONFIG__UNUSED__SHIFT 0x0
+#define SQ_CONFIG__DEBUG_EN_MASK 0x100
+#define SQ_CONFIG__DEBUG_EN__SHIFT 0x8
+#define SQ_CONFIG__DEBUG_SINGLE_MEMOP_MASK 0x200
+#define SQ_CONFIG__DEBUG_SINGLE_MEMOP__SHIFT 0x9
+#define SQ_CONFIG__DEBUG_ONE_INST_CLAUSE_MASK 0x400
+#define SQ_CONFIG__DEBUG_ONE_INST_CLAUSE__SHIFT 0xa
+#define SQ_CONFIG__EARLY_TA_DONE_DISABLE_MASK 0x1000
+#define SQ_CONFIG__EARLY_TA_DONE_DISABLE__SHIFT 0xc
+#define SQ_CONFIG__DUA_FLAT_LOCK_ENABLE_MASK 0x2000
+#define SQ_CONFIG__DUA_FLAT_LOCK_ENABLE__SHIFT 0xd
+#define SQ_CONFIG__DUA_LDS_BYPASS_DISABLE_MASK 0x4000
+#define SQ_CONFIG__DUA_LDS_BYPASS_DISABLE__SHIFT 0xe
+#define SQ_CONFIG__DUA_FLAT_LDS_PINGPONG_DISABLE_MASK 0x8000
+#define SQ_CONFIG__DUA_FLAT_LDS_PINGPONG_DISABLE__SHIFT 0xf
+#define SQ_CONFIG__DISABLE_VMEM_SOFT_CLAUSE_MASK 0x10000
+#define SQ_CONFIG__DISABLE_VMEM_SOFT_CLAUSE__SHIFT 0x10
+#define SQ_CONFIG__DISABLE_SMEM_SOFT_CLAUSE_MASK 0x20000
+#define SQ_CONFIG__DISABLE_SMEM_SOFT_CLAUSE__SHIFT 0x11
+#define SQ_CONFIG__ENABLE_HIPRIO_ON_EXP_RDY_VS_MASK 0x40000
+#define SQ_CONFIG__ENABLE_HIPRIO_ON_EXP_RDY_VS__SHIFT 0x12
+#define SQ_CONFIG__PRIO_VAL_ON_EXP_RDY_VS_MASK 0x180000
+#define SQ_CONFIG__PRIO_VAL_ON_EXP_RDY_VS__SHIFT 0x13
+#define SQ_CONFIG__REPLAY_SLEEP_CNT_MASK 0x1e00000
+#define SQ_CONFIG__REPLAY_SLEEP_CNT__SHIFT 0x15
+#define SQC_CONFIG__INST_CACHE_SIZE_MASK 0x3
+#define SQC_CONFIG__INST_CACHE_SIZE__SHIFT 0x0
+#define SQC_CONFIG__DATA_CACHE_SIZE_MASK 0xc
+#define SQC_CONFIG__DATA_CACHE_SIZE__SHIFT 0x2
+#define SQC_CONFIG__MISS_FIFO_DEPTH_MASK 0x30
+#define SQC_CONFIG__MISS_FIFO_DEPTH__SHIFT 0x4
+#define SQC_CONFIG__HIT_FIFO_DEPTH_MASK 0x40
+#define SQC_CONFIG__HIT_FIFO_DEPTH__SHIFT 0x6
+#define SQC_CONFIG__FORCE_ALWAYS_MISS_MASK 0x80
+#define SQC_CONFIG__FORCE_ALWAYS_MISS__SHIFT 0x7
+#define SQC_CONFIG__FORCE_IN_ORDER_MASK 0x100
+#define SQC_CONFIG__FORCE_IN_ORDER__SHIFT 0x8
+#define SQC_CONFIG__IDENTITY_HASH_BANK_MASK 0x200
+#define SQC_CONFIG__IDENTITY_HASH_BANK__SHIFT 0x9
+#define SQC_CONFIG__IDENTITY_HASH_SET_MASK 0x400
+#define SQC_CONFIG__IDENTITY_HASH_SET__SHIFT 0xa
+#define SQC_CONFIG__PER_VMID_INV_DISABLE_MASK 0x800
+#define SQC_CONFIG__PER_VMID_INV_DISABLE__SHIFT 0xb
+#define SQC_CONFIG__EVICT_LRU_MASK 0x3000
+#define SQC_CONFIG__EVICT_LRU__SHIFT 0xc
+#define SQC_CONFIG__FORCE_2_BANK_MASK 0x4000
+#define SQC_CONFIG__FORCE_2_BANK__SHIFT 0xe
+#define SQC_CONFIG__FORCE_1_BANK_MASK 0x8000
+#define SQC_CONFIG__FORCE_1_BANK__SHIFT 0xf
+#define SQC_CONFIG__LS_DISABLE_CLOCKS_MASK 0xff0000
+#define SQC_CONFIG__LS_DISABLE_CLOCKS__SHIFT 0x10
+#define SQC_CACHES__TARGET_INST_MASK 0x1
+#define SQC_CACHES__TARGET_INST__SHIFT 0x0
+#define SQC_CACHES__TARGET_DATA_MASK 0x2
+#define SQC_CACHES__TARGET_DATA__SHIFT 0x1
+#define SQC_CACHES__INVALIDATE_MASK 0x4
+#define SQC_CACHES__INVALIDATE__SHIFT 0x2
+#define SQC_CACHES__WRITEBACK_MASK 0x8
+#define SQC_CACHES__WRITEBACK__SHIFT 0x3
+#define SQC_CACHES__VOL_MASK 0x10
+#define SQC_CACHES__VOL__SHIFT 0x4
+#define SQC_CACHES__COMPLETE_MASK 0x10000
+#define SQC_CACHES__COMPLETE__SHIFT 0x10
+#define SQC_WRITEBACK__DWB_MASK 0x1
+#define SQC_WRITEBACK__DWB__SHIFT 0x0
+#define SQC_WRITEBACK__DIRTY_MASK 0x2
+#define SQC_WRITEBACK__DIRTY__SHIFT 0x1
+#define SQC_DSM_CNTL__SEL_DATA_ICACHE_BANKA_MASK 0x3
+#define SQC_DSM_CNTL__SEL_DATA_ICACHE_BANKA__SHIFT 0x0
+#define SQC_DSM_CNTL__EN_SINGLE_WR_ICACHE_BANKA_MASK 0x4
+#define SQC_DSM_CNTL__EN_SINGLE_WR_ICACHE_BANKA__SHIFT 0x2
+#define SQC_DSM_CNTL__SEL_DATA_ICACHE_BANKB_MASK 0x18
+#define SQC_DSM_CNTL__SEL_DATA_ICACHE_BANKB__SHIFT 0x3
+#define SQC_DSM_CNTL__EN_SINGLE_WR_ICACHE_BANKB_MASK 0x20
+#define SQC_DSM_CNTL__EN_SINGLE_WR_ICACHE_BANKB__SHIFT 0x5
+#define SQC_DSM_CNTL__SEL_DATA_ICACHE_BANKC_MASK 0xc0
+#define SQC_DSM_CNTL__SEL_DATA_ICACHE_BANKC__SHIFT 0x6
+#define SQC_DSM_CNTL__EN_SINGLE_WR_ICACHE_BANKC_MASK 0x100
+#define SQC_DSM_CNTL__EN_SINGLE_WR_ICACHE_BANKC__SHIFT 0x8
+#define SQC_DSM_CNTL__SEL_DATA_ICACHE_BANKD_MASK 0x600
+#define SQC_DSM_CNTL__SEL_DATA_ICACHE_BANKD__SHIFT 0x9
+#define SQC_DSM_CNTL__EN_SINGLE_WR_ICACHE_BANKD_MASK 0x800
+#define SQC_DSM_CNTL__EN_SINGLE_WR_ICACHE_BANKD__SHIFT 0xb
+#define SQC_DSM_CNTL__SEL_DATA_ICACHE_GATCL1_MASK 0x3000
+#define SQC_DSM_CNTL__SEL_DATA_ICACHE_GATCL1__SHIFT 0xc
+#define SQC_DSM_CNTL__EN_SINGLE_WR_ICACHE_GATCL1_MASK 0x4000
+#define SQC_DSM_CNTL__EN_SINGLE_WR_ICACHE_GATCL1__SHIFT 0xe
+#define SQC_DSM_CNTL__SEL_DATA_DCACHE_BANKA_MASK 0x18000
+#define SQC_DSM_CNTL__SEL_DATA_DCACHE_BANKA__SHIFT 0xf
+#define SQC_DSM_CNTL__EN_SINGLE_WR_DCACHE_BANKA_MASK 0x20000
+#define SQC_DSM_CNTL__EN_SINGLE_WR_DCACHE_BANKA__SHIFT 0x11
+#define SQC_DSM_CNTL__SEL_DATA_DCACHE_BANKB_MASK 0xc0000
+#define SQC_DSM_CNTL__SEL_DATA_DCACHE_BANKB__SHIFT 0x12
+#define SQC_DSM_CNTL__EN_SINGLE_WR_DCACHE_BANKB_MASK 0x100000
+#define SQC_DSM_CNTL__EN_SINGLE_WR_DCACHE_BANKB__SHIFT 0x14
+#define SQC_DSM_CNTL__SEL_DATA_DCACHE_BANKC_MASK 0x600000
+#define SQC_DSM_CNTL__SEL_DATA_DCACHE_BANKC__SHIFT 0x15
+#define SQC_DSM_CNTL__EN_SINGLE_WR_DCACHE_BANKC_MASK 0x800000
+#define SQC_DSM_CNTL__EN_SINGLE_WR_DCACHE_BANKC__SHIFT 0x17
+#define SQC_DSM_CNTL__SEL_DATA_DCACHE_BANKD_MASK 0x3000000
+#define SQC_DSM_CNTL__SEL_DATA_DCACHE_BANKD__SHIFT 0x18
+#define SQC_DSM_CNTL__EN_SINGLE_WR_DCACHE_BANKD_MASK 0x4000000
+#define SQC_DSM_CNTL__EN_SINGLE_WR_DCACHE_BANKD__SHIFT 0x1a
+#define SQC_DSM_CNTL__SEL_DATA_DCACHE_GATCL1_MASK 0x18000000
+#define SQC_DSM_CNTL__SEL_DATA_DCACHE_GATCL1__SHIFT 0x1b
+#define SQC_DSM_CNTL__EN_SINGLE_WR_DCACHE_GATCL1_MASK 0x20000000
+#define SQC_DSM_CNTL__EN_SINGLE_WR_DCACHE_GATCL1__SHIFT 0x1d
+#define SQ_RANDOM_WAVE_PRI__RET_MASK 0x7f
+#define SQ_RANDOM_WAVE_PRI__RET__SHIFT 0x0
+#define SQ_RANDOM_WAVE_PRI__RUI_MASK 0x380
+#define SQ_RANDOM_WAVE_PRI__RUI__SHIFT 0x7
+#define SQ_RANDOM_WAVE_PRI__RNG_MASK 0x1ffc00
+#define SQ_RANDOM_WAVE_PRI__RNG__SHIFT 0xa
+#define SQ_REG_CREDITS__SRBM_CREDITS_MASK 0x3f
+#define SQ_REG_CREDITS__SRBM_CREDITS__SHIFT 0x0
+#define SQ_REG_CREDITS__CMD_CREDITS_MASK 0xf00
+#define SQ_REG_CREDITS__CMD_CREDITS__SHIFT 0x8
+#define SQ_REG_CREDITS__REG_BUSY_MASK 0x10000000
+#define SQ_REG_CREDITS__REG_BUSY__SHIFT 0x1c
+#define SQ_REG_CREDITS__SRBM_OVERFLOW_MASK 0x20000000
+#define SQ_REG_CREDITS__SRBM_OVERFLOW__SHIFT 0x1d
+#define SQ_REG_CREDITS__IMMED_OVERFLOW_MASK 0x40000000
+#define SQ_REG_CREDITS__IMMED_OVERFLOW__SHIFT 0x1e
+#define SQ_REG_CREDITS__CMD_OVERFLOW_MASK 0x80000000
+#define SQ_REG_CREDITS__CMD_OVERFLOW__SHIFT 0x1f
+#define SQ_FIFO_SIZES__INTERRUPT_FIFO_SIZE_MASK 0xf
+#define SQ_FIFO_SIZES__INTERRUPT_FIFO_SIZE__SHIFT 0x0
+#define SQ_FIFO_SIZES__TTRACE_FIFO_SIZE_MASK 0xf00
+#define SQ_FIFO_SIZES__TTRACE_FIFO_SIZE__SHIFT 0x8
+#define SQ_FIFO_SIZES__EXPORT_BUF_SIZE_MASK 0x30000
+#define SQ_FIFO_SIZES__EXPORT_BUF_SIZE__SHIFT 0x10
+#define SQ_FIFO_SIZES__VMEM_DATA_FIFO_SIZE_MASK 0xc0000
+#define SQ_FIFO_SIZES__VMEM_DATA_FIFO_SIZE__SHIFT 0x12
+#define SQ_DSM_CNTL__WAVEFRONT_STALL_0_MASK 0x1
+#define SQ_DSM_CNTL__WAVEFRONT_STALL_0__SHIFT 0x0
+#define SQ_DSM_CNTL__WAVEFRONT_STALL_1_MASK 0x2
+#define SQ_DSM_CNTL__WAVEFRONT_STALL_1__SHIFT 0x1
+#define SQ_DSM_CNTL__SPI_BACKPRESSURE_0_MASK 0x4
+#define SQ_DSM_CNTL__SPI_BACKPRESSURE_0__SHIFT 0x2
+#define SQ_DSM_CNTL__SPI_BACKPRESSURE_1_MASK 0x8
+#define SQ_DSM_CNTL__SPI_BACKPRESSURE_1__SHIFT 0x3
+#define SQ_DSM_CNTL__SEL_DSM_SGPR_IRRITATOR_DATA0_MASK 0x100
+#define SQ_DSM_CNTL__SEL_DSM_SGPR_IRRITATOR_DATA0__SHIFT 0x8
+#define SQ_DSM_CNTL__SEL_DSM_SGPR_IRRITATOR_DATA1_MASK 0x200
+#define SQ_DSM_CNTL__SEL_DSM_SGPR_IRRITATOR_DATA1__SHIFT 0x9
+#define SQ_DSM_CNTL__SGPR_ENABLE_SINGLE_WRITE_MASK 0x400
+#define SQ_DSM_CNTL__SGPR_ENABLE_SINGLE_WRITE__SHIFT 0xa
+#define SQ_DSM_CNTL__SEL_DSM_LDS_IRRITATOR_DATA0_MASK 0x10000
+#define SQ_DSM_CNTL__SEL_DSM_LDS_IRRITATOR_DATA0__SHIFT 0x10
+#define SQ_DSM_CNTL__SEL_DSM_LDS_IRRITATOR_DATA1_MASK 0x20000
+#define SQ_DSM_CNTL__SEL_DSM_LDS_IRRITATOR_DATA1__SHIFT 0x11
+#define SQ_DSM_CNTL__LDS_ENABLE_SINGLE_WRITE01_MASK 0x40000
+#define SQ_DSM_CNTL__LDS_ENABLE_SINGLE_WRITE01__SHIFT 0x12
+#define SQ_DSM_CNTL__SEL_DSM_LDS_IRRITATOR_DATA2_MASK 0x80000
+#define SQ_DSM_CNTL__SEL_DSM_LDS_IRRITATOR_DATA2__SHIFT 0x13
+#define SQ_DSM_CNTL__SEL_DSM_LDS_IRRITATOR_DATA3_MASK 0x100000
+#define SQ_DSM_CNTL__SEL_DSM_LDS_IRRITATOR_DATA3__SHIFT 0x14
+#define SQ_DSM_CNTL__LDS_ENABLE_SINGLE_WRITE23_MASK 0x200000
+#define SQ_DSM_CNTL__LDS_ENABLE_SINGLE_WRITE23__SHIFT 0x15
+#define SQ_DSM_CNTL__SEL_DSM_SP_IRRITATOR_DATA0_MASK 0x1000000
+#define SQ_DSM_CNTL__SEL_DSM_SP_IRRITATOR_DATA0__SHIFT 0x18
+#define SQ_DSM_CNTL__SEL_DSM_SP_IRRITATOR_DATA1_MASK 0x2000000
+#define SQ_DSM_CNTL__SEL_DSM_SP_IRRITATOR_DATA1__SHIFT 0x19
+#define SQ_DSM_CNTL__SP_ENABLE_SINGLE_WRITE_MASK 0x4000000
+#define SQ_DSM_CNTL__SP_ENABLE_SINGLE_WRITE__SHIFT 0x1a
+#define CC_GC_SHADER_RATE_CONFIG__DPFP_RATE_MASK 0x6
+#define CC_GC_SHADER_RATE_CONFIG__DPFP_RATE__SHIFT 0x1
+#define CC_GC_SHADER_RATE_CONFIG__SQC_BALANCE_DISABLE_MASK 0x8
+#define CC_GC_SHADER_RATE_CONFIG__SQC_BALANCE_DISABLE__SHIFT 0x3
+#define CC_GC_SHADER_RATE_CONFIG__HALF_LDS_MASK 0x10
+#define CC_GC_SHADER_RATE_CONFIG__HALF_LDS__SHIFT 0x4
+#define GC_USER_SHADER_RATE_CONFIG__DPFP_RATE_MASK 0x6
+#define GC_USER_SHADER_RATE_CONFIG__DPFP_RATE__SHIFT 0x1
+#define GC_USER_SHADER_RATE_CONFIG__SQC_BALANCE_DISABLE_MASK 0x8
+#define GC_USER_SHADER_RATE_CONFIG__SQC_BALANCE_DISABLE__SHIFT 0x3
+#define GC_USER_SHADER_RATE_CONFIG__HALF_LDS_MASK 0x10
+#define GC_USER_SHADER_RATE_CONFIG__HALF_LDS__SHIFT 0x4
+#define SQ_INTERRUPT_AUTO_MASK__MASK_MASK 0xffffff
+#define SQ_INTERRUPT_AUTO_MASK__MASK__SHIFT 0x0
+#define SQ_INTERRUPT_MSG_CTRL__STALL_MASK 0x1
+#define SQ_INTERRUPT_MSG_CTRL__STALL__SHIFT 0x0
+#define SQ_PERFCOUNTER_CTRL__PS_EN_MASK 0x1
+#define SQ_PERFCOUNTER_CTRL__PS_EN__SHIFT 0x0
+#define SQ_PERFCOUNTER_CTRL__VS_EN_MASK 0x2
+#define SQ_PERFCOUNTER_CTRL__VS_EN__SHIFT 0x1
+#define SQ_PERFCOUNTER_CTRL__GS_EN_MASK 0x4
+#define SQ_PERFCOUNTER_CTRL__GS_EN__SHIFT 0x2
+#define SQ_PERFCOUNTER_CTRL__ES_EN_MASK 0x8
+#define SQ_PERFCOUNTER_CTRL__ES_EN__SHIFT 0x3
+#define SQ_PERFCOUNTER_CTRL__HS_EN_MASK 0x10
+#define SQ_PERFCOUNTER_CTRL__HS_EN__SHIFT 0x4
+#define SQ_PERFCOUNTER_CTRL__LS_EN_MASK 0x20
+#define SQ_PERFCOUNTER_CTRL__LS_EN__SHIFT 0x5
+#define SQ_PERFCOUNTER_CTRL__CS_EN_MASK 0x40
+#define SQ_PERFCOUNTER_CTRL__CS_EN__SHIFT 0x6
+#define SQ_PERFCOUNTER_CTRL__CNTR_RATE_MASK 0x1f00
+#define SQ_PERFCOUNTER_CTRL__CNTR_RATE__SHIFT 0x8
+#define SQ_PERFCOUNTER_CTRL__DISABLE_FLUSH_MASK 0x2000
+#define SQ_PERFCOUNTER_CTRL__DISABLE_FLUSH__SHIFT 0xd
+#define SQ_PERFCOUNTER_MASK__SH0_MASK_MASK 0xffff
+#define SQ_PERFCOUNTER_MASK__SH0_MASK__SHIFT 0x0
+#define SQ_PERFCOUNTER_MASK__SH1_MASK_MASK 0xffff0000
+#define SQ_PERFCOUNTER_MASK__SH1_MASK__SHIFT 0x10
+#define SQ_PERFCOUNTER_CTRL2__FORCE_EN_MASK 0x1
+#define SQ_PERFCOUNTER_CTRL2__FORCE_EN__SHIFT 0x0
+#define CC_SQC_BANK_DISABLE__SQC0_BANK_DISABLE_MASK 0xf0000
+#define CC_SQC_BANK_DISABLE__SQC0_BANK_DISABLE__SHIFT 0x10
+#define CC_SQC_BANK_DISABLE__SQC1_BANK_DISABLE_MASK 0xf00000
+#define CC_SQC_BANK_DISABLE__SQC1_BANK_DISABLE__SHIFT 0x14
+#define CC_SQC_BANK_DISABLE__SQC2_BANK_DISABLE_MASK 0xf000000
+#define CC_SQC_BANK_DISABLE__SQC2_BANK_DISABLE__SHIFT 0x18
+#define CC_SQC_BANK_DISABLE__SQC3_BANK_DISABLE_MASK 0xf0000000
+#define CC_SQC_BANK_DISABLE__SQC3_BANK_DISABLE__SHIFT 0x1c
+#define USER_SQC_BANK_DISABLE__SQC0_BANK_DISABLE_MASK 0xf0000
+#define USER_SQC_BANK_DISABLE__SQC0_BANK_DISABLE__SHIFT 0x10
+#define USER_SQC_BANK_DISABLE__SQC1_BANK_DISABLE_MASK 0xf00000
+#define USER_SQC_BANK_DISABLE__SQC1_BANK_DISABLE__SHIFT 0x14
+#define USER_SQC_BANK_DISABLE__SQC2_BANK_DISABLE_MASK 0xf000000
+#define USER_SQC_BANK_DISABLE__SQC2_BANK_DISABLE__SHIFT 0x18
+#define USER_SQC_BANK_DISABLE__SQC3_BANK_DISABLE_MASK 0xf0000000
+#define USER_SQC_BANK_DISABLE__SQC3_BANK_DISABLE__SHIFT 0x1c
+#define SQ_PERFCOUNTER0_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define SQ_PERFCOUNTER0_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define SQ_PERFCOUNTER1_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define SQ_PERFCOUNTER1_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define SQ_PERFCOUNTER2_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define SQ_PERFCOUNTER2_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define SQ_PERFCOUNTER3_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define SQ_PERFCOUNTER3_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define SQ_PERFCOUNTER4_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define SQ_PERFCOUNTER4_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define SQ_PERFCOUNTER5_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define SQ_PERFCOUNTER5_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define SQ_PERFCOUNTER6_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define SQ_PERFCOUNTER6_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define SQ_PERFCOUNTER7_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define SQ_PERFCOUNTER7_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define SQ_PERFCOUNTER8_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define SQ_PERFCOUNTER8_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define SQ_PERFCOUNTER9_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define SQ_PERFCOUNTER9_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define SQ_PERFCOUNTER10_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define SQ_PERFCOUNTER10_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define SQ_PERFCOUNTER11_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define SQ_PERFCOUNTER11_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define SQ_PERFCOUNTER12_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define SQ_PERFCOUNTER12_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define SQ_PERFCOUNTER13_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define SQ_PERFCOUNTER13_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define SQ_PERFCOUNTER14_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define SQ_PERFCOUNTER14_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define SQ_PERFCOUNTER15_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define SQ_PERFCOUNTER15_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define SQ_PERFCOUNTER0_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define SQ_PERFCOUNTER0_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define SQ_PERFCOUNTER1_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define SQ_PERFCOUNTER1_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define SQ_PERFCOUNTER2_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define SQ_PERFCOUNTER2_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define SQ_PERFCOUNTER3_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define SQ_PERFCOUNTER3_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define SQ_PERFCOUNTER4_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define SQ_PERFCOUNTER4_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define SQ_PERFCOUNTER5_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define SQ_PERFCOUNTER5_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define SQ_PERFCOUNTER6_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define SQ_PERFCOUNTER6_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define SQ_PERFCOUNTER7_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define SQ_PERFCOUNTER7_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define SQ_PERFCOUNTER8_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define SQ_PERFCOUNTER8_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define SQ_PERFCOUNTER9_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define SQ_PERFCOUNTER9_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define SQ_PERFCOUNTER10_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define SQ_PERFCOUNTER10_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define SQ_PERFCOUNTER11_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define SQ_PERFCOUNTER11_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define SQ_PERFCOUNTER12_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define SQ_PERFCOUNTER12_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define SQ_PERFCOUNTER13_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define SQ_PERFCOUNTER13_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define SQ_PERFCOUNTER14_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define SQ_PERFCOUNTER14_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define SQ_PERFCOUNTER15_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define SQ_PERFCOUNTER15_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define SQ_PERFCOUNTER0_SELECT__PERF_SEL_MASK 0x1ff
+#define SQ_PERFCOUNTER0_SELECT__PERF_SEL__SHIFT 0x0
+#define SQ_PERFCOUNTER0_SELECT__SQC_BANK_MASK_MASK 0xf000
+#define SQ_PERFCOUNTER0_SELECT__SQC_BANK_MASK__SHIFT 0xc
+#define SQ_PERFCOUNTER0_SELECT__SQC_CLIENT_MASK_MASK 0xf0000
+#define SQ_PERFCOUNTER0_SELECT__SQC_CLIENT_MASK__SHIFT 0x10
+#define SQ_PERFCOUNTER0_SELECT__SPM_MODE_MASK 0xf00000
+#define SQ_PERFCOUNTER0_SELECT__SPM_MODE__SHIFT 0x14
+#define SQ_PERFCOUNTER0_SELECT__SIMD_MASK_MASK 0xf000000
+#define SQ_PERFCOUNTER0_SELECT__SIMD_MASK__SHIFT 0x18
+#define SQ_PERFCOUNTER0_SELECT__PERF_MODE_MASK 0xf0000000
+#define SQ_PERFCOUNTER0_SELECT__PERF_MODE__SHIFT 0x1c
+#define SQ_PERFCOUNTER1_SELECT__PERF_SEL_MASK 0x1ff
+#define SQ_PERFCOUNTER1_SELECT__PERF_SEL__SHIFT 0x0
+#define SQ_PERFCOUNTER1_SELECT__SQC_BANK_MASK_MASK 0xf000
+#define SQ_PERFCOUNTER1_SELECT__SQC_BANK_MASK__SHIFT 0xc
+#define SQ_PERFCOUNTER1_SELECT__SQC_CLIENT_MASK_MASK 0xf0000
+#define SQ_PERFCOUNTER1_SELECT__SQC_CLIENT_MASK__SHIFT 0x10
+#define SQ_PERFCOUNTER1_SELECT__SPM_MODE_MASK 0xf00000
+#define SQ_PERFCOUNTER1_SELECT__SPM_MODE__SHIFT 0x14
+#define SQ_PERFCOUNTER1_SELECT__SIMD_MASK_MASK 0xf000000
+#define SQ_PERFCOUNTER1_SELECT__SIMD_MASK__SHIFT 0x18
+#define SQ_PERFCOUNTER1_SELECT__PERF_MODE_MASK 0xf0000000
+#define SQ_PERFCOUNTER1_SELECT__PERF_MODE__SHIFT 0x1c
+#define SQ_PERFCOUNTER2_SELECT__PERF_SEL_MASK 0x1ff
+#define SQ_PERFCOUNTER2_SELECT__PERF_SEL__SHIFT 0x0
+#define SQ_PERFCOUNTER2_SELECT__SQC_BANK_MASK_MASK 0xf000
+#define SQ_PERFCOUNTER2_SELECT__SQC_BANK_MASK__SHIFT 0xc
+#define SQ_PERFCOUNTER2_SELECT__SQC_CLIENT_MASK_MASK 0xf0000
+#define SQ_PERFCOUNTER2_SELECT__SQC_CLIENT_MASK__SHIFT 0x10
+#define SQ_PERFCOUNTER2_SELECT__SPM_MODE_MASK 0xf00000
+#define SQ_PERFCOUNTER2_SELECT__SPM_MODE__SHIFT 0x14
+#define SQ_PERFCOUNTER2_SELECT__SIMD_MASK_MASK 0xf000000
+#define SQ_PERFCOUNTER2_SELECT__SIMD_MASK__SHIFT 0x18
+#define SQ_PERFCOUNTER2_SELECT__PERF_MODE_MASK 0xf0000000
+#define SQ_PERFCOUNTER2_SELECT__PERF_MODE__SHIFT 0x1c
+#define SQ_PERFCOUNTER3_SELECT__PERF_SEL_MASK 0x1ff
+#define SQ_PERFCOUNTER3_SELECT__PERF_SEL__SHIFT 0x0
+#define SQ_PERFCOUNTER3_SELECT__SQC_BANK_MASK_MASK 0xf000
+#define SQ_PERFCOUNTER3_SELECT__SQC_BANK_MASK__SHIFT 0xc
+#define SQ_PERFCOUNTER3_SELECT__SQC_CLIENT_MASK_MASK 0xf0000
+#define SQ_PERFCOUNTER3_SELECT__SQC_CLIENT_MASK__SHIFT 0x10
+#define SQ_PERFCOUNTER3_SELECT__SPM_MODE_MASK 0xf00000
+#define SQ_PERFCOUNTER3_SELECT__SPM_MODE__SHIFT 0x14
+#define SQ_PERFCOUNTER3_SELECT__SIMD_MASK_MASK 0xf000000
+#define SQ_PERFCOUNTER3_SELECT__SIMD_MASK__SHIFT 0x18
+#define SQ_PERFCOUNTER3_SELECT__PERF_MODE_MASK 0xf0000000
+#define SQ_PERFCOUNTER3_SELECT__PERF_MODE__SHIFT 0x1c
+#define SQ_PERFCOUNTER4_SELECT__PERF_SEL_MASK 0x1ff
+#define SQ_PERFCOUNTER4_SELECT__PERF_SEL__SHIFT 0x0
+#define SQ_PERFCOUNTER4_SELECT__SQC_BANK_MASK_MASK 0xf000
+#define SQ_PERFCOUNTER4_SELECT__SQC_BANK_MASK__SHIFT 0xc
+#define SQ_PERFCOUNTER4_SELECT__SQC_CLIENT_MASK_MASK 0xf0000
+#define SQ_PERFCOUNTER4_SELECT__SQC_CLIENT_MASK__SHIFT 0x10
+#define SQ_PERFCOUNTER4_SELECT__SPM_MODE_MASK 0xf00000
+#define SQ_PERFCOUNTER4_SELECT__SPM_MODE__SHIFT 0x14
+#define SQ_PERFCOUNTER4_SELECT__SIMD_MASK_MASK 0xf000000
+#define SQ_PERFCOUNTER4_SELECT__SIMD_MASK__SHIFT 0x18
+#define SQ_PERFCOUNTER4_SELECT__PERF_MODE_MASK 0xf0000000
+#define SQ_PERFCOUNTER4_SELECT__PERF_MODE__SHIFT 0x1c
+#define SQ_PERFCOUNTER5_SELECT__PERF_SEL_MASK 0x1ff
+#define SQ_PERFCOUNTER5_SELECT__PERF_SEL__SHIFT 0x0
+#define SQ_PERFCOUNTER5_SELECT__SQC_BANK_MASK_MASK 0xf000
+#define SQ_PERFCOUNTER5_SELECT__SQC_BANK_MASK__SHIFT 0xc
+#define SQ_PERFCOUNTER5_SELECT__SQC_CLIENT_MASK_MASK 0xf0000
+#define SQ_PERFCOUNTER5_SELECT__SQC_CLIENT_MASK__SHIFT 0x10
+#define SQ_PERFCOUNTER5_SELECT__SPM_MODE_MASK 0xf00000
+#define SQ_PERFCOUNTER5_SELECT__SPM_MODE__SHIFT 0x14
+#define SQ_PERFCOUNTER5_SELECT__SIMD_MASK_MASK 0xf000000
+#define SQ_PERFCOUNTER5_SELECT__SIMD_MASK__SHIFT 0x18
+#define SQ_PERFCOUNTER5_SELECT__PERF_MODE_MASK 0xf0000000
+#define SQ_PERFCOUNTER5_SELECT__PERF_MODE__SHIFT 0x1c
+#define SQ_PERFCOUNTER6_SELECT__PERF_SEL_MASK 0x1ff
+#define SQ_PERFCOUNTER6_SELECT__PERF_SEL__SHIFT 0x0
+#define SQ_PERFCOUNTER6_SELECT__SQC_BANK_MASK_MASK 0xf000
+#define SQ_PERFCOUNTER6_SELECT__SQC_BANK_MASK__SHIFT 0xc
+#define SQ_PERFCOUNTER6_SELECT__SQC_CLIENT_MASK_MASK 0xf0000
+#define SQ_PERFCOUNTER6_SELECT__SQC_CLIENT_MASK__SHIFT 0x10
+#define SQ_PERFCOUNTER6_SELECT__SPM_MODE_MASK 0xf00000
+#define SQ_PERFCOUNTER6_SELECT__SPM_MODE__SHIFT 0x14
+#define SQ_PERFCOUNTER6_SELECT__SIMD_MASK_MASK 0xf000000
+#define SQ_PERFCOUNTER6_SELECT__SIMD_MASK__SHIFT 0x18
+#define SQ_PERFCOUNTER6_SELECT__PERF_MODE_MASK 0xf0000000
+#define SQ_PERFCOUNTER6_SELECT__PERF_MODE__SHIFT 0x1c
+#define SQ_PERFCOUNTER7_SELECT__PERF_SEL_MASK 0x1ff
+#define SQ_PERFCOUNTER7_SELECT__PERF_SEL__SHIFT 0x0
+#define SQ_PERFCOUNTER7_SELECT__SQC_BANK_MASK_MASK 0xf000
+#define SQ_PERFCOUNTER7_SELECT__SQC_BANK_MASK__SHIFT 0xc
+#define SQ_PERFCOUNTER7_SELECT__SQC_CLIENT_MASK_MASK 0xf0000
+#define SQ_PERFCOUNTER7_SELECT__SQC_CLIENT_MASK__SHIFT 0x10
+#define SQ_PERFCOUNTER7_SELECT__SPM_MODE_MASK 0xf00000
+#define SQ_PERFCOUNTER7_SELECT__SPM_MODE__SHIFT 0x14
+#define SQ_PERFCOUNTER7_SELECT__SIMD_MASK_MASK 0xf000000
+#define SQ_PERFCOUNTER7_SELECT__SIMD_MASK__SHIFT 0x18
+#define SQ_PERFCOUNTER7_SELECT__PERF_MODE_MASK 0xf0000000
+#define SQ_PERFCOUNTER7_SELECT__PERF_MODE__SHIFT 0x1c
+#define SQ_PERFCOUNTER8_SELECT__PERF_SEL_MASK 0x1ff
+#define SQ_PERFCOUNTER8_SELECT__PERF_SEL__SHIFT 0x0
+#define SQ_PERFCOUNTER8_SELECT__SQC_BANK_MASK_MASK 0xf000
+#define SQ_PERFCOUNTER8_SELECT__SQC_BANK_MASK__SHIFT 0xc
+#define SQ_PERFCOUNTER8_SELECT__SQC_CLIENT_MASK_MASK 0xf0000
+#define SQ_PERFCOUNTER8_SELECT__SQC_CLIENT_MASK__SHIFT 0x10
+#define SQ_PERFCOUNTER8_SELECT__SPM_MODE_MASK 0xf00000
+#define SQ_PERFCOUNTER8_SELECT__SPM_MODE__SHIFT 0x14
+#define SQ_PERFCOUNTER8_SELECT__SIMD_MASK_MASK 0xf000000
+#define SQ_PERFCOUNTER8_SELECT__SIMD_MASK__SHIFT 0x18
+#define SQ_PERFCOUNTER8_SELECT__PERF_MODE_MASK 0xf0000000
+#define SQ_PERFCOUNTER8_SELECT__PERF_MODE__SHIFT 0x1c
+#define SQ_PERFCOUNTER9_SELECT__PERF_SEL_MASK 0x1ff
+#define SQ_PERFCOUNTER9_SELECT__PERF_SEL__SHIFT 0x0
+#define SQ_PERFCOUNTER9_SELECT__SQC_BANK_MASK_MASK 0xf000
+#define SQ_PERFCOUNTER9_SELECT__SQC_BANK_MASK__SHIFT 0xc
+#define SQ_PERFCOUNTER9_SELECT__SQC_CLIENT_MASK_MASK 0xf0000
+#define SQ_PERFCOUNTER9_SELECT__SQC_CLIENT_MASK__SHIFT 0x10
+#define SQ_PERFCOUNTER9_SELECT__SPM_MODE_MASK 0xf00000
+#define SQ_PERFCOUNTER9_SELECT__SPM_MODE__SHIFT 0x14
+#define SQ_PERFCOUNTER9_SELECT__SIMD_MASK_MASK 0xf000000
+#define SQ_PERFCOUNTER9_SELECT__SIMD_MASK__SHIFT 0x18
+#define SQ_PERFCOUNTER9_SELECT__PERF_MODE_MASK 0xf0000000
+#define SQ_PERFCOUNTER9_SELECT__PERF_MODE__SHIFT 0x1c
+#define SQ_PERFCOUNTER10_SELECT__PERF_SEL_MASK 0x1ff
+#define SQ_PERFCOUNTER10_SELECT__PERF_SEL__SHIFT 0x0
+#define SQ_PERFCOUNTER10_SELECT__SQC_BANK_MASK_MASK 0xf000
+#define SQ_PERFCOUNTER10_SELECT__SQC_BANK_MASK__SHIFT 0xc
+#define SQ_PERFCOUNTER10_SELECT__SQC_CLIENT_MASK_MASK 0xf0000
+#define SQ_PERFCOUNTER10_SELECT__SQC_CLIENT_MASK__SHIFT 0x10
+#define SQ_PERFCOUNTER10_SELECT__SPM_MODE_MASK 0xf00000
+#define SQ_PERFCOUNTER10_SELECT__SPM_MODE__SHIFT 0x14
+#define SQ_PERFCOUNTER10_SELECT__SIMD_MASK_MASK 0xf000000
+#define SQ_PERFCOUNTER10_SELECT__SIMD_MASK__SHIFT 0x18
+#define SQ_PERFCOUNTER10_SELECT__PERF_MODE_MASK 0xf0000000
+#define SQ_PERFCOUNTER10_SELECT__PERF_MODE__SHIFT 0x1c
+#define SQ_PERFCOUNTER11_SELECT__PERF_SEL_MASK 0x1ff
+#define SQ_PERFCOUNTER11_SELECT__PERF_SEL__SHIFT 0x0
+#define SQ_PERFCOUNTER11_SELECT__SQC_BANK_MASK_MASK 0xf000
+#define SQ_PERFCOUNTER11_SELECT__SQC_BANK_MASK__SHIFT 0xc
+#define SQ_PERFCOUNTER11_SELECT__SQC_CLIENT_MASK_MASK 0xf0000
+#define SQ_PERFCOUNTER11_SELECT__SQC_CLIENT_MASK__SHIFT 0x10
+#define SQ_PERFCOUNTER11_SELECT__SPM_MODE_MASK 0xf00000
+#define SQ_PERFCOUNTER11_SELECT__SPM_MODE__SHIFT 0x14
+#define SQ_PERFCOUNTER11_SELECT__SIMD_MASK_MASK 0xf000000
+#define SQ_PERFCOUNTER11_SELECT__SIMD_MASK__SHIFT 0x18
+#define SQ_PERFCOUNTER11_SELECT__PERF_MODE_MASK 0xf0000000
+#define SQ_PERFCOUNTER11_SELECT__PERF_MODE__SHIFT 0x1c
+#define SQ_PERFCOUNTER12_SELECT__PERF_SEL_MASK 0x1ff
+#define SQ_PERFCOUNTER12_SELECT__PERF_SEL__SHIFT 0x0
+#define SQ_PERFCOUNTER12_SELECT__SQC_BANK_MASK_MASK 0xf000
+#define SQ_PERFCOUNTER12_SELECT__SQC_BANK_MASK__SHIFT 0xc
+#define SQ_PERFCOUNTER12_SELECT__SQC_CLIENT_MASK_MASK 0xf0000
+#define SQ_PERFCOUNTER12_SELECT__SQC_CLIENT_MASK__SHIFT 0x10
+#define SQ_PERFCOUNTER12_SELECT__SPM_MODE_MASK 0xf00000
+#define SQ_PERFCOUNTER12_SELECT__SPM_MODE__SHIFT 0x14
+#define SQ_PERFCOUNTER12_SELECT__SIMD_MASK_MASK 0xf000000
+#define SQ_PERFCOUNTER12_SELECT__SIMD_MASK__SHIFT 0x18
+#define SQ_PERFCOUNTER12_SELECT__PERF_MODE_MASK 0xf0000000
+#define SQ_PERFCOUNTER12_SELECT__PERF_MODE__SHIFT 0x1c
+#define SQ_PERFCOUNTER13_SELECT__PERF_SEL_MASK 0x1ff
+#define SQ_PERFCOUNTER13_SELECT__PERF_SEL__SHIFT 0x0
+#define SQ_PERFCOUNTER13_SELECT__SQC_BANK_MASK_MASK 0xf000
+#define SQ_PERFCOUNTER13_SELECT__SQC_BANK_MASK__SHIFT 0xc
+#define SQ_PERFCOUNTER13_SELECT__SQC_CLIENT_MASK_MASK 0xf0000
+#define SQ_PERFCOUNTER13_SELECT__SQC_CLIENT_MASK__SHIFT 0x10
+#define SQ_PERFCOUNTER13_SELECT__SPM_MODE_MASK 0xf00000
+#define SQ_PERFCOUNTER13_SELECT__SPM_MODE__SHIFT 0x14
+#define SQ_PERFCOUNTER13_SELECT__SIMD_MASK_MASK 0xf000000
+#define SQ_PERFCOUNTER13_SELECT__SIMD_MASK__SHIFT 0x18
+#define SQ_PERFCOUNTER13_SELECT__PERF_MODE_MASK 0xf0000000
+#define SQ_PERFCOUNTER13_SELECT__PERF_MODE__SHIFT 0x1c
+#define SQ_PERFCOUNTER14_SELECT__PERF_SEL_MASK 0x1ff
+#define SQ_PERFCOUNTER14_SELECT__PERF_SEL__SHIFT 0x0
+#define SQ_PERFCOUNTER14_SELECT__SQC_BANK_MASK_MASK 0xf000
+#define SQ_PERFCOUNTER14_SELECT__SQC_BANK_MASK__SHIFT 0xc
+#define SQ_PERFCOUNTER14_SELECT__SQC_CLIENT_MASK_MASK 0xf0000
+#define SQ_PERFCOUNTER14_SELECT__SQC_CLIENT_MASK__SHIFT 0x10
+#define SQ_PERFCOUNTER14_SELECT__SPM_MODE_MASK 0xf00000
+#define SQ_PERFCOUNTER14_SELECT__SPM_MODE__SHIFT 0x14
+#define SQ_PERFCOUNTER14_SELECT__SIMD_MASK_MASK 0xf000000
+#define SQ_PERFCOUNTER14_SELECT__SIMD_MASK__SHIFT 0x18
+#define SQ_PERFCOUNTER14_SELECT__PERF_MODE_MASK 0xf0000000
+#define SQ_PERFCOUNTER14_SELECT__PERF_MODE__SHIFT 0x1c
+#define SQ_PERFCOUNTER15_SELECT__PERF_SEL_MASK 0x1ff
+#define SQ_PERFCOUNTER15_SELECT__PERF_SEL__SHIFT 0x0
+#define SQ_PERFCOUNTER15_SELECT__SQC_BANK_MASK_MASK 0xf000
+#define SQ_PERFCOUNTER15_SELECT__SQC_BANK_MASK__SHIFT 0xc
+#define SQ_PERFCOUNTER15_SELECT__SQC_CLIENT_MASK_MASK 0xf0000
+#define SQ_PERFCOUNTER15_SELECT__SQC_CLIENT_MASK__SHIFT 0x10
+#define SQ_PERFCOUNTER15_SELECT__SPM_MODE_MASK 0xf00000
+#define SQ_PERFCOUNTER15_SELECT__SPM_MODE__SHIFT 0x14
+#define SQ_PERFCOUNTER15_SELECT__SIMD_MASK_MASK 0xf000000
+#define SQ_PERFCOUNTER15_SELECT__SIMD_MASK__SHIFT 0x18
+#define SQ_PERFCOUNTER15_SELECT__PERF_MODE_MASK 0xf0000000
+#define SQ_PERFCOUNTER15_SELECT__PERF_MODE__SHIFT 0x1c
+#define CGTT_SQ_CLK_CTRL__ON_DELAY_MASK 0xf
+#define CGTT_SQ_CLK_CTRL__ON_DELAY__SHIFT 0x0
+#define CGTT_SQ_CLK_CTRL__OFF_HYSTERESIS_MASK 0xff0
+#define CGTT_SQ_CLK_CTRL__OFF_HYSTERESIS__SHIFT 0x4
+#define CGTT_SQ_CLK_CTRL__PERFMON_OVERRIDE_MASK 0x20000000
+#define CGTT_SQ_CLK_CTRL__PERFMON_OVERRIDE__SHIFT 0x1d
+#define CGTT_SQ_CLK_CTRL__CORE_OVERRIDE_MASK 0x40000000
+#define CGTT_SQ_CLK_CTRL__CORE_OVERRIDE__SHIFT 0x1e
+#define CGTT_SQ_CLK_CTRL__REG_OVERRIDE_MASK 0x80000000
+#define CGTT_SQ_CLK_CTRL__REG_OVERRIDE__SHIFT 0x1f
+#define CGTT_SQG_CLK_CTRL__ON_DELAY_MASK 0xf
+#define CGTT_SQG_CLK_CTRL__ON_DELAY__SHIFT 0x0
+#define CGTT_SQG_CLK_CTRL__OFF_HYSTERESIS_MASK 0xff0
+#define CGTT_SQG_CLK_CTRL__OFF_HYSTERESIS__SHIFT 0x4
+#define CGTT_SQG_CLK_CTRL__TTRACE_OVERRIDE_MASK 0x10000000
+#define CGTT_SQG_CLK_CTRL__TTRACE_OVERRIDE__SHIFT 0x1c
+#define CGTT_SQG_CLK_CTRL__PERFMON_OVERRIDE_MASK 0x20000000
+#define CGTT_SQG_CLK_CTRL__PERFMON_OVERRIDE__SHIFT 0x1d
+#define CGTT_SQG_CLK_CTRL__CORE_OVERRIDE_MASK 0x40000000
+#define CGTT_SQG_CLK_CTRL__CORE_OVERRIDE__SHIFT 0x1e
+#define CGTT_SQG_CLK_CTRL__REG_OVERRIDE_MASK 0x80000000
+#define CGTT_SQG_CLK_CTRL__REG_OVERRIDE__SHIFT 0x1f
+#define SQ_ALU_CLK_CTRL__FORCE_CU_ON_SH0_MASK 0xffff
+#define SQ_ALU_CLK_CTRL__FORCE_CU_ON_SH0__SHIFT 0x0
+#define SQ_ALU_CLK_CTRL__FORCE_CU_ON_SH1_MASK 0xffff0000
+#define SQ_ALU_CLK_CTRL__FORCE_CU_ON_SH1__SHIFT 0x10
+#define SQ_TEX_CLK_CTRL__FORCE_CU_ON_SH0_MASK 0xffff
+#define SQ_TEX_CLK_CTRL__FORCE_CU_ON_SH0__SHIFT 0x0
+#define SQ_TEX_CLK_CTRL__FORCE_CU_ON_SH1_MASK 0xffff0000
+#define SQ_TEX_CLK_CTRL__FORCE_CU_ON_SH1__SHIFT 0x10
+#define SQ_LDS_CLK_CTRL__FORCE_CU_ON_SH0_MASK 0xffff
+#define SQ_LDS_CLK_CTRL__FORCE_CU_ON_SH0__SHIFT 0x0
+#define SQ_LDS_CLK_CTRL__FORCE_CU_ON_SH1_MASK 0xffff0000
+#define SQ_LDS_CLK_CTRL__FORCE_CU_ON_SH1__SHIFT 0x10
+#define SQ_POWER_THROTTLE__MIN_POWER_MASK 0x3fff
+#define SQ_POWER_THROTTLE__MIN_POWER__SHIFT 0x0
+#define SQ_POWER_THROTTLE__MAX_POWER_MASK 0x3fff0000
+#define SQ_POWER_THROTTLE__MAX_POWER__SHIFT 0x10
+#define SQ_POWER_THROTTLE__PHASE_OFFSET_MASK 0xc0000000
+#define SQ_POWER_THROTTLE__PHASE_OFFSET__SHIFT 0x1e
+#define SQ_POWER_THROTTLE2__MAX_POWER_DELTA_MASK 0x3fff
+#define SQ_POWER_THROTTLE2__MAX_POWER_DELTA__SHIFT 0x0
+#define SQ_POWER_THROTTLE2__SHORT_TERM_INTERVAL_SIZE_MASK 0x3ff0000
+#define SQ_POWER_THROTTLE2__SHORT_TERM_INTERVAL_SIZE__SHIFT 0x10
+#define SQ_POWER_THROTTLE2__LONG_TERM_INTERVAL_RATIO_MASK 0x78000000
+#define SQ_POWER_THROTTLE2__LONG_TERM_INTERVAL_RATIO__SHIFT 0x1b
+#define SQ_POWER_THROTTLE2__USE_REF_CLOCK_MASK 0x80000000
+#define SQ_POWER_THROTTLE2__USE_REF_CLOCK__SHIFT 0x1f
+#define SQ_TIME_HI__TIME_MASK 0xffffffff
+#define SQ_TIME_HI__TIME__SHIFT 0x0
+#define SQ_TIME_LO__TIME_MASK 0xffffffff
+#define SQ_TIME_LO__TIME__SHIFT 0x0
+#define SQ_THREAD_TRACE_BASE__ADDR_MASK 0xffffffff
+#define SQ_THREAD_TRACE_BASE__ADDR__SHIFT 0x0
+#define SQ_THREAD_TRACE_BASE2__ADDR_HI_MASK 0xf
+#define SQ_THREAD_TRACE_BASE2__ADDR_HI__SHIFT 0x0
+#define SQ_THREAD_TRACE_SIZE__SIZE_MASK 0x3fffff
+#define SQ_THREAD_TRACE_SIZE__SIZE__SHIFT 0x0
+#define SQ_THREAD_TRACE_MASK__CU_SEL_MASK 0x1f
+#define SQ_THREAD_TRACE_MASK__CU_SEL__SHIFT 0x0
+#define SQ_THREAD_TRACE_MASK__SH_SEL_MASK 0x20
+#define SQ_THREAD_TRACE_MASK__SH_SEL__SHIFT 0x5
+#define SQ_THREAD_TRACE_MASK__REG_STALL_EN_MASK 0x80
+#define SQ_THREAD_TRACE_MASK__REG_STALL_EN__SHIFT 0x7
+#define SQ_THREAD_TRACE_MASK__SIMD_EN_MASK 0xf00
+#define SQ_THREAD_TRACE_MASK__SIMD_EN__SHIFT 0x8
+#define SQ_THREAD_TRACE_MASK__VM_ID_MASK_MASK 0x3000
+#define SQ_THREAD_TRACE_MASK__VM_ID_MASK__SHIFT 0xc
+#define SQ_THREAD_TRACE_MASK__SPI_STALL_EN_MASK 0x4000
+#define SQ_THREAD_TRACE_MASK__SPI_STALL_EN__SHIFT 0xe
+#define SQ_THREAD_TRACE_MASK__SQ_STALL_EN_MASK 0x8000
+#define SQ_THREAD_TRACE_MASK__SQ_STALL_EN__SHIFT 0xf
+#define SQ_THREAD_TRACE_USERDATA_0__DATA_MASK 0xffffffff
+#define SQ_THREAD_TRACE_USERDATA_0__DATA__SHIFT 0x0
+#define SQ_THREAD_TRACE_USERDATA_1__DATA_MASK 0xffffffff
+#define SQ_THREAD_TRACE_USERDATA_1__DATA__SHIFT 0x0
+#define SQ_THREAD_TRACE_USERDATA_2__DATA_MASK 0xffffffff
+#define SQ_THREAD_TRACE_USERDATA_2__DATA__SHIFT 0x0
+#define SQ_THREAD_TRACE_USERDATA_3__DATA_MASK 0xffffffff
+#define SQ_THREAD_TRACE_USERDATA_3__DATA__SHIFT 0x0
+#define SQ_THREAD_TRACE_MODE__MASK_PS_MASK 0x7
+#define SQ_THREAD_TRACE_MODE__MASK_PS__SHIFT 0x0
+#define SQ_THREAD_TRACE_MODE__MASK_VS_MASK 0x38
+#define SQ_THREAD_TRACE_MODE__MASK_VS__SHIFT 0x3
+#define SQ_THREAD_TRACE_MODE__MASK_GS_MASK 0x1c0
+#define SQ_THREAD_TRACE_MODE__MASK_GS__SHIFT 0x6
+#define SQ_THREAD_TRACE_MODE__MASK_ES_MASK 0xe00
+#define SQ_THREAD_TRACE_MODE__MASK_ES__SHIFT 0x9
+#define SQ_THREAD_TRACE_MODE__MASK_HS_MASK 0x7000
+#define SQ_THREAD_TRACE_MODE__MASK_HS__SHIFT 0xc
+#define SQ_THREAD_TRACE_MODE__MASK_LS_MASK 0x38000
+#define SQ_THREAD_TRACE_MODE__MASK_LS__SHIFT 0xf
+#define SQ_THREAD_TRACE_MODE__MASK_CS_MASK 0x1c0000
+#define SQ_THREAD_TRACE_MODE__MASK_CS__SHIFT 0x12
+#define SQ_THREAD_TRACE_MODE__MODE_MASK 0x600000
+#define SQ_THREAD_TRACE_MODE__MODE__SHIFT 0x15
+#define SQ_THREAD_TRACE_MODE__CAPTURE_MODE_MASK 0x1800000
+#define SQ_THREAD_TRACE_MODE__CAPTURE_MODE__SHIFT 0x17
+#define SQ_THREAD_TRACE_MODE__AUTOFLUSH_EN_MASK 0x2000000
+#define SQ_THREAD_TRACE_MODE__AUTOFLUSH_EN__SHIFT 0x19
+#define SQ_THREAD_TRACE_MODE__PRIV_MASK 0x4000000
+#define SQ_THREAD_TRACE_MODE__PRIV__SHIFT 0x1a
+#define SQ_THREAD_TRACE_MODE__ISSUE_MASK_MASK 0x18000000
+#define SQ_THREAD_TRACE_MODE__ISSUE_MASK__SHIFT 0x1b
+#define SQ_THREAD_TRACE_MODE__TEST_MODE_MASK 0x20000000
+#define SQ_THREAD_TRACE_MODE__TEST_MODE__SHIFT 0x1d
+#define SQ_THREAD_TRACE_MODE__INTERRUPT_EN_MASK 0x40000000
+#define SQ_THREAD_TRACE_MODE__INTERRUPT_EN__SHIFT 0x1e
+#define SQ_THREAD_TRACE_MODE__WRAP_MASK 0x80000000
+#define SQ_THREAD_TRACE_MODE__WRAP__SHIFT 0x1f
+#define SQ_THREAD_TRACE_CTRL__RESET_BUFFER_MASK 0x80000000
+#define SQ_THREAD_TRACE_CTRL__RESET_BUFFER__SHIFT 0x1f
+#define SQ_THREAD_TRACE_TOKEN_MASK__TOKEN_MASK_MASK 0xffff
+#define SQ_THREAD_TRACE_TOKEN_MASK__TOKEN_MASK__SHIFT 0x0
+#define SQ_THREAD_TRACE_TOKEN_MASK__REG_MASK_MASK 0xff0000
+#define SQ_THREAD_TRACE_TOKEN_MASK__REG_MASK__SHIFT 0x10
+#define SQ_THREAD_TRACE_TOKEN_MASK__REG_DROP_ON_STALL_MASK 0x1000000
+#define SQ_THREAD_TRACE_TOKEN_MASK__REG_DROP_ON_STALL__SHIFT 0x18
+#define SQ_THREAD_TRACE_TOKEN_MASK2__INST_MASK_MASK 0xffffffff
+#define SQ_THREAD_TRACE_TOKEN_MASK2__INST_MASK__SHIFT 0x0
+#define SQ_THREAD_TRACE_PERF_MASK__SH0_MASK_MASK 0xffff
+#define SQ_THREAD_TRACE_PERF_MASK__SH0_MASK__SHIFT 0x0
+#define SQ_THREAD_TRACE_PERF_MASK__SH1_MASK_MASK 0xffff0000
+#define SQ_THREAD_TRACE_PERF_MASK__SH1_MASK__SHIFT 0x10
+#define SQ_THREAD_TRACE_WPTR__WPTR_MASK 0x3fffffff
+#define SQ_THREAD_TRACE_WPTR__WPTR__SHIFT 0x0
+#define SQ_THREAD_TRACE_WPTR__READ_OFFSET_MASK 0xc0000000
+#define SQ_THREAD_TRACE_WPTR__READ_OFFSET__SHIFT 0x1e
+#define SQ_THREAD_TRACE_STATUS__FINISH_PENDING_MASK 0x3ff
+#define SQ_THREAD_TRACE_STATUS__FINISH_PENDING__SHIFT 0x0
+#define SQ_THREAD_TRACE_STATUS__FINISH_DONE_MASK 0x3ff0000
+#define SQ_THREAD_TRACE_STATUS__FINISH_DONE__SHIFT 0x10
+#define SQ_THREAD_TRACE_STATUS__NEW_BUF_MASK 0x20000000
+#define SQ_THREAD_TRACE_STATUS__NEW_BUF__SHIFT 0x1d
+#define SQ_THREAD_TRACE_STATUS__BUSY_MASK 0x40000000
+#define SQ_THREAD_TRACE_STATUS__BUSY__SHIFT 0x1e
+#define SQ_THREAD_TRACE_STATUS__FULL_MASK 0x80000000
+#define SQ_THREAD_TRACE_STATUS__FULL__SHIFT 0x1f
+#define SQ_THREAD_TRACE_CNTR__CNTR_MASK 0xffffffff
+#define SQ_THREAD_TRACE_CNTR__CNTR__SHIFT 0x0
+#define SQ_THREAD_TRACE_HIWATER__HIWATER_MASK 0x7
+#define SQ_THREAD_TRACE_HIWATER__HIWATER__SHIFT 0x0
+#define SQ_LB_CTR_CTRL__START_MASK 0x1
+#define SQ_LB_CTR_CTRL__START__SHIFT 0x0
+#define SQ_LB_CTR_CTRL__LOAD_MASK 0x2
+#define SQ_LB_CTR_CTRL__LOAD__SHIFT 0x1
+#define SQ_LB_CTR_CTRL__CLEAR_MASK 0x4
+#define SQ_LB_CTR_CTRL__CLEAR__SHIFT 0x2
+#define SQ_LB_DATA_ALU_CYCLES__DATA_MASK 0xffffffff
+#define SQ_LB_DATA_ALU_CYCLES__DATA__SHIFT 0x0
+#define SQ_LB_DATA_TEX_CYCLES__DATA_MASK 0xffffffff
+#define SQ_LB_DATA_TEX_CYCLES__DATA__SHIFT 0x0
+#define SQ_LB_DATA_ALU_STALLS__DATA_MASK 0xffffffff
+#define SQ_LB_DATA_ALU_STALLS__DATA__SHIFT 0x0
+#define SQ_LB_DATA_TEX_STALLS__DATA_MASK 0xffffffff
+#define SQ_LB_DATA_TEX_STALLS__DATA__SHIFT 0x0
+#define SQC_EDC_CNT__INST_SEC_MASK 0xff
+#define SQC_EDC_CNT__INST_SEC__SHIFT 0x0
+#define SQC_EDC_CNT__INST_DED_MASK 0xff00
+#define SQC_EDC_CNT__INST_DED__SHIFT 0x8
+#define SQC_EDC_CNT__DATA_SEC_MASK 0xff0000
+#define SQC_EDC_CNT__DATA_SEC__SHIFT 0x10
+#define SQC_EDC_CNT__DATA_DED_MASK 0xff000000
+#define SQC_EDC_CNT__DATA_DED__SHIFT 0x18
+#define SQ_EDC_SEC_CNT__LDS_SEC_MASK 0xff
+#define SQ_EDC_SEC_CNT__LDS_SEC__SHIFT 0x0
+#define SQ_EDC_SEC_CNT__SGPR_SEC_MASK 0xff00
+#define SQ_EDC_SEC_CNT__SGPR_SEC__SHIFT 0x8
+#define SQ_EDC_SEC_CNT__VGPR_SEC_MASK 0xff0000
+#define SQ_EDC_SEC_CNT__VGPR_SEC__SHIFT 0x10
+#define SQ_EDC_DED_CNT__LDS_DED_MASK 0xff
+#define SQ_EDC_DED_CNT__LDS_DED__SHIFT 0x0
+#define SQ_EDC_DED_CNT__SGPR_DED_MASK 0xff00
+#define SQ_EDC_DED_CNT__SGPR_DED__SHIFT 0x8
+#define SQ_EDC_DED_CNT__VGPR_DED_MASK 0xff0000
+#define SQ_EDC_DED_CNT__VGPR_DED__SHIFT 0x10
+#define SQ_EDC_INFO__WAVE_ID_MASK 0xf
+#define SQ_EDC_INFO__WAVE_ID__SHIFT 0x0
+#define SQ_EDC_INFO__SIMD_ID_MASK 0x30
+#define SQ_EDC_INFO__SIMD_ID__SHIFT 0x4
+#define SQ_EDC_INFO__SOURCE_MASK 0x1c0
+#define SQ_EDC_INFO__SOURCE__SHIFT 0x6
+#define SQ_EDC_INFO__VM_ID_MASK 0x1e00
+#define SQ_EDC_INFO__VM_ID__SHIFT 0x9
+#define SQ_BUF_RSRC_WORD0__BASE_ADDRESS_MASK 0xffffffff
+#define SQ_BUF_RSRC_WORD0__BASE_ADDRESS__SHIFT 0x0
+#define SQ_BUF_RSRC_WORD1__BASE_ADDRESS_HI_MASK 0xffff
+#define SQ_BUF_RSRC_WORD1__BASE_ADDRESS_HI__SHIFT 0x0
+#define SQ_BUF_RSRC_WORD1__STRIDE_MASK 0x3fff0000
+#define SQ_BUF_RSRC_WORD1__STRIDE__SHIFT 0x10
+#define SQ_BUF_RSRC_WORD1__CACHE_SWIZZLE_MASK 0x40000000
+#define SQ_BUF_RSRC_WORD1__CACHE_SWIZZLE__SHIFT 0x1e
+#define SQ_BUF_RSRC_WORD1__SWIZZLE_ENABLE_MASK 0x80000000
+#define SQ_BUF_RSRC_WORD1__SWIZZLE_ENABLE__SHIFT 0x1f
+#define SQ_BUF_RSRC_WORD2__NUM_RECORDS_MASK 0xffffffff
+#define SQ_BUF_RSRC_WORD2__NUM_RECORDS__SHIFT 0x0
+#define SQ_BUF_RSRC_WORD3__DST_SEL_X_MASK 0x7
+#define SQ_BUF_RSRC_WORD3__DST_SEL_X__SHIFT 0x0
+#define SQ_BUF_RSRC_WORD3__DST_SEL_Y_MASK 0x38
+#define SQ_BUF_RSRC_WORD3__DST_SEL_Y__SHIFT 0x3
+#define SQ_BUF_RSRC_WORD3__DST_SEL_Z_MASK 0x1c0
+#define SQ_BUF_RSRC_WORD3__DST_SEL_Z__SHIFT 0x6
+#define SQ_BUF_RSRC_WORD3__DST_SEL_W_MASK 0xe00
+#define SQ_BUF_RSRC_WORD3__DST_SEL_W__SHIFT 0x9
+#define SQ_BUF_RSRC_WORD3__NUM_FORMAT_MASK 0x7000
+#define SQ_BUF_RSRC_WORD3__NUM_FORMAT__SHIFT 0xc
+#define SQ_BUF_RSRC_WORD3__DATA_FORMAT_MASK 0x78000
+#define SQ_BUF_RSRC_WORD3__DATA_FORMAT__SHIFT 0xf
+#define SQ_BUF_RSRC_WORD3__ELEMENT_SIZE_MASK 0x180000
+#define SQ_BUF_RSRC_WORD3__ELEMENT_SIZE__SHIFT 0x13
+#define SQ_BUF_RSRC_WORD3__INDEX_STRIDE_MASK 0x600000
+#define SQ_BUF_RSRC_WORD3__INDEX_STRIDE__SHIFT 0x15
+#define SQ_BUF_RSRC_WORD3__ADD_TID_ENABLE_MASK 0x800000
+#define SQ_BUF_RSRC_WORD3__ADD_TID_ENABLE__SHIFT 0x17
+#define SQ_BUF_RSRC_WORD3__ATC_MASK 0x1000000
+#define SQ_BUF_RSRC_WORD3__ATC__SHIFT 0x18
+#define SQ_BUF_RSRC_WORD3__HASH_ENABLE_MASK 0x2000000
+#define SQ_BUF_RSRC_WORD3__HASH_ENABLE__SHIFT 0x19
+#define SQ_BUF_RSRC_WORD3__HEAP_MASK 0x4000000
+#define SQ_BUF_RSRC_WORD3__HEAP__SHIFT 0x1a
+#define SQ_BUF_RSRC_WORD3__MTYPE_MASK 0x38000000
+#define SQ_BUF_RSRC_WORD3__MTYPE__SHIFT 0x1b
+#define SQ_BUF_RSRC_WORD3__TYPE_MASK 0xc0000000
+#define SQ_BUF_RSRC_WORD3__TYPE__SHIFT 0x1e
+#define SQ_IMG_RSRC_WORD0__BASE_ADDRESS_MASK 0xffffffff
+#define SQ_IMG_RSRC_WORD0__BASE_ADDRESS__SHIFT 0x0
+#define SQ_IMG_RSRC_WORD1__BASE_ADDRESS_HI_MASK 0xff
+#define SQ_IMG_RSRC_WORD1__BASE_ADDRESS_HI__SHIFT 0x0
+#define SQ_IMG_RSRC_WORD1__MIN_LOD_MASK 0xfff00
+#define SQ_IMG_RSRC_WORD1__MIN_LOD__SHIFT 0x8
+#define SQ_IMG_RSRC_WORD1__DATA_FORMAT_MASK 0x3f00000
+#define SQ_IMG_RSRC_WORD1__DATA_FORMAT__SHIFT 0x14
+#define SQ_IMG_RSRC_WORD1__NUM_FORMAT_MASK 0x3c000000
+#define SQ_IMG_RSRC_WORD1__NUM_FORMAT__SHIFT 0x1a
+#define SQ_IMG_RSRC_WORD1__MTYPE_MASK 0xc0000000
+#define SQ_IMG_RSRC_WORD1__MTYPE__SHIFT 0x1e
+#define SQ_IMG_RSRC_WORD2__WIDTH_MASK 0x3fff
+#define SQ_IMG_RSRC_WORD2__WIDTH__SHIFT 0x0
+#define SQ_IMG_RSRC_WORD2__HEIGHT_MASK 0xfffc000
+#define SQ_IMG_RSRC_WORD2__HEIGHT__SHIFT 0xe
+#define SQ_IMG_RSRC_WORD2__PERF_MOD_MASK 0x70000000
+#define SQ_IMG_RSRC_WORD2__PERF_MOD__SHIFT 0x1c
+#define SQ_IMG_RSRC_WORD2__INTERLACED_MASK 0x80000000
+#define SQ_IMG_RSRC_WORD2__INTERLACED__SHIFT 0x1f
+#define SQ_IMG_RSRC_WORD3__DST_SEL_X_MASK 0x7
+#define SQ_IMG_RSRC_WORD3__DST_SEL_X__SHIFT 0x0
+#define SQ_IMG_RSRC_WORD3__DST_SEL_Y_MASK 0x38
+#define SQ_IMG_RSRC_WORD3__DST_SEL_Y__SHIFT 0x3
+#define SQ_IMG_RSRC_WORD3__DST_SEL_Z_MASK 0x1c0
+#define SQ_IMG_RSRC_WORD3__DST_SEL_Z__SHIFT 0x6
+#define SQ_IMG_RSRC_WORD3__DST_SEL_W_MASK 0xe00
+#define SQ_IMG_RSRC_WORD3__DST_SEL_W__SHIFT 0x9
+#define SQ_IMG_RSRC_WORD3__BASE_LEVEL_MASK 0xf000
+#define SQ_IMG_RSRC_WORD3__BASE_LEVEL__SHIFT 0xc
+#define SQ_IMG_RSRC_WORD3__LAST_LEVEL_MASK 0xf0000
+#define SQ_IMG_RSRC_WORD3__LAST_LEVEL__SHIFT 0x10
+#define SQ_IMG_RSRC_WORD3__TILING_INDEX_MASK 0x1f00000
+#define SQ_IMG_RSRC_WORD3__TILING_INDEX__SHIFT 0x14
+#define SQ_IMG_RSRC_WORD3__POW2_PAD_MASK 0x2000000
+#define SQ_IMG_RSRC_WORD3__POW2_PAD__SHIFT 0x19
+#define SQ_IMG_RSRC_WORD3__MTYPE_MASK 0x4000000
+#define SQ_IMG_RSRC_WORD3__MTYPE__SHIFT 0x1a
+#define SQ_IMG_RSRC_WORD3__ATC_MASK 0x8000000
+#define SQ_IMG_RSRC_WORD3__ATC__SHIFT 0x1b
+#define SQ_IMG_RSRC_WORD3__TYPE_MASK 0xf0000000
+#define SQ_IMG_RSRC_WORD3__TYPE__SHIFT 0x1c
+#define SQ_IMG_RSRC_WORD4__DEPTH_MASK 0x1fff
+#define SQ_IMG_RSRC_WORD4__DEPTH__SHIFT 0x0
+#define SQ_IMG_RSRC_WORD4__PITCH_MASK 0x7ffe000
+#define SQ_IMG_RSRC_WORD4__PITCH__SHIFT 0xd
+#define SQ_IMG_RSRC_WORD5__BASE_ARRAY_MASK 0x1fff
+#define SQ_IMG_RSRC_WORD5__BASE_ARRAY__SHIFT 0x0
+#define SQ_IMG_RSRC_WORD5__LAST_ARRAY_MASK 0x3ffe000
+#define SQ_IMG_RSRC_WORD5__LAST_ARRAY__SHIFT 0xd
+#define SQ_IMG_RSRC_WORD6__MIN_LOD_WARN_MASK 0xfff
+#define SQ_IMG_RSRC_WORD6__MIN_LOD_WARN__SHIFT 0x0
+#define SQ_IMG_RSRC_WORD6__COUNTER_BANK_ID_MASK 0xff000
+#define SQ_IMG_RSRC_WORD6__COUNTER_BANK_ID__SHIFT 0xc
+#define SQ_IMG_RSRC_WORD6__LOD_HDW_CNT_EN_MASK 0x100000
+#define SQ_IMG_RSRC_WORD6__LOD_HDW_CNT_EN__SHIFT 0x14
+#define SQ_IMG_RSRC_WORD6__COMPRESSION_EN_MASK 0x200000
+#define SQ_IMG_RSRC_WORD6__COMPRESSION_EN__SHIFT 0x15
+#define SQ_IMG_RSRC_WORD6__ALPHA_IS_ON_MSB_MASK 0x400000
+#define SQ_IMG_RSRC_WORD6__ALPHA_IS_ON_MSB__SHIFT 0x16
+#define SQ_IMG_RSRC_WORD6__COLOR_TRANSFORM_MASK 0x800000
+#define SQ_IMG_RSRC_WORD6__COLOR_TRANSFORM__SHIFT 0x17
+#define SQ_IMG_RSRC_WORD6__LOST_ALPHA_BITS_MASK 0xf000000
+#define SQ_IMG_RSRC_WORD6__LOST_ALPHA_BITS__SHIFT 0x18
+#define SQ_IMG_RSRC_WORD6__LOST_COLOR_BITS_MASK 0xf0000000
+#define SQ_IMG_RSRC_WORD6__LOST_COLOR_BITS__SHIFT 0x1c
+#define SQ_IMG_RSRC_WORD7__META_DATA_ADDRESS_MASK 0xffffffff
+#define SQ_IMG_RSRC_WORD7__META_DATA_ADDRESS__SHIFT 0x0
+#define SQ_IMG_SAMP_WORD0__CLAMP_X_MASK 0x7
+#define SQ_IMG_SAMP_WORD0__CLAMP_X__SHIFT 0x0
+#define SQ_IMG_SAMP_WORD0__CLAMP_Y_MASK 0x38
+#define SQ_IMG_SAMP_WORD0__CLAMP_Y__SHIFT 0x3
+#define SQ_IMG_SAMP_WORD0__CLAMP_Z_MASK 0x1c0
+#define SQ_IMG_SAMP_WORD0__CLAMP_Z__SHIFT 0x6
+#define SQ_IMG_SAMP_WORD0__MAX_ANISO_RATIO_MASK 0xe00
+#define SQ_IMG_SAMP_WORD0__MAX_ANISO_RATIO__SHIFT 0x9
+#define SQ_IMG_SAMP_WORD0__DEPTH_COMPARE_FUNC_MASK 0x7000
+#define SQ_IMG_SAMP_WORD0__DEPTH_COMPARE_FUNC__SHIFT 0xc
+#define SQ_IMG_SAMP_WORD0__FORCE_UNNORMALIZED_MASK 0x8000
+#define SQ_IMG_SAMP_WORD0__FORCE_UNNORMALIZED__SHIFT 0xf
+#define SQ_IMG_SAMP_WORD0__ANISO_THRESHOLD_MASK 0x70000
+#define SQ_IMG_SAMP_WORD0__ANISO_THRESHOLD__SHIFT 0x10
+#define SQ_IMG_SAMP_WORD0__MC_COORD_TRUNC_MASK 0x80000
+#define SQ_IMG_SAMP_WORD0__MC_COORD_TRUNC__SHIFT 0x13
+#define SQ_IMG_SAMP_WORD0__FORCE_DEGAMMA_MASK 0x100000
+#define SQ_IMG_SAMP_WORD0__FORCE_DEGAMMA__SHIFT 0x14
+#define SQ_IMG_SAMP_WORD0__ANISO_BIAS_MASK 0x7e00000
+#define SQ_IMG_SAMP_WORD0__ANISO_BIAS__SHIFT 0x15
+#define SQ_IMG_SAMP_WORD0__TRUNC_COORD_MASK 0x8000000
+#define SQ_IMG_SAMP_WORD0__TRUNC_COORD__SHIFT 0x1b
+#define SQ_IMG_SAMP_WORD0__DISABLE_CUBE_WRAP_MASK 0x10000000
+#define SQ_IMG_SAMP_WORD0__DISABLE_CUBE_WRAP__SHIFT 0x1c
+#define SQ_IMG_SAMP_WORD0__FILTER_MODE_MASK 0x60000000
+#define SQ_IMG_SAMP_WORD0__FILTER_MODE__SHIFT 0x1d
+#define SQ_IMG_SAMP_WORD0__COMPAT_MODE_MASK 0x80000000
+#define SQ_IMG_SAMP_WORD0__COMPAT_MODE__SHIFT 0x1f
+#define SQ_IMG_SAMP_WORD1__MIN_LOD_MASK 0xfff
+#define SQ_IMG_SAMP_WORD1__MIN_LOD__SHIFT 0x0
+#define SQ_IMG_SAMP_WORD1__MAX_LOD_MASK 0xfff000
+#define SQ_IMG_SAMP_WORD1__MAX_LOD__SHIFT 0xc
+#define SQ_IMG_SAMP_WORD1__PERF_MIP_MASK 0xf000000
+#define SQ_IMG_SAMP_WORD1__PERF_MIP__SHIFT 0x18
+#define SQ_IMG_SAMP_WORD1__PERF_Z_MASK 0xf0000000
+#define SQ_IMG_SAMP_WORD1__PERF_Z__SHIFT 0x1c
+#define SQ_IMG_SAMP_WORD2__LOD_BIAS_MASK 0x3fff
+#define SQ_IMG_SAMP_WORD2__LOD_BIAS__SHIFT 0x0
+#define SQ_IMG_SAMP_WORD2__LOD_BIAS_SEC_MASK 0xfc000
+#define SQ_IMG_SAMP_WORD2__LOD_BIAS_SEC__SHIFT 0xe
+#define SQ_IMG_SAMP_WORD2__XY_MAG_FILTER_MASK 0x300000
+#define SQ_IMG_SAMP_WORD2__XY_MAG_FILTER__SHIFT 0x14
+#define SQ_IMG_SAMP_WORD2__XY_MIN_FILTER_MASK 0xc00000
+#define SQ_IMG_SAMP_WORD2__XY_MIN_FILTER__SHIFT 0x16
+#define SQ_IMG_SAMP_WORD2__Z_FILTER_MASK 0x3000000
+#define SQ_IMG_SAMP_WORD2__Z_FILTER__SHIFT 0x18
+#define SQ_IMG_SAMP_WORD2__MIP_FILTER_MASK 0xc000000
+#define SQ_IMG_SAMP_WORD2__MIP_FILTER__SHIFT 0x1a
+#define SQ_IMG_SAMP_WORD2__MIP_POINT_PRECLAMP_MASK 0x10000000
+#define SQ_IMG_SAMP_WORD2__MIP_POINT_PRECLAMP__SHIFT 0x1c
+#define SQ_IMG_SAMP_WORD2__DISABLE_LSB_CEIL_MASK 0x20000000
+#define SQ_IMG_SAMP_WORD2__DISABLE_LSB_CEIL__SHIFT 0x1d
+#define SQ_IMG_SAMP_WORD2__FILTER_PREC_FIX_MASK 0x40000000
+#define SQ_IMG_SAMP_WORD2__FILTER_PREC_FIX__SHIFT 0x1e
+#define SQ_IMG_SAMP_WORD2__ANISO_OVERRIDE_MASK 0x80000000
+#define SQ_IMG_SAMP_WORD2__ANISO_OVERRIDE__SHIFT 0x1f
+#define SQ_IMG_SAMP_WORD3__BORDER_COLOR_PTR_MASK 0xfff
+#define SQ_IMG_SAMP_WORD3__BORDER_COLOR_PTR__SHIFT 0x0
+#define SQ_IMG_SAMP_WORD3__BORDER_COLOR_TYPE_MASK 0xc0000000
+#define SQ_IMG_SAMP_WORD3__BORDER_COLOR_TYPE__SHIFT 0x1e
+#define SQ_FLAT_SCRATCH_WORD0__SIZE_MASK 0x7ffff
+#define SQ_FLAT_SCRATCH_WORD0__SIZE__SHIFT 0x0
+#define SQ_FLAT_SCRATCH_WORD1__OFFSET_MASK 0xffffff
+#define SQ_FLAT_SCRATCH_WORD1__OFFSET__SHIFT 0x0
+#define SQ_M0_GPR_IDX_WORD__INDEX_MASK 0xff
+#define SQ_M0_GPR_IDX_WORD__INDEX__SHIFT 0x0
+#define SQ_M0_GPR_IDX_WORD__VSRC0_REL_MASK 0x1000
+#define SQ_M0_GPR_IDX_WORD__VSRC0_REL__SHIFT 0xc
+#define SQ_M0_GPR_IDX_WORD__VSRC1_REL_MASK 0x2000
+#define SQ_M0_GPR_IDX_WORD__VSRC1_REL__SHIFT 0xd
+#define SQ_M0_GPR_IDX_WORD__VSRC2_REL_MASK 0x4000
+#define SQ_M0_GPR_IDX_WORD__VSRC2_REL__SHIFT 0xe
+#define SQ_M0_GPR_IDX_WORD__VDST_REL_MASK 0x8000
+#define SQ_M0_GPR_IDX_WORD__VDST_REL__SHIFT 0xf
+#define SQ_IND_INDEX__WAVE_ID_MASK 0xf
+#define SQ_IND_INDEX__WAVE_ID__SHIFT 0x0
+#define SQ_IND_INDEX__SIMD_ID_MASK 0x30
+#define SQ_IND_INDEX__SIMD_ID__SHIFT 0x4
+#define SQ_IND_INDEX__THREAD_ID_MASK 0xfc0
+#define SQ_IND_INDEX__THREAD_ID__SHIFT 0x6
+#define SQ_IND_INDEX__AUTO_INCR_MASK 0x1000
+#define SQ_IND_INDEX__AUTO_INCR__SHIFT 0xc
+#define SQ_IND_INDEX__FORCE_READ_MASK 0x2000
+#define SQ_IND_INDEX__FORCE_READ__SHIFT 0xd
+#define SQ_IND_INDEX__READ_TIMEOUT_MASK 0x4000
+#define SQ_IND_INDEX__READ_TIMEOUT__SHIFT 0xe
+#define SQ_IND_INDEX__UNINDEXED_MASK 0x8000
+#define SQ_IND_INDEX__UNINDEXED__SHIFT 0xf
+#define SQ_IND_INDEX__INDEX_MASK 0xffff0000
+#define SQ_IND_INDEX__INDEX__SHIFT 0x10
+#define SQ_CMD__CMD_MASK 0x7
+#define SQ_CMD__CMD__SHIFT 0x0
+#define SQ_CMD__MODE_MASK 0x70
+#define SQ_CMD__MODE__SHIFT 0x4
+#define SQ_CMD__CHECK_VMID_MASK 0x80
+#define SQ_CMD__CHECK_VMID__SHIFT 0x7
+#define SQ_CMD__DATA_MASK 0x700
+#define SQ_CMD__DATA__SHIFT 0x8
+#define SQ_CMD__WAVE_ID_MASK 0xf0000
+#define SQ_CMD__WAVE_ID__SHIFT 0x10
+#define SQ_CMD__SIMD_ID_MASK 0x300000
+#define SQ_CMD__SIMD_ID__SHIFT 0x14
+#define SQ_CMD__QUEUE_ID_MASK 0x7000000
+#define SQ_CMD__QUEUE_ID__SHIFT 0x18
+#define SQ_CMD__VM_ID_MASK 0xf0000000
+#define SQ_CMD__VM_ID__SHIFT 0x1c
+#define SQ_IND_DATA__DATA_MASK 0xffffffff
+#define SQ_IND_DATA__DATA__SHIFT 0x0
+#define SQ_REG_TIMESTAMP__TIMESTAMP_MASK 0xff
+#define SQ_REG_TIMESTAMP__TIMESTAMP__SHIFT 0x0
+#define SQ_CMD_TIMESTAMP__TIMESTAMP_MASK 0xff
+#define SQ_CMD_TIMESTAMP__TIMESTAMP__SHIFT 0x0
+#define SQ_HV_VMID_CTRL__DEFAULT_VMID_MASK 0xf
+#define SQ_HV_VMID_CTRL__DEFAULT_VMID__SHIFT 0x0
+#define SQ_HV_VMID_CTRL__ALLOWED_VMID_MASK_MASK 0xffff0
+#define SQ_HV_VMID_CTRL__ALLOWED_VMID_MASK__SHIFT 0x4
+#define SQ_WAVE_INST_DW0__INST_DW0_MASK 0xffffffff
+#define SQ_WAVE_INST_DW0__INST_DW0__SHIFT 0x0
+#define SQ_WAVE_INST_DW1__INST_DW1_MASK 0xffffffff
+#define SQ_WAVE_INST_DW1__INST_DW1__SHIFT 0x0
+#define SQ_WAVE_PC_LO__PC_LO_MASK 0xffffffff
+#define SQ_WAVE_PC_LO__PC_LO__SHIFT 0x0
+#define SQ_WAVE_PC_HI__PC_HI_MASK 0xffff
+#define SQ_WAVE_PC_HI__PC_HI__SHIFT 0x0
+#define SQ_WAVE_IB_DBG0__IBUF_ST_MASK 0x7
+#define SQ_WAVE_IB_DBG0__IBUF_ST__SHIFT 0x0
+#define SQ_WAVE_IB_DBG0__PC_INVALID_MASK 0x8
+#define SQ_WAVE_IB_DBG0__PC_INVALID__SHIFT 0x3
+#define SQ_WAVE_IB_DBG0__NEED_NEXT_DW_MASK 0x10
+#define SQ_WAVE_IB_DBG0__NEED_NEXT_DW__SHIFT 0x4
+#define SQ_WAVE_IB_DBG0__NO_PREFETCH_CNT_MASK 0xe0
+#define SQ_WAVE_IB_DBG0__NO_PREFETCH_CNT__SHIFT 0x5
+#define SQ_WAVE_IB_DBG0__IBUF_RPTR_MASK 0x300
+#define SQ_WAVE_IB_DBG0__IBUF_RPTR__SHIFT 0x8
+#define SQ_WAVE_IB_DBG0__IBUF_WPTR_MASK 0xc00
+#define SQ_WAVE_IB_DBG0__IBUF_WPTR__SHIFT 0xa
+#define SQ_WAVE_IB_DBG0__INST_STR_ST_MASK 0xf0000
+#define SQ_WAVE_IB_DBG0__INST_STR_ST__SHIFT 0x10
+#define SQ_WAVE_IB_DBG0__MISC_CNT_MASK 0xf00000
+#define SQ_WAVE_IB_DBG0__MISC_CNT__SHIFT 0x14
+#define SQ_WAVE_IB_DBG0__ECC_ST_MASK 0x3000000
+#define SQ_WAVE_IB_DBG0__ECC_ST__SHIFT 0x18
+#define SQ_WAVE_IB_DBG0__IS_HYB_MASK 0x4000000
+#define SQ_WAVE_IB_DBG0__IS_HYB__SHIFT 0x1a
+#define SQ_WAVE_IB_DBG0__HYB_CNT_MASK 0x18000000
+#define SQ_WAVE_IB_DBG0__HYB_CNT__SHIFT 0x1b
+#define SQ_WAVE_IB_DBG0__KILL_MASK 0x20000000
+#define SQ_WAVE_IB_DBG0__KILL__SHIFT 0x1d
+#define SQ_WAVE_IB_DBG0__NEED_KILL_IFETCH_MASK 0x40000000
+#define SQ_WAVE_IB_DBG0__NEED_KILL_IFETCH__SHIFT 0x1e
+#define SQ_WAVE_IB_DBG1__IXNACK_MASK 0x1
+#define SQ_WAVE_IB_DBG1__IXNACK__SHIFT 0x0
+#define SQ_WAVE_IB_DBG1__XNACK_MASK 0x2
+#define SQ_WAVE_IB_DBG1__XNACK__SHIFT 0x1
+#define SQ_WAVE_IB_DBG1__TA_NEED_RESET_MASK 0x4
+#define SQ_WAVE_IB_DBG1__TA_NEED_RESET__SHIFT 0x2
+#define SQ_WAVE_IB_DBG1__XCNT_MASK 0xf0
+#define SQ_WAVE_IB_DBG1__XCNT__SHIFT 0x4
+#define SQ_WAVE_IB_DBG1__QCNT_MASK 0xf00
+#define SQ_WAVE_IB_DBG1__QCNT__SHIFT 0x8
+#define SQ_WAVE_EXEC_LO__EXEC_LO_MASK 0xffffffff
+#define SQ_WAVE_EXEC_LO__EXEC_LO__SHIFT 0x0
+#define SQ_WAVE_EXEC_HI__EXEC_HI_MASK 0xffffffff
+#define SQ_WAVE_EXEC_HI__EXEC_HI__SHIFT 0x0
+#define SQ_WAVE_STATUS__SCC_MASK 0x1
+#define SQ_WAVE_STATUS__SCC__SHIFT 0x0
+#define SQ_WAVE_STATUS__SPI_PRIO_MASK 0x6
+#define SQ_WAVE_STATUS__SPI_PRIO__SHIFT 0x1
+#define SQ_WAVE_STATUS__USER_PRIO_MASK 0x18
+#define SQ_WAVE_STATUS__USER_PRIO__SHIFT 0x3
+#define SQ_WAVE_STATUS__PRIV_MASK 0x20
+#define SQ_WAVE_STATUS__PRIV__SHIFT 0x5
+#define SQ_WAVE_STATUS__TRAP_EN_MASK 0x40
+#define SQ_WAVE_STATUS__TRAP_EN__SHIFT 0x6
+#define SQ_WAVE_STATUS__TTRACE_EN_MASK 0x80
+#define SQ_WAVE_STATUS__TTRACE_EN__SHIFT 0x7
+#define SQ_WAVE_STATUS__EXPORT_RDY_MASK 0x100
+#define SQ_WAVE_STATUS__EXPORT_RDY__SHIFT 0x8
+#define SQ_WAVE_STATUS__EXECZ_MASK 0x200
+#define SQ_WAVE_STATUS__EXECZ__SHIFT 0x9
+#define SQ_WAVE_STATUS__VCCZ_MASK 0x400
+#define SQ_WAVE_STATUS__VCCZ__SHIFT 0xa
+#define SQ_WAVE_STATUS__IN_TG_MASK 0x800
+#define SQ_WAVE_STATUS__IN_TG__SHIFT 0xb
+#define SQ_WAVE_STATUS__IN_BARRIER_MASK 0x1000
+#define SQ_WAVE_STATUS__IN_BARRIER__SHIFT 0xc
+#define SQ_WAVE_STATUS__HALT_MASK 0x2000
+#define SQ_WAVE_STATUS__HALT__SHIFT 0xd
+#define SQ_WAVE_STATUS__TRAP_MASK 0x4000
+#define SQ_WAVE_STATUS__TRAP__SHIFT 0xe
+#define SQ_WAVE_STATUS__TTRACE_CU_EN_MASK 0x8000
+#define SQ_WAVE_STATUS__TTRACE_CU_EN__SHIFT 0xf
+#define SQ_WAVE_STATUS__VALID_MASK 0x10000
+#define SQ_WAVE_STATUS__VALID__SHIFT 0x10
+#define SQ_WAVE_STATUS__ECC_ERR_MASK 0x20000
+#define SQ_WAVE_STATUS__ECC_ERR__SHIFT 0x11
+#define SQ_WAVE_STATUS__SKIP_EXPORT_MASK 0x40000
+#define SQ_WAVE_STATUS__SKIP_EXPORT__SHIFT 0x12
+#define SQ_WAVE_STATUS__PERF_EN_MASK 0x80000
+#define SQ_WAVE_STATUS__PERF_EN__SHIFT 0x13
+#define SQ_WAVE_STATUS__COND_DBG_USER_MASK 0x100000
+#define SQ_WAVE_STATUS__COND_DBG_USER__SHIFT 0x14
+#define SQ_WAVE_STATUS__COND_DBG_SYS_MASK 0x200000
+#define SQ_WAVE_STATUS__COND_DBG_SYS__SHIFT 0x15
+#define SQ_WAVE_STATUS__ALLOW_REPLAY_MASK 0x400000
+#define SQ_WAVE_STATUS__ALLOW_REPLAY__SHIFT 0x16
+#define SQ_WAVE_STATUS__INST_ATC_MASK 0x800000
+#define SQ_WAVE_STATUS__INST_ATC__SHIFT 0x17
+#define SQ_WAVE_STATUS__MUST_EXPORT_MASK 0x8000000
+#define SQ_WAVE_STATUS__MUST_EXPORT__SHIFT 0x1b
+#define SQ_WAVE_MODE__FP_ROUND_MASK 0xf
+#define SQ_WAVE_MODE__FP_ROUND__SHIFT 0x0
+#define SQ_WAVE_MODE__FP_DENORM_MASK 0xf0
+#define SQ_WAVE_MODE__FP_DENORM__SHIFT 0x4
+#define SQ_WAVE_MODE__DX10_CLAMP_MASK 0x100
+#define SQ_WAVE_MODE__DX10_CLAMP__SHIFT 0x8
+#define SQ_WAVE_MODE__IEEE_MASK 0x200
+#define SQ_WAVE_MODE__IEEE__SHIFT 0x9
+#define SQ_WAVE_MODE__LOD_CLAMPED_MASK 0x400
+#define SQ_WAVE_MODE__LOD_CLAMPED__SHIFT 0xa
+#define SQ_WAVE_MODE__DEBUG_EN_MASK 0x800
+#define SQ_WAVE_MODE__DEBUG_EN__SHIFT 0xb
+#define SQ_WAVE_MODE__EXCP_EN_MASK 0x1ff000
+#define SQ_WAVE_MODE__EXCP_EN__SHIFT 0xc
+#define SQ_WAVE_MODE__GPR_IDX_EN_MASK 0x8000000
+#define SQ_WAVE_MODE__GPR_IDX_EN__SHIFT 0x1b
+#define SQ_WAVE_MODE__VSKIP_MASK 0x10000000
+#define SQ_WAVE_MODE__VSKIP__SHIFT 0x1c
+#define SQ_WAVE_MODE__CSP_MASK 0xe0000000
+#define SQ_WAVE_MODE__CSP__SHIFT 0x1d
+#define SQ_WAVE_TRAPSTS__EXCP_MASK 0x1ff
+#define SQ_WAVE_TRAPSTS__EXCP__SHIFT 0x0
+#define SQ_WAVE_TRAPSTS__SAVECTX_MASK 0x400
+#define SQ_WAVE_TRAPSTS__SAVECTX__SHIFT 0xa
+#define SQ_WAVE_TRAPSTS__EXCP_CYCLE_MASK 0x3f0000
+#define SQ_WAVE_TRAPSTS__EXCP_CYCLE__SHIFT 0x10
+#define SQ_WAVE_TRAPSTS__DP_RATE_MASK 0xe0000000
+#define SQ_WAVE_TRAPSTS__DP_RATE__SHIFT 0x1d
+#define SQ_WAVE_HW_ID__WAVE_ID_MASK 0xf
+#define SQ_WAVE_HW_ID__WAVE_ID__SHIFT 0x0
+#define SQ_WAVE_HW_ID__SIMD_ID_MASK 0x30
+#define SQ_WAVE_HW_ID__SIMD_ID__SHIFT 0x4
+#define SQ_WAVE_HW_ID__PIPE_ID_MASK 0xc0
+#define SQ_WAVE_HW_ID__PIPE_ID__SHIFT 0x6
+#define SQ_WAVE_HW_ID__CU_ID_MASK 0xf00
+#define SQ_WAVE_HW_ID__CU_ID__SHIFT 0x8
+#define SQ_WAVE_HW_ID__SH_ID_MASK 0x1000
+#define SQ_WAVE_HW_ID__SH_ID__SHIFT 0xc
+#define SQ_WAVE_HW_ID__SE_ID_MASK 0x6000
+#define SQ_WAVE_HW_ID__SE_ID__SHIFT 0xd
+#define SQ_WAVE_HW_ID__TG_ID_MASK 0xf0000
+#define SQ_WAVE_HW_ID__TG_ID__SHIFT 0x10
+#define SQ_WAVE_HW_ID__VM_ID_MASK 0xf00000
+#define SQ_WAVE_HW_ID__VM_ID__SHIFT 0x14
+#define SQ_WAVE_HW_ID__QUEUE_ID_MASK 0x7000000
+#define SQ_WAVE_HW_ID__QUEUE_ID__SHIFT 0x18
+#define SQ_WAVE_HW_ID__STATE_ID_MASK 0x38000000
+#define SQ_WAVE_HW_ID__STATE_ID__SHIFT 0x1b
+#define SQ_WAVE_HW_ID__ME_ID_MASK 0xc0000000
+#define SQ_WAVE_HW_ID__ME_ID__SHIFT 0x1e
+#define SQ_WAVE_GPR_ALLOC__VGPR_BASE_MASK 0x3f
+#define SQ_WAVE_GPR_ALLOC__VGPR_BASE__SHIFT 0x0
+#define SQ_WAVE_GPR_ALLOC__VGPR_SIZE_MASK 0x3f00
+#define SQ_WAVE_GPR_ALLOC__VGPR_SIZE__SHIFT 0x8
+#define SQ_WAVE_GPR_ALLOC__SGPR_BASE_MASK 0x3f0000
+#define SQ_WAVE_GPR_ALLOC__SGPR_BASE__SHIFT 0x10
+#define SQ_WAVE_GPR_ALLOC__SGPR_SIZE_MASK 0xf000000
+#define SQ_WAVE_GPR_ALLOC__SGPR_SIZE__SHIFT 0x18
+#define SQ_WAVE_LDS_ALLOC__LDS_BASE_MASK 0xff
+#define SQ_WAVE_LDS_ALLOC__LDS_BASE__SHIFT 0x0
+#define SQ_WAVE_LDS_ALLOC__LDS_SIZE_MASK 0x1ff000
+#define SQ_WAVE_LDS_ALLOC__LDS_SIZE__SHIFT 0xc
+#define SQ_WAVE_IB_STS__VM_CNT_MASK 0xf
+#define SQ_WAVE_IB_STS__VM_CNT__SHIFT 0x0
+#define SQ_WAVE_IB_STS__EXP_CNT_MASK 0x70
+#define SQ_WAVE_IB_STS__EXP_CNT__SHIFT 0x4
+#define SQ_WAVE_IB_STS__LGKM_CNT_MASK 0xf00
+#define SQ_WAVE_IB_STS__LGKM_CNT__SHIFT 0x8
+#define SQ_WAVE_IB_STS__VALU_CNT_MASK 0x7000
+#define SQ_WAVE_IB_STS__VALU_CNT__SHIFT 0xc
+#define SQ_WAVE_IB_STS__FIRST_REPLAY_MASK 0x8000
+#define SQ_WAVE_IB_STS__FIRST_REPLAY__SHIFT 0xf
+#define SQ_WAVE_IB_STS__RCNT_MASK 0xf0000
+#define SQ_WAVE_IB_STS__RCNT__SHIFT 0x10
+#define SQ_WAVE_M0__M0_MASK 0xffffffff
+#define SQ_WAVE_M0__M0__SHIFT 0x0
+#define SQ_WAVE_TBA_LO__ADDR_LO_MASK 0xffffffff
+#define SQ_WAVE_TBA_LO__ADDR_LO__SHIFT 0x0
+#define SQ_WAVE_TBA_HI__ADDR_HI_MASK 0xff
+#define SQ_WAVE_TBA_HI__ADDR_HI__SHIFT 0x0
+#define SQ_WAVE_TMA_LO__ADDR_LO_MASK 0xffffffff
+#define SQ_WAVE_TMA_LO__ADDR_LO__SHIFT 0x0
+#define SQ_WAVE_TMA_HI__ADDR_HI_MASK 0xff
+#define SQ_WAVE_TMA_HI__ADDR_HI__SHIFT 0x0
+#define SQ_WAVE_TTMP0__DATA_MASK 0xffffffff
+#define SQ_WAVE_TTMP0__DATA__SHIFT 0x0
+#define SQ_WAVE_TTMP1__DATA_MASK 0xffffffff
+#define SQ_WAVE_TTMP1__DATA__SHIFT 0x0
+#define SQ_WAVE_TTMP2__DATA_MASK 0xffffffff
+#define SQ_WAVE_TTMP2__DATA__SHIFT 0x0
+#define SQ_WAVE_TTMP3__DATA_MASK 0xffffffff
+#define SQ_WAVE_TTMP3__DATA__SHIFT 0x0
+#define SQ_WAVE_TTMP4__DATA_MASK 0xffffffff
+#define SQ_WAVE_TTMP4__DATA__SHIFT 0x0
+#define SQ_WAVE_TTMP5__DATA_MASK 0xffffffff
+#define SQ_WAVE_TTMP5__DATA__SHIFT 0x0
+#define SQ_WAVE_TTMP6__DATA_MASK 0xffffffff
+#define SQ_WAVE_TTMP6__DATA__SHIFT 0x0
+#define SQ_WAVE_TTMP7__DATA_MASK 0xffffffff
+#define SQ_WAVE_TTMP7__DATA__SHIFT 0x0
+#define SQ_WAVE_TTMP8__DATA_MASK 0xffffffff
+#define SQ_WAVE_TTMP8__DATA__SHIFT 0x0
+#define SQ_WAVE_TTMP9__DATA_MASK 0xffffffff
+#define SQ_WAVE_TTMP9__DATA__SHIFT 0x0
+#define SQ_WAVE_TTMP10__DATA_MASK 0xffffffff
+#define SQ_WAVE_TTMP10__DATA__SHIFT 0x0
+#define SQ_WAVE_TTMP11__DATA_MASK 0xffffffff
+#define SQ_WAVE_TTMP11__DATA__SHIFT 0x0
+#define SQ_DEBUG_STS_GLOBAL__BUSY_MASK 0x1
+#define SQ_DEBUG_STS_GLOBAL__BUSY__SHIFT 0x0
+#define SQ_DEBUG_STS_GLOBAL__INTERRUPT_MSG_BUSY_MASK 0x2
+#define SQ_DEBUG_STS_GLOBAL__INTERRUPT_MSG_BUSY__SHIFT 0x1
+#define SQ_DEBUG_STS_GLOBAL__WAVE_LEVEL_SH0_MASK 0xfff0
+#define SQ_DEBUG_STS_GLOBAL__WAVE_LEVEL_SH0__SHIFT 0x4
+#define SQ_DEBUG_STS_GLOBAL__WAVE_LEVEL_SH1_MASK 0xfff0000
+#define SQ_DEBUG_STS_GLOBAL__WAVE_LEVEL_SH1__SHIFT 0x10
+#define SQ_DEBUG_STS_GLOBAL2__FIFO_LEVEL_GFX0_MASK 0xff
+#define SQ_DEBUG_STS_GLOBAL2__FIFO_LEVEL_GFX0__SHIFT 0x0
+#define SQ_DEBUG_STS_GLOBAL2__FIFO_LEVEL_GFX1_MASK 0xff00
+#define SQ_DEBUG_STS_GLOBAL2__FIFO_LEVEL_GFX1__SHIFT 0x8
+#define SQ_DEBUG_STS_GLOBAL2__FIFO_LEVEL_IMMED_MASK 0xff0000
+#define SQ_DEBUG_STS_GLOBAL2__FIFO_LEVEL_IMMED__SHIFT 0x10
+#define SQ_DEBUG_STS_GLOBAL2__FIFO_LEVEL_HOST_MASK 0xff000000
+#define SQ_DEBUG_STS_GLOBAL2__FIFO_LEVEL_HOST__SHIFT 0x18
+#define SQ_DEBUG_STS_GLOBAL3__FIFO_LEVEL_HOST_CMD_MASK 0xf
+#define SQ_DEBUG_STS_GLOBAL3__FIFO_LEVEL_HOST_CMD__SHIFT 0x0
+#define SQ_DEBUG_STS_GLOBAL3__FIFO_LEVEL_HOST_REG_MASK 0x3f0
+#define SQ_DEBUG_STS_GLOBAL3__FIFO_LEVEL_HOST_REG__SHIFT 0x4
+#define SQ_DEBUG_STS_LOCAL__BUSY_MASK 0x1
+#define SQ_DEBUG_STS_LOCAL__BUSY__SHIFT 0x0
+#define SQ_DEBUG_STS_LOCAL__WAVE_LEVEL_MASK 0x3f0
+#define SQ_DEBUG_STS_LOCAL__WAVE_LEVEL__SHIFT 0x4
+#define SQ_DEBUG_CTRL_LOCAL__UNUSED_MASK 0xff
+#define SQ_DEBUG_CTRL_LOCAL__UNUSED__SHIFT 0x0
+#define SH_MEM_BASES__PRIVATE_BASE_MASK 0xffff
+#define SH_MEM_BASES__PRIVATE_BASE__SHIFT 0x0
+#define SH_MEM_BASES__SHARED_BASE_MASK 0xffff0000
+#define SH_MEM_BASES__SHARED_BASE__SHIFT 0x10
+#define SH_MEM_APE1_BASE__BASE_MASK 0xffffffff
+#define SH_MEM_APE1_BASE__BASE__SHIFT 0x0
+#define SH_MEM_APE1_LIMIT__LIMIT_MASK 0xffffffff
+#define SH_MEM_APE1_LIMIT__LIMIT__SHIFT 0x0
+#define SH_MEM_CONFIG__ADDRESS_MODE_MASK 0x3
+#define SH_MEM_CONFIG__ADDRESS_MODE__SHIFT 0x0
+#define SH_MEM_CONFIG__PRIVATE_ATC_MASK 0x4
+#define SH_MEM_CONFIG__PRIVATE_ATC__SHIFT 0x2
+#define SH_MEM_CONFIG__ALIGNMENT_MODE_MASK 0x18
+#define SH_MEM_CONFIG__ALIGNMENT_MODE__SHIFT 0x3
+#define SH_MEM_CONFIG__DEFAULT_MTYPE_MASK 0xe0
+#define SH_MEM_CONFIG__DEFAULT_MTYPE__SHIFT 0x5
+#define SH_MEM_CONFIG__APE1_MTYPE_MASK 0x700
+#define SH_MEM_CONFIG__APE1_MTYPE__SHIFT 0x8
+#define SH_MEM_CONFIG__APE1_ATC_MASK 0x800
+#define SH_MEM_CONFIG__APE1_ATC__SHIFT 0xb
+#define SQ_THREAD_TRACE_WORD_CMN__TOKEN_TYPE_MASK 0xf
+#define SQ_THREAD_TRACE_WORD_CMN__TOKEN_TYPE__SHIFT 0x0
+#define SQ_THREAD_TRACE_WORD_CMN__TIME_DELTA_MASK 0x10
+#define SQ_THREAD_TRACE_WORD_CMN__TIME_DELTA__SHIFT 0x4
+#define SQ_THREAD_TRACE_WORD_INST__TOKEN_TYPE_MASK 0xf
+#define SQ_THREAD_TRACE_WORD_INST__TOKEN_TYPE__SHIFT 0x0
+#define SQ_THREAD_TRACE_WORD_INST__TIME_DELTA_MASK 0x10
+#define SQ_THREAD_TRACE_WORD_INST__TIME_DELTA__SHIFT 0x4
+#define SQ_THREAD_TRACE_WORD_INST__WAVE_ID_MASK 0x1e0
+#define SQ_THREAD_TRACE_WORD_INST__WAVE_ID__SHIFT 0x5
+#define SQ_THREAD_TRACE_WORD_INST__SIMD_ID_MASK 0x600
+#define SQ_THREAD_TRACE_WORD_INST__SIMD_ID__SHIFT 0x9
+#define SQ_THREAD_TRACE_WORD_INST__INST_TYPE_MASK 0xf800
+#define SQ_THREAD_TRACE_WORD_INST__INST_TYPE__SHIFT 0xb
+#define SQ_THREAD_TRACE_WORD_INST_PC_1_OF_2__TOKEN_TYPE_MASK 0xf
+#define SQ_THREAD_TRACE_WORD_INST_PC_1_OF_2__TOKEN_TYPE__SHIFT 0x0
+#define SQ_THREAD_TRACE_WORD_INST_PC_1_OF_2__TIME_DELTA_MASK 0x10
+#define SQ_THREAD_TRACE_WORD_INST_PC_1_OF_2__TIME_DELTA__SHIFT 0x4
+#define SQ_THREAD_TRACE_WORD_INST_PC_1_OF_2__WAVE_ID_MASK 0x1e0
+#define SQ_THREAD_TRACE_WORD_INST_PC_1_OF_2__WAVE_ID__SHIFT 0x5
+#define SQ_THREAD_TRACE_WORD_INST_PC_1_OF_2__SIMD_ID_MASK 0x600
+#define SQ_THREAD_TRACE_WORD_INST_PC_1_OF_2__SIMD_ID__SHIFT 0x9
+#define SQ_THREAD_TRACE_WORD_INST_PC_1_OF_2__PC_LO_MASK 0xffff0000
+#define SQ_THREAD_TRACE_WORD_INST_PC_1_OF_2__PC_LO__SHIFT 0x10
+#define SQ_THREAD_TRACE_WORD_INST_PC_2_OF_2__PC_HI_MASK 0xffffff
+#define SQ_THREAD_TRACE_WORD_INST_PC_2_OF_2__PC_HI__SHIFT 0x0
+#define SQ_THREAD_TRACE_WORD_INST_USERDATA_1_OF_2__TOKEN_TYPE_MASK 0xf
+#define SQ_THREAD_TRACE_WORD_INST_USERDATA_1_OF_2__TOKEN_TYPE__SHIFT 0x0
+#define SQ_THREAD_TRACE_WORD_INST_USERDATA_1_OF_2__TIME_DELTA_MASK 0x10
+#define SQ_THREAD_TRACE_WORD_INST_USERDATA_1_OF_2__TIME_DELTA__SHIFT 0x4
+#define SQ_THREAD_TRACE_WORD_INST_USERDATA_1_OF_2__SH_ID_MASK 0x20
+#define SQ_THREAD_TRACE_WORD_INST_USERDATA_1_OF_2__SH_ID__SHIFT 0x5
+#define SQ_THREAD_TRACE_WORD_INST_USERDATA_1_OF_2__CU_ID_MASK 0x3c0
+#define SQ_THREAD_TRACE_WORD_INST_USERDATA_1_OF_2__CU_ID__SHIFT 0x6
+#define SQ_THREAD_TRACE_WORD_INST_USERDATA_1_OF_2__WAVE_ID_MASK 0x3c00
+#define SQ_THREAD_TRACE_WORD_INST_USERDATA_1_OF_2__WAVE_ID__SHIFT 0xa
+#define SQ_THREAD_TRACE_WORD_INST_USERDATA_1_OF_2__SIMD_ID_MASK 0xc000
+#define SQ_THREAD_TRACE_WORD_INST_USERDATA_1_OF_2__SIMD_ID__SHIFT 0xe
+#define SQ_THREAD_TRACE_WORD_INST_USERDATA_1_OF_2__DATA_LO_MASK 0xffff0000
+#define SQ_THREAD_TRACE_WORD_INST_USERDATA_1_OF_2__DATA_LO__SHIFT 0x10
+#define SQ_THREAD_TRACE_WORD_INST_USERDATA_2_OF_2__DATA_HI_MASK 0xffff
+#define SQ_THREAD_TRACE_WORD_INST_USERDATA_2_OF_2__DATA_HI__SHIFT 0x0
+#define SQ_THREAD_TRACE_WORD_TIMESTAMP_1_OF_2__TOKEN_TYPE_MASK 0xf
+#define SQ_THREAD_TRACE_WORD_TIMESTAMP_1_OF_2__TOKEN_TYPE__SHIFT 0x0
+#define SQ_THREAD_TRACE_WORD_TIMESTAMP_1_OF_2__TIME_LO_MASK 0xffff0000
+#define SQ_THREAD_TRACE_WORD_TIMESTAMP_1_OF_2__TIME_LO__SHIFT 0x10
+#define SQ_THREAD_TRACE_WORD_TIMESTAMP_2_OF_2__TIME_HI_MASK 0xffffffff
+#define SQ_THREAD_TRACE_WORD_TIMESTAMP_2_OF_2__TIME_HI__SHIFT 0x0
+#define SQ_THREAD_TRACE_WORD_WAVE__TOKEN_TYPE_MASK 0xf
+#define SQ_THREAD_TRACE_WORD_WAVE__TOKEN_TYPE__SHIFT 0x0
+#define SQ_THREAD_TRACE_WORD_WAVE__TIME_DELTA_MASK 0x10
+#define SQ_THREAD_TRACE_WORD_WAVE__TIME_DELTA__SHIFT 0x4
+#define SQ_THREAD_TRACE_WORD_WAVE__SH_ID_MASK 0x20
+#define SQ_THREAD_TRACE_WORD_WAVE__SH_ID__SHIFT 0x5
+#define SQ_THREAD_TRACE_WORD_WAVE__CU_ID_MASK 0x3c0
+#define SQ_THREAD_TRACE_WORD_WAVE__CU_ID__SHIFT 0x6
+#define SQ_THREAD_TRACE_WORD_WAVE__WAVE_ID_MASK 0x3c00
+#define SQ_THREAD_TRACE_WORD_WAVE__WAVE_ID__SHIFT 0xa
+#define SQ_THREAD_TRACE_WORD_WAVE__SIMD_ID_MASK 0xc000
+#define SQ_THREAD_TRACE_WORD_WAVE__SIMD_ID__SHIFT 0xe
+#define SQ_THREAD_TRACE_WORD_MISC__TOKEN_TYPE_MASK 0xf
+#define SQ_THREAD_TRACE_WORD_MISC__TOKEN_TYPE__SHIFT 0x0
+#define SQ_THREAD_TRACE_WORD_MISC__TIME_DELTA_MASK 0xff0
+#define SQ_THREAD_TRACE_WORD_MISC__TIME_DELTA__SHIFT 0x4
+#define SQ_THREAD_TRACE_WORD_MISC__SH_ID_MASK 0x1000
+#define SQ_THREAD_TRACE_WORD_MISC__SH_ID__SHIFT 0xc
+#define SQ_THREAD_TRACE_WORD_MISC__MISC_TOKEN_TYPE_MASK 0xe000
+#define SQ_THREAD_TRACE_WORD_MISC__MISC_TOKEN_TYPE__SHIFT 0xd
+#define SQ_THREAD_TRACE_WORD_WAVE_START__TOKEN_TYPE_MASK 0xf
+#define SQ_THREAD_TRACE_WORD_WAVE_START__TOKEN_TYPE__SHIFT 0x0
+#define SQ_THREAD_TRACE_WORD_WAVE_START__TIME_DELTA_MASK 0x10
+#define SQ_THREAD_TRACE_WORD_WAVE_START__TIME_DELTA__SHIFT 0x4
+#define SQ_THREAD_TRACE_WORD_WAVE_START__SH_ID_MASK 0x20
+#define SQ_THREAD_TRACE_WORD_WAVE_START__SH_ID__SHIFT 0x5
+#define SQ_THREAD_TRACE_WORD_WAVE_START__CU_ID_MASK 0x3c0
+#define SQ_THREAD_TRACE_WORD_WAVE_START__CU_ID__SHIFT 0x6
+#define SQ_THREAD_TRACE_WORD_WAVE_START__WAVE_ID_MASK 0x3c00
+#define SQ_THREAD_TRACE_WORD_WAVE_START__WAVE_ID__SHIFT 0xa
+#define SQ_THREAD_TRACE_WORD_WAVE_START__SIMD_ID_MASK 0xc000
+#define SQ_THREAD_TRACE_WORD_WAVE_START__SIMD_ID__SHIFT 0xe
+#define SQ_THREAD_TRACE_WORD_WAVE_START__DISPATCHER_MASK 0x1f0000
+#define SQ_THREAD_TRACE_WORD_WAVE_START__DISPATCHER__SHIFT 0x10
+#define SQ_THREAD_TRACE_WORD_WAVE_START__VS_NO_ALLOC_OR_GROUPED_MASK 0x200000
+#define SQ_THREAD_TRACE_WORD_WAVE_START__VS_NO_ALLOC_OR_GROUPED__SHIFT 0x15
+#define SQ_THREAD_TRACE_WORD_WAVE_START__COUNT_MASK 0x1fc00000
+#define SQ_THREAD_TRACE_WORD_WAVE_START__COUNT__SHIFT 0x16
+#define SQ_THREAD_TRACE_WORD_WAVE_START__TG_ID_MASK 0xe0000000
+#define SQ_THREAD_TRACE_WORD_WAVE_START__TG_ID__SHIFT 0x1d
+#define SQ_THREAD_TRACE_WORD_REG_1_OF_2__TOKEN_TYPE_MASK 0xf
+#define SQ_THREAD_TRACE_WORD_REG_1_OF_2__TOKEN_TYPE__SHIFT 0x0
+#define SQ_THREAD_TRACE_WORD_REG_1_OF_2__TIME_DELTA_MASK 0x10
+#define SQ_THREAD_TRACE_WORD_REG_1_OF_2__TIME_DELTA__SHIFT 0x4
+#define SQ_THREAD_TRACE_WORD_REG_1_OF_2__PIPE_ID_MASK 0x60
+#define SQ_THREAD_TRACE_WORD_REG_1_OF_2__PIPE_ID__SHIFT 0x5
+#define SQ_THREAD_TRACE_WORD_REG_1_OF_2__ME_ID_MASK 0x180
+#define SQ_THREAD_TRACE_WORD_REG_1_OF_2__ME_ID__SHIFT 0x7
+#define SQ_THREAD_TRACE_WORD_REG_1_OF_2__REG_DROPPED_PREV_MASK 0x200
+#define SQ_THREAD_TRACE_WORD_REG_1_OF_2__REG_DROPPED_PREV__SHIFT 0x9
+#define SQ_THREAD_TRACE_WORD_REG_1_OF_2__REG_TYPE_MASK 0x1c00
+#define SQ_THREAD_TRACE_WORD_REG_1_OF_2__REG_TYPE__SHIFT 0xa
+#define SQ_THREAD_TRACE_WORD_REG_1_OF_2__REG_PRIV_MASK 0x4000
+#define SQ_THREAD_TRACE_WORD_REG_1_OF_2__REG_PRIV__SHIFT 0xe
+#define SQ_THREAD_TRACE_WORD_REG_1_OF_2__REG_OP_MASK 0x8000
+#define SQ_THREAD_TRACE_WORD_REG_1_OF_2__REG_OP__SHIFT 0xf
+#define SQ_THREAD_TRACE_WORD_REG_1_OF_2__REG_ADDR_MASK 0xffff0000
+#define SQ_THREAD_TRACE_WORD_REG_1_OF_2__REG_ADDR__SHIFT 0x10
+#define SQ_THREAD_TRACE_WORD_REG_2_OF_2__DATA_MASK 0xffffffff
+#define SQ_THREAD_TRACE_WORD_REG_2_OF_2__DATA__SHIFT 0x0
+#define SQ_THREAD_TRACE_WORD_REG_CS_1_OF_2__TOKEN_TYPE_MASK 0xf
+#define SQ_THREAD_TRACE_WORD_REG_CS_1_OF_2__TOKEN_TYPE__SHIFT 0x0
+#define SQ_THREAD_TRACE_WORD_REG_CS_1_OF_2__TIME_DELTA_MASK 0x10
+#define SQ_THREAD_TRACE_WORD_REG_CS_1_OF_2__TIME_DELTA__SHIFT 0x4
+#define SQ_THREAD_TRACE_WORD_REG_CS_1_OF_2__PIPE_ID_MASK 0x60
+#define SQ_THREAD_TRACE_WORD_REG_CS_1_OF_2__PIPE_ID__SHIFT 0x5
+#define SQ_THREAD_TRACE_WORD_REG_CS_1_OF_2__ME_ID_MASK 0x180
+#define SQ_THREAD_TRACE_WORD_REG_CS_1_OF_2__ME_ID__SHIFT 0x7
+#define SQ_THREAD_TRACE_WORD_REG_CS_1_OF_2__REG_ADDR_MASK 0xfe00
+#define SQ_THREAD_TRACE_WORD_REG_CS_1_OF_2__REG_ADDR__SHIFT 0x9
+#define SQ_THREAD_TRACE_WORD_REG_CS_1_OF_2__DATA_LO_MASK 0xffff0000
+#define SQ_THREAD_TRACE_WORD_REG_CS_1_OF_2__DATA_LO__SHIFT 0x10
+#define SQ_THREAD_TRACE_WORD_REG_CS_2_OF_2__DATA_HI_MASK 0xffff
+#define SQ_THREAD_TRACE_WORD_REG_CS_2_OF_2__DATA_HI__SHIFT 0x0
+#define SQ_THREAD_TRACE_WORD_EVENT__TOKEN_TYPE_MASK 0xf
+#define SQ_THREAD_TRACE_WORD_EVENT__TOKEN_TYPE__SHIFT 0x0
+#define SQ_THREAD_TRACE_WORD_EVENT__TIME_DELTA_MASK 0x10
+#define SQ_THREAD_TRACE_WORD_EVENT__TIME_DELTA__SHIFT 0x4
+#define SQ_THREAD_TRACE_WORD_EVENT__SH_ID_MASK 0x20
+#define SQ_THREAD_TRACE_WORD_EVENT__SH_ID__SHIFT 0x5
+#define SQ_THREAD_TRACE_WORD_EVENT__STAGE_MASK 0x1c0
+#define SQ_THREAD_TRACE_WORD_EVENT__STAGE__SHIFT 0x6
+#define SQ_THREAD_TRACE_WORD_EVENT__EVENT_TYPE_MASK 0xfc00
+#define SQ_THREAD_TRACE_WORD_EVENT__EVENT_TYPE__SHIFT 0xa
+#define SQ_THREAD_TRACE_WORD_ISSUE__TOKEN_TYPE_MASK 0xf
+#define SQ_THREAD_TRACE_WORD_ISSUE__TOKEN_TYPE__SHIFT 0x0
+#define SQ_THREAD_TRACE_WORD_ISSUE__TIME_DELTA_MASK 0x10
+#define SQ_THREAD_TRACE_WORD_ISSUE__TIME_DELTA__SHIFT 0x4
+#define SQ_THREAD_TRACE_WORD_ISSUE__SIMD_ID_MASK 0x60
+#define SQ_THREAD_TRACE_WORD_ISSUE__SIMD_ID__SHIFT 0x5
+#define SQ_THREAD_TRACE_WORD_ISSUE__INST0_MASK 0x300
+#define SQ_THREAD_TRACE_WORD_ISSUE__INST0__SHIFT 0x8
+#define SQ_THREAD_TRACE_WORD_ISSUE__INST1_MASK 0xc00
+#define SQ_THREAD_TRACE_WORD_ISSUE__INST1__SHIFT 0xa
+#define SQ_THREAD_TRACE_WORD_ISSUE__INST2_MASK 0x3000
+#define SQ_THREAD_TRACE_WORD_ISSUE__INST2__SHIFT 0xc
+#define SQ_THREAD_TRACE_WORD_ISSUE__INST3_MASK 0xc000
+#define SQ_THREAD_TRACE_WORD_ISSUE__INST3__SHIFT 0xe
+#define SQ_THREAD_TRACE_WORD_ISSUE__INST4_MASK 0x30000
+#define SQ_THREAD_TRACE_WORD_ISSUE__INST4__SHIFT 0x10
+#define SQ_THREAD_TRACE_WORD_ISSUE__INST5_MASK 0xc0000
+#define SQ_THREAD_TRACE_WORD_ISSUE__INST5__SHIFT 0x12
+#define SQ_THREAD_TRACE_WORD_ISSUE__INST6_MASK 0x300000
+#define SQ_THREAD_TRACE_WORD_ISSUE__INST6__SHIFT 0x14
+#define SQ_THREAD_TRACE_WORD_ISSUE__INST7_MASK 0xc00000
+#define SQ_THREAD_TRACE_WORD_ISSUE__INST7__SHIFT 0x16
+#define SQ_THREAD_TRACE_WORD_ISSUE__INST8_MASK 0x3000000
+#define SQ_THREAD_TRACE_WORD_ISSUE__INST8__SHIFT 0x18
+#define SQ_THREAD_TRACE_WORD_ISSUE__INST9_MASK 0xc000000
+#define SQ_THREAD_TRACE_WORD_ISSUE__INST9__SHIFT 0x1a
+#define SQ_THREAD_TRACE_WORD_PERF_1_OF_2__TOKEN_TYPE_MASK 0xf
+#define SQ_THREAD_TRACE_WORD_PERF_1_OF_2__TOKEN_TYPE__SHIFT 0x0
+#define SQ_THREAD_TRACE_WORD_PERF_1_OF_2__TIME_DELTA_MASK 0x10
+#define SQ_THREAD_TRACE_WORD_PERF_1_OF_2__TIME_DELTA__SHIFT 0x4
+#define SQ_THREAD_TRACE_WORD_PERF_1_OF_2__SH_ID_MASK 0x20
+#define SQ_THREAD_TRACE_WORD_PERF_1_OF_2__SH_ID__SHIFT 0x5
+#define SQ_THREAD_TRACE_WORD_PERF_1_OF_2__CU_ID_MASK 0x3c0
+#define SQ_THREAD_TRACE_WORD_PERF_1_OF_2__CU_ID__SHIFT 0x6
+#define SQ_THREAD_TRACE_WORD_PERF_1_OF_2__CNTR_BANK_MASK 0xc00
+#define SQ_THREAD_TRACE_WORD_PERF_1_OF_2__CNTR_BANK__SHIFT 0xa
+#define SQ_THREAD_TRACE_WORD_PERF_1_OF_2__CNTR0_MASK 0x1fff000
+#define SQ_THREAD_TRACE_WORD_PERF_1_OF_2__CNTR0__SHIFT 0xc
+#define SQ_THREAD_TRACE_WORD_PERF_1_OF_2__CNTR1_LO_MASK 0xfe000000
+#define SQ_THREAD_TRACE_WORD_PERF_1_OF_2__CNTR1_LO__SHIFT 0x19
+#define SQ_THREAD_TRACE_WORD_PERF_2_OF_2__CNTR1_HI_MASK 0x3f
+#define SQ_THREAD_TRACE_WORD_PERF_2_OF_2__CNTR1_HI__SHIFT 0x0
+#define SQ_THREAD_TRACE_WORD_PERF_2_OF_2__CNTR2_MASK 0x7ffc0
+#define SQ_THREAD_TRACE_WORD_PERF_2_OF_2__CNTR2__SHIFT 0x6
+#define SQ_THREAD_TRACE_WORD_PERF_2_OF_2__CNTR3_MASK 0xfff80000
+#define SQ_THREAD_TRACE_WORD_PERF_2_OF_2__CNTR3__SHIFT 0x13
+#define SQ_WREXEC_EXEC_LO__ADDR_LO_MASK 0xffffffff
+#define SQ_WREXEC_EXEC_LO__ADDR_LO__SHIFT 0x0
+#define SQ_WREXEC_EXEC_HI__ADDR_HI_MASK 0xffff
+#define SQ_WREXEC_EXEC_HI__ADDR_HI__SHIFT 0x0
+#define SQ_WREXEC_EXEC_HI__FIRST_WAVE_MASK 0x4000000
+#define SQ_WREXEC_EXEC_HI__FIRST_WAVE__SHIFT 0x1a
+#define SQ_WREXEC_EXEC_HI__ATC_MASK 0x8000000
+#define SQ_WREXEC_EXEC_HI__ATC__SHIFT 0x1b
+#define SQ_WREXEC_EXEC_HI__MTYPE_MASK 0x70000000
+#define SQ_WREXEC_EXEC_HI__MTYPE__SHIFT 0x1c
+#define SQ_WREXEC_EXEC_HI__MSB_MASK 0x80000000
+#define SQ_WREXEC_EXEC_HI__MSB__SHIFT 0x1f
+#define SQC_GATCL1_CNTL__RESERVED_MASK 0x3ffff
+#define SQC_GATCL1_CNTL__RESERVED__SHIFT 0x0
+#define SQC_GATCL1_CNTL__DCACHE_INVALIDATE_ALL_VMID_MASK 0x40000
+#define SQC_GATCL1_CNTL__DCACHE_INVALIDATE_ALL_VMID__SHIFT 0x12
+#define SQC_GATCL1_CNTL__DCACHE_FORCE_MISS_MASK 0x80000
+#define SQC_GATCL1_CNTL__DCACHE_FORCE_MISS__SHIFT 0x13
+#define SQC_GATCL1_CNTL__DCACHE_FORCE_IN_ORDER_MASK 0x100000
+#define SQC_GATCL1_CNTL__DCACHE_FORCE_IN_ORDER__SHIFT 0x14
+#define SQC_GATCL1_CNTL__DCACHE_REDUCE_FIFO_DEPTH_BY_2_MASK 0x600000
+#define SQC_GATCL1_CNTL__DCACHE_REDUCE_FIFO_DEPTH_BY_2__SHIFT 0x15
+#define SQC_GATCL1_CNTL__DCACHE_REDUCE_CACHE_SIZE_BY_2_MASK 0x1800000
+#define SQC_GATCL1_CNTL__DCACHE_REDUCE_CACHE_SIZE_BY_2__SHIFT 0x17
+#define SQC_GATCL1_CNTL__ICACHE_INVALIDATE_ALL_VMID_MASK 0x2000000
+#define SQC_GATCL1_CNTL__ICACHE_INVALIDATE_ALL_VMID__SHIFT 0x19
+#define SQC_GATCL1_CNTL__ICACHE_FORCE_MISS_MASK 0x4000000
+#define SQC_GATCL1_CNTL__ICACHE_FORCE_MISS__SHIFT 0x1a
+#define SQC_GATCL1_CNTL__ICACHE_FORCE_IN_ORDER_MASK 0x8000000
+#define SQC_GATCL1_CNTL__ICACHE_FORCE_IN_ORDER__SHIFT 0x1b
+#define SQC_GATCL1_CNTL__ICACHE_REDUCE_FIFO_DEPTH_BY_2_MASK 0x30000000
+#define SQC_GATCL1_CNTL__ICACHE_REDUCE_FIFO_DEPTH_BY_2__SHIFT 0x1c
+#define SQC_GATCL1_CNTL__ICACHE_REDUCE_CACHE_SIZE_BY_2_MASK 0xc0000000
+#define SQC_GATCL1_CNTL__ICACHE_REDUCE_CACHE_SIZE_BY_2__SHIFT 0x1e
+#define SQC_ATC_EDC_GATCL1_CNT__ICACHE_DATA_SEC_MASK 0xff
+#define SQC_ATC_EDC_GATCL1_CNT__ICACHE_DATA_SEC__SHIFT 0x0
+#define SQC_ATC_EDC_GATCL1_CNT__DCACHE_DATA_SEC_MASK 0xff0000
+#define SQC_ATC_EDC_GATCL1_CNT__DCACHE_DATA_SEC__SHIFT 0x10
+#define SQ_INTERRUPT_WORD_CMN__SE_ID_MASK 0x3000000
+#define SQ_INTERRUPT_WORD_CMN__SE_ID__SHIFT 0x18
+#define SQ_INTERRUPT_WORD_CMN__ENCODING_MASK 0xc000000
+#define SQ_INTERRUPT_WORD_CMN__ENCODING__SHIFT 0x1a
+#define SQ_INTERRUPT_WORD_AUTO__THREAD_TRACE_MASK 0x1
+#define SQ_INTERRUPT_WORD_AUTO__THREAD_TRACE__SHIFT 0x0
+#define SQ_INTERRUPT_WORD_AUTO__WLT_MASK 0x2
+#define SQ_INTERRUPT_WORD_AUTO__WLT__SHIFT 0x1
+#define SQ_INTERRUPT_WORD_AUTO__THREAD_TRACE_BUF_FULL_MASK 0x4
+#define SQ_INTERRUPT_WORD_AUTO__THREAD_TRACE_BUF_FULL__SHIFT 0x2
+#define SQ_INTERRUPT_WORD_AUTO__REG_TIMESTAMP_MASK 0x8
+#define SQ_INTERRUPT_WORD_AUTO__REG_TIMESTAMP__SHIFT 0x3
+#define SQ_INTERRUPT_WORD_AUTO__CMD_TIMESTAMP_MASK 0x10
+#define SQ_INTERRUPT_WORD_AUTO__CMD_TIMESTAMP__SHIFT 0x4
+#define SQ_INTERRUPT_WORD_AUTO__HOST_CMD_OVERFLOW_MASK 0x20
+#define SQ_INTERRUPT_WORD_AUTO__HOST_CMD_OVERFLOW__SHIFT 0x5
+#define SQ_INTERRUPT_WORD_AUTO__HOST_REG_OVERFLOW_MASK 0x40
+#define SQ_INTERRUPT_WORD_AUTO__HOST_REG_OVERFLOW__SHIFT 0x6
+#define SQ_INTERRUPT_WORD_AUTO__IMMED_OVERFLOW_MASK 0x80
+#define SQ_INTERRUPT_WORD_AUTO__IMMED_OVERFLOW__SHIFT 0x7
+#define SQ_INTERRUPT_WORD_AUTO__SE_ID_MASK 0x3000000
+#define SQ_INTERRUPT_WORD_AUTO__SE_ID__SHIFT 0x18
+#define SQ_INTERRUPT_WORD_AUTO__ENCODING_MASK 0xc000000
+#define SQ_INTERRUPT_WORD_AUTO__ENCODING__SHIFT 0x1a
+#define SQ_INTERRUPT_WORD_WAVE__DATA_MASK 0xff
+#define SQ_INTERRUPT_WORD_WAVE__DATA__SHIFT 0x0
+#define SQ_INTERRUPT_WORD_WAVE__SH_ID_MASK 0x100
+#define SQ_INTERRUPT_WORD_WAVE__SH_ID__SHIFT 0x8
+#define SQ_INTERRUPT_WORD_WAVE__PRIV_MASK 0x200
+#define SQ_INTERRUPT_WORD_WAVE__PRIV__SHIFT 0x9
+#define SQ_INTERRUPT_WORD_WAVE__VM_ID_MASK 0x3c00
+#define SQ_INTERRUPT_WORD_WAVE__VM_ID__SHIFT 0xa
+#define SQ_INTERRUPT_WORD_WAVE__WAVE_ID_MASK 0x3c000
+#define SQ_INTERRUPT_WORD_WAVE__WAVE_ID__SHIFT 0xe
+#define SQ_INTERRUPT_WORD_WAVE__SIMD_ID_MASK 0xc0000
+#define SQ_INTERRUPT_WORD_WAVE__SIMD_ID__SHIFT 0x12
+#define SQ_INTERRUPT_WORD_WAVE__CU_ID_MASK 0xf00000
+#define SQ_INTERRUPT_WORD_WAVE__CU_ID__SHIFT 0x14
+#define SQ_INTERRUPT_WORD_WAVE__SE_ID_MASK 0x3000000
+#define SQ_INTERRUPT_WORD_WAVE__SE_ID__SHIFT 0x18
+#define SQ_INTERRUPT_WORD_WAVE__ENCODING_MASK 0xc000000
+#define SQ_INTERRUPT_WORD_WAVE__ENCODING__SHIFT 0x1a
+#define SQ_SOP2__SSRC0_MASK 0xff
+#define SQ_SOP2__SSRC0__SHIFT 0x0
+#define SQ_SOP2__SSRC1_MASK 0xff00
+#define SQ_SOP2__SSRC1__SHIFT 0x8
+#define SQ_SOP2__SDST_MASK 0x7f0000
+#define SQ_SOP2__SDST__SHIFT 0x10
+#define SQ_SOP2__OP_MASK 0x3f800000
+#define SQ_SOP2__OP__SHIFT 0x17
+#define SQ_SOP2__ENCODING_MASK 0xc0000000
+#define SQ_SOP2__ENCODING__SHIFT 0x1e
+#define SQ_VOP1__SRC0_MASK 0x1ff
+#define SQ_VOP1__SRC0__SHIFT 0x0
+#define SQ_VOP1__OP_MASK 0x1fe00
+#define SQ_VOP1__OP__SHIFT 0x9
+#define SQ_VOP1__VDST_MASK 0x1fe0000
+#define SQ_VOP1__VDST__SHIFT 0x11
+#define SQ_VOP1__ENCODING_MASK 0xfe000000
+#define SQ_VOP1__ENCODING__SHIFT 0x19
+#define SQ_MTBUF_1__VADDR_MASK 0xff
+#define SQ_MTBUF_1__VADDR__SHIFT 0x0
+#define SQ_MTBUF_1__VDATA_MASK 0xff00
+#define SQ_MTBUF_1__VDATA__SHIFT 0x8
+#define SQ_MTBUF_1__SRSRC_MASK 0x1f0000
+#define SQ_MTBUF_1__SRSRC__SHIFT 0x10
+#define SQ_MTBUF_1__SLC_MASK 0x400000
+#define SQ_MTBUF_1__SLC__SHIFT 0x16
+#define SQ_MTBUF_1__TFE_MASK 0x800000
+#define SQ_MTBUF_1__TFE__SHIFT 0x17
+#define SQ_MTBUF_1__SOFFSET_MASK 0xff000000
+#define SQ_MTBUF_1__SOFFSET__SHIFT 0x18
+#define SQ_EXP_1__VSRC0_MASK 0xff
+#define SQ_EXP_1__VSRC0__SHIFT 0x0
+#define SQ_EXP_1__VSRC1_MASK 0xff00
+#define SQ_EXP_1__VSRC1__SHIFT 0x8
+#define SQ_EXP_1__VSRC2_MASK 0xff0000
+#define SQ_EXP_1__VSRC2__SHIFT 0x10
+#define SQ_EXP_1__VSRC3_MASK 0xff000000
+#define SQ_EXP_1__VSRC3__SHIFT 0x18
+#define SQ_MUBUF_1__VADDR_MASK 0xff
+#define SQ_MUBUF_1__VADDR__SHIFT 0x0
+#define SQ_MUBUF_1__VDATA_MASK 0xff00
+#define SQ_MUBUF_1__VDATA__SHIFT 0x8
+#define SQ_MUBUF_1__SRSRC_MASK 0x1f0000
+#define SQ_MUBUF_1__SRSRC__SHIFT 0x10
+#define SQ_MUBUF_1__TFE_MASK 0x800000
+#define SQ_MUBUF_1__TFE__SHIFT 0x17
+#define SQ_MUBUF_1__SOFFSET_MASK 0xff000000
+#define SQ_MUBUF_1__SOFFSET__SHIFT 0x18
+#define SQ_SMEM_1__OFFSET_MASK 0xfffff
+#define SQ_SMEM_1__OFFSET__SHIFT 0x0
+#define SQ_INST__ENCODING_MASK 0xffffffff
+#define SQ_INST__ENCODING__SHIFT 0x0
+#define SQ_EXP_0__EN_MASK 0xf
+#define SQ_EXP_0__EN__SHIFT 0x0
+#define SQ_EXP_0__TGT_MASK 0x3f0
+#define SQ_EXP_0__TGT__SHIFT 0x4
+#define SQ_EXP_0__COMPR_MASK 0x400
+#define SQ_EXP_0__COMPR__SHIFT 0xa
+#define SQ_EXP_0__DONE_MASK 0x800
+#define SQ_EXP_0__DONE__SHIFT 0xb
+#define SQ_EXP_0__VM_MASK 0x1000
+#define SQ_EXP_0__VM__SHIFT 0xc
+#define SQ_EXP_0__ENCODING_MASK 0xfc000000
+#define SQ_EXP_0__ENCODING__SHIFT 0x1a
+#define SQ_MUBUF_0__OFFSET_MASK 0xfff
+#define SQ_MUBUF_0__OFFSET__SHIFT 0x0
+#define SQ_MUBUF_0__OFFEN_MASK 0x1000
+#define SQ_MUBUF_0__OFFEN__SHIFT 0xc
+#define SQ_MUBUF_0__IDXEN_MASK 0x2000
+#define SQ_MUBUF_0__IDXEN__SHIFT 0xd
+#define SQ_MUBUF_0__GLC_MASK 0x4000
+#define SQ_MUBUF_0__GLC__SHIFT 0xe
+#define SQ_MUBUF_0__LDS_MASK 0x10000
+#define SQ_MUBUF_0__LDS__SHIFT 0x10
+#define SQ_MUBUF_0__SLC_MASK 0x20000
+#define SQ_MUBUF_0__SLC__SHIFT 0x11
+#define SQ_MUBUF_0__OP_MASK 0x1fc0000
+#define SQ_MUBUF_0__OP__SHIFT 0x12
+#define SQ_MUBUF_0__ENCODING_MASK 0xfc000000
+#define SQ_MUBUF_0__ENCODING__SHIFT 0x1a
+#define SQ_VOP_SDWA__SRC0_MASK 0xff
+#define SQ_VOP_SDWA__SRC0__SHIFT 0x0
+#define SQ_VOP_SDWA__DST_SEL_MASK 0x700
+#define SQ_VOP_SDWA__DST_SEL__SHIFT 0x8
+#define SQ_VOP_SDWA__DST_UNUSED_MASK 0x1800
+#define SQ_VOP_SDWA__DST_UNUSED__SHIFT 0xb
+#define SQ_VOP_SDWA__CLAMP_MASK 0x2000
+#define SQ_VOP_SDWA__CLAMP__SHIFT 0xd
+#define SQ_VOP_SDWA__SRC0_SEL_MASK 0x70000
+#define SQ_VOP_SDWA__SRC0_SEL__SHIFT 0x10
+#define SQ_VOP_SDWA__SRC0_SEXT_MASK 0x80000
+#define SQ_VOP_SDWA__SRC0_SEXT__SHIFT 0x13
+#define SQ_VOP_SDWA__SRC0_NEG_MASK 0x100000
+#define SQ_VOP_SDWA__SRC0_NEG__SHIFT 0x14
+#define SQ_VOP_SDWA__SRC0_ABS_MASK 0x200000
+#define SQ_VOP_SDWA__SRC0_ABS__SHIFT 0x15
+#define SQ_VOP_SDWA__SRC1_SEL_MASK 0x7000000
+#define SQ_VOP_SDWA__SRC1_SEL__SHIFT 0x18
+#define SQ_VOP_SDWA__SRC1_SEXT_MASK 0x8000000
+#define SQ_VOP_SDWA__SRC1_SEXT__SHIFT 0x1b
+#define SQ_VOP_SDWA__SRC1_NEG_MASK 0x10000000
+#define SQ_VOP_SDWA__SRC1_NEG__SHIFT 0x1c
+#define SQ_VOP_SDWA__SRC1_ABS_MASK 0x20000000
+#define SQ_VOP_SDWA__SRC1_ABS__SHIFT 0x1d
+#define SQ_VOP3_0__VDST_MASK 0xff
+#define SQ_VOP3_0__VDST__SHIFT 0x0
+#define SQ_VOP3_0__ABS_MASK 0x700
+#define SQ_VOP3_0__ABS__SHIFT 0x8
+#define SQ_VOP3_0__CLAMP_MASK 0x8000
+#define SQ_VOP3_0__CLAMP__SHIFT 0xf
+#define SQ_VOP3_0__OP_MASK 0x3ff0000
+#define SQ_VOP3_0__OP__SHIFT 0x10
+#define SQ_VOP3_0__ENCODING_MASK 0xfc000000
+#define SQ_VOP3_0__ENCODING__SHIFT 0x1a
+#define SQ_VOP2__SRC0_MASK 0x1ff
+#define SQ_VOP2__SRC0__SHIFT 0x0
+#define SQ_VOP2__VSRC1_MASK 0x1fe00
+#define SQ_VOP2__VSRC1__SHIFT 0x9
+#define SQ_VOP2__VDST_MASK 0x1fe0000
+#define SQ_VOP2__VDST__SHIFT 0x11
+#define SQ_VOP2__OP_MASK 0x7e000000
+#define SQ_VOP2__OP__SHIFT 0x19
+#define SQ_VOP2__ENCODING_MASK 0x80000000
+#define SQ_VOP2__ENCODING__SHIFT 0x1f
+#define SQ_MTBUF_0__OFFSET_MASK 0xfff
+#define SQ_MTBUF_0__OFFSET__SHIFT 0x0
+#define SQ_MTBUF_0__OFFEN_MASK 0x1000
+#define SQ_MTBUF_0__OFFEN__SHIFT 0xc
+#define SQ_MTBUF_0__IDXEN_MASK 0x2000
+#define SQ_MTBUF_0__IDXEN__SHIFT 0xd
+#define SQ_MTBUF_0__GLC_MASK 0x4000
+#define SQ_MTBUF_0__GLC__SHIFT 0xe
+#define SQ_MTBUF_0__OP_MASK 0x78000
+#define SQ_MTBUF_0__OP__SHIFT 0xf
+#define SQ_MTBUF_0__DFMT_MASK 0x780000
+#define SQ_MTBUF_0__DFMT__SHIFT 0x13
+#define SQ_MTBUF_0__NFMT_MASK 0x3800000
+#define SQ_MTBUF_0__NFMT__SHIFT 0x17
+#define SQ_MTBUF_0__ENCODING_MASK 0xfc000000
+#define SQ_MTBUF_0__ENCODING__SHIFT 0x1a
+#define SQ_SOPP__SIMM16_MASK 0xffff
+#define SQ_SOPP__SIMM16__SHIFT 0x0
+#define SQ_SOPP__OP_MASK 0x7f0000
+#define SQ_SOPP__OP__SHIFT 0x10
+#define SQ_SOPP__ENCODING_MASK 0xff800000
+#define SQ_SOPP__ENCODING__SHIFT 0x17
+#define SQ_FLAT_0__GLC_MASK 0x10000
+#define SQ_FLAT_0__GLC__SHIFT 0x10
+#define SQ_FLAT_0__SLC_MASK 0x20000
+#define SQ_FLAT_0__SLC__SHIFT 0x11
+#define SQ_FLAT_0__OP_MASK 0x1fc0000
+#define SQ_FLAT_0__OP__SHIFT 0x12
+#define SQ_FLAT_0__ENCODING_MASK 0xfc000000
+#define SQ_FLAT_0__ENCODING__SHIFT 0x1a
+#define SQ_VOP3_0_SDST_ENC__VDST_MASK 0xff
+#define SQ_VOP3_0_SDST_ENC__VDST__SHIFT 0x0
+#define SQ_VOP3_0_SDST_ENC__SDST_MASK 0x7f00
+#define SQ_VOP3_0_SDST_ENC__SDST__SHIFT 0x8
+#define SQ_VOP3_0_SDST_ENC__CLAMP_MASK 0x8000
+#define SQ_VOP3_0_SDST_ENC__CLAMP__SHIFT 0xf
+#define SQ_VOP3_0_SDST_ENC__OP_MASK 0x3ff0000
+#define SQ_VOP3_0_SDST_ENC__OP__SHIFT 0x10
+#define SQ_VOP3_0_SDST_ENC__ENCODING_MASK 0xfc000000
+#define SQ_VOP3_0_SDST_ENC__ENCODING__SHIFT 0x1a
+#define SQ_MIMG_1__VADDR_MASK 0xff
+#define SQ_MIMG_1__VADDR__SHIFT 0x0
+#define SQ_MIMG_1__VDATA_MASK 0xff00
+#define SQ_MIMG_1__VDATA__SHIFT 0x8
+#define SQ_MIMG_1__SRSRC_MASK 0x1f0000
+#define SQ_MIMG_1__SRSRC__SHIFT 0x10
+#define SQ_MIMG_1__SSAMP_MASK 0x3e00000
+#define SQ_MIMG_1__SSAMP__SHIFT 0x15
+#define SQ_MIMG_1__D16_MASK 0x80000000
+#define SQ_MIMG_1__D16__SHIFT 0x1f
+#define SQ_SOP1__SSRC0_MASK 0xff
+#define SQ_SOP1__SSRC0__SHIFT 0x0
+#define SQ_SOP1__OP_MASK 0xff00
+#define SQ_SOP1__OP__SHIFT 0x8
+#define SQ_SOP1__SDST_MASK 0x7f0000
+#define SQ_SOP1__SDST__SHIFT 0x10
+#define SQ_SOP1__ENCODING_MASK 0xff800000
+#define SQ_SOP1__ENCODING__SHIFT 0x17
+#define SQ_SOPC__SSRC0_MASK 0xff
+#define SQ_SOPC__SSRC0__SHIFT 0x0
+#define SQ_SOPC__SSRC1_MASK 0xff00
+#define SQ_SOPC__SSRC1__SHIFT 0x8
+#define SQ_SOPC__OP_MASK 0x7f0000
+#define SQ_SOPC__OP__SHIFT 0x10
+#define SQ_SOPC__ENCODING_MASK 0xff800000
+#define SQ_SOPC__ENCODING__SHIFT 0x17
+#define SQ_FLAT_1__ADDR_MASK 0xff
+#define SQ_FLAT_1__ADDR__SHIFT 0x0
+#define SQ_FLAT_1__DATA_MASK 0xff00
+#define SQ_FLAT_1__DATA__SHIFT 0x8
+#define SQ_FLAT_1__TFE_MASK 0x800000
+#define SQ_FLAT_1__TFE__SHIFT 0x17
+#define SQ_FLAT_1__VDST_MASK 0xff000000
+#define SQ_FLAT_1__VDST__SHIFT 0x18
+#define SQ_DS_1__ADDR_MASK 0xff
+#define SQ_DS_1__ADDR__SHIFT 0x0
+#define SQ_DS_1__DATA0_MASK 0xff00
+#define SQ_DS_1__DATA0__SHIFT 0x8
+#define SQ_DS_1__DATA1_MASK 0xff0000
+#define SQ_DS_1__DATA1__SHIFT 0x10
+#define SQ_DS_1__VDST_MASK 0xff000000
+#define SQ_DS_1__VDST__SHIFT 0x18
+#define SQ_VOP3_1__SRC0_MASK 0x1ff
+#define SQ_VOP3_1__SRC0__SHIFT 0x0
+#define SQ_VOP3_1__SRC1_MASK 0x3fe00
+#define SQ_VOP3_1__SRC1__SHIFT 0x9
+#define SQ_VOP3_1__SRC2_MASK 0x7fc0000
+#define SQ_VOP3_1__SRC2__SHIFT 0x12
+#define SQ_VOP3_1__OMOD_MASK 0x18000000
+#define SQ_VOP3_1__OMOD__SHIFT 0x1b
+#define SQ_VOP3_1__NEG_MASK 0xe0000000
+#define SQ_VOP3_1__NEG__SHIFT 0x1d
+#define SQ_SMEM_0__SBASE_MASK 0x3f
+#define SQ_SMEM_0__SBASE__SHIFT 0x0
+#define SQ_SMEM_0__SDATA_MASK 0x1fc0
+#define SQ_SMEM_0__SDATA__SHIFT 0x6
+#define SQ_SMEM_0__GLC_MASK 0x10000
+#define SQ_SMEM_0__GLC__SHIFT 0x10
+#define SQ_SMEM_0__IMM_MASK 0x20000
+#define SQ_SMEM_0__IMM__SHIFT 0x11
+#define SQ_SMEM_0__OP_MASK 0x3fc0000
+#define SQ_SMEM_0__OP__SHIFT 0x12
+#define SQ_SMEM_0__ENCODING_MASK 0xfc000000
+#define SQ_SMEM_0__ENCODING__SHIFT 0x1a
+#define SQ_MIMG_0__DMASK_MASK 0xf00
+#define SQ_MIMG_0__DMASK__SHIFT 0x8
+#define SQ_MIMG_0__UNORM_MASK 0x1000
+#define SQ_MIMG_0__UNORM__SHIFT 0xc
+#define SQ_MIMG_0__GLC_MASK 0x2000
+#define SQ_MIMG_0__GLC__SHIFT 0xd
+#define SQ_MIMG_0__DA_MASK 0x4000
+#define SQ_MIMG_0__DA__SHIFT 0xe
+#define SQ_MIMG_0__R128_MASK 0x8000
+#define SQ_MIMG_0__R128__SHIFT 0xf
+#define SQ_MIMG_0__TFE_MASK 0x10000
+#define SQ_MIMG_0__TFE__SHIFT 0x10
+#define SQ_MIMG_0__LWE_MASK 0x20000
+#define SQ_MIMG_0__LWE__SHIFT 0x11
+#define SQ_MIMG_0__OP_MASK 0x1fc0000
+#define SQ_MIMG_0__OP__SHIFT 0x12
+#define SQ_MIMG_0__SLC_MASK 0x2000000
+#define SQ_MIMG_0__SLC__SHIFT 0x19
+#define SQ_MIMG_0__ENCODING_MASK 0xfc000000
+#define SQ_MIMG_0__ENCODING__SHIFT 0x1a
+#define SQ_SOPK__SIMM16_MASK 0xffff
+#define SQ_SOPK__SIMM16__SHIFT 0x0
+#define SQ_SOPK__SDST_MASK 0x7f0000
+#define SQ_SOPK__SDST__SHIFT 0x10
+#define SQ_SOPK__OP_MASK 0xf800000
+#define SQ_SOPK__OP__SHIFT 0x17
+#define SQ_SOPK__ENCODING_MASK 0xf0000000
+#define SQ_SOPK__ENCODING__SHIFT 0x1c
+#define SQ_DS_0__OFFSET0_MASK 0xff
+#define SQ_DS_0__OFFSET0__SHIFT 0x0
+#define SQ_DS_0__OFFSET1_MASK 0xff00
+#define SQ_DS_0__OFFSET1__SHIFT 0x8
+#define SQ_DS_0__GDS_MASK 0x10000
+#define SQ_DS_0__GDS__SHIFT 0x10
+#define SQ_DS_0__OP_MASK 0x1fe0000
+#define SQ_DS_0__OP__SHIFT 0x11
+#define SQ_DS_0__ENCODING_MASK 0xfc000000
+#define SQ_DS_0__ENCODING__SHIFT 0x1a
+#define SQ_VOP_DPP__SRC0_MASK 0xff
+#define SQ_VOP_DPP__SRC0__SHIFT 0x0
+#define SQ_VOP_DPP__DPP_CTRL_MASK 0x1ff00
+#define SQ_VOP_DPP__DPP_CTRL__SHIFT 0x8
+#define SQ_VOP_DPP__BOUND_CTRL_MASK 0x80000
+#define SQ_VOP_DPP__BOUND_CTRL__SHIFT 0x13
+#define SQ_VOP_DPP__SRC0_NEG_MASK 0x100000
+#define SQ_VOP_DPP__SRC0_NEG__SHIFT 0x14
+#define SQ_VOP_DPP__SRC0_ABS_MASK 0x200000
+#define SQ_VOP_DPP__SRC0_ABS__SHIFT 0x15
+#define SQ_VOP_DPP__SRC1_NEG_MASK 0x400000
+#define SQ_VOP_DPP__SRC1_NEG__SHIFT 0x16
+#define SQ_VOP_DPP__SRC1_ABS_MASK 0x800000
+#define SQ_VOP_DPP__SRC1_ABS__SHIFT 0x17
+#define SQ_VOP_DPP__BANK_MASK_MASK 0xf000000
+#define SQ_VOP_DPP__BANK_MASK__SHIFT 0x18
+#define SQ_VOP_DPP__ROW_MASK_MASK 0xf0000000
+#define SQ_VOP_DPP__ROW_MASK__SHIFT 0x1c
+#define SQ_VOPC__SRC0_MASK 0x1ff
+#define SQ_VOPC__SRC0__SHIFT 0x0
+#define SQ_VOPC__VSRC1_MASK 0x1fe00
+#define SQ_VOPC__VSRC1__SHIFT 0x9
+#define SQ_VOPC__OP_MASK 0x1fe0000
+#define SQ_VOPC__OP__SHIFT 0x11
+#define SQ_VOPC__ENCODING_MASK 0xfe000000
+#define SQ_VOPC__ENCODING__SHIFT 0x19
+#define SQ_VINTRP__VSRC_MASK 0xff
+#define SQ_VINTRP__VSRC__SHIFT 0x0
+#define SQ_VINTRP__ATTRCHAN_MASK 0x300
+#define SQ_VINTRP__ATTRCHAN__SHIFT 0x8
+#define SQ_VINTRP__ATTR_MASK 0xfc00
+#define SQ_VINTRP__ATTR__SHIFT 0xa
+#define SQ_VINTRP__OP_MASK 0x30000
+#define SQ_VINTRP__OP__SHIFT 0x10
+#define SQ_VINTRP__VDST_MASK 0x3fc0000
+#define SQ_VINTRP__VDST__SHIFT 0x12
+#define SQ_VINTRP__ENCODING_MASK 0xfc000000
+#define SQ_VINTRP__ENCODING__SHIFT 0x1a
+#define CGTT_SX_CLK_CTRL0__ON_DELAY_MASK 0xf
+#define CGTT_SX_CLK_CTRL0__ON_DELAY__SHIFT 0x0
+#define CGTT_SX_CLK_CTRL0__OFF_HYSTERESIS_MASK 0xff0
+#define CGTT_SX_CLK_CTRL0__OFF_HYSTERESIS__SHIFT 0x4
+#define CGTT_SX_CLK_CTRL0__RESERVED_MASK 0xfff000
+#define CGTT_SX_CLK_CTRL0__RESERVED__SHIFT 0xc
+#define CGTT_SX_CLK_CTRL0__SOFT_OVERRIDE7_MASK 0x1000000
+#define CGTT_SX_CLK_CTRL0__SOFT_OVERRIDE7__SHIFT 0x18
+#define CGTT_SX_CLK_CTRL0__SOFT_OVERRIDE6_MASK 0x2000000
+#define CGTT_SX_CLK_CTRL0__SOFT_OVERRIDE6__SHIFT 0x19
+#define CGTT_SX_CLK_CTRL0__SOFT_OVERRIDE5_MASK 0x4000000
+#define CGTT_SX_CLK_CTRL0__SOFT_OVERRIDE5__SHIFT 0x1a
+#define CGTT_SX_CLK_CTRL0__SOFT_OVERRIDE4_MASK 0x8000000
+#define CGTT_SX_CLK_CTRL0__SOFT_OVERRIDE4__SHIFT 0x1b
+#define CGTT_SX_CLK_CTRL0__SOFT_OVERRIDE3_MASK 0x10000000
+#define CGTT_SX_CLK_CTRL0__SOFT_OVERRIDE3__SHIFT 0x1c
+#define CGTT_SX_CLK_CTRL0__SOFT_OVERRIDE2_MASK 0x20000000
+#define CGTT_SX_CLK_CTRL0__SOFT_OVERRIDE2__SHIFT 0x1d
+#define CGTT_SX_CLK_CTRL0__SOFT_OVERRIDE1_MASK 0x40000000
+#define CGTT_SX_CLK_CTRL0__SOFT_OVERRIDE1__SHIFT 0x1e
+#define CGTT_SX_CLK_CTRL0__SOFT_OVERRIDE0_MASK 0x80000000
+#define CGTT_SX_CLK_CTRL0__SOFT_OVERRIDE0__SHIFT 0x1f
+#define CGTT_SX_CLK_CTRL1__ON_DELAY_MASK 0xf
+#define CGTT_SX_CLK_CTRL1__ON_DELAY__SHIFT 0x0
+#define CGTT_SX_CLK_CTRL1__OFF_HYSTERESIS_MASK 0xff0
+#define CGTT_SX_CLK_CTRL1__OFF_HYSTERESIS__SHIFT 0x4
+#define CGTT_SX_CLK_CTRL1__RESERVED_MASK 0xfff000
+#define CGTT_SX_CLK_CTRL1__RESERVED__SHIFT 0xc
+#define CGTT_SX_CLK_CTRL1__DBG_EN_MASK 0x1000000
+#define CGTT_SX_CLK_CTRL1__DBG_EN__SHIFT 0x18
+#define CGTT_SX_CLK_CTRL1__SOFT_OVERRIDE6_MASK 0x2000000
+#define CGTT_SX_CLK_CTRL1__SOFT_OVERRIDE6__SHIFT 0x19
+#define CGTT_SX_CLK_CTRL1__SOFT_OVERRIDE5_MASK 0x4000000
+#define CGTT_SX_CLK_CTRL1__SOFT_OVERRIDE5__SHIFT 0x1a
+#define CGTT_SX_CLK_CTRL1__SOFT_OVERRIDE4_MASK 0x8000000
+#define CGTT_SX_CLK_CTRL1__SOFT_OVERRIDE4__SHIFT 0x1b
+#define CGTT_SX_CLK_CTRL1__SOFT_OVERRIDE3_MASK 0x10000000
+#define CGTT_SX_CLK_CTRL1__SOFT_OVERRIDE3__SHIFT 0x1c
+#define CGTT_SX_CLK_CTRL1__SOFT_OVERRIDE2_MASK 0x20000000
+#define CGTT_SX_CLK_CTRL1__SOFT_OVERRIDE2__SHIFT 0x1d
+#define CGTT_SX_CLK_CTRL1__SOFT_OVERRIDE1_MASK 0x40000000
+#define CGTT_SX_CLK_CTRL1__SOFT_OVERRIDE1__SHIFT 0x1e
+#define CGTT_SX_CLK_CTRL1__SOFT_OVERRIDE0_MASK 0x80000000
+#define CGTT_SX_CLK_CTRL1__SOFT_OVERRIDE0__SHIFT 0x1f
+#define CGTT_SX_CLK_CTRL2__ON_DELAY_MASK 0xf
+#define CGTT_SX_CLK_CTRL2__ON_DELAY__SHIFT 0x0
+#define CGTT_SX_CLK_CTRL2__OFF_HYSTERESIS_MASK 0xff0
+#define CGTT_SX_CLK_CTRL2__OFF_HYSTERESIS__SHIFT 0x4
+#define CGTT_SX_CLK_CTRL2__RESERVED_MASK 0xfff000
+#define CGTT_SX_CLK_CTRL2__RESERVED__SHIFT 0xc
+#define CGTT_SX_CLK_CTRL2__DBG_EN_MASK 0x1000000
+#define CGTT_SX_CLK_CTRL2__DBG_EN__SHIFT 0x18
+#define CGTT_SX_CLK_CTRL2__SOFT_OVERRIDE6_MASK 0x2000000
+#define CGTT_SX_CLK_CTRL2__SOFT_OVERRIDE6__SHIFT 0x19
+#define CGTT_SX_CLK_CTRL2__SOFT_OVERRIDE5_MASK 0x4000000
+#define CGTT_SX_CLK_CTRL2__SOFT_OVERRIDE5__SHIFT 0x1a
+#define CGTT_SX_CLK_CTRL2__SOFT_OVERRIDE4_MASK 0x8000000
+#define CGTT_SX_CLK_CTRL2__SOFT_OVERRIDE4__SHIFT 0x1b
+#define CGTT_SX_CLK_CTRL2__SOFT_OVERRIDE3_MASK 0x10000000
+#define CGTT_SX_CLK_CTRL2__SOFT_OVERRIDE3__SHIFT 0x1c
+#define CGTT_SX_CLK_CTRL2__SOFT_OVERRIDE2_MASK 0x20000000
+#define CGTT_SX_CLK_CTRL2__SOFT_OVERRIDE2__SHIFT 0x1d
+#define CGTT_SX_CLK_CTRL2__SOFT_OVERRIDE1_MASK 0x40000000
+#define CGTT_SX_CLK_CTRL2__SOFT_OVERRIDE1__SHIFT 0x1e
+#define CGTT_SX_CLK_CTRL2__SOFT_OVERRIDE0_MASK 0x80000000
+#define CGTT_SX_CLK_CTRL2__SOFT_OVERRIDE0__SHIFT 0x1f
+#define CGTT_SX_CLK_CTRL3__ON_DELAY_MASK 0xf
+#define CGTT_SX_CLK_CTRL3__ON_DELAY__SHIFT 0x0
+#define CGTT_SX_CLK_CTRL3__OFF_HYSTERESIS_MASK 0xff0
+#define CGTT_SX_CLK_CTRL3__OFF_HYSTERESIS__SHIFT 0x4
+#define CGTT_SX_CLK_CTRL3__RESERVED_MASK 0xfff000
+#define CGTT_SX_CLK_CTRL3__RESERVED__SHIFT 0xc
+#define CGTT_SX_CLK_CTRL3__DBG_EN_MASK 0x1000000
+#define CGTT_SX_CLK_CTRL3__DBG_EN__SHIFT 0x18
+#define CGTT_SX_CLK_CTRL3__SOFT_OVERRIDE6_MASK 0x2000000
+#define CGTT_SX_CLK_CTRL3__SOFT_OVERRIDE6__SHIFT 0x19
+#define CGTT_SX_CLK_CTRL3__SOFT_OVERRIDE5_MASK 0x4000000
+#define CGTT_SX_CLK_CTRL3__SOFT_OVERRIDE5__SHIFT 0x1a
+#define CGTT_SX_CLK_CTRL3__SOFT_OVERRIDE4_MASK 0x8000000
+#define CGTT_SX_CLK_CTRL3__SOFT_OVERRIDE4__SHIFT 0x1b
+#define CGTT_SX_CLK_CTRL3__SOFT_OVERRIDE3_MASK 0x10000000
+#define CGTT_SX_CLK_CTRL3__SOFT_OVERRIDE3__SHIFT 0x1c
+#define CGTT_SX_CLK_CTRL3__SOFT_OVERRIDE2_MASK 0x20000000
+#define CGTT_SX_CLK_CTRL3__SOFT_OVERRIDE2__SHIFT 0x1d
+#define CGTT_SX_CLK_CTRL3__SOFT_OVERRIDE1_MASK 0x40000000
+#define CGTT_SX_CLK_CTRL3__SOFT_OVERRIDE1__SHIFT 0x1e
+#define CGTT_SX_CLK_CTRL3__SOFT_OVERRIDE0_MASK 0x80000000
+#define CGTT_SX_CLK_CTRL3__SOFT_OVERRIDE0__SHIFT 0x1f
+#define CGTT_SX_CLK_CTRL4__ON_DELAY_MASK 0xf
+#define CGTT_SX_CLK_CTRL4__ON_DELAY__SHIFT 0x0
+#define CGTT_SX_CLK_CTRL4__OFF_HYSTERESIS_MASK 0xff0
+#define CGTT_SX_CLK_CTRL4__OFF_HYSTERESIS__SHIFT 0x4
+#define CGTT_SX_CLK_CTRL4__RESERVED_MASK 0xfff000
+#define CGTT_SX_CLK_CTRL4__RESERVED__SHIFT 0xc
+#define CGTT_SX_CLK_CTRL4__DBG_EN_MASK 0x1000000
+#define CGTT_SX_CLK_CTRL4__DBG_EN__SHIFT 0x18
+#define CGTT_SX_CLK_CTRL4__SOFT_OVERRIDE6_MASK 0x2000000
+#define CGTT_SX_CLK_CTRL4__SOFT_OVERRIDE6__SHIFT 0x19
+#define CGTT_SX_CLK_CTRL4__SOFT_OVERRIDE5_MASK 0x4000000
+#define CGTT_SX_CLK_CTRL4__SOFT_OVERRIDE5__SHIFT 0x1a
+#define CGTT_SX_CLK_CTRL4__SOFT_OVERRIDE4_MASK 0x8000000
+#define CGTT_SX_CLK_CTRL4__SOFT_OVERRIDE4__SHIFT 0x1b
+#define CGTT_SX_CLK_CTRL4__SOFT_OVERRIDE3_MASK 0x10000000
+#define CGTT_SX_CLK_CTRL4__SOFT_OVERRIDE3__SHIFT 0x1c
+#define CGTT_SX_CLK_CTRL4__SOFT_OVERRIDE2_MASK 0x20000000
+#define CGTT_SX_CLK_CTRL4__SOFT_OVERRIDE2__SHIFT 0x1d
+#define CGTT_SX_CLK_CTRL4__SOFT_OVERRIDE1_MASK 0x40000000
+#define CGTT_SX_CLK_CTRL4__SOFT_OVERRIDE1__SHIFT 0x1e
+#define CGTT_SX_CLK_CTRL4__SOFT_OVERRIDE0_MASK 0x80000000
+#define CGTT_SX_CLK_CTRL4__SOFT_OVERRIDE0__SHIFT 0x1f
+#define SX_DEBUG_BUSY__POS_FREE_OR_VALIDS_MASK 0x1
+#define SX_DEBUG_BUSY__POS_FREE_OR_VALIDS__SHIFT 0x0
+#define SX_DEBUG_BUSY__POS_REQUESTER_BUSY_MASK 0x2
+#define SX_DEBUG_BUSY__POS_REQUESTER_BUSY__SHIFT 0x1
+#define SX_DEBUG_BUSY__PA_SX_BUSY_MASK 0x4
+#define SX_DEBUG_BUSY__PA_SX_BUSY__SHIFT 0x2
+#define SX_DEBUG_BUSY__POS_SCBD_BUSY_MASK 0x8
+#define SX_DEBUG_BUSY__POS_SCBD_BUSY__SHIFT 0x3
+#define SX_DEBUG_BUSY__POS_BANK3VAL3_BUSY_MASK 0x10
+#define SX_DEBUG_BUSY__POS_BANK3VAL3_BUSY__SHIFT 0x4
+#define SX_DEBUG_BUSY__POS_BANK3VAL2_BUSY_MASK 0x20
+#define SX_DEBUG_BUSY__POS_BANK3VAL2_BUSY__SHIFT 0x5
+#define SX_DEBUG_BUSY__POS_BANK3VAL1_BUSY_MASK 0x40
+#define SX_DEBUG_BUSY__POS_BANK3VAL1_BUSY__SHIFT 0x6
+#define SX_DEBUG_BUSY__POS_BANK3VAL0_BUSY_MASK 0x80
+#define SX_DEBUG_BUSY__POS_BANK3VAL0_BUSY__SHIFT 0x7
+#define SX_DEBUG_BUSY__POS_BANK2VAL3_BUSY_MASK 0x100
+#define SX_DEBUG_BUSY__POS_BANK2VAL3_BUSY__SHIFT 0x8
+#define SX_DEBUG_BUSY__POS_BANK2VAL2_BUSY_MASK 0x200
+#define SX_DEBUG_BUSY__POS_BANK2VAL2_BUSY__SHIFT 0x9
+#define SX_DEBUG_BUSY__POS_BANK2VAL1_BUSY_MASK 0x400
+#define SX_DEBUG_BUSY__POS_BANK2VAL1_BUSY__SHIFT 0xa
+#define SX_DEBUG_BUSY__POS_BANK2VAL0_BUSY_MASK 0x800
+#define SX_DEBUG_BUSY__POS_BANK2VAL0_BUSY__SHIFT 0xb
+#define SX_DEBUG_BUSY__POS_BANK1VAL3_BUSY_MASK 0x1000
+#define SX_DEBUG_BUSY__POS_BANK1VAL3_BUSY__SHIFT 0xc
+#define SX_DEBUG_BUSY__POS_BANK1VAL2_BUSY_MASK 0x2000
+#define SX_DEBUG_BUSY__POS_BANK1VAL2_BUSY__SHIFT 0xd
+#define SX_DEBUG_BUSY__POS_BANK1VAL1_BUSY_MASK 0x4000
+#define SX_DEBUG_BUSY__POS_BANK1VAL1_BUSY__SHIFT 0xe
+#define SX_DEBUG_BUSY__POS_BANK1VAL0_BUSY_MASK 0x8000
+#define SX_DEBUG_BUSY__POS_BANK1VAL0_BUSY__SHIFT 0xf
+#define SX_DEBUG_BUSY__POS_BANK0VAL3_BUSY_MASK 0x10000
+#define SX_DEBUG_BUSY__POS_BANK0VAL3_BUSY__SHIFT 0x10
+#define SX_DEBUG_BUSY__POS_BANK0VAL2_BUSY_MASK 0x20000
+#define SX_DEBUG_BUSY__POS_BANK0VAL2_BUSY__SHIFT 0x11
+#define SX_DEBUG_BUSY__POS_BANK0VAL1_BUSY_MASK 0x40000
+#define SX_DEBUG_BUSY__POS_BANK0VAL1_BUSY__SHIFT 0x12
+#define SX_DEBUG_BUSY__POS_BANK0VAL0_BUSY_MASK 0x80000
+#define SX_DEBUG_BUSY__POS_BANK0VAL0_BUSY__SHIFT 0x13
+#define SX_DEBUG_BUSY__POS_INMUX_VALID_MASK 0x100000
+#define SX_DEBUG_BUSY__POS_INMUX_VALID__SHIFT 0x14
+#define SX_DEBUG_BUSY__WRCTRL1_VALIDQ3_MASK 0x200000
+#define SX_DEBUG_BUSY__WRCTRL1_VALIDQ3__SHIFT 0x15
+#define SX_DEBUG_BUSY__WRCTRL1_VALIDQ2_MASK 0x400000
+#define SX_DEBUG_BUSY__WRCTRL1_VALIDQ2__SHIFT 0x16
+#define SX_DEBUG_BUSY__WRCTRL1_VALIDQ1_MASK 0x800000
+#define SX_DEBUG_BUSY__WRCTRL1_VALIDQ1__SHIFT 0x17
+#define SX_DEBUG_BUSY__WRCTRL0_VALIDQ3_MASK 0x1000000
+#define SX_DEBUG_BUSY__WRCTRL0_VALIDQ3__SHIFT 0x18
+#define SX_DEBUG_BUSY__WRCTRL0_VALIDQ2_MASK 0x2000000
+#define SX_DEBUG_BUSY__WRCTRL0_VALIDQ2__SHIFT 0x19
+#define SX_DEBUG_BUSY__WRCTRL0_VALIDQ1_MASK 0x4000000
+#define SX_DEBUG_BUSY__WRCTRL0_VALIDQ1__SHIFT 0x1a
+#define SX_DEBUG_BUSY__PCCMD_VALID_MASK 0x8000000
+#define SX_DEBUG_BUSY__PCCMD_VALID__SHIFT 0x1b
+#define SX_DEBUG_BUSY__VDATA1_VALID_MASK 0x10000000
+#define SX_DEBUG_BUSY__VDATA1_VALID__SHIFT 0x1c
+#define SX_DEBUG_BUSY__VDATA0_VALID_MASK 0x20000000
+#define SX_DEBUG_BUSY__VDATA0_VALID__SHIFT 0x1d
+#define SX_DEBUG_BUSY__CMD_BUSYORVAL_MASK 0x40000000
+#define SX_DEBUG_BUSY__CMD_BUSYORVAL__SHIFT 0x1e
+#define SX_DEBUG_BUSY__ADDR_BUSYORVAL_MASK 0x80000000
+#define SX_DEBUG_BUSY__ADDR_BUSYORVAL__SHIFT 0x1f
+#define SX_DEBUG_BUSY_2__COL_SCBD_BUSY_MASK 0x1
+#define SX_DEBUG_BUSY_2__COL_SCBD_BUSY__SHIFT 0x0
+#define SX_DEBUG_BUSY_2__COL_REQ3_FREECNT_NE0_MASK 0x2
+#define SX_DEBUG_BUSY_2__COL_REQ3_FREECNT_NE0__SHIFT 0x1
+#define SX_DEBUG_BUSY_2__COL_REQ3_IDLE_MASK 0x4
+#define SX_DEBUG_BUSY_2__COL_REQ3_IDLE__SHIFT 0x2
+#define SX_DEBUG_BUSY_2__COL_REQ3_BUSY_MASK 0x8
+#define SX_DEBUG_BUSY_2__COL_REQ3_BUSY__SHIFT 0x3
+#define SX_DEBUG_BUSY_2__COL_REQ2_FREECNT_NE0_MASK 0x10
+#define SX_DEBUG_BUSY_2__COL_REQ2_FREECNT_NE0__SHIFT 0x4
+#define SX_DEBUG_BUSY_2__COL_REQ2_IDLE_MASK 0x20
+#define SX_DEBUG_BUSY_2__COL_REQ2_IDLE__SHIFT 0x5
+#define SX_DEBUG_BUSY_2__COL_REQ2_BUSY_MASK 0x40
+#define SX_DEBUG_BUSY_2__COL_REQ2_BUSY__SHIFT 0x6
+#define SX_DEBUG_BUSY_2__COL_REQ1_FREECNT_NE0_MASK 0x80
+#define SX_DEBUG_BUSY_2__COL_REQ1_FREECNT_NE0__SHIFT 0x7
+#define SX_DEBUG_BUSY_2__COL_REQ1_IDLE_MASK 0x100
+#define SX_DEBUG_BUSY_2__COL_REQ1_IDLE__SHIFT 0x8
+#define SX_DEBUG_BUSY_2__COL_REQ1_BUSY_MASK 0x200
+#define SX_DEBUG_BUSY_2__COL_REQ1_BUSY__SHIFT 0x9
+#define SX_DEBUG_BUSY_2__COL_REQ0_FREECNT_NE0_MASK 0x400
+#define SX_DEBUG_BUSY_2__COL_REQ0_FREECNT_NE0__SHIFT 0xa
+#define SX_DEBUG_BUSY_2__COL_REQ0_IDLE_MASK 0x800
+#define SX_DEBUG_BUSY_2__COL_REQ0_IDLE__SHIFT 0xb
+#define SX_DEBUG_BUSY_2__COL_REQ0_BUSY_MASK 0x1000
+#define SX_DEBUG_BUSY_2__COL_REQ0_BUSY__SHIFT 0xc
+#define SX_DEBUG_BUSY_2__COL_DBIF3_SENDFREE_BUSY_MASK 0x2000
+#define SX_DEBUG_BUSY_2__COL_DBIF3_SENDFREE_BUSY__SHIFT 0xd
+#define SX_DEBUG_BUSY_2__COL_DBIF3_FIFO_BUSY_MASK 0x4000
+#define SX_DEBUG_BUSY_2__COL_DBIF3_FIFO_BUSY__SHIFT 0xe
+#define SX_DEBUG_BUSY_2__COL_DBIF3_READ_VALID_MASK 0x8000
+#define SX_DEBUG_BUSY_2__COL_DBIF3_READ_VALID__SHIFT 0xf
+#define SX_DEBUG_BUSY_2__COL_DBIF2_SENDFREE_BUSY_MASK 0x10000
+#define SX_DEBUG_BUSY_2__COL_DBIF2_SENDFREE_BUSY__SHIFT 0x10
+#define SX_DEBUG_BUSY_2__COL_DBIF2_FIFO_BUSY_MASK 0x20000
+#define SX_DEBUG_BUSY_2__COL_DBIF2_FIFO_BUSY__SHIFT 0x11
+#define SX_DEBUG_BUSY_2__COL_DBIF2_READ_VALID_MASK 0x40000
+#define SX_DEBUG_BUSY_2__COL_DBIF2_READ_VALID__SHIFT 0x12
+#define SX_DEBUG_BUSY_2__COL_DBIF1_SENDFREE_BUSY_MASK 0x80000
+#define SX_DEBUG_BUSY_2__COL_DBIF1_SENDFREE_BUSY__SHIFT 0x13
+#define SX_DEBUG_BUSY_2__COL_DBIF1_FIFO_BUSY_MASK 0x100000
+#define SX_DEBUG_BUSY_2__COL_DBIF1_FIFO_BUSY__SHIFT 0x14
+#define SX_DEBUG_BUSY_2__COL_DBIF1_READ_VALID_MASK 0x200000
+#define SX_DEBUG_BUSY_2__COL_DBIF1_READ_VALID__SHIFT 0x15
+#define SX_DEBUG_BUSY_2__COL_DBIF0_SENDFREE_BUSY_MASK 0x400000
+#define SX_DEBUG_BUSY_2__COL_DBIF0_SENDFREE_BUSY__SHIFT 0x16
+#define SX_DEBUG_BUSY_2__COL_DBIF0_FIFO_BUSY_MASK 0x800000
+#define SX_DEBUG_BUSY_2__COL_DBIF0_FIFO_BUSY__SHIFT 0x17
+#define SX_DEBUG_BUSY_2__COL_DBIF0_READ_VALID_MASK 0x1000000
+#define SX_DEBUG_BUSY_2__COL_DBIF0_READ_VALID__SHIFT 0x18
+#define SX_DEBUG_BUSY_2__COL_BUFF3_BANK3_VAL3_BUSY_MASK 0x2000000
+#define SX_DEBUG_BUSY_2__COL_BUFF3_BANK3_VAL3_BUSY__SHIFT 0x19
+#define SX_DEBUG_BUSY_2__COL_BUFF3_BANK3_VAL2_BUSY_MASK 0x4000000
+#define SX_DEBUG_BUSY_2__COL_BUFF3_BANK3_VAL2_BUSY__SHIFT 0x1a
+#define SX_DEBUG_BUSY_2__COL_BUFF3_BANK3_VAL1_BUSY_MASK 0x8000000
+#define SX_DEBUG_BUSY_2__COL_BUFF3_BANK3_VAL1_BUSY__SHIFT 0x1b
+#define SX_DEBUG_BUSY_2__COL_BUFF3_BANK3_VAL0_BUSY_MASK 0x10000000
+#define SX_DEBUG_BUSY_2__COL_BUFF3_BANK3_VAL0_BUSY__SHIFT 0x1c
+#define SX_DEBUG_BUSY_2__COL_BUFF3_BANK2_VAL3_BUSY_MASK 0x20000000
+#define SX_DEBUG_BUSY_2__COL_BUFF3_BANK2_VAL3_BUSY__SHIFT 0x1d
+#define SX_DEBUG_BUSY_2__COL_BUFF3_BANK2_VAL2_BUSY_MASK 0x40000000
+#define SX_DEBUG_BUSY_2__COL_BUFF3_BANK2_VAL2_BUSY__SHIFT 0x1e
+#define SX_DEBUG_BUSY_2__COL_BUFF3_BANK2_VAL1_BUSY_MASK 0x80000000
+#define SX_DEBUG_BUSY_2__COL_BUFF3_BANK2_VAL1_BUSY__SHIFT 0x1f
+#define SX_DEBUG_BUSY_3__COL_BUFF3_BANK2_VAL0_BUSY_MASK 0x1
+#define SX_DEBUG_BUSY_3__COL_BUFF3_BANK2_VAL0_BUSY__SHIFT 0x0
+#define SX_DEBUG_BUSY_3__COL_BUFF3_BANK1_VAL3_BUSY_MASK 0x2
+#define SX_DEBUG_BUSY_3__COL_BUFF3_BANK1_VAL3_BUSY__SHIFT 0x1
+#define SX_DEBUG_BUSY_3__COL_BUFF3_BANK1_VAL2_BUSY_MASK 0x4
+#define SX_DEBUG_BUSY_3__COL_BUFF3_BANK1_VAL2_BUSY__SHIFT 0x2
+#define SX_DEBUG_BUSY_3__COL_BUFF3_BANK1_VAL1_BUSY_MASK 0x8
+#define SX_DEBUG_BUSY_3__COL_BUFF3_BANK1_VAL1_BUSY__SHIFT 0x3
+#define SX_DEBUG_BUSY_3__COL_BUFF3_BANK1_VAL0_BUSY_MASK 0x10
+#define SX_DEBUG_BUSY_3__COL_BUFF3_BANK1_VAL0_BUSY__SHIFT 0x4
+#define SX_DEBUG_BUSY_3__COL_BUFF3_BANK0_VAL3_BUSY_MASK 0x20
+#define SX_DEBUG_BUSY_3__COL_BUFF3_BANK0_VAL3_BUSY__SHIFT 0x5
+#define SX_DEBUG_BUSY_3__COL_BUFF3_BANK0_VAL2_BUSY_MASK 0x40
+#define SX_DEBUG_BUSY_3__COL_BUFF3_BANK0_VAL2_BUSY__SHIFT 0x6
+#define SX_DEBUG_BUSY_3__COL_BUFF3_BANK0_VAL1_BUSY_MASK 0x80
+#define SX_DEBUG_BUSY_3__COL_BUFF3_BANK0_VAL1_BUSY__SHIFT 0x7
+#define SX_DEBUG_BUSY_3__COL_BUFF3_BANK0_VAL0_BUSY_MASK 0x100
+#define SX_DEBUG_BUSY_3__COL_BUFF3_BANK0_VAL0_BUSY__SHIFT 0x8
+#define SX_DEBUG_BUSY_3__COL_BUFF2_BANK3_VAL3_BUSY_MASK 0x200
+#define SX_DEBUG_BUSY_3__COL_BUFF2_BANK3_VAL3_BUSY__SHIFT 0x9
+#define SX_DEBUG_BUSY_3__COL_BUFF2_BANK3_VAL2_BUSY_MASK 0x400
+#define SX_DEBUG_BUSY_3__COL_BUFF2_BANK3_VAL2_BUSY__SHIFT 0xa
+#define SX_DEBUG_BUSY_3__COL_BUFF2_BANK3_VAL1_BUSY_MASK 0x800
+#define SX_DEBUG_BUSY_3__COL_BUFF2_BANK3_VAL1_BUSY__SHIFT 0xb
+#define SX_DEBUG_BUSY_3__COL_BUFF2_BANK3_VAL0_BUSY_MASK 0x1000
+#define SX_DEBUG_BUSY_3__COL_BUFF2_BANK3_VAL0_BUSY__SHIFT 0xc
+#define SX_DEBUG_BUSY_3__COL_BUFF2_BANK2_VAL3_BUSY_MASK 0x2000
+#define SX_DEBUG_BUSY_3__COL_BUFF2_BANK2_VAL3_BUSY__SHIFT 0xd
+#define SX_DEBUG_BUSY_3__COL_BUFF2_BANK2_VAL2_BUSY_MASK 0x4000
+#define SX_DEBUG_BUSY_3__COL_BUFF2_BANK2_VAL2_BUSY__SHIFT 0xe
+#define SX_DEBUG_BUSY_3__COL_BUFF2_BANK2_VAL1_BUSY_MASK 0x8000
+#define SX_DEBUG_BUSY_3__COL_BUFF2_BANK2_VAL1_BUSY__SHIFT 0xf
+#define SX_DEBUG_BUSY_3__COL_BUFF2_BANK2_VAL0_BUSY_MASK 0x10000
+#define SX_DEBUG_BUSY_3__COL_BUFF2_BANK2_VAL0_BUSY__SHIFT 0x10
+#define SX_DEBUG_BUSY_3__COL_BUFF2_BANK1_VAL3_BUSY_MASK 0x20000
+#define SX_DEBUG_BUSY_3__COL_BUFF2_BANK1_VAL3_BUSY__SHIFT 0x11
+#define SX_DEBUG_BUSY_3__COL_BUFF2_BANK1_VAL2_BUSY_MASK 0x40000
+#define SX_DEBUG_BUSY_3__COL_BUFF2_BANK1_VAL2_BUSY__SHIFT 0x12
+#define SX_DEBUG_BUSY_3__COL_BUFF2_BANK1_VAL1_BUSY_MASK 0x80000
+#define SX_DEBUG_BUSY_3__COL_BUFF2_BANK1_VAL1_BUSY__SHIFT 0x13
+#define SX_DEBUG_BUSY_3__COL_BUFF2_BANK1_VAL0_BUSY_MASK 0x100000
+#define SX_DEBUG_BUSY_3__COL_BUFF2_BANK1_VAL0_BUSY__SHIFT 0x14
+#define SX_DEBUG_BUSY_3__COL_BUFF2_BANK0_VAL3_BUSY_MASK 0x200000
+#define SX_DEBUG_BUSY_3__COL_BUFF2_BANK0_VAL3_BUSY__SHIFT 0x15
+#define SX_DEBUG_BUSY_3__COL_BUFF2_BANK0_VAL2_BUSY_MASK 0x400000
+#define SX_DEBUG_BUSY_3__COL_BUFF2_BANK0_VAL2_BUSY__SHIFT 0x16
+#define SX_DEBUG_BUSY_3__COL_BUFF2_BANK0_VAL1_BUSY_MASK 0x800000
+#define SX_DEBUG_BUSY_3__COL_BUFF2_BANK0_VAL1_BUSY__SHIFT 0x17
+#define SX_DEBUG_BUSY_3__COL_BUFF2_BANK0_VAL0_BUSY_MASK 0x1000000
+#define SX_DEBUG_BUSY_3__COL_BUFF2_BANK0_VAL0_BUSY__SHIFT 0x18
+#define SX_DEBUG_BUSY_3__COL_BUFF1_BANK3_VAL3_BUSY_MASK 0x2000000
+#define SX_DEBUG_BUSY_3__COL_BUFF1_BANK3_VAL3_BUSY__SHIFT 0x19
+#define SX_DEBUG_BUSY_3__COL_BUFF1_BANK3_VAL2_BUSY_MASK 0x4000000
+#define SX_DEBUG_BUSY_3__COL_BUFF1_BANK3_VAL2_BUSY__SHIFT 0x1a
+#define SX_DEBUG_BUSY_3__COL_BUFF1_BANK3_VAL1_BUSY_MASK 0x8000000
+#define SX_DEBUG_BUSY_3__COL_BUFF1_BANK3_VAL1_BUSY__SHIFT 0x1b
+#define SX_DEBUG_BUSY_3__COL_BUFF1_BANK3_VAL0_BUSY_MASK 0x10000000
+#define SX_DEBUG_BUSY_3__COL_BUFF1_BANK3_VAL0_BUSY__SHIFT 0x1c
+#define SX_DEBUG_BUSY_3__COL_BUFF1_BANK2_VAL3_BUSY_MASK 0x20000000
+#define SX_DEBUG_BUSY_3__COL_BUFF1_BANK2_VAL3_BUSY__SHIFT 0x1d
+#define SX_DEBUG_BUSY_3__COL_BUFF1_BANK2_VAL2_BUSY_MASK 0x40000000
+#define SX_DEBUG_BUSY_3__COL_BUFF1_BANK2_VAL2_BUSY__SHIFT 0x1e
+#define SX_DEBUG_BUSY_3__COL_BUFF1_BANK2_VAL1_BUSY_MASK 0x80000000
+#define SX_DEBUG_BUSY_3__COL_BUFF1_BANK2_VAL1_BUSY__SHIFT 0x1f
+#define SX_DEBUG_BUSY_4__COL_BUFF1_BANK2_VAL0_BUSY_MASK 0x1
+#define SX_DEBUG_BUSY_4__COL_BUFF1_BANK2_VAL0_BUSY__SHIFT 0x0
+#define SX_DEBUG_BUSY_4__COL_BUFF1_BANK1_VAL3_BUSY_MASK 0x2
+#define SX_DEBUG_BUSY_4__COL_BUFF1_BANK1_VAL3_BUSY__SHIFT 0x1
+#define SX_DEBUG_BUSY_4__COL_BUFF1_BANK1_VAL2_BUSY_MASK 0x4
+#define SX_DEBUG_BUSY_4__COL_BUFF1_BANK1_VAL2_BUSY__SHIFT 0x2
+#define SX_DEBUG_BUSY_4__COL_BUFF1_BANK1_VAL1_BUSY_MASK 0x8
+#define SX_DEBUG_BUSY_4__COL_BUFF1_BANK1_VAL1_BUSY__SHIFT 0x3
+#define SX_DEBUG_BUSY_4__COL_BUFF1_BANK1_VAL0_BUSY_MASK 0x10
+#define SX_DEBUG_BUSY_4__COL_BUFF1_BANK1_VAL0_BUSY__SHIFT 0x4
+#define SX_DEBUG_BUSY_4__COL_BUFF1_BANK0_VAL3_BUSY_MASK 0x20
+#define SX_DEBUG_BUSY_4__COL_BUFF1_BANK0_VAL3_BUSY__SHIFT 0x5
+#define SX_DEBUG_BUSY_4__COL_BUFF1_BANK0_VAL2_BUSY_MASK 0x40
+#define SX_DEBUG_BUSY_4__COL_BUFF1_BANK0_VAL2_BUSY__SHIFT 0x6
+#define SX_DEBUG_BUSY_4__COL_BUFF1_BANK0_VAL1_BUSY_MASK 0x80
+#define SX_DEBUG_BUSY_4__COL_BUFF1_BANK0_VAL1_BUSY__SHIFT 0x7
+#define SX_DEBUG_BUSY_4__COL_BUFF1_BANK0_VAL0_BUSY_MASK 0x100
+#define SX_DEBUG_BUSY_4__COL_BUFF1_BANK0_VAL0_BUSY__SHIFT 0x8
+#define SX_DEBUG_BUSY_4__COL_BUFF0_BANK3_VAL3_BUSY_MASK 0x200
+#define SX_DEBUG_BUSY_4__COL_BUFF0_BANK3_VAL3_BUSY__SHIFT 0x9
+#define SX_DEBUG_BUSY_4__COL_BUFF0_BANK3_VAL2_BUSY_MASK 0x400
+#define SX_DEBUG_BUSY_4__COL_BUFF0_BANK3_VAL2_BUSY__SHIFT 0xa
+#define SX_DEBUG_BUSY_4__COL_BUFF0_BANK3_VAL1_BUSY_MASK 0x800
+#define SX_DEBUG_BUSY_4__COL_BUFF0_BANK3_VAL1_BUSY__SHIFT 0xb
+#define SX_DEBUG_BUSY_4__COL_BUFF0_BANK3_VAL0_BUSY_MASK 0x1000
+#define SX_DEBUG_BUSY_4__COL_BUFF0_BANK3_VAL0_BUSY__SHIFT 0xc
+#define SX_DEBUG_BUSY_4__COL_BUFF0_BANK2_VAL3_BUSY_MASK 0x2000
+#define SX_DEBUG_BUSY_4__COL_BUFF0_BANK2_VAL3_BUSY__SHIFT 0xd
+#define SX_DEBUG_BUSY_4__COL_BUFF0_BANK2_VAL2_BUSY_MASK 0x4000
+#define SX_DEBUG_BUSY_4__COL_BUFF0_BANK2_VAL2_BUSY__SHIFT 0xe
+#define SX_DEBUG_BUSY_4__COL_BUFF0_BANK2_VAL1_BUSY_MASK 0x8000
+#define SX_DEBUG_BUSY_4__COL_BUFF0_BANK2_VAL1_BUSY__SHIFT 0xf
+#define SX_DEBUG_BUSY_4__COL_BUFF0_BANK2_VAL0_BUSY_MASK 0x10000
+#define SX_DEBUG_BUSY_4__COL_BUFF0_BANK2_VAL0_BUSY__SHIFT 0x10
+#define SX_DEBUG_BUSY_4__COL_BUFF0_BANK1_VAL3_BUSY_MASK 0x20000
+#define SX_DEBUG_BUSY_4__COL_BUFF0_BANK1_VAL3_BUSY__SHIFT 0x11
+#define SX_DEBUG_BUSY_4__COL_BUFF0_BANK1_VAL2_BUSY_MASK 0x40000
+#define SX_DEBUG_BUSY_4__COL_BUFF0_BANK1_VAL2_BUSY__SHIFT 0x12
+#define SX_DEBUG_BUSY_4__COL_BUFF0_BANK1_VAL1_BUSY_MASK 0x80000
+#define SX_DEBUG_BUSY_4__COL_BUFF0_BANK1_VAL1_BUSY__SHIFT 0x13
+#define SX_DEBUG_BUSY_4__COL_BUFF0_BANK1_VAL0_BUSY_MASK 0x100000
+#define SX_DEBUG_BUSY_4__COL_BUFF0_BANK1_VAL0_BUSY__SHIFT 0x14
+#define SX_DEBUG_BUSY_4__COL_BUFF0_BANK0_VAL3_BUSY_MASK 0x200000
+#define SX_DEBUG_BUSY_4__COL_BUFF0_BANK0_VAL3_BUSY__SHIFT 0x15
+#define SX_DEBUG_BUSY_4__COL_BUFF0_BANK0_VAL2_BUSY_MASK 0x400000
+#define SX_DEBUG_BUSY_4__COL_BUFF0_BANK0_VAL2_BUSY__SHIFT 0x16
+#define SX_DEBUG_BUSY_4__COL_BUFF0_BANK0_VAL1_BUSY_MASK 0x800000
+#define SX_DEBUG_BUSY_4__COL_BUFF0_BANK0_VAL1_BUSY__SHIFT 0x17
+#define SX_DEBUG_BUSY_4__COL_BUFF0_BANK0_VAL0_BUSY_MASK 0x1000000
+#define SX_DEBUG_BUSY_4__COL_BUFF0_BANK0_VAL0_BUSY__SHIFT 0x18
+#define SX_DEBUG_BUSY_4__RESERVED_MASK 0xfe000000
+#define SX_DEBUG_BUSY_4__RESERVED__SHIFT 0x19
+#define SX_DEBUG_1__SX_DB_QUAD_CREDIT_MASK 0x7f
+#define SX_DEBUG_1__SX_DB_QUAD_CREDIT__SHIFT 0x0
+#define SX_DEBUG_1__DISABLE_BLEND_OPT_DONT_RD_DST_MASK 0x100
+#define SX_DEBUG_1__DISABLE_BLEND_OPT_DONT_RD_DST__SHIFT 0x8
+#define SX_DEBUG_1__DISABLE_BLEND_OPT_BYPASS_MASK 0x200
+#define SX_DEBUG_1__DISABLE_BLEND_OPT_BYPASS__SHIFT 0x9
+#define SX_DEBUG_1__DISABLE_BLEND_OPT_DISCARD_PIXEL_MASK 0x400
+#define SX_DEBUG_1__DISABLE_BLEND_OPT_DISCARD_PIXEL__SHIFT 0xa
+#define SX_DEBUG_1__DISABLE_QUAD_PAIR_OPT_MASK 0x800
+#define SX_DEBUG_1__DISABLE_QUAD_PAIR_OPT__SHIFT 0xb
+#define SX_DEBUG_1__DISABLE_PIX_EN_ZERO_OPT_MASK 0x1000
+#define SX_DEBUG_1__DISABLE_PIX_EN_ZERO_OPT__SHIFT 0xc
+#define SX_DEBUG_1__DEBUG_DATA_MASK 0xffffe000
+#define SX_DEBUG_1__DEBUG_DATA__SHIFT 0xd
+#define SX_PERFCOUNTER0_SELECT__PERFCOUNTER_SELECT_MASK 0x3ff
+#define SX_PERFCOUNTER0_SELECT__PERFCOUNTER_SELECT__SHIFT 0x0
+#define SX_PERFCOUNTER0_SELECT__PERFCOUNTER_SELECT1_MASK 0xffc00
+#define SX_PERFCOUNTER0_SELECT__PERFCOUNTER_SELECT1__SHIFT 0xa
+#define SX_PERFCOUNTER0_SELECT__CNTR_MODE_MASK 0xf00000
+#define SX_PERFCOUNTER0_SELECT__CNTR_MODE__SHIFT 0x14
+#define SX_PERFCOUNTER1_SELECT__PERFCOUNTER_SELECT_MASK 0x3ff
+#define SX_PERFCOUNTER1_SELECT__PERFCOUNTER_SELECT__SHIFT 0x0
+#define SX_PERFCOUNTER1_SELECT__PERFCOUNTER_SELECT1_MASK 0xffc00
+#define SX_PERFCOUNTER1_SELECT__PERFCOUNTER_SELECT1__SHIFT 0xa
+#define SX_PERFCOUNTER1_SELECT__CNTR_MODE_MASK 0xf00000
+#define SX_PERFCOUNTER1_SELECT__CNTR_MODE__SHIFT 0x14
+#define SX_PERFCOUNTER2_SELECT__PERFCOUNTER_SELECT_MASK 0x3ff
+#define SX_PERFCOUNTER2_SELECT__PERFCOUNTER_SELECT__SHIFT 0x0
+#define SX_PERFCOUNTER2_SELECT__PERFCOUNTER_SELECT1_MASK 0xffc00
+#define SX_PERFCOUNTER2_SELECT__PERFCOUNTER_SELECT1__SHIFT 0xa
+#define SX_PERFCOUNTER2_SELECT__CNTR_MODE_MASK 0xf00000
+#define SX_PERFCOUNTER2_SELECT__CNTR_MODE__SHIFT 0x14
+#define SX_PERFCOUNTER3_SELECT__PERFCOUNTER_SELECT_MASK 0x3ff
+#define SX_PERFCOUNTER3_SELECT__PERFCOUNTER_SELECT__SHIFT 0x0
+#define SX_PERFCOUNTER3_SELECT__PERFCOUNTER_SELECT1_MASK 0xffc00
+#define SX_PERFCOUNTER3_SELECT__PERFCOUNTER_SELECT1__SHIFT 0xa
+#define SX_PERFCOUNTER3_SELECT__CNTR_MODE_MASK 0xf00000
+#define SX_PERFCOUNTER3_SELECT__CNTR_MODE__SHIFT 0x14
+#define SX_PERFCOUNTER0_SELECT1__PERFCOUNTER_SELECT2_MASK 0x3ff
+#define SX_PERFCOUNTER0_SELECT1__PERFCOUNTER_SELECT2__SHIFT 0x0
+#define SX_PERFCOUNTER0_SELECT1__PERFCOUNTER_SELECT3_MASK 0xffc00
+#define SX_PERFCOUNTER0_SELECT1__PERFCOUNTER_SELECT3__SHIFT 0xa
+#define SX_PERFCOUNTER1_SELECT1__PERFCOUNTER_SELECT2_MASK 0x3ff
+#define SX_PERFCOUNTER1_SELECT1__PERFCOUNTER_SELECT2__SHIFT 0x0
+#define SX_PERFCOUNTER1_SELECT1__PERFCOUNTER_SELECT3_MASK 0xffc00
+#define SX_PERFCOUNTER1_SELECT1__PERFCOUNTER_SELECT3__SHIFT 0xa
+#define SX_PERFCOUNTER0_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define SX_PERFCOUNTER0_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define SX_PERFCOUNTER0_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define SX_PERFCOUNTER0_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define SX_PERFCOUNTER1_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define SX_PERFCOUNTER1_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define SX_PERFCOUNTER1_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define SX_PERFCOUNTER1_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define SX_PERFCOUNTER2_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define SX_PERFCOUNTER2_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define SX_PERFCOUNTER2_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define SX_PERFCOUNTER2_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define SX_PERFCOUNTER3_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define SX_PERFCOUNTER3_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define SX_PERFCOUNTER3_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define SX_PERFCOUNTER3_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define SX_PS_DOWNCONVERT__MRT0_MASK 0xf
+#define SX_PS_DOWNCONVERT__MRT0__SHIFT 0x0
+#define SX_PS_DOWNCONVERT__MRT1_MASK 0xf0
+#define SX_PS_DOWNCONVERT__MRT1__SHIFT 0x4
+#define SX_PS_DOWNCONVERT__MRT2_MASK 0xf00
+#define SX_PS_DOWNCONVERT__MRT2__SHIFT 0x8
+#define SX_PS_DOWNCONVERT__MRT3_MASK 0xf000
+#define SX_PS_DOWNCONVERT__MRT3__SHIFT 0xc
+#define SX_PS_DOWNCONVERT__MRT4_MASK 0xf0000
+#define SX_PS_DOWNCONVERT__MRT4__SHIFT 0x10
+#define SX_PS_DOWNCONVERT__MRT5_MASK 0xf00000
+#define SX_PS_DOWNCONVERT__MRT5__SHIFT 0x14
+#define SX_PS_DOWNCONVERT__MRT6_MASK 0xf000000
+#define SX_PS_DOWNCONVERT__MRT6__SHIFT 0x18
+#define SX_PS_DOWNCONVERT__MRT7_MASK 0xf0000000
+#define SX_PS_DOWNCONVERT__MRT7__SHIFT 0x1c
+#define SX_BLEND_OPT_EPSILON__MRT0_EPSILON_MASK 0xf
+#define SX_BLEND_OPT_EPSILON__MRT0_EPSILON__SHIFT 0x0
+#define SX_BLEND_OPT_EPSILON__MRT1_EPSILON_MASK 0xf0
+#define SX_BLEND_OPT_EPSILON__MRT1_EPSILON__SHIFT 0x4
+#define SX_BLEND_OPT_EPSILON__MRT2_EPSILON_MASK 0xf00
+#define SX_BLEND_OPT_EPSILON__MRT2_EPSILON__SHIFT 0x8
+#define SX_BLEND_OPT_EPSILON__MRT3_EPSILON_MASK 0xf000
+#define SX_BLEND_OPT_EPSILON__MRT3_EPSILON__SHIFT 0xc
+#define SX_BLEND_OPT_EPSILON__MRT4_EPSILON_MASK 0xf0000
+#define SX_BLEND_OPT_EPSILON__MRT4_EPSILON__SHIFT 0x10
+#define SX_BLEND_OPT_EPSILON__MRT5_EPSILON_MASK 0xf00000
+#define SX_BLEND_OPT_EPSILON__MRT5_EPSILON__SHIFT 0x14
+#define SX_BLEND_OPT_EPSILON__MRT6_EPSILON_MASK 0xf000000
+#define SX_BLEND_OPT_EPSILON__MRT6_EPSILON__SHIFT 0x18
+#define SX_BLEND_OPT_EPSILON__MRT7_EPSILON_MASK 0xf0000000
+#define SX_BLEND_OPT_EPSILON__MRT7_EPSILON__SHIFT 0x1c
+#define SX_BLEND_OPT_CONTROL__MRT0_COLOR_OPT_DISABLE_MASK 0x1
+#define SX_BLEND_OPT_CONTROL__MRT0_COLOR_OPT_DISABLE__SHIFT 0x0
+#define SX_BLEND_OPT_CONTROL__MRT0_ALPHA_OPT_DISABLE_MASK 0x2
+#define SX_BLEND_OPT_CONTROL__MRT0_ALPHA_OPT_DISABLE__SHIFT 0x1
+#define SX_BLEND_OPT_CONTROL__MRT1_COLOR_OPT_DISABLE_MASK 0x10
+#define SX_BLEND_OPT_CONTROL__MRT1_COLOR_OPT_DISABLE__SHIFT 0x4
+#define SX_BLEND_OPT_CONTROL__MRT1_ALPHA_OPT_DISABLE_MASK 0x20
+#define SX_BLEND_OPT_CONTROL__MRT1_ALPHA_OPT_DISABLE__SHIFT 0x5
+#define SX_BLEND_OPT_CONTROL__MRT2_COLOR_OPT_DISABLE_MASK 0x100
+#define SX_BLEND_OPT_CONTROL__MRT2_COLOR_OPT_DISABLE__SHIFT 0x8
+#define SX_BLEND_OPT_CONTROL__MRT2_ALPHA_OPT_DISABLE_MASK 0x200
+#define SX_BLEND_OPT_CONTROL__MRT2_ALPHA_OPT_DISABLE__SHIFT 0x9
+#define SX_BLEND_OPT_CONTROL__MRT3_COLOR_OPT_DISABLE_MASK 0x1000
+#define SX_BLEND_OPT_CONTROL__MRT3_COLOR_OPT_DISABLE__SHIFT 0xc
+#define SX_BLEND_OPT_CONTROL__MRT3_ALPHA_OPT_DISABLE_MASK 0x2000
+#define SX_BLEND_OPT_CONTROL__MRT3_ALPHA_OPT_DISABLE__SHIFT 0xd
+#define SX_BLEND_OPT_CONTROL__MRT4_COLOR_OPT_DISABLE_MASK 0x10000
+#define SX_BLEND_OPT_CONTROL__MRT4_COLOR_OPT_DISABLE__SHIFT 0x10
+#define SX_BLEND_OPT_CONTROL__MRT4_ALPHA_OPT_DISABLE_MASK 0x20000
+#define SX_BLEND_OPT_CONTROL__MRT4_ALPHA_OPT_DISABLE__SHIFT 0x11
+#define SX_BLEND_OPT_CONTROL__MRT5_COLOR_OPT_DISABLE_MASK 0x100000
+#define SX_BLEND_OPT_CONTROL__MRT5_COLOR_OPT_DISABLE__SHIFT 0x14
+#define SX_BLEND_OPT_CONTROL__MRT5_ALPHA_OPT_DISABLE_MASK 0x200000
+#define SX_BLEND_OPT_CONTROL__MRT5_ALPHA_OPT_DISABLE__SHIFT 0x15
+#define SX_BLEND_OPT_CONTROL__MRT6_COLOR_OPT_DISABLE_MASK 0x1000000
+#define SX_BLEND_OPT_CONTROL__MRT6_COLOR_OPT_DISABLE__SHIFT 0x18
+#define SX_BLEND_OPT_CONTROL__MRT6_ALPHA_OPT_DISABLE_MASK 0x2000000
+#define SX_BLEND_OPT_CONTROL__MRT6_ALPHA_OPT_DISABLE__SHIFT 0x19
+#define SX_BLEND_OPT_CONTROL__MRT7_COLOR_OPT_DISABLE_MASK 0x10000000
+#define SX_BLEND_OPT_CONTROL__MRT7_COLOR_OPT_DISABLE__SHIFT 0x1c
+#define SX_BLEND_OPT_CONTROL__MRT7_ALPHA_OPT_DISABLE_MASK 0x20000000
+#define SX_BLEND_OPT_CONTROL__MRT7_ALPHA_OPT_DISABLE__SHIFT 0x1d
+#define SX_BLEND_OPT_CONTROL__PIXEN_ZERO_OPT_DISABLE_MASK 0x80000000
+#define SX_BLEND_OPT_CONTROL__PIXEN_ZERO_OPT_DISABLE__SHIFT 0x1f
+#define SX_MRT0_BLEND_OPT__COLOR_SRC_OPT_MASK 0x7
+#define SX_MRT0_BLEND_OPT__COLOR_SRC_OPT__SHIFT 0x0
+#define SX_MRT0_BLEND_OPT__COLOR_DST_OPT_MASK 0x70
+#define SX_MRT0_BLEND_OPT__COLOR_DST_OPT__SHIFT 0x4
+#define SX_MRT0_BLEND_OPT__COLOR_COMB_FCN_MASK 0x700
+#define SX_MRT0_BLEND_OPT__COLOR_COMB_FCN__SHIFT 0x8
+#define SX_MRT0_BLEND_OPT__ALPHA_SRC_OPT_MASK 0x70000
+#define SX_MRT0_BLEND_OPT__ALPHA_SRC_OPT__SHIFT 0x10
+#define SX_MRT0_BLEND_OPT__ALPHA_DST_OPT_MASK 0x700000
+#define SX_MRT0_BLEND_OPT__ALPHA_DST_OPT__SHIFT 0x14
+#define SX_MRT0_BLEND_OPT__ALPHA_COMB_FCN_MASK 0x7000000
+#define SX_MRT0_BLEND_OPT__ALPHA_COMB_FCN__SHIFT 0x18
+#define SX_MRT1_BLEND_OPT__COLOR_SRC_OPT_MASK 0x7
+#define SX_MRT1_BLEND_OPT__COLOR_SRC_OPT__SHIFT 0x0
+#define SX_MRT1_BLEND_OPT__COLOR_DST_OPT_MASK 0x70
+#define SX_MRT1_BLEND_OPT__COLOR_DST_OPT__SHIFT 0x4
+#define SX_MRT1_BLEND_OPT__COLOR_COMB_FCN_MASK 0x700
+#define SX_MRT1_BLEND_OPT__COLOR_COMB_FCN__SHIFT 0x8
+#define SX_MRT1_BLEND_OPT__ALPHA_SRC_OPT_MASK 0x70000
+#define SX_MRT1_BLEND_OPT__ALPHA_SRC_OPT__SHIFT 0x10
+#define SX_MRT1_BLEND_OPT__ALPHA_DST_OPT_MASK 0x700000
+#define SX_MRT1_BLEND_OPT__ALPHA_DST_OPT__SHIFT 0x14
+#define SX_MRT1_BLEND_OPT__ALPHA_COMB_FCN_MASK 0x7000000
+#define SX_MRT1_BLEND_OPT__ALPHA_COMB_FCN__SHIFT 0x18
+#define SX_MRT2_BLEND_OPT__COLOR_SRC_OPT_MASK 0x7
+#define SX_MRT2_BLEND_OPT__COLOR_SRC_OPT__SHIFT 0x0
+#define SX_MRT2_BLEND_OPT__COLOR_DST_OPT_MASK 0x70
+#define SX_MRT2_BLEND_OPT__COLOR_DST_OPT__SHIFT 0x4
+#define SX_MRT2_BLEND_OPT__COLOR_COMB_FCN_MASK 0x700
+#define SX_MRT2_BLEND_OPT__COLOR_COMB_FCN__SHIFT 0x8
+#define SX_MRT2_BLEND_OPT__ALPHA_SRC_OPT_MASK 0x70000
+#define SX_MRT2_BLEND_OPT__ALPHA_SRC_OPT__SHIFT 0x10
+#define SX_MRT2_BLEND_OPT__ALPHA_DST_OPT_MASK 0x700000
+#define SX_MRT2_BLEND_OPT__ALPHA_DST_OPT__SHIFT 0x14
+#define SX_MRT2_BLEND_OPT__ALPHA_COMB_FCN_MASK 0x7000000
+#define SX_MRT2_BLEND_OPT__ALPHA_COMB_FCN__SHIFT 0x18
+#define SX_MRT3_BLEND_OPT__COLOR_SRC_OPT_MASK 0x7
+#define SX_MRT3_BLEND_OPT__COLOR_SRC_OPT__SHIFT 0x0
+#define SX_MRT3_BLEND_OPT__COLOR_DST_OPT_MASK 0x70
+#define SX_MRT3_BLEND_OPT__COLOR_DST_OPT__SHIFT 0x4
+#define SX_MRT3_BLEND_OPT__COLOR_COMB_FCN_MASK 0x700
+#define SX_MRT3_BLEND_OPT__COLOR_COMB_FCN__SHIFT 0x8
+#define SX_MRT3_BLEND_OPT__ALPHA_SRC_OPT_MASK 0x70000
+#define SX_MRT3_BLEND_OPT__ALPHA_SRC_OPT__SHIFT 0x10
+#define SX_MRT3_BLEND_OPT__ALPHA_DST_OPT_MASK 0x700000
+#define SX_MRT3_BLEND_OPT__ALPHA_DST_OPT__SHIFT 0x14
+#define SX_MRT3_BLEND_OPT__ALPHA_COMB_FCN_MASK 0x7000000
+#define SX_MRT3_BLEND_OPT__ALPHA_COMB_FCN__SHIFT 0x18
+#define SX_MRT4_BLEND_OPT__COLOR_SRC_OPT_MASK 0x7
+#define SX_MRT4_BLEND_OPT__COLOR_SRC_OPT__SHIFT 0x0
+#define SX_MRT4_BLEND_OPT__COLOR_DST_OPT_MASK 0x70
+#define SX_MRT4_BLEND_OPT__COLOR_DST_OPT__SHIFT 0x4
+#define SX_MRT4_BLEND_OPT__COLOR_COMB_FCN_MASK 0x700
+#define SX_MRT4_BLEND_OPT__COLOR_COMB_FCN__SHIFT 0x8
+#define SX_MRT4_BLEND_OPT__ALPHA_SRC_OPT_MASK 0x70000
+#define SX_MRT4_BLEND_OPT__ALPHA_SRC_OPT__SHIFT 0x10
+#define SX_MRT4_BLEND_OPT__ALPHA_DST_OPT_MASK 0x700000
+#define SX_MRT4_BLEND_OPT__ALPHA_DST_OPT__SHIFT 0x14
+#define SX_MRT4_BLEND_OPT__ALPHA_COMB_FCN_MASK 0x7000000
+#define SX_MRT4_BLEND_OPT__ALPHA_COMB_FCN__SHIFT 0x18
+#define SX_MRT5_BLEND_OPT__COLOR_SRC_OPT_MASK 0x7
+#define SX_MRT5_BLEND_OPT__COLOR_SRC_OPT__SHIFT 0x0
+#define SX_MRT5_BLEND_OPT__COLOR_DST_OPT_MASK 0x70
+#define SX_MRT5_BLEND_OPT__COLOR_DST_OPT__SHIFT 0x4
+#define SX_MRT5_BLEND_OPT__COLOR_COMB_FCN_MASK 0x700
+#define SX_MRT5_BLEND_OPT__COLOR_COMB_FCN__SHIFT 0x8
+#define SX_MRT5_BLEND_OPT__ALPHA_SRC_OPT_MASK 0x70000
+#define SX_MRT5_BLEND_OPT__ALPHA_SRC_OPT__SHIFT 0x10
+#define SX_MRT5_BLEND_OPT__ALPHA_DST_OPT_MASK 0x700000
+#define SX_MRT5_BLEND_OPT__ALPHA_DST_OPT__SHIFT 0x14
+#define SX_MRT5_BLEND_OPT__ALPHA_COMB_FCN_MASK 0x7000000
+#define SX_MRT5_BLEND_OPT__ALPHA_COMB_FCN__SHIFT 0x18
+#define SX_MRT6_BLEND_OPT__COLOR_SRC_OPT_MASK 0x7
+#define SX_MRT6_BLEND_OPT__COLOR_SRC_OPT__SHIFT 0x0
+#define SX_MRT6_BLEND_OPT__COLOR_DST_OPT_MASK 0x70
+#define SX_MRT6_BLEND_OPT__COLOR_DST_OPT__SHIFT 0x4
+#define SX_MRT6_BLEND_OPT__COLOR_COMB_FCN_MASK 0x700
+#define SX_MRT6_BLEND_OPT__COLOR_COMB_FCN__SHIFT 0x8
+#define SX_MRT6_BLEND_OPT__ALPHA_SRC_OPT_MASK 0x70000
+#define SX_MRT6_BLEND_OPT__ALPHA_SRC_OPT__SHIFT 0x10
+#define SX_MRT6_BLEND_OPT__ALPHA_DST_OPT_MASK 0x700000
+#define SX_MRT6_BLEND_OPT__ALPHA_DST_OPT__SHIFT 0x14
+#define SX_MRT6_BLEND_OPT__ALPHA_COMB_FCN_MASK 0x7000000
+#define SX_MRT6_BLEND_OPT__ALPHA_COMB_FCN__SHIFT 0x18
+#define SX_MRT7_BLEND_OPT__COLOR_SRC_OPT_MASK 0x7
+#define SX_MRT7_BLEND_OPT__COLOR_SRC_OPT__SHIFT 0x0
+#define SX_MRT7_BLEND_OPT__COLOR_DST_OPT_MASK 0x70
+#define SX_MRT7_BLEND_OPT__COLOR_DST_OPT__SHIFT 0x4
+#define SX_MRT7_BLEND_OPT__COLOR_COMB_FCN_MASK 0x700
+#define SX_MRT7_BLEND_OPT__COLOR_COMB_FCN__SHIFT 0x8
+#define SX_MRT7_BLEND_OPT__ALPHA_SRC_OPT_MASK 0x70000
+#define SX_MRT7_BLEND_OPT__ALPHA_SRC_OPT__SHIFT 0x10
+#define SX_MRT7_BLEND_OPT__ALPHA_DST_OPT_MASK 0x700000
+#define SX_MRT7_BLEND_OPT__ALPHA_DST_OPT__SHIFT 0x14
+#define SX_MRT7_BLEND_OPT__ALPHA_COMB_FCN_MASK 0x7000000
+#define SX_MRT7_BLEND_OPT__ALPHA_COMB_FCN__SHIFT 0x18
+#define TCC_CTRL__CACHE_SIZE_MASK 0x3
+#define TCC_CTRL__CACHE_SIZE__SHIFT 0x0
+#define TCC_CTRL__RATE_MASK 0xc
+#define TCC_CTRL__RATE__SHIFT 0x2
+#define TCC_CTRL__WRITEBACK_MARGIN_MASK 0xf0
+#define TCC_CTRL__WRITEBACK_MARGIN__SHIFT 0x4
+#define TCC_CTRL__METADATA_LATENCY_FIFO_SIZE_MASK 0xf00
+#define TCC_CTRL__METADATA_LATENCY_FIFO_SIZE__SHIFT 0x8
+#define TCC_CTRL__SRC_FIFO_SIZE_MASK 0xf000
+#define TCC_CTRL__SRC_FIFO_SIZE__SHIFT 0xc
+#define TCC_CTRL__LATENCY_FIFO_SIZE_MASK 0xf0000
+#define TCC_CTRL__LATENCY_FIFO_SIZE__SHIFT 0x10
+#define TCC_CTRL__WB_OR_INV_ALL_VMIDS_MASK 0x100000
+#define TCC_CTRL__WB_OR_INV_ALL_VMIDS__SHIFT 0x14
+#define TCC_CTRL__MDC_SIZE_MASK 0x3000000
+#define TCC_CTRL__MDC_SIZE__SHIFT 0x18
+#define TCC_CTRL__MDC_SECTOR_SIZE_MASK 0xc000000
+#define TCC_CTRL__MDC_SECTOR_SIZE__SHIFT 0x1a
+#define TCC_CTRL__MDC_SIDEBAND_FIFO_SIZE_MASK 0xf0000000
+#define TCC_CTRL__MDC_SIDEBAND_FIFO_SIZE__SHIFT 0x1c
+#define TCC_EDC_CNT__SEC_COUNT_MASK 0xff
+#define TCC_EDC_CNT__SEC_COUNT__SHIFT 0x0
+#define TCC_EDC_CNT__DED_COUNT_MASK 0xff0000
+#define TCC_EDC_CNT__DED_COUNT__SHIFT 0x10
+#define TCC_REDUNDANCY__MC_SEL0_MASK 0x1
+#define TCC_REDUNDANCY__MC_SEL0__SHIFT 0x0
+#define TCC_REDUNDANCY__MC_SEL1_MASK 0x2
+#define TCC_REDUNDANCY__MC_SEL1__SHIFT 0x1
+#define TCC_EXE_DISABLE__EXE_DISABLE_MASK 0x2
+#define TCC_EXE_DISABLE__EXE_DISABLE__SHIFT 0x1
+#define TCC_DSM_CNTL__CACHE_RAM_IRRITATOR_DATA_SEL_MASK 0x3
+#define TCC_DSM_CNTL__CACHE_RAM_IRRITATOR_DATA_SEL__SHIFT 0x0
+#define TCC_DSM_CNTL__CACHE_RAM_IRRITATOR_SINGLE_WRITE_MASK 0x4
+#define TCC_DSM_CNTL__CACHE_RAM_IRRITATOR_SINGLE_WRITE__SHIFT 0x2
+#define TCC_CGTT_SCLK_CTRL__ON_DELAY_MASK 0xf
+#define TCC_CGTT_SCLK_CTRL__ON_DELAY__SHIFT 0x0
+#define TCC_CGTT_SCLK_CTRL__OFF_HYSTERESIS_MASK 0xff0
+#define TCC_CGTT_SCLK_CTRL__OFF_HYSTERESIS__SHIFT 0x4
+#define TCC_CGTT_SCLK_CTRL__SOFT_OVERRIDE7_MASK 0x1000000
+#define TCC_CGTT_SCLK_CTRL__SOFT_OVERRIDE7__SHIFT 0x18
+#define TCC_CGTT_SCLK_CTRL__SOFT_OVERRIDE6_MASK 0x2000000
+#define TCC_CGTT_SCLK_CTRL__SOFT_OVERRIDE6__SHIFT 0x19
+#define TCC_CGTT_SCLK_CTRL__SOFT_OVERRIDE5_MASK 0x4000000
+#define TCC_CGTT_SCLK_CTRL__SOFT_OVERRIDE5__SHIFT 0x1a
+#define TCC_CGTT_SCLK_CTRL__SOFT_OVERRIDE4_MASK 0x8000000
+#define TCC_CGTT_SCLK_CTRL__SOFT_OVERRIDE4__SHIFT 0x1b
+#define TCC_CGTT_SCLK_CTRL__SOFT_OVERRIDE3_MASK 0x10000000
+#define TCC_CGTT_SCLK_CTRL__SOFT_OVERRIDE3__SHIFT 0x1c
+#define TCC_CGTT_SCLK_CTRL__SOFT_OVERRIDE2_MASK 0x20000000
+#define TCC_CGTT_SCLK_CTRL__SOFT_OVERRIDE2__SHIFT 0x1d
+#define TCC_CGTT_SCLK_CTRL__SOFT_OVERRIDE1_MASK 0x40000000
+#define TCC_CGTT_SCLK_CTRL__SOFT_OVERRIDE1__SHIFT 0x1e
+#define TCC_CGTT_SCLK_CTRL__SOFT_OVERRIDE0_MASK 0x80000000
+#define TCC_CGTT_SCLK_CTRL__SOFT_OVERRIDE0__SHIFT 0x1f
+#define TCA_CGTT_SCLK_CTRL__ON_DELAY_MASK 0xf
+#define TCA_CGTT_SCLK_CTRL__ON_DELAY__SHIFT 0x0
+#define TCA_CGTT_SCLK_CTRL__OFF_HYSTERESIS_MASK 0xff0
+#define TCA_CGTT_SCLK_CTRL__OFF_HYSTERESIS__SHIFT 0x4
+#define TCA_CGTT_SCLK_CTRL__SOFT_OVERRIDE7_MASK 0x1000000
+#define TCA_CGTT_SCLK_CTRL__SOFT_OVERRIDE7__SHIFT 0x18
+#define TCA_CGTT_SCLK_CTRL__SOFT_OVERRIDE6_MASK 0x2000000
+#define TCA_CGTT_SCLK_CTRL__SOFT_OVERRIDE6__SHIFT 0x19
+#define TCA_CGTT_SCLK_CTRL__SOFT_OVERRIDE5_MASK 0x4000000
+#define TCA_CGTT_SCLK_CTRL__SOFT_OVERRIDE5__SHIFT 0x1a
+#define TCA_CGTT_SCLK_CTRL__SOFT_OVERRIDE4_MASK 0x8000000
+#define TCA_CGTT_SCLK_CTRL__SOFT_OVERRIDE4__SHIFT 0x1b
+#define TCA_CGTT_SCLK_CTRL__SOFT_OVERRIDE3_MASK 0x10000000
+#define TCA_CGTT_SCLK_CTRL__SOFT_OVERRIDE3__SHIFT 0x1c
+#define TCA_CGTT_SCLK_CTRL__SOFT_OVERRIDE2_MASK 0x20000000
+#define TCA_CGTT_SCLK_CTRL__SOFT_OVERRIDE2__SHIFT 0x1d
+#define TCA_CGTT_SCLK_CTRL__SOFT_OVERRIDE1_MASK 0x40000000
+#define TCA_CGTT_SCLK_CTRL__SOFT_OVERRIDE1__SHIFT 0x1e
+#define TCA_CGTT_SCLK_CTRL__SOFT_OVERRIDE0_MASK 0x80000000
+#define TCA_CGTT_SCLK_CTRL__SOFT_OVERRIDE0__SHIFT 0x1f
+#define TCC_PERFCOUNTER0_SELECT__PERF_SEL_MASK 0x3ff
+#define TCC_PERFCOUNTER0_SELECT__PERF_SEL__SHIFT 0x0
+#define TCC_PERFCOUNTER0_SELECT__PERF_SEL1_MASK 0xffc00
+#define TCC_PERFCOUNTER0_SELECT__PERF_SEL1__SHIFT 0xa
+#define TCC_PERFCOUNTER0_SELECT__CNTR_MODE_MASK 0xf00000
+#define TCC_PERFCOUNTER0_SELECT__CNTR_MODE__SHIFT 0x14
+#define TCC_PERFCOUNTER0_SELECT__PERF_MODE1_MASK 0xf000000
+#define TCC_PERFCOUNTER0_SELECT__PERF_MODE1__SHIFT 0x18
+#define TCC_PERFCOUNTER0_SELECT__PERF_MODE_MASK 0xf0000000
+#define TCC_PERFCOUNTER0_SELECT__PERF_MODE__SHIFT 0x1c
+#define TCC_PERFCOUNTER1_SELECT__PERF_SEL_MASK 0x3ff
+#define TCC_PERFCOUNTER1_SELECT__PERF_SEL__SHIFT 0x0
+#define TCC_PERFCOUNTER1_SELECT__PERF_SEL1_MASK 0xffc00
+#define TCC_PERFCOUNTER1_SELECT__PERF_SEL1__SHIFT 0xa
+#define TCC_PERFCOUNTER1_SELECT__CNTR_MODE_MASK 0xf00000
+#define TCC_PERFCOUNTER1_SELECT__CNTR_MODE__SHIFT 0x14
+#define TCC_PERFCOUNTER1_SELECT__PERF_MODE1_MASK 0xf000000
+#define TCC_PERFCOUNTER1_SELECT__PERF_MODE1__SHIFT 0x18
+#define TCC_PERFCOUNTER1_SELECT__PERF_MODE_MASK 0xf0000000
+#define TCC_PERFCOUNTER1_SELECT__PERF_MODE__SHIFT 0x1c
+#define TCC_PERFCOUNTER0_SELECT1__PERF_SEL2_MASK 0x3ff
+#define TCC_PERFCOUNTER0_SELECT1__PERF_SEL2__SHIFT 0x0
+#define TCC_PERFCOUNTER0_SELECT1__PERF_SEL3_MASK 0xffc00
+#define TCC_PERFCOUNTER0_SELECT1__PERF_SEL3__SHIFT 0xa
+#define TCC_PERFCOUNTER0_SELECT1__PERF_MODE2_MASK 0xf000000
+#define TCC_PERFCOUNTER0_SELECT1__PERF_MODE2__SHIFT 0x18
+#define TCC_PERFCOUNTER0_SELECT1__PERF_MODE3_MASK 0xf0000000
+#define TCC_PERFCOUNTER0_SELECT1__PERF_MODE3__SHIFT 0x1c
+#define TCC_PERFCOUNTER1_SELECT1__PERF_SEL2_MASK 0x3ff
+#define TCC_PERFCOUNTER1_SELECT1__PERF_SEL2__SHIFT 0x0
+#define TCC_PERFCOUNTER1_SELECT1__PERF_SEL3_MASK 0xffc00
+#define TCC_PERFCOUNTER1_SELECT1__PERF_SEL3__SHIFT 0xa
+#define TCC_PERFCOUNTER1_SELECT1__PERF_MODE2_MASK 0xf000000
+#define TCC_PERFCOUNTER1_SELECT1__PERF_MODE2__SHIFT 0x18
+#define TCC_PERFCOUNTER1_SELECT1__PERF_MODE3_MASK 0xf0000000
+#define TCC_PERFCOUNTER1_SELECT1__PERF_MODE3__SHIFT 0x1c
+#define TCC_PERFCOUNTER2_SELECT__PERF_SEL_MASK 0x3ff
+#define TCC_PERFCOUNTER2_SELECT__PERF_SEL__SHIFT 0x0
+#define TCC_PERFCOUNTER2_SELECT__CNTR_MODE_MASK 0xf00000
+#define TCC_PERFCOUNTER2_SELECT__CNTR_MODE__SHIFT 0x14
+#define TCC_PERFCOUNTER2_SELECT__PERF_MODE_MASK 0xf0000000
+#define TCC_PERFCOUNTER2_SELECT__PERF_MODE__SHIFT 0x1c
+#define TCC_PERFCOUNTER3_SELECT__PERF_SEL_MASK 0x3ff
+#define TCC_PERFCOUNTER3_SELECT__PERF_SEL__SHIFT 0x0
+#define TCC_PERFCOUNTER3_SELECT__CNTR_MODE_MASK 0xf00000
+#define TCC_PERFCOUNTER3_SELECT__CNTR_MODE__SHIFT 0x14
+#define TCC_PERFCOUNTER3_SELECT__PERF_MODE_MASK 0xf0000000
+#define TCC_PERFCOUNTER3_SELECT__PERF_MODE__SHIFT 0x1c
+#define TCC_PERFCOUNTER0_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define TCC_PERFCOUNTER0_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define TCC_PERFCOUNTER1_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define TCC_PERFCOUNTER1_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define TCC_PERFCOUNTER2_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define TCC_PERFCOUNTER2_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define TCC_PERFCOUNTER3_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define TCC_PERFCOUNTER3_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define TCC_PERFCOUNTER0_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define TCC_PERFCOUNTER0_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define TCC_PERFCOUNTER1_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define TCC_PERFCOUNTER1_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define TCC_PERFCOUNTER2_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define TCC_PERFCOUNTER2_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define TCC_PERFCOUNTER3_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define TCC_PERFCOUNTER3_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define TCA_CTRL__HOLE_TIMEOUT_MASK 0xf
+#define TCA_CTRL__HOLE_TIMEOUT__SHIFT 0x0
+#define TCA_PERFCOUNTER0_SELECT__PERF_SEL_MASK 0x3ff
+#define TCA_PERFCOUNTER0_SELECT__PERF_SEL__SHIFT 0x0
+#define TCA_PERFCOUNTER0_SELECT__PERF_SEL1_MASK 0xffc00
+#define TCA_PERFCOUNTER0_SELECT__PERF_SEL1__SHIFT 0xa
+#define TCA_PERFCOUNTER0_SELECT__CNTR_MODE_MASK 0xf00000
+#define TCA_PERFCOUNTER0_SELECT__CNTR_MODE__SHIFT 0x14
+#define TCA_PERFCOUNTER0_SELECT__PERF_MODE1_MASK 0xf000000
+#define TCA_PERFCOUNTER0_SELECT__PERF_MODE1__SHIFT 0x18
+#define TCA_PERFCOUNTER0_SELECT__PERF_MODE_MASK 0xf0000000
+#define TCA_PERFCOUNTER0_SELECT__PERF_MODE__SHIFT 0x1c
+#define TCA_PERFCOUNTER1_SELECT__PERF_SEL_MASK 0x3ff
+#define TCA_PERFCOUNTER1_SELECT__PERF_SEL__SHIFT 0x0
+#define TCA_PERFCOUNTER1_SELECT__PERF_SEL1_MASK 0xffc00
+#define TCA_PERFCOUNTER1_SELECT__PERF_SEL1__SHIFT 0xa
+#define TCA_PERFCOUNTER1_SELECT__CNTR_MODE_MASK 0xf00000
+#define TCA_PERFCOUNTER1_SELECT__CNTR_MODE__SHIFT 0x14
+#define TCA_PERFCOUNTER1_SELECT__PERF_MODE1_MASK 0xf000000
+#define TCA_PERFCOUNTER1_SELECT__PERF_MODE1__SHIFT 0x18
+#define TCA_PERFCOUNTER1_SELECT__PERF_MODE_MASK 0xf0000000
+#define TCA_PERFCOUNTER1_SELECT__PERF_MODE__SHIFT 0x1c
+#define TCA_PERFCOUNTER0_SELECT1__PERF_SEL2_MASK 0x3ff
+#define TCA_PERFCOUNTER0_SELECT1__PERF_SEL2__SHIFT 0x0
+#define TCA_PERFCOUNTER0_SELECT1__PERF_SEL3_MASK 0xffc00
+#define TCA_PERFCOUNTER0_SELECT1__PERF_SEL3__SHIFT 0xa
+#define TCA_PERFCOUNTER0_SELECT1__PERF_MODE2_MASK 0xf000000
+#define TCA_PERFCOUNTER0_SELECT1__PERF_MODE2__SHIFT 0x18
+#define TCA_PERFCOUNTER0_SELECT1__PERF_MODE3_MASK 0xf0000000
+#define TCA_PERFCOUNTER0_SELECT1__PERF_MODE3__SHIFT 0x1c
+#define TCA_PERFCOUNTER1_SELECT1__PERF_SEL2_MASK 0x3ff
+#define TCA_PERFCOUNTER1_SELECT1__PERF_SEL2__SHIFT 0x0
+#define TCA_PERFCOUNTER1_SELECT1__PERF_SEL3_MASK 0xffc00
+#define TCA_PERFCOUNTER1_SELECT1__PERF_SEL3__SHIFT 0xa
+#define TCA_PERFCOUNTER1_SELECT1__PERF_MODE2_MASK 0xf000000
+#define TCA_PERFCOUNTER1_SELECT1__PERF_MODE2__SHIFT 0x18
+#define TCA_PERFCOUNTER1_SELECT1__PERF_MODE3_MASK 0xf0000000
+#define TCA_PERFCOUNTER1_SELECT1__PERF_MODE3__SHIFT 0x1c
+#define TCA_PERFCOUNTER2_SELECT__PERF_SEL_MASK 0x3ff
+#define TCA_PERFCOUNTER2_SELECT__PERF_SEL__SHIFT 0x0
+#define TCA_PERFCOUNTER2_SELECT__CNTR_MODE_MASK 0xf00000
+#define TCA_PERFCOUNTER2_SELECT__CNTR_MODE__SHIFT 0x14
+#define TCA_PERFCOUNTER2_SELECT__PERF_MODE_MASK 0xf0000000
+#define TCA_PERFCOUNTER2_SELECT__PERF_MODE__SHIFT 0x1c
+#define TCA_PERFCOUNTER3_SELECT__PERF_SEL_MASK 0x3ff
+#define TCA_PERFCOUNTER3_SELECT__PERF_SEL__SHIFT 0x0
+#define TCA_PERFCOUNTER3_SELECT__CNTR_MODE_MASK 0xf00000
+#define TCA_PERFCOUNTER3_SELECT__CNTR_MODE__SHIFT 0x14
+#define TCA_PERFCOUNTER3_SELECT__PERF_MODE_MASK 0xf0000000
+#define TCA_PERFCOUNTER3_SELECT__PERF_MODE__SHIFT 0x1c
+#define TCA_PERFCOUNTER0_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define TCA_PERFCOUNTER0_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define TCA_PERFCOUNTER1_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define TCA_PERFCOUNTER1_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define TCA_PERFCOUNTER2_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define TCA_PERFCOUNTER2_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define TCA_PERFCOUNTER3_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define TCA_PERFCOUNTER3_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define TCA_PERFCOUNTER0_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define TCA_PERFCOUNTER0_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define TCA_PERFCOUNTER1_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define TCA_PERFCOUNTER1_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define TCA_PERFCOUNTER2_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define TCA_PERFCOUNTER2_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define TCA_PERFCOUNTER3_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define TCA_PERFCOUNTER3_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define TA_BC_BASE_ADDR__ADDRESS_MASK 0xffffffff
+#define TA_BC_BASE_ADDR__ADDRESS__SHIFT 0x0
+#define TA_BC_BASE_ADDR_HI__ADDRESS_MASK 0xff
+#define TA_BC_BASE_ADDR_HI__ADDRESS__SHIFT 0x0
+#define TD_CNTL__SYNC_PHASE_SH_MASK 0x3
+#define TD_CNTL__SYNC_PHASE_SH__SHIFT 0x0
+#define TD_CNTL__SYNC_PHASE_VC_SMX_MASK 0x30
+#define TD_CNTL__SYNC_PHASE_VC_SMX__SHIFT 0x4
+#define TD_CNTL__PAD_STALL_EN_MASK 0x100
+#define TD_CNTL__PAD_STALL_EN__SHIFT 0x8
+#define TD_CNTL__EXTEND_LDS_STALL_MASK 0x600
+#define TD_CNTL__EXTEND_LDS_STALL__SHIFT 0x9
+#define TD_CNTL__LDS_STALL_PHASE_ADJUST_MASK 0x1800
+#define TD_CNTL__LDS_STALL_PHASE_ADJUST__SHIFT 0xb
+#define TD_CNTL__PRECISION_COMPATIBILITY_MASK 0x8000
+#define TD_CNTL__PRECISION_COMPATIBILITY__SHIFT 0xf
+#define TD_CNTL__GATHER4_FLOAT_MODE_MASK 0x10000
+#define TD_CNTL__GATHER4_FLOAT_MODE__SHIFT 0x10
+#define TD_CNTL__LD_FLOAT_MODE_MASK 0x40000
+#define TD_CNTL__LD_FLOAT_MODE__SHIFT 0x12
+#define TD_CNTL__GATHER4_DX9_MODE_MASK 0x80000
+#define TD_CNTL__GATHER4_DX9_MODE__SHIFT 0x13
+#define TD_CNTL__DISABLE_POWER_THROTTLE_MASK 0x100000
+#define TD_CNTL__DISABLE_POWER_THROTTLE__SHIFT 0x14
+#define TD_CNTL__ENABLE_ROUND_TO_ZERO_MASK 0x200000
+#define TD_CNTL__ENABLE_ROUND_TO_ZERO__SHIFT 0x15
+#define TD_CNTL__DISABLE_D16_PACKING_MASK 0x400000
+#define TD_CNTL__DISABLE_D16_PACKING__SHIFT 0x16
+#define TD_CNTL__DISABLE_2BIT_SIGNED_FORMAT_MASK 0x800000
+#define TD_CNTL__DISABLE_2BIT_SIGNED_FORMAT__SHIFT 0x17
+#define TD_STATUS__BUSY_MASK 0x80000000
+#define TD_STATUS__BUSY__SHIFT 0x1f
+#define TD_DEBUG_INDEX__INDEX_MASK 0x1f
+#define TD_DEBUG_INDEX__INDEX__SHIFT 0x0
+#define TD_DEBUG_DATA__DATA_MASK 0xffffffff
+#define TD_DEBUG_DATA__DATA__SHIFT 0x0
+#define TD_DSM_CNTL__FORCE_SEDB_0_MASK 0x1
+#define TD_DSM_CNTL__FORCE_SEDB_0__SHIFT 0x0
+#define TD_DSM_CNTL__FORCE_SEDB_1_MASK 0x2
+#define TD_DSM_CNTL__FORCE_SEDB_1__SHIFT 0x1
+#define TD_DSM_CNTL__EN_SINGLE_WR_SEDB_MASK 0x4
+#define TD_DSM_CNTL__EN_SINGLE_WR_SEDB__SHIFT 0x2
+#define TD_PERFCOUNTER0_SELECT__PERF_SEL_MASK 0xff
+#define TD_PERFCOUNTER0_SELECT__PERF_SEL__SHIFT 0x0
+#define TD_PERFCOUNTER0_SELECT__PERF_SEL1_MASK 0x3fc00
+#define TD_PERFCOUNTER0_SELECT__PERF_SEL1__SHIFT 0xa
+#define TD_PERFCOUNTER0_SELECT__CNTR_MODE_MASK 0xf00000
+#define TD_PERFCOUNTER0_SELECT__CNTR_MODE__SHIFT 0x14
+#define TD_PERFCOUNTER0_SELECT__PERF_MODE1_MASK 0xf000000
+#define TD_PERFCOUNTER0_SELECT__PERF_MODE1__SHIFT 0x18
+#define TD_PERFCOUNTER0_SELECT__PERF_MODE_MASK 0xf0000000
+#define TD_PERFCOUNTER0_SELECT__PERF_MODE__SHIFT 0x1c
+#define TD_PERFCOUNTER1_SELECT__PERF_SEL_MASK 0xff
+#define TD_PERFCOUNTER1_SELECT__PERF_SEL__SHIFT 0x0
+#define TD_PERFCOUNTER1_SELECT__PERF_SEL1_MASK 0x3fc00
+#define TD_PERFCOUNTER1_SELECT__PERF_SEL1__SHIFT 0xa
+#define TD_PERFCOUNTER1_SELECT__CNTR_MODE_MASK 0xf00000
+#define TD_PERFCOUNTER1_SELECT__CNTR_MODE__SHIFT 0x14
+#define TD_PERFCOUNTER1_SELECT__PERF_MODE1_MASK 0xf000000
+#define TD_PERFCOUNTER1_SELECT__PERF_MODE1__SHIFT 0x18
+#define TD_PERFCOUNTER1_SELECT__PERF_MODE_MASK 0xf0000000
+#define TD_PERFCOUNTER1_SELECT__PERF_MODE__SHIFT 0x1c
+#define TD_PERFCOUNTER0_SELECT1__PERF_SEL2_MASK 0xff
+#define TD_PERFCOUNTER0_SELECT1__PERF_SEL2__SHIFT 0x0
+#define TD_PERFCOUNTER0_SELECT1__PERF_SEL3_MASK 0x3fc00
+#define TD_PERFCOUNTER0_SELECT1__PERF_SEL3__SHIFT 0xa
+#define TD_PERFCOUNTER0_SELECT1__PERF_MODE3_MASK 0xf000000
+#define TD_PERFCOUNTER0_SELECT1__PERF_MODE3__SHIFT 0x18
+#define TD_PERFCOUNTER0_SELECT1__PERF_MODE2_MASK 0xf0000000
+#define TD_PERFCOUNTER0_SELECT1__PERF_MODE2__SHIFT 0x1c
+#define TD_PERFCOUNTER0_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define TD_PERFCOUNTER0_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define TD_PERFCOUNTER1_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define TD_PERFCOUNTER1_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define TD_PERFCOUNTER0_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define TD_PERFCOUNTER0_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define TD_PERFCOUNTER1_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define TD_PERFCOUNTER1_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define TD_SCRATCH__SCRATCH_MASK 0xffffffff
+#define TD_SCRATCH__SCRATCH__SHIFT 0x0
+#define TA_CNTL__FX_XNACK_CREDIT_MASK 0x7f
+#define TA_CNTL__FX_XNACK_CREDIT__SHIFT 0x0
+#define TA_CNTL__SQ_XNACK_CREDIT_MASK 0x1e00
+#define TA_CNTL__SQ_XNACK_CREDIT__SHIFT 0x9
+#define TA_CNTL__TC_DATA_CREDIT_MASK 0xe000
+#define TA_CNTL__TC_DATA_CREDIT__SHIFT 0xd
+#define TA_CNTL__ALIGNER_CREDIT_MASK 0x1f0000
+#define TA_CNTL__ALIGNER_CREDIT__SHIFT 0x10
+#define TA_CNTL__TD_FIFO_CREDIT_MASK 0xffc00000
+#define TA_CNTL__TD_FIFO_CREDIT__SHIFT 0x16
+#define TA_CNTL_AUX__SCOAL_DSWIZZLE_N_MASK 0x1
+#define TA_CNTL_AUX__SCOAL_DSWIZZLE_N__SHIFT 0x0
+#define TA_CNTL_AUX__RESERVED_MASK 0xe
+#define TA_CNTL_AUX__RESERVED__SHIFT 0x1
+#define TA_CNTL_AUX__D16_PACK_DISABLE_MASK 0x10
+#define TA_CNTL_AUX__D16_PACK_DISABLE__SHIFT 0x4
+#define TA_CNTL_AUX__ANISO_WEIGHT_MODE_MASK 0x10000
+#define TA_CNTL_AUX__ANISO_WEIGHT_MODE__SHIFT 0x10
+#define TA_CNTL_AUX__ANISO_RATIO_LUT_MASK 0x20000
+#define TA_CNTL_AUX__ANISO_RATIO_LUT__SHIFT 0x11
+#define TA_CNTL_AUX__ANISO_TAP_MASK 0x40000
+#define TA_CNTL_AUX__ANISO_TAP__SHIFT 0x12
+#define TA_CNTL_AUX__ANISO_MIP_ADJ_MODE_MASK 0x80000
+#define TA_CNTL_AUX__ANISO_MIP_ADJ_MODE__SHIFT 0x13
+#define TA_RESERVED_010C__Unused_MASK 0xffffffff
+#define TA_RESERVED_010C__Unused__SHIFT 0x0
+#define TA_CS_BC_BASE_ADDR__ADDRESS_MASK 0xffffffff
+#define TA_CS_BC_BASE_ADDR__ADDRESS__SHIFT 0x0
+#define TA_CS_BC_BASE_ADDR_HI__ADDRESS_MASK 0xff
+#define TA_CS_BC_BASE_ADDR_HI__ADDRESS__SHIFT 0x0
+#define TA_STATUS__FG_PFIFO_EMPTYB_MASK 0x1000
+#define TA_STATUS__FG_PFIFO_EMPTYB__SHIFT 0xc
+#define TA_STATUS__FG_LFIFO_EMPTYB_MASK 0x2000
+#define TA_STATUS__FG_LFIFO_EMPTYB__SHIFT 0xd
+#define TA_STATUS__FG_SFIFO_EMPTYB_MASK 0x4000
+#define TA_STATUS__FG_SFIFO_EMPTYB__SHIFT 0xe
+#define TA_STATUS__FL_PFIFO_EMPTYB_MASK 0x10000
+#define TA_STATUS__FL_PFIFO_EMPTYB__SHIFT 0x10
+#define TA_STATUS__FL_LFIFO_EMPTYB_MASK 0x20000
+#define TA_STATUS__FL_LFIFO_EMPTYB__SHIFT 0x11
+#define TA_STATUS__FL_SFIFO_EMPTYB_MASK 0x40000
+#define TA_STATUS__FL_SFIFO_EMPTYB__SHIFT 0x12
+#define TA_STATUS__FA_PFIFO_EMPTYB_MASK 0x100000
+#define TA_STATUS__FA_PFIFO_EMPTYB__SHIFT 0x14
+#define TA_STATUS__FA_LFIFO_EMPTYB_MASK 0x200000
+#define TA_STATUS__FA_LFIFO_EMPTYB__SHIFT 0x15
+#define TA_STATUS__FA_SFIFO_EMPTYB_MASK 0x400000
+#define TA_STATUS__FA_SFIFO_EMPTYB__SHIFT 0x16
+#define TA_STATUS__IN_BUSY_MASK 0x1000000
+#define TA_STATUS__IN_BUSY__SHIFT 0x18
+#define TA_STATUS__FG_BUSY_MASK 0x2000000
+#define TA_STATUS__FG_BUSY__SHIFT 0x19
+#define TA_STATUS__LA_BUSY_MASK 0x4000000
+#define TA_STATUS__LA_BUSY__SHIFT 0x1a
+#define TA_STATUS__FL_BUSY_MASK 0x8000000
+#define TA_STATUS__FL_BUSY__SHIFT 0x1b
+#define TA_STATUS__TA_BUSY_MASK 0x10000000
+#define TA_STATUS__TA_BUSY__SHIFT 0x1c
+#define TA_STATUS__FA_BUSY_MASK 0x20000000
+#define TA_STATUS__FA_BUSY__SHIFT 0x1d
+#define TA_STATUS__AL_BUSY_MASK 0x40000000
+#define TA_STATUS__AL_BUSY__SHIFT 0x1e
+#define TA_STATUS__BUSY_MASK 0x80000000
+#define TA_STATUS__BUSY__SHIFT 0x1f
+#define TA_DEBUG_INDEX__INDEX_MASK 0x1f
+#define TA_DEBUG_INDEX__INDEX__SHIFT 0x0
+#define TA_DEBUG_DATA__DATA_MASK 0xffffffff
+#define TA_DEBUG_DATA__DATA__SHIFT 0x0
+#define TA_PERFCOUNTER0_SELECT__PERF_SEL_MASK 0xff
+#define TA_PERFCOUNTER0_SELECT__PERF_SEL__SHIFT 0x0
+#define TA_PERFCOUNTER0_SELECT__PERF_SEL1_MASK 0x3fc00
+#define TA_PERFCOUNTER0_SELECT__PERF_SEL1__SHIFT 0xa
+#define TA_PERFCOUNTER0_SELECT__CNTR_MODE_MASK 0xf00000
+#define TA_PERFCOUNTER0_SELECT__CNTR_MODE__SHIFT 0x14
+#define TA_PERFCOUNTER0_SELECT__PERF_MODE1_MASK 0xf000000
+#define TA_PERFCOUNTER0_SELECT__PERF_MODE1__SHIFT 0x18
+#define TA_PERFCOUNTER0_SELECT__PERF_MODE_MASK 0xf0000000
+#define TA_PERFCOUNTER0_SELECT__PERF_MODE__SHIFT 0x1c
+#define TA_PERFCOUNTER1_SELECT__PERF_SEL_MASK 0xff
+#define TA_PERFCOUNTER1_SELECT__PERF_SEL__SHIFT 0x0
+#define TA_PERFCOUNTER1_SELECT__PERF_SEL1_MASK 0x3fc00
+#define TA_PERFCOUNTER1_SELECT__PERF_SEL1__SHIFT 0xa
+#define TA_PERFCOUNTER1_SELECT__CNTR_MODE_MASK 0xf00000
+#define TA_PERFCOUNTER1_SELECT__CNTR_MODE__SHIFT 0x14
+#define TA_PERFCOUNTER1_SELECT__PERF_MODE1_MASK 0xf000000
+#define TA_PERFCOUNTER1_SELECT__PERF_MODE1__SHIFT 0x18
+#define TA_PERFCOUNTER1_SELECT__PERF_MODE_MASK 0xf0000000
+#define TA_PERFCOUNTER1_SELECT__PERF_MODE__SHIFT 0x1c
+#define TA_PERFCOUNTER0_SELECT1__PERF_SEL2_MASK 0xff
+#define TA_PERFCOUNTER0_SELECT1__PERF_SEL2__SHIFT 0x0
+#define TA_PERFCOUNTER0_SELECT1__PERF_SEL3_MASK 0x3fc00
+#define TA_PERFCOUNTER0_SELECT1__PERF_SEL3__SHIFT 0xa
+#define TA_PERFCOUNTER0_SELECT1__PERF_MODE3_MASK 0xf000000
+#define TA_PERFCOUNTER0_SELECT1__PERF_MODE3__SHIFT 0x18
+#define TA_PERFCOUNTER0_SELECT1__PERF_MODE2_MASK 0xf0000000
+#define TA_PERFCOUNTER0_SELECT1__PERF_MODE2__SHIFT 0x1c
+#define TA_PERFCOUNTER0_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define TA_PERFCOUNTER0_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define TA_PERFCOUNTER1_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define TA_PERFCOUNTER1_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define TA_PERFCOUNTER0_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define TA_PERFCOUNTER0_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define TA_PERFCOUNTER1_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define TA_PERFCOUNTER1_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define TA_SCRATCH__SCRATCH_MASK 0xffffffff
+#define TA_SCRATCH__SCRATCH__SHIFT 0x0
+#define SH_HIDDEN_PRIVATE_BASE_VMID__ADDRESS_MASK 0xffffffff
+#define SH_HIDDEN_PRIVATE_BASE_VMID__ADDRESS__SHIFT 0x0
+#define SH_STATIC_MEM_CONFIG__SWIZZLE_ENABLE_MASK 0x1
+#define SH_STATIC_MEM_CONFIG__SWIZZLE_ENABLE__SHIFT 0x0
+#define SH_STATIC_MEM_CONFIG__ELEMENT_SIZE_MASK 0x6
+#define SH_STATIC_MEM_CONFIG__ELEMENT_SIZE__SHIFT 0x1
+#define SH_STATIC_MEM_CONFIG__INDEX_STRIDE_MASK 0x18
+#define SH_STATIC_MEM_CONFIG__INDEX_STRIDE__SHIFT 0x3
+#define SH_STATIC_MEM_CONFIG__PRIVATE_MTYPE_MASK 0xe0
+#define SH_STATIC_MEM_CONFIG__PRIVATE_MTYPE__SHIFT 0x5
+#define SH_STATIC_MEM_CONFIG__READ_ONLY_CNTL_MASK 0xff00
+#define SH_STATIC_MEM_CONFIG__READ_ONLY_CNTL__SHIFT 0x8
+#define TCP_INVALIDATE__START_MASK 0x1
+#define TCP_INVALIDATE__START__SHIFT 0x0
+#define TCP_STATUS__TCP_BUSY_MASK 0x1
+#define TCP_STATUS__TCP_BUSY__SHIFT 0x0
+#define TCP_STATUS__INPUT_BUSY_MASK 0x2
+#define TCP_STATUS__INPUT_BUSY__SHIFT 0x1
+#define TCP_STATUS__ADRS_BUSY_MASK 0x4
+#define TCP_STATUS__ADRS_BUSY__SHIFT 0x2
+#define TCP_STATUS__TAGRAMS_BUSY_MASK 0x8
+#define TCP_STATUS__TAGRAMS_BUSY__SHIFT 0x3
+#define TCP_STATUS__CNTRL_BUSY_MASK 0x10
+#define TCP_STATUS__CNTRL_BUSY__SHIFT 0x4
+#define TCP_STATUS__LFIFO_BUSY_MASK 0x20
+#define TCP_STATUS__LFIFO_BUSY__SHIFT 0x5
+#define TCP_STATUS__READ_BUSY_MASK 0x40
+#define TCP_STATUS__READ_BUSY__SHIFT 0x6
+#define TCP_STATUS__FORMAT_BUSY_MASK 0x80
+#define TCP_STATUS__FORMAT_BUSY__SHIFT 0x7
+#define TCP_CNTL__FORCE_HIT_MASK 0x1
+#define TCP_CNTL__FORCE_HIT__SHIFT 0x0
+#define TCP_CNTL__FORCE_MISS_MASK 0x2
+#define TCP_CNTL__FORCE_MISS__SHIFT 0x1
+#define TCP_CNTL__L1_SIZE_MASK 0xc
+#define TCP_CNTL__L1_SIZE__SHIFT 0x2
+#define TCP_CNTL__FLAT_BUF_HASH_ENABLE_MASK 0x10
+#define TCP_CNTL__FLAT_BUF_HASH_ENABLE__SHIFT 0x4
+#define TCP_CNTL__FLAT_BUF_CACHE_SWIZZLE_MASK 0x20
+#define TCP_CNTL__FLAT_BUF_CACHE_SWIZZLE__SHIFT 0x5
+#define TCP_CNTL__FORCE_EOW_TOTAL_CNT_MASK 0x1f8000
+#define TCP_CNTL__FORCE_EOW_TOTAL_CNT__SHIFT 0xf
+#define TCP_CNTL__FORCE_EOW_TAGRAM_CNT_MASK 0xfc00000
+#define TCP_CNTL__FORCE_EOW_TAGRAM_CNT__SHIFT 0x16
+#define TCP_CNTL__DISABLE_Z_MAP_MASK 0x10000000
+#define TCP_CNTL__DISABLE_Z_MAP__SHIFT 0x1c
+#define TCP_CNTL__INV_ALL_VMIDS_MASK 0x20000000
+#define TCP_CNTL__INV_ALL_VMIDS__SHIFT 0x1d
+#define TCP_CHAN_STEER_LO__CHAN0_MASK 0xf
+#define TCP_CHAN_STEER_LO__CHAN0__SHIFT 0x0
+#define TCP_CHAN_STEER_LO__CHAN1_MASK 0xf0
+#define TCP_CHAN_STEER_LO__CHAN1__SHIFT 0x4
+#define TCP_CHAN_STEER_LO__CHAN2_MASK 0xf00
+#define TCP_CHAN_STEER_LO__CHAN2__SHIFT 0x8
+#define TCP_CHAN_STEER_LO__CHAN3_MASK 0xf000
+#define TCP_CHAN_STEER_LO__CHAN3__SHIFT 0xc
+#define TCP_CHAN_STEER_LO__CHAN4_MASK 0xf0000
+#define TCP_CHAN_STEER_LO__CHAN4__SHIFT 0x10
+#define TCP_CHAN_STEER_LO__CHAN5_MASK 0xf00000
+#define TCP_CHAN_STEER_LO__CHAN5__SHIFT 0x14
+#define TCP_CHAN_STEER_LO__CHAN6_MASK 0xf000000
+#define TCP_CHAN_STEER_LO__CHAN6__SHIFT 0x18
+#define TCP_CHAN_STEER_LO__CHAN7_MASK 0xf0000000
+#define TCP_CHAN_STEER_LO__CHAN7__SHIFT 0x1c
+#define TCP_CHAN_STEER_HI__CHAN8_MASK 0xf
+#define TCP_CHAN_STEER_HI__CHAN8__SHIFT 0x0
+#define TCP_CHAN_STEER_HI__CHAN9_MASK 0xf0
+#define TCP_CHAN_STEER_HI__CHAN9__SHIFT 0x4
+#define TCP_CHAN_STEER_HI__CHANA_MASK 0xf00
+#define TCP_CHAN_STEER_HI__CHANA__SHIFT 0x8
+#define TCP_CHAN_STEER_HI__CHANB_MASK 0xf000
+#define TCP_CHAN_STEER_HI__CHANB__SHIFT 0xc
+#define TCP_CHAN_STEER_HI__CHANC_MASK 0xf0000
+#define TCP_CHAN_STEER_HI__CHANC__SHIFT 0x10
+#define TCP_CHAN_STEER_HI__CHAND_MASK 0xf00000
+#define TCP_CHAN_STEER_HI__CHAND__SHIFT 0x14
+#define TCP_CHAN_STEER_HI__CHANE_MASK 0xf000000
+#define TCP_CHAN_STEER_HI__CHANE__SHIFT 0x18
+#define TCP_CHAN_STEER_HI__CHANF_MASK 0xf0000000
+#define TCP_CHAN_STEER_HI__CHANF__SHIFT 0x1c
+#define TCP_ADDR_CONFIG__NUM_TCC_BANKS_MASK 0xf
+#define TCP_ADDR_CONFIG__NUM_TCC_BANKS__SHIFT 0x0
+#define TCP_ADDR_CONFIG__NUM_BANKS_MASK 0x30
+#define TCP_ADDR_CONFIG__NUM_BANKS__SHIFT 0x4
+#define TCP_ADDR_CONFIG__COLHI_WIDTH_MASK 0x1c0
+#define TCP_ADDR_CONFIG__COLHI_WIDTH__SHIFT 0x6
+#define TCP_ADDR_CONFIG__RB_SPLIT_COLHI_MASK 0x200
+#define TCP_ADDR_CONFIG__RB_SPLIT_COLHI__SHIFT 0x9
+#define TCP_CREDIT__LFIFO_CREDIT_MASK 0x3ff
+#define TCP_CREDIT__LFIFO_CREDIT__SHIFT 0x0
+#define TCP_CREDIT__REQ_FIFO_CREDIT_MASK 0x7f0000
+#define TCP_CREDIT__REQ_FIFO_CREDIT__SHIFT 0x10
+#define TCP_CREDIT__TD_CREDIT_MASK 0xe0000000
+#define TCP_CREDIT__TD_CREDIT__SHIFT 0x1d
+#define TCP_PERFCOUNTER0_SELECT__PERF_SEL_MASK 0x3ff
+#define TCP_PERFCOUNTER0_SELECT__PERF_SEL__SHIFT 0x0
+#define TCP_PERFCOUNTER0_SELECT__PERF_SEL1_MASK 0xffc00
+#define TCP_PERFCOUNTER0_SELECT__PERF_SEL1__SHIFT 0xa
+#define TCP_PERFCOUNTER0_SELECT__CNTR_MODE_MASK 0xf00000
+#define TCP_PERFCOUNTER0_SELECT__CNTR_MODE__SHIFT 0x14
+#define TCP_PERFCOUNTER0_SELECT__PERF_MODE1_MASK 0xf000000
+#define TCP_PERFCOUNTER0_SELECT__PERF_MODE1__SHIFT 0x18
+#define TCP_PERFCOUNTER0_SELECT__PERF_MODE_MASK 0xf0000000
+#define TCP_PERFCOUNTER0_SELECT__PERF_MODE__SHIFT 0x1c
+#define TCP_PERFCOUNTER1_SELECT__PERF_SEL_MASK 0x3ff
+#define TCP_PERFCOUNTER1_SELECT__PERF_SEL__SHIFT 0x0
+#define TCP_PERFCOUNTER1_SELECT__PERF_SEL1_MASK 0xffc00
+#define TCP_PERFCOUNTER1_SELECT__PERF_SEL1__SHIFT 0xa
+#define TCP_PERFCOUNTER1_SELECT__CNTR_MODE_MASK 0xf00000
+#define TCP_PERFCOUNTER1_SELECT__CNTR_MODE__SHIFT 0x14
+#define TCP_PERFCOUNTER1_SELECT__PERF_MODE1_MASK 0xf000000
+#define TCP_PERFCOUNTER1_SELECT__PERF_MODE1__SHIFT 0x18
+#define TCP_PERFCOUNTER1_SELECT__PERF_MODE_MASK 0xf0000000
+#define TCP_PERFCOUNTER1_SELECT__PERF_MODE__SHIFT 0x1c
+#define TCP_PERFCOUNTER0_SELECT1__PERF_SEL2_MASK 0x3ff
+#define TCP_PERFCOUNTER0_SELECT1__PERF_SEL2__SHIFT 0x0
+#define TCP_PERFCOUNTER0_SELECT1__PERF_SEL3_MASK 0xffc00
+#define TCP_PERFCOUNTER0_SELECT1__PERF_SEL3__SHIFT 0xa
+#define TCP_PERFCOUNTER0_SELECT1__PERF_MODE3_MASK 0xf000000
+#define TCP_PERFCOUNTER0_SELECT1__PERF_MODE3__SHIFT 0x18
+#define TCP_PERFCOUNTER0_SELECT1__PERF_MODE2_MASK 0xf0000000
+#define TCP_PERFCOUNTER0_SELECT1__PERF_MODE2__SHIFT 0x1c
+#define TCP_PERFCOUNTER1_SELECT1__PERF_SEL2_MASK 0x3ff
+#define TCP_PERFCOUNTER1_SELECT1__PERF_SEL2__SHIFT 0x0
+#define TCP_PERFCOUNTER1_SELECT1__PERF_SEL3_MASK 0xffc00
+#define TCP_PERFCOUNTER1_SELECT1__PERF_SEL3__SHIFT 0xa
+#define TCP_PERFCOUNTER1_SELECT1__PERF_MODE3_MASK 0xf000000
+#define TCP_PERFCOUNTER1_SELECT1__PERF_MODE3__SHIFT 0x18
+#define TCP_PERFCOUNTER1_SELECT1__PERF_MODE2_MASK 0xf0000000
+#define TCP_PERFCOUNTER1_SELECT1__PERF_MODE2__SHIFT 0x1c
+#define TCP_PERFCOUNTER2_SELECT__PERF_SEL_MASK 0x3ff
+#define TCP_PERFCOUNTER2_SELECT__PERF_SEL__SHIFT 0x0
+#define TCP_PERFCOUNTER2_SELECT__CNTR_MODE_MASK 0xf00000
+#define TCP_PERFCOUNTER2_SELECT__CNTR_MODE__SHIFT 0x14
+#define TCP_PERFCOUNTER2_SELECT__PERF_MODE_MASK 0xf0000000
+#define TCP_PERFCOUNTER2_SELECT__PERF_MODE__SHIFT 0x1c
+#define TCP_PERFCOUNTER3_SELECT__PERF_SEL_MASK 0x3ff
+#define TCP_PERFCOUNTER3_SELECT__PERF_SEL__SHIFT 0x0
+#define TCP_PERFCOUNTER3_SELECT__CNTR_MODE_MASK 0xf00000
+#define TCP_PERFCOUNTER3_SELECT__CNTR_MODE__SHIFT 0x14
+#define TCP_PERFCOUNTER3_SELECT__PERF_MODE_MASK 0xf0000000
+#define TCP_PERFCOUNTER3_SELECT__PERF_MODE__SHIFT 0x1c
+#define TCP_PERFCOUNTER0_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define TCP_PERFCOUNTER0_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define TCP_PERFCOUNTER1_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define TCP_PERFCOUNTER1_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define TCP_PERFCOUNTER2_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define TCP_PERFCOUNTER2_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define TCP_PERFCOUNTER3_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define TCP_PERFCOUNTER3_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define TCP_PERFCOUNTER0_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define TCP_PERFCOUNTER0_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define TCP_PERFCOUNTER1_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define TCP_PERFCOUNTER1_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define TCP_PERFCOUNTER2_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define TCP_PERFCOUNTER2_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define TCP_PERFCOUNTER3_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define TCP_PERFCOUNTER3_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define TCP_BUFFER_ADDR_HASH_CNTL__CHANNEL_BITS_MASK 0x7
+#define TCP_BUFFER_ADDR_HASH_CNTL__CHANNEL_BITS__SHIFT 0x0
+#define TCP_BUFFER_ADDR_HASH_CNTL__BANK_BITS_MASK 0x700
+#define TCP_BUFFER_ADDR_HASH_CNTL__BANK_BITS__SHIFT 0x8
+#define TCP_BUFFER_ADDR_HASH_CNTL__CHANNEL_XOR_COUNT_MASK 0x70000
+#define TCP_BUFFER_ADDR_HASH_CNTL__CHANNEL_XOR_COUNT__SHIFT 0x10
+#define TCP_BUFFER_ADDR_HASH_CNTL__BANK_XOR_COUNT_MASK 0x7000000
+#define TCP_BUFFER_ADDR_HASH_CNTL__BANK_XOR_COUNT__SHIFT 0x18
+#define TCP_EDC_CNT__SEC_COUNT_MASK 0xff
+#define TCP_EDC_CNT__SEC_COUNT__SHIFT 0x0
+#define TCP_EDC_CNT__LFIFO_SED_COUNT_MASK 0xff00
+#define TCP_EDC_CNT__LFIFO_SED_COUNT__SHIFT 0x8
+#define TCP_EDC_CNT__DED_COUNT_MASK 0xff0000
+#define TCP_EDC_CNT__DED_COUNT__SHIFT 0x10
+#define TCP_EDC_CNT__UNUSED_MASK 0xff000000
+#define TCP_EDC_CNT__UNUSED__SHIFT 0x18
+#define TC_CFG_L1_LOAD_POLICY0__POLICY_0_MASK 0x3
+#define TC_CFG_L1_LOAD_POLICY0__POLICY_0__SHIFT 0x0
+#define TC_CFG_L1_LOAD_POLICY0__POLICY_1_MASK 0xc
+#define TC_CFG_L1_LOAD_POLICY0__POLICY_1__SHIFT 0x2
+#define TC_CFG_L1_LOAD_POLICY0__POLICY_2_MASK 0x30
+#define TC_CFG_L1_LOAD_POLICY0__POLICY_2__SHIFT 0x4
+#define TC_CFG_L1_LOAD_POLICY0__POLICY_3_MASK 0xc0
+#define TC_CFG_L1_LOAD_POLICY0__POLICY_3__SHIFT 0x6
+#define TC_CFG_L1_LOAD_POLICY0__POLICY_4_MASK 0x300
+#define TC_CFG_L1_LOAD_POLICY0__POLICY_4__SHIFT 0x8
+#define TC_CFG_L1_LOAD_POLICY0__POLICY_5_MASK 0xc00
+#define TC_CFG_L1_LOAD_POLICY0__POLICY_5__SHIFT 0xa
+#define TC_CFG_L1_LOAD_POLICY0__POLICY_6_MASK 0x3000
+#define TC_CFG_L1_LOAD_POLICY0__POLICY_6__SHIFT 0xc
+#define TC_CFG_L1_LOAD_POLICY0__POLICY_7_MASK 0xc000
+#define TC_CFG_L1_LOAD_POLICY0__POLICY_7__SHIFT 0xe
+#define TC_CFG_L1_LOAD_POLICY0__POLICY_8_MASK 0x30000
+#define TC_CFG_L1_LOAD_POLICY0__POLICY_8__SHIFT 0x10
+#define TC_CFG_L1_LOAD_POLICY0__POLICY_9_MASK 0xc0000
+#define TC_CFG_L1_LOAD_POLICY0__POLICY_9__SHIFT 0x12
+#define TC_CFG_L1_LOAD_POLICY0__POLICY_10_MASK 0x300000
+#define TC_CFG_L1_LOAD_POLICY0__POLICY_10__SHIFT 0x14
+#define TC_CFG_L1_LOAD_POLICY0__POLICY_11_MASK 0xc00000
+#define TC_CFG_L1_LOAD_POLICY0__POLICY_11__SHIFT 0x16
+#define TC_CFG_L1_LOAD_POLICY0__POLICY_12_MASK 0x3000000
+#define TC_CFG_L1_LOAD_POLICY0__POLICY_12__SHIFT 0x18
+#define TC_CFG_L1_LOAD_POLICY0__POLICY_13_MASK 0xc000000
+#define TC_CFG_L1_LOAD_POLICY0__POLICY_13__SHIFT 0x1a
+#define TC_CFG_L1_LOAD_POLICY0__POLICY_14_MASK 0x30000000
+#define TC_CFG_L1_LOAD_POLICY0__POLICY_14__SHIFT 0x1c
+#define TC_CFG_L1_LOAD_POLICY0__POLICY_15_MASK 0xc0000000
+#define TC_CFG_L1_LOAD_POLICY0__POLICY_15__SHIFT 0x1e
+#define TC_CFG_L1_LOAD_POLICY1__POLICY_16_MASK 0x3
+#define TC_CFG_L1_LOAD_POLICY1__POLICY_16__SHIFT 0x0
+#define TC_CFG_L1_LOAD_POLICY1__POLICY_17_MASK 0xc
+#define TC_CFG_L1_LOAD_POLICY1__POLICY_17__SHIFT 0x2
+#define TC_CFG_L1_LOAD_POLICY1__POLICY_18_MASK 0x30
+#define TC_CFG_L1_LOAD_POLICY1__POLICY_18__SHIFT 0x4
+#define TC_CFG_L1_LOAD_POLICY1__POLICY_19_MASK 0xc0
+#define TC_CFG_L1_LOAD_POLICY1__POLICY_19__SHIFT 0x6
+#define TC_CFG_L1_LOAD_POLICY1__POLICY_20_MASK 0x300
+#define TC_CFG_L1_LOAD_POLICY1__POLICY_20__SHIFT 0x8
+#define TC_CFG_L1_LOAD_POLICY1__POLICY_21_MASK 0xc00
+#define TC_CFG_L1_LOAD_POLICY1__POLICY_21__SHIFT 0xa
+#define TC_CFG_L1_LOAD_POLICY1__POLICY_22_MASK 0x3000
+#define TC_CFG_L1_LOAD_POLICY1__POLICY_22__SHIFT 0xc
+#define TC_CFG_L1_LOAD_POLICY1__POLICY_23_MASK 0xc000
+#define TC_CFG_L1_LOAD_POLICY1__POLICY_23__SHIFT 0xe
+#define TC_CFG_L1_LOAD_POLICY1__POLICY_24_MASK 0x30000
+#define TC_CFG_L1_LOAD_POLICY1__POLICY_24__SHIFT 0x10
+#define TC_CFG_L1_LOAD_POLICY1__POLICY_25_MASK 0xc0000
+#define TC_CFG_L1_LOAD_POLICY1__POLICY_25__SHIFT 0x12
+#define TC_CFG_L1_LOAD_POLICY1__POLICY_26_MASK 0x300000
+#define TC_CFG_L1_LOAD_POLICY1__POLICY_26__SHIFT 0x14
+#define TC_CFG_L1_LOAD_POLICY1__POLICY_27_MASK 0xc00000
+#define TC_CFG_L1_LOAD_POLICY1__POLICY_27__SHIFT 0x16
+#define TC_CFG_L1_LOAD_POLICY1__POLICY_28_MASK 0x3000000
+#define TC_CFG_L1_LOAD_POLICY1__POLICY_28__SHIFT 0x18
+#define TC_CFG_L1_LOAD_POLICY1__POLICY_29_MASK 0xc000000
+#define TC_CFG_L1_LOAD_POLICY1__POLICY_29__SHIFT 0x1a
+#define TC_CFG_L1_LOAD_POLICY1__POLICY_30_MASK 0x30000000
+#define TC_CFG_L1_LOAD_POLICY1__POLICY_30__SHIFT 0x1c
+#define TC_CFG_L1_LOAD_POLICY1__POLICY_31_MASK 0xc0000000
+#define TC_CFG_L1_LOAD_POLICY1__POLICY_31__SHIFT 0x1e
+#define TC_CFG_L1_STORE_POLICY__POLICY_0_MASK 0x1
+#define TC_CFG_L1_STORE_POLICY__POLICY_0__SHIFT 0x0
+#define TC_CFG_L1_STORE_POLICY__POLICY_1_MASK 0x2
+#define TC_CFG_L1_STORE_POLICY__POLICY_1__SHIFT 0x1
+#define TC_CFG_L1_STORE_POLICY__POLICY_2_MASK 0x4
+#define TC_CFG_L1_STORE_POLICY__POLICY_2__SHIFT 0x2
+#define TC_CFG_L1_STORE_POLICY__POLICY_3_MASK 0x8
+#define TC_CFG_L1_STORE_POLICY__POLICY_3__SHIFT 0x3
+#define TC_CFG_L1_STORE_POLICY__POLICY_4_MASK 0x10
+#define TC_CFG_L1_STORE_POLICY__POLICY_4__SHIFT 0x4
+#define TC_CFG_L1_STORE_POLICY__POLICY_5_MASK 0x20
+#define TC_CFG_L1_STORE_POLICY__POLICY_5__SHIFT 0x5
+#define TC_CFG_L1_STORE_POLICY__POLICY_6_MASK 0x40
+#define TC_CFG_L1_STORE_POLICY__POLICY_6__SHIFT 0x6
+#define TC_CFG_L1_STORE_POLICY__POLICY_7_MASK 0x80
+#define TC_CFG_L1_STORE_POLICY__POLICY_7__SHIFT 0x7
+#define TC_CFG_L1_STORE_POLICY__POLICY_8_MASK 0x100
+#define TC_CFG_L1_STORE_POLICY__POLICY_8__SHIFT 0x8
+#define TC_CFG_L1_STORE_POLICY__POLICY_9_MASK 0x200
+#define TC_CFG_L1_STORE_POLICY__POLICY_9__SHIFT 0x9
+#define TC_CFG_L1_STORE_POLICY__POLICY_10_MASK 0x400
+#define TC_CFG_L1_STORE_POLICY__POLICY_10__SHIFT 0xa
+#define TC_CFG_L1_STORE_POLICY__POLICY_11_MASK 0x800
+#define TC_CFG_L1_STORE_POLICY__POLICY_11__SHIFT 0xb
+#define TC_CFG_L1_STORE_POLICY__POLICY_12_MASK 0x1000
+#define TC_CFG_L1_STORE_POLICY__POLICY_12__SHIFT 0xc
+#define TC_CFG_L1_STORE_POLICY__POLICY_13_MASK 0x2000
+#define TC_CFG_L1_STORE_POLICY__POLICY_13__SHIFT 0xd
+#define TC_CFG_L1_STORE_POLICY__POLICY_14_MASK 0x4000
+#define TC_CFG_L1_STORE_POLICY__POLICY_14__SHIFT 0xe
+#define TC_CFG_L1_STORE_POLICY__POLICY_15_MASK 0x8000
+#define TC_CFG_L1_STORE_POLICY__POLICY_15__SHIFT 0xf
+#define TC_CFG_L1_STORE_POLICY__POLICY_16_MASK 0x10000
+#define TC_CFG_L1_STORE_POLICY__POLICY_16__SHIFT 0x10
+#define TC_CFG_L1_STORE_POLICY__POLICY_17_MASK 0x20000
+#define TC_CFG_L1_STORE_POLICY__POLICY_17__SHIFT 0x11
+#define TC_CFG_L1_STORE_POLICY__POLICY_18_MASK 0x40000
+#define TC_CFG_L1_STORE_POLICY__POLICY_18__SHIFT 0x12
+#define TC_CFG_L1_STORE_POLICY__POLICY_19_MASK 0x80000
+#define TC_CFG_L1_STORE_POLICY__POLICY_19__SHIFT 0x13
+#define TC_CFG_L1_STORE_POLICY__POLICY_20_MASK 0x100000
+#define TC_CFG_L1_STORE_POLICY__POLICY_20__SHIFT 0x14
+#define TC_CFG_L1_STORE_POLICY__POLICY_21_MASK 0x200000
+#define TC_CFG_L1_STORE_POLICY__POLICY_21__SHIFT 0x15
+#define TC_CFG_L1_STORE_POLICY__POLICY_22_MASK 0x400000
+#define TC_CFG_L1_STORE_POLICY__POLICY_22__SHIFT 0x16
+#define TC_CFG_L1_STORE_POLICY__POLICY_23_MASK 0x800000
+#define TC_CFG_L1_STORE_POLICY__POLICY_23__SHIFT 0x17
+#define TC_CFG_L1_STORE_POLICY__POLICY_24_MASK 0x1000000
+#define TC_CFG_L1_STORE_POLICY__POLICY_24__SHIFT 0x18
+#define TC_CFG_L1_STORE_POLICY__POLICY_25_MASK 0x2000000
+#define TC_CFG_L1_STORE_POLICY__POLICY_25__SHIFT 0x19
+#define TC_CFG_L1_STORE_POLICY__POLICY_26_MASK 0x4000000
+#define TC_CFG_L1_STORE_POLICY__POLICY_26__SHIFT 0x1a
+#define TC_CFG_L1_STORE_POLICY__POLICY_27_MASK 0x8000000
+#define TC_CFG_L1_STORE_POLICY__POLICY_27__SHIFT 0x1b
+#define TC_CFG_L1_STORE_POLICY__POLICY_28_MASK 0x10000000
+#define TC_CFG_L1_STORE_POLICY__POLICY_28__SHIFT 0x1c
+#define TC_CFG_L1_STORE_POLICY__POLICY_29_MASK 0x20000000
+#define TC_CFG_L1_STORE_POLICY__POLICY_29__SHIFT 0x1d
+#define TC_CFG_L1_STORE_POLICY__POLICY_30_MASK 0x40000000
+#define TC_CFG_L1_STORE_POLICY__POLICY_30__SHIFT 0x1e
+#define TC_CFG_L1_STORE_POLICY__POLICY_31_MASK 0x80000000
+#define TC_CFG_L1_STORE_POLICY__POLICY_31__SHIFT 0x1f
+#define TC_CFG_L2_LOAD_POLICY0__POLICY_0_MASK 0x3
+#define TC_CFG_L2_LOAD_POLICY0__POLICY_0__SHIFT 0x0
+#define TC_CFG_L2_LOAD_POLICY0__POLICY_1_MASK 0xc
+#define TC_CFG_L2_LOAD_POLICY0__POLICY_1__SHIFT 0x2
+#define TC_CFG_L2_LOAD_POLICY0__POLICY_2_MASK 0x30
+#define TC_CFG_L2_LOAD_POLICY0__POLICY_2__SHIFT 0x4
+#define TC_CFG_L2_LOAD_POLICY0__POLICY_3_MASK 0xc0
+#define TC_CFG_L2_LOAD_POLICY0__POLICY_3__SHIFT 0x6
+#define TC_CFG_L2_LOAD_POLICY0__POLICY_4_MASK 0x300
+#define TC_CFG_L2_LOAD_POLICY0__POLICY_4__SHIFT 0x8
+#define TC_CFG_L2_LOAD_POLICY0__POLICY_5_MASK 0xc00
+#define TC_CFG_L2_LOAD_POLICY0__POLICY_5__SHIFT 0xa
+#define TC_CFG_L2_LOAD_POLICY0__POLICY_6_MASK 0x3000
+#define TC_CFG_L2_LOAD_POLICY0__POLICY_6__SHIFT 0xc
+#define TC_CFG_L2_LOAD_POLICY0__POLICY_7_MASK 0xc000
+#define TC_CFG_L2_LOAD_POLICY0__POLICY_7__SHIFT 0xe
+#define TC_CFG_L2_LOAD_POLICY0__POLICY_8_MASK 0x30000
+#define TC_CFG_L2_LOAD_POLICY0__POLICY_8__SHIFT 0x10
+#define TC_CFG_L2_LOAD_POLICY0__POLICY_9_MASK 0xc0000
+#define TC_CFG_L2_LOAD_POLICY0__POLICY_9__SHIFT 0x12
+#define TC_CFG_L2_LOAD_POLICY0__POLICY_10_MASK 0x300000
+#define TC_CFG_L2_LOAD_POLICY0__POLICY_10__SHIFT 0x14
+#define TC_CFG_L2_LOAD_POLICY0__POLICY_11_MASK 0xc00000
+#define TC_CFG_L2_LOAD_POLICY0__POLICY_11__SHIFT 0x16
+#define TC_CFG_L2_LOAD_POLICY0__POLICY_12_MASK 0x3000000
+#define TC_CFG_L2_LOAD_POLICY0__POLICY_12__SHIFT 0x18
+#define TC_CFG_L2_LOAD_POLICY0__POLICY_13_MASK 0xc000000
+#define TC_CFG_L2_LOAD_POLICY0__POLICY_13__SHIFT 0x1a
+#define TC_CFG_L2_LOAD_POLICY0__POLICY_14_MASK 0x30000000
+#define TC_CFG_L2_LOAD_POLICY0__POLICY_14__SHIFT 0x1c
+#define TC_CFG_L2_LOAD_POLICY0__POLICY_15_MASK 0xc0000000
+#define TC_CFG_L2_LOAD_POLICY0__POLICY_15__SHIFT 0x1e
+#define TC_CFG_L2_LOAD_POLICY1__POLICY_16_MASK 0x3
+#define TC_CFG_L2_LOAD_POLICY1__POLICY_16__SHIFT 0x0
+#define TC_CFG_L2_LOAD_POLICY1__POLICY_17_MASK 0xc
+#define TC_CFG_L2_LOAD_POLICY1__POLICY_17__SHIFT 0x2
+#define TC_CFG_L2_LOAD_POLICY1__POLICY_18_MASK 0x30
+#define TC_CFG_L2_LOAD_POLICY1__POLICY_18__SHIFT 0x4
+#define TC_CFG_L2_LOAD_POLICY1__POLICY_19_MASK 0xc0
+#define TC_CFG_L2_LOAD_POLICY1__POLICY_19__SHIFT 0x6
+#define TC_CFG_L2_LOAD_POLICY1__POLICY_20_MASK 0x300
+#define TC_CFG_L2_LOAD_POLICY1__POLICY_20__SHIFT 0x8
+#define TC_CFG_L2_LOAD_POLICY1__POLICY_21_MASK 0xc00
+#define TC_CFG_L2_LOAD_POLICY1__POLICY_21__SHIFT 0xa
+#define TC_CFG_L2_LOAD_POLICY1__POLICY_22_MASK 0x3000
+#define TC_CFG_L2_LOAD_POLICY1__POLICY_22__SHIFT 0xc
+#define TC_CFG_L2_LOAD_POLICY1__POLICY_23_MASK 0xc000
+#define TC_CFG_L2_LOAD_POLICY1__POLICY_23__SHIFT 0xe
+#define TC_CFG_L2_LOAD_POLICY1__POLICY_24_MASK 0x30000
+#define TC_CFG_L2_LOAD_POLICY1__POLICY_24__SHIFT 0x10
+#define TC_CFG_L2_LOAD_POLICY1__POLICY_25_MASK 0xc0000
+#define TC_CFG_L2_LOAD_POLICY1__POLICY_25__SHIFT 0x12
+#define TC_CFG_L2_LOAD_POLICY1__POLICY_26_MASK 0x300000
+#define TC_CFG_L2_LOAD_POLICY1__POLICY_26__SHIFT 0x14
+#define TC_CFG_L2_LOAD_POLICY1__POLICY_27_MASK 0xc00000
+#define TC_CFG_L2_LOAD_POLICY1__POLICY_27__SHIFT 0x16
+#define TC_CFG_L2_LOAD_POLICY1__POLICY_28_MASK 0x3000000
+#define TC_CFG_L2_LOAD_POLICY1__POLICY_28__SHIFT 0x18
+#define TC_CFG_L2_LOAD_POLICY1__POLICY_29_MASK 0xc000000
+#define TC_CFG_L2_LOAD_POLICY1__POLICY_29__SHIFT 0x1a
+#define TC_CFG_L2_LOAD_POLICY1__POLICY_30_MASK 0x30000000
+#define TC_CFG_L2_LOAD_POLICY1__POLICY_30__SHIFT 0x1c
+#define TC_CFG_L2_LOAD_POLICY1__POLICY_31_MASK 0xc0000000
+#define TC_CFG_L2_LOAD_POLICY1__POLICY_31__SHIFT 0x1e
+#define TC_CFG_L2_STORE_POLICY0__POLICY_0_MASK 0x3
+#define TC_CFG_L2_STORE_POLICY0__POLICY_0__SHIFT 0x0
+#define TC_CFG_L2_STORE_POLICY0__POLICY_1_MASK 0xc
+#define TC_CFG_L2_STORE_POLICY0__POLICY_1__SHIFT 0x2
+#define TC_CFG_L2_STORE_POLICY0__POLICY_2_MASK 0x30
+#define TC_CFG_L2_STORE_POLICY0__POLICY_2__SHIFT 0x4
+#define TC_CFG_L2_STORE_POLICY0__POLICY_3_MASK 0xc0
+#define TC_CFG_L2_STORE_POLICY0__POLICY_3__SHIFT 0x6
+#define TC_CFG_L2_STORE_POLICY0__POLICY_4_MASK 0x300
+#define TC_CFG_L2_STORE_POLICY0__POLICY_4__SHIFT 0x8
+#define TC_CFG_L2_STORE_POLICY0__POLICY_5_MASK 0xc00
+#define TC_CFG_L2_STORE_POLICY0__POLICY_5__SHIFT 0xa
+#define TC_CFG_L2_STORE_POLICY0__POLICY_6_MASK 0x3000
+#define TC_CFG_L2_STORE_POLICY0__POLICY_6__SHIFT 0xc
+#define TC_CFG_L2_STORE_POLICY0__POLICY_7_MASK 0xc000
+#define TC_CFG_L2_STORE_POLICY0__POLICY_7__SHIFT 0xe
+#define TC_CFG_L2_STORE_POLICY0__POLICY_8_MASK 0x30000
+#define TC_CFG_L2_STORE_POLICY0__POLICY_8__SHIFT 0x10
+#define TC_CFG_L2_STORE_POLICY0__POLICY_9_MASK 0xc0000
+#define TC_CFG_L2_STORE_POLICY0__POLICY_9__SHIFT 0x12
+#define TC_CFG_L2_STORE_POLICY0__POLICY_10_MASK 0x300000
+#define TC_CFG_L2_STORE_POLICY0__POLICY_10__SHIFT 0x14
+#define TC_CFG_L2_STORE_POLICY0__POLICY_11_MASK 0xc00000
+#define TC_CFG_L2_STORE_POLICY0__POLICY_11__SHIFT 0x16
+#define TC_CFG_L2_STORE_POLICY0__POLICY_12_MASK 0x3000000
+#define TC_CFG_L2_STORE_POLICY0__POLICY_12__SHIFT 0x18
+#define TC_CFG_L2_STORE_POLICY0__POLICY_13_MASK 0xc000000
+#define TC_CFG_L2_STORE_POLICY0__POLICY_13__SHIFT 0x1a
+#define TC_CFG_L2_STORE_POLICY0__POLICY_14_MASK 0x30000000
+#define TC_CFG_L2_STORE_POLICY0__POLICY_14__SHIFT 0x1c
+#define TC_CFG_L2_STORE_POLICY0__POLICY_15_MASK 0xc0000000
+#define TC_CFG_L2_STORE_POLICY0__POLICY_15__SHIFT 0x1e
+#define TC_CFG_L2_STORE_POLICY1__POLICY_16_MASK 0x3
+#define TC_CFG_L2_STORE_POLICY1__POLICY_16__SHIFT 0x0
+#define TC_CFG_L2_STORE_POLICY1__POLICY_17_MASK 0xc
+#define TC_CFG_L2_STORE_POLICY1__POLICY_17__SHIFT 0x2
+#define TC_CFG_L2_STORE_POLICY1__POLICY_18_MASK 0x30
+#define TC_CFG_L2_STORE_POLICY1__POLICY_18__SHIFT 0x4
+#define TC_CFG_L2_STORE_POLICY1__POLICY_19_MASK 0xc0
+#define TC_CFG_L2_STORE_POLICY1__POLICY_19__SHIFT 0x6
+#define TC_CFG_L2_STORE_POLICY1__POLICY_20_MASK 0x300
+#define TC_CFG_L2_STORE_POLICY1__POLICY_20__SHIFT 0x8
+#define TC_CFG_L2_STORE_POLICY1__POLICY_21_MASK 0xc00
+#define TC_CFG_L2_STORE_POLICY1__POLICY_21__SHIFT 0xa
+#define TC_CFG_L2_STORE_POLICY1__POLICY_22_MASK 0x3000
+#define TC_CFG_L2_STORE_POLICY1__POLICY_22__SHIFT 0xc
+#define TC_CFG_L2_STORE_POLICY1__POLICY_23_MASK 0xc000
+#define TC_CFG_L2_STORE_POLICY1__POLICY_23__SHIFT 0xe
+#define TC_CFG_L2_STORE_POLICY1__POLICY_24_MASK 0x30000
+#define TC_CFG_L2_STORE_POLICY1__POLICY_24__SHIFT 0x10
+#define TC_CFG_L2_STORE_POLICY1__POLICY_25_MASK 0xc0000
+#define TC_CFG_L2_STORE_POLICY1__POLICY_25__SHIFT 0x12
+#define TC_CFG_L2_STORE_POLICY1__POLICY_26_MASK 0x300000
+#define TC_CFG_L2_STORE_POLICY1__POLICY_26__SHIFT 0x14
+#define TC_CFG_L2_STORE_POLICY1__POLICY_27_MASK 0xc00000
+#define TC_CFG_L2_STORE_POLICY1__POLICY_27__SHIFT 0x16
+#define TC_CFG_L2_STORE_POLICY1__POLICY_28_MASK 0x3000000
+#define TC_CFG_L2_STORE_POLICY1__POLICY_28__SHIFT 0x18
+#define TC_CFG_L2_STORE_POLICY1__POLICY_29_MASK 0xc000000
+#define TC_CFG_L2_STORE_POLICY1__POLICY_29__SHIFT 0x1a
+#define TC_CFG_L2_STORE_POLICY1__POLICY_30_MASK 0x30000000
+#define TC_CFG_L2_STORE_POLICY1__POLICY_30__SHIFT 0x1c
+#define TC_CFG_L2_STORE_POLICY1__POLICY_31_MASK 0xc0000000
+#define TC_CFG_L2_STORE_POLICY1__POLICY_31__SHIFT 0x1e
+#define TC_CFG_L2_ATOMIC_POLICY__POLICY_0_MASK 0x3
+#define TC_CFG_L2_ATOMIC_POLICY__POLICY_0__SHIFT 0x0
+#define TC_CFG_L2_ATOMIC_POLICY__POLICY_1_MASK 0xc
+#define TC_CFG_L2_ATOMIC_POLICY__POLICY_1__SHIFT 0x2
+#define TC_CFG_L2_ATOMIC_POLICY__POLICY_2_MASK 0x30
+#define TC_CFG_L2_ATOMIC_POLICY__POLICY_2__SHIFT 0x4
+#define TC_CFG_L2_ATOMIC_POLICY__POLICY_3_MASK 0xc0
+#define TC_CFG_L2_ATOMIC_POLICY__POLICY_3__SHIFT 0x6
+#define TC_CFG_L2_ATOMIC_POLICY__POLICY_4_MASK 0x300
+#define TC_CFG_L2_ATOMIC_POLICY__POLICY_4__SHIFT 0x8
+#define TC_CFG_L2_ATOMIC_POLICY__POLICY_5_MASK 0xc00
+#define TC_CFG_L2_ATOMIC_POLICY__POLICY_5__SHIFT 0xa
+#define TC_CFG_L2_ATOMIC_POLICY__POLICY_6_MASK 0x3000
+#define TC_CFG_L2_ATOMIC_POLICY__POLICY_6__SHIFT 0xc
+#define TC_CFG_L2_ATOMIC_POLICY__POLICY_7_MASK 0xc000
+#define TC_CFG_L2_ATOMIC_POLICY__POLICY_7__SHIFT 0xe
+#define TC_CFG_L2_ATOMIC_POLICY__POLICY_8_MASK 0x30000
+#define TC_CFG_L2_ATOMIC_POLICY__POLICY_8__SHIFT 0x10
+#define TC_CFG_L2_ATOMIC_POLICY__POLICY_9_MASK 0xc0000
+#define TC_CFG_L2_ATOMIC_POLICY__POLICY_9__SHIFT 0x12
+#define TC_CFG_L2_ATOMIC_POLICY__POLICY_10_MASK 0x300000
+#define TC_CFG_L2_ATOMIC_POLICY__POLICY_10__SHIFT 0x14
+#define TC_CFG_L2_ATOMIC_POLICY__POLICY_11_MASK 0xc00000
+#define TC_CFG_L2_ATOMIC_POLICY__POLICY_11__SHIFT 0x16
+#define TC_CFG_L2_ATOMIC_POLICY__POLICY_12_MASK 0x3000000
+#define TC_CFG_L2_ATOMIC_POLICY__POLICY_12__SHIFT 0x18
+#define TC_CFG_L2_ATOMIC_POLICY__POLICY_13_MASK 0xc000000
+#define TC_CFG_L2_ATOMIC_POLICY__POLICY_13__SHIFT 0x1a
+#define TC_CFG_L2_ATOMIC_POLICY__POLICY_14_MASK 0x30000000
+#define TC_CFG_L2_ATOMIC_POLICY__POLICY_14__SHIFT 0x1c
+#define TC_CFG_L2_ATOMIC_POLICY__POLICY_15_MASK 0xc0000000
+#define TC_CFG_L2_ATOMIC_POLICY__POLICY_15__SHIFT 0x1e
+#define TC_CFG_L1_VOLATILE__VOL_MASK 0xf
+#define TC_CFG_L1_VOLATILE__VOL__SHIFT 0x0
+#define TC_CFG_L2_VOLATILE__VOL_MASK 0xf
+#define TC_CFG_L2_VOLATILE__VOL__SHIFT 0x0
+#define TCP_WATCH0_ADDR_H__ADDR_MASK 0xffff
+#define TCP_WATCH0_ADDR_H__ADDR__SHIFT 0x0
+#define TCP_WATCH1_ADDR_H__ADDR_MASK 0xffff
+#define TCP_WATCH1_ADDR_H__ADDR__SHIFT 0x0
+#define TCP_WATCH2_ADDR_H__ADDR_MASK 0xffff
+#define TCP_WATCH2_ADDR_H__ADDR__SHIFT 0x0
+#define TCP_WATCH3_ADDR_H__ADDR_MASK 0xffff
+#define TCP_WATCH3_ADDR_H__ADDR__SHIFT 0x0
+#define TCP_WATCH0_ADDR_L__ADDR_MASK 0xffffffc0
+#define TCP_WATCH0_ADDR_L__ADDR__SHIFT 0x6
+#define TCP_WATCH1_ADDR_L__ADDR_MASK 0xffffffc0
+#define TCP_WATCH1_ADDR_L__ADDR__SHIFT 0x6
+#define TCP_WATCH2_ADDR_L__ADDR_MASK 0xffffffc0
+#define TCP_WATCH2_ADDR_L__ADDR__SHIFT 0x6
+#define TCP_WATCH3_ADDR_L__ADDR_MASK 0xffffffc0
+#define TCP_WATCH3_ADDR_L__ADDR__SHIFT 0x6
+#define TCP_WATCH0_CNTL__MASK_MASK 0xffffff
+#define TCP_WATCH0_CNTL__MASK__SHIFT 0x0
+#define TCP_WATCH0_CNTL__VMID_MASK 0xf000000
+#define TCP_WATCH0_CNTL__VMID__SHIFT 0x18
+#define TCP_WATCH0_CNTL__ATC_MASK 0x10000000
+#define TCP_WATCH0_CNTL__ATC__SHIFT 0x1c
+#define TCP_WATCH0_CNTL__MODE_MASK 0x60000000
+#define TCP_WATCH0_CNTL__MODE__SHIFT 0x1d
+#define TCP_WATCH0_CNTL__VALID_MASK 0x80000000
+#define TCP_WATCH0_CNTL__VALID__SHIFT 0x1f
+#define TCP_WATCH1_CNTL__MASK_MASK 0xffffff
+#define TCP_WATCH1_CNTL__MASK__SHIFT 0x0
+#define TCP_WATCH1_CNTL__VMID_MASK 0xf000000
+#define TCP_WATCH1_CNTL__VMID__SHIFT 0x18
+#define TCP_WATCH1_CNTL__ATC_MASK 0x10000000
+#define TCP_WATCH1_CNTL__ATC__SHIFT 0x1c
+#define TCP_WATCH1_CNTL__MODE_MASK 0x60000000
+#define TCP_WATCH1_CNTL__MODE__SHIFT 0x1d
+#define TCP_WATCH1_CNTL__VALID_MASK 0x80000000
+#define TCP_WATCH1_CNTL__VALID__SHIFT 0x1f
+#define TCP_WATCH2_CNTL__MASK_MASK 0xffffff
+#define TCP_WATCH2_CNTL__MASK__SHIFT 0x0
+#define TCP_WATCH2_CNTL__VMID_MASK 0xf000000
+#define TCP_WATCH2_CNTL__VMID__SHIFT 0x18
+#define TCP_WATCH2_CNTL__ATC_MASK 0x10000000
+#define TCP_WATCH2_CNTL__ATC__SHIFT 0x1c
+#define TCP_WATCH2_CNTL__MODE_MASK 0x60000000
+#define TCP_WATCH2_CNTL__MODE__SHIFT 0x1d
+#define TCP_WATCH2_CNTL__VALID_MASK 0x80000000
+#define TCP_WATCH2_CNTL__VALID__SHIFT 0x1f
+#define TCP_WATCH3_CNTL__MASK_MASK 0xffffff
+#define TCP_WATCH3_CNTL__MASK__SHIFT 0x0
+#define TCP_WATCH3_CNTL__VMID_MASK 0xf000000
+#define TCP_WATCH3_CNTL__VMID__SHIFT 0x18
+#define TCP_WATCH3_CNTL__ATC_MASK 0x10000000
+#define TCP_WATCH3_CNTL__ATC__SHIFT 0x1c
+#define TCP_WATCH3_CNTL__MODE_MASK 0x60000000
+#define TCP_WATCH3_CNTL__MODE__SHIFT 0x1d
+#define TCP_WATCH3_CNTL__VALID_MASK 0x80000000
+#define TCP_WATCH3_CNTL__VALID__SHIFT 0x1f
+#define TCP_GATCL1_CNTL__INVALIDATE_ALL_VMID_MASK 0x2000000
+#define TCP_GATCL1_CNTL__INVALIDATE_ALL_VMID__SHIFT 0x19
+#define TCP_GATCL1_CNTL__FORCE_MISS_MASK 0x4000000
+#define TCP_GATCL1_CNTL__FORCE_MISS__SHIFT 0x1a
+#define TCP_GATCL1_CNTL__FORCE_IN_ORDER_MASK 0x8000000
+#define TCP_GATCL1_CNTL__FORCE_IN_ORDER__SHIFT 0x1b
+#define TCP_GATCL1_CNTL__REDUCE_FIFO_DEPTH_BY_2_MASK 0x30000000
+#define TCP_GATCL1_CNTL__REDUCE_FIFO_DEPTH_BY_2__SHIFT 0x1c
+#define TCP_GATCL1_CNTL__REDUCE_CACHE_SIZE_BY_2_MASK 0xc0000000
+#define TCP_GATCL1_CNTL__REDUCE_CACHE_SIZE_BY_2__SHIFT 0x1e
+#define TCP_ATC_EDC_GATCL1_CNT__DATA_SEC_MASK 0xff
+#define TCP_ATC_EDC_GATCL1_CNT__DATA_SEC__SHIFT 0x0
+#define TCP_GATCL1_DSM_CNTL__SEL_DSM_TCP_GATCL1_IRRITATOR_DATA_A0_MASK 0x1
+#define TCP_GATCL1_DSM_CNTL__SEL_DSM_TCP_GATCL1_IRRITATOR_DATA_A0__SHIFT 0x0
+#define TCP_GATCL1_DSM_CNTL__SEL_DSM_TCP_GATCL1_IRRITATOR_DATA_A1_MASK 0x2
+#define TCP_GATCL1_DSM_CNTL__SEL_DSM_TCP_GATCL1_IRRITATOR_DATA_A1__SHIFT 0x1
+#define TCP_GATCL1_DSM_CNTL__TCP_GATCL1_ENABLE_SINGLE_WRITE_A_MASK 0x4
+#define TCP_GATCL1_DSM_CNTL__TCP_GATCL1_ENABLE_SINGLE_WRITE_A__SHIFT 0x2
+#define TCP_DSM_CNTL__CACHE_RAM_IRRITATOR_DATA_SEL_MASK 0x3
+#define TCP_DSM_CNTL__CACHE_RAM_IRRITATOR_DATA_SEL__SHIFT 0x0
+#define TCP_DSM_CNTL__CACHE_RAM_IRRITATOR_SINGLE_WRITE_MASK 0x4
+#define TCP_DSM_CNTL__CACHE_RAM_IRRITATOR_SINGLE_WRITE__SHIFT 0x2
+#define TCP_DSM_CNTL__LFIFO_RAM_IRRITATOR_DATA_SEL_MASK 0x18
+#define TCP_DSM_CNTL__LFIFO_RAM_IRRITATOR_DATA_SEL__SHIFT 0x3
+#define TCP_DSM_CNTL__LFIFO_RAM_IRRITATOR_SINGLE_WRITE_MASK 0x20
+#define TCP_DSM_CNTL__LFIFO_RAM_IRRITATOR_SINGLE_WRITE__SHIFT 0x5
+#define TCP_CNTL2__LS_DISABLE_CLOCKS_MASK 0xff
+#define TCP_CNTL2__LS_DISABLE_CLOCKS__SHIFT 0x0
+#define TD_CGTT_CTRL__ON_DELAY_MASK 0xf
+#define TD_CGTT_CTRL__ON_DELAY__SHIFT 0x0
+#define TD_CGTT_CTRL__OFF_HYSTERESIS_MASK 0xff0
+#define TD_CGTT_CTRL__OFF_HYSTERESIS__SHIFT 0x4
+#define TD_CGTT_CTRL__SOFT_OVERRIDE7_MASK 0x1000000
+#define TD_CGTT_CTRL__SOFT_OVERRIDE7__SHIFT 0x18
+#define TD_CGTT_CTRL__SOFT_OVERRIDE6_MASK 0x2000000
+#define TD_CGTT_CTRL__SOFT_OVERRIDE6__SHIFT 0x19
+#define TD_CGTT_CTRL__SOFT_OVERRIDE5_MASK 0x4000000
+#define TD_CGTT_CTRL__SOFT_OVERRIDE5__SHIFT 0x1a
+#define TD_CGTT_CTRL__SOFT_OVERRIDE4_MASK 0x8000000
+#define TD_CGTT_CTRL__SOFT_OVERRIDE4__SHIFT 0x1b
+#define TD_CGTT_CTRL__SOFT_OVERRIDE3_MASK 0x10000000
+#define TD_CGTT_CTRL__SOFT_OVERRIDE3__SHIFT 0x1c
+#define TD_CGTT_CTRL__SOFT_OVERRIDE2_MASK 0x20000000
+#define TD_CGTT_CTRL__SOFT_OVERRIDE2__SHIFT 0x1d
+#define TD_CGTT_CTRL__SOFT_OVERRIDE1_MASK 0x40000000
+#define TD_CGTT_CTRL__SOFT_OVERRIDE1__SHIFT 0x1e
+#define TD_CGTT_CTRL__SOFT_OVERRIDE0_MASK 0x80000000
+#define TD_CGTT_CTRL__SOFT_OVERRIDE0__SHIFT 0x1f
+#define TA_CGTT_CTRL__ON_DELAY_MASK 0xf
+#define TA_CGTT_CTRL__ON_DELAY__SHIFT 0x0
+#define TA_CGTT_CTRL__OFF_HYSTERESIS_MASK 0xff0
+#define TA_CGTT_CTRL__OFF_HYSTERESIS__SHIFT 0x4
+#define TA_CGTT_CTRL__SOFT_OVERRIDE7_MASK 0x1000000
+#define TA_CGTT_CTRL__SOFT_OVERRIDE7__SHIFT 0x18
+#define TA_CGTT_CTRL__SOFT_OVERRIDE6_MASK 0x2000000
+#define TA_CGTT_CTRL__SOFT_OVERRIDE6__SHIFT 0x19
+#define TA_CGTT_CTRL__SOFT_OVERRIDE5_MASK 0x4000000
+#define TA_CGTT_CTRL__SOFT_OVERRIDE5__SHIFT 0x1a
+#define TA_CGTT_CTRL__SOFT_OVERRIDE4_MASK 0x8000000
+#define TA_CGTT_CTRL__SOFT_OVERRIDE4__SHIFT 0x1b
+#define TA_CGTT_CTRL__SOFT_OVERRIDE3_MASK 0x10000000
+#define TA_CGTT_CTRL__SOFT_OVERRIDE3__SHIFT 0x1c
+#define TA_CGTT_CTRL__SOFT_OVERRIDE2_MASK 0x20000000
+#define TA_CGTT_CTRL__SOFT_OVERRIDE2__SHIFT 0x1d
+#define TA_CGTT_CTRL__SOFT_OVERRIDE1_MASK 0x40000000
+#define TA_CGTT_CTRL__SOFT_OVERRIDE1__SHIFT 0x1e
+#define TA_CGTT_CTRL__SOFT_OVERRIDE0_MASK 0x80000000
+#define TA_CGTT_CTRL__SOFT_OVERRIDE0__SHIFT 0x1f
+#define CGTT_TCP_CLK_CTRL__ON_DELAY_MASK 0xf
+#define CGTT_TCP_CLK_CTRL__ON_DELAY__SHIFT 0x0
+#define CGTT_TCP_CLK_CTRL__OFF_HYSTERESIS_MASK 0xff0
+#define CGTT_TCP_CLK_CTRL__OFF_HYSTERESIS__SHIFT 0x4
+#define CGTT_TCP_CLK_CTRL__SOFT_OVERRIDE7_MASK 0x1000000
+#define CGTT_TCP_CLK_CTRL__SOFT_OVERRIDE7__SHIFT 0x18
+#define CGTT_TCP_CLK_CTRL__SOFT_OVERRIDE6_MASK 0x2000000
+#define CGTT_TCP_CLK_CTRL__SOFT_OVERRIDE6__SHIFT 0x19
+#define CGTT_TCP_CLK_CTRL__SOFT_OVERRIDE5_MASK 0x4000000
+#define CGTT_TCP_CLK_CTRL__SOFT_OVERRIDE5__SHIFT 0x1a
+#define CGTT_TCP_CLK_CTRL__SOFT_OVERRIDE4_MASK 0x8000000
+#define CGTT_TCP_CLK_CTRL__SOFT_OVERRIDE4__SHIFT 0x1b
+#define CGTT_TCP_CLK_CTRL__SOFT_OVERRIDE3_MASK 0x10000000
+#define CGTT_TCP_CLK_CTRL__SOFT_OVERRIDE3__SHIFT 0x1c
+#define CGTT_TCP_CLK_CTRL__SOFT_OVERRIDE2_MASK 0x20000000
+#define CGTT_TCP_CLK_CTRL__SOFT_OVERRIDE2__SHIFT 0x1d
+#define CGTT_TCP_CLK_CTRL__SOFT_OVERRIDE1_MASK 0x40000000
+#define CGTT_TCP_CLK_CTRL__SOFT_OVERRIDE1__SHIFT 0x1e
+#define CGTT_TCP_CLK_CTRL__SOFT_OVERRIDE0_MASK 0x80000000
+#define CGTT_TCP_CLK_CTRL__SOFT_OVERRIDE0__SHIFT 0x1f
+#define CGTT_TCI_CLK_CTRL__ON_DELAY_MASK 0xf
+#define CGTT_TCI_CLK_CTRL__ON_DELAY__SHIFT 0x0
+#define CGTT_TCI_CLK_CTRL__OFF_HYSTERESIS_MASK 0xff0
+#define CGTT_TCI_CLK_CTRL__OFF_HYSTERESIS__SHIFT 0x4
+#define CGTT_TCI_CLK_CTRL__SOFT_OVERRIDE7_MASK 0x1000000
+#define CGTT_TCI_CLK_CTRL__SOFT_OVERRIDE7__SHIFT 0x18
+#define CGTT_TCI_CLK_CTRL__SOFT_OVERRIDE6_MASK 0x2000000
+#define CGTT_TCI_CLK_CTRL__SOFT_OVERRIDE6__SHIFT 0x19
+#define CGTT_TCI_CLK_CTRL__SOFT_OVERRIDE5_MASK 0x4000000
+#define CGTT_TCI_CLK_CTRL__SOFT_OVERRIDE5__SHIFT 0x1a
+#define CGTT_TCI_CLK_CTRL__SOFT_OVERRIDE4_MASK 0x8000000
+#define CGTT_TCI_CLK_CTRL__SOFT_OVERRIDE4__SHIFT 0x1b
+#define CGTT_TCI_CLK_CTRL__SOFT_OVERRIDE3_MASK 0x10000000
+#define CGTT_TCI_CLK_CTRL__SOFT_OVERRIDE3__SHIFT 0x1c
+#define CGTT_TCI_CLK_CTRL__SOFT_OVERRIDE2_MASK 0x20000000
+#define CGTT_TCI_CLK_CTRL__SOFT_OVERRIDE2__SHIFT 0x1d
+#define CGTT_TCI_CLK_CTRL__SOFT_OVERRIDE1_MASK 0x40000000
+#define CGTT_TCI_CLK_CTRL__SOFT_OVERRIDE1__SHIFT 0x1e
+#define CGTT_TCI_CLK_CTRL__SOFT_OVERRIDE0_MASK 0x80000000
+#define CGTT_TCI_CLK_CTRL__SOFT_OVERRIDE0__SHIFT 0x1f
+#define TCI_STATUS__TCI_BUSY_MASK 0x1
+#define TCI_STATUS__TCI_BUSY__SHIFT 0x0
+#define TCI_CNTL_1__WBINVL1_NUM_CYCLES_MASK 0xffff
+#define TCI_CNTL_1__WBINVL1_NUM_CYCLES__SHIFT 0x0
+#define TCI_CNTL_1__REQ_FIFO_DEPTH_MASK 0xff0000
+#define TCI_CNTL_1__REQ_FIFO_DEPTH__SHIFT 0x10
+#define TCI_CNTL_1__WDATA_RAM_DEPTH_MASK 0xff000000
+#define TCI_CNTL_1__WDATA_RAM_DEPTH__SHIFT 0x18
+#define TCI_CNTL_2__L1_INVAL_ON_WBINVL2_MASK 0x1
+#define TCI_CNTL_2__L1_INVAL_ON_WBINVL2__SHIFT 0x0
+#define TCI_CNTL_2__TCA_MAX_CREDIT_MASK 0x1fe
+#define TCI_CNTL_2__TCA_MAX_CREDIT__SHIFT 0x1
+#define GDS_CONFIG__SH0_GPR_PHASE_SEL_MASK 0x6
+#define GDS_CONFIG__SH0_GPR_PHASE_SEL__SHIFT 0x1
+#define GDS_CONFIG__SH1_GPR_PHASE_SEL_MASK 0x18
+#define GDS_CONFIG__SH1_GPR_PHASE_SEL__SHIFT 0x3
+#define GDS_CONFIG__SH2_GPR_PHASE_SEL_MASK 0x60
+#define GDS_CONFIG__SH2_GPR_PHASE_SEL__SHIFT 0x5
+#define GDS_CONFIG__SH3_GPR_PHASE_SEL_MASK 0x180
+#define GDS_CONFIG__SH3_GPR_PHASE_SEL__SHIFT 0x7
+#define GDS_CNTL_STATUS__GDS_BUSY_MASK 0x1
+#define GDS_CNTL_STATUS__GDS_BUSY__SHIFT 0x0
+#define GDS_CNTL_STATUS__GRBM_WBUF_BUSY_MASK 0x2
+#define GDS_CNTL_STATUS__GRBM_WBUF_BUSY__SHIFT 0x1
+#define GDS_CNTL_STATUS__ORD_APP_BUSY_MASK 0x4
+#define GDS_CNTL_STATUS__ORD_APP_BUSY__SHIFT 0x2
+#define GDS_CNTL_STATUS__DS_BANK_CONFLICT_MASK 0x8
+#define GDS_CNTL_STATUS__DS_BANK_CONFLICT__SHIFT 0x3
+#define GDS_CNTL_STATUS__DS_ADDR_CONFLICT_MASK 0x10
+#define GDS_CNTL_STATUS__DS_ADDR_CONFLICT__SHIFT 0x4
+#define GDS_CNTL_STATUS__DS_WR_CLAMP_MASK 0x20
+#define GDS_CNTL_STATUS__DS_WR_CLAMP__SHIFT 0x5
+#define GDS_CNTL_STATUS__DS_RD_CLAMP_MASK 0x40
+#define GDS_CNTL_STATUS__DS_RD_CLAMP__SHIFT 0x6
+#define GDS_CNTL_STATUS__GRBM_RBUF_BUSY_MASK 0x80
+#define GDS_CNTL_STATUS__GRBM_RBUF_BUSY__SHIFT 0x7
+#define GDS_CNTL_STATUS__DS_BUSY_MASK 0x100
+#define GDS_CNTL_STATUS__DS_BUSY__SHIFT 0x8
+#define GDS_CNTL_STATUS__GWS_BUSY_MASK 0x200
+#define GDS_CNTL_STATUS__GWS_BUSY__SHIFT 0x9
+#define GDS_CNTL_STATUS__ORD_FIFO_BUSY_MASK 0x400
+#define GDS_CNTL_STATUS__ORD_FIFO_BUSY__SHIFT 0xa
+#define GDS_CNTL_STATUS__CREDIT_BUSY0_MASK 0x800
+#define GDS_CNTL_STATUS__CREDIT_BUSY0__SHIFT 0xb
+#define GDS_CNTL_STATUS__CREDIT_BUSY1_MASK 0x1000
+#define GDS_CNTL_STATUS__CREDIT_BUSY1__SHIFT 0xc
+#define GDS_CNTL_STATUS__CREDIT_BUSY2_MASK 0x2000
+#define GDS_CNTL_STATUS__CREDIT_BUSY2__SHIFT 0xd
+#define GDS_CNTL_STATUS__CREDIT_BUSY3_MASK 0x4000
+#define GDS_CNTL_STATUS__CREDIT_BUSY3__SHIFT 0xe
+#define GDS_ENHANCE2__MISC_MASK 0xffff
+#define GDS_ENHANCE2__MISC__SHIFT 0x0
+#define GDS_ENHANCE2__UNUSED_MASK 0xffff0000
+#define GDS_ENHANCE2__UNUSED__SHIFT 0x10
+#define GDS_PROTECTION_FAULT__WRITE_DIS_MASK 0x1
+#define GDS_PROTECTION_FAULT__WRITE_DIS__SHIFT 0x0
+#define GDS_PROTECTION_FAULT__FAULT_DETECTED_MASK 0x2
+#define GDS_PROTECTION_FAULT__FAULT_DETECTED__SHIFT 0x1
+#define GDS_PROTECTION_FAULT__GRBM_MASK 0x4
+#define GDS_PROTECTION_FAULT__GRBM__SHIFT 0x2
+#define GDS_PROTECTION_FAULT__SH_ID_MASK 0x38
+#define GDS_PROTECTION_FAULT__SH_ID__SHIFT 0x3
+#define GDS_PROTECTION_FAULT__CU_ID_MASK 0x3c0
+#define GDS_PROTECTION_FAULT__CU_ID__SHIFT 0x6
+#define GDS_PROTECTION_FAULT__SIMD_ID_MASK 0xc00
+#define GDS_PROTECTION_FAULT__SIMD_ID__SHIFT 0xa
+#define GDS_PROTECTION_FAULT__WAVE_ID_MASK 0xf000
+#define GDS_PROTECTION_FAULT__WAVE_ID__SHIFT 0xc
+#define GDS_PROTECTION_FAULT__ADDRESS_MASK 0xffff0000
+#define GDS_PROTECTION_FAULT__ADDRESS__SHIFT 0x10
+#define GDS_VM_PROTECTION_FAULT__WRITE_DIS_MASK 0x1
+#define GDS_VM_PROTECTION_FAULT__WRITE_DIS__SHIFT 0x0
+#define GDS_VM_PROTECTION_FAULT__FAULT_DETECTED_MASK 0x2
+#define GDS_VM_PROTECTION_FAULT__FAULT_DETECTED__SHIFT 0x1
+#define GDS_VM_PROTECTION_FAULT__GWS_MASK 0x4
+#define GDS_VM_PROTECTION_FAULT__GWS__SHIFT 0x2
+#define GDS_VM_PROTECTION_FAULT__OA_MASK 0x8
+#define GDS_VM_PROTECTION_FAULT__OA__SHIFT 0x3
+#define GDS_VM_PROTECTION_FAULT__GRBM_MASK 0x10
+#define GDS_VM_PROTECTION_FAULT__GRBM__SHIFT 0x4
+#define GDS_VM_PROTECTION_FAULT__VMID_MASK 0xf00
+#define GDS_VM_PROTECTION_FAULT__VMID__SHIFT 0x8
+#define GDS_VM_PROTECTION_FAULT__ADDRESS_MASK 0xffff0000
+#define GDS_VM_PROTECTION_FAULT__ADDRESS__SHIFT 0x10
+#define GDS_EDC_CNT__DED_MASK 0xff
+#define GDS_EDC_CNT__DED__SHIFT 0x0
+#define GDS_EDC_CNT__SED_MASK 0xff00
+#define GDS_EDC_CNT__SED__SHIFT 0x8
+#define GDS_EDC_CNT__SEC_MASK 0xff0000
+#define GDS_EDC_CNT__SEC__SHIFT 0x10
+#define GDS_EDC_GRBM_CNT__DED_MASK 0xff
+#define GDS_EDC_GRBM_CNT__DED__SHIFT 0x0
+#define GDS_EDC_GRBM_CNT__SEC_MASK 0xff0000
+#define GDS_EDC_GRBM_CNT__SEC__SHIFT 0x10
+#define GDS_EDC_OA_DED__ME0_GFXHP3D_PIX_DED_MASK 0x1
+#define GDS_EDC_OA_DED__ME0_GFXHP3D_PIX_DED__SHIFT 0x0
+#define GDS_EDC_OA_DED__ME0_GFXHP3D_VTX_DED_MASK 0x2
+#define GDS_EDC_OA_DED__ME0_GFXHP3D_VTX_DED__SHIFT 0x1
+#define GDS_EDC_OA_DED__ME0_CS_DED_MASK 0x4
+#define GDS_EDC_OA_DED__ME0_CS_DED__SHIFT 0x2
+#define GDS_EDC_OA_DED__UNUSED0_MASK 0x8
+#define GDS_EDC_OA_DED__UNUSED0__SHIFT 0x3
+#define GDS_EDC_OA_DED__ME1_PIPE0_DED_MASK 0x10
+#define GDS_EDC_OA_DED__ME1_PIPE0_DED__SHIFT 0x4
+#define GDS_EDC_OA_DED__ME1_PIPE1_DED_MASK 0x20
+#define GDS_EDC_OA_DED__ME1_PIPE1_DED__SHIFT 0x5
+#define GDS_EDC_OA_DED__ME1_PIPE2_DED_MASK 0x40
+#define GDS_EDC_OA_DED__ME1_PIPE2_DED__SHIFT 0x6
+#define GDS_EDC_OA_DED__ME1_PIPE3_DED_MASK 0x80
+#define GDS_EDC_OA_DED__ME1_PIPE3_DED__SHIFT 0x7
+#define GDS_EDC_OA_DED__ME2_PIPE0_DED_MASK 0x100
+#define GDS_EDC_OA_DED__ME2_PIPE0_DED__SHIFT 0x8
+#define GDS_EDC_OA_DED__ME2_PIPE1_DED_MASK 0x200
+#define GDS_EDC_OA_DED__ME2_PIPE1_DED__SHIFT 0x9
+#define GDS_EDC_OA_DED__ME2_PIPE2_DED_MASK 0x400
+#define GDS_EDC_OA_DED__ME2_PIPE2_DED__SHIFT 0xa
+#define GDS_EDC_OA_DED__ME2_PIPE3_DED_MASK 0x800
+#define GDS_EDC_OA_DED__ME2_PIPE3_DED__SHIFT 0xb
+#define GDS_EDC_OA_DED__UNUSED1_MASK 0xfffff000
+#define GDS_EDC_OA_DED__UNUSED1__SHIFT 0xc
+#define GDS_DEBUG_CNTL__GDS_DEBUG_INDX_MASK 0x1f
+#define GDS_DEBUG_CNTL__GDS_DEBUG_INDX__SHIFT 0x0
+#define GDS_DEBUG_CNTL__UNUSED_MASK 0xffffffe0
+#define GDS_DEBUG_CNTL__UNUSED__SHIFT 0x5
+#define GDS_DEBUG_DATA__DATA_MASK 0xffffffff
+#define GDS_DEBUG_DATA__DATA__SHIFT 0x0
+#define GDS_DSM_CNTL__SEL_DSM_GDS_IRRITATOR_DATA_A_0_MASK 0x1
+#define GDS_DSM_CNTL__SEL_DSM_GDS_IRRITATOR_DATA_A_0__SHIFT 0x0
+#define GDS_DSM_CNTL__SEL_DSM_GDS_IRRITATOR_DATA_A_1_MASK 0x2
+#define GDS_DSM_CNTL__SEL_DSM_GDS_IRRITATOR_DATA_A_1__SHIFT 0x1
+#define GDS_DSM_CNTL__GDS_ENABLE_SINGLE_WRITE_A_MASK 0x4
+#define GDS_DSM_CNTL__GDS_ENABLE_SINGLE_WRITE_A__SHIFT 0x2
+#define GDS_DSM_CNTL__SEL_DSM_GDS_IRRITATOR_DATA_B_0_MASK 0x8
+#define GDS_DSM_CNTL__SEL_DSM_GDS_IRRITATOR_DATA_B_0__SHIFT 0x3
+#define GDS_DSM_CNTL__SEL_DSM_GDS_IRRITATOR_DATA_B_1_MASK 0x10
+#define GDS_DSM_CNTL__SEL_DSM_GDS_IRRITATOR_DATA_B_1__SHIFT 0x4
+#define GDS_DSM_CNTL__GDS_ENABLE_SINGLE_WRITE_B_MASK 0x20
+#define GDS_DSM_CNTL__GDS_ENABLE_SINGLE_WRITE_B__SHIFT 0x5
+#define GDS_DSM_CNTL__UNUSED_MASK 0xffffffc0
+#define GDS_DSM_CNTL__UNUSED__SHIFT 0x6
+#define CGTT_GDS_CLK_CTRL__ON_DELAY_MASK 0xf
+#define CGTT_GDS_CLK_CTRL__ON_DELAY__SHIFT 0x0
+#define CGTT_GDS_CLK_CTRL__OFF_HYSTERESIS_MASK 0xff0
+#define CGTT_GDS_CLK_CTRL__OFF_HYSTERESIS__SHIFT 0x4
+#define CGTT_GDS_CLK_CTRL__SOFT_OVERRIDE7_MASK 0x1000000
+#define CGTT_GDS_CLK_CTRL__SOFT_OVERRIDE7__SHIFT 0x18
+#define CGTT_GDS_CLK_CTRL__SOFT_OVERRIDE6_MASK 0x2000000
+#define CGTT_GDS_CLK_CTRL__SOFT_OVERRIDE6__SHIFT 0x19
+#define CGTT_GDS_CLK_CTRL__SOFT_OVERRIDE5_MASK 0x4000000
+#define CGTT_GDS_CLK_CTRL__SOFT_OVERRIDE5__SHIFT 0x1a
+#define CGTT_GDS_CLK_CTRL__SOFT_OVERRIDE4_MASK 0x8000000
+#define CGTT_GDS_CLK_CTRL__SOFT_OVERRIDE4__SHIFT 0x1b
+#define CGTT_GDS_CLK_CTRL__SOFT_OVERRIDE3_MASK 0x10000000
+#define CGTT_GDS_CLK_CTRL__SOFT_OVERRIDE3__SHIFT 0x1c
+#define CGTT_GDS_CLK_CTRL__SOFT_OVERRIDE2_MASK 0x20000000
+#define CGTT_GDS_CLK_CTRL__SOFT_OVERRIDE2__SHIFT 0x1d
+#define CGTT_GDS_CLK_CTRL__SOFT_OVERRIDE1_MASK 0x40000000
+#define CGTT_GDS_CLK_CTRL__SOFT_OVERRIDE1__SHIFT 0x1e
+#define CGTT_GDS_CLK_CTRL__SOFT_OVERRIDE0_MASK 0x80000000
+#define CGTT_GDS_CLK_CTRL__SOFT_OVERRIDE0__SHIFT 0x1f
+#define GDS_RD_ADDR__READ_ADDR_MASK 0xffffffff
+#define GDS_RD_ADDR__READ_ADDR__SHIFT 0x0
+#define GDS_RD_DATA__READ_DATA_MASK 0xffffffff
+#define GDS_RD_DATA__READ_DATA__SHIFT 0x0
+#define GDS_RD_BURST_ADDR__BURST_ADDR_MASK 0xffffffff
+#define GDS_RD_BURST_ADDR__BURST_ADDR__SHIFT 0x0
+#define GDS_RD_BURST_COUNT__BURST_COUNT_MASK 0xffffffff
+#define GDS_RD_BURST_COUNT__BURST_COUNT__SHIFT 0x0
+#define GDS_RD_BURST_DATA__BURST_DATA_MASK 0xffffffff
+#define GDS_RD_BURST_DATA__BURST_DATA__SHIFT 0x0
+#define GDS_WR_ADDR__WRITE_ADDR_MASK 0xffffffff
+#define GDS_WR_ADDR__WRITE_ADDR__SHIFT 0x0
+#define GDS_WR_DATA__WRITE_DATA_MASK 0xffffffff
+#define GDS_WR_DATA__WRITE_DATA__SHIFT 0x0
+#define GDS_WR_BURST_ADDR__WRITE_ADDR_MASK 0xffffffff
+#define GDS_WR_BURST_ADDR__WRITE_ADDR__SHIFT 0x0
+#define GDS_WR_BURST_DATA__WRITE_DATA_MASK 0xffffffff
+#define GDS_WR_BURST_DATA__WRITE_DATA__SHIFT 0x0
+#define GDS_WRITE_COMPLETE__WRITE_COMPLETE_MASK 0xffffffff
+#define GDS_WRITE_COMPLETE__WRITE_COMPLETE__SHIFT 0x0
+#define GDS_ATOM_CNTL__AINC_MASK 0x3f
+#define GDS_ATOM_CNTL__AINC__SHIFT 0x0
+#define GDS_ATOM_CNTL__UNUSED1_MASK 0xc0
+#define GDS_ATOM_CNTL__UNUSED1__SHIFT 0x6
+#define GDS_ATOM_CNTL__DMODE_MASK 0x300
+#define GDS_ATOM_CNTL__DMODE__SHIFT 0x8
+#define GDS_ATOM_CNTL__UNUSED2_MASK 0xfffffc00
+#define GDS_ATOM_CNTL__UNUSED2__SHIFT 0xa
+#define GDS_ATOM_COMPLETE__COMPLETE_MASK 0x1
+#define GDS_ATOM_COMPLETE__COMPLETE__SHIFT 0x0
+#define GDS_ATOM_COMPLETE__UNUSED_MASK 0xfffffffe
+#define GDS_ATOM_COMPLETE__UNUSED__SHIFT 0x1
+#define GDS_ATOM_BASE__BASE_MASK 0xffff
+#define GDS_ATOM_BASE__BASE__SHIFT 0x0
+#define GDS_ATOM_BASE__UNUSED_MASK 0xffff0000
+#define GDS_ATOM_BASE__UNUSED__SHIFT 0x10
+#define GDS_ATOM_SIZE__SIZE_MASK 0xffff
+#define GDS_ATOM_SIZE__SIZE__SHIFT 0x0
+#define GDS_ATOM_SIZE__UNUSED_MASK 0xffff0000
+#define GDS_ATOM_SIZE__UNUSED__SHIFT 0x10
+#define GDS_ATOM_OFFSET0__OFFSET0_MASK 0xff
+#define GDS_ATOM_OFFSET0__OFFSET0__SHIFT 0x0
+#define GDS_ATOM_OFFSET0__UNUSED_MASK 0xffffff00
+#define GDS_ATOM_OFFSET0__UNUSED__SHIFT 0x8
+#define GDS_ATOM_OFFSET1__OFFSET1_MASK 0xff
+#define GDS_ATOM_OFFSET1__OFFSET1__SHIFT 0x0
+#define GDS_ATOM_OFFSET1__UNUSED_MASK 0xffffff00
+#define GDS_ATOM_OFFSET1__UNUSED__SHIFT 0x8
+#define GDS_ATOM_DST__DST_MASK 0xffffffff
+#define GDS_ATOM_DST__DST__SHIFT 0x0
+#define GDS_ATOM_OP__OP_MASK 0xff
+#define GDS_ATOM_OP__OP__SHIFT 0x0
+#define GDS_ATOM_OP__UNUSED_MASK 0xffffff00
+#define GDS_ATOM_OP__UNUSED__SHIFT 0x8
+#define GDS_ATOM_SRC0__DATA_MASK 0xffffffff
+#define GDS_ATOM_SRC0__DATA__SHIFT 0x0
+#define GDS_ATOM_SRC0_U__DATA_MASK 0xffffffff
+#define GDS_ATOM_SRC0_U__DATA__SHIFT 0x0
+#define GDS_ATOM_SRC1__DATA_MASK 0xffffffff
+#define GDS_ATOM_SRC1__DATA__SHIFT 0x0
+#define GDS_ATOM_SRC1_U__DATA_MASK 0xffffffff
+#define GDS_ATOM_SRC1_U__DATA__SHIFT 0x0
+#define GDS_ATOM_READ0__DATA_MASK 0xffffffff
+#define GDS_ATOM_READ0__DATA__SHIFT 0x0
+#define GDS_ATOM_READ0_U__DATA_MASK 0xffffffff
+#define GDS_ATOM_READ0_U__DATA__SHIFT 0x0
+#define GDS_ATOM_READ1__DATA_MASK 0xffffffff
+#define GDS_ATOM_READ1__DATA__SHIFT 0x0
+#define GDS_ATOM_READ1_U__DATA_MASK 0xffffffff
+#define GDS_ATOM_READ1_U__DATA__SHIFT 0x0
+#define GDS_GWS_RESOURCE_CNTL__INDEX_MASK 0x3f
+#define GDS_GWS_RESOURCE_CNTL__INDEX__SHIFT 0x0
+#define GDS_GWS_RESOURCE_CNTL__UNUSED_MASK 0xffffffc0
+#define GDS_GWS_RESOURCE_CNTL__UNUSED__SHIFT 0x6
+#define GDS_GWS_RESOURCE__FLAG_MASK 0x1
+#define GDS_GWS_RESOURCE__FLAG__SHIFT 0x0
+#define GDS_GWS_RESOURCE__COUNTER_MASK 0x1ffe
+#define GDS_GWS_RESOURCE__COUNTER__SHIFT 0x1
+#define GDS_GWS_RESOURCE__TYPE_MASK 0x2000
+#define GDS_GWS_RESOURCE__TYPE__SHIFT 0xd
+#define GDS_GWS_RESOURCE__DED_MASK 0x4000
+#define GDS_GWS_RESOURCE__DED__SHIFT 0xe
+#define GDS_GWS_RESOURCE__RELEASE_ALL_MASK 0x8000
+#define GDS_GWS_RESOURCE__RELEASE_ALL__SHIFT 0xf
+#define GDS_GWS_RESOURCE__HEAD_QUEUE_MASK 0xfff0000
+#define GDS_GWS_RESOURCE__HEAD_QUEUE__SHIFT 0x10
+#define GDS_GWS_RESOURCE__HEAD_VALID_MASK 0x10000000
+#define GDS_GWS_RESOURCE__HEAD_VALID__SHIFT 0x1c
+#define GDS_GWS_RESOURCE__HEAD_FLAG_MASK 0x20000000
+#define GDS_GWS_RESOURCE__HEAD_FLAG__SHIFT 0x1d
+#define GDS_GWS_RESOURCE__UNUSED1_MASK 0xc0000000
+#define GDS_GWS_RESOURCE__UNUSED1__SHIFT 0x1e
+#define GDS_GWS_RESOURCE_CNT__RESOURCE_CNT_MASK 0xffff
+#define GDS_GWS_RESOURCE_CNT__RESOURCE_CNT__SHIFT 0x0
+#define GDS_GWS_RESOURCE_CNT__UNUSED_MASK 0xffff0000
+#define GDS_GWS_RESOURCE_CNT__UNUSED__SHIFT 0x10
+#define GDS_OA_CNTL__INDEX_MASK 0xf
+#define GDS_OA_CNTL__INDEX__SHIFT 0x0
+#define GDS_OA_CNTL__UNUSED_MASK 0xfffffff0
+#define GDS_OA_CNTL__UNUSED__SHIFT 0x4
+#define GDS_OA_COUNTER__SPACE_AVAILABLE_MASK 0xffffffff
+#define GDS_OA_COUNTER__SPACE_AVAILABLE__SHIFT 0x0
+#define GDS_OA_ADDRESS__DS_ADDRESS_MASK 0xffff
+#define GDS_OA_ADDRESS__DS_ADDRESS__SHIFT 0x0
+#define GDS_OA_ADDRESS__CRAWLER_MASK 0xf0000
+#define GDS_OA_ADDRESS__CRAWLER__SHIFT 0x10
+#define GDS_OA_ADDRESS__CRAWLER_TYPE_MASK 0x300000
+#define GDS_OA_ADDRESS__CRAWLER_TYPE__SHIFT 0x14
+#define GDS_OA_ADDRESS__UNUSED_MASK 0x3fc00000
+#define GDS_OA_ADDRESS__UNUSED__SHIFT 0x16
+#define GDS_OA_ADDRESS__NO_ALLOC_MASK 0x40000000
+#define GDS_OA_ADDRESS__NO_ALLOC__SHIFT 0x1e
+#define GDS_OA_ADDRESS__ENABLE_MASK 0x80000000
+#define GDS_OA_ADDRESS__ENABLE__SHIFT 0x1f
+#define GDS_OA_INCDEC__VALUE_MASK 0x7fffffff
+#define GDS_OA_INCDEC__VALUE__SHIFT 0x0
+#define GDS_OA_INCDEC__INCDEC_MASK 0x80000000
+#define GDS_OA_INCDEC__INCDEC__SHIFT 0x1f
+#define GDS_OA_RING_SIZE__RING_SIZE_MASK 0xffffffff
+#define GDS_OA_RING_SIZE__RING_SIZE__SHIFT 0x0
+#define GDS_DEBUG_REG0__spare1_MASK 0x3f
+#define GDS_DEBUG_REG0__spare1__SHIFT 0x0
+#define GDS_DEBUG_REG0__write_buff_valid_MASK 0x40
+#define GDS_DEBUG_REG0__write_buff_valid__SHIFT 0x6
+#define GDS_DEBUG_REG0__wr_pixel_nxt_ptr_MASK 0xf80
+#define GDS_DEBUG_REG0__wr_pixel_nxt_ptr__SHIFT 0x7
+#define GDS_DEBUG_REG0__last_pixel_ptr_MASK 0x1000
+#define GDS_DEBUG_REG0__last_pixel_ptr__SHIFT 0xc
+#define GDS_DEBUG_REG0__cstate_MASK 0x1e000
+#define GDS_DEBUG_REG0__cstate__SHIFT 0xd
+#define GDS_DEBUG_REG0__buff_write_MASK 0x20000
+#define GDS_DEBUG_REG0__buff_write__SHIFT 0x11
+#define GDS_DEBUG_REG0__flush_request_MASK 0x40000
+#define GDS_DEBUG_REG0__flush_request__SHIFT 0x12
+#define GDS_DEBUG_REG0__wr_buffer_wr_complete_MASK 0x80000
+#define GDS_DEBUG_REG0__wr_buffer_wr_complete__SHIFT 0x13
+#define GDS_DEBUG_REG0__wbuf_fifo_empty_MASK 0x100000
+#define GDS_DEBUG_REG0__wbuf_fifo_empty__SHIFT 0x14
+#define GDS_DEBUG_REG0__wbuf_fifo_full_MASK 0x200000
+#define GDS_DEBUG_REG0__wbuf_fifo_full__SHIFT 0x15
+#define GDS_DEBUG_REG0__spare_MASK 0xffc00000
+#define GDS_DEBUG_REG0__spare__SHIFT 0x16
+#define GDS_DEBUG_REG1__tag_hit_MASK 0x1
+#define GDS_DEBUG_REG1__tag_hit__SHIFT 0x0
+#define GDS_DEBUG_REG1__tag_miss_MASK 0x2
+#define GDS_DEBUG_REG1__tag_miss__SHIFT 0x1
+#define GDS_DEBUG_REG1__pixel_addr_MASK 0x1fffc
+#define GDS_DEBUG_REG1__pixel_addr__SHIFT 0x2
+#define GDS_DEBUG_REG1__pixel_vld_MASK 0x20000
+#define GDS_DEBUG_REG1__pixel_vld__SHIFT 0x11
+#define GDS_DEBUG_REG1__data_ready_MASK 0x40000
+#define GDS_DEBUG_REG1__data_ready__SHIFT 0x12
+#define GDS_DEBUG_REG1__awaiting_data_MASK 0x80000
+#define GDS_DEBUG_REG1__awaiting_data__SHIFT 0x13
+#define GDS_DEBUG_REG1__addr_fifo_full_MASK 0x100000
+#define GDS_DEBUG_REG1__addr_fifo_full__SHIFT 0x14
+#define GDS_DEBUG_REG1__addr_fifo_empty_MASK 0x200000
+#define GDS_DEBUG_REG1__addr_fifo_empty__SHIFT 0x15
+#define GDS_DEBUG_REG1__buffer_loaded_MASK 0x400000
+#define GDS_DEBUG_REG1__buffer_loaded__SHIFT 0x16
+#define GDS_DEBUG_REG1__buffer_invalid_MASK 0x800000
+#define GDS_DEBUG_REG1__buffer_invalid__SHIFT 0x17
+#define GDS_DEBUG_REG1__spare_MASK 0xff000000
+#define GDS_DEBUG_REG1__spare__SHIFT 0x18
+#define GDS_DEBUG_REG2__ds_full_MASK 0x1
+#define GDS_DEBUG_REG2__ds_full__SHIFT 0x0
+#define GDS_DEBUG_REG2__ds_credit_avail_MASK 0x2
+#define GDS_DEBUG_REG2__ds_credit_avail__SHIFT 0x1
+#define GDS_DEBUG_REG2__ord_idx_free_MASK 0x4
+#define GDS_DEBUG_REG2__ord_idx_free__SHIFT 0x2
+#define GDS_DEBUG_REG2__cmd_write_MASK 0x8
+#define GDS_DEBUG_REG2__cmd_write__SHIFT 0x3
+#define GDS_DEBUG_REG2__app_sel_MASK 0xf0
+#define GDS_DEBUG_REG2__app_sel__SHIFT 0x4
+#define GDS_DEBUG_REG2__req_MASK 0x7fff00
+#define GDS_DEBUG_REG2__req__SHIFT 0x8
+#define GDS_DEBUG_REG2__spare_MASK 0xff800000
+#define GDS_DEBUG_REG2__spare__SHIFT 0x17
+#define GDS_DEBUG_REG3__pipe_num_busy_MASK 0x7ff
+#define GDS_DEBUG_REG3__pipe_num_busy__SHIFT 0x0
+#define GDS_DEBUG_REG3__pipe0_busy_num_MASK 0x7800
+#define GDS_DEBUG_REG3__pipe0_busy_num__SHIFT 0xb
+#define GDS_DEBUG_REG3__spare_MASK 0xffff8000
+#define GDS_DEBUG_REG3__spare__SHIFT 0xf
+#define GDS_DEBUG_REG4__gws_busy_MASK 0x1
+#define GDS_DEBUG_REG4__gws_busy__SHIFT 0x0
+#define GDS_DEBUG_REG4__gws_req_MASK 0x2
+#define GDS_DEBUG_REG4__gws_req__SHIFT 0x1
+#define GDS_DEBUG_REG4__gws_out_stall_MASK 0x4
+#define GDS_DEBUG_REG4__gws_out_stall__SHIFT 0x2
+#define GDS_DEBUG_REG4__cur_reso_MASK 0x1f8
+#define GDS_DEBUG_REG4__cur_reso__SHIFT 0x3
+#define GDS_DEBUG_REG4__cur_reso_head_valid_MASK 0x200
+#define GDS_DEBUG_REG4__cur_reso_head_valid__SHIFT 0x9
+#define GDS_DEBUG_REG4__cur_reso_head_dirty_MASK 0x400
+#define GDS_DEBUG_REG4__cur_reso_head_dirty__SHIFT 0xa
+#define GDS_DEBUG_REG4__cur_reso_head_flag_MASK 0x800
+#define GDS_DEBUG_REG4__cur_reso_head_flag__SHIFT 0xb
+#define GDS_DEBUG_REG4__cur_reso_fed_MASK 0x1000
+#define GDS_DEBUG_REG4__cur_reso_fed__SHIFT 0xc
+#define GDS_DEBUG_REG4__cur_reso_barrier_MASK 0x2000
+#define GDS_DEBUG_REG4__cur_reso_barrier__SHIFT 0xd
+#define GDS_DEBUG_REG4__cur_reso_flag_MASK 0x4000
+#define GDS_DEBUG_REG4__cur_reso_flag__SHIFT 0xe
+#define GDS_DEBUG_REG4__cur_reso_cnt_gt0_MASK 0x8000
+#define GDS_DEBUG_REG4__cur_reso_cnt_gt0__SHIFT 0xf
+#define GDS_DEBUG_REG4__credit_cnt_gt0_MASK 0x10000
+#define GDS_DEBUG_REG4__credit_cnt_gt0__SHIFT 0x10
+#define GDS_DEBUG_REG4__cmd_write_MASK 0x20000
+#define GDS_DEBUG_REG4__cmd_write__SHIFT 0x11
+#define GDS_DEBUG_REG4__grbm_gws_reso_wr_MASK 0x40000
+#define GDS_DEBUG_REG4__grbm_gws_reso_wr__SHIFT 0x12
+#define GDS_DEBUG_REG4__grbm_gws_reso_rd_MASK 0x80000
+#define GDS_DEBUG_REG4__grbm_gws_reso_rd__SHIFT 0x13
+#define GDS_DEBUG_REG4__ram_read_busy_MASK 0x100000
+#define GDS_DEBUG_REG4__ram_read_busy__SHIFT 0x14
+#define GDS_DEBUG_REG4__gws_bulkfree_MASK 0x200000
+#define GDS_DEBUG_REG4__gws_bulkfree__SHIFT 0x15
+#define GDS_DEBUG_REG4__ram_gws_re_MASK 0x400000
+#define GDS_DEBUG_REG4__ram_gws_re__SHIFT 0x16
+#define GDS_DEBUG_REG4__ram_gws_we_MASK 0x800000
+#define GDS_DEBUG_REG4__ram_gws_we__SHIFT 0x17
+#define GDS_DEBUG_REG4__spare_MASK 0xff000000
+#define GDS_DEBUG_REG4__spare__SHIFT 0x18
+#define GDS_DEBUG_REG5__write_dis_MASK 0x1
+#define GDS_DEBUG_REG5__write_dis__SHIFT 0x0
+#define GDS_DEBUG_REG5__dec_error_MASK 0x2
+#define GDS_DEBUG_REG5__dec_error__SHIFT 0x1
+#define GDS_DEBUG_REG5__alloc_opco_error_MASK 0x4
+#define GDS_DEBUG_REG5__alloc_opco_error__SHIFT 0x2
+#define GDS_DEBUG_REG5__dealloc_opco_error_MASK 0x8
+#define GDS_DEBUG_REG5__dealloc_opco_error__SHIFT 0x3
+#define GDS_DEBUG_REG5__wrap_opco_error_MASK 0x10
+#define GDS_DEBUG_REG5__wrap_opco_error__SHIFT 0x4
+#define GDS_DEBUG_REG5__spare_MASK 0xe0
+#define GDS_DEBUG_REG5__spare__SHIFT 0x5
+#define GDS_DEBUG_REG5__error_ds_address_MASK 0x3fff00
+#define GDS_DEBUG_REG5__error_ds_address__SHIFT 0x8
+#define GDS_DEBUG_REG5__spare1_MASK 0xffc00000
+#define GDS_DEBUG_REG5__spare1__SHIFT 0x16
+#define GDS_DEBUG_REG6__oa_busy_MASK 0x1
+#define GDS_DEBUG_REG6__oa_busy__SHIFT 0x0
+#define GDS_DEBUG_REG6__counters_enabled_MASK 0x1e
+#define GDS_DEBUG_REG6__counters_enabled__SHIFT 0x1
+#define GDS_DEBUG_REG6__counters_busy_MASK 0x1fffe0
+#define GDS_DEBUG_REG6__counters_busy__SHIFT 0x5
+#define GDS_DEBUG_REG6__spare_MASK 0xffe00000
+#define GDS_DEBUG_REG6__spare__SHIFT 0x15
+#define GDS_PERFCOUNTER0_SELECT__PERFCOUNTER_SELECT_MASK 0x3ff
+#define GDS_PERFCOUNTER0_SELECT__PERFCOUNTER_SELECT__SHIFT 0x0
+#define GDS_PERFCOUNTER0_SELECT__PERFCOUNTER_SELECT1_MASK 0xffc00
+#define GDS_PERFCOUNTER0_SELECT__PERFCOUNTER_SELECT1__SHIFT 0xa
+#define GDS_PERFCOUNTER0_SELECT__CNTR_MODE_MASK 0xf00000
+#define GDS_PERFCOUNTER0_SELECT__CNTR_MODE__SHIFT 0x14
+#define GDS_PERFCOUNTER1_SELECT__PERFCOUNTER_SELECT_MASK 0x3ff
+#define GDS_PERFCOUNTER1_SELECT__PERFCOUNTER_SELECT__SHIFT 0x0
+#define GDS_PERFCOUNTER1_SELECT__PERFCOUNTER_SELECT1_MASK 0xffc00
+#define GDS_PERFCOUNTER1_SELECT__PERFCOUNTER_SELECT1__SHIFT 0xa
+#define GDS_PERFCOUNTER1_SELECT__CNTR_MODE_MASK 0xf00000
+#define GDS_PERFCOUNTER1_SELECT__CNTR_MODE__SHIFT 0x14
+#define GDS_PERFCOUNTER2_SELECT__PERFCOUNTER_SELECT_MASK 0x3ff
+#define GDS_PERFCOUNTER2_SELECT__PERFCOUNTER_SELECT__SHIFT 0x0
+#define GDS_PERFCOUNTER2_SELECT__PERFCOUNTER_SELECT1_MASK 0xffc00
+#define GDS_PERFCOUNTER2_SELECT__PERFCOUNTER_SELECT1__SHIFT 0xa
+#define GDS_PERFCOUNTER2_SELECT__CNTR_MODE_MASK 0xf00000
+#define GDS_PERFCOUNTER2_SELECT__CNTR_MODE__SHIFT 0x14
+#define GDS_PERFCOUNTER3_SELECT__PERFCOUNTER_SELECT_MASK 0x3ff
+#define GDS_PERFCOUNTER3_SELECT__PERFCOUNTER_SELECT__SHIFT 0x0
+#define GDS_PERFCOUNTER3_SELECT__PERFCOUNTER_SELECT1_MASK 0xffc00
+#define GDS_PERFCOUNTER3_SELECT__PERFCOUNTER_SELECT1__SHIFT 0xa
+#define GDS_PERFCOUNTER3_SELECT__CNTR_MODE_MASK 0xf00000
+#define GDS_PERFCOUNTER3_SELECT__CNTR_MODE__SHIFT 0x14
+#define GDS_PERFCOUNTER0_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define GDS_PERFCOUNTER0_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define GDS_PERFCOUNTER1_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define GDS_PERFCOUNTER1_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define GDS_PERFCOUNTER2_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define GDS_PERFCOUNTER2_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define GDS_PERFCOUNTER3_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define GDS_PERFCOUNTER3_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define GDS_PERFCOUNTER0_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define GDS_PERFCOUNTER0_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define GDS_PERFCOUNTER1_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define GDS_PERFCOUNTER1_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define GDS_PERFCOUNTER2_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define GDS_PERFCOUNTER2_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define GDS_PERFCOUNTER3_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define GDS_PERFCOUNTER3_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define GDS_PERFCOUNTER0_SELECT1__PERFCOUNTER_SELECT2_MASK 0x3ff
+#define GDS_PERFCOUNTER0_SELECT1__PERFCOUNTER_SELECT2__SHIFT 0x0
+#define GDS_PERFCOUNTER0_SELECT1__PERFCOUNTER_SELECT3_MASK 0xffc00
+#define GDS_PERFCOUNTER0_SELECT1__PERFCOUNTER_SELECT3__SHIFT 0xa
+#define GDS_VMID0_BASE__BASE_MASK 0xffff
+#define GDS_VMID0_BASE__BASE__SHIFT 0x0
+#define GDS_VMID1_BASE__BASE_MASK 0xffff
+#define GDS_VMID1_BASE__BASE__SHIFT 0x0
+#define GDS_VMID2_BASE__BASE_MASK 0xffff
+#define GDS_VMID2_BASE__BASE__SHIFT 0x0
+#define GDS_VMID3_BASE__BASE_MASK 0xffff
+#define GDS_VMID3_BASE__BASE__SHIFT 0x0
+#define GDS_VMID4_BASE__BASE_MASK 0xffff
+#define GDS_VMID4_BASE__BASE__SHIFT 0x0
+#define GDS_VMID5_BASE__BASE_MASK 0xffff
+#define GDS_VMID5_BASE__BASE__SHIFT 0x0
+#define GDS_VMID6_BASE__BASE_MASK 0xffff
+#define GDS_VMID6_BASE__BASE__SHIFT 0x0
+#define GDS_VMID7_BASE__BASE_MASK 0xffff
+#define GDS_VMID7_BASE__BASE__SHIFT 0x0
+#define GDS_VMID8_BASE__BASE_MASK 0xffff
+#define GDS_VMID8_BASE__BASE__SHIFT 0x0
+#define GDS_VMID9_BASE__BASE_MASK 0xffff
+#define GDS_VMID9_BASE__BASE__SHIFT 0x0
+#define GDS_VMID10_BASE__BASE_MASK 0xffff
+#define GDS_VMID10_BASE__BASE__SHIFT 0x0
+#define GDS_VMID11_BASE__BASE_MASK 0xffff
+#define GDS_VMID11_BASE__BASE__SHIFT 0x0
+#define GDS_VMID12_BASE__BASE_MASK 0xffff
+#define GDS_VMID12_BASE__BASE__SHIFT 0x0
+#define GDS_VMID13_BASE__BASE_MASK 0xffff
+#define GDS_VMID13_BASE__BASE__SHIFT 0x0
+#define GDS_VMID14_BASE__BASE_MASK 0xffff
+#define GDS_VMID14_BASE__BASE__SHIFT 0x0
+#define GDS_VMID15_BASE__BASE_MASK 0xffff
+#define GDS_VMID15_BASE__BASE__SHIFT 0x0
+#define GDS_VMID0_SIZE__SIZE_MASK 0x1ffff
+#define GDS_VMID0_SIZE__SIZE__SHIFT 0x0
+#define GDS_VMID1_SIZE__SIZE_MASK 0x1ffff
+#define GDS_VMID1_SIZE__SIZE__SHIFT 0x0
+#define GDS_VMID2_SIZE__SIZE_MASK 0x1ffff
+#define GDS_VMID2_SIZE__SIZE__SHIFT 0x0
+#define GDS_VMID3_SIZE__SIZE_MASK 0x1ffff
+#define GDS_VMID3_SIZE__SIZE__SHIFT 0x0
+#define GDS_VMID4_SIZE__SIZE_MASK 0x1ffff
+#define GDS_VMID4_SIZE__SIZE__SHIFT 0x0
+#define GDS_VMID5_SIZE__SIZE_MASK 0x1ffff
+#define GDS_VMID5_SIZE__SIZE__SHIFT 0x0
+#define GDS_VMID6_SIZE__SIZE_MASK 0x1ffff
+#define GDS_VMID6_SIZE__SIZE__SHIFT 0x0
+#define GDS_VMID7_SIZE__SIZE_MASK 0x1ffff
+#define GDS_VMID7_SIZE__SIZE__SHIFT 0x0
+#define GDS_VMID8_SIZE__SIZE_MASK 0x1ffff
+#define GDS_VMID8_SIZE__SIZE__SHIFT 0x0
+#define GDS_VMID9_SIZE__SIZE_MASK 0x1ffff
+#define GDS_VMID9_SIZE__SIZE__SHIFT 0x0
+#define GDS_VMID10_SIZE__SIZE_MASK 0x1ffff
+#define GDS_VMID10_SIZE__SIZE__SHIFT 0x0
+#define GDS_VMID11_SIZE__SIZE_MASK 0x1ffff
+#define GDS_VMID11_SIZE__SIZE__SHIFT 0x0
+#define GDS_VMID12_SIZE__SIZE_MASK 0x1ffff
+#define GDS_VMID12_SIZE__SIZE__SHIFT 0x0
+#define GDS_VMID13_SIZE__SIZE_MASK 0x1ffff
+#define GDS_VMID13_SIZE__SIZE__SHIFT 0x0
+#define GDS_VMID14_SIZE__SIZE_MASK 0x1ffff
+#define GDS_VMID14_SIZE__SIZE__SHIFT 0x0
+#define GDS_VMID15_SIZE__SIZE_MASK 0x1ffff
+#define GDS_VMID15_SIZE__SIZE__SHIFT 0x0
+#define GDS_GWS_VMID0__BASE_MASK 0x3f
+#define GDS_GWS_VMID0__BASE__SHIFT 0x0
+#define GDS_GWS_VMID0__SIZE_MASK 0x7f0000
+#define GDS_GWS_VMID0__SIZE__SHIFT 0x10
+#define GDS_GWS_VMID1__BASE_MASK 0x3f
+#define GDS_GWS_VMID1__BASE__SHIFT 0x0
+#define GDS_GWS_VMID1__SIZE_MASK 0x7f0000
+#define GDS_GWS_VMID1__SIZE__SHIFT 0x10
+#define GDS_GWS_VMID2__BASE_MASK 0x3f
+#define GDS_GWS_VMID2__BASE__SHIFT 0x0
+#define GDS_GWS_VMID2__SIZE_MASK 0x7f0000
+#define GDS_GWS_VMID2__SIZE__SHIFT 0x10
+#define GDS_GWS_VMID3__BASE_MASK 0x3f
+#define GDS_GWS_VMID3__BASE__SHIFT 0x0
+#define GDS_GWS_VMID3__SIZE_MASK 0x7f0000
+#define GDS_GWS_VMID3__SIZE__SHIFT 0x10
+#define GDS_GWS_VMID4__BASE_MASK 0x3f
+#define GDS_GWS_VMID4__BASE__SHIFT 0x0
+#define GDS_GWS_VMID4__SIZE_MASK 0x7f0000
+#define GDS_GWS_VMID4__SIZE__SHIFT 0x10
+#define GDS_GWS_VMID5__BASE_MASK 0x3f
+#define GDS_GWS_VMID5__BASE__SHIFT 0x0
+#define GDS_GWS_VMID5__SIZE_MASK 0x7f0000
+#define GDS_GWS_VMID5__SIZE__SHIFT 0x10
+#define GDS_GWS_VMID6__BASE_MASK 0x3f
+#define GDS_GWS_VMID6__BASE__SHIFT 0x0
+#define GDS_GWS_VMID6__SIZE_MASK 0x7f0000
+#define GDS_GWS_VMID6__SIZE__SHIFT 0x10
+#define GDS_GWS_VMID7__BASE_MASK 0x3f
+#define GDS_GWS_VMID7__BASE__SHIFT 0x0
+#define GDS_GWS_VMID7__SIZE_MASK 0x7f0000
+#define GDS_GWS_VMID7__SIZE__SHIFT 0x10
+#define GDS_GWS_VMID8__BASE_MASK 0x3f
+#define GDS_GWS_VMID8__BASE__SHIFT 0x0
+#define GDS_GWS_VMID8__SIZE_MASK 0x7f0000
+#define GDS_GWS_VMID8__SIZE__SHIFT 0x10
+#define GDS_GWS_VMID9__BASE_MASK 0x3f
+#define GDS_GWS_VMID9__BASE__SHIFT 0x0
+#define GDS_GWS_VMID9__SIZE_MASK 0x7f0000
+#define GDS_GWS_VMID9__SIZE__SHIFT 0x10
+#define GDS_GWS_VMID10__BASE_MASK 0x3f
+#define GDS_GWS_VMID10__BASE__SHIFT 0x0
+#define GDS_GWS_VMID10__SIZE_MASK 0x7f0000
+#define GDS_GWS_VMID10__SIZE__SHIFT 0x10
+#define GDS_GWS_VMID11__BASE_MASK 0x3f
+#define GDS_GWS_VMID11__BASE__SHIFT 0x0
+#define GDS_GWS_VMID11__SIZE_MASK 0x7f0000
+#define GDS_GWS_VMID11__SIZE__SHIFT 0x10
+#define GDS_GWS_VMID12__BASE_MASK 0x3f
+#define GDS_GWS_VMID12__BASE__SHIFT 0x0
+#define GDS_GWS_VMID12__SIZE_MASK 0x7f0000
+#define GDS_GWS_VMID12__SIZE__SHIFT 0x10
+#define GDS_GWS_VMID13__BASE_MASK 0x3f
+#define GDS_GWS_VMID13__BASE__SHIFT 0x0
+#define GDS_GWS_VMID13__SIZE_MASK 0x7f0000
+#define GDS_GWS_VMID13__SIZE__SHIFT 0x10
+#define GDS_GWS_VMID14__BASE_MASK 0x3f
+#define GDS_GWS_VMID14__BASE__SHIFT 0x0
+#define GDS_GWS_VMID14__SIZE_MASK 0x7f0000
+#define GDS_GWS_VMID14__SIZE__SHIFT 0x10
+#define GDS_GWS_VMID15__BASE_MASK 0x3f
+#define GDS_GWS_VMID15__BASE__SHIFT 0x0
+#define GDS_GWS_VMID15__SIZE_MASK 0x7f0000
+#define GDS_GWS_VMID15__SIZE__SHIFT 0x10
+#define GDS_OA_VMID0__MASK_MASK 0xffff
+#define GDS_OA_VMID0__MASK__SHIFT 0x0
+#define GDS_OA_VMID0__UNUSED_MASK 0xffff0000
+#define GDS_OA_VMID0__UNUSED__SHIFT 0x10
+#define GDS_OA_VMID1__MASK_MASK 0xffff
+#define GDS_OA_VMID1__MASK__SHIFT 0x0
+#define GDS_OA_VMID1__UNUSED_MASK 0xffff0000
+#define GDS_OA_VMID1__UNUSED__SHIFT 0x10
+#define GDS_OA_VMID2__MASK_MASK 0xffff
+#define GDS_OA_VMID2__MASK__SHIFT 0x0
+#define GDS_OA_VMID2__UNUSED_MASK 0xffff0000
+#define GDS_OA_VMID2__UNUSED__SHIFT 0x10
+#define GDS_OA_VMID3__MASK_MASK 0xffff
+#define GDS_OA_VMID3__MASK__SHIFT 0x0
+#define GDS_OA_VMID3__UNUSED_MASK 0xffff0000
+#define GDS_OA_VMID3__UNUSED__SHIFT 0x10
+#define GDS_OA_VMID4__MASK_MASK 0xffff
+#define GDS_OA_VMID4__MASK__SHIFT 0x0
+#define GDS_OA_VMID4__UNUSED_MASK 0xffff0000
+#define GDS_OA_VMID4__UNUSED__SHIFT 0x10
+#define GDS_OA_VMID5__MASK_MASK 0xffff
+#define GDS_OA_VMID5__MASK__SHIFT 0x0
+#define GDS_OA_VMID5__UNUSED_MASK 0xffff0000
+#define GDS_OA_VMID5__UNUSED__SHIFT 0x10
+#define GDS_OA_VMID6__MASK_MASK 0xffff
+#define GDS_OA_VMID6__MASK__SHIFT 0x0
+#define GDS_OA_VMID6__UNUSED_MASK 0xffff0000
+#define GDS_OA_VMID6__UNUSED__SHIFT 0x10
+#define GDS_OA_VMID7__MASK_MASK 0xffff
+#define GDS_OA_VMID7__MASK__SHIFT 0x0
+#define GDS_OA_VMID7__UNUSED_MASK 0xffff0000
+#define GDS_OA_VMID7__UNUSED__SHIFT 0x10
+#define GDS_OA_VMID8__MASK_MASK 0xffff
+#define GDS_OA_VMID8__MASK__SHIFT 0x0
+#define GDS_OA_VMID8__UNUSED_MASK 0xffff0000
+#define GDS_OA_VMID8__UNUSED__SHIFT 0x10
+#define GDS_OA_VMID9__MASK_MASK 0xffff
+#define GDS_OA_VMID9__MASK__SHIFT 0x0
+#define GDS_OA_VMID9__UNUSED_MASK 0xffff0000
+#define GDS_OA_VMID9__UNUSED__SHIFT 0x10
+#define GDS_OA_VMID10__MASK_MASK 0xffff
+#define GDS_OA_VMID10__MASK__SHIFT 0x0
+#define GDS_OA_VMID10__UNUSED_MASK 0xffff0000
+#define GDS_OA_VMID10__UNUSED__SHIFT 0x10
+#define GDS_OA_VMID11__MASK_MASK 0xffff
+#define GDS_OA_VMID11__MASK__SHIFT 0x0
+#define GDS_OA_VMID11__UNUSED_MASK 0xffff0000
+#define GDS_OA_VMID11__UNUSED__SHIFT 0x10
+#define GDS_OA_VMID12__MASK_MASK 0xffff
+#define GDS_OA_VMID12__MASK__SHIFT 0x0
+#define GDS_OA_VMID12__UNUSED_MASK 0xffff0000
+#define GDS_OA_VMID12__UNUSED__SHIFT 0x10
+#define GDS_OA_VMID13__MASK_MASK 0xffff
+#define GDS_OA_VMID13__MASK__SHIFT 0x0
+#define GDS_OA_VMID13__UNUSED_MASK 0xffff0000
+#define GDS_OA_VMID13__UNUSED__SHIFT 0x10
+#define GDS_OA_VMID14__MASK_MASK 0xffff
+#define GDS_OA_VMID14__MASK__SHIFT 0x0
+#define GDS_OA_VMID14__UNUSED_MASK 0xffff0000
+#define GDS_OA_VMID14__UNUSED__SHIFT 0x10
+#define GDS_OA_VMID15__MASK_MASK 0xffff
+#define GDS_OA_VMID15__MASK__SHIFT 0x0
+#define GDS_OA_VMID15__UNUSED_MASK 0xffff0000
+#define GDS_OA_VMID15__UNUSED__SHIFT 0x10
+#define GDS_GWS_RESET0__RESOURCE0_RESET_MASK 0x1
+#define GDS_GWS_RESET0__RESOURCE0_RESET__SHIFT 0x0
+#define GDS_GWS_RESET0__RESOURCE1_RESET_MASK 0x2
+#define GDS_GWS_RESET0__RESOURCE1_RESET__SHIFT 0x1
+#define GDS_GWS_RESET0__RESOURCE2_RESET_MASK 0x4
+#define GDS_GWS_RESET0__RESOURCE2_RESET__SHIFT 0x2
+#define GDS_GWS_RESET0__RESOURCE3_RESET_MASK 0x8
+#define GDS_GWS_RESET0__RESOURCE3_RESET__SHIFT 0x3
+#define GDS_GWS_RESET0__RESOURCE4_RESET_MASK 0x10
+#define GDS_GWS_RESET0__RESOURCE4_RESET__SHIFT 0x4
+#define GDS_GWS_RESET0__RESOURCE5_RESET_MASK 0x20
+#define GDS_GWS_RESET0__RESOURCE5_RESET__SHIFT 0x5
+#define GDS_GWS_RESET0__RESOURCE6_RESET_MASK 0x40
+#define GDS_GWS_RESET0__RESOURCE6_RESET__SHIFT 0x6
+#define GDS_GWS_RESET0__RESOURCE7_RESET_MASK 0x80
+#define GDS_GWS_RESET0__RESOURCE7_RESET__SHIFT 0x7
+#define GDS_GWS_RESET0__RESOURCE8_RESET_MASK 0x100
+#define GDS_GWS_RESET0__RESOURCE8_RESET__SHIFT 0x8
+#define GDS_GWS_RESET0__RESOURCE9_RESET_MASK 0x200
+#define GDS_GWS_RESET0__RESOURCE9_RESET__SHIFT 0x9
+#define GDS_GWS_RESET0__RESOURCE10_RESET_MASK 0x400
+#define GDS_GWS_RESET0__RESOURCE10_RESET__SHIFT 0xa
+#define GDS_GWS_RESET0__RESOURCE11_RESET_MASK 0x800
+#define GDS_GWS_RESET0__RESOURCE11_RESET__SHIFT 0xb
+#define GDS_GWS_RESET0__RESOURCE12_RESET_MASK 0x1000
+#define GDS_GWS_RESET0__RESOURCE12_RESET__SHIFT 0xc
+#define GDS_GWS_RESET0__RESOURCE13_RESET_MASK 0x2000
+#define GDS_GWS_RESET0__RESOURCE13_RESET__SHIFT 0xd
+#define GDS_GWS_RESET0__RESOURCE14_RESET_MASK 0x4000
+#define GDS_GWS_RESET0__RESOURCE14_RESET__SHIFT 0xe
+#define GDS_GWS_RESET0__RESOURCE15_RESET_MASK 0x8000
+#define GDS_GWS_RESET0__RESOURCE15_RESET__SHIFT 0xf
+#define GDS_GWS_RESET0__RESOURCE16_RESET_MASK 0x10000
+#define GDS_GWS_RESET0__RESOURCE16_RESET__SHIFT 0x10
+#define GDS_GWS_RESET0__RESOURCE17_RESET_MASK 0x20000
+#define GDS_GWS_RESET0__RESOURCE17_RESET__SHIFT 0x11
+#define GDS_GWS_RESET0__RESOURCE18_RESET_MASK 0x40000
+#define GDS_GWS_RESET0__RESOURCE18_RESET__SHIFT 0x12
+#define GDS_GWS_RESET0__RESOURCE19_RESET_MASK 0x80000
+#define GDS_GWS_RESET0__RESOURCE19_RESET__SHIFT 0x13
+#define GDS_GWS_RESET0__RESOURCE20_RESET_MASK 0x100000
+#define GDS_GWS_RESET0__RESOURCE20_RESET__SHIFT 0x14
+#define GDS_GWS_RESET0__RESOURCE21_RESET_MASK 0x200000
+#define GDS_GWS_RESET0__RESOURCE21_RESET__SHIFT 0x15
+#define GDS_GWS_RESET0__RESOURCE22_RESET_MASK 0x400000
+#define GDS_GWS_RESET0__RESOURCE22_RESET__SHIFT 0x16
+#define GDS_GWS_RESET0__RESOURCE23_RESET_MASK 0x800000
+#define GDS_GWS_RESET0__RESOURCE23_RESET__SHIFT 0x17
+#define GDS_GWS_RESET0__RESOURCE24_RESET_MASK 0x1000000
+#define GDS_GWS_RESET0__RESOURCE24_RESET__SHIFT 0x18
+#define GDS_GWS_RESET0__RESOURCE25_RESET_MASK 0x2000000
+#define GDS_GWS_RESET0__RESOURCE25_RESET__SHIFT 0x19
+#define GDS_GWS_RESET0__RESOURCE26_RESET_MASK 0x4000000
+#define GDS_GWS_RESET0__RESOURCE26_RESET__SHIFT 0x1a
+#define GDS_GWS_RESET0__RESOURCE27_RESET_MASK 0x8000000
+#define GDS_GWS_RESET0__RESOURCE27_RESET__SHIFT 0x1b
+#define GDS_GWS_RESET0__RESOURCE28_RESET_MASK 0x10000000
+#define GDS_GWS_RESET0__RESOURCE28_RESET__SHIFT 0x1c
+#define GDS_GWS_RESET0__RESOURCE29_RESET_MASK 0x20000000
+#define GDS_GWS_RESET0__RESOURCE29_RESET__SHIFT 0x1d
+#define GDS_GWS_RESET0__RESOURCE30_RESET_MASK 0x40000000
+#define GDS_GWS_RESET0__RESOURCE30_RESET__SHIFT 0x1e
+#define GDS_GWS_RESET0__RESOURCE31_RESET_MASK 0x80000000
+#define GDS_GWS_RESET0__RESOURCE31_RESET__SHIFT 0x1f
+#define GDS_GWS_RESET1__RESOURCE32_RESET_MASK 0x1
+#define GDS_GWS_RESET1__RESOURCE32_RESET__SHIFT 0x0
+#define GDS_GWS_RESET1__RESOURCE33_RESET_MASK 0x2
+#define GDS_GWS_RESET1__RESOURCE33_RESET__SHIFT 0x1
+#define GDS_GWS_RESET1__RESOURCE34_RESET_MASK 0x4
+#define GDS_GWS_RESET1__RESOURCE34_RESET__SHIFT 0x2
+#define GDS_GWS_RESET1__RESOURCE35_RESET_MASK 0x8
+#define GDS_GWS_RESET1__RESOURCE35_RESET__SHIFT 0x3
+#define GDS_GWS_RESET1__RESOURCE36_RESET_MASK 0x10
+#define GDS_GWS_RESET1__RESOURCE36_RESET__SHIFT 0x4
+#define GDS_GWS_RESET1__RESOURCE37_RESET_MASK 0x20
+#define GDS_GWS_RESET1__RESOURCE37_RESET__SHIFT 0x5
+#define GDS_GWS_RESET1__RESOURCE38_RESET_MASK 0x40
+#define GDS_GWS_RESET1__RESOURCE38_RESET__SHIFT 0x6
+#define GDS_GWS_RESET1__RESOURCE39_RESET_MASK 0x80
+#define GDS_GWS_RESET1__RESOURCE39_RESET__SHIFT 0x7
+#define GDS_GWS_RESET1__RESOURCE40_RESET_MASK 0x100
+#define GDS_GWS_RESET1__RESOURCE40_RESET__SHIFT 0x8
+#define GDS_GWS_RESET1__RESOURCE41_RESET_MASK 0x200
+#define GDS_GWS_RESET1__RESOURCE41_RESET__SHIFT 0x9
+#define GDS_GWS_RESET1__RESOURCE42_RESET_MASK 0x400
+#define GDS_GWS_RESET1__RESOURCE42_RESET__SHIFT 0xa
+#define GDS_GWS_RESET1__RESOURCE43_RESET_MASK 0x800
+#define GDS_GWS_RESET1__RESOURCE43_RESET__SHIFT 0xb
+#define GDS_GWS_RESET1__RESOURCE44_RESET_MASK 0x1000
+#define GDS_GWS_RESET1__RESOURCE44_RESET__SHIFT 0xc
+#define GDS_GWS_RESET1__RESOURCE45_RESET_MASK 0x2000
+#define GDS_GWS_RESET1__RESOURCE45_RESET__SHIFT 0xd
+#define GDS_GWS_RESET1__RESOURCE46_RESET_MASK 0x4000
+#define GDS_GWS_RESET1__RESOURCE46_RESET__SHIFT 0xe
+#define GDS_GWS_RESET1__RESOURCE47_RESET_MASK 0x8000
+#define GDS_GWS_RESET1__RESOURCE47_RESET__SHIFT 0xf
+#define GDS_GWS_RESET1__RESOURCE48_RESET_MASK 0x10000
+#define GDS_GWS_RESET1__RESOURCE48_RESET__SHIFT 0x10
+#define GDS_GWS_RESET1__RESOURCE49_RESET_MASK 0x20000
+#define GDS_GWS_RESET1__RESOURCE49_RESET__SHIFT 0x11
+#define GDS_GWS_RESET1__RESOURCE50_RESET_MASK 0x40000
+#define GDS_GWS_RESET1__RESOURCE50_RESET__SHIFT 0x12
+#define GDS_GWS_RESET1__RESOURCE51_RESET_MASK 0x80000
+#define GDS_GWS_RESET1__RESOURCE51_RESET__SHIFT 0x13
+#define GDS_GWS_RESET1__RESOURCE52_RESET_MASK 0x100000
+#define GDS_GWS_RESET1__RESOURCE52_RESET__SHIFT 0x14
+#define GDS_GWS_RESET1__RESOURCE53_RESET_MASK 0x200000
+#define GDS_GWS_RESET1__RESOURCE53_RESET__SHIFT 0x15
+#define GDS_GWS_RESET1__RESOURCE54_RESET_MASK 0x400000
+#define GDS_GWS_RESET1__RESOURCE54_RESET__SHIFT 0x16
+#define GDS_GWS_RESET1__RESOURCE55_RESET_MASK 0x800000
+#define GDS_GWS_RESET1__RESOURCE55_RESET__SHIFT 0x17
+#define GDS_GWS_RESET1__RESOURCE56_RESET_MASK 0x1000000
+#define GDS_GWS_RESET1__RESOURCE56_RESET__SHIFT 0x18
+#define GDS_GWS_RESET1__RESOURCE57_RESET_MASK 0x2000000
+#define GDS_GWS_RESET1__RESOURCE57_RESET__SHIFT 0x19
+#define GDS_GWS_RESET1__RESOURCE58_RESET_MASK 0x4000000
+#define GDS_GWS_RESET1__RESOURCE58_RESET__SHIFT 0x1a
+#define GDS_GWS_RESET1__RESOURCE59_RESET_MASK 0x8000000
+#define GDS_GWS_RESET1__RESOURCE59_RESET__SHIFT 0x1b
+#define GDS_GWS_RESET1__RESOURCE60_RESET_MASK 0x10000000
+#define GDS_GWS_RESET1__RESOURCE60_RESET__SHIFT 0x1c
+#define GDS_GWS_RESET1__RESOURCE61_RESET_MASK 0x20000000
+#define GDS_GWS_RESET1__RESOURCE61_RESET__SHIFT 0x1d
+#define GDS_GWS_RESET1__RESOURCE62_RESET_MASK 0x40000000
+#define GDS_GWS_RESET1__RESOURCE62_RESET__SHIFT 0x1e
+#define GDS_GWS_RESET1__RESOURCE63_RESET_MASK 0x80000000
+#define GDS_GWS_RESET1__RESOURCE63_RESET__SHIFT 0x1f
+#define GDS_GWS_RESOURCE_RESET__RESET_MASK 0x1
+#define GDS_GWS_RESOURCE_RESET__RESET__SHIFT 0x0
+#define GDS_GWS_RESOURCE_RESET__RESOURCE_ID_MASK 0xff00
+#define GDS_GWS_RESOURCE_RESET__RESOURCE_ID__SHIFT 0x8
+#define GDS_COMPUTE_MAX_WAVE_ID__MAX_WAVE_ID_MASK 0xfff
+#define GDS_COMPUTE_MAX_WAVE_ID__MAX_WAVE_ID__SHIFT 0x0
+#define GDS_OA_RESET_MASK__ME0_GFXHP3D_PIX_RESET_MASK 0x1
+#define GDS_OA_RESET_MASK__ME0_GFXHP3D_PIX_RESET__SHIFT 0x0
+#define GDS_OA_RESET_MASK__ME0_GFXHP3D_VTX_RESET_MASK 0x2
+#define GDS_OA_RESET_MASK__ME0_GFXHP3D_VTX_RESET__SHIFT 0x1
+#define GDS_OA_RESET_MASK__ME0_CS_RESET_MASK 0x4
+#define GDS_OA_RESET_MASK__ME0_CS_RESET__SHIFT 0x2
+#define GDS_OA_RESET_MASK__UNUSED0_MASK 0x8
+#define GDS_OA_RESET_MASK__UNUSED0__SHIFT 0x3
+#define GDS_OA_RESET_MASK__ME1_PIPE0_RESET_MASK 0x10
+#define GDS_OA_RESET_MASK__ME1_PIPE0_RESET__SHIFT 0x4
+#define GDS_OA_RESET_MASK__ME1_PIPE1_RESET_MASK 0x20
+#define GDS_OA_RESET_MASK__ME1_PIPE1_RESET__SHIFT 0x5
+#define GDS_OA_RESET_MASK__ME1_PIPE2_RESET_MASK 0x40
+#define GDS_OA_RESET_MASK__ME1_PIPE2_RESET__SHIFT 0x6
+#define GDS_OA_RESET_MASK__ME1_PIPE3_RESET_MASK 0x80
+#define GDS_OA_RESET_MASK__ME1_PIPE3_RESET__SHIFT 0x7
+#define GDS_OA_RESET_MASK__ME2_PIPE0_RESET_MASK 0x100
+#define GDS_OA_RESET_MASK__ME2_PIPE0_RESET__SHIFT 0x8
+#define GDS_OA_RESET_MASK__ME2_PIPE1_RESET_MASK 0x200
+#define GDS_OA_RESET_MASK__ME2_PIPE1_RESET__SHIFT 0x9
+#define GDS_OA_RESET_MASK__ME2_PIPE2_RESET_MASK 0x400
+#define GDS_OA_RESET_MASK__ME2_PIPE2_RESET__SHIFT 0xa
+#define GDS_OA_RESET_MASK__ME2_PIPE3_RESET_MASK 0x800
+#define GDS_OA_RESET_MASK__ME2_PIPE3_RESET__SHIFT 0xb
+#define GDS_OA_RESET_MASK__UNUSED1_MASK 0xfffff000
+#define GDS_OA_RESET_MASK__UNUSED1__SHIFT 0xc
+#define GDS_OA_RESET__RESET_MASK 0x1
+#define GDS_OA_RESET__RESET__SHIFT 0x0
+#define GDS_OA_RESET__PIPE_ID_MASK 0xff00
+#define GDS_OA_RESET__PIPE_ID__SHIFT 0x8
+#define GDS_ENHANCE__MISC_MASK 0xffff
+#define GDS_ENHANCE__MISC__SHIFT 0x0
+#define GDS_ENHANCE__AUTO_INC_INDEX_MASK 0x10000
+#define GDS_ENHANCE__AUTO_INC_INDEX__SHIFT 0x10
+#define GDS_ENHANCE__CGPG_RESTORE_MASK 0x20000
+#define GDS_ENHANCE__CGPG_RESTORE__SHIFT 0x11
+#define GDS_ENHANCE__UNUSED_MASK 0xfffc0000
+#define GDS_ENHANCE__UNUSED__SHIFT 0x12
+#define GDS_OA_CGPG_RESTORE__VMID_MASK 0xff
+#define GDS_OA_CGPG_RESTORE__VMID__SHIFT 0x0
+#define GDS_OA_CGPG_RESTORE__MEID_MASK 0xf00
+#define GDS_OA_CGPG_RESTORE__MEID__SHIFT 0x8
+#define GDS_OA_CGPG_RESTORE__PIPEID_MASK 0xf000
+#define GDS_OA_CGPG_RESTORE__PIPEID__SHIFT 0xc
+#define GDS_OA_CGPG_RESTORE__QUEUEID_MASK 0xf0000
+#define GDS_OA_CGPG_RESTORE__QUEUEID__SHIFT 0x10
+#define GDS_OA_CGPG_RESTORE__UNUSED_MASK 0xfff00000
+#define GDS_OA_CGPG_RESTORE__UNUSED__SHIFT 0x14
+#define GDS_CS_CTXSW_STATUS__R_MASK 0x1
+#define GDS_CS_CTXSW_STATUS__R__SHIFT 0x0
+#define GDS_CS_CTXSW_STATUS__W_MASK 0x2
+#define GDS_CS_CTXSW_STATUS__W__SHIFT 0x1
+#define GDS_CS_CTXSW_STATUS__UNUSED_MASK 0xfffffffc
+#define GDS_CS_CTXSW_STATUS__UNUSED__SHIFT 0x2
+#define GDS_CS_CTXSW_CNT0__UPDN_MASK 0xffff
+#define GDS_CS_CTXSW_CNT0__UPDN__SHIFT 0x0
+#define GDS_CS_CTXSW_CNT0__PTR_MASK 0xffff0000
+#define GDS_CS_CTXSW_CNT0__PTR__SHIFT 0x10
+#define GDS_CS_CTXSW_CNT1__UPDN_MASK 0xffff
+#define GDS_CS_CTXSW_CNT1__UPDN__SHIFT 0x0
+#define GDS_CS_CTXSW_CNT1__PTR_MASK 0xffff0000
+#define GDS_CS_CTXSW_CNT1__PTR__SHIFT 0x10
+#define GDS_CS_CTXSW_CNT2__UPDN_MASK 0xffff
+#define GDS_CS_CTXSW_CNT2__UPDN__SHIFT 0x0
+#define GDS_CS_CTXSW_CNT2__PTR_MASK 0xffff0000
+#define GDS_CS_CTXSW_CNT2__PTR__SHIFT 0x10
+#define GDS_CS_CTXSW_CNT3__UPDN_MASK 0xffff
+#define GDS_CS_CTXSW_CNT3__UPDN__SHIFT 0x0
+#define GDS_CS_CTXSW_CNT3__PTR_MASK 0xffff0000
+#define GDS_CS_CTXSW_CNT3__PTR__SHIFT 0x10
+#define GDS_GFX_CTXSW_STATUS__R_MASK 0x1
+#define GDS_GFX_CTXSW_STATUS__R__SHIFT 0x0
+#define GDS_GFX_CTXSW_STATUS__W_MASK 0x2
+#define GDS_GFX_CTXSW_STATUS__W__SHIFT 0x1
+#define GDS_GFX_CTXSW_STATUS__UNUSED_MASK 0xfffffffc
+#define GDS_GFX_CTXSW_STATUS__UNUSED__SHIFT 0x2
+#define GDS_VS_CTXSW_CNT0__UPDN_MASK 0xffff
+#define GDS_VS_CTXSW_CNT0__UPDN__SHIFT 0x0
+#define GDS_VS_CTXSW_CNT0__PTR_MASK 0xffff0000
+#define GDS_VS_CTXSW_CNT0__PTR__SHIFT 0x10
+#define GDS_VS_CTXSW_CNT1__UPDN_MASK 0xffff
+#define GDS_VS_CTXSW_CNT1__UPDN__SHIFT 0x0
+#define GDS_VS_CTXSW_CNT1__PTR_MASK 0xffff0000
+#define GDS_VS_CTXSW_CNT1__PTR__SHIFT 0x10
+#define GDS_VS_CTXSW_CNT2__UPDN_MASK 0xffff
+#define GDS_VS_CTXSW_CNT2__UPDN__SHIFT 0x0
+#define GDS_VS_CTXSW_CNT2__PTR_MASK 0xffff0000
+#define GDS_VS_CTXSW_CNT2__PTR__SHIFT 0x10
+#define GDS_VS_CTXSW_CNT3__UPDN_MASK 0xffff
+#define GDS_VS_CTXSW_CNT3__UPDN__SHIFT 0x0
+#define GDS_VS_CTXSW_CNT3__PTR_MASK 0xffff0000
+#define GDS_VS_CTXSW_CNT3__PTR__SHIFT 0x10
+#define GDS_PS0_CTXSW_CNT0__UPDN_MASK 0xffff
+#define GDS_PS0_CTXSW_CNT0__UPDN__SHIFT 0x0
+#define GDS_PS0_CTXSW_CNT0__PTR_MASK 0xffff0000
+#define GDS_PS0_CTXSW_CNT0__PTR__SHIFT 0x10
+#define GDS_PS1_CTXSW_CNT0__UPDN_MASK 0xffff
+#define GDS_PS1_CTXSW_CNT0__UPDN__SHIFT 0x0
+#define GDS_PS1_CTXSW_CNT0__PTR_MASK 0xffff0000
+#define GDS_PS1_CTXSW_CNT0__PTR__SHIFT 0x10
+#define GDS_PS2_CTXSW_CNT0__UPDN_MASK 0xffff
+#define GDS_PS2_CTXSW_CNT0__UPDN__SHIFT 0x0
+#define GDS_PS2_CTXSW_CNT0__PTR_MASK 0xffff0000
+#define GDS_PS2_CTXSW_CNT0__PTR__SHIFT 0x10
+#define GDS_PS3_CTXSW_CNT0__UPDN_MASK 0xffff
+#define GDS_PS3_CTXSW_CNT0__UPDN__SHIFT 0x0
+#define GDS_PS3_CTXSW_CNT0__PTR_MASK 0xffff0000
+#define GDS_PS3_CTXSW_CNT0__PTR__SHIFT 0x10
+#define GDS_PS4_CTXSW_CNT0__UPDN_MASK 0xffff
+#define GDS_PS4_CTXSW_CNT0__UPDN__SHIFT 0x0
+#define GDS_PS4_CTXSW_CNT0__PTR_MASK 0xffff0000
+#define GDS_PS4_CTXSW_CNT0__PTR__SHIFT 0x10
+#define GDS_PS5_CTXSW_CNT0__UPDN_MASK 0xffff
+#define GDS_PS5_CTXSW_CNT0__UPDN__SHIFT 0x0
+#define GDS_PS5_CTXSW_CNT0__PTR_MASK 0xffff0000
+#define GDS_PS5_CTXSW_CNT0__PTR__SHIFT 0x10
+#define GDS_PS6_CTXSW_CNT0__UPDN_MASK 0xffff
+#define GDS_PS6_CTXSW_CNT0__UPDN__SHIFT 0x0
+#define GDS_PS6_CTXSW_CNT0__PTR_MASK 0xffff0000
+#define GDS_PS6_CTXSW_CNT0__PTR__SHIFT 0x10
+#define GDS_PS7_CTXSW_CNT0__UPDN_MASK 0xffff
+#define GDS_PS7_CTXSW_CNT0__UPDN__SHIFT 0x0
+#define GDS_PS7_CTXSW_CNT0__PTR_MASK 0xffff0000
+#define GDS_PS7_CTXSW_CNT0__PTR__SHIFT 0x10
+#define GDS_PS0_CTXSW_CNT1__UPDN_MASK 0xffff
+#define GDS_PS0_CTXSW_CNT1__UPDN__SHIFT 0x0
+#define GDS_PS0_CTXSW_CNT1__PTR_MASK 0xffff0000
+#define GDS_PS0_CTXSW_CNT1__PTR__SHIFT 0x10
+#define GDS_PS1_CTXSW_CNT1__UPDN_MASK 0xffff
+#define GDS_PS1_CTXSW_CNT1__UPDN__SHIFT 0x0
+#define GDS_PS1_CTXSW_CNT1__PTR_MASK 0xffff0000
+#define GDS_PS1_CTXSW_CNT1__PTR__SHIFT 0x10
+#define GDS_PS2_CTXSW_CNT1__UPDN_MASK 0xffff
+#define GDS_PS2_CTXSW_CNT1__UPDN__SHIFT 0x0
+#define GDS_PS2_CTXSW_CNT1__PTR_MASK 0xffff0000
+#define GDS_PS2_CTXSW_CNT1__PTR__SHIFT 0x10
+#define GDS_PS3_CTXSW_CNT1__UPDN_MASK 0xffff
+#define GDS_PS3_CTXSW_CNT1__UPDN__SHIFT 0x0
+#define GDS_PS3_CTXSW_CNT1__PTR_MASK 0xffff0000
+#define GDS_PS3_CTXSW_CNT1__PTR__SHIFT 0x10
+#define GDS_PS4_CTXSW_CNT1__UPDN_MASK 0xffff
+#define GDS_PS4_CTXSW_CNT1__UPDN__SHIFT 0x0
+#define GDS_PS4_CTXSW_CNT1__PTR_MASK 0xffff0000
+#define GDS_PS4_CTXSW_CNT1__PTR__SHIFT 0x10
+#define GDS_PS5_CTXSW_CNT1__UPDN_MASK 0xffff
+#define GDS_PS5_CTXSW_CNT1__UPDN__SHIFT 0x0
+#define GDS_PS5_CTXSW_CNT1__PTR_MASK 0xffff0000
+#define GDS_PS5_CTXSW_CNT1__PTR__SHIFT 0x10
+#define GDS_PS6_CTXSW_CNT1__UPDN_MASK 0xffff
+#define GDS_PS6_CTXSW_CNT1__UPDN__SHIFT 0x0
+#define GDS_PS6_CTXSW_CNT1__PTR_MASK 0xffff0000
+#define GDS_PS6_CTXSW_CNT1__PTR__SHIFT 0x10
+#define GDS_PS7_CTXSW_CNT1__UPDN_MASK 0xffff
+#define GDS_PS7_CTXSW_CNT1__UPDN__SHIFT 0x0
+#define GDS_PS7_CTXSW_CNT1__PTR_MASK 0xffff0000
+#define GDS_PS7_CTXSW_CNT1__PTR__SHIFT 0x10
+#define GDS_PS0_CTXSW_CNT2__UPDN_MASK 0xffff
+#define GDS_PS0_CTXSW_CNT2__UPDN__SHIFT 0x0
+#define GDS_PS0_CTXSW_CNT2__PTR_MASK 0xffff0000
+#define GDS_PS0_CTXSW_CNT2__PTR__SHIFT 0x10
+#define GDS_PS1_CTXSW_CNT2__UPDN_MASK 0xffff
+#define GDS_PS1_CTXSW_CNT2__UPDN__SHIFT 0x0
+#define GDS_PS1_CTXSW_CNT2__PTR_MASK 0xffff0000
+#define GDS_PS1_CTXSW_CNT2__PTR__SHIFT 0x10
+#define GDS_PS2_CTXSW_CNT2__UPDN_MASK 0xffff
+#define GDS_PS2_CTXSW_CNT2__UPDN__SHIFT 0x0
+#define GDS_PS2_CTXSW_CNT2__PTR_MASK 0xffff0000
+#define GDS_PS2_CTXSW_CNT2__PTR__SHIFT 0x10
+#define GDS_PS3_CTXSW_CNT2__UPDN_MASK 0xffff
+#define GDS_PS3_CTXSW_CNT2__UPDN__SHIFT 0x0
+#define GDS_PS3_CTXSW_CNT2__PTR_MASK 0xffff0000
+#define GDS_PS3_CTXSW_CNT2__PTR__SHIFT 0x10
+#define GDS_PS4_CTXSW_CNT2__UPDN_MASK 0xffff
+#define GDS_PS4_CTXSW_CNT2__UPDN__SHIFT 0x0
+#define GDS_PS4_CTXSW_CNT2__PTR_MASK 0xffff0000
+#define GDS_PS4_CTXSW_CNT2__PTR__SHIFT 0x10
+#define GDS_PS5_CTXSW_CNT2__UPDN_MASK 0xffff
+#define GDS_PS5_CTXSW_CNT2__UPDN__SHIFT 0x0
+#define GDS_PS5_CTXSW_CNT2__PTR_MASK 0xffff0000
+#define GDS_PS5_CTXSW_CNT2__PTR__SHIFT 0x10
+#define GDS_PS6_CTXSW_CNT2__UPDN_MASK 0xffff
+#define GDS_PS6_CTXSW_CNT2__UPDN__SHIFT 0x0
+#define GDS_PS6_CTXSW_CNT2__PTR_MASK 0xffff0000
+#define GDS_PS6_CTXSW_CNT2__PTR__SHIFT 0x10
+#define GDS_PS7_CTXSW_CNT2__UPDN_MASK 0xffff
+#define GDS_PS7_CTXSW_CNT2__UPDN__SHIFT 0x0
+#define GDS_PS7_CTXSW_CNT2__PTR_MASK 0xffff0000
+#define GDS_PS7_CTXSW_CNT2__PTR__SHIFT 0x10
+#define GDS_PS0_CTXSW_CNT3__UPDN_MASK 0xffff
+#define GDS_PS0_CTXSW_CNT3__UPDN__SHIFT 0x0
+#define GDS_PS0_CTXSW_CNT3__PTR_MASK 0xffff0000
+#define GDS_PS0_CTXSW_CNT3__PTR__SHIFT 0x10
+#define GDS_PS1_CTXSW_CNT3__UPDN_MASK 0xffff
+#define GDS_PS1_CTXSW_CNT3__UPDN__SHIFT 0x0
+#define GDS_PS1_CTXSW_CNT3__PTR_MASK 0xffff0000
+#define GDS_PS1_CTXSW_CNT3__PTR__SHIFT 0x10
+#define GDS_PS2_CTXSW_CNT3__UPDN_MASK 0xffff
+#define GDS_PS2_CTXSW_CNT3__UPDN__SHIFT 0x0
+#define GDS_PS2_CTXSW_CNT3__PTR_MASK 0xffff0000
+#define GDS_PS2_CTXSW_CNT3__PTR__SHIFT 0x10
+#define GDS_PS3_CTXSW_CNT3__UPDN_MASK 0xffff
+#define GDS_PS3_CTXSW_CNT3__UPDN__SHIFT 0x0
+#define GDS_PS3_CTXSW_CNT3__PTR_MASK 0xffff0000
+#define GDS_PS3_CTXSW_CNT3__PTR__SHIFT 0x10
+#define GDS_PS4_CTXSW_CNT3__UPDN_MASK 0xffff
+#define GDS_PS4_CTXSW_CNT3__UPDN__SHIFT 0x0
+#define GDS_PS4_CTXSW_CNT3__PTR_MASK 0xffff0000
+#define GDS_PS4_CTXSW_CNT3__PTR__SHIFT 0x10
+#define GDS_PS5_CTXSW_CNT3__UPDN_MASK 0xffff
+#define GDS_PS5_CTXSW_CNT3__UPDN__SHIFT 0x0
+#define GDS_PS5_CTXSW_CNT3__PTR_MASK 0xffff0000
+#define GDS_PS5_CTXSW_CNT3__PTR__SHIFT 0x10
+#define GDS_PS6_CTXSW_CNT3__UPDN_MASK 0xffff
+#define GDS_PS6_CTXSW_CNT3__UPDN__SHIFT 0x0
+#define GDS_PS6_CTXSW_CNT3__PTR_MASK 0xffff0000
+#define GDS_PS6_CTXSW_CNT3__PTR__SHIFT 0x10
+#define GDS_PS7_CTXSW_CNT3__UPDN_MASK 0xffff
+#define GDS_PS7_CTXSW_CNT3__UPDN__SHIFT 0x0
+#define GDS_PS7_CTXSW_CNT3__PTR_MASK 0xffff0000
+#define GDS_PS7_CTXSW_CNT3__PTR__SHIFT 0x10
+#define CS_COPY_STATE__SRC_STATE_ID_MASK 0x7
+#define CS_COPY_STATE__SRC_STATE_ID__SHIFT 0x0
+#define GFX_COPY_STATE__SRC_STATE_ID_MASK 0x7
+#define GFX_COPY_STATE__SRC_STATE_ID__SHIFT 0x0
+#define VGT_DRAW_INITIATOR__SOURCE_SELECT_MASK 0x3
+#define VGT_DRAW_INITIATOR__SOURCE_SELECT__SHIFT 0x0
+#define VGT_DRAW_INITIATOR__MAJOR_MODE_MASK 0xc
+#define VGT_DRAW_INITIATOR__MAJOR_MODE__SHIFT 0x2
+#define VGT_DRAW_INITIATOR__SPRITE_EN_R6XX_MASK 0x10
+#define VGT_DRAW_INITIATOR__SPRITE_EN_R6XX__SHIFT 0x4
+#define VGT_DRAW_INITIATOR__NOT_EOP_MASK 0x20
+#define VGT_DRAW_INITIATOR__NOT_EOP__SHIFT 0x5
+#define VGT_DRAW_INITIATOR__USE_OPAQUE_MASK 0x40
+#define VGT_DRAW_INITIATOR__USE_OPAQUE__SHIFT 0x6
+#define VGT_EVENT_INITIATOR__EVENT_TYPE_MASK 0x3f
+#define VGT_EVENT_INITIATOR__EVENT_TYPE__SHIFT 0x0
+#define VGT_EVENT_INITIATOR__ADDRESS_HI_MASK 0x7fc0000
+#define VGT_EVENT_INITIATOR__ADDRESS_HI__SHIFT 0x12
+#define VGT_EVENT_INITIATOR__EXTENDED_EVENT_MASK 0x8000000
+#define VGT_EVENT_INITIATOR__EXTENDED_EVENT__SHIFT 0x1b
+#define VGT_EVENT_ADDRESS_REG__ADDRESS_LOW_MASK 0xfffffff
+#define VGT_EVENT_ADDRESS_REG__ADDRESS_LOW__SHIFT 0x0
+#define VGT_DMA_BASE_HI__BASE_ADDR_MASK 0xff
+#define VGT_DMA_BASE_HI__BASE_ADDR__SHIFT 0x0
+#define VGT_DMA_BASE__BASE_ADDR_MASK 0xffffffff
+#define VGT_DMA_BASE__BASE_ADDR__SHIFT 0x0
+#define VGT_DMA_INDEX_TYPE__INDEX_TYPE_MASK 0x3
+#define VGT_DMA_INDEX_TYPE__INDEX_TYPE__SHIFT 0x0
+#define VGT_DMA_INDEX_TYPE__SWAP_MODE_MASK 0xc
+#define VGT_DMA_INDEX_TYPE__SWAP_MODE__SHIFT 0x2
+#define VGT_DMA_INDEX_TYPE__BUF_TYPE_MASK 0x30
+#define VGT_DMA_INDEX_TYPE__BUF_TYPE__SHIFT 0x4
+#define VGT_DMA_INDEX_TYPE__RDREQ_POLICY_MASK 0x40
+#define VGT_DMA_INDEX_TYPE__RDREQ_POLICY__SHIFT 0x6
+#define VGT_DMA_INDEX_TYPE__NOT_EOP_MASK 0x200
+#define VGT_DMA_INDEX_TYPE__NOT_EOP__SHIFT 0x9
+#define VGT_DMA_INDEX_TYPE__REQ_PATH_MASK 0x400
+#define VGT_DMA_INDEX_TYPE__REQ_PATH__SHIFT 0xa
+#define VGT_DMA_INDEX_TYPE__MTYPE_MASK 0x1800
+#define VGT_DMA_INDEX_TYPE__MTYPE__SHIFT 0xb
+#define VGT_DMA_NUM_INSTANCES__NUM_INSTANCES_MASK 0xffffffff
+#define VGT_DMA_NUM_INSTANCES__NUM_INSTANCES__SHIFT 0x0
+#define IA_ENHANCE__MISC_MASK 0xffffffff
+#define IA_ENHANCE__MISC__SHIFT 0x0
+#define VGT_DMA_SIZE__NUM_INDICES_MASK 0xffffffff
+#define VGT_DMA_SIZE__NUM_INDICES__SHIFT 0x0
+#define VGT_DMA_MAX_SIZE__MAX_SIZE_MASK 0xffffffff
+#define VGT_DMA_MAX_SIZE__MAX_SIZE__SHIFT 0x0
+#define VGT_DMA_PRIMITIVE_TYPE__PRIM_TYPE_MASK 0x3f
+#define VGT_DMA_PRIMITIVE_TYPE__PRIM_TYPE__SHIFT 0x0
+#define VGT_DMA_CONTROL__PRIMGROUP_SIZE_MASK 0xffff
+#define VGT_DMA_CONTROL__PRIMGROUP_SIZE__SHIFT 0x0
+#define VGT_DMA_CONTROL__IA_SWITCH_ON_EOP_MASK 0x20000
+#define VGT_DMA_CONTROL__IA_SWITCH_ON_EOP__SHIFT 0x11
+#define VGT_DMA_CONTROL__WD_SWITCH_ON_EOP_MASK 0x100000
+#define VGT_DMA_CONTROL__WD_SWITCH_ON_EOP__SHIFT 0x14
+#define VGT_IMMED_DATA__DATA_MASK 0xffffffff
+#define VGT_IMMED_DATA__DATA__SHIFT 0x0
+#define VGT_INDEX_TYPE__INDEX_TYPE_MASK 0x3
+#define VGT_INDEX_TYPE__INDEX_TYPE__SHIFT 0x0
+#define VGT_NUM_INDICES__NUM_INDICES_MASK 0xffffffff
+#define VGT_NUM_INDICES__NUM_INDICES__SHIFT 0x0
+#define VGT_NUM_INSTANCES__NUM_INSTANCES_MASK 0xffffffff
+#define VGT_NUM_INSTANCES__NUM_INSTANCES__SHIFT 0x0
+#define VGT_PRIMITIVE_TYPE__PRIM_TYPE_MASK 0x3f
+#define VGT_PRIMITIVE_TYPE__PRIM_TYPE__SHIFT 0x0
+#define VGT_PRIMITIVEID_EN__PRIMITIVEID_EN_MASK 0x1
+#define VGT_PRIMITIVEID_EN__PRIMITIVEID_EN__SHIFT 0x0
+#define VGT_PRIMITIVEID_EN__DISABLE_RESET_ON_EOI_MASK 0x2
+#define VGT_PRIMITIVEID_EN__DISABLE_RESET_ON_EOI__SHIFT 0x1
+#define VGT_PRIMITIVEID_RESET__VALUE_MASK 0xffffffff
+#define VGT_PRIMITIVEID_RESET__VALUE__SHIFT 0x0
+#define VGT_VTX_CNT_EN__VTX_CNT_EN_MASK 0x1
+#define VGT_VTX_CNT_EN__VTX_CNT_EN__SHIFT 0x0
+#define VGT_REUSE_OFF__REUSE_OFF_MASK 0x1
+#define VGT_REUSE_OFF__REUSE_OFF__SHIFT 0x0
+#define VGT_INSTANCE_STEP_RATE_0__STEP_RATE_MASK 0xffffffff
+#define VGT_INSTANCE_STEP_RATE_0__STEP_RATE__SHIFT 0x0
+#define VGT_INSTANCE_STEP_RATE_1__STEP_RATE_MASK 0xffffffff
+#define VGT_INSTANCE_STEP_RATE_1__STEP_RATE__SHIFT 0x0
+#define VGT_MAX_VTX_INDX__MAX_INDX_MASK 0xffffffff
+#define VGT_MAX_VTX_INDX__MAX_INDX__SHIFT 0x0
+#define VGT_MIN_VTX_INDX__MIN_INDX_MASK 0xffffffff
+#define VGT_MIN_VTX_INDX__MIN_INDX__SHIFT 0x0
+#define VGT_INDX_OFFSET__INDX_OFFSET_MASK 0xffffffff
+#define VGT_INDX_OFFSET__INDX_OFFSET__SHIFT 0x0
+#define VGT_VERTEX_REUSE_BLOCK_CNTL__VTX_REUSE_DEPTH_MASK 0xff
+#define VGT_VERTEX_REUSE_BLOCK_CNTL__VTX_REUSE_DEPTH__SHIFT 0x0
+#define VGT_OUT_DEALLOC_CNTL__DEALLOC_DIST_MASK 0x7f
+#define VGT_OUT_DEALLOC_CNTL__DEALLOC_DIST__SHIFT 0x0
+#define VGT_MULTI_PRIM_IB_RESET_INDX__RESET_INDX_MASK 0xffffffff
+#define VGT_MULTI_PRIM_IB_RESET_INDX__RESET_INDX__SHIFT 0x0
+#define VGT_MULTI_PRIM_IB_RESET_EN__RESET_EN_MASK 0x1
+#define VGT_MULTI_PRIM_IB_RESET_EN__RESET_EN__SHIFT 0x0
+#define VGT_ENHANCE__MISC_MASK 0xffffffff
+#define VGT_ENHANCE__MISC__SHIFT 0x0
+#define VGT_OUTPUT_PATH_CNTL__PATH_SELECT_MASK 0x7
+#define VGT_OUTPUT_PATH_CNTL__PATH_SELECT__SHIFT 0x0
+#define VGT_HOS_CNTL__TESS_MODE_MASK 0x3
+#define VGT_HOS_CNTL__TESS_MODE__SHIFT 0x0
+#define VGT_HOS_MAX_TESS_LEVEL__MAX_TESS_MASK 0xffffffff
+#define VGT_HOS_MAX_TESS_LEVEL__MAX_TESS__SHIFT 0x0
+#define VGT_HOS_MIN_TESS_LEVEL__MIN_TESS_MASK 0xffffffff
+#define VGT_HOS_MIN_TESS_LEVEL__MIN_TESS__SHIFT 0x0
+#define VGT_HOS_REUSE_DEPTH__REUSE_DEPTH_MASK 0xff
+#define VGT_HOS_REUSE_DEPTH__REUSE_DEPTH__SHIFT 0x0
+#define VGT_GROUP_PRIM_TYPE__PRIM_TYPE_MASK 0x1f
+#define VGT_GROUP_PRIM_TYPE__PRIM_TYPE__SHIFT 0x0
+#define VGT_GROUP_PRIM_TYPE__RETAIN_ORDER_MASK 0x4000
+#define VGT_GROUP_PRIM_TYPE__RETAIN_ORDER__SHIFT 0xe
+#define VGT_GROUP_PRIM_TYPE__RETAIN_QUADS_MASK 0x8000
+#define VGT_GROUP_PRIM_TYPE__RETAIN_QUADS__SHIFT 0xf
+#define VGT_GROUP_PRIM_TYPE__PRIM_ORDER_MASK 0x70000
+#define VGT_GROUP_PRIM_TYPE__PRIM_ORDER__SHIFT 0x10
+#define VGT_GROUP_FIRST_DECR__FIRST_DECR_MASK 0xf
+#define VGT_GROUP_FIRST_DECR__FIRST_DECR__SHIFT 0x0
+#define VGT_GROUP_DECR__DECR_MASK 0xf
+#define VGT_GROUP_DECR__DECR__SHIFT 0x0
+#define VGT_GROUP_VECT_0_CNTL__COMP_X_EN_MASK 0x1
+#define VGT_GROUP_VECT_0_CNTL__COMP_X_EN__SHIFT 0x0
+#define VGT_GROUP_VECT_0_CNTL__COMP_Y_EN_MASK 0x2
+#define VGT_GROUP_VECT_0_CNTL__COMP_Y_EN__SHIFT 0x1
+#define VGT_GROUP_VECT_0_CNTL__COMP_Z_EN_MASK 0x4
+#define VGT_GROUP_VECT_0_CNTL__COMP_Z_EN__SHIFT 0x2
+#define VGT_GROUP_VECT_0_CNTL__COMP_W_EN_MASK 0x8
+#define VGT_GROUP_VECT_0_CNTL__COMP_W_EN__SHIFT 0x3
+#define VGT_GROUP_VECT_0_CNTL__STRIDE_MASK 0xff00
+#define VGT_GROUP_VECT_0_CNTL__STRIDE__SHIFT 0x8
+#define VGT_GROUP_VECT_0_CNTL__SHIFT_MASK 0xff0000
+#define VGT_GROUP_VECT_0_CNTL__SHIFT__SHIFT 0x10
+#define VGT_GROUP_VECT_1_CNTL__COMP_X_EN_MASK 0x1
+#define VGT_GROUP_VECT_1_CNTL__COMP_X_EN__SHIFT 0x0
+#define VGT_GROUP_VECT_1_CNTL__COMP_Y_EN_MASK 0x2
+#define VGT_GROUP_VECT_1_CNTL__COMP_Y_EN__SHIFT 0x1
+#define VGT_GROUP_VECT_1_CNTL__COMP_Z_EN_MASK 0x4
+#define VGT_GROUP_VECT_1_CNTL__COMP_Z_EN__SHIFT 0x2
+#define VGT_GROUP_VECT_1_CNTL__COMP_W_EN_MASK 0x8
+#define VGT_GROUP_VECT_1_CNTL__COMP_W_EN__SHIFT 0x3
+#define VGT_GROUP_VECT_1_CNTL__STRIDE_MASK 0xff00
+#define VGT_GROUP_VECT_1_CNTL__STRIDE__SHIFT 0x8
+#define VGT_GROUP_VECT_1_CNTL__SHIFT_MASK 0xff0000
+#define VGT_GROUP_VECT_1_CNTL__SHIFT__SHIFT 0x10
+#define VGT_GROUP_VECT_0_FMT_CNTL__X_CONV_MASK 0xf
+#define VGT_GROUP_VECT_0_FMT_CNTL__X_CONV__SHIFT 0x0
+#define VGT_GROUP_VECT_0_FMT_CNTL__X_OFFSET_MASK 0xf0
+#define VGT_GROUP_VECT_0_FMT_CNTL__X_OFFSET__SHIFT 0x4
+#define VGT_GROUP_VECT_0_FMT_CNTL__Y_CONV_MASK 0xf00
+#define VGT_GROUP_VECT_0_FMT_CNTL__Y_CONV__SHIFT 0x8
+#define VGT_GROUP_VECT_0_FMT_CNTL__Y_OFFSET_MASK 0xf000
+#define VGT_GROUP_VECT_0_FMT_CNTL__Y_OFFSET__SHIFT 0xc
+#define VGT_GROUP_VECT_0_FMT_CNTL__Z_CONV_MASK 0xf0000
+#define VGT_GROUP_VECT_0_FMT_CNTL__Z_CONV__SHIFT 0x10
+#define VGT_GROUP_VECT_0_FMT_CNTL__Z_OFFSET_MASK 0xf00000
+#define VGT_GROUP_VECT_0_FMT_CNTL__Z_OFFSET__SHIFT 0x14
+#define VGT_GROUP_VECT_0_FMT_CNTL__W_CONV_MASK 0xf000000
+#define VGT_GROUP_VECT_0_FMT_CNTL__W_CONV__SHIFT 0x18
+#define VGT_GROUP_VECT_0_FMT_CNTL__W_OFFSET_MASK 0xf0000000
+#define VGT_GROUP_VECT_0_FMT_CNTL__W_OFFSET__SHIFT 0x1c
+#define VGT_GROUP_VECT_1_FMT_CNTL__X_CONV_MASK 0xf
+#define VGT_GROUP_VECT_1_FMT_CNTL__X_CONV__SHIFT 0x0
+#define VGT_GROUP_VECT_1_FMT_CNTL__X_OFFSET_MASK 0xf0
+#define VGT_GROUP_VECT_1_FMT_CNTL__X_OFFSET__SHIFT 0x4
+#define VGT_GROUP_VECT_1_FMT_CNTL__Y_CONV_MASK 0xf00
+#define VGT_GROUP_VECT_1_FMT_CNTL__Y_CONV__SHIFT 0x8
+#define VGT_GROUP_VECT_1_FMT_CNTL__Y_OFFSET_MASK 0xf000
+#define VGT_GROUP_VECT_1_FMT_CNTL__Y_OFFSET__SHIFT 0xc
+#define VGT_GROUP_VECT_1_FMT_CNTL__Z_CONV_MASK 0xf0000
+#define VGT_GROUP_VECT_1_FMT_CNTL__Z_CONV__SHIFT 0x10
+#define VGT_GROUP_VECT_1_FMT_CNTL__Z_OFFSET_MASK 0xf00000
+#define VGT_GROUP_VECT_1_FMT_CNTL__Z_OFFSET__SHIFT 0x14
+#define VGT_GROUP_VECT_1_FMT_CNTL__W_CONV_MASK 0xf000000
+#define VGT_GROUP_VECT_1_FMT_CNTL__W_CONV__SHIFT 0x18
+#define VGT_GROUP_VECT_1_FMT_CNTL__W_OFFSET_MASK 0xf0000000
+#define VGT_GROUP_VECT_1_FMT_CNTL__W_OFFSET__SHIFT 0x1c
+#define VGT_VTX_VECT_EJECT_REG__PRIM_COUNT_MASK 0x3ff
+#define VGT_VTX_VECT_EJECT_REG__PRIM_COUNT__SHIFT 0x0
+#define VGT_DMA_DATA_FIFO_DEPTH__DMA_DATA_FIFO_DEPTH_MASK 0x1ff
+#define VGT_DMA_DATA_FIFO_DEPTH__DMA_DATA_FIFO_DEPTH__SHIFT 0x0
+#define VGT_DMA_DATA_FIFO_DEPTH__DMA2DRAW_FIFO_DEPTH_MASK 0x7fe00
+#define VGT_DMA_DATA_FIFO_DEPTH__DMA2DRAW_FIFO_DEPTH__SHIFT 0x9
+#define VGT_DMA_REQ_FIFO_DEPTH__DMA_REQ_FIFO_DEPTH_MASK 0x3f
+#define VGT_DMA_REQ_FIFO_DEPTH__DMA_REQ_FIFO_DEPTH__SHIFT 0x0
+#define VGT_DRAW_INIT_FIFO_DEPTH__DRAW_INIT_FIFO_DEPTH_MASK 0x3f
+#define VGT_DRAW_INIT_FIFO_DEPTH__DRAW_INIT_FIFO_DEPTH__SHIFT 0x0
+#define VGT_LAST_COPY_STATE__SRC_STATE_ID_MASK 0x7
+#define VGT_LAST_COPY_STATE__SRC_STATE_ID__SHIFT 0x0
+#define VGT_LAST_COPY_STATE__DST_STATE_ID_MASK 0x70000
+#define VGT_LAST_COPY_STATE__DST_STATE_ID__SHIFT 0x10
+#define CC_GC_SHADER_ARRAY_CONFIG__INACTIVE_CUS_MASK 0xffff0000
+#define CC_GC_SHADER_ARRAY_CONFIG__INACTIVE_CUS__SHIFT 0x10
+#define GC_USER_SHADER_ARRAY_CONFIG__INACTIVE_CUS_MASK 0xffff0000
+#define GC_USER_SHADER_ARRAY_CONFIG__INACTIVE_CUS__SHIFT 0x10
+#define VGT_GS_MODE__MODE_MASK 0x7
+#define VGT_GS_MODE__MODE__SHIFT 0x0
+#define VGT_GS_MODE__RESERVED_0_MASK 0x8
+#define VGT_GS_MODE__RESERVED_0__SHIFT 0x3
+#define VGT_GS_MODE__CUT_MODE_MASK 0x30
+#define VGT_GS_MODE__CUT_MODE__SHIFT 0x4
+#define VGT_GS_MODE__RESERVED_1_MASK 0x7c0
+#define VGT_GS_MODE__RESERVED_1__SHIFT 0x6
+#define VGT_GS_MODE__GS_C_PACK_EN_MASK 0x800
+#define VGT_GS_MODE__GS_C_PACK_EN__SHIFT 0xb
+#define VGT_GS_MODE__RESERVED_2_MASK 0x1000
+#define VGT_GS_MODE__RESERVED_2__SHIFT 0xc
+#define VGT_GS_MODE__ES_PASSTHRU_MASK 0x2000
+#define VGT_GS_MODE__ES_PASSTHRU__SHIFT 0xd
+#define VGT_GS_MODE__RESERVED_3_MASK 0x4000
+#define VGT_GS_MODE__RESERVED_3__SHIFT 0xe
+#define VGT_GS_MODE__RESERVED_4_MASK 0x8000
+#define VGT_GS_MODE__RESERVED_4__SHIFT 0xf
+#define VGT_GS_MODE__RESERVED_5_MASK 0x10000
+#define VGT_GS_MODE__RESERVED_5__SHIFT 0x10
+#define VGT_GS_MODE__PARTIAL_THD_AT_EOI_MASK 0x20000
+#define VGT_GS_MODE__PARTIAL_THD_AT_EOI__SHIFT 0x11
+#define VGT_GS_MODE__SUPPRESS_CUTS_MASK 0x40000
+#define VGT_GS_MODE__SUPPRESS_CUTS__SHIFT 0x12
+#define VGT_GS_MODE__ES_WRITE_OPTIMIZE_MASK 0x80000
+#define VGT_GS_MODE__ES_WRITE_OPTIMIZE__SHIFT 0x13
+#define VGT_GS_MODE__GS_WRITE_OPTIMIZE_MASK 0x100000
+#define VGT_GS_MODE__GS_WRITE_OPTIMIZE__SHIFT 0x14
+#define VGT_GS_MODE__ONCHIP_MASK 0x600000
+#define VGT_GS_MODE__ONCHIP__SHIFT 0x15
+#define VGT_GS_ONCHIP_CNTL__ES_VERTS_PER_SUBGRP_MASK 0x7ff
+#define VGT_GS_ONCHIP_CNTL__ES_VERTS_PER_SUBGRP__SHIFT 0x0
+#define VGT_GS_ONCHIP_CNTL__GS_PRIMS_PER_SUBGRP_MASK 0x3ff800
+#define VGT_GS_ONCHIP_CNTL__GS_PRIMS_PER_SUBGRP__SHIFT 0xb
+#define VGT_GS_OUT_PRIM_TYPE__OUTPRIM_TYPE_MASK 0x3f
+#define VGT_GS_OUT_PRIM_TYPE__OUTPRIM_TYPE__SHIFT 0x0
+#define VGT_GS_OUT_PRIM_TYPE__OUTPRIM_TYPE_1_MASK 0x3f00
+#define VGT_GS_OUT_PRIM_TYPE__OUTPRIM_TYPE_1__SHIFT 0x8
+#define VGT_GS_OUT_PRIM_TYPE__OUTPRIM_TYPE_2_MASK 0x3f0000
+#define VGT_GS_OUT_PRIM_TYPE__OUTPRIM_TYPE_2__SHIFT 0x10
+#define VGT_GS_OUT_PRIM_TYPE__OUTPRIM_TYPE_3_MASK 0xfc00000
+#define VGT_GS_OUT_PRIM_TYPE__OUTPRIM_TYPE_3__SHIFT 0x16
+#define VGT_GS_OUT_PRIM_TYPE__UNIQUE_TYPE_PER_STREAM_MASK 0x80000000
+#define VGT_GS_OUT_PRIM_TYPE__UNIQUE_TYPE_PER_STREAM__SHIFT 0x1f
+#define VGT_CACHE_INVALIDATION__CACHE_INVALIDATION_MASK 0x3
+#define VGT_CACHE_INVALIDATION__CACHE_INVALIDATION__SHIFT 0x0
+#define VGT_CACHE_INVALIDATION__DIS_INSTANCING_OPT_MASK 0x10
+#define VGT_CACHE_INVALIDATION__DIS_INSTANCING_OPT__SHIFT 0x4
+#define VGT_CACHE_INVALIDATION__VS_NO_EXTRA_BUFFER_MASK 0x20
+#define VGT_CACHE_INVALIDATION__VS_NO_EXTRA_BUFFER__SHIFT 0x5
+#define VGT_CACHE_INVALIDATION__AUTO_INVLD_EN_MASK 0xc0
+#define VGT_CACHE_INVALIDATION__AUTO_INVLD_EN__SHIFT 0x6
+#define VGT_CACHE_INVALIDATION__USE_GS_DONE_MASK 0x200
+#define VGT_CACHE_INVALIDATION__USE_GS_DONE__SHIFT 0x9
+#define VGT_CACHE_INVALIDATION__DIS_RANGE_FULL_INVLD_MASK 0x800
+#define VGT_CACHE_INVALIDATION__DIS_RANGE_FULL_INVLD__SHIFT 0xb
+#define VGT_CACHE_INVALIDATION__GS_LATE_ALLOC_EN_MASK 0x1000
+#define VGT_CACHE_INVALIDATION__GS_LATE_ALLOC_EN__SHIFT 0xc
+#define VGT_CACHE_INVALIDATION__STREAMOUT_FULL_FLUSH_MASK 0x2000
+#define VGT_CACHE_INVALIDATION__STREAMOUT_FULL_FLUSH__SHIFT 0xd
+#define VGT_CACHE_INVALIDATION__ES_LIMIT_MASK 0x1f0000
+#define VGT_CACHE_INVALIDATION__ES_LIMIT__SHIFT 0x10
+#define VGT_RESET_DEBUG__GS_DISABLE_MASK 0x1
+#define VGT_RESET_DEBUG__GS_DISABLE__SHIFT 0x0
+#define VGT_RESET_DEBUG__TESS_DISABLE_MASK 0x2
+#define VGT_RESET_DEBUG__TESS_DISABLE__SHIFT 0x1
+#define VGT_RESET_DEBUG__WD_DISABLE_MASK 0x4
+#define VGT_RESET_DEBUG__WD_DISABLE__SHIFT 0x2
+#define VGT_STRMOUT_DELAY__SKIP_DELAY_MASK 0xff
+#define VGT_STRMOUT_DELAY__SKIP_DELAY__SHIFT 0x0
+#define VGT_STRMOUT_DELAY__SE0_WD_DELAY_MASK 0x700
+#define VGT_STRMOUT_DELAY__SE0_WD_DELAY__SHIFT 0x8
+#define VGT_STRMOUT_DELAY__SE1_WD_DELAY_MASK 0x3800
+#define VGT_STRMOUT_DELAY__SE1_WD_DELAY__SHIFT 0xb
+#define VGT_STRMOUT_DELAY__SE2_WD_DELAY_MASK 0x1c000
+#define VGT_STRMOUT_DELAY__SE2_WD_DELAY__SHIFT 0xe
+#define VGT_STRMOUT_DELAY__SE3_WD_DELAY_MASK 0xe0000
+#define VGT_STRMOUT_DELAY__SE3_WD_DELAY__SHIFT 0x11
+#define VGT_FIFO_DEPTHS__VS_DEALLOC_TBL_DEPTH_MASK 0x7f
+#define VGT_FIFO_DEPTHS__VS_DEALLOC_TBL_DEPTH__SHIFT 0x0
+#define VGT_FIFO_DEPTHS__RESERVED_0_MASK 0x80
+#define VGT_FIFO_DEPTHS__RESERVED_0__SHIFT 0x7
+#define VGT_FIFO_DEPTHS__CLIPP_FIFO_DEPTH_MASK 0x3fff00
+#define VGT_FIFO_DEPTHS__CLIPP_FIFO_DEPTH__SHIFT 0x8
+#define VGT_FIFO_DEPTHS__HSINPUT_FIFO_DEPTH_MASK 0xfc00000
+#define VGT_FIFO_DEPTHS__HSINPUT_FIFO_DEPTH__SHIFT 0x16
+#define VGT_GS_PER_ES__GS_PER_ES_MASK 0x7ff
+#define VGT_GS_PER_ES__GS_PER_ES__SHIFT 0x0
+#define VGT_ES_PER_GS__ES_PER_GS_MASK 0x7ff
+#define VGT_ES_PER_GS__ES_PER_GS__SHIFT 0x0
+#define VGT_GS_PER_VS__GS_PER_VS_MASK 0xf
+#define VGT_GS_PER_VS__GS_PER_VS__SHIFT 0x0
+#define VGT_GS_VERTEX_REUSE__VERT_REUSE_MASK 0x1f
+#define VGT_GS_VERTEX_REUSE__VERT_REUSE__SHIFT 0x0
+#define VGT_MC_LAT_CNTL__MC_TIME_STAMP_RES_MASK 0x3
+#define VGT_MC_LAT_CNTL__MC_TIME_STAMP_RES__SHIFT 0x0
+#define IA_CNTL_STATUS__IA_BUSY_MASK 0x1
+#define IA_CNTL_STATUS__IA_BUSY__SHIFT 0x0
+#define IA_CNTL_STATUS__IA_DMA_BUSY_MASK 0x2
+#define IA_CNTL_STATUS__IA_DMA_BUSY__SHIFT 0x1
+#define IA_CNTL_STATUS__IA_DMA_REQ_BUSY_MASK 0x4
+#define IA_CNTL_STATUS__IA_DMA_REQ_BUSY__SHIFT 0x2
+#define IA_CNTL_STATUS__IA_GRP_BUSY_MASK 0x8
+#define IA_CNTL_STATUS__IA_GRP_BUSY__SHIFT 0x3
+#define IA_CNTL_STATUS__IA_ADC_BUSY_MASK 0x10
+#define IA_CNTL_STATUS__IA_ADC_BUSY__SHIFT 0x4
+#define VGT_STRMOUT_CONFIG__STREAMOUT_0_EN_MASK 0x1
+#define VGT_STRMOUT_CONFIG__STREAMOUT_0_EN__SHIFT 0x0
+#define VGT_STRMOUT_CONFIG__STREAMOUT_1_EN_MASK 0x2
+#define VGT_STRMOUT_CONFIG__STREAMOUT_1_EN__SHIFT 0x1
+#define VGT_STRMOUT_CONFIG__STREAMOUT_2_EN_MASK 0x4
+#define VGT_STRMOUT_CONFIG__STREAMOUT_2_EN__SHIFT 0x2
+#define VGT_STRMOUT_CONFIG__STREAMOUT_3_EN_MASK 0x8
+#define VGT_STRMOUT_CONFIG__STREAMOUT_3_EN__SHIFT 0x3
+#define VGT_STRMOUT_CONFIG__RAST_STREAM_MASK 0x70
+#define VGT_STRMOUT_CONFIG__RAST_STREAM__SHIFT 0x4
+#define VGT_STRMOUT_CONFIG__RAST_STREAM_MASK_MASK 0xf00
+#define VGT_STRMOUT_CONFIG__RAST_STREAM_MASK__SHIFT 0x8
+#define VGT_STRMOUT_CONFIG__USE_RAST_STREAM_MASK_MASK 0x80000000
+#define VGT_STRMOUT_CONFIG__USE_RAST_STREAM_MASK__SHIFT 0x1f
+#define VGT_STRMOUT_BUFFER_SIZE_0__SIZE_MASK 0xffffffff
+#define VGT_STRMOUT_BUFFER_SIZE_0__SIZE__SHIFT 0x0
+#define VGT_STRMOUT_BUFFER_SIZE_1__SIZE_MASK 0xffffffff
+#define VGT_STRMOUT_BUFFER_SIZE_1__SIZE__SHIFT 0x0
+#define VGT_STRMOUT_BUFFER_SIZE_2__SIZE_MASK 0xffffffff
+#define VGT_STRMOUT_BUFFER_SIZE_2__SIZE__SHIFT 0x0
+#define VGT_STRMOUT_BUFFER_SIZE_3__SIZE_MASK 0xffffffff
+#define VGT_STRMOUT_BUFFER_SIZE_3__SIZE__SHIFT 0x0
+#define VGT_STRMOUT_BUFFER_OFFSET_0__OFFSET_MASK 0xffffffff
+#define VGT_STRMOUT_BUFFER_OFFSET_0__OFFSET__SHIFT 0x0
+#define VGT_STRMOUT_BUFFER_OFFSET_1__OFFSET_MASK 0xffffffff
+#define VGT_STRMOUT_BUFFER_OFFSET_1__OFFSET__SHIFT 0x0
+#define VGT_STRMOUT_BUFFER_OFFSET_2__OFFSET_MASK 0xffffffff
+#define VGT_STRMOUT_BUFFER_OFFSET_2__OFFSET__SHIFT 0x0
+#define VGT_STRMOUT_BUFFER_OFFSET_3__OFFSET_MASK 0xffffffff
+#define VGT_STRMOUT_BUFFER_OFFSET_3__OFFSET__SHIFT 0x0
+#define VGT_STRMOUT_VTX_STRIDE_0__STRIDE_MASK 0x3ff
+#define VGT_STRMOUT_VTX_STRIDE_0__STRIDE__SHIFT 0x0
+#define VGT_STRMOUT_VTX_STRIDE_1__STRIDE_MASK 0x3ff
+#define VGT_STRMOUT_VTX_STRIDE_1__STRIDE__SHIFT 0x0
+#define VGT_STRMOUT_VTX_STRIDE_2__STRIDE_MASK 0x3ff
+#define VGT_STRMOUT_VTX_STRIDE_2__STRIDE__SHIFT 0x0
+#define VGT_STRMOUT_VTX_STRIDE_3__STRIDE_MASK 0x3ff
+#define VGT_STRMOUT_VTX_STRIDE_3__STRIDE__SHIFT 0x0
+#define VGT_STRMOUT_BUFFER_CONFIG__STREAM_0_BUFFER_EN_MASK 0xf
+#define VGT_STRMOUT_BUFFER_CONFIG__STREAM_0_BUFFER_EN__SHIFT 0x0
+#define VGT_STRMOUT_BUFFER_CONFIG__STREAM_1_BUFFER_EN_MASK 0xf0
+#define VGT_STRMOUT_BUFFER_CONFIG__STREAM_1_BUFFER_EN__SHIFT 0x4
+#define VGT_STRMOUT_BUFFER_CONFIG__STREAM_2_BUFFER_EN_MASK 0xf00
+#define VGT_STRMOUT_BUFFER_CONFIG__STREAM_2_BUFFER_EN__SHIFT 0x8
+#define VGT_STRMOUT_BUFFER_CONFIG__STREAM_3_BUFFER_EN_MASK 0xf000
+#define VGT_STRMOUT_BUFFER_CONFIG__STREAM_3_BUFFER_EN__SHIFT 0xc
+#define VGT_STRMOUT_BUFFER_FILLED_SIZE_0__SIZE_MASK 0xffffffff
+#define VGT_STRMOUT_BUFFER_FILLED_SIZE_0__SIZE__SHIFT 0x0
+#define VGT_STRMOUT_BUFFER_FILLED_SIZE_1__SIZE_MASK 0xffffffff
+#define VGT_STRMOUT_BUFFER_FILLED_SIZE_1__SIZE__SHIFT 0x0
+#define VGT_STRMOUT_BUFFER_FILLED_SIZE_2__SIZE_MASK 0xffffffff
+#define VGT_STRMOUT_BUFFER_FILLED_SIZE_2__SIZE__SHIFT 0x0
+#define VGT_STRMOUT_BUFFER_FILLED_SIZE_3__SIZE_MASK 0xffffffff
+#define VGT_STRMOUT_BUFFER_FILLED_SIZE_3__SIZE__SHIFT 0x0
+#define VGT_STRMOUT_DRAW_OPAQUE_OFFSET__OFFSET_MASK 0xffffffff
+#define VGT_STRMOUT_DRAW_OPAQUE_OFFSET__OFFSET__SHIFT 0x0
+#define VGT_STRMOUT_DRAW_OPAQUE_BUFFER_FILLED_SIZE__SIZE_MASK 0xffffffff
+#define VGT_STRMOUT_DRAW_OPAQUE_BUFFER_FILLED_SIZE__SIZE__SHIFT 0x0
+#define VGT_STRMOUT_DRAW_OPAQUE_VERTEX_STRIDE__VERTEX_STRIDE_MASK 0x1ff
+#define VGT_STRMOUT_DRAW_OPAQUE_VERTEX_STRIDE__VERTEX_STRIDE__SHIFT 0x0
+#define VGT_GS_MAX_VERT_OUT__MAX_VERT_OUT_MASK 0x7ff
+#define VGT_GS_MAX_VERT_OUT__MAX_VERT_OUT__SHIFT 0x0
+#define VGT_SHADER_STAGES_EN__LS_EN_MASK 0x3
+#define VGT_SHADER_STAGES_EN__LS_EN__SHIFT 0x0
+#define VGT_SHADER_STAGES_EN__HS_EN_MASK 0x4
+#define VGT_SHADER_STAGES_EN__HS_EN__SHIFT 0x2
+#define VGT_SHADER_STAGES_EN__ES_EN_MASK 0x18
+#define VGT_SHADER_STAGES_EN__ES_EN__SHIFT 0x3
+#define VGT_SHADER_STAGES_EN__GS_EN_MASK 0x20
+#define VGT_SHADER_STAGES_EN__GS_EN__SHIFT 0x5
+#define VGT_SHADER_STAGES_EN__VS_EN_MASK 0xc0
+#define VGT_SHADER_STAGES_EN__VS_EN__SHIFT 0x6
+#define VGT_SHADER_STAGES_EN__DYNAMIC_HS_MASK 0x100
+#define VGT_SHADER_STAGES_EN__DYNAMIC_HS__SHIFT 0x8
+#define VGT_SHADER_STAGES_EN__DISPATCH_DRAW_EN_MASK 0x200
+#define VGT_SHADER_STAGES_EN__DISPATCH_DRAW_EN__SHIFT 0x9
+#define VGT_SHADER_STAGES_EN__DIS_DEALLOC_ACCUM_0_MASK 0x400
+#define VGT_SHADER_STAGES_EN__DIS_DEALLOC_ACCUM_0__SHIFT 0xa
+#define VGT_SHADER_STAGES_EN__DIS_DEALLOC_ACCUM_1_MASK 0x800
+#define VGT_SHADER_STAGES_EN__DIS_DEALLOC_ACCUM_1__SHIFT 0xb
+#define VGT_SHADER_STAGES_EN__VS_WAVE_ID_EN_MASK 0x1000
+#define VGT_SHADER_STAGES_EN__VS_WAVE_ID_EN__SHIFT 0xc
+#define VGT_DISPATCH_DRAW_INDEX__MATCH_INDEX_MASK 0xffffffff
+#define VGT_DISPATCH_DRAW_INDEX__MATCH_INDEX__SHIFT 0x0
+#define VGT_LS_HS_CONFIG__NUM_PATCHES_MASK 0xff
+#define VGT_LS_HS_CONFIG__NUM_PATCHES__SHIFT 0x0
+#define VGT_LS_HS_CONFIG__HS_NUM_INPUT_CP_MASK 0x3f00
+#define VGT_LS_HS_CONFIG__HS_NUM_INPUT_CP__SHIFT 0x8
+#define VGT_LS_HS_CONFIG__HS_NUM_OUTPUT_CP_MASK 0xfc000
+#define VGT_LS_HS_CONFIG__HS_NUM_OUTPUT_CP__SHIFT 0xe
+#define VGT_DMA_LS_HS_CONFIG__HS_NUM_INPUT_CP_MASK 0x3f00
+#define VGT_DMA_LS_HS_CONFIG__HS_NUM_INPUT_CP__SHIFT 0x8
+#define VGT_TF_PARAM__TYPE_MASK 0x3
+#define VGT_TF_PARAM__TYPE__SHIFT 0x0
+#define VGT_TF_PARAM__PARTITIONING_MASK 0x1c
+#define VGT_TF_PARAM__PARTITIONING__SHIFT 0x2
+#define VGT_TF_PARAM__TOPOLOGY_MASK 0xe0
+#define VGT_TF_PARAM__TOPOLOGY__SHIFT 0x5
+#define VGT_TF_PARAM__RESERVED_REDUC_AXIS_MASK 0x100
+#define VGT_TF_PARAM__RESERVED_REDUC_AXIS__SHIFT 0x8
+#define VGT_TF_PARAM__DEPRECATED_MASK 0x200
+#define VGT_TF_PARAM__DEPRECATED__SHIFT 0x9
+#define VGT_TF_PARAM__NUM_DS_WAVES_PER_SIMD_MASK 0x3c00
+#define VGT_TF_PARAM__NUM_DS_WAVES_PER_SIMD__SHIFT 0xa
+#define VGT_TF_PARAM__DISABLE_DONUTS_MASK 0x4000
+#define VGT_TF_PARAM__DISABLE_DONUTS__SHIFT 0xe
+#define VGT_TF_PARAM__RDREQ_POLICY_MASK 0x8000
+#define VGT_TF_PARAM__RDREQ_POLICY__SHIFT 0xf
+#define VGT_TF_PARAM__DISTRIBUTION_MODE_MASK 0x60000
+#define VGT_TF_PARAM__DISTRIBUTION_MODE__SHIFT 0x11
+#define VGT_TF_PARAM__MTYPE_MASK 0x180000
+#define VGT_TF_PARAM__MTYPE__SHIFT 0x13
+#define VGT_TESS_DISTRIBUTION__ACCUM_ISOLINE_MASK 0xff
+#define VGT_TESS_DISTRIBUTION__ACCUM_ISOLINE__SHIFT 0x0
+#define VGT_TESS_DISTRIBUTION__ACCUM_TRI_MASK 0xff00
+#define VGT_TESS_DISTRIBUTION__ACCUM_TRI__SHIFT 0x8
+#define VGT_TESS_DISTRIBUTION__ACCUM_QUAD_MASK 0xff0000
+#define VGT_TESS_DISTRIBUTION__ACCUM_QUAD__SHIFT 0x10
+#define VGT_TESS_DISTRIBUTION__DONUT_SPLIT_MASK 0xff000000
+#define VGT_TESS_DISTRIBUTION__DONUT_SPLIT__SHIFT 0x18
+#define VGT_TF_RING_SIZE__SIZE_MASK 0xffff
+#define VGT_TF_RING_SIZE__SIZE__SHIFT 0x0
+#define VGT_SYS_CONFIG__DUAL_CORE_EN_MASK 0x1
+#define VGT_SYS_CONFIG__DUAL_CORE_EN__SHIFT 0x0
+#define VGT_SYS_CONFIG__MAX_LS_HS_THDGRP_MASK 0x7e
+#define VGT_SYS_CONFIG__MAX_LS_HS_THDGRP__SHIFT 0x1
+#define VGT_SYS_CONFIG__ADC_EVENT_FILTER_DISABLE_MASK 0x80
+#define VGT_SYS_CONFIG__ADC_EVENT_FILTER_DISABLE__SHIFT 0x7
+#define VGT_HS_OFFCHIP_PARAM__OFFCHIP_BUFFERING_MASK 0x1ff
+#define VGT_HS_OFFCHIP_PARAM__OFFCHIP_BUFFERING__SHIFT 0x0
+#define VGT_HS_OFFCHIP_PARAM__OFFCHIP_GRANULARITY_MASK 0x600
+#define VGT_HS_OFFCHIP_PARAM__OFFCHIP_GRANULARITY__SHIFT 0x9
+#define VGT_TF_MEMORY_BASE__BASE_MASK 0xffffffff
+#define VGT_TF_MEMORY_BASE__BASE__SHIFT 0x0
+#define VGT_GS_INSTANCE_CNT__ENABLE_MASK 0x1
+#define VGT_GS_INSTANCE_CNT__ENABLE__SHIFT 0x0
+#define VGT_GS_INSTANCE_CNT__CNT_MASK 0x1fc
+#define VGT_GS_INSTANCE_CNT__CNT__SHIFT 0x2
+#define IA_MULTI_VGT_PARAM__PRIMGROUP_SIZE_MASK 0xffff
+#define IA_MULTI_VGT_PARAM__PRIMGROUP_SIZE__SHIFT 0x0
+#define IA_MULTI_VGT_PARAM__PARTIAL_VS_WAVE_ON_MASK 0x10000
+#define IA_MULTI_VGT_PARAM__PARTIAL_VS_WAVE_ON__SHIFT 0x10
+#define IA_MULTI_VGT_PARAM__SWITCH_ON_EOP_MASK 0x20000
+#define IA_MULTI_VGT_PARAM__SWITCH_ON_EOP__SHIFT 0x11
+#define IA_MULTI_VGT_PARAM__PARTIAL_ES_WAVE_ON_MASK 0x40000
+#define IA_MULTI_VGT_PARAM__PARTIAL_ES_WAVE_ON__SHIFT 0x12
+#define IA_MULTI_VGT_PARAM__SWITCH_ON_EOI_MASK 0x80000
+#define IA_MULTI_VGT_PARAM__SWITCH_ON_EOI__SHIFT 0x13
+#define IA_MULTI_VGT_PARAM__WD_SWITCH_ON_EOP_MASK 0x100000
+#define IA_MULTI_VGT_PARAM__WD_SWITCH_ON_EOP__SHIFT 0x14
+#define IA_MULTI_VGT_PARAM__MAX_PRIMGRP_IN_WAVE_MASK 0xf0000000
+#define IA_MULTI_VGT_PARAM__MAX_PRIMGRP_IN_WAVE__SHIFT 0x1c
+#define VGT_VS_MAX_WAVE_ID__MAX_WAVE_ID_MASK 0xfff
+#define VGT_VS_MAX_WAVE_ID__MAX_WAVE_ID__SHIFT 0x0
+#define VGT_ESGS_RING_SIZE__MEM_SIZE_MASK 0xffffffff
+#define VGT_ESGS_RING_SIZE__MEM_SIZE__SHIFT 0x0
+#define VGT_GSVS_RING_SIZE__MEM_SIZE_MASK 0xffffffff
+#define VGT_GSVS_RING_SIZE__MEM_SIZE__SHIFT 0x0
+#define VGT_GSVS_RING_OFFSET_1__OFFSET_MASK 0x7fff
+#define VGT_GSVS_RING_OFFSET_1__OFFSET__SHIFT 0x0
+#define VGT_GSVS_RING_OFFSET_2__OFFSET_MASK 0x7fff
+#define VGT_GSVS_RING_OFFSET_2__OFFSET__SHIFT 0x0
+#define VGT_GSVS_RING_OFFSET_3__OFFSET_MASK 0x7fff
+#define VGT_GSVS_RING_OFFSET_3__OFFSET__SHIFT 0x0
+#define VGT_ESGS_RING_ITEMSIZE__ITEMSIZE_MASK 0x7fff
+#define VGT_ESGS_RING_ITEMSIZE__ITEMSIZE__SHIFT 0x0
+#define VGT_GSVS_RING_ITEMSIZE__ITEMSIZE_MASK 0x7fff
+#define VGT_GSVS_RING_ITEMSIZE__ITEMSIZE__SHIFT 0x0
+#define VGT_GS_VERT_ITEMSIZE__ITEMSIZE_MASK 0x7fff
+#define VGT_GS_VERT_ITEMSIZE__ITEMSIZE__SHIFT 0x0
+#define VGT_GS_VERT_ITEMSIZE_1__ITEMSIZE_MASK 0x7fff
+#define VGT_GS_VERT_ITEMSIZE_1__ITEMSIZE__SHIFT 0x0
+#define VGT_GS_VERT_ITEMSIZE_2__ITEMSIZE_MASK 0x7fff
+#define VGT_GS_VERT_ITEMSIZE_2__ITEMSIZE__SHIFT 0x0
+#define VGT_GS_VERT_ITEMSIZE_3__ITEMSIZE_MASK 0x7fff
+#define VGT_GS_VERT_ITEMSIZE_3__ITEMSIZE__SHIFT 0x0
+#define WD_CNTL_STATUS__WD_BUSY_MASK 0x1
+#define WD_CNTL_STATUS__WD_BUSY__SHIFT 0x0
+#define WD_CNTL_STATUS__WD_SPL_DMA_BUSY_MASK 0x2
+#define WD_CNTL_STATUS__WD_SPL_DMA_BUSY__SHIFT 0x1
+#define WD_CNTL_STATUS__WD_SPL_DI_BUSY_MASK 0x4
+#define WD_CNTL_STATUS__WD_SPL_DI_BUSY__SHIFT 0x2
+#define WD_CNTL_STATUS__WD_ADC_BUSY_MASK 0x8
+#define WD_CNTL_STATUS__WD_ADC_BUSY__SHIFT 0x3
+#define WD_ENHANCE__MISC_MASK 0xffffffff
+#define WD_ENHANCE__MISC__SHIFT 0x0
+#define GFX_PIPE_CONTROL__HYSTERESIS_CNT_MASK 0x1fff
+#define GFX_PIPE_CONTROL__HYSTERESIS_CNT__SHIFT 0x0
+#define GFX_PIPE_CONTROL__RESERVED_MASK 0xe000
+#define GFX_PIPE_CONTROL__RESERVED__SHIFT 0xd
+#define GFX_PIPE_CONTROL__CONTEXT_SUSPEND_EN_MASK 0x10000
+#define GFX_PIPE_CONTROL__CONTEXT_SUSPEND_EN__SHIFT 0x10
+#define CGTT_VGT_CLK_CTRL__ON_DELAY_MASK 0xf
+#define CGTT_VGT_CLK_CTRL__ON_DELAY__SHIFT 0x0
+#define CGTT_VGT_CLK_CTRL__OFF_HYSTERESIS_MASK 0xff0
+#define CGTT_VGT_CLK_CTRL__OFF_HYSTERESIS__SHIFT 0x4
+#define CGTT_VGT_CLK_CTRL__SOFT_OVERRIDE7_MASK 0x1000000
+#define CGTT_VGT_CLK_CTRL__SOFT_OVERRIDE7__SHIFT 0x18
+#define CGTT_VGT_CLK_CTRL__PERF_ENABLE_MASK 0x2000000
+#define CGTT_VGT_CLK_CTRL__PERF_ENABLE__SHIFT 0x19
+#define CGTT_VGT_CLK_CTRL__DBG_ENABLE_MASK 0x4000000
+#define CGTT_VGT_CLK_CTRL__DBG_ENABLE__SHIFT 0x1a
+#define CGTT_VGT_CLK_CTRL__SOFT_OVERRIDE4_MASK 0x8000000
+#define CGTT_VGT_CLK_CTRL__SOFT_OVERRIDE4__SHIFT 0x1b
+#define CGTT_VGT_CLK_CTRL__TESS_OVERRIDE_MASK 0x10000000
+#define CGTT_VGT_CLK_CTRL__TESS_OVERRIDE__SHIFT 0x1c
+#define CGTT_VGT_CLK_CTRL__GS_OVERRIDE_MASK 0x20000000
+#define CGTT_VGT_CLK_CTRL__GS_OVERRIDE__SHIFT 0x1d
+#define CGTT_VGT_CLK_CTRL__CORE_OVERRIDE_MASK 0x40000000
+#define CGTT_VGT_CLK_CTRL__CORE_OVERRIDE__SHIFT 0x1e
+#define CGTT_VGT_CLK_CTRL__REG_OVERRIDE_MASK 0x80000000
+#define CGTT_VGT_CLK_CTRL__REG_OVERRIDE__SHIFT 0x1f
+#define CGTT_IA_CLK_CTRL__ON_DELAY_MASK 0xf
+#define CGTT_IA_CLK_CTRL__ON_DELAY__SHIFT 0x0
+#define CGTT_IA_CLK_CTRL__OFF_HYSTERESIS_MASK 0xff0
+#define CGTT_IA_CLK_CTRL__OFF_HYSTERESIS__SHIFT 0x4
+#define CGTT_IA_CLK_CTRL__SOFT_OVERRIDE7_MASK 0x1000000
+#define CGTT_IA_CLK_CTRL__SOFT_OVERRIDE7__SHIFT 0x18
+#define CGTT_IA_CLK_CTRL__PERF_ENABLE_MASK 0x2000000
+#define CGTT_IA_CLK_CTRL__PERF_ENABLE__SHIFT 0x19
+#define CGTT_IA_CLK_CTRL__DBG_ENABLE_MASK 0x4000000
+#define CGTT_IA_CLK_CTRL__DBG_ENABLE__SHIFT 0x1a
+#define CGTT_IA_CLK_CTRL__SOFT_OVERRIDE4_MASK 0x8000000
+#define CGTT_IA_CLK_CTRL__SOFT_OVERRIDE4__SHIFT 0x1b
+#define CGTT_IA_CLK_CTRL__SOFT_OVERRIDE3_MASK 0x10000000
+#define CGTT_IA_CLK_CTRL__SOFT_OVERRIDE3__SHIFT 0x1c
+#define CGTT_IA_CLK_CTRL__SOFT_OVERRIDE2_MASK 0x20000000
+#define CGTT_IA_CLK_CTRL__SOFT_OVERRIDE2__SHIFT 0x1d
+#define CGTT_IA_CLK_CTRL__CORE_OVERRIDE_MASK 0x40000000
+#define CGTT_IA_CLK_CTRL__CORE_OVERRIDE__SHIFT 0x1e
+#define CGTT_IA_CLK_CTRL__REG_OVERRIDE_MASK 0x80000000
+#define CGTT_IA_CLK_CTRL__REG_OVERRIDE__SHIFT 0x1f
+#define CGTT_WD_CLK_CTRL__ON_DELAY_MASK 0xf
+#define CGTT_WD_CLK_CTRL__ON_DELAY__SHIFT 0x0
+#define CGTT_WD_CLK_CTRL__OFF_HYSTERESIS_MASK 0xff0
+#define CGTT_WD_CLK_CTRL__OFF_HYSTERESIS__SHIFT 0x4
+#define CGTT_WD_CLK_CTRL__SOFT_OVERRIDE7_MASK 0x1000000
+#define CGTT_WD_CLK_CTRL__SOFT_OVERRIDE7__SHIFT 0x18
+#define CGTT_WD_CLK_CTRL__PERF_ENABLE_MASK 0x2000000
+#define CGTT_WD_CLK_CTRL__PERF_ENABLE__SHIFT 0x19
+#define CGTT_WD_CLK_CTRL__DBG_ENABLE_MASK 0x4000000
+#define CGTT_WD_CLK_CTRL__DBG_ENABLE__SHIFT 0x1a
+#define CGTT_WD_CLK_CTRL__SOFT_OVERRIDE4_MASK 0x8000000
+#define CGTT_WD_CLK_CTRL__SOFT_OVERRIDE4__SHIFT 0x1b
+#define CGTT_WD_CLK_CTRL__TESS_OVERRIDE_MASK 0x10000000
+#define CGTT_WD_CLK_CTRL__TESS_OVERRIDE__SHIFT 0x1c
+#define CGTT_WD_CLK_CTRL__CORE_OVERRIDE_MASK 0x20000000
+#define CGTT_WD_CLK_CTRL__CORE_OVERRIDE__SHIFT 0x1d
+#define CGTT_WD_CLK_CTRL__RBIU_INPUT_OVERRIDE_MASK 0x40000000
+#define CGTT_WD_CLK_CTRL__RBIU_INPUT_OVERRIDE__SHIFT 0x1e
+#define CGTT_WD_CLK_CTRL__REG_OVERRIDE_MASK 0x80000000
+#define CGTT_WD_CLK_CTRL__REG_OVERRIDE__SHIFT 0x1f
+#define VGT_DEBUG_CNTL__VGT_DEBUG_INDX_MASK 0x3f
+#define VGT_DEBUG_CNTL__VGT_DEBUG_INDX__SHIFT 0x0
+#define VGT_DEBUG_CNTL__VGT_DEBUG_SEL_BUS_B_MASK 0x40
+#define VGT_DEBUG_CNTL__VGT_DEBUG_SEL_BUS_B__SHIFT 0x6
+#define VGT_DEBUG_DATA__DATA_MASK 0xffffffff
+#define VGT_DEBUG_DATA__DATA__SHIFT 0x0
+#define IA_DEBUG_CNTL__IA_DEBUG_INDX_MASK 0x3f
+#define IA_DEBUG_CNTL__IA_DEBUG_INDX__SHIFT 0x0
+#define IA_DEBUG_CNTL__IA_DEBUG_SEL_BUS_B_MASK 0x40
+#define IA_DEBUG_CNTL__IA_DEBUG_SEL_BUS_B__SHIFT 0x6
+#define IA_DEBUG_DATA__DATA_MASK 0xffffffff
+#define IA_DEBUG_DATA__DATA__SHIFT 0x0
+#define VGT_CNTL_STATUS__VGT_BUSY_MASK 0x1
+#define VGT_CNTL_STATUS__VGT_BUSY__SHIFT 0x0
+#define VGT_CNTL_STATUS__VGT_OUT_INDX_BUSY_MASK 0x2
+#define VGT_CNTL_STATUS__VGT_OUT_INDX_BUSY__SHIFT 0x1
+#define VGT_CNTL_STATUS__VGT_OUT_BUSY_MASK 0x4
+#define VGT_CNTL_STATUS__VGT_OUT_BUSY__SHIFT 0x2
+#define VGT_CNTL_STATUS__VGT_PT_BUSY_MASK 0x8
+#define VGT_CNTL_STATUS__VGT_PT_BUSY__SHIFT 0x3
+#define VGT_CNTL_STATUS__VGT_TE_BUSY_MASK 0x10
+#define VGT_CNTL_STATUS__VGT_TE_BUSY__SHIFT 0x4
+#define VGT_CNTL_STATUS__VGT_VR_BUSY_MASK 0x20
+#define VGT_CNTL_STATUS__VGT_VR_BUSY__SHIFT 0x5
+#define VGT_CNTL_STATUS__VGT_PI_BUSY_MASK 0x40
+#define VGT_CNTL_STATUS__VGT_PI_BUSY__SHIFT 0x6
+#define VGT_CNTL_STATUS__VGT_GS_BUSY_MASK 0x80
+#define VGT_CNTL_STATUS__VGT_GS_BUSY__SHIFT 0x7
+#define VGT_CNTL_STATUS__VGT_HS_BUSY_MASK 0x100
+#define VGT_CNTL_STATUS__VGT_HS_BUSY__SHIFT 0x8
+#define VGT_CNTL_STATUS__VGT_TE11_BUSY_MASK 0x200
+#define VGT_CNTL_STATUS__VGT_TE11_BUSY__SHIFT 0x9
+#define WD_DEBUG_CNTL__WD_DEBUG_INDX_MASK 0x3f
+#define WD_DEBUG_CNTL__WD_DEBUG_INDX__SHIFT 0x0
+#define WD_DEBUG_CNTL__WD_DEBUG_SEL_BUS_B_MASK 0x40
+#define WD_DEBUG_CNTL__WD_DEBUG_SEL_BUS_B__SHIFT 0x6
+#define WD_DEBUG_DATA__DATA_MASK 0xffffffff
+#define WD_DEBUG_DATA__DATA__SHIFT 0x0
+#define WD_QOS__DRAW_STALL_MASK 0x1
+#define WD_QOS__DRAW_STALL__SHIFT 0x0
+#define CC_GC_PRIM_CONFIG__INACTIVE_IA_MASK 0x30000
+#define CC_GC_PRIM_CONFIG__INACTIVE_IA__SHIFT 0x10
+#define CC_GC_PRIM_CONFIG__INACTIVE_VGT_PA_MASK 0xf000000
+#define CC_GC_PRIM_CONFIG__INACTIVE_VGT_PA__SHIFT 0x18
+#define GC_USER_PRIM_CONFIG__INACTIVE_IA_MASK 0x30000
+#define GC_USER_PRIM_CONFIG__INACTIVE_IA__SHIFT 0x10
+#define GC_USER_PRIM_CONFIG__INACTIVE_VGT_PA_MASK 0xf000000
+#define GC_USER_PRIM_CONFIG__INACTIVE_VGT_PA__SHIFT 0x18
+#define WD_DEBUG_REG0__wd_busy_extended_MASK 0x1
+#define WD_DEBUG_REG0__wd_busy_extended__SHIFT 0x0
+#define WD_DEBUG_REG0__wd_nodma_busy_extended_MASK 0x2
+#define WD_DEBUG_REG0__wd_nodma_busy_extended__SHIFT 0x1
+#define WD_DEBUG_REG0__wd_busy_MASK 0x4
+#define WD_DEBUG_REG0__wd_busy__SHIFT 0x2
+#define WD_DEBUG_REG0__wd_nodma_busy_MASK 0x8
+#define WD_DEBUG_REG0__wd_nodma_busy__SHIFT 0x3
+#define WD_DEBUG_REG0__rbiu_busy_MASK 0x10
+#define WD_DEBUG_REG0__rbiu_busy__SHIFT 0x4
+#define WD_DEBUG_REG0__spl_dma_busy_MASK 0x20
+#define WD_DEBUG_REG0__spl_dma_busy__SHIFT 0x5
+#define WD_DEBUG_REG0__spl_di_busy_MASK 0x40
+#define WD_DEBUG_REG0__spl_di_busy__SHIFT 0x6
+#define WD_DEBUG_REG0__vgt0_active_q_MASK 0x80
+#define WD_DEBUG_REG0__vgt0_active_q__SHIFT 0x7
+#define WD_DEBUG_REG0__vgt1_active_q_MASK 0x100
+#define WD_DEBUG_REG0__vgt1_active_q__SHIFT 0x8
+#define WD_DEBUG_REG0__spl_dma_p1_busy_MASK 0x200
+#define WD_DEBUG_REG0__spl_dma_p1_busy__SHIFT 0x9
+#define WD_DEBUG_REG0__rbiu_dr_p1_fifo_busy_MASK 0x400
+#define WD_DEBUG_REG0__rbiu_dr_p1_fifo_busy__SHIFT 0xa
+#define WD_DEBUG_REG0__rbiu_di_p1_fifo_busy_MASK 0x800
+#define WD_DEBUG_REG0__rbiu_di_p1_fifo_busy__SHIFT 0xb
+#define WD_DEBUG_REG0__SPARE2_MASK 0x1000
+#define WD_DEBUG_REG0__SPARE2__SHIFT 0xc
+#define WD_DEBUG_REG0__rbiu_dr_fifo_busy_MASK 0x2000
+#define WD_DEBUG_REG0__rbiu_dr_fifo_busy__SHIFT 0xd
+#define WD_DEBUG_REG0__rbiu_spl_dr_valid_MASK 0x4000
+#define WD_DEBUG_REG0__rbiu_spl_dr_valid__SHIFT 0xe
+#define WD_DEBUG_REG0__spl_rbiu_dr_read_MASK 0x8000
+#define WD_DEBUG_REG0__spl_rbiu_dr_read__SHIFT 0xf
+#define WD_DEBUG_REG0__SPARE3_MASK 0x10000
+#define WD_DEBUG_REG0__SPARE3__SHIFT 0x10
+#define WD_DEBUG_REG0__rbiu_di_fifo_busy_MASK 0x20000
+#define WD_DEBUG_REG0__rbiu_di_fifo_busy__SHIFT 0x11
+#define WD_DEBUG_REG0__rbiu_spl_di_valid_MASK 0x40000
+#define WD_DEBUG_REG0__rbiu_spl_di_valid__SHIFT 0x12
+#define WD_DEBUG_REG0__spl_rbiu_di_read_MASK 0x80000
+#define WD_DEBUG_REG0__spl_rbiu_di_read__SHIFT 0x13
+#define WD_DEBUG_REG0__se0_synced_q_MASK 0x100000
+#define WD_DEBUG_REG0__se0_synced_q__SHIFT 0x14
+#define WD_DEBUG_REG0__se1_synced_q_MASK 0x200000
+#define WD_DEBUG_REG0__se1_synced_q__SHIFT 0x15
+#define WD_DEBUG_REG0__se2_synced_q_MASK 0x400000
+#define WD_DEBUG_REG0__se2_synced_q__SHIFT 0x16
+#define WD_DEBUG_REG0__se3_synced_q_MASK 0x800000
+#define WD_DEBUG_REG0__se3_synced_q__SHIFT 0x17
+#define WD_DEBUG_REG0__reg_clk_busy_MASK 0x1000000
+#define WD_DEBUG_REG0__reg_clk_busy__SHIFT 0x18
+#define WD_DEBUG_REG0__input_clk_busy_MASK 0x2000000
+#define WD_DEBUG_REG0__input_clk_busy__SHIFT 0x19
+#define WD_DEBUG_REG0__core_clk_busy_MASK 0x4000000
+#define WD_DEBUG_REG0__core_clk_busy__SHIFT 0x1a
+#define WD_DEBUG_REG0__vgt2_active_q_MASK 0x8000000
+#define WD_DEBUG_REG0__vgt2_active_q__SHIFT 0x1b
+#define WD_DEBUG_REG0__sclk_reg_vld_MASK 0x10000000
+#define WD_DEBUG_REG0__sclk_reg_vld__SHIFT 0x1c
+#define WD_DEBUG_REG0__sclk_input_vld_MASK 0x20000000
+#define WD_DEBUG_REG0__sclk_input_vld__SHIFT 0x1d
+#define WD_DEBUG_REG0__sclk_core_vld_MASK 0x40000000
+#define WD_DEBUG_REG0__sclk_core_vld__SHIFT 0x1e
+#define WD_DEBUG_REG0__vgt3_active_q_MASK 0x80000000
+#define WD_DEBUG_REG0__vgt3_active_q__SHIFT 0x1f
+#define WD_DEBUG_REG1__grbm_fifo_empty_MASK 0x1
+#define WD_DEBUG_REG1__grbm_fifo_empty__SHIFT 0x0
+#define WD_DEBUG_REG1__grbm_fifo_full_MASK 0x2
+#define WD_DEBUG_REG1__grbm_fifo_full__SHIFT 0x1
+#define WD_DEBUG_REG1__grbm_fifo_we_MASK 0x4
+#define WD_DEBUG_REG1__grbm_fifo_we__SHIFT 0x2
+#define WD_DEBUG_REG1__grbm_fifo_re_MASK 0x8
+#define WD_DEBUG_REG1__grbm_fifo_re__SHIFT 0x3
+#define WD_DEBUG_REG1__draw_initiator_valid_q_MASK 0x10
+#define WD_DEBUG_REG1__draw_initiator_valid_q__SHIFT 0x4
+#define WD_DEBUG_REG1__event_initiator_valid_q_MASK 0x20
+#define WD_DEBUG_REG1__event_initiator_valid_q__SHIFT 0x5
+#define WD_DEBUG_REG1__event_addr_valid_q_MASK 0x40
+#define WD_DEBUG_REG1__event_addr_valid_q__SHIFT 0x6
+#define WD_DEBUG_REG1__dma_request_valid_q_MASK 0x80
+#define WD_DEBUG_REG1__dma_request_valid_q__SHIFT 0x7
+#define WD_DEBUG_REG1__SPARE0_MASK 0x100
+#define WD_DEBUG_REG1__SPARE0__SHIFT 0x8
+#define WD_DEBUG_REG1__min_indx_valid_q_MASK 0x200
+#define WD_DEBUG_REG1__min_indx_valid_q__SHIFT 0x9
+#define WD_DEBUG_REG1__max_indx_valid_q_MASK 0x400
+#define WD_DEBUG_REG1__max_indx_valid_q__SHIFT 0xa
+#define WD_DEBUG_REG1__indx_offset_valid_q_MASK 0x800
+#define WD_DEBUG_REG1__indx_offset_valid_q__SHIFT 0xb
+#define WD_DEBUG_REG1__grbm_fifo_rdata_reg_id_MASK 0x1f000
+#define WD_DEBUG_REG1__grbm_fifo_rdata_reg_id__SHIFT 0xc
+#define WD_DEBUG_REG1__grbm_fifo_rdata_state_MASK 0xe0000
+#define WD_DEBUG_REG1__grbm_fifo_rdata_state__SHIFT 0x11
+#define WD_DEBUG_REG1__free_cnt_q_MASK 0x3f00000
+#define WD_DEBUG_REG1__free_cnt_q__SHIFT 0x14
+#define WD_DEBUG_REG1__rbiu_di_fifo_we_MASK 0x4000000
+#define WD_DEBUG_REG1__rbiu_di_fifo_we__SHIFT 0x1a
+#define WD_DEBUG_REG1__rbiu_dr_fifo_we_MASK 0x8000000
+#define WD_DEBUG_REG1__rbiu_dr_fifo_we__SHIFT 0x1b
+#define WD_DEBUG_REG1__rbiu_di_fifo_empty_MASK 0x10000000
+#define WD_DEBUG_REG1__rbiu_di_fifo_empty__SHIFT 0x1c
+#define WD_DEBUG_REG1__rbiu_di_fifo_full_MASK 0x20000000
+#define WD_DEBUG_REG1__rbiu_di_fifo_full__SHIFT 0x1d
+#define WD_DEBUG_REG1__rbiu_dr_fifo_empty_MASK 0x40000000
+#define WD_DEBUG_REG1__rbiu_dr_fifo_empty__SHIFT 0x1e
+#define WD_DEBUG_REG1__rbiu_dr_fifo_full_MASK 0x80000000
+#define WD_DEBUG_REG1__rbiu_dr_fifo_full__SHIFT 0x1f
+#define WD_DEBUG_REG2__p1_grbm_fifo_empty_MASK 0x1
+#define WD_DEBUG_REG2__p1_grbm_fifo_empty__SHIFT 0x0
+#define WD_DEBUG_REG2__p1_grbm_fifo_full_MASK 0x2
+#define WD_DEBUG_REG2__p1_grbm_fifo_full__SHIFT 0x1
+#define WD_DEBUG_REG2__p1_grbm_fifo_we_MASK 0x4
+#define WD_DEBUG_REG2__p1_grbm_fifo_we__SHIFT 0x2
+#define WD_DEBUG_REG2__p1_grbm_fifo_re_MASK 0x8
+#define WD_DEBUG_REG2__p1_grbm_fifo_re__SHIFT 0x3
+#define WD_DEBUG_REG2__p1_draw_initiator_valid_q_MASK 0x10
+#define WD_DEBUG_REG2__p1_draw_initiator_valid_q__SHIFT 0x4
+#define WD_DEBUG_REG2__p1_event_initiator_valid_q_MASK 0x20
+#define WD_DEBUG_REG2__p1_event_initiator_valid_q__SHIFT 0x5
+#define WD_DEBUG_REG2__p1_event_addr_valid_q_MASK 0x40
+#define WD_DEBUG_REG2__p1_event_addr_valid_q__SHIFT 0x6
+#define WD_DEBUG_REG2__p1_dma_request_valid_q_MASK 0x80
+#define WD_DEBUG_REG2__p1_dma_request_valid_q__SHIFT 0x7
+#define WD_DEBUG_REG2__SPARE0_MASK 0x100
+#define WD_DEBUG_REG2__SPARE0__SHIFT 0x8
+#define WD_DEBUG_REG2__p1_min_indx_valid_q_MASK 0x200
+#define WD_DEBUG_REG2__p1_min_indx_valid_q__SHIFT 0x9
+#define WD_DEBUG_REG2__p1_max_indx_valid_q_MASK 0x400
+#define WD_DEBUG_REG2__p1_max_indx_valid_q__SHIFT 0xa
+#define WD_DEBUG_REG2__p1_indx_offset_valid_q_MASK 0x800
+#define WD_DEBUG_REG2__p1_indx_offset_valid_q__SHIFT 0xb
+#define WD_DEBUG_REG2__p1_grbm_fifo_rdata_reg_id_MASK 0x1f000
+#define WD_DEBUG_REG2__p1_grbm_fifo_rdata_reg_id__SHIFT 0xc
+#define WD_DEBUG_REG2__p1_grbm_fifo_rdata_state_MASK 0xe0000
+#define WD_DEBUG_REG2__p1_grbm_fifo_rdata_state__SHIFT 0x11
+#define WD_DEBUG_REG2__p1_free_cnt_q_MASK 0x3f00000
+#define WD_DEBUG_REG2__p1_free_cnt_q__SHIFT 0x14
+#define WD_DEBUG_REG2__p1_rbiu_di_fifo_we_MASK 0x4000000
+#define WD_DEBUG_REG2__p1_rbiu_di_fifo_we__SHIFT 0x1a
+#define WD_DEBUG_REG2__p1_rbiu_dr_fifo_we_MASK 0x8000000
+#define WD_DEBUG_REG2__p1_rbiu_dr_fifo_we__SHIFT 0x1b
+#define WD_DEBUG_REG2__p1_rbiu_di_fifo_empty_MASK 0x10000000
+#define WD_DEBUG_REG2__p1_rbiu_di_fifo_empty__SHIFT 0x1c
+#define WD_DEBUG_REG2__p1_rbiu_di_fifo_full_MASK 0x20000000
+#define WD_DEBUG_REG2__p1_rbiu_di_fifo_full__SHIFT 0x1d
+#define WD_DEBUG_REG2__p1_rbiu_dr_fifo_empty_MASK 0x40000000
+#define WD_DEBUG_REG2__p1_rbiu_dr_fifo_empty__SHIFT 0x1e
+#define WD_DEBUG_REG2__p1_rbiu_dr_fifo_full_MASK 0x80000000
+#define WD_DEBUG_REG2__p1_rbiu_dr_fifo_full__SHIFT 0x1f
+#define WD_DEBUG_REG3__rbiu_spl_dr_valid_MASK 0x1
+#define WD_DEBUG_REG3__rbiu_spl_dr_valid__SHIFT 0x0
+#define WD_DEBUG_REG3__SPARE0_MASK 0x2
+#define WD_DEBUG_REG3__SPARE0__SHIFT 0x1
+#define WD_DEBUG_REG3__pipe0_dr_MASK 0x4
+#define WD_DEBUG_REG3__pipe0_dr__SHIFT 0x2
+#define WD_DEBUG_REG3__pipe0_rtr_MASK 0x8
+#define WD_DEBUG_REG3__pipe0_rtr__SHIFT 0x3
+#define WD_DEBUG_REG3__pipe1_dr_MASK 0x10
+#define WD_DEBUG_REG3__pipe1_dr__SHIFT 0x4
+#define WD_DEBUG_REG3__pipe1_rtr_MASK 0x20
+#define WD_DEBUG_REG3__pipe1_rtr__SHIFT 0x5
+#define WD_DEBUG_REG3__wd_subdma_fifo_empty_MASK 0x40
+#define WD_DEBUG_REG3__wd_subdma_fifo_empty__SHIFT 0x6
+#define WD_DEBUG_REG3__wd_subdma_fifo_full_MASK 0x80
+#define WD_DEBUG_REG3__wd_subdma_fifo_full__SHIFT 0x7
+#define WD_DEBUG_REG3__dma_buf_type_p0_q_MASK 0x300
+#define WD_DEBUG_REG3__dma_buf_type_p0_q__SHIFT 0x8
+#define WD_DEBUG_REG3__dma_zero_indices_p0_q_MASK 0x400
+#define WD_DEBUG_REG3__dma_zero_indices_p0_q__SHIFT 0xa
+#define WD_DEBUG_REG3__dma_req_path_p3_q_MASK 0x800
+#define WD_DEBUG_REG3__dma_req_path_p3_q__SHIFT 0xb
+#define WD_DEBUG_REG3__dma_not_eop_p1_q_MASK 0x1000
+#define WD_DEBUG_REG3__dma_not_eop_p1_q__SHIFT 0xc
+#define WD_DEBUG_REG3__out_of_range_p4_MASK 0x2000
+#define WD_DEBUG_REG3__out_of_range_p4__SHIFT 0xd
+#define WD_DEBUG_REG3__last_sub_dma_p3_q_MASK 0x4000
+#define WD_DEBUG_REG3__last_sub_dma_p3_q__SHIFT 0xe
+#define WD_DEBUG_REG3__last_rdreq_of_sub_dma_p4_MASK 0x8000
+#define WD_DEBUG_REG3__last_rdreq_of_sub_dma_p4__SHIFT 0xf
+#define WD_DEBUG_REG3__WD_IA_dma_send_d_MASK 0x10000
+#define WD_DEBUG_REG3__WD_IA_dma_send_d__SHIFT 0x10
+#define WD_DEBUG_REG3__WD_IA_dma_rtr_MASK 0x20000
+#define WD_DEBUG_REG3__WD_IA_dma_rtr__SHIFT 0x11
+#define WD_DEBUG_REG3__WD_IA1_dma_send_d_MASK 0x40000
+#define WD_DEBUG_REG3__WD_IA1_dma_send_d__SHIFT 0x12
+#define WD_DEBUG_REG3__WD_IA1_dma_rtr_MASK 0x80000
+#define WD_DEBUG_REG3__WD_IA1_dma_rtr__SHIFT 0x13
+#define WD_DEBUG_REG3__last_inst_of_dma_p2_MASK 0x100000
+#define WD_DEBUG_REG3__last_inst_of_dma_p2__SHIFT 0x14
+#define WD_DEBUG_REG3__last_sd_of_inst_p2_MASK 0x200000
+#define WD_DEBUG_REG3__last_sd_of_inst_p2__SHIFT 0x15
+#define WD_DEBUG_REG3__last_sd_of_dma_p2_MASK 0x400000
+#define WD_DEBUG_REG3__last_sd_of_dma_p2__SHIFT 0x16
+#define WD_DEBUG_REG3__SPARE1_MASK 0x800000
+#define WD_DEBUG_REG3__SPARE1__SHIFT 0x17
+#define WD_DEBUG_REG3__WD_IA_dma_busy_MASK 0x1000000
+#define WD_DEBUG_REG3__WD_IA_dma_busy__SHIFT 0x18
+#define WD_DEBUG_REG3__WD_IA1_dma_busy_MASK 0x2000000
+#define WD_DEBUG_REG3__WD_IA1_dma_busy__SHIFT 0x19
+#define WD_DEBUG_REG3__send_to_ia1_p3_q_MASK 0x4000000
+#define WD_DEBUG_REG3__send_to_ia1_p3_q__SHIFT 0x1a
+#define WD_DEBUG_REG3__dma_wd_switch_on_eop_p3_q_MASK 0x8000000
+#define WD_DEBUG_REG3__dma_wd_switch_on_eop_p3_q__SHIFT 0x1b
+#define WD_DEBUG_REG3__pipe3_dr_MASK 0x10000000
+#define WD_DEBUG_REG3__pipe3_dr__SHIFT 0x1c
+#define WD_DEBUG_REG3__pipe3_rtr_MASK 0x20000000
+#define WD_DEBUG_REG3__pipe3_rtr__SHIFT 0x1d
+#define WD_DEBUG_REG3__wd_dma2draw_fifo_empty_MASK 0x40000000
+#define WD_DEBUG_REG3__wd_dma2draw_fifo_empty__SHIFT 0x1e
+#define WD_DEBUG_REG3__wd_dma2draw_fifo_full_MASK 0x80000000
+#define WD_DEBUG_REG3__wd_dma2draw_fifo_full__SHIFT 0x1f
+#define WD_DEBUG_REG4__rbiu_spl_di_valid_MASK 0x1
+#define WD_DEBUG_REG4__rbiu_spl_di_valid__SHIFT 0x0
+#define WD_DEBUG_REG4__spl_rbiu_di_read_MASK 0x2
+#define WD_DEBUG_REG4__spl_rbiu_di_read__SHIFT 0x1
+#define WD_DEBUG_REG4__rbiu_spl_p1_di_valid_MASK 0x4
+#define WD_DEBUG_REG4__rbiu_spl_p1_di_valid__SHIFT 0x2
+#define WD_DEBUG_REG4__spl_rbiu_p1_di_read_MASK 0x8
+#define WD_DEBUG_REG4__spl_rbiu_p1_di_read__SHIFT 0x3
+#define WD_DEBUG_REG4__pipe0_dr_MASK 0x10
+#define WD_DEBUG_REG4__pipe0_dr__SHIFT 0x4
+#define WD_DEBUG_REG4__pipe0_rtr_MASK 0x20
+#define WD_DEBUG_REG4__pipe0_rtr__SHIFT 0x5
+#define WD_DEBUG_REG4__pipe1_dr_MASK 0x40
+#define WD_DEBUG_REG4__pipe1_dr__SHIFT 0x6
+#define WD_DEBUG_REG4__pipe1_rtr_MASK 0x80
+#define WD_DEBUG_REG4__pipe1_rtr__SHIFT 0x7
+#define WD_DEBUG_REG4__pipe2_dr_MASK 0x100
+#define WD_DEBUG_REG4__pipe2_dr__SHIFT 0x8
+#define WD_DEBUG_REG4__pipe2_rtr_MASK 0x200
+#define WD_DEBUG_REG4__pipe2_rtr__SHIFT 0x9
+#define WD_DEBUG_REG4__pipe3_ld_MASK 0x400
+#define WD_DEBUG_REG4__pipe3_ld__SHIFT 0xa
+#define WD_DEBUG_REG4__pipe3_rtr_MASK 0x800
+#define WD_DEBUG_REG4__pipe3_rtr__SHIFT 0xb
+#define WD_DEBUG_REG4__WD_IA_draw_send_d_MASK 0x1000
+#define WD_DEBUG_REG4__WD_IA_draw_send_d__SHIFT 0xc
+#define WD_DEBUG_REG4__WD_IA_draw_rtr_MASK 0x2000
+#define WD_DEBUG_REG4__WD_IA_draw_rtr__SHIFT 0xd
+#define WD_DEBUG_REG4__di_type_p0_MASK 0xc000
+#define WD_DEBUG_REG4__di_type_p0__SHIFT 0xe
+#define WD_DEBUG_REG4__di_state_sel_p1_q_MASK 0x70000
+#define WD_DEBUG_REG4__di_state_sel_p1_q__SHIFT 0x10
+#define WD_DEBUG_REG4__di_wd_switch_on_eop_p1_q_MASK 0x80000
+#define WD_DEBUG_REG4__di_wd_switch_on_eop_p1_q__SHIFT 0x13
+#define WD_DEBUG_REG4__rbiu_spl_pipe0_lockout_MASK 0x100000
+#define WD_DEBUG_REG4__rbiu_spl_pipe0_lockout__SHIFT 0x14
+#define WD_DEBUG_REG4__last_inst_of_di_p2_MASK 0x200000
+#define WD_DEBUG_REG4__last_inst_of_di_p2__SHIFT 0x15
+#define WD_DEBUG_REG4__last_sd_of_inst_p2_MASK 0x400000
+#define WD_DEBUG_REG4__last_sd_of_inst_p2__SHIFT 0x16
+#define WD_DEBUG_REG4__last_sd_of_di_p2_MASK 0x800000
+#define WD_DEBUG_REG4__last_sd_of_di_p2__SHIFT 0x17
+#define WD_DEBUG_REG4__not_eop_wait_p1_q_MASK 0x1000000
+#define WD_DEBUG_REG4__not_eop_wait_p1_q__SHIFT 0x18
+#define WD_DEBUG_REG4__not_eop_wait_q_MASK 0x2000000
+#define WD_DEBUG_REG4__not_eop_wait_q__SHIFT 0x19
+#define WD_DEBUG_REG4__ext_event_wait_p1_q_MASK 0x4000000
+#define WD_DEBUG_REG4__ext_event_wait_p1_q__SHIFT 0x1a
+#define WD_DEBUG_REG4__ext_event_wait_q_MASK 0x8000000
+#define WD_DEBUG_REG4__ext_event_wait_q__SHIFT 0x1b
+#define WD_DEBUG_REG4__WD_IA1_draw_send_d_MASK 0x10000000
+#define WD_DEBUG_REG4__WD_IA1_draw_send_d__SHIFT 0x1c
+#define WD_DEBUG_REG4__WD_IA1_draw_rtr_MASK 0x20000000
+#define WD_DEBUG_REG4__WD_IA1_draw_rtr__SHIFT 0x1d
+#define WD_DEBUG_REG4__send_to_ia1_q_MASK 0x40000000
+#define WD_DEBUG_REG4__send_to_ia1_q__SHIFT 0x1e
+#define WD_DEBUG_REG4__dual_ia_mode_MASK 0x80000000
+#define WD_DEBUG_REG4__dual_ia_mode__SHIFT 0x1f
+#define WD_DEBUG_REG5__p1_rbiu_spl_dr_valid_MASK 0x1
+#define WD_DEBUG_REG5__p1_rbiu_spl_dr_valid__SHIFT 0x0
+#define WD_DEBUG_REG5__SPARE0_MASK 0x2
+#define WD_DEBUG_REG5__SPARE0__SHIFT 0x1
+#define WD_DEBUG_REG5__p1_pipe0_dr_MASK 0x4
+#define WD_DEBUG_REG5__p1_pipe0_dr__SHIFT 0x2
+#define WD_DEBUG_REG5__p1_pipe0_rtr_MASK 0x8
+#define WD_DEBUG_REG5__p1_pipe0_rtr__SHIFT 0x3
+#define WD_DEBUG_REG5__p1_pipe1_dr_MASK 0x10
+#define WD_DEBUG_REG5__p1_pipe1_dr__SHIFT 0x4
+#define WD_DEBUG_REG5__p1_pipe1_rtr_MASK 0x20
+#define WD_DEBUG_REG5__p1_pipe1_rtr__SHIFT 0x5
+#define WD_DEBUG_REG5__p1_wd_subdma_fifo_empty_MASK 0x40
+#define WD_DEBUG_REG5__p1_wd_subdma_fifo_empty__SHIFT 0x6
+#define WD_DEBUG_REG5__p1_wd_subdma_fifo_full_MASK 0x80
+#define WD_DEBUG_REG5__p1_wd_subdma_fifo_full__SHIFT 0x7
+#define WD_DEBUG_REG5__p1_dma_buf_type_p0_q_MASK 0x300
+#define WD_DEBUG_REG5__p1_dma_buf_type_p0_q__SHIFT 0x8
+#define WD_DEBUG_REG5__p1_dma_zero_indices_p0_q_MASK 0x400
+#define WD_DEBUG_REG5__p1_dma_zero_indices_p0_q__SHIFT 0xa
+#define WD_DEBUG_REG5__p1_dma_req_path_p3_q_MASK 0x800
+#define WD_DEBUG_REG5__p1_dma_req_path_p3_q__SHIFT 0xb
+#define WD_DEBUG_REG5__p1_dma_not_eop_p1_q_MASK 0x1000
+#define WD_DEBUG_REG5__p1_dma_not_eop_p1_q__SHIFT 0xc
+#define WD_DEBUG_REG5__p1_out_of_range_p4_MASK 0x2000
+#define WD_DEBUG_REG5__p1_out_of_range_p4__SHIFT 0xd
+#define WD_DEBUG_REG5__p1_last_sub_dma_p3_q_MASK 0x4000
+#define WD_DEBUG_REG5__p1_last_sub_dma_p3_q__SHIFT 0xe
+#define WD_DEBUG_REG5__p1_last_rdreq_of_sub_dma_p4_MASK 0x8000
+#define WD_DEBUG_REG5__p1_last_rdreq_of_sub_dma_p4__SHIFT 0xf
+#define WD_DEBUG_REG5__p1_WD_IA_dma_send_d_MASK 0x10000
+#define WD_DEBUG_REG5__p1_WD_IA_dma_send_d__SHIFT 0x10
+#define WD_DEBUG_REG5__p1_WD_IA_dma_rtr_MASK 0x20000
+#define WD_DEBUG_REG5__p1_WD_IA_dma_rtr__SHIFT 0x11
+#define WD_DEBUG_REG5__p1_WD_IA1_dma_send_d_MASK 0x40000
+#define WD_DEBUG_REG5__p1_WD_IA1_dma_send_d__SHIFT 0x12
+#define WD_DEBUG_REG5__p1_WD_IA1_dma_rtr_MASK 0x80000
+#define WD_DEBUG_REG5__p1_WD_IA1_dma_rtr__SHIFT 0x13
+#define WD_DEBUG_REG5__p1_last_inst_of_dma_p2_MASK 0x100000
+#define WD_DEBUG_REG5__p1_last_inst_of_dma_p2__SHIFT 0x14
+#define WD_DEBUG_REG5__p1_last_sd_of_inst_p2_MASK 0x200000
+#define WD_DEBUG_REG5__p1_last_sd_of_inst_p2__SHIFT 0x15
+#define WD_DEBUG_REG5__p1_last_sd_of_dma_p2_MASK 0x400000
+#define WD_DEBUG_REG5__p1_last_sd_of_dma_p2__SHIFT 0x16
+#define WD_DEBUG_REG5__SPARE1_MASK 0x800000
+#define WD_DEBUG_REG5__SPARE1__SHIFT 0x17
+#define WD_DEBUG_REG5__p1_WD_IA_dma_busy_MASK 0x1000000
+#define WD_DEBUG_REG5__p1_WD_IA_dma_busy__SHIFT 0x18
+#define WD_DEBUG_REG5__p1_WD_IA1_dma_busy_MASK 0x2000000
+#define WD_DEBUG_REG5__p1_WD_IA1_dma_busy__SHIFT 0x19
+#define WD_DEBUG_REG5__p1_send_to_ia1_p3_q_MASK 0x4000000
+#define WD_DEBUG_REG5__p1_send_to_ia1_p3_q__SHIFT 0x1a
+#define WD_DEBUG_REG5__p1_dma_wd_switch_on_eop_p3_q_MASK 0x8000000
+#define WD_DEBUG_REG5__p1_dma_wd_switch_on_eop_p3_q__SHIFT 0x1b
+#define WD_DEBUG_REG5__p1_pipe3_dr_MASK 0x10000000
+#define WD_DEBUG_REG5__p1_pipe3_dr__SHIFT 0x1c
+#define WD_DEBUG_REG5__p1_pipe3_rtr_MASK 0x20000000
+#define WD_DEBUG_REG5__p1_pipe3_rtr__SHIFT 0x1d
+#define WD_DEBUG_REG5__p1_wd_dma2draw_fifo_empty_MASK 0x40000000
+#define WD_DEBUG_REG5__p1_wd_dma2draw_fifo_empty__SHIFT 0x1e
+#define WD_DEBUG_REG5__p1_wd_dma2draw_fifo_full_MASK 0x80000000
+#define WD_DEBUG_REG5__p1_wd_dma2draw_fifo_full__SHIFT 0x1f
+#define WD_DEBUG_REG6__WD_IA_draw_eop_MASK 0xffffffff
+#define WD_DEBUG_REG6__WD_IA_draw_eop__SHIFT 0x0
+#define WD_DEBUG_REG7__SE0VGT_WD_thdgrp_send_in_MASK 0x1
+#define WD_DEBUG_REG7__SE0VGT_WD_thdgrp_send_in__SHIFT 0x0
+#define WD_DEBUG_REG7__wd_arb_se0_input_fifo_re_MASK 0x2
+#define WD_DEBUG_REG7__wd_arb_se0_input_fifo_re__SHIFT 0x1
+#define WD_DEBUG_REG7__wd_arb_se0_input_fifo_empty_MASK 0x4
+#define WD_DEBUG_REG7__wd_arb_se0_input_fifo_empty__SHIFT 0x2
+#define WD_DEBUG_REG7__wd_arb_se0_input_fifo_full_MASK 0x8
+#define WD_DEBUG_REG7__wd_arb_se0_input_fifo_full__SHIFT 0x3
+#define WD_DEBUG_REG7__SE1VGT_WD_thdgrp_send_in_MASK 0x10
+#define WD_DEBUG_REG7__SE1VGT_WD_thdgrp_send_in__SHIFT 0x4
+#define WD_DEBUG_REG7__wd_arb_se1_input_fifo_re_MASK 0x20
+#define WD_DEBUG_REG7__wd_arb_se1_input_fifo_re__SHIFT 0x5
+#define WD_DEBUG_REG7__wd_arb_se1_input_fifo_empty_MASK 0x40
+#define WD_DEBUG_REG7__wd_arb_se1_input_fifo_empty__SHIFT 0x6
+#define WD_DEBUG_REG7__wd_arb_se1_input_fifo_full_MASK 0x80
+#define WD_DEBUG_REG7__wd_arb_se1_input_fifo_full__SHIFT 0x7
+#define WD_DEBUG_REG7__SPARE1_MASK 0xf00
+#define WD_DEBUG_REG7__SPARE1__SHIFT 0x8
+#define WD_DEBUG_REG7__SPARE2_MASK 0xf000
+#define WD_DEBUG_REG7__SPARE2__SHIFT 0xc
+#define WD_DEBUG_REG7__te11_arb_state_q_MASK 0x70000
+#define WD_DEBUG_REG7__te11_arb_state_q__SHIFT 0x10
+#define WD_DEBUG_REG7__SPARE5_MASK 0x80000
+#define WD_DEBUG_REG7__SPARE5__SHIFT 0x13
+#define WD_DEBUG_REG7__se0_thdgrp_is_event_MASK 0x100000
+#define WD_DEBUG_REG7__se0_thdgrp_is_event__SHIFT 0x14
+#define WD_DEBUG_REG7__se0_thdgrp_eop_MASK 0x200000
+#define WD_DEBUG_REG7__se0_thdgrp_eop__SHIFT 0x15
+#define WD_DEBUG_REG7__se1_thdgrp_is_event_MASK 0x400000
+#define WD_DEBUG_REG7__se1_thdgrp_is_event__SHIFT 0x16
+#define WD_DEBUG_REG7__se1_thdgrp_eop_MASK 0x800000
+#define WD_DEBUG_REG7__se1_thdgrp_eop__SHIFT 0x17
+#define WD_DEBUG_REG7__SPARE6_MASK 0xf000000
+#define WD_DEBUG_REG7__SPARE6__SHIFT 0x18
+#define WD_DEBUG_REG7__tfreq_arb_tgroup_rtr_MASK 0x10000000
+#define WD_DEBUG_REG7__tfreq_arb_tgroup_rtr__SHIFT 0x1c
+#define WD_DEBUG_REG7__arb_tfreq_tgroup_rts_MASK 0x20000000
+#define WD_DEBUG_REG7__arb_tfreq_tgroup_rts__SHIFT 0x1d
+#define WD_DEBUG_REG7__arb_tfreq_tgroup_event_MASK 0x40000000
+#define WD_DEBUG_REG7__arb_tfreq_tgroup_event__SHIFT 0x1e
+#define WD_DEBUG_REG7__te11_arb_busy_MASK 0x80000000
+#define WD_DEBUG_REG7__te11_arb_busy__SHIFT 0x1f
+#define WD_DEBUG_REG8__pipe0_dr_MASK 0x1
+#define WD_DEBUG_REG8__pipe0_dr__SHIFT 0x0
+#define WD_DEBUG_REG8__pipe1_dr_MASK 0x2
+#define WD_DEBUG_REG8__pipe1_dr__SHIFT 0x1
+#define WD_DEBUG_REG8__pipe0_rtr_MASK 0x4
+#define WD_DEBUG_REG8__pipe0_rtr__SHIFT 0x2
+#define WD_DEBUG_REG8__pipe1_rtr_MASK 0x8
+#define WD_DEBUG_REG8__pipe1_rtr__SHIFT 0x3
+#define WD_DEBUG_REG8__tfreq_tg_fifo_empty_MASK 0x10
+#define WD_DEBUG_REG8__tfreq_tg_fifo_empty__SHIFT 0x4
+#define WD_DEBUG_REG8__tfreq_tg_fifo_full_MASK 0x20
+#define WD_DEBUG_REG8__tfreq_tg_fifo_full__SHIFT 0x5
+#define WD_DEBUG_REG8__tf_data_fifo_busy_q_MASK 0x40
+#define WD_DEBUG_REG8__tf_data_fifo_busy_q__SHIFT 0x6
+#define WD_DEBUG_REG8__tf_data_fifo_rtr_q_MASK 0x80
+#define WD_DEBUG_REG8__tf_data_fifo_rtr_q__SHIFT 0x7
+#define WD_DEBUG_REG8__tf_skid_fifo_empty_MASK 0x100
+#define WD_DEBUG_REG8__tf_skid_fifo_empty__SHIFT 0x8
+#define WD_DEBUG_REG8__tf_skid_fifo_full_MASK 0x200
+#define WD_DEBUG_REG8__tf_skid_fifo_full__SHIFT 0x9
+#define WD_DEBUG_REG8__wd_tc_rdreq_rtr_q_MASK 0x400
+#define WD_DEBUG_REG8__wd_tc_rdreq_rtr_q__SHIFT 0xa
+#define WD_DEBUG_REG8__last_req_of_tg_p2_MASK 0x800
+#define WD_DEBUG_REG8__last_req_of_tg_p2__SHIFT 0xb
+#define WD_DEBUG_REG8__se0spi_wd_hs_done_cnt_q_MASK 0x3f000
+#define WD_DEBUG_REG8__se0spi_wd_hs_done_cnt_q__SHIFT 0xc
+#define WD_DEBUG_REG8__event_flag_p1_q_MASK 0x40000
+#define WD_DEBUG_REG8__event_flag_p1_q__SHIFT 0x12
+#define WD_DEBUG_REG8__null_flag_p1_q_MASK 0x80000
+#define WD_DEBUG_REG8__null_flag_p1_q__SHIFT 0x13
+#define WD_DEBUG_REG8__tf_data_fifo_cnt_q_MASK 0x7f00000
+#define WD_DEBUG_REG8__tf_data_fifo_cnt_q__SHIFT 0x14
+#define WD_DEBUG_REG8__second_tf_ret_data_q_MASK 0x8000000
+#define WD_DEBUG_REG8__second_tf_ret_data_q__SHIFT 0x1b
+#define WD_DEBUG_REG8__first_req_of_tg_p1_q_MASK 0x10000000
+#define WD_DEBUG_REG8__first_req_of_tg_p1_q__SHIFT 0x1c
+#define WD_DEBUG_REG8__WD_TC_rdreq_send_out_MASK 0x20000000
+#define WD_DEBUG_REG8__WD_TC_rdreq_send_out__SHIFT 0x1d
+#define WD_DEBUG_REG8__WD_TC_rdnfo_stall_out_MASK 0x40000000
+#define WD_DEBUG_REG8__WD_TC_rdnfo_stall_out__SHIFT 0x1e
+#define WD_DEBUG_REG8__TC_WD_rdret_valid_in_MASK 0x80000000
+#define WD_DEBUG_REG8__TC_WD_rdret_valid_in__SHIFT 0x1f
+#define WD_DEBUG_REG9__pipe0_dr_MASK 0x1
+#define WD_DEBUG_REG9__pipe0_dr__SHIFT 0x0
+#define WD_DEBUG_REG9__pipec_tf_dr_MASK 0x2
+#define WD_DEBUG_REG9__pipec_tf_dr__SHIFT 0x1
+#define WD_DEBUG_REG9__pipe2_dr_MASK 0x4
+#define WD_DEBUG_REG9__pipe2_dr__SHIFT 0x2
+#define WD_DEBUG_REG9__event_or_null_flags_p0_q_MASK 0x8
+#define WD_DEBUG_REG9__event_or_null_flags_p0_q__SHIFT 0x3
+#define WD_DEBUG_REG9__pipe0_rtr_MASK 0x10
+#define WD_DEBUG_REG9__pipe0_rtr__SHIFT 0x4
+#define WD_DEBUG_REG9__pipe1_rtr_MASK 0x20
+#define WD_DEBUG_REG9__pipe1_rtr__SHIFT 0x5
+#define WD_DEBUG_REG9__pipec_tf_rtr_MASK 0x40
+#define WD_DEBUG_REG9__pipec_tf_rtr__SHIFT 0x6
+#define WD_DEBUG_REG9__pipe2_rtr_MASK 0x80
+#define WD_DEBUG_REG9__pipe2_rtr__SHIFT 0x7
+#define WD_DEBUG_REG9__ttp_patch_fifo_full_MASK 0x100
+#define WD_DEBUG_REG9__ttp_patch_fifo_full__SHIFT 0x8
+#define WD_DEBUG_REG9__ttp_patch_fifo_empty_MASK 0x200
+#define WD_DEBUG_REG9__ttp_patch_fifo_empty__SHIFT 0x9
+#define WD_DEBUG_REG9__ttp_tf_fifo_empty_MASK 0x400
+#define WD_DEBUG_REG9__ttp_tf_fifo_empty__SHIFT 0xa
+#define WD_DEBUG_REG9__SPARE0_MASK 0xf800
+#define WD_DEBUG_REG9__SPARE0__SHIFT 0xb
+#define WD_DEBUG_REG9__tf_fetch_state_q_MASK 0x70000
+#define WD_DEBUG_REG9__tf_fetch_state_q__SHIFT 0x10
+#define WD_DEBUG_REG9__last_patch_of_tg_MASK 0x80000
+#define WD_DEBUG_REG9__last_patch_of_tg__SHIFT 0x13
+#define WD_DEBUG_REG9__tf_pointer_p0_q_MASK 0xf00000
+#define WD_DEBUG_REG9__tf_pointer_p0_q__SHIFT 0x14
+#define WD_DEBUG_REG9__dynamic_hs_p0_q_MASK 0x1000000
+#define WD_DEBUG_REG9__dynamic_hs_p0_q__SHIFT 0x18
+#define WD_DEBUG_REG9__first_fetch_of_tg_p0_q_MASK 0x2000000
+#define WD_DEBUG_REG9__first_fetch_of_tg_p0_q__SHIFT 0x19
+#define WD_DEBUG_REG9__mem_is_even_MASK 0x4000000
+#define WD_DEBUG_REG9__mem_is_even__SHIFT 0x1a
+#define WD_DEBUG_REG9__SPARE1_MASK 0x8000000
+#define WD_DEBUG_REG9__SPARE1__SHIFT 0x1b
+#define WD_DEBUG_REG9__SPARE2_MASK 0x30000000
+#define WD_DEBUG_REG9__SPARE2__SHIFT 0x1c
+#define WD_DEBUG_REG9__pipe4_dr_MASK 0x40000000
+#define WD_DEBUG_REG9__pipe4_dr__SHIFT 0x1e
+#define WD_DEBUG_REG9__pipe4_rtr_MASK 0x80000000
+#define WD_DEBUG_REG9__pipe4_rtr__SHIFT 0x1f
+#define WD_DEBUG_REG10__ttp_pd_patch_rts_MASK 0x1
+#define WD_DEBUG_REG10__ttp_pd_patch_rts__SHIFT 0x0
+#define WD_DEBUG_REG10__ttp_pd_is_event_MASK 0x2
+#define WD_DEBUG_REG10__ttp_pd_is_event__SHIFT 0x1
+#define WD_DEBUG_REG10__ttp_pd_eopg_MASK 0x4
+#define WD_DEBUG_REG10__ttp_pd_eopg__SHIFT 0x2
+#define WD_DEBUG_REG10__ttp_pd_eop_MASK 0x8
+#define WD_DEBUG_REG10__ttp_pd_eop__SHIFT 0x3
+#define WD_DEBUG_REG10__pipe0_dr_MASK 0x10
+#define WD_DEBUG_REG10__pipe0_dr__SHIFT 0x4
+#define WD_DEBUG_REG10__pipe1_dr_MASK 0x20
+#define WD_DEBUG_REG10__pipe1_dr__SHIFT 0x5
+#define WD_DEBUG_REG10__pipe0_rtr_MASK 0x40
+#define WD_DEBUG_REG10__pipe0_rtr__SHIFT 0x6
+#define WD_DEBUG_REG10__pipe1_rtr_MASK 0x80
+#define WD_DEBUG_REG10__pipe1_rtr__SHIFT 0x7
+#define WD_DEBUG_REG10__donut_en_p1_q_MASK 0x100
+#define WD_DEBUG_REG10__donut_en_p1_q__SHIFT 0x8
+#define WD_DEBUG_REG10__donut_se_switch_p2_MASK 0x200
+#define WD_DEBUG_REG10__donut_se_switch_p2__SHIFT 0x9
+#define WD_DEBUG_REG10__patch_se_switch_p2_MASK 0x400
+#define WD_DEBUG_REG10__patch_se_switch_p2__SHIFT 0xa
+#define WD_DEBUG_REG10__last_donut_switch_p2_MASK 0x800
+#define WD_DEBUG_REG10__last_donut_switch_p2__SHIFT 0xb
+#define WD_DEBUG_REG10__last_donut_of_patch_p2_MASK 0x1000
+#define WD_DEBUG_REG10__last_donut_of_patch_p2__SHIFT 0xc
+#define WD_DEBUG_REG10__is_event_p1_q_MASK 0x2000
+#define WD_DEBUG_REG10__is_event_p1_q__SHIFT 0xd
+#define WD_DEBUG_REG10__eopg_p1_q_MASK 0x4000
+#define WD_DEBUG_REG10__eopg_p1_q__SHIFT 0xe
+#define WD_DEBUG_REG10__eop_p1_q_MASK 0x8000
+#define WD_DEBUG_REG10__eop_p1_q__SHIFT 0xf
+#define WD_DEBUG_REG10__patch_accum_q_MASK 0xff0000
+#define WD_DEBUG_REG10__patch_accum_q__SHIFT 0x10
+#define WD_DEBUG_REG10__wd_te11_out_se0_fifo_full_MASK 0x1000000
+#define WD_DEBUG_REG10__wd_te11_out_se0_fifo_full__SHIFT 0x18
+#define WD_DEBUG_REG10__wd_te11_out_se0_fifo_empty_MASK 0x2000000
+#define WD_DEBUG_REG10__wd_te11_out_se0_fifo_empty__SHIFT 0x19
+#define WD_DEBUG_REG10__wd_te11_out_se1_fifo_full_MASK 0x4000000
+#define WD_DEBUG_REG10__wd_te11_out_se1_fifo_full__SHIFT 0x1a
+#define WD_DEBUG_REG10__wd_te11_out_se1_fifo_empty_MASK 0x8000000
+#define WD_DEBUG_REG10__wd_te11_out_se1_fifo_empty__SHIFT 0x1b
+#define WD_DEBUG_REG10__wd_te11_out_se2_fifo_full_MASK 0x10000000
+#define WD_DEBUG_REG10__wd_te11_out_se2_fifo_full__SHIFT 0x1c
+#define WD_DEBUG_REG10__wd_te11_out_se2_fifo_empty_MASK 0x20000000
+#define WD_DEBUG_REG10__wd_te11_out_se2_fifo_empty__SHIFT 0x1d
+#define WD_DEBUG_REG10__wd_te11_out_se3_fifo_full_MASK 0x40000000
+#define WD_DEBUG_REG10__wd_te11_out_se3_fifo_full__SHIFT 0x1e
+#define WD_DEBUG_REG10__wd_te11_out_se3_fifo_empty_MASK 0x80000000
+#define WD_DEBUG_REG10__wd_te11_out_se3_fifo_empty__SHIFT 0x1f
+#define IA_DEBUG_REG0__ia_busy_extended_MASK 0x1
+#define IA_DEBUG_REG0__ia_busy_extended__SHIFT 0x0
+#define IA_DEBUG_REG0__ia_nodma_busy_extended_MASK 0x2
+#define IA_DEBUG_REG0__ia_nodma_busy_extended__SHIFT 0x1
+#define IA_DEBUG_REG0__ia_busy_MASK 0x4
+#define IA_DEBUG_REG0__ia_busy__SHIFT 0x2
+#define IA_DEBUG_REG0__ia_nodma_busy_MASK 0x8
+#define IA_DEBUG_REG0__ia_nodma_busy__SHIFT 0x3
+#define IA_DEBUG_REG0__SPARE0_MASK 0x10
+#define IA_DEBUG_REG0__SPARE0__SHIFT 0x4
+#define IA_DEBUG_REG0__dma_req_busy_MASK 0x20
+#define IA_DEBUG_REG0__dma_req_busy__SHIFT 0x5
+#define IA_DEBUG_REG0__dma_busy_MASK 0x40
+#define IA_DEBUG_REG0__dma_busy__SHIFT 0x6
+#define IA_DEBUG_REG0__mc_xl8r_busy_MASK 0x80
+#define IA_DEBUG_REG0__mc_xl8r_busy__SHIFT 0x7
+#define IA_DEBUG_REG0__grp_busy_MASK 0x100
+#define IA_DEBUG_REG0__grp_busy__SHIFT 0x8
+#define IA_DEBUG_REG0__SPARE1_MASK 0x200
+#define IA_DEBUG_REG0__SPARE1__SHIFT 0x9
+#define IA_DEBUG_REG0__dma_grp_valid_MASK 0x400
+#define IA_DEBUG_REG0__dma_grp_valid__SHIFT 0xa
+#define IA_DEBUG_REG0__grp_dma_read_MASK 0x800
+#define IA_DEBUG_REG0__grp_dma_read__SHIFT 0xb
+#define IA_DEBUG_REG0__dma_grp_hp_valid_MASK 0x1000
+#define IA_DEBUG_REG0__dma_grp_hp_valid__SHIFT 0xc
+#define IA_DEBUG_REG0__grp_dma_hp_read_MASK 0x2000
+#define IA_DEBUG_REG0__grp_dma_hp_read__SHIFT 0xd
+#define IA_DEBUG_REG0__SPARE2_MASK 0xffc000
+#define IA_DEBUG_REG0__SPARE2__SHIFT 0xe
+#define IA_DEBUG_REG0__reg_clk_busy_MASK 0x1000000
+#define IA_DEBUG_REG0__reg_clk_busy__SHIFT 0x18
+#define IA_DEBUG_REG0__core_clk_busy_MASK 0x2000000
+#define IA_DEBUG_REG0__core_clk_busy__SHIFT 0x19
+#define IA_DEBUG_REG0__SPARE3_MASK 0x4000000
+#define IA_DEBUG_REG0__SPARE3__SHIFT 0x1a
+#define IA_DEBUG_REG0__SPARE4_MASK 0x8000000
+#define IA_DEBUG_REG0__SPARE4__SHIFT 0x1b
+#define IA_DEBUG_REG0__sclk_reg_vld_MASK 0x10000000
+#define IA_DEBUG_REG0__sclk_reg_vld__SHIFT 0x1c
+#define IA_DEBUG_REG0__sclk_core_vld_MASK 0x20000000
+#define IA_DEBUG_REG0__sclk_core_vld__SHIFT 0x1d
+#define IA_DEBUG_REG0__SPARE5_MASK 0x40000000
+#define IA_DEBUG_REG0__SPARE5__SHIFT 0x1e
+#define IA_DEBUG_REG0__SPARE6_MASK 0x80000000
+#define IA_DEBUG_REG0__SPARE6__SHIFT 0x1f
+#define IA_DEBUG_REG1__dma_input_fifo_empty_MASK 0x1
+#define IA_DEBUG_REG1__dma_input_fifo_empty__SHIFT 0x0
+#define IA_DEBUG_REG1__dma_input_fifo_full_MASK 0x2
+#define IA_DEBUG_REG1__dma_input_fifo_full__SHIFT 0x1
+#define IA_DEBUG_REG1__start_new_packet_MASK 0x4
+#define IA_DEBUG_REG1__start_new_packet__SHIFT 0x2
+#define IA_DEBUG_REG1__dma_rdreq_dr_q_MASK 0x8
+#define IA_DEBUG_REG1__dma_rdreq_dr_q__SHIFT 0x3
+#define IA_DEBUG_REG1__dma_zero_indices_q_MASK 0x10
+#define IA_DEBUG_REG1__dma_zero_indices_q__SHIFT 0x4
+#define IA_DEBUG_REG1__dma_buf_type_q_MASK 0x60
+#define IA_DEBUG_REG1__dma_buf_type_q__SHIFT 0x5
+#define IA_DEBUG_REG1__dma_req_path_q_MASK 0x80
+#define IA_DEBUG_REG1__dma_req_path_q__SHIFT 0x7
+#define IA_DEBUG_REG1__discard_1st_chunk_MASK 0x100
+#define IA_DEBUG_REG1__discard_1st_chunk__SHIFT 0x8
+#define IA_DEBUG_REG1__discard_2nd_chunk_MASK 0x200
+#define IA_DEBUG_REG1__discard_2nd_chunk__SHIFT 0x9
+#define IA_DEBUG_REG1__second_tc_ret_data_q_MASK 0x400
+#define IA_DEBUG_REG1__second_tc_ret_data_q__SHIFT 0xa
+#define IA_DEBUG_REG1__dma_tc_ret_sel_q_MASK 0x800
+#define IA_DEBUG_REG1__dma_tc_ret_sel_q__SHIFT 0xb
+#define IA_DEBUG_REG1__last_rdreq_in_dma_op_MASK 0x1000
+#define IA_DEBUG_REG1__last_rdreq_in_dma_op__SHIFT 0xc
+#define IA_DEBUG_REG1__dma_mask_fifo_empty_MASK 0x2000
+#define IA_DEBUG_REG1__dma_mask_fifo_empty__SHIFT 0xd
+#define IA_DEBUG_REG1__dma_data_fifo_empty_q_MASK 0x4000
+#define IA_DEBUG_REG1__dma_data_fifo_empty_q__SHIFT 0xe
+#define IA_DEBUG_REG1__dma_data_fifo_full_MASK 0x8000
+#define IA_DEBUG_REG1__dma_data_fifo_full__SHIFT 0xf
+#define IA_DEBUG_REG1__dma_req_fifo_empty_MASK 0x10000
+#define IA_DEBUG_REG1__dma_req_fifo_empty__SHIFT 0x10
+#define IA_DEBUG_REG1__dma_req_fifo_full_MASK 0x20000
+#define IA_DEBUG_REG1__dma_req_fifo_full__SHIFT 0x11
+#define IA_DEBUG_REG1__stage2_dr_MASK 0x40000
+#define IA_DEBUG_REG1__stage2_dr__SHIFT 0x12
+#define IA_DEBUG_REG1__stage2_rtr_MASK 0x80000
+#define IA_DEBUG_REG1__stage2_rtr__SHIFT 0x13
+#define IA_DEBUG_REG1__stage3_dr_MASK 0x100000
+#define IA_DEBUG_REG1__stage3_dr__SHIFT 0x14
+#define IA_DEBUG_REG1__stage3_rtr_MASK 0x200000
+#define IA_DEBUG_REG1__stage3_rtr__SHIFT 0x15
+#define IA_DEBUG_REG1__stage4_dr_MASK 0x400000
+#define IA_DEBUG_REG1__stage4_dr__SHIFT 0x16
+#define IA_DEBUG_REG1__stage4_rtr_MASK 0x800000
+#define IA_DEBUG_REG1__stage4_rtr__SHIFT 0x17
+#define IA_DEBUG_REG1__dma_skid_fifo_empty_MASK 0x1000000
+#define IA_DEBUG_REG1__dma_skid_fifo_empty__SHIFT 0x18
+#define IA_DEBUG_REG1__dma_skid_fifo_full_MASK 0x2000000
+#define IA_DEBUG_REG1__dma_skid_fifo_full__SHIFT 0x19
+#define IA_DEBUG_REG1__dma_grp_valid_MASK 0x4000000
+#define IA_DEBUG_REG1__dma_grp_valid__SHIFT 0x1a
+#define IA_DEBUG_REG1__grp_dma_read_MASK 0x8000000
+#define IA_DEBUG_REG1__grp_dma_read__SHIFT 0x1b
+#define IA_DEBUG_REG1__current_data_valid_MASK 0x10000000
+#define IA_DEBUG_REG1__current_data_valid__SHIFT 0x1c
+#define IA_DEBUG_REG1__out_of_range_r2_q_MASK 0x20000000
+#define IA_DEBUG_REG1__out_of_range_r2_q__SHIFT 0x1d
+#define IA_DEBUG_REG1__dma_mask_fifo_we_MASK 0x40000000
+#define IA_DEBUG_REG1__dma_mask_fifo_we__SHIFT 0x1e
+#define IA_DEBUG_REG1__dma_ret_data_we_q_MASK 0x80000000
+#define IA_DEBUG_REG1__dma_ret_data_we_q__SHIFT 0x1f
+#define IA_DEBUG_REG2__hp_dma_input_fifo_empty_MASK 0x1
+#define IA_DEBUG_REG2__hp_dma_input_fifo_empty__SHIFT 0x0
+#define IA_DEBUG_REG2__hp_dma_input_fifo_full_MASK 0x2
+#define IA_DEBUG_REG2__hp_dma_input_fifo_full__SHIFT 0x1
+#define IA_DEBUG_REG2__hp_start_new_packet_MASK 0x4
+#define IA_DEBUG_REG2__hp_start_new_packet__SHIFT 0x2
+#define IA_DEBUG_REG2__hp_dma_rdreq_dr_q_MASK 0x8
+#define IA_DEBUG_REG2__hp_dma_rdreq_dr_q__SHIFT 0x3
+#define IA_DEBUG_REG2__hp_dma_zero_indices_q_MASK 0x10
+#define IA_DEBUG_REG2__hp_dma_zero_indices_q__SHIFT 0x4
+#define IA_DEBUG_REG2__hp_dma_buf_type_q_MASK 0x60
+#define IA_DEBUG_REG2__hp_dma_buf_type_q__SHIFT 0x5
+#define IA_DEBUG_REG2__hp_dma_req_path_q_MASK 0x80
+#define IA_DEBUG_REG2__hp_dma_req_path_q__SHIFT 0x7
+#define IA_DEBUG_REG2__hp_discard_1st_chunk_MASK 0x100
+#define IA_DEBUG_REG2__hp_discard_1st_chunk__SHIFT 0x8
+#define IA_DEBUG_REG2__hp_discard_2nd_chunk_MASK 0x200
+#define IA_DEBUG_REG2__hp_discard_2nd_chunk__SHIFT 0x9
+#define IA_DEBUG_REG2__hp_second_tc_ret_data_q_MASK 0x400
+#define IA_DEBUG_REG2__hp_second_tc_ret_data_q__SHIFT 0xa
+#define IA_DEBUG_REG2__hp_dma_tc_ret_sel_q_MASK 0x800
+#define IA_DEBUG_REG2__hp_dma_tc_ret_sel_q__SHIFT 0xb
+#define IA_DEBUG_REG2__hp_last_rdreq_in_dma_op_MASK 0x1000
+#define IA_DEBUG_REG2__hp_last_rdreq_in_dma_op__SHIFT 0xc
+#define IA_DEBUG_REG2__hp_dma_mask_fifo_empty_MASK 0x2000
+#define IA_DEBUG_REG2__hp_dma_mask_fifo_empty__SHIFT 0xd
+#define IA_DEBUG_REG2__hp_dma_data_fifo_empty_q_MASK 0x4000
+#define IA_DEBUG_REG2__hp_dma_data_fifo_empty_q__SHIFT 0xe
+#define IA_DEBUG_REG2__hp_dma_data_fifo_full_MASK 0x8000
+#define IA_DEBUG_REG2__hp_dma_data_fifo_full__SHIFT 0xf
+#define IA_DEBUG_REG2__hp_dma_req_fifo_empty_MASK 0x10000
+#define IA_DEBUG_REG2__hp_dma_req_fifo_empty__SHIFT 0x10
+#define IA_DEBUG_REG2__hp_dma_req_fifo_full_MASK 0x20000
+#define IA_DEBUG_REG2__hp_dma_req_fifo_full__SHIFT 0x11
+#define IA_DEBUG_REG2__hp_stage2_dr_MASK 0x40000
+#define IA_DEBUG_REG2__hp_stage2_dr__SHIFT 0x12
+#define IA_DEBUG_REG2__hp_stage2_rtr_MASK 0x80000
+#define IA_DEBUG_REG2__hp_stage2_rtr__SHIFT 0x13
+#define IA_DEBUG_REG2__hp_stage3_dr_MASK 0x100000
+#define IA_DEBUG_REG2__hp_stage3_dr__SHIFT 0x14
+#define IA_DEBUG_REG2__hp_stage3_rtr_MASK 0x200000
+#define IA_DEBUG_REG2__hp_stage3_rtr__SHIFT 0x15
+#define IA_DEBUG_REG2__hp_stage4_dr_MASK 0x400000
+#define IA_DEBUG_REG2__hp_stage4_dr__SHIFT 0x16
+#define IA_DEBUG_REG2__hp_stage4_rtr_MASK 0x800000
+#define IA_DEBUG_REG2__hp_stage4_rtr__SHIFT 0x17
+#define IA_DEBUG_REG2__hp_dma_skid_fifo_empty_MASK 0x1000000
+#define IA_DEBUG_REG2__hp_dma_skid_fifo_empty__SHIFT 0x18
+#define IA_DEBUG_REG2__hp_dma_skid_fifo_full_MASK 0x2000000
+#define IA_DEBUG_REG2__hp_dma_skid_fifo_full__SHIFT 0x19
+#define IA_DEBUG_REG2__hp_dma_grp_valid_MASK 0x4000000
+#define IA_DEBUG_REG2__hp_dma_grp_valid__SHIFT 0x1a
+#define IA_DEBUG_REG2__hp_grp_dma_read_MASK 0x8000000
+#define IA_DEBUG_REG2__hp_grp_dma_read__SHIFT 0x1b
+#define IA_DEBUG_REG2__hp_current_data_valid_MASK 0x10000000
+#define IA_DEBUG_REG2__hp_current_data_valid__SHIFT 0x1c
+#define IA_DEBUG_REG2__hp_out_of_range_r2_q_MASK 0x20000000
+#define IA_DEBUG_REG2__hp_out_of_range_r2_q__SHIFT 0x1d
+#define IA_DEBUG_REG2__hp_dma_mask_fifo_we_MASK 0x40000000
+#define IA_DEBUG_REG2__hp_dma_mask_fifo_we__SHIFT 0x1e
+#define IA_DEBUG_REG2__hp_dma_ret_data_we_q_MASK 0x80000000
+#define IA_DEBUG_REG2__hp_dma_ret_data_we_q__SHIFT 0x1f
+#define IA_DEBUG_REG3__dma_pipe0_rdreq_valid_MASK 0x1
+#define IA_DEBUG_REG3__dma_pipe0_rdreq_valid__SHIFT 0x0
+#define IA_DEBUG_REG3__dma_pipe0_rdreq_read_MASK 0x2
+#define IA_DEBUG_REG3__dma_pipe0_rdreq_read__SHIFT 0x1
+#define IA_DEBUG_REG3__dma_pipe0_rdreq_null_out_MASK 0x4
+#define IA_DEBUG_REG3__dma_pipe0_rdreq_null_out__SHIFT 0x2
+#define IA_DEBUG_REG3__dma_pipe0_rdreq_eop_out_MASK 0x8
+#define IA_DEBUG_REG3__dma_pipe0_rdreq_eop_out__SHIFT 0x3
+#define IA_DEBUG_REG3__dma_pipe0_rdreq_use_tc_out_MASK 0x10
+#define IA_DEBUG_REG3__dma_pipe0_rdreq_use_tc_out__SHIFT 0x4
+#define IA_DEBUG_REG3__grp_dma_draw_is_pipe0_MASK 0x20
+#define IA_DEBUG_REG3__grp_dma_draw_is_pipe0__SHIFT 0x5
+#define IA_DEBUG_REG3__must_service_pipe0_req_MASK 0x40
+#define IA_DEBUG_REG3__must_service_pipe0_req__SHIFT 0x6
+#define IA_DEBUG_REG3__send_pipe1_req_MASK 0x80
+#define IA_DEBUG_REG3__send_pipe1_req__SHIFT 0x7
+#define IA_DEBUG_REG3__dma_pipe1_rdreq_valid_MASK 0x100
+#define IA_DEBUG_REG3__dma_pipe1_rdreq_valid__SHIFT 0x8
+#define IA_DEBUG_REG3__dma_pipe1_rdreq_read_MASK 0x200
+#define IA_DEBUG_REG3__dma_pipe1_rdreq_read__SHIFT 0x9
+#define IA_DEBUG_REG3__dma_pipe1_rdreq_null_out_MASK 0x400
+#define IA_DEBUG_REG3__dma_pipe1_rdreq_null_out__SHIFT 0xa
+#define IA_DEBUG_REG3__dma_pipe1_rdreq_eop_out_MASK 0x800
+#define IA_DEBUG_REG3__dma_pipe1_rdreq_eop_out__SHIFT 0xb
+#define IA_DEBUG_REG3__dma_pipe1_rdreq_use_tc_out_MASK 0x1000
+#define IA_DEBUG_REG3__dma_pipe1_rdreq_use_tc_out__SHIFT 0xc
+#define IA_DEBUG_REG3__ia_mc_rdreq_rtr_q_MASK 0x2000
+#define IA_DEBUG_REG3__ia_mc_rdreq_rtr_q__SHIFT 0xd
+#define IA_DEBUG_REG3__mc_out_rtr_MASK 0x4000
+#define IA_DEBUG_REG3__mc_out_rtr__SHIFT 0xe
+#define IA_DEBUG_REG3__dma_rdreq_send_out_MASK 0x8000
+#define IA_DEBUG_REG3__dma_rdreq_send_out__SHIFT 0xf
+#define IA_DEBUG_REG3__pipe0_dr_MASK 0x10000
+#define IA_DEBUG_REG3__pipe0_dr__SHIFT 0x10
+#define IA_DEBUG_REG3__pipe0_rtr_MASK 0x20000
+#define IA_DEBUG_REG3__pipe0_rtr__SHIFT 0x11
+#define IA_DEBUG_REG3__ia_tc_rdreq_rtr_q_MASK 0x40000
+#define IA_DEBUG_REG3__ia_tc_rdreq_rtr_q__SHIFT 0x12
+#define IA_DEBUG_REG3__tc_out_rtr_MASK 0x80000
+#define IA_DEBUG_REG3__tc_out_rtr__SHIFT 0x13
+#define IA_DEBUG_REG3__pair0_valid_p1_MASK 0x100000
+#define IA_DEBUG_REG3__pair0_valid_p1__SHIFT 0x14
+#define IA_DEBUG_REG3__pair1_valid_p1_MASK 0x200000
+#define IA_DEBUG_REG3__pair1_valid_p1__SHIFT 0x15
+#define IA_DEBUG_REG3__pair2_valid_p1_MASK 0x400000
+#define IA_DEBUG_REG3__pair2_valid_p1__SHIFT 0x16
+#define IA_DEBUG_REG3__pair3_valid_p1_MASK 0x800000
+#define IA_DEBUG_REG3__pair3_valid_p1__SHIFT 0x17
+#define IA_DEBUG_REG3__tc_req_count_q_MASK 0x3000000
+#define IA_DEBUG_REG3__tc_req_count_q__SHIFT 0x18
+#define IA_DEBUG_REG3__discard_1st_chunk_MASK 0x4000000
+#define IA_DEBUG_REG3__discard_1st_chunk__SHIFT 0x1a
+#define IA_DEBUG_REG3__discard_2nd_chunk_MASK 0x8000000
+#define IA_DEBUG_REG3__discard_2nd_chunk__SHIFT 0x1b
+#define IA_DEBUG_REG3__last_tc_req_p1_MASK 0x10000000
+#define IA_DEBUG_REG3__last_tc_req_p1__SHIFT 0x1c
+#define IA_DEBUG_REG3__IA_TC_rdreq_send_out_MASK 0x20000000
+#define IA_DEBUG_REG3__IA_TC_rdreq_send_out__SHIFT 0x1d
+#define IA_DEBUG_REG3__TC_IA_rdret_valid_in_MASK 0x40000000
+#define IA_DEBUG_REG3__TC_IA_rdret_valid_in__SHIFT 0x1e
+#define IA_DEBUG_REG3__TAP_IA_rdret_vld_in_MASK 0x80000000
+#define IA_DEBUG_REG3__TAP_IA_rdret_vld_in__SHIFT 0x1f
+#define IA_DEBUG_REG4__pipe0_dr_MASK 0x1
+#define IA_DEBUG_REG4__pipe0_dr__SHIFT 0x0
+#define IA_DEBUG_REG4__pipe1_dr_MASK 0x2
+#define IA_DEBUG_REG4__pipe1_dr__SHIFT 0x1
+#define IA_DEBUG_REG4__pipe2_dr_MASK 0x4
+#define IA_DEBUG_REG4__pipe2_dr__SHIFT 0x2
+#define IA_DEBUG_REG4__pipe3_dr_MASK 0x8
+#define IA_DEBUG_REG4__pipe3_dr__SHIFT 0x3
+#define IA_DEBUG_REG4__pipe4_dr_MASK 0x10
+#define IA_DEBUG_REG4__pipe4_dr__SHIFT 0x4
+#define IA_DEBUG_REG4__pipe5_dr_MASK 0x20
+#define IA_DEBUG_REG4__pipe5_dr__SHIFT 0x5
+#define IA_DEBUG_REG4__grp_se0_fifo_empty_MASK 0x40
+#define IA_DEBUG_REG4__grp_se0_fifo_empty__SHIFT 0x6
+#define IA_DEBUG_REG4__grp_se0_fifo_full_MASK 0x80
+#define IA_DEBUG_REG4__grp_se0_fifo_full__SHIFT 0x7
+#define IA_DEBUG_REG4__pipe0_rtr_MASK 0x100
+#define IA_DEBUG_REG4__pipe0_rtr__SHIFT 0x8
+#define IA_DEBUG_REG4__pipe1_rtr_MASK 0x200
+#define IA_DEBUG_REG4__pipe1_rtr__SHIFT 0x9
+#define IA_DEBUG_REG4__pipe2_rtr_MASK 0x400
+#define IA_DEBUG_REG4__pipe2_rtr__SHIFT 0xa
+#define IA_DEBUG_REG4__pipe3_rtr_MASK 0x800
+#define IA_DEBUG_REG4__pipe3_rtr__SHIFT 0xb
+#define IA_DEBUG_REG4__pipe4_rtr_MASK 0x1000
+#define IA_DEBUG_REG4__pipe4_rtr__SHIFT 0xc
+#define IA_DEBUG_REG4__pipe5_rtr_MASK 0x2000
+#define IA_DEBUG_REG4__pipe5_rtr__SHIFT 0xd
+#define IA_DEBUG_REG4__ia_vgt_prim_rtr_q_MASK 0x4000
+#define IA_DEBUG_REG4__ia_vgt_prim_rtr_q__SHIFT 0xe
+#define IA_DEBUG_REG4__ia_se1vgt_prim_rtr_q_MASK 0x8000
+#define IA_DEBUG_REG4__ia_se1vgt_prim_rtr_q__SHIFT 0xf
+#define IA_DEBUG_REG4__di_major_mode_p1_q_MASK 0x10000
+#define IA_DEBUG_REG4__di_major_mode_p1_q__SHIFT 0x10
+#define IA_DEBUG_REG4__gs_mode_p1_q_MASK 0xe0000
+#define IA_DEBUG_REG4__gs_mode_p1_q__SHIFT 0x11
+#define IA_DEBUG_REG4__di_event_flag_p1_q_MASK 0x100000
+#define IA_DEBUG_REG4__di_event_flag_p1_q__SHIFT 0x14
+#define IA_DEBUG_REG4__di_state_sel_p1_q_MASK 0xe00000
+#define IA_DEBUG_REG4__di_state_sel_p1_q__SHIFT 0x15
+#define IA_DEBUG_REG4__draw_opaq_en_p1_q_MASK 0x1000000
+#define IA_DEBUG_REG4__draw_opaq_en_p1_q__SHIFT 0x18
+#define IA_DEBUG_REG4__draw_opaq_active_q_MASK 0x2000000
+#define IA_DEBUG_REG4__draw_opaq_active_q__SHIFT 0x19
+#define IA_DEBUG_REG4__di_source_select_p1_q_MASK 0xc000000
+#define IA_DEBUG_REG4__di_source_select_p1_q__SHIFT 0x1a
+#define IA_DEBUG_REG4__ready_to_read_di_MASK 0x10000000
+#define IA_DEBUG_REG4__ready_to_read_di__SHIFT 0x1c
+#define IA_DEBUG_REG4__di_first_group_of_draw_q_MASK 0x20000000
+#define IA_DEBUG_REG4__di_first_group_of_draw_q__SHIFT 0x1d
+#define IA_DEBUG_REG4__last_shift_of_draw_MASK 0x40000000
+#define IA_DEBUG_REG4__last_shift_of_draw__SHIFT 0x1e
+#define IA_DEBUG_REG4__current_shift_is_vect1_q_MASK 0x80000000
+#define IA_DEBUG_REG4__current_shift_is_vect1_q__SHIFT 0x1f
+#define IA_DEBUG_REG5__di_index_counter_q_15_0_MASK 0xffff
+#define IA_DEBUG_REG5__di_index_counter_q_15_0__SHIFT 0x0
+#define IA_DEBUG_REG5__instanceid_13_0_MASK 0x3fff0000
+#define IA_DEBUG_REG5__instanceid_13_0__SHIFT 0x10
+#define IA_DEBUG_REG5__draw_input_fifo_full_MASK 0x40000000
+#define IA_DEBUG_REG5__draw_input_fifo_full__SHIFT 0x1e
+#define IA_DEBUG_REG5__draw_input_fifo_empty_MASK 0x80000000
+#define IA_DEBUG_REG5__draw_input_fifo_empty__SHIFT 0x1f
+#define IA_DEBUG_REG6__current_shift_q_MASK 0xf
+#define IA_DEBUG_REG6__current_shift_q__SHIFT 0x0
+#define IA_DEBUG_REG6__current_stride_pre_MASK 0xf0
+#define IA_DEBUG_REG6__current_stride_pre__SHIFT 0x4
+#define IA_DEBUG_REG6__current_stride_q_MASK 0x1f00
+#define IA_DEBUG_REG6__current_stride_q__SHIFT 0x8
+#define IA_DEBUG_REG6__first_group_partial_MASK 0x2000
+#define IA_DEBUG_REG6__first_group_partial__SHIFT 0xd
+#define IA_DEBUG_REG6__second_group_partial_MASK 0x4000
+#define IA_DEBUG_REG6__second_group_partial__SHIFT 0xe
+#define IA_DEBUG_REG6__curr_prim_partial_MASK 0x8000
+#define IA_DEBUG_REG6__curr_prim_partial__SHIFT 0xf
+#define IA_DEBUG_REG6__next_stride_q_MASK 0x1f0000
+#define IA_DEBUG_REG6__next_stride_q__SHIFT 0x10
+#define IA_DEBUG_REG6__next_group_partial_MASK 0x200000
+#define IA_DEBUG_REG6__next_group_partial__SHIFT 0x15
+#define IA_DEBUG_REG6__after_group_partial_MASK 0x400000
+#define IA_DEBUG_REG6__after_group_partial__SHIFT 0x16
+#define IA_DEBUG_REG6__extract_group_MASK 0x800000
+#define IA_DEBUG_REG6__extract_group__SHIFT 0x17
+#define IA_DEBUG_REG6__grp_shift_debug_data_MASK 0xff000000
+#define IA_DEBUG_REG6__grp_shift_debug_data__SHIFT 0x18
+#define IA_DEBUG_REG7__reset_indx_state_q_MASK 0xf
+#define IA_DEBUG_REG7__reset_indx_state_q__SHIFT 0x0
+#define IA_DEBUG_REG7__shift_vect_valid_p2_q_MASK 0xf0
+#define IA_DEBUG_REG7__shift_vect_valid_p2_q__SHIFT 0x4
+#define IA_DEBUG_REG7__shift_vect1_valid_p2_q_MASK 0xf00
+#define IA_DEBUG_REG7__shift_vect1_valid_p2_q__SHIFT 0x8
+#define IA_DEBUG_REG7__shift_vect0_reset_match_p2_q_MASK 0xf000
+#define IA_DEBUG_REG7__shift_vect0_reset_match_p2_q__SHIFT 0xc
+#define IA_DEBUG_REG7__shift_vect1_reset_match_p2_q_MASK 0xf0000
+#define IA_DEBUG_REG7__shift_vect1_reset_match_p2_q__SHIFT 0x10
+#define IA_DEBUG_REG7__num_indx_in_group_p2_q_MASK 0x700000
+#define IA_DEBUG_REG7__num_indx_in_group_p2_q__SHIFT 0x14
+#define IA_DEBUG_REG7__last_group_of_draw_p2_q_MASK 0x800000
+#define IA_DEBUG_REG7__last_group_of_draw_p2_q__SHIFT 0x17
+#define IA_DEBUG_REG7__shift_event_flag_p2_q_MASK 0x1000000
+#define IA_DEBUG_REG7__shift_event_flag_p2_q__SHIFT 0x18
+#define IA_DEBUG_REG7__indx_shift_is_one_p2_q_MASK 0x2000000
+#define IA_DEBUG_REG7__indx_shift_is_one_p2_q__SHIFT 0x19
+#define IA_DEBUG_REG7__indx_shift_is_two_p2_q_MASK 0x4000000
+#define IA_DEBUG_REG7__indx_shift_is_two_p2_q__SHIFT 0x1a
+#define IA_DEBUG_REG7__indx_stride_is_four_p2_q_MASK 0x8000000
+#define IA_DEBUG_REG7__indx_stride_is_four_p2_q__SHIFT 0x1b
+#define IA_DEBUG_REG7__shift_prim1_reset_p3_q_MASK 0x10000000
+#define IA_DEBUG_REG7__shift_prim1_reset_p3_q__SHIFT 0x1c
+#define IA_DEBUG_REG7__shift_prim1_partial_p3_q_MASK 0x20000000
+#define IA_DEBUG_REG7__shift_prim1_partial_p3_q__SHIFT 0x1d
+#define IA_DEBUG_REG7__shift_prim0_reset_p3_q_MASK 0x40000000
+#define IA_DEBUG_REG7__shift_prim0_reset_p3_q__SHIFT 0x1e
+#define IA_DEBUG_REG7__shift_prim0_partial_p3_q_MASK 0x80000000
+#define IA_DEBUG_REG7__shift_prim0_partial_p3_q__SHIFT 0x1f
+#define IA_DEBUG_REG8__di_prim_type_p1_q_MASK 0x1f
+#define IA_DEBUG_REG8__di_prim_type_p1_q__SHIFT 0x0
+#define IA_DEBUG_REG8__two_cycle_xfer_p1_q_MASK 0x20
+#define IA_DEBUG_REG8__two_cycle_xfer_p1_q__SHIFT 0x5
+#define IA_DEBUG_REG8__two_prim_input_p1_q_MASK 0x40
+#define IA_DEBUG_REG8__two_prim_input_p1_q__SHIFT 0x6
+#define IA_DEBUG_REG8__shift_vect_end_of_packet_p5_q_MASK 0x80
+#define IA_DEBUG_REG8__shift_vect_end_of_packet_p5_q__SHIFT 0x7
+#define IA_DEBUG_REG8__last_group_of_inst_p5_q_MASK 0x100
+#define IA_DEBUG_REG8__last_group_of_inst_p5_q__SHIFT 0x8
+#define IA_DEBUG_REG8__shift_prim1_null_flag_p5_q_MASK 0x200
+#define IA_DEBUG_REG8__shift_prim1_null_flag_p5_q__SHIFT 0x9
+#define IA_DEBUG_REG8__shift_prim0_null_flag_p5_q_MASK 0x400
+#define IA_DEBUG_REG8__shift_prim0_null_flag_p5_q__SHIFT 0xa
+#define IA_DEBUG_REG8__grp_continued_MASK 0x800
+#define IA_DEBUG_REG8__grp_continued__SHIFT 0xb
+#define IA_DEBUG_REG8__grp_state_sel_MASK 0x7000
+#define IA_DEBUG_REG8__grp_state_sel__SHIFT 0xc
+#define IA_DEBUG_REG8__grp_sub_prim_type_MASK 0x1f8000
+#define IA_DEBUG_REG8__grp_sub_prim_type__SHIFT 0xf
+#define IA_DEBUG_REG8__grp_output_path_MASK 0xe00000
+#define IA_DEBUG_REG8__grp_output_path__SHIFT 0x15
+#define IA_DEBUG_REG8__grp_null_primitive_MASK 0x1000000
+#define IA_DEBUG_REG8__grp_null_primitive__SHIFT 0x18
+#define IA_DEBUG_REG8__grp_eop_MASK 0x2000000
+#define IA_DEBUG_REG8__grp_eop__SHIFT 0x19
+#define IA_DEBUG_REG8__grp_eopg_MASK 0x4000000
+#define IA_DEBUG_REG8__grp_eopg__SHIFT 0x1a
+#define IA_DEBUG_REG8__grp_event_flag_MASK 0x8000000
+#define IA_DEBUG_REG8__grp_event_flag__SHIFT 0x1b
+#define IA_DEBUG_REG8__grp_components_valid_MASK 0xf0000000
+#define IA_DEBUG_REG8__grp_components_valid__SHIFT 0x1c
+#define IA_DEBUG_REG9__send_to_se1_p6_MASK 0x1
+#define IA_DEBUG_REG9__send_to_se1_p6__SHIFT 0x0
+#define IA_DEBUG_REG9__gfx_se_switch_p6_MASK 0x2
+#define IA_DEBUG_REG9__gfx_se_switch_p6__SHIFT 0x1
+#define IA_DEBUG_REG9__null_eoi_xfer_prim1_p6_MASK 0x4
+#define IA_DEBUG_REG9__null_eoi_xfer_prim1_p6__SHIFT 0x2
+#define IA_DEBUG_REG9__null_eoi_xfer_prim0_p6_MASK 0x8
+#define IA_DEBUG_REG9__null_eoi_xfer_prim0_p6__SHIFT 0x3
+#define IA_DEBUG_REG9__prim1_eoi_p6_MASK 0x10
+#define IA_DEBUG_REG9__prim1_eoi_p6__SHIFT 0x4
+#define IA_DEBUG_REG9__prim0_eoi_p6_MASK 0x20
+#define IA_DEBUG_REG9__prim0_eoi_p6__SHIFT 0x5
+#define IA_DEBUG_REG9__prim1_valid_eopg_p6_MASK 0x40
+#define IA_DEBUG_REG9__prim1_valid_eopg_p6__SHIFT 0x6
+#define IA_DEBUG_REG9__prim0_valid_eopg_p6_MASK 0x80
+#define IA_DEBUG_REG9__prim0_valid_eopg_p6__SHIFT 0x7
+#define IA_DEBUG_REG9__prim1_to_other_se_p6_MASK 0x100
+#define IA_DEBUG_REG9__prim1_to_other_se_p6__SHIFT 0x8
+#define IA_DEBUG_REG9__eopg_on_last_prim_p6_MASK 0x200
+#define IA_DEBUG_REG9__eopg_on_last_prim_p6__SHIFT 0x9
+#define IA_DEBUG_REG9__eopg_between_prims_p6_MASK 0x400
+#define IA_DEBUG_REG9__eopg_between_prims_p6__SHIFT 0xa
+#define IA_DEBUG_REG9__prim_count_eq_group_size_p6_MASK 0x800
+#define IA_DEBUG_REG9__prim_count_eq_group_size_p6__SHIFT 0xb
+#define IA_DEBUG_REG9__prim_count_gt_group_size_p6_MASK 0x1000
+#define IA_DEBUG_REG9__prim_count_gt_group_size_p6__SHIFT 0xc
+#define IA_DEBUG_REG9__two_prim_output_p5_q_MASK 0x2000
+#define IA_DEBUG_REG9__two_prim_output_p5_q__SHIFT 0xd
+#define IA_DEBUG_REG9__SPARE0_MASK 0x4000
+#define IA_DEBUG_REG9__SPARE0__SHIFT 0xe
+#define IA_DEBUG_REG9__SPARE1_MASK 0x8000
+#define IA_DEBUG_REG9__SPARE1__SHIFT 0xf
+#define IA_DEBUG_REG9__shift_vect_end_of_packet_p5_q_MASK 0x10000
+#define IA_DEBUG_REG9__shift_vect_end_of_packet_p5_q__SHIFT 0x10
+#define IA_DEBUG_REG9__prim1_xfer_p6_MASK 0x20000
+#define IA_DEBUG_REG9__prim1_xfer_p6__SHIFT 0x11
+#define IA_DEBUG_REG9__grp_se1_fifo_empty_MASK 0x40000
+#define IA_DEBUG_REG9__grp_se1_fifo_empty__SHIFT 0x12
+#define IA_DEBUG_REG9__grp_se1_fifo_full_MASK 0x80000
+#define IA_DEBUG_REG9__grp_se1_fifo_full__SHIFT 0x13
+#define IA_DEBUG_REG9__prim_counter_q_MASK 0xfff00000
+#define IA_DEBUG_REG9__prim_counter_q__SHIFT 0x14
+#define VGT_DEBUG_REG0__vgt_busy_extended_MASK 0x1
+#define VGT_DEBUG_REG0__vgt_busy_extended__SHIFT 0x0
+#define VGT_DEBUG_REG0__SPARE9_MASK 0x2
+#define VGT_DEBUG_REG0__SPARE9__SHIFT 0x1
+#define VGT_DEBUG_REG0__vgt_busy_MASK 0x4
+#define VGT_DEBUG_REG0__vgt_busy__SHIFT 0x2
+#define VGT_DEBUG_REG0__SPARE8_MASK 0x8
+#define VGT_DEBUG_REG0__SPARE8__SHIFT 0x3
+#define VGT_DEBUG_REG0__SPARE7_MASK 0x10
+#define VGT_DEBUG_REG0__SPARE7__SHIFT 0x4
+#define VGT_DEBUG_REG0__SPARE6_MASK 0x20
+#define VGT_DEBUG_REG0__SPARE6__SHIFT 0x5
+#define VGT_DEBUG_REG0__SPARE5_MASK 0x40
+#define VGT_DEBUG_REG0__SPARE5__SHIFT 0x6
+#define VGT_DEBUG_REG0__SPARE4_MASK 0x80
+#define VGT_DEBUG_REG0__SPARE4__SHIFT 0x7
+#define VGT_DEBUG_REG0__pi_busy_MASK 0x100
+#define VGT_DEBUG_REG0__pi_busy__SHIFT 0x8
+#define VGT_DEBUG_REG0__vr_pi_busy_MASK 0x200
+#define VGT_DEBUG_REG0__vr_pi_busy__SHIFT 0x9
+#define VGT_DEBUG_REG0__pt_pi_busy_MASK 0x400
+#define VGT_DEBUG_REG0__pt_pi_busy__SHIFT 0xa
+#define VGT_DEBUG_REG0__te_pi_busy_MASK 0x800
+#define VGT_DEBUG_REG0__te_pi_busy__SHIFT 0xb
+#define VGT_DEBUG_REG0__gs_busy_MASK 0x1000
+#define VGT_DEBUG_REG0__gs_busy__SHIFT 0xc
+#define VGT_DEBUG_REG0__rcm_busy_MASK 0x2000
+#define VGT_DEBUG_REG0__rcm_busy__SHIFT 0xd
+#define VGT_DEBUG_REG0__tm_busy_MASK 0x4000
+#define VGT_DEBUG_REG0__tm_busy__SHIFT 0xe
+#define VGT_DEBUG_REG0__cm_busy_MASK 0x8000
+#define VGT_DEBUG_REG0__cm_busy__SHIFT 0xf
+#define VGT_DEBUG_REG0__gog_busy_MASK 0x10000
+#define VGT_DEBUG_REG0__gog_busy__SHIFT 0x10
+#define VGT_DEBUG_REG0__frmt_busy_MASK 0x20000
+#define VGT_DEBUG_REG0__frmt_busy__SHIFT 0x11
+#define VGT_DEBUG_REG0__SPARE10_MASK 0x40000
+#define VGT_DEBUG_REG0__SPARE10__SHIFT 0x12
+#define VGT_DEBUG_REG0__te11_pi_busy_MASK 0x80000
+#define VGT_DEBUG_REG0__te11_pi_busy__SHIFT 0x13
+#define VGT_DEBUG_REG0__SPARE3_MASK 0x100000
+#define VGT_DEBUG_REG0__SPARE3__SHIFT 0x14
+#define VGT_DEBUG_REG0__combined_out_busy_MASK 0x200000
+#define VGT_DEBUG_REG0__combined_out_busy__SHIFT 0x15
+#define VGT_DEBUG_REG0__spi_vs_interfaces_busy_MASK 0x400000
+#define VGT_DEBUG_REG0__spi_vs_interfaces_busy__SHIFT 0x16
+#define VGT_DEBUG_REG0__pa_interfaces_busy_MASK 0x800000
+#define VGT_DEBUG_REG0__pa_interfaces_busy__SHIFT 0x17
+#define VGT_DEBUG_REG0__reg_clk_busy_MASK 0x1000000
+#define VGT_DEBUG_REG0__reg_clk_busy__SHIFT 0x18
+#define VGT_DEBUG_REG0__SPARE2_MASK 0x2000000
+#define VGT_DEBUG_REG0__SPARE2__SHIFT 0x19
+#define VGT_DEBUG_REG0__core_clk_busy_MASK 0x4000000
+#define VGT_DEBUG_REG0__core_clk_busy__SHIFT 0x1a
+#define VGT_DEBUG_REG0__gs_clk_busy_MASK 0x8000000
+#define VGT_DEBUG_REG0__gs_clk_busy__SHIFT 0x1b
+#define VGT_DEBUG_REG0__SPARE1_MASK 0x10000000
+#define VGT_DEBUG_REG0__SPARE1__SHIFT 0x1c
+#define VGT_DEBUG_REG0__sclk_core_vld_MASK 0x20000000
+#define VGT_DEBUG_REG0__sclk_core_vld__SHIFT 0x1d
+#define VGT_DEBUG_REG0__sclk_gs_vld_MASK 0x40000000
+#define VGT_DEBUG_REG0__sclk_gs_vld__SHIFT 0x1e
+#define VGT_DEBUG_REG0__SPARE0_MASK 0x80000000
+#define VGT_DEBUG_REG0__SPARE0__SHIFT 0x1f
+#define VGT_DEBUG_REG1__SPARE9_MASK 0x1
+#define VGT_DEBUG_REG1__SPARE9__SHIFT 0x0
+#define VGT_DEBUG_REG1__SPARE8_MASK 0x2
+#define VGT_DEBUG_REG1__SPARE8__SHIFT 0x1
+#define VGT_DEBUG_REG1__SPARE7_MASK 0x4
+#define VGT_DEBUG_REG1__SPARE7__SHIFT 0x2
+#define VGT_DEBUG_REG1__SPARE6_MASK 0x8
+#define VGT_DEBUG_REG1__SPARE6__SHIFT 0x3
+#define VGT_DEBUG_REG1__SPARE5_MASK 0x10
+#define VGT_DEBUG_REG1__SPARE5__SHIFT 0x4
+#define VGT_DEBUG_REG1__SPARE4_MASK 0x20
+#define VGT_DEBUG_REG1__SPARE4__SHIFT 0x5
+#define VGT_DEBUG_REG1__SPARE3_MASK 0x40
+#define VGT_DEBUG_REG1__SPARE3__SHIFT 0x6
+#define VGT_DEBUG_REG1__SPARE2_MASK 0x80
+#define VGT_DEBUG_REG1__SPARE2__SHIFT 0x7
+#define VGT_DEBUG_REG1__SPARE1_MASK 0x100
+#define VGT_DEBUG_REG1__SPARE1__SHIFT 0x8
+#define VGT_DEBUG_REG1__SPARE0_MASK 0x200
+#define VGT_DEBUG_REG1__SPARE0__SHIFT 0x9
+#define VGT_DEBUG_REG1__pi_vr_valid_MASK 0x400
+#define VGT_DEBUG_REG1__pi_vr_valid__SHIFT 0xa
+#define VGT_DEBUG_REG1__vr_pi_read_MASK 0x800
+#define VGT_DEBUG_REG1__vr_pi_read__SHIFT 0xb
+#define VGT_DEBUG_REG1__pi_pt_valid_MASK 0x1000
+#define VGT_DEBUG_REG1__pi_pt_valid__SHIFT 0xc
+#define VGT_DEBUG_REG1__pt_pi_read_MASK 0x2000
+#define VGT_DEBUG_REG1__pt_pi_read__SHIFT 0xd
+#define VGT_DEBUG_REG1__pi_te_valid_MASK 0x4000
+#define VGT_DEBUG_REG1__pi_te_valid__SHIFT 0xe
+#define VGT_DEBUG_REG1__te_grp_read_MASK 0x8000
+#define VGT_DEBUG_REG1__te_grp_read__SHIFT 0xf
+#define VGT_DEBUG_REG1__vr_out_indx_valid_MASK 0x10000
+#define VGT_DEBUG_REG1__vr_out_indx_valid__SHIFT 0x10
+#define VGT_DEBUG_REG1__SPARE12_MASK 0x20000
+#define VGT_DEBUG_REG1__SPARE12__SHIFT 0x11
+#define VGT_DEBUG_REG1__vr_out_prim_valid_MASK 0x40000
+#define VGT_DEBUG_REG1__vr_out_prim_valid__SHIFT 0x12
+#define VGT_DEBUG_REG1__SPARE11_MASK 0x80000
+#define VGT_DEBUG_REG1__SPARE11__SHIFT 0x13
+#define VGT_DEBUG_REG1__pt_out_indx_valid_MASK 0x100000
+#define VGT_DEBUG_REG1__pt_out_indx_valid__SHIFT 0x14
+#define VGT_DEBUG_REG1__SPARE10_MASK 0x200000
+#define VGT_DEBUG_REG1__SPARE10__SHIFT 0x15
+#define VGT_DEBUG_REG1__pt_out_prim_valid_MASK 0x400000
+#define VGT_DEBUG_REG1__pt_out_prim_valid__SHIFT 0x16
+#define VGT_DEBUG_REG1__SPARE23_MASK 0x800000
+#define VGT_DEBUG_REG1__SPARE23__SHIFT 0x17
+#define VGT_DEBUG_REG1__te_out_data_valid_MASK 0x1000000
+#define VGT_DEBUG_REG1__te_out_data_valid__SHIFT 0x18
+#define VGT_DEBUG_REG1__SPARE25_MASK 0x2000000
+#define VGT_DEBUG_REG1__SPARE25__SHIFT 0x19
+#define VGT_DEBUG_REG1__pi_gs_valid_MASK 0x4000000
+#define VGT_DEBUG_REG1__pi_gs_valid__SHIFT 0x1a
+#define VGT_DEBUG_REG1__gs_pi_read_MASK 0x8000000
+#define VGT_DEBUG_REG1__gs_pi_read__SHIFT 0x1b
+#define VGT_DEBUG_REG1__gog_out_indx_valid_MASK 0x10000000
+#define VGT_DEBUG_REG1__gog_out_indx_valid__SHIFT 0x1c
+#define VGT_DEBUG_REG1__out_indx_read_MASK 0x20000000
+#define VGT_DEBUG_REG1__out_indx_read__SHIFT 0x1d
+#define VGT_DEBUG_REG1__gog_out_prim_valid_MASK 0x40000000
+#define VGT_DEBUG_REG1__gog_out_prim_valid__SHIFT 0x1e
+#define VGT_DEBUG_REG1__out_prim_read_MASK 0x80000000
+#define VGT_DEBUG_REG1__out_prim_read__SHIFT 0x1f
+#define VGT_DEBUG_REG2__hs_grp_busy_MASK 0x1
+#define VGT_DEBUG_REG2__hs_grp_busy__SHIFT 0x0
+#define VGT_DEBUG_REG2__hs_noif_busy_MASK 0x2
+#define VGT_DEBUG_REG2__hs_noif_busy__SHIFT 0x1
+#define VGT_DEBUG_REG2__tfmmIsBusy_MASK 0x4
+#define VGT_DEBUG_REG2__tfmmIsBusy__SHIFT 0x2
+#define VGT_DEBUG_REG2__lsVertIfBusy_0_MASK 0x8
+#define VGT_DEBUG_REG2__lsVertIfBusy_0__SHIFT 0x3
+#define VGT_DEBUG_REG2__te11_hs_tess_input_rtr_MASK 0x10
+#define VGT_DEBUG_REG2__te11_hs_tess_input_rtr__SHIFT 0x4
+#define VGT_DEBUG_REG2__lsWaveIfBusy_0_MASK 0x20
+#define VGT_DEBUG_REG2__lsWaveIfBusy_0__SHIFT 0x5
+#define VGT_DEBUG_REG2__hs_te11_tess_input_rts_MASK 0x40
+#define VGT_DEBUG_REG2__hs_te11_tess_input_rts__SHIFT 0x6
+#define VGT_DEBUG_REG2__grpModBusy_MASK 0x80
+#define VGT_DEBUG_REG2__grpModBusy__SHIFT 0x7
+#define VGT_DEBUG_REG2__lsVertFifoEmpty_MASK 0x100
+#define VGT_DEBUG_REG2__lsVertFifoEmpty__SHIFT 0x8
+#define VGT_DEBUG_REG2__lsWaveFifoEmpty_MASK 0x200
+#define VGT_DEBUG_REG2__lsWaveFifoEmpty__SHIFT 0x9
+#define VGT_DEBUG_REG2__hsVertFifoEmpty_MASK 0x400
+#define VGT_DEBUG_REG2__hsVertFifoEmpty__SHIFT 0xa
+#define VGT_DEBUG_REG2__hsWaveFifoEmpty_MASK 0x800
+#define VGT_DEBUG_REG2__hsWaveFifoEmpty__SHIFT 0xb
+#define VGT_DEBUG_REG2__hsInputFifoEmpty_MASK 0x1000
+#define VGT_DEBUG_REG2__hsInputFifoEmpty__SHIFT 0xc
+#define VGT_DEBUG_REG2__hsTifFifoEmpty_MASK 0x2000
+#define VGT_DEBUG_REG2__hsTifFifoEmpty__SHIFT 0xd
+#define VGT_DEBUG_REG2__lsVertFifoFull_MASK 0x4000
+#define VGT_DEBUG_REG2__lsVertFifoFull__SHIFT 0xe
+#define VGT_DEBUG_REG2__lsWaveFifoFull_MASK 0x8000
+#define VGT_DEBUG_REG2__lsWaveFifoFull__SHIFT 0xf
+#define VGT_DEBUG_REG2__hsVertFifoFull_MASK 0x10000
+#define VGT_DEBUG_REG2__hsVertFifoFull__SHIFT 0x10
+#define VGT_DEBUG_REG2__hsWaveFifoFull_MASK 0x20000
+#define VGT_DEBUG_REG2__hsWaveFifoFull__SHIFT 0x11
+#define VGT_DEBUG_REG2__hsInputFifoFull_MASK 0x40000
+#define VGT_DEBUG_REG2__hsInputFifoFull__SHIFT 0x12
+#define VGT_DEBUG_REG2__hsTifFifoFull_MASK 0x80000
+#define VGT_DEBUG_REG2__hsTifFifoFull__SHIFT 0x13
+#define VGT_DEBUG_REG2__p0_rtr_MASK 0x100000
+#define VGT_DEBUG_REG2__p0_rtr__SHIFT 0x14
+#define VGT_DEBUG_REG2__p1_rtr_MASK 0x200000
+#define VGT_DEBUG_REG2__p1_rtr__SHIFT 0x15
+#define VGT_DEBUG_REG2__p0_dr_MASK 0x400000
+#define VGT_DEBUG_REG2__p0_dr__SHIFT 0x16
+#define VGT_DEBUG_REG2__p1_dr_MASK 0x800000
+#define VGT_DEBUG_REG2__p1_dr__SHIFT 0x17
+#define VGT_DEBUG_REG2__p0_rts_MASK 0x1000000
+#define VGT_DEBUG_REG2__p0_rts__SHIFT 0x18
+#define VGT_DEBUG_REG2__p1_rts_MASK 0x2000000
+#define VGT_DEBUG_REG2__p1_rts__SHIFT 0x19
+#define VGT_DEBUG_REG2__ls_sh_id_MASK 0x4000000
+#define VGT_DEBUG_REG2__ls_sh_id__SHIFT 0x1a
+#define VGT_DEBUG_REG2__lsFwaveFlag_MASK 0x8000000
+#define VGT_DEBUG_REG2__lsFwaveFlag__SHIFT 0x1b
+#define VGT_DEBUG_REG2__lsWaveSendFlush_MASK 0x10000000
+#define VGT_DEBUG_REG2__lsWaveSendFlush__SHIFT 0x1c
+#define VGT_DEBUG_REG2__SPARE_MASK 0xe0000000
+#define VGT_DEBUG_REG2__SPARE__SHIFT 0x1d
+#define VGT_DEBUG_REG3__lsTgRelInd_MASK 0xfff
+#define VGT_DEBUG_REG3__lsTgRelInd__SHIFT 0x0
+#define VGT_DEBUG_REG3__lsWaveRelInd_MASK 0x3f000
+#define VGT_DEBUG_REG3__lsWaveRelInd__SHIFT 0xc
+#define VGT_DEBUG_REG3__lsPatchCnt_MASK 0x3fc0000
+#define VGT_DEBUG_REG3__lsPatchCnt__SHIFT 0x12
+#define VGT_DEBUG_REG3__hsWaveRelInd_MASK 0xfc000000
+#define VGT_DEBUG_REG3__hsWaveRelInd__SHIFT 0x1a
+#define VGT_DEBUG_REG4__hsPatchCnt_MASK 0xff
+#define VGT_DEBUG_REG4__hsPatchCnt__SHIFT 0x0
+#define VGT_DEBUG_REG4__hsPrimId_15_0_MASK 0xffff00
+#define VGT_DEBUG_REG4__hsPrimId_15_0__SHIFT 0x8
+#define VGT_DEBUG_REG4__hsCpCnt_MASK 0x1f000000
+#define VGT_DEBUG_REG4__hsCpCnt__SHIFT 0x18
+#define VGT_DEBUG_REG4__hsWaveSendFlush_MASK 0x20000000
+#define VGT_DEBUG_REG4__hsWaveSendFlush__SHIFT 0x1d
+#define VGT_DEBUG_REG4__hsFwaveFlag_MASK 0x40000000
+#define VGT_DEBUG_REG4__hsFwaveFlag__SHIFT 0x1e
+#define VGT_DEBUG_REG4__SPARE_MASK 0x80000000
+#define VGT_DEBUG_REG4__SPARE__SHIFT 0x1f
+#define VGT_DEBUG_REG5__SPARE4_MASK 0x7
+#define VGT_DEBUG_REG5__SPARE4__SHIFT 0x0
+#define VGT_DEBUG_REG5__hsWaveCreditCnt_0_MASK 0xf8
+#define VGT_DEBUG_REG5__hsWaveCreditCnt_0__SHIFT 0x3
+#define VGT_DEBUG_REG5__SPARE3_MASK 0x700
+#define VGT_DEBUG_REG5__SPARE3__SHIFT 0x8
+#define VGT_DEBUG_REG5__hsVertCreditCnt_0_MASK 0xf800
+#define VGT_DEBUG_REG5__hsVertCreditCnt_0__SHIFT 0xb
+#define VGT_DEBUG_REG5__SPARE2_MASK 0x70000
+#define VGT_DEBUG_REG5__SPARE2__SHIFT 0x10
+#define VGT_DEBUG_REG5__lsWaveCreditCnt_0_MASK 0xf80000
+#define VGT_DEBUG_REG5__lsWaveCreditCnt_0__SHIFT 0x13
+#define VGT_DEBUG_REG5__SPARE1_MASK 0x7000000
+#define VGT_DEBUG_REG5__SPARE1__SHIFT 0x18
+#define VGT_DEBUG_REG5__lsVertCreditCnt_0_MASK 0xf8000000
+#define VGT_DEBUG_REG5__lsVertCreditCnt_0__SHIFT 0x1b
+#define VGT_DEBUG_REG6__debug_BASE_MASK 0xffff
+#define VGT_DEBUG_REG6__debug_BASE__SHIFT 0x0
+#define VGT_DEBUG_REG6__debug_SIZE_MASK 0xffff0000
+#define VGT_DEBUG_REG6__debug_SIZE__SHIFT 0x10
+#define VGT_DEBUG_REG7__debug_tfmmFifoEmpty_MASK 0x1
+#define VGT_DEBUG_REG7__debug_tfmmFifoEmpty__SHIFT 0x0
+#define VGT_DEBUG_REG7__debug_tfmmFifoFull_MASK 0x2
+#define VGT_DEBUG_REG7__debug_tfmmFifoFull__SHIFT 0x1
+#define VGT_DEBUG_REG7__hs_pipe0_dr_MASK 0x4
+#define VGT_DEBUG_REG7__hs_pipe0_dr__SHIFT 0x2
+#define VGT_DEBUG_REG7__hs_pipe0_rtr_MASK 0x8
+#define VGT_DEBUG_REG7__hs_pipe0_rtr__SHIFT 0x3
+#define VGT_DEBUG_REG7__hs_pipe1_rtr_MASK 0x10
+#define VGT_DEBUG_REG7__hs_pipe1_rtr__SHIFT 0x4
+#define VGT_DEBUG_REG7__SPARE_MASK 0xffe0
+#define VGT_DEBUG_REG7__SPARE__SHIFT 0x5
+#define VGT_DEBUG_REG7__TF_addr_MASK 0xffff0000
+#define VGT_DEBUG_REG7__TF_addr__SHIFT 0x10
+#define VGT_DEBUG_REG8__rcm_busy_q_MASK 0x1
+#define VGT_DEBUG_REG8__rcm_busy_q__SHIFT 0x0
+#define VGT_DEBUG_REG8__rcm_noif_busy_q_MASK 0x2
+#define VGT_DEBUG_REG8__rcm_noif_busy_q__SHIFT 0x1
+#define VGT_DEBUG_REG8__r1_inst_rtr_MASK 0x4
+#define VGT_DEBUG_REG8__r1_inst_rtr__SHIFT 0x2
+#define VGT_DEBUG_REG8__spi_gsprim_fifo_busy_q_MASK 0x8
+#define VGT_DEBUG_REG8__spi_gsprim_fifo_busy_q__SHIFT 0x3
+#define VGT_DEBUG_REG8__spi_esvert_fifo_busy_q_MASK 0x10
+#define VGT_DEBUG_REG8__spi_esvert_fifo_busy_q__SHIFT 0x4
+#define VGT_DEBUG_REG8__gs_tbl_valid_r3_q_MASK 0x20
+#define VGT_DEBUG_REG8__gs_tbl_valid_r3_q__SHIFT 0x5
+#define VGT_DEBUG_REG8__valid_r0_q_MASK 0x40
+#define VGT_DEBUG_REG8__valid_r0_q__SHIFT 0x6
+#define VGT_DEBUG_REG8__valid_r1_q_MASK 0x80
+#define VGT_DEBUG_REG8__valid_r1_q__SHIFT 0x7
+#define VGT_DEBUG_REG8__valid_r2_MASK 0x100
+#define VGT_DEBUG_REG8__valid_r2__SHIFT 0x8
+#define VGT_DEBUG_REG8__valid_r2_q_MASK 0x200
+#define VGT_DEBUG_REG8__valid_r2_q__SHIFT 0x9
+#define VGT_DEBUG_REG8__r0_rtr_MASK 0x400
+#define VGT_DEBUG_REG8__r0_rtr__SHIFT 0xa
+#define VGT_DEBUG_REG8__r1_rtr_MASK 0x800
+#define VGT_DEBUG_REG8__r1_rtr__SHIFT 0xb
+#define VGT_DEBUG_REG8__r2_indx_rtr_MASK 0x1000
+#define VGT_DEBUG_REG8__r2_indx_rtr__SHIFT 0xc
+#define VGT_DEBUG_REG8__r2_rtr_MASK 0x2000
+#define VGT_DEBUG_REG8__r2_rtr__SHIFT 0xd
+#define VGT_DEBUG_REG8__es_gs_rtr_MASK 0x4000
+#define VGT_DEBUG_REG8__es_gs_rtr__SHIFT 0xe
+#define VGT_DEBUG_REG8__gs_event_fifo_rtr_MASK 0x8000
+#define VGT_DEBUG_REG8__gs_event_fifo_rtr__SHIFT 0xf
+#define VGT_DEBUG_REG8__tm_rcm_gs_event_rtr_MASK 0x10000
+#define VGT_DEBUG_REG8__tm_rcm_gs_event_rtr__SHIFT 0x10
+#define VGT_DEBUG_REG8__gs_tbl_r3_rtr_MASK 0x20000
+#define VGT_DEBUG_REG8__gs_tbl_r3_rtr__SHIFT 0x11
+#define VGT_DEBUG_REG8__prim_skid_fifo_empty_MASK 0x40000
+#define VGT_DEBUG_REG8__prim_skid_fifo_empty__SHIFT 0x12
+#define VGT_DEBUG_REG8__VGT_SPI_gsprim_rtr_q_MASK 0x80000
+#define VGT_DEBUG_REG8__VGT_SPI_gsprim_rtr_q__SHIFT 0x13
+#define VGT_DEBUG_REG8__tm_rcm_gs_tbl_rtr_MASK 0x100000
+#define VGT_DEBUG_REG8__tm_rcm_gs_tbl_rtr__SHIFT 0x14
+#define VGT_DEBUG_REG8__tm_rcm_es_tbl_rtr_MASK 0x200000
+#define VGT_DEBUG_REG8__tm_rcm_es_tbl_rtr__SHIFT 0x15
+#define VGT_DEBUG_REG8__VGT_SPI_esvert_rtr_q_MASK 0x400000
+#define VGT_DEBUG_REG8__VGT_SPI_esvert_rtr_q__SHIFT 0x16
+#define VGT_DEBUG_REG8__r2_no_bp_rtr_MASK 0x800000
+#define VGT_DEBUG_REG8__r2_no_bp_rtr__SHIFT 0x17
+#define VGT_DEBUG_REG8__hold_for_es_flush_MASK 0x1000000
+#define VGT_DEBUG_REG8__hold_for_es_flush__SHIFT 0x18
+#define VGT_DEBUG_REG8__gs_event_fifo_empty_MASK 0x2000000
+#define VGT_DEBUG_REG8__gs_event_fifo_empty__SHIFT 0x19
+#define VGT_DEBUG_REG8__gsprim_buff_empty_q_MASK 0x4000000
+#define VGT_DEBUG_REG8__gsprim_buff_empty_q__SHIFT 0x1a
+#define VGT_DEBUG_REG8__gsprim_buff_full_q_MASK 0x8000000
+#define VGT_DEBUG_REG8__gsprim_buff_full_q__SHIFT 0x1b
+#define VGT_DEBUG_REG8__te_prim_fifo_empty_MASK 0x10000000
+#define VGT_DEBUG_REG8__te_prim_fifo_empty__SHIFT 0x1c
+#define VGT_DEBUG_REG8__te_prim_fifo_full_MASK 0x20000000
+#define VGT_DEBUG_REG8__te_prim_fifo_full__SHIFT 0x1d
+#define VGT_DEBUG_REG8__te_vert_fifo_empty_MASK 0x40000000
+#define VGT_DEBUG_REG8__te_vert_fifo_empty__SHIFT 0x1e
+#define VGT_DEBUG_REG8__te_vert_fifo_full_MASK 0x80000000
+#define VGT_DEBUG_REG8__te_vert_fifo_full__SHIFT 0x1f
+#define VGT_DEBUG_REG9__indices_to_send_r2_q_MASK 0x3
+#define VGT_DEBUG_REG9__indices_to_send_r2_q__SHIFT 0x0
+#define VGT_DEBUG_REG9__valid_indices_r3_MASK 0x4
+#define VGT_DEBUG_REG9__valid_indices_r3__SHIFT 0x2
+#define VGT_DEBUG_REG9__gs_eov_r3_MASK 0x8
+#define VGT_DEBUG_REG9__gs_eov_r3__SHIFT 0x3
+#define VGT_DEBUG_REG9__eop_indx_r3_MASK 0x10
+#define VGT_DEBUG_REG9__eop_indx_r3__SHIFT 0x4
+#define VGT_DEBUG_REG9__eop_prim_r3_MASK 0x20
+#define VGT_DEBUG_REG9__eop_prim_r3__SHIFT 0x5
+#define VGT_DEBUG_REG9__es_eov_r3_MASK 0x40
+#define VGT_DEBUG_REG9__es_eov_r3__SHIFT 0x6
+#define VGT_DEBUG_REG9__es_tbl_state_r3_q_0_MASK 0x80
+#define VGT_DEBUG_REG9__es_tbl_state_r3_q_0__SHIFT 0x7
+#define VGT_DEBUG_REG9__pending_es_send_r3_q_MASK 0x100
+#define VGT_DEBUG_REG9__pending_es_send_r3_q__SHIFT 0x8
+#define VGT_DEBUG_REG9__pending_es_flush_r3_MASK 0x200
+#define VGT_DEBUG_REG9__pending_es_flush_r3__SHIFT 0x9
+#define VGT_DEBUG_REG9__gs_tbl_num_es_per_gs_r3_q_not_0_MASK 0x400
+#define VGT_DEBUG_REG9__gs_tbl_num_es_per_gs_r3_q_not_0__SHIFT 0xa
+#define VGT_DEBUG_REG9__gs_tbl_prim_cnt_r3_q_MASK 0x3f800
+#define VGT_DEBUG_REG9__gs_tbl_prim_cnt_r3_q__SHIFT 0xb
+#define VGT_DEBUG_REG9__gs_tbl_eop_r3_q_MASK 0x40000
+#define VGT_DEBUG_REG9__gs_tbl_eop_r3_q__SHIFT 0x12
+#define VGT_DEBUG_REG9__gs_tbl_state_r3_q_MASK 0x380000
+#define VGT_DEBUG_REG9__gs_tbl_state_r3_q__SHIFT 0x13
+#define VGT_DEBUG_REG9__gs_pending_state_r3_q_MASK 0x400000
+#define VGT_DEBUG_REG9__gs_pending_state_r3_q__SHIFT 0x16
+#define VGT_DEBUG_REG9__invalidate_rb_roll_over_q_MASK 0x800000
+#define VGT_DEBUG_REG9__invalidate_rb_roll_over_q__SHIFT 0x17
+#define VGT_DEBUG_REG9__gs_instancing_state_q_MASK 0x1000000
+#define VGT_DEBUG_REG9__gs_instancing_state_q__SHIFT 0x18
+#define VGT_DEBUG_REG9__es_per_gs_vert_cnt_r3_q_not_0_MASK 0x2000000
+#define VGT_DEBUG_REG9__es_per_gs_vert_cnt_r3_q_not_0__SHIFT 0x19
+#define VGT_DEBUG_REG9__gs_prim_per_es_ctr_r3_q_not_0_MASK 0x4000000
+#define VGT_DEBUG_REG9__gs_prim_per_es_ctr_r3_q_not_0__SHIFT 0x1a
+#define VGT_DEBUG_REG9__pre_r0_rtr_MASK 0x8000000
+#define VGT_DEBUG_REG9__pre_r0_rtr__SHIFT 0x1b
+#define VGT_DEBUG_REG9__valid_r3_q_MASK 0x10000000
+#define VGT_DEBUG_REG9__valid_r3_q__SHIFT 0x1c
+#define VGT_DEBUG_REG9__valid_pre_r0_q_MASK 0x20000000
+#define VGT_DEBUG_REG9__valid_pre_r0_q__SHIFT 0x1d
+#define VGT_DEBUG_REG9__SPARE0_MASK 0x40000000
+#define VGT_DEBUG_REG9__SPARE0__SHIFT 0x1e
+#define VGT_DEBUG_REG9__off_chip_hs_r2_q_MASK 0x80000000
+#define VGT_DEBUG_REG9__off_chip_hs_r2_q__SHIFT 0x1f
+#define VGT_DEBUG_REG10__index_buffer_depth_r1_q_MASK 0x1f
+#define VGT_DEBUG_REG10__index_buffer_depth_r1_q__SHIFT 0x0
+#define VGT_DEBUG_REG10__eopg_r2_q_MASK 0x20
+#define VGT_DEBUG_REG10__eopg_r2_q__SHIFT 0x5
+#define VGT_DEBUG_REG10__eotg_r2_q_MASK 0x40
+#define VGT_DEBUG_REG10__eotg_r2_q__SHIFT 0x6
+#define VGT_DEBUG_REG10__onchip_gs_en_r0_q_MASK 0x180
+#define VGT_DEBUG_REG10__onchip_gs_en_r0_q__SHIFT 0x7
+#define VGT_DEBUG_REG10__SPARE2_MASK 0x600
+#define VGT_DEBUG_REG10__SPARE2__SHIFT 0x9
+#define VGT_DEBUG_REG10__rcm_mem_gsprim_re_qq_MASK 0x800
+#define VGT_DEBUG_REG10__rcm_mem_gsprim_re_qq__SHIFT 0xb
+#define VGT_DEBUG_REG10__rcm_mem_gsprim_re_q_MASK 0x1000
+#define VGT_DEBUG_REG10__rcm_mem_gsprim_re_q__SHIFT 0xc
+#define VGT_DEBUG_REG10__gs_rb_space_avail_r3_q_9_0_MASK 0x7fe000
+#define VGT_DEBUG_REG10__gs_rb_space_avail_r3_q_9_0__SHIFT 0xd
+#define VGT_DEBUG_REG10__es_rb_space_avail_r2_q_8_0_MASK 0xff800000
+#define VGT_DEBUG_REG10__es_rb_space_avail_r2_q_8_0__SHIFT 0x17
+#define VGT_DEBUG_REG11__tm_busy_q_MASK 0x1
+#define VGT_DEBUG_REG11__tm_busy_q__SHIFT 0x0
+#define VGT_DEBUG_REG11__tm_noif_busy_q_MASK 0x2
+#define VGT_DEBUG_REG11__tm_noif_busy_q__SHIFT 0x1
+#define VGT_DEBUG_REG11__tm_out_busy_q_MASK 0x4
+#define VGT_DEBUG_REG11__tm_out_busy_q__SHIFT 0x2
+#define VGT_DEBUG_REG11__es_rb_dealloc_fifo_busy_MASK 0x8
+#define VGT_DEBUG_REG11__es_rb_dealloc_fifo_busy__SHIFT 0x3
+#define VGT_DEBUG_REG11__vs_dealloc_tbl_busy_MASK 0x10
+#define VGT_DEBUG_REG11__vs_dealloc_tbl_busy__SHIFT 0x4
+#define VGT_DEBUG_REG11__SPARE1_MASK 0x20
+#define VGT_DEBUG_REG11__SPARE1__SHIFT 0x5
+#define VGT_DEBUG_REG11__spi_gsthread_fifo_busy_MASK 0x40
+#define VGT_DEBUG_REG11__spi_gsthread_fifo_busy__SHIFT 0x6
+#define VGT_DEBUG_REG11__spi_esthread_fifo_busy_MASK 0x80
+#define VGT_DEBUG_REG11__spi_esthread_fifo_busy__SHIFT 0x7
+#define VGT_DEBUG_REG11__hold_eswave_MASK 0x100
+#define VGT_DEBUG_REG11__hold_eswave__SHIFT 0x8
+#define VGT_DEBUG_REG11__es_rb_roll_over_r3_MASK 0x200
+#define VGT_DEBUG_REG11__es_rb_roll_over_r3__SHIFT 0x9
+#define VGT_DEBUG_REG11__counters_busy_r0_MASK 0x400
+#define VGT_DEBUG_REG11__counters_busy_r0__SHIFT 0xa
+#define VGT_DEBUG_REG11__counters_avail_r0_MASK 0x800
+#define VGT_DEBUG_REG11__counters_avail_r0__SHIFT 0xb
+#define VGT_DEBUG_REG11__counters_available_r0_MASK 0x1000
+#define VGT_DEBUG_REG11__counters_available_r0__SHIFT 0xc
+#define VGT_DEBUG_REG11__vs_event_fifo_rtr_MASK 0x2000
+#define VGT_DEBUG_REG11__vs_event_fifo_rtr__SHIFT 0xd
+#define VGT_DEBUG_REG11__VGT_SPI_gsthread_rtr_q_MASK 0x4000
+#define VGT_DEBUG_REG11__VGT_SPI_gsthread_rtr_q__SHIFT 0xe
+#define VGT_DEBUG_REG11__VGT_SPI_esthread_rtr_q_MASK 0x8000
+#define VGT_DEBUG_REG11__VGT_SPI_esthread_rtr_q__SHIFT 0xf
+#define VGT_DEBUG_REG11__gs_issue_rtr_MASK 0x10000
+#define VGT_DEBUG_REG11__gs_issue_rtr__SHIFT 0x10
+#define VGT_DEBUG_REG11__tm_pt_event_rtr_MASK 0x20000
+#define VGT_DEBUG_REG11__tm_pt_event_rtr__SHIFT 0x11
+#define VGT_DEBUG_REG11__SPARE0_MASK 0x40000
+#define VGT_DEBUG_REG11__SPARE0__SHIFT 0x12
+#define VGT_DEBUG_REG11__gs_r0_rtr_MASK 0x80000
+#define VGT_DEBUG_REG11__gs_r0_rtr__SHIFT 0x13
+#define VGT_DEBUG_REG11__es_r0_rtr_MASK 0x100000
+#define VGT_DEBUG_REG11__es_r0_rtr__SHIFT 0x14
+#define VGT_DEBUG_REG11__gog_tm_vs_event_rtr_MASK 0x200000
+#define VGT_DEBUG_REG11__gog_tm_vs_event_rtr__SHIFT 0x15
+#define VGT_DEBUG_REG11__tm_rcm_gs_event_rtr_MASK 0x400000
+#define VGT_DEBUG_REG11__tm_rcm_gs_event_rtr__SHIFT 0x16
+#define VGT_DEBUG_REG11__tm_rcm_gs_tbl_rtr_MASK 0x800000
+#define VGT_DEBUG_REG11__tm_rcm_gs_tbl_rtr__SHIFT 0x17
+#define VGT_DEBUG_REG11__tm_rcm_es_tbl_rtr_MASK 0x1000000
+#define VGT_DEBUG_REG11__tm_rcm_es_tbl_rtr__SHIFT 0x18
+#define VGT_DEBUG_REG11__vs_event_fifo_empty_MASK 0x2000000
+#define VGT_DEBUG_REG11__vs_event_fifo_empty__SHIFT 0x19
+#define VGT_DEBUG_REG11__vs_event_fifo_full_MASK 0x4000000
+#define VGT_DEBUG_REG11__vs_event_fifo_full__SHIFT 0x1a
+#define VGT_DEBUG_REG11__es_rb_dealloc_fifo_full_MASK 0x8000000
+#define VGT_DEBUG_REG11__es_rb_dealloc_fifo_full__SHIFT 0x1b
+#define VGT_DEBUG_REG11__vs_dealloc_tbl_full_MASK 0x10000000
+#define VGT_DEBUG_REG11__vs_dealloc_tbl_full__SHIFT 0x1c
+#define VGT_DEBUG_REG11__send_event_q_MASK 0x20000000
+#define VGT_DEBUG_REG11__send_event_q__SHIFT 0x1d
+#define VGT_DEBUG_REG11__es_tbl_empty_MASK 0x40000000
+#define VGT_DEBUG_REG11__es_tbl_empty__SHIFT 0x1e
+#define VGT_DEBUG_REG11__no_active_states_r0_MASK 0x80000000
+#define VGT_DEBUG_REG11__no_active_states_r0__SHIFT 0x1f
+#define VGT_DEBUG_REG12__gs_state0_r0_q_MASK 0x7
+#define VGT_DEBUG_REG12__gs_state0_r0_q__SHIFT 0x0
+#define VGT_DEBUG_REG12__gs_state1_r0_q_MASK 0x38
+#define VGT_DEBUG_REG12__gs_state1_r0_q__SHIFT 0x3
+#define VGT_DEBUG_REG12__gs_state2_r0_q_MASK 0x1c0
+#define VGT_DEBUG_REG12__gs_state2_r0_q__SHIFT 0x6
+#define VGT_DEBUG_REG12__gs_state3_r0_q_MASK 0xe00
+#define VGT_DEBUG_REG12__gs_state3_r0_q__SHIFT 0x9
+#define VGT_DEBUG_REG12__gs_state4_r0_q_MASK 0x7000
+#define VGT_DEBUG_REG12__gs_state4_r0_q__SHIFT 0xc
+#define VGT_DEBUG_REG12__gs_state5_r0_q_MASK 0x38000
+#define VGT_DEBUG_REG12__gs_state5_r0_q__SHIFT 0xf
+#define VGT_DEBUG_REG12__gs_state6_r0_q_MASK 0x1c0000
+#define VGT_DEBUG_REG12__gs_state6_r0_q__SHIFT 0x12
+#define VGT_DEBUG_REG12__gs_state7_r0_q_MASK 0xe00000
+#define VGT_DEBUG_REG12__gs_state7_r0_q__SHIFT 0x15
+#define VGT_DEBUG_REG12__gs_state8_r0_q_MASK 0x7000000
+#define VGT_DEBUG_REG12__gs_state8_r0_q__SHIFT 0x18
+#define VGT_DEBUG_REG12__gs_state9_r0_q_MASK 0x38000000
+#define VGT_DEBUG_REG12__gs_state9_r0_q__SHIFT 0x1b
+#define VGT_DEBUG_REG12__hold_eswave_eop_MASK 0x40000000
+#define VGT_DEBUG_REG12__hold_eswave_eop__SHIFT 0x1e
+#define VGT_DEBUG_REG12__SPARE0_MASK 0x80000000
+#define VGT_DEBUG_REG12__SPARE0__SHIFT 0x1f
+#define VGT_DEBUG_REG13__gs_state10_r0_q_MASK 0x7
+#define VGT_DEBUG_REG13__gs_state10_r0_q__SHIFT 0x0
+#define VGT_DEBUG_REG13__gs_state11_r0_q_MASK 0x38
+#define VGT_DEBUG_REG13__gs_state11_r0_q__SHIFT 0x3
+#define VGT_DEBUG_REG13__gs_state12_r0_q_MASK 0x1c0
+#define VGT_DEBUG_REG13__gs_state12_r0_q__SHIFT 0x6
+#define VGT_DEBUG_REG13__gs_state13_r0_q_MASK 0xe00
+#define VGT_DEBUG_REG13__gs_state13_r0_q__SHIFT 0x9
+#define VGT_DEBUG_REG13__gs_state14_r0_q_MASK 0x7000
+#define VGT_DEBUG_REG13__gs_state14_r0_q__SHIFT 0xc
+#define VGT_DEBUG_REG13__gs_state15_r0_q_MASK 0x38000
+#define VGT_DEBUG_REG13__gs_state15_r0_q__SHIFT 0xf
+#define VGT_DEBUG_REG13__gs_tbl_wrptr_r0_q_3_0_MASK 0x3c0000
+#define VGT_DEBUG_REG13__gs_tbl_wrptr_r0_q_3_0__SHIFT 0x12
+#define VGT_DEBUG_REG13__gsfetch_done_fifo_cnt_q_not_0_MASK 0x400000
+#define VGT_DEBUG_REG13__gsfetch_done_fifo_cnt_q_not_0__SHIFT 0x16
+#define VGT_DEBUG_REG13__gsfetch_done_cnt_q_not_0_MASK 0x800000
+#define VGT_DEBUG_REG13__gsfetch_done_cnt_q_not_0__SHIFT 0x17
+#define VGT_DEBUG_REG13__es_tbl_full_MASK 0x1000000
+#define VGT_DEBUG_REG13__es_tbl_full__SHIFT 0x18
+#define VGT_DEBUG_REG13__SPARE1_MASK 0x2000000
+#define VGT_DEBUG_REG13__SPARE1__SHIFT 0x19
+#define VGT_DEBUG_REG13__SPARE0_MASK 0x4000000
+#define VGT_DEBUG_REG13__SPARE0__SHIFT 0x1a
+#define VGT_DEBUG_REG13__active_cm_sm_r0_q_MASK 0xf8000000
+#define VGT_DEBUG_REG13__active_cm_sm_r0_q__SHIFT 0x1b
+#define VGT_DEBUG_REG14__SPARE3_MASK 0xf
+#define VGT_DEBUG_REG14__SPARE3__SHIFT 0x0
+#define VGT_DEBUG_REG14__gsfetch_done_fifo_full_MASK 0x10
+#define VGT_DEBUG_REG14__gsfetch_done_fifo_full__SHIFT 0x4
+#define VGT_DEBUG_REG14__gs_rb_space_avail_r0_MASK 0x20
+#define VGT_DEBUG_REG14__gs_rb_space_avail_r0__SHIFT 0x5
+#define VGT_DEBUG_REG14__smx_es_done_cnt_r0_q_not_0_MASK 0x40
+#define VGT_DEBUG_REG14__smx_es_done_cnt_r0_q_not_0__SHIFT 0x6
+#define VGT_DEBUG_REG14__SPARE8_MASK 0x180
+#define VGT_DEBUG_REG14__SPARE8__SHIFT 0x7
+#define VGT_DEBUG_REG14__vs_done_cnt_q_not_0_MASK 0x200
+#define VGT_DEBUG_REG14__vs_done_cnt_q_not_0__SHIFT 0x9
+#define VGT_DEBUG_REG14__es_flush_cnt_busy_q_MASK 0x400
+#define VGT_DEBUG_REG14__es_flush_cnt_busy_q__SHIFT 0xa
+#define VGT_DEBUG_REG14__gs_tbl_full_r0_MASK 0x800
+#define VGT_DEBUG_REG14__gs_tbl_full_r0__SHIFT 0xb
+#define VGT_DEBUG_REG14__SPARE2_MASK 0x1ff000
+#define VGT_DEBUG_REG14__SPARE2__SHIFT 0xc
+#define VGT_DEBUG_REG14__se1spi_gsthread_fifo_busy_MASK 0x200000
+#define VGT_DEBUG_REG14__se1spi_gsthread_fifo_busy__SHIFT 0x15
+#define VGT_DEBUG_REG14__SPARE_MASK 0x1c00000
+#define VGT_DEBUG_REG14__SPARE__SHIFT 0x16
+#define VGT_DEBUG_REG14__VGT_SE1SPI_gsthread_rtr_q_MASK 0x2000000
+#define VGT_DEBUG_REG14__VGT_SE1SPI_gsthread_rtr_q__SHIFT 0x19
+#define VGT_DEBUG_REG14__smx1_es_done_cnt_r0_q_not_0_MASK 0x4000000
+#define VGT_DEBUG_REG14__smx1_es_done_cnt_r0_q_not_0__SHIFT 0x1a
+#define VGT_DEBUG_REG14__se1spi_esthread_fifo_busy_MASK 0x8000000
+#define VGT_DEBUG_REG14__se1spi_esthread_fifo_busy__SHIFT 0x1b
+#define VGT_DEBUG_REG14__SPARE1_MASK 0x10000000
+#define VGT_DEBUG_REG14__SPARE1__SHIFT 0x1c
+#define VGT_DEBUG_REG14__gsfetch_done_se1_cnt_q_not_0_MASK 0x20000000
+#define VGT_DEBUG_REG14__gsfetch_done_se1_cnt_q_not_0__SHIFT 0x1d
+#define VGT_DEBUG_REG14__SPARE0_MASK 0x40000000
+#define VGT_DEBUG_REG14__SPARE0__SHIFT 0x1e
+#define VGT_DEBUG_REG14__VGT_SE1SPI_esthread_rtr_q_MASK 0x80000000
+#define VGT_DEBUG_REG14__VGT_SE1SPI_esthread_rtr_q__SHIFT 0x1f
+#define VGT_DEBUG_REG15__cm_busy_q_MASK 0x1
+#define VGT_DEBUG_REG15__cm_busy_q__SHIFT 0x0
+#define VGT_DEBUG_REG15__counters_busy_q_MASK 0x2
+#define VGT_DEBUG_REG15__counters_busy_q__SHIFT 0x1
+#define VGT_DEBUG_REG15__output_fifo_empty_MASK 0x4
+#define VGT_DEBUG_REG15__output_fifo_empty__SHIFT 0x2
+#define VGT_DEBUG_REG15__output_fifo_full_MASK 0x8
+#define VGT_DEBUG_REG15__output_fifo_full__SHIFT 0x3
+#define VGT_DEBUG_REG15__counters_full_MASK 0x10
+#define VGT_DEBUG_REG15__counters_full__SHIFT 0x4
+#define VGT_DEBUG_REG15__active_sm_q_MASK 0x3e0
+#define VGT_DEBUG_REG15__active_sm_q__SHIFT 0x5
+#define VGT_DEBUG_REG15__entry_rdptr_q_MASK 0x7c00
+#define VGT_DEBUG_REG15__entry_rdptr_q__SHIFT 0xa
+#define VGT_DEBUG_REG15__cntr_tbl_wrptr_q_MASK 0xf8000
+#define VGT_DEBUG_REG15__cntr_tbl_wrptr_q__SHIFT 0xf
+#define VGT_DEBUG_REG15__SPARE25_MASK 0x3f00000
+#define VGT_DEBUG_REG15__SPARE25__SHIFT 0x14
+#define VGT_DEBUG_REG15__st_cut_mode_q_MASK 0xc000000
+#define VGT_DEBUG_REG15__st_cut_mode_q__SHIFT 0x1a
+#define VGT_DEBUG_REG15__gs_done_array_q_not_0_MASK 0x10000000
+#define VGT_DEBUG_REG15__gs_done_array_q_not_0__SHIFT 0x1c
+#define VGT_DEBUG_REG15__SPARE31_MASK 0xe0000000
+#define VGT_DEBUG_REG15__SPARE31__SHIFT 0x1d
+#define VGT_DEBUG_REG16__gog_busy_MASK 0x1
+#define VGT_DEBUG_REG16__gog_busy__SHIFT 0x0
+#define VGT_DEBUG_REG16__gog_state_q_MASK 0xe
+#define VGT_DEBUG_REG16__gog_state_q__SHIFT 0x1
+#define VGT_DEBUG_REG16__r0_rtr_MASK 0x10
+#define VGT_DEBUG_REG16__r0_rtr__SHIFT 0x4
+#define VGT_DEBUG_REG16__r1_rtr_MASK 0x20
+#define VGT_DEBUG_REG16__r1_rtr__SHIFT 0x5
+#define VGT_DEBUG_REG16__r1_upstream_rtr_MASK 0x40
+#define VGT_DEBUG_REG16__r1_upstream_rtr__SHIFT 0x6
+#define VGT_DEBUG_REG16__r2_vs_tbl_rtr_MASK 0x80
+#define VGT_DEBUG_REG16__r2_vs_tbl_rtr__SHIFT 0x7
+#define VGT_DEBUG_REG16__r2_prim_rtr_MASK 0x100
+#define VGT_DEBUG_REG16__r2_prim_rtr__SHIFT 0x8
+#define VGT_DEBUG_REG16__r2_indx_rtr_MASK 0x200
+#define VGT_DEBUG_REG16__r2_indx_rtr__SHIFT 0x9
+#define VGT_DEBUG_REG16__r2_rtr_MASK 0x400
+#define VGT_DEBUG_REG16__r2_rtr__SHIFT 0xa
+#define VGT_DEBUG_REG16__gog_tm_vs_event_rtr_MASK 0x800
+#define VGT_DEBUG_REG16__gog_tm_vs_event_rtr__SHIFT 0xb
+#define VGT_DEBUG_REG16__r3_force_vs_tbl_we_rtr_MASK 0x1000
+#define VGT_DEBUG_REG16__r3_force_vs_tbl_we_rtr__SHIFT 0xc
+#define VGT_DEBUG_REG16__indx_valid_r2_q_MASK 0x2000
+#define VGT_DEBUG_REG16__indx_valid_r2_q__SHIFT 0xd
+#define VGT_DEBUG_REG16__prim_valid_r2_q_MASK 0x4000
+#define VGT_DEBUG_REG16__prim_valid_r2_q__SHIFT 0xe
+#define VGT_DEBUG_REG16__valid_r2_q_MASK 0x8000
+#define VGT_DEBUG_REG16__valid_r2_q__SHIFT 0xf
+#define VGT_DEBUG_REG16__prim_valid_r1_q_MASK 0x10000
+#define VGT_DEBUG_REG16__prim_valid_r1_q__SHIFT 0x10
+#define VGT_DEBUG_REG16__indx_valid_r1_q_MASK 0x20000
+#define VGT_DEBUG_REG16__indx_valid_r1_q__SHIFT 0x11
+#define VGT_DEBUG_REG16__valid_r1_q_MASK 0x40000
+#define VGT_DEBUG_REG16__valid_r1_q__SHIFT 0x12
+#define VGT_DEBUG_REG16__indx_valid_r0_q_MASK 0x80000
+#define VGT_DEBUG_REG16__indx_valid_r0_q__SHIFT 0x13
+#define VGT_DEBUG_REG16__prim_valid_r0_q_MASK 0x100000
+#define VGT_DEBUG_REG16__prim_valid_r0_q__SHIFT 0x14
+#define VGT_DEBUG_REG16__valid_r0_q_MASK 0x200000
+#define VGT_DEBUG_REG16__valid_r0_q__SHIFT 0x15
+#define VGT_DEBUG_REG16__send_event_q_MASK 0x400000
+#define VGT_DEBUG_REG16__send_event_q__SHIFT 0x16
+#define VGT_DEBUG_REG16__SPARE24_MASK 0x800000
+#define VGT_DEBUG_REG16__SPARE24__SHIFT 0x17
+#define VGT_DEBUG_REG16__vert_seen_since_sopg_r2_q_MASK 0x1000000
+#define VGT_DEBUG_REG16__vert_seen_since_sopg_r2_q__SHIFT 0x18
+#define VGT_DEBUG_REG16__gog_out_prim_state_sel_MASK 0xe000000
+#define VGT_DEBUG_REG16__gog_out_prim_state_sel__SHIFT 0x19
+#define VGT_DEBUG_REG16__multiple_streams_en_r1_q_MASK 0x10000000
+#define VGT_DEBUG_REG16__multiple_streams_en_r1_q__SHIFT 0x1c
+#define VGT_DEBUG_REG16__vs_vert_count_r2_q_not_0_MASK 0x20000000
+#define VGT_DEBUG_REG16__vs_vert_count_r2_q_not_0__SHIFT 0x1d
+#define VGT_DEBUG_REG16__num_gs_r2_q_not_0_MASK 0x40000000
+#define VGT_DEBUG_REG16__num_gs_r2_q_not_0__SHIFT 0x1e
+#define VGT_DEBUG_REG16__new_vs_thread_r2_MASK 0x80000000
+#define VGT_DEBUG_REG16__new_vs_thread_r2__SHIFT 0x1f
+#define VGT_DEBUG_REG17__gog_out_prim_rel_indx2_5_0_MASK 0x3f
+#define VGT_DEBUG_REG17__gog_out_prim_rel_indx2_5_0__SHIFT 0x0
+#define VGT_DEBUG_REG17__gog_out_prim_rel_indx1_5_0_MASK 0xfc0
+#define VGT_DEBUG_REG17__gog_out_prim_rel_indx1_5_0__SHIFT 0x6
+#define VGT_DEBUG_REG17__gog_out_prim_rel_indx0_5_0_MASK 0x3f000
+#define VGT_DEBUG_REG17__gog_out_prim_rel_indx0_5_0__SHIFT 0xc
+#define VGT_DEBUG_REG17__gog_out_indx_13_0_MASK 0xfffc0000
+#define VGT_DEBUG_REG17__gog_out_indx_13_0__SHIFT 0x12
+#define VGT_DEBUG_REG18__grp_vr_valid_MASK 0x1
+#define VGT_DEBUG_REG18__grp_vr_valid__SHIFT 0x0
+#define VGT_DEBUG_REG18__pipe0_dr_MASK 0x2
+#define VGT_DEBUG_REG18__pipe0_dr__SHIFT 0x1
+#define VGT_DEBUG_REG18__pipe1_dr_MASK 0x4
+#define VGT_DEBUG_REG18__pipe1_dr__SHIFT 0x2
+#define VGT_DEBUG_REG18__vr_grp_read_MASK 0x8
+#define VGT_DEBUG_REG18__vr_grp_read__SHIFT 0x3
+#define VGT_DEBUG_REG18__pipe0_rtr_MASK 0x10
+#define VGT_DEBUG_REG18__pipe0_rtr__SHIFT 0x4
+#define VGT_DEBUG_REG18__pipe1_rtr_MASK 0x20
+#define VGT_DEBUG_REG18__pipe1_rtr__SHIFT 0x5
+#define VGT_DEBUG_REG18__out_vr_indx_read_MASK 0x40
+#define VGT_DEBUG_REG18__out_vr_indx_read__SHIFT 0x6
+#define VGT_DEBUG_REG18__out_vr_prim_read_MASK 0x80
+#define VGT_DEBUG_REG18__out_vr_prim_read__SHIFT 0x7
+#define VGT_DEBUG_REG18__indices_to_send_q_MASK 0x700
+#define VGT_DEBUG_REG18__indices_to_send_q__SHIFT 0x8
+#define VGT_DEBUG_REG18__valid_indices_MASK 0x800
+#define VGT_DEBUG_REG18__valid_indices__SHIFT 0xb
+#define VGT_DEBUG_REG18__last_indx_of_prim_MASK 0x1000
+#define VGT_DEBUG_REG18__last_indx_of_prim__SHIFT 0xc
+#define VGT_DEBUG_REG18__indx0_new_d_MASK 0x2000
+#define VGT_DEBUG_REG18__indx0_new_d__SHIFT 0xd
+#define VGT_DEBUG_REG18__indx1_new_d_MASK 0x4000
+#define VGT_DEBUG_REG18__indx1_new_d__SHIFT 0xe
+#define VGT_DEBUG_REG18__indx2_new_d_MASK 0x8000
+#define VGT_DEBUG_REG18__indx2_new_d__SHIFT 0xf
+#define VGT_DEBUG_REG18__indx2_hit_d_MASK 0x10000
+#define VGT_DEBUG_REG18__indx2_hit_d__SHIFT 0x10
+#define VGT_DEBUG_REG18__indx1_hit_d_MASK 0x20000
+#define VGT_DEBUG_REG18__indx1_hit_d__SHIFT 0x11
+#define VGT_DEBUG_REG18__indx0_hit_d_MASK 0x40000
+#define VGT_DEBUG_REG18__indx0_hit_d__SHIFT 0x12
+#define VGT_DEBUG_REG18__st_vertex_reuse_off_r0_q_MASK 0x80000
+#define VGT_DEBUG_REG18__st_vertex_reuse_off_r0_q__SHIFT 0x13
+#define VGT_DEBUG_REG18__last_group_of_instance_r0_q_MASK 0x100000
+#define VGT_DEBUG_REG18__last_group_of_instance_r0_q__SHIFT 0x14
+#define VGT_DEBUG_REG18__null_primitive_r0_q_MASK 0x200000
+#define VGT_DEBUG_REG18__null_primitive_r0_q__SHIFT 0x15
+#define VGT_DEBUG_REG18__eop_r0_q_MASK 0x400000
+#define VGT_DEBUG_REG18__eop_r0_q__SHIFT 0x16
+#define VGT_DEBUG_REG18__eject_vtx_vect_r1_d_MASK 0x800000
+#define VGT_DEBUG_REG18__eject_vtx_vect_r1_d__SHIFT 0x17
+#define VGT_DEBUG_REG18__sub_prim_type_r0_q_MASK 0x7000000
+#define VGT_DEBUG_REG18__sub_prim_type_r0_q__SHIFT 0x18
+#define VGT_DEBUG_REG18__gs_scenario_a_r0_q_MASK 0x8000000
+#define VGT_DEBUG_REG18__gs_scenario_a_r0_q__SHIFT 0x1b
+#define VGT_DEBUG_REG18__gs_scenario_b_r0_q_MASK 0x10000000
+#define VGT_DEBUG_REG18__gs_scenario_b_r0_q__SHIFT 0x1c
+#define VGT_DEBUG_REG18__components_valid_r0_q_MASK 0xe0000000
+#define VGT_DEBUG_REG18__components_valid_r0_q__SHIFT 0x1d
+#define VGT_DEBUG_REG19__separate_out_busy_q_MASK 0x1
+#define VGT_DEBUG_REG19__separate_out_busy_q__SHIFT 0x0
+#define VGT_DEBUG_REG19__separate_out_indx_busy_q_MASK 0x2
+#define VGT_DEBUG_REG19__separate_out_indx_busy_q__SHIFT 0x1
+#define VGT_DEBUG_REG19__prim_buffer_empty_MASK 0x4
+#define VGT_DEBUG_REG19__prim_buffer_empty__SHIFT 0x2
+#define VGT_DEBUG_REG19__prim_buffer_full_MASK 0x8
+#define VGT_DEBUG_REG19__prim_buffer_full__SHIFT 0x3
+#define VGT_DEBUG_REG19__pa_clips_fifo_busy_q_MASK 0x10
+#define VGT_DEBUG_REG19__pa_clips_fifo_busy_q__SHIFT 0x4
+#define VGT_DEBUG_REG19__pa_clipp_fifo_busy_q_MASK 0x20
+#define VGT_DEBUG_REG19__pa_clipp_fifo_busy_q__SHIFT 0x5
+#define VGT_DEBUG_REG19__VGT_PA_clips_rtr_q_MASK 0x40
+#define VGT_DEBUG_REG19__VGT_PA_clips_rtr_q__SHIFT 0x6
+#define VGT_DEBUG_REG19__VGT_PA_clipp_rtr_q_MASK 0x80
+#define VGT_DEBUG_REG19__VGT_PA_clipp_rtr_q__SHIFT 0x7
+#define VGT_DEBUG_REG19__spi_vsthread_fifo_busy_q_MASK 0x100
+#define VGT_DEBUG_REG19__spi_vsthread_fifo_busy_q__SHIFT 0x8
+#define VGT_DEBUG_REG19__spi_vsvert_fifo_busy_q_MASK 0x200
+#define VGT_DEBUG_REG19__spi_vsvert_fifo_busy_q__SHIFT 0x9
+#define VGT_DEBUG_REG19__pa_clipv_fifo_busy_q_MASK 0x400
+#define VGT_DEBUG_REG19__pa_clipv_fifo_busy_q__SHIFT 0xa
+#define VGT_DEBUG_REG19__hold_prim_MASK 0x800
+#define VGT_DEBUG_REG19__hold_prim__SHIFT 0xb
+#define VGT_DEBUG_REG19__VGT_SPI_vsthread_rtr_q_MASK 0x1000
+#define VGT_DEBUG_REG19__VGT_SPI_vsthread_rtr_q__SHIFT 0xc
+#define VGT_DEBUG_REG19__VGT_SPI_vsvert_rtr_q_MASK 0x2000
+#define VGT_DEBUG_REG19__VGT_SPI_vsvert_rtr_q__SHIFT 0xd
+#define VGT_DEBUG_REG19__VGT_PA_clipv_rtr_q_MASK 0x4000
+#define VGT_DEBUG_REG19__VGT_PA_clipv_rtr_q__SHIFT 0xe
+#define VGT_DEBUG_REG19__new_packet_q_MASK 0x8000
+#define VGT_DEBUG_REG19__new_packet_q__SHIFT 0xf
+#define VGT_DEBUG_REG19__buffered_prim_event_MASK 0x10000
+#define VGT_DEBUG_REG19__buffered_prim_event__SHIFT 0x10
+#define VGT_DEBUG_REG19__buffered_prim_null_primitive_MASK 0x20000
+#define VGT_DEBUG_REG19__buffered_prim_null_primitive__SHIFT 0x11
+#define VGT_DEBUG_REG19__buffered_prim_eop_MASK 0x40000
+#define VGT_DEBUG_REG19__buffered_prim_eop__SHIFT 0x12
+#define VGT_DEBUG_REG19__buffered_prim_eject_vtx_vect_MASK 0x80000
+#define VGT_DEBUG_REG19__buffered_prim_eject_vtx_vect__SHIFT 0x13
+#define VGT_DEBUG_REG19__buffered_prim_type_event_MASK 0x3f00000
+#define VGT_DEBUG_REG19__buffered_prim_type_event__SHIFT 0x14
+#define VGT_DEBUG_REG19__VGT_SE1SPI_vswave_rtr_q_MASK 0x4000000
+#define VGT_DEBUG_REG19__VGT_SE1SPI_vswave_rtr_q__SHIFT 0x1a
+#define VGT_DEBUG_REG19__VGT_SE1SPI_vsvert_rtr_q_MASK 0x8000000
+#define VGT_DEBUG_REG19__VGT_SE1SPI_vsvert_rtr_q__SHIFT 0x1b
+#define VGT_DEBUG_REG19__num_new_unique_rel_indx_MASK 0x30000000
+#define VGT_DEBUG_REG19__num_new_unique_rel_indx__SHIFT 0x1c
+#define VGT_DEBUG_REG19__null_terminate_vtx_vector_MASK 0x40000000
+#define VGT_DEBUG_REG19__null_terminate_vtx_vector__SHIFT 0x1e
+#define VGT_DEBUG_REG19__filter_event_MASK 0x80000000
+#define VGT_DEBUG_REG19__filter_event__SHIFT 0x1f
+#define VGT_DEBUG_REG20__dbg_VGT_SPI_vsthread_sovertexindex_MASK 0xffff
+#define VGT_DEBUG_REG20__dbg_VGT_SPI_vsthread_sovertexindex__SHIFT 0x0
+#define VGT_DEBUG_REG20__dbg_VGT_SPI_vsthread_sovertexcount_not_0_MASK 0x10000
+#define VGT_DEBUG_REG20__dbg_VGT_SPI_vsthread_sovertexcount_not_0__SHIFT 0x10
+#define VGT_DEBUG_REG20__SPARE17_MASK 0x20000
+#define VGT_DEBUG_REG20__SPARE17__SHIFT 0x11
+#define VGT_DEBUG_REG20__alloc_counter_q_MASK 0x3c0000
+#define VGT_DEBUG_REG20__alloc_counter_q__SHIFT 0x12
+#define VGT_DEBUG_REG20__curr_dealloc_distance_q_MASK 0x1fc00000
+#define VGT_DEBUG_REG20__curr_dealloc_distance_q__SHIFT 0x16
+#define VGT_DEBUG_REG20__new_allocate_q_MASK 0x20000000
+#define VGT_DEBUG_REG20__new_allocate_q__SHIFT 0x1d
+#define VGT_DEBUG_REG20__curr_slot_in_vtx_vect_q_not_0_MASK 0x40000000
+#define VGT_DEBUG_REG20__curr_slot_in_vtx_vect_q_not_0__SHIFT 0x1e
+#define VGT_DEBUG_REG20__int_vtx_counter_q_not_0_MASK 0x80000000
+#define VGT_DEBUG_REG20__int_vtx_counter_q_not_0__SHIFT 0x1f
+#define VGT_DEBUG_REG21__out_indx_fifo_empty_MASK 0x1
+#define VGT_DEBUG_REG21__out_indx_fifo_empty__SHIFT 0x0
+#define VGT_DEBUG_REG21__indx_side_fifo_empty_MASK 0x2
+#define VGT_DEBUG_REG21__indx_side_fifo_empty__SHIFT 0x1
+#define VGT_DEBUG_REG21__pipe0_dr_MASK 0x4
+#define VGT_DEBUG_REG21__pipe0_dr__SHIFT 0x2
+#define VGT_DEBUG_REG21__pipe1_dr_MASK 0x8
+#define VGT_DEBUG_REG21__pipe1_dr__SHIFT 0x3
+#define VGT_DEBUG_REG21__pipe2_dr_MASK 0x10
+#define VGT_DEBUG_REG21__pipe2_dr__SHIFT 0x4
+#define VGT_DEBUG_REG21__vsthread_buff_empty_MASK 0x20
+#define VGT_DEBUG_REG21__vsthread_buff_empty__SHIFT 0x5
+#define VGT_DEBUG_REG21__out_indx_fifo_full_MASK 0x40
+#define VGT_DEBUG_REG21__out_indx_fifo_full__SHIFT 0x6
+#define VGT_DEBUG_REG21__indx_side_fifo_full_MASK 0x80
+#define VGT_DEBUG_REG21__indx_side_fifo_full__SHIFT 0x7
+#define VGT_DEBUG_REG21__pipe0_rtr_MASK 0x100
+#define VGT_DEBUG_REG21__pipe0_rtr__SHIFT 0x8
+#define VGT_DEBUG_REG21__pipe1_rtr_MASK 0x200
+#define VGT_DEBUG_REG21__pipe1_rtr__SHIFT 0x9
+#define VGT_DEBUG_REG21__pipe2_rtr_MASK 0x400
+#define VGT_DEBUG_REG21__pipe2_rtr__SHIFT 0xa
+#define VGT_DEBUG_REG21__vsthread_buff_full_MASK 0x800
+#define VGT_DEBUG_REG21__vsthread_buff_full__SHIFT 0xb
+#define VGT_DEBUG_REG21__interfaces_rtr_MASK 0x1000
+#define VGT_DEBUG_REG21__interfaces_rtr__SHIFT 0xc
+#define VGT_DEBUG_REG21__indx_count_q_not_0_MASK 0x2000
+#define VGT_DEBUG_REG21__indx_count_q_not_0__SHIFT 0xd
+#define VGT_DEBUG_REG21__wait_for_external_eopg_q_MASK 0x4000
+#define VGT_DEBUG_REG21__wait_for_external_eopg_q__SHIFT 0xe
+#define VGT_DEBUG_REG21__full_state_p1_q_MASK 0x8000
+#define VGT_DEBUG_REG21__full_state_p1_q__SHIFT 0xf
+#define VGT_DEBUG_REG21__indx_side_indx_valid_MASK 0x10000
+#define VGT_DEBUG_REG21__indx_side_indx_valid__SHIFT 0x10
+#define VGT_DEBUG_REG21__stateid_p0_q_MASK 0xe0000
+#define VGT_DEBUG_REG21__stateid_p0_q__SHIFT 0x11
+#define VGT_DEBUG_REG21__is_event_p0_q_MASK 0x100000
+#define VGT_DEBUG_REG21__is_event_p0_q__SHIFT 0x14
+#define VGT_DEBUG_REG21__lshs_dealloc_p1_MASK 0x200000
+#define VGT_DEBUG_REG21__lshs_dealloc_p1__SHIFT 0x15
+#define VGT_DEBUG_REG21__stream_id_r2_q_MASK 0x400000
+#define VGT_DEBUG_REG21__stream_id_r2_q__SHIFT 0x16
+#define VGT_DEBUG_REG21__vtx_vect_counter_q_not_0_MASK 0x800000
+#define VGT_DEBUG_REG21__vtx_vect_counter_q_not_0__SHIFT 0x17
+#define VGT_DEBUG_REG21__buff_full_p1_MASK 0x1000000
+#define VGT_DEBUG_REG21__buff_full_p1__SHIFT 0x18
+#define VGT_DEBUG_REG21__strmout_valid_p1_MASK 0x2000000
+#define VGT_DEBUG_REG21__strmout_valid_p1__SHIFT 0x19
+#define VGT_DEBUG_REG21__eotg_r2_q_MASK 0x4000000
+#define VGT_DEBUG_REG21__eotg_r2_q__SHIFT 0x1a
+#define VGT_DEBUG_REG21__null_r2_q_MASK 0x8000000
+#define VGT_DEBUG_REG21__null_r2_q__SHIFT 0x1b
+#define VGT_DEBUG_REG21__p0_dr_MASK 0x10000000
+#define VGT_DEBUG_REG21__p0_dr__SHIFT 0x1c
+#define VGT_DEBUG_REG21__p0_rtr_MASK 0x20000000
+#define VGT_DEBUG_REG21__p0_rtr__SHIFT 0x1d
+#define VGT_DEBUG_REG21__eopg_p0_q_MASK 0x40000000
+#define VGT_DEBUG_REG21__eopg_p0_q__SHIFT 0x1e
+#define VGT_DEBUG_REG21__p0_nobp_MASK 0x80000000
+#define VGT_DEBUG_REG21__p0_nobp__SHIFT 0x1f
+#define VGT_DEBUG_REG22__cm_state16_MASK 0x3
+#define VGT_DEBUG_REG22__cm_state16__SHIFT 0x0
+#define VGT_DEBUG_REG22__cm_state17_MASK 0xc
+#define VGT_DEBUG_REG22__cm_state17__SHIFT 0x2
+#define VGT_DEBUG_REG22__cm_state18_MASK 0x30
+#define VGT_DEBUG_REG22__cm_state18__SHIFT 0x4
+#define VGT_DEBUG_REG22__cm_state19_MASK 0xc0
+#define VGT_DEBUG_REG22__cm_state19__SHIFT 0x6
+#define VGT_DEBUG_REG22__cm_state20_MASK 0x300
+#define VGT_DEBUG_REG22__cm_state20__SHIFT 0x8
+#define VGT_DEBUG_REG22__cm_state21_MASK 0xc00
+#define VGT_DEBUG_REG22__cm_state21__SHIFT 0xa
+#define VGT_DEBUG_REG22__cm_state22_MASK 0x3000
+#define VGT_DEBUG_REG22__cm_state22__SHIFT 0xc
+#define VGT_DEBUG_REG22__cm_state23_MASK 0xc000
+#define VGT_DEBUG_REG22__cm_state23__SHIFT 0xe
+#define VGT_DEBUG_REG22__cm_state24_MASK 0x30000
+#define VGT_DEBUG_REG22__cm_state24__SHIFT 0x10
+#define VGT_DEBUG_REG22__cm_state25_MASK 0xc0000
+#define VGT_DEBUG_REG22__cm_state25__SHIFT 0x12
+#define VGT_DEBUG_REG22__cm_state26_MASK 0x300000
+#define VGT_DEBUG_REG22__cm_state26__SHIFT 0x14
+#define VGT_DEBUG_REG22__cm_state27_MASK 0xc00000
+#define VGT_DEBUG_REG22__cm_state27__SHIFT 0x16
+#define VGT_DEBUG_REG22__cm_state28_MASK 0x3000000
+#define VGT_DEBUG_REG22__cm_state28__SHIFT 0x18
+#define VGT_DEBUG_REG22__cm_state29_MASK 0xc000000
+#define VGT_DEBUG_REG22__cm_state29__SHIFT 0x1a
+#define VGT_DEBUG_REG22__cm_state30_MASK 0x30000000
+#define VGT_DEBUG_REG22__cm_state30__SHIFT 0x1c
+#define VGT_DEBUG_REG22__cm_state31_MASK 0xc0000000
+#define VGT_DEBUG_REG22__cm_state31__SHIFT 0x1e
+#define VGT_DEBUG_REG23__frmt_busy_MASK 0x1
+#define VGT_DEBUG_REG23__frmt_busy__SHIFT 0x0
+#define VGT_DEBUG_REG23__rcm_frmt_vert_rtr_MASK 0x2
+#define VGT_DEBUG_REG23__rcm_frmt_vert_rtr__SHIFT 0x1
+#define VGT_DEBUG_REG23__rcm_frmt_prim_rtr_MASK 0x4
+#define VGT_DEBUG_REG23__rcm_frmt_prim_rtr__SHIFT 0x2
+#define VGT_DEBUG_REG23__prim_r3_rtr_MASK 0x8
+#define VGT_DEBUG_REG23__prim_r3_rtr__SHIFT 0x3
+#define VGT_DEBUG_REG23__prim_r2_rtr_MASK 0x10
+#define VGT_DEBUG_REG23__prim_r2_rtr__SHIFT 0x4
+#define VGT_DEBUG_REG23__vert_r3_rtr_MASK 0x20
+#define VGT_DEBUG_REG23__vert_r3_rtr__SHIFT 0x5
+#define VGT_DEBUG_REG23__vert_r2_rtr_MASK 0x40
+#define VGT_DEBUG_REG23__vert_r2_rtr__SHIFT 0x6
+#define VGT_DEBUG_REG23__vert_r1_rtr_MASK 0x80
+#define VGT_DEBUG_REG23__vert_r1_rtr__SHIFT 0x7
+#define VGT_DEBUG_REG23__vert_r0_rtr_MASK 0x100
+#define VGT_DEBUG_REG23__vert_r0_rtr__SHIFT 0x8
+#define VGT_DEBUG_REG23__prim_fifo_empty_MASK 0x200
+#define VGT_DEBUG_REG23__prim_fifo_empty__SHIFT 0x9
+#define VGT_DEBUG_REG23__prim_fifo_full_MASK 0x400
+#define VGT_DEBUG_REG23__prim_fifo_full__SHIFT 0xa
+#define VGT_DEBUG_REG23__vert_dr_r2_q_MASK 0x800
+#define VGT_DEBUG_REG23__vert_dr_r2_q__SHIFT 0xb
+#define VGT_DEBUG_REG23__prim_dr_r2_q_MASK 0x1000
+#define VGT_DEBUG_REG23__prim_dr_r2_q__SHIFT 0xc
+#define VGT_DEBUG_REG23__vert_dr_r1_q_MASK 0x2000
+#define VGT_DEBUG_REG23__vert_dr_r1_q__SHIFT 0xd
+#define VGT_DEBUG_REG23__vert_dr_r0_q_MASK 0x4000
+#define VGT_DEBUG_REG23__vert_dr_r0_q__SHIFT 0xe
+#define VGT_DEBUG_REG23__new_verts_r2_q_MASK 0x18000
+#define VGT_DEBUG_REG23__new_verts_r2_q__SHIFT 0xf
+#define VGT_DEBUG_REG23__verts_sent_r2_q_MASK 0x1e0000
+#define VGT_DEBUG_REG23__verts_sent_r2_q__SHIFT 0x11
+#define VGT_DEBUG_REG23__prim_state_sel_r2_q_MASK 0xe00000
+#define VGT_DEBUG_REG23__prim_state_sel_r2_q__SHIFT 0x15
+#define VGT_DEBUG_REG23__SPARE_MASK 0xff000000
+#define VGT_DEBUG_REG23__SPARE__SHIFT 0x18
+#define VGT_DEBUG_REG24__avail_es_rb_space_r0_q_23_0_MASK 0xffffff
+#define VGT_DEBUG_REG24__avail_es_rb_space_r0_q_23_0__SHIFT 0x0
+#define VGT_DEBUG_REG24__dependent_st_cut_mode_q_MASK 0x3000000
+#define VGT_DEBUG_REG24__dependent_st_cut_mode_q__SHIFT 0x18
+#define VGT_DEBUG_REG24__SPARE31_MASK 0xfc000000
+#define VGT_DEBUG_REG24__SPARE31__SHIFT 0x1a
+#define VGT_DEBUG_REG25__avail_gs_rb_space_r0_q_25_0_MASK 0x3ffffff
+#define VGT_DEBUG_REG25__avail_gs_rb_space_r0_q_25_0__SHIFT 0x0
+#define VGT_DEBUG_REG25__active_sm_r0_q_MASK 0x3c000000
+#define VGT_DEBUG_REG25__active_sm_r0_q__SHIFT 0x1a
+#define VGT_DEBUG_REG25__add_gs_rb_space_r1_q_MASK 0x40000000
+#define VGT_DEBUG_REG25__add_gs_rb_space_r1_q__SHIFT 0x1e
+#define VGT_DEBUG_REG25__add_gs_rb_space_r0_q_MASK 0x80000000
+#define VGT_DEBUG_REG25__add_gs_rb_space_r0_q__SHIFT 0x1f
+#define VGT_DEBUG_REG26__cm_state0_MASK 0x3
+#define VGT_DEBUG_REG26__cm_state0__SHIFT 0x0
+#define VGT_DEBUG_REG26__cm_state1_MASK 0xc
+#define VGT_DEBUG_REG26__cm_state1__SHIFT 0x2
+#define VGT_DEBUG_REG26__cm_state2_MASK 0x30
+#define VGT_DEBUG_REG26__cm_state2__SHIFT 0x4
+#define VGT_DEBUG_REG26__cm_state3_MASK 0xc0
+#define VGT_DEBUG_REG26__cm_state3__SHIFT 0x6
+#define VGT_DEBUG_REG26__cm_state4_MASK 0x300
+#define VGT_DEBUG_REG26__cm_state4__SHIFT 0x8
+#define VGT_DEBUG_REG26__cm_state5_MASK 0xc00
+#define VGT_DEBUG_REG26__cm_state5__SHIFT 0xa
+#define VGT_DEBUG_REG26__cm_state6_MASK 0x3000
+#define VGT_DEBUG_REG26__cm_state6__SHIFT 0xc
+#define VGT_DEBUG_REG26__cm_state7_MASK 0xc000
+#define VGT_DEBUG_REG26__cm_state7__SHIFT 0xe
+#define VGT_DEBUG_REG26__cm_state8_MASK 0x30000
+#define VGT_DEBUG_REG26__cm_state8__SHIFT 0x10
+#define VGT_DEBUG_REG26__cm_state9_MASK 0xc0000
+#define VGT_DEBUG_REG26__cm_state9__SHIFT 0x12
+#define VGT_DEBUG_REG26__cm_state10_MASK 0x300000
+#define VGT_DEBUG_REG26__cm_state10__SHIFT 0x14
+#define VGT_DEBUG_REG26__cm_state11_MASK 0xc00000
+#define VGT_DEBUG_REG26__cm_state11__SHIFT 0x16
+#define VGT_DEBUG_REG26__cm_state12_MASK 0x3000000
+#define VGT_DEBUG_REG26__cm_state12__SHIFT 0x18
+#define VGT_DEBUG_REG26__cm_state13_MASK 0xc000000
+#define VGT_DEBUG_REG26__cm_state13__SHIFT 0x1a
+#define VGT_DEBUG_REG26__cm_state14_MASK 0x30000000
+#define VGT_DEBUG_REG26__cm_state14__SHIFT 0x1c
+#define VGT_DEBUG_REG26__cm_state15_MASK 0xc0000000
+#define VGT_DEBUG_REG26__cm_state15__SHIFT 0x1e
+#define VGT_DEBUG_REG27__pipe0_dr_MASK 0x1
+#define VGT_DEBUG_REG27__pipe0_dr__SHIFT 0x0
+#define VGT_DEBUG_REG27__gsc0_dr_MASK 0x2
+#define VGT_DEBUG_REG27__gsc0_dr__SHIFT 0x1
+#define VGT_DEBUG_REG27__pipe1_dr_MASK 0x4
+#define VGT_DEBUG_REG27__pipe1_dr__SHIFT 0x2
+#define VGT_DEBUG_REG27__tm_pt_event_rtr_MASK 0x8
+#define VGT_DEBUG_REG27__tm_pt_event_rtr__SHIFT 0x3
+#define VGT_DEBUG_REG27__pipe0_rtr_MASK 0x10
+#define VGT_DEBUG_REG27__pipe0_rtr__SHIFT 0x4
+#define VGT_DEBUG_REG27__gsc0_rtr_MASK 0x20
+#define VGT_DEBUG_REG27__gsc0_rtr__SHIFT 0x5
+#define VGT_DEBUG_REG27__pipe1_rtr_MASK 0x40
+#define VGT_DEBUG_REG27__pipe1_rtr__SHIFT 0x6
+#define VGT_DEBUG_REG27__last_indx_of_prim_p1_q_MASK 0x80
+#define VGT_DEBUG_REG27__last_indx_of_prim_p1_q__SHIFT 0x7
+#define VGT_DEBUG_REG27__indices_to_send_p0_q_MASK 0x300
+#define VGT_DEBUG_REG27__indices_to_send_p0_q__SHIFT 0x8
+#define VGT_DEBUG_REG27__event_flag_p1_q_MASK 0x400
+#define VGT_DEBUG_REG27__event_flag_p1_q__SHIFT 0xa
+#define VGT_DEBUG_REG27__eop_p1_q_MASK 0x800
+#define VGT_DEBUG_REG27__eop_p1_q__SHIFT 0xb
+#define VGT_DEBUG_REG27__gs_out_prim_type_p0_q_MASK 0x3000
+#define VGT_DEBUG_REG27__gs_out_prim_type_p0_q__SHIFT 0xc
+#define VGT_DEBUG_REG27__gsc_null_primitive_p0_q_MASK 0x4000
+#define VGT_DEBUG_REG27__gsc_null_primitive_p0_q__SHIFT 0xe
+#define VGT_DEBUG_REG27__gsc_eop_p0_q_MASK 0x8000
+#define VGT_DEBUG_REG27__gsc_eop_p0_q__SHIFT 0xf
+#define VGT_DEBUG_REG27__gsc_2cycle_output_MASK 0x10000
+#define VGT_DEBUG_REG27__gsc_2cycle_output__SHIFT 0x10
+#define VGT_DEBUG_REG27__gsc_2nd_cycle_p0_q_MASK 0x20000
+#define VGT_DEBUG_REG27__gsc_2nd_cycle_p0_q__SHIFT 0x11
+#define VGT_DEBUG_REG27__last_indx_of_vsprim_MASK 0x40000
+#define VGT_DEBUG_REG27__last_indx_of_vsprim__SHIFT 0x12
+#define VGT_DEBUG_REG27__first_vsprim_of_gsprim_p0_q_MASK 0x80000
+#define VGT_DEBUG_REG27__first_vsprim_of_gsprim_p0_q__SHIFT 0x13
+#define VGT_DEBUG_REG27__gsc_indx_count_p0_q_MASK 0x7ff00000
+#define VGT_DEBUG_REG27__gsc_indx_count_p0_q__SHIFT 0x14
+#define VGT_DEBUG_REG27__last_vsprim_of_gsprim_MASK 0x80000000
+#define VGT_DEBUG_REG27__last_vsprim_of_gsprim__SHIFT 0x1f
+#define VGT_DEBUG_REG28__con_state_q_MASK 0xf
+#define VGT_DEBUG_REG28__con_state_q__SHIFT 0x0
+#define VGT_DEBUG_REG28__second_cycle_q_MASK 0x10
+#define VGT_DEBUG_REG28__second_cycle_q__SHIFT 0x4
+#define VGT_DEBUG_REG28__process_tri_middle_p0_q_MASK 0x20
+#define VGT_DEBUG_REG28__process_tri_middle_p0_q__SHIFT 0x5
+#define VGT_DEBUG_REG28__process_tri_1st_2nd_half_p0_q_MASK 0x40
+#define VGT_DEBUG_REG28__process_tri_1st_2nd_half_p0_q__SHIFT 0x6
+#define VGT_DEBUG_REG28__process_tri_center_poly_p0_q_MASK 0x80
+#define VGT_DEBUG_REG28__process_tri_center_poly_p0_q__SHIFT 0x7
+#define VGT_DEBUG_REG28__pipe0_patch_dr_MASK 0x100
+#define VGT_DEBUG_REG28__pipe0_patch_dr__SHIFT 0x8
+#define VGT_DEBUG_REG28__pipe0_edge_dr_MASK 0x200
+#define VGT_DEBUG_REG28__pipe0_edge_dr__SHIFT 0x9
+#define VGT_DEBUG_REG28__pipe1_dr_MASK 0x400
+#define VGT_DEBUG_REG28__pipe1_dr__SHIFT 0xa
+#define VGT_DEBUG_REG28__pipe0_patch_rtr_MASK 0x800
+#define VGT_DEBUG_REG28__pipe0_patch_rtr__SHIFT 0xb
+#define VGT_DEBUG_REG28__pipe0_edge_rtr_MASK 0x1000
+#define VGT_DEBUG_REG28__pipe0_edge_rtr__SHIFT 0xc
+#define VGT_DEBUG_REG28__pipe1_rtr_MASK 0x2000
+#define VGT_DEBUG_REG28__pipe1_rtr__SHIFT 0xd
+#define VGT_DEBUG_REG28__outer_parity_p0_q_MASK 0x4000
+#define VGT_DEBUG_REG28__outer_parity_p0_q__SHIFT 0xe
+#define VGT_DEBUG_REG28__parallel_parity_p0_q_MASK 0x8000
+#define VGT_DEBUG_REG28__parallel_parity_p0_q__SHIFT 0xf
+#define VGT_DEBUG_REG28__first_ring_of_patch_p0_q_MASK 0x10000
+#define VGT_DEBUG_REG28__first_ring_of_patch_p0_q__SHIFT 0x10
+#define VGT_DEBUG_REG28__last_ring_of_patch_p0_q_MASK 0x20000
+#define VGT_DEBUG_REG28__last_ring_of_patch_p0_q__SHIFT 0x11
+#define VGT_DEBUG_REG28__last_edge_of_outer_ring_p0_q_MASK 0x40000
+#define VGT_DEBUG_REG28__last_edge_of_outer_ring_p0_q__SHIFT 0x12
+#define VGT_DEBUG_REG28__last_point_of_outer_ring_p1_MASK 0x80000
+#define VGT_DEBUG_REG28__last_point_of_outer_ring_p1__SHIFT 0x13
+#define VGT_DEBUG_REG28__last_point_of_inner_ring_p1_MASK 0x100000
+#define VGT_DEBUG_REG28__last_point_of_inner_ring_p1__SHIFT 0x14
+#define VGT_DEBUG_REG28__outer_edge_tf_eq_one_p0_q_MASK 0x200000
+#define VGT_DEBUG_REG28__outer_edge_tf_eq_one_p0_q__SHIFT 0x15
+#define VGT_DEBUG_REG28__advance_outer_point_p1_MASK 0x400000
+#define VGT_DEBUG_REG28__advance_outer_point_p1__SHIFT 0x16
+#define VGT_DEBUG_REG28__advance_inner_point_p1_MASK 0x800000
+#define VGT_DEBUG_REG28__advance_inner_point_p1__SHIFT 0x17
+#define VGT_DEBUG_REG28__next_ring_is_rect_p0_q_MASK 0x1000000
+#define VGT_DEBUG_REG28__next_ring_is_rect_p0_q__SHIFT 0x18
+#define VGT_DEBUG_REG28__pipe1_outer1_rtr_MASK 0x2000000
+#define VGT_DEBUG_REG28__pipe1_outer1_rtr__SHIFT 0x19
+#define VGT_DEBUG_REG28__pipe1_outer2_rtr_MASK 0x4000000
+#define VGT_DEBUG_REG28__pipe1_outer2_rtr__SHIFT 0x1a
+#define VGT_DEBUG_REG28__pipe1_inner1_rtr_MASK 0x8000000
+#define VGT_DEBUG_REG28__pipe1_inner1_rtr__SHIFT 0x1b
+#define VGT_DEBUG_REG28__pipe1_inner2_rtr_MASK 0x10000000
+#define VGT_DEBUG_REG28__pipe1_inner2_rtr__SHIFT 0x1c
+#define VGT_DEBUG_REG28__pipe1_patch_rtr_MASK 0x20000000
+#define VGT_DEBUG_REG28__pipe1_patch_rtr__SHIFT 0x1d
+#define VGT_DEBUG_REG28__pipe1_edge_rtr_MASK 0x40000000
+#define VGT_DEBUG_REG28__pipe1_edge_rtr__SHIFT 0x1e
+#define VGT_DEBUG_REG28__use_stored_inner_q_ring2_MASK 0x80000000
+#define VGT_DEBUG_REG28__use_stored_inner_q_ring2__SHIFT 0x1f
+#define VGT_DEBUG_REG29__con_state_q_MASK 0xf
+#define VGT_DEBUG_REG29__con_state_q__SHIFT 0x0
+#define VGT_DEBUG_REG29__second_cycle_q_MASK 0x10
+#define VGT_DEBUG_REG29__second_cycle_q__SHIFT 0x4
+#define VGT_DEBUG_REG29__process_tri_middle_p0_q_MASK 0x20
+#define VGT_DEBUG_REG29__process_tri_middle_p0_q__SHIFT 0x5
+#define VGT_DEBUG_REG29__process_tri_1st_2nd_half_p0_q_MASK 0x40
+#define VGT_DEBUG_REG29__process_tri_1st_2nd_half_p0_q__SHIFT 0x6
+#define VGT_DEBUG_REG29__process_tri_center_poly_p0_q_MASK 0x80
+#define VGT_DEBUG_REG29__process_tri_center_poly_p0_q__SHIFT 0x7
+#define VGT_DEBUG_REG29__pipe0_patch_dr_MASK 0x100
+#define VGT_DEBUG_REG29__pipe0_patch_dr__SHIFT 0x8
+#define VGT_DEBUG_REG29__pipe0_edge_dr_MASK 0x200
+#define VGT_DEBUG_REG29__pipe0_edge_dr__SHIFT 0x9
+#define VGT_DEBUG_REG29__pipe1_dr_MASK 0x400
+#define VGT_DEBUG_REG29__pipe1_dr__SHIFT 0xa
+#define VGT_DEBUG_REG29__pipe0_patch_rtr_MASK 0x800
+#define VGT_DEBUG_REG29__pipe0_patch_rtr__SHIFT 0xb
+#define VGT_DEBUG_REG29__pipe0_edge_rtr_MASK 0x1000
+#define VGT_DEBUG_REG29__pipe0_edge_rtr__SHIFT 0xc
+#define VGT_DEBUG_REG29__pipe1_rtr_MASK 0x2000
+#define VGT_DEBUG_REG29__pipe1_rtr__SHIFT 0xd
+#define VGT_DEBUG_REG29__outer_parity_p0_q_MASK 0x4000
+#define VGT_DEBUG_REG29__outer_parity_p0_q__SHIFT 0xe
+#define VGT_DEBUG_REG29__parallel_parity_p0_q_MASK 0x8000
+#define VGT_DEBUG_REG29__parallel_parity_p0_q__SHIFT 0xf
+#define VGT_DEBUG_REG29__first_ring_of_patch_p0_q_MASK 0x10000
+#define VGT_DEBUG_REG29__first_ring_of_patch_p0_q__SHIFT 0x10
+#define VGT_DEBUG_REG29__last_ring_of_patch_p0_q_MASK 0x20000
+#define VGT_DEBUG_REG29__last_ring_of_patch_p0_q__SHIFT 0x11
+#define VGT_DEBUG_REG29__last_edge_of_outer_ring_p0_q_MASK 0x40000
+#define VGT_DEBUG_REG29__last_edge_of_outer_ring_p0_q__SHIFT 0x12
+#define VGT_DEBUG_REG29__last_point_of_outer_ring_p1_MASK 0x80000
+#define VGT_DEBUG_REG29__last_point_of_outer_ring_p1__SHIFT 0x13
+#define VGT_DEBUG_REG29__last_point_of_inner_ring_p1_MASK 0x100000
+#define VGT_DEBUG_REG29__last_point_of_inner_ring_p1__SHIFT 0x14
+#define VGT_DEBUG_REG29__outer_edge_tf_eq_one_p0_q_MASK 0x200000
+#define VGT_DEBUG_REG29__outer_edge_tf_eq_one_p0_q__SHIFT 0x15
+#define VGT_DEBUG_REG29__advance_outer_point_p1_MASK 0x400000
+#define VGT_DEBUG_REG29__advance_outer_point_p1__SHIFT 0x16
+#define VGT_DEBUG_REG29__advance_inner_point_p1_MASK 0x800000
+#define VGT_DEBUG_REG29__advance_inner_point_p1__SHIFT 0x17
+#define VGT_DEBUG_REG29__next_ring_is_rect_p0_q_MASK 0x1000000
+#define VGT_DEBUG_REG29__next_ring_is_rect_p0_q__SHIFT 0x18
+#define VGT_DEBUG_REG29__pipe1_outer1_rtr_MASK 0x2000000
+#define VGT_DEBUG_REG29__pipe1_outer1_rtr__SHIFT 0x19
+#define VGT_DEBUG_REG29__pipe1_outer2_rtr_MASK 0x4000000
+#define VGT_DEBUG_REG29__pipe1_outer2_rtr__SHIFT 0x1a
+#define VGT_DEBUG_REG29__pipe1_inner1_rtr_MASK 0x8000000
+#define VGT_DEBUG_REG29__pipe1_inner1_rtr__SHIFT 0x1b
+#define VGT_DEBUG_REG29__pipe1_inner2_rtr_MASK 0x10000000
+#define VGT_DEBUG_REG29__pipe1_inner2_rtr__SHIFT 0x1c
+#define VGT_DEBUG_REG29__pipe1_patch_rtr_MASK 0x20000000
+#define VGT_DEBUG_REG29__pipe1_patch_rtr__SHIFT 0x1d
+#define VGT_DEBUG_REG29__pipe1_edge_rtr_MASK 0x40000000
+#define VGT_DEBUG_REG29__pipe1_edge_rtr__SHIFT 0x1e
+#define VGT_DEBUG_REG29__use_stored_inner_q_ring3_MASK 0x80000000
+#define VGT_DEBUG_REG29__use_stored_inner_q_ring3__SHIFT 0x1f
+#define VGT_DEBUG_REG31__pipe0_dr_MASK 0x1
+#define VGT_DEBUG_REG31__pipe0_dr__SHIFT 0x0
+#define VGT_DEBUG_REG31__pipe0_rtr_MASK 0x2
+#define VGT_DEBUG_REG31__pipe0_rtr__SHIFT 0x1
+#define VGT_DEBUG_REG31__pipe1_outer_dr_MASK 0x4
+#define VGT_DEBUG_REG31__pipe1_outer_dr__SHIFT 0x2
+#define VGT_DEBUG_REG31__pipe1_inner_dr_MASK 0x8
+#define VGT_DEBUG_REG31__pipe1_inner_dr__SHIFT 0x3
+#define VGT_DEBUG_REG31__pipe2_outer_dr_MASK 0x10
+#define VGT_DEBUG_REG31__pipe2_outer_dr__SHIFT 0x4
+#define VGT_DEBUG_REG31__pipe2_inner_dr_MASK 0x20
+#define VGT_DEBUG_REG31__pipe2_inner_dr__SHIFT 0x5
+#define VGT_DEBUG_REG31__pipe3_outer_dr_MASK 0x40
+#define VGT_DEBUG_REG31__pipe3_outer_dr__SHIFT 0x6
+#define VGT_DEBUG_REG31__pipe3_inner_dr_MASK 0x80
+#define VGT_DEBUG_REG31__pipe3_inner_dr__SHIFT 0x7
+#define VGT_DEBUG_REG31__pipe4_outer_dr_MASK 0x100
+#define VGT_DEBUG_REG31__pipe4_outer_dr__SHIFT 0x8
+#define VGT_DEBUG_REG31__pipe4_inner_dr_MASK 0x200
+#define VGT_DEBUG_REG31__pipe4_inner_dr__SHIFT 0x9
+#define VGT_DEBUG_REG31__pipe5_outer_dr_MASK 0x400
+#define VGT_DEBUG_REG31__pipe5_outer_dr__SHIFT 0xa
+#define VGT_DEBUG_REG31__pipe5_inner_dr_MASK 0x800
+#define VGT_DEBUG_REG31__pipe5_inner_dr__SHIFT 0xb
+#define VGT_DEBUG_REG31__pipe2_outer_rtr_MASK 0x1000
+#define VGT_DEBUG_REG31__pipe2_outer_rtr__SHIFT 0xc
+#define VGT_DEBUG_REG31__pipe2_inner_rtr_MASK 0x2000
+#define VGT_DEBUG_REG31__pipe2_inner_rtr__SHIFT 0xd
+#define VGT_DEBUG_REG31__pipe3_outer_rtr_MASK 0x4000
+#define VGT_DEBUG_REG31__pipe3_outer_rtr__SHIFT 0xe
+#define VGT_DEBUG_REG31__pipe3_inner_rtr_MASK 0x8000
+#define VGT_DEBUG_REG31__pipe3_inner_rtr__SHIFT 0xf
+#define VGT_DEBUG_REG31__pipe4_outer_rtr_MASK 0x10000
+#define VGT_DEBUG_REG31__pipe4_outer_rtr__SHIFT 0x10
+#define VGT_DEBUG_REG31__pipe4_inner_rtr_MASK 0x20000
+#define VGT_DEBUG_REG31__pipe4_inner_rtr__SHIFT 0x11
+#define VGT_DEBUG_REG31__pipe5_outer_rtr_MASK 0x40000
+#define VGT_DEBUG_REG31__pipe5_outer_rtr__SHIFT 0x12
+#define VGT_DEBUG_REG31__pipe5_inner_rtr_MASK 0x80000
+#define VGT_DEBUG_REG31__pipe5_inner_rtr__SHIFT 0x13
+#define VGT_DEBUG_REG31__pg_con_outer_point1_rts_MASK 0x100000
+#define VGT_DEBUG_REG31__pg_con_outer_point1_rts__SHIFT 0x14
+#define VGT_DEBUG_REG31__pg_con_outer_point2_rts_MASK 0x200000
+#define VGT_DEBUG_REG31__pg_con_outer_point2_rts__SHIFT 0x15
+#define VGT_DEBUG_REG31__pg_con_inner_point1_rts_MASK 0x400000
+#define VGT_DEBUG_REG31__pg_con_inner_point1_rts__SHIFT 0x16
+#define VGT_DEBUG_REG31__pg_con_inner_point2_rts_MASK 0x800000
+#define VGT_DEBUG_REG31__pg_con_inner_point2_rts__SHIFT 0x17
+#define VGT_DEBUG_REG31__pg_patch_fifo_empty_MASK 0x1000000
+#define VGT_DEBUG_REG31__pg_patch_fifo_empty__SHIFT 0x18
+#define VGT_DEBUG_REG31__pg_edge_fifo_empty_MASK 0x2000000
+#define VGT_DEBUG_REG31__pg_edge_fifo_empty__SHIFT 0x19
+#define VGT_DEBUG_REG31__pg_inner3_perp_fifo_empty_MASK 0x4000000
+#define VGT_DEBUG_REG31__pg_inner3_perp_fifo_empty__SHIFT 0x1a
+#define VGT_DEBUG_REG31__pg_patch_fifo_full_MASK 0x8000000
+#define VGT_DEBUG_REG31__pg_patch_fifo_full__SHIFT 0x1b
+#define VGT_DEBUG_REG31__pg_edge_fifo_full_MASK 0x10000000
+#define VGT_DEBUG_REG31__pg_edge_fifo_full__SHIFT 0x1c
+#define VGT_DEBUG_REG31__pg_inner_perp_fifo_full_MASK 0x20000000
+#define VGT_DEBUG_REG31__pg_inner_perp_fifo_full__SHIFT 0x1d
+#define VGT_DEBUG_REG31__outer_ring_done_q_MASK 0x40000000
+#define VGT_DEBUG_REG31__outer_ring_done_q__SHIFT 0x1e
+#define VGT_DEBUG_REG31__inner_ring_done_q_MASK 0x80000000
+#define VGT_DEBUG_REG31__inner_ring_done_q__SHIFT 0x1f
+#define VGT_DEBUG_REG32__first_ring_of_patch_MASK 0x1
+#define VGT_DEBUG_REG32__first_ring_of_patch__SHIFT 0x0
+#define VGT_DEBUG_REG32__last_ring_of_patch_MASK 0x2
+#define VGT_DEBUG_REG32__last_ring_of_patch__SHIFT 0x1
+#define VGT_DEBUG_REG32__last_edge_of_outer_ring_MASK 0x4
+#define VGT_DEBUG_REG32__last_edge_of_outer_ring__SHIFT 0x2
+#define VGT_DEBUG_REG32__last_point_of_outer_edge_MASK 0x8
+#define VGT_DEBUG_REG32__last_point_of_outer_edge__SHIFT 0x3
+#define VGT_DEBUG_REG32__last_edge_of_inner_ring_MASK 0x10
+#define VGT_DEBUG_REG32__last_edge_of_inner_ring__SHIFT 0x4
+#define VGT_DEBUG_REG32__last_point_of_inner_edge_MASK 0x20
+#define VGT_DEBUG_REG32__last_point_of_inner_edge__SHIFT 0x5
+#define VGT_DEBUG_REG32__last_patch_of_tg_p0_q_MASK 0x40
+#define VGT_DEBUG_REG32__last_patch_of_tg_p0_q__SHIFT 0x6
+#define VGT_DEBUG_REG32__event_null_special_p0_q_MASK 0x80
+#define VGT_DEBUG_REG32__event_null_special_p0_q__SHIFT 0x7
+#define VGT_DEBUG_REG32__event_flag_p5_q_MASK 0x100
+#define VGT_DEBUG_REG32__event_flag_p5_q__SHIFT 0x8
+#define VGT_DEBUG_REG32__first_point_of_patch_p5_q_MASK 0x200
+#define VGT_DEBUG_REG32__first_point_of_patch_p5_q__SHIFT 0x9
+#define VGT_DEBUG_REG32__first_point_of_edge_p5_q_MASK 0x400
+#define VGT_DEBUG_REG32__first_point_of_edge_p5_q__SHIFT 0xa
+#define VGT_DEBUG_REG32__last_patch_of_tg_p5_q_MASK 0x800
+#define VGT_DEBUG_REG32__last_patch_of_tg_p5_q__SHIFT 0xb
+#define VGT_DEBUG_REG32__tess_topology_p5_q_MASK 0x3000
+#define VGT_DEBUG_REG32__tess_topology_p5_q__SHIFT 0xc
+#define VGT_DEBUG_REG32__pipe5_inner3_rtr_MASK 0x4000
+#define VGT_DEBUG_REG32__pipe5_inner3_rtr__SHIFT 0xe
+#define VGT_DEBUG_REG32__pipe5_inner2_rtr_MASK 0x8000
+#define VGT_DEBUG_REG32__pipe5_inner2_rtr__SHIFT 0xf
+#define VGT_DEBUG_REG32__pg_edge_fifo3_full_MASK 0x10000
+#define VGT_DEBUG_REG32__pg_edge_fifo3_full__SHIFT 0x10
+#define VGT_DEBUG_REG32__pg_edge_fifo2_full_MASK 0x20000
+#define VGT_DEBUG_REG32__pg_edge_fifo2_full__SHIFT 0x11
+#define VGT_DEBUG_REG32__pg_inner3_point_fifo_full_MASK 0x40000
+#define VGT_DEBUG_REG32__pg_inner3_point_fifo_full__SHIFT 0x12
+#define VGT_DEBUG_REG32__pg_outer3_point_fifo_full_MASK 0x80000
+#define VGT_DEBUG_REG32__pg_outer3_point_fifo_full__SHIFT 0x13
+#define VGT_DEBUG_REG32__pg_inner2_point_fifo_full_MASK 0x100000
+#define VGT_DEBUG_REG32__pg_inner2_point_fifo_full__SHIFT 0x14
+#define VGT_DEBUG_REG32__pg_outer2_point_fifo_full_MASK 0x200000
+#define VGT_DEBUG_REG32__pg_outer2_point_fifo_full__SHIFT 0x15
+#define VGT_DEBUG_REG32__pg_inner_point_fifo_full_MASK 0x400000
+#define VGT_DEBUG_REG32__pg_inner_point_fifo_full__SHIFT 0x16
+#define VGT_DEBUG_REG32__pg_outer_point_fifo_full_MASK 0x800000
+#define VGT_DEBUG_REG32__pg_outer_point_fifo_full__SHIFT 0x17
+#define VGT_DEBUG_REG32__inner2_fifos_rtr_MASK 0x1000000
+#define VGT_DEBUG_REG32__inner2_fifos_rtr__SHIFT 0x18
+#define VGT_DEBUG_REG32__inner_fifos_rtr_MASK 0x2000000
+#define VGT_DEBUG_REG32__inner_fifos_rtr__SHIFT 0x19
+#define VGT_DEBUG_REG32__outer_fifos_rtr_MASK 0x4000000
+#define VGT_DEBUG_REG32__outer_fifos_rtr__SHIFT 0x1a
+#define VGT_DEBUG_REG32__fifos_rtr_MASK 0x8000000
+#define VGT_DEBUG_REG32__fifos_rtr__SHIFT 0x1b
+#define VGT_DEBUG_REG32__SPARE_MASK 0xf0000000
+#define VGT_DEBUG_REG32__SPARE__SHIFT 0x1c
+#define VGT_DEBUG_REG33__pipe0_patch_dr_MASK 0x1
+#define VGT_DEBUG_REG33__pipe0_patch_dr__SHIFT 0x0
+#define VGT_DEBUG_REG33__ring3_pipe1_dr_MASK 0x2
+#define VGT_DEBUG_REG33__ring3_pipe1_dr__SHIFT 0x1
+#define VGT_DEBUG_REG33__pipe1_dr_MASK 0x4
+#define VGT_DEBUG_REG33__pipe1_dr__SHIFT 0x2
+#define VGT_DEBUG_REG33__pipe2_dr_MASK 0x8
+#define VGT_DEBUG_REG33__pipe2_dr__SHIFT 0x3
+#define VGT_DEBUG_REG33__pipe0_patch_rtr_MASK 0x10
+#define VGT_DEBUG_REG33__pipe0_patch_rtr__SHIFT 0x4
+#define VGT_DEBUG_REG33__ring2_pipe1_dr_MASK 0x20
+#define VGT_DEBUG_REG33__ring2_pipe1_dr__SHIFT 0x5
+#define VGT_DEBUG_REG33__ring1_pipe1_dr_MASK 0x40
+#define VGT_DEBUG_REG33__ring1_pipe1_dr__SHIFT 0x6
+#define VGT_DEBUG_REG33__pipe2_rtr_MASK 0x80
+#define VGT_DEBUG_REG33__pipe2_rtr__SHIFT 0x7
+#define VGT_DEBUG_REG33__pipe3_dr_MASK 0x100
+#define VGT_DEBUG_REG33__pipe3_dr__SHIFT 0x8
+#define VGT_DEBUG_REG33__pipe3_rtr_MASK 0x200
+#define VGT_DEBUG_REG33__pipe3_rtr__SHIFT 0x9
+#define VGT_DEBUG_REG33__ring2_in_sync_q_MASK 0x400
+#define VGT_DEBUG_REG33__ring2_in_sync_q__SHIFT 0xa
+#define VGT_DEBUG_REG33__ring1_in_sync_q_MASK 0x800
+#define VGT_DEBUG_REG33__ring1_in_sync_q__SHIFT 0xb
+#define VGT_DEBUG_REG33__pipe1_patch_rtr_MASK 0x1000
+#define VGT_DEBUG_REG33__pipe1_patch_rtr__SHIFT 0xc
+#define VGT_DEBUG_REG33__ring3_in_sync_q_MASK 0x2000
+#define VGT_DEBUG_REG33__ring3_in_sync_q__SHIFT 0xd
+#define VGT_DEBUG_REG33__tm_te11_event_rtr_MASK 0x4000
+#define VGT_DEBUG_REG33__tm_te11_event_rtr__SHIFT 0xe
+#define VGT_DEBUG_REG33__first_prim_of_patch_q_MASK 0x8000
+#define VGT_DEBUG_REG33__first_prim_of_patch_q__SHIFT 0xf
+#define VGT_DEBUG_REG33__con_prim_fifo_full_MASK 0x10000
+#define VGT_DEBUG_REG33__con_prim_fifo_full__SHIFT 0x10
+#define VGT_DEBUG_REG33__con_vert_fifo_full_MASK 0x20000
+#define VGT_DEBUG_REG33__con_vert_fifo_full__SHIFT 0x11
+#define VGT_DEBUG_REG33__con_prim_fifo_empty_MASK 0x40000
+#define VGT_DEBUG_REG33__con_prim_fifo_empty__SHIFT 0x12
+#define VGT_DEBUG_REG33__con_vert_fifo_empty_MASK 0x80000
+#define VGT_DEBUG_REG33__con_vert_fifo_empty__SHIFT 0x13
+#define VGT_DEBUG_REG33__last_patch_of_tg_p0_q_MASK 0x100000
+#define VGT_DEBUG_REG33__last_patch_of_tg_p0_q__SHIFT 0x14
+#define VGT_DEBUG_REG33__ring3_valid_p2_MASK 0x200000
+#define VGT_DEBUG_REG33__ring3_valid_p2__SHIFT 0x15
+#define VGT_DEBUG_REG33__ring2_valid_p2_MASK 0x400000
+#define VGT_DEBUG_REG33__ring2_valid_p2__SHIFT 0x16
+#define VGT_DEBUG_REG33__ring1_valid_p2_MASK 0x800000
+#define VGT_DEBUG_REG33__ring1_valid_p2__SHIFT 0x17
+#define VGT_DEBUG_REG33__tess_type_p0_q_MASK 0x3000000
+#define VGT_DEBUG_REG33__tess_type_p0_q__SHIFT 0x18
+#define VGT_DEBUG_REG33__tess_topology_p0_q_MASK 0xc000000
+#define VGT_DEBUG_REG33__tess_topology_p0_q__SHIFT 0x1a
+#define VGT_DEBUG_REG33__te11_out_vert_gs_en_MASK 0x10000000
+#define VGT_DEBUG_REG33__te11_out_vert_gs_en__SHIFT 0x1c
+#define VGT_DEBUG_REG33__con_ring3_busy_MASK 0x20000000
+#define VGT_DEBUG_REG33__con_ring3_busy__SHIFT 0x1d
+#define VGT_DEBUG_REG33__con_ring2_busy_MASK 0x40000000
+#define VGT_DEBUG_REG33__con_ring2_busy__SHIFT 0x1e
+#define VGT_DEBUG_REG33__con_ring1_busy_MASK 0x80000000
+#define VGT_DEBUG_REG33__con_ring1_busy__SHIFT 0x1f
+#define VGT_DEBUG_REG34__con_state_q_MASK 0xf
+#define VGT_DEBUG_REG34__con_state_q__SHIFT 0x0
+#define VGT_DEBUG_REG34__second_cycle_q_MASK 0x10
+#define VGT_DEBUG_REG34__second_cycle_q__SHIFT 0x4
+#define VGT_DEBUG_REG34__process_tri_middle_p0_q_MASK 0x20
+#define VGT_DEBUG_REG34__process_tri_middle_p0_q__SHIFT 0x5
+#define VGT_DEBUG_REG34__process_tri_1st_2nd_half_p0_q_MASK 0x40
+#define VGT_DEBUG_REG34__process_tri_1st_2nd_half_p0_q__SHIFT 0x6
+#define VGT_DEBUG_REG34__process_tri_center_poly_p0_q_MASK 0x80
+#define VGT_DEBUG_REG34__process_tri_center_poly_p0_q__SHIFT 0x7
+#define VGT_DEBUG_REG34__pipe0_patch_dr_MASK 0x100
+#define VGT_DEBUG_REG34__pipe0_patch_dr__SHIFT 0x8
+#define VGT_DEBUG_REG34__pipe0_edge_dr_MASK 0x200
+#define VGT_DEBUG_REG34__pipe0_edge_dr__SHIFT 0x9
+#define VGT_DEBUG_REG34__pipe1_dr_MASK 0x400
+#define VGT_DEBUG_REG34__pipe1_dr__SHIFT 0xa
+#define VGT_DEBUG_REG34__pipe0_patch_rtr_MASK 0x800
+#define VGT_DEBUG_REG34__pipe0_patch_rtr__SHIFT 0xb
+#define VGT_DEBUG_REG34__pipe0_edge_rtr_MASK 0x1000
+#define VGT_DEBUG_REG34__pipe0_edge_rtr__SHIFT 0xc
+#define VGT_DEBUG_REG34__pipe1_rtr_MASK 0x2000
+#define VGT_DEBUG_REG34__pipe1_rtr__SHIFT 0xd
+#define VGT_DEBUG_REG34__outer_parity_p0_q_MASK 0x4000
+#define VGT_DEBUG_REG34__outer_parity_p0_q__SHIFT 0xe
+#define VGT_DEBUG_REG34__parallel_parity_p0_q_MASK 0x8000
+#define VGT_DEBUG_REG34__parallel_parity_p0_q__SHIFT 0xf
+#define VGT_DEBUG_REG34__first_ring_of_patch_p0_q_MASK 0x10000
+#define VGT_DEBUG_REG34__first_ring_of_patch_p0_q__SHIFT 0x10
+#define VGT_DEBUG_REG34__last_ring_of_patch_p0_q_MASK 0x20000
+#define VGT_DEBUG_REG34__last_ring_of_patch_p0_q__SHIFT 0x11
+#define VGT_DEBUG_REG34__last_edge_of_outer_ring_p0_q_MASK 0x40000
+#define VGT_DEBUG_REG34__last_edge_of_outer_ring_p0_q__SHIFT 0x12
+#define VGT_DEBUG_REG34__last_point_of_outer_ring_p1_MASK 0x80000
+#define VGT_DEBUG_REG34__last_point_of_outer_ring_p1__SHIFT 0x13
+#define VGT_DEBUG_REG34__last_point_of_inner_ring_p1_MASK 0x100000
+#define VGT_DEBUG_REG34__last_point_of_inner_ring_p1__SHIFT 0x14
+#define VGT_DEBUG_REG34__outer_edge_tf_eq_one_p0_q_MASK 0x200000
+#define VGT_DEBUG_REG34__outer_edge_tf_eq_one_p0_q__SHIFT 0x15
+#define VGT_DEBUG_REG34__advance_outer_point_p1_MASK 0x400000
+#define VGT_DEBUG_REG34__advance_outer_point_p1__SHIFT 0x16
+#define VGT_DEBUG_REG34__advance_inner_point_p1_MASK 0x800000
+#define VGT_DEBUG_REG34__advance_inner_point_p1__SHIFT 0x17
+#define VGT_DEBUG_REG34__next_ring_is_rect_p0_q_MASK 0x1000000
+#define VGT_DEBUG_REG34__next_ring_is_rect_p0_q__SHIFT 0x18
+#define VGT_DEBUG_REG34__pipe1_outer1_rtr_MASK 0x2000000
+#define VGT_DEBUG_REG34__pipe1_outer1_rtr__SHIFT 0x19
+#define VGT_DEBUG_REG34__pipe1_outer2_rtr_MASK 0x4000000
+#define VGT_DEBUG_REG34__pipe1_outer2_rtr__SHIFT 0x1a
+#define VGT_DEBUG_REG34__pipe1_inner1_rtr_MASK 0x8000000
+#define VGT_DEBUG_REG34__pipe1_inner1_rtr__SHIFT 0x1b
+#define VGT_DEBUG_REG34__pipe1_inner2_rtr_MASK 0x10000000
+#define VGT_DEBUG_REG34__pipe1_inner2_rtr__SHIFT 0x1c
+#define VGT_DEBUG_REG34__pipe1_patch_rtr_MASK 0x20000000
+#define VGT_DEBUG_REG34__pipe1_patch_rtr__SHIFT 0x1d
+#define VGT_DEBUG_REG34__pipe1_edge_rtr_MASK 0x40000000
+#define VGT_DEBUG_REG34__pipe1_edge_rtr__SHIFT 0x1e
+#define VGT_DEBUG_REG34__use_stored_inner_q_ring1_MASK 0x80000000
+#define VGT_DEBUG_REG34__use_stored_inner_q_ring1__SHIFT 0x1f
+#define VGT_DEBUG_REG36__VGT_PA_clipp_eop_MASK 0xffffffff
+#define VGT_DEBUG_REG36__VGT_PA_clipp_eop__SHIFT 0x0
+#define VGT_PERFCOUNTER_SEID_MASK__PERF_SEID_IGNORE_MASK_MASK 0xff
+#define VGT_PERFCOUNTER_SEID_MASK__PERF_SEID_IGNORE_MASK__SHIFT 0x0
+#define VGT_PERFCOUNTER0_SELECT__PERF_SEL_MASK 0x3ff
+#define VGT_PERFCOUNTER0_SELECT__PERF_SEL__SHIFT 0x0
+#define VGT_PERFCOUNTER0_SELECT__PERF_SEL1_MASK 0xffc00
+#define VGT_PERFCOUNTER0_SELECT__PERF_SEL1__SHIFT 0xa
+#define VGT_PERFCOUNTER0_SELECT__CNTR_MODE_MASK 0xf00000
+#define VGT_PERFCOUNTER0_SELECT__CNTR_MODE__SHIFT 0x14
+#define VGT_PERFCOUNTER0_SELECT__PERF_MODE1_MASK 0xf000000
+#define VGT_PERFCOUNTER0_SELECT__PERF_MODE1__SHIFT 0x18
+#define VGT_PERFCOUNTER0_SELECT__PERF_MODE_MASK 0xf0000000
+#define VGT_PERFCOUNTER0_SELECT__PERF_MODE__SHIFT 0x1c
+#define VGT_PERFCOUNTER1_SELECT__PERF_SEL_MASK 0x3ff
+#define VGT_PERFCOUNTER1_SELECT__PERF_SEL__SHIFT 0x0
+#define VGT_PERFCOUNTER1_SELECT__PERF_SEL1_MASK 0xffc00
+#define VGT_PERFCOUNTER1_SELECT__PERF_SEL1__SHIFT 0xa
+#define VGT_PERFCOUNTER1_SELECT__CNTR_MODE_MASK 0xf00000
+#define VGT_PERFCOUNTER1_SELECT__CNTR_MODE__SHIFT 0x14
+#define VGT_PERFCOUNTER1_SELECT__PERF_MODE1_MASK 0xf000000
+#define VGT_PERFCOUNTER1_SELECT__PERF_MODE1__SHIFT 0x18
+#define VGT_PERFCOUNTER1_SELECT__PERF_MODE_MASK 0xf0000000
+#define VGT_PERFCOUNTER1_SELECT__PERF_MODE__SHIFT 0x1c
+#define VGT_PERFCOUNTER2_SELECT__PERF_SEL_MASK 0xff
+#define VGT_PERFCOUNTER2_SELECT__PERF_SEL__SHIFT 0x0
+#define VGT_PERFCOUNTER2_SELECT__PERF_MODE_MASK 0xf0000000
+#define VGT_PERFCOUNTER2_SELECT__PERF_MODE__SHIFT 0x1c
+#define VGT_PERFCOUNTER3_SELECT__PERF_SEL_MASK 0xff
+#define VGT_PERFCOUNTER3_SELECT__PERF_SEL__SHIFT 0x0
+#define VGT_PERFCOUNTER3_SELECT__PERF_MODE_MASK 0xf0000000
+#define VGT_PERFCOUNTER3_SELECT__PERF_MODE__SHIFT 0x1c
+#define VGT_PERFCOUNTER0_SELECT1__PERF_SEL2_MASK 0x3ff
+#define VGT_PERFCOUNTER0_SELECT1__PERF_SEL2__SHIFT 0x0
+#define VGT_PERFCOUNTER0_SELECT1__PERF_SEL3_MASK 0xffc00
+#define VGT_PERFCOUNTER0_SELECT1__PERF_SEL3__SHIFT 0xa
+#define VGT_PERFCOUNTER0_SELECT1__PERF_MODE3_MASK 0xf000000
+#define VGT_PERFCOUNTER0_SELECT1__PERF_MODE3__SHIFT 0x18
+#define VGT_PERFCOUNTER0_SELECT1__PERF_MODE2_MASK 0xf0000000
+#define VGT_PERFCOUNTER0_SELECT1__PERF_MODE2__SHIFT 0x1c
+#define VGT_PERFCOUNTER1_SELECT1__PERF_SEL2_MASK 0x3ff
+#define VGT_PERFCOUNTER1_SELECT1__PERF_SEL2__SHIFT 0x0
+#define VGT_PERFCOUNTER1_SELECT1__PERF_SEL3_MASK 0xffc00
+#define VGT_PERFCOUNTER1_SELECT1__PERF_SEL3__SHIFT 0xa
+#define VGT_PERFCOUNTER1_SELECT1__PERF_MODE3_MASK 0xf000000
+#define VGT_PERFCOUNTER1_SELECT1__PERF_MODE3__SHIFT 0x18
+#define VGT_PERFCOUNTER1_SELECT1__PERF_MODE2_MASK 0xf0000000
+#define VGT_PERFCOUNTER1_SELECT1__PERF_MODE2__SHIFT 0x1c
+#define VGT_PERFCOUNTER0_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define VGT_PERFCOUNTER0_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define VGT_PERFCOUNTER1_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define VGT_PERFCOUNTER1_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define VGT_PERFCOUNTER2_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define VGT_PERFCOUNTER2_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define VGT_PERFCOUNTER3_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define VGT_PERFCOUNTER3_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define VGT_PERFCOUNTER0_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define VGT_PERFCOUNTER0_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define VGT_PERFCOUNTER1_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define VGT_PERFCOUNTER1_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define VGT_PERFCOUNTER2_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define VGT_PERFCOUNTER2_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define VGT_PERFCOUNTER3_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define VGT_PERFCOUNTER3_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define IA_PERFCOUNTER0_SELECT__PERF_SEL_MASK 0x3ff
+#define IA_PERFCOUNTER0_SELECT__PERF_SEL__SHIFT 0x0
+#define IA_PERFCOUNTER0_SELECT__PERF_SEL1_MASK 0xffc00
+#define IA_PERFCOUNTER0_SELECT__PERF_SEL1__SHIFT 0xa
+#define IA_PERFCOUNTER0_SELECT__CNTR_MODE_MASK 0xf00000
+#define IA_PERFCOUNTER0_SELECT__CNTR_MODE__SHIFT 0x14
+#define IA_PERFCOUNTER0_SELECT__PERF_MODE1_MASK 0xf000000
+#define IA_PERFCOUNTER0_SELECT__PERF_MODE1__SHIFT 0x18
+#define IA_PERFCOUNTER0_SELECT__PERF_MODE_MASK 0xf0000000
+#define IA_PERFCOUNTER0_SELECT__PERF_MODE__SHIFT 0x1c
+#define IA_PERFCOUNTER1_SELECT__PERF_SEL_MASK 0xff
+#define IA_PERFCOUNTER1_SELECT__PERF_SEL__SHIFT 0x0
+#define IA_PERFCOUNTER1_SELECT__PERF_MODE_MASK 0xf0000000
+#define IA_PERFCOUNTER1_SELECT__PERF_MODE__SHIFT 0x1c
+#define IA_PERFCOUNTER2_SELECT__PERF_SEL_MASK 0xff
+#define IA_PERFCOUNTER2_SELECT__PERF_SEL__SHIFT 0x0
+#define IA_PERFCOUNTER2_SELECT__PERF_MODE_MASK 0xf0000000
+#define IA_PERFCOUNTER2_SELECT__PERF_MODE__SHIFT 0x1c
+#define IA_PERFCOUNTER3_SELECT__PERF_SEL_MASK 0xff
+#define IA_PERFCOUNTER3_SELECT__PERF_SEL__SHIFT 0x0
+#define IA_PERFCOUNTER3_SELECT__PERF_MODE_MASK 0xf0000000
+#define IA_PERFCOUNTER3_SELECT__PERF_MODE__SHIFT 0x1c
+#define IA_PERFCOUNTER0_SELECT1__PERF_SEL2_MASK 0x3ff
+#define IA_PERFCOUNTER0_SELECT1__PERF_SEL2__SHIFT 0x0
+#define IA_PERFCOUNTER0_SELECT1__PERF_SEL3_MASK 0xffc00
+#define IA_PERFCOUNTER0_SELECT1__PERF_SEL3__SHIFT 0xa
+#define IA_PERFCOUNTER0_SELECT1__PERF_MODE3_MASK 0xf000000
+#define IA_PERFCOUNTER0_SELECT1__PERF_MODE3__SHIFT 0x18
+#define IA_PERFCOUNTER0_SELECT1__PERF_MODE2_MASK 0xf0000000
+#define IA_PERFCOUNTER0_SELECT1__PERF_MODE2__SHIFT 0x1c
+#define IA_PERFCOUNTER0_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define IA_PERFCOUNTER0_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define IA_PERFCOUNTER1_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define IA_PERFCOUNTER1_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define IA_PERFCOUNTER2_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define IA_PERFCOUNTER2_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define IA_PERFCOUNTER3_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define IA_PERFCOUNTER3_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define IA_PERFCOUNTER0_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define IA_PERFCOUNTER0_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define IA_PERFCOUNTER1_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define IA_PERFCOUNTER1_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define IA_PERFCOUNTER2_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define IA_PERFCOUNTER2_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define IA_PERFCOUNTER3_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define IA_PERFCOUNTER3_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define WD_PERFCOUNTER0_SELECT__PERF_SEL_MASK 0xff
+#define WD_PERFCOUNTER0_SELECT__PERF_SEL__SHIFT 0x0
+#define WD_PERFCOUNTER0_SELECT__PERF_MODE_MASK 0xf0000000
+#define WD_PERFCOUNTER0_SELECT__PERF_MODE__SHIFT 0x1c
+#define WD_PERFCOUNTER1_SELECT__PERF_SEL_MASK 0xff
+#define WD_PERFCOUNTER1_SELECT__PERF_SEL__SHIFT 0x0
+#define WD_PERFCOUNTER1_SELECT__PERF_MODE_MASK 0xf0000000
+#define WD_PERFCOUNTER1_SELECT__PERF_MODE__SHIFT 0x1c
+#define WD_PERFCOUNTER2_SELECT__PERF_SEL_MASK 0xff
+#define WD_PERFCOUNTER2_SELECT__PERF_SEL__SHIFT 0x0
+#define WD_PERFCOUNTER2_SELECT__PERF_MODE_MASK 0xf0000000
+#define WD_PERFCOUNTER2_SELECT__PERF_MODE__SHIFT 0x1c
+#define WD_PERFCOUNTER3_SELECT__PERF_SEL_MASK 0xff
+#define WD_PERFCOUNTER3_SELECT__PERF_SEL__SHIFT 0x0
+#define WD_PERFCOUNTER3_SELECT__PERF_MODE_MASK 0xf0000000
+#define WD_PERFCOUNTER3_SELECT__PERF_MODE__SHIFT 0x1c
+#define WD_PERFCOUNTER0_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define WD_PERFCOUNTER0_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define WD_PERFCOUNTER1_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define WD_PERFCOUNTER1_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define WD_PERFCOUNTER2_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define WD_PERFCOUNTER2_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define WD_PERFCOUNTER3_LO__PERFCOUNTER_LO_MASK 0xffffffff
+#define WD_PERFCOUNTER3_LO__PERFCOUNTER_LO__SHIFT 0x0
+#define WD_PERFCOUNTER0_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define WD_PERFCOUNTER0_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define WD_PERFCOUNTER1_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define WD_PERFCOUNTER1_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define WD_PERFCOUNTER2_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define WD_PERFCOUNTER2_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define WD_PERFCOUNTER3_HI__PERFCOUNTER_HI_MASK 0xffffffff
+#define WD_PERFCOUNTER3_HI__PERFCOUNTER_HI__SHIFT 0x0
+#define DIDT_IND_INDEX__DIDT_IND_INDEX_MASK 0xffffffff
+#define DIDT_IND_INDEX__DIDT_IND_INDEX__SHIFT 0x0
+#define DIDT_IND_DATA__DIDT_IND_DATA_MASK 0xffffffff
+#define DIDT_IND_DATA__DIDT_IND_DATA__SHIFT 0x0
+#define DIDT_SQ_CTRL0__DIDT_CTRL_EN_MASK 0x1
+#define DIDT_SQ_CTRL0__DIDT_CTRL_EN__SHIFT 0x0
+#define DIDT_SQ_CTRL0__USE_REF_CLOCK_MASK 0x2
+#define DIDT_SQ_CTRL0__USE_REF_CLOCK__SHIFT 0x1
+#define DIDT_SQ_CTRL0__PHASE_OFFSET_MASK 0xc
+#define DIDT_SQ_CTRL0__PHASE_OFFSET__SHIFT 0x2
+#define DIDT_SQ_CTRL0__DIDT_CTRL_RST_MASK 0x10
+#define DIDT_SQ_CTRL0__DIDT_CTRL_RST__SHIFT 0x4
+#define DIDT_SQ_CTRL0__DIDT_CLK_EN_OVERRIDE_MASK 0x20
+#define DIDT_SQ_CTRL0__DIDT_CLK_EN_OVERRIDE__SHIFT 0x5
+#define DIDT_SQ_CTRL0__UNUSED_0_MASK 0xffffffc0
+#define DIDT_SQ_CTRL0__UNUSED_0__SHIFT 0x6
+#define DIDT_SQ_CTRL1__MIN_POWER_MASK 0xffff
+#define DIDT_SQ_CTRL1__MIN_POWER__SHIFT 0x0
+#define DIDT_SQ_CTRL1__MAX_POWER_MASK 0xffff0000
+#define DIDT_SQ_CTRL1__MAX_POWER__SHIFT 0x10
+#define DIDT_SQ_CTRL2__MAX_POWER_DELTA_MASK 0x3fff
+#define DIDT_SQ_CTRL2__MAX_POWER_DELTA__SHIFT 0x0
+#define DIDT_SQ_CTRL2__UNUSED_0_MASK 0xc000
+#define DIDT_SQ_CTRL2__UNUSED_0__SHIFT 0xe
+#define DIDT_SQ_CTRL2__SHORT_TERM_INTERVAL_SIZE_MASK 0x3ff0000
+#define DIDT_SQ_CTRL2__SHORT_TERM_INTERVAL_SIZE__SHIFT 0x10
+#define DIDT_SQ_CTRL2__UNUSED_1_MASK 0x4000000
+#define DIDT_SQ_CTRL2__UNUSED_1__SHIFT 0x1a
+#define DIDT_SQ_CTRL2__LONG_TERM_INTERVAL_RATIO_MASK 0x78000000
+#define DIDT_SQ_CTRL2__LONG_TERM_INTERVAL_RATIO__SHIFT 0x1b
+#define DIDT_SQ_CTRL2__UNUSED_2_MASK 0x80000000
+#define DIDT_SQ_CTRL2__UNUSED_2__SHIFT 0x1f
+#define DIDT_SQ_CTRL_OCP__UNUSED_0_MASK 0xffff
+#define DIDT_SQ_CTRL_OCP__UNUSED_0__SHIFT 0x0
+#define DIDT_SQ_CTRL_OCP__OCP_MAX_POWER_MASK 0xffff0000
+#define DIDT_SQ_CTRL_OCP__OCP_MAX_POWER__SHIFT 0x10
+#define DIDT_SQ_WEIGHT0_3__WEIGHT0_MASK 0xff
+#define DIDT_SQ_WEIGHT0_3__WEIGHT0__SHIFT 0x0
+#define DIDT_SQ_WEIGHT0_3__WEIGHT1_MASK 0xff00
+#define DIDT_SQ_WEIGHT0_3__WEIGHT1__SHIFT 0x8
+#define DIDT_SQ_WEIGHT0_3__WEIGHT2_MASK 0xff0000
+#define DIDT_SQ_WEIGHT0_3__WEIGHT2__SHIFT 0x10
+#define DIDT_SQ_WEIGHT0_3__WEIGHT3_MASK 0xff000000
+#define DIDT_SQ_WEIGHT0_3__WEIGHT3__SHIFT 0x18
+#define DIDT_SQ_WEIGHT4_7__WEIGHT4_MASK 0xff
+#define DIDT_SQ_WEIGHT4_7__WEIGHT4__SHIFT 0x0
+#define DIDT_SQ_WEIGHT4_7__WEIGHT5_MASK 0xff00
+#define DIDT_SQ_WEIGHT4_7__WEIGHT5__SHIFT 0x8
+#define DIDT_SQ_WEIGHT4_7__WEIGHT6_MASK 0xff0000
+#define DIDT_SQ_WEIGHT4_7__WEIGHT6__SHIFT 0x10
+#define DIDT_SQ_WEIGHT4_7__WEIGHT7_MASK 0xff000000
+#define DIDT_SQ_WEIGHT4_7__WEIGHT7__SHIFT 0x18
+#define DIDT_SQ_WEIGHT8_11__WEIGHT8_MASK 0xff
+#define DIDT_SQ_WEIGHT8_11__WEIGHT8__SHIFT 0x0
+#define DIDT_SQ_WEIGHT8_11__WEIGHT9_MASK 0xff00
+#define DIDT_SQ_WEIGHT8_11__WEIGHT9__SHIFT 0x8
+#define DIDT_SQ_WEIGHT8_11__WEIGHT10_MASK 0xff0000
+#define DIDT_SQ_WEIGHT8_11__WEIGHT10__SHIFT 0x10
+#define DIDT_SQ_WEIGHT8_11__WEIGHT11_MASK 0xff000000
+#define DIDT_SQ_WEIGHT8_11__WEIGHT11__SHIFT 0x18
+#define DIDT_DB_CTRL0__DIDT_CTRL_EN_MASK 0x1
+#define DIDT_DB_CTRL0__DIDT_CTRL_EN__SHIFT 0x0
+#define DIDT_DB_CTRL0__USE_REF_CLOCK_MASK 0x2
+#define DIDT_DB_CTRL0__USE_REF_CLOCK__SHIFT 0x1
+#define DIDT_DB_CTRL0__PHASE_OFFSET_MASK 0xc
+#define DIDT_DB_CTRL0__PHASE_OFFSET__SHIFT 0x2
+#define DIDT_DB_CTRL0__DIDT_CTRL_RST_MASK 0x10
+#define DIDT_DB_CTRL0__DIDT_CTRL_RST__SHIFT 0x4
+#define DIDT_DB_CTRL0__DIDT_CLK_EN_OVERRIDE_MASK 0x20
+#define DIDT_DB_CTRL0__DIDT_CLK_EN_OVERRIDE__SHIFT 0x5
+#define DIDT_DB_CTRL0__UNUSED_0_MASK 0xffffffc0
+#define DIDT_DB_CTRL0__UNUSED_0__SHIFT 0x6
+#define DIDT_DB_CTRL1__MIN_POWER_MASK 0xffff
+#define DIDT_DB_CTRL1__MIN_POWER__SHIFT 0x0
+#define DIDT_DB_CTRL1__MAX_POWER_MASK 0xffff0000
+#define DIDT_DB_CTRL1__MAX_POWER__SHIFT 0x10
+#define DIDT_DB_CTRL2__MAX_POWER_DELTA_MASK 0x3fff
+#define DIDT_DB_CTRL2__MAX_POWER_DELTA__SHIFT 0x0
+#define DIDT_DB_CTRL2__UNUSED_0_MASK 0xc000
+#define DIDT_DB_CTRL2__UNUSED_0__SHIFT 0xe
+#define DIDT_DB_CTRL2__SHORT_TERM_INTERVAL_SIZE_MASK 0x3ff0000
+#define DIDT_DB_CTRL2__SHORT_TERM_INTERVAL_SIZE__SHIFT 0x10
+#define DIDT_DB_CTRL2__UNUSED_1_MASK 0x4000000
+#define DIDT_DB_CTRL2__UNUSED_1__SHIFT 0x1a
+#define DIDT_DB_CTRL2__LONG_TERM_INTERVAL_RATIO_MASK 0x78000000
+#define DIDT_DB_CTRL2__LONG_TERM_INTERVAL_RATIO__SHIFT 0x1b
+#define DIDT_DB_CTRL2__UNUSED_2_MASK 0x80000000
+#define DIDT_DB_CTRL2__UNUSED_2__SHIFT 0x1f
+#define DIDT_DB_CTRL_OCP__UNUSED_0_MASK 0xffff
+#define DIDT_DB_CTRL_OCP__UNUSED_0__SHIFT 0x0
+#define DIDT_DB_CTRL_OCP__OCP_MAX_POWER_MASK 0xffff0000
+#define DIDT_DB_CTRL_OCP__OCP_MAX_POWER__SHIFT 0x10
+#define DIDT_DB_WEIGHT0_3__WEIGHT0_MASK 0xff
+#define DIDT_DB_WEIGHT0_3__WEIGHT0__SHIFT 0x0
+#define DIDT_DB_WEIGHT0_3__WEIGHT1_MASK 0xff00
+#define DIDT_DB_WEIGHT0_3__WEIGHT1__SHIFT 0x8
+#define DIDT_DB_WEIGHT0_3__WEIGHT2_MASK 0xff0000
+#define DIDT_DB_WEIGHT0_3__WEIGHT2__SHIFT 0x10
+#define DIDT_DB_WEIGHT0_3__WEIGHT3_MASK 0xff000000
+#define DIDT_DB_WEIGHT0_3__WEIGHT3__SHIFT 0x18
+#define DIDT_DB_WEIGHT4_7__WEIGHT4_MASK 0xff
+#define DIDT_DB_WEIGHT4_7__WEIGHT4__SHIFT 0x0
+#define DIDT_DB_WEIGHT4_7__WEIGHT5_MASK 0xff00
+#define DIDT_DB_WEIGHT4_7__WEIGHT5__SHIFT 0x8
+#define DIDT_DB_WEIGHT4_7__WEIGHT6_MASK 0xff0000
+#define DIDT_DB_WEIGHT4_7__WEIGHT6__SHIFT 0x10
+#define DIDT_DB_WEIGHT4_7__WEIGHT7_MASK 0xff000000
+#define DIDT_DB_WEIGHT4_7__WEIGHT7__SHIFT 0x18
+#define DIDT_DB_WEIGHT8_11__WEIGHT8_MASK 0xff
+#define DIDT_DB_WEIGHT8_11__WEIGHT8__SHIFT 0x0
+#define DIDT_DB_WEIGHT8_11__WEIGHT9_MASK 0xff00
+#define DIDT_DB_WEIGHT8_11__WEIGHT9__SHIFT 0x8
+#define DIDT_DB_WEIGHT8_11__WEIGHT10_MASK 0xff0000
+#define DIDT_DB_WEIGHT8_11__WEIGHT10__SHIFT 0x10
+#define DIDT_DB_WEIGHT8_11__WEIGHT11_MASK 0xff000000
+#define DIDT_DB_WEIGHT8_11__WEIGHT11__SHIFT 0x18
+#define DIDT_TD_CTRL0__DIDT_CTRL_EN_MASK 0x1
+#define DIDT_TD_CTRL0__DIDT_CTRL_EN__SHIFT 0x0
+#define DIDT_TD_CTRL0__USE_REF_CLOCK_MASK 0x2
+#define DIDT_TD_CTRL0__USE_REF_CLOCK__SHIFT 0x1
+#define DIDT_TD_CTRL0__PHASE_OFFSET_MASK 0xc
+#define DIDT_TD_CTRL0__PHASE_OFFSET__SHIFT 0x2
+#define DIDT_TD_CTRL0__DIDT_CTRL_RST_MASK 0x10
+#define DIDT_TD_CTRL0__DIDT_CTRL_RST__SHIFT 0x4
+#define DIDT_TD_CTRL0__DIDT_CLK_EN_OVERRIDE_MASK 0x20
+#define DIDT_TD_CTRL0__DIDT_CLK_EN_OVERRIDE__SHIFT 0x5
+#define DIDT_TD_CTRL0__UNUSED_0_MASK 0xffffffc0
+#define DIDT_TD_CTRL0__UNUSED_0__SHIFT 0x6
+#define DIDT_TD_CTRL1__MIN_POWER_MASK 0xffff
+#define DIDT_TD_CTRL1__MIN_POWER__SHIFT 0x0
+#define DIDT_TD_CTRL1__MAX_POWER_MASK 0xffff0000
+#define DIDT_TD_CTRL1__MAX_POWER__SHIFT 0x10
+#define DIDT_TD_CTRL2__MAX_POWER_DELTA_MASK 0x3fff
+#define DIDT_TD_CTRL2__MAX_POWER_DELTA__SHIFT 0x0
+#define DIDT_TD_CTRL2__UNUSED_0_MASK 0xc000
+#define DIDT_TD_CTRL2__UNUSED_0__SHIFT 0xe
+#define DIDT_TD_CTRL2__SHORT_TERM_INTERVAL_SIZE_MASK 0x3ff0000
+#define DIDT_TD_CTRL2__SHORT_TERM_INTERVAL_SIZE__SHIFT 0x10
+#define DIDT_TD_CTRL2__UNUSED_1_MASK 0x4000000
+#define DIDT_TD_CTRL2__UNUSED_1__SHIFT 0x1a
+#define DIDT_TD_CTRL2__LONG_TERM_INTERVAL_RATIO_MASK 0x78000000
+#define DIDT_TD_CTRL2__LONG_TERM_INTERVAL_RATIO__SHIFT 0x1b
+#define DIDT_TD_CTRL2__UNUSED_2_MASK 0x80000000
+#define DIDT_TD_CTRL2__UNUSED_2__SHIFT 0x1f
+#define DIDT_TD_CTRL_OCP__UNUSED_0_MASK 0xffff
+#define DIDT_TD_CTRL_OCP__UNUSED_0__SHIFT 0x0
+#define DIDT_TD_CTRL_OCP__OCP_MAX_POWER_MASK 0xffff0000
+#define DIDT_TD_CTRL_OCP__OCP_MAX_POWER__SHIFT 0x10
+#define DIDT_TD_WEIGHT0_3__WEIGHT0_MASK 0xff
+#define DIDT_TD_WEIGHT0_3__WEIGHT0__SHIFT 0x0
+#define DIDT_TD_WEIGHT0_3__WEIGHT1_MASK 0xff00
+#define DIDT_TD_WEIGHT0_3__WEIGHT1__SHIFT 0x8
+#define DIDT_TD_WEIGHT0_3__WEIGHT2_MASK 0xff0000
+#define DIDT_TD_WEIGHT0_3__WEIGHT2__SHIFT 0x10
+#define DIDT_TD_WEIGHT0_3__WEIGHT3_MASK 0xff000000
+#define DIDT_TD_WEIGHT0_3__WEIGHT3__SHIFT 0x18
+#define DIDT_TD_WEIGHT4_7__WEIGHT4_MASK 0xff
+#define DIDT_TD_WEIGHT4_7__WEIGHT4__SHIFT 0x0
+#define DIDT_TD_WEIGHT4_7__WEIGHT5_MASK 0xff00
+#define DIDT_TD_WEIGHT4_7__WEIGHT5__SHIFT 0x8
+#define DIDT_TD_WEIGHT4_7__WEIGHT6_MASK 0xff0000
+#define DIDT_TD_WEIGHT4_7__WEIGHT6__SHIFT 0x10
+#define DIDT_TD_WEIGHT4_7__WEIGHT7_MASK 0xff000000
+#define DIDT_TD_WEIGHT4_7__WEIGHT7__SHIFT 0x18
+#define DIDT_TD_WEIGHT8_11__WEIGHT8_MASK 0xff
+#define DIDT_TD_WEIGHT8_11__WEIGHT8__SHIFT 0x0
+#define DIDT_TD_WEIGHT8_11__WEIGHT9_MASK 0xff00
+#define DIDT_TD_WEIGHT8_11__WEIGHT9__SHIFT 0x8
+#define DIDT_TD_WEIGHT8_11__WEIGHT10_MASK 0xff0000
+#define DIDT_TD_WEIGHT8_11__WEIGHT10__SHIFT 0x10
+#define DIDT_TD_WEIGHT8_11__WEIGHT11_MASK 0xff000000
+#define DIDT_TD_WEIGHT8_11__WEIGHT11__SHIFT 0x18
+#define DIDT_TCP_CTRL0__DIDT_CTRL_EN_MASK 0x1
+#define DIDT_TCP_CTRL0__DIDT_CTRL_EN__SHIFT 0x0
+#define DIDT_TCP_CTRL0__USE_REF_CLOCK_MASK 0x2
+#define DIDT_TCP_CTRL0__USE_REF_CLOCK__SHIFT 0x1
+#define DIDT_TCP_CTRL0__PHASE_OFFSET_MASK 0xc
+#define DIDT_TCP_CTRL0__PHASE_OFFSET__SHIFT 0x2
+#define DIDT_TCP_CTRL0__DIDT_CTRL_RST_MASK 0x10
+#define DIDT_TCP_CTRL0__DIDT_CTRL_RST__SHIFT 0x4
+#define DIDT_TCP_CTRL0__DIDT_CLK_EN_OVERRIDE_MASK 0x20
+#define DIDT_TCP_CTRL0__DIDT_CLK_EN_OVERRIDE__SHIFT 0x5
+#define DIDT_TCP_CTRL0__UNUSED_0_MASK 0xffffffc0
+#define DIDT_TCP_CTRL0__UNUSED_0__SHIFT 0x6
+#define DIDT_TCP_CTRL1__MIN_POWER_MASK 0xffff
+#define DIDT_TCP_CTRL1__MIN_POWER__SHIFT 0x0
+#define DIDT_TCP_CTRL1__MAX_POWER_MASK 0xffff0000
+#define DIDT_TCP_CTRL1__MAX_POWER__SHIFT 0x10
+#define DIDT_TCP_CTRL2__MAX_POWER_DELTA_MASK 0x3fff
+#define DIDT_TCP_CTRL2__MAX_POWER_DELTA__SHIFT 0x0
+#define DIDT_TCP_CTRL2__UNUSED_0_MASK 0xc000
+#define DIDT_TCP_CTRL2__UNUSED_0__SHIFT 0xe
+#define DIDT_TCP_CTRL2__SHORT_TERM_INTERVAL_SIZE_MASK 0x3ff0000
+#define DIDT_TCP_CTRL2__SHORT_TERM_INTERVAL_SIZE__SHIFT 0x10
+#define DIDT_TCP_CTRL2__UNUSED_1_MASK 0x4000000
+#define DIDT_TCP_CTRL2__UNUSED_1__SHIFT 0x1a
+#define DIDT_TCP_CTRL2__LONG_TERM_INTERVAL_RATIO_MASK 0x78000000
+#define DIDT_TCP_CTRL2__LONG_TERM_INTERVAL_RATIO__SHIFT 0x1b
+#define DIDT_TCP_CTRL2__UNUSED_2_MASK 0x80000000
+#define DIDT_TCP_CTRL2__UNUSED_2__SHIFT 0x1f
+#define DIDT_TCP_CTRL_OCP__UNUSED_0_MASK 0xffff
+#define DIDT_TCP_CTRL_OCP__UNUSED_0__SHIFT 0x0
+#define DIDT_TCP_CTRL_OCP__OCP_MAX_POWER_MASK 0xffff0000
+#define DIDT_TCP_CTRL_OCP__OCP_MAX_POWER__SHIFT 0x10
+#define DIDT_TCP_WEIGHT0_3__WEIGHT0_MASK 0xff
+#define DIDT_TCP_WEIGHT0_3__WEIGHT0__SHIFT 0x0
+#define DIDT_TCP_WEIGHT0_3__WEIGHT1_MASK 0xff00
+#define DIDT_TCP_WEIGHT0_3__WEIGHT1__SHIFT 0x8
+#define DIDT_TCP_WEIGHT0_3__WEIGHT2_MASK 0xff0000
+#define DIDT_TCP_WEIGHT0_3__WEIGHT2__SHIFT 0x10
+#define DIDT_TCP_WEIGHT0_3__WEIGHT3_MASK 0xff000000
+#define DIDT_TCP_WEIGHT0_3__WEIGHT3__SHIFT 0x18
+#define DIDT_TCP_WEIGHT4_7__WEIGHT4_MASK 0xff
+#define DIDT_TCP_WEIGHT4_7__WEIGHT4__SHIFT 0x0
+#define DIDT_TCP_WEIGHT4_7__WEIGHT5_MASK 0xff00
+#define DIDT_TCP_WEIGHT4_7__WEIGHT5__SHIFT 0x8
+#define DIDT_TCP_WEIGHT4_7__WEIGHT6_MASK 0xff0000
+#define DIDT_TCP_WEIGHT4_7__WEIGHT6__SHIFT 0x10
+#define DIDT_TCP_WEIGHT4_7__WEIGHT7_MASK 0xff000000
+#define DIDT_TCP_WEIGHT4_7__WEIGHT7__SHIFT 0x18
+#define DIDT_TCP_WEIGHT8_11__WEIGHT8_MASK 0xff
+#define DIDT_TCP_WEIGHT8_11__WEIGHT8__SHIFT 0x0
+#define DIDT_TCP_WEIGHT8_11__WEIGHT9_MASK 0xff00
+#define DIDT_TCP_WEIGHT8_11__WEIGHT9__SHIFT 0x8
+#define DIDT_TCP_WEIGHT8_11__WEIGHT10_MASK 0xff0000
+#define DIDT_TCP_WEIGHT8_11__WEIGHT10__SHIFT 0x10
+#define DIDT_TCP_WEIGHT8_11__WEIGHT11_MASK 0xff000000
+#define DIDT_TCP_WEIGHT8_11__WEIGHT11__SHIFT 0x18
+#define DIDT_DBR_CTRL0__DIDT_CTRL_EN_MASK 0x1
+#define DIDT_DBR_CTRL0__DIDT_CTRL_EN__SHIFT 0x0
+#define DIDT_DBR_CTRL0__USE_REF_CLOCK_MASK 0x2
+#define DIDT_DBR_CTRL0__USE_REF_CLOCK__SHIFT 0x1
+#define DIDT_DBR_CTRL0__PHASE_OFFSET_MASK 0xc
+#define DIDT_DBR_CTRL0__PHASE_OFFSET__SHIFT 0x2
+#define DIDT_DBR_CTRL0__DIDT_CTRL_RST_MASK 0x10
+#define DIDT_DBR_CTRL0__DIDT_CTRL_RST__SHIFT 0x4
+#define DIDT_DBR_CTRL0__DIDT_CLK_EN_OVERRIDE_MASK 0x20
+#define DIDT_DBR_CTRL0__DIDT_CLK_EN_OVERRIDE__SHIFT 0x5
+#define DIDT_DBR_CTRL0__UNUSED_0_MASK 0xffffffc0
+#define DIDT_DBR_CTRL0__UNUSED_0__SHIFT 0x6
+#define DIDT_DBR_CTRL1__MIN_POWER_MASK 0xffff
+#define DIDT_DBR_CTRL1__MIN_POWER__SHIFT 0x0
+#define DIDT_DBR_CTRL1__MAX_POWER_MASK 0xffff0000
+#define DIDT_DBR_CTRL1__MAX_POWER__SHIFT 0x10
+#define DIDT_DBR_CTRL2__MAX_POWER_DELTA_MASK 0x3fff
+#define DIDT_DBR_CTRL2__MAX_POWER_DELTA__SHIFT 0x0
+#define DIDT_DBR_CTRL2__UNUSED_0_MASK 0xc000
+#define DIDT_DBR_CTRL2__UNUSED_0__SHIFT 0xe
+#define DIDT_DBR_CTRL2__SHORT_TERM_INTERVAL_SIZE_MASK 0x3ff0000
+#define DIDT_DBR_CTRL2__SHORT_TERM_INTERVAL_SIZE__SHIFT 0x10
+#define DIDT_DBR_CTRL2__UNUSED_1_MASK 0x4000000
+#define DIDT_DBR_CTRL2__UNUSED_1__SHIFT 0x1a
+#define DIDT_DBR_CTRL2__LONG_TERM_INTERVAL_RATIO_MASK 0x78000000
+#define DIDT_DBR_CTRL2__LONG_TERM_INTERVAL_RATIO__SHIFT 0x1b
+#define DIDT_DBR_CTRL2__UNUSED_2_MASK 0x80000000
+#define DIDT_DBR_CTRL2__UNUSED_2__SHIFT 0x1f
+#define DIDT_DBR_CTRL_OCP__UNUSED_0_MASK 0xffff
+#define DIDT_DBR_CTRL_OCP__UNUSED_0__SHIFT 0x0
+#define DIDT_DBR_CTRL_OCP__OCP_MAX_POWER_MASK 0xffff0000
+#define DIDT_DBR_CTRL_OCP__OCP_MAX_POWER__SHIFT 0x10
+#define DIDT_DBR_WEIGHT0_3__WEIGHT0_MASK 0xff
+#define DIDT_DBR_WEIGHT0_3__WEIGHT0__SHIFT 0x0
+#define DIDT_DBR_WEIGHT0_3__WEIGHT1_MASK 0xff00
+#define DIDT_DBR_WEIGHT0_3__WEIGHT1__SHIFT 0x8
+#define DIDT_DBR_WEIGHT0_3__WEIGHT2_MASK 0xff0000
+#define DIDT_DBR_WEIGHT0_3__WEIGHT2__SHIFT 0x10
+#define DIDT_DBR_WEIGHT0_3__WEIGHT3_MASK 0xff000000
+#define DIDT_DBR_WEIGHT0_3__WEIGHT3__SHIFT 0x18
+#define DIDT_DBR_WEIGHT4_7__WEIGHT4_MASK 0xff
+#define DIDT_DBR_WEIGHT4_7__WEIGHT4__SHIFT 0x0
+#define DIDT_DBR_WEIGHT4_7__WEIGHT5_MASK 0xff00
+#define DIDT_DBR_WEIGHT4_7__WEIGHT5__SHIFT 0x8
+#define DIDT_DBR_WEIGHT4_7__WEIGHT6_MASK 0xff0000
+#define DIDT_DBR_WEIGHT4_7__WEIGHT6__SHIFT 0x10
+#define DIDT_DBR_WEIGHT4_7__WEIGHT7_MASK 0xff000000
+#define DIDT_DBR_WEIGHT4_7__WEIGHT7__SHIFT 0x18
+#define DIDT_DBR_WEIGHT8_11__WEIGHT8_MASK 0xff
+#define DIDT_DBR_WEIGHT8_11__WEIGHT8__SHIFT 0x0
+#define DIDT_DBR_WEIGHT8_11__WEIGHT9_MASK 0xff00
+#define DIDT_DBR_WEIGHT8_11__WEIGHT9__SHIFT 0x8
+#define DIDT_DBR_WEIGHT8_11__WEIGHT10_MASK 0xff0000
+#define DIDT_DBR_WEIGHT8_11__WEIGHT10__SHIFT 0x10
+#define DIDT_DBR_WEIGHT8_11__WEIGHT11_MASK 0xff000000
+#define DIDT_DBR_WEIGHT8_11__WEIGHT11__SHIFT 0x18
+
+#endif /* GFX_8_1_SH_MASK_H */
-- 
cgit v0.10.2


From fe537d003f9a97c65848e47b3b9acbb0c5002fd9 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Tue, 27 Oct 2015 17:00:17 +0100
Subject: drm/amdgpu: ignore scheduler fences from the same entity
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We are going to submit them before the job anyway.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>

diff --git a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c
index 7fa1d7a..8dd7316 100644
--- a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c
+++ b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c
@@ -222,6 +222,12 @@ amd_sched_entity_pop_job(struct amd_sched_entity *entity)
 
 	while ((entity->dependency = sched->ops->dependency(sched_job))) {
 
+		if (entity->dependency->context == entity->fence_context) {
+			/* We can ignore fences from ourself */
+			fence_put(entity->dependency);
+			continue;
+		}
+
 		if (fence_add_callback(entity->dependency, &entity->cb,
 				       amd_sched_entity_wakeup))
 			fence_put(entity->dependency);
-- 
cgit v0.10.2


From a519435a96597d8cd96123246fea4ae5a6c90b02 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Tue, 20 Oct 2015 16:34:16 +0200
Subject: dma-buf/fence: add fence_wait_any_timeout function v2
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Waiting for the first fence in an array of fences to signal.

This is useful for device driver specific resource managers
and also Vulkan needs something similar.

v2: more parameter checks, handling for timeout==0,
    remove NULL entry support, better callback removal.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>

diff --git a/drivers/dma-buf/fence.c b/drivers/dma-buf/fence.c
index 50ef8bd..7b05dbe 100644
--- a/drivers/dma-buf/fence.c
+++ b/drivers/dma-buf/fence.c
@@ -397,6 +397,104 @@ out:
 }
 EXPORT_SYMBOL(fence_default_wait);
 
+static bool
+fence_test_signaled_any(struct fence **fences, uint32_t count)
+{
+	int i;
+
+	for (i = 0; i < count; ++i) {
+		struct fence *fence = fences[i];
+		if (test_bit(FENCE_FLAG_SIGNALED_BIT, &fence->flags))
+			return true;
+	}
+	return false;
+}
+
+/**
+ * fence_wait_any_timeout - sleep until any fence gets signaled
+ * or until timeout elapses
+ * @fences:	[in]	array of fences to wait on
+ * @count:	[in]	number of fences to wait on
+ * @intr:	[in]	if true, do an interruptible wait
+ * @timeout:	[in]	timeout value in jiffies, or MAX_SCHEDULE_TIMEOUT
+ *
+ * Returns -EINVAL on custom fence wait implementation, -ERESTARTSYS if
+ * interrupted, 0 if the wait timed out, or the remaining timeout in jiffies
+ * on success.
+ *
+ * Synchronous waits for the first fence in the array to be signaled. The
+ * caller needs to hold a reference to all fences in the array, otherwise a
+ * fence might be freed before return, resulting in undefined behavior.
+ */
+signed long
+fence_wait_any_timeout(struct fence **fences, uint32_t count,
+		       bool intr, signed long timeout)
+{
+	struct default_wait_cb *cb;
+	signed long ret = timeout;
+	unsigned i;
+
+	if (WARN_ON(!fences || !count || timeout < 0))
+		return -EINVAL;
+
+	if (timeout == 0) {
+		for (i = 0; i < count; ++i)
+			if (fence_is_signaled(fences[i]))
+				return 1;
+
+		return 0;
+	}
+
+	cb = kcalloc(count, sizeof(struct default_wait_cb), GFP_KERNEL);
+	if (cb == NULL) {
+		ret = -ENOMEM;
+		goto err_free_cb;
+	}
+
+	for (i = 0; i < count; ++i) {
+		struct fence *fence = fences[i];
+
+		if (fence->ops->wait != fence_default_wait) {
+			ret = -EINVAL;
+			goto fence_rm_cb;
+		}
+
+		cb[i].task = current;
+		if (fence_add_callback(fence, &cb[i].base,
+				       fence_default_wait_cb)) {
+			/* This fence is already signaled */
+			goto fence_rm_cb;
+		}
+	}
+
+	while (ret > 0) {
+		if (intr)
+			set_current_state(TASK_INTERRUPTIBLE);
+		else
+			set_current_state(TASK_UNINTERRUPTIBLE);
+
+		if (fence_test_signaled_any(fences, count))
+			break;
+
+		ret = schedule_timeout(ret);
+
+		if (ret > 0 && intr && signal_pending(current))
+			ret = -ERESTARTSYS;
+	}
+
+	__set_current_state(TASK_RUNNING);
+
+fence_rm_cb:
+	while (i-- > 0)
+		fence_remove_callback(fences[i], &cb[i].base);
+
+err_free_cb:
+	kfree(cb);
+
+	return ret;
+}
+EXPORT_SYMBOL(fence_wait_any_timeout);
+
 /**
  * fence_init - Initialize a custom fence.
  * @fence:	[in]	the fence to initialize
diff --git a/include/linux/fence.h b/include/linux/fence.h
index 39efee1..a4084d6 100644
--- a/include/linux/fence.h
+++ b/include/linux/fence.h
@@ -305,7 +305,8 @@ static inline struct fence *fence_later(struct fence *f1, struct fence *f2)
 }
 
 signed long fence_wait_timeout(struct fence *, bool intr, signed long timeout);
-
+signed long fence_wait_any_timeout(struct fence **fences, uint32_t count,
+				   bool intr, signed long timeout);
 
 /**
  * fence_wait - sleep until the fence gets signaled
-- 
cgit v0.10.2


From 6c455ac17bcf4beae6c094a1007b976b60b4bb57 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Wed, 21 Oct 2015 12:58:17 +0200
Subject: dma-buf/fence: add fence_is_later()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Return true when fence 1 is later than fence 2 without
checking if any of them are signaled.

Useful for driver specific resource handling based on fences.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/include/linux/fence.h b/include/linux/fence.h
index a4084d6..bb52201 100644
--- a/include/linux/fence.h
+++ b/include/linux/fence.h
@@ -280,6 +280,22 @@ fence_is_signaled(struct fence *fence)
 }
 
 /**
+ * fence_is_later - return if f1 is chronologically later than f2
+ * @f1:	[in]	the first fence from the same context
+ * @f2:	[in]	the second fence from the same context
+ *
+ * Returns true if f1 is chronologically later than f2. Both fences must be
+ * from the same context, since a seqno is not re-used across contexts.
+ */
+static inline bool fence_is_later(struct fence *f1, struct fence *f2)
+{
+	if (WARN_ON(f1->context != f2->context))
+		return false;
+
+	return f1->seqno - f2->seqno < INT_MAX;
+}
+
+/**
  * fence_later - return the chronologically later fence
  * @f1:	[in]	the first fence from the same context
  * @f2:	[in]	the second fence from the same context
@@ -298,10 +314,10 @@ static inline struct fence *fence_later(struct fence *f1, struct fence *f2)
 	 * set if enable_signaling wasn't called, and enabling that here is
 	 * overkill.
 	 */
-	if (f2->seqno - f1->seqno <= INT_MAX)
-		return fence_is_signaled(f2) ? NULL : f2;
-	else
+	if (fence_is_later(f1, f2))
 		return fence_is_signaled(f1) ? NULL : f1;
+	else
+		return fence_is_signaled(f2) ? NULL : f2;
 }
 
 signed long fence_wait_timeout(struct fence *, bool intr, signed long timeout);
-- 
cgit v0.10.2


From 318cd340c5573a1f021f5b7711893133fe5e8480 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Thu, 15 Oct 2015 17:58:09 +0200
Subject: drm/amdgpu: remove unneeded fence functions
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

amdgpu_fence_default_wait isn't needed any more the default wait does the same
thing and amdgpu_test_signaled is dead as well.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
index 003a219..663caa9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
@@ -822,11 +822,6 @@ static const char *amdgpu_fence_get_timeline_name(struct fence *f)
 	return (const char *)fence->ring->name;
 }
 
-static inline bool amdgpu_test_signaled(struct amdgpu_fence *fence)
-{
-	return test_bit(FENCE_FLAG_SIGNALED_BIT, &fence->base.flags);
-}
-
 static bool amdgpu_test_signaled_any(struct fence **fences, uint32_t count)
 {
 	int idx;
@@ -854,12 +849,6 @@ static void amdgpu_fence_wait_cb(struct fence *fence, struct fence_cb *cb)
 	wake_up_process(wait->task);
 }
 
-static signed long amdgpu_fence_default_wait(struct fence *f, bool intr,
-					     signed long t)
-{
-	return amdgpu_fence_wait_any(&f, 1, intr, t);
-}
-
 /**
  * Wait the fence array with timeout
  *
@@ -936,6 +925,6 @@ const struct fence_ops amdgpu_fence_ops = {
 	.get_timeline_name = amdgpu_fence_get_timeline_name,
 	.enable_signaling = amdgpu_fence_enable_signaling,
 	.signaled = amdgpu_fence_is_signaled,
-	.wait = amdgpu_fence_default_wait,
+	.wait = fence_default_wait,
 	.release = NULL,
 };
-- 
cgit v0.10.2


From ee327caf1a2cdba9313167c36db2d7ff02d534bc Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Tue, 20 Oct 2015 17:38:07 +0200
Subject: drm/amdgpu: switch to common fence_wait_any_timeout v2
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

No need to duplicate the functionality any more.

v2: fix handling if no fence is available.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com> (v1)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index a9c0def..dd7d2ce 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -447,10 +447,6 @@ int amdgpu_fence_wait_next(struct amdgpu_ring *ring);
 int amdgpu_fence_wait_empty(struct amdgpu_ring *ring);
 unsigned amdgpu_fence_count_emitted(struct amdgpu_ring *ring);
 
-signed long amdgpu_fence_wait_any(struct fence **array,
-				  uint32_t count,
-				  bool intr,
-				  signed long t);
 struct amdgpu_fence *amdgpu_fence_ref(struct amdgpu_fence *fence);
 void amdgpu_fence_unref(struct amdgpu_fence **fence);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
index 663caa9..c4bb282 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
@@ -822,104 +822,6 @@ static const char *amdgpu_fence_get_timeline_name(struct fence *f)
 	return (const char *)fence->ring->name;
 }
 
-static bool amdgpu_test_signaled_any(struct fence **fences, uint32_t count)
-{
-	int idx;
-	struct fence *fence;
-
-	for (idx = 0; idx < count; ++idx) {
-		fence = fences[idx];
-		if (fence) {
-			if (test_bit(FENCE_FLAG_SIGNALED_BIT, &fence->flags))
-				return true;
-		}
-	}
-	return false;
-}
-
-struct amdgpu_wait_cb {
-	struct fence_cb base;
-	struct task_struct *task;
-};
-
-static void amdgpu_fence_wait_cb(struct fence *fence, struct fence_cb *cb)
-{
-	struct amdgpu_wait_cb *wait =
-		container_of(cb, struct amdgpu_wait_cb, base);
-	wake_up_process(wait->task);
-}
-
-/**
- * Wait the fence array with timeout
- *
- * @array:    the fence array with amdgpu fence pointer
- * @count:    the number of the fence array
- * @intr:     when sleep, set the current task interruptable or not
- * @t:        timeout to wait
- *
- * It will return when any fence is signaled or timeout.
- */
-signed long amdgpu_fence_wait_any(struct fence **array, uint32_t count,
-				  bool intr, signed long t)
-{
-	struct amdgpu_wait_cb *cb;
-	struct fence *fence;
-	unsigned idx;
-
-	BUG_ON(!array);
-
-	cb = kcalloc(count, sizeof(struct amdgpu_wait_cb), GFP_KERNEL);
-	if (cb == NULL) {
-		t = -ENOMEM;
-		goto err_free_cb;
-	}
-
-	for (idx = 0; idx < count; ++idx) {
-		fence = array[idx];
-		if (fence) {
-			cb[idx].task = current;
-			if (fence_add_callback(fence,
-					&cb[idx].base, amdgpu_fence_wait_cb)) {
-				/* The fence is already signaled */
-				goto fence_rm_cb;
-			}
-		}
-	}
-
-	while (t > 0) {
-		if (intr)
-			set_current_state(TASK_INTERRUPTIBLE);
-		else
-			set_current_state(TASK_UNINTERRUPTIBLE);
-
-		/*
-		 * amdgpu_test_signaled_any must be called after
-		 * set_current_state to prevent a race with wake_up_process
-		 */
-		if (amdgpu_test_signaled_any(array, count))
-			break;
-
-		t = schedule_timeout(t);
-
-		if (t > 0 && intr && signal_pending(current))
-			t = -ERESTARTSYS;
-	}
-
-	__set_current_state(TASK_RUNNING);
-
-fence_rm_cb:
-	for (idx = 0; idx < count; ++idx) {
-		fence = array[idx];
-		if (fence && cb[idx].base.func)
-			fence_remove_callback(fence, &cb[idx].base);
-	}
-
-err_free_cb:
-	kfree(cb);
-
-	return t;
-}
-
 const struct fence_ops amdgpu_fence_ops = {
 	.get_driver_name = amdgpu_fence_get_driver_name,
 	.get_timeline_name = amdgpu_fence_get_timeline_name,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c
index 5cb27d5..3f48759 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c
@@ -337,6 +337,7 @@ int amdgpu_sa_bo_new(struct amdgpu_device *adev,
 {
 	struct fence *fences[AMDGPU_MAX_RINGS];
 	unsigned tries[AMDGPU_MAX_RINGS];
+	unsigned count;
 	int i, r;
 	signed long t;
 
@@ -371,13 +372,18 @@ int amdgpu_sa_bo_new(struct amdgpu_device *adev,
 			/* see if we can skip over some allocations */
 		} while (amdgpu_sa_bo_next_hole(sa_manager, fences, tries));
 
-		spin_unlock(&sa_manager->wq.lock);
-		t = amdgpu_fence_wait_any(fences, AMDGPU_MAX_RINGS,
-					  false, MAX_SCHEDULE_TIMEOUT);
-		r = (t > 0) ? 0 : t;
-		spin_lock(&sa_manager->wq.lock);
-		/* if we have nothing to wait for block */
-		if (r == -ENOENT) {
+		for (i = 0, count = 0; i < AMDGPU_MAX_RINGS; ++i)
+			if (fences[i])
+				fences[count++] = fences[i];
+
+		if (count) {
+			spin_unlock(&sa_manager->wq.lock);
+			t = fence_wait_any_timeout(fences, count, false,
+						   MAX_SCHEDULE_TIMEOUT);
+			r = (t > 0) ? 0 : t;
+			spin_lock(&sa_manager->wq.lock);
+		} else {
+			/* if we have nothing to wait for block */
 			r = wait_event_interruptible_locked(
 				sa_manager->wq,
 				amdgpu_sa_event(sa_manager, size, align)
-- 
cgit v0.10.2


From 8120b61fdf3d3aa31d6b7e4a4f633bec1d2edd38 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Thu, 22 Oct 2015 11:29:33 +0200
Subject: drm/amdgpu: move ring_from_fence to common code
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Going to need that elsewhere as well.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index dd7d2ce..6d9c929 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -1231,6 +1231,7 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring,
 		     struct amdgpu_irq_src *irq_src, unsigned irq_type,
 		     enum amdgpu_ring_type ring_type);
 void amdgpu_ring_fini(struct amdgpu_ring *ring);
+struct amdgpu_ring *amdgpu_ring_from_fence(struct fence *f);
 
 /*
  * CS.
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
index b2df348..78e9b0f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
@@ -436,6 +436,30 @@ void amdgpu_ring_fini(struct amdgpu_ring *ring)
 	}
 }
 
+/**
+ * amdgpu_ring_from_fence - get ring from fence
+ *
+ * @f: fence structure
+ *
+ * Extract the ring a fence belongs to. Handles both scheduler as
+ * well as hardware fences.
+ */
+struct amdgpu_ring *amdgpu_ring_from_fence(struct fence *f)
+{
+	struct amdgpu_fence *a_fence;
+	struct amd_sched_fence *s_fence;
+
+	s_fence = to_amd_sched_fence(f);
+	if (s_fence)
+		return container_of(s_fence->sched, struct amdgpu_ring, sched);
+
+	a_fence = to_amdgpu_fence(f);
+	if (a_fence)
+		return a_fence->ring;
+
+	return NULL;
+}
+
 /*
  * Debugfs info
  */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c
index 3f48759..0212b31 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c
@@ -139,25 +139,6 @@ int amdgpu_sa_bo_manager_suspend(struct amdgpu_device *adev,
 	return r;
 }
 
-static uint32_t amdgpu_sa_get_ring_from_fence(struct fence *f)
-{
-	struct amdgpu_fence *a_fence;
-	struct amd_sched_fence *s_fence;
-
-	s_fence = to_amd_sched_fence(f);
-	if (s_fence) {
-		struct amdgpu_ring *ring;
-
-		ring = container_of(s_fence->sched, struct amdgpu_ring, sched);
-		return ring->idx;
-	}
-
-	a_fence = to_amdgpu_fence(f);
-	if (a_fence)
-		return a_fence->ring->idx;
-	return 0;
-}
-
 static void amdgpu_sa_bo_remove_locked(struct amdgpu_sa_bo *sa_bo)
 {
 	struct amdgpu_sa_manager *sa_manager = sa_bo->manager;
@@ -318,7 +299,7 @@ static bool amdgpu_sa_bo_next_hole(struct amdgpu_sa_manager *sa_manager,
 	}
 
 	if (best_bo) {
-		uint32_t idx = amdgpu_sa_get_ring_from_fence(best_bo->fence);
+		uint32_t idx = amdgpu_ring_from_fence(best_bo->fence)->idx;
 		++tries[idx];
 		sa_manager->hole = best_bo->olist.prev;
 
@@ -412,7 +393,7 @@ void amdgpu_sa_bo_free(struct amdgpu_device *adev, struct amdgpu_sa_bo **sa_bo,
 	if (fence && !fence_is_signaled(fence)) {
 		uint32_t idx;
 		(*sa_bo)->fence = fence_get(fence);
-		idx = amdgpu_sa_get_ring_from_fence(fence);
+		idx = amdgpu_ring_from_fence(fence)->idx;
 		list_add_tail(&(*sa_bo)->flist, &sa_manager->flist[idx]);
 	} else {
 		amdgpu_sa_bo_remove_locked(*sa_bo);
-- 
cgit v0.10.2


From d52832986aa52ea18940e817d5d04c081c2f42f3 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Thu, 22 Oct 2015 11:55:58 +0200
Subject: drm/amdgpu: use common fences for VMID management v2
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

v2: add missing NULL check.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 6d9c929..371994c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -970,7 +970,7 @@ struct amdgpu_vm_id {
 	/* last flushed PD/PT update */
 	struct fence	        *flushed_updates;
 	/* last use of vmid */
-	struct amdgpu_fence	*last_id_use;
+	struct fence		*last_id_use;
 };
 
 struct amdgpu_vm {
@@ -1003,7 +1003,7 @@ struct amdgpu_vm {
 };
 
 struct amdgpu_vm_manager {
-	struct amdgpu_fence		*active[AMDGPU_NUM_VM];
+	struct fence			*active[AMDGPU_NUM_VM];
 	uint32_t			max_pfn;
 	/* number of VMIDs */
 	unsigned			nvm;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index 06e207f..a12c726 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -135,7 +135,7 @@ struct amdgpu_bo_list_entry *amdgpu_vm_get_bos(struct amdgpu_device *adev,
 int amdgpu_vm_grab_id(struct amdgpu_vm *vm, struct amdgpu_ring *ring,
 		      struct amdgpu_sync *sync)
 {
-	struct amdgpu_fence *best[AMDGPU_MAX_RINGS] = {};
+	struct fence *best[AMDGPU_MAX_RINGS] = {};
 	struct amdgpu_vm_id *vm_id = &vm->ids[ring->idx];
 	struct amdgpu_device *adev = ring->adev;
 
@@ -154,7 +154,8 @@ int amdgpu_vm_grab_id(struct amdgpu_vm *vm, struct amdgpu_ring *ring,
 
 	/* skip over VMID 0, since it is the system VM */
 	for (i = 1; i < adev->vm_manager.nvm; ++i) {
-		struct amdgpu_fence *fence = adev->vm_manager.active[i];
+		struct fence *fence = adev->vm_manager.active[i];
+		struct amdgpu_ring *fring;
 
 		if (fence == NULL) {
 			/* found a free one */
@@ -163,21 +164,23 @@ int amdgpu_vm_grab_id(struct amdgpu_vm *vm, struct amdgpu_ring *ring,
 			return 0;
 		}
 
-		if (amdgpu_fence_is_earlier(fence, best[fence->ring->idx])) {
-			best[fence->ring->idx] = fence;
-			choices[fence->ring == ring ? 0 : 1] = i;
+		fring = amdgpu_ring_from_fence(fence);
+		if (best[fring->idx] == NULL ||
+		    fence_is_later(best[fring->idx], fence)) {
+			best[fring->idx] = fence;
+			choices[fring == ring ? 0 : 1] = i;
 		}
 	}
 
 	for (i = 0; i < 2; ++i) {
 		if (choices[i]) {
-			struct amdgpu_fence *fence;
+			struct fence *fence;
 
 			fence  = adev->vm_manager.active[choices[i]];
 			vm_id->id = choices[i];
 
 			trace_amdgpu_vm_grab_id(choices[i], ring->idx);
-			return amdgpu_sync_fence(ring->adev, sync, &fence->base);
+			return amdgpu_sync_fence(ring->adev, sync, fence);
 		}
 	}
 
@@ -246,11 +249,11 @@ void amdgpu_vm_fence(struct amdgpu_device *adev,
 	unsigned ridx = fence->ring->idx;
 	unsigned vm_id = vm->ids[ridx].id;
 
-	amdgpu_fence_unref(&adev->vm_manager.active[vm_id]);
-	adev->vm_manager.active[vm_id] = amdgpu_fence_ref(fence);
+	fence_put(adev->vm_manager.active[vm_id]);
+	adev->vm_manager.active[vm_id] = fence_get(&fence->base);
 
-	amdgpu_fence_unref(&vm->ids[ridx].last_id_use);
-	vm->ids[ridx].last_id_use = amdgpu_fence_ref(fence);
+	fence_put(vm->ids[ridx].last_id_use);
+	vm->ids[ridx].last_id_use = fence_get(&fence->base);
 }
 
 /**
@@ -1311,7 +1314,7 @@ void amdgpu_vm_fini(struct amdgpu_device *adev, struct amdgpu_vm *vm)
 
 	for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
 		fence_put(vm->ids[i].flushed_updates);
-		amdgpu_fence_unref(&vm->ids[i].last_id_use);
+		fence_put(vm->ids[i].last_id_use);
 	}
 
 	mutex_destroy(&vm->mutex);
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
index 4883482..85bbcdc 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
@@ -965,7 +965,7 @@ static int gmc_v7_0_sw_fini(void *handle)
 
 	if (adev->vm_manager.enabled) {
 		for (i = 0; i < AMDGPU_NUM_VM; ++i)
-			amdgpu_fence_unref(&adev->vm_manager.active[i]);
+			fence_put(adev->vm_manager.active[i]);
 		gmc_v7_0_vm_fini(adev);
 		adev->vm_manager.enabled = false;
 	}
@@ -1015,7 +1015,7 @@ static int gmc_v7_0_suspend(void *handle)
 
 	if (adev->vm_manager.enabled) {
 		for (i = 0; i < AMDGPU_NUM_VM; ++i)
-			amdgpu_fence_unref(&adev->vm_manager.active[i]);
+			fence_put(adev->vm_manager.active[i]);
 		gmc_v7_0_vm_fini(adev);
 		adev->vm_manager.enabled = false;
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
index 72e977b..1bcc4e7 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
@@ -984,7 +984,7 @@ static int gmc_v8_0_sw_fini(void *handle)
 
 	if (adev->vm_manager.enabled) {
 		for (i = 0; i < AMDGPU_NUM_VM; ++i)
-			amdgpu_fence_unref(&adev->vm_manager.active[i]);
+			fence_put(adev->vm_manager.active[i]);
 		gmc_v8_0_vm_fini(adev);
 		adev->vm_manager.enabled = false;
 	}
@@ -1036,7 +1036,7 @@ static int gmc_v8_0_suspend(void *handle)
 
 	if (adev->vm_manager.enabled) {
 		for (i = 0; i < AMDGPU_NUM_VM; ++i)
-			amdgpu_fence_unref(&adev->vm_manager.active[i]);
+			fence_put(adev->vm_manager.active[i]);
 		gmc_v8_0_vm_fini(adev);
 		adev->vm_manager.enabled = false;
 	}
-- 
cgit v0.10.2


From 24233860b97c46eb943215d761e03f6f2e1285a6 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Thu, 22 Oct 2015 10:53:16 +0200
Subject: drm/amdgpu: use the new fence_is_later
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Instead of coding the check ourself.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c
index 4921de1..0499d2b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c
@@ -87,6 +87,15 @@ static bool amdgpu_sync_test_owner(struct fence *f, void *owner)
 	return false;
 }
 
+static void amdgpu_sync_keep_later(struct fence **keep, struct fence *fence)
+{
+	if (*keep && fence_is_later(*keep, fence))
+		return;
+
+	fence_put(*keep);
+	*keep = fence_get(fence);
+}
+
 /**
  * amdgpu_sync_fence - remember to sync to this fence
  *
@@ -100,34 +109,21 @@ int amdgpu_sync_fence(struct amdgpu_device *adev, struct amdgpu_sync *sync,
 	struct amdgpu_sync_entry *e;
 	struct amdgpu_fence *fence;
 	struct amdgpu_fence *other;
-	struct fence *tmp, *later;
 
 	if (!f)
 		return 0;
 
 	if (amdgpu_sync_same_dev(adev, f) &&
-	    amdgpu_sync_test_owner(f, AMDGPU_FENCE_OWNER_VM)) {
-		if (sync->last_vm_update) {
-			tmp = sync->last_vm_update;
-			BUG_ON(f->context != tmp->context);
-			later = (f->seqno - tmp->seqno <= INT_MAX) ? f : tmp;
-			sync->last_vm_update = fence_get(later);
-			fence_put(tmp);
-		} else
-			sync->last_vm_update = fence_get(f);
-	}
+	    amdgpu_sync_test_owner(f, AMDGPU_FENCE_OWNER_VM))
+		amdgpu_sync_keep_later(&sync->last_vm_update, f);
 
 	fence = to_amdgpu_fence(f);
 	if (!fence || fence->ring->adev != adev) {
 		hash_for_each_possible(sync->fences, e, node, f->context) {
-			struct fence *new;
 			if (unlikely(e->fence->context != f->context))
 				continue;
-			new = fence_get(fence_later(e->fence, f));
-			if (new) {
-				fence_put(e->fence);
-				e->fence = new;
-			}
+
+			amdgpu_sync_keep_later(&e->fence, f);
 			return 0;
 		}
 
-- 
cgit v0.10.2


From 16545c32a5242634a51c6906533750cf1db54879 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Thu, 22 Oct 2015 15:04:50 +0200
Subject: drm/amdgpu: use common fence for sync
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Stop using the driver internal functions.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 371994c..b986ea1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -704,7 +704,7 @@ void amdgpu_semaphore_free(struct amdgpu_device *adev,
  */
 struct amdgpu_sync {
 	struct amdgpu_semaphore *semaphores[AMDGPU_NUM_SYNCS];
-	struct amdgpu_fence	*sync_to[AMDGPU_MAX_RINGS];
+	struct fence		*sync_to[AMDGPU_MAX_RINGS];
 	DECLARE_HASHTABLE(fences, 4);
 	struct fence	        *last_vm_update;
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c
index 0499d2b..c6489b1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c
@@ -108,7 +108,6 @@ int amdgpu_sync_fence(struct amdgpu_device *adev, struct amdgpu_sync *sync,
 {
 	struct amdgpu_sync_entry *e;
 	struct amdgpu_fence *fence;
-	struct amdgpu_fence *other;
 
 	if (!f)
 		return 0;
@@ -136,10 +135,7 @@ int amdgpu_sync_fence(struct amdgpu_device *adev, struct amdgpu_sync *sync,
 		return 0;
 	}
 
-	other = sync->sync_to[fence->ring->idx];
-	sync->sync_to[fence->ring->idx] = amdgpu_fence_ref(
-		amdgpu_fence_later(fence, other));
-	amdgpu_fence_unref(&other);
+	amdgpu_sync_keep_later(&sync->sync_to[fence->ring->idx], f);
 
 	return 0;
 }
@@ -258,11 +254,11 @@ int amdgpu_sync_wait(struct amdgpu_sync *sync)
 		return 0;
 
 	for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
-		struct amdgpu_fence *fence = sync->sync_to[i];
+		struct fence *fence = sync->sync_to[i];
 		if (!fence)
 			continue;
 
-		r = fence_wait(&fence->base, false);
+		r = fence_wait(fence, false);
 		if (r)
 			return r;
 	}
@@ -287,9 +283,14 @@ int amdgpu_sync_rings(struct amdgpu_sync *sync,
 	int i, r;
 
 	for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
-		struct amdgpu_fence *fence = sync->sync_to[i];
-		struct amdgpu_semaphore *semaphore;
 		struct amdgpu_ring *other = adev->rings[i];
+		struct amdgpu_semaphore *semaphore;
+		struct amdgpu_fence *fence;
+
+		if (!sync->sync_to[i])
+			continue;
+
+		fence = to_amdgpu_fence(sync->sync_to[i]);
 
 		/* check if we really need to sync */
 		if (!amdgpu_fence_need_sync(fence, ring))
@@ -374,7 +375,7 @@ void amdgpu_sync_free(struct amdgpu_device *adev,
 		amdgpu_semaphore_free(adev, &sync->semaphores[i], fence);
 
 	for (i = 0; i < AMDGPU_MAX_RINGS; ++i)
-		amdgpu_fence_unref(&sync->sync_to[i]);
+		fence_put(sync->sync_to[i]);
 
 	fence_put(sync->last_vm_update);
 }
-- 
cgit v0.10.2


From 6ef68c17d40e1e7e291ca513627a0d2a13ae095f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Thu, 22 Oct 2015 15:16:22 +0200
Subject: drm/amdgpu: remove amdgpu_fence_ref/unref
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Just move the remaining users to fence_put/get.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index b986ea1..8305a6c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -447,9 +447,6 @@ int amdgpu_fence_wait_next(struct amdgpu_ring *ring);
 int amdgpu_fence_wait_empty(struct amdgpu_ring *ring);
 unsigned amdgpu_fence_count_emitted(struct amdgpu_ring *ring);
 
-struct amdgpu_fence *amdgpu_fence_ref(struct amdgpu_fence *fence);
-void amdgpu_fence_unref(struct amdgpu_fence **fence);
-
 bool amdgpu_fence_need_sync(struct amdgpu_fence *fence,
 			    struct amdgpu_ring *ring);
 void amdgpu_fence_note_sync(struct amdgpu_fence *fence,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
index c4bb282..b912539 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
@@ -417,36 +417,6 @@ int amdgpu_fence_wait_empty(struct amdgpu_ring *ring)
 }
 
 /**
- * amdgpu_fence_ref - take a ref on a fence
- *
- * @fence: amdgpu fence object
- *
- * Take a reference on a fence (all asics).
- * Returns the fence.
- */
-struct amdgpu_fence *amdgpu_fence_ref(struct amdgpu_fence *fence)
-{
-	fence_get(&fence->base);
-	return fence;
-}
-
-/**
- * amdgpu_fence_unref - remove a ref on a fence
- *
- * @fence: amdgpu fence object
- *
- * Remove a reference on a fence (all asics).
- */
-void amdgpu_fence_unref(struct amdgpu_fence **fence)
-{
-	struct amdgpu_fence *tmp = *fence;
-
-	*fence = NULL;
-	if (tmp)
-		fence_put(&tmp->base);
-}
-
-/**
  * amdgpu_fence_count_emitted - get the count of emitted fences
  *
  * @ring: ring the fence is associated with
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
index aad4c1c..e659877 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
@@ -95,7 +95,8 @@ void amdgpu_ib_free(struct amdgpu_device *adev, struct amdgpu_ib *ib)
 {
 	amdgpu_sync_free(adev, &ib->sync, &ib->fence->base);
 	amdgpu_sa_bo_free(adev, &ib->sa_bo, &ib->fence->base);
-	amdgpu_fence_unref(&ib->fence);
+	if (ib->fence)
+		fence_put(&ib->fence->base);
 }
 
 /**
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c
index 2e946b2..dcf4a8a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c
@@ -54,7 +54,8 @@ static struct fence *amdgpu_sched_run_job(struct amd_sched_job *sched_job)
 		goto err;
 	}
 
-	fence = amdgpu_fence_ref(job->ibs[job->num_ibs - 1].fence);
+	fence = job->ibs[job->num_ibs - 1].fence;
+	fence_get(&fence->base);
 
 err:
 	if (job->free_job)
-- 
cgit v0.10.2


From c8a8067f7d66bf77a0627e0b237b0b4486862a73 Mon Sep 17 00:00:00 2001
From: Thierry Reding <treding@nvidia.com>
Date: Tue, 11 Aug 2015 15:38:43 +0200
Subject: drm/tegra: fb: Remove gratuituous blank line

This was introduced by mistake in commit 0f7d9052fb70 ("drm/tegra: Use
new drm_fb_helper functions"). Remove it.

Signed-off-by: Thierry Reding <treding@nvidia.com>

diff --git a/drivers/gpu/drm/tegra/fb.c b/drivers/gpu/drm/tegra/fb.c
index 07c844b..1004075 100644
--- a/drivers/gpu/drm/tegra/fb.c
+++ b/drivers/gpu/drm/tegra/fb.c
@@ -341,7 +341,6 @@ fini:
 
 static void tegra_fbdev_exit(struct tegra_fbdev *fbdev)
 {
-
 	drm_fb_helper_unregister_fbi(&fbdev->base);
 	drm_fb_helper_release_fbi(&fbdev->base);
 
-- 
cgit v0.10.2


From 6dd0d0e1ae7f3e129c6943cd351d47d194bde238 Mon Sep 17 00:00:00 2001
From: Fabio Estevam <fabio.estevam@freescale.com>
Date: Sat, 9 May 2015 13:08:25 -0300
Subject: gpu: ipu-v3: ipu-csi: bool test doesn't need a comparison to false

Bool initializations should use true and false. Bool tests don't need
comparisons. Based on contributions from Joe Perches, Rusty Russell
and Bruce W Allan.

The semantic patch that makes this change is available
in scripts/coccinelle/misc/boolinit.cocci.

More information about semantic patching is available at
http://coccinelle.lip6.fr/

Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>

diff --git a/drivers/gpu/ipu-v3/ipu-csi.c b/drivers/gpu/ipu-v3/ipu-csi.c
index 752cdd2..eec5adb 100644
--- a/drivers/gpu/ipu-v3/ipu-csi.c
+++ b/drivers/gpu/ipu-v3/ipu-csi.c
@@ -538,7 +538,7 @@ void ipu_csi_set_test_generator(struct ipu_csi *csi, bool active,
 
 	temp = ipu_csi_read(csi, CSI_TST_CTRL);
 
-	if (active == false) {
+	if (!active) {
 		temp &= ~CSI_TEST_GEN_MODE_EN;
 		ipu_csi_write(csi, temp, CSI_TST_CTRL);
 	} else {
-- 
cgit v0.10.2


From ef739aa4c68be8a05b7f0b91cf87747acbe0b424 Mon Sep 17 00:00:00 2001
From: Philipp Zabel <p.zabel@pengutronix.de>
Date: Fri, 19 Jun 2015 16:22:38 +0200
Subject: MAINTAINERS: Add IPUv3 core driver to the i.MX DRM driver section

The ipu-v3 core driver is maintained in and merged through
the same trees as the imx-drm drivers that are using it.
This patch adds the drivers/gpu/ipu-v3/ directory to the
i.MX DRM driver section.

Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>

diff --git a/MAINTAINERS b/MAINTAINERS
index 5f46784..f998dd4 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -3617,6 +3617,7 @@ M:	Philipp Zabel <p.zabel@pengutronix.de>
 L:	dri-devel@lists.freedesktop.org
 S:	Maintained
 F:	drivers/gpu/drm/imx/
+F:	drivers/gpu/ipu-v3/
 F:	Documentation/devicetree/bindings/drm/imx/
 
 DRM DRIVERS FOR NVIDIA TEGRA
-- 
cgit v0.10.2


From 0cb8b75792d6b752df945993cc5d455c75bf6884 Mon Sep 17 00:00:00 2001
From: Philipp Zabel <p.zabel@pengutronix.de>
Date: Fri, 12 Dec 2014 13:40:14 +0100
Subject: gpu: ipu-v3: add support for 15-bit RGB with 1-bit alpha formats

This patch adds support for ARGB1555, ABGR1555, RGBA5551, and BGRA5551
in-memory formats.

Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>

diff --git a/drivers/gpu/ipu-v3/ipu-common.c b/drivers/gpu/ipu-v3/ipu-common.c
index e5a38d2..c0b1916 100644
--- a/drivers/gpu/ipu-v3/ipu-common.c
+++ b/drivers/gpu/ipu-v3/ipu-common.c
@@ -57,6 +57,10 @@ EXPORT_SYMBOL_GPL(ipu_srm_dp_sync_update);
 enum ipu_color_space ipu_drm_fourcc_to_colorspace(u32 drm_fourcc)
 {
 	switch (drm_fourcc) {
+	case DRM_FORMAT_ARGB1555:
+	case DRM_FORMAT_ABGR1555:
+	case DRM_FORMAT_RGBA5551:
+	case DRM_FORMAT_BGRA5551:
 	case DRM_FORMAT_RGB565:
 	case DRM_FORMAT_BGR565:
 	case DRM_FORMAT_RGB888:
diff --git a/drivers/gpu/ipu-v3/ipu-cpmem.c b/drivers/gpu/ipu-v3/ipu-cpmem.c
index 3bf05bc..d26b8be 100644
--- a/drivers/gpu/ipu-v3/ipu-cpmem.c
+++ b/drivers/gpu/ipu-v3/ipu-cpmem.c
@@ -500,6 +500,38 @@ static const struct ipu_rgb def_bgr_16 = {
 	.bits_per_pixel = 16,
 };
 
+static const struct ipu_rgb def_argb_16 = {
+	.red	= { .offset = 10, .length = 5, },
+	.green	= { .offset =  5, .length = 5, },
+	.blue	= { .offset =  0, .length = 5, },
+	.transp = { .offset = 15, .length = 1, },
+	.bits_per_pixel = 16,
+};
+
+static const struct ipu_rgb def_abgr_16 = {
+	.red	= { .offset =  0, .length = 5, },
+	.green	= { .offset =  5, .length = 5, },
+	.blue	= { .offset = 10, .length = 5, },
+	.transp = { .offset = 15, .length = 1, },
+	.bits_per_pixel = 16,
+};
+
+static const struct ipu_rgb def_rgba_16 = {
+	.red	= { .offset = 11, .length = 5, },
+	.green	= { .offset =  6, .length = 5, },
+	.blue	= { .offset =  1, .length = 5, },
+	.transp = { .offset =  0, .length = 1, },
+	.bits_per_pixel = 16,
+};
+
+static const struct ipu_rgb def_bgra_16 = {
+	.red	= { .offset =  1, .length = 5, },
+	.green	= { .offset =  6, .length = 5, },
+	.blue	= { .offset = 11, .length = 5, },
+	.transp = { .offset =  0, .length = 1, },
+	.bits_per_pixel = 16,
+};
+
 #define Y_OFFSET(pix, x, y)	((x) + pix->width * (y))
 #define U_OFFSET(pix, x, y)	((pix->width * pix->height) +		\
 				 (pix->width * (y) / 4) + (x) / 2)
@@ -581,6 +613,18 @@ int ipu_cpmem_set_fmt(struct ipuv3_channel *ch, u32 drm_fourcc)
 	case DRM_FORMAT_BGR565:
 		ipu_cpmem_set_format_rgb(ch, &def_bgr_16);
 		break;
+	case DRM_FORMAT_ARGB1555:
+		ipu_cpmem_set_format_rgb(ch, &def_argb_16);
+		break;
+	case DRM_FORMAT_ABGR1555:
+		ipu_cpmem_set_format_rgb(ch, &def_abgr_16);
+		break;
+	case DRM_FORMAT_RGBA5551:
+		ipu_cpmem_set_format_rgb(ch, &def_rgba_16);
+		break;
+	case DRM_FORMAT_BGRA5551:
+		ipu_cpmem_set_format_rgb(ch, &def_bgra_16);
+		break;
 	default:
 		return -EINVAL;
 	}
-- 
cgit v0.10.2


From c639a1cfc4930684300860abdac5ebcb619523ea Mon Sep 17 00:00:00 2001
From: Philipp Zabel <p.zabel@pengutronix.de>
Date: Fri, 12 Dec 2014 13:40:38 +0100
Subject: drm/imx: enable 15-bit RGB with 1-bit alpha formats

This patch enables the ARGB1555, ABGR1555, RGBA5551,
and BGRA5551 formats to be used on planes.

Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>

diff --git a/drivers/gpu/drm/imx/ipuv3-plane.c b/drivers/gpu/drm/imx/ipuv3-plane.c
index 878a643..d13dbb6 100644
--- a/drivers/gpu/drm/imx/ipuv3-plane.c
+++ b/drivers/gpu/drm/imx/ipuv3-plane.c
@@ -23,8 +23,12 @@
 #define to_ipu_plane(x)	container_of(x, struct ipu_plane, base)
 
 static const uint32_t ipu_plane_formats[] = {
+	DRM_FORMAT_ARGB1555,
 	DRM_FORMAT_XRGB1555,
+	DRM_FORMAT_ABGR1555,
 	DRM_FORMAT_XBGR1555,
+	DRM_FORMAT_RGBA5551,
+	DRM_FORMAT_BGRA5551,
 	DRM_FORMAT_ARGB8888,
 	DRM_FORMAT_XRGB8888,
 	DRM_FORMAT_ABGR8888,
@@ -175,6 +179,10 @@ int ipu_plane_mode_set(struct ipu_plane *ipu_plane, struct drm_crtc *crtc,
 		ipu_dp_set_window_pos(ipu_plane->dp, crtc_x, crtc_y);
 		/* Enable local alpha on partial plane */
 		switch (fb->pixel_format) {
+		case DRM_FORMAT_ARGB1555:
+		case DRM_FORMAT_ABGR1555:
+		case DRM_FORMAT_RGBA5551:
+		case DRM_FORMAT_BGRA5551:
 		case DRM_FORMAT_ARGB8888:
 		case DRM_FORMAT_ABGR8888:
 			ipu_dp_set_global_alpha(ipu_plane->dp, false, 0, false);
-- 
cgit v0.10.2


From 067f4aa402ac2c41e70e25abda97af30333d9581 Mon Sep 17 00:00:00 2001
From: Philipp Zabel <p.zabel@pengutronix.de>
Date: Thu, 16 Apr 2015 15:27:04 +0200
Subject: gpu: ipu-v3: add support for RGBX8888 and RGBA8888 pixel formats

This patch adds support for the RGBA8888, RGBX8888, BGRA8888, and
BGRX8888 in-memory formats.

Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>

diff --git a/drivers/gpu/ipu-v3/ipu-cpmem.c b/drivers/gpu/ipu-v3/ipu-cpmem.c
index d26b8be..0e6b868 100644
--- a/drivers/gpu/ipu-v3/ipu-cpmem.c
+++ b/drivers/gpu/ipu-v3/ipu-cpmem.c
@@ -452,7 +452,7 @@ void ipu_cpmem_set_yuv_planar(struct ipuv3_channel *ch,
 }
 EXPORT_SYMBOL_GPL(ipu_cpmem_set_yuv_planar);
 
-static const struct ipu_rgb def_rgb_32 = {
+static const struct ipu_rgb def_xrgb_32 = {
 	.red	= { .offset = 16, .length = 8, },
 	.green	= { .offset =  8, .length = 8, },
 	.blue	= { .offset =  0, .length = 8, },
@@ -460,7 +460,7 @@ static const struct ipu_rgb def_rgb_32 = {
 	.bits_per_pixel = 32,
 };
 
-static const struct ipu_rgb def_bgr_32 = {
+static const struct ipu_rgb def_xbgr_32 = {
 	.red	= { .offset =  0, .length = 8, },
 	.green	= { .offset =  8, .length = 8, },
 	.blue	= { .offset = 16, .length = 8, },
@@ -468,6 +468,22 @@ static const struct ipu_rgb def_bgr_32 = {
 	.bits_per_pixel = 32,
 };
 
+static const struct ipu_rgb def_rgbx_32 = {
+	.red	= { .offset = 24, .length = 8, },
+	.green	= { .offset = 16, .length = 8, },
+	.blue	= { .offset =  8, .length = 8, },
+	.transp = { .offset =  0, .length = 8, },
+	.bits_per_pixel = 32,
+};
+
+static const struct ipu_rgb def_bgrx_32 = {
+	.red	= { .offset =  8, .length = 8, },
+	.green	= { .offset = 16, .length = 8, },
+	.blue	= { .offset = 24, .length = 8, },
+	.transp = { .offset =  0, .length = 8, },
+	.bits_per_pixel = 32,
+};
+
 static const struct ipu_rgb def_rgb_24 = {
 	.red	= { .offset = 16, .length = 8, },
 	.green	= { .offset =  8, .length = 8, },
@@ -595,11 +611,19 @@ int ipu_cpmem_set_fmt(struct ipuv3_channel *ch, u32 drm_fourcc)
 		break;
 	case DRM_FORMAT_ABGR8888:
 	case DRM_FORMAT_XBGR8888:
-		ipu_cpmem_set_format_rgb(ch, &def_bgr_32);
+		ipu_cpmem_set_format_rgb(ch, &def_xbgr_32);
 		break;
 	case DRM_FORMAT_ARGB8888:
 	case DRM_FORMAT_XRGB8888:
-		ipu_cpmem_set_format_rgb(ch, &def_rgb_32);
+		ipu_cpmem_set_format_rgb(ch, &def_xrgb_32);
+		break;
+	case DRM_FORMAT_RGBA8888:
+	case DRM_FORMAT_RGBX8888:
+		ipu_cpmem_set_format_rgb(ch, &def_rgbx_32);
+		break;
+	case DRM_FORMAT_BGRA8888:
+	case DRM_FORMAT_BGRX8888:
+		ipu_cpmem_set_format_rgb(ch, &def_bgrx_32);
 		break;
 	case DRM_FORMAT_BGR888:
 		ipu_cpmem_set_format_rgb(ch, &def_bgr_24);
-- 
cgit v0.10.2


From 59d6b7189a968d627af37fc26a410dced0854b99 Mon Sep 17 00:00:00 2001
From: Philipp Zabel <p.zabel@pengutronix.de>
Date: Thu, 16 Apr 2015 15:56:40 +0200
Subject: drm/imx: ipuv3-plane: enable support for RGBX8888 and RGBA8888 pixel
 formats

This patch allows to use the RGBX and RGBA 8:8:8:8 formats.

Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>

diff --git a/drivers/gpu/drm/imx/ipuv3-plane.c b/drivers/gpu/drm/imx/ipuv3-plane.c
index d13dbb6..d030990 100644
--- a/drivers/gpu/drm/imx/ipuv3-plane.c
+++ b/drivers/gpu/drm/imx/ipuv3-plane.c
@@ -33,6 +33,10 @@ static const uint32_t ipu_plane_formats[] = {
 	DRM_FORMAT_XRGB8888,
 	DRM_FORMAT_ABGR8888,
 	DRM_FORMAT_XBGR8888,
+	DRM_FORMAT_RGBA8888,
+	DRM_FORMAT_RGBX8888,
+	DRM_FORMAT_BGRA8888,
+	DRM_FORMAT_BGRA8888,
 	DRM_FORMAT_YUYV,
 	DRM_FORMAT_YVYU,
 	DRM_FORMAT_YUV420,
@@ -185,6 +189,8 @@ int ipu_plane_mode_set(struct ipu_plane *ipu_plane, struct drm_crtc *crtc,
 		case DRM_FORMAT_BGRA5551:
 		case DRM_FORMAT_ARGB8888:
 		case DRM_FORMAT_ABGR8888:
+		case DRM_FORMAT_RGBA8888:
+		case DRM_FORMAT_BGRA8888:
 			ipu_dp_set_global_alpha(ipu_plane->dp, false, 0, false);
 			break;
 		default:
-- 
cgit v0.10.2


From 7d2e8a204aa04b83aca2beb5f1441262c3aa26c8 Mon Sep 17 00:00:00 2001
From: Lucas Stach <l.stach@pengutronix.de>
Date: Tue, 4 Aug 2015 17:21:04 +0200
Subject: gpu: ipu-v3: add support for ARGB4444 16-bit color format

This patch adds support for the ARGB4444 color format.

Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>

diff --git a/drivers/gpu/ipu-v3/ipu-common.c b/drivers/gpu/ipu-v3/ipu-common.c
index c0b1916..ba47b30 100644
--- a/drivers/gpu/ipu-v3/ipu-common.c
+++ b/drivers/gpu/ipu-v3/ipu-common.c
@@ -65,6 +65,7 @@ enum ipu_color_space ipu_drm_fourcc_to_colorspace(u32 drm_fourcc)
 	case DRM_FORMAT_BGR565:
 	case DRM_FORMAT_RGB888:
 	case DRM_FORMAT_BGR888:
+	case DRM_FORMAT_ARGB4444:
 	case DRM_FORMAT_XRGB8888:
 	case DRM_FORMAT_XBGR8888:
 	case DRM_FORMAT_RGBX8888:
diff --git a/drivers/gpu/ipu-v3/ipu-cpmem.c b/drivers/gpu/ipu-v3/ipu-cpmem.c
index 0e6b868..63eb16b 100644
--- a/drivers/gpu/ipu-v3/ipu-cpmem.c
+++ b/drivers/gpu/ipu-v3/ipu-cpmem.c
@@ -524,6 +524,14 @@ static const struct ipu_rgb def_argb_16 = {
 	.bits_per_pixel = 16,
 };
 
+static const struct ipu_rgb def_argb_16_4444 = {
+	.red	= { .offset =  8, .length = 4, },
+	.green	= { .offset =  4, .length = 4, },
+	.blue	= { .offset =  0, .length = 4, },
+	.transp = { .offset = 12, .length = 4, },
+	.bits_per_pixel = 16,
+};
+
 static const struct ipu_rgb def_abgr_16 = {
 	.red	= { .offset =  0, .length = 5, },
 	.green	= { .offset =  5, .length = 5, },
@@ -649,6 +657,9 @@ int ipu_cpmem_set_fmt(struct ipuv3_channel *ch, u32 drm_fourcc)
 	case DRM_FORMAT_BGRA5551:
 		ipu_cpmem_set_format_rgb(ch, &def_bgra_16);
 		break;
+	case DRM_FORMAT_ARGB4444:
+		ipu_cpmem_set_format_rgb(ch, &def_argb_16_4444);
+		break;
 	default:
 		return -EINVAL;
 	}
-- 
cgit v0.10.2


From cb166a302589f1494a62b6f1ca108fddb3925e31 Mon Sep 17 00:00:00 2001
From: Lucas Stach <l.stach@pengutronix.de>
Date: Tue, 4 Aug 2015 17:22:06 +0200
Subject: drm/imx: enable ARGB4444 16-bit color format

This patch allows to use the ARGB4444 color format on planes.

Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>

diff --git a/drivers/gpu/drm/imx/ipuv3-plane.c b/drivers/gpu/drm/imx/ipuv3-plane.c
index d030990..575f4c8 100644
--- a/drivers/gpu/drm/imx/ipuv3-plane.c
+++ b/drivers/gpu/drm/imx/ipuv3-plane.c
@@ -29,6 +29,7 @@ static const uint32_t ipu_plane_formats[] = {
 	DRM_FORMAT_XBGR1555,
 	DRM_FORMAT_RGBA5551,
 	DRM_FORMAT_BGRA5551,
+	DRM_FORMAT_ARGB4444,
 	DRM_FORMAT_ARGB8888,
 	DRM_FORMAT_XRGB8888,
 	DRM_FORMAT_ABGR8888,
@@ -187,6 +188,7 @@ int ipu_plane_mode_set(struct ipu_plane *ipu_plane, struct drm_crtc *crtc,
 		case DRM_FORMAT_ABGR1555:
 		case DRM_FORMAT_RGBA5551:
 		case DRM_FORMAT_BGRA5551:
+		case DRM_FORMAT_ARGB4444:
 		case DRM_FORMAT_ARGB8888:
 		case DRM_FORMAT_ABGR8888:
 		case DRM_FORMAT_RGBA8888:
-- 
cgit v0.10.2


From 3e8b2246c36f4a340febc97d140db5dd727bb3c6 Mon Sep 17 00:00:00 2001
From: Philippe De Muyter <phdm@macqel.be>
Date: Fri, 18 Sep 2015 15:41:55 +0200
Subject: gpu: ipu-v3: csi: add support for 8 bpp grayscale sensors.

Enable the MEDIA_BUS_FMT_Y8_1X8 format.

Signed-off-by: Philippe De Muyter <phdm@macqel.be>
Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>

diff --git a/drivers/gpu/ipu-v3/ipu-csi.c b/drivers/gpu/ipu-v3/ipu-csi.c
index eec5adb..3bdb705 100644
--- a/drivers/gpu/ipu-v3/ipu-csi.c
+++ b/drivers/gpu/ipu-v3/ipu-csi.c
@@ -271,6 +271,7 @@ static int mbus_code_to_bus_cfg(struct ipu_csi_bus_config *cfg, u32 mbus_code)
 	case MEDIA_BUS_FMT_SGBRG8_1X8:
 	case MEDIA_BUS_FMT_SGRBG8_1X8:
 	case MEDIA_BUS_FMT_SRGGB8_1X8:
+	case MEDIA_BUS_FMT_Y8_1X8:
 		cfg->data_fmt = CSI_SENS_CONF_DATA_FMT_BAYER;
 		cfg->mipi_dt = MIPI_DT_RAW8;
 		cfg->data_width = IPU_CSI_DATA_WIDTH_8;
-- 
cgit v0.10.2


From 9b6d0d33c848661c859ccee331b9d2c2ba74489b Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <a.hajda@samsung.com>
Date: Mon, 21 Sep 2015 15:33:48 +0200
Subject: gpu: ipu-v3: fix div_ratio type

The variable can be negative.

The problem has been detected using proposed semantic patch
scripts/coccinelle/tests/unsigned_lesser_than_zero.cocci [1].

[1]: http://permalink.gmane.org/gmane.linux.kernel/2038576

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>

diff --git a/drivers/gpu/ipu-v3/ipu-csi.c b/drivers/gpu/ipu-v3/ipu-csi.c
index 3bdb705..06631ac 100644
--- a/drivers/gpu/ipu-v3/ipu-csi.c
+++ b/drivers/gpu/ipu-v3/ipu-csi.c
@@ -202,7 +202,7 @@ static int ipu_csi_set_testgen_mclk(struct ipu_csi *csi, u32 pixel_clk,
 					u32 ipu_clk)
 {
 	u32 temp;
-	u32 div_ratio;
+	int div_ratio;
 
 	div_ratio = (ipu_clk / pixel_clk) - 1;
 
-- 
cgit v0.10.2


From a5f4185c4b8c131c0ccafa6b1b00cd4e5413e47e Mon Sep 17 00:00:00 2001
From: Lucas Stach <l.stach@pengutronix.de>
Date: Thu, 15 Oct 2015 15:42:17 +0200
Subject: drm/imx: hdmi: fix HDMI setup to allow modes larger than FullHD

This worked before the dw-hdmi bridge code was changed to validate
the setup data more strictly. Add back support for modes with a
pixel clock up to 216MHz. Even higher clocks should work, but we
are missing the required setup data for now.

Also change the mode validate callbacks to disallow modes with
higher pixelclocks, so we don't end up failing the modeset later
on.

Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>

diff --git a/drivers/gpu/drm/imx/dw_hdmi-imx.c b/drivers/gpu/drm/imx/dw_hdmi-imx.c
index 644edf6..98605ea 100644
--- a/drivers/gpu/drm/imx/dw_hdmi-imx.c
+++ b/drivers/gpu/drm/imx/dw_hdmi-imx.c
@@ -48,11 +48,17 @@ static const struct dw_hdmi_mpll_config imx_mpll_cfg[] = {
 			{ 0x40a2, 0x000a },
 		},
 	}, {
-		~0UL, {
+		216000000, {
 			{ 0x00a0, 0x000a },
 			{ 0x2001, 0x000f },
 			{ 0x4002, 0x000f },
 		},
+	}, {
+		~0UL, {
+			{ 0x0000, 0x0000 },
+			{ 0x0000, 0x0000 },
+			{ 0x0000, 0x0000 },
+		},
 	}
 };
 
@@ -82,7 +88,7 @@ static const struct dw_hdmi_curr_ctrl imx_cur_ctr[] = {
  */
 static const struct dw_hdmi_phy_config imx_phy_config[] = {
 	/*pixelclk   symbol   term   vlev */
-	{ 148500000, 0x800d, 0x0005, 0x01ad},
+	{ 216000000, 0x800d, 0x0005, 0x01ad},
 	{ ~0UL,      0x0000, 0x0000, 0x0000}
 };
 
@@ -148,7 +154,8 @@ static enum drm_mode_status imx6q_hdmi_mode_valid(struct drm_connector *con,
 {
 	if (mode->clock < 13500)
 		return MODE_CLOCK_LOW;
-	if (mode->clock > 266000)
+	/* FIXME: Hardware is capable of 266MHz, but setup data is missing. */
+	if (mode->clock > 216000)
 		return MODE_CLOCK_HIGH;
 
 	return MODE_OK;
@@ -159,7 +166,8 @@ static enum drm_mode_status imx6dl_hdmi_mode_valid(struct drm_connector *con,
 {
 	if (mode->clock < 13500)
 		return MODE_CLOCK_LOW;
-	if (mode->clock > 270000)
+	/* FIXME: Hardware is capable of 270MHz, but setup data is missing. */
+	if (mode->clock > 216000)
 		return MODE_CLOCK_HIGH;
 
 	return MODE_OK;
-- 
cgit v0.10.2


From ab0162b8fd338efad23b9d6fa6a0d2a0ff5e416f Mon Sep 17 00:00:00 2001
From: kbuild test robot <fengguang.wu@intel.com>
Date: Fri, 21 Aug 2015 19:05:39 +0800
Subject: drm/bridge: ps8622: Fix coccinelle warnings

The platform_no_drv_owner.cocci coccinelle script generates the
following warning:

	drivers/gpu/drm/bridge/parade-ps8622.c:671:3-8: No need to set .owner here. The core will do it.

Remove .owner field if calls are used which set it automatically. Patch
generated by scripts/coccinelle/api/platform_no_drv_owner.cocci.

Signed-off-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>

diff --git a/drivers/gpu/drm/bridge/parade-ps8622.c b/drivers/gpu/drm/bridge/parade-ps8622.c
index 1a6607b..be881e9 100644
--- a/drivers/gpu/drm/bridge/parade-ps8622.c
+++ b/drivers/gpu/drm/bridge/parade-ps8622.c
@@ -668,7 +668,6 @@ static struct i2c_driver ps8622_driver = {
 	.remove		= ps8622_remove,
 	.driver		= {
 		.name	= "ps8622",
-		.owner	= THIS_MODULE,
 		.of_match_table = ps8622_devices,
 	},
 };
-- 
cgit v0.10.2


From f1811a8a641329d0dada7c4db7e29ce5ae32a10b Mon Sep 17 00:00:00 2001
From: kbuild test robot <fengguang.wu@intel.com>
Date: Fri, 21 Aug 2015 19:05:39 +0800
Subject: drm/bridge: ptn3460: Fix coccinelle warnings

The platform_no_drv_owner.cocci coccinelle script generates the
following warning:

	drivers/gpu/drm/bridge/nxp-ptn3460.c:403:3-8: No need to set .owner here. The core will do it.

Remove .owner field if calls are used which set it automatically. Patch
generated by scripts/coccinelle/api/platform_no_drv_owner.cocci.

Signed-off-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>

diff --git a/drivers/gpu/drm/bridge/nxp-ptn3460.c b/drivers/gpu/drm/bridge/nxp-ptn3460.c
index 1b1bf23..0ffa3a6 100644
--- a/drivers/gpu/drm/bridge/nxp-ptn3460.c
+++ b/drivers/gpu/drm/bridge/nxp-ptn3460.c
@@ -400,7 +400,6 @@ static struct i2c_driver ptn3460_driver = {
 	.remove		= ptn3460_remove,
 	.driver		= {
 		.name	= "nxp,ptn3460",
-		.owner	= THIS_MODULE,
 		.of_match_table = ptn3460_match,
 	},
 };
-- 
cgit v0.10.2


From 2bcdcbfae2895764372ef973a969f395a548f1ea Mon Sep 17 00:00:00 2001
From: Thierry Reding <treding@nvidia.com>
Date: Mon, 24 Aug 2015 14:47:10 +0200
Subject: drm/tegra: dc: Request/free syncpoint at init/exit

syncpoints are resources provided by host1x and their lifetime is tied
to the host1x device. They are not properly reference counted either, so
removing the host1x device before any of its clients causes a use-after-
free error. Adding proper reference counting would be a major enterprise
so work around it for now by requesting and freeing the syncpoint at
init and exit time, respectively. The host1x device is guaranteed to be
around at this point.

Signed-off-by: Thierry Reding <treding@nvidia.com>

diff --git a/drivers/gpu/drm/tegra/dc.c b/drivers/gpu/drm/tegra/dc.c
index ddefb85..b463fd9 100644
--- a/drivers/gpu/drm/tegra/dc.c
+++ b/drivers/gpu/drm/tegra/dc.c
@@ -1696,6 +1696,7 @@ static int tegra_dc_debugfs_exit(struct tegra_dc *dc)
 static int tegra_dc_init(struct host1x_client *client)
 {
 	struct drm_device *drm = dev_get_drvdata(client->parent);
+	unsigned long flags = HOST1X_SYNCPT_CLIENT_MANAGED;
 	struct tegra_dc *dc = host1x_client_to_dc(client);
 	struct tegra_drm *tegra = drm->dev_private;
 	struct drm_plane *primary = NULL;
@@ -1703,6 +1704,10 @@ static int tegra_dc_init(struct host1x_client *client)
 	u32 value;
 	int err;
 
+	dc->syncpt = host1x_syncpt_request(dc->dev, flags);
+	if (!dc->syncpt)
+		dev_warn(dc->dev, "failed to allocate syncpoint\n");
+
 	if (tegra->domain) {
 		err = iommu_attach_device(tegra->domain, dc->dev);
 		if (err < 0) {
@@ -1849,6 +1854,8 @@ static int tegra_dc_exit(struct host1x_client *client)
 		dc->domain = NULL;
 	}
 
+	host1x_syncpt_free(dc->syncpt);
+
 	return 0;
 }
 
@@ -1961,7 +1968,6 @@ static int tegra_dc_parse_dt(struct tegra_dc *dc)
 
 static int tegra_dc_probe(struct platform_device *pdev)
 {
-	unsigned long flags = HOST1X_SYNCPT_CLIENT_MANAGED;
 	const struct of_device_id *id;
 	struct resource *regs;
 	struct tegra_dc *dc;
@@ -2036,10 +2042,6 @@ static int tegra_dc_probe(struct platform_device *pdev)
 		return -ENXIO;
 	}
 
-	dc->syncpt = host1x_syncpt_request(&pdev->dev, flags);
-	if (!dc->syncpt)
-		dev_warn(&pdev->dev, "failed to allocate syncpoint\n");
-
 	INIT_LIST_HEAD(&dc->client.list);
 	dc->client.ops = &dc_client_ops;
 	dc->client.dev = &pdev->dev;
@@ -2067,8 +2069,6 @@ static int tegra_dc_remove(struct platform_device *pdev)
 	struct tegra_dc *dc = platform_get_drvdata(pdev);
 	int err;
 
-	host1x_syncpt_free(dc->syncpt);
-
 	err = host1x_client_unregister(&dc->client);
 	if (err < 0) {
 		dev_err(&pdev->dev, "failed to unregister host1x client: %d\n",
-- 
cgit v0.10.2


From b10f1b20171945b49988b2b1fe68cb312cc36d32 Mon Sep 17 00:00:00 2001
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Date: Thu, 22 Oct 2015 13:56:34 +0200
Subject: drm/i915/skl: Prevent unclaimed register writes on skylake.

I'm getting unclaimed register writes when checking the WM registers
after the crtc is disabled. So I would imagine those are guarded by
the crtc power well. Fix this by not reading out wm state when the
power well is off.

Cc: stable@vger.kernel.org # v4.3
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=92181
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>

diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index 80175a3..d52a15d 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -2818,7 +2818,12 @@ void skl_ddb_get_hw_state(struct drm_i915_private *dev_priv,
 	int plane;
 	u32 val;
 
+	memset(ddb, 0, sizeof(*ddb));
+
 	for_each_pipe(dev_priv, pipe) {
+		if (!intel_display_power_is_enabled(dev_priv, POWER_DOMAIN_PIPE(pipe)))
+			continue;
+
 		for_each_plane(dev_priv, pipe, plane) {
 			val = I915_READ(PLANE_BUF_CFG(pipe, plane));
 			skl_ddb_entry_init_from_hw(&ddb->plane[pipe][plane],
-- 
cgit v0.10.2


From 5ad9dceee97477a7bdaa8c57b67cb802627a579d Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Wed, 28 Oct 2015 00:19:45 -0700
Subject: drm/vmwgfx: switch from ioremap_cache to memremap

Per commit 2e586a7e017a "drm/vmwgfx: Map the fifo as cached" the driver
expects the fifo registers to be cacheable.  In preparation for
deprecating ioremap_cache() convert its usage in vmwgfx to memremap().

Cc: David Airlie <airlied@linux.ie>
Cc: Thomas Hellstrom <thellstrom@vmware.com>
Cc: Sinclair Yeh <syeh@vmware.com>
Cc: dri-devel@lists.freedesktop.org
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Reviewed-by: Thomas Hellstrom <thellstrom@vmware.com>

diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
index b7525f7..bee0a45 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
@@ -752,8 +752,14 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset)
 	ttm_lock_set_kill(&dev_priv->fbdev_master.lock, false, SIGTERM);
 	dev_priv->active_master = &dev_priv->fbdev_master;
 
-	dev_priv->mmio_virt = ioremap_cache(dev_priv->mmio_start,
-					    dev_priv->mmio_size);
+	/*
+	 * Force __iomem for this mapping until the implied compiler
+	 * barriers and {READ|WRITE}_ONCE semantics from the
+	 * io{read|write}32() accessors can be replaced with explicit
+	 * barriers.
+	 */
+	dev_priv->mmio_virt = (void __iomem *) memremap(dev_priv->mmio_start,
+					    dev_priv->mmio_size, MEMREMAP_WB);
 
 	if (unlikely(dev_priv->mmio_virt == NULL)) {
 		ret = -ENOMEM;
@@ -907,7 +913,7 @@ out_no_irq:
 out_no_device:
 	ttm_object_device_release(&dev_priv->tdev);
 out_err4:
-	iounmap(dev_priv->mmio_virt);
+	memunmap((void __force *) dev_priv->mmio_virt);
 out_err3:
 	vmw_ttm_global_release(dev_priv);
 out_err0:
@@ -958,7 +964,7 @@ static int vmw_driver_unload(struct drm_device *dev)
 		pci_release_regions(dev->pdev);
 
 	ttm_object_device_release(&dev_priv->tdev);
-	iounmap(dev_priv->mmio_virt);
+	memunmap((void __force *) dev_priv->mmio_virt);
 	if (dev_priv->ctx.staged_bindings)
 		vmw_binding_state_free(dev_priv->ctx.staged_bindings);
 	vmw_ttm_global_release(dev_priv);
-- 
cgit v0.10.2


From 53c1e53571b1552204b2111b9fff5086f091bc9a Mon Sep 17 00:00:00 2001
From: Thomas Hellstrom <thellstrom@vmware.com>
Date: Mon, 26 Oct 2015 04:42:31 -0700
Subject: drm/vmwgfx: Turn off support for multisample count != 0 v2

Do this until we know how much MOB memory to allocate for these surfaces.
v2: Mask also non-DX multisample.

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: Brian Paul <brianp@vmware.com>

diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c
index 0a970af..a3e3c83 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c
@@ -122,6 +122,22 @@ int vmw_getparam_ioctl(struct drm_device *dev, void *data,
 	return 0;
 }
 
+static u32 vmw_mask_multisample(unsigned int cap, u32 fmt_value)
+{
+	/* If the header is updated, update the format test as well! */
+	BUILD_BUG_ON(SVGA3D_DEVCAP_DXFMT_BC5_UNORM + 1 != SVGA3D_DEVCAP_MAX);
+
+	if (cap >= SVGA3D_DEVCAP_DXFMT_X8R8G8B8 &&
+	    cap <= SVGA3D_DEVCAP_DXFMT_BC5_UNORM)
+		fmt_value &= ~(SVGADX_DXFMT_MULTISAMPLE_2 |
+			       SVGADX_DXFMT_MULTISAMPLE_4 |
+			       SVGADX_DXFMT_MULTISAMPLE_8);
+	else if (cap == SVGA3D_DEVCAP_MULTISAMPLE_MASKABLESAMPLES)
+		return 0;
+
+	return fmt_value;
+}
+
 static int vmw_fill_compat_cap(struct vmw_private *dev_priv, void *bounce,
 			       size_t size)
 {
@@ -147,7 +163,8 @@ static int vmw_fill_compat_cap(struct vmw_private *dev_priv, void *bounce,
 	for (i = 0; i < max_size; ++i) {
 		vmw_write(dev_priv, SVGA_REG_DEV_CAP, i);
 		compat_cap->pairs[i][0] = i;
-		compat_cap->pairs[i][1] = vmw_read(dev_priv, SVGA_REG_DEV_CAP);
+		compat_cap->pairs[i][1] = vmw_mask_multisample
+			(i, vmw_read(dev_priv, SVGA_REG_DEV_CAP));
 	}
 	spin_unlock(&dev_priv->cap_lock);
 
@@ -202,7 +219,8 @@ int vmw_get_cap_3d_ioctl(struct drm_device *dev, void *data,
 		spin_lock(&dev_priv->cap_lock);
 		for (i = 0; i < num; ++i) {
 			vmw_write(dev_priv, SVGA_REG_DEV_CAP, i);
-			*bounce32++ = vmw_read(dev_priv, SVGA_REG_DEV_CAP);
+			*bounce32++ = vmw_mask_multisample
+				(i, vmw_read(dev_priv, SVGA_REG_DEV_CAP));
 		}
 		spin_unlock(&dev_priv->cap_lock);
 	} else if (gb_objects) {
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c b/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c
index 64b5040..a279863 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_surface.c
@@ -1290,6 +1290,8 @@ int vmw_gb_surface_define_ioctl(struct drm_device *dev, void *data,
 	uint32_t size;
 	uint32_t backup_handle;
 
+	if (req->multisample_count != 0)
+		return -EINVAL;
 
 	if (unlikely(vmw_user_surface_size == 0))
 		vmw_user_surface_size = ttm_round_pot(sizeof(*user_srf)) +
-- 
cgit v0.10.2


From b76ff5ea1cf6fe648e7ce2b84e636f8a95849a0b Mon Sep 17 00:00:00 2001
From: Thomas Hellstrom <thellstrom@vmware.com>
Date: Wed, 28 Oct 2015 10:44:04 +0100
Subject: drm/vmwgfx: Replace iowrite/ioread with volatile memory accesses

Now that we use memremap instead of ioremap, Use WRITE_ONCE / READ_ONCE
instead of iowrite / ioread.

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: Sinclair Yeh <syeh@vmware.com>

diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
index bee0a45..d1c34ab 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
@@ -752,14 +752,8 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset)
 	ttm_lock_set_kill(&dev_priv->fbdev_master.lock, false, SIGTERM);
 	dev_priv->active_master = &dev_priv->fbdev_master;
 
-	/*
-	 * Force __iomem for this mapping until the implied compiler
-	 * barriers and {READ|WRITE}_ONCE semantics from the
-	 * io{read|write}32() accessors can be replaced with explicit
-	 * barriers.
-	 */
-	dev_priv->mmio_virt = (void __iomem *) memremap(dev_priv->mmio_start,
-					    dev_priv->mmio_size, MEMREMAP_WB);
+	dev_priv->mmio_virt = memremap(dev_priv->mmio_start,
+				       dev_priv->mmio_size, MEMREMAP_WB);
 
 	if (unlikely(dev_priv->mmio_virt == NULL)) {
 		ret = -ENOMEM;
@@ -913,7 +907,7 @@ out_no_irq:
 out_no_device:
 	ttm_object_device_release(&dev_priv->tdev);
 out_err4:
-	memunmap((void __force *) dev_priv->mmio_virt);
+	memunmap(dev_priv->mmio_virt);
 out_err3:
 	vmw_ttm_global_release(dev_priv);
 out_err0:
@@ -964,7 +958,7 @@ static int vmw_driver_unload(struct drm_device *dev)
 		pci_release_regions(dev->pdev);
 
 	ttm_object_device_release(&dev_priv->tdev);
-	memunmap((void __force *) dev_priv->mmio_virt);
+	memunmap(dev_priv->mmio_virt);
 	if (dev_priv->ctx.staged_bindings)
 		vmw_binding_state_free(dev_priv->ctx.staged_bindings);
 	vmw_ttm_global_release(dev_priv);
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
index a613bd4..198c8b1 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
@@ -375,7 +375,7 @@ struct vmw_private {
 	uint32_t stdu_max_height;
 	uint32_t initial_width;
 	uint32_t initial_height;
-	u32 __iomem *mmio_virt;
+	u32 *mmio_virt;
 	uint32_t capabilities;
 	uint32_t max_gmr_ids;
 	uint32_t max_gmr_pages;
@@ -1206,4 +1206,30 @@ static inline void vmw_fifo_resource_dec(struct vmw_private *dev_priv)
 {
 	atomic_dec(&dev_priv->num_fifo_resources);
 }
+
+/**
+ * vmw_mmio_read - Perform a MMIO read from volatile memory
+ *
+ * @addr: The address to read from
+ *
+ * This function is intended to be equivalent to ioread32() on
+ * memremap'd memory, but without byteswapping.
+ */
+static inline u32 vmw_mmio_read(u32 *addr)
+{
+	return READ_ONCE(*addr);
+}
+
+/**
+ * vmw_mmio_write - Perform a MMIO write to volatile memory
+ *
+ * @addr: The address to write to
+ *
+ * This function is intended to be equivalent to iowrite32 on
+ * memremap'd memory, but without byteswapping.
+ */
+static inline void vmw_mmio_write(u32 value, u32 *addr)
+{
+	WRITE_ONCE(*addr, value);
+}
 #endif
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
index 567dded..8e689b4 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
@@ -142,8 +142,8 @@ static bool vmw_fence_enable_signaling(struct fence *f)
 	struct vmw_fence_manager *fman = fman_from_fence(fence);
 	struct vmw_private *dev_priv = fman->dev_priv;
 
-	u32 __iomem *fifo_mem = dev_priv->mmio_virt;
-	u32 seqno = ioread32(fifo_mem + SVGA_FIFO_FENCE);
+	u32 *fifo_mem = dev_priv->mmio_virt;
+	u32 seqno = vmw_mmio_read(fifo_mem + SVGA_FIFO_FENCE);
 	if (seqno - fence->base.seqno < VMW_FENCE_WRAP)
 		return false;
 
@@ -386,14 +386,14 @@ static bool vmw_fence_goal_new_locked(struct vmw_fence_manager *fman,
 				      u32 passed_seqno)
 {
 	u32 goal_seqno;
-	u32 __iomem *fifo_mem;
+	u32 *fifo_mem;
 	struct vmw_fence_obj *fence;
 
 	if (likely(!fman->seqno_valid))
 		return false;
 
 	fifo_mem = fman->dev_priv->mmio_virt;
-	goal_seqno = ioread32(fifo_mem + SVGA_FIFO_FENCE_GOAL);
+	goal_seqno = vmw_mmio_read(fifo_mem + SVGA_FIFO_FENCE_GOAL);
 	if (likely(passed_seqno - goal_seqno >= VMW_FENCE_WRAP))
 		return false;
 
@@ -401,8 +401,8 @@ static bool vmw_fence_goal_new_locked(struct vmw_fence_manager *fman,
 	list_for_each_entry(fence, &fman->fence_list, head) {
 		if (!list_empty(&fence->seq_passed_actions)) {
 			fman->seqno_valid = true;
-			iowrite32(fence->base.seqno,
-				  fifo_mem + SVGA_FIFO_FENCE_GOAL);
+			vmw_mmio_write(fence->base.seqno,
+				       fifo_mem + SVGA_FIFO_FENCE_GOAL);
 			break;
 		}
 	}
@@ -430,18 +430,18 @@ static bool vmw_fence_goal_check_locked(struct vmw_fence_obj *fence)
 {
 	struct vmw_fence_manager *fman = fman_from_fence(fence);
 	u32 goal_seqno;
-	u32 __iomem *fifo_mem;
+	u32 *fifo_mem;
 
 	if (fence_is_signaled_locked(&fence->base))
 		return false;
 
 	fifo_mem = fman->dev_priv->mmio_virt;
-	goal_seqno = ioread32(fifo_mem + SVGA_FIFO_FENCE_GOAL);
+	goal_seqno = vmw_mmio_read(fifo_mem + SVGA_FIFO_FENCE_GOAL);
 	if (likely(fman->seqno_valid &&
 		   goal_seqno - fence->base.seqno < VMW_FENCE_WRAP))
 		return false;
 
-	iowrite32(fence->base.seqno, fifo_mem + SVGA_FIFO_FENCE_GOAL);
+	vmw_mmio_write(fence->base.seqno, fifo_mem + SVGA_FIFO_FENCE_GOAL);
 	fman->seqno_valid = true;
 
 	return true;
@@ -453,9 +453,9 @@ static void __vmw_fences_update(struct vmw_fence_manager *fman)
 	struct list_head action_list;
 	bool needs_rerun;
 	uint32_t seqno, new_seqno;
-	u32 __iomem *fifo_mem = fman->dev_priv->mmio_virt;
+	u32 *fifo_mem = fman->dev_priv->mmio_virt;
 
-	seqno = ioread32(fifo_mem + SVGA_FIFO_FENCE);
+	seqno = vmw_mmio_read(fifo_mem + SVGA_FIFO_FENCE);
 rerun:
 	list_for_each_entry_safe(fence, next_fence, &fman->fence_list, head) {
 		if (seqno - fence->base.seqno < VMW_FENCE_WRAP) {
@@ -477,7 +477,7 @@ rerun:
 
 	needs_rerun = vmw_fence_goal_new_locked(fman, seqno);
 	if (unlikely(needs_rerun)) {
-		new_seqno = ioread32(fifo_mem + SVGA_FIFO_FENCE);
+		new_seqno = vmw_mmio_read(fifo_mem + SVGA_FIFO_FENCE);
 		if (new_seqno != seqno) {
 			seqno = new_seqno;
 			goto rerun;
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c b/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c
index 80c40c3..0cbaf88 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c
@@ -36,7 +36,7 @@ struct vmw_temp_set_context {
 
 bool vmw_fifo_have_3d(struct vmw_private *dev_priv)
 {
-	u32 __iomem *fifo_mem = dev_priv->mmio_virt;
+	u32 *fifo_mem = dev_priv->mmio_virt;
 	uint32_t fifo_min, hwversion;
 	const struct vmw_fifo_state *fifo = &dev_priv->fifo;
 
@@ -60,15 +60,15 @@ bool vmw_fifo_have_3d(struct vmw_private *dev_priv)
 	if (!(dev_priv->capabilities & SVGA_CAP_EXTENDED_FIFO))
 		return false;
 
-	fifo_min = ioread32(fifo_mem  + SVGA_FIFO_MIN);
+	fifo_min = vmw_mmio_read(fifo_mem  + SVGA_FIFO_MIN);
 	if (fifo_min <= SVGA_FIFO_3D_HWVERSION * sizeof(unsigned int))
 		return false;
 
-	hwversion = ioread32(fifo_mem +
-			     ((fifo->capabilities &
-			       SVGA_FIFO_CAP_3D_HWVERSION_REVISED) ?
-			      SVGA_FIFO_3D_HWVERSION_REVISED :
-			      SVGA_FIFO_3D_HWVERSION));
+	hwversion = vmw_mmio_read(fifo_mem +
+				  ((fifo->capabilities &
+				    SVGA_FIFO_CAP_3D_HWVERSION_REVISED) ?
+				   SVGA_FIFO_3D_HWVERSION_REVISED :
+				   SVGA_FIFO_3D_HWVERSION));
 
 	if (hwversion == 0)
 		return false;
@@ -85,13 +85,13 @@ bool vmw_fifo_have_3d(struct vmw_private *dev_priv)
 
 bool vmw_fifo_have_pitchlock(struct vmw_private *dev_priv)
 {
-	u32 __iomem *fifo_mem = dev_priv->mmio_virt;
+	u32  *fifo_mem = dev_priv->mmio_virt;
 	uint32_t caps;
 
 	if (!(dev_priv->capabilities & SVGA_CAP_EXTENDED_FIFO))
 		return false;
 
-	caps = ioread32(fifo_mem + SVGA_FIFO_CAPABILITIES);
+	caps = vmw_mmio_read(fifo_mem + SVGA_FIFO_CAPABILITIES);
 	if (caps & SVGA_FIFO_CAP_PITCHLOCK)
 		return true;
 
@@ -100,7 +100,7 @@ bool vmw_fifo_have_pitchlock(struct vmw_private *dev_priv)
 
 int vmw_fifo_init(struct vmw_private *dev_priv, struct vmw_fifo_state *fifo)
 {
-	u32 __iomem *fifo_mem = dev_priv->mmio_virt;
+	u32  *fifo_mem = dev_priv->mmio_virt;
 	uint32_t max;
 	uint32_t min;
 
@@ -137,19 +137,19 @@ int vmw_fifo_init(struct vmw_private *dev_priv, struct vmw_fifo_state *fifo)
 	if (min < PAGE_SIZE)
 		min = PAGE_SIZE;
 
-	iowrite32(min, fifo_mem + SVGA_FIFO_MIN);
-	iowrite32(dev_priv->mmio_size, fifo_mem + SVGA_FIFO_MAX);
+	vmw_mmio_write(min, fifo_mem + SVGA_FIFO_MIN);
+	vmw_mmio_write(dev_priv->mmio_size, fifo_mem + SVGA_FIFO_MAX);
 	wmb();
-	iowrite32(min,  fifo_mem + SVGA_FIFO_NEXT_CMD);
-	iowrite32(min,  fifo_mem + SVGA_FIFO_STOP);
-	iowrite32(0, fifo_mem + SVGA_FIFO_BUSY);
+	vmw_mmio_write(min,  fifo_mem + SVGA_FIFO_NEXT_CMD);
+	vmw_mmio_write(min,  fifo_mem + SVGA_FIFO_STOP);
+	vmw_mmio_write(0, fifo_mem + SVGA_FIFO_BUSY);
 	mb();
 
 	vmw_write(dev_priv, SVGA_REG_CONFIG_DONE, 1);
 
-	max = ioread32(fifo_mem + SVGA_FIFO_MAX);
-	min = ioread32(fifo_mem  + SVGA_FIFO_MIN);
-	fifo->capabilities = ioread32(fifo_mem + SVGA_FIFO_CAPABILITIES);
+	max = vmw_mmio_read(fifo_mem + SVGA_FIFO_MAX);
+	min = vmw_mmio_read(fifo_mem  + SVGA_FIFO_MIN);
+	fifo->capabilities = vmw_mmio_read(fifo_mem + SVGA_FIFO_CAPABILITIES);
 
 	DRM_INFO("Fifo max 0x%08x min 0x%08x cap 0x%08x\n",
 		 (unsigned int) max,
@@ -157,7 +157,7 @@ int vmw_fifo_init(struct vmw_private *dev_priv, struct vmw_fifo_state *fifo)
 		 (unsigned int) fifo->capabilities);
 
 	atomic_set(&dev_priv->marker_seq, dev_priv->last_read_seqno);
-	iowrite32(dev_priv->last_read_seqno, fifo_mem + SVGA_FIFO_FENCE);
+	vmw_mmio_write(dev_priv->last_read_seqno, fifo_mem + SVGA_FIFO_FENCE);
 	vmw_marker_queue_init(&fifo->marker_queue);
 
 	return 0;
@@ -165,31 +165,23 @@ int vmw_fifo_init(struct vmw_private *dev_priv, struct vmw_fifo_state *fifo)
 
 void vmw_fifo_ping_host(struct vmw_private *dev_priv, uint32_t reason)
 {
-	u32 __iomem *fifo_mem = dev_priv->mmio_virt;
-	static DEFINE_SPINLOCK(ping_lock);
-	unsigned long irq_flags;
+	u32 *fifo_mem = dev_priv->mmio_virt;
 
-	/*
-	 * The ping_lock is needed because we don't have an atomic
-	 * test-and-set of the SVGA_FIFO_BUSY register.
-	 */
-	spin_lock_irqsave(&ping_lock, irq_flags);
-	if (unlikely(ioread32(fifo_mem + SVGA_FIFO_BUSY) == 0)) {
-		iowrite32(1, fifo_mem + SVGA_FIFO_BUSY);
+	preempt_disable();
+	if (cmpxchg(fifo_mem + SVGA_FIFO_BUSY, 0, 1) == 0)
 		vmw_write(dev_priv, SVGA_REG_SYNC, reason);
-	}
-	spin_unlock_irqrestore(&ping_lock, irq_flags);
+	preempt_enable();
 }
 
 void vmw_fifo_release(struct vmw_private *dev_priv, struct vmw_fifo_state *fifo)
 {
-	u32 __iomem *fifo_mem = dev_priv->mmio_virt;
+	u32  *fifo_mem = dev_priv->mmio_virt;
 
 	vmw_write(dev_priv, SVGA_REG_SYNC, SVGA_SYNC_GENERIC);
 	while (vmw_read(dev_priv, SVGA_REG_BUSY) != 0)
 		;
 
-	dev_priv->last_read_seqno = ioread32(fifo_mem + SVGA_FIFO_FENCE);
+	dev_priv->last_read_seqno = vmw_mmio_read(fifo_mem + SVGA_FIFO_FENCE);
 
 	vmw_write(dev_priv, SVGA_REG_CONFIG_DONE,
 		  dev_priv->config_done_state);
@@ -213,11 +205,11 @@ void vmw_fifo_release(struct vmw_private *dev_priv, struct vmw_fifo_state *fifo)
 
 static bool vmw_fifo_is_full(struct vmw_private *dev_priv, uint32_t bytes)
 {
-	u32 __iomem *fifo_mem = dev_priv->mmio_virt;
-	uint32_t max = ioread32(fifo_mem + SVGA_FIFO_MAX);
-	uint32_t next_cmd = ioread32(fifo_mem + SVGA_FIFO_NEXT_CMD);
-	uint32_t min = ioread32(fifo_mem + SVGA_FIFO_MIN);
-	uint32_t stop = ioread32(fifo_mem + SVGA_FIFO_STOP);
+	u32  *fifo_mem = dev_priv->mmio_virt;
+	uint32_t max = vmw_mmio_read(fifo_mem + SVGA_FIFO_MAX);
+	uint32_t next_cmd = vmw_mmio_read(fifo_mem + SVGA_FIFO_NEXT_CMD);
+	uint32_t min = vmw_mmio_read(fifo_mem + SVGA_FIFO_MIN);
+	uint32_t stop = vmw_mmio_read(fifo_mem + SVGA_FIFO_STOP);
 
 	return ((max - next_cmd) + (stop - min) <= bytes);
 }
@@ -321,7 +313,7 @@ static void *vmw_local_fifo_reserve(struct vmw_private *dev_priv,
 				    uint32_t bytes)
 {
 	struct vmw_fifo_state *fifo_state = &dev_priv->fifo;
-	u32 __iomem *fifo_mem = dev_priv->mmio_virt;
+	u32  *fifo_mem = dev_priv->mmio_virt;
 	uint32_t max;
 	uint32_t min;
 	uint32_t next_cmd;
@@ -329,9 +321,9 @@ static void *vmw_local_fifo_reserve(struct vmw_private *dev_priv,
 	int ret;
 
 	mutex_lock(&fifo_state->fifo_mutex);
-	max = ioread32(fifo_mem + SVGA_FIFO_MAX);
-	min = ioread32(fifo_mem + SVGA_FIFO_MIN);
-	next_cmd = ioread32(fifo_mem + SVGA_FIFO_NEXT_CMD);
+	max = vmw_mmio_read(fifo_mem + SVGA_FIFO_MAX);
+	min = vmw_mmio_read(fifo_mem + SVGA_FIFO_MIN);
+	next_cmd = vmw_mmio_read(fifo_mem + SVGA_FIFO_NEXT_CMD);
 
 	if (unlikely(bytes >= (max - min)))
 		goto out_err;
@@ -342,7 +334,7 @@ static void *vmw_local_fifo_reserve(struct vmw_private *dev_priv,
 	fifo_state->reserved_size = bytes;
 
 	while (1) {
-		uint32_t stop = ioread32(fifo_mem + SVGA_FIFO_STOP);
+		uint32_t stop = vmw_mmio_read(fifo_mem + SVGA_FIFO_STOP);
 		bool need_bounce = false;
 		bool reserve_in_place = false;
 
@@ -376,8 +368,8 @@ static void *vmw_local_fifo_reserve(struct vmw_private *dev_priv,
 				fifo_state->using_bounce_buffer = false;
 
 				if (reserveable)
-					iowrite32(bytes, fifo_mem +
-						  SVGA_FIFO_RESERVED);
+					vmw_mmio_write(bytes, fifo_mem +
+						       SVGA_FIFO_RESERVED);
 				return (void __force *) (fifo_mem +
 							 (next_cmd >> 2));
 			} else {
@@ -427,7 +419,7 @@ void *vmw_fifo_reserve_dx(struct vmw_private *dev_priv, uint32_t bytes,
 }
 
 static void vmw_fifo_res_copy(struct vmw_fifo_state *fifo_state,
-			      u32 __iomem *fifo_mem,
+			      u32  *fifo_mem,
 			      uint32_t next_cmd,
 			      uint32_t max, uint32_t min, uint32_t bytes)
 {
@@ -439,17 +431,16 @@ static void vmw_fifo_res_copy(struct vmw_fifo_state *fifo_state,
 	if (bytes < chunk_size)
 		chunk_size = bytes;
 
-	iowrite32(bytes, fifo_mem + SVGA_FIFO_RESERVED);
+	vmw_mmio_write(bytes, fifo_mem + SVGA_FIFO_RESERVED);
 	mb();
-	memcpy_toio(fifo_mem + (next_cmd >> 2), buffer, chunk_size);
+	memcpy(fifo_mem + (next_cmd >> 2), buffer, chunk_size);
 	rest = bytes - chunk_size;
 	if (rest)
-		memcpy_toio(fifo_mem + (min >> 2), buffer + (chunk_size >> 2),
-			    rest);
+		memcpy(fifo_mem + (min >> 2), buffer + (chunk_size >> 2), rest);
 }
 
 static void vmw_fifo_slow_copy(struct vmw_fifo_state *fifo_state,
-			       u32 __iomem *fifo_mem,
+			       u32  *fifo_mem,
 			       uint32_t next_cmd,
 			       uint32_t max, uint32_t min, uint32_t bytes)
 {
@@ -457,12 +448,12 @@ static void vmw_fifo_slow_copy(struct vmw_fifo_state *fifo_state,
 	    fifo_state->dynamic_buffer : fifo_state->static_buffer;
 
 	while (bytes > 0) {
-		iowrite32(*buffer++, fifo_mem + (next_cmd >> 2));
+		vmw_mmio_write(*buffer++, fifo_mem + (next_cmd >> 2));
 		next_cmd += sizeof(uint32_t);
 		if (unlikely(next_cmd == max))
 			next_cmd = min;
 		mb();
-		iowrite32(next_cmd, fifo_mem + SVGA_FIFO_NEXT_CMD);
+		vmw_mmio_write(next_cmd, fifo_mem + SVGA_FIFO_NEXT_CMD);
 		mb();
 		bytes -= sizeof(uint32_t);
 	}
@@ -471,10 +462,10 @@ static void vmw_fifo_slow_copy(struct vmw_fifo_state *fifo_state,
 static void vmw_local_fifo_commit(struct vmw_private *dev_priv, uint32_t bytes)
 {
 	struct vmw_fifo_state *fifo_state = &dev_priv->fifo;
-	u32 __iomem *fifo_mem = dev_priv->mmio_virt;
-	uint32_t next_cmd = ioread32(fifo_mem + SVGA_FIFO_NEXT_CMD);
-	uint32_t max = ioread32(fifo_mem + SVGA_FIFO_MAX);
-	uint32_t min = ioread32(fifo_mem + SVGA_FIFO_MIN);
+	u32  *fifo_mem = dev_priv->mmio_virt;
+	uint32_t next_cmd = vmw_mmio_read(fifo_mem + SVGA_FIFO_NEXT_CMD);
+	uint32_t max = vmw_mmio_read(fifo_mem + SVGA_FIFO_MAX);
+	uint32_t min = vmw_mmio_read(fifo_mem + SVGA_FIFO_MIN);
 	bool reserveable = fifo_state->capabilities & SVGA_FIFO_CAP_RESERVE;
 
 	if (fifo_state->dx)
@@ -507,11 +498,11 @@ static void vmw_local_fifo_commit(struct vmw_private *dev_priv, uint32_t bytes)
 		if (next_cmd >= max)
 			next_cmd -= max - min;
 		mb();
-		iowrite32(next_cmd, fifo_mem + SVGA_FIFO_NEXT_CMD);
+		vmw_mmio_write(next_cmd, fifo_mem + SVGA_FIFO_NEXT_CMD);
 	}
 
 	if (reserveable)
-		iowrite32(0, fifo_mem + SVGA_FIFO_RESERVED);
+		vmw_mmio_write(0, fifo_mem + SVGA_FIFO_RESERVED);
 	mb();
 	up_write(&fifo_state->rwsem);
 	vmw_fifo_ping_host(dev_priv, SVGA_SYNC_GENERIC);
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c
index a3e3c83..b8c6a03 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ioctl.c
@@ -64,7 +64,7 @@ int vmw_getparam_ioctl(struct drm_device *dev, void *data,
 		break;
 	case DRM_VMW_PARAM_FIFO_HW_VERSION:
 	{
-		u32 __iomem *fifo_mem = dev_priv->mmio_virt;
+		u32 *fifo_mem = dev_priv->mmio_virt;
 		const struct vmw_fifo_state *fifo = &dev_priv->fifo;
 
 		if ((dev_priv->capabilities & SVGA_CAP_GBOBJECTS)) {
@@ -73,11 +73,11 @@ int vmw_getparam_ioctl(struct drm_device *dev, void *data,
 		}
 
 		param->value =
-			ioread32(fifo_mem +
-				 ((fifo->capabilities &
-				   SVGA_FIFO_CAP_3D_HWVERSION_REVISED) ?
-				  SVGA_FIFO_3D_HWVERSION_REVISED :
-				  SVGA_FIFO_3D_HWVERSION));
+			vmw_mmio_read(fifo_mem +
+				      ((fifo->capabilities &
+					SVGA_FIFO_CAP_3D_HWVERSION_REVISED) ?
+				       SVGA_FIFO_3D_HWVERSION_REVISED :
+				       SVGA_FIFO_3D_HWVERSION));
 		break;
 	}
 	case DRM_VMW_PARAM_MAX_SURF_MEMORY:
@@ -179,7 +179,7 @@ int vmw_get_cap_3d_ioctl(struct drm_device *dev, void *data,
 		(struct drm_vmw_get_3d_cap_arg *) data;
 	struct vmw_private *dev_priv = vmw_priv(dev);
 	uint32_t size;
-	u32 __iomem *fifo_mem;
+	u32 *fifo_mem;
 	void __user *buffer = (void __user *)((unsigned long)(arg->buffer));
 	void *bounce;
 	int ret;
@@ -229,7 +229,7 @@ int vmw_get_cap_3d_ioctl(struct drm_device *dev, void *data,
 			goto out_err;
 	} else {
 		fifo_mem = dev_priv->mmio_virt;
-		memcpy_fromio(bounce, &fifo_mem[SVGA_FIFO_3D_CAPS], size);
+		memcpy(bounce, &fifo_mem[SVGA_FIFO_3D_CAPS], size);
 	}
 
 	ret = copy_to_user(buffer, bounce, size);
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_irq.c b/drivers/gpu/drm/vmwgfx/vmwgfx_irq.c
index 9498a5e..ac3eccd 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_irq.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_irq.c
@@ -72,8 +72,8 @@ static bool vmw_fifo_idle(struct vmw_private *dev_priv, uint32_t seqno)
 void vmw_update_seqno(struct vmw_private *dev_priv,
 			 struct vmw_fifo_state *fifo_state)
 {
-	u32 __iomem *fifo_mem = dev_priv->mmio_virt;
-	uint32_t seqno = ioread32(fifo_mem + SVGA_FIFO_FENCE);
+	u32 *fifo_mem = dev_priv->mmio_virt;
+	uint32_t seqno = vmw_mmio_read(fifo_mem + SVGA_FIFO_FENCE);
 
 	if (dev_priv->last_read_seqno != seqno) {
 		dev_priv->last_read_seqno = seqno;
@@ -178,8 +178,9 @@ int vmw_fallback_wait(struct vmw_private *dev_priv,
 	}
 	finish_wait(&dev_priv->fence_queue, &__wait);
 	if (ret == 0 && fifo_idle) {
-		u32 __iomem *fifo_mem = dev_priv->mmio_virt;
-		iowrite32(signal_seq, fifo_mem + SVGA_FIFO_FENCE);
+		u32 *fifo_mem = dev_priv->mmio_virt;
+
+		vmw_mmio_write(signal_seq, fifo_mem + SVGA_FIFO_FENCE);
 	}
 	wake_up_all(&dev_priv->fence_queue);
 out_err:
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
index 03ffab2..a94b24d 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
@@ -123,14 +123,14 @@ err_unreserve:
 void vmw_cursor_update_position(struct vmw_private *dev_priv,
 				bool show, int x, int y)
 {
-	u32 __iomem *fifo_mem = dev_priv->mmio_virt;
+	u32 *fifo_mem = dev_priv->mmio_virt;
 	uint32_t count;
 
-	iowrite32(show ? 1 : 0, fifo_mem + SVGA_FIFO_CURSOR_ON);
-	iowrite32(x, fifo_mem + SVGA_FIFO_CURSOR_X);
-	iowrite32(y, fifo_mem + SVGA_FIFO_CURSOR_Y);
-	count = ioread32(fifo_mem + SVGA_FIFO_CURSOR_COUNT);
-	iowrite32(++count, fifo_mem + SVGA_FIFO_CURSOR_COUNT);
+	vmw_mmio_write(show ? 1 : 0, fifo_mem + SVGA_FIFO_CURSOR_ON);
+	vmw_mmio_write(x, fifo_mem + SVGA_FIFO_CURSOR_X);
+	vmw_mmio_write(y, fifo_mem + SVGA_FIFO_CURSOR_Y);
+	count = vmw_mmio_read(fifo_mem + SVGA_FIFO_CURSOR_COUNT);
+	vmw_mmio_write(++count, fifo_mem + SVGA_FIFO_CURSOR_COUNT);
 }
 
 int vmw_du_crtc_cursor_set(struct drm_crtc *crtc, struct drm_file *file_priv,
@@ -1155,7 +1155,8 @@ int vmw_kms_write_svga(struct vmw_private *vmw_priv,
 	if (vmw_priv->capabilities & SVGA_CAP_PITCHLOCK)
 		vmw_write(vmw_priv, SVGA_REG_PITCHLOCK, pitch);
 	else if (vmw_fifo_have_pitchlock(vmw_priv))
-		iowrite32(pitch, vmw_priv->mmio_virt + SVGA_FIFO_PITCHLOCK);
+		vmw_mmio_write(pitch, vmw_priv->mmio_virt +
+			       SVGA_FIFO_PITCHLOCK);
 	vmw_write(vmw_priv, SVGA_REG_WIDTH, width);
 	vmw_write(vmw_priv, SVGA_REG_HEIGHT, height);
 	vmw_write(vmw_priv, SVGA_REG_BITS_PER_PIXEL, bpp);
@@ -1181,8 +1182,8 @@ int vmw_kms_save_vga(struct vmw_private *vmw_priv)
 		vmw_priv->vga_pitchlock =
 		  vmw_read(vmw_priv, SVGA_REG_PITCHLOCK);
 	else if (vmw_fifo_have_pitchlock(vmw_priv))
-		vmw_priv->vga_pitchlock = ioread32(vmw_priv->mmio_virt +
-						   SVGA_FIFO_PITCHLOCK);
+		vmw_priv->vga_pitchlock = vmw_mmio_read(vmw_priv->mmio_virt +
+							SVGA_FIFO_PITCHLOCK);
 
 	if (!(vmw_priv->capabilities & SVGA_CAP_DISPLAY_TOPOLOGY))
 		return 0;
@@ -1230,8 +1231,8 @@ int vmw_kms_restore_vga(struct vmw_private *vmw_priv)
 		vmw_write(vmw_priv, SVGA_REG_PITCHLOCK,
 			  vmw_priv->vga_pitchlock);
 	else if (vmw_fifo_have_pitchlock(vmw_priv))
-		iowrite32(vmw_priv->vga_pitchlock,
-			  vmw_priv->mmio_virt + SVGA_FIFO_PITCHLOCK);
+		vmw_mmio_write(vmw_priv->vga_pitchlock,
+			       vmw_priv->mmio_virt + SVGA_FIFO_PITCHLOCK);
 
 	if (!(vmw_priv->capabilities & SVGA_CAP_DISPLAY_TOPOLOGY))
 		return 0;
-- 
cgit v0.10.2


From 9be64eee3a87dc03218ca9a12834d1150a57b8a8 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 30 Oct 2015 14:50:24 +0200
Subject: drm/i915: add quirk to enable backlight on Dell Chromebook 11 (2015)

Reported-by: Keith Webb <khwebb@gmail.com>
Suggested-by: Keith Webb <khwebb@gmail.com>
Cc: stable@vger.kernel.org
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=106671
Reviewed-by: Clint Taylor <Clinton.A.Taylor@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1446209424-28801-1-git-send-email-jani.nikula@intel.com

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 82128b9..81358ca 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -14695,6 +14695,9 @@ static struct intel_quirk intel_quirks[] = {
 
 	/* Dell Chromebook 11 */
 	{ 0x0a06, 0x1028, 0x0a35, quirk_backlight_present },
+
+	/* Dell Chromebook 11 (2015 version) */
+	{ 0x0a16, 0x1028, 0x0a35, quirk_backlight_present },
 };
 
 static void intel_init_quirks(struct drm_device *dev)
-- 
cgit v0.10.2


From 1d1106b0f6b5cb4bc1b88d7bd4c41d0413331c5d Mon Sep 17 00:00:00 2001
From: Harry Wentland <harry.wentland@amd.com>
Date: Wed, 15 Jul 2015 07:10:41 -0400
Subject: drm/amdgpu: Make amdgpu_mn functions inline
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Unused amdgpu_mn functions threw warnings for every file that includes
amdgpu.h. It makes sense to inline this amdgpu_mn stubs to avoid the warning.

Signed-off-by: Harry Wentland <harry.wentland@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Cc: stable@vger.kernel.org

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 8305a6c..510cade 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -1752,11 +1752,11 @@ void amdgpu_test_syncing(struct amdgpu_device *adev);
 int amdgpu_mn_register(struct amdgpu_bo *bo, unsigned long addr);
 void amdgpu_mn_unregister(struct amdgpu_bo *bo);
 #else
-static int amdgpu_mn_register(struct amdgpu_bo *bo, unsigned long addr)
+static inline int amdgpu_mn_register(struct amdgpu_bo *bo, unsigned long addr)
 {
 	return -ENODEV;
 }
-static void amdgpu_mn_unregister(struct amdgpu_bo *bo) {}
+static inline void amdgpu_mn_unregister(struct amdgpu_bo *bo) {}
 #endif
 
 /*
-- 
cgit v0.10.2


From 4f54f21cd68eb7b2ed355b9de90a2d58b96fdda9 Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <a.hajda@samsung.com>
Date: Tue, 20 Oct 2015 11:22:34 +0200
Subject: drm/exynos/decon5433: add PCLK clock

PCLK clock is used by DECON IP. The patch also replaces magic number with
number of clocks in array definition.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
index 1ea26dbb..b25d764 100644
--- a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
+++ b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
@@ -27,13 +27,23 @@
 #define CURSOR_WIN	2
 #define MIN_FB_WIDTH_FOR_16WORD_BURST	128
 
+static const char * const decon_clks_name[] = {
+	"pclk",
+	"aclk_decon",
+	"aclk_smmu_decon0x",
+	"aclk_xiu_decon0x",
+	"pclk_smmu_decon0x",
+	"sclk_decon_vclk",
+	"sclk_decon_eclk",
+};
+
 struct decon_context {
 	struct device			*dev;
 	struct drm_device		*drm_dev;
 	struct exynos_drm_crtc		*crtc;
 	struct exynos_drm_plane		planes[WINDOWS_NR];
 	void __iomem			*addr;
-	struct clk			*clks[6];
+	struct clk			*clks[ARRAY_SIZE(decon_clks_name)];
 	unsigned long			irq_flags;
 	int				pipe;
 	bool				suspended;
@@ -45,15 +55,6 @@ struct decon_context {
 	atomic_t			win_updated;
 };
 
-static const char * const decon_clks_name[] = {
-	"aclk_decon",
-	"aclk_smmu_decon0x",
-	"aclk_xiu_decon0x",
-	"pclk_smmu_decon0x",
-	"sclk_decon_vclk",
-	"sclk_decon_eclk",
-};
-
 static const uint32_t decon_formats[] = {
 	DRM_FORMAT_XRGB1555,
 	DRM_FORMAT_RGB565,
-- 
cgit v0.10.2


From 85de275ad9aff7dae5811fc583a6a43e7b5cc0f1 Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <a.hajda@samsung.com>
Date: Tue, 20 Oct 2015 11:22:36 +0200
Subject: drm/exynos/decon5433: fix timing registers writes

All timing registers should contain values decreased by one.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
index b25d764..83e0939 100644
--- a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
+++ b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
@@ -104,7 +104,7 @@ static void decon_setup_trigger(struct decon_context *ctx)
 static void decon_commit(struct exynos_drm_crtc *crtc)
 {
 	struct decon_context *ctx = crtc->ctx;
-	struct drm_display_mode *mode = &crtc->base.mode;
+	struct drm_display_mode *m = &crtc->base.mode;
 	u32 val;
 
 	if (ctx->suspended)
@@ -122,29 +122,29 @@ static void decon_commit(struct exynos_drm_crtc *crtc)
 		val |= VIDOUT_RGB_IF;
 	writel(val, ctx->addr + DECON_VIDOUTCON0);
 
-	val = VIDTCON2_LINEVAL(mode->vdisplay - 1) |
-		VIDTCON2_HOZVAL(mode->hdisplay - 1);
+	val = VIDTCON2_LINEVAL(m->vdisplay - 1) |
+		VIDTCON2_HOZVAL(m->hdisplay - 1);
 	writel(val, ctx->addr + DECON_VIDTCON2);
 
 	if (!ctx->i80_if) {
 		val = VIDTCON00_VBPD_F(
-				mode->crtc_vtotal - mode->crtc_vsync_end) |
+				m->crtc_vtotal - m->crtc_vsync_end - 1) |
 			VIDTCON00_VFPD_F(
-				mode->crtc_vsync_start - mode->crtc_vdisplay);
+				m->crtc_vsync_start - m->crtc_vdisplay - 1);
 		writel(val, ctx->addr + DECON_VIDTCON00);
 
 		val = VIDTCON01_VSPW_F(
-				mode->crtc_vsync_end - mode->crtc_vsync_start);
+				m->crtc_vsync_end - m->crtc_vsync_start - 1);
 		writel(val, ctx->addr + DECON_VIDTCON01);
 
 		val = VIDTCON10_HBPD_F(
-				mode->crtc_htotal - mode->crtc_hsync_end) |
+				m->crtc_htotal - m->crtc_hsync_end - 1) |
 			VIDTCON10_HFPD_F(
-				mode->crtc_hsync_start - mode->crtc_hdisplay);
+				m->crtc_hsync_start - m->crtc_hdisplay - 1);
 		writel(val, ctx->addr + DECON_VIDTCON10);
 
 		val = VIDTCON11_HSPW_F(
-				mode->crtc_hsync_end - mode->crtc_hsync_start);
+				m->crtc_hsync_end - m->crtc_hsync_start - 1);
 		writel(val, ctx->addr + DECON_VIDTCON11);
 	}
 
-- 
cgit v0.10.2


From b219207385dbd575ebd7e1cbf560636c8e780467 Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <a.hajda@samsung.com>
Date: Tue, 20 Oct 2015 11:22:37 +0200
Subject: drm/exynos/decon5433: add function to set particular register bits

The driver often sets only particular bits of configuration registers.
Using separate function to such action simplifies the code.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
index 83e0939..722c11a 100644
--- a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
+++ b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
@@ -62,6 +62,13 @@ static const uint32_t decon_formats[] = {
 	DRM_FORMAT_ARGB8888,
 };
 
+static inline void decon_set_bits(struct decon_context *ctx, u32 reg, u32 mask,
+				  u32 val)
+{
+	val = (val & mask) | (readl(ctx->addr + reg) & ~mask);
+	writel(val, ctx->addr + reg);
+}
+
 static int decon_enable_vblank(struct exynos_drm_crtc *crtc)
 {
 	struct decon_context *ctx = crtc->ctx;
@@ -215,16 +222,8 @@ static void decon_win_set_pixfmt(struct decon_context *ctx, unsigned int win,
 static void decon_shadow_protect_win(struct decon_context *ctx, int win,
 					bool protect)
 {
-	u32 val;
-
-	val = readl(ctx->addr + DECON_SHADOWCON);
-
-	if (protect)
-		val |= SHADOWCON_Wx_PROTECT(win);
-	else
-		val &= ~SHADOWCON_Wx_PROTECT(win);
-
-	writel(val, ctx->addr + DECON_SHADOWCON);
+	decon_set_bits(ctx, DECON_SHADOWCON, SHADOWCON_Wx_PROTECT(win),
+		       protect ? ~0 : 0);
 }
 
 static void decon_atomic_begin(struct exynos_drm_crtc *crtc,
@@ -278,14 +277,10 @@ static void decon_update_plane(struct exynos_drm_crtc *crtc,
 	decon_win_set_pixfmt(ctx, win, state->fb);
 
 	/* window enable */
-	val = readl(ctx->addr + DECON_WINCONx(win));
-	val |= WINCONx_ENWIN_F;
-	writel(val, ctx->addr + DECON_WINCONx(win));
+	decon_set_bits(ctx, DECON_WINCONx(win), WINCONx_ENWIN_F, ~0);
 
 	/* standalone update */
-	val = readl(ctx->addr + DECON_UPDATE);
-	val |= STANDALONE_UPDATE_F;
-	writel(val, ctx->addr + DECON_UPDATE);
+	decon_set_bits(ctx, DECON_UPDATE, STANDALONE_UPDATE_F, ~0);
 }
 
 static void decon_disable_plane(struct exynos_drm_crtc *crtc,
@@ -293,7 +288,6 @@ static void decon_disable_plane(struct exynos_drm_crtc *crtc,
 {
 	struct decon_context *ctx = crtc->ctx;
 	unsigned int win = plane->zpos;
-	u32 val;
 
 	if (ctx->suspended)
 		return;
@@ -301,16 +295,12 @@ static void decon_disable_plane(struct exynos_drm_crtc *crtc,
 	decon_shadow_protect_win(ctx, win, true);
 
 	/* window disable */
-	val = readl(ctx->addr + DECON_WINCONx(win));
-	val &= ~WINCONx_ENWIN_F;
-	writel(val, ctx->addr + DECON_WINCONx(win));
+	decon_set_bits(ctx, DECON_WINCONx(win), WINCONx_ENWIN_F, 0);
 
 	decon_shadow_protect_win(ctx, win, false);
 
 	/* standalone update */
-	val = readl(ctx->addr + DECON_UPDATE);
-	val |= STANDALONE_UPDATE_F;
-	writel(val, ctx->addr + DECON_UPDATE);
+	decon_set_bits(ctx, DECON_UPDATE, STANDALONE_UPDATE_F, ~0);
 }
 
 static void decon_atomic_flush(struct exynos_drm_crtc *crtc,
@@ -416,17 +406,12 @@ static void decon_disable(struct exynos_drm_crtc *crtc)
 void decon_te_irq_handler(struct exynos_drm_crtc *crtc)
 {
 	struct decon_context *ctx = crtc->ctx;
-	u32 val;
 
 	if (!test_bit(BIT_CLKS_ENABLED, &ctx->enabled))
 		return;
 
-	if (atomic_add_unless(&ctx->win_updated, -1, 0)) {
-		/* trigger */
-		val = readl(ctx->addr + DECON_TRIGCON);
-		val |= TRIGCON_SWTRIGCMD;
-		writel(val, ctx->addr + DECON_TRIGCON);
-	}
+	if (atomic_add_unless(&ctx->win_updated, -1, 0))
+		decon_set_bits(ctx, DECON_TRIGCON, TRIGCON_SWTRIGCMD, ~0);
 
 	drm_crtc_handle_vblank(&ctx->crtc->base);
 }
@@ -435,7 +420,6 @@ static void decon_clear_channels(struct exynos_drm_crtc *crtc)
 {
 	struct decon_context *ctx = crtc->ctx;
 	int win, i, ret;
-	u32 val;
 
 	DRM_DEBUG_KMS("%s\n", __FILE__);
 
@@ -446,25 +430,10 @@ static void decon_clear_channels(struct exynos_drm_crtc *crtc)
 	}
 
 	for (win = 0; win < WINDOWS_NR; win++) {
-		/* shadow update disable */
-		val = readl(ctx->addr + DECON_SHADOWCON);
-		val |= SHADOWCON_Wx_PROTECT(win);
-		writel(val, ctx->addr + DECON_SHADOWCON);
-
-		/* window disable */
-		val = readl(ctx->addr + DECON_WINCONx(win));
-		val &= ~WINCONx_ENWIN_F;
-		writel(val, ctx->addr + DECON_WINCONx(win));
-
-		/* shadow update enable */
-		val = readl(ctx->addr + DECON_SHADOWCON);
-		val &= ~SHADOWCON_Wx_PROTECT(win);
-		writel(val, ctx->addr + DECON_SHADOWCON);
-
-		/* standalone update */
-		val = readl(ctx->addr + DECON_UPDATE);
-		val |= STANDALONE_UPDATE_F;
-		writel(val, ctx->addr + DECON_UPDATE);
+		decon_shadow_protect_win(ctx, win, true);
+		decon_set_bits(ctx, DECON_WINCONx(win), WINCONx_ENWIN_F, 0);
+		decon_shadow_protect_win(ctx, win, false);
+		decon_set_bits(ctx, DECON_UPDATE, STANDALONE_UPDATE_F, ~0);
 	}
 	/* TODO: wait for possible vsync */
 	msleep(50);
-- 
cgit v0.10.2


From 7b6bb6ed01e9f9c20c86b722c733d7219c4c55f6 Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <a.hajda@samsung.com>
Date: Tue, 20 Oct 2015 11:22:38 +0200
Subject: drm/exynos/decon5433: merge different flag fields

Driver uses four different fields for internal flags. They can be merged
into one.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
index 722c11a..265a77f 100644
--- a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
+++ b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
@@ -37,6 +37,13 @@ static const char * const decon_clks_name[] = {
 	"sclk_decon_eclk",
 };
 
+enum decon_flag_bits {
+	BIT_CLKS_ENABLED,
+	BIT_IRQS_ENABLED,
+	BIT_WIN_UPDATED,
+	BIT_SUSPENDED
+};
+
 struct decon_context {
 	struct device			*dev;
 	struct drm_device		*drm_dev;
@@ -44,15 +51,9 @@ struct decon_context {
 	struct exynos_drm_plane		planes[WINDOWS_NR];
 	void __iomem			*addr;
 	struct clk			*clks[ARRAY_SIZE(decon_clks_name)];
-	unsigned long			irq_flags;
 	int				pipe;
-	bool				suspended;
-
-#define BIT_CLKS_ENABLED		0
-#define BIT_IRQS_ENABLED		1
-	unsigned long			enabled;
+	unsigned long			flags;
 	bool				i80_if;
-	atomic_t			win_updated;
 };
 
 static const uint32_t decon_formats[] = {
@@ -74,10 +75,10 @@ static int decon_enable_vblank(struct exynos_drm_crtc *crtc)
 	struct decon_context *ctx = crtc->ctx;
 	u32 val;
 
-	if (ctx->suspended)
+	if (test_bit(BIT_SUSPENDED, &ctx->flags))
 		return -EPERM;
 
-	if (test_and_set_bit(0, &ctx->irq_flags)) {
+	if (test_and_set_bit(BIT_IRQS_ENABLED, &ctx->flags)) {
 		val = VIDINTCON0_INTEN;
 		if (ctx->i80_if)
 			val |= VIDINTCON0_FRAMEDONE;
@@ -94,10 +95,10 @@ static void decon_disable_vblank(struct exynos_drm_crtc *crtc)
 {
 	struct decon_context *ctx = crtc->ctx;
 
-	if (ctx->suspended)
+	if (test_bit(BIT_SUSPENDED, &ctx->flags))
 		return;
 
-	if (test_and_clear_bit(0, &ctx->irq_flags))
+	if (test_and_clear_bit(BIT_IRQS_ENABLED, &ctx->flags))
 		writel(0, ctx->addr + DECON_VIDINTCON0);
 }
 
@@ -114,7 +115,7 @@ static void decon_commit(struct exynos_drm_crtc *crtc)
 	struct drm_display_mode *m = &crtc->base.mode;
 	u32 val;
 
-	if (ctx->suspended)
+	if (test_bit(BIT_SUSPENDED, &ctx->flags))
 		return;
 
 	/* enable clock gate */
@@ -231,7 +232,7 @@ static void decon_atomic_begin(struct exynos_drm_crtc *crtc,
 {
 	struct decon_context *ctx = crtc->ctx;
 
-	if (ctx->suspended)
+	if (test_bit(BIT_SUSPENDED, &ctx->flags))
 		return;
 
 	decon_shadow_protect_win(ctx, plane->zpos, true);
@@ -247,7 +248,7 @@ static void decon_update_plane(struct exynos_drm_crtc *crtc,
 	unsigned int pitch = state->fb->pitches[0];
 	u32 val;
 
-	if (ctx->suspended)
+	if (test_bit(BIT_SUSPENDED, &ctx->flags))
 		return;
 
 	val = COORDINATE_X(plane->crtc_x) | COORDINATE_Y(plane->crtc_y);
@@ -289,7 +290,7 @@ static void decon_disable_plane(struct exynos_drm_crtc *crtc,
 	struct decon_context *ctx = crtc->ctx;
 	unsigned int win = plane->zpos;
 
-	if (ctx->suspended)
+	if (test_bit(BIT_SUSPENDED, &ctx->flags))
 		return;
 
 	decon_shadow_protect_win(ctx, win, true);
@@ -308,13 +309,13 @@ static void decon_atomic_flush(struct exynos_drm_crtc *crtc,
 {
 	struct decon_context *ctx = crtc->ctx;
 
-	if (ctx->suspended)
+	if (test_bit(BIT_SUSPENDED, &ctx->flags))
 		return;
 
 	decon_shadow_protect_win(ctx, plane->zpos, false);
 
 	if (ctx->i80_if)
-		atomic_set(&ctx->win_updated, 1);
+		set_bit(BIT_WIN_UPDATED, &ctx->flags);
 }
 
 static void decon_swreset(struct decon_context *ctx)
@@ -346,11 +347,9 @@ static void decon_enable(struct exynos_drm_crtc *crtc)
 	int ret;
 	int i;
 
-	if (!ctx->suspended)
+	if (!test_and_clear_bit(BIT_SUSPENDED, &ctx->flags))
 		return;
 
-	ctx->suspended = false;
-
 	pm_runtime_get_sync(ctx->dev);
 
 	for (i = 0; i < ARRAY_SIZE(decon_clks_name); i++) {
@@ -359,10 +358,10 @@ static void decon_enable(struct exynos_drm_crtc *crtc)
 			goto err;
 	}
 
-	set_bit(BIT_CLKS_ENABLED, &ctx->enabled);
+	set_bit(BIT_CLKS_ENABLED, &ctx->flags);
 
 	/* if vblank was enabled status, enable it again. */
-	if (test_and_clear_bit(0, &ctx->irq_flags))
+	if (test_and_clear_bit(BIT_IRQS_ENABLED, &ctx->flags))
 		decon_enable_vblank(ctx->crtc);
 
 	decon_commit(ctx->crtc);
@@ -372,7 +371,7 @@ err:
 	while (--i >= 0)
 		clk_disable_unprepare(ctx->clks[i]);
 
-	ctx->suspended = true;
+	set_bit(BIT_SUSPENDED, &ctx->flags);
 }
 
 static void decon_disable(struct exynos_drm_crtc *crtc)
@@ -380,7 +379,7 @@ static void decon_disable(struct exynos_drm_crtc *crtc)
 	struct decon_context *ctx = crtc->ctx;
 	int i;
 
-	if (ctx->suspended)
+	if (test_bit(BIT_SUSPENDED, &ctx->flags))
 		return;
 
 	/*
@@ -396,21 +395,21 @@ static void decon_disable(struct exynos_drm_crtc *crtc)
 	for (i = 0; i < ARRAY_SIZE(decon_clks_name); i++)
 		clk_disable_unprepare(ctx->clks[i]);
 
-	clear_bit(BIT_CLKS_ENABLED, &ctx->enabled);
+	clear_bit(BIT_CLKS_ENABLED, &ctx->flags);
 
 	pm_runtime_put_sync(ctx->dev);
 
-	ctx->suspended = true;
+	set_bit(BIT_SUSPENDED, &ctx->flags);
 }
 
 void decon_te_irq_handler(struct exynos_drm_crtc *crtc)
 {
 	struct decon_context *ctx = crtc->ctx;
 
-	if (!test_bit(BIT_CLKS_ENABLED, &ctx->enabled))
+	if (!test_bit(BIT_CLKS_ENABLED, &ctx->flags))
 		return;
 
-	if (atomic_add_unless(&ctx->win_updated, -1, 0))
+	if (test_and_clear_bit(BIT_WIN_UPDATED, &ctx->flags))
 		decon_set_bits(ctx, DECON_TRIGCON, TRIGCON_SWTRIGCMD, ~0);
 
 	drm_crtc_handle_vblank(&ctx->crtc->base);
@@ -520,7 +519,7 @@ static irqreturn_t decon_vsync_irq_handler(int irq, void *dev_id)
 	struct decon_context *ctx = dev_id;
 	u32 val;
 
-	if (!test_bit(BIT_CLKS_ENABLED, &ctx->enabled))
+	if (!test_bit(BIT_CLKS_ENABLED, &ctx->flags))
 		goto out;
 
 	val = readl(ctx->addr + DECON_VIDINTCON1);
@@ -541,7 +540,7 @@ static irqreturn_t decon_lcd_sys_irq_handler(int irq, void *dev_id)
 	u32 val;
 	int win;
 
-	if (!test_bit(BIT_CLKS_ENABLED, &ctx->enabled))
+	if (!test_bit(BIT_CLKS_ENABLED, &ctx->flags))
 		goto out;
 
 	val = readl(ctx->addr + DECON_VIDINTCON1);
@@ -576,7 +575,7 @@ static int exynos5433_decon_probe(struct platform_device *pdev)
 	if (!ctx)
 		return -ENOMEM;
 
-	ctx->suspended = true;
+	__set_bit(BIT_SUSPENDED, &ctx->flags);
 	ctx->dev = dev;
 	if (of_get_child_by_name(dev->of_node, "i80-if-timings"))
 		ctx->i80_if = true;
-- 
cgit v0.10.2


From 5d929ba50a1ea69fba1a92e24fc54d484583a109 Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <a.hajda@samsung.com>
Date: Tue, 20 Oct 2015 11:22:39 +0200
Subject: drm/exynos/decon5433: remove duplicated initialization

Field .commit is already initialized few lines above.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
index 265a77f..3c9aa4e 100644
--- a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
+++ b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
@@ -448,7 +448,6 @@ static struct exynos_drm_crtc_ops decon_crtc_ops = {
 	.commit			= decon_commit,
 	.enable_vblank		= decon_enable_vblank,
 	.disable_vblank		= decon_disable_vblank,
-	.commit			= decon_commit,
 	.atomic_begin		= decon_atomic_begin,
 	.update_plane		= decon_update_plane,
 	.disable_plane		= decon_disable_plane,
-- 
cgit v0.10.2


From b8182832c5a9d9ce645d53be84e5db07f8aa5302 Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <a.hajda@samsung.com>
Date: Tue, 20 Oct 2015 18:22:41 +0900
Subject: drm/exynos/decon5433: add support for DECON-TV

DECON-TV IP is responsible for generating video stream which is transferred
to HDMI IP. It is almost fully compatible with DECON IP.

The patch is based on initial work of Hyungwon Hwang.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
index 3c9aa4e..fbe1b31 100644
--- a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
+++ b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
@@ -13,6 +13,7 @@
 #include <linux/platform_device.h>
 #include <linux/clk.h>
 #include <linux/component.h>
+#include <linux/of_device.h>
 #include <linux/of_gpio.h>
 #include <linux/pm_runtime.h>
 
@@ -37,6 +38,12 @@ static const char * const decon_clks_name[] = {
 	"sclk_decon_eclk",
 };
 
+enum decon_iftype {
+	IFTYPE_RGB,
+	IFTYPE_I80,
+	IFTYPE_HDMI
+};
+
 enum decon_flag_bits {
 	BIT_CLKS_ENABLED,
 	BIT_IRQS_ENABLED,
@@ -53,7 +60,8 @@ struct decon_context {
 	struct clk			*clks[ARRAY_SIZE(decon_clks_name)];
 	int				pipe;
 	unsigned long			flags;
-	bool				i80_if;
+	enum decon_iftype		out_type;
+	int				first_win;
 };
 
 static const uint32_t decon_formats[] = {
@@ -80,7 +88,7 @@ static int decon_enable_vblank(struct exynos_drm_crtc *crtc)
 
 	if (test_and_set_bit(BIT_IRQS_ENABLED, &ctx->flags)) {
 		val = VIDINTCON0_INTEN;
-		if (ctx->i80_if)
+		if (ctx->out_type == IFTYPE_I80)
 			val |= VIDINTCON0_FRAMEDONE;
 		else
 			val |= VIDINTCON0_INTFRMEN;
@@ -104,8 +112,11 @@ static void decon_disable_vblank(struct exynos_drm_crtc *crtc)
 
 static void decon_setup_trigger(struct decon_context *ctx)
 {
-	u32 val = TRIGCON_TRIGEN_PER_F | TRIGCON_TRIGEN_F |
-			TRIGCON_TE_AUTO_MASK | TRIGCON_SWTRIGEN;
+	u32 val = (ctx->out_type != IFTYPE_HDMI)
+		? TRIGCON_TRIGEN_PER_F | TRIGCON_TRIGEN_F |
+		  TRIGCON_TE_AUTO_MASK | TRIGCON_SWTRIGEN
+		: TRIGCON_TRIGEN_PER_F | TRIGCON_TRIGEN_F |
+		  TRIGCON_HWTRIGMASK_I80_RGB | TRIGCON_HWTRIGEN_I80_RGB;
 	writel(val, ctx->addr + DECON_TRIGCON);
 }
 
@@ -118,13 +129,22 @@ static void decon_commit(struct exynos_drm_crtc *crtc)
 	if (test_bit(BIT_SUSPENDED, &ctx->flags))
 		return;
 
+	if (ctx->out_type == IFTYPE_HDMI) {
+		m->crtc_hsync_start = m->crtc_hdisplay + 10;
+		m->crtc_hsync_end = m->crtc_htotal - 92;
+		m->crtc_vsync_start = m->crtc_vdisplay + 1;
+		m->crtc_vsync_end = m->crtc_vsync_start + 1;
+	}
+
+	decon_set_bits(ctx, DECON_VIDCON0, VIDCON0_ENVID, 0);
+
 	/* enable clock gate */
 	val = CMU_CLKGAGE_MODE_SFR_F | CMU_CLKGAGE_MODE_MEM_F;
 	writel(val, ctx->addr + DECON_CMU);
 
 	/* lcd on and use command if */
 	val = VIDOUT_LCD_ON;
-	if (ctx->i80_if)
+	if (ctx->out_type == IFTYPE_I80)
 		val |= VIDOUT_COMMAND_IF;
 	else
 		val |= VIDOUT_RGB_IF;
@@ -134,7 +154,7 @@ static void decon_commit(struct exynos_drm_crtc *crtc)
 		VIDTCON2_HOZVAL(m->hdisplay - 1);
 	writel(val, ctx->addr + DECON_VIDTCON2);
 
-	if (!ctx->i80_if) {
+	if (ctx->out_type != IFTYPE_I80) {
 		val = VIDTCON00_VBPD_F(
 				m->crtc_vtotal - m->crtc_vsync_end - 1) |
 			VIDTCON00_VFPD_F(
@@ -159,15 +179,9 @@ static void decon_commit(struct exynos_drm_crtc *crtc)
 	decon_setup_trigger(ctx);
 
 	/* enable output and display signal */
-	val = VIDCON0_ENVID | VIDCON0_ENVID_F;
-	writel(val, ctx->addr + DECON_VIDCON0);
+	decon_set_bits(ctx, DECON_VIDCON0, VIDCON0_ENVID | VIDCON0_ENVID_F, ~0);
 }
 
-#define COORDINATE_X(x)		(((x) & 0xfff) << 12)
-#define COORDINATE_Y(x)		((x) & 0xfff)
-#define OFFSIZE(x)		(((x) & 0x3fff) << 14)
-#define PAGEWIDTH(x)		((x) & 0x3fff)
-
 static void decon_win_set_pixfmt(struct decon_context *ctx, unsigned int win,
 				 struct drm_framebuffer *fb)
 {
@@ -238,6 +252,10 @@ static void decon_atomic_begin(struct exynos_drm_crtc *crtc,
 	decon_shadow_protect_win(ctx, plane->zpos, true);
 }
 
+#define BIT_VAL(x, e, s) (((x) & ((1 << ((e) - (s) + 1)) - 1)) << (s))
+#define COORDINATE_X(x) BIT_VAL((x), 23, 12)
+#define COORDINATE_Y(x) BIT_VAL((x), 11, 0)
+
 static void decon_update_plane(struct exynos_drm_crtc *crtc,
 			       struct exynos_drm_plane *plane)
 {
@@ -271,8 +289,12 @@ static void decon_update_plane(struct exynos_drm_crtc *crtc,
 	val = plane->dma_addr[0] + pitch * plane->crtc_h;
 	writel(val, ctx->addr + DECON_VIDW0xADD1B0(win));
 
-	val = OFFSIZE(pitch - plane->crtc_w * bpp)
-		| PAGEWIDTH(plane->crtc_w * bpp);
+	if (ctx->out_type != IFTYPE_HDMI)
+		val = BIT_VAL(pitch - plane->crtc_w * bpp, 27, 14)
+			| BIT_VAL(plane->crtc_w * bpp, 13, 0);
+	else
+		val = BIT_VAL(pitch - plane->crtc_w * bpp, 29, 15)
+			| BIT_VAL(plane->crtc_w * bpp, 14, 0);
 	writel(val, ctx->addr + DECON_VIDW0xADD2(win));
 
 	decon_win_set_pixfmt(ctx, win, state->fb);
@@ -314,7 +336,7 @@ static void decon_atomic_flush(struct exynos_drm_crtc *crtc,
 
 	decon_shadow_protect_win(ctx, plane->zpos, false);
 
-	if (ctx->i80_if)
+	if (ctx->out_type == IFTYPE_I80)
 		set_bit(BIT_WIN_UPDATED, &ctx->flags);
 }
 
@@ -339,6 +361,17 @@ static void decon_swreset(struct decon_context *ctx)
 	}
 
 	WARN(tries == 0, "failed to software reset DECON\n");
+
+	if (ctx->out_type != IFTYPE_HDMI)
+		return;
+
+	writel(VIDCON0_CLKVALUP | VIDCON0_VLCKFREE, ctx->addr + DECON_VIDCON0);
+	decon_set_bits(ctx, DECON_CMU,
+		       CMU_CLKGAGE_MODE_SFR_F | CMU_CLKGAGE_MODE_MEM_F, ~0);
+	writel(VIDCON1_VCLK_RUN_VDEN_DISABLE, ctx->addr + DECON_VIDCON1);
+	writel(CRCCTRL_CRCEN | CRCCTRL_CRCSTART_F | CRCCTRL_CRCCLKEN,
+	       ctx->addr + DECON_CRCCTRL);
+	decon_setup_trigger(ctx);
 }
 
 static void decon_enable(struct exynos_drm_crtc *crtc)
@@ -387,7 +420,7 @@ static void decon_disable(struct exynos_drm_crtc *crtc)
 	 * suspend that connector. Otherwise we might try to scan from
 	 * a destroyed buffer later.
 	 */
-	for (i = 0; i < WINDOWS_NR; i++)
+	for (i = ctx->first_win; i < WINDOWS_NR; i++)
 		decon_disable_plane(crtc, &ctx->planes[i]);
 
 	decon_swreset(ctx);
@@ -461,25 +494,30 @@ static int decon_bind(struct device *dev, struct device *master, void *data)
 	struct drm_device *drm_dev = data;
 	struct exynos_drm_private *priv = drm_dev->dev_private;
 	struct exynos_drm_plane *exynos_plane;
+	enum exynos_drm_output_type out_type;
 	enum drm_plane_type type;
-	unsigned int zpos;
+	unsigned int win;
 	int ret;
 
 	ctx->drm_dev = drm_dev;
 	ctx->pipe = priv->pipe++;
 
-	for (zpos = 0; zpos < WINDOWS_NR; zpos++) {
-		type = exynos_plane_get_type(zpos, CURSOR_WIN);
-		ret = exynos_plane_init(drm_dev, &ctx->planes[zpos],
+	for (win = ctx->first_win; win < WINDOWS_NR; win++) {
+		int tmp = (win == ctx->first_win) ? 0 : win;
+
+		type = exynos_plane_get_type(tmp, CURSOR_WIN);
+		ret = exynos_plane_init(drm_dev, &ctx->planes[win],
 				1 << ctx->pipe, type, decon_formats,
-				ARRAY_SIZE(decon_formats), zpos);
+				ARRAY_SIZE(decon_formats), win);
 		if (ret)
 			return ret;
 	}
 
-	exynos_plane = &ctx->planes[DEFAULT_WIN];
+	exynos_plane = &ctx->planes[ctx->first_win];
+	out_type = (ctx->out_type == IFTYPE_HDMI) ? EXYNOS_DISPLAY_TYPE_HDMI
+						  : EXYNOS_DISPLAY_TYPE_LCD;
 	ctx->crtc = exynos_drm_crtc_create(drm_dev, &exynos_plane->base,
-					ctx->pipe, EXYNOS_DISPLAY_TYPE_LCD,
+					ctx->pipe, out_type,
 					&decon_crtc_ops, ctx);
 	if (IS_ERR(ctx->crtc)) {
 		ret = PTR_ERR(ctx->crtc);
@@ -513,27 +551,7 @@ static const struct component_ops decon_component_ops = {
 	.unbind = decon_unbind,
 };
 
-static irqreturn_t decon_vsync_irq_handler(int irq, void *dev_id)
-{
-	struct decon_context *ctx = dev_id;
-	u32 val;
-
-	if (!test_bit(BIT_CLKS_ENABLED, &ctx->flags))
-		goto out;
-
-	val = readl(ctx->addr + DECON_VIDINTCON1);
-	if (val & VIDINTCON1_INTFRMPEND) {
-		drm_crtc_handle_vblank(&ctx->crtc->base);
-
-		/* clear */
-		writel(VIDINTCON1_INTFRMPEND, ctx->addr + DECON_VIDINTCON1);
-	}
-
-out:
-	return IRQ_HANDLED;
-}
-
-static irqreturn_t decon_lcd_sys_irq_handler(int irq, void *dev_id)
+static irqreturn_t decon_irq_handler(int irq, void *dev_id)
 {
 	struct decon_context *ctx = dev_id;
 	u32 val;
@@ -543,8 +561,10 @@ static irqreturn_t decon_lcd_sys_irq_handler(int irq, void *dev_id)
 		goto out;
 
 	val = readl(ctx->addr + DECON_VIDINTCON1);
-	if (val & VIDINTCON1_INTFRMDONEPEND) {
-		for (win = 0 ; win < WINDOWS_NR ; win++) {
+	val &= VIDINTCON1_INTFRMDONEPEND | VIDINTCON1_INTFRMPEND;
+
+	if (val) {
+		for (win = ctx->first_win; win < WINDOWS_NR ; win++) {
 			struct exynos_drm_plane *plane = &ctx->planes[win];
 
 			if (!plane->pending_fb)
@@ -554,16 +574,29 @@ static irqreturn_t decon_lcd_sys_irq_handler(int irq, void *dev_id)
 		}
 
 		/* clear */
-		writel(VIDINTCON1_INTFRMDONEPEND,
-				ctx->addr + DECON_VIDINTCON1);
+		writel(val, ctx->addr + DECON_VIDINTCON1);
 	}
 
 out:
 	return IRQ_HANDLED;
 }
 
+static const struct of_device_id exynos5433_decon_driver_dt_match[] = {
+	{
+		.compatible = "samsung,exynos5433-decon",
+		.data = (void *)IFTYPE_RGB
+	},
+	{
+		.compatible = "samsung,exynos5433-decon-tv",
+		.data = (void *)IFTYPE_HDMI
+	},
+	{},
+};
+MODULE_DEVICE_TABLE(of, exynos5433_decon_driver_dt_match);
+
 static int exynos5433_decon_probe(struct platform_device *pdev)
 {
+	const struct of_device_id *of_id;
 	struct device *dev = &pdev->dev;
 	struct decon_context *ctx;
 	struct resource *res;
@@ -576,8 +609,14 @@ static int exynos5433_decon_probe(struct platform_device *pdev)
 
 	__set_bit(BIT_SUSPENDED, &ctx->flags);
 	ctx->dev = dev;
-	if (of_get_child_by_name(dev->of_node, "i80-if-timings"))
-		ctx->i80_if = true;
+
+	of_id = of_match_device(exynos5433_decon_driver_dt_match, &pdev->dev);
+	ctx->out_type = (enum decon_iftype)of_id->data;
+
+	if (ctx->out_type == IFTYPE_HDMI)
+		ctx->first_win = 1;
+	else if (of_get_child_by_name(dev->of_node, "i80-if-timings"))
+		ctx->out_type = IFTYPE_I80;
 
 	for (i = 0; i < ARRAY_SIZE(decon_clks_name); i++) {
 		struct clk *clk;
@@ -602,15 +641,14 @@ static int exynos5433_decon_probe(struct platform_device *pdev)
 	}
 
 	res = platform_get_resource_byname(pdev, IORESOURCE_IRQ,
-			ctx->i80_if ? "lcd_sys" : "vsync");
+			(ctx->out_type == IFTYPE_I80) ? "lcd_sys" : "vsync");
 	if (!res) {
 		dev_err(dev, "cannot find IRQ resource\n");
 		return -ENXIO;
 	}
 
-	ret = devm_request_irq(dev, res->start, ctx->i80_if ?
-			decon_lcd_sys_irq_handler : decon_vsync_irq_handler, 0,
-			"drm_decon", ctx);
+	ret = devm_request_irq(dev, res->start, decon_irq_handler, 0,
+			       "drm_decon", ctx);
 	if (ret < 0) {
 		dev_err(dev, "lcd_sys irq request failed\n");
 		return ret;
@@ -641,12 +679,6 @@ static int exynos5433_decon_remove(struct platform_device *pdev)
 	return 0;
 }
 
-static const struct of_device_id exynos5433_decon_driver_dt_match[] = {
-	{ .compatible = "samsung,exynos5433-decon" },
-	{},
-};
-MODULE_DEVICE_TABLE(of, exynos5433_decon_driver_dt_match);
-
 struct platform_driver exynos5433_decon_driver = {
 	.probe		= exynos5433_decon_probe,
 	.remove		= exynos5433_decon_remove,
diff --git a/include/video/exynos5433_decon.h b/include/video/exynos5433_decon.h
index 3696575..c1c1ca1 100644
--- a/include/video/exynos5433_decon.h
+++ b/include/video/exynos5433_decon.h
@@ -82,6 +82,8 @@
 
 /* VIDCON0 */
 #define VIDCON0_SWRESET			(1 << 28)
+#define VIDCON0_CLKVALUP		(1 << 14)
+#define VIDCON0_VLCKFREE		(1 << 5)
 #define VIDCON0_STOP_STATUS		(1 << 2)
 #define VIDCON0_ENVID			(1 << 1)
 #define VIDCON0_ENVID_F			(1 << 0)
@@ -137,6 +139,13 @@
 /* DECON_UPDATE */
 #define STANDALONE_UPDATE_F		(1 << 0)
 
+/* DECON_VIDCON1 */
+#define VIDCON1_VCLK_MASK		(0x3 << 9)
+#define VIDCON1_VCLK_RUN_VDEN_DISABLE	(0x3 << 9)
+#define VIDCON1_VCLK_HOLD		(0x0 << 9)
+#define VIDCON1_VCLK_RUN		(0x1 << 9)
+
+
 /* DECON_VIDTCON00 */
 #define VIDTCON00_VBPD_F(x)		(((x) & 0xfff) << 16)
 #define VIDTCON00_VFPD_F(x)		((x) & 0xfff)
@@ -159,7 +168,27 @@
 #define TRIGCON_TRIGEN_PER_F		(1 << 31)
 #define TRIGCON_TRIGEN_F		(1 << 30)
 #define TRIGCON_TE_AUTO_MASK		(1 << 29)
+#define TRIGCON_WB_SWTRIGCMD		(1 << 28)
+#define TRIGCON_SWTRIGCMD_W4BUF		(1 << 26)
+#define TRIGCON_TRIGMODE_W4BUF		(1 << 25)
+#define TRIGCON_SWTRIGCMD_W3BUF		(1 << 21)
+#define TRIGCON_TRIGMODE_W3BUF		(1 << 20)
+#define TRIGCON_SWTRIGCMD_W2BUF		(1 << 16)
+#define TRIGCON_TRIGMODE_W2BUF		(1 << 15)
+#define TRIGCON_SWTRIGCMD_W1BUF		(1 << 11)
+#define TRIGCON_TRIGMODE_W1BUF		(1 << 10)
+#define TRIGCON_SWTRIGCMD_W0BUF		(1 << 6)
+#define TRIGCON_TRIGMODE_W0BUF		(1 << 5)
+#define TRIGCON_HWTRIGMASK_I80_RGB	(1 << 4)
+#define TRIGCON_HWTRIGEN_I80_RGB	(1 << 3)
+#define TRIGCON_HWTRIG_INV_I80_RGB	(1 << 2)
 #define TRIGCON_SWTRIGCMD		(1 << 1)
 #define TRIGCON_SWTRIGEN		(1 << 0)
 
+/* DECON_CRCCTRL */
+#define CRCCTRL_CRCCLKEN		(0x1 << 2)
+#define CRCCTRL_CRCSTART_F		(0x1 << 1)
+#define CRCCTRL_CRCEN			(0x1 << 0)
+#define CRCCTRL_MASK			(0x7)
+
 #endif /* EXYNOS_REGS_DECON_H */
-- 
cgit v0.10.2


From 5625b3418a5bf977e3561bb12229d4c3e96b44d8 Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <a.hajda@samsung.com>
Date: Mon, 26 Oct 2015 13:03:39 +0100
Subject: drm/exynos: add atomic_check callback to exynos_crtc

Some CRTCs needs mode validation, this patch adds neccessary
callback to Exynos DRM framework. It is called from DRM core
via atomic_check helper for drm_crtc.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/exynos_drm_crtc.c b/drivers/gpu/drm/exynos/exynos_drm_crtc.c
index 50dec0d..b3ba27f 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_crtc.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_crtc.c
@@ -50,6 +50,17 @@ exynos_drm_crtc_mode_set_nofb(struct drm_crtc *crtc)
 		exynos_crtc->ops->commit(exynos_crtc);
 }
 
+static int exynos_crtc_atomic_check(struct drm_crtc *crtc,
+				     struct drm_crtc_state *state)
+{
+	struct exynos_drm_crtc *exynos_crtc = to_exynos_crtc(crtc);
+
+	if (exynos_crtc->ops->atomic_check)
+		return exynos_crtc->ops->atomic_check(exynos_crtc, state);
+
+	return 0;
+}
+
 static void exynos_crtc_atomic_begin(struct drm_crtc *crtc,
 				     struct drm_crtc_state *old_crtc_state)
 {
@@ -86,6 +97,7 @@ static struct drm_crtc_helper_funcs exynos_crtc_helper_funcs = {
 	.enable		= exynos_drm_crtc_enable,
 	.disable	= exynos_drm_crtc_disable,
 	.mode_set_nofb	= exynos_drm_crtc_mode_set_nofb,
+	.atomic_check	= exynos_crtc_atomic_check,
 	.atomic_begin	= exynos_crtc_atomic_begin,
 	.atomic_flush	= exynos_crtc_atomic_flush,
 };
diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.h b/drivers/gpu/drm/exynos/exynos_drm_drv.h
index 638fc43..f1eda7f 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_drv.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_drv.h
@@ -89,6 +89,7 @@ struct exynos_drm_plane {
  * @disable_vblank: specific driver callback for disabling vblank interrupt.
  * @wait_for_vblank: wait for vblank interrupt to make sure that
  *	hardware overlay is updated.
+ * @atomic_check: validate state
  * @atomic_begin: prepare a window to receive a update
  * @atomic_flush: mark the end of a window update
  * @update_plane: apply hardware specific overlay data to registers.
@@ -108,6 +109,8 @@ struct exynos_drm_crtc_ops {
 	int (*enable_vblank)(struct exynos_drm_crtc *crtc);
 	void (*disable_vblank)(struct exynos_drm_crtc *crtc);
 	void (*wait_for_vblank)(struct exynos_drm_crtc *crtc);
+	int (*atomic_check)(struct exynos_drm_crtc *crtc,
+			    struct drm_crtc_state *state);
 	void (*atomic_begin)(struct exynos_drm_crtc *crtc,
 			      struct exynos_drm_plane *plane);
 	void (*update_plane)(struct exynos_drm_crtc *crtc,
-- 
cgit v0.10.2


From 3ae24362e02df0aa27ed1dc9d85b720514c99d20 Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <a.hajda@samsung.com>
Date: Mon, 26 Oct 2015 13:03:40 +0100
Subject: drm/exynos/mixer: replace direct cross-driver call with drm mode
 validation

HDMI driver called directly function from MIXER driver to invalidate modes
not supported by MIXER. The patch replaces the hack with proper .atomic_check
callback.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c
index b0f5ff4..57b6755 100644
--- a/drivers/gpu/drm/exynos/exynos_hdmi.c
+++ b/drivers/gpu/drm/exynos/exynos_hdmi.c
@@ -44,7 +44,6 @@
 
 #include "exynos_drm_drv.h"
 #include "exynos_drm_crtc.h"
-#include "exynos_mixer.h"
 
 #define HOTPLUG_DEBOUNCE_MS		1100
 
@@ -1017,10 +1016,6 @@ static int hdmi_mode_valid(struct drm_connector *connector,
 		(mode->flags & DRM_MODE_FLAG_INTERLACE) ? true :
 		false, mode->clock * 1000);
 
-	ret = mixer_check_mode(mode);
-	if (ret)
-		return MODE_BAD;
-
 	ret = hdmi_find_phy_conf(hdata, mode->clock * 1000);
 	if (ret < 0)
 		return MODE_BAD;
diff --git a/drivers/gpu/drm/exynos/exynos_mixer.c b/drivers/gpu/drm/exynos/exynos_mixer.c
index 3f9f072..d09f8f9 100644
--- a/drivers/gpu/drm/exynos/exynos_mixer.c
+++ b/drivers/gpu/drm/exynos/exynos_mixer.c
@@ -39,7 +39,6 @@
 #include "exynos_drm_crtc.h"
 #include "exynos_drm_plane.h"
 #include "exynos_drm_iommu.h"
-#include "exynos_mixer.h"
 
 #define MIXER_WIN_NR		3
 #define VP_DEFAULT_WIN		2
@@ -1096,8 +1095,10 @@ static void mixer_disable(struct exynos_drm_crtc *crtc)
 }
 
 /* Only valid for Mixer version 16.0.33.0 */
-int mixer_check_mode(struct drm_display_mode *mode)
+static int mixer_atomic_check(struct exynos_drm_crtc *crtc,
+		       struct drm_crtc_state *state)
 {
+	struct drm_display_mode *mode = &state->adjusted_mode;
 	u32 w, h;
 
 	w = mode->hdisplay;
@@ -1123,6 +1124,7 @@ static const struct exynos_drm_crtc_ops mixer_crtc_ops = {
 	.wait_for_vblank	= mixer_wait_for_vblank,
 	.update_plane		= mixer_update_plane,
 	.disable_plane		= mixer_disable_plane,
+	.atomic_check		= mixer_atomic_check,
 };
 
 static struct mixer_drv_data exynos5420_mxr_drv_data = {
diff --git a/drivers/gpu/drm/exynos/exynos_mixer.h b/drivers/gpu/drm/exynos/exynos_mixer.h
deleted file mode 100644
index 3811e41..0000000
--- a/drivers/gpu/drm/exynos/exynos_mixer.h
+++ /dev/null
@@ -1,20 +0,0 @@
-/*
- * Copyright (C) 2013 Google, Inc.
- *
- * This software is licensed under the terms of the GNU General Public
- * License version 2, as published by the Free Software Foundation, and
- * may be copied, distributed, and modified under those terms.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- */
-
-#ifndef _EXYNOS_MIXER_H_
-#define _EXYNOS_MIXER_H_
-
-/* This function returns 0 if the given timing is valid for the mixer */
-int mixer_check_mode(struct drm_display_mode *mode);
-
-#endif
-- 
cgit v0.10.2


From 3cb02b4a9e3e4f71ca7fefaff96ae47dd42a0adc Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <a.hajda@samsung.com>
Date: Mon, 26 Oct 2015 13:03:42 +0100
Subject: drm/exynos: separate Mixer and HDMI drivers

Latest Exynos SoCs does not have Mixer IP, but they still have HDMI IP.
Their drivers should be configurable separately.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/Kconfig b/drivers/gpu/drm/exynos/Kconfig
index bd1a415..1d7d9b8 100644
--- a/drivers/gpu/drm/exynos/Kconfig
+++ b/drivers/gpu/drm/exynos/Kconfig
@@ -62,9 +62,15 @@ config DRM_EXYNOS_DP
 	help
 	  This enables support for DP device.
 
+config DRM_EXYNOS_MIXER
+	bool "Exynos DRM Mixer"
+	depends on DRM_EXYNOS && !VIDEO_SAMSUNG_S5P_TV
+	help
+	  Choose this option if you want to use Exynos Mixer for DRM.
+
 config DRM_EXYNOS_HDMI
 	bool "Exynos DRM HDMI"
-	depends on DRM_EXYNOS && !VIDEO_SAMSUNG_S5P_TV
+	depends on DRM_EXYNOS && !VIDEO_SAMSUNG_S5P_TV && (DRM_EXYNOS_MIXER || DRM_EXYNOS5433_DECON)
 	help
 	  Choose this option if you want to use Exynos HDMI for DRM.
 
diff --git a/drivers/gpu/drm/exynos/Makefile b/drivers/gpu/drm/exynos/Makefile
index 02aecfe..6496532 100644
--- a/drivers/gpu/drm/exynos/Makefile
+++ b/drivers/gpu/drm/exynos/Makefile
@@ -14,7 +14,8 @@ exynosdrm-$(CONFIG_DRM_EXYNOS7_DECON)	+= exynos7_drm_decon.o
 exynosdrm-$(CONFIG_DRM_EXYNOS_DPI)	+= exynos_drm_dpi.o
 exynosdrm-$(CONFIG_DRM_EXYNOS_DSI)	+= exynos_drm_dsi.o
 exynosdrm-$(CONFIG_DRM_EXYNOS_DP)	+= exynos_dp_core.o exynos_dp_reg.o
-exynosdrm-$(CONFIG_DRM_EXYNOS_HDMI)	+= exynos_hdmi.o exynos_mixer.o
+exynosdrm-$(CONFIG_DRM_EXYNOS_MIXER)	+= exynos_mixer.o
+exynosdrm-$(CONFIG_DRM_EXYNOS_HDMI)	+= exynos_hdmi.o
 exynosdrm-$(CONFIG_DRM_EXYNOS_VIDI)	+= exynos_drm_vidi.o
 exynosdrm-$(CONFIG_DRM_EXYNOS_G2D)	+= exynos_drm_g2d.o
 exynosdrm-$(CONFIG_DRM_EXYNOS_IPP)	+= exynos_drm_ipp.o
diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.c b/drivers/gpu/drm/exynos/exynos_drm_drv.c
index 09c4c6a..2c6019d 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_drv.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_drv.c
@@ -529,8 +529,10 @@ static struct platform_driver *const exynos_drm_kms_drivers[] = {
 #ifdef CONFIG_DRM_EXYNOS_DSI
 	&dsi_driver,
 #endif
-#ifdef CONFIG_DRM_EXYNOS_HDMI
+#ifdef CONFIG_DRM_EXYNOS_MIXER
 	&mixer_driver,
+#endif
+#ifdef CONFIG_DRM_EXYNOS_HDMI
 	&hdmi_driver,
 #endif
 #ifdef CONFIG_DRM_EXYNOS_VIDI
-- 
cgit v0.10.2


From dba6c5280d62cef3df9c67cafaddf1c7df0d5889 Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <a.hajda@samsung.com>
Date: Mon, 26 Oct 2015 13:03:43 +0100
Subject: drm/exynos: abstract out common dependency

All options depends on DRM_EXYNOS so it can be moved to enclosing if clause.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/Kconfig b/drivers/gpu/drm/exynos/Kconfig
index 1d7d9b8..b0107f0 100644
--- a/drivers/gpu/drm/exynos/Kconfig
+++ b/drivers/gpu/drm/exynos/Kconfig
@@ -11,14 +11,16 @@ config DRM_EXYNOS
 	  Choose this option if you have a Samsung SoC EXYNOS chipset.
 	  If M is selected the module will be called exynosdrm.
 
+if DRM_EXYNOS
+
 config DRM_EXYNOS_IOMMU
 	bool
-	depends on DRM_EXYNOS && EXYNOS_IOMMU && ARM_DMA_USE_IOMMU
+	depends on EXYNOS_IOMMU && ARM_DMA_USE_IOMMU
 	default y
 
 config DRM_EXYNOS_FIMD
 	bool "Exynos DRM FIMD"
-	depends on DRM_EXYNOS && !FB_S3C
+	depends on !FB_S3C
 	select FB_MODE_HELPERS
 	select MFD_SYSCON
 	help
@@ -26,20 +28,19 @@ config DRM_EXYNOS_FIMD
 
 config DRM_EXYNOS5433_DECON
 	bool "Exynos5433 DRM DECON"
-	depends on DRM_EXYNOS
 	help
 	  Choose this option if you want to use Exynos5433 DECON for DRM.
 
 config DRM_EXYNOS7_DECON
 	bool "Exynos7 DRM DECON"
-	depends on DRM_EXYNOS && !FB_S3C
+	depends on !FB_S3C
 	select FB_MODE_HELPERS
 	help
 	  Choose this option if you want to use Exynos DECON for DRM.
 
 config DRM_EXYNOS_DPI
 	bool "EXYNOS DRM parallel output support"
-	depends on DRM_EXYNOS && (DRM_EXYNOS_FIMD || DRM_EXYNOS7_DECON)
+	depends on DRM_EXYNOS_FIMD || DRM_EXYNOS7_DECON
 	select DRM_PANEL
 	default n
 	help
@@ -47,7 +48,7 @@ config DRM_EXYNOS_DPI
 
 config DRM_EXYNOS_DSI
 	bool "EXYNOS DRM MIPI-DSI driver support"
-	depends on DRM_EXYNOS && (DRM_EXYNOS_FIMD || DRM_EXYNOS5433_DECON || DRM_EXYNOS7_DECON)
+	depends on DRM_EXYNOS_FIMD || DRM_EXYNOS5433_DECON || DRM_EXYNOS7_DECON
 	select DRM_MIPI_DSI
 	select DRM_PANEL
 	default n
@@ -56,7 +57,7 @@ config DRM_EXYNOS_DSI
 
 config DRM_EXYNOS_DP
 	bool "EXYNOS DRM DP driver support"
-	depends on DRM_EXYNOS && (DRM_EXYNOS_FIMD || DRM_EXYNOS7_DECON)
+	depends on DRM_EXYNOS_FIMD || DRM_EXYNOS7_DECON
 	default DRM_EXYNOS
 	select DRM_PANEL
 	help
@@ -64,32 +65,30 @@ config DRM_EXYNOS_DP
 
 config DRM_EXYNOS_MIXER
 	bool "Exynos DRM Mixer"
-	depends on DRM_EXYNOS && !VIDEO_SAMSUNG_S5P_TV
+	depends on !VIDEO_SAMSUNG_S5P_TV
 	help
 	  Choose this option if you want to use Exynos Mixer for DRM.
 
 config DRM_EXYNOS_HDMI
 	bool "Exynos DRM HDMI"
-	depends on DRM_EXYNOS && !VIDEO_SAMSUNG_S5P_TV && (DRM_EXYNOS_MIXER || DRM_EXYNOS5433_DECON)
+	depends on !VIDEO_SAMSUNG_S5P_TV && (DRM_EXYNOS_MIXER || DRM_EXYNOS5433_DECON)
 	help
 	  Choose this option if you want to use Exynos HDMI for DRM.
 
 config DRM_EXYNOS_VIDI
 	bool "Exynos DRM Virtual Display"
-	depends on DRM_EXYNOS
 	help
 	  Choose this option if you want to use Exynos VIDI for DRM.
 
 config DRM_EXYNOS_G2D
 	bool "Exynos DRM G2D"
-	depends on DRM_EXYNOS && !VIDEO_SAMSUNG_S5P_G2D
+	depends on !VIDEO_SAMSUNG_S5P_G2D
 	select FRAME_VECTOR
 	help
 	  Choose this option if you want to use Exynos G2D for DRM.
 
 config DRM_EXYNOS_IPP
 	bool "Exynos DRM IPP"
-	depends on DRM_EXYNOS
 	help
 	  Choose this option if you want to use IPP feature for DRM.
 
@@ -113,6 +112,8 @@ config DRM_EXYNOS_GSC
 
 config DRM_EXYNOS_MIC
 	bool "Exynos DRM MIC"
-	depends on (DRM_EXYNOS && DRM_EXYNOS5433_DECON)
+	depends on DRM_EXYNOS5433_DECON
 	help
 	  Choose this option if you want to use Exynos MIC for DRM.
+
+endif
-- 
cgit v0.10.2


From ea9776465d8f1617c99ffe7f90aa245fba474544 Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <a.hajda@samsung.com>
Date: Mon, 26 Oct 2015 13:03:44 +0100
Subject: drm/exynos: re-arrange Kconfig entries

Exynos DRM driver have quite big number of components and options.
The patch re-arranges them into three logical groups:
- CRTCs,
- Encoders and Bridges,
- Sub-drivers.
It should make driver options more clear.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/Kconfig b/drivers/gpu/drm/exynos/Kconfig
index b0107f0..b2d6cbd 100644
--- a/drivers/gpu/drm/exynos/Kconfig
+++ b/drivers/gpu/drm/exynos/Kconfig
@@ -18,6 +18,8 @@ config DRM_EXYNOS_IOMMU
 	depends on EXYNOS_IOMMU && ARM_DMA_USE_IOMMU
 	default y
 
+comment "CRTCs"
+
 config DRM_EXYNOS_FIMD
 	bool "Exynos DRM FIMD"
 	depends on !FB_S3C
@@ -38,9 +40,22 @@ config DRM_EXYNOS7_DECON
 	help
 	  Choose this option if you want to use Exynos DECON for DRM.
 
+config DRM_EXYNOS_MIXER
+	bool "Exynos DRM Mixer"
+	depends on !VIDEO_SAMSUNG_S5P_TV
+	help
+	  Choose this option if you want to use Exynos Mixer for DRM.
+
+config DRM_EXYNOS_VIDI
+	bool "Exynos DRM Virtual Display"
+	help
+	  Choose this option if you want to use Exynos VIDI for DRM.
+
+comment "Encoders and Bridges"
+
 config DRM_EXYNOS_DPI
 	bool "EXYNOS DRM parallel output support"
-	depends on DRM_EXYNOS_FIMD || DRM_EXYNOS7_DECON
+	depends on DRM_EXYNOS_FIMD
 	select DRM_PANEL
 	default n
 	help
@@ -63,22 +78,19 @@ config DRM_EXYNOS_DP
 	help
 	  This enables support for DP device.
 
-config DRM_EXYNOS_MIXER
-	bool "Exynos DRM Mixer"
-	depends on !VIDEO_SAMSUNG_S5P_TV
-	help
-	  Choose this option if you want to use Exynos Mixer for DRM.
-
 config DRM_EXYNOS_HDMI
 	bool "Exynos DRM HDMI"
 	depends on !VIDEO_SAMSUNG_S5P_TV && (DRM_EXYNOS_MIXER || DRM_EXYNOS5433_DECON)
 	help
 	  Choose this option if you want to use Exynos HDMI for DRM.
 
-config DRM_EXYNOS_VIDI
-	bool "Exynos DRM Virtual Display"
+config DRM_EXYNOS_MIC
+	bool "Exynos DRM MIC"
+	depends on DRM_EXYNOS5433_DECON
 	help
-	  Choose this option if you want to use Exynos VIDI for DRM.
+	  Choose this option if you want to use Exynos MIC for DRM.
+
+comment "Sub-drivers"
 
 config DRM_EXYNOS_G2D
 	bool "Exynos DRM G2D"
@@ -110,10 +122,4 @@ config DRM_EXYNOS_GSC
 	help
 	  Choose this option if you want to use Exynos GSC for DRM.
 
-config DRM_EXYNOS_MIC
-	bool "Exynos DRM MIC"
-	depends on DRM_EXYNOS5433_DECON
-	help
-	  Choose this option if you want to use Exynos MIC for DRM.
-
 endif
-- 
cgit v0.10.2


From 5a3c35b377ce5b1d438792ee4ebb8e6222dcd50d Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <a.hajda@samsung.com>
Date: Mon, 26 Oct 2015 13:03:45 +0100
Subject: drm/exynos: simplify Kconfig component names

Many Exynos DRM sub-options mentions Exynos DRM in their titles.
It is redundant and can be safely shortened. The patch additionally
makes some entries more descriptive.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/Kconfig b/drivers/gpu/drm/exynos/Kconfig
index b2d6cbd..96e86cf 100644
--- a/drivers/gpu/drm/exynos/Kconfig
+++ b/drivers/gpu/drm/exynos/Kconfig
@@ -21,7 +21,7 @@ config DRM_EXYNOS_IOMMU
 comment "CRTCs"
 
 config DRM_EXYNOS_FIMD
-	bool "Exynos DRM FIMD"
+	bool "FIMD"
 	depends on !FB_S3C
 	select FB_MODE_HELPERS
 	select MFD_SYSCON
@@ -29,32 +29,32 @@ config DRM_EXYNOS_FIMD
 	  Choose this option if you want to use Exynos FIMD for DRM.
 
 config DRM_EXYNOS5433_DECON
-	bool "Exynos5433 DRM DECON"
+	bool "DECON on Exynos5433"
 	help
 	  Choose this option if you want to use Exynos5433 DECON for DRM.
 
 config DRM_EXYNOS7_DECON
-	bool "Exynos7 DRM DECON"
+	bool "DECON on Exynos7"
 	depends on !FB_S3C
 	select FB_MODE_HELPERS
 	help
 	  Choose this option if you want to use Exynos DECON for DRM.
 
 config DRM_EXYNOS_MIXER
-	bool "Exynos DRM Mixer"
+	bool "Mixer"
 	depends on !VIDEO_SAMSUNG_S5P_TV
 	help
 	  Choose this option if you want to use Exynos Mixer for DRM.
 
 config DRM_EXYNOS_VIDI
-	bool "Exynos DRM Virtual Display"
+	bool "Virtual Display"
 	help
 	  Choose this option if you want to use Exynos VIDI for DRM.
 
 comment "Encoders and Bridges"
 
 config DRM_EXYNOS_DPI
-	bool "EXYNOS DRM parallel output support"
+	bool "Parallel output"
 	depends on DRM_EXYNOS_FIMD
 	select DRM_PANEL
 	default n
@@ -62,7 +62,7 @@ config DRM_EXYNOS_DPI
 	  This enables support for Exynos parallel output.
 
 config DRM_EXYNOS_DSI
-	bool "EXYNOS DRM MIPI-DSI driver support"
+	bool "MIPI-DSI host"
 	depends on DRM_EXYNOS_FIMD || DRM_EXYNOS5433_DECON || DRM_EXYNOS7_DECON
 	select DRM_MIPI_DSI
 	select DRM_PANEL
@@ -71,7 +71,7 @@ config DRM_EXYNOS_DSI
 	  This enables support for Exynos MIPI-DSI device.
 
 config DRM_EXYNOS_DP
-	bool "EXYNOS DRM DP driver support"
+	bool "Display Port"
 	depends on DRM_EXYNOS_FIMD || DRM_EXYNOS7_DECON
 	default DRM_EXYNOS
 	select DRM_PANEL
@@ -79,13 +79,13 @@ config DRM_EXYNOS_DP
 	  This enables support for DP device.
 
 config DRM_EXYNOS_HDMI
-	bool "Exynos DRM HDMI"
+	bool "HDMI"
 	depends on !VIDEO_SAMSUNG_S5P_TV && (DRM_EXYNOS_MIXER || DRM_EXYNOS5433_DECON)
 	help
 	  Choose this option if you want to use Exynos HDMI for DRM.
 
 config DRM_EXYNOS_MIC
-	bool "Exynos DRM MIC"
+	bool "Mobile Image Compressor"
 	depends on DRM_EXYNOS5433_DECON
 	help
 	  Choose this option if you want to use Exynos MIC for DRM.
@@ -93,31 +93,31 @@ config DRM_EXYNOS_MIC
 comment "Sub-drivers"
 
 config DRM_EXYNOS_G2D
-	bool "Exynos DRM G2D"
+	bool "G2D"
 	depends on !VIDEO_SAMSUNG_S5P_G2D
 	select FRAME_VECTOR
 	help
 	  Choose this option if you want to use Exynos G2D for DRM.
 
 config DRM_EXYNOS_IPP
-	bool "Exynos DRM IPP"
+	bool "Image Post Processor"
 	help
 	  Choose this option if you want to use IPP feature for DRM.
 
 config DRM_EXYNOS_FIMC
-	bool "Exynos DRM FIMC"
+	bool "FIMC"
 	depends on DRM_EXYNOS_IPP && MFD_SYSCON
 	help
 	  Choose this option if you want to use Exynos FIMC for DRM.
 
 config DRM_EXYNOS_ROTATOR
-	bool "Exynos DRM Rotator"
+	bool "Rotator"
 	depends on DRM_EXYNOS_IPP
 	help
 	  Choose this option if you want to use Exynos Rotator for DRM.
 
 config DRM_EXYNOS_GSC
-	bool "Exynos DRM GSC"
+	bool "GScaler"
 	depends on DRM_EXYNOS_IPP && ARCH_EXYNOS5 && !ARCH_MULTIPLATFORM
 	help
 	  Choose this option if you want to use Exynos GSC for DRM.
-- 
cgit v0.10.2


From 01351315465fe58b9ab990554467773367f74cab Mon Sep 17 00:00:00 2001
From: Andrzej Hajda <a.hajda@samsung.com>
Date: Mon, 26 Oct 2015 13:03:41 +0100
Subject: ARM: exynos_defconfig: enable Exynos DRM Mixer driver

Mixer driver is selected by CONFIG_DRM_EXYNOS_HDMI option. Since Exynos5433
HDMI does not require Mixer. There will be separate options to select Mixer
and HDMI. Adding new option to defconfig before Kconfig will allow to keep
bisectability.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>
Acked-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/arch/arm/configs/exynos_defconfig b/arch/arm/configs/exynos_defconfig
index 1ff2bfa..af79a21 100644
--- a/arch/arm/configs/exynos_defconfig
+++ b/arch/arm/configs/exynos_defconfig
@@ -132,6 +132,7 @@ CONFIG_DRM_PARADE_PS8622=y
 CONFIG_DRM_EXYNOS=y
 CONFIG_DRM_EXYNOS_FIMD=y
 CONFIG_DRM_EXYNOS_DSI=y
+CONFIG_DRM_EXYNOS_MIXER=y
 CONFIG_DRM_EXYNOS_HDMI=y
 CONFIG_DRM_PANEL_SIMPLE=y
 CONFIG_DRM_PANEL_SAMSUNG_S6E8AA0=y
-- 
cgit v0.10.2


From df547bf7735a623500eedff9cc6716ac1d82b95d Mon Sep 17 00:00:00 2001
From: Marek Szyprowski <m.szyprowski@samsung.com>
Date: Tue, 13 Oct 2015 13:47:20 +0200
Subject: drm/exynos/gem: remove DMA-mapping hacks used for constructing page
 array

Exynos GEM objects contains an array of pointers to the pages, which the
allocated buffer consists of. Till now the code used some hacks (like
relying on DMA-mapping internal structures or using ARM-specific
dma_to_pfn helper) to build this array. This patch fixes this by adding
proper call to dma_get_sgtable_attrs() and using the acquired scatter-list
to construct needed array. This approach is more portable (work also for
ARM64) and finally fixes the layering violation that was present in this
code.

Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>

diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.c b/drivers/gpu/drm/exynos/exynos_drm_gem.c
index f1dcdd0..252eb30 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_gem.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_gem.c
@@ -25,6 +25,8 @@ static int exynos_drm_alloc_buf(struct exynos_drm_gem *exynos_gem)
 	struct drm_device *dev = exynos_gem->base.dev;
 	enum dma_attr attr;
 	unsigned int nr_pages;
+	struct sg_table sgt;
+	int ret = -ENOMEM;
 
 	if (exynos_gem->dma_addr) {
 		DRM_DEBUG_KMS("already allocated.\n");
@@ -56,13 +58,10 @@ static int exynos_drm_alloc_buf(struct exynos_drm_gem *exynos_gem)
 
 	nr_pages = exynos_gem->size >> PAGE_SHIFT;
 
-	if (!is_drm_iommu_supported(dev)) {
-		exynos_gem->pages = drm_calloc_large(nr_pages,
-						     sizeof(struct page *));
-		if (!exynos_gem->pages) {
-			DRM_ERROR("failed to allocate pages.\n");
-			return -ENOMEM;
-		}
+	exynos_gem->pages = drm_calloc_large(nr_pages, sizeof(struct page *));
+	if (!exynos_gem->pages) {
+		DRM_ERROR("failed to allocate pages.\n");
+		return -ENOMEM;
 	}
 
 	exynos_gem->cookie = dma_alloc_attrs(dev->dev, exynos_gem->size,
@@ -70,30 +69,40 @@ static int exynos_drm_alloc_buf(struct exynos_drm_gem *exynos_gem)
 					     &exynos_gem->dma_attrs);
 	if (!exynos_gem->cookie) {
 		DRM_ERROR("failed to allocate buffer.\n");
-		if (exynos_gem->pages)
-			drm_free_large(exynos_gem->pages);
-		return -ENOMEM;
+		goto err_free;
 	}
 
-	if (exynos_gem->pages) {
-		dma_addr_t start_addr;
-		unsigned int i = 0;
-
-		start_addr = exynos_gem->dma_addr;
-		while (i < nr_pages) {
-			exynos_gem->pages[i] =
-				pfn_to_page(dma_to_pfn(dev->dev, start_addr));
-			start_addr += PAGE_SIZE;
-			i++;
-		}
-	} else {
-		exynos_gem->pages = exynos_gem->cookie;
+	ret = dma_get_sgtable_attrs(dev->dev, &sgt, exynos_gem->cookie,
+				    exynos_gem->dma_addr, exynos_gem->size,
+				    &exynos_gem->dma_attrs);
+	if (ret < 0) {
+		DRM_ERROR("failed to get sgtable.\n");
+		goto err_dma_free;
+	}
+
+	if (drm_prime_sg_to_page_addr_arrays(&sgt, exynos_gem->pages, NULL,
+					     nr_pages)) {
+		DRM_ERROR("invalid sgtable.\n");
+		ret = -EINVAL;
+		goto err_sgt_free;
 	}
 
+	sg_free_table(&sgt);
+
 	DRM_DEBUG_KMS("dma_addr(0x%lx), size(0x%lx)\n",
 			(unsigned long)exynos_gem->dma_addr, exynos_gem->size);
 
 	return 0;
+
+err_sgt_free:
+	sg_free_table(&sgt);
+err_dma_free:
+	dma_free_attrs(dev->dev, exynos_gem->size, exynos_gem->cookie,
+		       exynos_gem->dma_addr, &exynos_gem->dma_attrs);
+err_free:
+	drm_free_large(exynos_gem->pages);
+
+	return ret;
 }
 
 static void exynos_drm_free_buf(struct exynos_drm_gem *exynos_gem)
@@ -112,8 +121,7 @@ static void exynos_drm_free_buf(struct exynos_drm_gem *exynos_gem)
 			(dma_addr_t)exynos_gem->dma_addr,
 			&exynos_gem->dma_attrs);
 
-	if (!is_drm_iommu_supported(dev))
-		drm_free_large(exynos_gem->pages);
+	drm_free_large(exynos_gem->pages);
 }
 
 static int exynos_drm_gem_handle_create(struct drm_gem_object *obj,
-- 
cgit v0.10.2


From 04b8a4bd8e01e25b9fa9fa7b1c957a7346ae83c1 Mon Sep 17 00:00:00 2001
From: Ilia Mirkin <imirkin@alum.mit.edu>
Date: Tue, 20 Oct 2015 01:15:39 -0400
Subject: drm/nouveau/gem: return only valid domain when there's only one

On nv50+, we restrict the valid domains to just the one where the buffer
was originally created. However after the buffer is evicted to system
memory, we might move it back to a different domain that was not
originally valid. When sharing the buffer and retrieving its GEM_INFO
data, we still want the domain that will be valid for this buffer in a
pushbuf, not the one where it currently happens to be.

This resolves fdo#92504 and several others. These are due to suspend
evicting all buffers, making it more likely that they temporarily end up
in the wrong place.

Cc: stable@vger.kernel.org
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=92504
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/nouveau_gem.c b/drivers/gpu/drm/nouveau/nouveau_gem.c
index 2c99815..41be584 100644
--- a/drivers/gpu/drm/nouveau/nouveau_gem.c
+++ b/drivers/gpu/drm/nouveau/nouveau_gem.c
@@ -227,11 +227,12 @@ nouveau_gem_info(struct drm_file *file_priv, struct drm_gem_object *gem,
 	struct nouveau_bo *nvbo = nouveau_gem_object(gem);
 	struct nvkm_vma *vma;
 
-	if (nvbo->bo.mem.mem_type == TTM_PL_TT)
+	if (is_power_of_2(nvbo->valid_domains))
+		rep->domain = nvbo->valid_domains;
+	else if (nvbo->bo.mem.mem_type == TTM_PL_TT)
 		rep->domain = NOUVEAU_GEM_DOMAIN_GART;
 	else
 		rep->domain = NOUVEAU_GEM_DOMAIN_VRAM;
-
 	rep->offset = nvbo->bo.offset;
 	if (cli->vm) {
 		vma = nouveau_bo_vma_find(nvbo, cli->vm);
-- 
cgit v0.10.2


From 579b7c58215329803ce184704463de09f0f310ac Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Thu, 3 Sep 2015 17:39:52 +0900
Subject: drm/nouveau/pmu: do not assume a PMU is present

Some devices may not have a PMU. Avoid a NULL pointer dereference in
such cases by checking whether the pointer given to nvkm_pmu_pgob() is
valid.

Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/base.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/base.c
index 27a79c0..d95eb86 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/base.c
@@ -28,7 +28,7 @@
 void
 nvkm_pmu_pgob(struct nvkm_pmu *pmu, bool enable)
 {
-	if (pmu->func->pgob)
+	if (pmu && pmu->func->pgob)
 		pmu->func->pgob(pmu, enable);
 }
 
-- 
cgit v0.10.2


From 542f60dc84c6f382a56a0b42dc09aa183198a55f Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Thu, 3 Sep 2015 17:48:13 +0900
Subject: drm/nouveau/timer: re-introduce nvkm_wait_xsec macros

Reintroduce macros allowing us to test a register against a certain
mask, since this is the most common usage pattern for the more generic
nvkm_xsec macros and makes the code more concise and readable.

Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/timer.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/timer.h
index 62ed088..82d3e28 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/timer.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/timer.h
@@ -59,6 +59,16 @@ void nvkm_timer_alarm_cancel(struct nvkm_timer *, struct nvkm_alarm *);
 #define nvkm_usec(d,u,cond...) nvkm_nsec((d), (u) * 1000, ##cond)
 #define nvkm_msec(d,m,cond...) nvkm_usec((d), (m) * 1000, ##cond)
 
+#define nvkm_wait_nsec(d,n,addr,mask,data)                                     \
+	nvkm_nsec(d, n,                                                        \
+		if ((nvkm_rd32(d, (addr)) & (mask)) == (data))                 \
+			break;                                                 \
+		)
+#define nvkm_wait_usec(d,u,addr,mask,data)                                     \
+	nvkm_wait_nsec((d), (u) * 1000, (addr), (mask), (data))
+#define nvkm_wait_msec(d,m,addr,mask,data)                                     \
+	nvkm_wait_usec((d), (m) * 1000, (addr), (mask), (data))
+
 int nv04_timer_new(struct nvkm_device *, int, struct nvkm_timer **);
 int nv40_timer_new(struct nvkm_device *, int, struct nvkm_timer **);
 int nv41_timer_new(struct nvkm_device *, int, struct nvkm_timer **);
-- 
cgit v0.10.2


From 38a8fc78d00749328888bb0e37ad93535e7e9992 Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Thu, 3 Sep 2015 17:48:14 +0900
Subject: drm/nouveau/ltc: add hooks for invalidate and flush

These are useful for systems without a coherent CPU/GPU bus. For such
systems we may need to maintain the L2 ourselves.

Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/ltc.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/ltc.h
index c773b5e..5464fcf 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/ltc.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/ltc.h
@@ -30,6 +30,9 @@ void nvkm_ltc_tags_clear(struct nvkm_ltc *, u32 first, u32 count);
 int nvkm_ltc_zbc_color_get(struct nvkm_ltc *, int index, const u32[4]);
 int nvkm_ltc_zbc_depth_get(struct nvkm_ltc *, int index, const u32);
 
+void nvkm_ltc_invalidate(struct nvkm_ltc *);
+void nvkm_ltc_flush(struct nvkm_ltc *);
+
 int gf100_ltc_new(struct nvkm_device *, int, struct nvkm_ltc **);
 int gk104_ltc_new(struct nvkm_device *, int, struct nvkm_ltc **);
 int gm107_ltc_new(struct nvkm_device *, int, struct nvkm_ltc **);
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/base.c b/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/base.c
index 930d25b..85b1464 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/base.c
@@ -67,6 +67,20 @@ nvkm_ltc_zbc_depth_get(struct nvkm_ltc *ltc, int index, const u32 depth)
 	return index;
 }
 
+void
+nvkm_ltc_invalidate(struct nvkm_ltc *ltc)
+{
+	if (ltc->func->invalidate)
+		ltc->func->invalidate(ltc);
+}
+
+void
+nvkm_ltc_flush(struct nvkm_ltc *ltc)
+{
+	if (ltc->func->flush)
+		ltc->func->flush(ltc);
+}
+
 static void
 nvkm_ltc_intr(struct nvkm_subdev *subdev)
 {
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/priv.h b/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/priv.h
index 4e05037..6f66bd0 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/priv.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/priv.h
@@ -17,6 +17,9 @@ struct nvkm_ltc_func {
 	int zbc;
 	void (*zbc_clear_color)(struct nvkm_ltc *, int, const u32[4]);
 	void (*zbc_clear_depth)(struct nvkm_ltc *, int, const u32);
+
+	void (*invalidate)(struct nvkm_ltc *);
+	void (*flush)(struct nvkm_ltc *);
 };
 
 int gf100_ltc_oneinit(struct nvkm_ltc *);
-- 
cgit v0.10.2


From a0a49bac2fe18375b7ccb9de4671960531294470 Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Thu, 3 Sep 2015 17:48:15 +0900
Subject: drm/nouveau/ltc/gf100: add flush/invalidate functions

Allow clients to manually flush and invalidate L2. This will be useful
for Tegra systems for which we want to write instmem using the CPU.

Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/ltc.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/ltc.h
index 5464fcf..3d4dbbf 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/ltc.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/ltc.h
@@ -35,5 +35,6 @@ void nvkm_ltc_flush(struct nvkm_ltc *);
 
 int gf100_ltc_new(struct nvkm_device *, int, struct nvkm_ltc **);
 int gk104_ltc_new(struct nvkm_device *, int, struct nvkm_ltc **);
+int gk20a_ltc_new(struct nvkm_device *, int, struct nvkm_ltc **);
 int gm107_ltc_new(struct nvkm_device *, int, struct nvkm_ltc **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/gf100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/gf100.c
index 45ac765..fb0de83 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/gf100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/gf100.c
@@ -122,6 +122,36 @@ gf100_ltc_intr(struct nvkm_ltc *ltc)
 	}
 }
 
+void
+gf100_ltc_invalidate(struct nvkm_ltc *ltc)
+{
+	struct nvkm_device *device = ltc->subdev.device;
+	s64 taken;
+
+	nvkm_wr32(device, 0x70004, 0x00000001);
+	taken = nvkm_wait_msec(device, 2, 0x70004, 0x00000003, 0x00000000);
+	if (taken < 0)
+		nvkm_warn(&ltc->subdev, "LTC invalidate timeout\n");
+
+	if (taken > 0)
+		nvkm_debug(&ltc->subdev, "LTC invalidate took %lld ns\n", taken);
+}
+
+void
+gf100_ltc_flush(struct nvkm_ltc *ltc)
+{
+	struct nvkm_device *device = ltc->subdev.device;
+	s64 taken;
+
+	nvkm_wr32(device, 0x70010, 0x00000001);
+	taken = nvkm_wait_msec(device, 2, 0x70010, 0x00000003, 0x00000000);
+	if (taken < 0)
+		nvkm_warn(&ltc->subdev, "LTC flush timeout\n");
+
+	if (taken > 0)
+		nvkm_debug(&ltc->subdev, "LTC flush took %lld ns\n", taken);
+}
+
 /* TODO: Figure out tag memory details and drop the over-cautious allocation.
  */
 int
@@ -215,6 +245,8 @@ gf100_ltc = {
 	.zbc = 16,
 	.zbc_clear_color = gf100_ltc_zbc_clear_color,
 	.zbc_clear_depth = gf100_ltc_zbc_clear_depth,
+	.invalidate = gf100_ltc_invalidate,
+	.flush = gf100_ltc_flush,
 };
 
 int
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/gk104.c b/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/gk104.c
index 839e6b4..b4f6e00 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/gk104.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/gk104.c
@@ -45,6 +45,8 @@ gk104_ltc = {
 	.zbc = 16,
 	.zbc_clear_color = gf100_ltc_zbc_clear_color,
 	.zbc_clear_depth = gf100_ltc_zbc_clear_depth,
+	.invalidate = gf100_ltc_invalidate,
+	.flush = gf100_ltc_flush,
 };
 
 int
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/gm107.c b/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/gm107.c
index 389331b..3043bbf 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/gm107.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/gm107.c
@@ -138,6 +138,8 @@ gm107_ltc = {
 	.zbc = 16,
 	.zbc_clear_color = gm107_ltc_zbc_clear_color,
 	.zbc_clear_depth = gm107_ltc_zbc_clear_depth,
+	.invalidate = gf100_ltc_invalidate,
+	.flush = gf100_ltc_flush,
 };
 
 int
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/priv.h b/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/priv.h
index 6f66bd0..4e3755b 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/priv.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/priv.h
@@ -29,4 +29,6 @@ void gf100_ltc_cbc_clear(struct nvkm_ltc *, u32, u32);
 void gf100_ltc_cbc_wait(struct nvkm_ltc *);
 void gf100_ltc_zbc_clear_color(struct nvkm_ltc *, int, const u32[4]);
 void gf100_ltc_zbc_clear_depth(struct nvkm_ltc *, int, const u32);
+void gf100_ltc_invalidate(struct nvkm_ltc *);
+void gf100_ltc_flush(struct nvkm_ltc *);
 #endif
-- 
cgit v0.10.2


From fcf3f91c34105c3551741febbfc1066aaa7f1db7 Mon Sep 17 00:00:00 2001
From: Ben Skeggs <bskeggs@redhat.com>
Date: Fri, 4 Sep 2015 14:40:32 +1000
Subject: drm/nouveau: remove unnecessary usage of object handles

No longer required in a lot of cases, as objects are identified over NVIF
via an alternate mechanism since the rework.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/nouveau_abi16.c b/drivers/gpu/drm/nouveau/nouveau_abi16.c
index d336c22..0b3c8ab 100644
--- a/drivers/gpu/drm/nouveau/nouveau_abi16.c
+++ b/drivers/gpu/drm/nouveau/nouveau_abi16.c
@@ -51,8 +51,7 @@ nouveau_abi16_get(struct drm_file *file_priv, struct drm_device *dev)
 			 * device (ie. the one that belongs to the fd it
 			 * opened)
 			 */
-			if (nvif_device_init(&cli->base.object,
-					     NOUVEAU_ABI16_DEVICE, NV_DEVICE,
+			if (nvif_device_init(&cli->base.object, 0, NV_DEVICE,
 					     &args, sizeof(args),
 					     &abi16->device) == 0)
 				return cli->abi16;
@@ -133,7 +132,6 @@ nouveau_abi16_chan_fini(struct nouveau_abi16 *abi16,
 
 	/* destroy channel object, all children will be killed too */
 	if (chan->chan) {
-		abi16->handles &= ~(1ULL << (chan->chan->user.handle & 0xffff));
 		nouveau_channel_idle(chan->chan);
 		nouveau_channel_del(&chan->chan);
 	}
@@ -268,26 +266,21 @@ nouveau_abi16_ioctl_channel_alloc(ABI16_IOCTL_ARGS)
 		return nouveau_abi16_put(abi16, -EINVAL);
 
 	/* allocate "abi16 channel" data and make up a handle for it */
-	init->channel = __ffs64(~abi16->handles);
-	if (~abi16->handles == 0)
-		return nouveau_abi16_put(abi16, -ENOSPC);
-
 	chan = kzalloc(sizeof(*chan), GFP_KERNEL);
 	if (!chan)
 		return nouveau_abi16_put(abi16, -ENOMEM);
 
 	INIT_LIST_HEAD(&chan->notifiers);
 	list_add(&chan->head, &abi16->channels);
-	abi16->handles |= (1ULL << init->channel);
 
 	/* create channel object and initialise dma and fence management */
-	ret = nouveau_channel_new(drm, device,
-				  NOUVEAU_ABI16_CHAN(init->channel),
-				  init->fb_ctxdma_handle,
+	ret = nouveau_channel_new(drm, device, init->fb_ctxdma_handle,
 				  init->tt_ctxdma_handle, &chan->chan);
 	if (ret)
 		goto done;
 
+	init->channel = chan->chan->chid;
+
 	if (device->info.family >= NV_DEVICE_INFO_V0_TESLA)
 		init->pushbuf_domains = NOUVEAU_GEM_DOMAIN_VRAM |
 					NOUVEAU_GEM_DOMAIN_GART;
@@ -338,7 +331,7 @@ nouveau_abi16_chan(struct nouveau_abi16 *abi16, int channel)
 	struct nouveau_abi16_chan *chan;
 
 	list_for_each_entry(chan, &abi16->channels, head) {
-		if (chan->chan->user.handle == NOUVEAU_ABI16_CHAN(channel))
+		if (chan->chan->chid == channel)
 			return chan;
 	}
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_chan.c b/drivers/gpu/drm/nouveau/nouveau_chan.c
index ff5e59d..1860f38 100644
--- a/drivers/gpu/drm/nouveau/nouveau_chan.c
+++ b/drivers/gpu/drm/nouveau/nouveau_chan.c
@@ -55,10 +55,8 @@ nouveau_channel_idle(struct nouveau_channel *chan)
 		}
 
 		if (ret) {
-			NV_PRINTK(err, cli, "failed to idle channel "
-					    "0x%08x [%s]\n",
-				  chan->user.handle,
-				  nvxx_client(&cli->base)->name);
+			NV_PRINTK(err, cli, "failed to idle channel %d [%s]\n",
+				  chan->chid, nvxx_client(&cli->base)->name);
 			return ret;
 		}
 	}
@@ -89,7 +87,7 @@ nouveau_channel_del(struct nouveau_channel **pchan)
 
 static int
 nouveau_channel_prep(struct nouveau_drm *drm, struct nvif_device *device,
-		     u32 handle, u32 size, struct nouveau_channel **pchan)
+		     u32 size, struct nouveau_channel **pchan)
 {
 	struct nouveau_cli *cli = (void *)device->object.client;
 	struct nvkm_mmu *mmu = nvxx_mmu(device);
@@ -174,8 +172,7 @@ nouveau_channel_prep(struct nouveau_drm *drm, struct nvif_device *device,
 		}
 	}
 
-	ret = nvif_object_init(&device->object, NVDRM_PUSH |
-			       (handle & 0xffff), NV_DMA_FROM_MEMORY,
+	ret = nvif_object_init(&device->object, 0, NV_DMA_FROM_MEMORY,
 			       &args, sizeof(args), &chan->push.ctxdma);
 	if (ret) {
 		nouveau_channel_del(pchan);
@@ -187,7 +184,7 @@ nouveau_channel_prep(struct nouveau_drm *drm, struct nvif_device *device,
 
 static int
 nouveau_channel_ind(struct nouveau_drm *drm, struct nvif_device *device,
-		    u32 handle, u32 engine, struct nouveau_channel **pchan)
+		    u32 engine, struct nouveau_channel **pchan)
 {
 	static const u16 oclasses[] = { MAXWELL_CHANNEL_GPFIFO_A,
 					KEPLER_CHANNEL_GPFIFO_A,
@@ -206,7 +203,7 @@ nouveau_channel_ind(struct nouveau_drm *drm, struct nvif_device *device,
 	int ret;
 
 	/* allocate dma push buffer */
-	ret = nouveau_channel_prep(drm, device, handle, 0x12000, &chan);
+	ret = nouveau_channel_prep(drm, device, 0x12000, &chan);
 	*pchan = chan;
 	if (ret)
 		return ret;
@@ -236,7 +233,7 @@ nouveau_channel_ind(struct nouveau_drm *drm, struct nvif_device *device,
 			size = sizeof(args.nv50);
 		}
 
-		ret = nvif_object_init(&device->object, handle, *oclass++,
+		ret = nvif_object_init(&device->object, 0, *oclass++,
 				       &args, size, &chan->user);
 		if (ret == 0) {
 			if (chan->user.oclass >= KEPLER_CHANNEL_GPFIFO_A)
@@ -256,7 +253,7 @@ nouveau_channel_ind(struct nouveau_drm *drm, struct nvif_device *device,
 
 static int
 nouveau_channel_dma(struct nouveau_drm *drm, struct nvif_device *device,
-		    u32 handle, struct nouveau_channel **pchan)
+		    struct nouveau_channel **pchan)
 {
 	static const u16 oclasses[] = { NV40_CHANNEL_DMA,
 					NV17_CHANNEL_DMA,
@@ -269,7 +266,7 @@ nouveau_channel_dma(struct nouveau_drm *drm, struct nvif_device *device,
 	int ret;
 
 	/* allocate dma push buffer */
-	ret = nouveau_channel_prep(drm, device, handle, 0x10000, &chan);
+	ret = nouveau_channel_prep(drm, device, 0x10000, &chan);
 	*pchan = chan;
 	if (ret)
 		return ret;
@@ -280,7 +277,7 @@ nouveau_channel_dma(struct nouveau_drm *drm, struct nvif_device *device,
 	args.offset = chan->push.vma.offset;
 
 	do {
-		ret = nvif_object_init(&device->object, handle, *oclass++,
+		ret = nvif_object_init(&device->object, 0, *oclass++,
 				       &args, sizeof(args), &chan->user);
 		if (ret == 0) {
 			chan->chid = args.chid;
@@ -401,8 +398,7 @@ nouveau_channel_init(struct nouveau_channel *chan, u32 vram, u32 gart)
 
 int
 nouveau_channel_new(struct nouveau_drm *drm, struct nvif_device *device,
-		    u32 handle, u32 arg0, u32 arg1,
-		    struct nouveau_channel **pchan)
+		    u32 arg0, u32 arg1, struct nouveau_channel **pchan)
 {
 	struct nouveau_cli *cli = (void *)device->object.client;
 	bool super;
@@ -412,10 +408,10 @@ nouveau_channel_new(struct nouveau_drm *drm, struct nvif_device *device,
 	super = cli->base.super;
 	cli->base.super = true;
 
-	ret = nouveau_channel_ind(drm, device, handle, arg0, pchan);
+	ret = nouveau_channel_ind(drm, device, arg0, pchan);
 	if (ret) {
 		NV_PRINTK(dbg, cli, "ib channel create, %d\n", ret);
-		ret = nouveau_channel_dma(drm, device, handle, pchan);
+		ret = nouveau_channel_dma(drm, device, pchan);
 		if (ret) {
 			NV_PRINTK(dbg, cli, "dma channel create, %d\n", ret);
 			goto done;
diff --git a/drivers/gpu/drm/nouveau/nouveau_chan.h b/drivers/gpu/drm/nouveau/nouveau_chan.h
index 2ed3241..48062c9 100644
--- a/drivers/gpu/drm/nouveau/nouveau_chan.h
+++ b/drivers/gpu/drm/nouveau/nouveau_chan.h
@@ -42,8 +42,7 @@ struct nouveau_channel {
 
 
 int  nouveau_channel_new(struct nouveau_drm *, struct nvif_device *,
-			 u32 handle, u32 arg0, u32 arg1,
-			 struct nouveau_channel **);
+			 u32 arg0, u32 arg1, struct nouveau_channel **);
 void nouveau_channel_del(struct nouveau_channel **);
 int  nouveau_channel_idle(struct nouveau_channel *);
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_display.c b/drivers/gpu/drm/nouveau/nouveau_display.c
index 614b32e..db6bc67 100644
--- a/drivers/gpu/drm/nouveau/nouveau_display.c
+++ b/drivers/gpu/drm/nouveau/nouveau_display.c
@@ -509,9 +509,8 @@ nouveau_display_create(struct drm_device *dev)
 		int i;
 
 		for (i = 0, ret = -ENODEV; ret && i < ARRAY_SIZE(oclass); i++) {
-			ret = nvif_object_init(&drm->device.object,
-					       NVDRM_DISPLAY, oclass[i],
-					       NULL, 0, &disp->disp);
+			ret = nvif_object_init(&drm->device.object, 0,
+					       oclass[i], NULL, 0, &disp->disp);
 		}
 
 		if (ret == 0) {
diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c
index 45ba678..f932665 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_drm.c
@@ -208,7 +208,7 @@ nouveau_accel_init(struct nouveau_drm *drm)
 	}
 
 	if (device->info.family >= NV_DEVICE_INFO_V0_KEPLER) {
-		ret = nouveau_channel_new(drm, &drm->device, NVDRM_CHAN + 1,
+		ret = nouveau_channel_new(drm, &drm->device,
 					  KEPLER_CHANNEL_GPFIFO_A_V0_ENGINE_CE0|
 					  KEPLER_CHANNEL_GPFIFO_A_V0_ENGINE_CE1,
 					  0, &drm->cechan);
@@ -221,7 +221,7 @@ nouveau_accel_init(struct nouveau_drm *drm)
 	if (device->info.chipset >= 0xa3 &&
 	    device->info.chipset != 0xaa &&
 	    device->info.chipset != 0xac) {
-		ret = nouveau_channel_new(drm, &drm->device, NVDRM_CHAN + 1,
+		ret = nouveau_channel_new(drm, &drm->device,
 					  NvDmaFB, NvDmaTT, &drm->cechan);
 		if (ret)
 			NV_ERROR(drm, "failed to create ce channel, %d\n", ret);
@@ -233,8 +233,7 @@ nouveau_accel_init(struct nouveau_drm *drm)
 		arg1 = NvDmaTT;
 	}
 
-	ret = nouveau_channel_new(drm, &drm->device, NVDRM_CHAN, arg0, arg1,
-				 &drm->channel);
+	ret = nouveau_channel_new(drm, &drm->device, arg0, arg1, &drm->channel);
 	if (ret) {
 		NV_ERROR(drm, "failed to create kernel channel, %d\n", ret);
 		nouveau_accel_fini(drm);
@@ -403,8 +402,7 @@ nouveau_drm_load(struct drm_device *dev, unsigned long flags)
 
 	nouveau_get_hdmi_dev(drm);
 
-	ret = nvif_device_init(&drm->client.base.object,
-			       NVDRM_DEVICE, NV_DEVICE,
+	ret = nvif_device_init(&drm->client.base.object, 0, NV_DEVICE,
 			       &(struct nv_device_v0) {
 					.device = ~0,
 			       }, sizeof(struct nv_device_v0),
diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.h b/drivers/gpu/drm/nouveau/nouveau_drm.h
index 3c902c2..acfa03b 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drm.h
+++ b/drivers/gpu/drm/nouveau/nouveau_drm.h
@@ -74,11 +74,6 @@ enum nouveau_drm_notify_route {
 };
 
 enum nouveau_drm_handle {
-	NVDRM_CLIENT  = 0xffffffff,
-	NVDRM_DEVICE  = 0xdddddddd,
-	NVDRM_CONTROL = 0xdddddddc,
-	NVDRM_DISPLAY = 0xd1500000,
-	NVDRM_PUSH    = 0xbbbb0000, /* |= client chid */
 	NVDRM_CHAN    = 0xcccc0000, /* |= client chid */
 	NVDRM_NVSW    = 0x55550000,
 };
diff --git a/drivers/gpu/drm/nouveau/nouveau_gem.c b/drivers/gpu/drm/nouveau/nouveau_gem.c
index 41be584..6bf1a78 100644
--- a/drivers/gpu/drm/nouveau/nouveau_gem.c
+++ b/drivers/gpu/drm/nouveau/nouveau_gem.c
@@ -682,7 +682,7 @@ nouveau_gem_ioctl_pushbuf(struct drm_device *dev, void *data,
 		return -ENOMEM;
 
 	list_for_each_entry(temp, &abi16->channels, head) {
-		if (temp->chan->user.handle == (NVDRM_CHAN | req->channel)) {
+		if (temp->chan->chid == req->channel) {
 			chan = temp->chan;
 			break;
 		}
diff --git a/drivers/gpu/drm/nouveau/nouveau_sysfs.c b/drivers/gpu/drm/nouveau/nouveau_sysfs.c
index d12a5fa..5dac354 100644
--- a/drivers/gpu/drm/nouveau/nouveau_sysfs.c
+++ b/drivers/gpu/drm/nouveau/nouveau_sysfs.c
@@ -188,9 +188,8 @@ nouveau_sysfs_init(struct drm_device *dev)
 	if (!sysfs)
 		return -ENOMEM;
 
-	ret = nvif_object_init(&device->object, NVDRM_CONTROL,
-			       NVIF_IOCTL_NEW_V0_CONTROL, NULL, 0,
-			       &sysfs->ctrl);
+	ret = nvif_object_init(&device->object, 0, NVIF_IOCTL_NEW_V0_CONTROL,
+			       NULL, 0, &sysfs->ctrl);
 	if (ret == 0)
 		device_create_file(nvxx_device(device)->dev, &dev_attr_pstate);
 
diff --git a/drivers/gpu/drm/nouveau/nv50_display.c b/drivers/gpu/drm/nouveau/nv50_display.c
index 4ae87ae..c053c50 100644
--- a/drivers/gpu/drm/nouveau/nv50_display.c
+++ b/drivers/gpu/drm/nouveau/nv50_display.c
@@ -68,7 +68,6 @@ nv50_chan_create(struct nvif_device *device, struct nvif_object *disp,
 		 const s32 *oclass, u8 head, void *data, u32 size,
 		 struct nv50_chan *chan)
 {
-	const u32 handle = (oclass[0] << 16) | head;
 	struct nvif_sclass *sclass;
 	int ret, i, n;
 
@@ -81,7 +80,7 @@ nv50_chan_create(struct nvif_device *device, struct nvif_object *disp,
 	while (oclass[0]) {
 		for (i = 0; i < n; i++) {
 			if (sclass[i].oclass == oclass[0]) {
-				ret = nvif_object_init(disp, handle, oclass[0],
+				ret = nvif_object_init(disp, 0, oclass[0],
 						       data, size, &chan->user);
 				if (ret == 0)
 					nvif_object_map(&chan->user);
@@ -231,8 +230,8 @@ nv50_dmac_create(struct nvif_device *device, struct nvif_object *disp,
 	if (!dmac->ptr)
 		return -ENOMEM;
 
-	ret = nvif_object_init(&device->object, 0xd0000000,
-			       NV_DMA_FROM_MEMORY, &(struct nv_dma_v0) {
+	ret = nvif_object_init(&device->object, 0, NV_DMA_FROM_MEMORY,
+			       &(struct nv_dma_v0) {
 					.target = NV_DMA_V0_TARGET_PCI_US,
 					.access = NV_DMA_V0_ACCESS_RD,
 					.start = dmac->handle + 0x0000,
diff --git a/include/uapi/drm/nouveau_drm.h b/include/uapi/drm/nouveau_drm.h
index 5507eea..fd594cc 100644
--- a/include/uapi/drm/nouveau_drm.h
+++ b/include/uapi/drm/nouveau_drm.h
@@ -27,14 +27,6 @@
 
 #define DRM_NOUVEAU_EVENT_NVIF                                       0x80000000
 
-/* reserved object handles when using deprecated object APIs - these
- * are here so that libdrm can allow interoperability with the new
- * object APIs
- */
-#define NOUVEAU_ABI16_CLIENT   0xffffffff
-#define NOUVEAU_ABI16_DEVICE   0xdddddddd
-#define NOUVEAU_ABI16_CHAN(n) (0xcccc0000 | (n))
-
 #define NOUVEAU_GEM_DOMAIN_CPU       (1 << 0)
 #define NOUVEAU_GEM_DOMAIN_VRAM      (1 << 1)
 #define NOUVEAU_GEM_DOMAIN_GART      (1 << 2)
-- 
cgit v0.10.2


From 69c4938249fb48aeed32fd76c67972e71f471cd2 Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Fri, 4 Sep 2015 19:52:11 +0900
Subject: drm/nouveau/instmem/gk20a: use direct CPU access

The Great Nouveau Refactoring Take II brought us a lot of goodness,
including acquire/release methods that are called before and after an
instobj is modified. These functions can be used as synchronization
points to manage CPU/GPU coherency if we modify an instobj using the
CPU.

This patch replaces the legacy and slow PRAMIN access for gk20a instmem
with CPU mappings and writes. A LRU list is used to unmap unused
mappings after a certain threshold (currently 1MB) of mapped instobjs is
reached. This allows mappings to be reused most of the time.

Accessing instobjs using the CPU requires to maintain the GPU L2 cache,
which we do in the acquire/release functions. This triggers a lot of L2
flushes/invalidates, but most of them are performed on an empty cache
(and thus return immediately), and overall context setup performance
greatly benefits from this (from 250ms to 160ms on Jetson TK1 for a
simple libdrm program).

Making L2 management more explicit should allow us to grab some more
performance in the future.

Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/gk20a.c b/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/gk20a.c
index cd7feb1..a2921ac 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/gk20a.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/gk20a.c
@@ -23,35 +23,42 @@
 /*
  * GK20A does not have dedicated video memory, and to accurately represent this
  * fact Nouveau will not create a RAM device for it. Therefore its instmem
- * implementation must be done directly on top of system memory, while providing
- * coherent read and write operations.
+ * implementation must be done directly on top of system memory, while
+ * preserving coherency for read and write operations.
  *
  * Instmem can be allocated through two means:
- * 1) If an IOMMU mapping has been probed, the IOMMU API is used to make memory
+ * 1) If an IOMMU unit has been probed, the IOMMU API is used to make memory
  *    pages contiguous to the GPU. This is the preferred way.
- * 2) If no IOMMU mapping is probed, the DMA API is used to allocate physically
+ * 2) If no IOMMU unit is probed, the DMA API is used to allocate physically
  *    contiguous memory.
  *
- * In both cases CPU read and writes are performed using PRAMIN (i.e. using the
- * GPU path) to ensure these operations are coherent for the GPU. This allows us
- * to use more "relaxed" allocation parameters when using the DMA API, since we
- * never need a kernel mapping.
+ * In both cases CPU read and writes are performed by creating a write-combined
+ * mapping. The GPU L2 cache must thus be flushed/invalidated when required. To
+ * be conservative we do this every time we acquire or release an instobj, but
+ * ideally L2 management should be handled at a higher level.
+ *
+ * To improve performance, CPU mappings are not removed upon instobj release.
+ * Instead they are placed into a LRU list to be recycled when the mapped space
+ * goes beyond a certain threshold. At the moment this limit is 1MB.
  */
-#define gk20a_instmem(p) container_of((p), struct gk20a_instmem, base)
 #include "priv.h"
 
 #include <core/memory.h>
 #include <core/mm.h>
 #include <core/tegra.h>
 #include <subdev/fb.h>
-
-#define gk20a_instobj(p) container_of((p), struct gk20a_instobj, memory)
+#include <subdev/ltc.h>
 
 struct gk20a_instobj {
 	struct nvkm_memory memory;
-	struct gk20a_instmem *imem;
 	struct nvkm_mem mem;
+	struct gk20a_instmem *imem;
+
+	/* CPU mapping */
+	u32 *vaddr;
+	struct list_head vaddr_node;
 };
+#define gk20a_instobj(p) container_of((p), struct gk20a_instobj, memory)
 
 /*
  * Used for objects allocated using the DMA API
@@ -59,10 +66,12 @@ struct gk20a_instobj {
 struct gk20a_instobj_dma {
 	struct gk20a_instobj base;
 
-	void *cpuaddr;
+	u32 *cpuaddr;
 	dma_addr_t handle;
 	struct nvkm_mm_node r;
 };
+#define gk20a_instobj_dma(p) \
+	container_of(gk20a_instobj(p), struct gk20a_instobj_dma, base)
 
 /*
  * Used for objects flattened using the IOMMU API
@@ -70,15 +79,24 @@ struct gk20a_instobj_dma {
 struct gk20a_instobj_iommu {
 	struct gk20a_instobj base;
 
-	/* array of base.mem->size pages */
+	/* will point to the higher half of pages */
+	dma_addr_t *dma_addrs;
+	/* array of base.mem->size pages (+ dma_addr_ts) */
 	struct page *pages[];
 };
+#define gk20a_instobj_iommu(p) \
+	container_of(gk20a_instobj(p), struct gk20a_instobj_iommu, base)
 
 struct gk20a_instmem {
 	struct nvkm_instmem base;
-	unsigned long lock_flags;
+
+	/* protects vaddr_* and gk20a_instobj::vaddr* */
 	spinlock_t lock;
-	u64 addr;
+
+	/* CPU mappings LRU */
+	unsigned int vaddr_use;
+	unsigned int vaddr_max;
+	struct list_head vaddr_lru;
 
 	/* Only used if IOMMU if present */
 	struct mutex *mm_mutex;
@@ -88,7 +106,10 @@ struct gk20a_instmem {
 
 	/* Only used by DMA API */
 	struct dma_attrs attrs;
+
+	void __iomem * (*cpu_map)(struct nvkm_memory *);
 };
+#define gk20a_instmem(p) container_of((p), struct gk20a_instmem, base)
 
 static enum nvkm_memory_target
 gk20a_instobj_target(struct nvkm_memory *memory)
@@ -100,7 +121,6 @@ static u64
 gk20a_instobj_addr(struct nvkm_memory *memory)
 {
 	return gk20a_instobj(memory)->mem.offset;
-
 }
 
 static u64
@@ -110,107 +130,217 @@ gk20a_instobj_size(struct nvkm_memory *memory)
 }
 
 static void __iomem *
+gk20a_instobj_cpu_map_dma(struct nvkm_memory *memory)
+{
+	struct gk20a_instobj_dma *node = gk20a_instobj_dma(memory);
+	struct device *dev = node->base.imem->base.subdev.device->dev;
+	int npages = nvkm_memory_size(memory) >> 12;
+	struct page *pages[npages];
+	int i;
+
+	/* phys_to_page does not exist on all platforms... */
+	pages[0] = pfn_to_page(dma_to_phys(dev, node->handle) >> PAGE_SHIFT);
+	for (i = 1; i < npages; i++)
+		pages[i] = pages[0] + i;
+
+	return vmap(pages, npages, VM_MAP, pgprot_writecombine(PAGE_KERNEL));
+}
+
+static void __iomem *
+gk20a_instobj_cpu_map_iommu(struct nvkm_memory *memory)
+{
+	struct gk20a_instobj_iommu *node = gk20a_instobj_iommu(memory);
+	int npages = nvkm_memory_size(memory) >> 12;
+
+	return vmap(node->pages, npages, VM_MAP,
+		    pgprot_writecombine(PAGE_KERNEL));
+}
+
+/*
+ * Must be called while holding gk20a_instmem_lock
+ */
+static void
+gk20a_instmem_vaddr_gc(struct gk20a_instmem *imem, const u64 size)
+{
+	while (imem->vaddr_use + size > imem->vaddr_max) {
+		struct gk20a_instobj *obj;
+
+		/* no candidate that can be unmapped, abort... */
+		if (list_empty(&imem->vaddr_lru))
+			break;
+
+		obj = list_first_entry(&imem->vaddr_lru, struct gk20a_instobj,
+				       vaddr_node);
+		list_del(&obj->vaddr_node);
+		vunmap(obj->vaddr);
+		obj->vaddr = NULL;
+		imem->vaddr_use -= nvkm_memory_size(&obj->memory);
+		nvkm_debug(&imem->base.subdev, "(GC) vaddr used: %x/%x\n",
+			   imem->vaddr_use, imem->vaddr_max);
+
+	}
+}
+
+static void __iomem *
 gk20a_instobj_acquire(struct nvkm_memory *memory)
 {
-	struct gk20a_instmem *imem = gk20a_instobj(memory)->imem;
+	struct gk20a_instobj *node = gk20a_instobj(memory);
+	struct gk20a_instmem *imem = node->imem;
+	struct nvkm_ltc *ltc = imem->base.subdev.device->ltc;
+	const u64 size = nvkm_memory_size(memory);
 	unsigned long flags;
+
+	nvkm_ltc_flush(ltc);
+
 	spin_lock_irqsave(&imem->lock, flags);
-	imem->lock_flags = flags;
-	return NULL;
+
+	if (node->vaddr) {
+		/* remove us from the LRU list since we cannot be unmapped */
+		list_del(&node->vaddr_node);
+
+		goto out;
+	}
+
+	/* try to free some address space if we reached the limit */
+	gk20a_instmem_vaddr_gc(imem, size);
+
+	node->vaddr = imem->cpu_map(memory);
+
+	if (!node->vaddr) {
+		nvkm_error(&imem->base.subdev, "cannot map instobj - "
+			   "this is not going to end well...\n");
+		goto out;
+	}
+
+	imem->vaddr_use += size;
+	nvkm_debug(&imem->base.subdev, "vaddr used: %x/%x\n",
+		   imem->vaddr_use, imem->vaddr_max);
+
+out:
+	spin_unlock_irqrestore(&imem->lock, flags);
+
+	return node->vaddr;
 }
 
 static void
 gk20a_instobj_release(struct nvkm_memory *memory)
 {
-	struct gk20a_instmem *imem = gk20a_instobj(memory)->imem;
-	spin_unlock_irqrestore(&imem->lock, imem->lock_flags);
-}
+	struct gk20a_instobj *node = gk20a_instobj(memory);
+	struct gk20a_instmem *imem = node->imem;
+	struct nvkm_ltc *ltc = imem->base.subdev.device->ltc;
+	unsigned long flags;
 
-/*
- * Use PRAMIN to read/write data and avoid coherency issues.
- * PRAMIN uses the GPU path and ensures data will always be coherent.
- *
- * A dynamic mapping based solution would be desirable in the future, but
- * the issue remains of how to maintain coherency efficiently. On ARM it is
- * not easy (if possible at all?) to create uncached temporary mappings.
- */
+	spin_lock_irqsave(&imem->lock, flags);
+
+	/* add ourselves to the LRU list so our CPU mapping can be freed */
+	list_add_tail(&node->vaddr_node, &imem->vaddr_lru);
+
+	spin_unlock_irqrestore(&imem->lock, flags);
+
+	wmb();
+	nvkm_ltc_invalidate(ltc);
+}
 
 static u32
 gk20a_instobj_rd32(struct nvkm_memory *memory, u64 offset)
 {
 	struct gk20a_instobj *node = gk20a_instobj(memory);
-	struct gk20a_instmem *imem = node->imem;
-	struct nvkm_device *device = imem->base.subdev.device;
-	u64 base = (node->mem.offset + offset) & 0xffffff00000ULL;
-	u64 addr = (node->mem.offset + offset) & 0x000000fffffULL;
-	u32 data;
-
-	if (unlikely(imem->addr != base)) {
-		nvkm_wr32(device, 0x001700, base >> 16);
-		imem->addr = base;
-	}
-	data = nvkm_rd32(device, 0x700000 + addr);
-	return data;
+
+	return node->vaddr[offset / 4];
 }
 
 static void
 gk20a_instobj_wr32(struct nvkm_memory *memory, u64 offset, u32 data)
 {
 	struct gk20a_instobj *node = gk20a_instobj(memory);
-	struct gk20a_instmem *imem = node->imem;
-	struct nvkm_device *device = imem->base.subdev.device;
-	u64 base = (node->mem.offset + offset) & 0xffffff00000ULL;
-	u64 addr = (node->mem.offset + offset) & 0x000000fffffULL;
 
-	if (unlikely(imem->addr != base)) {
-		nvkm_wr32(device, 0x001700, base >> 16);
-		imem->addr = base;
-	}
-	nvkm_wr32(device, 0x700000 + addr, data);
+	node->vaddr[offset / 4] = data;
 }
 
 static void
 gk20a_instobj_map(struct nvkm_memory *memory, struct nvkm_vma *vma, u64 offset)
 {
 	struct gk20a_instobj *node = gk20a_instobj(memory);
+
 	nvkm_vm_map_at(vma, offset, &node->mem);
 }
 
+/*
+ * Clear the CPU mapping of an instobj if it exists
+ */
 static void
-gk20a_instobj_dtor_dma(struct gk20a_instobj *_node)
+gk20a_instobj_dtor(struct gk20a_instobj *node)
+{
+	struct gk20a_instmem *imem = node->imem;
+	struct gk20a_instobj *obj;
+	unsigned long flags;
+
+	spin_lock_irqsave(&imem->lock, flags);
+
+	if (!node->vaddr)
+		goto out;
+
+	list_for_each_entry(obj, &imem->vaddr_lru, vaddr_node) {
+		if (obj == node) {
+			list_del(&obj->vaddr_node);
+			break;
+		}
+	}
+	vunmap(node->vaddr);
+	node->vaddr = NULL;
+	imem->vaddr_use -= nvkm_memory_size(&node->memory);
+	nvkm_debug(&imem->base.subdev, "vaddr used: %x/%x\n",
+		   imem->vaddr_use, imem->vaddr_max);
+
+out:
+	spin_unlock_irqrestore(&imem->lock, flags);
+}
+
+static void *
+gk20a_instobj_dtor_dma(struct nvkm_memory *memory)
 {
-	struct gk20a_instobj_dma *node = (void *)_node;
-	struct gk20a_instmem *imem = _node->imem;
+	struct gk20a_instobj_dma *node = gk20a_instobj_dma(memory);
+	struct gk20a_instmem *imem = node->base.imem;
 	struct device *dev = imem->base.subdev.device->dev;
 
+	gk20a_instobj_dtor(&node->base);
+
 	if (unlikely(!node->cpuaddr))
-		return;
+		goto out;
 
-	dma_free_attrs(dev, _node->mem.size << PAGE_SHIFT, node->cpuaddr,
+	dma_free_attrs(dev, node->base.mem.size << PAGE_SHIFT, node->cpuaddr,
 		       node->handle, &imem->attrs);
+
+out:
+	return node;
 }
 
-static void
-gk20a_instobj_dtor_iommu(struct gk20a_instobj *_node)
+static void *
+gk20a_instobj_dtor_iommu(struct nvkm_memory *memory)
 {
-	struct gk20a_instobj_iommu *node = (void *)_node;
-	struct gk20a_instmem *imem = _node->imem;
+	struct gk20a_instobj_iommu *node = gk20a_instobj_iommu(memory);
+	struct gk20a_instmem *imem = node->base.imem;
+	struct device *dev = imem->base.subdev.device->dev;
 	struct nvkm_mm_node *r;
 	int i;
 
-	if (unlikely(list_empty(&_node->mem.regions)))
-		return;
+	gk20a_instobj_dtor(&node->base);
 
-	r = list_first_entry(&_node->mem.regions, struct nvkm_mm_node,
+	if (unlikely(list_empty(&node->base.mem.regions)))
+		goto out;
+
+	r = list_first_entry(&node->base.mem.regions, struct nvkm_mm_node,
 			     rl_entry);
 
 	/* clear bit 34 to unmap pages */
 	r->offset &= ~BIT(34 - imem->iommu_pgshift);
 
 	/* Unmap pages from GPU address space and free them */
-	for (i = 0; i < _node->mem.size; i++) {
+	for (i = 0; i < node->base.mem.size; i++) {
 		iommu_unmap(imem->domain,
 			    (r->offset + i) << imem->iommu_pgshift, PAGE_SIZE);
+		dma_unmap_page(dev, node->dma_addrs[i], PAGE_SIZE,
+			       DMA_BIDIRECTIONAL);
 		__free_page(node->pages[i]);
 	}
 
@@ -218,25 +348,27 @@ gk20a_instobj_dtor_iommu(struct gk20a_instobj *_node)
 	mutex_lock(imem->mm_mutex);
 	nvkm_mm_free(imem->mm, &r);
 	mutex_unlock(imem->mm_mutex);
-}
-
-static void *
-gk20a_instobj_dtor(struct nvkm_memory *memory)
-{
-	struct gk20a_instobj *node = gk20a_instobj(memory);
-	struct gk20a_instmem *imem = node->imem;
-
-	if (imem->domain)
-		gk20a_instobj_dtor_iommu(node);
-	else
-		gk20a_instobj_dtor_dma(node);
 
+out:
 	return node;
 }
 
 static const struct nvkm_memory_func
-gk20a_instobj_func = {
-	.dtor = gk20a_instobj_dtor,
+gk20a_instobj_func_dma = {
+	.dtor = gk20a_instobj_dtor_dma,
+	.target = gk20a_instobj_target,
+	.addr = gk20a_instobj_addr,
+	.size = gk20a_instobj_size,
+	.acquire = gk20a_instobj_acquire,
+	.release = gk20a_instobj_release,
+	.rd32 = gk20a_instobj_rd32,
+	.wr32 = gk20a_instobj_wr32,
+	.map = gk20a_instobj_map,
+};
+
+static const struct nvkm_memory_func
+gk20a_instobj_func_iommu = {
+	.dtor = gk20a_instobj_dtor_iommu,
 	.target = gk20a_instobj_target,
 	.addr = gk20a_instobj_addr,
 	.size = gk20a_instobj_size,
@@ -259,6 +391,8 @@ gk20a_instobj_ctor_dma(struct gk20a_instmem *imem, u32 npages, u32 align,
 		return -ENOMEM;
 	*_node = &node->base;
 
+	nvkm_memory_ctor(&gk20a_instobj_func_dma, &node->base.memory);
+
 	node->cpuaddr = dma_alloc_attrs(dev, npages << PAGE_SHIFT,
 					&node->handle, GFP_KERNEL,
 					&imem->attrs);
@@ -292,24 +426,40 @@ gk20a_instobj_ctor_iommu(struct gk20a_instmem *imem, u32 npages, u32 align,
 {
 	struct gk20a_instobj_iommu *node;
 	struct nvkm_subdev *subdev = &imem->base.subdev;
+	struct device *dev = subdev->device->dev;
 	struct nvkm_mm_node *r;
 	int ret;
 	int i;
 
-	if (!(node = kzalloc(sizeof(*node) +
-			     sizeof( node->pages[0]) * npages, GFP_KERNEL)))
+	/*
+	 * despite their variable size, instmem allocations are small enough
+	 * (< 1 page) to be handled by kzalloc
+	 */
+	if (!(node = kzalloc(sizeof(*node) + ((sizeof(node->pages[0]) +
+			     sizeof(*node->dma_addrs)) * npages), GFP_KERNEL)))
 		return -ENOMEM;
 	*_node = &node->base;
+	node->dma_addrs = (void *)(node->pages + npages);
+
+	nvkm_memory_ctor(&gk20a_instobj_func_iommu, &node->base.memory);
 
 	/* Allocate backing memory */
 	for (i = 0; i < npages; i++) {
 		struct page *p = alloc_page(GFP_KERNEL);
+		dma_addr_t dma_adr;
 
 		if (p == NULL) {
 			ret = -ENOMEM;
 			goto free_pages;
 		}
 		node->pages[i] = p;
+		dma_adr = dma_map_page(dev, p, 0, PAGE_SIZE, DMA_BIDIRECTIONAL);
+		if (dma_mapping_error(dev, dma_adr)) {
+			nvkm_error(subdev, "DMA mapping error!\n");
+			ret = -ENOMEM;
+			goto free_pages;
+		}
+		node->dma_addrs[i] = dma_adr;
 	}
 
 	mutex_lock(imem->mm_mutex);
@@ -318,16 +468,15 @@ gk20a_instobj_ctor_iommu(struct gk20a_instmem *imem, u32 npages, u32 align,
 			   align >> imem->iommu_pgshift, &r);
 	mutex_unlock(imem->mm_mutex);
 	if (ret) {
-		nvkm_error(subdev, "virtual space is full!\n");
+		nvkm_error(subdev, "IOMMU space is full!\n");
 		goto free_pages;
 	}
 
 	/* Map into GPU address space */
 	for (i = 0; i < npages; i++) {
-		struct page *p = node->pages[i];
 		u32 offset = (r->offset + i) << imem->iommu_pgshift;
 
-		ret = iommu_map(imem->domain, offset, page_to_phys(p),
+		ret = iommu_map(imem->domain, offset, node->dma_addrs[i],
 				PAGE_SIZE, IOMMU_READ | IOMMU_WRITE);
 		if (ret < 0) {
 			nvkm_error(subdev, "IOMMU mapping failure: %d\n", ret);
@@ -356,8 +505,13 @@ release_area:
 	mutex_unlock(imem->mm_mutex);
 
 free_pages:
-	for (i = 0; i < npages && node->pages[i] != NULL; i++)
+	for (i = 0; i < npages && node->pages[i] != NULL; i++) {
+		dma_addr_t dma_addr = node->dma_addrs[i];
+		if (dma_addr)
+			dma_unmap_page(dev, dma_addr, PAGE_SIZE,
+				       DMA_BIDIRECTIONAL);
 		__free_page(node->pages[i]);
+	}
 
 	return ret;
 }
@@ -367,8 +521,8 @@ gk20a_instobj_new(struct nvkm_instmem *base, u32 size, u32 align, bool zero,
 		  struct nvkm_memory **pmemory)
 {
 	struct gk20a_instmem *imem = gk20a_instmem(base);
-	struct gk20a_instobj *node = NULL;
 	struct nvkm_subdev *subdev = &imem->base.subdev;
+	struct gk20a_instobj *node = NULL;
 	int ret;
 
 	nvkm_debug(subdev, "%s (%s): size: %x align: %x\n", __func__,
@@ -388,7 +542,6 @@ gk20a_instobj_new(struct nvkm_instmem *base, u32 size, u32 align, bool zero,
 	if (ret)
 		return ret;
 
-	nvkm_memory_ctor(&gk20a_instobj_func, &node->memory);
 	node->imem = imem;
 
 	/* present memory for being mapped using small pages */
@@ -402,15 +555,25 @@ gk20a_instobj_new(struct nvkm_instmem *base, u32 size, u32 align, bool zero,
 	return 0;
 }
 
-static void
-gk20a_instmem_fini(struct nvkm_instmem *base)
+static void *
+gk20a_instmem_dtor(struct nvkm_instmem *base)
 {
-	gk20a_instmem(base)->addr = ~0ULL;
+	struct gk20a_instmem *imem = gk20a_instmem(base);
+
+	/* perform some sanity checks... */
+	if (!list_empty(&imem->vaddr_lru))
+		nvkm_warn(&base->subdev, "instobj LRU not empty!\n");
+
+	if (imem->vaddr_use != 0)
+		nvkm_warn(&base->subdev, "instobj vmap area not empty! "
+			  "0x%x bytes still mapped\n", imem->vaddr_use);
+
+	return imem;
 }
 
 static const struct nvkm_instmem_func
 gk20a_instmem = {
-	.fini = gk20a_instmem_fini,
+	.dtor = gk20a_instmem_dtor,
 	.memory_new = gk20a_instobj_new,
 	.persistent = true,
 	.zero = false,
@@ -429,23 +592,27 @@ gk20a_instmem_new(struct nvkm_device *device, int index,
 	spin_lock_init(&imem->lock);
 	*pimem = &imem->base;
 
+	/* do not allow more than 1MB of CPU-mapped instmem */
+	imem->vaddr_use = 0;
+	imem->vaddr_max = 0x100000;
+	INIT_LIST_HEAD(&imem->vaddr_lru);
+
 	if (tdev->iommu.domain) {
-		imem->domain = tdev->iommu.domain;
+		imem->mm_mutex = &tdev->iommu.mutex;
 		imem->mm = &tdev->iommu.mm;
+		imem->domain = tdev->iommu.domain;
 		imem->iommu_pgshift = tdev->iommu.pgshift;
-		imem->mm_mutex = &tdev->iommu.mutex;
+		imem->cpu_map = gk20a_instobj_cpu_map_iommu;
 
 		nvkm_info(&imem->base.subdev, "using IOMMU\n");
 	} else {
 		init_dma_attrs(&imem->attrs);
-		/*
-		 * We will access instmem through PRAMIN and thus do not need a
-		 * consistent CPU pointer or kernel mapping
-		 */
+		/* We will access the memory through our own mapping */
 		dma_set_attr(DMA_ATTR_NON_CONSISTENT, &imem->attrs);
 		dma_set_attr(DMA_ATTR_WEAK_ORDERING, &imem->attrs);
 		dma_set_attr(DMA_ATTR_WRITE_COMBINE, &imem->attrs);
 		dma_set_attr(DMA_ATTR_NO_KERNEL_MAPPING, &imem->attrs);
+		imem->cpu_map = gk20a_instobj_cpu_map_dma;
 
 		nvkm_info(&imem->base.subdev, "using DMA API\n");
 	}
-- 
cgit v0.10.2


From e396ecd178c6c8373e765d78b5b1a29e38837d3a Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Fri, 4 Sep 2015 19:59:31 +0900
Subject: drm/nouveau/platform: allow to specify the IOMMU bit

Current Tegra code taking advantage of the IOMMU assumes a hardcoded
value for the IOMMU bit. Make it a platform property instead for
flexibility.

v2 (Ben Skeggs): remove nvkm dependence on drm structures

Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/include/nvif/os.h b/drivers/gpu/drm/nouveau/include/nvif/os.h
index 3accc99..9fcab67 100644
--- a/drivers/gpu/drm/nouveau/include/nvif/os.h
+++ b/drivers/gpu/drm/nouveau/include/nvif/os.h
@@ -27,6 +27,7 @@
 #include <linux/agp_backend.h>
 #include <linux/reset.h>
 #include <linux/iommu.h>
+#include <linux/of_device.h>
 
 #include <asm/unaligned.h>
 
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/core/tegra.h b/drivers/gpu/drm/nouveau/include/nvkm/core/tegra.h
index 5aa2480..16641ce 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/core/tegra.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/core/tegra.h
@@ -4,6 +4,7 @@
 #include <core/mm.h>
 
 struct nvkm_device_tegra {
+	const struct nvkm_device_tegra_func *func;
 	struct nvkm_device device;
 	struct platform_device *pdev;
 	int irq;
@@ -28,7 +29,17 @@ struct nvkm_device_tegra {
 	int gpu_speedo;
 };
 
-int nvkm_device_tegra_new(struct platform_device *,
+struct nvkm_device_tegra_func {
+	/*
+	 * If an IOMMU is used, indicates which address bit will trigger a
+	 * IOMMU translation when set (when this bit is not set, IOMMU is
+	 * bypassed). A value of 0 means an IOMMU is never used.
+	 */
+	u8 iommu_bit;
+};
+
+int nvkm_device_tegra_new(const struct nvkm_device_tegra_func *,
+			  struct platform_device *,
 			  const char *cfg, const char *dbg,
 			  bool detect, bool mmio, u64 subdev_mask,
 			  struct nvkm_device **);
diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c
index f932665..1d3ee51 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_drm.c
@@ -1028,13 +1028,14 @@ nouveau_drm_pci_driver = {
 };
 
 struct drm_device *
-nouveau_platform_device_create(struct platform_device *pdev,
+nouveau_platform_device_create(const struct nvkm_device_tegra_func *func,
+			       struct platform_device *pdev,
 			       struct nvkm_device **pdevice)
 {
 	struct drm_device *drm;
 	int err;
 
-	err = nvkm_device_tegra_new(pdev, nouveau_config, nouveau_debug,
+	err = nvkm_device_tegra_new(func, pdev, nouveau_config, nouveau_debug,
 				    true, true, ~0ULL, pdevice);
 	if (err)
 		goto err_free;
diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.h b/drivers/gpu/drm/nouveau/nouveau_drm.h
index acfa03b..51027a9 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drm.h
+++ b/drivers/gpu/drm/nouveau/nouveau_drm.h
@@ -178,8 +178,11 @@ nouveau_drm(struct drm_device *dev)
 int nouveau_pmops_suspend(struct device *);
 int nouveau_pmops_resume(struct device *);
 
+#include <nvkm/core/tegra.h>
+
 struct drm_device *
-nouveau_platform_device_create(struct platform_device *, struct nvkm_device **);
+nouveau_platform_device_create(const struct nvkm_device_tegra_func *,
+			       struct platform_device *, struct nvkm_device **);
 void nouveau_drm_device_remove(struct drm_device *dev);
 
 #define NV_PRINTK(l,c,f,a...) do {                                             \
diff --git a/drivers/gpu/drm/nouveau/nouveau_platform.c b/drivers/gpu/drm/nouveau/nouveau_platform.c
index 3eb6654..60e32c4 100644
--- a/drivers/gpu/drm/nouveau/nouveau_platform.c
+++ b/drivers/gpu/drm/nouveau/nouveau_platform.c
@@ -23,11 +23,14 @@
 
 static int nouveau_platform_probe(struct platform_device *pdev)
 {
+	const struct nvkm_device_tegra_func *func;
 	struct nvkm_device *device;
 	struct drm_device *drm;
 	int ret;
 
-	drm = nouveau_platform_device_create(pdev, &device);
+	func = of_device_get_match_data(&pdev->dev);
+
+	drm = nouveau_platform_device_create(func, pdev, &device);
 	if (IS_ERR(drm))
 		return PTR_ERR(drm);
 
@@ -48,9 +51,19 @@ static int nouveau_platform_remove(struct platform_device *pdev)
 }
 
 #if IS_ENABLED(CONFIG_OF)
+static const struct nvkm_device_tegra_func gk20a_platform_data = {
+	.iommu_bit = 34,
+};
+
 static const struct of_device_id nouveau_platform_match[] = {
-	{ .compatible = "nvidia,gk20a" },
-	{ .compatible = "nvidia,gm20b" },
+	{
+		.compatible = "nvidia,gk20a",
+		.data = &gk20a_platform_data,
+	},
+	{
+		.compatible = "nvidia,gm20b",
+		.data = &gk20a_platform_data,
+	},
 	{ }
 };
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/tegra.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/tegra.c
index da57c8a..7f8a427 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/device/tegra.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/tegra.c
@@ -85,6 +85,9 @@ nvkm_device_tegra_probe_iommu(struct nvkm_device_tegra *tdev)
 	unsigned long pgsize_bitmap;
 	int ret;
 
+	if (!tdev->func->iommu_bit)
+		return;
+
 	mutex_init(&tdev->iommu.mutex);
 
 	if (iommu_present(&platform_bus_type)) {
@@ -114,7 +117,8 @@ nvkm_device_tegra_probe_iommu(struct nvkm_device_tegra *tdev)
 			goto free_domain;
 
 		ret = nvkm_mm_init(&tdev->iommu.mm, 0,
-				   (1ULL << 40) >> tdev->iommu.pgshift, 1);
+				   (1ULL << tdev->func->iommu_bit) >>
+				   tdev->iommu.pgshift, 1);
 		if (ret)
 			goto detach_device;
 	}
@@ -237,7 +241,8 @@ nvkm_device_tegra_func = {
 };
 
 int
-nvkm_device_tegra_new(struct platform_device *pdev,
+nvkm_device_tegra_new(const struct nvkm_device_tegra_func *func,
+		      struct platform_device *pdev,
 		      const char *cfg, const char *dbg,
 		      bool detect, bool mmio, u64 subdev_mask,
 		      struct nvkm_device **pdevice)
@@ -248,6 +253,7 @@ nvkm_device_tegra_new(struct platform_device *pdev,
 	if (!(tdev = kzalloc(sizeof(*tdev), GFP_KERNEL)))
 		return -ENOMEM;
 	*pdevice = &tdev->device;
+	tdev->func = func;
 	tdev->pdev = pdev;
 	tdev->irq = -1;
 
@@ -285,7 +291,8 @@ nvkm_device_tegra_new(struct platform_device *pdev,
 }
 #else
 int
-nvkm_device_tegra_new(struct platform_device *pdev,
+nvkm_device_tegra_new(const struct nvkm_device_tegra_func *func,
+		      struct platform_device *pdev,
 		      const char *cfg, const char *dbg,
 		      bool detect, bool mmio, u64 subdev_mask,
 		      struct nvkm_device **pdevice)
-- 
cgit v0.10.2


From 68b566534c1275facf64969cfd643fa6a724b351 Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Fri, 4 Sep 2015 19:59:32 +0900
Subject: drm/nouveau/instmem/gk20a: make use of the IOMMU bit

Use the IOMMU bit specified in platform data instead of hardcoding it to
the bit used by current Tegra GPUs.

Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/gk20a.c b/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/gk20a.c
index a2921ac..fc419bb 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/gk20a.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/gk20a.c
@@ -103,6 +103,7 @@ struct gk20a_instmem {
 	struct nvkm_mm *mm;
 	struct iommu_domain *domain;
 	unsigned long iommu_pgshift;
+	u16 iommu_bit;
 
 	/* Only used by DMA API */
 	struct dma_attrs attrs;
@@ -332,8 +333,8 @@ gk20a_instobj_dtor_iommu(struct nvkm_memory *memory)
 	r = list_first_entry(&node->base.mem.regions, struct nvkm_mm_node,
 			     rl_entry);
 
-	/* clear bit 34 to unmap pages */
-	r->offset &= ~BIT(34 - imem->iommu_pgshift);
+	/* clear IOMMU bit to unmap pages */
+	r->offset &= ~BIT(imem->iommu_bit - imem->iommu_pgshift);
 
 	/* Unmap pages from GPU address space and free them */
 	for (i = 0; i < node->base.mem.size; i++) {
@@ -489,8 +490,8 @@ gk20a_instobj_ctor_iommu(struct gk20a_instmem *imem, u32 npages, u32 align,
 		}
 	}
 
-	/* Bit 34 tells that an address is to be resolved through the IOMMU */
-	r->offset |= BIT(34 - imem->iommu_pgshift);
+	/* IOMMU bit tells that an address is to be resolved through the IOMMU */
+	r->offset |= BIT(imem->iommu_bit - imem->iommu_pgshift);
 
 	node->base.mem.offset = ((u64)r->offset) << imem->iommu_pgshift;
 
@@ -603,6 +604,7 @@ gk20a_instmem_new(struct nvkm_device *device, int index,
 		imem->domain = tdev->iommu.domain;
 		imem->iommu_pgshift = tdev->iommu.pgshift;
 		imem->cpu_map = gk20a_instobj_cpu_map_iommu;
+		imem->iommu_bit = tdev->func->iommu_bit;
 
 		nvkm_info(&imem->base.subdev, "using IOMMU\n");
 	} else {
-- 
cgit v0.10.2


From 524883bb48464ed76bd635819989284a249bf917 Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Fri, 4 Sep 2015 19:59:33 +0900
Subject: drm/nouveau/ttm: convert to DMA API

The pci_dma_* functions are now superseeded in the kernel by the DMA
API. Make the conversion to this more generic API.

Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/nouveau_ttm.c b/drivers/gpu/drm/nouveau/nouveau_ttm.c
index 3f0fb55..bd287c2 100644
--- a/drivers/gpu/drm/nouveau/nouveau_ttm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_ttm.c
@@ -338,7 +338,7 @@ nouveau_ttm_init(struct nouveau_drm *drm)
 	struct nvkm_device *device = nvxx_device(&drm->device);
 	struct nvkm_pci *pci = device->pci;
 	struct drm_device *dev = drm->dev;
-	u32 bits;
+	u8 bits;
 	int ret;
 
 	if (pci && pci->agp.bridge) {
@@ -351,18 +351,16 @@ nouveau_ttm_init(struct nouveau_drm *drm)
 	bits = nvxx_mmu(&drm->device)->dma_bits;
 	if (nvxx_device(&drm->device)->func->pci) {
 		if (drm->agp.bridge ||
-		     !pci_dma_supported(dev->pdev, DMA_BIT_MASK(bits)))
+		     !dma_supported(dev->dev, DMA_BIT_MASK(bits)))
 			bits = 32;
 
-		ret = pci_set_dma_mask(dev->pdev, DMA_BIT_MASK(bits));
+		ret = dma_set_mask(dev->dev, DMA_BIT_MASK(bits));
 		if (ret)
 			return ret;
 
-		ret = pci_set_consistent_dma_mask(dev->pdev,
-						  DMA_BIT_MASK(bits));
+		ret = dma_set_coherent_mask(dev->dev, DMA_BIT_MASK(bits));
 		if (ret)
-			pci_set_consistent_dma_mask(dev->pdev,
-						    DMA_BIT_MASK(32));
+			dma_set_coherent_mask(dev->dev, DMA_BIT_MASK(32));
 	}
 
 	ret = nouveau_ttm_global_init(drm);
-- 
cgit v0.10.2


From b31cf78b93243f8ff64297c1f77a4d030c32ca56 Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Fri, 4 Sep 2015 19:59:34 +0900
Subject: drm/nouveau/ttm: set the DMA mask for platform devices

So far the DMA mask was not set for platform devices, which limited them
to a 32-bit physical space. Allow dma_set_mask() to be called for
non-PCI devices, and also take the IOMMU bit into account since it could
restrict the physically addressable space.

Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/nouveau_ttm.c b/drivers/gpu/drm/nouveau/nouveau_ttm.c
index bd287c2..3f713c1 100644
--- a/drivers/gpu/drm/nouveau/nouveau_ttm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_ttm.c
@@ -29,6 +29,9 @@
 #include "nouveau_gem.h"
 
 #include "drm_legacy.h"
+
+#include <core/tegra.h>
+
 static int
 nouveau_vram_manager_init(struct ttm_mem_type_manager *man, unsigned long psize)
 {
@@ -353,16 +356,26 @@ nouveau_ttm_init(struct nouveau_drm *drm)
 		if (drm->agp.bridge ||
 		     !dma_supported(dev->dev, DMA_BIT_MASK(bits)))
 			bits = 32;
+	} else if (device->func->tegra) {
+		struct nvkm_device_tegra *tegra = device->func->tegra(device);
 
-		ret = dma_set_mask(dev->dev, DMA_BIT_MASK(bits));
-		if (ret)
-			return ret;
+		/*
+		 * If the platform can use a IOMMU, then the addressable DMA
+		 * space is constrained by the IOMMU bit
+		 */
+		if (tegra->func->iommu_bit)
+			bits = min(bits, tegra->func->iommu_bit);
 
-		ret = dma_set_coherent_mask(dev->dev, DMA_BIT_MASK(bits));
-		if (ret)
-			dma_set_coherent_mask(dev->dev, DMA_BIT_MASK(32));
 	}
 
+	ret = dma_set_mask(dev->dev, DMA_BIT_MASK(bits));
+	if (ret)
+		return ret;
+
+	ret = dma_set_coherent_mask(dev->dev, DMA_BIT_MASK(bits));
+	if (ret)
+		dma_set_coherent_mask(dev->dev, DMA_BIT_MASK(32));
+
 	ret = nouveau_ttm_global_init(drm);
 	if (ret)
 		return ret;
-- 
cgit v0.10.2


From 7bddeba9663962a5bd507e636c84361852314f04 Mon Sep 17 00:00:00 2001
From: Martin Peres <martin.peres@free.fr>
Date: Wed, 9 Sep 2015 04:05:50 +0300
Subject: drm/nouveau/bios/volt: add support for pwm-based volt management

Signed-off-by: Martin Peres <martin.peres@free.fr>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/volt.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/volt.h
index eb2de4b..b0df610 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/volt.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/volt.h
@@ -1,11 +1,24 @@
 #ifndef __NVBIOS_VOLT_H__
 #define __NVBIOS_VOLT_H__
+
+enum nvbios_volt_type {
+	NVBIOS_VOLT_GPIO = 0,
+	NVBIOS_VOLT_PWM,
+};
+
 struct nvbios_volt {
-	u8  vidmask;
+	enum nvbios_volt_type type;
 	u32 min;
 	u32 max;
 	u32 base;
+
+	/* GPIO mode */
+	u8  vidmask;
 	s16 step;
+
+	/* PWM mode */
+	u32 pwm_freq;
+	u32 pwm_range;
 };
 
 u16 nvbios_volt_table(struct nvkm_bios *, u8 *ver, u8 *hdr, u8 *cnt, u8 *len);
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/volt.c b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/volt.c
index 615804c..6e0a336 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/volt.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/volt.c
@@ -73,15 +73,19 @@ nvbios_volt_parse(struct nvkm_bios *bios, u8 *ver, u8 *hdr, u8 *cnt, u8 *len,
 	memset(info, 0x00, sizeof(*info));
 	switch (!!volt * *ver) {
 	case 0x12:
+		info->type    = NVBIOS_VOLT_GPIO;
 		info->vidmask = nvbios_rd08(bios, volt + 0x04);
 		break;
 	case 0x20:
+		info->type    = NVBIOS_VOLT_GPIO;
 		info->vidmask = nvbios_rd08(bios, volt + 0x05);
 		break;
 	case 0x30:
+		info->type    = NVBIOS_VOLT_GPIO;
 		info->vidmask = nvbios_rd08(bios, volt + 0x04);
 		break;
 	case 0x40:
+		info->type    = NVBIOS_VOLT_GPIO;
 		info->base    = nvbios_rd32(bios, volt + 0x04);
 		info->step    = nvbios_rd16(bios, volt + 0x08);
 		info->vidmask = nvbios_rd08(bios, volt + 0x0b);
@@ -90,11 +94,20 @@ nvbios_volt_parse(struct nvkm_bios *bios, u8 *ver, u8 *hdr, u8 *cnt, u8 *len,
 		info->max     = info->base;
 		break;
 	case 0x50:
-		info->vidmask = nvbios_rd08(bios, volt + 0x06);
 		info->min     = nvbios_rd32(bios, volt + 0x0a);
 		info->max     = nvbios_rd32(bios, volt + 0x0e);
 		info->base    = nvbios_rd32(bios, volt + 0x12) & 0x00ffffff;
-		info->step    = nvbios_rd16(bios, volt + 0x16);
+
+		/* offset 4 seems to be a flag byte */
+		if (nvbios_rd32(bios, volt + 0x4) & 1) {
+			info->type      = NVBIOS_VOLT_PWM;
+			info->pwm_freq  = nvbios_rd32(bios, volt + 0x5) / 1000;
+			info->pwm_range = nvbios_rd32(bios, volt + 0x16);
+		} else {
+			info->type      = NVBIOS_VOLT_GPIO;
+			info->vidmask   = nvbios_rd08(bios, volt + 0x06);
+			info->step      = nvbios_rd16(bios, volt + 0x16);
+		}
 		break;
 	}
 	return volt;
-- 
cgit v0.10.2


From 4c58a05b4a4709db7a1667d740f3ac2725fa3ce4 Mon Sep 17 00:00:00 2001
From: Martin Peres <martin.peres@free.fr>
Date: Wed, 9 Sep 2015 04:05:51 +0300
Subject: drm/nouveau/volt: add support for non-vid-based voltage controllers

This patch is not ideal but it definitely beats a rewrite of the current
interface and is very self-contained.

Signed-off-by: Martin Peres <martin.peres@free.fr>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/volt/base.c b/drivers/gpu/drm/nouveau/nvkm/subdev/volt/base.c
index 4752dbd..50b5649 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/volt/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/volt/base.c
@@ -30,7 +30,12 @@
 int
 nvkm_volt_get(struct nvkm_volt *volt)
 {
-	int ret = volt->func->vid_get(volt), i;
+	int ret, i;
+
+	if (volt->func->volt_get)
+		return volt->func->volt_get(volt);
+
+	ret = volt->func->vid_get(volt);
 	if (ret >= 0) {
 		for (i = 0; i < volt->vid_nr; i++) {
 			if (volt->vid[i].vid == ret)
@@ -46,6 +51,10 @@ nvkm_volt_set(struct nvkm_volt *volt, u32 uv)
 {
 	struct nvkm_subdev *subdev = &volt->subdev;
 	int i, ret = -EINVAL;
+
+	if (volt->func->volt_set)
+		return volt->func->volt_set(volt, uv);
+
 	for (i = 0; i < volt->vid_nr; i++) {
 		if (volt->vid[i].uv == uv) {
 			ret = volt->func->vid_set(volt, volt->vid[i].vid);
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/volt/priv.h b/drivers/gpu/drm/nouveau/nvkm/subdev/volt/priv.h
index 394f37c..cdb3d9f 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/volt/priv.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/volt/priv.h
@@ -9,6 +9,8 @@ int nvkm_volt_new_(const struct nvkm_volt_func *, struct nvkm_device *,
 		   int index, struct nvkm_volt **);
 
 struct nvkm_volt_func {
+	int (*volt_get)(struct nvkm_volt *);
+	int (*volt_set)(struct nvkm_volt *, u32 uv);
 	int (*vid_get)(struct nvkm_volt *);
 	int (*vid_set)(struct nvkm_volt *, u8 vid);
 	int (*set_id)(struct nvkm_volt *, u8 id, int condition);
-- 
cgit v0.10.2


From 1531dbbb56820d4e0fee4f9baa9fc84485bb6623 Mon Sep 17 00:00:00 2001
From: Martin Peres <martin.peres@free.fr>
Date: Wed, 9 Sep 2015 00:34:33 +0200
Subject: drm/nouveau/volt/gk104: add support for pwm and gpio modes

Most Keplers actually use the GPIO-based voltage management instead of the new
PWM-based one. Use the GPIO mode as a fallback as it already gracefully handles
the case where no GPIOs exist.

All the Maxwells seem to use the PWM method though.

v2:
 - Do not forget to commit the PWM configuration change!

Signed-off-by: Martin Peres <martin.peres@free.fr>

diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/gpio.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/gpio.h
index 33be260..a47d46dd 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/gpio.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/gpio.h
@@ -15,6 +15,7 @@ enum dcb_gpio_func_name {
 	DCB_GPIO_VID5 = 0x74,
 	DCB_GPIO_VID6 = 0x75,
 	DCB_GPIO_VID7 = 0x76,
+	DCB_GPIO_VID_PWM = 0x81,
 };
 
 #define DCB_GPIO_LOG_DIR     0x02
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/volt.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/volt.h
index 5c8a3f1..b458d04 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/volt.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/volt.h
@@ -18,5 +18,6 @@ int nvkm_volt_get(struct nvkm_volt *);
 int nvkm_volt_set_id(struct nvkm_volt *, u8 id, int condition);
 
 int nv40_volt_new(struct nvkm_device *, int, struct nvkm_volt **);
+int gk104_volt_new(struct nvkm_device *, int, struct nvkm_volt **);
 int gk20a_volt_new(struct nvkm_device *, int, struct nvkm_volt **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
index 94a906b..7c35f1f 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
@@ -1673,7 +1673,7 @@ nve4_chipset = {
 	.pmu = gk104_pmu_new,
 	.therm = gf119_therm_new,
 	.timer = nv41_timer_new,
-	.volt = nv40_volt_new,
+	.volt = gk104_volt_new,
 	.ce[0] = gk104_ce_new,
 	.ce[1] = gk104_ce_new,
 	.ce[2] = gk104_ce_new,
@@ -1710,7 +1710,7 @@ nve6_chipset = {
 	.pmu = gk104_pmu_new,
 	.therm = gf119_therm_new,
 	.timer = nv41_timer_new,
-	.volt = nv40_volt_new,
+	.volt = gk104_volt_new,
 	.ce[0] = gk104_ce_new,
 	.ce[1] = gk104_ce_new,
 	.ce[2] = gk104_ce_new,
@@ -1747,7 +1747,7 @@ nve7_chipset = {
 	.pmu = gf119_pmu_new,
 	.therm = gf119_therm_new,
 	.timer = nv41_timer_new,
-	.volt = nv40_volt_new,
+	.volt = gk104_volt_new,
 	.ce[0] = gk104_ce_new,
 	.ce[1] = gk104_ce_new,
 	.ce[2] = gk104_ce_new,
@@ -1808,7 +1808,7 @@ nvf0_chipset = {
 	.pmu = gk110_pmu_new,
 	.therm = gf119_therm_new,
 	.timer = nv41_timer_new,
-	.volt = nv40_volt_new,
+	.volt = gk104_volt_new,
 	.ce[0] = gk104_ce_new,
 	.ce[1] = gk104_ce_new,
 	.ce[2] = gk104_ce_new,
@@ -1844,7 +1844,7 @@ nvf1_chipset = {
 	.pmu = gk110_pmu_new,
 	.therm = gf119_therm_new,
 	.timer = nv41_timer_new,
-	.volt = nv40_volt_new,
+	.volt = gk104_volt_new,
 	.ce[0] = gk104_ce_new,
 	.ce[1] = gk104_ce_new,
 	.ce[2] = gk104_ce_new,
@@ -1880,7 +1880,7 @@ nv106_chipset = {
 	.pmu = gk208_pmu_new,
 	.therm = gf119_therm_new,
 	.timer = nv41_timer_new,
-	.volt = nv40_volt_new,
+	.volt = gk104_volt_new,
 	.ce[0] = gk104_ce_new,
 	.ce[1] = gk104_ce_new,
 	.ce[2] = gk104_ce_new,
@@ -1916,7 +1916,7 @@ nv108_chipset = {
 	.pmu = gk208_pmu_new,
 	.therm = gf119_therm_new,
 	.timer = nv41_timer_new,
-	.volt = nv40_volt_new,
+	.volt = gk104_volt_new,
 	.ce[0] = gk104_ce_new,
 	.ce[1] = gk104_ce_new,
 	.ce[2] = gk104_ce_new,
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/volt/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/volt/Kbuild
index 6b46ff4..b035c6e 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/volt/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/volt/Kbuild
@@ -1,4 +1,5 @@
 nvkm-y += nvkm/subdev/volt/base.o
 nvkm-y += nvkm/subdev/volt/gpio.o
 nvkm-y += nvkm/subdev/volt/nv40.o
+nvkm-y += nvkm/subdev/volt/gk104.o
 nvkm-y += nvkm/subdev/volt/gk20a.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/volt/gk104.c b/drivers/gpu/drm/nouveau/nvkm/subdev/volt/gk104.c
new file mode 100644
index 0000000..b61509e
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/volt/gk104.c
@@ -0,0 +1,119 @@
+/*
+ * Copyright 2015 Martin Peres
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Martin Peres
+ */
+#include "priv.h"
+
+#include <subdev/volt.h>
+#include <subdev/gpio.h>
+#include <subdev/bios.h>
+#include <subdev/bios/volt.h>
+
+#define gk104_volt(p) container_of((p), struct gk104_volt, base)
+struct gk104_volt {
+	struct nvkm_volt base;
+	struct nvbios_volt bios;
+};
+
+int
+gk104_volt_get(struct nvkm_volt *base)
+{
+	struct nvbios_volt *bios = &gk104_volt(base)->bios;
+	struct nvkm_device *device = base->subdev.device;
+	u32 div, duty;
+
+	div  = nvkm_rd32(device, 0x20340);
+	duty = nvkm_rd32(device, 0x20344);
+
+	return bios->base + bios->pwm_range * duty / div;
+}
+
+int
+gk104_volt_set(struct nvkm_volt *base, u32 uv)
+{
+	struct nvbios_volt *bios = &gk104_volt(base)->bios;
+	struct nvkm_device *device = base->subdev.device;
+	u32 div, duty;
+
+	/* the blob uses this crystal frequency, let's use it too. */
+	div = 27648000 / bios->pwm_freq;
+	duty = (uv - bios->base) * div / bios->pwm_range;
+
+	nvkm_wr32(device, 0x20340, div);
+	nvkm_wr32(device, 0x20344, 0x8000000 | duty);
+
+	return 0;
+}
+
+static const struct nvkm_volt_func
+gk104_volt_pwm = {
+	.volt_get = gk104_volt_get,
+	.volt_set = gk104_volt_set,
+}, gk104_volt_gpio = {
+	.vid_get = nvkm_voltgpio_get,
+	.vid_set = nvkm_voltgpio_set,
+};
+
+int
+gk104_volt_new(struct nvkm_device *device, int index, struct nvkm_volt **pvolt)
+{
+	const struct nvkm_volt_func *volt_func = &gk104_volt_gpio;
+	struct dcb_gpio_func gpio;
+	struct nvbios_volt bios;
+	struct gk104_volt *volt;
+	u8 ver, hdr, cnt, len;
+	const char *mode;
+
+	if (!nvbios_volt_parse(device->bios, &ver, &hdr, &cnt, &len, &bios))
+		return 0;
+
+	if (!nvkm_gpio_find(device->gpio, 0, DCB_GPIO_VID_PWM, 0xff, &gpio) &&
+	    bios.type == NVBIOS_VOLT_PWM) {
+		volt_func = &gk104_volt_pwm;
+	}
+
+	if (!(volt = kzalloc(sizeof(*volt), GFP_KERNEL)))
+		return -ENOMEM;
+	nvkm_volt_ctor(volt_func, device, index, &volt->base);
+	*pvolt = &volt->base;
+	volt->bios = bios;
+
+	/* now that we have a subdev, we can show an error if we found through
+	 * the voltage table that we were supposed to use the PWN mode but we
+	 * did not find the right GPIO for it.
+	 */
+	if (bios.type == NVBIOS_VOLT_PWM && volt_func != &gk104_volt_pwm) {
+		nvkm_error(&volt->base.subdev,
+			   "Type mismatch between the voltage table type and "
+			   "the GPIO table. Fallback to GPIO mode.\n");
+	}
+
+	if (volt_func == &gk104_volt_gpio) {
+		nvkm_voltgpio_init(&volt->base);
+		mode = "GPIO";
+	} else
+		mode = "PWM";
+
+	nvkm_debug(&volt->base.subdev, "Using %s mode\n", mode);
+
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/volt/priv.h b/drivers/gpu/drm/nouveau/nvkm/subdev/volt/priv.h
index cdb3d9f..d5140d9 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/volt/priv.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/volt/priv.h
@@ -19,4 +19,8 @@ struct nvkm_volt_func {
 int nvkm_voltgpio_init(struct nvkm_volt *);
 int nvkm_voltgpio_get(struct nvkm_volt *);
 int nvkm_voltgpio_set(struct nvkm_volt *, u8);
+
+int nvkm_voltpwm_init(struct nvkm_volt *volt);
+int nvkm_voltpwm_get(struct nvkm_volt *volt);
+int nvkm_voltpwm_set(struct nvkm_volt *volt, u32 uv);
 #endif
-- 
cgit v0.10.2


From dc47700f7d965ca0c6abeccc8cf467de8a1ce768 Mon Sep 17 00:00:00 2001
From: Martin Peres <martin.peres@free.fr>
Date: Wed, 9 Sep 2015 02:13:30 +0200
Subject: drm/nouveau/gm107: add voltage control using the new gk104 volt class

Let's ignore the other desktop Maxwells until I get my hands on one and confirm
that we still can change the voltage.

Signed-off-by: Martin Peres <martin.peres@free.fr>

diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
index 7c35f1f..b324fd1 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
@@ -1952,6 +1952,7 @@ nv117_chipset = {
 	.pmu = gm107_pmu_new,
 	.therm = gm107_therm_new,
 	.timer = gk20a_timer_new,
+	.volt = gk104_volt_new,
 	.ce[0] = gk104_ce_new,
 	.ce[2] = gk104_ce_new,
 	.disp = gm107_disp_new,
-- 
cgit v0.10.2


From 24580d1c310d5987f72adbb49bf3e976a5e5313b Mon Sep 17 00:00:00 2001
From: Martin Peres <martin.peres@free.fr>
Date: Wed, 16 Sep 2015 22:45:33 +0300
Subject: drm/nouveau/gm204/6: add voltage control using the new gk104 volt
 class

I got confirmation that we can read and change the voltage with the same code.
The divider is also computed correctly on the gm204 we got our hands on.

Thanks to Yoshimo on IRC for executing the tests on his gm204!

Signed-off-by: Martin Peres <martin.peres@free.fr>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
index b324fd1..8e9b5e8 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
@@ -1982,6 +1982,7 @@ nv124_chipset = {
 	.pci = nv40_pci_new,
 	.pmu = gm107_pmu_new,
 	.timer = gk20a_timer_new,
+	.volt = gk104_volt_new,
 	.ce[0] = gm204_ce_new,
 	.ce[1] = gm204_ce_new,
 	.ce[2] = gm204_ce_new,
@@ -2012,6 +2013,7 @@ nv126_chipset = {
 	.pci = nv40_pci_new,
 	.pmu = gm107_pmu_new,
 	.timer = gk20a_timer_new,
+	.volt = gk104_volt_new,
 	.ce[0] = gm204_ce_new,
 	.ce[1] = gm204_ce_new,
 	.ce[2] = gm204_ce_new,
-- 
cgit v0.10.2


From b6afa2650cb3ed4ec8ba0c2ccb29c1dc99e039d5 Mon Sep 17 00:00:00 2001
From: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Date: Thu, 24 Sep 2015 20:26:15 +0200
Subject: drm/nouveau/ibus/gf100: increase wait timeout to avoid read faults

Increase clock timeout of some unknown engines in order to avoid failure
at high gpcclk rate.

This fixes IBUS read faults on my GF119 when reclocking is manually
enabled. Note that memory reclocking is completely broken and NvMemExec
has to be disabled to allow core clock reclocking only.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/ibus.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/ibus.h
index 9d512cd5..c4dcd26 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/ibus.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/ibus.h
@@ -3,6 +3,7 @@
 #include <core/subdev.h>
 
 int gf100_ibus_new(struct nvkm_device *, int, struct nvkm_subdev **);
+int gf117_ibus_new(struct nvkm_device *, int, struct nvkm_subdev **);
 int gk104_ibus_new(struct nvkm_device *, int, struct nvkm_subdev **);
 int gk20a_ibus_new(struct nvkm_device *, int, struct nvkm_subdev **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
index 8e9b5e8..acc2fe9 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
@@ -1595,7 +1595,7 @@ nvd7_chipset = {
 	.fuse = gf100_fuse_new,
 	.gpio = gf119_gpio_new,
 	.i2c = gf117_i2c_new,
-	.ibus = gf100_ibus_new,
+	.ibus = gf117_ibus_new,
 	.imem = nv50_instmem_new,
 	.ltc = gf100_ltc_new,
 	.mc = gf100_mc_new,
@@ -1628,7 +1628,7 @@ nvd9_chipset = {
 	.fuse = gf100_fuse_new,
 	.gpio = gf119_gpio_new,
 	.i2c = gf119_i2c_new,
-	.ibus = gf100_ibus_new,
+	.ibus = gf117_ibus_new,
 	.imem = nv50_instmem_new,
 	.ltc = gf100_ltc_new,
 	.mc = gf100_mc_new,
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/Kbuild
index a0b12d2..de888fa 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/Kbuild
@@ -1,3 +1,4 @@
 nvkm-y += nvkm/subdev/ibus/gf100.o
+nvkm-y += nvkm/subdev/ibus/gf117.o
 nvkm-y += nvkm/subdev/ibus/gk104.o
 nvkm-y += nvkm/subdev/ibus/gk20a.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/gf100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/gf100.c
index 37a0496..72d6330 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/gf100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/gf100.c
@@ -21,7 +21,7 @@
  *
  * Authors: Ben Skeggs
  */
-#include <subdev/ibus.h>
+#include "priv.h"
 
 static void
 gf100_ibus_intr_hub(struct nvkm_subdev *ibus, int i)
@@ -56,7 +56,7 @@ gf100_ibus_intr_gpc(struct nvkm_subdev *ibus, int i)
 	nvkm_mask(device, 0x128128 + (i * 0x0400), 0x00000200, 0x00000000);
 }
 
-static void
+void
 gf100_ibus_intr(struct nvkm_subdev *ibus)
 {
 	struct nvkm_device *device = ibus->device;
@@ -92,8 +92,21 @@ gf100_ibus_intr(struct nvkm_subdev *ibus)
 	}
 }
 
+static int
+gf100_ibus_init(struct nvkm_subdev *ibus)
+{
+	struct nvkm_device *device = ibus->device;
+	nvkm_mask(device, 0x122310, 0x0003ffff, 0x00000800);
+	nvkm_wr32(device, 0x12232c, 0x00100064);
+	nvkm_wr32(device, 0x122330, 0x00100064);
+	nvkm_wr32(device, 0x122334, 0x00100064);
+	nvkm_mask(device, 0x122348, 0x0003ffff, 0x00000100);
+	return 0;
+}
+
 static const struct nvkm_subdev_func
 gf100_ibus = {
+	.init = gf100_ibus_init,
 	.intr = gf100_ibus_intr,
 };
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/gf117.c b/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/gf117.c
new file mode 100644
index 0000000..f69f263
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/gf117.c
@@ -0,0 +1,51 @@
+/*
+ * Copyright 2015 Samuel Pitosiet
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Samuel Pitoiset
+ */
+#include "priv.h"
+
+static int
+gf117_ibus_init(struct nvkm_subdev *ibus)
+{
+	struct nvkm_device *device = ibus->device;
+	nvkm_mask(device, 0x122310, 0x0003ffff, 0x00000800);
+	nvkm_mask(device, 0x122348, 0x0003ffff, 0x00000100);
+	nvkm_mask(device, 0x1223b0, 0x0003ffff, 0x00000fff);
+	return 0;
+}
+
+static const struct nvkm_subdev_func
+gf117_ibus = {
+	.init = gf117_ibus_init,
+	.intr = gf100_ibus_intr,
+};
+
+int
+gf117_ibus_new(struct nvkm_device *device, int index,
+	       struct nvkm_subdev **pibus)
+{
+	struct nvkm_subdev *ibus;
+	if (!(ibus = *pibus = kzalloc(sizeof(*ibus), GFP_KERNEL)))
+		return -ENOMEM;
+	nvkm_subdev_ctor(&gf117_ibus, device, index, 0, ibus);
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/priv.h b/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/priv.h
new file mode 100644
index 0000000..48e1b63
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/priv.h
@@ -0,0 +1,7 @@
+#ifndef __NVKM_IBUS_PRIV_H__
+#define __NVKM_IBUS_PRIV_H__
+
+#include <subdev/ibus.h>
+
+void gf100_ibus_intr(struct nvkm_subdev *);
+#endif
-- 
cgit v0.10.2


From 3e55b53bc7039f1a6b051aa1a5ba15eec6dce2e7 Mon Sep 17 00:00:00 2001
From: Ben Skeggs <bskeggs@redhat.com>
Date: Thu, 1 Oct 2015 09:29:58 +1000
Subject: drm/nouveau/pci/g84: split implementation from nv50

An upcoming patch will implement functionality that we don't use on the
original NV50.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/pci.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/pci.h
index 5b3c054..b5b004b 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/pci.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/pci.h
@@ -30,5 +30,6 @@ int nv04_pci_new(struct nvkm_device *, int, struct nvkm_pci **);
 int nv40_pci_new(struct nvkm_device *, int, struct nvkm_pci **);
 int nv4c_pci_new(struct nvkm_device *, int, struct nvkm_pci **);
 int nv50_pci_new(struct nvkm_device *, int, struct nvkm_pci **);
+int g84_pci_new(struct nvkm_device *, int, struct nvkm_pci **);
 int gf100_pci_new(struct nvkm_device *, int, struct nvkm_pci **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
index acc2fe9..13774c1 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
@@ -929,7 +929,7 @@ nv84_chipset = {
 	.mc = nv50_mc_new,
 	.mmu = nv50_mmu_new,
 	.mxm = nv50_mxm_new,
-	.pci = nv50_pci_new,
+	.pci = g84_pci_new,
 	.therm = g84_therm_new,
 	.timer = nv41_timer_new,
 	.volt = nv40_volt_new,
@@ -961,7 +961,7 @@ nv86_chipset = {
 	.mc = nv50_mc_new,
 	.mmu = nv50_mmu_new,
 	.mxm = nv50_mxm_new,
-	.pci = nv50_pci_new,
+	.pci = g84_pci_new,
 	.therm = g84_therm_new,
 	.timer = nv41_timer_new,
 	.volt = nv40_volt_new,
@@ -993,7 +993,7 @@ nv92_chipset = {
 	.mc = nv50_mc_new,
 	.mmu = nv50_mmu_new,
 	.mxm = nv50_mxm_new,
-	.pci = nv50_pci_new,
+	.pci = g84_pci_new,
 	.therm = g84_therm_new,
 	.timer = nv41_timer_new,
 	.volt = nv40_volt_new,
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/Kbuild
index 99672c3..57652f5 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/Kbuild
@@ -4,4 +4,5 @@ nvkm-y += nvkm/subdev/pci/nv04.o
 nvkm-y += nvkm/subdev/pci/nv40.o
 nvkm-y += nvkm/subdev/pci/nv4c.o
 nvkm-y += nvkm/subdev/pci/nv50.o
+nvkm-y += nvkm/subdev/pci/g84.o
 nvkm-y += nvkm/subdev/pci/gf100.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/g84.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/g84.c
new file mode 100644
index 0000000..30be277
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/g84.c
@@ -0,0 +1,40 @@
+/*
+ * Copyright 2015 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs <bskeggs@redhat.com>
+ */
+#include "priv.h"
+
+#include <core/pci.h>
+
+static const struct nvkm_pci_func
+g84_pci_func = {
+	.rd32 = nv40_pci_rd32,
+	.wr08 = nv40_pci_wr08,
+	.wr32 = nv40_pci_wr32,
+	.msi_rearm = nv50_pci_msi_rearm,
+};
+
+int
+g84_pci_new(struct nvkm_device *device, int index, struct nvkm_pci **ppci)
+{
+	return nvkm_pci_new_(&g84_pci_func, device, index, ppci);
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/nv50.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/nv50.c
index 3e167d4..0274815 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/nv50.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/nv50.c
@@ -25,10 +25,10 @@
 
 #include <core/pci.h>
 
-/* MSI re-arm through the PRI appears to be broken on the original G80,
+/* MSI re-arm through the PRI appears to be broken on NV50/G84/G86/G92,
  * so we access it via alternate PCI config space mechanisms.
  */
-static void
+void
 nv50_pci_msi_rearm(struct nvkm_pci *pci)
 {
 	struct nvkm_device *device = pci->subdev.device;
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/priv.h b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/priv.h
index d22c2c1..d7622aa 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/priv.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/priv.h
@@ -16,4 +16,6 @@ struct nvkm_pci_func {
 u32 nv40_pci_rd32(struct nvkm_pci *, u16);
 void nv40_pci_wr08(struct nvkm_pci *, u16, u8);
 void nv40_pci_wr32(struct nvkm_pci *, u16, u32);
+
+void nv50_pci_msi_rearm(struct nvkm_pci *);
 #endif
-- 
cgit v0.10.2


From b31505c472f2451c6143bf4727e79ea177a35ca5 Mon Sep 17 00:00:00 2001
From: Ben Skeggs <bskeggs@redhat.com>
Date: Thu, 1 Oct 2015 09:34:45 +1000
Subject: drm/nouveau/pci/g94: split implementation from nv40

An upcoming patch will implement functionality that we don't use on any
NV40 chipset.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/pci.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/pci.h
index b5b004b..9c0b5e1 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/pci.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/pci.h
@@ -31,5 +31,6 @@ int nv40_pci_new(struct nvkm_device *, int, struct nvkm_pci **);
 int nv4c_pci_new(struct nvkm_device *, int, struct nvkm_pci **);
 int nv50_pci_new(struct nvkm_device *, int, struct nvkm_pci **);
 int g84_pci_new(struct nvkm_device *, int, struct nvkm_pci **);
+int g94_pci_new(struct nvkm_device *, int, struct nvkm_pci **);
 int gf100_pci_new(struct nvkm_device *, int, struct nvkm_pci **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
index 13774c1..727e18e 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
@@ -1025,7 +1025,7 @@ nv94_chipset = {
 	.mc = nv50_mc_new,
 	.mmu = nv50_mmu_new,
 	.mxm = nv50_mxm_new,
-	.pci = nv40_pci_new,
+	.pci = g94_pci_new,
 	.therm = g84_therm_new,
 	.timer = nv41_timer_new,
 	.volt = nv40_volt_new,
@@ -1057,7 +1057,7 @@ nv96_chipset = {
 	.mc = nv50_mc_new,
 	.mmu = nv50_mmu_new,
 	.mxm = nv50_mxm_new,
-	.pci = nv40_pci_new,
+	.pci = g94_pci_new,
 	.therm = g84_therm_new,
 	.timer = nv41_timer_new,
 	.volt = nv40_volt_new,
@@ -1089,7 +1089,7 @@ nv98_chipset = {
 	.mc = g98_mc_new,
 	.mmu = nv50_mmu_new,
 	.mxm = nv50_mxm_new,
-	.pci = nv40_pci_new,
+	.pci = g94_pci_new,
 	.therm = g84_therm_new,
 	.timer = nv41_timer_new,
 	.volt = nv40_volt_new,
@@ -1121,7 +1121,7 @@ nva0_chipset = {
 	.mc = g98_mc_new,
 	.mmu = nv50_mmu_new,
 	.mxm = nv50_mxm_new,
-	.pci = nv40_pci_new,
+	.pci = g94_pci_new,
 	.therm = g84_therm_new,
 	.timer = nv41_timer_new,
 	.volt = nv40_volt_new,
@@ -1153,7 +1153,7 @@ nva3_chipset = {
 	.mc = g98_mc_new,
 	.mmu = nv50_mmu_new,
 	.mxm = nv50_mxm_new,
-	.pci = nv40_pci_new,
+	.pci = g94_pci_new,
 	.pmu = gt215_pmu_new,
 	.therm = gt215_therm_new,
 	.timer = nv41_timer_new,
@@ -1187,7 +1187,7 @@ nva5_chipset = {
 	.mc = g98_mc_new,
 	.mmu = nv50_mmu_new,
 	.mxm = nv50_mxm_new,
-	.pci = nv40_pci_new,
+	.pci = g94_pci_new,
 	.pmu = gt215_pmu_new,
 	.therm = gt215_therm_new,
 	.timer = nv41_timer_new,
@@ -1220,7 +1220,7 @@ nva8_chipset = {
 	.mc = g98_mc_new,
 	.mmu = nv50_mmu_new,
 	.mxm = nv50_mxm_new,
-	.pci = nv40_pci_new,
+	.pci = g94_pci_new,
 	.pmu = gt215_pmu_new,
 	.therm = gt215_therm_new,
 	.timer = nv41_timer_new,
@@ -1253,7 +1253,7 @@ nvaa_chipset = {
 	.mc = g98_mc_new,
 	.mmu = nv50_mmu_new,
 	.mxm = nv50_mxm_new,
-	.pci = nv40_pci_new,
+	.pci = g94_pci_new,
 	.therm = g84_therm_new,
 	.timer = nv41_timer_new,
 	.volt = nv40_volt_new,
@@ -1285,7 +1285,7 @@ nvac_chipset = {
 	.mc = g98_mc_new,
 	.mmu = nv50_mmu_new,
 	.mxm = nv50_mxm_new,
-	.pci = nv40_pci_new,
+	.pci = g94_pci_new,
 	.therm = g84_therm_new,
 	.timer = nv41_timer_new,
 	.volt = nv40_volt_new,
@@ -1317,7 +1317,7 @@ nvaf_chipset = {
 	.mc = g98_mc_new,
 	.mmu = nv50_mmu_new,
 	.mxm = nv50_mxm_new,
-	.pci = nv40_pci_new,
+	.pci = g94_pci_new,
 	.pmu = gt215_pmu_new,
 	.therm = gt215_therm_new,
 	.timer = nv41_timer_new,
@@ -1388,7 +1388,7 @@ nvc1_chipset = {
 	.mc = gf100_mc_new,
 	.mmu = gf100_mmu_new,
 	.mxm = nv50_mxm_new,
-	.pci = nv40_pci_new,
+	.pci = g94_pci_new,
 	.pmu = gf100_pmu_new,
 	.therm = gt215_therm_new,
 	.timer = nv41_timer_new,
@@ -1423,7 +1423,7 @@ nvc3_chipset = {
 	.mc = gf100_mc_new,
 	.mmu = gf100_mmu_new,
 	.mxm = nv50_mxm_new,
-	.pci = nv40_pci_new,
+	.pci = g94_pci_new,
 	.pmu = gf100_pmu_new,
 	.therm = gt215_therm_new,
 	.timer = nv41_timer_new,
@@ -1566,7 +1566,7 @@ nvcf_chipset = {
 	.mc = gf100_mc_new,
 	.mmu = gf100_mmu_new,
 	.mxm = nv50_mxm_new,
-	.pci = nv40_pci_new,
+	.pci = g94_pci_new,
 	.pmu = gf100_pmu_new,
 	.therm = gt215_therm_new,
 	.timer = nv41_timer_new,
@@ -1601,7 +1601,7 @@ nvd7_chipset = {
 	.mc = gf100_mc_new,
 	.mmu = gf100_mmu_new,
 	.mxm = nv50_mxm_new,
-	.pci = nv40_pci_new,
+	.pci = g94_pci_new,
 	.therm = gf119_therm_new,
 	.timer = nv41_timer_new,
 	.ce[0] = gf100_ce_new,
@@ -1634,7 +1634,7 @@ nvd9_chipset = {
 	.mc = gf100_mc_new,
 	.mmu = gf100_mmu_new,
 	.mxm = nv50_mxm_new,
-	.pci = nv40_pci_new,
+	.pci = g94_pci_new,
 	.pmu = gf119_pmu_new,
 	.therm = gf119_therm_new,
 	.timer = nv41_timer_new,
@@ -1669,7 +1669,7 @@ nve4_chipset = {
 	.mc = gf100_mc_new,
 	.mmu = gf100_mmu_new,
 	.mxm = nv50_mxm_new,
-	.pci = nv40_pci_new,
+	.pci = g94_pci_new,
 	.pmu = gk104_pmu_new,
 	.therm = gf119_therm_new,
 	.timer = nv41_timer_new,
@@ -1706,7 +1706,7 @@ nve6_chipset = {
 	.mc = gf100_mc_new,
 	.mmu = gf100_mmu_new,
 	.mxm = nv50_mxm_new,
-	.pci = nv40_pci_new,
+	.pci = g94_pci_new,
 	.pmu = gk104_pmu_new,
 	.therm = gf119_therm_new,
 	.timer = nv41_timer_new,
@@ -1743,7 +1743,7 @@ nve7_chipset = {
 	.mc = gf100_mc_new,
 	.mmu = gf100_mmu_new,
 	.mxm = nv50_mxm_new,
-	.pci = nv40_pci_new,
+	.pci = g94_pci_new,
 	.pmu = gf119_pmu_new,
 	.therm = gf119_therm_new,
 	.timer = nv41_timer_new,
@@ -1804,7 +1804,7 @@ nvf0_chipset = {
 	.mc = gf100_mc_new,
 	.mmu = gf100_mmu_new,
 	.mxm = nv50_mxm_new,
-	.pci = nv40_pci_new,
+	.pci = g94_pci_new,
 	.pmu = gk110_pmu_new,
 	.therm = gf119_therm_new,
 	.timer = nv41_timer_new,
@@ -1840,7 +1840,7 @@ nvf1_chipset = {
 	.mc = gf100_mc_new,
 	.mmu = gf100_mmu_new,
 	.mxm = nv50_mxm_new,
-	.pci = nv40_pci_new,
+	.pci = g94_pci_new,
 	.pmu = gk110_pmu_new,
 	.therm = gf119_therm_new,
 	.timer = nv41_timer_new,
@@ -1876,7 +1876,7 @@ nv106_chipset = {
 	.mc = gk20a_mc_new,
 	.mmu = gf100_mmu_new,
 	.mxm = nv50_mxm_new,
-	.pci = nv40_pci_new,
+	.pci = g94_pci_new,
 	.pmu = gk208_pmu_new,
 	.therm = gf119_therm_new,
 	.timer = nv41_timer_new,
@@ -1912,7 +1912,7 @@ nv108_chipset = {
 	.mc = gk20a_mc_new,
 	.mmu = gf100_mmu_new,
 	.mxm = nv50_mxm_new,
-	.pci = nv40_pci_new,
+	.pci = g94_pci_new,
 	.pmu = gk208_pmu_new,
 	.therm = gf119_therm_new,
 	.timer = nv41_timer_new,
@@ -1948,7 +1948,7 @@ nv117_chipset = {
 	.mc = gk20a_mc_new,
 	.mmu = gf100_mmu_new,
 	.mxm = nv50_mxm_new,
-	.pci = nv40_pci_new,
+	.pci = g94_pci_new,
 	.pmu = gm107_pmu_new,
 	.therm = gm107_therm_new,
 	.timer = gk20a_timer_new,
@@ -1979,7 +1979,7 @@ nv124_chipset = {
 	.mc = gk20a_mc_new,
 	.mmu = gf100_mmu_new,
 	.mxm = nv50_mxm_new,
-	.pci = nv40_pci_new,
+	.pci = g94_pci_new,
 	.pmu = gm107_pmu_new,
 	.timer = gk20a_timer_new,
 	.volt = gk104_volt_new,
@@ -2010,7 +2010,7 @@ nv126_chipset = {
 	.mc = gk20a_mc_new,
 	.mmu = gf100_mmu_new,
 	.mxm = nv50_mxm_new,
-	.pci = nv40_pci_new,
+	.pci = g94_pci_new,
 	.pmu = gm107_pmu_new,
 	.timer = gk20a_timer_new,
 	.volt = gk104_volt_new,
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/Kbuild
index 57652f5..85cb525 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/Kbuild
@@ -5,4 +5,5 @@ nvkm-y += nvkm/subdev/pci/nv40.o
 nvkm-y += nvkm/subdev/pci/nv4c.o
 nvkm-y += nvkm/subdev/pci/nv50.o
 nvkm-y += nvkm/subdev/pci/g84.o
+nvkm-y += nvkm/subdev/pci/g94.o
 nvkm-y += nvkm/subdev/pci/gf100.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/g94.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/g94.c
new file mode 100644
index 0000000..1714421
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/g94.c
@@ -0,0 +1,38 @@
+/*
+ * Copyright 2015 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs <bskeggs@redhat.com>
+ */
+#include "priv.h"
+
+static const struct nvkm_pci_func
+g94_pci_func = {
+	.rd32 = nv40_pci_rd32,
+	.wr08 = nv40_pci_wr08,
+	.wr32 = nv40_pci_wr32,
+	.msi_rearm = nv40_pci_msi_rearm,
+};
+
+int
+g94_pci_new(struct nvkm_device *device, int index, struct nvkm_pci **ppci)
+{
+	return nvkm_pci_new_(&g94_pci_func, device, index, ppci);
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/nv40.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/nv40.c
index 090a187..6eb4177 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/nv40.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/nv40.c
@@ -44,7 +44,7 @@ nv40_pci_wr32(struct nvkm_pci *pci, u16 addr, u32 data)
 	nvkm_wr32(device, 0x088000 + addr, data);
 }
 
-static void
+void
 nv40_pci_msi_rearm(struct nvkm_pci *pci)
 {
 	nvkm_pci_wr08(pci, 0x0068, 0xff);
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/priv.h b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/priv.h
index d7622aa..68ce8e5 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/priv.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/priv.h
@@ -16,6 +16,7 @@ struct nvkm_pci_func {
 u32 nv40_pci_rd32(struct nvkm_pci *, u16);
 void nv40_pci_wr08(struct nvkm_pci *, u16, u8);
 void nv40_pci_wr32(struct nvkm_pci *, u16, u32);
+void nv40_pci_msi_rearm(struct nvkm_pci *);
 
 void nv50_pci_msi_rearm(struct nvkm_pci *);
 #endif
-- 
cgit v0.10.2


From c4266a9c7b0ee66b3d1ca22745d8eb9472e21b4b Mon Sep 17 00:00:00 2001
From: Ben Skeggs <bskeggs@redhat.com>
Date: Thu, 1 Oct 2015 11:36:58 +1000
Subject: drm/nouveau/pci/nv46: attempt to fix msi, and re-enable by default

Was not able to obtain a trace of NVRM due to kernel version annoyances,
however, experimentally confirmed that the WAR we use on NV50/G8x boards
works here too.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/pci.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/pci.h
index 9c0b5e1..39ca88f 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/pci.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/pci.h
@@ -28,8 +28,8 @@ void nvkm_pci_rom_shadow(struct nvkm_pci *, bool shadow);
 
 int nv04_pci_new(struct nvkm_device *, int, struct nvkm_pci **);
 int nv40_pci_new(struct nvkm_device *, int, struct nvkm_pci **);
+int nv46_pci_new(struct nvkm_device *, int, struct nvkm_pci **);
 int nv4c_pci_new(struct nvkm_device *, int, struct nvkm_pci **);
-int nv50_pci_new(struct nvkm_device *, int, struct nvkm_pci **);
 int g84_pci_new(struct nvkm_device *, int, struct nvkm_pci **);
 int g94_pci_new(struct nvkm_device *, int, struct nvkm_pci **);
 int gf100_pci_new(struct nvkm_device *, int, struct nvkm_pci **);
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
index 727e18e..2cff79c 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
@@ -637,7 +637,7 @@ nv46_chipset = {
 	.imem = nv40_instmem_new,
 	.mc = nv44_mc_new,
 	.mmu = nv44_mmu_new,
-	.pci = nv4c_pci_new,
+	.pci = nv46_pci_new,
 	.therm = nv40_therm_new,
 	.timer = nv41_timer_new,
 	.volt = nv40_volt_new,
@@ -822,7 +822,7 @@ nv50_chipset = {
 	.mc = nv50_mc_new,
 	.mmu = nv50_mmu_new,
 	.mxm = nv50_mxm_new,
-	.pci = nv50_pci_new,
+	.pci = nv46_pci_new,
 	.therm = nv50_therm_new,
 	.timer = nv41_timer_new,
 	.volt = nv40_volt_new,
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/Kbuild
index 85cb525..4476ef7 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/Kbuild
@@ -2,8 +2,8 @@ nvkm-y += nvkm/subdev/pci/agp.o
 nvkm-y += nvkm/subdev/pci/base.o
 nvkm-y += nvkm/subdev/pci/nv04.o
 nvkm-y += nvkm/subdev/pci/nv40.o
+nvkm-y += nvkm/subdev/pci/nv46.o
 nvkm-y += nvkm/subdev/pci/nv4c.o
-nvkm-y += nvkm/subdev/pci/nv50.o
 nvkm-y += nvkm/subdev/pci/g84.o
 nvkm-y += nvkm/subdev/pci/g94.o
 nvkm-y += nvkm/subdev/pci/gf100.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/g84.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/g84.c
index 30be277..8f3b001 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/g84.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/g84.c
@@ -30,7 +30,7 @@ g84_pci_func = {
 	.rd32 = nv40_pci_rd32,
 	.wr08 = nv40_pci_wr08,
 	.wr32 = nv40_pci_wr32,
-	.msi_rearm = nv50_pci_msi_rearm,
+	.msi_rearm = nv46_pci_msi_rearm,
 };
 
 int
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/nv46.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/nv46.c
new file mode 100644
index 0000000..fc617e4
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/nv46.c
@@ -0,0 +1,51 @@
+/*
+ * Copyright 2015 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs <bskeggs@redhat.com>
+ */
+#include "priv.h"
+
+#include <core/pci.h>
+
+/* MSI re-arm through the PRI appears to be broken on NV46/NV50/G84/G86/G92,
+ * so we access it via alternate PCI config space mechanisms.
+ */
+void
+nv46_pci_msi_rearm(struct nvkm_pci *pci)
+{
+	struct nvkm_device *device = pci->subdev.device;
+	struct pci_dev *pdev = device->func->pci(device)->pdev;
+	pci_write_config_byte(pdev, 0x68, 0xff);
+}
+
+static const struct nvkm_pci_func
+nv46_pci_func = {
+	.rd32 = nv40_pci_rd32,
+	.wr08 = nv40_pci_wr08,
+	.wr32 = nv40_pci_wr32,
+	.msi_rearm = nv46_pci_msi_rearm,
+};
+
+int
+nv46_pci_new(struct nvkm_device *device, int index, struct nvkm_pci **ppci)
+{
+	return nvkm_pci_new_(&nv46_pci_func, device, index, ppci);
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/nv50.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/nv50.c
deleted file mode 100644
index 0274815..0000000
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/nv50.c
+++ /dev/null
@@ -1,51 +0,0 @@
-/*
- * Copyright 2015 Red Hat Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
- * OTHER DEALINGS IN THE SOFTWARE.
- *
- * Authors: Ben Skeggs <bskeggs@redhat.com>
- */
-#include "priv.h"
-
-#include <core/pci.h>
-
-/* MSI re-arm through the PRI appears to be broken on NV50/G84/G86/G92,
- * so we access it via alternate PCI config space mechanisms.
- */
-void
-nv50_pci_msi_rearm(struct nvkm_pci *pci)
-{
-	struct nvkm_device *device = pci->subdev.device;
-	struct pci_dev *pdev = device->func->pci(device)->pdev;
-	pci_write_config_byte(pdev, 0x68, 0xff);
-}
-
-static const struct nvkm_pci_func
-nv50_pci_func = {
-	.rd32 = nv40_pci_rd32,
-	.wr08 = nv40_pci_wr08,
-	.wr32 = nv40_pci_wr32,
-	.msi_rearm = nv50_pci_msi_rearm,
-};
-
-int
-nv50_pci_new(struct nvkm_device *device, int index, struct nvkm_pci **ppci)
-{
-	return nvkm_pci_new_(&nv50_pci_func, device, index, ppci);
-}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/priv.h b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/priv.h
index 68ce8e5..473b8ab 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/priv.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/priv.h
@@ -18,5 +18,5 @@ void nv40_pci_wr08(struct nvkm_pci *, u16, u8);
 void nv40_pci_wr32(struct nvkm_pci *, u16, u32);
 void nv40_pci_msi_rearm(struct nvkm_pci *);
 
-void nv50_pci_msi_rearm(struct nvkm_pci *);
+void nv46_pci_msi_rearm(struct nvkm_pci *);
 #endif
-- 
cgit v0.10.2


From 779d16aacc192bc70232304ffc6b86dcedad1b45 Mon Sep 17 00:00:00 2001
From: Ben Skeggs <bskeggs@redhat.com>
Date: Thu, 1 Oct 2015 09:42:54 +1000
Subject: drm/nouveau/pci: prepare for chipset-specific initialisation tasks

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/base.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/base.c
index d1c148e..2110622 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/base.c
@@ -111,6 +111,9 @@ nvkm_pci_init(struct nvkm_subdev *subdev)
 			return ret;
 	}
 
+	if (pci->func->init)
+		pci->func->init(pci);
+
 	ret = request_irq(pdev->irq, nvkm_pci_intr, IRQF_SHARED, "nvkm", pci);
 	if (ret)
 		return ret;
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/priv.h b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/priv.h
index 473b8ab..1acd4bc 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/priv.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/priv.h
@@ -7,6 +7,7 @@ int nvkm_pci_new_(const struct nvkm_pci_func *, struct nvkm_device *,
 		  int index, struct nvkm_pci **);
 
 struct nvkm_pci_func {
+	void (*init)(struct nvkm_pci *);
 	u32 (*rd32)(struct nvkm_pci *, u16 addr);
 	void (*wr08)(struct nvkm_pci *, u16 addr, u8 data);
 	void (*wr32)(struct nvkm_pci *, u16 addr, u32 data);
-- 
cgit v0.10.2


From 560f989fe4260f0c729d3fc8a8691a3a02815f25 Mon Sep 17 00:00:00 2001
From: Ben Skeggs <bskeggs@redhat.com>
Date: Thu, 1 Oct 2015 14:58:04 +1000
Subject: drm/nouveau/pmu/gk104: check fuse to determine presence of PGOB

Not 100% confirmed, but seems to match from the few boards I've looked
at so far.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gk104.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gk104.c
index e33f5c0..d942fa7 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gk104.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/gk104.c
@@ -27,6 +27,7 @@
 #include "fuc/gf119.fuc4.h"
 
 #include <core/option.h>
+#include <subdev/fuse.h>
 #include <subdev/timer.h>
 
 static void
@@ -57,6 +58,9 @@ gk104_pmu_pgob(struct nvkm_pmu *pmu, bool enable)
 {
 	struct nvkm_device *device = pmu->subdev.device;
 
+	if (!(nvkm_fuse_read(device->fuse, 0x31c) & 0x00000001))
+		return;
+
 	nvkm_mask(device, 0x000200, 0x00001000, 0x00000000);
 	nvkm_rd32(device, 0x000200);
 	nvkm_mask(device, 0x000200, 0x08000000, 0x08000000);
-- 
cgit v0.10.2


From 3c9aca318150ba1152e957a37473ff67d8ebba30 Mon Sep 17 00:00:00 2001
From: Ben Skeggs <bskeggs@redhat.com>
Date: Thu, 1 Oct 2015 15:00:23 +1000
Subject: drm/nouveau/pmu/gk107: enable PGOB codepaths

Reported to be needed as per fdo#70354 comment #61.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
index 2cff79c..bbc9824 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
@@ -1744,7 +1744,7 @@ nve7_chipset = {
 	.mmu = gf100_mmu_new,
 	.mxm = nv50_mxm_new,
 	.pci = g94_pci_new,
-	.pmu = gf119_pmu_new,
+	.pmu = gk104_pmu_new,
 	.therm = gf119_therm_new,
 	.timer = nv41_timer_new,
 	.volt = gk104_volt_new,
-- 
cgit v0.10.2


From 4458c5639ba827d494ab06a81d1bb129ca5aea91 Mon Sep 17 00:00:00 2001
From: Sudip Mukherjee <sudipm.mukherjee@gmail.com>
Date: Wed, 2 Sep 2015 12:08:08 +0530
Subject: drm/nouveau: remove unused function

coverity.com reported that memset was using a buffer of size 0, on
checking the code it turned out that the function was not being used. So
remove it.

Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/pmu.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/pmu.h
index d606875..3a643df 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/pmu.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/pmu.h
@@ -4,8 +4,6 @@ struct nvbios_pmuT {
 };
 
 u32 nvbios_pmuTe(struct nvkm_bios *, u8 *ver, u8 *hdr, u8 *cnt, u8 *len);
-u32 nvbios_pmuTp(struct nvkm_bios *, u8 *ver, u8 *hdr, u8 *cnt, u8 *len,
-		 struct nvbios_pmuT *);
 
 struct nvbios_pmuE {
 	u8  type;
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/pmu.c b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/pmu.c
index 441ec45..c268e5a 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/pmu.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/pmu.c
@@ -62,19 +62,6 @@ nvbios_pmuTe(struct nvkm_bios *bios, u8 *ver, u8 *hdr, u8 *cnt, u8 *len)
 }
 
 u32
-nvbios_pmuTp(struct nvkm_bios *bios, u8 *ver, u8 *hdr, u8 *cnt, u8 *len,
-	     struct nvbios_pmuT *info)
-{
-	u32 data = nvbios_pmuTe(bios, ver, hdr, cnt, len);
-	memset(info, 0x00, sizeof(*info));
-	switch (!!data * *ver) {
-	default:
-		break;
-	}
-	return data;
-}
-
-u32
 nvbios_pmuEe(struct nvkm_bios *bios, int idx, u8 *ver, u8 *hdr)
 {
 	u8  cnt, len;
-- 
cgit v0.10.2


From bad4274a69328256eb3eaad4baaefd1b33491f7e Mon Sep 17 00:00:00 2001
From: Sudip Mukherjee <sudipm.mukherjee@gmail.com>
Date: Fri, 11 Sep 2015 15:00:56 +0530
Subject: drm/nouveau: fix memory leak

If pm_runtime_get_sync() we were going to "out" but we missed freeing
vma.

Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/nouveau_gem.c b/drivers/gpu/drm/nouveau/nouveau_gem.c
index 6bf1a78..a108cc3 100644
--- a/drivers/gpu/drm/nouveau/nouveau_gem.c
+++ b/drivers/gpu/drm/nouveau/nouveau_gem.c
@@ -84,8 +84,10 @@ nouveau_gem_object_open(struct drm_gem_object *gem, struct drm_file *file_priv)
 		}
 
 		ret = pm_runtime_get_sync(dev);
-		if (ret < 0 && ret != -EACCES)
+		if (ret < 0 && ret != -EACCES) {
+			kfree(vma);
 			goto out;
+		}
 
 		ret = nouveau_bo_vma_add(nvbo, cli->vm, vma);
 		if (ret)
-- 
cgit v0.10.2


From 3988f645f053a6889d00324dac3e57bd62cb8900 Mon Sep 17 00:00:00 2001
From: Ilia Mirkin <imirkin@alum.mit.edu>
Date: Wed, 7 Oct 2015 18:39:32 -0400
Subject: drm/nouveau/gr: document mp error 0x10

NVIDIA provided the documentation for mp error 0x10, INVALID_ADDR_SPACE,
which apparently happens when trying to use an atomic operation on
local or shared memory (instead of global memory).

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c
index f1358a5..dda7a7d 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c
@@ -882,6 +882,7 @@ static const struct nvkm_enum gf100_mp_warp_error[] = {
 	{ 0x0d, "GPR_OUT_OF_BOUNDS" },
 	{ 0x0e, "MEM_OUT_OF_BOUNDS" },
 	{ 0x0f, "UNALIGNED_MEM_ACCESS" },
+	{ 0x10, "INVALID_ADDR_SPACE" },
 	{ 0x11, "INVALID_PARAM" },
 	{}
 };
-- 
cgit v0.10.2


From 2e69a5cd25a131cf283d3b655c01fb6db15dd60c Mon Sep 17 00:00:00 2001
From: Ilia Mirkin <imirkin@alum.mit.edu>
Date: Wed, 7 Oct 2015 18:39:33 -0400
Subject: drm/nouveau/gr: add FERMI_COMPUTE_B class to GF110+

GF110+ supports both the A and B compute classes, make sure to accept
both.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf110.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf110.c
index d131874..d081ee4 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf110.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf110.c
@@ -98,6 +98,7 @@ gf110_gr = {
 		{ -1, -1, FERMI_B, &gf100_fermi },
 		{ -1, -1, FERMI_C, &gf100_fermi },
 		{ -1, -1, FERMI_COMPUTE_A },
+		{ -1, -1, FERMI_COMPUTE_B },
 		{}
 	}
 };
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf117.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf117.c
index 28483d8..d8e8af4 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf117.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf117.c
@@ -135,6 +135,7 @@ gf117_gr = {
 		{ -1, -1, FERMI_B, &gf100_fermi },
 		{ -1, -1, FERMI_C, &gf100_fermi },
 		{ -1, -1, FERMI_COMPUTE_A },
+		{ -1, -1, FERMI_COMPUTE_B },
 		{}
 	}
 };
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf119.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf119.c
index 9811a72..01faf9a 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf119.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf119.c
@@ -189,6 +189,7 @@ gf119_gr = {
 		{ -1, -1, FERMI_B, &gf100_fermi },
 		{ -1, -1, FERMI_C, &gf100_fermi },
 		{ -1, -1, FERMI_COMPUTE_A },
+		{ -1, -1, FERMI_COMPUTE_B },
 		{}
 	}
 };
-- 
cgit v0.10.2


From 354a22496613366833edfe29300a6bfe6482255c Mon Sep 17 00:00:00 2001
From: Julia Lawall <Julia.Lawall@lip6.fr>
Date: Sun, 11 Oct 2015 14:18:09 +0200
Subject: drm/nouveau/disp,pm: constify nvkm_object_func structures

These nvkm_object_func structures are never modified.  All other
nvkm_object_func structures are declared as const.

Done with the help of Coccinelle.

Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootnv04.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootnv04.c
index 62d3fb6..2be8463 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootnv04.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootnv04.c
@@ -109,7 +109,7 @@ nv04_disp_mthd(struct nvkm_object *object, u32 mthd, void *data, u32 size)
 	return -EINVAL;
 }
 
-static struct nvkm_object_func
+static const struct nvkm_object_func
 nv04_disp_root = {
 	.mthd = nv04_disp_mthd,
 	.ntfy = nvkm_disp_ntfy,
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/pm/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/pm/base.c
index 0db9be2..2721592 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/pm/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/pm/base.c
@@ -633,7 +633,7 @@ nvkm_perfmon_dtor(struct nvkm_object *object)
 	return perfmon;
 }
 
-static struct nvkm_object_func
+static const struct nvkm_object_func
 nvkm_perfmon = {
 	.dtor = nvkm_perfmon_dtor,
 	.mthd = nvkm_perfmon_mthd,
-- 
cgit v0.10.2


From 5d5b43f59bd214506265ea05a013229ff5291101 Mon Sep 17 00:00:00 2001
From: Pierre Moreau <pierre.morrow@free.fr>
Date: Sat, 3 Oct 2015 21:35:16 +0200
Subject: drm/nouveau/pci: Handle 5-bit and 8-bit tag field

If the hardware supports extended tag field (8-bit ones), then enable it.

This is usually done by the VBIOS, but not on some MBPs (see fdo#86537).

In case extended tag field is not supported, 5-bit tag field is used which
limits the possible number of requests to 32. Apparently bits 7:0 of
0x08841c stores some number of outstanding requests, so cap it to 32 if
extended tag is unsupported.

Fixes: fdo#86537

v2: Restrict changes to chipsets >= 0x84
v3:
  * Add nvkm_pci_mask to pci.h
  * Mask bit 8 before setting it
v4:
  * Rename `add` argument of nvkm_pci_mask to `value`
  * Move code from nvkm_pci_init to g84_pci_init and remove PCIe and chipset
    checks
v5:
  * Rebase code on latest PCI structure
  * Restore PCIe check
  * Fix namings in nvkm_pci_mask
  * Rephrase part of the commit message

Signed-off-by: Pierre Moreau <pierre.morrow@free.fr>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/pci.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/pci.h
index 39ca88f..fee0a97 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/pci.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/pci.h
@@ -24,6 +24,7 @@ struct nvkm_pci {
 u32 nvkm_pci_rd32(struct nvkm_pci *, u16 addr);
 void nvkm_pci_wr08(struct nvkm_pci *, u16 addr, u8 data);
 void nvkm_pci_wr32(struct nvkm_pci *, u16 addr, u32 data);
+u32 nvkm_pci_mask(struct nvkm_pci *, u16 addr, u32 mask, u32 value);
 void nvkm_pci_rom_shadow(struct nvkm_pci *, bool shadow);
 
 int nv04_pci_new(struct nvkm_device *, int, struct nvkm_pci **);
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/base.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/base.c
index 2110622..d671dcf 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/base.c
@@ -46,6 +46,14 @@ nvkm_pci_wr32(struct nvkm_pci *pci, u16 addr, u32 data)
 	pci->func->wr32(pci, addr, data);
 }
 
+u32
+nvkm_pci_mask(struct nvkm_pci *pci, u16 addr, u32 mask, u32 value)
+{
+	u32 data = pci->func->rd32(pci, addr);
+	pci->func->wr32(pci, addr, (data & ~mask) | value);
+	return data;
+}
+
 void
 nvkm_pci_rom_shadow(struct nvkm_pci *pci, bool shadow)
 {
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/g84.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/g84.c
index 8f3b001..3faa6bf 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/g84.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/g84.c
@@ -25,8 +25,32 @@
 
 #include <core/pci.h>
 
+void
+g84_pci_init(struct nvkm_pci *pci)
+{
+	/* The following only concerns PCIe cards. */
+	if (!pci_is_pcie(pci->pdev))
+		return;
+
+	/* Tag field is 8-bit long, regardless of EXT_TAG.
+	 * However, if EXT_TAG is disabled, only the lower 5 bits of the tag
+	 * field should be used, limiting the number of request to 32.
+	 *
+	 * Apparently, 0x041c stores some limit on the number of requests
+	 * possible, so if EXT_TAG is disabled, limit that requests number to
+	 * 32
+	 *
+	 * Fixes fdo#86537
+	 */
+	if (nvkm_pci_rd32(pci, 0x007c) & 0x00000020)
+		nvkm_pci_mask(pci, 0x0080, 0x00000100, 0x00000100);
+	else
+		nvkm_pci_mask(pci, 0x041c, 0x00000060, 0x00000000);
+}
+
 static const struct nvkm_pci_func
 g84_pci_func = {
+	.init = g84_pci_init,
 	.rd32 = nv40_pci_rd32,
 	.wr08 = nv40_pci_wr08,
 	.wr32 = nv40_pci_wr32,
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/g94.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/g94.c
index 1714421..cd311ee 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/g94.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/g94.c
@@ -25,6 +25,7 @@
 
 static const struct nvkm_pci_func
 g94_pci_func = {
+	.init = g84_pci_init,
 	.rd32 = nv40_pci_rd32,
 	.wr08 = nv40_pci_wr08,
 	.wr32 = nv40_pci_wr32,
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/gf100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/gf100.c
index 86f8226..25e1ae7 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/gf100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/gf100.c
@@ -31,6 +31,7 @@ gf100_pci_msi_rearm(struct nvkm_pci *pci)
 
 static const struct nvkm_pci_func
 gf100_pci_func = {
+	.init = g84_pci_init,
 	.rd32 = nv40_pci_rd32,
 	.wr08 = nv40_pci_wr08,
 	.wr32 = nv40_pci_wr32,
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/priv.h b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/priv.h
index 1acd4bc..cf46d38 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pci/priv.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pci/priv.h
@@ -20,4 +20,6 @@ void nv40_pci_wr32(struct nvkm_pci *, u16, u32);
 void nv40_pci_msi_rearm(struct nvkm_pci *);
 
 void nv46_pci_msi_rearm(struct nvkm_pci *);
+
+void g84_pci_init(struct nvkm_pci *pci);
 #endif
-- 
cgit v0.10.2


From b4f2bf33bb1d50f1b52fae673dc85dfd750e6712 Mon Sep 17 00:00:00 2001
From: Roy Spliet <rspliet@eclipso.eu>
Date: Wed, 30 Sep 2015 00:23:44 +0100
Subject: drm/nouveau/bios/rammap: Identify DLLoff for >= GF100

Signed-off-by: Roy Spliet <rspliet@eclipso.eu>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/ramcfg.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/ramcfg.h
index 3a9abd3..dd48db7 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/ramcfg.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/ramcfg.h
@@ -78,7 +78,6 @@ struct nvbios_ramcfg {
 			unsigned ramcfg_11_01_04:1;
 			unsigned ramcfg_11_01_08:1;
 			unsigned ramcfg_11_01_10:1;
-			unsigned ramcfg_11_01_20:1;
 			unsigned ramcfg_11_01_40:1;
 			unsigned ramcfg_11_01_80:1;
 			unsigned ramcfg_11_02_03:2;
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/rammap.c b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/rammap.c
index f0e1fc74..3bbb1a7 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/rammap.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/rammap.c
@@ -219,7 +219,7 @@ nvbios_rammapSp(struct nvkm_bios *bios, u32 data,
 		p->ramcfg_11_01_04 = (nvbios_rd08(bios, data + 0x01) & 0x04) >> 2;
 		p->ramcfg_11_01_08 = (nvbios_rd08(bios, data + 0x01) & 0x08) >> 3;
 		p->ramcfg_11_01_10 = (nvbios_rd08(bios, data + 0x01) & 0x10) >> 4;
-		p->ramcfg_11_01_20 = (nvbios_rd08(bios, data + 0x01) & 0x20) >> 5;
+		p->ramcfg_DLLoff =   (nvbios_rd08(bios, data + 0x01) & 0x20) >> 5;
 		p->ramcfg_11_01_40 = (nvbios_rd08(bios, data + 0x01) & 0x40) >> 6;
 		p->ramcfg_11_01_80 = (nvbios_rd08(bios, data + 0x01) & 0x80) >> 7;
 		p->ramcfg_11_02_03 = (nvbios_rd08(bios, data + 0x02) & 0x03) >> 0;
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gddr5.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gddr5.c
index 24f83b0..2cc074d 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gddr5.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gddr5.c
@@ -38,11 +38,12 @@ nvkm_gddr5_calc(struct nvkm_ram *ram, bool nuts)
 	int WL, CL, WR, at[2], dt, ds;
 	int rq = ram->freq < 1000000; /* XXX */
 
+	xd = !ram->next->bios.ramcfg_DLLoff;
+
 	switch (ram->next->bios.ramcfg_ver) {
 	case 0x11:
 		pd =  ram->next->bios.ramcfg_11_01_80;
 		lf =  ram->next->bios.ramcfg_11_01_40;
-		xd = !ram->next->bios.ramcfg_11_01_20;
 		vh =  ram->next->bios.ramcfg_11_02_10;
 		vr =  ram->next->bios.ramcfg_11_02_04;
 		vo =  ram->next->bios.ramcfg_11_06;
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramgk104.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramgk104.c
index 9893556..0d20563 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramgk104.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramgk104.c
@@ -673,6 +673,25 @@ gk104_ram_calc_gddr5(struct gk104_ram *ram, u32 freq)
  * DDR3
  ******************************************************************************/
 
+static void
+nvkm_sddr3_dll_reset(struct gk104_ramfuc *fuc)
+{
+	ram_nuke(fuc, mr[0]);
+	ram_mask(fuc, mr[0], 0x100, 0x100);
+	ram_mask(fuc, mr[0], 0x100, 0x000);
+}
+
+static void
+nvkm_sddr3_dll_disable(struct gk104_ramfuc *fuc)
+{
+	u32 mr1_old = ram_rd32(fuc, mr[1]);
+
+	if (!(mr1_old & 0x1)) {
+		ram_mask(fuc, mr[1], 0x1, 0x1);
+		ram_nsec(fuc, 1000);
+	}
+}
+
 static int
 gk104_ram_calc_sddr3(struct gk104_ram *ram, u32 freq)
 {
@@ -702,6 +721,10 @@ gk104_ram_calc_sddr3(struct gk104_ram *ram, u32 freq)
 		ram_mask(fuc, 0x10f808, 0x04000000, 0x04000000);
 
 	ram_wr32(fuc, 0x10f314, 0x00000001); /* PRECHARGE */
+
+	if (next->bios.ramcfg_DLLoff)
+		nvkm_sddr3_dll_disable(fuc);
+
 	ram_wr32(fuc, 0x10f210, 0x00000000); /* REFRESH_AUTO = 0 */
 	ram_wr32(fuc, 0x10f310, 0x00000001); /* REFRESH */
 	ram_mask(fuc, 0x10f200, 0x80000000, 0x80000000);
@@ -879,17 +902,20 @@ gk104_ram_calc_sddr3(struct gk104_ram *ram, u32 freq)
 	ram_wr32(fuc, 0x10f210, 0x80000000); /* REFRESH_AUTO = 1 */
 	ram_nsec(fuc, 1000);
 
-	ram_nuke(fuc, mr[0]);
-	ram_mask(fuc, mr[0], 0x100, 0x100);
-	ram_mask(fuc, mr[0], 0x100, 0x000);
+	if (!next->bios.ramcfg_DLLoff) {
+		ram_mask(fuc, mr[1], 0x1, 0x0);
+		nvkm_sddr3_dll_reset(fuc);
+	}
 
-	ram_mask(fuc, mr[2], 0xfff, ram->base.mr[2]);
+	ram_mask(fuc, mr[2], 0x00000fff, ram->base.mr[2]);
+	ram_mask(fuc, mr[1], 0xffffffff, ram->base.mr[1]);
 	ram_wr32(fuc, mr[0], ram->base.mr[0]);
 	ram_nsec(fuc, 1000);
 
-	ram_nuke(fuc, mr[0]);
-	ram_mask(fuc, mr[0], 0x100, 0x100);
-	ram_mask(fuc, mr[0], 0x100, 0x000);
+	if (!next->bios.ramcfg_DLLoff) {
+		nvkm_sddr3_dll_reset(fuc);
+		ram_nsec(fuc, 1000);
+	}
 
 	if (vc == 0 && ram_have(fuc, gpio2E)) {
 		u32 temp  = ram_mask(fuc, gpio2E, 0x3000, fuc->r_func2E[0]);
@@ -1600,6 +1626,7 @@ gk104_ram_new(struct nvkm_fb *fb, struct nvkm_ram **pram)
 		break;
 	case NVKM_RAM_TYPE_DDR3:
 		ram->fuc.r_mr[0] = ramfuc_reg(0x10f300);
+		ram->fuc.r_mr[1] = ramfuc_reg(0x10f304);
 		ram->fuc.r_mr[2] = ramfuc_reg(0x10f320);
 		break;
 	default:
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/sddr3.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/sddr3.c
index b4edc97..2690033 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/sddr3.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/sddr3.c
@@ -70,6 +70,8 @@ nvkm_sddr3_calc(struct nvkm_ram *ram)
 {
 	int CWL, CL, WR, DLL = 0, ODT = 0;
 
+	DLL = !ram->next->bios.ramcfg_DLLoff;
+
 	switch (ram->next->bios.timing_ver) {
 	case 0x10:
 		if (ram->next->bios.timing_hdr < 0x17) {
@@ -79,7 +81,6 @@ nvkm_sddr3_calc(struct nvkm_ram *ram)
 		CWL = ram->next->bios.timing_10_CWL;
 		CL  = ram->next->bios.timing_10_CL;
 		WR  = ram->next->bios.timing_10_WR;
-		DLL = !ram->next->bios.ramcfg_DLLoff;
 		ODT = ram->next->bios.timing_10_ODT;
 		break;
 	case 0x20:
@@ -87,7 +88,6 @@ nvkm_sddr3_calc(struct nvkm_ram *ram)
 		CL  = (ram->next->bios.timing[1] & 0x0000001f) >> 0;
 		WR  = (ram->next->bios.timing[2] & 0x007f0000) >> 16;
 		/* XXX: Get these values from the VBIOS instead */
-		DLL = !(ram->mr[1] & 0x1);
 		ODT =   (ram->mr[1] & 0x004) >> 2 |
 			(ram->mr[1] & 0x040) >> 5 |
 		        (ram->mr[1] & 0x200) >> 7;
-- 
cgit v0.10.2


From e0a37f85fc95e3f2550446316bc4a27d00d75993 Mon Sep 17 00:00:00 2001
From: Roy Spliet <rspliet@eclipso.eu>
Date: Wed, 30 Sep 2015 00:23:45 +0100
Subject: drm/nouveau/fb/ramgt215: Transform GPIO ramfuc method from
 FBVREF-specific to generic

In preparation of changing FBVDDQ, as observed on at least one GDDR3 card.
While at it, adhere to func.log[1] properly for consistency.

Signed-off-by: Roy Spliet <rspliet@eclipso.eu>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramgt215.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramgt215.c
index 5c08ae8..0c28f38 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramgt215.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramgt215.c
@@ -34,9 +34,6 @@
 #include <subdev/clk/gt215.h>
 #include <subdev/gpio.h>
 
-/* XXX: Remove when memx gains GPIO support */
-extern int nv50_gpio_location(int line, u32 *reg, u32 *shift);
-
 struct gt215_ramfuc {
 	struct ramfuc base;
 	struct ramfuc_reg r_0x001610;
@@ -75,7 +72,7 @@ struct gt215_ramfuc {
 	struct ramfuc_reg r_0x111400;
 	struct ramfuc_reg r_0x611200;
 	struct ramfuc_reg r_mr[4];
-	struct ramfuc_reg r_gpioFBVREF;
+	struct ramfuc_reg r_gpio[4];
 };
 
 struct gt215_ltrain {
@@ -466,24 +463,27 @@ gt215_ram_lock_pll(struct gt215_ramfuc *fuc, struct gt215_clk_info *mclk)
 }
 
 static void
-gt215_ram_fbvref(struct gt215_ramfuc *fuc, u32 val)
+gt215_ram_gpio(struct gt215_ramfuc *fuc, u8 tag, u32 val)
 {
 	struct nvkm_gpio *gpio = fuc->base.fb->subdev.device->gpio;
 	struct dcb_gpio_func func;
 	u32 reg, sh, gpio_val;
 	int ret;
 
-	if (nvkm_gpio_get(gpio, 0, 0x2e, DCB_GPIO_UNUSED) != val) {
-		ret = nvkm_gpio_find(gpio, 0, 0x2e, DCB_GPIO_UNUSED, &func);
+	if (nvkm_gpio_get(gpio, 0, tag, DCB_GPIO_UNUSED) != val) {
+		ret = nvkm_gpio_find(gpio, 0, tag, DCB_GPIO_UNUSED, &func);
 		if (ret)
 			return;
 
-		nv50_gpio_location(func.line, &reg, &sh);
-		gpio_val = ram_rd32(fuc, gpioFBVREF);
+		reg = func.line >> 3;
+		sh = (func.line & 0x7) << 2;
+		gpio_val = ram_rd32(fuc, gpio[reg]);
 		if (gpio_val & (8 << sh))
 			val = !val;
+		if (!(func.log[1] & 1))
+			val = !val;
 
-		ram_mask(fuc, gpioFBVREF, (0x3 << sh), ((val | 0x2) << sh));
+		ram_mask(fuc, gpio[reg], (0x3 << sh), ((val | 0x2) << sh));
 		ram_nsec(fuc, 20000);
 	}
 }
@@ -642,8 +642,8 @@ gt215_ram_calc(struct nvkm_ram *base, u32 freq)
 		break;
 	}
 
-	if (fuc->r_gpioFBVREF.addr && next->bios.timing_10_ODT)
-		gt215_ram_fbvref(fuc, 0);
+	if (next->bios.timing_10_ODT)
+		gt215_ram_gpio(fuc, 0x2e, 1);
 
 	/* Brace RAM for impact */
 	ram_wr32(fuc, 0x1002d4, 0x00000001);
@@ -809,8 +809,8 @@ gt215_ram_calc(struct nvkm_ram *base, u32 freq)
 	ram_mask(fuc, 0x100718, 0xffffffff, unk718);
 	ram_mask(fuc, 0x111100, 0xffffffff, r111100);
 
-	if (fuc->r_gpioFBVREF.addr && !next->bios.timing_10_ODT)
-		gt215_ram_fbvref(fuc, 1);
+	if (!next->bios.timing_10_ODT)
+		gt215_ram_gpio(fuc, 0x2e, 0);
 
 	/* Reset DLL */
 	if (!next->bios.ramcfg_DLLoff)
@@ -919,10 +919,7 @@ gt215_ram_func = {
 int
 gt215_ram_new(struct nvkm_fb *fb, struct nvkm_ram **pram)
 {
-	struct nvkm_gpio *gpio = fb->subdev.device->gpio;
-	struct dcb_gpio_func func;
 	struct gt215_ram *ram;
-	u32 reg, shift;
 	int ret, i;
 
 	if (!(ram = kzalloc(sizeof(*ram), GFP_KERNEL)))
@@ -981,12 +978,10 @@ gt215_ram_new(struct nvkm_fb *fb, struct nvkm_ram **pram)
 		ram->fuc.r_mr[2] = ramfuc_reg(0x1002e0);
 		ram->fuc.r_mr[3] = ramfuc_reg(0x1002e4);
 	}
-
-	ret = nvkm_gpio_find(gpio, 0, 0x2e, DCB_GPIO_UNUSED, &func);
-	if (ret == 0) {
-		nv50_gpio_location(func.line, &reg, &shift);
-		ram->fuc.r_gpioFBVREF = ramfuc_reg(reg);
-	}
+	ram->fuc.r_gpio[0] = ramfuc_reg(0x00e104);
+	ram->fuc.r_gpio[1] = ramfuc_reg(0x00e108);
+	ram->fuc.r_gpio[2] = ramfuc_reg(0x00e120);
+	ram->fuc.r_gpio[3] = ramfuc_reg(0x00e124);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gpio/nv50.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gpio/nv50.c
index 8996649..73923fd 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/gpio/nv50.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gpio/nv50.c
@@ -54,7 +54,7 @@ nv50_gpio_reset(struct nvkm_gpio *gpio, u8 match)
 	}
 }
 
-int
+static int
 nv50_gpio_location(int line, u32 *reg, u32 *shift)
 {
 	const u32 nv50_gpio_reg[4] = { 0xe104, 0xe108, 0xe280, 0xe284 };
-- 
cgit v0.10.2


From ef6e8f4c7fd017ee66fc4b0fd3cfeae48c2e26d5 Mon Sep 17 00:00:00 2001
From: Roy Spliet <rspliet@eclipso.eu>
Date: Wed, 30 Sep 2015 00:23:46 +0100
Subject: drm/nouveau/fb/ramgt215: Change FBVDD/Q when BIOS asks for it

Signed-off-by: Roy Spliet <rspliet@eclipso.eu>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/ramcfg.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/ramcfg.h
index dd48db7..dca6c06 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/ramcfg.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/ramcfg.h
@@ -39,6 +39,7 @@ struct nvbios_ramcfg {
 	unsigned ramcfg_timing;
 	unsigned ramcfg_DLLoff;
 	unsigned ramcfg_RON;
+	unsigned ramcfg_FBVDDQ;
 	union {
 		struct {
 			unsigned ramcfg_00_03_01:1;
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/rammap.c b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/rammap.c
index 3bbb1a7..74a4ab5 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/rammap.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/rammap.c
@@ -205,6 +205,7 @@ nvbios_rammapSp(struct nvkm_bios *bios, u32 data,
 		p->ramcfg_DLLoff   = (nvbios_rd08(bios, data + 0x02) & 0x40) >> 6;
 		p->ramcfg_10_03_0f = (nvbios_rd08(bios, data + 0x03) & 0x0f) >> 0;
 		p->ramcfg_10_04_01 = (nvbios_rd08(bios, data + 0x04) & 0x01) >> 0;
+		p->ramcfg_FBVDDQ   = (nvbios_rd08(bios, data + 0x04) & 0x08) >> 3;
 		p->ramcfg_10_05    = (nvbios_rd08(bios, data + 0x05) & 0xff) >> 0;
 		p->ramcfg_10_06    = (nvbios_rd08(bios, data + 0x06) & 0xff) >> 0;
 		p->ramcfg_10_07    = (nvbios_rd08(bios, data + 0x07) & 0xff) >> 0;
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramgt215.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramgt215.c
index 0c28f38..8d81bf1 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramgt215.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramgt215.c
@@ -498,6 +498,7 @@ gt215_ram_calc(struct nvkm_ram *base, u32 freq)
 	struct nvkm_device *device = subdev->device;
 	struct nvkm_bios *bios = device->bios;
 	struct gt215_clk_info mclk;
+	struct nvkm_gpio *gpio = device->gpio;
 	struct nvkm_ram_data *next;
 	u8  ver, hdr, cnt, len, strap;
 	u32 data;
@@ -656,6 +657,23 @@ gt215_ram_calc(struct nvkm_ram *base, u32 freq)
 	if (device->chipset == 0xa3 && freq <= 500000)
 		ram_mask(fuc, 0x100700, 0x00000006, 0x00000006);
 
+	/* Alter FBVDD/Q, apparently must be done with PLL disabled, thus
+	 * set it to bypass */
+	if (nvkm_gpio_get(gpio, 0, 0x18, DCB_GPIO_UNUSED) ==
+			next->bios.ramcfg_FBVDDQ) {
+		data = ram_rd32(fuc, 0x004000) & 0x9;
+
+		if (data == 0x1)
+			ram_mask(fuc, 0x004000, 0x8, 0x8);
+		if (data & 0x1)
+			ram_mask(fuc, 0x004000, 0x1, 0x0);
+
+		gt215_ram_gpio(fuc, 0x18, !next->bios.ramcfg_FBVDDQ);
+
+		if (data & 0x1)
+			ram_mask(fuc, 0x004000, 0x1, 0x1);
+	}
+
 	/* Fiddle with clocks */
 	/* There's 4 scenario's
 	 * pll->pll: first switch to a 324MHz clock, set up new PLL, switch
-- 
cgit v0.10.2


From 0b0b78cd7def0fc001fc8cefa8621823bff6a8e5 Mon Sep 17 00:00:00 2001
From: Roy Spliet <rspliet@eclipso.eu>
Date: Wed, 30 Sep 2015 00:23:47 +0100
Subject: drm/nouveau/fb/ramgt215: Restructure r111100 calculation for DDR2

Seems to be mostly equal to DDR3 on < GT218, should improve stability for
DDR2 reclocks.

Signed-off-by: Roy Spliet <rspliet@eclipso.eu>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramgt215.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramgt215.c
index 8d81bf1..d15ea88 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramgt215.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramgt215.c
@@ -771,39 +771,43 @@ gt215_ram_calc(struct nvkm_ram *base, u32 freq)
 	unk71c  = ram_rd32(fuc, 0x10071c) & ~0x00000100;
 	r111100 = ram_rd32(fuc, 0x111100) & ~0x3a800000;
 
-	if (next->bios.ramcfg_10_02_04) {
-		switch (ram->base.type) {
-		case NVKM_RAM_TYPE_DDR3:
-			if (device->chipset != 0xa8)
-				r111100 |= 0x00000004;
-			/* no break */
-		case NVKM_RAM_TYPE_DDR2:
-			r111100 |= 0x08000000;
-			break;
-		default:
-			break;
-		}
-	} else {
-		switch (ram->base.type) {
-		case NVKM_RAM_TYPE_DDR2:
-			r111100 |= 0x1a800000;
+	/* NVA8 seems to skip various bits related to ramcfg_10_02_04 */
+	if (device->chipset == 0xa8) {
+		r111100 |= 0x08000000;
+		if (!next->bios.ramcfg_10_02_04)
 			unk714  |= 0x00000010;
-			break;
-		case NVKM_RAM_TYPE_DDR3:
-			if (device->chipset == 0xa8) {
-				r111100 |=  0x08000000;
-			} else {
-				r111100 &= ~0x00000004;
+	} else {
+		if (next->bios.ramcfg_10_02_04) {
+			switch (ram->base.type) {
+			case NVKM_RAM_TYPE_DDR2:
+			case NVKM_RAM_TYPE_DDR3:
+				r111100 &= ~0x00000020;
+				if (next->bios.ramcfg_10_02_10)
+					r111100 |= 0x08000004;
+				else
+					r111100 |= 0x00000024;
+				break;
+			default:
+				break;
+			}
+		} else {
+			switch (ram->base.type) {
+			case NVKM_RAM_TYPE_DDR2:
+			case NVKM_RAM_TYPE_DDR3:
+				r111100 &= ~0x00000024;
 				r111100 |=  0x12800000;
+
+				if (next->bios.ramcfg_10_02_10)
+					r111100 |= 0x08000000;
+				unk714  |= 0x00000010;
+				break;
+			case NVKM_RAM_TYPE_GDDR3:
+				r111100 |= 0x30000000;
+				unk714  |= 0x00000020;
+				break;
+			default:
+				break;
 			}
-			unk714  |= 0x00000010;
-			break;
-		case NVKM_RAM_TYPE_GDDR3:
-			r111100 |= 0x30000000;
-			unk714  |= 0x00000020;
-			break;
-		default:
-			break;
 		}
 	}
 
-- 
cgit v0.10.2


From 1cf688dd1b6f8f3ab998000e0f14b37041c77e30 Mon Sep 17 00:00:00 2001
From: Roy Spliet <rspliet@eclipso.eu>
Date: Wed, 30 Sep 2015 00:23:48 +0100
Subject: drm/nouveau/fb/ramnv50: Voltage GPIOs

Does not seem to be necessary for NVA0, hence untested by me.

Signed-off-by: Roy Spliet <rspliet@eclipso.eu>
Tested-by: Pierre Moreau <pierre.morrow@free.fr>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/rammap.c b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/rammap.c
index 74a4ab5..d0ae745 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/rammap.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/rammap.c
@@ -171,6 +171,7 @@ nvbios_rammapSp_from_perf(struct nvkm_bios *bios, u32 data, u8 size, int idx,
 	p->ramcfg_DLLoff   = (nvbios_rd08(bios, data + 0x03) & 0x04) >> 2;
 	p->ramcfg_00_03_08 = (nvbios_rd08(bios, data + 0x03) & 0x08) >> 3;
 	p->ramcfg_RON      = (nvbios_rd08(bios, data + 0x03) & 0x10) >> 3;
+	p->ramcfg_FBVDDQ   = (nvbios_rd08(bios, data + 0x03) & 0x80) >> 7;
 	p->ramcfg_00_04_02 = (nvbios_rd08(bios, data + 0x04) & 0x02) >> 1;
 	p->ramcfg_00_04_04 = (nvbios_rd08(bios, data + 0x04) & 0x04) >> 2;
 	p->ramcfg_00_04_20 = (nvbios_rd08(bios, data + 0x04) & 0x20) >> 5;
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramnv50.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramnv50.c
index 9197e0e..ae6b0c4 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramnv50.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramnv50.c
@@ -33,6 +33,7 @@
 #include <subdev/bios/rammap.h>
 #include <subdev/bios/timing.h>
 #include <subdev/clk/pll.h>
+#include <subdev/gpio.h>
 
 struct nv50_ramseq {
 	struct hwsq base;
@@ -59,6 +60,7 @@ struct nv50_ramseq {
 	struct hwsq_reg r_0x611200;
 	struct hwsq_reg r_timing[9];
 	struct hwsq_reg r_mr[4];
+	struct hwsq_reg r_gpio[4];
 };
 
 struct nv50_ram {
@@ -154,6 +156,33 @@ nvkm_sddr2_dll_reset(struct nv50_ramseq *hwsq)
 	ram_nsec(hwsq, 24000);
 }
 
+static void
+nv50_ram_gpio(struct nv50_ramseq *hwsq, u8 tag, u32 val)
+{
+	struct nvkm_gpio *gpio = hwsq->base.subdev->device->gpio;
+	struct dcb_gpio_func func;
+	u32 reg, sh, gpio_val;
+	int ret;
+
+	if (nvkm_gpio_get(gpio, 0, tag, DCB_GPIO_UNUSED) != val) {
+		ret = nvkm_gpio_find(gpio, 0, tag, DCB_GPIO_UNUSED, &func);
+		if (ret)
+			return;
+
+		reg = func.line >> 3;
+		sh = (func.line & 0x7) << 2;
+		gpio_val = ram_rd32(hwsq, gpio[reg]);
+
+		if (gpio_val & (8 << sh))
+			val = !val;
+		if (!(func.log[1] & 1))
+			val = !val;
+
+		ram_mask(hwsq, gpio[reg], (0x3 << sh), ((val | 0x2) << sh));
+		ram_nsec(hwsq, 20000);
+	}
+}
+
 static int
 nv50_ram_calc(struct nvkm_ram *base, u32 freq)
 {
@@ -250,6 +279,9 @@ nv50_ram_calc(struct nvkm_ram *base, u32 freq)
 	ram_wait(hwsq, 0x00, 0x01); /* wait for fb disabled */
 	ram_nsec(hwsq, 2000);
 
+	if (next->bios.timing_10_ODT)
+		nv50_ram_gpio(hwsq, 0x2e, 1);
+
 	ram_wr32(hwsq, 0x1002d4, 0x00000001); /* precharge */
 	ram_wr32(hwsq, 0x1002d0, 0x00000001); /* refresh */
 	ram_wr32(hwsq, 0x1002d0, 0x00000001); /* refresh */
@@ -288,6 +320,7 @@ nv50_ram_calc(struct nvkm_ram *base, u32 freq)
 	ram_mask(hwsq, 0x004008, 0x91ff0000, r004008);
 	if (subdev->device->chipset >= 0x96)
 		ram_wr32(hwsq, 0x100da0, r100da0);
+	nv50_ram_gpio(hwsq, 0x18, !next->bios.ramcfg_FBVDDQ);
 	ram_nsec(hwsq, 64000); /*XXX*/
 	ram_nsec(hwsq, 32000); /*XXX*/
 
@@ -364,6 +397,9 @@ nv50_ram_calc(struct nvkm_ram *base, u32 freq)
 	}
 	ram_mask(hwsq, mr[1], 0xffffffff, ram->base.mr[1]);
 
+	if (!next->bios.timing_10_ODT)
+		nv50_ram_gpio(hwsq, 0x2e, 0);
+
 	/* Reset DLL */
 	if (!next->bios.ramcfg_DLLoff)
 		nvkm_sddr2_dll_reset(hwsq);
@@ -634,5 +670,10 @@ nv50_ram_new(struct nvkm_fb *fb, struct nvkm_ram **pram)
 		ram->hwsq.r_mr[3] = hwsq_reg(0x1002e4);
 	}
 
+	ram->hwsq.r_gpio[0] = hwsq_reg(0x00e104);
+	ram->hwsq.r_gpio[1] = hwsq_reg(0x00e108);
+	ram->hwsq.r_gpio[2] = hwsq_reg(0x00e120);
+	ram->hwsq.r_gpio[3] = hwsq_reg(0x00e124);
+
 	return 0;
 }
-- 
cgit v0.10.2


From 797eb6ed8f13bb2f88bd605515990ef795a1eff7 Mon Sep 17 00:00:00 2001
From: Roy Spliet <rspliet@eclipso.eu>
Date: Wed, 30 Sep 2015 00:23:49 +0100
Subject: drm/nouveau/fb/ramnv50: Deal with cards without timing entries

Like Pierre's G94. We might want to structure Kepler similarly in a follow-up.

Signed-off-by: Roy Spliet <rspliet@eclipso.eu>
Tested-by: Pierre Moreau <pierre.morrow@free.fr>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gddr3.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gddr3.c
index 79b523a..60ece0a 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gddr3.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/gddr3.c
@@ -63,7 +63,7 @@ ramgddr3_wr_lo[] = {
 	{ 5, 2 }, { 7, 4 }, { 8, 5 }, { 9, 6 }, { 10, 7 },
 	{ 11, 0 }, { 13 , 1 },
 	/* the below are mentioned in some, but not all, gddr3 docs */
-	{ 4, 1 }, { 6, 3 }, { 12, 1 },
+	{ 4, 0 }, { 6, 3 }, { 12, 1 },
 	{ -1 }
 };
 
@@ -87,15 +87,17 @@ nvkm_gddr3_calc(struct nvkm_ram *ram)
 		WR  = (ram->next->bios.timing[2] & 0x007f0000) >> 16;
 		/* XXX: Get these values from the VBIOS instead */
 		DLL = !(ram->mr[1] & 0x1);
-		ODT =  (ram->mr[1] & 0x004) >> 2 |
-		       (ram->mr[1] & 0x040) >> 5 |
-		       (ram->mr[1] & 0x200) >> 7;
 		RON = !(ram->mr[1] & 0x300) >> 8;
 		break;
 	default:
 		return -ENOSYS;
 	}
 
+	if (ram->next->bios.timing_ver == 0x20 ||
+	    ram->next->bios.ramcfg_timing == 0xff) {
+		ODT =  (ram->mr[1] & 0xc) >> 2;
+	}
+
 	hi = ram->mr[2] & 0x1;
 	CL  = ramxlat(hi ? ramgddr3_cl_hi : ramgddr3_cl_lo, CL);
 	WR  = ramxlat(ramgddr3_wr_lo, WR);
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramnv50.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramnv50.c
index ae6b0c4..1c6ae6b 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramnv50.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramnv50.c
@@ -146,6 +146,38 @@ nv50_ram_timing_calc(struct nv50_ram *ram, u32 *timing)
 	nvkm_debug(subdev, " 240: %08x\n", timing[8]);
 	return 0;
 }
+
+static int
+nv50_ram_timing_read(struct nv50_ram *ram, u32 *timing)
+{
+	unsigned int i;
+	struct nvbios_ramcfg *cfg = &ram->base.target.bios;
+	struct nvkm_subdev *subdev = &ram->base.fb->subdev;
+	struct nvkm_device *device = subdev->device;
+
+	for (i = 0; i <= 8; i++)
+		timing[i] = nvkm_rd32(device, 0x100220 + (i * 4));
+
+	/* Derive the bare minimum for the MR calculation to succeed */
+	cfg->timing_ver = 0x10;
+	T(CL) = (timing[3] & 0xff) + 1;
+
+	switch (ram->base.type) {
+	case NVKM_RAM_TYPE_DDR2:
+		T(CWL) = T(CL) - 1;
+		break;
+	case NVKM_RAM_TYPE_GDDR3:
+		T(CWL) = ((timing[2] & 0xff000000) >> 24) + 1;
+		break;
+	default:
+		return -ENOSYS;
+		break;
+	}
+
+	T(WR) = ((timing[1] >> 24) & 0xff) - 1 - T(CWL);
+
+	return 0;
+}
 #undef T
 
 static void
@@ -242,10 +274,11 @@ nv50_ram_calc(struct nvkm_ram *base, u32 freq)
 				 strap, data, ver, hdr);
 			return -EINVAL;
 		}
+		nv50_ram_timing_calc(ram, timing);
+	} else {
+		nv50_ram_timing_read(ram, timing);
 	}
 
-	nv50_ram_timing_calc(ram, timing);
-
 	ret = ram_init(hwsq, subdev);
 	if (ret)
 		return ret;
@@ -264,8 +297,10 @@ nv50_ram_calc(struct nvkm_ram *base, u32 freq)
 		break;
 	}
 
-	if (ret)
+	if (ret) {
+		nvkm_error(subdev, "Could not calculate MR\n");
 		return ret;
+	}
 
 	/* Always disable this bit during reclock */
 	ram_mask(hwsq, 0x100200, 0x00000800, 0x00000000);
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/sddr2.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/sddr2.c
index 86bf674..b9f1ffd 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/sddr2.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/sddr2.c
@@ -76,6 +76,12 @@ nvkm_sddr2_calc(struct nvkm_ram *ram)
 		return -ENOSYS;
 	}
 
+	if (ram->next->bios.timing_ver == 0x20 ||
+	    ram->next->bios.ramcfg_timing == 0xff) {
+		ODT =  (ram->mr[1] & 0x004) >> 2 |
+		       (ram->mr[1] & 0x040) >> 5;
+	}
+
 	CL  = ramxlat(ramddr2_cl, CL);
 	WR  = ramxlat(ramddr2_wr, WR);
 	if (CL < 0 || WR < 0)
-- 
cgit v0.10.2


From 4d9faafa0fdda2f4ba04b5cdffc0af1bab2312f4 Mon Sep 17 00:00:00 2001
From: Roy Spliet <rspliet@eclipso.eu>
Date: Wed, 30 Sep 2015 00:23:50 +0100
Subject: drm/nouveau/fb/ramnv50: Script changes for G94 and up

10053c is not even read on some cards, and I have no idea exactly what the
criteria are. Likely NVIDIA pre-scans the VBIOS and in their driver disables
all features that are never used. The practical effect should be the same
as this implementation though.

Signed-off-by: Roy Spliet <rspliet@eclipso.eu>
Tested-by: Pierre Moreau <pierre.morrow@free.fr>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramnv50.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramnv50.c
index 1c6ae6b..d98d306 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramnv50.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramnv50.c
@@ -302,6 +302,9 @@ nv50_ram_calc(struct nvkm_ram *base, u32 freq)
 		return ret;
 	}
 
+	if (subdev->device->chipset <= 0x96 && !next->bios.ramcfg_00_03_02)
+		ram_mask(hwsq, 0x100710, 0x00000200, 0x00000000);
+
 	/* Always disable this bit during reclock */
 	ram_mask(hwsq, 0x100200, 0x00000800, 0x00000000);
 
@@ -353,8 +356,11 @@ nv50_ram_calc(struct nvkm_ram *base, u32 freq)
 			next->bios.rammap_00_16_40 << 14);
 	ram_mask(hwsq, 0x00400c, 0x0000ffff, (N1 << 8) | M1);
 	ram_mask(hwsq, 0x004008, 0x91ff0000, r004008);
-	if (subdev->device->chipset >= 0x96)
+
+	/* XXX: GDDR3 only? */
+	if (subdev->device->chipset >= 0x92)
 		ram_wr32(hwsq, 0x100da0, r100da0);
+
 	nv50_ram_gpio(hwsq, 0x18, !next->bios.ramcfg_FBVDDQ);
 	ram_nsec(hwsq, 64000); /*XXX*/
 	ram_nsec(hwsq, 32000); /*XXX*/
@@ -397,19 +403,33 @@ nv50_ram_calc(struct nvkm_ram *base, u32 freq)
 	ram_mask(hwsq, 0x100200, 0x00001000, !next->bios.ramcfg_00_04_02 << 12);
 
 	/* XXX: A lot of this could be "chipset"/"ram type" specific stuff */
-	unk710  = ram_rd32(hwsq, 0x100710) & ~0x00000101;
+	unk710  = ram_rd32(hwsq, 0x100710) & ~0x00000100;
 	unk714  = ram_rd32(hwsq, 0x100714) & ~0xf0000020;
 	unk718  = ram_rd32(hwsq, 0x100718) & ~0x00000100;
 	unk71c  = ram_rd32(hwsq, 0x10071c) & ~0x00000100;
+	if (subdev->device->chipset <= 0x96) {
+		unk710 &= ~0x0000006e;
+		unk714 &= ~0x00000100;
+
+		if (!next->bios.ramcfg_00_03_08)
+			unk710 |= 0x00000060;
+		if (!next->bios.ramcfg_FBVDDQ)
+			unk714 |= 0x00000100;
+		if ( next->bios.ramcfg_00_04_04)
+			unk710 |= 0x0000000e;
+	} else {
+		unk710 &= ~0x00000001;
+
+		if (!next->bios.ramcfg_00_03_08)
+			unk710 |= 0x00000001;
+	}
 
 	if ( next->bios.ramcfg_00_03_01)
 		unk71c |= 0x00000100;
 	if ( next->bios.ramcfg_00_03_02)
 		unk710 |= 0x00000100;
-	if (!next->bios.ramcfg_00_03_08) {
-		unk710 |= 0x1;
-		unk714 |= 0x20;
-	}
+	if (!next->bios.ramcfg_00_03_08)
+		unk714 |= 0x00000020;
 	if ( next->bios.ramcfg_00_04_04)
 		unk714 |= 0x70000000;
 	if ( next->bios.ramcfg_00_04_20)
@@ -420,6 +440,8 @@ nv50_ram_calc(struct nvkm_ram *base, u32 freq)
 	ram_mask(hwsq, 0x100718, 0xffffffff, unk718);
 	ram_mask(hwsq, 0x100710, 0xffffffff, unk710);
 
+	/* XXX: G94 does not even test these regs in trace. Harmless we do it,
+	 * but why is it omitted? */
 	if (next->bios.rammap_00_16_20) {
 		ram_wr32(hwsq, 0x1005a0, next->bios.ramcfg_00_07 << 16 |
 					 next->bios.ramcfg_00_06 << 8 |
@@ -450,6 +472,8 @@ nv50_ram_calc(struct nvkm_ram *base, u32 freq)
 		ram_mask(hwsq, 0x004008, 0x00004000, 0x00000000);
 	if (next->bios.ramcfg_00_03_02)
 		ram_mask(hwsq, 0x10021c, 0x00010000, 0x00010000);
+	if (subdev->device->chipset <= 0x96 && next->bios.ramcfg_00_03_02)
+		ram_mask(hwsq, 0x100710, 0x00000200, 0x00000200);
 
 	return 0;
 }
-- 
cgit v0.10.2


From 271c27665c2d2f719a2fca6a3530a82984f22cca Mon Sep 17 00:00:00 2001
From: Roy Spliet <rspliet@eclipso.eu>
Date: Wed, 30 Sep 2015 00:23:51 +0100
Subject: drm/nouveau/bus/hwsq: Implement VBLANK waiting heuristic

Avoids waiting for VBLANKS that never arrive on headless or otherwise
unconventional set-ups. Strategy taken from MEMX.

Signed-off-by: Roy Spliet <rspliet@eclipso.eu>
Tested-by: Pierre Moreau <pierre.morrow@free.fr>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/bus.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/bus.h
index 6a04d9c..33a057c 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/bus.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/bus.h
@@ -14,6 +14,7 @@ int  nvkm_hwsq_fini(struct nvkm_hwsq **, bool exec);
 void nvkm_hwsq_wr32(struct nvkm_hwsq *, u32 addr, u32 data);
 void nvkm_hwsq_setf(struct nvkm_hwsq *, u8 flag, int data);
 void nvkm_hwsq_wait(struct nvkm_hwsq *, u8 flag, u8 data);
+void nvkm_hwsq_wait_vblank(struct nvkm_hwsq *);
 void nvkm_hwsq_nsec(struct nvkm_hwsq *, u32 nsec);
 
 int nv04_bus_new(struct nvkm_device *, int, struct nvkm_bus **);
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bus/hwsq.c b/drivers/gpu/drm/nouveau/nvkm/subdev/bus/hwsq.c
index 79f1cf5..2a56689 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/bus/hwsq.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bus/hwsq.c
@@ -132,6 +132,38 @@ nvkm_hwsq_wait(struct nvkm_hwsq *hwsq, u8 flag, u8 data)
 }
 
 void
+nvkm_hwsq_wait_vblank(struct nvkm_hwsq *hwsq)
+{
+	struct nvkm_subdev *subdev = hwsq->subdev;
+	struct nvkm_device *device = subdev->device;
+	u32 heads, x, y, px = 0;
+	int i, head_sync;
+
+	heads = nvkm_rd32(device, 0x610050);
+	for (i = 0; i < 2; i++) {
+		/* Heuristic: sync to head with biggest resolution */
+		if (heads & (2 << (i << 3))) {
+			x = nvkm_rd32(device, 0x610b40 + (0x540 * i));
+			y = (x & 0xffff0000) >> 16;
+			x &= 0x0000ffff;
+			if ((x * y) > px) {
+				px = (x * y);
+				head_sync = i;
+			}
+		}
+	}
+
+	if (px == 0) {
+		nvkm_debug(subdev, "WAIT VBLANK !NO ACTIVE HEAD\n");
+		return;
+	}
+
+	nvkm_debug(subdev, "WAIT VBLANK HEAD%d\n", head_sync);
+	nvkm_hwsq_wait(hwsq, head_sync ? 0x3 : 0x1, 0x0);
+	nvkm_hwsq_wait(hwsq, head_sync ? 0x3 : 0x1, 0x1);
+}
+
+void
 nvkm_hwsq_nsec(struct nvkm_hwsq *hwsq, u32 nsec)
 {
 	u8 shift = 0, usec = nsec / 1000;
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bus/hwsq.h b/drivers/gpu/drm/nouveau/nvkm/subdev/bus/hwsq.h
index 8117ec5..54ec3b1 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/bus/hwsq.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bus/hwsq.h
@@ -134,6 +134,12 @@ hwsq_wait(struct hwsq *ram, u8 flag, u8 data)
 }
 
 static inline void
+hwsq_wait_vblank(struct hwsq *ram)
+{
+	nvkm_hwsq_wait_vblank(ram->hwsq);
+}
+
+static inline void
 hwsq_nsec(struct hwsq *ram, u32 nsec)
 {
 	nvkm_hwsq_nsec(ram->hwsq, nsec);
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramnv50.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramnv50.c
index d98d306..87bde8f 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramnv50.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramnv50.c
@@ -308,8 +308,7 @@ nv50_ram_calc(struct nvkm_ram *base, u32 freq)
 	/* Always disable this bit during reclock */
 	ram_mask(hwsq, 0x100200, 0x00000800, 0x00000000);
 
-	ram_wait(hwsq, 0x01, 0x00); /* wait for !vblank */
-	ram_wait(hwsq, 0x01, 0x01); /* wait for vblank */
+	ram_wait_vblank(hwsq);
 	ram_wr32(hwsq, 0x611200, 0x00003300);
 	ram_wr32(hwsq, 0x002504, 0x00000001); /* block fifo */
 	ram_nsec(hwsq, 8000);
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramseq.h b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramseq.h
index 0f1f97c..8df7306 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramseq.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramseq.h
@@ -11,5 +11,6 @@
 #define ram_mask(s,r,m,d)   hwsq_mask(&(s)->base, &(s)->r_##r, (m), (d))
 #define ram_setf(s,f,d)     hwsq_setf(&(s)->base, (f), (d))
 #define ram_wait(s,f,d)     hwsq_wait(&(s)->base, (f), (d))
+#define ram_wait_vblank(s)  hwsq_wait_vblank(&(s)->base)
 #define ram_nsec(s,n)       hwsq_nsec(&(s)->base, (n))
 #endif
-- 
cgit v0.10.2


From 0d42743dfa908a2ca4e349f883361906ebb4db95 Mon Sep 17 00:00:00 2001
From: Roy Spliet <rspliet@eclipso.eu>
Date: Wed, 30 Sep 2015 00:23:52 +0100
Subject: drm/nouveau/clk/g84: Enable reclocking for GDDR3 G94-G200

Your milage may vary, as it's only been tested on a single G94 and one G96.

Signed-off-by: Roy Spliet <rspliet@eclipso.eu>
Tested-by: Pierre Moreau <pierre.morrow@free.fr>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/clk/g84.c b/drivers/gpu/drm/nouveau/nvkm/subdev/clk/g84.c
index 347da9e..f97e3ec 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/clk/g84.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/clk/g84.c
@@ -44,5 +44,5 @@ int
 g84_clk_new(struct nvkm_device *device, int index, struct nvkm_clk **pclk)
 {
 	return nv50_clk_new_(&g84_clk, device, index,
-			     (device->chipset == 0xa0), pclk);
+			     (device->chipset >= 0x94), pclk);
 }
-- 
cgit v0.10.2


From 78eaf335e4c8224e74e5d512f20ec48109db9dac Mon Sep 17 00:00:00 2001
From: Karol Herbst <nouveau@karolherbst.de>
Date: Sun, 16 Aug 2015 10:19:25 +0200
Subject: drm/nouveau/pll/gk104: fix PLL instability due to bad configuration
 with gddr5

This patch uses an approach closer to the nvidia driver to configure
both PLLs for high gddr5 memory clocks (usually above 2400MHz)

Previously nouveau used the one PLL as it was used for the lower clocks
and just adjusted the second PLL to get as close as possible to the
requested clock.  This means for my card, that I got a 4050 MHz clock
although 4008 MHz was requested.

Now the driver iterates over a list of PLL configuration also used by
the nvidia driver and then adjust the second PLL to get near the
requested clock.  Also it hold to some restriction I found while
analyzing the PLL configurations

This won't fix all gddr5 high clock issues itself, but it should be
fine on hybrid gpu systems as found on many laptops these days.  Also
switching while normal desktop usage should be a lot more stable than
before.

v2: move the pll code into ramgk104

Signed-off-by: Karol Herbst <nouveau@karolherbst.de>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramgk104.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramgk104.c
index 0d20563..9df4503 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramgk104.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramgk104.c
@@ -971,6 +971,67 @@ gk104_ram_calc_data(struct gk104_ram *ram, u32 khz, struct nvkm_ram_data *data)
 }
 
 static int
+gk104_calc_pll_output(int fN, int M, int N, int P, int clk)
+{
+	return ((clk * N) + (((u16)(fN + 4096) * clk) >> 13)) / (M * P);
+}
+
+static int
+gk104_pll_calc_hiclk(int target_khz, int crystal,
+		int *N1, int *fN1, int *M1, int *P1,
+		int *N2, int *M2, int *P2)
+{
+	int best_clk = 0, best_err = target_khz, p_ref, n_ref;
+	bool upper = false;
+
+	*M1 = 1;
+	/* M has to be 1, otherwise it gets unstable */
+	*M2 = 1;
+	/* can be 1 or 2, sticking with 1 for simplicity */
+	*P2 = 1;
+
+	for (p_ref = 0x7; p_ref >= 0x5; --p_ref) {
+		for (n_ref = 0x25; n_ref <= 0x2b; ++n_ref) {
+			int cur_N, cur_clk, cur_err;
+
+			cur_clk = gk104_calc_pll_output(0, 1, n_ref, p_ref, crystal);
+			cur_N = target_khz / cur_clk;
+			cur_err = target_khz
+				- gk104_calc_pll_output(0xf000, 1, cur_N, 1, cur_clk);
+
+			/* we found a better combination */
+			if (cur_err < best_err) {
+				best_err = cur_err;
+				best_clk = cur_clk;
+				*N2 = cur_N;
+				*N1 = n_ref;
+				*P1 = p_ref;
+				upper = false;
+			}
+
+			cur_N += 1;
+			cur_err = gk104_calc_pll_output(0xf000, 1, cur_N, 1, cur_clk)
+				- target_khz;
+			if (cur_err < best_err) {
+				best_err = cur_err;
+				best_clk = cur_clk;
+				*N2 = cur_N;
+				*N1 = n_ref;
+				*P1 = p_ref;
+				upper = true;
+			}
+		}
+	}
+
+	/* adjust fN to get closer to the target clock */
+	*fN1 = (u16)((((best_err / *N2 * *P2) * (*P1 * *M1)) << 13) / crystal);
+	if (upper)
+		*fN1 = (u16)(1 - *fN1);
+
+	return gk104_calc_pll_output(*fN1, 1, *N1, *P1, crystal);
+}
+
+static int
 gk104_ram_calc_xits(struct gk104_ram *ram, struct nvkm_ram_data *next)
 {
 	struct gk104_ramfuc *fuc = &ram->fuc;
@@ -994,31 +1055,24 @@ gk104_ram_calc_xits(struct gk104_ram *ram, struct nvkm_ram_data *next)
 	 * kepler boards, no idea how/why they're chosen.
 	 */
 	refclk = next->freq;
-	if (ram->mode == 2)
-		refclk = fuc->mempll.refclk;
-
-	/* calculate refpll coefficients */
-	ret = gt215_pll_calc(subdev, &fuc->refpll, refclk, &ram->N1,
-			     &ram->fN1, &ram->M1, &ram->P1);
-	fuc->mempll.refclk = ret;
-	if (ret <= 0) {
-		nvkm_error(subdev, "unable to calc refpll\n");
-		return -EINVAL;
-	}
-
-	/* calculate mempll coefficients, if we're using it */
 	if (ram->mode == 2) {
-		/* post-divider doesn't work... the reg takes the values but
-		 * appears to completely ignore it.  there *is* a bit at
-		 * bit 28 that appears to divide the clock by 2 if set.
-		 */
-		fuc->mempll.min_p = 1;
-		fuc->mempll.max_p = 2;
-
-		ret = gt215_pll_calc(subdev, &fuc->mempll, next->freq,
-				     &ram->N2, NULL, &ram->M2, &ram->P2);
+		ret = gk104_pll_calc_hiclk(next->freq, subdev->device->crystal,
+				&ram->N1, &ram->fN1, &ram->M1, &ram->P1,
+				&ram->N2, &ram->M2, &ram->P2);
+		fuc->mempll.refclk = ret;
+		if (ret <= 0) {
+			nvkm_error(subdev, "unable to calc plls\n");
+			return -EINVAL;
+		}
+		nvkm_debug(subdev, "sucessfully calced PLLs for clock %i kHz"
+				" (refclock: %i kHz)\n", next->freq, ret);
+	} else {
+		/* calculate refpll coefficients */
+		ret = gt215_pll_calc(subdev, &fuc->refpll, refclk, &ram->N1,
+				     &ram->fN1, &ram->M1, &ram->P1);
+		fuc->mempll.refclk = ret;
 		if (ret <= 0) {
-			nvkm_error(subdev, "unable to calc mempll\n");
+			nvkm_error(subdev, "unable to calc refpll\n");
 			return -EINVAL;
 		}
 	}
-- 
cgit v0.10.2


From a2736b07df1f17c24c8890eb7b2dac529d9725d3 Mon Sep 17 00:00:00 2001
From: Ilia Mirkin <imirkin@alum.mit.edu>
Date: Tue, 27 Oct 2015 17:39:49 -0400
Subject: drm/nouveau/pci: enable c800 magic for Lenovo Y510P

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=70354#c75
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/pci.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/pci.c
index e8eb14e..c5fc909 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/device/pci.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/pci.c
@@ -259,6 +259,12 @@ nvkm_device_pci_10de_0df4[] = {
 };
 
 static const struct nvkm_device_pci_vendor
+nvkm_device_pci_10de_0fcd[] = {
+	{ 0x17aa, 0x3801, NULL, { .War00C800_0 = true } }, /* Lenovo Y510P */
+	{}
+};
+
+static const struct nvkm_device_pci_vendor
 nvkm_device_pci_10de_0fd2[] = {
 	{ 0x1028, 0x0595, "GeForce GT 640M LE" },
 	{ 0x1028, 0x05b2, "GeForce GT 640M LE" },
@@ -1349,7 +1355,7 @@ nvkm_device_pci_10de[] = {
 	{ 0x0fc6, "GeForce GTX 650" },
 	{ 0x0fc8, "GeForce GT 740" },
 	{ 0x0fc9, "GeForce GT 730" },
-	{ 0x0fcd, "GeForce GT 755M" },
+	{ 0x0fcd, "GeForce GT 755M", nvkm_device_pci_10de_0fcd },
 	{ 0x0fce, "GeForce GT 640M LE" },
 	{ 0x0fd1, "GeForce GT 650M" },
 	{ 0x0fd2, "GeForce GT 640M", nvkm_device_pci_10de_0fd2 },
-- 
cgit v0.10.2


From b41c48571d504f293a9a6a6cfd6c8e5dcb4ccf83 Mon Sep 17 00:00:00 2001
From: Ilia Mirkin <imirkin@alum.mit.edu>
Date: Sat, 31 Oct 2015 15:06:11 -0400
Subject: drm/nouveau/pci: enable c800 magic for Medion Erazer X7827

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=91557
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/pci.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/pci.c
index c5fc909..e3c783d 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/device/pci.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/pci.c
@@ -684,6 +684,7 @@ nvkm_device_pci_10de_1189[] = {
 static const struct nvkm_device_pci_vendor
 nvkm_device_pci_10de_1199[] = {
 	{ 0x1458, 0xd001, "GeForce GTX 760" },
+	{ 0x1462, 0x1106, "GeForce GTX 780M", { .War00C800_0 = true } }, /* Medion Erazer X7827 */
 	{}
 };
 
-- 
cgit v0.10.2


From 09433f24e6e16e590b289489f1f4e16d92e79080 Mon Sep 17 00:00:00 2001
From: Ben Skeggs <bskeggs@redhat.com>
Date: Tue, 3 Nov 2015 10:17:49 +1000
Subject: drm/nouveau/abi16: remove unused argument from nouveau_abi16_get()

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/nouveau_abi16.c b/drivers/gpu/drm/nouveau/nouveau_abi16.c
index 0b3c8ab..c3dea33 100644
--- a/drivers/gpu/drm/nouveau/nouveau_abi16.c
+++ b/drivers/gpu/drm/nouveau/nouveau_abi16.c
@@ -33,7 +33,7 @@
 #include "nouveau_abi16.h"
 
 struct nouveau_abi16 *
-nouveau_abi16_get(struct drm_file *file_priv, struct drm_device *dev)
+nouveau_abi16_get(struct drm_file *file_priv)
 {
 	struct nouveau_cli *cli = nouveau_cli(file_priv);
 	mutex_lock(&cli->mutex);
@@ -236,7 +236,7 @@ nouveau_abi16_ioctl_channel_alloc(ABI16_IOCTL_ARGS)
 	struct drm_nouveau_channel_alloc *init = data;
 	struct nouveau_cli *cli = nouveau_cli(file_priv);
 	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct nouveau_abi16 *abi16 = nouveau_abi16_get(file_priv, dev);
+	struct nouveau_abi16 *abi16 = nouveau_abi16_get(file_priv);
 	struct nouveau_abi16_chan *chan;
 	struct nvif_device *device;
 	int ret;
@@ -342,7 +342,7 @@ int
 nouveau_abi16_ioctl_channel_free(ABI16_IOCTL_ARGS)
 {
 	struct drm_nouveau_channel_free *req = data;
-	struct nouveau_abi16 *abi16 = nouveau_abi16_get(file_priv, dev);
+	struct nouveau_abi16 *abi16 = nouveau_abi16_get(file_priv);
 	struct nouveau_abi16_chan *chan;
 
 	if (unlikely(!abi16))
@@ -359,7 +359,7 @@ int
 nouveau_abi16_ioctl_grobj_alloc(ABI16_IOCTL_ARGS)
 {
 	struct drm_nouveau_grobj_alloc *init = data;
-	struct nouveau_abi16 *abi16 = nouveau_abi16_get(file_priv, dev);
+	struct nouveau_abi16 *abi16 = nouveau_abi16_get(file_priv);
 	struct nouveau_abi16_chan *chan;
 	struct nouveau_abi16_ntfy *ntfy;
 	struct nvif_client *client;
@@ -452,7 +452,7 @@ nouveau_abi16_ioctl_notifierobj_alloc(ABI16_IOCTL_ARGS)
 {
 	struct drm_nouveau_notifierobj_alloc *info = data;
 	struct nouveau_drm *drm = nouveau_drm(dev);
-	struct nouveau_abi16 *abi16 = nouveau_abi16_get(file_priv, dev);
+	struct nouveau_abi16 *abi16 = nouveau_abi16_get(file_priv);
 	struct nouveau_abi16_chan *chan;
 	struct nouveau_abi16_ntfy *ntfy;
 	struct nvif_device *device = &abi16->device;
@@ -524,7 +524,7 @@ int
 nouveau_abi16_ioctl_gpuobj_free(ABI16_IOCTL_ARGS)
 {
 	struct drm_nouveau_gpuobj_free *fini = data;
-	struct nouveau_abi16 *abi16 = nouveau_abi16_get(file_priv, dev);
+	struct nouveau_abi16 *abi16 = nouveau_abi16_get(file_priv);
 	struct nouveau_abi16_chan *chan;
 	struct nouveau_abi16_ntfy *ntfy;
 	int ret = -ENOENT;
diff --git a/drivers/gpu/drm/nouveau/nouveau_abi16.h b/drivers/gpu/drm/nouveau/nouveau_abi16.h
index 6584557..fa1d856 100644
--- a/drivers/gpu/drm/nouveau/nouveau_abi16.h
+++ b/drivers/gpu/drm/nouveau/nouveau_abi16.h
@@ -33,8 +33,7 @@ struct nouveau_abi16 {
 	u64 handles;
 };
 
-struct nouveau_drm;
-struct nouveau_abi16 *nouveau_abi16_get(struct drm_file *, struct drm_device *);
+struct nouveau_abi16 *nouveau_abi16_get(struct drm_file *);
 int  nouveau_abi16_put(struct nouveau_abi16 *, int);
 void nouveau_abi16_fini(struct nouveau_abi16 *);
 s32  nouveau_abi16_swclass(struct nouveau_drm *);
diff --git a/drivers/gpu/drm/nouveau/nouveau_gem.c b/drivers/gpu/drm/nouveau/nouveau_gem.c
index a108cc3..a0865c4 100644
--- a/drivers/gpu/drm/nouveau/nouveau_gem.c
+++ b/drivers/gpu/drm/nouveau/nouveau_gem.c
@@ -668,7 +668,7 @@ int
 nouveau_gem_ioctl_pushbuf(struct drm_device *dev, void *data,
 			  struct drm_file *file_priv)
 {
-	struct nouveau_abi16 *abi16 = nouveau_abi16_get(file_priv, dev);
+	struct nouveau_abi16 *abi16 = nouveau_abi16_get(file_priv);
 	struct nouveau_cli *cli = nouveau_cli(file_priv);
 	struct nouveau_abi16_chan *temp;
 	struct nouveau_drm *drm = nouveau_drm(dev);
-- 
cgit v0.10.2


From 786a57ef2cebb2d09d7f152b0ed4f1da1d368073 Mon Sep 17 00:00:00 2001
From: Ben Skeggs <bskeggs@redhat.com>
Date: Tue, 3 Nov 2015 10:55:45 +1000
Subject: drm/nouveau/abi16: introduce locked variant of nouveau_abi16_get()

USIF already takes the client mutex, but will need access to ABI16 data
in order to provide some limited interoperability.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/nouveau_abi16.c b/drivers/gpu/drm/nouveau/nouveau_abi16.c
index c3dea33..998f5cb 100644
--- a/drivers/gpu/drm/nouveau/nouveau_abi16.c
+++ b/drivers/gpu/drm/nouveau/nouveau_abi16.c
@@ -32,11 +32,10 @@
 #include "nouveau_chan.h"
 #include "nouveau_abi16.h"
 
-struct nouveau_abi16 *
-nouveau_abi16_get(struct drm_file *file_priv)
+static struct nouveau_abi16 *
+nouveau_abi16(struct drm_file *file_priv)
 {
 	struct nouveau_cli *cli = nouveau_cli(file_priv);
-	mutex_lock(&cli->mutex);
 	if (!cli->abi16) {
 		struct nouveau_abi16 *abi16;
 		cli->abi16 = abi16 = kzalloc(sizeof(*abi16), GFP_KERNEL);
@@ -59,12 +58,21 @@ nouveau_abi16_get(struct drm_file *file_priv)
 			kfree(cli->abi16);
 			cli->abi16 = NULL;
 		}
-
-		mutex_unlock(&cli->mutex);
 	}
 	return cli->abi16;
 }
 
+struct nouveau_abi16 *
+nouveau_abi16_get(struct drm_file *file_priv)
+{
+	struct nouveau_cli *cli = nouveau_cli(file_priv);
+	mutex_lock(&cli->mutex);
+	if (nouveau_abi16(file_priv))
+		return cli->abi16;
+	mutex_unlock(&cli->mutex);
+	return NULL;
+}
+
 int
 nouveau_abi16_put(struct nouveau_abi16 *abi16, int ret)
 {
-- 
cgit v0.10.2


From 2621a41647fe783be809e789faa5d8b6b06c8072 Mon Sep 17 00:00:00 2001
From: Ben Skeggs <bskeggs@redhat.com>
Date: Tue, 3 Nov 2015 11:21:43 +1000
Subject: drm/nouveau/abi16: implement limited interoperability with usif/nvif

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/nouveau_abi16.c b/drivers/gpu/drm/nouveau/nouveau_abi16.c
index 998f5cb..7f50cf5 100644
--- a/drivers/gpu/drm/nouveau/nouveau_abi16.c
+++ b/drivers/gpu/drm/nouveau/nouveau_abi16.c
@@ -25,6 +25,7 @@
 #include <nvif/driver.h>
 #include <nvif/ioctl.h>
 #include <nvif/class.h>
+#include <nvif/unpack.h>
 
 #include "nouveau_drm.h"
 #include "nouveau_dma.h"
@@ -347,6 +348,44 @@ nouveau_abi16_chan(struct nouveau_abi16 *abi16, int channel)
 }
 
 int
+nouveau_abi16_usif(struct drm_file *file_priv, void *data, u32 size)
+{
+	union {
+		struct nvif_ioctl_v0 v0;
+	} *args = data;
+	struct nouveau_abi16_chan *chan;
+	struct nouveau_abi16 *abi16;
+	int ret;
+
+	if (nvif_unpack(args->v0, 0, 0, true)) {
+		switch (args->v0.type) {
+		case NVIF_IOCTL_V0_NEW:
+		case NVIF_IOCTL_V0_MTHD:
+		case NVIF_IOCTL_V0_SCLASS:
+			break;
+		default:
+			return -EACCES;
+		}
+	} else
+		return ret;
+
+	if (!(abi16 = nouveau_abi16(file_priv)))
+		return -ENOMEM;
+
+	if (args->v0.token != ~0ULL) {
+		if (!(chan = nouveau_abi16_chan(abi16, args->v0.token)))
+			return -EINVAL;
+		args->v0.object = nvif_handle(&chan->chan->user);
+		args->v0.owner  = NVIF_IOCTL_V0_OWNER_ANY;
+		return 0;
+	}
+
+	args->v0.object = nvif_handle(&abi16->device.object);
+	args->v0.owner  = NVIF_IOCTL_V0_OWNER_ANY;
+	return 0;
+}
+
+int
 nouveau_abi16_ioctl_channel_free(ABI16_IOCTL_ARGS)
 {
 	struct drm_nouveau_channel_free *req = data;
diff --git a/drivers/gpu/drm/nouveau/nouveau_abi16.h b/drivers/gpu/drm/nouveau/nouveau_abi16.h
index fa1d856..841cc55 100644
--- a/drivers/gpu/drm/nouveau/nouveau_abi16.h
+++ b/drivers/gpu/drm/nouveau/nouveau_abi16.h
@@ -37,6 +37,7 @@ struct nouveau_abi16 *nouveau_abi16_get(struct drm_file *);
 int  nouveau_abi16_put(struct nouveau_abi16 *, int);
 void nouveau_abi16_fini(struct nouveau_abi16 *);
 s32  nouveau_abi16_swclass(struct nouveau_drm *);
+int  nouveau_abi16_usif(struct drm_file *, void *data, u32 size);
 
 #define NOUVEAU_GEM_DOMAIN_VRAM      (1 << 1)
 #define NOUVEAU_GEM_DOMAIN_GART      (1 << 2)
diff --git a/drivers/gpu/drm/nouveau/nouveau_usif.c b/drivers/gpu/drm/nouveau/nouveau_usif.c
index cb1182d..89dc4ce 100644
--- a/drivers/gpu/drm/nouveau/nouveau_usif.c
+++ b/drivers/gpu/drm/nouveau/nouveau_usif.c
@@ -24,6 +24,7 @@
 
 #include "nouveau_drm.h"
 #include "nouveau_usif.h"
+#include "nouveau_abi16.h"
 
 #include <nvif/notify.h>
 #include <nvif/unpack.h>
@@ -316,11 +317,21 @@ usif_ioctl(struct drm_file *filp, void __user *user, u32 argc)
 	} else
 		goto done;
 
+	/* USIF slightly abuses some return-only ioctl members in order
+	 * to provide interoperability with the older ABI16 objects
+	 */
 	mutex_lock(&cli->mutex);
+	if (argv->v0.route) {
+		if (ret = -EINVAL, argv->v0.route == 0xff)
+			ret = nouveau_abi16_usif(filp, argv, argc);
+		if (ret) {
+			mutex_unlock(&cli->mutex);
+			goto done;
+		}
+	}
+
 	switch (argv->v0.type) {
 	case NVIF_IOCTL_V0_NEW:
-		/* ... except if we're creating children */
-		argv->v0.owner = NVIF_IOCTL_V0_OWNER_ANY;
 		ret = usif_object_new(filp, data, size, argv, argc);
 		break;
 	case NVIF_IOCTL_V0_NTFY_NEW:
-- 
cgit v0.10.2


From 79ef5dca5e5cd5a33662d64c927c1b9786d4edee Mon Sep 17 00:00:00 2001
From: Ben Skeggs <bskeggs@redhat.com>
Date: Tue, 3 Nov 2015 12:40:13 +1000
Subject: drm/nouveau: bump patchlevel to indicate availability of abi16/nvif
 interop

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.h b/drivers/gpu/drm/nouveau/nouveau_drm.h
index 51027a9..3050042e 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drm.h
+++ b/drivers/gpu/drm/nouveau/nouveau_drm.h
@@ -10,7 +10,7 @@
 
 #define DRIVER_MAJOR		1
 #define DRIVER_MINOR		3
-#define DRIVER_PATCHLEVEL	0
+#define DRIVER_PATCHLEVEL	1
 
 /*
  * 1.1.1:
@@ -33,6 +33,8 @@
  * 1.3.0:
  *      - NVIF ABI modified, safe because only (current) users are test
  *        programs that get directly linked with NVKM.
+ * 1.3.1:
+ *      - implemented limited ABI16/NVIF interop
  */
 
 #include <nvif/client.h>
-- 
cgit v0.10.2


From df7989fe1ef5ae39902deb4eed498cf8e51cb7df Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Mon, 2 Nov 2015 10:52:32 -0500
Subject: drm/amdgpu: don't VT switch on suspend
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c
index 8a122b1..8574bb2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fb.c
@@ -207,6 +207,7 @@ static int amdgpufb_create(struct drm_fb_helper *helper,
 	}
 
 	info->par = rfbdev;
+	info->skip_vt_switch = true;
 
 	ret = amdgpu_framebuffer_init(adev->ddev, &rfbdev->rfb, &mode_cmd, gobj);
 	if (ret) {
-- 
cgit v0.10.2


From fe295b27150afe5ddc5e8bb6670149970b33834d Mon Sep 17 00:00:00 2001
From: Dave Airlie <airlied@redhat.com>
Date: Tue, 3 Nov 2015 11:07:11 -0500
Subject: drm/amdgpu: don't oops on failure to load (v2)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

In two places amdgpu tries to tear down something it hasn't
initalised when failing. This is what happens when you
enable experimental support on topaz which then fails in
ring init.

This patch allows it to fail cleanly.

v2 (agd): split out scheduler change into a separate patch

Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
index e0b80cc..fec65f0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
@@ -69,6 +69,9 @@ void amdgpu_ctx_fini(struct amdgpu_ctx *ctx)
 	struct amdgpu_device *adev = ctx->adev;
 	unsigned i, j;
 
+	if (!adev)
+		return;
+
 	for (i = 0; i < AMDGPU_MAX_RINGS; ++i)
 		for (j = 0; j < AMDGPU_CTX_MAX_CS_PENDING; ++j)
 			fence_put(ctx->rings[i].fences[j]);
-- 
cgit v0.10.2


From 32544d021515f767e651648e5702472d48ab9436 Mon Sep 17 00:00:00 2001
From: Dave Airlie <airlied@redhat.com>
Date: Tue, 3 Nov 2015 11:10:03 -0500
Subject: drm/amd/scheduler: don't oops on failure to load
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

In two places amdgpu tries to tear down something it hasn't
initalised when failing. This is what happens when you
enable experimental support on topaz which then fails in
ring init.

This patch allows it to fail cleanly.

agd: Split out from from the original patch since the
scheduler is a driver independent.

Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org

diff --git a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c
index 8dd7316..b3e84d0 100644
--- a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c
+++ b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c
@@ -468,5 +468,6 @@ int amd_sched_init(struct amd_gpu_scheduler *sched,
  */
 void amd_sched_fini(struct amd_gpu_scheduler *sched)
 {
-	kthread_stop(sched->thread);
+	if (sched->thread)
+		kthread_stop(sched->thread);
 }
-- 
cgit v0.10.2


From a3d08fa56b2b867460e58ecdb649b49804aaa5fd Mon Sep 17 00:00:00 2001
From: Flora Cui <Flora.Cui@amd.com>
Date: Mon, 2 Nov 2015 21:15:55 +0800
Subject: drm/amdgpu: extract common code in vi_common_early_init

Change-Id: I9ed25353c559e27bc1b1d5b50f977b0ff03de87f
Signed-off-by: Flora Cui <Flora.Cui@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/vi.c b/drivers/gpu/drm/amd/amdgpu/vi.c
index 9904761..2e52f3e 100644
--- a/drivers/gpu/drm/amd/amdgpu/vi.c
+++ b/drivers/gpu/drm/amd/amdgpu/vi.c
@@ -1410,8 +1410,6 @@ static int vi_common_early_init(void *handle)
 		adev->cg_flags = 0;
 		adev->pg_flags = 0;
 		adev->external_rev_id = 0x1;
-		if (amdgpu_smc_load_fw && smc_enabled)
-			adev->firmware.smu_load = true;
 		break;
 	case CHIP_FIJI:
 	case CHIP_TONGA:
@@ -1419,8 +1417,6 @@ static int vi_common_early_init(void *handle)
 		adev->cg_flags = 0;
 		adev->pg_flags = 0;
 		adev->external_rev_id = adev->rev_id + 0x14;
-		if (amdgpu_smc_load_fw && smc_enabled)
-			adev->firmware.smu_load = true;
 		break;
 	case CHIP_CARRIZO:
 	case CHIP_STONEY:
@@ -1429,14 +1425,15 @@ static int vi_common_early_init(void *handle)
 		/* Disable UVD pg */
 		adev->pg_flags = /* AMDGPU_PG_SUPPORT_UVD | */AMDGPU_PG_SUPPORT_VCE;
 		adev->external_rev_id = adev->rev_id + 0x1;
-		if (amdgpu_smc_load_fw && smc_enabled)
-			adev->firmware.smu_load = true;
 		break;
 	default:
 		/* FIXME: not supported yet */
 		return -EINVAL;
 	}
 
+	if (amdgpu_smc_load_fw && smc_enabled)
+		adev->firmware.smu_load = true;
+
 	return 0;
 }
 
-- 
cgit v0.10.2


From b6bc28ff2995f5c2c29dafc12a7e716de620cc8d Mon Sep 17 00:00:00 2001
From: Flora Cui <Flora.Cui@amd.com>
Date: Mon, 2 Nov 2015 21:21:34 +0800
Subject: drm/amdgpu: update Fiji's rev id

Change-Id: I0018e2b72feb771683c57960ba3ce942bec5d3ab
Signed-off-by: Flora Cui <Flora.Cui@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/vi.c b/drivers/gpu/drm/amd/amdgpu/vi.c
index 2e52f3e..b4f1fbc 100644
--- a/drivers/gpu/drm/amd/amdgpu/vi.c
+++ b/drivers/gpu/drm/amd/amdgpu/vi.c
@@ -1412,6 +1412,11 @@ static int vi_common_early_init(void *handle)
 		adev->external_rev_id = 0x1;
 		break;
 	case CHIP_FIJI:
+		adev->has_uvd = true;
+		adev->cg_flags = 0;
+		adev->pg_flags = 0;
+		adev->external_rev_id = adev->rev_id + 0x3c;
+		break;
 	case CHIP_TONGA:
 		adev->has_uvd = true;
 		adev->cg_flags = 0;
-- 
cgit v0.10.2


From a7ca8ef93001eaab4c5c7f4f3accfbe6ffcf5f26 Mon Sep 17 00:00:00 2001
From: Flora Cui <Flora.Cui@amd.com>
Date: Mon, 2 Nov 2015 14:27:45 +0800
Subject: drm/amdgpu: update Fiji's Golden setting

Change-Id: Ic3f3bfce4767cc05d04f6eb24e22a0f3e7ceacaa
Signed-off-by: Flora Cui <Flora.Cui@amd.com>
Reviewed-by: Jammy Zhou <Jammy.Zhou@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index f092400..6776cf7 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -235,11 +235,13 @@ static const u32 fiji_golden_common_all[] =
 	mmGRBM_GFX_INDEX, 0xffffffff, 0xe0000000,
 	mmPA_SC_RASTER_CONFIG, 0xffffffff, 0x3a00161a,
 	mmPA_SC_RASTER_CONFIG_1, 0xffffffff, 0x0000002e,
-	mmGB_ADDR_CONFIG, 0xffffffff, 0x12011003,
+	mmGB_ADDR_CONFIG, 0xffffffff, 0x22011003,
 	mmSPI_RESOURCE_RESERVE_CU_0, 0xffffffff, 0x00000800,
 	mmSPI_RESOURCE_RESERVE_CU_1, 0xffffffff, 0x00000800,
 	mmSPI_RESOURCE_RESERVE_EN_CU_0, 0xffffffff, 0x00007FBF,
-	mmSPI_RESOURCE_RESERVE_EN_CU_1, 0xffffffff, 0x00007FAF
+	mmSPI_RESOURCE_RESERVE_EN_CU_1, 0xffffffff, 0x00007FAF,
+	mmGRBM_GFX_INDEX, 0xffffffff, 0xe0000000,
+	mmSPI_CONFIG_CNTL_1, 0x0000000f, 0x00000009,
 };
 
 static const u32 golden_settings_fiji_a10[] =
@@ -247,24 +249,26 @@ static const u32 golden_settings_fiji_a10[] =
 	mmCB_HW_CONTROL_3, 0x000001ff, 0x00000040,
 	mmDB_DEBUG2, 0xf00fffff, 0x00000400,
 	mmPA_SC_ENHANCE, 0xffffffff, 0x20000001,
-	mmPA_SC_FIFO_DEPTH_CNTL, 0x000003ff, 0x00000100,
 	mmPA_SC_LINE_STIPPLE_STATE, 0x0000ff0f, 0x00000000,
+	mmRLC_CGCG_CGLS_CTRL, 0x00000003, 0x0001003c,
+	mmSQ_RANDOM_WAVE_PRI, 0x001fffff, 0x000006fd,
 	mmTA_CNTL_AUX, 0x000f000f, 0x000b0000,
-	mmTCC_CTRL, 0x00100000, 0xf30fff7f,
+	mmTCC_CTRL, 0x00100000, 0xf31fff7f,
+	mmTCC_EXE_DISABLE, 0x00000002, 0x00000002,
 	mmTCP_ADDR_CONFIG, 0x000003ff, 0x000000ff,
-	mmTCP_CHAN_STEER_HI, 0xffffffff, 0x7d6cf5e4,
-	mmTCP_CHAN_STEER_LO, 0xffffffff, 0x3928b1a0,
+	mmVGT_RESET_DEBUG, 0x00000004, 0x00000004,
 };
 
 static const u32 fiji_mgcg_cgcg_init[] =
 {
-	mmRLC_CGTT_MGCG_OVERRIDE, 0xffffffff, 0xffffffc0,
+	mmRLC_CGTT_MGCG_OVERRIDE, 0xffffffff, 0xffffffff,
 	mmGRBM_GFX_INDEX, 0xffffffff, 0xe0000000,
 	mmCB_CGTT_SCLK_CTRL, 0xffffffff, 0x00000100,
 	mmCGTT_BCI_CLK_CTRL, 0xffffffff, 0x00000100,
 	mmCGTT_CP_CLK_CTRL, 0xffffffff, 0x00000100,
 	mmCGTT_CPC_CLK_CTRL, 0xffffffff, 0x00000100,
 	mmCGTT_CPF_CLK_CTRL, 0xffffffff, 0x40000100,
+	mmCGTT_DRM_CLK_CTRL0, 0xffffffff, 0x00600100,
 	mmCGTT_GDS_CLK_CTRL, 0xffffffff, 0x00000100,
 	mmCGTT_IA_CLK_CTRL, 0xffffffff, 0x06000100,
 	mmCGTT_PA_CLK_CTRL, 0xffffffff, 0x00000100,
@@ -292,6 +296,10 @@ static const u32 fiji_mgcg_cgcg_init[] =
 	mmCGTS_SM_CTRL_REG, 0xffffffff, 0x96e00200,
 	mmCP_RB_WPTR_POLL_CNTL, 0xffffffff, 0x00900100,
 	mmRLC_CGCG_CGLS_CTRL, 0xffffffff, 0x0020003c,
+	mmPCIE_INDEX, 0xffffffff, 0x0140001c,
+	mmPCIE_DATA, 0x000f0000, 0x00000000,
+	mmCGTT_DRM_CLK_CTRL0, 0xff000fff, 0x00000100,
+	mmHDP_XDP_CGTT_BLK_CTRL, 0xc0000fff, 0x00000104,
 	mmCP_MEM_SLP_CNTL, 0x00000001, 0x00000001,
 };
 
-- 
cgit v0.10.2


From bf248ca1f5c7ba1e535ba4bd517a15a1ae965c69 Mon Sep 17 00:00:00 2001
From: Daniel Stone <daniels@collabora.com>
Date: Tue, 3 Nov 2015 21:42:31 +0000
Subject: drm/i915: Fix locking around GuC firmware load

The GuC firmware load requires struct_mutex to create a GEM object,
but this collides badly with request_firmware. Move struct_mutex
locking down into the loader itself, so we don't hold it across the
entire load process, including request_firmware.

[   20.451400] ======================================================
[   20.451420] [ INFO: possible circular locking dependency detected ]
[   20.451441] 4.3.0-rc5+ #1 Tainted: G        W
[   20.451457] -------------------------------------------------------
[   20.451477] plymouthd/371 is trying to acquire lock:
[   20.451494]  (&dev->struct_mutex){+.+.+.}, at: [<ffffffffa0093c62>]
drm_gem_mmap+0x112/0x290 [drm]
[   20.451538]
               but task is already holding lock:
[   20.451557]  (&mm->mmap_sem){++++++}, at: [<ffffffff811fd9ac>]
vm_mmap_pgoff+0x8c/0xf0
[   20.451591]
               which lock already depends on the new lock.

[   20.451617]
               the existing dependency chain (in reverse order) is:
[   20.451640]
               -> #3 (&mm->mmap_sem){++++++}:
[   20.451661]        [<ffffffff8110644e>] lock_acquire+0xce/0x1c0
[   20.451683]        [<ffffffff8120ec9a>] __might_fault+0x7a/0xa0
[   20.451705]        [<ffffffff8127e34e>] filldir+0x9e/0x130
[   20.451726]        [<ffffffff81295b86>] dcache_readdir+0x186/0x230
[   20.451748]        [<ffffffff8127e117>] iterate_dir+0x97/0x130
[   20.451769]        [<ffffffff8127e66a>] SyS_getdents+0x9a/0x130
[   20.451790]        [<ffffffff8184f2f2>] entry_SYSCALL_64_fastpath+0x12/0x76
[   20.451829]
               -> #2 (&sb->s_type->i_mutex_key#2){+.+.+.}:
[   20.451852]        [<ffffffff8110644e>] lock_acquire+0xce/0x1c0
[   20.451872]        [<ffffffff8184b516>] mutex_lock_nested+0x86/0x400
[   20.451893]        [<ffffffff81277790>] walk_component+0x1d0/0x2a0
[   20.451914]        [<ffffffff812779f0>] link_path_walk+0x190/0x5a0
[   20.451935]        [<ffffffff8127803b>] path_openat+0xab/0x1260
[   20.451955]        [<ffffffff8127a651>] do_filp_open+0x91/0x100
[   20.451975]        [<ffffffff81267e67>] file_open_name+0xf7/0x150
[   20.451995]        [<ffffffff81267ef3>] filp_open+0x33/0x60
[   20.452014]        [<ffffffff8157e1e7>] _request_firmware+0x277/0x880
[   20.452038]        [<ffffffff8157e9e4>] request_firmware_work_func+0x34/0x80
[   20.452060]        [<ffffffff810c7020>] process_one_work+0x230/0x680
[   20.452082]        [<ffffffff810c74be>] worker_thread+0x4e/0x450
[   20.452102]        [<ffffffff810ce511>] kthread+0x101/0x120
[   20.452121]        [<ffffffff8184f66f>] ret_from_fork+0x3f/0x70
[   20.452140]
               -> #1 (umhelper_sem){++++.+}:
[   20.452159]        [<ffffffff8110644e>] lock_acquire+0xce/0x1c0
[   20.452178]        [<ffffffff8184c5c1>] down_read+0x51/0xa0
[   20.452197]        [<ffffffff810c203b>]
usermodehelper_read_trylock+0x5b/0x130
[   20.452221]        [<ffffffff8157e147>] _request_firmware+0x1d7/0x880
[   20.452242]        [<ffffffff8157e821>] request_firmware+0x31/0x50
[   20.452262]        [<ffffffffa01b54a4>]
intel_guc_ucode_init+0xf4/0x400 [i915]
[   20.452305]        [<ffffffffa0213913>] i915_driver_load+0xd63/0x16e0 [i915]
[   20.452343]        [<ffffffffa00987d9>] drm_dev_register+0xa9/0xc0 [drm]
[   20.452369]        [<ffffffffa009ae3d>] drm_get_pci_dev+0x8d/0x1e0 [drm]
[   20.452396]        [<ffffffffa01521e4>] i915_pci_probe+0x34/0x50 [i915]
[   20.452421]        [<ffffffff81464675>] local_pci_probe+0x45/0xa0
[   20.452443]        [<ffffffff81465a6d>] pci_device_probe+0xfd/0x140
[   20.452464]        [<ffffffff8156a2e4>] driver_probe_device+0x224/0x480
[   20.452486]        [<ffffffff8156a5c8>] __driver_attach+0x88/0x90
[   20.452505]        [<ffffffff81567cf3>] bus_for_each_dev+0x73/0xc0
[   20.452526]        [<ffffffff81569a7e>] driver_attach+0x1e/0x20
[   20.452546]        [<ffffffff815695ae>] bus_add_driver+0x1ee/0x280
[   20.452566]        [<ffffffff8156b100>] driver_register+0x60/0xe0
[   20.453197]        [<ffffffff81464050>] __pci_register_driver+0x60/0x70
[   20.453845]        [<ffffffffa009b070>] drm_pci_init+0xe0/0x110 [drm]
[   20.454497]        [<ffffffffa027f092>] 0xffffffffa027f092
[   20.455156]        [<ffffffff81002123>] do_one_initcall+0xb3/0x200
[   20.455796]        [<ffffffff811d8c01>] do_init_module+0x5f/0x1e7
[   20.456434]        [<ffffffff8114c4e6>] load_module+0x2126/0x27d0
[   20.457071]        [<ffffffff8114cdf9>] SyS_finit_module+0xb9/0xf0
[   20.457738]        [<ffffffff8184f2f2>] entry_SYSCALL_64_fastpath+0x12/0x76
[   20.458370]
               -> #0 (&dev->struct_mutex){+.+.+.}:
[   20.459773]        [<ffffffff8110584f>] __lock_acquire+0x191f/0x1ba0
[   20.460451]        [<ffffffff8110644e>] lock_acquire+0xce/0x1c0
[   20.461074]        [<ffffffffa0093c88>] drm_gem_mmap+0x138/0x290 [drm]
[   20.461693]        [<ffffffff8121a5ec>] mmap_region+0x3ec/0x670
[   20.462298]        [<ffffffff8121abb2>] do_mmap+0x342/0x420
[   20.462901]        [<ffffffff811fd9d2>] vm_mmap_pgoff+0xb2/0xf0
[   20.463532]        [<ffffffff81218f62>] SyS_mmap_pgoff+0x1f2/0x290
[   20.464118]        [<ffffffff8102187b>] SyS_mmap+0x1b/0x30
[   20.464702]        [<ffffffff8184f2f2>] entry_SYSCALL_64_fastpath+0x12/0x76
[   20.465289]
               other info that might help us debug this:

[   20.467179] Chain exists of:
                 &dev->struct_mutex --> &sb->s_type->i_mutex_key#2 -->
&mm->mmap_sem

[   20.468928]  Possible unsafe locking scenario:

[   20.470161]        CPU0                    CPU1
[   20.470745]        ----                    ----
[   20.471325]   lock(&mm->mmap_sem);
[   20.471902]                                lock(&sb->s_type->i_mutex_key#2);
[   20.472538]                                lock(&mm->mmap_sem);
[   20.473118]   lock(&dev->struct_mutex);
[   20.473704]
                *** DEADLOCK ***

Signed-off-by: Daniel Stone <daniels@collabora.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>

diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index 2336af9..b4741d1 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -406,10 +406,7 @@ static int i915_load_modeset_init(struct drm_device *dev)
 	 * working irqs for e.g. gmbus and dp aux transfers. */
 	intel_modeset_init(dev);
 
-	/* intel_guc_ucode_init() needs the mutex to allocate GEM objects */
-	mutex_lock(&dev->struct_mutex);
 	intel_guc_ucode_init(dev);
-	mutex_unlock(&dev->struct_mutex);
 
 	ret = i915_gem_init(dev);
 	if (ret)
@@ -452,9 +449,7 @@ cleanup_gem:
 	i915_gem_context_fini(dev);
 	mutex_unlock(&dev->struct_mutex);
 cleanup_irq:
-	mutex_lock(&dev->struct_mutex);
 	intel_guc_ucode_fini(dev);
-	mutex_unlock(&dev->struct_mutex);
 	drm_irq_uninstall(dev);
 cleanup_gem_stolen:
 	i915_gem_cleanup_stolen(dev);
@@ -1193,8 +1188,8 @@ int i915_driver_unload(struct drm_device *dev)
 	/* Flush any outstanding unpin_work. */
 	flush_workqueue(dev_priv->wq);
 
-	mutex_lock(&dev->struct_mutex);
 	intel_guc_ucode_fini(dev);
+	mutex_lock(&dev->struct_mutex);
 	i915_gem_cleanup_ringbuffer(dev);
 	i915_gem_context_fini(dev);
 	mutex_unlock(&dev->struct_mutex);
diff --git a/drivers/gpu/drm/i915/intel_guc_loader.c b/drivers/gpu/drm/i915/intel_guc_loader.c
index a17b6a5..3541f76 100644
--- a/drivers/gpu/drm/i915/intel_guc_loader.c
+++ b/drivers/gpu/drm/i915/intel_guc_loader.c
@@ -504,7 +504,9 @@ static void guc_fw_fetch(struct drm_device *dev, struct intel_guc_fw *guc_fw)
 			guc_fw->guc_fw_major_found, guc_fw->guc_fw_minor_found,
 			guc_fw->guc_fw_major_wanted, guc_fw->guc_fw_minor_wanted);
 
+	mutex_lock(&dev->struct_mutex);
 	obj = i915_gem_object_create_from_data(dev, fw->data, fw->size);
+	mutex_unlock(&dev->struct_mutex);
 	if (IS_ERR_OR_NULL(obj)) {
 		err = obj ? PTR_ERR(obj) : -ENOMEM;
 		goto fail;
@@ -540,8 +542,6 @@ fail:
  * @dev:	drm device
  *
  * Called early during driver load, but after GEM is initialised.
- * The device struct_mutex must be held by the caller, as we're
- * going to allocate a GEM object to hold the firmware image.
  *
  * The firmware will be transferred to the GuC's memory later,
  * when intel_guc_ucode_load() is called.
@@ -598,9 +598,11 @@ void intel_guc_ucode_fini(struct drm_device *dev)
 	direct_interrupts_to_host(dev_priv);
 	i915_guc_submission_fini(dev);
 
+	mutex_lock(&dev->struct_mutex);
 	if (guc_fw->guc_fw_obj)
 		drm_gem_object_unreference(&guc_fw->guc_fw_obj->base);
 	guc_fw->guc_fw_obj = NULL;
+	mutex_unlock(&dev->struct_mutex);
 
 	guc_fw->guc_fw_fetch_status = GUC_FIRMWARE_NONE;
 }
-- 
cgit v0.10.2


From 4cfdcd9cecc05646b08805ba449811a645b3f4a8 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@oracle.com>
Date: Wed, 4 Nov 2015 16:25:09 +0300
Subject: drm/amdgpu: cleanup on error in amdgpu_cs_ioctl()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We recently changed the locking in this function and now there is a
missing unlock on error.  Also there are some other resources that we
should probably release as well...

Fixes: f48b2659f521 ('drm/amdgpu: fix the broken vm->mutex V2')
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index 8727c30..233ac9c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -875,8 +875,10 @@ int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 		struct amdgpu_job *job;
 		struct amdgpu_ring * ring =  parser->ibs->ring;
 		job = kzalloc(sizeof(struct amdgpu_job), GFP_KERNEL);
-		if (!job)
-			return -ENOMEM;
+		if (!job) {
+			r = -ENOMEM;
+			goto out;
+		}
 		job->base.sched = &ring->sched;
 		job->base.s_entity = &parser->ctx->rings[ring->idx].entity;
 		job->adev = parser->adev;
-- 
cgit v0.10.2


From 424839a6a913f2d2f473ce6d5b6465aeddf694db Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Mon, 2 Nov 2015 16:25:10 +0100
Subject: drm/amdgpu: fix stoping the scheduler timeout
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

cancel_delayed_work_sync is forbidden in interrupt context.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c
index b3e84d0..89619a5 100644
--- a/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c
+++ b/drivers/gpu/drm/amd/scheduler/gpu_scheduler.c
@@ -338,7 +338,7 @@ static void amd_sched_process_job(struct fence *f, struct fence_cb *cb)
 	atomic_dec(&sched->hw_rq_count);
 	amd_sched_fence_signal(s_fence);
 	if (sched->timeout != MAX_SCHEDULE_TIMEOUT) {
-		cancel_delayed_work_sync(&s_fence->dwork);
+		cancel_delayed_work(&s_fence->dwork);
 		spin_lock_irqsave(&sched->fence_list_lock, flags);
 		list_del_init(&s_fence->list);
 		spin_unlock_irqrestore(&sched->fence_list_lock, flags);
-- 
cgit v0.10.2


From 24372447e5cfe65d43a462acbf201c7882551b84 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Tue, 3 Nov 2015 11:26:42 +0100
Subject: drm/amdgpu: fix fence fallback check
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Interrupts are notorious unreliable, enable the fallback at
a couple of more places.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
index b912539..2c485a2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
@@ -342,6 +342,7 @@ static bool amdgpu_fence_enable_signaling(struct fence *f)
 	fence->fence_wake.func = amdgpu_fence_check_signaled;
 	__add_wait_queue(&ring->fence_drv.fence_queue, &fence->fence_wake);
 	fence_get(f);
+	amdgpu_fence_schedule_check(ring);
 	FENCE_TRACE(&fence->base, "armed on ring %i!\n", ring->idx);
 	return true;
 }
@@ -367,6 +368,7 @@ static int amdgpu_fence_ring_wait_seq(struct amdgpu_ring *ring, uint64_t seq)
 	if (atomic64_read(&ring->fence_drv.last_seq) >= seq)
 		return 0;
 
+	amdgpu_fence_schedule_check(ring);
 	wait_event(ring->fence_drv.fence_queue, (
 		   (signaled = amdgpu_fence_seq_signaled(ring, seq))));
 
-- 
cgit v0.10.2


From a67bcd34f6257d5519971a84705e9101ed1160e8 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Mon, 26 Oct 2015 13:22:33 +0100
Subject: drm/amdgpu: remove now unused fence functions
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 510cade..e615fff 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -452,45 +452,6 @@ bool amdgpu_fence_need_sync(struct amdgpu_fence *fence,
 void amdgpu_fence_note_sync(struct amdgpu_fence *fence,
 			    struct amdgpu_ring *ring);
 
-static inline struct amdgpu_fence *amdgpu_fence_later(struct amdgpu_fence *a,
-						      struct amdgpu_fence *b)
-{
-	if (!a) {
-		return b;
-	}
-
-	if (!b) {
-		return a;
-	}
-
-	BUG_ON(a->ring != b->ring);
-
-	if (a->seq > b->seq) {
-		return a;
-	} else {
-		return b;
-	}
-}
-
-static inline bool amdgpu_fence_is_earlier(struct amdgpu_fence *a,
-					   struct amdgpu_fence *b)
-{
-	if (!a) {
-		return false;
-	}
-
-	if (!b) {
-		return true;
-	}
-
-	BUG_ON(a->ring != b->ring);
-
-	return a->seq < b->seq;
-}
-
-int amdgpu_user_fence_emit(struct amdgpu_ring *ring, struct amdgpu_user_fence *user,
-			   void *owner, struct amdgpu_fence **fence);
-
 /*
  * TTM.
  */
-- 
cgit v0.10.2


From 7a91d6cb3c8f93288865f98ffa03485aff6dbab4 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Tue, 27 Oct 2015 17:28:24 +0100
Subject: drm/amdgpu: remove AMDGPU_FENCE_OWNER_MOVE
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Moves are exclusive operations anyway, just use the undefined owner for those.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index e615fff..5961d63 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -405,7 +405,6 @@ struct amdgpu_fence_driver {
 /* some special values for the owner field */
 #define AMDGPU_FENCE_OWNER_UNDEFINED	((void*)0ul)
 #define AMDGPU_FENCE_OWNER_VM		((void*)1ul)
-#define AMDGPU_FENCE_OWNER_MOVE		((void*)2ul)
 
 #define AMDGPU_FENCE_FLAG_64BIT         (1 << 0)
 #define AMDGPU_FENCE_FLAG_INT           (1 << 1)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c
index c6489b1..a6697fd 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c
@@ -191,8 +191,8 @@ int amdgpu_sync_resv(struct amdgpu_device *adev,
 			 * for other VM updates and moves.
 			 */
 			fence_owner = amdgpu_sync_get_owner(f);
-			if ((owner != AMDGPU_FENCE_OWNER_MOVE) &&
-			    (fence_owner != AMDGPU_FENCE_OWNER_MOVE) &&
+			if ((owner != AMDGPU_FENCE_OWNER_UNDEFINED) &&
+			    (fence_owner != AMDGPU_FENCE_OWNER_UNDEFINED) &&
 			    ((owner == AMDGPU_FENCE_OWNER_VM) !=
 			     (fence_owner == AMDGPU_FENCE_OWNER_VM)))
 				continue;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index a089e69..81bb8e9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -1041,7 +1041,7 @@ int amdgpu_copy_buffer(struct amdgpu_ring *ring,
 	WARN_ON(ib->length_dw > num_dw);
 	r = amdgpu_sched_ib_submit_kernel_helper(adev, ring, ib, 1,
 						 &amdgpu_vm_free_job,
-						 AMDGPU_FENCE_OWNER_MOVE,
+						 AMDGPU_FENCE_OWNER_UNDEFINED,
 						 fence);
 	if (r)
 		goto error_free;
-- 
cgit v0.10.2


From a95e264254dca5b6bfb331d5902930d0787bd7e1 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Tue, 3 Nov 2015 12:21:57 +0100
Subject: drm/amdgpu: group together common fence implementation
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

And also add some missing function documentation. No functional change.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
index 2c485a2..257d722 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
@@ -137,42 +137,6 @@ int amdgpu_fence_emit(struct amdgpu_ring *ring, void *owner,
 }
 
 /**
- * amdgpu_fence_check_signaled - callback from fence_queue
- *
- * this function is called with fence_queue lock held, which is also used
- * for the fence locking itself, so unlocked variants are used for
- * fence_signal, and remove_wait_queue.
- */
-static int amdgpu_fence_check_signaled(wait_queue_t *wait, unsigned mode, int flags, void *key)
-{
-	struct amdgpu_fence *fence;
-	struct amdgpu_device *adev;
-	u64 seq;
-	int ret;
-
-	fence = container_of(wait, struct amdgpu_fence, fence_wake);
-	adev = fence->ring->adev;
-
-	/*
-	 * We cannot use amdgpu_fence_process here because we're already
-	 * in the waitqueue, in a call from wake_up_all.
-	 */
-	seq = atomic64_read(&fence->ring->fence_drv.last_seq);
-	if (seq >= fence->seq) {
-		ret = fence_signal_locked(&fence->base);
-		if (!ret)
-			FENCE_TRACE(&fence->base, "signaled from irq context\n");
-		else
-			FENCE_TRACE(&fence->base, "was already signaled\n");
-
-		__remove_wait_queue(&fence->ring->fence_drv.fence_queue, &fence->fence_wake);
-		fence_put(&fence->base);
-	} else
-		FENCE_TRACE(&fence->base, "pending\n");
-	return 0;
-}
-
-/**
  * amdgpu_fence_activity - check for fence activity
  *
  * @ring: pointer to struct amdgpu_ring
@@ -305,48 +269,6 @@ static bool amdgpu_fence_seq_signaled(struct amdgpu_ring *ring, u64 seq)
 	return false;
 }
 
-static bool amdgpu_fence_is_signaled(struct fence *f)
-{
-	struct amdgpu_fence *fence = to_amdgpu_fence(f);
-	struct amdgpu_ring *ring = fence->ring;
-
-	if (atomic64_read(&ring->fence_drv.last_seq) >= fence->seq)
-		return true;
-
-	amdgpu_fence_process(ring);
-
-	if (atomic64_read(&ring->fence_drv.last_seq) >= fence->seq)
-		return true;
-
-	return false;
-}
-
-/**
- * amdgpu_fence_enable_signaling - enable signalling on fence
- * @fence: fence
- *
- * This function is called with fence_queue lock held, and adds a callback
- * to fence_queue that checks if this fence is signaled, and if so it
- * signals the fence and removes itself.
- */
-static bool amdgpu_fence_enable_signaling(struct fence *f)
-{
-	struct amdgpu_fence *fence = to_amdgpu_fence(f);
-	struct amdgpu_ring *ring = fence->ring;
-
-	if (atomic64_read(&ring->fence_drv.last_seq) >= fence->seq)
-		return false;
-
-	fence->fence_wake.flags = 0;
-	fence->fence_wake.private = NULL;
-	fence->fence_wake.func = amdgpu_fence_check_signaled;
-	__add_wait_queue(&ring->fence_drv.fence_queue, &fence->fence_wake);
-	fence_get(f);
-	amdgpu_fence_schedule_check(ring);
-	FENCE_TRACE(&fence->base, "armed on ring %i!\n", ring->idx);
-	return true;
-}
-
 /*
  * amdgpu_ring_wait_seq_timeout - wait for seq of the specific ring to signal
  * @ring: ring to wait on for the seq number
@@ -733,6 +655,115 @@ void amdgpu_fence_driver_force_completion(struct amdgpu_device *adev)
 	}
 }
 
+/*
+ * Common fence implementation
+ */
+
+static const char *amdgpu_fence_get_driver_name(struct fence *fence)
+{
+	return "amdgpu";
+}
+
+static const char *amdgpu_fence_get_timeline_name(struct fence *f)
+{
+	struct amdgpu_fence *fence = to_amdgpu_fence(f);
+	return (const char *)fence->ring->name;
+}
+
+/**
+ * amdgpu_fence_is_signaled - test if fence is signaled
+ *
+ * @f: fence to test
+ *
+ * Test the fence sequence number if it is already signaled. If it isn't
+ * signaled start fence processing. Returns True if the fence is signaled.
+ */
+static bool amdgpu_fence_is_signaled(struct fence *f)
+{
+	struct amdgpu_fence *fence = to_amdgpu_fence(f);
+	struct amdgpu_ring *ring = fence->ring;
+
+	if (atomic64_read(&ring->fence_drv.last_seq) >= fence->seq)
+		return true;
+
+	amdgpu_fence_process(ring);
+
+	if (atomic64_read(&ring->fence_drv.last_seq) >= fence->seq)
+		return true;
+
+	return false;
+}
+
+/**
+ * amdgpu_fence_check_signaled - callback from fence_queue
+ *
+ * this function is called with fence_queue lock held, which is also used
+ * for the fence locking itself, so unlocked variants are used for
+ * fence_signal, and remove_wait_queue.
+ */
+static int amdgpu_fence_check_signaled(wait_queue_t *wait, unsigned mode, int flags, void *key)
+{
+	struct amdgpu_fence *fence;
+	struct amdgpu_device *adev;
+	u64 seq;
+	int ret;
+
+	fence = container_of(wait, struct amdgpu_fence, fence_wake);
+	adev = fence->ring->adev;
+
+	/*
+	 * We cannot use amdgpu_fence_process here because we're already
+	 * in the waitqueue, in a call from wake_up_all.
+	 */
+	seq = atomic64_read(&fence->ring->fence_drv.last_seq);
+	if (seq >= fence->seq) {
+		ret = fence_signal_locked(&fence->base);
+		if (!ret)
+			FENCE_TRACE(&fence->base, "signaled from irq context\n");
+		else
+			FENCE_TRACE(&fence->base, "was already signaled\n");
+
+		__remove_wait_queue(&fence->ring->fence_drv.fence_queue, &fence->fence_wake);
+		fence_put(&fence->base);
+	} else
+		FENCE_TRACE(&fence->base, "pending\n");
+	return 0;
+}
+
+/**
+ * amdgpu_fence_enable_signaling - enable signalling on fence
+ * @fence: fence
+ *
+ * This function is called with fence_queue lock held, and adds a callback
+ * to fence_queue that checks if this fence is signaled, and if so it
+ * signals the fence and removes itself.
+ */
+static bool amdgpu_fence_enable_signaling(struct fence *f)
+{
+	struct amdgpu_fence *fence = to_amdgpu_fence(f);
+	struct amdgpu_ring *ring = fence->ring;
+
+	if (atomic64_read(&ring->fence_drv.last_seq) >= fence->seq)
+		return false;
+
+	fence->fence_wake.flags = 0;
+	fence->fence_wake.private = NULL;
+	fence->fence_wake.func = amdgpu_fence_check_signaled;
+	__add_wait_queue(&ring->fence_drv.fence_queue, &fence->fence_wake);
+	fence_get(f);
+	amdgpu_fence_schedule_check(ring);
+	FENCE_TRACE(&fence->base, "armed on ring %i!\n", ring->idx);
+	return true;
+}
+
+const struct fence_ops amdgpu_fence_ops = {
+	.get_driver_name = amdgpu_fence_get_driver_name,
+	.get_timeline_name = amdgpu_fence_get_timeline_name,
+	.enable_signaling = amdgpu_fence_enable_signaling,
+	.signaled = amdgpu_fence_is_signaled,
+	.wait = fence_default_wait,
+	.release = NULL,
+};
 
 /*
  * Fence debugfs
@@ -783,22 +814,3 @@ int amdgpu_debugfs_fence_init(struct amdgpu_device *adev)
 #endif
 }
 
-static const char *amdgpu_fence_get_driver_name(struct fence *fence)
-{
-	return "amdgpu";
-}
-
-static const char *amdgpu_fence_get_timeline_name(struct fence *f)
-{
-	struct amdgpu_fence *fence = to_amdgpu_fence(f);
-	return (const char *)fence->ring->name;
-}
-
-const struct fence_ops amdgpu_fence_ops = {
-	.get_driver_name = amdgpu_fence_get_driver_name,
-	.get_timeline_name = amdgpu_fence_get_timeline_name,
-	.enable_signaling = amdgpu_fence_enable_signaling,
-	.signaled = amdgpu_fence_is_signaled,
-	.wait = fence_default_wait,
-	.release = NULL,
-};
-- 
cgit v0.10.2


From e02328f47bd75fde9decf9657ec7d769b370f857 Mon Sep 17 00:00:00 2001
From: Lukas Wunner <lukas@wunner.de>
Date: Tue, 8 Sep 2015 14:17:47 +0200
Subject: vga_switcheroo: Drop client power state VGA_SWITCHEROO_INIT

hda_intel.c:azx_probe() defers initialization of an audio controller
on the discrete GPU if the GPU is powered off. The power state of the
GPU is determined by calling vga_switcheroo_get_client_state().

vga_switcheroo_get_client_state() returns VGA_SWITCHEROO_INIT if
vga_switcheroo is not enabled, i.e. if no second GPU or no handler
has registered.

This can go wrong in the following scenario:
- Driver for the integrated GPU is not loaded.
- Driver for the discrete GPU registers with vga_switcheroo, uses driver
  power control to power down the GPU, handler cuts power to the GPU.
- Driver for the audio controller gets loaded after the GPU was powered
  down, calls vga_switcheroo_get_client_state() which returns
  VGA_SWITCHEROO_INIT instead of VGA_SWITCHEROO_OFF.
- Consequence: azx_probe() tries to initialize the audio controller even
  though the GPU is powered down.

The power state VGA_SWITCHEROO_INIT was introduced by c8e9cf7bb240
("vga_switcheroo: Add a helper function to get the client state").
It is not apparent what its benefit might be. The idea seems to
be to initialize the audio controller even if the power state is
VGA_SWITCHEROO_OFF (were vga_switcheroo enabled), but as shown
above this can fail.

Drop VGA_SWITCHEROO_INIT to solve this.

Acked-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Dave Airlie <airlied@redhat.com>

diff --git a/drivers/gpu/vga/vga_switcheroo.c b/drivers/gpu/vga/vga_switcheroo.c
index 56bbbd6..41edd5a 100644
--- a/drivers/gpu/vga/vga_switcheroo.c
+++ b/drivers/gpu/vga/vga_switcheroo.c
@@ -356,8 +356,6 @@ enum vga_switcheroo_state vga_switcheroo_get_client_state(struct pci_dev *pdev)
 	client = find_client_from_pci(&vgasr_priv.clients, pdev);
 	if (!client)
 		ret = VGA_SWITCHEROO_NOT_FOUND;
-	else if (!vgasr_priv.active)
-		ret = VGA_SWITCHEROO_INIT;
 	else
 		ret = client->pwr_state;
 	mutex_unlock(&vgasr_mutex);
diff --git a/include/linux/vga_switcheroo.h b/include/linux/vga_switcheroo.h
index 786bc93..69e1d4a1 100644
--- a/include/linux/vga_switcheroo.h
+++ b/include/linux/vga_switcheroo.h
@@ -39,10 +39,6 @@ struct pci_dev;
  * enum vga_switcheroo_state - client power state
  * @VGA_SWITCHEROO_OFF: off
  * @VGA_SWITCHEROO_ON: on
- * @VGA_SWITCHEROO_INIT: client has registered with vga_switcheroo but
- * 	vga_switcheroo is not enabled, i.e. no second client or no handler
- * 	has registered. Only used in vga_switcheroo_get_client_state() which
- * 	in turn is only called from hda_intel.c
  * @VGA_SWITCHEROO_NOT_FOUND: client has not registered with vga_switcheroo.
  * 	Only used in vga_switcheroo_get_client_state() which in turn is only
  * 	called from hda_intel.c
@@ -53,7 +49,6 @@ enum vga_switcheroo_state {
 	VGA_SWITCHEROO_OFF,
 	VGA_SWITCHEROO_ON,
 	/* below are referred only from vga_switcheroo_get_client_state() */
-	VGA_SWITCHEROO_INIT,
 	VGA_SWITCHEROO_NOT_FOUND,
 };
 
-- 
cgit v0.10.2


From 9807216f585fc69a6cad3668414a2936a84444c8 Mon Sep 17 00:00:00 2001
From: Vivek Kasireddy <vivek.kasireddy@intel.com>
Date: Thu, 29 Oct 2015 18:54:38 -0700
Subject: drm/i915: Skip fence installation for objects with rotated views (v4)

While pinning a fb object to the display plane, only install a fence
if the object is using a normal view. This corresponds with the
behavior found in i915_gem_object_do_pin() where the fencability
criteria is determined only for objects with normal views.

v2:
Look at the object's map_and_fenceable flag to determine whether to
install a fence or not (Chris).

v3:
Pin and unpin a fence only if the current view type is normal.

v4:
Extend the "view type is normal" check for pin_fence as well.

Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Ville Syrjala <ville.syrjala@linux.intel.com>
Signed-off-by: Vivek Kasireddy <vivek.kasireddy@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1446170078-20792-1-git-send-email-vivek.kasireddy@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 81358ca..6795df5 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -2389,22 +2389,24 @@ intel_pin_and_fence_fb_obj(struct drm_plane *plane,
 	 * framebuffer compression.  For simplicity, we always install
 	 * a fence as the cost is not that onerous.
 	 */
-	ret = i915_gem_object_get_fence(obj);
-	if (ret == -EDEADLK) {
-		/*
-		 * -EDEADLK means there are no free fences
-		 * no pending flips.
-		 *
-		 * This is propagated to atomic, but it uses
-		 * -EDEADLK to force a locking recovery, so
-		 * change the returned error to -EBUSY.
-		 */
-		ret = -EBUSY;
-		goto err_unpin;
-	} else if (ret)
-		goto err_unpin;
+	if (view.type == I915_GGTT_VIEW_NORMAL) {
+		ret = i915_gem_object_get_fence(obj);
+		if (ret == -EDEADLK) {
+			/*
+			 * -EDEADLK means there are no free fences
+			 * no pending flips.
+			 *
+			 * This is propagated to atomic, but it uses
+			 * -EDEADLK to force a locking recovery, so
+			 * change the returned error to -EBUSY.
+			 */
+			ret = -EBUSY;
+			goto err_unpin;
+		} else if (ret)
+			goto err_unpin;
 
-	i915_gem_object_pin_fence(obj);
+		i915_gem_object_pin_fence(obj);
+	}
 
 	dev_priv->mm.interruptible = true;
 	intel_runtime_pm_put(dev_priv);
@@ -2430,7 +2432,9 @@ static void intel_unpin_fb_obj(struct drm_framebuffer *fb,
 	ret = intel_fill_fb_ggtt_view(&view, fb, plane_state);
 	WARN_ONCE(ret, "Couldn't get view from plane state!");
 
-	i915_gem_object_unpin_fence(obj);
+	if (view.type == I915_GGTT_VIEW_NORMAL)
+		i915_gem_object_unpin_fence(obj);
+
 	i915_gem_object_unpin_from_display_plane(obj, &view);
 }
 
-- 
cgit v0.10.2


From 99106bc17e667989b4c0af0a6afcbd6ddbada8fb Mon Sep 17 00:00:00 2001
From: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Date: Thu, 5 Nov 2015 13:11:38 +0200
Subject: drm/i915: Do graphics device reset under forcewake

We have a timed release of a forcewake when using
I915_READ/WRITE macros. wait_for() macro will go to quite
long sleep if the first read doesn't satisfy the condition for
successful exit. With these two interacting, it is possible that
we lose the forcewake during the wait_for() and the subsequent read
will reaquire forcewake.

Further experiments with skl shows that when we lose forcewake,
we lose the reset request we submitted. So reset request register
is not power context saved.

Grab forcewakes for all engines before starting the reset/request
dance so that all requests stay valid for the duration of reset
requisition across all the engines.

v2: Add comment on power well sleeps. Wrap the reset handling
    under forcewake instead of just reset requests (Chris)

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=92774
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Tested-by: Tomi Sarvela <tomix.p.sarvela@intel.com> (v1, v2)
Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1446721898-1450-1-git-send-email-mika.kuoppala@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>

diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c
index 1663ea5..43cba12 100644
--- a/drivers/gpu/drm/i915/intel_uncore.c
+++ b/drivers/gpu/drm/i915/intel_uncore.c
@@ -1531,13 +1531,22 @@ static int (*intel_get_gpu_reset(struct drm_device *dev))(struct drm_device *)
 
 int intel_gpu_reset(struct drm_device *dev)
 {
+	struct drm_i915_private *dev_priv = to_i915(dev);
 	int (*reset)(struct drm_device *);
+	int ret;
 
 	reset = intel_get_gpu_reset(dev);
 	if (reset == NULL)
 		return -ENODEV;
 
-	return reset(dev);
+	/* If the power well sleeps during the reset, the reset
+	 * request may be dropped and never completes (causing -EIO).
+	 */
+	intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL);
+	ret = reset(dev);
+	intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL);
+
+	return ret;
 }
 
 bool intel_has_gpu_reset(struct drm_device *dev)
-- 
cgit v0.10.2


From b291681926a142958112eedde62823230d6afb84 Mon Sep 17 00:00:00 2001
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Date: Tue, 3 Nov 2015 08:31:41 +0100
Subject: drm/i915: Extend DSL readout fix to BDW and SKL.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Those platforms have the same bug as haswell, and the same fix applies
to them.

The original HSW fix that this extends is

commit 41b578fb0e8b930f2470d3f673b0fa279e77a7b8
Author: Jesse Barnes <jbarnes@virtuousgeek.org>
Date:   Tue Sep 22 12:15:54 2015 -0700

    drm/i915: workaround bad DSL readout v3

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Acked-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: stable@vger.kernel.org # v4.3
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=91579
Link: http://patchwork.freedesktop.org/patch/msgid/1446535913-31970-3-git-send-email-maarten.lankhorst@linux.intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 5d8ec7a..684a0d9 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -749,7 +749,7 @@ static int __intel_get_crtc_scanline(struct intel_crtc *crtc)
 	 * problem.  We may need to extend this to include other platforms,
 	 * but so far testing only shows the problem on HSW.
 	 */
-	if (IS_HASWELL(dev) && !position) {
+	if (HAS_DDI(dev) && !position) {
 		int i, temp;
 
 		for (i = 0; i < 100; i++) {
-- 
cgit v0.10.2


From 1b0e3a049efe471c399674fd954500ce97438d30 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Thu, 5 Nov 2015 23:04:11 +0200
Subject: drm/i915/skl: disable display side power well support for now
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The display power well support on this platform is in a somewhat broken
state atm, so disable it by default.

This in effect will get rid of incorrect assert WARNs about the CSR/DMC
firmware not being loaded during power well toggling. It also removes a
problem during driver loading where a register is accessed while its
backing power well is down, resulting in another WARN. Until we come up
with the root cause of the second problem and the proper fix for both
issues, keep all display side power wells on.

Also clarify a bit the option description.

Reported-by: Dave Airlie <airlied@redhat.com>
Reference: http://mid.gmane.org/CAPM=9tyjBQjSBTKa49cRr6SYkpNW7Pq-fUFznZZ8Y1snvvk7mA@mail.gmail.com
Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1446757451-2777-1-git-send-email-imre.deak@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>

diff --git a/drivers/gpu/drm/i915/i915_params.c b/drivers/gpu/drm/i915/i915_params.c
index ca9b8f6..96bb238 100644
--- a/drivers/gpu/drm/i915/i915_params.c
+++ b/drivers/gpu/drm/i915/i915_params.c
@@ -38,7 +38,7 @@ struct i915_params i915 __read_mostly = {
 	.enable_ppgtt = -1,
 	.enable_psr = 0,
 	.preliminary_hw_support = IS_ENABLED(CONFIG_DRM_I915_PRELIMINARY_HW_SUPPORT),
-	.disable_power_well = 1,
+	.disable_power_well = -1,
 	.enable_ips = 1,
 	.prefault_disable = 0,
 	.load_detect_test = 0,
@@ -127,7 +127,8 @@ MODULE_PARM_DESC(preliminary_hw_support,
 
 module_param_named_unsafe(disable_power_well, i915.disable_power_well, int, 0600);
 MODULE_PARM_DESC(disable_power_well,
-	"Disable the power well when possible (default: true)");
+	"Disable display power wells when possible "
+	"(-1=auto [default], 0=power wells always on, 1=power wells disabled when possible)");
 
 module_param_named_unsafe(enable_ips, i915.enable_ips, int, 0600);
 MODULE_PARM_DESC(enable_ips, "Enable IPS (default: true)");
diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c
index 0cfe4c1..1aedf27 100644
--- a/drivers/gpu/drm/i915/intel_runtime_pm.c
+++ b/drivers/gpu/drm/i915/intel_runtime_pm.c
@@ -1810,6 +1810,21 @@ static struct i915_power_well bxt_power_wells[] = {
 	}
 };
 
+static int
+sanitize_disable_power_well_option(const struct drm_i915_private *dev_priv,
+				   int disable_power_well)
+{
+	if (disable_power_well >= 0)
+		return !!disable_power_well;
+
+	if (IS_SKYLAKE(dev_priv)) {
+		DRM_DEBUG_KMS("Disabling display power well support\n");
+		return 0;
+	}
+
+	return 1;
+}
+
 #define set_power_wells(power_domains, __power_wells) ({		\
 	(power_domains)->power_wells = (__power_wells);			\
 	(power_domains)->power_well_count = ARRAY_SIZE(__power_wells);	\
@@ -1826,6 +1841,9 @@ int intel_power_domains_init(struct drm_i915_private *dev_priv)
 {
 	struct i915_power_domains *power_domains = &dev_priv->power_domains;
 
+	i915.disable_power_well = sanitize_disable_power_well_option(dev_priv,
+						     i915.disable_power_well);
+
 	mutex_init(&power_domains->lock);
 
 	/*
-- 
cgit v0.10.2


From 4e0858a682c008a2a8f5f4fb3aec705500575ae1 Mon Sep 17 00:00:00 2001
From: Thomas Hellstrom <thellstrom@vmware.com>
Date: Thu, 5 Nov 2015 02:18:55 -0800
Subject: drm/vmwgfx: Properly flush cursor updates and page-flips

With the introduction of the new command buffer mechanism,
proper care wasn't taken to flush cursor image updates and
event-less screen-target page-flips.

Fix this by introducing explicit flush points.

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: Brian Paul <brianp@vmware.com>

diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
index a94b24d..9fcd7f8 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
@@ -78,7 +78,7 @@ int vmw_cursor_update_image(struct vmw_private *dev_priv,
 	cmd->cursor.hotspotX = hotspotX;
 	cmd->cursor.hotspotY = hotspotY;
 
-	vmw_fifo_commit(dev_priv, cmd_size);
+	vmw_fifo_commit_flush(dev_priv, cmd_size);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c b/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c
index c22e2df..b1fc1c0 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c
@@ -717,6 +717,8 @@ static int vmw_stdu_crtc_page_flip(struct drm_crtc *crtc,
 						   &event->event.tv_usec,
 						   true);
 		vmw_fence_obj_unreference(&fence);
+	} else {
+		vmw_fifo_flush(dev_priv, false);
 	}
 
 	return ret;
-- 
cgit v0.10.2


From d2e8851a909c76c6f0b34553f2c9993808bc1d58 Mon Sep 17 00:00:00 2001
From: Thomas Hellstrom <thellstrom@vmware.com>
Date: Wed, 28 Oct 2015 19:07:35 +0100
Subject: drm/vmwgfx: Relax irq locking somewhat

Relax locking with the goal of reducing the number of locking cycles and
time spent with irqs disabled.

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: Sinclair Yeh <syeh@vmware.com>

diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
index d1c34ab..a09cf85 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
@@ -643,7 +643,7 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset)
 	init_waitqueue_head(&dev_priv->fence_queue);
 	init_waitqueue_head(&dev_priv->fifo_queue);
 	dev_priv->fence_queue_waiters = 0;
-	atomic_set(&dev_priv->fifo_queue_waiters, 0);
+	dev_priv->fifo_queue_waiters = 0;
 
 	dev_priv->used_memory_size = 0;
 
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
index 198c8b1..a8ae9df 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
@@ -440,13 +440,12 @@ struct vmw_private {
 	spinlock_t waiter_lock;
 	int fence_queue_waiters; /* Protected by waiter_lock */
 	int goal_queue_waiters; /* Protected by waiter_lock */
-	int cmdbuf_waiters; /* Protected by irq_lock */
-	int error_waiters; /* Protected by irq_lock */
-	atomic_t fifo_queue_waiters;
+	int cmdbuf_waiters; /* Protected by waiter_lock */
+	int error_waiters; /* Protected by waiter_lock */
+	int fifo_queue_waiters; /* Protected by waiter_lock */
 	uint32_t last_read_seqno;
-	spinlock_t irq_lock;
 	struct vmw_fence_manager *fman;
-	uint32_t irq_mask;
+	uint32_t irq_mask; /* Updates protected by waiter_lock */
 
 	/*
 	 * Device state
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c b/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c
index 0cbaf88..a8baf5f 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_fifo.c
@@ -252,7 +252,6 @@ static int vmw_fifo_wait(struct vmw_private *dev_priv,
 			 unsigned long timeout)
 {
 	long ret = 1L;
-	unsigned long irq_flags;
 
 	if (likely(!vmw_fifo_is_full(dev_priv, bytes)))
 		return 0;
@@ -262,16 +261,8 @@ static int vmw_fifo_wait(struct vmw_private *dev_priv,
 		return vmw_fifo_wait_noirq(dev_priv, bytes,
 					   interruptible, timeout);
 
-	spin_lock(&dev_priv->waiter_lock);
-	if (atomic_add_return(1, &dev_priv->fifo_queue_waiters) > 0) {
-		spin_lock_irqsave(&dev_priv->irq_lock, irq_flags);
-		outl(SVGA_IRQFLAG_FIFO_PROGRESS,
-		     dev_priv->io_start + VMWGFX_IRQSTATUS_PORT);
-		dev_priv->irq_mask |= SVGA_IRQFLAG_FIFO_PROGRESS;
-		vmw_write(dev_priv, SVGA_REG_IRQMASK, dev_priv->irq_mask);
-		spin_unlock_irqrestore(&dev_priv->irq_lock, irq_flags);
-	}
-	spin_unlock(&dev_priv->waiter_lock);
+	vmw_generic_waiter_add(dev_priv, SVGA_IRQFLAG_FIFO_PROGRESS,
+			       &dev_priv->fifo_queue_waiters);
 
 	if (interruptible)
 		ret = wait_event_interruptible_timeout
@@ -287,14 +278,8 @@ static int vmw_fifo_wait(struct vmw_private *dev_priv,
 	else if (likely(ret > 0))
 		ret = 0;
 
-	spin_lock(&dev_priv->waiter_lock);
-	if (atomic_dec_and_test(&dev_priv->fifo_queue_waiters)) {
-		spin_lock_irqsave(&dev_priv->irq_lock, irq_flags);
-		dev_priv->irq_mask &= ~SVGA_IRQFLAG_FIFO_PROGRESS;
-		vmw_write(dev_priv, SVGA_REG_IRQMASK, dev_priv->irq_mask);
-		spin_unlock_irqrestore(&dev_priv->irq_lock, irq_flags);
-	}
-	spin_unlock(&dev_priv->waiter_lock);
+	vmw_generic_waiter_remove(dev_priv, SVGA_IRQFLAG_FIFO_PROGRESS,
+				  &dev_priv->fifo_queue_waiters);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_irq.c b/drivers/gpu/drm/vmwgfx/vmwgfx_irq.c
index ac3eccd..0c7e172 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_irq.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_irq.c
@@ -36,15 +36,13 @@ irqreturn_t vmw_irq_handler(int irq, void *arg)
 	struct vmw_private *dev_priv = vmw_priv(dev);
 	uint32_t status, masked_status;
 
-	spin_lock(&dev_priv->irq_lock);
 	status = inl(dev_priv->io_start + VMWGFX_IRQSTATUS_PORT);
-	masked_status = status & dev_priv->irq_mask;
-	spin_unlock(&dev_priv->irq_lock);
+	masked_status = status & READ_ONCE(dev_priv->irq_mask);
 
 	if (likely(status))
 		outl(status, dev_priv->io_start + VMWGFX_IRQSTATUS_PORT);
 
-	if (!masked_status)
+	if (!status)
 		return IRQ_NONE;
 
 	if (masked_status & (SVGA_IRQFLAG_ANY_FENCE |
@@ -190,65 +188,51 @@ out_err:
 	return ret;
 }
 
-void vmw_seqno_waiter_add(struct vmw_private *dev_priv)
+void vmw_generic_waiter_add(struct vmw_private *dev_priv,
+			    u32 flag, int *waiter_count)
 {
-	spin_lock(&dev_priv->waiter_lock);
-	if (dev_priv->fence_queue_waiters++ == 0) {
-		unsigned long irq_flags;
-
-		spin_lock_irqsave(&dev_priv->irq_lock, irq_flags);
-		outl(SVGA_IRQFLAG_ANY_FENCE,
-		     dev_priv->io_start + VMWGFX_IRQSTATUS_PORT);
-		dev_priv->irq_mask |= SVGA_IRQFLAG_ANY_FENCE;
+	spin_lock_bh(&dev_priv->waiter_lock);
+	if ((*waiter_count)++ == 0) {
+		outl(flag, dev_priv->io_start + VMWGFX_IRQSTATUS_PORT);
+		dev_priv->irq_mask |= flag;
 		vmw_write(dev_priv, SVGA_REG_IRQMASK, dev_priv->irq_mask);
-		spin_unlock_irqrestore(&dev_priv->irq_lock, irq_flags);
 	}
-	spin_unlock(&dev_priv->waiter_lock);
+	spin_unlock_bh(&dev_priv->waiter_lock);
 }
 
-void vmw_seqno_waiter_remove(struct vmw_private *dev_priv)
+void vmw_generic_waiter_remove(struct vmw_private *dev_priv,
+			       u32 flag, int *waiter_count)
 {
-	spin_lock(&dev_priv->waiter_lock);
-	if (--dev_priv->fence_queue_waiters == 0) {
-		unsigned long irq_flags;
-
-		spin_lock_irqsave(&dev_priv->irq_lock, irq_flags);
-		dev_priv->irq_mask &= ~SVGA_IRQFLAG_ANY_FENCE;
+	spin_lock_bh(&dev_priv->waiter_lock);
+	if (--(*waiter_count) == 0) {
+		dev_priv->irq_mask &= ~flag;
 		vmw_write(dev_priv, SVGA_REG_IRQMASK, dev_priv->irq_mask);
-		spin_unlock_irqrestore(&dev_priv->irq_lock, irq_flags);
 	}
-	spin_unlock(&dev_priv->waiter_lock);
+	spin_unlock_bh(&dev_priv->waiter_lock);
 }
 
+void vmw_seqno_waiter_add(struct vmw_private *dev_priv)
+{
+	vmw_generic_waiter_add(dev_priv, SVGA_IRQFLAG_ANY_FENCE,
+			       &dev_priv->fence_queue_waiters);
+}
+
+void vmw_seqno_waiter_remove(struct vmw_private *dev_priv)
+{
+	vmw_generic_waiter_remove(dev_priv, SVGA_IRQFLAG_ANY_FENCE,
+				  &dev_priv->fence_queue_waiters);
+}
 
 void vmw_goal_waiter_add(struct vmw_private *dev_priv)
 {
-	spin_lock(&dev_priv->waiter_lock);
-	if (dev_priv->goal_queue_waiters++ == 0) {
-		unsigned long irq_flags;
-
-		spin_lock_irqsave(&dev_priv->irq_lock, irq_flags);
-		outl(SVGA_IRQFLAG_FENCE_GOAL,
-		     dev_priv->io_start + VMWGFX_IRQSTATUS_PORT);
-		dev_priv->irq_mask |= SVGA_IRQFLAG_FENCE_GOAL;
-		vmw_write(dev_priv, SVGA_REG_IRQMASK, dev_priv->irq_mask);
-		spin_unlock_irqrestore(&dev_priv->irq_lock, irq_flags);
-	}
-	spin_unlock(&dev_priv->waiter_lock);
+	vmw_generic_waiter_add(dev_priv, SVGA_IRQFLAG_FENCE_GOAL,
+			       &dev_priv->goal_queue_waiters);
 }
 
 void vmw_goal_waiter_remove(struct vmw_private *dev_priv)
 {
-	spin_lock(&dev_priv->waiter_lock);
-	if (--dev_priv->goal_queue_waiters == 0) {
-		unsigned long irq_flags;
-
-		spin_lock_irqsave(&dev_priv->irq_lock, irq_flags);
-		dev_priv->irq_mask &= ~SVGA_IRQFLAG_FENCE_GOAL;
-		vmw_write(dev_priv, SVGA_REG_IRQMASK, dev_priv->irq_mask);
-		spin_unlock_irqrestore(&dev_priv->irq_lock, irq_flags);
-	}
-	spin_unlock(&dev_priv->waiter_lock);
+	vmw_generic_waiter_remove(dev_priv, SVGA_IRQFLAG_FENCE_GOAL,
+				  &dev_priv->goal_queue_waiters);
 }
 
 int vmw_wait_seqno(struct vmw_private *dev_priv,
@@ -305,7 +289,6 @@ void vmw_irq_preinstall(struct drm_device *dev)
 	if (!(dev_priv->capabilities & SVGA_CAP_IRQMASK))
 		return;
 
-	spin_lock_init(&dev_priv->irq_lock);
 	status = inl(dev_priv->io_start + VMWGFX_IRQSTATUS_PORT);
 	outl(status, dev_priv->io_start + VMWGFX_IRQSTATUS_PORT);
 }
@@ -328,30 +311,3 @@ void vmw_irq_uninstall(struct drm_device *dev)
 	status = inl(dev_priv->io_start + VMWGFX_IRQSTATUS_PORT);
 	outl(status, dev_priv->io_start + VMWGFX_IRQSTATUS_PORT);
 }
-
-void vmw_generic_waiter_add(struct vmw_private *dev_priv,
-			    u32 flag, int *waiter_count)
-{
-	unsigned long irq_flags;
-
-	spin_lock_irqsave(&dev_priv->irq_lock, irq_flags);
-	if ((*waiter_count)++ == 0) {
-		outl(flag, dev_priv->io_start + VMWGFX_IRQSTATUS_PORT);
-		dev_priv->irq_mask |= flag;
-		vmw_write(dev_priv, SVGA_REG_IRQMASK, dev_priv->irq_mask);
-	}
-	spin_unlock_irqrestore(&dev_priv->irq_lock, irq_flags);
-}
-
-void vmw_generic_waiter_remove(struct vmw_private *dev_priv,
-			       u32 flag, int *waiter_count)
-{
-	unsigned long irq_flags;
-
-	spin_lock_irqsave(&dev_priv->irq_lock, irq_flags);
-	if (--(*waiter_count) == 0) {
-		dev_priv->irq_mask &= ~flag;
-		vmw_write(dev_priv, SVGA_REG_IRQMASK, dev_priv->irq_mask);
-	}
-	spin_unlock_irqrestore(&dev_priv->irq_lock, irq_flags);
-}
-- 
cgit v0.10.2