diff options
| author | Dave Airlie <airlied@redhat.com> | 2012-05-02 08:21:50 (GMT) |
|---|---|---|
| committer | Dave Airlie <airlied@redhat.com> | 2012-05-02 08:22:29 (GMT) |
| commit | 5bc69bf9aeb73547cad8e1ce683a103fe9728282 (patch) | |
| tree | d3ef275532fc4391cb645f8b4d45d39d7fbb73f4 /drivers/gpu | |
| parent | c6543a6e64ad8e456674a1c4a01dd024e38b665f (diff) | |
| parent | a85d4bcb8a0cd5b3c754f98ff91ef2b9b3a73bc5 (diff) | |
| download | linux-5bc69bf9aeb73547cad8e1ce683a103fe9728282.tar.xz | |
Merge tag 'drm-intel-next-2012-04-23' of git://people.freedesktop.org/~danvet/drm-intel into drm-core-next
Daniel Vetter writes:
A new drm-intel-next pull. Highlights:
- More gmbus patches from Daniel Kurtz, I think gmbus is now ready, all
known issues fixed.
- Fencing cleanup and pipelined fencing removal from Chris.
- rc6 residency interface from Ben, useful for powertop.
- Cleanups and code reorg around the ringbuffer code (Ben&me).
- Use hw semaphores in the pageflip code from Ben.
- More vlv stuff from Jesse, unfortunately his vlv cpu is doa, so less
merged than I've hoped for - we still have the unused function warning :(
- More hsw patches from Eugeni, again, not yet enabled fully.
- intel_pm.c refactoring from Eugeni.
- Ironlake sprite support from Chris.
- And various smaller improvements/fixes all over the place.
Note that this pull request also contains a backmerge of -rc3 to sort out
a few things in -next. I've also had to frob the shortlog a bit to exclude
anything that -rc3 brings in with this pull.
Regression wise we have a few strange bugs going on, but for all of them
closer inspection revealed that they've been pre-existing, just now
slightly more likely to be hit. And for most of them we have a patch
already. Otherwise QA has not reported any regressions, and I'm also not
aware of anything bad happening in 3.4.
* tag 'drm-intel-next-2012-04-23' of git://people.freedesktop.org/~danvet/drm-intel: (420 commits)
drm/i915: rc6 residency (fix the fix)
drm/i915/tv: fix open-coded ARRAY_SIZE.
drm/i915: invalidate render cache on gen2
drm/i915: Silence the change of LVDS sync polarity
drm/i915: add generic power management initialization
drm/i915: move clock gating functionality into intel_pm module
drm/i915: move emon functionality into intel_pm module
drm/i915: move drps, rps and rc6-related functions to intel_pm
drm/i915: fix line breaks in intel_pm
drm/i915: move watermarks settings into intel_pm module
drm/i915: move fbc-related functionality into intel_pm module
drm/i915: Refactor get_fence() to use the common fence writing routine
drm/i915: Refactor fence clearing to use the common fence writing routine
drm/i915: Refactor put_fence() to use the common fence writing routine
drm/i915: Prepare to consolidate fence writing
drm/i915: Remove the unsightly "optimisation" from flush_fence()
drm/i915: Simplify fence finding
drm/i915: Discard the unused obj->last_fenced_ring
drm/i915: Remove unused ring->setup_seqno
drm/i915: Remove fence pipelining
...
Diffstat (limited to 'drivers/gpu')
48 files changed, 4588 insertions, 4247 deletions
diff --git a/drivers/gpu/drm/exynos/exynos_drm_buf.c b/drivers/gpu/drm/exynos/exynos_drm_buf.c index 4a3a5f7..de8d209 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_buf.c +++ b/drivers/gpu/drm/exynos/exynos_drm_buf.c @@ -34,14 +34,14 @@ static int lowlevel_buffer_allocate(struct drm_device *dev, unsigned int flags, struct exynos_drm_gem_buf *buf) { - dma_addr_t start_addr, end_addr; + dma_addr_t start_addr; unsigned int npages, page_size, i = 0; struct scatterlist *sgl; int ret = 0; DRM_DEBUG_KMS("%s\n", __FILE__); - if (flags & EXYNOS_BO_NONCONTIG) { + if (IS_NONCONTIG_BUFFER(flags)) { DRM_DEBUG_KMS("not support allocation type.\n"); return -EINVAL; } @@ -52,13 +52,13 @@ static int lowlevel_buffer_allocate(struct drm_device *dev, } if (buf->size >= SZ_1M) { - npages = (buf->size >> SECTION_SHIFT) + 1; + npages = buf->size >> SECTION_SHIFT; page_size = SECTION_SIZE; } else if (buf->size >= SZ_64K) { - npages = (buf->size >> 16) + 1; + npages = buf->size >> 16; page_size = SZ_64K; } else { - npages = (buf->size >> PAGE_SHIFT) + 1; + npages = buf->size >> PAGE_SHIFT; page_size = PAGE_SIZE; } @@ -76,26 +76,13 @@ static int lowlevel_buffer_allocate(struct drm_device *dev, return -ENOMEM; } - buf->kvaddr = dma_alloc_writecombine(dev->dev, buf->size, - &buf->dma_addr, GFP_KERNEL); - if (!buf->kvaddr) { - DRM_ERROR("failed to allocate buffer.\n"); - ret = -ENOMEM; - goto err1; - } - - start_addr = buf->dma_addr; - end_addr = buf->dma_addr + buf->size; - - buf->pages = kzalloc(sizeof(struct page) * npages, GFP_KERNEL); - if (!buf->pages) { - DRM_ERROR("failed to allocate pages.\n"); - ret = -ENOMEM; - goto err2; - } - - start_addr = buf->dma_addr; - end_addr = buf->dma_addr + buf->size; + buf->kvaddr = dma_alloc_writecombine(dev->dev, buf->size, + &buf->dma_addr, GFP_KERNEL); + if (!buf->kvaddr) { + DRM_ERROR("failed to allocate buffer.\n"); + ret = -ENOMEM; + goto err1; + } buf->pages = kzalloc(sizeof(struct page) * npages, GFP_KERNEL); if (!buf->pages) { @@ -105,23 +92,17 @@ static int lowlevel_buffer_allocate(struct drm_device *dev, } sgl = buf->sgt->sgl; + start_addr = buf->dma_addr; while (i < npages) { buf->pages[i] = phys_to_page(start_addr); sg_set_page(sgl, buf->pages[i], page_size, 0); sg_dma_address(sgl) = start_addr; start_addr += page_size; - if (end_addr - start_addr < page_size) - break; sgl = sg_next(sgl); i++; } - buf->pages[i] = phys_to_page(start_addr); - - sgl = sg_next(sgl); - sg_set_page(sgl, buf->pages[i+1], end_addr - start_addr, 0); - DRM_DEBUG_KMS("vaddr(0x%lx), dma_addr(0x%lx), size(0x%lx)\n", (unsigned long)buf->kvaddr, (unsigned long)buf->dma_addr, @@ -150,7 +131,7 @@ static void lowlevel_buffer_deallocate(struct drm_device *dev, * non-continuous memory would be released by exynos * gem framework. */ - if (flags & EXYNOS_BO_NONCONTIG) { + if (IS_NONCONTIG_BUFFER(flags)) { DRM_DEBUG_KMS("not support allocation type.\n"); return; } diff --git a/drivers/gpu/drm/exynos/exynos_drm_core.c b/drivers/gpu/drm/exynos/exynos_drm_core.c index 411832e..eaf630d 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_core.c +++ b/drivers/gpu/drm/exynos/exynos_drm_core.c @@ -54,16 +54,18 @@ static int exynos_drm_subdrv_probe(struct drm_device *dev, * * P.S. note that this driver is considered for modularization. */ - ret = subdrv->probe(dev, subdrv->manager.dev); + ret = subdrv->probe(dev, subdrv->dev); if (ret) return ret; } - if (subdrv->is_local) + if (!subdrv->manager) return 0; + subdrv->manager->dev = subdrv->dev; + /* create and initialize a encoder for this sub driver. */ - encoder = exynos_drm_encoder_create(dev, &subdrv->manager, + encoder = exynos_drm_encoder_create(dev, subdrv->manager, (1 << MAX_CRTC) - 1); if (!encoder) { DRM_ERROR("failed to create encoder\n"); @@ -186,7 +188,7 @@ int exynos_drm_subdrv_open(struct drm_device *dev, struct drm_file *file) list_for_each_entry(subdrv, &exynos_drm_subdrv_list, list) { if (subdrv->open) { - ret = subdrv->open(dev, subdrv->manager.dev, file); + ret = subdrv->open(dev, subdrv->dev, file); if (ret) goto err; } @@ -197,7 +199,7 @@ int exynos_drm_subdrv_open(struct drm_device *dev, struct drm_file *file) err: list_for_each_entry_reverse(subdrv, &subdrv->list, list) { if (subdrv->close) - subdrv->close(dev, subdrv->manager.dev, file); + subdrv->close(dev, subdrv->dev, file); } return ret; } @@ -209,7 +211,7 @@ void exynos_drm_subdrv_close(struct drm_device *dev, struct drm_file *file) list_for_each_entry(subdrv, &exynos_drm_subdrv_list, list) { if (subdrv->close) - subdrv->close(dev, subdrv->manager.dev, file); + subdrv->close(dev, subdrv->dev, file); } } EXPORT_SYMBOL_GPL(exynos_drm_subdrv_close); diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.h b/drivers/gpu/drm/exynos/exynos_drm_drv.h index fbd0a23..1d81417 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_drv.h +++ b/drivers/gpu/drm/exynos/exynos_drm_drv.h @@ -225,24 +225,25 @@ struct exynos_drm_private { * Exynos drm sub driver structure. * * @list: sub driver has its own list object to register to exynos drm driver. + * @dev: pointer to device object for subdrv device driver. * @drm_dev: pointer to drm_device and this pointer would be set * when sub driver calls exynos_drm_subdrv_register(). - * @is_local: appear encoder and connector disrelated device. + * @manager: subdrv has its own manager to control a hardware appropriately + * and we can access a hardware drawing on this manager. * @probe: this callback would be called by exynos drm driver after * subdrv is registered to it. * @remove: this callback is used to release resources created * by probe callback. * @open: this would be called with drm device file open. * @close: this would be called with drm device file close. - * @manager: subdrv has its own manager to control a hardware appropriately - * and we can access a hardware drawing on this manager. * @encoder: encoder object owned by this sub driver. * @connector: connector object owned by this sub driver. */ struct exynos_drm_subdrv { struct list_head list; + struct device *dev; struct drm_device *drm_dev; - bool is_local; + struct exynos_drm_manager *manager; int (*probe)(struct drm_device *drm_dev, struct device *dev); void (*remove)(struct drm_device *dev); @@ -251,7 +252,6 @@ struct exynos_drm_subdrv { void (*close)(struct drm_device *drm_dev, struct device *dev, struct drm_file *file); - struct exynos_drm_manager manager; struct drm_encoder *encoder; struct drm_connector *connector; }; diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimd.c b/drivers/gpu/drm/exynos/exynos_drm_fimd.c index ecb6db2..29fdbfe 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fimd.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fimd.c @@ -172,7 +172,7 @@ static void fimd_dpms(struct device *subdrv_dev, int mode) static void fimd_apply(struct device *subdrv_dev) { struct fimd_context *ctx = get_fimd_context(subdrv_dev); - struct exynos_drm_manager *mgr = &ctx->subdrv.manager; + struct exynos_drm_manager *mgr = ctx->subdrv.manager; struct exynos_drm_manager_ops *mgr_ops = mgr->ops; struct exynos_drm_overlay_ops *ovl_ops = mgr->overlay_ops; struct fimd_win_data *win_data; @@ -577,6 +577,13 @@ static struct exynos_drm_overlay_ops fimd_overlay_ops = { .disable = fimd_win_disable, }; +static struct exynos_drm_manager fimd_manager = { + .pipe = -1, + .ops = &fimd_manager_ops, + .overlay_ops = &fimd_overlay_ops, + .display_ops = &fimd_display_ops, +}; + static void fimd_finish_pageflip(struct drm_device *drm_dev, int crtc) { struct exynos_drm_private *dev_priv = drm_dev->dev_private; @@ -628,7 +635,7 @@ static irqreturn_t fimd_irq_handler(int irq, void *dev_id) struct fimd_context *ctx = (struct fimd_context *)dev_id; struct exynos_drm_subdrv *subdrv = &ctx->subdrv; struct drm_device *drm_dev = subdrv->drm_dev; - struct exynos_drm_manager *manager = &subdrv->manager; + struct exynos_drm_manager *manager = subdrv->manager; u32 val; val = readl(ctx->regs + VIDINTCON1); @@ -744,7 +751,7 @@ static void fimd_clear_win(struct fimd_context *ctx, int win) static int fimd_power_on(struct fimd_context *ctx, bool enable) { struct exynos_drm_subdrv *subdrv = &ctx->subdrv; - struct device *dev = subdrv->manager.dev; + struct device *dev = subdrv->dev; DRM_DEBUG_KMS("%s\n", __FILE__); @@ -867,13 +874,10 @@ static int __devinit fimd_probe(struct platform_device *pdev) subdrv = &ctx->subdrv; + subdrv->dev = dev; + subdrv->manager = &fimd_manager; subdrv->probe = fimd_subdrv_probe; subdrv->remove = fimd_subdrv_remove; - subdrv->manager.pipe = -1; - subdrv->manager.ops = &fimd_manager_ops; - subdrv->manager.overlay_ops = &fimd_overlay_ops; - subdrv->manager.display_ops = &fimd_display_ops; - subdrv->manager.dev = dev; mutex_init(&ctx->lock); diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.c b/drivers/gpu/drm/exynos/exynos_drm_gem.c index fa1aa94..26d5197 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gem.c +++ b/drivers/gpu/drm/exynos/exynos_drm_gem.c @@ -56,9 +56,28 @@ static unsigned int convert_to_vm_err_msg(int msg) return out_msg; } -static unsigned int mask_gem_flags(unsigned int flags) +static int check_gem_flags(unsigned int flags) { - return flags &= EXYNOS_BO_NONCONTIG; + if (flags & ~(EXYNOS_BO_MASK)) { + DRM_ERROR("invalid flags.\n"); + return -EINVAL; + } + + return 0; +} + +static unsigned long roundup_gem_size(unsigned long size, unsigned int flags) +{ + if (!IS_NONCONTIG_BUFFER(flags)) { + if (size >= SZ_1M) + return roundup(size, SECTION_SIZE); + else if (size >= SZ_64K) + return roundup(size, SZ_64K); + else + goto out; + } +out: + return roundup(size, PAGE_SIZE); } static struct page **exynos_gem_get_pages(struct drm_gem_object *obj, @@ -319,10 +338,17 @@ struct exynos_drm_gem_obj *exynos_drm_gem_create(struct drm_device *dev, struct exynos_drm_gem_buf *buf; int ret; - size = roundup(size, PAGE_SIZE); - DRM_DEBUG_KMS("%s: size = 0x%lx\n", __FILE__, size); + if (!size) { + DRM_ERROR("invalid size.\n"); + return ERR_PTR(-EINVAL); + } - flags = mask_gem_flags(flags); + size = roundup_gem_size(size, flags); + DRM_DEBUG_KMS("%s\n", __FILE__); + + ret = check_gem_flags(flags); + if (ret) + return ERR_PTR(ret); buf = exynos_drm_init_buf(dev, size); if (!buf) @@ -331,7 +357,7 @@ struct exynos_drm_gem_obj *exynos_drm_gem_create(struct drm_device *dev, exynos_gem_obj = exynos_drm_gem_init(dev, size); if (!exynos_gem_obj) { ret = -ENOMEM; - goto err; + goto err_fini_buf; } exynos_gem_obj->buffer = buf; @@ -347,18 +373,19 @@ struct exynos_drm_gem_obj *exynos_drm_gem_create(struct drm_device *dev, ret = exynos_drm_gem_get_pages(&exynos_gem_obj->base); if (ret < 0) { drm_gem_object_release(&exynos_gem_obj->base); - goto err; + goto err_fini_buf; } } else { ret = exynos_drm_alloc_buf(dev, buf, flags); if (ret < 0) { drm_gem_object_release(&exynos_gem_obj->base); - goto err; + goto err_fini_buf; } } return exynos_gem_obj; -err: + +err_fini_buf: exynos_drm_fini_buf(dev, buf); 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 e40fbad..4ed8420 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gem.h +++ b/drivers/gpu/drm/exynos/exynos_drm_gem.h @@ -29,6 +29,8 @@ #define to_exynos_gem_obj(x) container_of(x,\ struct exynos_drm_gem_obj, base) +#define IS_NONCONTIG_BUFFER(f) (f & EXYNOS_BO_NONCONTIG) + /* * exynos drm gem buffer structure. * diff --git a/drivers/gpu/drm/exynos/exynos_drm_hdmi.c b/drivers/gpu/drm/exynos/exynos_drm_hdmi.c index 14eb26b..3424463 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_hdmi.c +++ b/drivers/gpu/drm/exynos/exynos_drm_hdmi.c @@ -30,9 +30,8 @@ struct drm_hdmi_context, subdrv); /* these callback points shoud be set by specific drivers. */ -static struct exynos_hdmi_display_ops *hdmi_display_ops; -static struct exynos_hdmi_manager_ops *hdmi_manager_ops; -static struct exynos_hdmi_overlay_ops *hdmi_overlay_ops; +static struct exynos_hdmi_ops *hdmi_ops; +static struct exynos_mixer_ops *mixer_ops; struct drm_hdmi_context { struct exynos_drm_subdrv subdrv; @@ -40,31 +39,20 @@ struct drm_hdmi_context { struct exynos_drm_hdmi_context *mixer_ctx; }; -void exynos_drm_display_ops_register(struct exynos_hdmi_display_ops - *display_ops) +void exynos_hdmi_ops_register(struct exynos_hdmi_ops *ops) { DRM_DEBUG_KMS("%s\n", __FILE__); - if (display_ops) - hdmi_display_ops = display_ops; + if (ops) + hdmi_ops = ops; } -void exynos_drm_manager_ops_register(struct exynos_hdmi_manager_ops - *manager_ops) +void exynos_mixer_ops_register(struct exynos_mixer_ops *ops) { DRM_DEBUG_KMS("%s\n", __FILE__); - if (manager_ops) - hdmi_manager_ops = manager_ops; -} - -void exynos_drm_overlay_ops_register(struct exynos_hdmi_overlay_ops - *overlay_ops) -{ - DRM_DEBUG_KMS("%s\n", __FILE__); - - if (overlay_ops) - hdmi_overlay_ops = overlay_ops; + if (ops) + mixer_ops = ops; } static bool drm_hdmi_is_connected(struct device *dev) @@ -73,8 +61,8 @@ static bool drm_hdmi_is_connected(struct device *dev) DRM_DEBUG_KMS("%s\n", __FILE__); - if (hdmi_display_ops && hdmi_display_ops->is_connected) - return hdmi_display_ops->is_connected(ctx->hdmi_ctx->ctx); + if (hdmi_ops && hdmi_ops->is_connected) + return hdmi_ops->is_connected(ctx->hdmi_ctx->ctx); return false; } @@ -86,9 +74,9 @@ static int drm_hdmi_get_edid(struct device *dev, DRM_DEBUG_KMS("%s\n", __FILE__); - if (hdmi_display_ops && hdmi_display_ops->get_edid) - return hdmi_display_ops->get_edid(ctx->hdmi_ctx->ctx, - connector, edid, len); + if (hdmi_ops && hdmi_ops->get_edid) + return hdmi_ops->get_edid(ctx->hdmi_ctx->ctx, connector, edid, + len); return 0; } @@ -99,9 +87,8 @@ static int drm_hdmi_check_timing(struct device *dev, void *timing) DRM_DEBUG_KMS("%s\n", __FILE__); - if (hdmi_display_ops && hdmi_display_ops->check_timing) - return hdmi_display_ops->check_timing(ctx->hdmi_ctx->ctx, - timing); + if (hdmi_ops && hdmi_ops->check_timing) + return hdmi_ops->check_timing(ctx->hdmi_ctx->ctx, timing); return 0; } @@ -112,8 +99,8 @@ static int drm_hdmi_power_on(struct device *dev, int mode) DRM_DEBUG_KMS("%s\n", __FILE__); - if (hdmi_display_ops && hdmi_display_ops->power_on) - return hdmi_display_ops->power_on(ctx->hdmi_ctx->ctx, mode); + if (hdmi_ops && hdmi_ops->power_on) + return hdmi_ops->power_on(ctx->hdmi_ctx->ctx, mode); return 0; } @@ -130,13 +117,13 @@ static int drm_hdmi_enable_vblank(struct device *subdrv_dev) { struct drm_hdmi_context *ctx = to_context(subdrv_dev); struct exynos_drm_subdrv *subdrv = &ctx->subdrv; - struct exynos_drm_manager *manager = &subdrv->manager; + struct exynos_drm_manager *manager = subdrv->manager; DRM_DEBUG_KMS("%s\n", __FILE__); - if (hdmi_overlay_ops && hdmi_overlay_ops->enable_vblank) - return hdmi_overlay_ops->enable_vblank(ctx->mixer_ctx->ctx, - manager->pipe); + if (mixer_ops && mixer_ops->enable_vblank) + return mixer_ops->enable_vblank(ctx->mixer_ctx->ctx, + manager->pipe); return 0; } @@ -147,8 +134,8 @@ static void drm_hdmi_disable_vblank(struct device *subdrv_dev) DRM_DEBUG_KMS("%s\n", __FILE__); - if (hdmi_overlay_ops && hdmi_overlay_ops->disable_vblank) - return hdmi_overlay_ops->disable_vblank(ctx->mixer_ctx->ctx); + if (mixer_ops && mixer_ops->disable_vblank) + return mixer_ops->disable_vblank(ctx->mixer_ctx->ctx); } static void drm_hdmi_mode_fixup(struct device *subdrv_dev, @@ -160,9 +147,9 @@ static void drm_hdmi_mode_fixup(struct device *subdrv_dev, DRM_DEBUG_KMS("%s\n", __FILE__); - if (hdmi_manager_ops && hdmi_manager_ops->mode_fixup) - hdmi_manager_ops->mode_fixup(ctx->hdmi_ctx->ctx, connector, - mode, adjusted_mode); + if (hdmi_ops && hdmi_ops->mode_fixup) + hdmi_ops->mode_fixup(ctx->hdmi_ctx->ctx, connector, mode, + adjusted_mode); } static void drm_hdmi_mode_set(struct device *subdrv_dev, void *mode) @@ -171,8 +158,8 @@ static void drm_hdmi_mode_set(struct device *subdrv_dev, void *mode) DRM_DEBUG_KMS("%s\n", __FILE__); - if (hdmi_manager_ops && hdmi_manager_ops->mode_set) - hdmi_manager_ops->mode_set(ctx->hdmi_ctx->ctx, mode); + if (hdmi_ops && hdmi_ops->mode_set) + hdmi_ops->mode_set(ctx->hdmi_ctx->ctx, mode); } static void drm_hdmi_get_max_resol(struct device *subdrv_dev, @@ -182,9 +169,8 @@ static void drm_hdmi_get_max_resol(struct device *subdrv_dev, DRM_DEBUG_KMS("%s\n", __FILE__); - if (hdmi_manager_ops && hdmi_manager_ops->get_max_resol) - hdmi_manager_ops->get_max_resol(ctx->hdmi_ctx->ctx, width, - height); + if (hdmi_ops && hdmi_ops->get_max_resol) + hdmi_ops->get_max_resol(ctx->hdmi_ctx->ctx, width, height); } static void drm_hdmi_commit(struct device *subdrv_dev) @@ -193,8 +179,8 @@ static void drm_hdmi_commit(struct device *subdrv_dev) DRM_DEBUG_KMS("%s\n", __FILE__); - if (hdmi_manager_ops && hdmi_manager_ops->commit) - hdmi_manager_ops->commit(ctx->hdmi_ctx->ctx); + if (hdmi_ops && hdmi_ops->commit) + hdmi_ops->commit(ctx->hdmi_ctx->ctx); } static void drm_hdmi_dpms(struct device *subdrv_dev, int mode) @@ -209,8 +195,8 @@ static void drm_hdmi_dpms(struct device *subdrv_dev, int mode) case DRM_MODE_DPMS_STANDBY: case DRM_MODE_DPMS_SUSPEND: case DRM_MODE_DPMS_OFF: - if (hdmi_manager_ops && hdmi_manager_ops->disable) - hdmi_manager_ops->disable(ctx->hdmi_ctx->ctx); + if (hdmi_ops && hdmi_ops->disable) + hdmi_ops->disable(ctx->hdmi_ctx->ctx); break; default: DRM_DEBUG_KMS("unkown dps mode: %d\n", mode); @@ -235,8 +221,8 @@ static void drm_mixer_mode_set(struct device *subdrv_dev, DRM_DEBUG_KMS("%s\n", __FILE__); - if (hdmi_overlay_ops && hdmi_overlay_ops->win_mode_set) - hdmi_overlay_ops->win_mode_set(ctx->mixer_ctx->ctx, overlay); + if (mixer_ops && mixer_ops->win_mode_set) + mixer_ops->win_mode_set(ctx->mixer_ctx->ctx, overlay); } static void drm_mixer_commit(struct device *subdrv_dev, int zpos) @@ -245,8 +231,8 @@ static void drm_mixer_commit(struct device *subdrv_dev, int zpos) DRM_DEBUG_KMS("%s\n", __FILE__); - if (hdmi_overlay_ops && hdmi_overlay_ops->win_commit) - hdmi_overlay_ops->win_commit(ctx->mixer_ctx->ctx, zpos); + if (mixer_ops && mixer_ops->win_commit) + mixer_ops->win_commit(ctx->mixer_ctx->ctx, zpos); } static void drm_mixer_disable(struct device *subdrv_dev, int zpos) @@ -255,8 +241,8 @@ static void drm_mixer_disable(struct device *subdrv_dev, int zpos) DRM_DEBUG_KMS("%s\n", __FILE__); - if (hdmi_overlay_ops && hdmi_overlay_ops->win_disable) - hdmi_overlay_ops->win_disable(ctx->mixer_ctx->ctx, zpos); + if (mixer_ops && mixer_ops->win_disable) + mixer_ops->win_disable(ctx->mixer_ctx->ctx, zpos); } static struct exynos_drm_overlay_ops drm_hdmi_overlay_ops = { @@ -265,6 +251,12 @@ static struct exynos_drm_overlay_ops drm_hdmi_overlay_ops = { .disable = drm_mixer_disable, }; +static struct exynos_drm_manager hdmi_manager = { + .pipe = -1, + .ops = &drm_hdmi_manager_ops, + .overlay_ops = &drm_hdmi_overlay_ops, + .display_ops = &drm_hdmi_display_ops, +}; static int hdmi_subdrv_probe(struct drm_device *drm_dev, struct device *dev) @@ -332,12 +324,9 @@ static int __devinit exynos_drm_hdmi_probe(struct platform_device *pdev) subdrv = &ctx->subdrv; + subdrv->dev = dev; + subdrv->manager = &hdmi_manager; subdrv->probe = hdmi_subdrv_probe; - subdrv->manager.pipe = -1; - subdrv->manager.ops = &drm_hdmi_manager_ops; - subdrv->manager.overlay_ops = &drm_hdmi_overlay_ops; - subdrv->manager.display_ops = &drm_hdmi_display_ops; - subdrv->manager.dev = dev; platform_set_drvdata(pdev, subdrv); diff --git a/drivers/gpu/drm/exynos/exynos_drm_hdmi.h b/drivers/gpu/drm/exynos/exynos_drm_hdmi.h index 44497cf..f3ae192 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_hdmi.h +++ b/drivers/gpu/drm/exynos/exynos_drm_hdmi.h @@ -38,15 +38,15 @@ struct exynos_drm_hdmi_context { void *ctx; }; -struct exynos_hdmi_display_ops { +struct exynos_hdmi_ops { + /* display */ bool (*is_connected)(void *ctx); int (*get_edid)(void *ctx, struct drm_connector *connector, u8 *edid, int len); int (*check_timing)(void *ctx, void *timing); int (*power_on)(void *ctx, int mode); -}; -struct exynos_hdmi_manager_ops { + /* manager */ void (*mode_fixup)(void *ctx, struct drm_connector *connector, struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode); @@ -57,22 +57,17 @@ struct exynos_hdmi_manager_ops { void (*disable)(void *ctx); }; -struct exynos_hdmi_overlay_ops { +struct exynos_mixer_ops { + /* manager */ int (*enable_vblank)(void *ctx, int pipe); void (*disable_vblank)(void *ctx); + + /* overlay */ void (*win_mode_set)(void *ctx, struct exynos_drm_overlay *overlay); void (*win_commit)(void *ctx, int zpos); void (*win_disable)(void *ctx, int zpos); }; -extern struct platform_driver hdmi_driver; -extern struct platform_driver mixer_driver; - -void exynos_drm_display_ops_register(struct exynos_hdmi_display_ops - *display_ops); -void exynos_drm_manager_ops_register(struct exynos_hdmi_manager_ops - *manager_ops); -void exynos_drm_overlay_ops_register(struct exynos_hdmi_overlay_ops - *overlay_ops); - +void exynos_hdmi_ops_register(struct exynos_hdmi_ops *ops); +void exynos_mixer_ops_register(struct exynos_mixer_ops *ops); #endif diff --git a/drivers/gpu/drm/exynos/exynos_drm_plane.c b/drivers/gpu/drm/exynos/exynos_drm_plane.c index c277a3a..f92fe4c 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_plane.c +++ b/drivers/gpu/drm/exynos/exynos_drm_plane.c @@ -24,6 +24,10 @@ struct exynos_plane { static const uint32_t formats[] = { DRM_FORMAT_XRGB8888, + DRM_FORMAT_ARGB8888, + DRM_FORMAT_NV12, + DRM_FORMAT_NV12M, + DRM_FORMAT_NV12MT, }; static int diff --git a/drivers/gpu/drm/exynos/exynos_drm_vidi.c b/drivers/gpu/drm/exynos/exynos_drm_vidi.c index 8e1339f..7b9c153 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_vidi.c +++ b/drivers/gpu/drm/exynos/exynos_drm_vidi.c @@ -199,7 +199,7 @@ static void vidi_dpms(struct device *subdrv_dev, int mode) static void vidi_apply(struct device *subdrv_dev) { struct vidi_context *ctx = get_vidi_context(subdrv_dev); - struct exynos_drm_manager *mgr = &ctx->subdrv.manager; + struct exynos_drm_manager *mgr = ctx->subdrv.manager; struct exynos_drm_manager_ops *mgr_ops = mgr->ops; struct exynos_drm_overlay_ops *ovl_ops = mgr->overlay_ops; struct vidi_win_data *win_data; @@ -374,6 +374,13 @@ static struct exynos_drm_overlay_ops vidi_overlay_ops = { .disable = vidi_win_disable, }; +static struct exynos_drm_manager vidi_manager = { + .pipe = -1, + .ops = &vidi_manager_ops, + .overlay_ops = &vidi_overlay_ops, + .display_ops = &vidi_display_ops, +}; + static void vidi_finish_pageflip(struct drm_device *drm_dev, int crtc) { struct exynos_drm_private *dev_priv = drm_dev->dev_private; @@ -425,7 +432,7 @@ static void vidi_fake_vblank_handler(struct work_struct *work) struct vidi_context *ctx = container_of(work, struct vidi_context, work); struct exynos_drm_subdrv *subdrv = &ctx->subdrv; - struct exynos_drm_manager *manager = &subdrv->manager; + struct exynos_drm_manager *manager = subdrv->manager; if (manager->pipe < 0) return; @@ -471,7 +478,7 @@ static void vidi_subdrv_remove(struct drm_device *drm_dev) static int vidi_power_on(struct vidi_context *ctx, bool enable) { struct exynos_drm_subdrv *subdrv = &ctx->subdrv; - struct device *dev = subdrv->manager.dev; + struct device *dev = subdrv->dev; DRM_DEBUG_KMS("%s\n", __FILE__); @@ -611,13 +618,10 @@ static int __devinit vidi_probe(struct platform_device *pdev) ctx->raw_edid = (struct edid *)fake_edid_info; subdrv = &ctx->subdrv; + subdrv->dev = dev; + subdrv->manager = &vidi_manager; subdrv->probe = vidi_subdrv_probe; subdrv->remove = vidi_subdrv_remove; - subdrv->manager.pipe = -1; - subdrv->manager.ops = &vidi_manager_ops; - subdrv->manager.overlay_ops = &vidi_overlay_ops; - subdrv->manager.display_ops = &vidi_display_ops; - subdrv->manager.dev = dev; mutex_init(&ctx->lock); diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c index 575a8cb..b003538 100644 --- a/drivers/gpu/drm/exynos/exynos_hdmi.c +++ b/drivers/gpu/drm/exynos/exynos_hdmi.c @@ -40,7 +40,6 @@ #include "exynos_hdmi.h" -#define HDMI_OVERLAY_NUMBER 3 #define MAX_WIDTH 1920 #define MAX_HEIGHT 1080 #define get_hdmi_context(dev) platform_get_drvdata(to_platform_device(dev)) @@ -1194,7 +1193,7 @@ static int hdmi_conf_index(struct hdmi_context *hdata, static bool hdmi_is_connected(void *ctx) { - struct hdmi_context *hdata = (struct hdmi_context *)ctx; + struct hdmi_context *hdata = ctx; u32 val = hdmi_reg_read(hdata, HDMI_HPD_STATUS); if (val) @@ -1207,7 +1206,7 @@ static int hdmi_get_edid(void *ctx, struct drm_connector *connector, u8 *edid, int len) { struct edid *raw_edid; - struct hdmi_context *hdata = (struct hdmi_context *)ctx; + struct hdmi_context *hdata = ctx; DRM_DEBUG_KMS("[%d] %s\n", __LINE__, __func__); @@ -1275,7 +1274,7 @@ static int hdmi_v14_check_timing(struct fb_videomode *check_timing) static int hdmi_check_timing(void *ctx, void *timing) { - struct hdmi_context *hdata = (struct hdmi_context *)ctx; + struct hdmi_context *hdata = ctx; struct fb_videomode *check_timing = timing; DRM_DEBUG_KMS("[%d] %s\n", __LINE__, __func__); @@ -1312,13 +1311,6 @@ static int hdmi_display_power_on(void *ctx, int mode) return 0; } -static struct exynos_hdmi_display_ops display_ops = { - .is_connected = hdmi_is_connected, - .get_edid = hdmi_get_edid, - .check_timing = hdmi_check_timing, - .power_on = hdmi_display_power_on, -}; - static void hdmi_set_acr(u32 freq, u8 *acr) { u32 n, cts; @@ -1914,7 +1906,7 @@ static void hdmi_mode_fixup(void *ctx, struct drm_connector *connector, struct drm_display_mode *adjusted_mode) { struct drm_display_mode *m; - struct hdmi_context *hdata = (struct hdmi_context *)ctx; + struct hdmi_context *hdata = ctx; int index; DRM_DEBUG_KMS("[%d] %s\n", __LINE__, __func__); @@ -1951,7 +1943,7 @@ static void hdmi_mode_fixup(void *ctx, struct drm_connector *connector, static void hdmi_mode_set(void *ctx, void *mode) { - struct hdmi_context *hdata = (struct hdmi_context *)ctx; + struct hdmi_context *hdata = ctx; int conf_idx; DRM_DEBUG_KMS("[%d] %s\n", __LINE__, __func__); @@ -1974,7 +1966,7 @@ static void hdmi_get_max_resol(void *ctx, unsigned int *width, static void hdmi_commit(void *ctx) { - struct hdmi_context *hdata = (struct hdmi_context *)ctx; + struct hdmi_context *hdata = ctx; DRM_DEBUG_KMS("[%d] %s\n", __LINE__, __func__); @@ -1985,7 +1977,7 @@ static void hdmi_commit(void *ctx) static void hdmi_disable(void *ctx) { - struct hdmi_context *hdata = (struct hdmi_context *)ctx; + struct hdmi_context *hdata = ctx; DRM_DEBUG_KMS("[%d] %s\n", __LINE__, __func__); @@ -1996,7 +1988,14 @@ static void hdmi_disable(void *ctx) } } -static struct exynos_hdmi_manager_ops manager_ops = { +static struct exynos_hdmi_ops hdmi_ops = { + /* display */ + .is_connected = hdmi_is_connected, + .get_edid = hdmi_get_edid, + .check_timing = hdmi_check_timing, + .power_on = hdmi_display_power_on, + + /* manager */ .mode_fixup = hdmi_mode_fixup, .mode_set = hdmi_mode_set, .get_max_resol = hdmi_get_max_resol, @@ -2020,7 +2019,7 @@ static void hdmi_hotplug_func(struct work_struct *work) static irqreturn_t hdmi_irq_handler(int irq, void *arg) { struct exynos_drm_hdmi_context *ctx = arg; - struct hdmi_context *hdata = (struct hdmi_context *)ctx->ctx; + struct hdmi_context *hdata = ctx->ctx; u32 intc_flag; intc_flag = hdmi_reg_read(hdata, HDMI_INTC_FLAG); @@ -2173,7 +2172,7 @@ static int hdmi_runtime_suspend(struct device *dev) DRM_DEBUG_KMS("%s\n", __func__); - hdmi_resource_poweroff((struct hdmi_context *)ctx->ctx); + hdmi_resource_poweroff(ctx->ctx); return 0; } @@ -2184,7 +2183,7 @@ static int hdmi_runtime_resume(struct device *dev) DRM_DEBUG_KMS("%s\n", __func__); - hdmi_resource_poweron((struct hdmi_context *)ctx->ctx); + hdmi_resource_poweron(ctx->ctx); return 0; } @@ -2322,8 +2321,7 @@ static int __devinit hdmi_probe(struct platform_device *pdev) hdata->irq = res->start; /* register specific callbacks to common hdmi. */ - exynos_drm_display_ops_register(&display_ops); - exynos_drm_manager_ops_register(&manager_ops); + exynos_hdmi_ops_register(&hdmi_ops); hdmi_resource_poweron(hdata); @@ -2351,7 +2349,7 @@ err_data: static int __devexit hdmi_remove(struct platform_device *pdev) { struct exynos_drm_hdmi_context *ctx = platform_get_drvdata(pdev); - struct hdmi_context *hdata = (struct hdmi_context *)ctx->ctx; + struct hdmi_context *hdata = ctx->ctx; DRM_DEBUG_KMS("[%d] %s\n", __LINE__, __func__); diff --git a/drivers/gpu/drm/exynos/exynos_mixer.c b/drivers/gpu/drm/exynos/exynos_mixer.c index 4d5f41e..e15438c 100644 --- a/drivers/gpu/drm/exynos/exynos_mixer.c +++ b/drivers/gpu/drm/exynos/exynos_mixer.c @@ -37,7 +37,8 @@ #include "exynos_drm_drv.h" #include "exynos_drm_hdmi.h" -#define HDMI_OVERLAY_NUMBER 3 +#define MIXER_WIN_NR 3 +#define MIXER_DEFAULT_WIN 0 #define get_mixer_context(dev) platform_get_drvdata(to_platform_device(dev)) @@ -75,16 +76,12 @@ struct mixer_resources { }; struct mixer_context { - struct fb_videomode *default_timing; - unsigned int default_win; - unsigned int default_bpp; unsigned int irq; int pipe; bool interlace; - bool vp_enabled; struct mixer_resources mixer_res; - struct hdmi_win_data win_data[HDMI_OVERLAY_NUMBER]; + struct hdmi_win_data win_data[MIXER_WIN_NR]; }; static const u8 filter_y_horiz_tap8[] = { @@ -643,9 +640,9 @@ static void mixer_win_mode_set(void *ctx, win = overlay->zpos; if (win == DEFAULT_ZPOS) - win = mixer_ctx->default_win; + win = MIXER_DEFAULT_WIN; - if (win < 0 || win > HDMI_OVERLAY_NUMBER) { + if (win < 0 || win > MIXER_WIN_NR) { DRM_ERROR("overlay plane[%d] is wrong\n", win); return; } @@ -683,9 +680,9 @@ static void mixer_win_commit(void *ctx, int zpos) DRM_DEBUG_KMS("[%d] %s, win: %d\n", __LINE__, __func__, win); if (win == DEFAULT_ZPOS) - win = mixer_ctx->default_win; + win = MIXER_DEFAULT_WIN; - if (win < 0 || win > HDMI_OVERLAY_NUMBER) { + if (win < 0 || win > MIXER_WIN_NR) { DRM_ERROR("overlay plane[%d] is wrong\n", win); return; } @@ -706,9 +703,9 @@ static void mixer_win_disable(void *ctx, int zpos) DRM_DEBUG_KMS("[%d] %s, win: %d\n", __LINE__, __func__, win); if (win == DEFAULT_ZPOS) - win = mixer_ctx->default_win; + win = MIXER_DEFAULT_WIN; - if (win < 0 || win > HDMI_OVERLAY_NUMBER) { + if (win < 0 || win > MIXER_WIN_NR) { DRM_ERROR("overlay plane[%d] is wrong\n", win); return; } @@ -722,9 +719,12 @@ static void mixer_win_disable(void *ctx, int zpos) spin_unlock_irqrestore(&res->reg_slock, flags); } -static struct exynos_hdmi_overlay_ops overlay_ops = { +static struct exynos_mixer_ops mixer_ops = { + /* manager */ .enable_vblank = mixer_enable_vblank, .disable_vblank = mixer_disable_vblank, + + /* overlay */ .win_mode_set = mixer_win_mode_set, .win_commit = mixer_win_commit, .win_disable = mixer_win_disable, @@ -771,8 +771,7 @@ static void mixer_finish_pageflip(struct drm_device *drm_dev, int crtc) static irqreturn_t mixer_irq_handler(int irq, void *arg) { struct exynos_drm_hdmi_context *drm_hdmi_ctx = arg; - struct mixer_context *ctx = - (struct mixer_context *)drm_hdmi_ctx->ctx; + struct mixer_context *ctx = drm_hdmi_ctx->ctx; struct mixer_resources *res = &ctx->mixer_res; u32 val, val_base; @@ -902,7 +901,7 @@ static int mixer_runtime_resume(struct device *dev) DRM_DEBUG_KMS("resume - start\n"); - mixer_resource_poweron((struct mixer_context *)ctx->ctx); + mixer_resource_poweron(ctx->ctx); return 0; } @@ -913,7 +912,7 @@ static int mixer_runtime_suspend(struct device *dev) DRM_DEBUG_KMS("suspend - start\n"); - mixer_resource_poweroff((struct mixer_context *)ctx->ctx); + mixer_resource_poweroff(ctx->ctx); return 0; } @@ -926,8 +925,7 @@ static const struct dev_pm_ops mixer_pm_ops = { static int __devinit mixer_resources_init(struct exynos_drm_hdmi_context *ctx, struct platform_device *pdev) { - struct mixer_context *mixer_ctx = - (struct mixer_context *)ctx->ctx; + struct mixer_context *mixer_ctx = ctx->ctx; struct device *dev = &pdev->dev; struct mixer_resources *mixer_res = &mixer_ctx->mixer_res; struct resource *res; @@ -1076,7 +1074,7 @@ static int __devinit mixer_probe(struct platform_device *pdev) goto fail; /* register specific callback point to common hdmi. */ - exynos_drm_overlay_ops_register(&overlay_ops); + exynos_mixer_ops_register(&mixer_ops); mixer_resource_poweron(ctx); @@ -1093,7 +1091,7 @@ static int mixer_remove(struct platform_device *pdev) struct device *dev = &pdev->dev; struct exynos_drm_hdmi_context *drm_hdmi_ctx = platform_get_drvdata(pdev); - struct mixer_context *ctx = (struct mixer_context *)drm_hdmi_ctx->ctx; + struct mixer_context *ctx = drm_hdmi_ctx->ctx; dev_info(dev, "remove successful\n"); diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile index ce7fc77..b65c06f 100644 --- a/drivers/gpu/drm/i915/Makefile +++ b/drivers/gpu/drm/i915/Makefile @@ -12,6 +12,7 @@ i915-y := i915_drv.o i915_dma.o i915_irq.o \ i915_gem_execbuffer.o \ i915_gem_gtt.o \ i915_gem_tiling.o \ + i915_sysfs.o \ i915_trace_points.o \ intel_display.o \ intel_crt.o \ @@ -22,6 +23,7 @@ i915-y := i915_drv.o i915_dma.o i915_irq.o \ intel_sdvo.o \ intel_modes.o \ intel_panel.o \ + intel_pm.o \ intel_i2c.o \ intel_fb.o \ intel_tv.o \ diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c index 967fb92..35462df 100644 --- a/drivers/gpu/drm/i915/i915_debugfs.c +++ b/drivers/gpu/drm/i915/i915_debugfs.c @@ -1171,6 +1171,17 @@ static int gen6_drpc_info(struct seq_file *m) seq_printf(m, "Core Power Down: %s\n", yesno(gt_core_status & GEN6_CORE_CPD_STATE_MASK)); + + /* Not exactly sure what this is */ + seq_printf(m, "RC6 \"Locked to RPn\" residency since boot: %u\n", + I915_READ(GEN6_GT_GFX_RC6_LOCKED)); + seq_printf(m, "RC6 residency since boot: %u\n", + I915_READ(GEN6_GT_GFX_RC6)); + seq_printf(m, "RC6+ residency since boot: %u\n", + I915_READ(GEN6_GT_GFX_RC6p)); + seq_printf(m, "RC6++ residency since boot: %u\n", + I915_READ(GEN6_GT_GFX_RC6pp)); + return 0; } @@ -1821,7 +1832,7 @@ static int i915_forcewake_open(struct inode *inode, struct file *file) return 0; } -int i915_forcewake_release(struct inode *inode, struct file *file) +static int i915_forcewake_release(struct inode *inode, struct file *file) { struct drm_device *dev = inode->i_private; struct drm_i915_private *dev_priv = dev->dev_private; diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c index 652f43f..a813f65 100644 --- a/drivers/gpu/drm/i915/i915_dma.c +++ b/drivers/gpu/drm/i915/i915_dma.c @@ -2115,7 +2115,7 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags) spin_lock_init(&dev_priv->error_lock); spin_lock_init(&dev_priv->rps_lock); - if (IS_IVYBRIDGE(dev)) + if (IS_IVYBRIDGE(dev) || IS_HASWELL(dev)) dev_priv->num_pipe = 3; else if (IS_MOBILE(dev) || !IS_GEN2(dev)) dev_priv->num_pipe = 2; @@ -2139,6 +2139,8 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags) } } + i915_setup_sysfs(dev); + /* Must be done after probing outputs */ intel_opregion_init(dev); acpi_video_register(); @@ -2190,6 +2192,8 @@ int i915_driver_unload(struct drm_device *dev) i915_mch_dev = NULL; spin_unlock(&mchdev_lock); + i915_teardown_sysfs(dev); + if (dev_priv->mm.inactive_shrinker.shrink) unregister_shrinker(&dev_priv->mm.inactive_shrinker); diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index c33b0a4..3effcf7 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -64,7 +64,7 @@ MODULE_PARM_DESC(semaphores, "Use semaphores for inter-ring sync (default: -1 (use per-chip defaults))"); int i915_enable_rc6 __read_mostly = -1; -module_param_named(i915_enable_rc6, i915_enable_rc6, int, 0600); +module_param_named(i915_enable_rc6, i915_enable_rc6, int, 0400); MODULE_PARM_DESC(i915_enable_rc6, "Enable power-saving render C-state 6. " "Different stages can be selected via bitmask values " @@ -394,6 +394,21 @@ void intel_detect_pch(struct drm_device *dev) } } +bool i915_semaphore_is_enabled(struct drm_device *dev) +{ + if (INTEL_INFO(dev)->gen < 6) + return 0; + + if (i915_semaphores >= 0) + return i915_semaphores; + + /* Enable semaphores on SNB when IO remapping is off */ + if (INTEL_INFO(dev)->gen == 6) + return !intel_iommu_enabled; + + return 1; +} + void __gen6_gt_force_wake_get(struct drm_i915_private *dev_priv) { int count; @@ -836,9 +851,14 @@ int i915_reset(struct drm_device *dev, u8 flags) i915_gem_init_ppgtt(dev); mutex_unlock(&dev->struct_mutex); + + if (drm_core_check_feature(dev, DRIVER_MODESET)) + intel_modeset_init_hw(dev); + drm_irq_uninstall(dev); drm_mode_config_reset(dev); drm_irq_install(dev); + mutex_lock(&dev->struct_mutex); } diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 92e496a..69e1539 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -38,6 +38,7 @@ #include <linux/i2c-algo-bit.h> #include <drm/intel-gtt.h> #include <linux/backlight.h> +#include <linux/intel-iommu.h> /* General customization: */ @@ -145,7 +146,6 @@ struct drm_i915_master_private { struct drm_i915_fence_reg { struct list_head lru_list; struct drm_i915_gem_object *obj; - uint32_t setup_seqno; int pin_count; }; @@ -930,13 +930,12 @@ struct drm_i915_gem_object { */ uint32_t gtt_offset; - /** Breadcrumb of last rendering to the buffer. */ - uint32_t last_rendering_seqno; struct intel_ring_buffer *ring; + /** Breadcrumb of last rendering to the buffer. */ + uint32_t last_rendering_seqno; /** Breadcrumb of last fenced GPU access to the buffer. */ uint32_t last_fenced_seqno; - struct intel_ring_buffer *last_fenced_ring; /** Current tiling stride for the object, if it's tiled. */ uint32_t stride; @@ -1127,8 +1126,10 @@ extern void i915_driver_preclose(struct drm_device *dev, extern void i915_driver_postclose(struct drm_device *dev, struct drm_file *file_priv); extern int i915_driver_device_is_agp(struct drm_device * dev); +#ifdef CONFIG_COMPAT extern long i915_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg); +#endif extern int i915_emit_box(struct drm_device *dev, struct drm_clip_rect *box, int DR1, int DR4); @@ -1230,6 +1231,8 @@ void i915_gem_lastclose(struct drm_device *dev); int __must_check i915_mutex_lock_interruptible(struct drm_device *dev); int __must_check i915_gem_object_wait_rendering(struct drm_i915_gem_object *obj); +int i915_gem_object_sync(struct drm_i915_gem_object *obj, + struct intel_ring_buffer *to); void i915_gem_object_move_to_active(struct drm_i915_gem_object *obj, struct intel_ring_buffer *ring, u32 seqno); @@ -1252,17 +1255,18 @@ i915_seqno_passed(uint32_t seq1, uint32_t seq2) u32 i915_gem_next_request_seqno(struct intel_ring_buffer *ring); -int __must_check i915_gem_object_get_fence(struct drm_i915_gem_object *obj, - struct intel_ring_buffer *pipelined); +int __must_check i915_gem_object_get_fence(struct drm_i915_gem_object *obj); int __must_check i915_gem_object_put_fence(struct drm_i915_gem_object *obj); -static inline void +static inline bool i915_gem_object_pin_fence(struct drm_i915_gem_object *obj) { if (obj->fence_reg != I915_FENCE_REG_NONE) { struct drm_i915_private *dev_priv = obj->base.dev->dev_private; dev_priv->fence_regs[obj->fence_reg].pin_count++; - } + return true; + } else + return false; } static inline void @@ -1380,6 +1384,10 @@ extern int i915_restore_state(struct drm_device *dev); extern int i915_save_state(struct drm_device *dev); extern int i915_restore_state(struct drm_device *dev); +/* i915_sysfs.c */ +void i915_setup_sysfs(struct drm_device *dev_priv); +void i915_teardown_sysfs(struct drm_device *dev_priv); + /* intel_i2c.c */ extern int intel_setup_gmbus(struct drm_device *dev); extern void intel_teardown_gmbus(struct drm_device *dev); @@ -1424,6 +1432,7 @@ static inline void intel_unregister_dsm_handler(void) { return; } #endif /* CONFIG_ACPI */ /* modesetting */ +extern void intel_modeset_init_hw(struct drm_device *dev); extern void intel_modeset_init(struct drm_device *dev); extern void intel_modeset_gem_init(struct drm_device *dev); extern void intel_modeset_cleanup(struct drm_device *dev); @@ -1436,7 +1445,9 @@ extern void ironlake_enable_rc6(struct drm_device *dev); extern void gen6_set_rps(struct drm_device *dev, u8 val); extern void intel_detect_pch(struct drm_device *dev); extern int intel_trans_dp_port_sel(struct drm_crtc *crtc); +extern int intel_enable_rc6(const struct drm_device *dev); +extern bool i915_semaphore_is_enabled(struct drm_device *dev); extern void __gen6_gt_force_wake_get(struct drm_i915_private *dev_priv); extern void __gen6_gt_force_wake_mt_get(struct drm_i915_private *dev_priv); extern void __gen6_gt_force_wake_put(struct drm_i915_private *dev_priv); diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index b851bd3..7bc4a40 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -42,18 +42,34 @@ static void i915_gem_object_flush_cpu_write_domain(struct drm_i915_gem_object *o static __must_check int i915_gem_object_bind_to_gtt(struct drm_i915_gem_object *obj, unsigned alignment, bool map_and_fenceable); -static void i915_gem_clear_fence_reg(struct drm_device *dev, - struct drm_i915_fence_reg *reg); static int i915_gem_phys_pwrite(struct drm_device *dev, struct drm_i915_gem_object *obj, struct drm_i915_gem_pwrite *args, struct drm_file *file); static void i915_gem_free_object_tail(struct drm_i915_gem_object *obj); +static void i915_gem_write_fence(struct drm_device *dev, int reg, + struct drm_i915_gem_object *obj); +static void i915_gem_object_update_fence(struct drm_i915_gem_object *obj, + struct drm_i915_fence_reg *fence, + bool enable); + static int i915_gem_inactive_shrink(struct shrinker *shrinker, struct shrink_control *sc); static void i915_gem_object_truncate(struct drm_i915_gem_object *obj); +static inline void i915_gem_object_fence_lost(struct drm_i915_gem_object *obj) +{ + if (obj->tiling_mode) + i915_gem_release_mmap(obj); + + /* As we do not have an associated fence register, we will force + * a tiling change if we ever need to acquire one. + */ + obj->tiling_changed = false; + obj->fence_reg = I915_FENCE_REG_NONE; +} + /* some bookkeeping */ static void i915_gem_info_add_obj(struct drm_i915_private *dev_priv, size_t size) @@ -876,6 +892,7 @@ i915_gem_pwrite_ioctl(struct drm_device *dev, void *data, if (obj->gtt_space && obj->cache_level == I915_CACHE_NONE && + obj->tiling_mode == I915_TILING_NONE && obj->map_and_fenceable && obj->base.write_domain != I915_GEM_DOMAIN_CPU) { ret = i915_gem_gtt_pwrite_fast(dev, obj, args, file); @@ -1078,10 +1095,7 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf) if (!obj->has_global_gtt_mapping) i915_gem_gtt_bind_object(obj, obj->cache_level); - if (obj->tiling_mode == I915_TILING_NONE) - ret = i915_gem_object_put_fence(obj); - else - ret = i915_gem_object_get_fence(obj, NULL); + ret = i915_gem_object_get_fence(obj); if (ret) goto unlock; @@ -1400,7 +1414,6 @@ i915_gem_object_move_to_active(struct drm_i915_gem_object *obj, if (obj->fenced_gpu_access) { obj->last_fenced_seqno = seqno; - obj->last_fenced_ring = ring; /* Bump MRU to take account of the delayed flush */ if (obj->fence_reg != I915_FENCE_REG_NONE) { @@ -1418,6 +1431,7 @@ i915_gem_object_move_off_active(struct drm_i915_gem_object *obj) { list_del_init(&obj->ring_list); obj->last_rendering_seqno = 0; + obj->last_fenced_seqno = 0; } static void @@ -1639,20 +1653,18 @@ static void i915_gem_reset_fences(struct drm_device *dev) for (i = 0; i < dev_priv->num_fence_regs; i++) { struct drm_i915_fence_reg *reg = &dev_priv->fence_regs[i]; - struct drm_i915_gem_object *obj = reg->obj; - if (!obj) - continue; + i915_gem_write_fence(dev, i, NULL); - if (obj->tiling_mode) - i915_gem_release_mmap(obj); + if (reg->obj) + i915_gem_object_fence_lost(reg->obj); - reg->obj->fence_reg = I915_FENCE_REG_NONE; - reg->obj->fenced_gpu_access = false; - reg->obj->last_fenced_seqno = 0; - reg->obj->last_fenced_ring = NULL; - i915_gem_clear_fence_reg(dev, reg); + reg->pin_count = 0; + reg->obj = NULL; + INIT_LIST_HEAD(®->lru_list); } + + INIT_LIST_HEAD(&dev_priv->mm.fence_list); } void i915_gem_reset(struct drm_device *dev) @@ -1956,6 +1968,62 @@ i915_gem_object_wait_rendering(struct drm_i915_gem_object *obj) return 0; } +/** + * i915_gem_object_sync - sync an object to a ring. + * + * @obj: object which may be in use on another ring. + * @to: ring we wish to use the object on. May be NULL. + * + * This code is meant to abstract object synchronization with the GPU. + * Calling with NULL implies synchronizing the object with the CPU + * rather than a particular GPU ring. + * + * Returns 0 if successful, else propagates up the lower layer error. + */ +int +i915_gem_object_sync(struct drm_i915_gem_object *obj, + struct intel_ring_buffer *to) +{ + struct intel_ring_buffer *from = obj->ring; + u32 seqno; + int ret, idx; + + if (from == NULL || to == from) + return 0; + + if (to == NULL || !i915_semaphore_is_enabled(obj->base.dev)) + return i915_gem_object_wait_rendering(obj); + + idx = intel_ring_sync_index(from, to); + + seqno = obj->last_rendering_seqno; + if (seqno <= from->sync_seqno[idx]) + return 0; + + if (seqno == from->outstanding_lazy_request) { + struct drm_i915_gem_request *request; + + request = kzalloc(sizeof(*request), GFP_KERNEL); + if (request == NULL) + return -ENOMEM; + + ret = i915_add_request(from, NULL, request); + if (ret) { + kfree(request); + return ret; + } + + seqno = request->seqno; + } + + + ret = to->sync_to(to, from, seqno); + if (!ret) + from->sync_seqno[idx] = seqno; + + return ret; +} + static void i915_gem_object_finish_gtt(struct drm_i915_gem_object *obj) { u32 old_write_domain, old_read_domains; @@ -2110,189 +2178,178 @@ int i915_gpu_idle(struct drm_device *dev, bool do_retire) return 0; } -static int sandybridge_write_fence_reg(struct drm_i915_gem_object *obj, - struct intel_ring_buffer *pipelined) +static void sandybridge_write_fence_reg(struct drm_device *dev, int reg, + struct drm_i915_gem_object *obj) { - struct drm_device *dev = obj->base.dev; drm_i915_private_t *dev_priv = dev->dev_private; - u32 size = obj->gtt_space->size; - int regnum = obj->fence_reg; uint64_t val; - val = (uint64_t)((obj->gtt_offset + size - 4096) & - 0xfffff000) << 32; - val |= obj->gtt_offset & 0xfffff000; - val |= (uint64_t)((obj->stride / 128) - 1) << - SANDYBRIDGE_FENCE_PITCH_SHIFT; - - if (obj->tiling_mode == I915_TILING_Y) - val |= 1 << I965_FENCE_TILING_Y_SHIFT; - val |= I965_FENCE_REG_VALID; + if (obj) { + u32 size = obj->gtt_space->size; - if (pipelined) { - int ret = intel_ring_begin(pipelined, 6); - if (ret) - return ret; + val = (uint64_t)((obj->gtt_offset + size - 4096) & + 0xfffff000) << 32; + val |= obj->gtt_offset & 0xfffff000; + val |= (uint64_t)((obj->stride / 128) - 1) << + SANDYBRIDGE_FENCE_PITCH_SHIFT; - intel_ring_emit(pipelined, MI_NOOP); - intel_ring_emit(pipelined, MI_LOAD_REGISTER_IMM(2)); - intel_ring_emit(pipelined, FENCE_REG_SANDYBRIDGE_0 + regnum*8); - intel_ring_emit(pipelined, (u32)val); - intel_ring_emit(pipelined, FENCE_REG_SANDYBRIDGE_0 + regnum*8 + 4); - intel_ring_emit(pipelined, (u32)(val >> 32)); - intel_ring_advance(pipelined); + if (obj->tiling_mode == I915_TILING_Y) + val |= 1 << I965_FENCE_TILING_Y_SHIFT; + val |= I965_FENCE_REG_VALID; } else - I915_WRITE64(FENCE_REG_SANDYBRIDGE_0 + regnum * 8, val); + val = 0; - return 0; + I915_WRITE64(FENCE_REG_SANDYBRIDGE_0 + reg * 8, val); + POSTING_READ(FENCE_REG_SANDYBRIDGE_0 + reg * 8); } -static int i965_write_fence_reg(struct drm_i915_gem_object *obj, - struct intel_ring_buffer *pipelined) +static void i965_write_fence_reg(struct drm_device *dev, int reg, + struct drm_i915_gem_object *obj) { - struct drm_device *dev = obj->base.dev; drm_i915_private_t *dev_priv = dev->dev_private; - u32 size = obj->gtt_space->size; - int regnum = obj->fence_reg; uint64_t val; - val = (uint64_t)((obj->gtt_offset + size - 4096) & - 0xfffff000) << 32; - val |= obj->gtt_offset & 0xfffff000; - val |= ((obj->stride / 128) - 1) << I965_FENCE_PITCH_SHIFT; - if (obj->tiling_mode == I915_TILING_Y) - val |= 1 << I965_FENCE_TILING_Y_SHIFT; - val |= I965_FENCE_REG_VALID; - - if (pipelined) { - int ret = intel_ring_begin(pipelined, 6); - if (ret) - return ret; + if (obj) { + u32 size = obj->gtt_space->size; - intel_ring_emit(pipelined, MI_NOOP); - intel_ring_emit(pipelined, MI_LOAD_REGISTER_IMM(2)); - intel_ring_emit(pipelined, FENCE_REG_965_0 + regnum*8); - intel_ring_emit(pipelined, (u32)val); - intel_ring_emit(pipelined, FENCE_REG_965_0 + regnum*8 + 4); - intel_ring_emit(pipelined, (u32)(val >> 32)); - intel_ring_advance(pipelined); + val = (uint64_t)((obj->gtt_offset + size - 4096) & + 0xfffff000) << 32; + val |= obj->gtt_offset & 0xfffff000; + val |= ((obj->stride / 128) - 1) << I965_FENCE_PITCH_SHIFT; + if (obj->tiling_mode == I915_TILING_Y) + val |= 1 << I965_FENCE_TILING_Y_SHIFT; + val |= I965_FENCE_REG_VALID; } else - I915_WRITE64(FENCE_REG_965_0 + regnum * 8, val); + val = 0; - return 0; + I915_WRITE64(FENCE_REG_965_0 + reg * 8, val); + POSTING_READ(FENCE_REG_965_0 + reg * 8); } -static int i915_write_fence_reg(struct drm_i915_gem_object *obj, - struct intel_ring_buffer *pipelined) +static void i915_write_fence_reg(struct drm_device *dev, int reg, + struct drm_i915_gem_object *obj) { - struct drm_device *dev = obj->base.dev; drm_i915_private_t *dev_priv = dev->dev_private; - u32 size = obj->gtt_space->size; - u32 fence_reg, val, pitch_val; - int tile_width; - - if (WARN((obj->gtt_offset & ~I915_FENCE_START_MASK) || - (size & -size) != size || - (obj->gtt_offset & (size - 1)), - "object 0x%08x [fenceable? %d] not 1M or pot-size (0x%08x) aligned\n", - obj->gtt_offset, obj->map_and_fenceable, size)) - return -EINVAL; + u32 val; - if (obj->tiling_mode == I915_TILING_Y && HAS_128_BYTE_Y_TILING(dev)) - tile_width = 128; - else - tile_width = 512; - - /* Note: pitch better be a power of two tile widths */ - pitch_val = obj->stride / tile_width; - pitch_val = ffs(pitch_val) - 1; - - val = obj->gtt_offset; - if (obj->tiling_mode == I915_TILING_Y) - val |= 1 << I830_FENCE_TILING_Y_SHIFT; - val |= I915_FENCE_SIZE_BITS(size); - val |= pitch_val << I830_FENCE_PITCH_SHIFT; - val |= I830_FENCE_REG_VALID; - - fence_reg = obj->fence_reg; - if (fence_reg < 8) - fence_reg = FENCE_REG_830_0 + fence_reg * 4; - else - fence_reg = FENCE_REG_945_8 + (fence_reg - 8) * 4; + if (obj) { + u32 size = obj->gtt_space->size; + int pitch_val; + int tile_width; - if (pipelined) { - int ret = intel_ring_begin(pipelined, 4); - if (ret) - return ret; + WARN((obj->gtt_offset & ~I915_FENCE_START_MASK) || + (size & -size) != size || + (obj->gtt_offset & (size - 1)), + "object 0x%08x [fenceable? %d] not 1M or pot-size (0x%08x) aligned\n", + obj->gtt_offset, obj->map_and_fenceable, size); - intel_ring_emit(pipelined, MI_NOOP); - intel_ring_emit(pipelined, MI_LOAD_REGISTER_IMM(1)); - intel_ring_emit(pipelined, fence_reg); - intel_ring_emit(pipelined, val); - intel_ring_advance(pipelined); + if (obj->tiling_mode == I915_TILING_Y && HAS_128_BYTE_Y_TILING(dev)) + tile_width = 128; + else + tile_width = 512; + + /* Note: pitch better be a power of two tile widths */ + pitch_val = obj->stride / tile_width; + pitch_val = ffs(pitch_val) - 1; + + val = obj->gtt_offset; + if (obj->tiling_mode == I915_TILING_Y) + val |= 1 << I830_FENCE_TILING_Y_SHIFT; + val |= I915_FENCE_SIZE_BITS(size); + val |= pitch_val << I830_FENCE_PITCH_SHIFT; + val |= I830_FENCE_REG_VALID; } else - I915_WRITE(fence_reg, val); + val = 0; - return 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); } -static int i830_write_fence_reg(struct drm_i915_gem_object *obj, - struct intel_ring_buffer *pipelined) +static void i830_write_fence_reg(struct drm_device *dev, int reg, + struct drm_i915_gem_object *obj) { - struct drm_device *dev = obj->base.dev; drm_i915_private_t *dev_priv = dev->dev_private; - u32 size = obj->gtt_space->size; - int regnum = obj->fence_reg; uint32_t val; - uint32_t pitch_val; - if (WARN((obj->gtt_offset & ~I830_FENCE_START_MASK) || - (size & -size) != size || - (obj->gtt_offset & (size - 1)), - "object 0x%08x not 512K or pot-size 0x%08x aligned\n", - obj->gtt_offset, size)) - return -EINVAL; - - pitch_val = obj->stride / 128; - pitch_val = ffs(pitch_val) - 1; - - val = obj->gtt_offset; - if (obj->tiling_mode == I915_TILING_Y) - val |= 1 << I830_FENCE_TILING_Y_SHIFT; - val |= I830_FENCE_SIZE_BITS(size); - val |= pitch_val << I830_FENCE_PITCH_SHIFT; - val |= I830_FENCE_REG_VALID; + if (obj) { + u32 size = obj->gtt_space->size; + uint32_t pitch_val; + + WARN((obj->gtt_offset & ~I830_FENCE_START_MASK) || + (size & -size) != size || + (obj->gtt_offset & (size - 1)), + "object 0x%08x not 512K or pot-size 0x%08x aligned\n", + obj->gtt_offset, size); + + pitch_val = obj->stride / 128; + pitch_val = ffs(pitch_val) - 1; + + val = obj->gtt_offset; + if (obj->tiling_mode == I915_TILING_Y) + val |= 1 << I830_FENCE_TILING_Y_SHIFT; + val |= I830_FENCE_SIZE_BITS(size); + val |= pitch_val << I830_FENCE_PITCH_SHIFT; + val |= I830_FENCE_REG_VALID; + } else + val = 0; - if (pipelined) { - int ret = intel_ring_begin(pipelined, 4); - if (ret) - return ret; + I915_WRITE(FENCE_REG_830_0 + reg * 4, val); + POSTING_READ(FENCE_REG_830_0 + reg * 4); +} - intel_ring_emit(pipelined, MI_NOOP); - intel_ring_emit(pipelined, MI_LOAD_REGISTER_IMM(1)); - intel_ring_emit(pipelined, FENCE_REG_830_0 + regnum*4); - intel_ring_emit(pipelined, val); - intel_ring_advance(pipelined); - } else - I915_WRITE(FENCE_REG_830_0 + regnum * 4, val); +static void i915_gem_write_fence(struct drm_device *dev, int reg, + struct drm_i915_gem_object *obj) +{ + switch (INTEL_INFO(dev)->gen) { + case 7: + case 6: sandybridge_write_fence_reg(dev, reg, obj); break; + case 5: + case 4: i965_write_fence_reg(dev, reg, obj); break; + case 3: i915_write_fence_reg(dev, reg, obj); break; + case 2: i830_write_fence_reg(dev, reg, obj); break; + default: break; + } +} - return 0; +static inline int fence_number(struct drm_i915_private *dev_priv, + struct drm_i915_fence_reg *fence) +{ + return fence - dev_priv->fence_regs; } -static bool ring_passed_seqno(struct intel_ring_buffer *ring, u32 seqno) +static void i915_gem_object_update_fence(struct drm_i915_gem_object *obj, + struct drm_i915_fence_reg *fence, + bool enable) { - return i915_seqno_passed(ring->get_seqno(ring), seqno); + struct drm_i915_private *dev_priv = obj->base.dev->dev_private; + int reg = fence_number(dev_priv, fence); + + i915_gem_write_fence(obj->base.dev, reg, enable ? obj : NULL); + + if (enable) { + obj->fence_reg = reg; + fence->obj = obj; + list_move_tail(&fence->lru_list, &dev_priv->mm.fence_list); + } else { + obj->fence_reg = I915_FENCE_REG_NONE; + fence->obj = NULL; + list_del_init(&fence->lru_list); + } } static int -i915_gem_object_flush_fence(struct drm_i915_gem_object *obj, - struct intel_ring_buffer *pipelined) +i915_gem_object_flush_fence(struct drm_i915_gem_object *obj) { int ret; if (obj->fenced_gpu_access) { if (obj->base.write_domain & I915_GEM_GPU_DOMAINS) { - ret = i915_gem_flush_ring(obj->last_fenced_ring, + ret = i915_gem_flush_ring(obj->ring, 0, obj->base.write_domain); if (ret) return ret; @@ -2301,18 +2358,14 @@ i915_gem_object_flush_fence(struct drm_i915_gem_object *obj, obj->fenced_gpu_access = false; } - if (obj->last_fenced_seqno && pipelined != obj->last_fenced_ring) { - if (!ring_passed_seqno(obj->last_fenced_ring, - obj->last_fenced_seqno)) { - ret = i915_wait_request(obj->last_fenced_ring, - obj->last_fenced_seqno, - true); - if (ret) - return ret; - } + if (obj->last_fenced_seqno) { + ret = i915_wait_request(obj->ring, + obj->last_fenced_seqno, + false); + if (ret) + return ret; obj->last_fenced_seqno = 0; - obj->last_fenced_ring = NULL; } /* Ensure that all CPU reads are completed before installing a fence @@ -2327,34 +2380,29 @@ i915_gem_object_flush_fence(struct drm_i915_gem_object *obj, int i915_gem_object_put_fence(struct drm_i915_gem_object *obj) { + struct drm_i915_private *dev_priv = obj->base.dev->dev_private; int ret; - if (obj->tiling_mode) - i915_gem_release_mmap(obj); - - ret = i915_gem_object_flush_fence(obj, NULL); + ret = i915_gem_object_flush_fence(obj); if (ret) return ret; - if (obj->fence_reg != I915_FENCE_REG_NONE) { - struct drm_i915_private *dev_priv = obj->base.dev->dev_private; - - WARN_ON(dev_priv->fence_regs[obj->fence_reg].pin_count); - i915_gem_clear_fence_reg(obj->base.dev, - &dev_priv->fence_regs[obj->fence_reg]); + if (obj->fence_reg == I915_FENCE_REG_NONE) + return 0; - obj->fence_reg = I915_FENCE_REG_NONE; - } + i915_gem_object_update_fence(obj, + &dev_priv->fence_regs[obj->fence_reg], + false); + i915_gem_object_fence_lost(obj); return 0; } static struct drm_i915_fence_reg * -i915_find_fence_reg(struct drm_device *dev, - struct intel_ring_buffer *pipelined) +i915_find_fence_reg(struct drm_device *dev) { struct drm_i915_private *dev_priv = dev->dev_private; - struct drm_i915_fence_reg *reg, *first, *avail; + struct drm_i915_fence_reg *reg, *avail; int i; /* First try to find a free reg */ @@ -2372,204 +2420,77 @@ i915_find_fence_reg(struct drm_device *dev, return NULL; /* None available, try to steal one or wait for a user to finish */ - avail = first = NULL; list_for_each_entry(reg, &dev_priv->mm.fence_list, lru_list) { if (reg->pin_count) continue; - if (first == NULL) - first = reg; - - if (!pipelined || - !reg->obj->last_fenced_ring || - reg->obj->last_fenced_ring == pipelined) { - avail = reg; - break; - } + return reg; } - if (avail == NULL) - avail = first; - - return avail; + return NULL; } /** - * i915_gem_object_get_fence - set up a fence reg for an object + * i915_gem_object_get_fence - set up fencing for an object * @obj: object to map through a fence reg - * @pipelined: ring on which to queue the change, or NULL for CPU access - * @interruptible: must we wait uninterruptibly for the register to retire? * * When mapping objects through the GTT, userspace wants to be able to write * to them without having to worry about swizzling if the object is tiled. - * * This function walks the fence regs looking for a free one for @obj, * stealing one if it can't find any. * * It then sets up the reg based on the object's properties: address, pitch * and tiling format. + * + * For an untiled surface, this removes any existing fence. */ int -i915_gem_object_get_fence(struct drm_i915_gem_object *obj, - struct intel_ring_buffer *pipelined) +i915_gem_object_get_fence(struct drm_i915_gem_object *obj) { struct drm_device *dev = obj->base.dev; struct drm_i915_private *dev_priv = dev->dev_private; + bool enable = obj->tiling_mode != I915_TILING_NONE; struct drm_i915_fence_reg *reg; int ret; - /* XXX disable pipelining. There are bugs. Shocking. */ - pipelined = NULL; + /* Have we updated the tiling parameters upon the object and so + * will need to serialise the write to the associated fence register? + */ + if (obj->tiling_changed) { + ret = i915_gem_object_flush_fence(obj); + if (ret) + return ret; + } /* Just update our place in the LRU if our fence is getting reused. */ if (obj->fence_reg != I915_FENCE_REG_NONE) { reg = &dev_priv->fence_regs[obj->fence_reg]; - list_move_tail(®->lru_list, &dev_priv->mm.fence_list); - - if (obj->tiling_changed) { - ret = i915_gem_object_flush_fence(obj, pipelined); - if (ret) - return ret; - - if (!obj->fenced_gpu_access && !obj->last_fenced_seqno) - pipelined = NULL; - - if (pipelined) { - reg->setup_seqno = - i915_gem_next_request_seqno(pipelined); - obj->last_fenced_seqno = reg->setup_seqno; - obj->last_fenced_ring = pipelined; - } - - goto update; + if (!obj->tiling_changed) { + list_move_tail(®->lru_list, + &dev_priv->mm.fence_list); + return 0; } + } else if (enable) { + reg = i915_find_fence_reg(dev); + if (reg == NULL) + return -EDEADLK; - if (!pipelined) { - if (reg->setup_seqno) { - if (!ring_passed_seqno(obj->last_fenced_ring, - reg->setup_seqno)) { - ret = i915_wait_request(obj->last_fenced_ring, - reg->setup_seqno, - true); - if (ret) - return ret; - } + if (reg->obj) { + struct drm_i915_gem_object *old = reg->obj; - reg->setup_seqno = 0; - } - } else if (obj->last_fenced_ring && - obj->last_fenced_ring != pipelined) { - ret = i915_gem_object_flush_fence(obj, pipelined); + ret = i915_gem_object_flush_fence(old); if (ret) return ret; - } - - return 0; - } - - reg = i915_find_fence_reg(dev, pipelined); - if (reg == NULL) - return -EDEADLK; - - ret = i915_gem_object_flush_fence(obj, pipelined); - if (ret) - return ret; - - if (reg->obj) { - struct drm_i915_gem_object *old = reg->obj; - - drm_gem_object_reference(&old->base); - - if (old->tiling_mode) - i915_gem_release_mmap(old); - ret = i915_gem_object_flush_fence(old, pipelined); - if (ret) { - drm_gem_object_unreference(&old->base); - return ret; + i915_gem_object_fence_lost(old); } + } else + return 0; - if (old->last_fenced_seqno == 0 && obj->last_fenced_seqno == 0) - pipelined = NULL; - - old->fence_reg = I915_FENCE_REG_NONE; - old->last_fenced_ring = pipelined; - old->last_fenced_seqno = - pipelined ? i915_gem_next_request_seqno(pipelined) : 0; - - drm_gem_object_unreference(&old->base); - } else if (obj->last_fenced_seqno == 0) - pipelined = NULL; - - reg->obj = obj; - list_move_tail(®->lru_list, &dev_priv->mm.fence_list); - obj->fence_reg = reg - dev_priv->fence_regs; - obj->last_fenced_ring = pipelined; - - reg->setup_seqno = - pipelined ? i915_gem_next_request_seqno(pipelined) : 0; - obj->last_fenced_seqno = reg->setup_seqno; - -update: + i915_gem_object_update_fence(obj, reg, enable); obj->tiling_changed = false; - switch (INTEL_INFO(dev)->gen) { - case 7: - case 6: - ret = sandybridge_write_fence_reg(obj, pipelined); - break; - case 5: - case 4: - ret = i965_write_fence_reg(obj, pipelined); - break; - case 3: - ret = i915_write_fence_reg(obj, pipelined); - break; - case 2: - ret = i830_write_fence_reg(obj, pipelined); - break; - } - - return ret; -} - -/** - * i915_gem_clear_fence_reg - clear out fence register info - * @obj: object to clear - * - * Zeroes out the fence register itself and clears out the associated - * data structures in dev_priv and obj. - */ -static void -i915_gem_clear_fence_reg(struct drm_device *dev, - struct drm_i915_fence_reg *reg) -{ - drm_i915_private_t *dev_priv = dev->dev_private; - uint32_t fence_reg = reg - dev_priv->fence_regs; - switch (INTEL_INFO(dev)->gen) { - case 7: - case 6: - I915_WRITE64(FENCE_REG_SANDYBRIDGE_0 + fence_reg*8, 0); - break; - case 5: - case 4: - I915_WRITE64(FENCE_REG_965_0 + fence_reg*8, 0); - break; - case 3: - if (fence_reg >= 8) - fence_reg = FENCE_REG_945_8 + (fence_reg - 8) * 4; - else - case 2: - fence_reg = FENCE_REG_830_0 + fence_reg * 4; - - I915_WRITE(fence_reg, 0); - break; - } - - list_del_init(®->lru_list); - reg->obj = NULL; - reg->setup_seqno = 0; - reg->pin_count = 0; + return 0; } /** @@ -2926,11 +2847,6 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj, * Prepare buffer for display plane (scanout, cursors, etc). * Can be called from an uninterruptible phase (modesetting) and allows * any flushes to be pipelined (for pageflips). - * - * For the display plane, we want to be in the GTT but out of any write - * domains. So in many ways this looks like set_to_gtt_domain() apart from the - * ability to pipeline the waits, pinning and any additional subtleties - * that may differentiate the display plane from ordinary buffers. */ int i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj, @@ -2945,8 +2861,8 @@ i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj, return ret; if (pipelined != obj->ring) { - ret = i915_gem_object_wait_rendering(obj); - if (ret == -ERESTARTSYS) + ret = i915_gem_object_sync(obj, pipelined); + if (ret) return ret; } @@ -3031,9 +2947,11 @@ i915_gem_object_set_to_cpu_domain(struct drm_i915_gem_object *obj, bool write) if (ret) return ret; - ret = i915_gem_object_wait_rendering(obj); - if (ret) - return ret; + if (write || obj->pending_gpu_write) { + ret = i915_gem_object_wait_rendering(obj); + if (ret) + return ret; + } i915_gem_object_flush_gtt_write_domain(obj); @@ -3613,7 +3531,15 @@ void i915_gem_init_ppgtt(struct drm_device *dev) pd_offset <<= 16; if (INTEL_INFO(dev)->gen == 6) { - uint32_t ecochk = I915_READ(GAM_ECOCHK); + uint32_t ecochk, gab_ctl, ecobits; + + ecobits = I915_READ(GAC_ECO_BITS); + I915_WRITE(GAC_ECO_BITS, ecobits | ECOBITS_PPGTT_CACHE64B); + + gab_ctl = I915_READ(GAB_CTL); + I915_WRITE(GAB_CTL, gab_ctl | GAB_CTL_CONT_AFTER_PAGEFAULT); + + ecochk = I915_READ(GAM_ECOCHK); I915_WRITE(GAM_ECOCHK, ecochk | ECOCHK_SNB_BIT | ECOCHK_PPGTT_CACHE64B); I915_WRITE(GFX_MODE, GFX_MODE_ENABLE(GFX_PPGTT_ENABLE)); @@ -3804,9 +3730,7 @@ i915_gem_load(struct drm_device *dev) dev_priv->num_fence_regs = 8; /* Initialize fence registers to zero */ - for (i = 0; i < dev_priv->num_fence_regs; i++) { - i915_gem_clear_fence_reg(dev, &dev_priv->fence_regs[i]); - } + i915_gem_reset_fences(dev); i915_gem_detect_bit_6_swizzle(dev); init_waitqueue_head(&dev_priv->pending_flip_queue); diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c index 254e2f6..68ec013 100644 --- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c +++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c @@ -381,7 +381,11 @@ i915_gem_execbuffer_relocate_entry(struct drm_i915_gem_object *obj, uint32_t __iomem *reloc_entry; void __iomem *reloc_page; - ret = i915_gem_object_set_to_gtt_domain(obj, 1); + ret = i915_gem_object_set_to_gtt_domain(obj, true); + if (ret) + return ret; + + ret = i915_gem_object_put_fence(obj); if (ret) return ret; @@ -530,18 +534,13 @@ pin_and_fence_object(struct drm_i915_gem_object *obj, if (has_fenced_gpu_access) { if (entry->flags & EXEC_OBJECT_NEEDS_FENCE) { - if (obj->tiling_mode) { - ret = i915_gem_object_get_fence(obj, ring); - if (ret) - goto err_unpin; + ret = i915_gem_object_get_fence(obj); + if (ret) + goto err_unpin; + if (i915_gem_object_pin_fence(obj)) entry->flags |= __EXEC_OBJECT_HAS_FENCE; - i915_gem_object_pin_fence(obj); - } else { - ret = i915_gem_object_put_fence(obj); - if (ret) - goto err_unpin; - } + obj->pending_fenced_gpu_access = true; } } @@ -840,64 +839,6 @@ i915_gem_execbuffer_flush(struct drm_device *dev, return 0; } -static bool -intel_enable_semaphores(struct drm_device *dev) -{ - if (INTEL_INFO(dev)->gen < 6) - return 0; - - if (i915_semaphores >= 0) - return i915_semaphores; - - /* Disable semaphores on SNB */ - if (INTEL_INFO(dev)->gen == 6) - return 0; - - return 1; -} - -static int -i915_gem_execbuffer_sync_rings(struct drm_i915_gem_object *obj, - struct intel_ring_buffer *to) -{ - struct intel_ring_buffer *from = obj->ring; - u32 seqno; - int ret, idx; - - if (from == NULL || to == from) - return 0; - - /* XXX gpu semaphores are implicated in various hard hangs on SNB */ - if (!intel_enable_semaphores(obj->base.dev)) - return i915_gem_object_wait_rendering(obj); - - idx = intel_ring_sync_index(from, to); - - seqno = obj->last_rendering_seqno; - if (seqno <= from->sync_seqno[idx]) - return 0; - - if (seqno == from->outstanding_lazy_request) { - struct drm_i915_gem_request *request; - - request = kzalloc(sizeof(*request), GFP_KERNEL); - if (request == NULL) - return -ENOMEM; - - ret = i915_add_request(from, NULL, request); - if (ret) { - kfree(request); - return ret; - } - - seqno = request->seqno; - } - - from->sync_seqno[idx] = seqno; - - return to->sync_to(to, from, seqno - 1); -} - static int i915_gem_execbuffer_wait_for_flips(struct intel_ring_buffer *ring, u32 flips) { @@ -959,7 +900,7 @@ i915_gem_execbuffer_move_to_gpu(struct intel_ring_buffer *ring, } list_for_each_entry(obj, objects, exec_list) { - ret = i915_gem_execbuffer_sync_rings(obj, ring); + ret = i915_gem_object_sync(obj, ring); if (ret) return ret; } diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c index 4fb875d..25c8bf9 100644 --- a/drivers/gpu/drm/i915/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/i915_gem_gtt.c @@ -96,11 +96,10 @@ int i915_gem_init_aliasing_ppgtt(struct drm_device *dev) GFP_KERNEL); if (!ppgtt->pt_dma_addr) goto err_pt_alloc; - } - for (i = 0; i < ppgtt->num_pd_entries; i++) { - dma_addr_t pt_addr; - if (dev_priv->mm.gtt->needs_dmar) { + for (i = 0; i < ppgtt->num_pd_entries; i++) { + dma_addr_t pt_addr; + pt_addr = pci_map_page(dev->pdev, ppgtt->pt_pages[i], 0, 4096, PCI_DMA_BIDIRECTIONAL); @@ -112,8 +111,7 @@ int i915_gem_init_aliasing_ppgtt(struct drm_device *dev) } ppgtt->pt_dma_addr[i] = pt_addr; - } else - pt_addr = page_to_phys(ppgtt->pt_pages[i]); + } } ppgtt->scratch_page_dma_addr = dev_priv->mm.gtt->scratch_page_dma; diff --git a/drivers/gpu/drm/i915/i915_ioc32.c b/drivers/gpu/drm/i915/i915_ioc32.c index 13b0289..0e72abb 100644 --- a/drivers/gpu/drm/i915/i915_ioc32.c +++ b/drivers/gpu/drm/i915/i915_ioc32.c @@ -34,6 +34,7 @@ #include "drmP.h" #include "drm.h" #include "i915_drm.h" +#include "i915_drv.h" typedef struct _drm_i915_batchbuffer32 { int start; /* agp offset */ @@ -181,7 +182,7 @@ static int compat_i915_alloc(struct file *file, unsigned int cmd, (unsigned long)request); } -drm_ioctl_compat_t *i915_compat_ioctls[] = { +static drm_ioctl_compat_t *i915_compat_ioctls[] = { [DRM_I915_BATCHBUFFER] = compat_i915_batchbuffer, [DRM_I915_CMDBUFFER] = compat_i915_cmdbuffer, [DRM_I915_GETPARAM] = compat_i915_getparam, @@ -189,6 +190,7 @@ drm_ioctl_compat_t *i915_compat_ioctls[] = { [DRM_I915_ALLOC] = compat_i915_alloc }; +#ifdef CONFIG_COMPAT /** * Called whenever a 32-bit process running under a 64-bit kernel * performs an ioctl on /dev/dri/card<n>. @@ -217,3 +219,4 @@ long i915_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) return ret; } +#endif diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index febddc2..ab023ca 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -451,6 +451,31 @@ static void snb_gt_irq_handler(struct drm_device *dev, } } +static void gen6_queue_rps_work(struct drm_i915_private *dev_priv, + u32 pm_iir) +{ + unsigned long flags; + + /* + * IIR bits should never already be set because IMR should + * prevent an interrupt from being shown in IIR. The warning + * displays a case where we've unsafely cleared + * dev_priv->pm_iir. Although missing an interrupt of the same + * type is not a problem, it displays a problem in the logic. + * + * The mask bit in IMR is cleared by rps_work. + */ + + spin_lock_irqsave(&dev_priv->rps_lock, flags); + WARN(dev_priv->pm_iir & pm_iir, "Missed a PM interrupt\n"); + dev_priv->pm_iir |= pm_iir; + I915_WRITE(GEN6_PMIMR, dev_priv->pm_iir); + POSTING_READ(GEN6_PMIMR); + spin_unlock_irqrestore(&dev_priv->rps_lock, flags); + + queue_work(dev_priv->wq, &dev_priv->rps_work); +} + static irqreturn_t valleyview_irq_handler(DRM_IRQ_ARGS) { struct drm_device *dev = (struct drm_device *) arg; @@ -532,16 +557,8 @@ static irqreturn_t valleyview_irq_handler(DRM_IRQ_ARGS) if (pipe_stats[pipe] & PIPE_LEGACY_BLC_EVENT_STATUS) blc_event = true; - if (pm_iir & GEN6_PM_DEFERRED_EVENTS) { - unsigned long flags; - spin_lock_irqsave(&dev_priv->rps_lock, flags); - WARN(dev_priv->pm_iir & pm_iir, "Missed a PM interrupt\n"); - dev_priv->pm_iir |= pm_iir; - I915_WRITE(GEN6_PMIMR, dev_priv->pm_iir); - POSTING_READ(GEN6_PMIMR); - spin_unlock_irqrestore(&dev_priv->rps_lock, flags); - queue_work(dev_priv->wq, &dev_priv->rps_work); - } + if (pm_iir & GEN6_PM_DEFERRED_EVENTS) + gen6_queue_rps_work(dev_priv, pm_iir); I915_WRITE(GTIIR, gt_iir); I915_WRITE(GEN6_PMIIR, pm_iir); @@ -655,16 +672,8 @@ static irqreturn_t ivybridge_irq_handler(DRM_IRQ_ARGS) pch_irq_handler(dev); } - if (pm_iir & GEN6_PM_DEFERRED_EVENTS) { - unsigned long flags; - spin_lock_irqsave(&dev_priv->rps_lock, flags); - WARN(dev_priv->pm_iir & pm_iir, "Missed a PM interrupt\n"); - dev_priv->pm_iir |= pm_iir; - I915_WRITE(GEN6_PMIMR, dev_priv->pm_iir); - POSTING_READ(GEN6_PMIMR); - spin_unlock_irqrestore(&dev_priv->rps_lock, flags); - queue_work(dev_priv->wq, &dev_priv->rps_work); - } + if (pm_iir & GEN6_PM_DEFERRED_EVENTS) + gen6_queue_rps_work(dev_priv, pm_iir); /* should clear PCH hotplug event before clear CPU irq */ I915_WRITE(SDEIIR, pch_iir); @@ -764,25 +773,8 @@ static irqreturn_t ironlake_irq_handler(DRM_IRQ_ARGS) i915_handle_rps_change(dev); } - if (IS_GEN6(dev) && pm_iir & GEN6_PM_DEFERRED_EVENTS) { - /* - * IIR bits should never already be set because IMR should - * prevent an interrupt from being shown in IIR. The warning - * displays a case where we've unsafely cleared - * dev_priv->pm_iir. Although missing an interrupt of the same - * type is not a problem, it displays a problem in the logic. - * - * The mask bit in IMR is cleared by rps_work. - */ - unsigned long flags; - spin_lock_irqsave(&dev_priv->rps_lock, flags); - WARN(dev_priv->pm_iir & pm_iir, "Missed a PM interrupt\n"); - dev_priv->pm_iir |= pm_iir; - I915_WRITE(GEN6_PMIMR, dev_priv->pm_iir); - POSTING_READ(GEN6_PMIMR); - spin_unlock_irqrestore(&dev_priv->rps_lock, flags); - queue_work(dev_priv->wq, &dev_priv->rps_work); - } + if (IS_GEN6(dev) && pm_iir & GEN6_PM_DEFERRED_EVENTS) + gen6_queue_rps_work(dev_priv, pm_iir); /* should clear PCH hotplug event before clear CPU irq */ I915_WRITE(SDEIIR, pch_iir); @@ -1376,7 +1368,8 @@ static void i915_pageflip_stall_check(struct drm_device *dev, int pipe) obj = work->pending_flip_obj; if (INTEL_INFO(dev)->gen >= 4) { int dspsurf = DSPSURF(intel_crtc->plane); - stall_detected = I915_READ(dspsurf) == obj->gtt_offset; + stall_detected = I915_HI_DISPBASE(I915_READ(dspsurf)) == + obj->gtt_offset; } else { int dspaddr = DSPADDR(intel_crtc->plane); stall_detected = I915_READ(dspaddr) == (obj->gtt_offset + @@ -1875,6 +1868,36 @@ static bool kick_ring(struct intel_ring_buffer *ring) return false; } +static bool i915_hangcheck_hung(struct drm_device *dev) +{ + drm_i915_private_t *dev_priv = dev->dev_private; + + if (dev_priv->hangcheck_count++ > 1) { + DRM_ERROR("Hangcheck timer elapsed... GPU hung\n"); + i915_handle_error(dev, true); + + if (!IS_GEN2(dev)) { + /* Is the chip hanging on a WAIT_FOR_EVENT? + * If so we can simply poke the RB_WAIT bit + * and break the hang. This should work on + * all but the second generation chipsets. + */ + if (kick_ring(&dev_priv->ring[RCS])) + return false; + + if (HAS_BSD(dev) && kick_ring(&dev_priv->ring[VCS])) + return false; + + if (HAS_BLT(dev) && kick_ring(&dev_priv->ring[BCS])) + return false; + } + + return true; + } + + return false; +} + /** * This is called when the chip hasn't reported back with completed * batchbuffers in a long time. The first time this is called we simply record @@ -1895,9 +1918,14 @@ void i915_hangcheck_elapsed(unsigned long data) if (i915_hangcheck_ring_idle(&dev_priv->ring[RCS], &err) && i915_hangcheck_ring_idle(&dev_priv->ring[VCS], &err) && i915_hangcheck_ring_idle(&dev_priv->ring[BCS], &err)) { - dev_priv->hangcheck_count = 0; - if (err) + if (err) { + if (i915_hangcheck_hung(dev)) + return; + goto repeat; + } + + dev_priv->hangcheck_count = 0; return; } @@ -1919,30 +1947,8 @@ void i915_hangcheck_elapsed(unsigned long data) dev_priv->last_acthd_blt == acthd_blt && dev_priv->last_instdone == instdone && dev_priv->last_instdone1 == instdone1) { - if (dev_priv->hangcheck_count++ > 1) { - DRM_ERROR("Hangcheck timer elapsed... GPU hung\n"); - i915_handle_error(dev, true); - - if (!IS_GEN2(dev)) { - /* Is the chip hanging on a WAIT_FOR_EVENT? - * If so we can simply poke the RB_WAIT bit - * and break the hang. This should work on - * all but the second generation chipsets. - */ - if (kick_ring(&dev_priv->ring[RCS])) - goto repeat; - - if (HAS_BSD(dev) && - kick_ring(&dev_priv->ring[VCS])) - goto repeat; - - if (HAS_BLT(dev) && - kick_ring(&dev_priv->ring[BCS])) - goto repeat; - } - + if (i915_hangcheck_hung(dev)) return; - } } else { dev_priv->hangcheck_count = 0; diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index 6924f44..5ac9837 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -127,6 +127,13 @@ #define ECOCHK_PPGTT_CACHE64B (0x3<<3) #define ECOCHK_PPGTT_CACHE4B (0x0<<3) +#define GAC_ECO_BITS 0x14090 +#define ECOBITS_PPGTT_CACHE64B (3<<8) +#define ECOBITS_PPGTT_CACHE4B (0<<8) + +#define GAB_CTL 0x24000 +#define GAB_CTL_CONT_AFTER_PAGEFAULT (1<<8) + /* VGA stuff */ #define VGA_ST01_MDA 0x3ba @@ -224,6 +231,7 @@ #define MI_BATCH_NON_SECURE (1) #define MI_BATCH_NON_SECURE_I965 (1<<8) #define MI_BATCH_BUFFER_START MI_INSTR(0x31, 0) +#define MI_BATCH_GTT (2<<6) /* aliased with (1<<7) on gen4 */ #define MI_SEMAPHORE_MBOX MI_INSTR(0x16, 1) /* gen6+ */ #define MI_SEMAPHORE_GLOBAL_GTT (1<<22) #define MI_SEMAPHORE_UPDATE (1<<21) @@ -490,6 +498,7 @@ */ # define _3D_CHICKEN2_WM_READ_PIPELINED (1 << 14) #define _3D_CHICKEN3 0x02090 +#define _3D_CHICKEN_SF_DISABLE_FASTCLIP_CULL (1 << 5) #define MI_MODE 0x0209c # define VS_TIMER_DISPATCH (1 << 6) @@ -631,6 +640,7 @@ #define CM0_MASK_SHIFT 16 #define CM0_IZ_OPT_DISABLE (1<<6) #define CM0_ZR_OPT_DISABLE (1<<5) +#define CM0_STC_EVICT_DISABLE_LRA_SNB (1<<5) #define CM0_DEPTH_EVICT_DISABLE (1<<4) #define CM0_COLOR_EVICT_DISABLE (1<<3) #define CM0_DEPTH_WRITE_DISABLE (1<<1) @@ -682,6 +692,21 @@ #define GEN6_BSD_RNCID 0x12198 +#define GEN7_FF_THREAD_MODE 0x20a0 +#define GEN7_FF_SCHED_MASK 0x0077070 +#define GEN7_FF_TS_SCHED_HS1 (0x5<<16) +#define GEN7_FF_TS_SCHED_HS0 (0x3<<16) +#define GEN7_FF_TS_SCHED_LOAD_BALANCE (0x1<<16) +#define GEN7_FF_TS_SCHED_HW (0x0<<16) /* Default */ +#define GEN7_FF_VS_SCHED_HS1 (0x5<<12) +#define GEN7_FF_VS_SCHED_HS0 (0x3<<12) +#define GEN7_FF_VS_SCHED_LOAD_BALANCE (0x1<<12) /* Default */ +#define GEN7_FF_VS_SCHED_HW (0x0<<12) +#define GEN7_FF_DS_SCHED_HS1 (0x5<<4) +#define GEN7_FF_DS_SCHED_HS0 (0x3<<4) +#define GEN7_FF_DS_SCHED_LOAD_BALANCE (0x1<<4) /* Default */ +#define GEN7_FF_DS_SCHED_HW (0x0<<4) + /* * Framebuffer compression (915+ only) */ @@ -2860,6 +2885,13 @@ #define DSPSURF(plane) _PIPE(plane, _DSPASURF, _DSPBSURF) #define DSPTILEOFF(plane) _PIPE(plane, _DSPATILEOFF, _DSPBTILEOFF) +/* Display/Sprite base address macros */ +#define DISP_BASEADDR_MASK (0xfffff000) +#define I915_LO_DISPBASE(val) (val & ~DISP_BASEADDR_MASK) +#define I915_HI_DISPBASE(val) (val & DISP_BASEADDR_MASK) +#define I915_MODIFY_DISPBASE(reg, gfx_addr) \ + (I915_WRITE(reg, gfx_addr | I915_LO_DISPBASE(I915_READ(reg)))) + /* VBIOS flags */ #define SWF00 0x71410 #define SWF01 0x71414 @@ -3648,6 +3680,9 @@ #define FDI_LINK_TRAIN_PATTERN_IDLE_CPT (2<<8) #define FDI_LINK_TRAIN_NORMAL_CPT (3<<8) #define FDI_LINK_TRAIN_PATTERN_MASK_CPT (3<<8) +/* LPT */ +#define FDI_PORT_WIDTH_2X_LPT (1<<19) +#define FDI_PORT_WIDTH_1X_LPT (0<<19) #define _FDI_RXA_MISC 0xf0010 #define _FDI_RXB_MISC 0xf1010 @@ -3891,6 +3926,10 @@ #define GT_FIFO_FREE_ENTRIES 0x120008 #define GT_FIFO_NUM_RESERVED_ENTRIES 20 +#define GEN6_UCGCTL1 0x9400 +# define GEN6_BLBUNIT_CLOCK_GATE_DISABLE (1 << 5) +# define GEN6_CSUNIT_CLOCK_GATE_DISABLE (1 << 7) + #define GEN6_UCGCTL2 0x9404 # define GEN6_RCZUNIT_CLOCK_GATE_DISABLE (1 << 13) # define GEN6_RCPBUNIT_CLOCK_GATE_DISABLE (1 << 12) @@ -3970,6 +4009,11 @@ GEN6_PM_RP_DOWN_THRESHOLD | \ GEN6_PM_RP_DOWN_TIMEOUT) +#define GEN6_GT_GFX_RC6_LOCKED 0x138104 +#define GEN6_GT_GFX_RC6 0x138108 +#define GEN6_GT_GFX_RC6p 0x13810C +#define GEN6_GT_GFX_RC6pp 0x138110 + #define GEN6_PCODE_MAILBOX 0x138124 #define GEN6_PCODE_READY (1<<31) #define GEN6_READ_OC_PARAMS 0xc @@ -4170,6 +4214,10 @@ #define WRPLL_PLL_SELECT_SSC (0x01<<28) #define WRPLL_PLL_SELECT_NON_SCC (0x02<<28) #define WRPLL_PLL_SELECT_LCPLL_2700 (0x03<<28) +/* WRPLL divider programming */ +#define WRPLL_DIVIDER_REFERENCE(x) ((x)<<0) +#define WRPLL_DIVIDER_POST(x) ((x)<<8) +#define WRPLL_DIVIDER_FEEDBACK(x) ((x)<<16) /* Port clock selection */ #define PORT_CLK_SEL_A 0x46100 diff --git a/drivers/gpu/drm/i915/i915_suspend.c b/drivers/gpu/drm/i915/i915_suspend.c index 2b5eb22..0c3e3bf 100644 --- a/drivers/gpu/drm/i915/i915_suspend.c +++ b/drivers/gpu/drm/i915/i915_suspend.c @@ -879,17 +879,7 @@ int i915_restore_state(struct drm_device *dev) mutex_unlock(&dev->struct_mutex); if (drm_core_check_feature(dev, DRIVER_MODESET)) - intel_init_clock_gating(dev); - - if (IS_IRONLAKE_M(dev)) { - ironlake_enable_drps(dev); - intel_init_emon(dev); - } - - if (INTEL_INFO(dev)->gen >= 6) { - gen6_enable_rps(dev_priv); - gen6_update_ring_freq(dev_priv); - } + intel_modeset_init_hw(dev); mutex_lock(&dev->struct_mutex); diff --git a/drivers/gpu/drm/i915/i915_sysfs.c b/drivers/gpu/drm/i915/i915_sysfs.c new file mode 100644 index 0000000..79f8344 --- /dev/null +++ b/drivers/gpu/drm/i915/i915_sysfs.c @@ -0,0 +1,111 @@ +/* + * Copyright © 2012 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: + * Ben Widawsky <ben@bwidawsk.net> + * + */ + +#include <linux/device.h> +#include <linux/module.h> +#include <linux/stat.h> +#include <linux/sysfs.h> +#include "i915_drv.h" + +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 */ + + if (!intel_enable_rc6(dev)) + return 0; + + raw_time = I915_READ(reg) * 128ULL; + return DIV_ROUND_UP_ULL(raw_time, 100000); +} + +static ssize_t +show_rc6_mask(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct drm_minor *dminor = container_of(dev, struct drm_minor, kdev); + return snprintf(buf, PAGE_SIZE, "%x", intel_enable_rc6(dminor->dev)); +} + +static ssize_t +show_rc6_ms(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct drm_minor *dminor = container_of(dev, struct drm_minor, kdev); + u32 rc6_residency = calc_residency(dminor->dev, GEN6_GT_GFX_RC6); + return snprintf(buf, PAGE_SIZE, "%u", rc6_residency); +} + +static ssize_t +show_rc6p_ms(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct drm_minor *dminor = container_of(dev, struct drm_minor, kdev); + u32 rc6p_residency = calc_residency(dminor->dev, GEN6_GT_GFX_RC6p); + return snprintf(buf, PAGE_SIZE, "%u", rc6p_residency); +} + +static ssize_t +show_rc6pp_ms(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct drm_minor *dminor = container_of(dev, struct drm_minor, kdev); + u32 rc6pp_residency = calc_residency(dminor->dev, GEN6_GT_GFX_RC6pp); + return snprintf(buf, PAGE_SIZE, "%u", rc6pp_residency); +} + +static DEVICE_ATTR(rc6_enable, S_IRUGO, show_rc6_mask, NULL); +static DEVICE_ATTR(rc6_residency_ms, S_IRUGO, show_rc6_ms, NULL); +static DEVICE_ATTR(rc6p_residency_ms, S_IRUGO, show_rc6p_ms, NULL); +static DEVICE_ATTR(rc6pp_residency_ms, S_IRUGO, show_rc6pp_ms, NULL); + +static struct attribute *rc6_attrs[] = { + &dev_attr_rc6_enable.attr, + &dev_attr_rc6_residency_ms.attr, + &dev_attr_rc6p_residency_ms.attr, + &dev_attr_rc6pp_residency_ms.attr, + NULL +}; + +static struct attribute_group rc6_attr_group = { + .name = power_group_name, + .attrs = rc6_attrs +}; + +void i915_setup_sysfs(struct drm_device *dev) +{ + int ret; + + /* ILK doesn't have any residency information */ + if (INTEL_INFO(dev)->gen < 6) + return; + + ret = sysfs_merge_group(&dev->primary->kdev.kobj, &rc6_attr_group); + if (ret) + DRM_ERROR("sysfs setup failed\n"); +} + +void i915_teardown_sysfs(struct drm_device *dev) +{ + sysfs_unmerge_group(&dev->primary->kdev.kobj, &rc6_attr_group); +} diff --git a/drivers/gpu/drm/i915/i915_trace_points.c b/drivers/gpu/drm/i915/i915_trace_points.c index ead876e..f1df2bd 100644 --- a/drivers/gpu/drm/i915/i915_trace_points.c +++ b/drivers/gpu/drm/i915/i915_trace_points.c @@ -7,5 +7,7 @@ #include "i915_drv.h" +#ifndef __CHECKER__ #define CREATE_TRACE_POINTS #include "i915_trace.h" +#endif diff --git a/drivers/gpu/drm/i915/intel_acpi.c b/drivers/gpu/drm/i915/intel_acpi.c index f152b2a..f413899 100644 --- a/drivers/gpu/drm/i915/intel_acpi.c +++ b/drivers/gpu/drm/i915/intel_acpi.c @@ -9,6 +9,7 @@ #include <acpi/acpi_drivers.h> #include "drmP.h" +#include "i915_drv.h" #define INTEL_DSM_REVISION_ID 1 /* For Calpella anyway... */ diff --git a/drivers/gpu/drm/i915/intel_crt.c b/drivers/gpu/drm/i915/intel_crt.c index 70b0f1a..0976137 100644 --- a/drivers/gpu/drm/i915/intel_crt.c +++ b/drivers/gpu/drm/i915/intel_crt.c @@ -55,18 +55,36 @@ static struct intel_crt *intel_attached_crt(struct drm_connector *connector) struct intel_crt, base); } -static void intel_crt_dpms(struct drm_encoder *encoder, int mode) +static void pch_crt_dpms(struct drm_encoder *encoder, int mode) { struct drm_device *dev = encoder->dev; struct drm_i915_private *dev_priv = dev->dev_private; - u32 temp, reg; + u32 temp; - if (HAS_PCH_SPLIT(dev)) - reg = PCH_ADPA; - else - reg = ADPA; + temp = I915_READ(PCH_ADPA); + temp &= ~ADPA_DAC_ENABLE; + + switch (mode) { + case DRM_MODE_DPMS_ON: + temp |= ADPA_DAC_ENABLE; + break; + case DRM_MODE_DPMS_STANDBY: + case DRM_MODE_DPMS_SUSPEND: + case DRM_MODE_DPMS_OFF: + /* Just leave port enable cleared */ + break; + } + + I915_WRITE(PCH_ADPA, temp); +} - temp = I915_READ(reg); +static void gmch_crt_dpms(struct drm_encoder *encoder, int mode) +{ + struct drm_device *dev = encoder->dev; + struct drm_i915_private *dev_priv = dev->dev_private; + u32 temp; + + temp = I915_READ(ADPA); temp &= ~(ADPA_HSYNC_CNTL_DISABLE | ADPA_VSYNC_CNTL_DISABLE); temp &= ~ADPA_DAC_ENABLE; @@ -85,7 +103,7 @@ static void intel_crt_dpms(struct drm_encoder *encoder, int mode) break; } - I915_WRITE(reg, temp); + I915_WRITE(ADPA, temp); } static int intel_crt_mode_valid(struct drm_connector *connector, @@ -516,12 +534,20 @@ static void intel_crt_reset(struct drm_connector *connector) * Routines for controlling stuff on the analog port */ -static const struct drm_encoder_helper_funcs intel_crt_helper_funcs = { - .dpms = intel_crt_dpms, +static const struct drm_encoder_helper_funcs pch_encoder_funcs = { + .mode_fixup = intel_crt_mode_fixup, + .prepare = intel_encoder_prepare, + .commit = intel_encoder_commit, + .mode_set = intel_crt_mode_set, + .dpms = pch_crt_dpms, +}; + +static const struct drm_encoder_helper_funcs gmch_encoder_funcs = { .mode_fixup = intel_crt_mode_fixup, .prepare = intel_encoder_prepare, .commit = intel_encoder_commit, .mode_set = intel_crt_mode_set, + .dpms = gmch_crt_dpms, }; static const struct drm_connector_funcs intel_crt_connector_funcs = { @@ -567,6 +593,7 @@ void intel_crt_init(struct drm_device *dev) struct intel_crt *crt; struct intel_connector *intel_connector; struct drm_i915_private *dev_priv = dev->dev_private; + const struct drm_encoder_helper_funcs *encoder_helper_funcs; /* Skip machines without VGA that falsely report hotplug events */ if (dmi_check_system(intel_no_crt)) @@ -602,7 +629,12 @@ void intel_crt_init(struct drm_device *dev) connector->interlace_allowed = 1; connector->doublescan_allowed = 0; - drm_encoder_helper_add(&crt->base.base, &intel_crt_helper_funcs); + if (HAS_PCH_SPLIT(dev)) + encoder_helper_funcs = &pch_encoder_funcs; + else + encoder_helper_funcs = &gmch_encoder_funcs; + + drm_encoder_helper_add(&crt->base.base, encoder_helper_funcs); drm_connector_helper_add(connector, &intel_crt_connector_helper_funcs); drm_sysfs_connector_add(connector); diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 37514a5..4c844c6 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -25,7 +25,6 @@ */ #include <linux/dmi.h> -#include <linux/cpufreq.h> #include <linux/module.h> #include <linux/input.h> #include <linux/i2c.h> @@ -45,7 +44,6 @@ #define HAS_eDP (intel_pipe_has_type(crtc, INTEL_OUTPUT_EDP)) bool intel_pipe_has_type(struct drm_crtc *crtc, int type); -static void intel_update_watermarks(struct drm_device *dev); static void intel_increase_pllclock(struct drm_crtc *crtc); static void intel_crtc_update_cursor(struct drm_crtc *crtc, bool on); @@ -1517,7 +1515,7 @@ static void intel_disable_pipe(struct drm_i915_private *dev_priv, * Plane regs are double buffered, going from enabled->disabled needs a * trigger in order to latch. The display address reg provides this. */ -static void intel_flush_display_plane(struct drm_i915_private *dev_priv, +void intel_flush_display_plane(struct drm_i915_private *dev_priv, enum plane plane) { I915_WRITE(DSPADDR(plane), I915_READ(DSPADDR(plane))); @@ -1628,490 +1626,6 @@ static void intel_disable_pch_ports(struct drm_i915_private *dev_priv, disable_pch_hdmi(dev_priv, pipe, HDMID); } -static void i8xx_disable_fbc(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - u32 fbc_ctl; - - /* Disable compression */ - fbc_ctl = I915_READ(FBC_CONTROL); - if ((fbc_ctl & FBC_CTL_EN) == 0) - return; - - fbc_ctl &= ~FBC_CTL_EN; - I915_WRITE(FBC_CONTROL, fbc_ctl); - - /* Wait for compressing bit to clear */ - if (wait_for((I915_READ(FBC_STATUS) & FBC_STAT_COMPRESSING) == 0, 10)) { - DRM_DEBUG_KMS("FBC idle timed out\n"); - return; - } - - DRM_DEBUG_KMS("disabled FBC\n"); -} - -static void i8xx_enable_fbc(struct drm_crtc *crtc, unsigned long interval) -{ - struct drm_device *dev = crtc->dev; - struct drm_i915_private *dev_priv = dev->dev_private; - struct drm_framebuffer *fb = crtc->fb; - struct intel_framebuffer *intel_fb = to_intel_framebuffer(fb); - struct drm_i915_gem_object *obj = intel_fb->obj; - struct intel_crtc *intel_crtc = to_intel_crtc(crtc); - int cfb_pitch; - int plane, i; - u32 fbc_ctl, fbc_ctl2; - - cfb_pitch = dev_priv->cfb_size / FBC_LL_SIZE; - if (fb->pitches[0] < cfb_pitch) - cfb_pitch = fb->pitches[0]; - - /* FBC_CTL wants 64B units */ - cfb_pitch = (cfb_pitch / 64) - 1; - plane = intel_crtc->plane == 0 ? FBC_CTL_PLANEA : FBC_CTL_PLANEB; - - /* Clear old tags */ - for (i = 0; i < (FBC_LL_SIZE / 32) + 1; i++) - I915_WRITE(FBC_TAG + (i * 4), 0); - - /* Set it up... */ - fbc_ctl2 = FBC_CTL_FENCE_DBL | FBC_CTL_IDLE_IMM | FBC_CTL_CPU_FENCE; - fbc_ctl2 |= plane; - I915_WRITE(FBC_CONTROL2, fbc_ctl2); - I915_WRITE(FBC_FENCE_OFF, crtc->y); - - /* enable it... */ - fbc_ctl = FBC_CTL_EN | FBC_CTL_PERIODIC; - if (IS_I945GM(dev)) - fbc_ctl |= FBC_CTL_C3_IDLE; /* 945 needs special SR handling */ - fbc_ctl |= (cfb_pitch & 0xff) << FBC_CTL_STRIDE_SHIFT; - fbc_ctl |= (interval & 0x2fff) << FBC_CTL_INTERVAL_SHIFT; - fbc_ctl |= obj->fence_reg; - I915_WRITE(FBC_CONTROL, fbc_ctl); - - DRM_DEBUG_KMS("enabled FBC, pitch %d, yoff %d, plane %d, ", - cfb_pitch, crtc->y, intel_crtc->plane); -} - -static bool i8xx_fbc_enabled(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - - return I915_READ(FBC_CONTROL) & FBC_CTL_EN; -} - -static void g4x_enable_fbc(struct drm_crtc *crtc, unsigned long interval) -{ - struct drm_device *dev = crtc->dev; - struct drm_i915_private *dev_priv = dev->dev_private; - struct drm_framebuffer *fb = crtc->fb; - struct intel_framebuffer *intel_fb = to_intel_framebuffer(fb); - struct drm_i915_gem_object *obj = intel_fb->obj; - struct intel_crtc *intel_crtc = to_intel_crtc(crtc); - int plane = intel_crtc->plane == 0 ? DPFC_CTL_PLANEA : DPFC_CTL_PLANEB; - unsigned long stall_watermark = 200; - u32 dpfc_ctl; - - dpfc_ctl = plane | DPFC_SR_EN | DPFC_CTL_LIMIT_1X; - dpfc_ctl |= DPFC_CTL_FENCE_EN | obj->fence_reg; - I915_WRITE(DPFC_CHICKEN, DPFC_HT_MODIFY); - - I915_WRITE(DPFC_RECOMP_CTL, DPFC_RECOMP_STALL_EN | - (stall_watermark << DPFC_RECOMP_STALL_WM_SHIFT) | - (interval << DPFC_RECOMP_TIMER_COUNT_SHIFT)); - I915_WRITE(DPFC_FENCE_YOFF, crtc->y); - - /* enable it... */ - I915_WRITE(DPFC_CONTROL, I915_READ(DPFC_CONTROL) | DPFC_CTL_EN); - - DRM_DEBUG_KMS("enabled fbc on plane %d\n", intel_crtc->plane); -} - -static void g4x_disable_fbc(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - u32 dpfc_ctl; - - /* Disable compression */ - dpfc_ctl = I915_READ(DPFC_CONTROL); - if (dpfc_ctl & DPFC_CTL_EN) { - dpfc_ctl &= ~DPFC_CTL_EN; - I915_WRITE(DPFC_CONTROL, dpfc_ctl); - - DRM_DEBUG_KMS("disabled FBC\n"); - } -} - -static bool g4x_fbc_enabled(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - - return I915_READ(DPFC_CONTROL) & DPFC_CTL_EN; -} - -static void sandybridge_blit_fbc_update(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - u32 blt_ecoskpd; - - /* Make sure blitter notifies FBC of writes */ - gen6_gt_force_wake_get(dev_priv); - blt_ecoskpd = I915_READ(GEN6_BLITTER_ECOSKPD); - blt_ecoskpd |= GEN6_BLITTER_FBC_NOTIFY << - GEN6_BLITTER_LOCK_SHIFT; - I915_WRITE(GEN6_BLITTER_ECOSKPD, blt_ecoskpd); - blt_ecoskpd |= GEN6_BLITTER_FBC_NOTIFY; - I915_WRITE(GEN6_BLITTER_ECOSKPD, blt_ecoskpd); - blt_ecoskpd &= ~(GEN6_BLITTER_FBC_NOTIFY << - GEN6_BLITTER_LOCK_SHIFT); - I915_WRITE(GEN6_BLITTER_ECOSKPD, blt_ecoskpd); - POSTING_READ(GEN6_BLITTER_ECOSKPD); - gen6_gt_force_wake_put(dev_priv); -} - -static void ironlake_enable_fbc(struct drm_crtc *crtc, unsigned long interval) -{ - struct drm_device *dev = crtc->dev; - struct drm_i915_private *dev_priv = dev->dev_private; - struct drm_framebuffer *fb = crtc->fb; - struct intel_framebuffer *intel_fb = to_intel_framebuffer(fb); - struct drm_i915_gem_object *obj = intel_fb->obj; - struct intel_crtc *intel_crtc = to_intel_crtc(crtc); - int plane = intel_crtc->plane == 0 ? DPFC_CTL_PLANEA : DPFC_CTL_PLANEB; - unsigned long stall_watermark = 200; - u32 dpfc_ctl; - - dpfc_ctl = I915_READ(ILK_DPFC_CONTROL); - dpfc_ctl &= DPFC_RESERVED; - dpfc_ctl |= (plane | DPFC_CTL_LIMIT_1X); - /* Set persistent mode for front-buffer rendering, ala X. */ - dpfc_ctl |= DPFC_CTL_PERSISTENT_MODE; - dpfc_ctl |= (DPFC_CTL_FENCE_EN | obj->fence_reg); - I915_WRITE(ILK_DPFC_CHICKEN, DPFC_HT_MODIFY); - - I915_WRITE(ILK_DPFC_RECOMP_CTL, DPFC_RECOMP_STALL_EN | - (stall_watermark << DPFC_RECOMP_STALL_WM_SHIFT) | - (interval << DPFC_RECOMP_TIMER_COUNT_SHIFT)); - I915_WRITE(ILK_DPFC_FENCE_YOFF, crtc->y); - I915_WRITE(ILK_FBC_RT_BASE, obj->gtt_offset | ILK_FBC_RT_VALID); - /* enable it... */ - I915_WRITE(ILK_DPFC_CONTROL, dpfc_ctl | DPFC_CTL_EN); - - if (IS_GEN6(dev)) { - I915_WRITE(SNB_DPFC_CTL_SA, - SNB_CPU_FENCE_ENABLE | obj->fence_reg); - I915_WRITE(DPFC_CPU_FENCE_OFFSET, crtc->y); - sandybridge_blit_fbc_update(dev); - } - - DRM_DEBUG_KMS("enabled fbc on plane %d\n", intel_crtc->plane); -} - -static void ironlake_disable_fbc(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - u32 dpfc_ctl; - - /* Disable compression */ - dpfc_ctl = I915_READ(ILK_DPFC_CONTROL); - if (dpfc_ctl & DPFC_CTL_EN) { - dpfc_ctl &= ~DPFC_CTL_EN; - I915_WRITE(ILK_DPFC_CONTROL, dpfc_ctl); - - DRM_DEBUG_KMS("disabled FBC\n"); - } -} - -static bool ironlake_fbc_enabled(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - - return I915_READ(ILK_DPFC_CONTROL) & DPFC_CTL_EN; -} - -bool intel_fbc_enabled(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - - if (!dev_priv->display.fbc_enabled) - return false; - - return dev_priv->display.fbc_enabled(dev); -} - -static void intel_fbc_work_fn(struct work_struct *__work) -{ - struct intel_fbc_work *work = - container_of(to_delayed_work(__work), - struct intel_fbc_work, work); - struct drm_device *dev = work->crtc->dev; - struct drm_i915_private *dev_priv = dev->dev_private; - - mutex_lock(&dev->struct_mutex); - if (work == dev_priv->fbc_work) { - /* Double check that we haven't switched fb without cancelling - * the prior work. - */ - if (work->crtc->fb == work->fb) { - dev_priv->display.enable_fbc(work->crtc, - work->interval); - - dev_priv->cfb_plane = to_intel_crtc(work->crtc)->plane; - dev_priv->cfb_fb = work->crtc->fb->base.id; - dev_priv->cfb_y = work->crtc->y; - } - - dev_priv->fbc_work = NULL; - } - mutex_unlock(&dev->struct_mutex); - - kfree(work); -} - -static void intel_cancel_fbc_work(struct drm_i915_private *dev_priv) -{ - if (dev_priv->fbc_work == NULL) - return; - - DRM_DEBUG_KMS("cancelling pending FBC enable\n"); - - /* Synchronisation is provided by struct_mutex and checking of - * dev_priv->fbc_work, so we can perform the cancellation - * entirely asynchronously. - */ - if (cancel_delayed_work(&dev_priv->fbc_work->work)) - /* tasklet was killed before being run, clean up */ - kfree(dev_priv->fbc_work); - - /* Mark the work as no longer wanted so that if it does - * wake-up (because the work was already running and waiting - * for our mutex), it will discover that is no longer - * necessary to run. - */ - dev_priv->fbc_work = NULL; -} - -static void intel_enable_fbc(struct drm_crtc *crtc, unsigned long interval) -{ - struct intel_fbc_work *work; - struct drm_device *dev = crtc->dev; - struct drm_i915_private *dev_priv = dev->dev_private; - - if (!dev_priv->display.enable_fbc) - return; - - intel_cancel_fbc_work(dev_priv); - - work = kzalloc(sizeof *work, GFP_KERNEL); - if (work == NULL) { - dev_priv->display.enable_fbc(crtc, interval); - return; - } - - work->crtc = crtc; - work->fb = crtc->fb; - work->interval = interval; - INIT_DELAYED_WORK(&work->work, intel_fbc_work_fn); - - dev_priv->fbc_work = work; - - DRM_DEBUG_KMS("scheduling delayed FBC enable\n"); - - /* Delay the actual enabling to let pageflipping cease and the - * display to settle before starting the compression. Note that - * this delay also serves a second purpose: it allows for a - * vblank to pass after disabling the FBC before we attempt - * to modify the control registers. - * - * A more complicated solution would involve tracking vblanks - * following the termination of the page-flipping sequence - * and indeed performing the enable as a co-routine and not - * waiting synchronously upon the vblank. - */ - schedule_delayed_work(&work->work, msecs_to_jiffies(50)); -} - -void intel_disable_fbc(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - - intel_cancel_fbc_work(dev_priv); - - if (!dev_priv->display.disable_fbc) - return; - - dev_priv->display.disable_fbc(dev); - dev_priv->cfb_plane = -1; -} - -/** - * intel_update_fbc - enable/disable FBC as needed - * @dev: the drm_device - * - * Set up the framebuffer compression hardware at mode set time. We - * enable it if possible: - * - plane A only (on pre-965) - * - no pixel mulitply/line duplication - * - no alpha buffer discard - * - no dual wide - * - framebuffer <= 2048 in width, 1536 in height - * - * We can't assume that any compression will take place (worst case), - * so the compressed buffer has to be the same size as the uncompressed - * one. It also must reside (along with the line length buffer) in - * stolen memory. - * - * We need to enable/disable FBC on a global basis. - */ -static void intel_update_fbc(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - struct drm_crtc *crtc = NULL, *tmp_crtc; - struct intel_crtc *intel_crtc; - struct drm_framebuffer *fb; - struct intel_framebuffer *intel_fb; - struct drm_i915_gem_object *obj; - int enable_fbc; - - DRM_DEBUG_KMS("\n"); - - if (!i915_powersave) - return; - - if (!I915_HAS_FBC(dev)) - return; - - /* - * If FBC is already on, we just have to verify that we can - * keep it that way... - * Need to disable if: - * - more than one pipe is active - * - changing FBC params (stride, fence, mode) - * - new fb is too large to fit in compressed buffer - * - going to an unsupported config (interlace, pixel multiply, etc.) - */ - list_for_each_entry(tmp_crtc, &dev->mode_config.crtc_list, head) { - if (tmp_crtc->enabled && tmp_crtc->fb) { - if (crtc) { - DRM_DEBUG_KMS("more than one pipe active, disabling compression\n"); - dev_priv->no_fbc_reason = FBC_MULTIPLE_PIPES; - goto out_disable; - } - crtc = tmp_crtc; - } - } - - if (!crtc || crtc->fb == NULL) { - DRM_DEBUG_KMS("no output, disabling\n"); - dev_priv->no_fbc_reason = FBC_NO_OUTPUT; - goto out_disable; - } - - intel_crtc = to_intel_crtc(crtc); - fb = crtc->fb; - intel_fb = to_intel_framebuffer(fb); - obj = intel_fb->obj; - - enable_fbc = i915_enable_fbc; - if (enable_fbc < 0) { - DRM_DEBUG_KMS("fbc set to per-chip default\n"); - enable_fbc = 1; - if (INTEL_INFO(dev)->gen <= 6) - enable_fbc = 0; - } - if (!enable_fbc) { - DRM_DEBUG_KMS("fbc disabled per module param\n"); - dev_priv->no_fbc_reason = FBC_MODULE_PARAM; - goto out_disable; - } - if (intel_fb->obj->base.size > dev_priv->cfb_size) { - DRM_DEBUG_KMS("framebuffer too large, disabling " - "compression\n"); - dev_priv->no_fbc_reason = FBC_STOLEN_TOO_SMALL; - goto out_disable; - } - if ((crtc->mode.flags & DRM_MODE_FLAG_INTERLACE) || - (crtc->mode.flags & DRM_MODE_FLAG_DBLSCAN)) { - DRM_DEBUG_KMS("mode incompatible with compression, " - "disabling\n"); - dev_priv->no_fbc_reason = FBC_UNSUPPORTED_MODE; - goto out_disable; - } - if ((crtc->mode.hdisplay > 2048) || - (crtc->mode.vdisplay > 1536)) { - DRM_DEBUG_KMS("mode too large for compression, disabling\n"); - dev_priv->no_fbc_reason = FBC_MODE_TOO_LARGE; - goto out_disable; - } - if ((IS_I915GM(dev) || IS_I945GM(dev)) && intel_crtc->plane != 0) { - DRM_DEBUG_KMS("plane not 0, disabling compression\n"); - dev_priv->no_fbc_reason = FBC_BAD_PLANE; - goto out_disable; - } - - /* The use of a CPU fence is mandatory in order to detect writes - * by the CPU to the scanout and trigger updates to the FBC. - */ - if (obj->tiling_mode != I915_TILING_X || - obj->fence_reg == I915_FENCE_REG_NONE) { - DRM_DEBUG_KMS("framebuffer not tiled or fenced, disabling compression\n"); - dev_priv->no_fbc_reason = FBC_NOT_TILED; - goto out_disable; - } - - /* If the kernel debugger is active, always disable compression */ - if (in_dbg_master()) - goto out_disable; - - /* If the scanout has not changed, don't modify the FBC settings. - * Note that we make the fundamental assumption that the fb->obj - * cannot be unpinned (and have its GTT offset and fence revoked) - * without first being decoupled from the scanout and FBC disabled. - */ - if (dev_priv->cfb_plane == intel_crtc->plane && - dev_priv->cfb_fb == fb->base.id && - dev_priv->cfb_y == crtc->y) - return; - - if (intel_fbc_enabled(dev)) { - /* We update FBC along two paths, after changing fb/crtc - * configuration (modeswitching) and after page-flipping - * finishes. For the latter, we know that not only did - * we disable the FBC at the start of the page-flip - * sequence, but also more than one vblank has passed. - * - * For the former case of modeswitching, it is possible - * to switch between two FBC valid configurations - * instantaneously so we do need to disable the FBC - * before we can modify its control registers. We also - * have to wait for the next vblank for that to take - * effect. However, since we delay enabling FBC we can - * assume that a vblank has passed since disabling and - * that we can safely alter the registers in the deferred - * callback. - * - * In the scenario that we go from a valid to invalid - * and then back to valid FBC configuration we have - * no strict enforcement that a vblank occurred since - * disabling the FBC. However, along all current pipe - * disabling paths we do need to wait for a vblank at - * some point. And we wait before enabling FBC anyway. - */ - DRM_DEBUG_KMS("disabling active FBC for update\n"); - intel_disable_fbc(dev); - } - - intel_enable_fbc(crtc, 500); - return; - -out_disable: - /* Multiple disables should be harmless */ - if (intel_fbc_enabled(dev)) { - DRM_DEBUG_KMS("unsupported config, disabling FBC\n"); - intel_disable_fbc(dev); - } -} - int intel_pin_and_fence_fb_obj(struct drm_device *dev, struct drm_i915_gem_object *obj, @@ -2152,13 +1666,11 @@ intel_pin_and_fence_fb_obj(struct drm_device *dev, * framebuffer compression. For simplicity, we always install * a fence as the cost is not that onerous. */ - if (obj->tiling_mode != I915_TILING_NONE) { - ret = i915_gem_object_get_fence(obj, pipelined); - if (ret) - goto err_unpin; + ret = i915_gem_object_get_fence(obj); + if (ret) + goto err_unpin; - i915_gem_object_pin_fence(obj); - } + i915_gem_object_pin_fence(obj); dev_priv->mm.interruptible = true; return 0; @@ -2239,7 +1751,7 @@ static int i9xx_update_plane(struct drm_crtc *crtc, struct drm_framebuffer *fb, Start, Offset, x, y, fb->pitches[0]); I915_WRITE(DSPSTRIDE(plane), fb->pitches[0]); if (INTEL_INFO(dev)->gen >= 4) { - I915_WRITE(DSPSURF(plane), Start); + I915_MODIFY_DISPBASE(DSPSURF(plane), Start); I915_WRITE(DSPTILEOFF(plane), (y << 16) | x); I915_WRITE(DSPADDR(plane), Offset); } else @@ -2319,7 +1831,7 @@ static int ironlake_update_plane(struct drm_crtc *crtc, DRM_DEBUG_KMS("Writing base %08lX %08lX %d %d %d\n", Start, Offset, x, y, fb->pitches[0]); I915_WRITE(DSPSTRIDE(plane), fb->pitches[0]); - I915_WRITE(DSPSURF(plane), Start); + I915_MODIFY_DISPBASE(DSPSURF(plane), Start); I915_WRITE(DSPTILEOFF(plane), (y << 16) | x); I915_WRITE(DSPADDR(plane), Offset); POSTING_READ(reg); @@ -2334,16 +1846,39 @@ intel_pipe_set_base_atomic(struct drm_crtc *crtc, struct drm_framebuffer *fb, { struct drm_device *dev = crtc->dev; struct drm_i915_private *dev_priv = dev->dev_private; + + if (dev_priv->display.disable_fbc) + dev_priv->display.disable_fbc(dev); + intel_increase_pllclock(crtc); + + return dev_priv->display.update_plane(crtc, fb, x, y); +} + +static int +intel_finish_fb(struct drm_framebuffer *old_fb) +{ + struct drm_i915_gem_object *obj = to_intel_framebuffer(old_fb)->obj; + struct drm_i915_private *dev_priv = obj->base.dev->dev_private; + bool was_interruptible = dev_priv->mm.interruptible; int ret; - ret = dev_priv->display.update_plane(crtc, fb, x, y); - if (ret) - return ret; + wait_event(dev_priv->pending_flip_queue, + atomic_read(&dev_priv->mm.wedged) || + atomic_read(&obj->pending_flip) == 0); - intel_update_fbc(dev); - intel_increase_pllclock(crtc); + /* Big Hammer, we also need to ensure that any pending + * MI_WAIT_FOR_EVENT inside a user batch buffer on the + * current scanout is retired before unpinning the old + * framebuffer. + * + * This should only fail upon a hung GPU, in which case we + * can safely continue. + */ + dev_priv->mm.interruptible = false; + ret = i915_gem_object_finish_gpu(obj); + dev_priv->mm.interruptible = was_interruptible; - return 0; + return ret; } static int @@ -2351,6 +1886,7 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y, struct drm_framebuffer *old_fb) { struct drm_device *dev = crtc->dev; + struct drm_i915_private *dev_priv = dev->dev_private; struct drm_i915_master_private *master_priv; struct intel_crtc *intel_crtc = to_intel_crtc(crtc); int ret; @@ -2384,28 +1920,10 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y, return ret; } - if (old_fb) { - struct drm_i915_private *dev_priv = dev->dev_private; - struct drm_i915_gem_object *obj = to_intel_framebuffer(old_fb)->obj; - - wait_event(dev_priv->pending_flip_queue, - atomic_read(&dev_priv->mm.wedged) || - atomic_read(&obj->pending_flip) == 0); - - /* Big Hammer, we also need to ensure that any pending - * MI_WAIT_FOR_EVENT inside a user batch buffer on the - * current scanout is retired before unpinning the old - * framebuffer. - * - * This should only fail upon a hung GPU, in which case we - * can safely continue. - */ - ret = i915_gem_object_finish_gpu(obj); - (void) ret; - } + if (old_fb) + intel_finish_fb(old_fb); - ret = intel_pipe_set_base_atomic(crtc, crtc->fb, x, y, - LEAVE_ATOMIC_MODE_SET); + ret = dev_priv->display.update_plane(crtc, crtc->fb, x, y); if (ret) { intel_unpin_fb_obj(to_intel_framebuffer(crtc->fb)->obj); mutex_unlock(&dev->struct_mutex); @@ -2418,6 +1936,7 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y, intel_unpin_fb_obj(to_intel_framebuffer(old_fb)->obj); } + intel_update_fbc(dev); mutex_unlock(&dev->struct_mutex); if (!dev->primary->master) @@ -3010,16 +2529,14 @@ static void intel_clear_scanline_wait(struct drm_device *dev) static void intel_crtc_wait_for_pending_flips(struct drm_crtc *crtc) { - struct drm_i915_gem_object *obj; - struct drm_i915_private *dev_priv; + struct drm_device *dev = crtc->dev; if (crtc->fb == NULL) return; - obj = to_intel_framebuffer(crtc->fb)->obj; - dev_priv = crtc->dev->dev_private; - wait_event(dev_priv->pending_flip_queue, - atomic_read(&obj->pending_flip) == 0); + mutex_lock(&dev->struct_mutex); + intel_finish_fb(crtc->fb); + mutex_unlock(&dev->struct_mutex); } static bool intel_crtc_driving_pch(struct drm_crtc *crtc) @@ -3669,1482 +3186,6 @@ ironlake_compute_m_n(int bits_per_pixel, int nlanes, int pixel_clock, fdi_reduce_ratio(&m_n->link_m, &m_n->link_n); } - -struct intel_watermark_params { - unsigned long fifo_size; - unsigned long max_wm; - unsigned long default_wm; - unsigned long guard_size; - unsigned long cacheline_size; -}; - -/* Pineview has different values for various configs */ -static const struct intel_watermark_params pineview_display_wm = { - PINEVIEW_DISPLAY_FIFO, - PINEVIEW_MAX_WM, - PINEVIEW_DFT_WM, - PINEVIEW_GUARD_WM, - PINEVIEW_FIFO_LINE_SIZE -}; -static const struct intel_watermark_params pineview_display_hplloff_wm = { - PINEVIEW_DISPLAY_FIFO, - PINEVIEW_MAX_WM, - PINEVIEW_DFT_HPLLOFF_WM, - PINEVIEW_GUARD_WM, - PINEVIEW_FIFO_LINE_SIZE -}; -static const struct intel_watermark_params pineview_cursor_wm = { - PINEVIEW_CURSOR_FIFO, - PINEVIEW_CURSOR_MAX_WM, - PINEVIEW_CURSOR_DFT_WM, - PINEVIEW_CURSOR_GUARD_WM, - PINEVIEW_FIFO_LINE_SIZE, -}; -static const struct intel_watermark_params pineview_cursor_hplloff_wm = { - PINEVIEW_CURSOR_FIFO, - PINEVIEW_CURSOR_MAX_WM, - PINEVIEW_CURSOR_DFT_WM, - PINEVIEW_CURSOR_GUARD_WM, - PINEVIEW_FIFO_LINE_SIZE -}; -static const struct intel_watermark_params g4x_wm_info = { - G4X_FIFO_SIZE, - G4X_MAX_WM, - G4X_MAX_WM, - 2, - G4X_FIFO_LINE_SIZE, -}; -static const struct intel_watermark_params g4x_cursor_wm_info = { - I965_CURSOR_FIFO, - I965_CURSOR_MAX_WM, - I965_CURSOR_DFT_WM, - 2, - G4X_FIFO_LINE_SIZE, -}; -static const struct intel_watermark_params valleyview_wm_info = { - VALLEYVIEW_FIFO_SIZE, - VALLEYVIEW_MAX_WM, - VALLEYVIEW_MAX_WM, - 2, - G4X_FIFO_LINE_SIZE, -}; -static const struct intel_watermark_params valleyview_cursor_wm_info = { - I965_CURSOR_FIFO, - VALLEYVIEW_CURSOR_MAX_WM, - I965_CURSOR_DFT_WM, - 2, - G4X_FIFO_LINE_SIZE, -}; -static const struct intel_watermark_params i965_cursor_wm_info = { - I965_CURSOR_FIFO, - I965_CURSOR_MAX_WM, - I965_CURSOR_DFT_WM, - 2, - I915_FIFO_LINE_SIZE, -}; -static const struct intel_watermark_params i945_wm_info = { - I945_FIFO_SIZE, - I915_MAX_WM, - 1, - 2, - I915_FIFO_LINE_SIZE -}; -static const struct intel_watermark_params i915_wm_info = { - I915_FIFO_SIZE, - I915_MAX_WM, - 1, - 2, - I915_FIFO_LINE_SIZE -}; -static const struct intel_watermark_params i855_wm_info = { - I855GM_FIFO_SIZE, - I915_MAX_WM, - 1, - 2, - I830_FIFO_LINE_SIZE -}; -static const struct intel_watermark_params i830_wm_info = { - I830_FIFO_SIZE, - I915_MAX_WM, - 1, - 2, - I830_FIFO_LINE_SIZE -}; - -static const struct intel_watermark_params ironlake_display_wm_info = { - ILK_DISPLAY_FIFO, - ILK_DISPLAY_MAXWM, - ILK_DISPLAY_DFTWM, - 2, - ILK_FIFO_LINE_SIZE -}; -static const struct intel_watermark_params ironlake_cursor_wm_info = { - ILK_CURSOR_FIFO, - ILK_CURSOR_MAXWM, - ILK_CURSOR_DFTWM, - 2, - ILK_FIFO_LINE_SIZE -}; -static const struct intel_watermark_params ironlake_display_srwm_info = { - ILK_DISPLAY_SR_FIFO, - ILK_DISPLAY_MAX_SRWM, - ILK_DISPLAY_DFT_SRWM, - 2, - ILK_FIFO_LINE_SIZE -}; -static const struct intel_watermark_params ironlake_cursor_srwm_info = { - ILK_CURSOR_SR_FIFO, - ILK_CURSOR_MAX_SRWM, - ILK_CURSOR_DFT_SRWM, - 2, - ILK_FIFO_LINE_SIZE -}; - -static const struct intel_watermark_params sandybridge_display_wm_info = { - SNB_DISPLAY_FIFO, - SNB_DISPLAY_MAXWM, - SNB_DISPLAY_DFTWM, - 2, - SNB_FIFO_LINE_SIZE -}; -static const struct intel_watermark_params sandybridge_cursor_wm_info = { - SNB_CURSOR_FIFO, - SNB_CURSOR_MAXWM, - SNB_CURSOR_DFTWM, - 2, - SNB_FIFO_LINE_SIZE -}; -static const struct intel_watermark_params sandybridge_display_srwm_info = { - SNB_DISPLAY_SR_FIFO, - SNB_DISPLAY_MAX_SRWM, - SNB_DISPLAY_DFT_SRWM, - 2, - SNB_FIFO_LINE_SIZE -}; -static const struct intel_watermark_params sandybridge_cursor_srwm_info = { - SNB_CURSOR_SR_FIFO, - SNB_CURSOR_MAX_SRWM, - SNB_CURSOR_DFT_SRWM, - 2, - SNB_FIFO_LINE_SIZE -}; - - -/** - * intel_calculate_wm - calculate watermark level - * @clock_in_khz: pixel clock - * @wm: chip FIFO params - * @pixel_size: display pixel size - * @latency_ns: memory latency for the platform - * - * Calculate the watermark level (the level at which the display plane will - * start fetching from memory again). Each chip has a different display - * FIFO size and allocation, so the caller needs to figure that out and pass - * in the correct intel_watermark_params structure. - * - * As the pixel clock runs, the FIFO will be drained at a rate that depends - * on the pixel size. When it reaches the watermark level, it'll start - * fetching FIFO line sized based chunks from memory until the FIFO fills - * past the watermark point. If the FIFO drains completely, a FIFO underrun - * will occur, and a display engine hang could result. - */ -static unsigned long intel_calculate_wm(unsigned long clock_in_khz, - const struct intel_watermark_params *wm, - int fifo_size, - int pixel_size, - unsigned long latency_ns) -{ - long entries_required, wm_size; - - /* - * Note: we need to make sure we don't overflow for various clock & - * latency values. - * clocks go from a few thousand to several hundred thousand. - * latency is usually a few thousand - */ - entries_required = ((clock_in_khz / 1000) * pixel_size * latency_ns) / - 1000; - entries_required = DIV_ROUND_UP(entries_required, wm->cacheline_size); - - DRM_DEBUG_KMS("FIFO entries required for mode: %ld\n", entries_required); - - wm_size = fifo_size - (entries_required + wm->guard_size); - - DRM_DEBUG_KMS("FIFO watermark level: %ld\n", wm_size); - - /* Don't promote wm_size to unsigned... */ - if (wm_size > (long)wm->max_wm) - wm_size = wm->max_wm; - if (wm_size <= 0) - wm_size = wm->default_wm; - return wm_size; -} - -struct cxsr_latency { - int is_desktop; - int is_ddr3; - unsigned long fsb_freq; - unsigned long mem_freq; - unsigned long display_sr; - unsigned long display_hpll_disable; - unsigned long cursor_sr; - unsigned long cursor_hpll_disable; -}; - -static const struct cxsr_latency cxsr_latency_table[] = { - {1, 0, 800, 400, 3382, 33382, 3983, 33983}, /* DDR2-400 SC */ - {1, 0, 800, 667, 3354, 33354, 3807, 33807}, /* DDR2-667 SC */ - {1, 0, 800, 800, 3347, 33347, 3763, 33763}, /* DDR2-800 SC */ - {1, 1, 800, 667, 6420, 36420, 6873, 36873}, /* DDR3-667 SC */ - {1, 1, 800, 800, 5902, 35902, 6318, 36318}, /* DDR3-800 SC */ - - {1, 0, 667, 400, 3400, 33400, 4021, 34021}, /* DDR2-400 SC */ - {1, 0, 667, 667, 3372, 33372, 3845, 33845}, /* DDR2-667 SC */ - {1, 0, 667, 800, 3386, 33386, 3822, 33822}, /* DDR2-800 SC */ - {1, 1, 667, 667, 6438, 36438, 6911, 36911}, /* DDR3-667 SC */ - {1, 1, 667, 800, 5941, 35941, 6377, 36377}, /* DDR3-800 SC */ - - {1, 0, 400, 400, 3472, 33472, 4173, 34173}, /* DDR2-400 SC */ - {1, 0, 400, 667, 3443, 33443, 3996, 33996}, /* DDR2-667 SC */ - {1, 0, 400, 800, 3430, 33430, 3946, 33946}, /* DDR2-800 SC */ - {1, 1, 400, 667, 6509, 36509, 7062, 37062}, /* DDR3-667 SC */ - {1, 1, 400, 800, 5985, 35985, 6501, 36501}, /* DDR3-800 SC */ - - {0, 0, 800, 400, 3438, 33438, 4065, 34065}, /* DDR2-400 SC */ - {0, 0, 800, 667, 3410, 33410, 3889, 33889}, /* DDR2-667 SC */ - {0, 0, 800, 800, 3403, 33403, 3845, 33845}, /* DDR2-800 SC */ - {0, 1, 800, 667, 6476, 36476, 6955, 36955}, /* DDR3-667 SC */ - {0, 1, 800, 800, 5958, 35958, 6400, 36400}, /* DDR3-800 SC */ - - {0, 0, 667, 400, 3456, 33456, 4103, 34106}, /* DDR2-400 SC */ - {0, 0, 667, 667, 3428, 33428, 3927, 33927}, /* DDR2-667 SC */ - {0, 0, 667, 800, 3443, 33443, 3905, 33905}, /* DDR2-800 SC */ - {0, 1, 667, 667, 6494, 36494, 6993, 36993}, /* DDR3-667 SC */ - {0, 1, 667, 800, 5998, 35998, 6460, 36460}, /* DDR3-800 SC */ - - {0, 0, 400, 400, 3528, 33528, 4255, 34255}, /* DDR2-400 SC */ - {0, 0, 400, 667, 3500, 33500, 4079, 34079}, /* DDR2-667 SC */ - {0, 0, 400, 800, 3487, 33487, 4029, 34029}, /* DDR2-800 SC */ - {0, 1, 400, 667, 6566, 36566, 7145, 37145}, /* DDR3-667 SC */ - {0, 1, 400, 800, 6042, 36042, 6584, 36584}, /* DDR3-800 SC */ -}; - -static const struct cxsr_latency *intel_get_cxsr_latency(int is_desktop, - int is_ddr3, - int fsb, - int mem) -{ - const struct cxsr_latency *latency; - int i; - - if (fsb == 0 || mem == 0) - return NULL; - - for (i = 0; i < ARRAY_SIZE(cxsr_latency_table); i++) { - latency = &cxsr_latency_table[i]; - if (is_desktop == latency->is_desktop && - is_ddr3 == latency->is_ddr3 && - fsb == latency->fsb_freq && mem == latency->mem_freq) - return latency; - } - - DRM_DEBUG_KMS("Unknown FSB/MEM found, disable CxSR\n"); - - return NULL; -} - -static void pineview_disable_cxsr(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - - /* deactivate cxsr */ - I915_WRITE(DSPFW3, I915_READ(DSPFW3) & ~PINEVIEW_SELF_REFRESH_EN); -} - -/* - * Latency for FIFO fetches is dependent on several factors: - * - memory configuration (speed, channels) - * - chipset - * - current MCH state - * It can be fairly high in some situations, so here we assume a fairly - * pessimal value. It's a tradeoff between extra memory fetches (if we - * set this value too high, the FIFO will fetch frequently to stay full) - * and power consumption (set it too low to save power and we might see - * FIFO underruns and display "flicker"). - * - * A value of 5us seems to be a good balance; safe for very low end - * platforms but not overly aggressive on lower latency configs. - */ -static const int latency_ns = 5000; - -static int i9xx_get_fifo_size(struct drm_device *dev, int plane) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - uint32_t dsparb = I915_READ(DSPARB); - int size; - - size = dsparb & 0x7f; - if (plane) - size = ((dsparb >> DSPARB_CSTART_SHIFT) & 0x7f) - size; - - DRM_DEBUG_KMS("FIFO size - (0x%08x) %s: %d\n", dsparb, - plane ? "B" : "A", size); - - return size; -} - -static int i85x_get_fifo_size(struct drm_device *dev, int plane) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - uint32_t dsparb = I915_READ(DSPARB); - int size; - - size = dsparb & 0x1ff; - if (plane) - size = ((dsparb >> DSPARB_BEND_SHIFT) & 0x1ff) - size; - size >>= 1; /* Convert to cachelines */ - - DRM_DEBUG_KMS("FIFO size - (0x%08x) %s: %d\n", dsparb, - plane ? "B" : "A", size); - - return size; -} - -static int i845_get_fifo_size(struct drm_device *dev, int plane) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - uint32_t dsparb = I915_READ(DSPARB); - int size; - - size = dsparb & 0x7f; - size >>= 2; /* Convert to cachelines */ - - DRM_DEBUG_KMS("FIFO size - (0x%08x) %s: %d\n", dsparb, - plane ? "B" : "A", - size); - - return size; -} - -static int i830_get_fifo_size(struct drm_device *dev, int plane) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - uint32_t dsparb = I915_READ(DSPARB); - int size; - - size = dsparb & 0x7f; - size >>= 1; /* Convert to cachelines */ - - DRM_DEBUG_KMS("FIFO size - (0x%08x) %s: %d\n", dsparb, - plane ? "B" : "A", size); - - return size; -} - -static struct drm_crtc *single_enabled_crtc(struct drm_device *dev) -{ - struct drm_crtc *crtc, *enabled = NULL; - - list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) { - if (crtc->enabled && crtc->fb) { - if (enabled) - return NULL; - enabled = crtc; - } - } - - return enabled; -} - -static void pineview_update_wm(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - struct drm_crtc *crtc; - const struct cxsr_latency *latency; - u32 reg; - unsigned long wm; - - latency = intel_get_cxsr_latency(IS_PINEVIEW_G(dev), dev_priv->is_ddr3, - dev_priv->fsb_freq, dev_priv->mem_freq); - if (!latency) { - DRM_DEBUG_KMS("Unknown FSB/MEM found, disable CxSR\n"); - pineview_disable_cxsr(dev); - return; - } - - crtc = single_enabled_crtc(dev); - if (crtc) { - int clock = crtc->mode.clock; - int pixel_size = crtc->fb->bits_per_pixel / 8; - - /* Display SR */ - wm = intel_calculate_wm(clock, &pineview_display_wm, - pineview_display_wm.fifo_size, - pixel_size, latency->display_sr); - reg = I915_READ(DSPFW1); - reg &= ~DSPFW_SR_MASK; - reg |= wm << DSPFW_SR_SHIFT; - I915_WRITE(DSPFW1, reg); - DRM_DEBUG_KMS("DSPFW1 register is %x\n", reg); - - /* cursor SR */ - wm = intel_calculate_wm(clock, &pineview_cursor_wm, - pineview_display_wm.fifo_size, - pixel_size, latency->cursor_sr); - reg = I915_READ(DSPFW3); - reg &= ~DSPFW_CURSOR_SR_MASK; - reg |= (wm & 0x3f) << DSPFW_CURSOR_SR_SHIFT; - I915_WRITE(DSPFW3, reg); - - /* Display HPLL off SR */ - wm = intel_calculate_wm(clock, &pineview_display_hplloff_wm, - pineview_display_hplloff_wm.fifo_size, - pixel_size, latency->display_hpll_disable); - reg = I915_READ(DSPFW3); - reg &= ~DSPFW_HPLL_SR_MASK; - reg |= wm & DSPFW_HPLL_SR_MASK; - I915_WRITE(DSPFW3, reg); - - /* cursor HPLL off SR */ - wm = intel_calculate_wm(clock, &pineview_cursor_hplloff_wm, - pineview_display_hplloff_wm.fifo_size, - pixel_size, latency->cursor_hpll_disable); - reg = I915_READ(DSPFW3); - reg &= ~DSPFW_HPLL_CURSOR_MASK; - reg |= (wm & 0x3f) << DSPFW_HPLL_CURSOR_SHIFT; - I915_WRITE(DSPFW3, reg); - DRM_DEBUG_KMS("DSPFW3 register is %x\n", reg); - - /* activate cxsr */ - I915_WRITE(DSPFW3, - I915_READ(DSPFW3) | PINEVIEW_SELF_REFRESH_EN); - DRM_DEBUG_KMS("Self-refresh is enabled\n"); - } else { - pineview_disable_cxsr(dev); - DRM_DEBUG_KMS("Self-refresh is disabled\n"); - } -} - -static bool g4x_compute_wm0(struct drm_device *dev, - int plane, - const struct intel_watermark_params *display, - int display_latency_ns, - const struct intel_watermark_params *cursor, - int cursor_latency_ns, - int *plane_wm, - int *cursor_wm) -{ - struct drm_crtc *crtc; - int htotal, hdisplay, clock, pixel_size; - int line_time_us, line_count; - int entries, tlb_miss; - - crtc = intel_get_crtc_for_plane(dev, plane); - if (crtc->fb == NULL || !crtc->enabled) { - *cursor_wm = cursor->guard_size; - *plane_wm = display->guard_size; - return false; - } - - htotal = crtc->mode.htotal; - hdisplay = crtc->mode.hdisplay; - clock = crtc->mode.clock; - pixel_size = crtc->fb->bits_per_pixel / 8; - - /* Use the small buffer method to calculate plane watermark */ - entries = ((clock * pixel_size / 1000) * display_latency_ns) / 1000; - tlb_miss = display->fifo_size*display->cacheline_size - hdisplay * 8; - if (tlb_miss > 0) - entries += tlb_miss; - entries = DIV_ROUND_UP(entries, display->cacheline_size); - *plane_wm = entries + display->guard_size; - if (*plane_wm > (int)display->max_wm) - *plane_wm = display->max_wm; - - /* Use the large buffer method to calculate cursor watermark */ - line_time_us = ((htotal * 1000) / clock); - line_count = (cursor_latency_ns / line_time_us + 1000) / 1000; - entries = line_count * 64 * pixel_size; - tlb_miss = cursor->fifo_size*cursor->cacheline_size - hdisplay * 8; - if (tlb_miss > 0) - entries += tlb_miss; - entries = DIV_ROUND_UP(entries, cursor->cacheline_size); - *cursor_wm = entries + cursor->guard_size; - if (*cursor_wm > (int)cursor->max_wm) - *cursor_wm = (int)cursor->max_wm; - - return true; -} - -/* - * Check the wm result. - * - * If any calculated watermark values is larger than the maximum value that - * can be programmed into the associated watermark register, that watermark - * must be disabled. - */ -static bool g4x_check_srwm(struct drm_device *dev, - int display_wm, int cursor_wm, - const struct intel_watermark_params *display, - const struct intel_watermark_params *cursor) -{ - DRM_DEBUG_KMS("SR watermark: display plane %d, cursor %d\n", - display_wm, cursor_wm); - - if (display_wm > display->max_wm) { - DRM_DEBUG_KMS("display watermark is too large(%d/%ld), disabling\n", - display_wm, display->max_wm); - return false; - } - - if (cursor_wm > cursor->max_wm) { - DRM_DEBUG_KMS("cursor watermark is too large(%d/%ld), disabling\n", - cursor_wm, cursor->max_wm); - return false; - } - - if (!(display_wm || cursor_wm)) { - DRM_DEBUG_KMS("SR latency is 0, disabling\n"); - return false; - } - - return true; -} - -static bool g4x_compute_srwm(struct drm_device *dev, - int plane, - int latency_ns, - const struct intel_watermark_params *display, - const struct intel_watermark_params *cursor, - int *display_wm, int *cursor_wm) -{ - struct drm_crtc *crtc; - int hdisplay, htotal, pixel_size, clock; - unsigned long line_time_us; - int line_count, line_size; - int small, large; - int entries; - - if (!latency_ns) { - *display_wm = *cursor_wm = 0; - return false; - } - - crtc = intel_get_crtc_for_plane(dev, plane); - hdisplay = crtc->mode.hdisplay; - htotal = crtc->mode.htotal; - clock = crtc->mode.clock; - pixel_size = crtc->fb->bits_per_pixel / 8; - - line_time_us = (htotal * 1000) / clock; - line_count = (latency_ns / line_time_us + 1000) / 1000; - line_size = hdisplay * pixel_size; - - /* Use the minimum of the small and large buffer method for primary */ - small = ((clock * pixel_size / 1000) * latency_ns) / 1000; - large = line_count * line_size; - - entries = DIV_ROUND_UP(min(small, large), display->cacheline_size); - *display_wm = entries + display->guard_size; - - /* calculate the self-refresh watermark for display cursor */ - entries = line_count * pixel_size * 64; - entries = DIV_ROUND_UP(entries, cursor->cacheline_size); - *cursor_wm = entries + cursor->guard_size; - - return g4x_check_srwm(dev, - *display_wm, *cursor_wm, - display, cursor); -} - -static bool vlv_compute_drain_latency(struct drm_device *dev, - int plane, - int *plane_prec_mult, - int *plane_dl, - int *cursor_prec_mult, - int *cursor_dl) -{ - struct drm_crtc *crtc; - int clock, pixel_size; - int entries; - - crtc = intel_get_crtc_for_plane(dev, plane); - if (crtc->fb == NULL || !crtc->enabled) - return false; - - clock = crtc->mode.clock; /* VESA DOT Clock */ - pixel_size = crtc->fb->bits_per_pixel / 8; /* BPP */ - - entries = (clock / 1000) * pixel_size; - *plane_prec_mult = (entries > 256) ? - DRAIN_LATENCY_PRECISION_32 : DRAIN_LATENCY_PRECISION_16; - *plane_dl = (64 * (*plane_prec_mult) * 4) / ((clock / 1000) * - pixel_size); - - entries = (clock / 1000) * 4; /* BPP is always 4 for cursor */ - *cursor_prec_mult = (entries > 256) ? - DRAIN_LATENCY_PRECISION_32 : DRAIN_LATENCY_PRECISION_16; - *cursor_dl = (64 * (*cursor_prec_mult) * 4) / ((clock / 1000) * 4); - - return true; -} - -/* - * Update drain latency registers of memory arbiter - * - * Valleyview SoC has a new memory arbiter and needs drain latency registers - * to be programmed. Each plane has a drain latency multiplier and a drain - * latency value. - */ - -static void vlv_update_drain_latency(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - int planea_prec, planea_dl, planeb_prec, planeb_dl; - int cursora_prec, cursora_dl, cursorb_prec, cursorb_dl; - int plane_prec_mult, cursor_prec_mult; /* Precision multiplier is - either 16 or 32 */ - - /* For plane A, Cursor A */ - if (vlv_compute_drain_latency(dev, 0, &plane_prec_mult, &planea_dl, - &cursor_prec_mult, &cursora_dl)) { - cursora_prec = (cursor_prec_mult == DRAIN_LATENCY_PRECISION_32) ? - DDL_CURSORA_PRECISION_32 : DDL_CURSORA_PRECISION_16; - planea_prec = (plane_prec_mult == DRAIN_LATENCY_PRECISION_32) ? - DDL_PLANEA_PRECISION_32 : DDL_PLANEA_PRECISION_16; - - I915_WRITE(VLV_DDL1, cursora_prec | - (cursora_dl << DDL_CURSORA_SHIFT) | - planea_prec | planea_dl); - } - - /* For plane B, Cursor B */ - if (vlv_compute_drain_latency(dev, 1, &plane_prec_mult, &planeb_dl, - &cursor_prec_mult, &cursorb_dl)) { - cursorb_prec = (cursor_prec_mult == DRAIN_LATENCY_PRECISION_32) ? - DDL_CURSORB_PRECISION_32 : DDL_CURSORB_PRECISION_16; - planeb_prec = (plane_prec_mult == DRAIN_LATENCY_PRECISION_32) ? - DDL_PLANEB_PRECISION_32 : DDL_PLANEB_PRECISION_16; - - I915_WRITE(VLV_DDL2, cursorb_prec | - (cursorb_dl << DDL_CURSORB_SHIFT) | - planeb_prec | planeb_dl); - } -} - -#define single_plane_enabled(mask) is_power_of_2(mask) - -static void valleyview_update_wm(struct drm_device *dev) -{ - static const int sr_latency_ns = 12000; - struct drm_i915_private *dev_priv = dev->dev_private; - int planea_wm, planeb_wm, cursora_wm, cursorb_wm; - int plane_sr, cursor_sr; - unsigned int enabled = 0; - - vlv_update_drain_latency(dev); - - if (g4x_compute_wm0(dev, 0, - &valleyview_wm_info, latency_ns, - &valleyview_cursor_wm_info, latency_ns, - &planea_wm, &cursora_wm)) - enabled |= 1; - - if (g4x_compute_wm0(dev, 1, - &valleyview_wm_info, latency_ns, - &valleyview_cursor_wm_info, latency_ns, - &planeb_wm, &cursorb_wm)) - enabled |= 2; - - plane_sr = cursor_sr = 0; - if (single_plane_enabled(enabled) && - g4x_compute_srwm(dev, ffs(enabled) - 1, - sr_latency_ns, - &valleyview_wm_info, - &valleyview_cursor_wm_info, - &plane_sr, &cursor_sr)) - I915_WRITE(FW_BLC_SELF_VLV, FW_CSPWRDWNEN); - else - I915_WRITE(FW_BLC_SELF_VLV, - I915_READ(FW_BLC_SELF_VLV) & ~FW_CSPWRDWNEN); - - DRM_DEBUG_KMS("Setting FIFO watermarks - A: plane=%d, cursor=%d, B: plane=%d, cursor=%d, SR: plane=%d, cursor=%d\n", - planea_wm, cursora_wm, - planeb_wm, cursorb_wm, - plane_sr, cursor_sr); - - I915_WRITE(DSPFW1, - (plane_sr << DSPFW_SR_SHIFT) | - (cursorb_wm << DSPFW_CURSORB_SHIFT) | - (planeb_wm << DSPFW_PLANEB_SHIFT) | - planea_wm); - I915_WRITE(DSPFW2, - (I915_READ(DSPFW2) & DSPFW_CURSORA_MASK) | - (cursora_wm << DSPFW_CURSORA_SHIFT)); - I915_WRITE(DSPFW3, - (I915_READ(DSPFW3) | (cursor_sr << DSPFW_CURSOR_SR_SHIFT))); -} - -static void g4x_update_wm(struct drm_device *dev) -{ - static const int sr_latency_ns = 12000; - struct drm_i915_private *dev_priv = dev->dev_private; - int planea_wm, planeb_wm, cursora_wm, cursorb_wm; - int plane_sr, cursor_sr; - unsigned int enabled = 0; - - if (g4x_compute_wm0(dev, 0, - &g4x_wm_info, latency_ns, - &g4x_cursor_wm_info, latency_ns, - &planea_wm, &cursora_wm)) - enabled |= 1; - - if (g4x_compute_wm0(dev, 1, - &g4x_wm_info, latency_ns, - &g4x_cursor_wm_info, latency_ns, - &planeb_wm, &cursorb_wm)) - enabled |= 2; - - plane_sr = cursor_sr = 0; - if (single_plane_enabled(enabled) && - g4x_compute_srwm(dev, ffs(enabled) - 1, - sr_latency_ns, - &g4x_wm_info, - &g4x_cursor_wm_info, - &plane_sr, &cursor_sr)) - I915_WRITE(FW_BLC_SELF, FW_BLC_SELF_EN); - else - I915_WRITE(FW_BLC_SELF, - I915_READ(FW_BLC_SELF) & ~FW_BLC_SELF_EN); - - DRM_DEBUG_KMS("Setting FIFO watermarks - A: plane=%d, cursor=%d, B: plane=%d, cursor=%d, SR: plane=%d, cursor=%d\n", - planea_wm, cursora_wm, - planeb_wm, cursorb_wm, - plane_sr, cursor_sr); - - I915_WRITE(DSPFW1, - (plane_sr << DSPFW_SR_SHIFT) | - (cursorb_wm << DSPFW_CURSORB_SHIFT) | - (planeb_wm << DSPFW_PLANEB_SHIFT) | - planea_wm); - I915_WRITE(DSPFW2, - (I915_READ(DSPFW2) & DSPFW_CURSORA_MASK) | - (cursora_wm << DSPFW_CURSORA_SHIFT)); - /* HPLL off in SR has some issues on G4x... disable it */ - I915_WRITE(DSPFW3, - (I915_READ(DSPFW3) & ~DSPFW_HPLL_SR_EN) | - (cursor_sr << DSPFW_CURSOR_SR_SHIFT)); -} - -static void i965_update_wm(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - struct drm_crtc *crtc; - int srwm = 1; - int cursor_sr = 16; - - /* Calc sr entries for one plane configs */ - crtc = single_enabled_crtc(dev); - if (crtc) { - /* self-refresh has much higher latency */ - static const int sr_latency_ns = 12000; - int clock = crtc->mode.clock; - int htotal = crtc->mode.htotal; - int hdisplay = crtc->mode.hdisplay; - int pixel_size = crtc->fb->bits_per_pixel / 8; - unsigned long line_time_us; - int entries; - - line_time_us = ((htotal * 1000) / clock); - - /* Use ns/us then divide to preserve precision */ - entries = (((sr_latency_ns / line_time_us) + 1000) / 1000) * - pixel_size * hdisplay; - entries = DIV_ROUND_UP(entries, I915_FIFO_LINE_SIZE); - srwm = I965_FIFO_SIZE - entries; - if (srwm < 0) - srwm = 1; - srwm &= 0x1ff; - DRM_DEBUG_KMS("self-refresh entries: %d, wm: %d\n", - entries, srwm); - - entries = (((sr_latency_ns / line_time_us) + 1000) / 1000) * - pixel_size * 64; - entries = DIV_ROUND_UP(entries, - i965_cursor_wm_info.cacheline_size); - cursor_sr = i965_cursor_wm_info.fifo_size - - (entries + i965_cursor_wm_info.guard_size); - - if (cursor_sr > i965_cursor_wm_info.max_wm) - cursor_sr = i965_cursor_wm_info.max_wm; - - DRM_DEBUG_KMS("self-refresh watermark: display plane %d " - "cursor %d\n", srwm, cursor_sr); - - if (IS_CRESTLINE(dev)) - I915_WRITE(FW_BLC_SELF, FW_BLC_SELF_EN); - } else { - /* Turn off self refresh if both pipes are enabled */ - if (IS_CRESTLINE(dev)) - I915_WRITE(FW_BLC_SELF, I915_READ(FW_BLC_SELF) - & ~FW_BLC_SELF_EN); - } - - DRM_DEBUG_KMS("Setting FIFO watermarks - A: 8, B: 8, C: 8, SR %d\n", - srwm); - - /* 965 has limitations... */ - I915_WRITE(DSPFW1, (srwm << DSPFW_SR_SHIFT) | - (8 << 16) | (8 << 8) | (8 << 0)); - I915_WRITE(DSPFW2, (8 << 8) | (8 << 0)); - /* update cursor SR watermark */ - I915_WRITE(DSPFW3, (cursor_sr << DSPFW_CURSOR_SR_SHIFT)); -} - -static void i9xx_update_wm(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - const struct intel_watermark_params *wm_info; - uint32_t fwater_lo; - uint32_t fwater_hi; - int cwm, srwm = 1; - int fifo_size; - int planea_wm, planeb_wm; - struct drm_crtc *crtc, *enabled = NULL; - - if (IS_I945GM(dev)) - wm_info = &i945_wm_info; - else if (!IS_GEN2(dev)) - wm_info = &i915_wm_info; - else - wm_info = &i855_wm_info; - - fifo_size = dev_priv->display.get_fifo_size(dev, 0); - crtc = intel_get_crtc_for_plane(dev, 0); - if (crtc->enabled && crtc->fb) { - planea_wm = intel_calculate_wm(crtc->mode.clock, - wm_info, fifo_size, - crtc->fb->bits_per_pixel / 8, - latency_ns); - enabled = crtc; - } else - planea_wm = fifo_size - wm_info->guard_size; - - fifo_size = dev_priv->display.get_fifo_size(dev, 1); - crtc = intel_get_crtc_for_plane(dev, 1); - if (crtc->enabled && crtc->fb) { - planeb_wm = intel_calculate_wm(crtc->mode.clock, - wm_info, fifo_size, - crtc->fb->bits_per_pixel / 8, - latency_ns); - if (enabled == NULL) - enabled = crtc; - else - enabled = NULL; - } else - planeb_wm = fifo_size - wm_info->guard_size; - - DRM_DEBUG_KMS("FIFO watermarks - A: %d, B: %d\n", planea_wm, planeb_wm); - - /* - * Overlay gets an aggressive default since video jitter is bad. - */ - cwm = 2; - - /* Play safe and disable self-refresh before adjusting watermarks. */ - if (IS_I945G(dev) || IS_I945GM(dev)) - I915_WRITE(FW_BLC_SELF, FW_BLC_SELF_EN_MASK | 0); - else if (IS_I915GM(dev)) - I915_WRITE(INSTPM, I915_READ(INSTPM) & ~INSTPM_SELF_EN); - - /* Calc sr entries for one plane configs */ - if (HAS_FW_BLC(dev) && enabled) { - /* self-refresh has much higher latency */ - static const int sr_latency_ns = 6000; - int clock = enabled->mode.clock; - int htotal = enabled->mode.htotal; - int hdisplay = enabled->mode.hdisplay; - int pixel_size = enabled->fb->bits_per_pixel / 8; - unsigned long line_time_us; - int entries; - - line_time_us = (htotal * 1000) / clock; - - /* Use ns/us then divide to preserve precision */ - entries = (((sr_latency_ns / line_time_us) + 1000) / 1000) * - pixel_size * hdisplay; - entries = DIV_ROUND_UP(entries, wm_info->cacheline_size); - DRM_DEBUG_KMS("self-refresh entries: %d\n", entries); - srwm = wm_info->fifo_size - entries; - if (srwm < 0) - srwm = 1; - - if (IS_I945G(dev) || IS_I945GM(dev)) - I915_WRITE(FW_BLC_SELF, - FW_BLC_SELF_FIFO_MASK | (srwm & 0xff)); - else if (IS_I915GM(dev)) - I915_WRITE(FW_BLC_SELF, srwm & 0x3f); - } - - DRM_DEBUG_KMS("Setting FIFO watermarks - A: %d, B: %d, C: %d, SR %d\n", - planea_wm, planeb_wm, cwm, srwm); - - fwater_lo = ((planeb_wm & 0x3f) << 16) | (planea_wm & 0x3f); - fwater_hi = (cwm & 0x1f); - - /* Set request length to 8 cachelines per fetch */ - fwater_lo = fwater_lo | (1 << 24) | (1 << 8); - fwater_hi = fwater_hi | (1 << 8); - - I915_WRITE(FW_BLC, fwater_lo); - I915_WRITE(FW_BLC2, fwater_hi); - - if (HAS_FW_BLC(dev)) { - if (enabled) { - if (IS_I945G(dev) || IS_I945GM(dev)) - I915_WRITE(FW_BLC_SELF, - FW_BLC_SELF_EN_MASK | FW_BLC_SELF_EN); - else if (IS_I915GM(dev)) - I915_WRITE(INSTPM, I915_READ(INSTPM) | INSTPM_SELF_EN); - DRM_DEBUG_KMS("memory self refresh enabled\n"); - } else - DRM_DEBUG_KMS("memory self refresh disabled\n"); - } -} - -static void i830_update_wm(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - struct drm_crtc *crtc; - uint32_t fwater_lo; - int planea_wm; - - crtc = single_enabled_crtc(dev); - if (crtc == NULL) - return; - - planea_wm = intel_calculate_wm(crtc->mode.clock, &i830_wm_info, - dev_priv->display.get_fifo_size(dev, 0), - crtc->fb->bits_per_pixel / 8, - latency_ns); - fwater_lo = I915_READ(FW_BLC) & ~0xfff; - fwater_lo |= (3<<8) | planea_wm; - - DRM_DEBUG_KMS("Setting FIFO watermarks - A: %d\n", planea_wm); - - I915_WRITE(FW_BLC, fwater_lo); -} - -#define ILK_LP0_PLANE_LATENCY 700 -#define ILK_LP0_CURSOR_LATENCY 1300 - -/* - * Check the wm result. - * - * If any calculated watermark values is larger than the maximum value that - * can be programmed into the associated watermark register, that watermark - * must be disabled. - */ -static bool ironlake_check_srwm(struct drm_device *dev, int level, - int fbc_wm, int display_wm, int cursor_wm, - const struct intel_watermark_params *display, - const struct intel_watermark_params *cursor) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - - DRM_DEBUG_KMS("watermark %d: display plane %d, fbc lines %d," - " cursor %d\n", level, display_wm, fbc_wm, cursor_wm); - - if (fbc_wm > SNB_FBC_MAX_SRWM) { - DRM_DEBUG_KMS("fbc watermark(%d) is too large(%d), disabling wm%d+\n", - fbc_wm, SNB_FBC_MAX_SRWM, level); - - /* fbc has it's own way to disable FBC WM */ - I915_WRITE(DISP_ARB_CTL, - I915_READ(DISP_ARB_CTL) | DISP_FBC_WM_DIS); - return false; - } - - if (display_wm > display->max_wm) { - DRM_DEBUG_KMS("display watermark(%d) is too large(%d), disabling wm%d+\n", - display_wm, SNB_DISPLAY_MAX_SRWM, level); - return false; - } - - if (cursor_wm > cursor->max_wm) { - DRM_DEBUG_KMS("cursor watermark(%d) is too large(%d), disabling wm%d+\n", - cursor_wm, SNB_CURSOR_MAX_SRWM, level); - return false; - } - - if (!(fbc_wm || display_wm || cursor_wm)) { - DRM_DEBUG_KMS("latency %d is 0, disabling wm%d+\n", level, level); - return false; - } - - return true; -} - -/* - * Compute watermark values of WM[1-3], - */ -static bool ironlake_compute_srwm(struct drm_device *dev, int level, int plane, - int latency_ns, - const struct intel_watermark_params *display, - const struct intel_watermark_params *cursor, - int *fbc_wm, int *display_wm, int *cursor_wm) -{ - struct drm_crtc *crtc; - unsigned long line_time_us; - int hdisplay, htotal, pixel_size, clock; - int line_count, line_size; - int small, large; - int entries; - - if (!latency_ns) { - *fbc_wm = *display_wm = *cursor_wm = 0; - return false; - } - - crtc = intel_get_crtc_for_plane(dev, plane); - hdisplay = crtc->mode.hdisplay; - htotal = crtc->mode.htotal; - clock = crtc->mode.clock; - pixel_size = crtc->fb->bits_per_pixel / 8; - - line_time_us = (htotal * 1000) / clock; - line_count = (latency_ns / line_time_us + 1000) / 1000; - line_size = hdisplay * pixel_size; - - /* Use the minimum of the small and large buffer method for primary */ - small = ((clock * pixel_size / 1000) * latency_ns) / 1000; - large = line_count * line_size; - - entries = DIV_ROUND_UP(min(small, large), display->cacheline_size); - *display_wm = entries + display->guard_size; - - /* - * Spec says: - * FBC WM = ((Final Primary WM * 64) / number of bytes per line) + 2 - */ - *fbc_wm = DIV_ROUND_UP(*display_wm * 64, line_size) + 2; - - /* calculate the self-refresh watermark for display cursor */ - entries = line_count * pixel_size * 64; - entries = DIV_ROUND_UP(entries, cursor->cacheline_size); - *cursor_wm = entries + cursor->guard_size; - - return ironlake_check_srwm(dev, level, - *fbc_wm, *display_wm, *cursor_wm, - display, cursor); -} - -static void ironlake_update_wm(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - int fbc_wm, plane_wm, cursor_wm; - unsigned int enabled; - - enabled = 0; - if (g4x_compute_wm0(dev, 0, - &ironlake_display_wm_info, - ILK_LP0_PLANE_LATENCY, - &ironlake_cursor_wm_info, - ILK_LP0_CURSOR_LATENCY, - &plane_wm, &cursor_wm)) { - I915_WRITE(WM0_PIPEA_ILK, - (plane_wm << WM0_PIPE_PLANE_SHIFT) | cursor_wm); - DRM_DEBUG_KMS("FIFO watermarks For pipe A -" - " plane %d, " "cursor: %d\n", - plane_wm, cursor_wm); - enabled |= 1; - } - - if (g4x_compute_wm0(dev, 1, - &ironlake_display_wm_info, - ILK_LP0_PLANE_LATENCY, - &ironlake_cursor_wm_info, - ILK_LP0_CURSOR_LATENCY, - &plane_wm, &cursor_wm)) { - I915_WRITE(WM0_PIPEB_ILK, - (plane_wm << WM0_PIPE_PLANE_SHIFT) | cursor_wm); - DRM_DEBUG_KMS("FIFO watermarks For pipe B -" - " plane %d, cursor: %d\n", - plane_wm, cursor_wm); - enabled |= 2; - } - - /* - * Calculate and update the self-refresh watermark only when one - * display plane is used. - */ - I915_WRITE(WM3_LP_ILK, 0); - I915_WRITE(WM2_LP_ILK, 0); - I915_WRITE(WM1_LP_ILK, 0); - - if (!single_plane_enabled(enabled)) - return; - enabled = ffs(enabled) - 1; - - /* WM1 */ - if (!ironlake_compute_srwm(dev, 1, enabled, - ILK_READ_WM1_LATENCY() * 500, - &ironlake_display_srwm_info, - &ironlake_cursor_srwm_info, - &fbc_wm, &plane_wm, &cursor_wm)) - return; - - I915_WRITE(WM1_LP_ILK, - WM1_LP_SR_EN | - (ILK_READ_WM1_LATENCY() << WM1_LP_LATENCY_SHIFT) | - (fbc_wm << WM1_LP_FBC_SHIFT) | - (plane_wm << WM1_LP_SR_SHIFT) | - cursor_wm); - - /* WM2 */ - if (!ironlake_compute_srwm(dev, 2, enabled, - ILK_READ_WM2_LATENCY() * 500, - &ironlake_display_srwm_info, - &ironlake_cursor_srwm_info, - &fbc_wm, &plane_wm, &cursor_wm)) - return; - - I915_WRITE(WM2_LP_ILK, - WM2_LP_EN | - (ILK_READ_WM2_LATENCY() << WM1_LP_LATENCY_SHIFT) | - (fbc_wm << WM1_LP_FBC_SHIFT) | - (plane_wm << WM1_LP_SR_SHIFT) | - cursor_wm); - - /* - * WM3 is unsupported on ILK, probably because we don't have latency - * data for that power state - */ -} - -void sandybridge_update_wm(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - int latency = SNB_READ_WM0_LATENCY() * 100; /* In unit 0.1us */ - u32 val; - int fbc_wm, plane_wm, cursor_wm; - unsigned int enabled; - - enabled = 0; - if (g4x_compute_wm0(dev, 0, - &sandybridge_display_wm_info, latency, - &sandybridge_cursor_wm_info, latency, - &plane_wm, &cursor_wm)) { - val = I915_READ(WM0_PIPEA_ILK); - val &= ~(WM0_PIPE_PLANE_MASK | WM0_PIPE_CURSOR_MASK); - I915_WRITE(WM0_PIPEA_ILK, val | - ((plane_wm << WM0_PIPE_PLANE_SHIFT) | cursor_wm)); - DRM_DEBUG_KMS("FIFO watermarks For pipe A -" - " plane %d, " "cursor: %d\n", - plane_wm, cursor_wm); - enabled |= 1; - } - - if (g4x_compute_wm0(dev, 1, - &sandybridge_display_wm_info, latency, - &sandybridge_cursor_wm_info, latency, - &plane_wm, &cursor_wm)) { - val = I915_READ(WM0_PIPEB_ILK); - val &= ~(WM0_PIPE_PLANE_MASK | WM0_PIPE_CURSOR_MASK); - I915_WRITE(WM0_PIPEB_ILK, val | - ((plane_wm << WM0_PIPE_PLANE_SHIFT) | cursor_wm)); - DRM_DEBUG_KMS("FIFO watermarks For pipe B -" - " plane %d, cursor: %d\n", - plane_wm, cursor_wm); - enabled |= 2; - } - - /* IVB has 3 pipes */ - if (IS_IVYBRIDGE(dev) && - g4x_compute_wm0(dev, 2, - &sandybridge_display_wm_info, latency, - &sandybridge_cursor_wm_info, latency, - &plane_wm, &cursor_wm)) { - val = I915_READ(WM0_PIPEC_IVB); - val &= ~(WM0_PIPE_PLANE_MASK | WM0_PIPE_CURSOR_MASK); - I915_WRITE(WM0_PIPEC_IVB, val | - ((plane_wm << WM0_PIPE_PLANE_SHIFT) | cursor_wm)); - DRM_DEBUG_KMS("FIFO watermarks For pipe C -" - " plane %d, cursor: %d\n", - plane_wm, cursor_wm); - enabled |= 3; - } - - /* - * Calculate and update the self-refresh watermark only when one - * display plane is used. - * - * SNB support 3 levels of watermark. - * - * WM1/WM2/WM2 watermarks have to be enabled in the ascending order, - * and disabled in the descending order - * - */ - I915_WRITE(WM3_LP_ILK, 0); - I915_WRITE(WM2_LP_ILK, 0); - I915_WRITE(WM1_LP_ILK, 0); - - if (!single_plane_enabled(enabled) || - dev_priv->sprite_scaling_enabled) - return; - enabled = ffs(enabled) - 1; - - /* WM1 */ - if (!ironlake_compute_srwm(dev, 1, enabled, - SNB_READ_WM1_LATENCY() * 500, - &sandybridge_display_srwm_info, - &sandybridge_cursor_srwm_info, - &fbc_wm, &plane_wm, &cursor_wm)) - return; - - I915_WRITE(WM1_LP_ILK, - WM1_LP_SR_EN | - (SNB_READ_WM1_LATENCY() << WM1_LP_LATENCY_SHIFT) | - (fbc_wm << WM1_LP_FBC_SHIFT) | - (plane_wm << WM1_LP_SR_SHIFT) | - cursor_wm); - - /* WM2 */ - if (!ironlake_compute_srwm(dev, 2, enabled, - SNB_READ_WM2_LATENCY() * 500, - &sandybridge_display_srwm_info, - &sandybridge_cursor_srwm_info, - &fbc_wm, &plane_wm, &cursor_wm)) - return; - - I915_WRITE(WM2_LP_ILK, - WM2_LP_EN | - (SNB_READ_WM2_LATENCY() << WM1_LP_LATENCY_SHIFT) | - (fbc_wm << WM1_LP_FBC_SHIFT) | - (plane_wm << WM1_LP_SR_SHIFT) | - cursor_wm); - - /* WM3 */ - if (!ironlake_compute_srwm(dev, 3, enabled, - SNB_READ_WM3_LATENCY() * 500, - &sandybridge_display_srwm_info, - &sandybridge_cursor_srwm_info, - &fbc_wm, &plane_wm, &cursor_wm)) - return; - - I915_WRITE(WM3_LP_ILK, - WM3_LP_EN | - (SNB_READ_WM3_LATENCY() << WM1_LP_LATENCY_SHIFT) | - (fbc_wm << WM1_LP_FBC_SHIFT) | - (plane_wm << WM1_LP_SR_SHIFT) | - cursor_wm); -} - -static bool -sandybridge_compute_sprite_wm(struct drm_device *dev, int plane, - uint32_t sprite_width, int pixel_size, - const struct intel_watermark_params *display, - int display_latency_ns, int *sprite_wm) -{ - struct drm_crtc *crtc; - int clock; - int entries, tlb_miss; - - crtc = intel_get_crtc_for_plane(dev, plane); - if (crtc->fb == NULL || !crtc->enabled) { - *sprite_wm = display->guard_size; - return false; - } - - clock = crtc->mode.clock; - - /* Use the small buffer method to calculate the sprite watermark */ - entries = ((clock * pixel_size / 1000) * display_latency_ns) / 1000; - tlb_miss = display->fifo_size*display->cacheline_size - - sprite_width * 8; - if (tlb_miss > 0) - entries += tlb_miss; - entries = DIV_ROUND_UP(entries, display->cacheline_size); - *sprite_wm = entries + display->guard_size; - if (*sprite_wm > (int)display->max_wm) - *sprite_wm = display->max_wm; - - return true; -} - -static bool -sandybridge_compute_sprite_srwm(struct drm_device *dev, int plane, - uint32_t sprite_width, int pixel_size, - const struct intel_watermark_params *display, - int latency_ns, int *sprite_wm) -{ - struct drm_crtc *crtc; - unsigned long line_time_us; - int clock; - int line_count, line_size; - int small, large; - int entries; - - if (!latency_ns) { - *sprite_wm = 0; - return false; - } - - crtc = intel_get_crtc_for_plane(dev, plane); - clock = crtc->mode.clock; - if (!clock) { - *sprite_wm = 0; - return false; - } - - line_time_us = (sprite_width * 1000) / clock; - if (!line_time_us) { - *sprite_wm = 0; - return false; - } - - line_count = (latency_ns / line_time_us + 1000) / 1000; - line_size = sprite_width * pixel_size; - - /* Use the minimum of the small and large buffer method for primary */ - small = ((clock * pixel_size / 1000) * latency_ns) / 1000; - large = line_count * line_size; - - entries = DIV_ROUND_UP(min(small, large), display->cacheline_size); - *sprite_wm = entries + display->guard_size; - - return *sprite_wm > 0x3ff ? false : true; -} - -static void sandybridge_update_sprite_wm(struct drm_device *dev, int pipe, - uint32_t sprite_width, int pixel_size) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - int latency = SNB_READ_WM0_LATENCY() * 100; /* In unit 0.1us */ - u32 val; - int sprite_wm, reg; - int ret; - - switch (pipe) { - case 0: - reg = WM0_PIPEA_ILK; - break; - case 1: - reg = WM0_PIPEB_ILK; - break; - case 2: - reg = WM0_PIPEC_IVB; - break; - default: - return; /* bad pipe */ - } - - ret = sandybridge_compute_sprite_wm(dev, pipe, sprite_width, pixel_size, - &sandybridge_display_wm_info, - latency, &sprite_wm); - if (!ret) { - DRM_DEBUG_KMS("failed to compute sprite wm for pipe %d\n", - pipe); - return; - } - - val = I915_READ(reg); - val &= ~WM0_PIPE_SPRITE_MASK; - I915_WRITE(reg, val | (sprite_wm << WM0_PIPE_SPRITE_SHIFT)); - DRM_DEBUG_KMS("sprite watermarks For pipe %d - %d\n", pipe, sprite_wm); - - - ret = sandybridge_compute_sprite_srwm(dev, pipe, sprite_width, - pixel_size, - &sandybridge_display_srwm_info, - SNB_READ_WM1_LATENCY() * 500, - &sprite_wm); - if (!ret) { - DRM_DEBUG_KMS("failed to compute sprite lp1 wm on pipe %d\n", - pipe); - return; - } - I915_WRITE(WM1S_LP_ILK, sprite_wm); - - /* Only IVB has two more LP watermarks for sprite */ - if (!IS_IVYBRIDGE(dev)) - return; - - ret = sandybridge_compute_sprite_srwm(dev, pipe, sprite_width, - pixel_size, - &sandybridge_display_srwm_info, - SNB_READ_WM2_LATENCY() * 500, - &sprite_wm); - if (!ret) { - DRM_DEBUG_KMS("failed to compute sprite lp2 wm on pipe %d\n", - pipe); - return; - } - I915_WRITE(WM2S_LP_IVB, sprite_wm); - - ret = sandybridge_compute_sprite_srwm(dev, pipe, sprite_width, - pixel_size, - &sandybridge_display_srwm_info, - SNB_READ_WM3_LATENCY() * 500, - &sprite_wm); - if (!ret) { - DRM_DEBUG_KMS("failed to compute sprite lp3 wm on pipe %d\n", - pipe); - return; - } - I915_WRITE(WM3S_LP_IVB, sprite_wm); -} - -/** - * intel_update_watermarks - update FIFO watermark values based on current modes - * - * Calculate watermark values for the various WM regs based on current mode - * and plane configuration. - * - * There are several cases to deal with here: - * - normal (i.e. non-self-refresh) - * - self-refresh (SR) mode - * - lines are large relative to FIFO size (buffer can hold up to 2) - * - lines are small relative to FIFO size (buffer can hold more than 2 - * lines), so need to account for TLB latency - * - * The normal calculation is: - * watermark = dotclock * bytes per pixel * latency - * where latency is platform & configuration dependent (we assume pessimal - * values here). - * - * The SR calculation is: - * watermark = (trunc(latency/line time)+1) * surface width * - * bytes per pixel - * where - * line time = htotal / dotclock - * surface width = hdisplay for normal plane and 64 for cursor - * and latency is assumed to be high, as above. - * - * The final value programmed to the register should always be rounded up, - * and include an extra 2 entries to account for clock crossings. - * - * We don't use the sprite, so we can ignore that. And on Crestline we have - * to set the non-SR watermarks to 8. - */ -static void intel_update_watermarks(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - - if (dev_priv->display.update_wm) - dev_priv->display.update_wm(dev); -} - -void intel_update_sprite_watermarks(struct drm_device *dev, int pipe, - uint32_t sprite_width, int pixel_size) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - - if (dev_priv->display.update_sprite_wm) - dev_priv->display.update_sprite_wm(dev, pipe, sprite_width, - pixel_size); -} - static inline bool intel_panel_use_ssc(struct drm_i915_private *dev_priv) { if (i915_panel_use_ssc >= 0) @@ -5375,7 +3416,7 @@ static void intel_update_lvds(struct drm_crtc *crtc, intel_clock_t *clock, struct drm_i915_private *dev_priv = dev->dev_private; struct intel_crtc *intel_crtc = to_intel_crtc(crtc); int pipe = intel_crtc->pipe; - u32 temp, lvds_sync = 0; + u32 temp; temp = I915_READ(LVDS); temp |= LVDS_PORT_EN | LVDS_A0A2_CLKA_POWER_UP; @@ -5405,22 +3446,11 @@ static void intel_update_lvds(struct drm_crtc *crtc, intel_clock_t *clock, else temp &= ~LVDS_ENABLE_DITHER; } + temp &= ~(LVDS_HSYNC_POLARITY | LVDS_VSYNC_POLARITY); if (adjusted_mode->flags & DRM_MODE_FLAG_NHSYNC) - lvds_sync |= LVDS_HSYNC_POLARITY; + temp |= LVDS_HSYNC_POLARITY; if (adjusted_mode->flags & DRM_MODE_FLAG_NVSYNC) - lvds_sync |= LVDS_VSYNC_POLARITY; - if ((temp & (LVDS_HSYNC_POLARITY | LVDS_VSYNC_POLARITY)) - != lvds_sync) { - char flags[2] = "-+"; - DRM_INFO("Changing LVDS panel from " - "(%chsync, %cvsync) to (%chsync, %cvsync)\n", - flags[!(temp & LVDS_HSYNC_POLARITY)], - flags[!(temp & LVDS_VSYNC_POLARITY)], - flags[!(lvds_sync & LVDS_HSYNC_POLARITY)], - flags[!(lvds_sync & LVDS_VSYNC_POLARITY)]); - temp &= ~(LVDS_HSYNC_POLARITY | LVDS_VSYNC_POLARITY); - temp |= lvds_sync; - } + temp |= LVDS_VSYNC_POLARITY; I915_WRITE(LVDS, temp); } @@ -5965,17 +3995,16 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc, u32 dpll, fp = 0, fp2 = 0, dspcntr, pipeconf; bool ok, has_reduced_clock = false, is_sdvo = false; bool is_crt = false, is_lvds = false, is_tv = false, is_dp = false; - struct intel_encoder *has_edp_encoder = NULL; struct drm_mode_config *mode_config = &dev->mode_config; - struct intel_encoder *encoder; + struct intel_encoder *encoder, *edp_encoder = NULL; const intel_limit_t *limit; int ret; struct fdi_m_n m_n = {0}; u32 temp; - u32 lvds_sync = 0; int target_clock, pixel_multiplier, lane, link_bw, factor; unsigned int pipe_bpp; bool dither; + bool is_cpu_edp = false, is_pch_edp = false; list_for_each_entry(encoder, &mode_config->encoder_list, base.head) { if (encoder->base.crtc != crtc) @@ -6001,7 +4030,12 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc, is_dp = true; break; case INTEL_OUTPUT_EDP: - has_edp_encoder = encoder; + is_dp = true; + if (intel_encoder_is_pch_edp(&encoder->base)) + is_pch_edp = true; + else + is_cpu_edp = true; + edp_encoder = encoder; break; } @@ -6064,15 +4098,13 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc, lane = 0; /* CPU eDP doesn't require FDI link, so just set DP M/N according to current link config */ - if (has_edp_encoder && - !intel_encoder_is_pch_edp(&has_edp_encoder->base)) { + if (is_cpu_edp) { target_clock = mode->clock; - intel_edp_link_config(has_edp_encoder, - &lane, &link_bw); + intel_edp_link_config(edp_encoder, &lane, &link_bw); } else { /* [e]DP over FDI requires target mode clock instead of link clock */ - if (is_dp || intel_encoder_is_pch_edp(&has_edp_encoder->base)) + if (is_dp) target_clock = mode->clock; else target_clock = adjusted_mode->clock; @@ -6163,7 +4195,7 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc, } dpll |= DPLL_DVO_HIGH_SPEED; } - if (is_dp || intel_encoder_is_pch_edp(&has_edp_encoder->base)) + if (is_dp && !is_cpu_edp) dpll |= DPLL_DVO_HIGH_SPEED; /* compute bitmask from p1 value */ @@ -6208,8 +4240,7 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc, /* PCH eDP needs FDI, but CPU eDP does not */ if (!intel_crtc->no_pll) { - if (!has_edp_encoder || - intel_encoder_is_pch_edp(&has_edp_encoder->base)) { + if (!is_cpu_edp) { I915_WRITE(PCH_FP0(pipe), fp); I915_WRITE(PCH_DPLL(pipe), dpll & ~DPLL_VCO_ENABLE); @@ -6262,22 +4293,11 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc, * appropriately here, but we need to look more thoroughly into how * panels behave in the two modes. */ + temp &= ~(LVDS_HSYNC_POLARITY | LVDS_VSYNC_POLARITY); if (adjusted_mode->flags & DRM_MODE_FLAG_NHSYNC) - lvds_sync |= LVDS_HSYNC_POLARITY; + temp |= LVDS_HSYNC_POLARITY; if (adjusted_mode->flags & DRM_MODE_FLAG_NVSYNC) - lvds_sync |= LVDS_VSYNC_POLARITY; - if ((temp & (LVDS_HSYNC_POLARITY | LVDS_VSYNC_POLARITY)) - != lvds_sync) { - char flags[2] = "-+"; - DRM_INFO("Changing LVDS panel from " - "(%chsync, %cvsync) to (%chsync, %cvsync)\n", - flags[!(temp & LVDS_HSYNC_POLARITY)], - flags[!(temp & LVDS_VSYNC_POLARITY)], - flags[!(lvds_sync & LVDS_HSYNC_POLARITY)], - flags[!(lvds_sync & LVDS_VSYNC_POLARITY)]); - temp &= ~(LVDS_HSYNC_POLARITY | LVDS_VSYNC_POLARITY); - temp |= lvds_sync; - } + temp |= LVDS_VSYNC_POLARITY; I915_WRITE(PCH_LVDS, temp); } @@ -6287,7 +4307,7 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc, pipeconf |= PIPECONF_DITHER_EN; pipeconf |= PIPECONF_DITHER_TYPE_SP; } - if (is_dp || intel_encoder_is_pch_edp(&has_edp_encoder->base)) { + if (is_dp && !is_cpu_edp) { intel_dp_set_m_n(crtc, mode, adjusted_mode); } else { /* For non-DP output, clear any trans DP clock recovery setting.*/ @@ -6297,9 +4317,7 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc, I915_WRITE(TRANSDPLINK_N1(pipe), 0); } - if (!intel_crtc->no_pll && - (!has_edp_encoder || - intel_encoder_is_pch_edp(&has_edp_encoder->base))) { + if (!intel_crtc->no_pll && (!edp_encoder || is_pch_edp)) { I915_WRITE(PCH_DPLL(pipe), dpll); /* Wait for the clocks to stabilize. */ @@ -6377,10 +4395,8 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc, I915_WRITE(PIPE_LINK_M1(pipe), m_n.link_m); I915_WRITE(PIPE_LINK_N1(pipe), m_n.link_n); - if (has_edp_encoder && - !intel_encoder_is_pch_edp(&has_edp_encoder->base)) { + if (is_cpu_edp) ironlake_set_pll_edp(crtc, adjusted_mode->clock); - } I915_WRITE(PIPECONF(pipe), pipeconf); POSTING_READ(PIPECONF(pipe)); @@ -6748,7 +4764,7 @@ static void intel_crtc_update_cursor(struct drm_crtc *crtc, if (!visible && !intel_crtc->cursor_visible) return; - if (IS_IVYBRIDGE(dev)) { + if (IS_IVYBRIDGE(dev) || IS_HASWELL(dev)) { I915_WRITE(CURPOS_IVB(pipe), pos); ivb_update_cursor(crtc, base); } else { @@ -7636,14 +5652,14 @@ static int intel_gen2_queue_flip(struct drm_device *dev, ret = intel_pin_and_fence_fb_obj(dev, obj, LP_RING(dev_priv)); if (ret) - goto out; + goto err; /* Offset into the new buffer for cases of shared fbs between CRTCs */ offset = crtc->y * fb->pitches[0] + crtc->x * fb->bits_per_pixel/8; ret = BEGIN_LP_RING(6); if (ret) - goto out; + goto err_unpin; /* Can't queue multiple flips, so wait for the previous * one to finish before executing the next. @@ -7660,7 +5676,11 @@ static int intel_gen2_queue_flip(struct drm_device *dev, OUT_RING(obj->gtt_offset + offset); OUT_RING(0); /* aux display base address, unused */ ADVANCE_LP_RING(); -out: + return 0; + +err_unpin: + intel_unpin_fb_obj(obj); +err: return ret; } @@ -7677,14 +5697,14 @@ static int intel_gen3_queue_flip(struct drm_device *dev, ret = intel_pin_and_fence_fb_obj(dev, obj, LP_RING(dev_priv)); if (ret) - goto out; + goto err; /* Offset into the new buffer for cases of shared fbs between CRTCs */ offset = crtc->y * fb->pitches[0] + crtc->x * fb->bits_per_pixel/8; ret = BEGIN_LP_RING(6); if (ret) - goto out; + goto err_unpin; if (intel_crtc->plane) flip_mask = MI_WAIT_FOR_PLANE_B_FLIP; @@ -7699,7 +5719,11 @@ static int intel_gen3_queue_flip(struct drm_device *dev, OUT_RING(MI_NOOP); ADVANCE_LP_RING(); -out: + return 0; + +err_unpin: + intel_unpin_fb_obj(obj); +err: return ret; } @@ -7715,11 +5739,11 @@ static int intel_gen4_queue_flip(struct drm_device *dev, ret = intel_pin_and_fence_fb_obj(dev, obj, LP_RING(dev_priv)); if (ret) - goto out; + goto err; ret = BEGIN_LP_RING(4); if (ret) - goto out; + goto err_unpin; /* i965+ uses the linear or tiled offsets from the * Display Registers (which do not change across a page-flip) @@ -7738,7 +5762,11 @@ static int intel_gen4_queue_flip(struct drm_device *dev, pipesrc = I915_READ(PIPESRC(intel_crtc->pipe)) & 0x0fff0fff; OUT_RING(pf | pipesrc); ADVANCE_LP_RING(); -out: + return 0; + +err_unpin: + intel_unpin_fb_obj(obj); +err: return ret; } @@ -7754,11 +5782,11 @@ static int intel_gen6_queue_flip(struct drm_device *dev, ret = intel_pin_and_fence_fb_obj(dev, obj, LP_RING(dev_priv)); if (ret) - goto out; + goto err; ret = BEGIN_LP_RING(4); if (ret) - goto out; + goto err_unpin; OUT_RING(MI_DISPLAY_FLIP | MI_DISPLAY_FLIP_PLANE(intel_crtc->plane)); @@ -7769,7 +5797,11 @@ static int intel_gen6_queue_flip(struct drm_device *dev, pipesrc = I915_READ(PIPESRC(intel_crtc->pipe)) & 0x0fff0fff; OUT_RING(pf | pipesrc); ADVANCE_LP_RING(); -out: + return 0; + +err_unpin: + intel_unpin_fb_obj(obj); +err: return ret; } @@ -7791,18 +5823,22 @@ static int intel_gen7_queue_flip(struct drm_device *dev, ret = intel_pin_and_fence_fb_obj(dev, obj, ring); if (ret) - goto out; + goto err; ret = intel_ring_begin(ring, 4); if (ret) - goto out; + goto err_unpin; intel_ring_emit(ring, MI_DISPLAY_FLIP_I915 | (intel_crtc->plane << 19)); intel_ring_emit(ring, (fb->pitches[0] | obj->tiling_mode)); intel_ring_emit(ring, (obj->gtt_offset)); intel_ring_emit(ring, (MI_NOOP)); intel_ring_advance(ring); -out: + return 0; + +err_unpin: + intel_unpin_fb_obj(obj); +err: return ret; } @@ -8292,926 +6328,6 @@ static const struct drm_mode_config_funcs intel_mode_funcs = { .output_poll_changed = intel_fb_output_poll_changed, }; -static struct drm_i915_gem_object * -intel_alloc_context_page(struct drm_device *dev) -{ - struct drm_i915_gem_object *ctx; - int ret; - - WARN_ON(!mutex_is_locked(&dev->struct_mutex)); - - ctx = i915_gem_alloc_object(dev, 4096); - if (!ctx) { - DRM_DEBUG("failed to alloc power context, RC6 disabled\n"); - return NULL; - } - - ret = i915_gem_object_pin(ctx, 4096, true); - if (ret) { - DRM_ERROR("failed to pin power context: %d\n", ret); - goto err_unref; - } - - ret = i915_gem_object_set_to_gtt_domain(ctx, 1); - if (ret) { - DRM_ERROR("failed to set-domain on power context: %d\n", ret); - goto err_unpin; - } - - return ctx; - -err_unpin: - i915_gem_object_unpin(ctx); -err_unref: - drm_gem_object_unreference(&ctx->base); - mutex_unlock(&dev->struct_mutex); - return NULL; -} - -bool ironlake_set_drps(struct drm_device *dev, u8 val) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - u16 rgvswctl; - - rgvswctl = I915_READ16(MEMSWCTL); - if (rgvswctl & MEMCTL_CMD_STS) { - DRM_DEBUG("gpu busy, RCS change rejected\n"); - return false; /* still busy with another command */ - } - - rgvswctl = (MEMCTL_CMD_CHFREQ << MEMCTL_CMD_SHIFT) | - (val << MEMCTL_FREQ_SHIFT) | MEMCTL_SFCAVM; - I915_WRITE16(MEMSWCTL, rgvswctl); - POSTING_READ16(MEMSWCTL); - - rgvswctl |= MEMCTL_CMD_STS; - I915_WRITE16(MEMSWCTL, rgvswctl); - - return true; -} - -void ironlake_enable_drps(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - u32 rgvmodectl = I915_READ(MEMMODECTL); - u8 fmax, fmin, fstart, vstart; - - /* Enable temp reporting */ - I915_WRITE16(PMMISC, I915_READ(PMMISC) | MCPPCE_EN); - I915_WRITE16(TSC1, I915_READ(TSC1) | TSE); - - /* 100ms RC evaluation intervals */ - I915_WRITE(RCUPEI, 100000); - I915_WRITE(RCDNEI, 100000); - - /* Set max/min thresholds to 90ms and 80ms respectively */ - I915_WRITE(RCBMAXAVG, 90000); - I915_WRITE(RCBMINAVG, 80000); - - I915_WRITE(MEMIHYST, 1); - - /* Set up min, max, and cur for interrupt handling */ - fmax = (rgvmodectl & MEMMODE_FMAX_MASK) >> MEMMODE_FMAX_SHIFT; - fmin = (rgvmodectl & MEMMODE_FMIN_MASK); - fstart = (rgvmodectl & MEMMODE_FSTART_MASK) >> - MEMMODE_FSTART_SHIFT; - - vstart = (I915_READ(PXVFREQ_BASE + (fstart * 4)) & PXVFREQ_PX_MASK) >> - PXVFREQ_PX_SHIFT; - - dev_priv->fmax = fmax; /* IPS callback will increase this */ - dev_priv->fstart = fstart; - - dev_priv->max_delay = fstart; - dev_priv->min_delay = fmin; - dev_priv->cur_delay = fstart; - - DRM_DEBUG_DRIVER("fmax: %d, fmin: %d, fstart: %d\n", - fmax, fmin, fstart); - - I915_WRITE(MEMINTREN, MEMINT_CX_SUPR_EN | MEMINT_EVAL_CHG_EN); - - /* - * Interrupts will be enabled in ironlake_irq_postinstall - */ - - I915_WRITE(VIDSTART, vstart); - POSTING_READ(VIDSTART); - - rgvmodectl |= MEMMODE_SWMODE_EN; - I915_WRITE(MEMMODECTL, rgvmodectl); - - if (wait_for((I915_READ(MEMSWCTL) & MEMCTL_CMD_STS) == 0, 10)) - DRM_ERROR("stuck trying to change perf mode\n"); - msleep(1); - - ironlake_set_drps(dev, fstart); - - dev_priv->last_count1 = I915_READ(0x112e4) + I915_READ(0x112e8) + - I915_READ(0x112e0); - dev_priv->last_time1 = jiffies_to_msecs(jiffies); - dev_priv->last_count2 = I915_READ(0x112f4); - getrawmonotonic(&dev_priv->last_time2); -} - -void ironlake_disable_drps(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - u16 rgvswctl = I915_READ16(MEMSWCTL); - - /* Ack interrupts, disable EFC interrupt */ - I915_WRITE(MEMINTREN, I915_READ(MEMINTREN) & ~MEMINT_EVAL_CHG_EN); - I915_WRITE(MEMINTRSTS, MEMINT_EVAL_CHG); - I915_WRITE(DEIER, I915_READ(DEIER) & ~DE_PCU_EVENT); - I915_WRITE(DEIIR, DE_PCU_EVENT); - I915_WRITE(DEIMR, I915_READ(DEIMR) | DE_PCU_EVENT); - - /* Go back to the starting frequency */ - ironlake_set_drps(dev, dev_priv->fstart); - msleep(1); - rgvswctl |= MEMCTL_CMD_STS; - I915_WRITE(MEMSWCTL, rgvswctl); - msleep(1); - -} - -void gen6_set_rps(struct drm_device *dev, u8 val) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - u32 swreq; - - swreq = (val & 0x3ff) << 25; - I915_WRITE(GEN6_RPNSWREQ, swreq); -} - -void gen6_disable_rps(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - - I915_WRITE(GEN6_RPNSWREQ, 1 << 31); - I915_WRITE(GEN6_PMINTRMSK, 0xffffffff); - I915_WRITE(GEN6_PMIER, 0); - /* Complete PM interrupt masking here doesn't race with the rps work - * item again unmasking PM interrupts because that is using a different - * register (PMIMR) to mask PM interrupts. The only risk is in leaving - * stale bits in PMIIR and PMIMR which gen6_enable_rps will clean up. */ - - spin_lock_irq(&dev_priv->rps_lock); - dev_priv->pm_iir = 0; - spin_unlock_irq(&dev_priv->rps_lock); - - I915_WRITE(GEN6_PMIIR, I915_READ(GEN6_PMIIR)); -} - -static unsigned long intel_pxfreq(u32 vidfreq) -{ - unsigned long freq; - int div = (vidfreq & 0x3f0000) >> 16; - int post = (vidfreq & 0x3000) >> 12; - int pre = (vidfreq & 0x7); - - if (!pre) - return 0; - - freq = ((div * 133333) / ((1<<post) * pre)); - - return freq; -} - -void intel_init_emon(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - u32 lcfuse; - u8 pxw[16]; - int i; - - /* Disable to program */ - I915_WRITE(ECR, 0); - POSTING_READ(ECR); - - /* Program energy weights for various events */ - I915_WRITE(SDEW, 0x15040d00); - I915_WRITE(CSIEW0, 0x007f0000); - I915_WRITE(CSIEW1, 0x1e220004); - I915_WRITE(CSIEW2, 0x04000004); - - for (i = 0; i < 5; i++) - I915_WRITE(PEW + (i * 4), 0); - for (i = 0; i < 3; i++) - I915_WRITE(DEW + (i * 4), 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)); - unsigned long freq = intel_pxfreq(pxvidfreq); - unsigned long vid = (pxvidfreq & PXVFREQ_PX_MASK) >> - PXVFREQ_PX_SHIFT; - unsigned long val; - - val = vid * vid; - val *= (freq / 1000); - val *= 255; - val /= (127*127*900); - if (val > 0xff) - DRM_ERROR("bad pxval: %ld\n", val); - pxw[i] = val; - } - /* Render standby states get 0 weight */ - pxw[14] = 0; - pxw[15] = 0; - - 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); - } - - /* Adjust magic regs to magic values (more experimental results) */ - I915_WRITE(OGW0, 0); - I915_WRITE(OGW1, 0); - I915_WRITE(EG0, 0x00007f00); - I915_WRITE(EG1, 0x0000000e); - I915_WRITE(EG2, 0x000e0000); - I915_WRITE(EG3, 0x68000300); - I915_WRITE(EG4, 0x42000000); - I915_WRITE(EG5, 0x00140031); - I915_WRITE(EG6, 0); - I915_WRITE(EG7, 0); - - for (i = 0; i < 8; i++) - I915_WRITE(PXWL + (i * 4), 0); - - /* Enable PMON + select events */ - I915_WRITE(ECR, 0x80000019); - - lcfuse = I915_READ(LCFUSE02); - - dev_priv->corr = (lcfuse & LCFUSE_HIV_MASK); -} - -static int intel_enable_rc6(struct drm_device *dev) -{ - /* - * Respect the kernel parameter if it is set - */ - if (i915_enable_rc6 >= 0) - return i915_enable_rc6; - - /* - * Disable RC6 on Ironlake - */ - if (INTEL_INFO(dev)->gen == 5) - return 0; - - /* - * Disable rc6 on Sandybridge - */ - if (INTEL_INFO(dev)->gen == 6) { - DRM_DEBUG_DRIVER("Sandybridge: deep RC6 disabled\n"); - return INTEL_RC6_ENABLE; - } - DRM_DEBUG_DRIVER("RC6 and deep RC6 enabled\n"); - return (INTEL_RC6_ENABLE | INTEL_RC6p_ENABLE); -} - -void gen6_enable_rps(struct drm_i915_private *dev_priv) -{ - u32 rp_state_cap = I915_READ(GEN6_RP_STATE_CAP); - u32 gt_perf_status = I915_READ(GEN6_GT_PERF_STATUS); - u32 pcu_mbox, rc6_mask = 0; - u32 gtfifodbg; - int cur_freq, min_freq, max_freq; - int rc6_mode; - int i; - - /* Here begins a magic sequence of register writes to enable - * auto-downclocking. - * - * Perhaps there might be some value in exposing these to - * userspace... - */ - I915_WRITE(GEN6_RC_STATE, 0); - mutex_lock(&dev_priv->dev->struct_mutex); - - /* Clear the DBG now so we don't confuse earlier errors */ - if ((gtfifodbg = I915_READ(GTFIFODBG))) { - DRM_ERROR("GT fifo had a previous error %x\n", gtfifodbg); - I915_WRITE(GTFIFODBG, gtfifodbg); - } - - gen6_gt_force_wake_get(dev_priv); - - /* disable the counters and set deterministic thresholds */ - I915_WRITE(GEN6_RC_CONTROL, 0); - - I915_WRITE(GEN6_RC1_WAKE_RATE_LIMIT, 1000 << 16); - I915_WRITE(GEN6_RC6_WAKE_RATE_LIMIT, 40 << 16 | 30); - I915_WRITE(GEN6_RC6pp_WAKE_RATE_LIMIT, 30); - I915_WRITE(GEN6_RC_EVALUATION_INTERVAL, 125000); - I915_WRITE(GEN6_RC_IDLE_HYSTERSIS, 25); - - for (i = 0; i < I915_NUM_RINGS; i++) - I915_WRITE(RING_MAX_IDLE(dev_priv->ring[i].mmio_base), 10); - - I915_WRITE(GEN6_RC_SLEEP, 0); - I915_WRITE(GEN6_RC1e_THRESHOLD, 1000); - I915_WRITE(GEN6_RC6_THRESHOLD, 50000); - I915_WRITE(GEN6_RC6p_THRESHOLD, 100000); - I915_WRITE(GEN6_RC6pp_THRESHOLD, 64000); /* unused */ - - rc6_mode = intel_enable_rc6(dev_priv->dev); - if (rc6_mode & INTEL_RC6_ENABLE) - rc6_mask |= GEN6_RC_CTL_RC6_ENABLE; - - if (rc6_mode & INTEL_RC6p_ENABLE) - rc6_mask |= GEN6_RC_CTL_RC6p_ENABLE; - - if (rc6_mode & INTEL_RC6pp_ENABLE) - rc6_mask |= GEN6_RC_CTL_RC6pp_ENABLE; - - DRM_INFO("Enabling RC6 states: RC6 %s, RC6p %s, RC6pp %s\n", - (rc6_mode & INTEL_RC6_ENABLE) ? "on" : "off", - (rc6_mode & INTEL_RC6p_ENABLE) ? "on" : "off", - (rc6_mode & INTEL_RC6pp_ENABLE) ? "on" : "off"); - - I915_WRITE(GEN6_RC_CONTROL, - rc6_mask | - GEN6_RC_CTL_EI_MODE(1) | - GEN6_RC_CTL_HW_ENABLE); - - I915_WRITE(GEN6_RPNSWREQ, - GEN6_FREQUENCY(10) | - GEN6_OFFSET(0) | - GEN6_AGGRESSIVE_TURBO); - I915_WRITE(GEN6_RC_VIDEO_FREQ, - GEN6_FREQUENCY(12)); - - I915_WRITE(GEN6_RP_DOWN_TIMEOUT, 1000000); - I915_WRITE(GEN6_RP_INTERRUPT_LIMITS, - 18 << 24 | - 6 << 16); - I915_WRITE(GEN6_RP_UP_THRESHOLD, 10000); - I915_WRITE(GEN6_RP_DOWN_THRESHOLD, 1000000); - I915_WRITE(GEN6_RP_UP_EI, 100000); - I915_WRITE(GEN6_RP_DOWN_EI, 5000000); - I915_WRITE(GEN6_RP_IDLE_HYSTERSIS, 10); - I915_WRITE(GEN6_RP_CONTROL, - GEN6_RP_MEDIA_TURBO | - GEN6_RP_MEDIA_HW_MODE | - GEN6_RP_MEDIA_IS_GFX | - GEN6_RP_ENABLE | - GEN6_RP_UP_BUSY_AVG | - GEN6_RP_DOWN_IDLE_CONT); - - if (wait_for((I915_READ(GEN6_PCODE_MAILBOX) & GEN6_PCODE_READY) == 0, - 500)) - DRM_ERROR("timeout waiting for pcode mailbox to become idle\n"); - - I915_WRITE(GEN6_PCODE_DATA, 0); - I915_WRITE(GEN6_PCODE_MAILBOX, - GEN6_PCODE_READY | - GEN6_PCODE_WRITE_MIN_FREQ_TABLE); - if (wait_for((I915_READ(GEN6_PCODE_MAILBOX) & GEN6_PCODE_READY) == 0, - 500)) - DRM_ERROR("timeout waiting for pcode mailbox to finish\n"); - - min_freq = (rp_state_cap & 0xff0000) >> 16; - max_freq = rp_state_cap & 0xff; - cur_freq = (gt_perf_status & 0xff00) >> 8; - - /* Check for overclock support */ - if (wait_for((I915_READ(GEN6_PCODE_MAILBOX) & GEN6_PCODE_READY) == 0, - 500)) - DRM_ERROR("timeout waiting for pcode mailbox to become idle\n"); - I915_WRITE(GEN6_PCODE_MAILBOX, GEN6_READ_OC_PARAMS); - pcu_mbox = I915_READ(GEN6_PCODE_DATA); - if (wait_for((I915_READ(GEN6_PCODE_MAILBOX) & GEN6_PCODE_READY) == 0, - 500)) - DRM_ERROR("timeout waiting for pcode mailbox to finish\n"); - if (pcu_mbox & (1<<31)) { /* OC supported */ - max_freq = pcu_mbox & 0xff; - DRM_DEBUG_DRIVER("overclocking supported, adjusting frequency max to %dMHz\n", pcu_mbox * 50); - } - - /* In units of 100MHz */ - dev_priv->max_delay = max_freq; - dev_priv->min_delay = min_freq; - dev_priv->cur_delay = cur_freq; - - /* requires MSI enabled */ - I915_WRITE(GEN6_PMIER, - GEN6_PM_MBOX_EVENT | - GEN6_PM_THERMAL_EVENT | - GEN6_PM_RP_DOWN_TIMEOUT | - GEN6_PM_RP_UP_THRESHOLD | - GEN6_PM_RP_DOWN_THRESHOLD | - GEN6_PM_RP_UP_EI_EXPIRED | - GEN6_PM_RP_DOWN_EI_EXPIRED); - spin_lock_irq(&dev_priv->rps_lock); - WARN_ON(dev_priv->pm_iir != 0); - I915_WRITE(GEN6_PMIMR, 0); - spin_unlock_irq(&dev_priv->rps_lock); - /* enable all PM interrupts */ - I915_WRITE(GEN6_PMINTRMSK, 0); - - gen6_gt_force_wake_put(dev_priv); - mutex_unlock(&dev_priv->dev->struct_mutex); -} - -void gen6_update_ring_freq(struct drm_i915_private *dev_priv) -{ - int min_freq = 15; - int gpu_freq, ia_freq, max_ia_freq; - int scaling_factor = 180; - - max_ia_freq = cpufreq_quick_get_max(0); - /* - * Default to measured freq if none found, PCU will ensure we don't go - * over - */ - if (!max_ia_freq) - max_ia_freq = tsc_khz; - - /* Convert from kHz to MHz */ - max_ia_freq /= 1000; - - mutex_lock(&dev_priv->dev->struct_mutex); - - /* - * For each potential GPU frequency, load a ring frequency we'd like - * to use for memory access. We do this by specifying the IA frequency - * the PCU should use as a reference to determine the ring frequency. - */ - for (gpu_freq = dev_priv->max_delay; gpu_freq >= dev_priv->min_delay; - gpu_freq--) { - int diff = dev_priv->max_delay - gpu_freq; - - /* - * For GPU frequencies less than 750MHz, just use the lowest - * ring freq. - */ - if (gpu_freq < min_freq) - ia_freq = 800; - else - ia_freq = max_ia_freq - ((diff * scaling_factor) / 2); - ia_freq = DIV_ROUND_CLOSEST(ia_freq, 100); - - I915_WRITE(GEN6_PCODE_DATA, - (ia_freq << GEN6_PCODE_FREQ_IA_RATIO_SHIFT) | - gpu_freq); - I915_WRITE(GEN6_PCODE_MAILBOX, GEN6_PCODE_READY | - GEN6_PCODE_WRITE_MIN_FREQ_TABLE); - if (wait_for((I915_READ(GEN6_PCODE_MAILBOX) & - GEN6_PCODE_READY) == 0, 10)) { - DRM_ERROR("pcode write of freq table timed out\n"); - continue; - } - } - - mutex_unlock(&dev_priv->dev->struct_mutex); -} - -static void ironlake_init_clock_gating(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - uint32_t dspclk_gate = VRHUNIT_CLOCK_GATE_DISABLE; - - /* Required for FBC */ - dspclk_gate |= DPFCUNIT_CLOCK_GATE_DISABLE | - DPFCRUNIT_CLOCK_GATE_DISABLE | - DPFDUNIT_CLOCK_GATE_DISABLE; - /* Required for CxSR */ - dspclk_gate |= DPARBUNIT_CLOCK_GATE_DISABLE; - - I915_WRITE(PCH_3DCGDIS0, - MARIUNIT_CLOCK_GATE_DISABLE | - SVSMUNIT_CLOCK_GATE_DISABLE); - I915_WRITE(PCH_3DCGDIS1, - VFMUNIT_CLOCK_GATE_DISABLE); - - I915_WRITE(PCH_DSPCLK_GATE_D, dspclk_gate); - - /* - * According to the spec the following bits should be set in - * order to enable memory self-refresh - * The bit 22/21 of 0x42004 - * The bit 5 of 0x42020 - * The bit 15 of 0x45000 - */ - I915_WRITE(ILK_DISPLAY_CHICKEN2, - (I915_READ(ILK_DISPLAY_CHICKEN2) | - ILK_DPARB_GATE | ILK_VSDPFD_FULL)); - I915_WRITE(ILK_DSPCLK_GATE, - (I915_READ(ILK_DSPCLK_GATE) | - ILK_DPARB_CLK_GATE)); - I915_WRITE(DISP_ARB_CTL, - (I915_READ(DISP_ARB_CTL) | - DISP_FBC_WM_DIS)); - I915_WRITE(WM3_LP_ILK, 0); - I915_WRITE(WM2_LP_ILK, 0); - I915_WRITE(WM1_LP_ILK, 0); - - /* - * Based on the document from hardware guys the following bits - * should be set unconditionally in order to enable FBC. - * The bit 22 of 0x42000 - * The bit 22 of 0x42004 - * The bit 7,8,9 of 0x42020. - */ - if (IS_IRONLAKE_M(dev)) { - I915_WRITE(ILK_DISPLAY_CHICKEN1, - I915_READ(ILK_DISPLAY_CHICKEN1) | - ILK_FBCQ_DIS); - I915_WRITE(ILK_DISPLAY_CHICKEN2, - I915_READ(ILK_DISPLAY_CHICKEN2) | - ILK_DPARB_GATE); - I915_WRITE(ILK_DSPCLK_GATE, - I915_READ(ILK_DSPCLK_GATE) | - ILK_DPFC_DIS1 | - ILK_DPFC_DIS2 | - ILK_CLK_FBC); - } - - I915_WRITE(ILK_DISPLAY_CHICKEN2, - I915_READ(ILK_DISPLAY_CHICKEN2) | - ILK_ELPIN_409_SELECT); - I915_WRITE(_3D_CHICKEN2, - _3D_CHICKEN2_WM_READ_PIPELINED << 16 | - _3D_CHICKEN2_WM_READ_PIPELINED); -} - -static void gen6_init_clock_gating(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - int pipe; - uint32_t dspclk_gate = VRHUNIT_CLOCK_GATE_DISABLE; - - I915_WRITE(PCH_DSPCLK_GATE_D, dspclk_gate); - - I915_WRITE(ILK_DISPLAY_CHICKEN2, - I915_READ(ILK_DISPLAY_CHICKEN2) | - ILK_ELPIN_409_SELECT); - - I915_WRITE(WM3_LP_ILK, 0); - I915_WRITE(WM2_LP_ILK, 0); - I915_WRITE(WM1_LP_ILK, 0); - - /* According to the BSpec vol1g, bit 12 (RCPBUNIT) clock - * gating disable must be set. Failure to set it results in - * flickering pixels due to Z write ordering failures after - * some amount of runtime in the Mesa "fire" demo, and Unigine - * Sanctuary and Tropics, and apparently anything else with - * alpha test or pixel discard. - * - * According to the spec, bit 11 (RCCUNIT) must also be set, - * but we didn't debug actual testcases to find it out. - */ - I915_WRITE(GEN6_UCGCTL2, - GEN6_RCPBUNIT_CLOCK_GATE_DISABLE | - GEN6_RCCUNIT_CLOCK_GATE_DISABLE); - - /* - * According to the spec the following bits should be - * set in order to enable memory self-refresh and fbc: - * The bit21 and bit22 of 0x42000 - * The bit21 and bit22 of 0x42004 - * The bit5 and bit7 of 0x42020 - * The bit14 of 0x70180 - * The bit14 of 0x71180 - */ - I915_WRITE(ILK_DISPLAY_CHICKEN1, - I915_READ(ILK_DISPLAY_CHICKEN1) | - ILK_FBCQ_DIS | ILK_PABSTRETCH_DIS); - I915_WRITE(ILK_DISPLAY_CHICKEN2, - I915_READ(ILK_DISPLAY_CHICKEN2) | - ILK_DPARB_GATE | ILK_VSDPFD_FULL); - I915_WRITE(ILK_DSPCLK_GATE, - I915_READ(ILK_DSPCLK_GATE) | - ILK_DPARB_CLK_GATE | - ILK_DPFD_CLK_GATE); - - for_each_pipe(pipe) { - I915_WRITE(DSPCNTR(pipe), - I915_READ(DSPCNTR(pipe)) | - DISPPLANE_TRICKLE_FEED_DISABLE); - intel_flush_display_plane(dev_priv, pipe); - } -} - -static void ivybridge_init_clock_gating(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - int pipe; - uint32_t dspclk_gate = VRHUNIT_CLOCK_GATE_DISABLE; - - I915_WRITE(PCH_DSPCLK_GATE_D, dspclk_gate); - - I915_WRITE(WM3_LP_ILK, 0); - I915_WRITE(WM2_LP_ILK, 0); - I915_WRITE(WM1_LP_ILK, 0); - - /* According to the spec, bit 13 (RCZUNIT) must be set on IVB. - * This implements the WaDisableRCZUnitClockGating workaround. - */ - I915_WRITE(GEN6_UCGCTL2, GEN6_RCZUNIT_CLOCK_GATE_DISABLE); - - I915_WRITE(ILK_DSPCLK_GATE, IVB_VRHUNIT_CLK_GATE); - - I915_WRITE(IVB_CHICKEN3, - CHICKEN3_DGMG_REQ_OUT_FIX_DISABLE | - CHICKEN3_DGMG_DONE_FIX_DISABLE); - - /* Apply the WaDisableRHWOOptimizationForRenderHang workaround. */ - I915_WRITE(GEN7_COMMON_SLICE_CHICKEN1, - GEN7_CSC1_RHWO_OPT_DISABLE_IN_RCC); - - /* WaApplyL3ControlAndL3ChickenMode requires those two on Ivy Bridge */ - I915_WRITE(GEN7_L3CNTLREG1, - GEN7_WA_FOR_GEN7_L3_CONTROL); - I915_WRITE(GEN7_L3_CHICKEN_MODE_REGISTER, - GEN7_WA_L3_CHICKEN_MODE); - - /* This is required by WaCatErrorRejectionIssue */ - I915_WRITE(GEN7_SQ_CHICKEN_MBCUNIT_CONFIG, - I915_READ(GEN7_SQ_CHICKEN_MBCUNIT_CONFIG) | - GEN7_SQ_CHICKEN_MBCUNIT_SQINTMOB); - - for_each_pipe(pipe) { - I915_WRITE(DSPCNTR(pipe), - I915_READ(DSPCNTR(pipe)) | - DISPPLANE_TRICKLE_FEED_DISABLE); - intel_flush_display_plane(dev_priv, pipe); - } -} - -static void valleyview_init_clock_gating(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - int pipe; - uint32_t dspclk_gate = VRHUNIT_CLOCK_GATE_DISABLE; - - I915_WRITE(PCH_DSPCLK_GATE_D, dspclk_gate); - - I915_WRITE(WM3_LP_ILK, 0); - I915_WRITE(WM2_LP_ILK, 0); - I915_WRITE(WM1_LP_ILK, 0); - - /* According to the spec, bit 13 (RCZUNIT) must be set on IVB. - * This implements the WaDisableRCZUnitClockGating workaround. - */ - I915_WRITE(GEN6_UCGCTL2, GEN6_RCZUNIT_CLOCK_GATE_DISABLE); - - I915_WRITE(ILK_DSPCLK_GATE, IVB_VRHUNIT_CLK_GATE); - - I915_WRITE(IVB_CHICKEN3, - CHICKEN3_DGMG_REQ_OUT_FIX_DISABLE | - CHICKEN3_DGMG_DONE_FIX_DISABLE); - - /* Apply the WaDisableRHWOOptimizationForRenderHang workaround. */ - I915_WRITE(GEN7_COMMON_SLICE_CHICKEN1, - GEN7_CSC1_RHWO_OPT_DISABLE_IN_RCC); - - /* WaApplyL3ControlAndL3ChickenMode requires those two on Ivy Bridge */ - I915_WRITE(GEN7_L3CNTLREG1, GEN7_WA_FOR_GEN7_L3_CONTROL); - I915_WRITE(GEN7_L3_CHICKEN_MODE_REGISTER, GEN7_WA_L3_CHICKEN_MODE); - - /* This is required by WaCatErrorRejectionIssue */ - I915_WRITE(GEN7_SQ_CHICKEN_MBCUNIT_CONFIG, - I915_READ(GEN7_SQ_CHICKEN_MBCUNIT_CONFIG) | - GEN7_SQ_CHICKEN_MBCUNIT_SQINTMOB); - - for_each_pipe(pipe) { - I915_WRITE(DSPCNTR(pipe), - I915_READ(DSPCNTR(pipe)) | - DISPPLANE_TRICKLE_FEED_DISABLE); - intel_flush_display_plane(dev_priv, pipe); - } - - I915_WRITE(CACHE_MODE_1, I915_READ(CACHE_MODE_1) | - (PIXEL_SUBSPAN_COLLECT_OPT_DISABLE << 16) | - PIXEL_SUBSPAN_COLLECT_OPT_DISABLE); -} - -static void g4x_init_clock_gating(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - uint32_t dspclk_gate; - - I915_WRITE(RENCLK_GATE_D1, 0); - I915_WRITE(RENCLK_GATE_D2, VF_UNIT_CLOCK_GATE_DISABLE | - GS_UNIT_CLOCK_GATE_DISABLE | - CL_UNIT_CLOCK_GATE_DISABLE); - I915_WRITE(RAMCLK_GATE_D, 0); - dspclk_gate = VRHUNIT_CLOCK_GATE_DISABLE | - OVRUNIT_CLOCK_GATE_DISABLE | - OVCUNIT_CLOCK_GATE_DISABLE; - if (IS_GM45(dev)) - dspclk_gate |= DSSUNIT_CLOCK_GATE_DISABLE; - I915_WRITE(DSPCLK_GATE_D, dspclk_gate); -} - -static void crestline_init_clock_gating(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - - I915_WRITE(RENCLK_GATE_D1, I965_RCC_CLOCK_GATE_DISABLE); - I915_WRITE(RENCLK_GATE_D2, 0); - I915_WRITE(DSPCLK_GATE_D, 0); - I915_WRITE(RAMCLK_GATE_D, 0); - I915_WRITE16(DEUC, 0); -} - -static void broadwater_init_clock_gating(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - - I915_WRITE(RENCLK_GATE_D1, I965_RCZ_CLOCK_GATE_DISABLE | - I965_RCC_CLOCK_GATE_DISABLE | - I965_RCPB_CLOCK_GATE_DISABLE | - I965_ISC_CLOCK_GATE_DISABLE | - I965_FBC_CLOCK_GATE_DISABLE); - I915_WRITE(RENCLK_GATE_D2, 0); -} - -static void gen3_init_clock_gating(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - u32 dstate = I915_READ(D_STATE); - - dstate |= DSTATE_PLL_D3_OFF | DSTATE_GFX_CLOCK_GATING | - DSTATE_DOT_CLOCK_GATING; - I915_WRITE(D_STATE, dstate); -} - -static void i85x_init_clock_gating(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - - I915_WRITE(RENCLK_GATE_D1, SV_CLOCK_GATE_DISABLE); -} - -static void i830_init_clock_gating(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - - I915_WRITE(DSPCLK_GATE_D, OVRUNIT_CLOCK_GATE_DISABLE); -} - -static void ibx_init_clock_gating(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - - /* - * On Ibex Peak and Cougar Point, we need to disable clock - * gating for the panel power sequencer or it will fail to - * start up when no ports are active. - */ - I915_WRITE(SOUTH_DSPCLK_GATE_D, PCH_DPLSUNIT_CLOCK_GATE_DISABLE); -} - -static void cpt_init_clock_gating(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - int pipe; - - /* - * On Ibex Peak and Cougar Point, we need to disable clock - * gating for the panel power sequencer or it will fail to - * start up when no ports are active. - */ - I915_WRITE(SOUTH_DSPCLK_GATE_D, PCH_DPLSUNIT_CLOCK_GATE_DISABLE); - I915_WRITE(SOUTH_CHICKEN2, I915_READ(SOUTH_CHICKEN2) | - DPLS_EDP_PPS_FIX_DIS); - /* Without this, mode sets may fail silently on FDI */ - for_each_pipe(pipe) - I915_WRITE(TRANS_CHICKEN2(pipe), TRANS_AUTOTRAIN_GEN_STALL_DIS); -} - -static void ironlake_teardown_rc6(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - - if (dev_priv->renderctx) { - i915_gem_object_unpin(dev_priv->renderctx); - drm_gem_object_unreference(&dev_priv->renderctx->base); - dev_priv->renderctx = NULL; - } - - if (dev_priv->pwrctx) { - i915_gem_object_unpin(dev_priv->pwrctx); - drm_gem_object_unreference(&dev_priv->pwrctx->base); - dev_priv->pwrctx = NULL; - } -} - -static void ironlake_disable_rc6(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - - if (I915_READ(PWRCTXA)) { - /* Wake the GPU, prevent RC6, then restore RSTDBYCTL */ - I915_WRITE(RSTDBYCTL, I915_READ(RSTDBYCTL) | RCX_SW_EXIT); - wait_for(((I915_READ(RSTDBYCTL) & RSX_STATUS_MASK) == RSX_STATUS_ON), - 50); - - I915_WRITE(PWRCTXA, 0); - POSTING_READ(PWRCTXA); - - I915_WRITE(RSTDBYCTL, I915_READ(RSTDBYCTL) & ~RCX_SW_EXIT); - POSTING_READ(RSTDBYCTL); - } - - ironlake_teardown_rc6(dev); -} - -static int ironlake_setup_rc6(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - - if (dev_priv->renderctx == NULL) - dev_priv->renderctx = intel_alloc_context_page(dev); - if (!dev_priv->renderctx) - return -ENOMEM; - - if (dev_priv->pwrctx == NULL) - dev_priv->pwrctx = intel_alloc_context_page(dev); - if (!dev_priv->pwrctx) { - ironlake_teardown_rc6(dev); - return -ENOMEM; - } - - return 0; -} - -void ironlake_enable_rc6(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - int ret; - - /* rc6 disabled by default due to repeated reports of hanging during - * boot and resume. - */ - if (!intel_enable_rc6(dev)) - return; - - mutex_lock(&dev->struct_mutex); - ret = ironlake_setup_rc6(dev); - if (ret) { - mutex_unlock(&dev->struct_mutex); - return; - } - - /* - * GPU can automatically power down the render unit if given a page - * to save state. - */ - ret = BEGIN_LP_RING(6); - if (ret) { - ironlake_teardown_rc6(dev); - mutex_unlock(&dev->struct_mutex); - return; - } - - OUT_RING(MI_SUSPEND_FLUSH | MI_SUSPEND_FLUSH_EN); - OUT_RING(MI_SET_CONTEXT); - OUT_RING(dev_priv->renderctx->gtt_offset | - MI_MM_SPACE_GTT | - MI_SAVE_EXT_STATE_EN | - MI_RESTORE_EXT_STATE_EN | - MI_RESTORE_INHIBIT); - OUT_RING(MI_SUSPEND_FLUSH); - OUT_RING(MI_NOOP); - OUT_RING(MI_FLUSH); - ADVANCE_LP_RING(); - - /* - * Wait for the command parser to advance past MI_SET_CONTEXT. The HW - * does an implicit flush, combined with MI_FLUSH above, it should be - * safe to assume that renderctx is valid - */ - ret = intel_wait_ring_idle(LP_RING(dev_priv)); - if (ret) { - DRM_ERROR("failed to enable ironlake power power savings\n"); - ironlake_teardown_rc6(dev); - mutex_unlock(&dev->struct_mutex); - return; - } - - I915_WRITE(PWRCTXA, dev_priv->pwrctx->gtt_offset | PWRCTX_EN); - I915_WRITE(RSTDBYCTL, I915_READ(RSTDBYCTL) & ~RCX_SW_EXIT); - mutex_unlock(&dev->struct_mutex); -} - -void intel_init_clock_gating(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - - dev_priv->display.init_clock_gating(dev); - - if (dev_priv->display.init_pch_clock_gating) - dev_priv->display.init_pch_clock_gating(dev); -} - /* Set up chip specific display functions */ static void intel_init_display(struct drm_device *dev) { @@ -9228,23 +6344,6 @@ static void intel_init_display(struct drm_device *dev) dev_priv->display.update_plane = i9xx_update_plane; } - if (I915_HAS_FBC(dev)) { - if (HAS_PCH_SPLIT(dev)) { - dev_priv->display.fbc_enabled = ironlake_fbc_enabled; - dev_priv->display.enable_fbc = ironlake_enable_fbc; - dev_priv->display.disable_fbc = ironlake_disable_fbc; - } else if (IS_GM45(dev)) { - dev_priv->display.fbc_enabled = g4x_fbc_enabled; - dev_priv->display.enable_fbc = g4x_enable_fbc; - dev_priv->display.disable_fbc = g4x_disable_fbc; - } else if (IS_CRESTLINE(dev)) { - dev_priv->display.fbc_enabled = i8xx_fbc_enabled; - dev_priv->display.enable_fbc = i8xx_enable_fbc; - dev_priv->display.disable_fbc = i8xx_disable_fbc; - } - /* 855GM needs testing */ - } - /* Returns the core display clock speed */ if (IS_VALLEYVIEW(dev)) dev_priv->display.get_display_clock_speed = @@ -9271,130 +6370,24 @@ static void intel_init_display(struct drm_device *dev) dev_priv->display.get_display_clock_speed = i830_get_display_clock_speed; - /* For FIFO watermark updates */ if (HAS_PCH_SPLIT(dev)) { - dev_priv->display.force_wake_get = __gen6_gt_force_wake_get; - dev_priv->display.force_wake_put = __gen6_gt_force_wake_put; - - /* IVB configs may use multi-threaded forcewake */ - if (IS_IVYBRIDGE(dev)) { - u32 ecobus; - - /* A small trick here - if the bios hasn't configured MT forcewake, - * and if the device is in RC6, then force_wake_mt_get will not wake - * the device and the ECOBUS read will return zero. Which will be - * (correctly) interpreted by the test below as MT forcewake being - * disabled. - */ - mutex_lock(&dev->struct_mutex); - __gen6_gt_force_wake_mt_get(dev_priv); - ecobus = I915_READ_NOTRACE(ECOBUS); - __gen6_gt_force_wake_mt_put(dev_priv); - mutex_unlock(&dev->struct_mutex); - - if (ecobus & FORCEWAKE_MT_ENABLE) { - DRM_DEBUG_KMS("Using MT version of forcewake\n"); - dev_priv->display.force_wake_get = - __gen6_gt_force_wake_mt_get; - dev_priv->display.force_wake_put = - __gen6_gt_force_wake_mt_put; - } - } - - if (HAS_PCH_IBX(dev)) - dev_priv->display.init_pch_clock_gating = ibx_init_clock_gating; - else if (HAS_PCH_CPT(dev)) - dev_priv->display.init_pch_clock_gating = cpt_init_clock_gating; - if (IS_GEN5(dev)) { - if (I915_READ(MLTR_ILK) & ILK_SRLT_MASK) - dev_priv->display.update_wm = ironlake_update_wm; - else { - DRM_DEBUG_KMS("Failed to get proper latency. " - "Disable CxSR\n"); - dev_priv->display.update_wm = NULL; - } dev_priv->display.fdi_link_train = ironlake_fdi_link_train; - dev_priv->display.init_clock_gating = ironlake_init_clock_gating; dev_priv->display.write_eld = ironlake_write_eld; } else if (IS_GEN6(dev)) { - if (SNB_READ_WM0_LATENCY()) { - dev_priv->display.update_wm = sandybridge_update_wm; - dev_priv->display.update_sprite_wm = sandybridge_update_sprite_wm; - } else { - DRM_DEBUG_KMS("Failed to read display plane latency. " - "Disable CxSR\n"); - dev_priv->display.update_wm = NULL; - } dev_priv->display.fdi_link_train = gen6_fdi_link_train; - dev_priv->display.init_clock_gating = gen6_init_clock_gating; dev_priv->display.write_eld = ironlake_write_eld; } else if (IS_IVYBRIDGE(dev)) { /* FIXME: detect B0+ stepping and use auto training */ dev_priv->display.fdi_link_train = ivb_manual_fdi_link_train; - if (SNB_READ_WM0_LATENCY()) { - dev_priv->display.update_wm = sandybridge_update_wm; - dev_priv->display.update_sprite_wm = sandybridge_update_sprite_wm; - } else { - DRM_DEBUG_KMS("Failed to read display plane latency. " - "Disable CxSR\n"); - dev_priv->display.update_wm = NULL; - } - dev_priv->display.init_clock_gating = ivybridge_init_clock_gating; dev_priv->display.write_eld = ironlake_write_eld; } else dev_priv->display.update_wm = NULL; } else if (IS_VALLEYVIEW(dev)) { - dev_priv->display.update_wm = valleyview_update_wm; - dev_priv->display.init_clock_gating = - valleyview_init_clock_gating; dev_priv->display.force_wake_get = vlv_force_wake_get; dev_priv->display.force_wake_put = vlv_force_wake_put; - } else if (IS_PINEVIEW(dev)) { - if (!intel_get_cxsr_latency(IS_PINEVIEW_G(dev), - dev_priv->is_ddr3, - dev_priv->fsb_freq, - dev_priv->mem_freq)) { - DRM_INFO("failed to find known CxSR latency " - "(found ddr%s fsb freq %d, mem freq %d), " - "disabling CxSR\n", - (dev_priv->is_ddr3 == 1) ? "3" : "2", - dev_priv->fsb_freq, dev_priv->mem_freq); - /* Disable CxSR and never update its watermark again */ - pineview_disable_cxsr(dev); - dev_priv->display.update_wm = NULL; - } else - dev_priv->display.update_wm = pineview_update_wm; - dev_priv->display.init_clock_gating = gen3_init_clock_gating; } else if (IS_G4X(dev)) { dev_priv->display.write_eld = g4x_write_eld; - dev_priv->display.update_wm = g4x_update_wm; - dev_priv->display.init_clock_gating = g4x_init_clock_gating; - } else if (IS_GEN4(dev)) { - dev_priv->display.update_wm = i965_update_wm; - if (IS_CRESTLINE(dev)) - dev_priv->display.init_clock_gating = crestline_init_clock_gating; - else if (IS_BROADWATER(dev)) - dev_priv->display.init_clock_gating = broadwater_init_clock_gating; - } else if (IS_GEN3(dev)) { - dev_priv->display.update_wm = i9xx_update_wm; - dev_priv->display.get_fifo_size = i9xx_get_fifo_size; - dev_priv->display.init_clock_gating = gen3_init_clock_gating; - } else if (IS_I865G(dev)) { - dev_priv->display.update_wm = i830_update_wm; - dev_priv->display.init_clock_gating = i85x_init_clock_gating; - dev_priv->display.get_fifo_size = i830_get_fifo_size; - } else if (IS_I85X(dev)) { - dev_priv->display.update_wm = i9xx_update_wm; - dev_priv->display.get_fifo_size = i85x_get_fifo_size; - dev_priv->display.init_clock_gating = i85x_init_clock_gating; - } else { - dev_priv->display.update_wm = i830_update_wm; - dev_priv->display.init_clock_gating = i830_init_clock_gating; - if (IS_845G(dev)) - dev_priv->display.get_fifo_size = i845_get_fifo_size; - else - dev_priv->display.get_fifo_size = i830_get_fifo_size; } /* Default just returns -ENODEV to indicate unsupported */ @@ -9464,7 +6457,7 @@ struct intel_quirk { void (*hook)(struct drm_device *dev); }; -struct intel_quirk intel_quirks[] = { +static struct intel_quirk intel_quirks[] = { /* HP Mini needs pipe A force quirk (LP: #322104) */ { 0x27ae, 0x103c, 0x361a, quirk_pipea_force }, @@ -9524,7 +6517,7 @@ static void i915_disable_vga(struct drm_device *dev) vga_reg = VGACNTRL; vga_get_uninterruptible(dev->pdev, VGA_RSRC_LEGACY_IO); - outb(1, VGA_SR_INDEX); + outb(SR01, VGA_SR_INDEX); sr1 = inb(VGA_SR_DATA); outb(sr1 | 1<<5, VGA_SR_DATA); vga_put(dev->pdev, VGA_RSRC_LEGACY_IO); @@ -9534,6 +6527,39 @@ static void i915_disable_vga(struct drm_device *dev) POSTING_READ(vga_reg); } +static void ivb_pch_pwm_override(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + + /* + * IVB has CPU eDP backlight regs too, set things up to let the + * PCH regs control the backlight + */ + I915_WRITE(BLC_PWM_CPU_CTL2, PWM_ENABLE); + I915_WRITE(BLC_PWM_CPU_CTL, 0); + I915_WRITE(BLC_PWM_PCH_CTL1, PWM_ENABLE | (1<<30)); +} + +void intel_modeset_init_hw(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + + intel_init_clock_gating(dev); + + if (IS_IRONLAKE_M(dev)) { + ironlake_enable_drps(dev); + intel_init_emon(dev); + } + + if ((IS_GEN6(dev) || IS_GEN7(dev)) && !IS_VALLEYVIEW(dev)) { + gen6_enable_rps(dev_priv); + gen6_update_ring_freq(dev_priv); + } + + if (IS_IVYBRIDGE(dev)) + ivb_pch_pwm_override(dev); +} + void intel_modeset_init(struct drm_device *dev) { struct drm_i915_private *dev_priv = dev->dev_private; @@ -9551,6 +6577,8 @@ void intel_modeset_init(struct drm_device *dev) intel_init_quirks(dev); + intel_init_pm(dev); + intel_init_display(dev); if (IS_GEN2(dev)) { @@ -9579,17 +6607,7 @@ void intel_modeset_init(struct drm_device *dev) i915_disable_vga(dev); intel_setup_outputs(dev); - intel_init_clock_gating(dev); - - if (IS_IRONLAKE_M(dev)) { - ironlake_enable_drps(dev); - intel_init_emon(dev); - } - - if (IS_GEN6(dev) || IS_GEN7(dev)) { - gen6_enable_rps(dev_priv); - gen6_update_ring_freq(dev_priv); - } + intel_modeset_init_hw(dev); INIT_WORK(&dev_priv->idle_work, intel_idle_update); setup_timer(&dev_priv->idle_timer, intel_gpu_idle_timer, @@ -9629,7 +6647,7 @@ void intel_modeset_cleanup(struct drm_device *dev) if (IS_IRONLAKE_M(dev)) ironlake_disable_drps(dev); - if (IS_GEN6(dev) || IS_GEN7(dev)) + if ((IS_GEN6(dev) || IS_GEN7(dev)) && !IS_VALLEYVIEW(dev)) gen6_disable_rps(dev); if (IS_IRONLAKE_M(dev)) diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index 110552f..44cf32c 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -219,14 +219,38 @@ intel_dp_max_data_rate(int max_link_clock, int max_lanes) return (max_link_clock * max_lanes * 8) / 10; } +static bool +intel_dp_adjust_dithering(struct intel_dp *intel_dp, + struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode) +{ + int max_link_clock = intel_dp_link_clock(intel_dp_max_link_bw(intel_dp)); + int max_lanes = intel_dp_max_lane_count(intel_dp); + int max_rate, mode_rate; + + mode_rate = intel_dp_link_required(mode->clock, 24); + max_rate = intel_dp_max_data_rate(max_link_clock, max_lanes); + + if (mode_rate > max_rate) { + mode_rate = intel_dp_link_required(mode->clock, 18); + if (mode_rate > max_rate) + return false; + + if (adjusted_mode) + adjusted_mode->private_flags + |= INTEL_MODE_DP_FORCE_6BPC; + + return true; + } + + return true; +} + static int intel_dp_mode_valid(struct drm_connector *connector, struct drm_display_mode *mode) { struct intel_dp *intel_dp = intel_attached_dp(connector); - int max_link_clock = intel_dp_link_clock(intel_dp_max_link_bw(intel_dp)); - int max_lanes = intel_dp_max_lane_count(intel_dp); - int max_rate, mode_rate; if (is_edp(intel_dp) && intel_dp->panel_fixed_mode) { if (mode->hdisplay > intel_dp->panel_fixed_mode->hdisplay) @@ -236,16 +260,8 @@ intel_dp_mode_valid(struct drm_connector *connector, return MODE_PANEL; } - mode_rate = intel_dp_link_required(mode->clock, 24); - max_rate = intel_dp_max_data_rate(max_link_clock, max_lanes); - - if (mode_rate > max_rate) { - mode_rate = intel_dp_link_required(mode->clock, 18); - if (mode_rate > max_rate) - return MODE_CLOCK_HIGH; - else - mode->private_flags |= INTEL_MODE_DP_FORCE_6BPC; - } + if (!intel_dp_adjust_dithering(intel_dp, mode, NULL)) + return MODE_CLOCK_HIGH; if (mode->clock < 10000) return MODE_CLOCK_LOW; @@ -672,7 +688,7 @@ intel_dp_mode_fixup(struct drm_encoder *encoder, struct drm_display_mode *mode, int lane_count, clock; int max_lane_count = intel_dp_max_lane_count(intel_dp); int max_clock = intel_dp_max_link_bw(intel_dp) == DP_LINK_BW_2_7 ? 1 : 0; - int bpp = mode->private_flags & INTEL_MODE_DP_FORCE_6BPC ? 18 : 24; + int bpp; static int bws[2] = { DP_LINK_BW_1_62, DP_LINK_BW_2_7 }; if (is_edp(intel_dp) && intel_dp->panel_fixed_mode) { @@ -686,6 +702,11 @@ intel_dp_mode_fixup(struct drm_encoder *encoder, struct drm_display_mode *mode, mode->clock = intel_dp->panel_fixed_mode->clock; } + if (!intel_dp_adjust_dithering(intel_dp, mode, adjusted_mode)) + return false; + + bpp = adjusted_mode->private_flags & INTEL_MODE_DP_FORCE_6BPC ? 18 : 24; + for (lane_count = 1; lane_count <= max_lane_count; lane_count <<= 1) { for (clock = 0; clock <= max_clock; clock++) { int link_avail = intel_dp_max_data_rate(intel_dp_link_clock(bws[clock]), lane_count); @@ -1128,6 +1149,7 @@ static void ironlake_edp_panel_off(struct intel_dp *intel_dp) DRM_DEBUG_KMS("Turn eDP power off\n"); WARN(intel_dp->want_panel_vdd, "Cannot turn power off while VDD is on\n"); + ironlake_panel_vdd_off_sync(intel_dp); /* finish any pending work */ pp = ironlake_get_pp_control(dev_priv); pp &= ~(POWER_TARGET_ON | EDP_FORCE_VDD | PANEL_POWER_RESET | EDP_BLC_ENABLE); @@ -2462,6 +2484,13 @@ intel_dp_init(struct drm_device *dev, int output_reg) pp_off = I915_READ(PCH_PP_OFF_DELAYS); pp_div = I915_READ(PCH_PP_DIVISOR); + if (!pp_on || !pp_off || !pp_div) { + DRM_INFO("bad panel power sequencing delays, disabling panel\n"); + intel_dp_encoder_destroy(&intel_dp->base.base); + intel_dp_destroy(&intel_connector->base); + return; + } + /* Pull timing values out of registers */ cur.t1_t3 = (pp_on & PANEL_POWER_UP_DELAY_MASK) >> PANEL_POWER_UP_DELAY_SHIFT; diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h index 79cabf5..c5bf8be 100644 --- a/drivers/gpu/drm/i915/intel_drv.h +++ b/drivers/gpu/drm/i915/intel_drv.h @@ -204,6 +204,25 @@ struct intel_plane { struct drm_intel_sprite_colorkey *key); }; +struct intel_watermark_params { + unsigned long fifo_size; + unsigned long max_wm; + unsigned long default_wm; + unsigned long guard_size; + unsigned long cacheline_size; +}; + +struct cxsr_latency { + int is_desktop; + int is_ddr3; + unsigned long fsb_freq; + unsigned long mem_freq; + unsigned long display_sr; + unsigned long display_hpll_disable; + unsigned long cursor_sr; + unsigned long cursor_hpll_disable; +}; + #define to_intel_crtc(x) container_of(x, struct intel_crtc, base) #define to_intel_connector(x) container_of(x, struct intel_connector, base) #define to_intel_encoder(x) container_of(x, struct intel_encoder, base) @@ -320,6 +339,8 @@ extern bool intel_dpd_is_edp(struct drm_device *dev); extern void intel_edp_link_config(struct intel_encoder *, int *, int *); extern bool intel_encoder_is_pch_edp(struct drm_encoder *encoder); extern int intel_plane_init(struct drm_device *dev, enum pipe pipe); +extern void intel_flush_display_plane(struct drm_i915_private *dev_priv, + enum plane plane); /* intel_panel.c */ extern void intel_fixed_panel_mode(struct drm_display_mode *fixed_mode, @@ -377,6 +398,7 @@ extern void intel_crtc_fb_gamma_set(struct drm_crtc *crtc, u16 red, u16 green, extern void intel_crtc_fb_gamma_get(struct drm_crtc *crtc, u16 *red, u16 *green, u16 *blue, int regno); extern void intel_enable_clock_gating(struct drm_device *dev); +extern void ironlake_disable_rc6(struct drm_device *dev); extern void ironlake_enable_drps(struct drm_device *dev); extern void ironlake_disable_drps(struct drm_device *dev); extern void gen6_enable_rps(struct drm_i915_private *dev_priv); @@ -422,7 +444,7 @@ extern void intel_write_eld(struct drm_encoder *encoder, extern void intel_cpt_verify_modeset(struct drm_device *dev, int pipe); /* For use by IVB LP watermark workaround in intel_sprite.c */ -extern void sandybridge_update_wm(struct drm_device *dev); +extern void intel_update_watermarks(struct drm_device *dev); extern void intel_update_sprite_watermarks(struct drm_device *dev, int pipe, uint32_t sprite_width, int pixel_size); @@ -434,4 +456,11 @@ extern int intel_sprite_get_colorkey(struct drm_device *dev, void *data, extern u32 intel_dpio_read(struct drm_i915_private *dev_priv, int reg); +/* Power-related functions, located in intel_pm.c */ +extern void intel_init_pm(struct drm_device *dev); +/* FBC */ +extern bool intel_fbc_enabled(struct drm_device *dev); +extern void intel_enable_fbc(struct drm_crtc *crtc, unsigned long interval); +extern void intel_update_fbc(struct drm_device *dev); + #endif /* __INTEL_DRV_H__ */ diff --git a/drivers/gpu/drm/i915/intel_fb.c b/drivers/gpu/drm/i915/intel_fb.c index 19ecd78..71ef289 100644 --- a/drivers/gpu/drm/i915/intel_fb.c +++ b/drivers/gpu/drm/i915/intel_fb.c @@ -94,7 +94,7 @@ static int intelfb_create(struct intel_fbdev *ifbdev, mutex_lock(&dev->struct_mutex); /* Flush everything out, we'll be doing GTT only from now on */ - ret = intel_pin_and_fence_fb_obj(dev, obj, false); + ret = intel_pin_and_fence_fb_obj(dev, obj, NULL); if (ret) { DRM_ERROR("failed to pin fb: %d\n", ret); goto out_unref; diff --git a/drivers/gpu/drm/i915/intel_i2c.c b/drivers/gpu/drm/i915/intel_i2c.c index c12db72..e04255e 100644 --- a/drivers/gpu/drm/i915/intel_i2c.c +++ b/drivers/gpu/drm/i915/intel_i2c.c @@ -205,27 +205,29 @@ intel_gpio_setup(struct intel_gmbus *bus, u32 pin) static int gmbus_xfer_read(struct drm_i915_private *dev_priv, struct i2c_msg *msg, - bool last) + u32 gmbus1_index) { int reg_offset = dev_priv->gpio_mmio_base; u16 len = msg->len; u8 *buf = msg->buf; I915_WRITE(GMBUS1 + reg_offset, + gmbus1_index | GMBUS_CYCLE_WAIT | - (last ? GMBUS_CYCLE_STOP : 0) | (len << GMBUS_BYTE_COUNT_SHIFT) | (msg->addr << GMBUS_SLAVE_ADDR_SHIFT) | GMBUS_SLAVE_READ | GMBUS_SW_RDY); - POSTING_READ(GMBUS2 + reg_offset); - do { + while (len) { + int ret; u32 val, loop = 0; + u32 gmbus2; - if (wait_for(I915_READ(GMBUS2 + reg_offset) & - (GMBUS_SATOER | GMBUS_HW_RDY), - 50)) + ret = wait_for((gmbus2 = I915_READ(GMBUS2 + reg_offset)) & + (GMBUS_SATOER | GMBUS_HW_RDY), + 50); + if (ret) return -ETIMEDOUT; - if (I915_READ(GMBUS2 + reg_offset) & GMBUS_SATOER) + if (gmbus2 & GMBUS_SATOER) return -ENXIO; val = I915_READ(GMBUS3 + reg_offset); @@ -233,14 +235,13 @@ gmbus_xfer_read(struct drm_i915_private *dev_priv, struct i2c_msg *msg, *buf++ = val & 0xff; val >>= 8; } while (--len && ++loop < 4); - } while (len); + } return 0; } static int -gmbus_xfer_write(struct drm_i915_private *dev_priv, struct i2c_msg *msg, - bool last) +gmbus_xfer_write(struct drm_i915_private *dev_priv, struct i2c_msg *msg) { int reg_offset = dev_priv->gpio_mmio_base; u16 len = msg->len; @@ -248,25 +249,20 @@ gmbus_xfer_write(struct drm_i915_private *dev_priv, struct i2c_msg *msg, u32 val, loop; val = loop = 0; - do { - val |= *buf++ << (8 * loop); - } while (--len && ++loop < 4); + while (len && loop < 4) { + val |= *buf++ << (8 * loop++); + len -= 1; + } I915_WRITE(GMBUS3 + reg_offset, val); I915_WRITE(GMBUS1 + reg_offset, GMBUS_CYCLE_WAIT | - (last ? GMBUS_CYCLE_STOP : 0) | (msg->len << GMBUS_BYTE_COUNT_SHIFT) | (msg->addr << GMBUS_SLAVE_ADDR_SHIFT) | GMBUS_SLAVE_WRITE | GMBUS_SW_RDY); - POSTING_READ(GMBUS2 + reg_offset); while (len) { - if (wait_for(I915_READ(GMBUS2 + reg_offset) & - (GMBUS_SATOER | GMBUS_HW_RDY), - 50)) - return -ETIMEDOUT; - if (I915_READ(GMBUS2 + reg_offset) & GMBUS_SATOER) - return -ENXIO; + int ret; + u32 gmbus2; val = loop = 0; do { @@ -274,11 +270,58 @@ gmbus_xfer_write(struct drm_i915_private *dev_priv, struct i2c_msg *msg, } while (--len && ++loop < 4); I915_WRITE(GMBUS3 + reg_offset, val); - POSTING_READ(GMBUS2 + reg_offset); + + ret = wait_for((gmbus2 = I915_READ(GMBUS2 + reg_offset)) & + (GMBUS_SATOER | GMBUS_HW_RDY), + 50); + if (ret) + return -ETIMEDOUT; + if (gmbus2 & GMBUS_SATOER) + return -ENXIO; } return 0; } +/* + * The gmbus controller can combine a 1 or 2 byte write with a read that + * immediately follows it by using an "INDEX" cycle. + */ +static bool +gmbus_is_index_read(struct i2c_msg *msgs, int i, int num) +{ + return (i + 1 < num && + !(msgs[i].flags & I2C_M_RD) && msgs[i].len <= 2 && + (msgs[i + 1].flags & I2C_M_RD)); +} + +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; + + if (msgs[0].len == 2) + gmbus5 = GMBUS_2BYTE_INDEX_EN | + msgs[0].buf[1] | (msgs[0].buf[0] << 8); + if (msgs[0].len == 1) + gmbus1_index = GMBUS_CYCLE_INDEX | + (msgs[0].buf[0] << GMBUS_SLAVE_INDEX_SHIFT); + + /* GMBUS5 holds 16-bit index */ + if (gmbus5) + I915_WRITE(GMBUS5 + reg_offset, 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); + + return ret; +} + static int gmbus_xfer(struct i2c_adapter *adapter, struct i2c_msg *msgs, @@ -288,7 +331,8 @@ gmbus_xfer(struct i2c_adapter *adapter, struct intel_gmbus, adapter); struct drm_i915_private *dev_priv = bus->dev_priv; - int i, reg_offset, ret; + int i, reg_offset; + int ret = 0; mutex_lock(&dev_priv->gmbus_mutex); @@ -302,47 +346,82 @@ gmbus_xfer(struct i2c_adapter *adapter, I915_WRITE(GMBUS0 + reg_offset, bus->reg0); for (i = 0; i < num; i++) { - bool last = i + 1 == num; + u32 gmbus2; - if (msgs[i].flags & I2C_M_RD) - ret = gmbus_xfer_read(dev_priv, &msgs[i], last); - else - ret = gmbus_xfer_write(dev_priv, &msgs[i], last); + if (gmbus_is_index_read(msgs, i, num)) { + ret = gmbus_xfer_index_read(dev_priv, &msgs[i]); + i += 1; /* set i to the index of the read xfer */ + } else if (msgs[i].flags & I2C_M_RD) { + ret = gmbus_xfer_read(dev_priv, &msgs[i], 0); + } else { + ret = gmbus_xfer_write(dev_priv, &msgs[i]); + } if (ret == -ETIMEDOUT) goto timeout; if (ret == -ENXIO) goto clear_err; - if (!last && - wait_for(I915_READ(GMBUS2 + reg_offset) & - (GMBUS_SATOER | GMBUS_HW_WAIT_PHASE), - 50)) + ret = wait_for((gmbus2 = I915_READ(GMBUS2 + reg_offset)) & + (GMBUS_SATOER | GMBUS_HW_WAIT_PHASE), + 50); + if (ret) goto timeout; - if (I915_READ(GMBUS2 + reg_offset) & GMBUS_SATOER) + if (gmbus2 & GMBUS_SATOER) goto clear_err; } - goto done; + /* Generate a STOP condition on the bus. Note that gmbus can't generata + * 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); + + /* Mark the GMBUS interface as disabled after waiting for idle. + * We will re-enable it at the start of the next xfer, + * till then let it sleep. + */ + if (wait_for((I915_READ(GMBUS2 + reg_offset) & GMBUS_ACTIVE) == 0, + 10)) { + DRM_DEBUG_KMS("GMBUS [%s] timed out waiting for idle\n", + adapter->name); + ret = -ETIMEDOUT; + } + I915_WRITE(GMBUS0 + reg_offset, 0); + ret = ret ?: i; + goto out; clear_err: + /* + * Wait for bus to IDLE before clearing NAK. + * If we clear the NAK while bus is still active, then it will stay + * active and the next transaction may fail. + */ + if (wait_for((I915_READ(GMBUS2 + reg_offset) & GMBUS_ACTIVE) == 0, + 10)) + DRM_DEBUG_KMS("GMBUS [%s] timed out after NAK\n", + adapter->name); + /* Toggle the Software Clear Interrupt bit. This has the effect * 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); -done: - /* Mark the GMBUS interface as disabled after waiting for idle. - * We will re-enable it at the start of the next xfer, - * till then let it sleep. + DRM_DEBUG_KMS("GMBUS [%s] NAK for addr: %04x %c(%d)\n", + adapter->name, msgs[i].addr, + (msgs[i].flags & I2C_M_RD) ? 'r' : 'w', msgs[i].len); + + /* + * If no ACK is received during the address phase of a transaction, + * the adapter must report -ENXIO. + * It is not clear what to return if no ACK is received at other times. + * So, we always return -ENXIO in all NAK cases, to ensure we send + * it at least during the one case that is specified. */ - if (wait_for((I915_READ(GMBUS2 + reg_offset) & GMBUS_ACTIVE) == 0, 10)) - DRM_INFO("GMBUS [%s] timed out waiting for idle\n", - bus->adapter.name); - I915_WRITE(GMBUS0 + reg_offset, 0); - ret = i; + ret = -ENXIO; goto out; timeout: diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c new file mode 100644 index 0000000..36940a3 --- /dev/null +++ b/drivers/gpu/drm/i915/intel_pm.c @@ -0,0 +1,3075 @@ +/* + * Copyright © 2012 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: + * Eugeni Dodonov <eugeni.dodonov@intel.com> + * + */ + +#include <linux/cpufreq.h> +#include "i915_drv.h" +#include "intel_drv.h" + +/* FBC, or Frame Buffer Compression, is a technique employed to compress the + * framebuffer contents in-memory, aiming at reducing the required bandwidth + * during in-memory transfers and, therefore, reduce the power packet. + * + * The benefits of FBC are mostly visible with solid backgrounds and + * variation-less patterns. + * + * FBC-related functionality can be enabled by the means of the + * i915.i915_enable_fbc parameter + */ + +static void i8xx_disable_fbc(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + u32 fbc_ctl; + + /* Disable compression */ + fbc_ctl = I915_READ(FBC_CONTROL); + if ((fbc_ctl & FBC_CTL_EN) == 0) + return; + + fbc_ctl &= ~FBC_CTL_EN; + I915_WRITE(FBC_CONTROL, fbc_ctl); + + /* Wait for compressing bit to clear */ + if (wait_for((I915_READ(FBC_STATUS) & FBC_STAT_COMPRESSING) == 0, 10)) { + DRM_DEBUG_KMS("FBC idle timed out\n"); + return; + } + + DRM_DEBUG_KMS("disabled FBC\n"); +} + +static void i8xx_enable_fbc(struct drm_crtc *crtc, unsigned long interval) +{ + struct drm_device *dev = crtc->dev; + struct drm_i915_private *dev_priv = dev->dev_private; + struct drm_framebuffer *fb = crtc->fb; + struct intel_framebuffer *intel_fb = to_intel_framebuffer(fb); + struct drm_i915_gem_object *obj = intel_fb->obj; + struct intel_crtc *intel_crtc = to_intel_crtc(crtc); + int cfb_pitch; + int plane, i; + u32 fbc_ctl, fbc_ctl2; + + cfb_pitch = dev_priv->cfb_size / FBC_LL_SIZE; + if (fb->pitches[0] < cfb_pitch) + cfb_pitch = fb->pitches[0]; + + /* FBC_CTL wants 64B units */ + cfb_pitch = (cfb_pitch / 64) - 1; + plane = intel_crtc->plane == 0 ? FBC_CTL_PLANEA : FBC_CTL_PLANEB; + + /* Clear old tags */ + for (i = 0; i < (FBC_LL_SIZE / 32) + 1; i++) + I915_WRITE(FBC_TAG + (i * 4), 0); + + /* Set it up... */ + fbc_ctl2 = FBC_CTL_FENCE_DBL | FBC_CTL_IDLE_IMM | FBC_CTL_CPU_FENCE; + fbc_ctl2 |= plane; + I915_WRITE(FBC_CONTROL2, fbc_ctl2); + I915_WRITE(FBC_FENCE_OFF, crtc->y); + + /* enable it... */ + fbc_ctl = FBC_CTL_EN | FBC_CTL_PERIODIC; + if (IS_I945GM(dev)) + fbc_ctl |= FBC_CTL_C3_IDLE; /* 945 needs special SR handling */ + fbc_ctl |= (cfb_pitch & 0xff) << FBC_CTL_STRIDE_SHIFT; + fbc_ctl |= (interval & 0x2fff) << FBC_CTL_INTERVAL_SHIFT; + fbc_ctl |= obj->fence_reg; + I915_WRITE(FBC_CONTROL, fbc_ctl); + + DRM_DEBUG_KMS("enabled FBC, pitch %d, yoff %d, plane %d, ", + cfb_pitch, crtc->y, intel_crtc->plane); +} + +static bool i8xx_fbc_enabled(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + + return I915_READ(FBC_CONTROL) & FBC_CTL_EN; +} + +static void g4x_enable_fbc(struct drm_crtc *crtc, unsigned long interval) +{ + struct drm_device *dev = crtc->dev; + struct drm_i915_private *dev_priv = dev->dev_private; + struct drm_framebuffer *fb = crtc->fb; + struct intel_framebuffer *intel_fb = to_intel_framebuffer(fb); + struct drm_i915_gem_object *obj = intel_fb->obj; + struct intel_crtc *intel_crtc = to_intel_crtc(crtc); + int plane = intel_crtc->plane == 0 ? DPFC_CTL_PLANEA : DPFC_CTL_PLANEB; + unsigned long stall_watermark = 200; + u32 dpfc_ctl; + + dpfc_ctl = plane | DPFC_SR_EN | DPFC_CTL_LIMIT_1X; + dpfc_ctl |= DPFC_CTL_FENCE_EN | obj->fence_reg; + I915_WRITE(DPFC_CHICKEN, DPFC_HT_MODIFY); + + I915_WRITE(DPFC_RECOMP_CTL, DPFC_RECOMP_STALL_EN | + (stall_watermark << DPFC_RECOMP_STALL_WM_SHIFT) | + (interval << DPFC_RECOMP_TIMER_COUNT_SHIFT)); + I915_WRITE(DPFC_FENCE_YOFF, crtc->y); + + /* enable it... */ + I915_WRITE(DPFC_CONTROL, I915_READ(DPFC_CONTROL) | DPFC_CTL_EN); + + DRM_DEBUG_KMS("enabled fbc on plane %d\n", intel_crtc->plane); +} + +static void g4x_disable_fbc(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + u32 dpfc_ctl; + + /* Disable compression */ + dpfc_ctl = I915_READ(DPFC_CONTROL); + if (dpfc_ctl & DPFC_CTL_EN) { + dpfc_ctl &= ~DPFC_CTL_EN; + I915_WRITE(DPFC_CONTROL, dpfc_ctl); + + DRM_DEBUG_KMS("disabled FBC\n"); + } +} + +static bool g4x_fbc_enabled(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + + return I915_READ(DPFC_CONTROL) & DPFC_CTL_EN; +} + +static void sandybridge_blit_fbc_update(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + u32 blt_ecoskpd; + + /* Make sure blitter notifies FBC of writes */ + gen6_gt_force_wake_get(dev_priv); + blt_ecoskpd = I915_READ(GEN6_BLITTER_ECOSKPD); + blt_ecoskpd |= GEN6_BLITTER_FBC_NOTIFY << + GEN6_BLITTER_LOCK_SHIFT; + I915_WRITE(GEN6_BLITTER_ECOSKPD, blt_ecoskpd); + blt_ecoskpd |= GEN6_BLITTER_FBC_NOTIFY; + I915_WRITE(GEN6_BLITTER_ECOSKPD, blt_ecoskpd); + blt_ecoskpd &= ~(GEN6_BLITTER_FBC_NOTIFY << + GEN6_BLITTER_LOCK_SHIFT); + I915_WRITE(GEN6_BLITTER_ECOSKPD, blt_ecoskpd); + POSTING_READ(GEN6_BLITTER_ECOSKPD); + gen6_gt_force_wake_put(dev_priv); +} + +static void ironlake_enable_fbc(struct drm_crtc *crtc, unsigned long interval) +{ + struct drm_device *dev = crtc->dev; + struct drm_i915_private *dev_priv = dev->dev_private; + struct drm_framebuffer *fb = crtc->fb; + struct intel_framebuffer *intel_fb = to_intel_framebuffer(fb); + struct drm_i915_gem_object *obj = intel_fb->obj; + struct intel_crtc *intel_crtc = to_intel_crtc(crtc); + int plane = intel_crtc->plane == 0 ? DPFC_CTL_PLANEA : DPFC_CTL_PLANEB; + unsigned long stall_watermark = 200; + u32 dpfc_ctl; + + dpfc_ctl = I915_READ(ILK_DPFC_CONTROL); + dpfc_ctl &= DPFC_RESERVED; + dpfc_ctl |= (plane | DPFC_CTL_LIMIT_1X); + /* Set persistent mode for front-buffer rendering, ala X. */ + dpfc_ctl |= DPFC_CTL_PERSISTENT_MODE; + dpfc_ctl |= (DPFC_CTL_FENCE_EN | obj->fence_reg); + I915_WRITE(ILK_DPFC_CHICKEN, DPFC_HT_MODIFY); + + I915_WRITE(ILK_DPFC_RECOMP_CTL, DPFC_RECOMP_STALL_EN | + (stall_watermark << DPFC_RECOMP_STALL_WM_SHIFT) | + (interval << DPFC_RECOMP_TIMER_COUNT_SHIFT)); + I915_WRITE(ILK_DPFC_FENCE_YOFF, crtc->y); + I915_WRITE(ILK_FBC_RT_BASE, obj->gtt_offset | ILK_FBC_RT_VALID); + /* enable it... */ + I915_WRITE(ILK_DPFC_CONTROL, dpfc_ctl | DPFC_CTL_EN); + + if (IS_GEN6(dev)) { + I915_WRITE(SNB_DPFC_CTL_SA, + SNB_CPU_FENCE_ENABLE | obj->fence_reg); + I915_WRITE(DPFC_CPU_FENCE_OFFSET, crtc->y); + sandybridge_blit_fbc_update(dev); + } + + DRM_DEBUG_KMS("enabled fbc on plane %d\n", intel_crtc->plane); +} + +static void ironlake_disable_fbc(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + u32 dpfc_ctl; + + /* Disable compression */ + dpfc_ctl = I915_READ(ILK_DPFC_CONTROL); + if (dpfc_ctl & DPFC_CTL_EN) { + dpfc_ctl &= ~DPFC_CTL_EN; + I915_WRITE(ILK_DPFC_CONTROL, dpfc_ctl); + + DRM_DEBUG_KMS("disabled FBC\n"); + } +} + +static bool ironlake_fbc_enabled(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + + return I915_READ(ILK_DPFC_CONTROL) & DPFC_CTL_EN; +} + +bool intel_fbc_enabled(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + + if (!dev_priv->display.fbc_enabled) + return false; + + return dev_priv->display.fbc_enabled(dev); +} + +static void intel_fbc_work_fn(struct work_struct *__work) +{ + struct intel_fbc_work *work = + container_of(to_delayed_work(__work), + struct intel_fbc_work, work); + struct drm_device *dev = work->crtc->dev; + struct drm_i915_private *dev_priv = dev->dev_private; + + mutex_lock(&dev->struct_mutex); + if (work == dev_priv->fbc_work) { + /* Double check that we haven't switched fb without cancelling + * the prior work. + */ + if (work->crtc->fb == work->fb) { + dev_priv->display.enable_fbc(work->crtc, + work->interval); + + dev_priv->cfb_plane = to_intel_crtc(work->crtc)->plane; + dev_priv->cfb_fb = work->crtc->fb->base.id; + dev_priv->cfb_y = work->crtc->y; + } + + dev_priv->fbc_work = NULL; + } + mutex_unlock(&dev->struct_mutex); + + kfree(work); +} + +static void intel_cancel_fbc_work(struct drm_i915_private *dev_priv) +{ + if (dev_priv->fbc_work == NULL) + return; + + DRM_DEBUG_KMS("cancelling pending FBC enable\n"); + + /* Synchronisation is provided by struct_mutex and checking of + * dev_priv->fbc_work, so we can perform the cancellation + * entirely asynchronously. + */ + if (cancel_delayed_work(&dev_priv->fbc_work->work)) + /* tasklet was killed before being run, clean up */ + kfree(dev_priv->fbc_work); + + /* Mark the work as no longer wanted so that if it does + * wake-up (because the work was already running and waiting + * for our mutex), it will discover that is no longer + * necessary to run. + */ + dev_priv->fbc_work = NULL; +} + +void intel_enable_fbc(struct drm_crtc *crtc, unsigned long interval) +{ + struct intel_fbc_work *work; + struct drm_device *dev = crtc->dev; + struct drm_i915_private *dev_priv = dev->dev_private; + + if (!dev_priv->display.enable_fbc) + return; + + intel_cancel_fbc_work(dev_priv); + + work = kzalloc(sizeof *work, GFP_KERNEL); + if (work == NULL) { + dev_priv->display.enable_fbc(crtc, interval); + return; + } + + work->crtc = crtc; + work->fb = crtc->fb; + work->interval = interval; + INIT_DELAYED_WORK(&work->work, intel_fbc_work_fn); + + dev_priv->fbc_work = work; + + DRM_DEBUG_KMS("scheduling delayed FBC enable\n"); + + /* Delay the actual enabling to let pageflipping cease and the + * display to settle before starting the compression. Note that + * this delay also serves a second purpose: it allows for a + * vblank to pass after disabling the FBC before we attempt + * to modify the control registers. + * + * A more complicated solution would involve tracking vblanks + * following the termination of the page-flipping sequence + * and indeed performing the enable as a co-routine and not + * waiting synchronously upon the vblank. + */ + schedule_delayed_work(&work->work, msecs_to_jiffies(50)); +} + +void intel_disable_fbc(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + + intel_cancel_fbc_work(dev_priv); + + if (!dev_priv->display.disable_fbc) + return; + + dev_priv->display.disable_fbc(dev); + dev_priv->cfb_plane = -1; +} + +/** + * intel_update_fbc - enable/disable FBC as needed + * @dev: the drm_device + * + * Set up the framebuffer compression hardware at mode set time. We + * enable it if possible: + * - plane A only (on pre-965) + * - no pixel mulitply/line duplication + * - no alpha buffer discard + * - no dual wide + * - framebuffer <= 2048 in width, 1536 in height + * + * We can't assume that any compression will take place (worst case), + * so the compressed buffer has to be the same size as the uncompressed + * one. It also must reside (along with the line length buffer) in + * stolen memory. + * + * We need to enable/disable FBC on a global basis. + */ +void intel_update_fbc(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + struct drm_crtc *crtc = NULL, *tmp_crtc; + struct intel_crtc *intel_crtc; + struct drm_framebuffer *fb; + struct intel_framebuffer *intel_fb; + struct drm_i915_gem_object *obj; + int enable_fbc; + + DRM_DEBUG_KMS("\n"); + + if (!i915_powersave) + return; + + if (!I915_HAS_FBC(dev)) + return; + + /* + * If FBC is already on, we just have to verify that we can + * keep it that way... + * Need to disable if: + * - more than one pipe is active + * - changing FBC params (stride, fence, mode) + * - new fb is too large to fit in compressed buffer + * - going to an unsupported config (interlace, pixel multiply, etc.) + */ + list_for_each_entry(tmp_crtc, &dev->mode_config.crtc_list, head) { + if (tmp_crtc->enabled && tmp_crtc->fb) { + if (crtc) { + DRM_DEBUG_KMS("more than one pipe active, disabling compression\n"); + dev_priv->no_fbc_reason = FBC_MULTIPLE_PIPES; + goto out_disable; + } + crtc = tmp_crtc; + } + } + + if (!crtc || crtc->fb == NULL) { + DRM_DEBUG_KMS("no output, disabling\n"); + dev_priv->no_fbc_reason = FBC_NO_OUTPUT; + goto out_disable; + } + + intel_crtc = to_intel_crtc(crtc); + fb = crtc->fb; + intel_fb = to_intel_framebuffer(fb); + obj = intel_fb->obj; + + enable_fbc = i915_enable_fbc; + if (enable_fbc < 0) { + DRM_DEBUG_KMS("fbc set to per-chip default\n"); + enable_fbc = 1; + if (INTEL_INFO(dev)->gen <= 6) + enable_fbc = 0; + } + if (!enable_fbc) { + DRM_DEBUG_KMS("fbc disabled per module param\n"); + dev_priv->no_fbc_reason = FBC_MODULE_PARAM; + goto out_disable; + } + if (intel_fb->obj->base.size > dev_priv->cfb_size) { + DRM_DEBUG_KMS("framebuffer too large, disabling " + "compression\n"); + dev_priv->no_fbc_reason = FBC_STOLEN_TOO_SMALL; + goto out_disable; + } + if ((crtc->mode.flags & DRM_MODE_FLAG_INTERLACE) || + (crtc->mode.flags & DRM_MODE_FLAG_DBLSCAN)) { + DRM_DEBUG_KMS("mode incompatible with compression, " + "disabling\n"); + dev_priv->no_fbc_reason = FBC_UNSUPPORTED_MODE; + goto out_disable; + } + if ((crtc->mode.hdisplay > 2048) || + (crtc->mode.vdisplay > 1536)) { + DRM_DEBUG_KMS("mode too large for compression, disabling\n"); + dev_priv->no_fbc_reason = FBC_MODE_TOO_LARGE; + goto out_disable; + } + if ((IS_I915GM(dev) || IS_I945GM(dev)) && intel_crtc->plane != 0) { + DRM_DEBUG_KMS("plane not 0, disabling compression\n"); + dev_priv->no_fbc_reason = FBC_BAD_PLANE; + goto out_disable; + } + + /* The use of a CPU fence is mandatory in order to detect writes + * by the CPU to the scanout and trigger updates to the FBC. + */ + if (obj->tiling_mode != I915_TILING_X || + obj->fence_reg == I915_FENCE_REG_NONE) { + DRM_DEBUG_KMS("framebuffer not tiled or fenced, disabling compression\n"); + dev_priv->no_fbc_reason = FBC_NOT_TILED; + goto out_disable; + } + + /* If the kernel debugger is active, always disable compression */ + if (in_dbg_master()) + goto out_disable; + + /* If the scanout has not changed, don't modify the FBC settings. + * Note that we make the fundamental assumption that the fb->obj + * cannot be unpinned (and have its GTT offset and fence revoked) + * without first being decoupled from the scanout and FBC disabled. + */ + if (dev_priv->cfb_plane == intel_crtc->plane && + dev_priv->cfb_fb == fb->base.id && + dev_priv->cfb_y == crtc->y) + return; + + if (intel_fbc_enabled(dev)) { + /* We update FBC along two paths, after changing fb/crtc + * configuration (modeswitching) and after page-flipping + * finishes. For the latter, we know that not only did + * we disable the FBC at the start of the page-flip + * sequence, but also more than one vblank has passed. + * + * For the former case of modeswitching, it is possible + * to switch between two FBC valid configurations + * instantaneously so we do need to disable the FBC + * before we can modify its control registers. We also + * have to wait for the next vblank for that to take + * effect. However, since we delay enabling FBC we can + * assume that a vblank has passed since disabling and + * that we can safely alter the registers in the deferred + * callback. + * + * In the scenario that we go from a valid to invalid + * and then back to valid FBC configuration we have + * no strict enforcement that a vblank occurred since + * disabling the FBC. However, along all current pipe + * disabling paths we do need to wait for a vblank at + * some point. And we wait before enabling FBC anyway. + */ + DRM_DEBUG_KMS("disabling active FBC for update\n"); + intel_disable_fbc(dev); + } + + intel_enable_fbc(crtc, 500); + return; + +out_disable: + /* Multiple disables should be harmless */ + if (intel_fbc_enabled(dev)) { + DRM_DEBUG_KMS("unsupported config, disabling FBC\n"); + intel_disable_fbc(dev); + } +} + +static const struct cxsr_latency cxsr_latency_table[] = { + {1, 0, 800, 400, 3382, 33382, 3983, 33983}, /* DDR2-400 SC */ + {1, 0, 800, 667, 3354, 33354, 3807, 33807}, /* DDR2-667 SC */ + {1, 0, 800, 800, 3347, 33347, 3763, 33763}, /* DDR2-800 SC */ + {1, 1, 800, 667, 6420, 36420, 6873, 36873}, /* DDR3-667 SC */ + {1, 1, 800, 800, 5902, 35902, 6318, 36318}, /* DDR3-800 SC */ + + {1, 0, 667, 400, 3400, 33400, 4021, 34021}, /* DDR2-400 SC */ + {1, 0, 667, 667, 3372, 33372, 3845, 33845}, /* DDR2-667 SC */ + {1, 0, 667, 800, 3386, 33386, 3822, 33822}, /* DDR2-800 SC */ + {1, 1, 667, 667, 6438, 36438, 6911, 36911}, /* DDR3-667 SC */ + {1, 1, 667, 800, 5941, 35941, 6377, 36377}, /* DDR3-800 SC */ + + {1, 0, 400, 400, 3472, 33472, 4173, 34173}, /* DDR2-400 SC */ + {1, 0, 400, 667, 3443, 33443, 3996, 33996}, /* DDR2-667 SC */ + {1, 0, 400, 800, 3430, 33430, 3946, 33946}, /* DDR2-800 SC */ + {1, 1, 400, 667, 6509, 36509, 7062, 37062}, /* DDR3-667 SC */ + {1, 1, 400, 800, 5985, 35985, 6501, 36501}, /* DDR3-800 SC */ + + {0, 0, 800, 400, 3438, 33438, 4065, 34065}, /* DDR2-400 SC */ + {0, 0, 800, 667, 3410, 33410, 3889, 33889}, /* DDR2-667 SC */ + {0, 0, 800, 800, 3403, 33403, 3845, 33845}, /* DDR2-800 SC */ + {0, 1, 800, 667, 6476, 36476, 6955, 36955}, /* DDR3-667 SC */ + {0, 1, 800, 800, 5958, 35958, 6400, 36400}, /* DDR3-800 SC */ + + {0, 0, 667, 400, 3456, 33456, 4103, 34106}, /* DDR2-400 SC */ + {0, 0, 667, 667, 3428, 33428, 3927, 33927}, /* DDR2-667 SC */ + {0, 0, 667, 800, 3443, 33443, 3905, 33905}, /* DDR2-800 SC */ + {0, 1, 667, 667, 6494, 36494, 6993, 36993}, /* DDR3-667 SC */ + {0, 1, 667, 800, 5998, 35998, 6460, 36460}, /* DDR3-800 SC */ + + {0, 0, 400, 400, 3528, 33528, 4255, 34255}, /* DDR2-400 SC */ + {0, 0, 400, 667, 3500, 33500, 4079, 34079}, /* DDR2-667 SC */ + {0, 0, 400, 800, 3487, 33487, 4029, 34029}, /* DDR2-800 SC */ + {0, 1, 400, 667, 6566, 36566, 7145, 37145}, /* DDR3-667 SC */ + {0, 1, 400, 800, 6042, 36042, 6584, 36584}, /* DDR3-800 SC */ +}; + +const struct cxsr_latency *intel_get_cxsr_latency(int is_desktop, + int is_ddr3, + int fsb, + int mem) +{ + const struct cxsr_latency *latency; + int i; + + if (fsb == 0 || mem == 0) + return NULL; + + for (i = 0; i < ARRAY_SIZE(cxsr_latency_table); i++) { + latency = &cxsr_latency_table[i]; + if (is_desktop == latency->is_desktop && + is_ddr3 == latency->is_ddr3 && + fsb == latency->fsb_freq && mem == latency->mem_freq) + return latency; + } + + DRM_DEBUG_KMS("Unknown FSB/MEM found, disable CxSR\n"); + + return NULL; +} + +static void pineview_disable_cxsr(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + + /* deactivate cxsr */ + I915_WRITE(DSPFW3, I915_READ(DSPFW3) & ~PINEVIEW_SELF_REFRESH_EN); +} + +/* + * Latency for FIFO fetches is dependent on several factors: + * - memory configuration (speed, channels) + * - chipset + * - current MCH state + * It can be fairly high in some situations, so here we assume a fairly + * pessimal value. It's a tradeoff between extra memory fetches (if we + * set this value too high, the FIFO will fetch frequently to stay full) + * and power consumption (set it too low to save power and we might see + * FIFO underruns and display "flicker"). + * + * A value of 5us seems to be a good balance; safe for very low end + * platforms but not overly aggressive on lower latency configs. + */ +static const int latency_ns = 5000; + +static int i9xx_get_fifo_size(struct drm_device *dev, int plane) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + uint32_t dsparb = I915_READ(DSPARB); + int size; + + size = dsparb & 0x7f; + if (plane) + size = ((dsparb >> DSPARB_CSTART_SHIFT) & 0x7f) - size; + + DRM_DEBUG_KMS("FIFO size - (0x%08x) %s: %d\n", dsparb, + plane ? "B" : "A", size); + + return size; +} + +static int i85x_get_fifo_size(struct drm_device *dev, int plane) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + uint32_t dsparb = I915_READ(DSPARB); + int size; + + size = dsparb & 0x1ff; + if (plane) + size = ((dsparb >> DSPARB_BEND_SHIFT) & 0x1ff) - size; + size >>= 1; /* Convert to cachelines */ + + DRM_DEBUG_KMS("FIFO size - (0x%08x) %s: %d\n", dsparb, + plane ? "B" : "A", size); + + return size; +} + +static int i845_get_fifo_size(struct drm_device *dev, int plane) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + uint32_t dsparb = I915_READ(DSPARB); + int size; + + size = dsparb & 0x7f; + size >>= 2; /* Convert to cachelines */ + + DRM_DEBUG_KMS("FIFO size - (0x%08x) %s: %d\n", dsparb, + plane ? "B" : "A", + size); + + return size; +} + +static int i830_get_fifo_size(struct drm_device *dev, int plane) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + uint32_t dsparb = I915_READ(DSPARB); + int size; + + size = dsparb & 0x7f; + size >>= 1; /* Convert to cachelines */ + + DRM_DEBUG_KMS("FIFO size - (0x%08x) %s: %d\n", dsparb, + plane ? "B" : "A", size); + + return size; +} + +/* Pineview has different values for various configs */ +static const struct intel_watermark_params pineview_display_wm = { + PINEVIEW_DISPLAY_FIFO, + PINEVIEW_MAX_WM, + PINEVIEW_DFT_WM, + PINEVIEW_GUARD_WM, + PINEVIEW_FIFO_LINE_SIZE +}; +static const struct intel_watermark_params pineview_display_hplloff_wm = { + PINEVIEW_DISPLAY_FIFO, + PINEVIEW_MAX_WM, + PINEVIEW_DFT_HPLLOFF_WM, + PINEVIEW_GUARD_WM, + PINEVIEW_FIFO_LINE_SIZE +}; +static const struct intel_watermark_params pineview_cursor_wm = { + PINEVIEW_CURSOR_FIFO, + PINEVIEW_CURSOR_MAX_WM, + PINEVIEW_CURSOR_DFT_WM, + PINEVIEW_CURSOR_GUARD_WM, + PINEVIEW_FIFO_LINE_SIZE, +}; +static const struct intel_watermark_params pineview_cursor_hplloff_wm = { + PINEVIEW_CURSOR_FIFO, + PINEVIEW_CURSOR_MAX_WM, + PINEVIEW_CURSOR_DFT_WM, + PINEVIEW_CURSOR_GUARD_WM, + PINEVIEW_FIFO_LINE_SIZE +}; +static const struct intel_watermark_params g4x_wm_info = { + G4X_FIFO_SIZE, + G4X_MAX_WM, + G4X_MAX_WM, + 2, + G4X_FIFO_LINE_SIZE, +}; +static const struct intel_watermark_params g4x_cursor_wm_info = { + I965_CURSOR_FIFO, + I965_CURSOR_MAX_WM, + I965_CURSOR_DFT_WM, + 2, + G4X_FIFO_LINE_SIZE, +}; +static const struct intel_watermark_params valleyview_wm_info = { + VALLEYVIEW_FIFO_SIZE, + VALLEYVIEW_MAX_WM, + VALLEYVIEW_MAX_WM, + 2, + G4X_FIFO_LINE_SIZE, +}; +static const struct intel_watermark_params valleyview_cursor_wm_info = { + I965_CURSOR_FIFO, + VALLEYVIEW_CURSOR_MAX_WM, + I965_CURSOR_DFT_WM, + 2, + G4X_FIFO_LINE_SIZE, +}; +static const struct intel_watermark_params i965_cursor_wm_info = { + I965_CURSOR_FIFO, + I965_CURSOR_MAX_WM, + I965_CURSOR_DFT_WM, + 2, + I915_FIFO_LINE_SIZE, +}; +static const struct intel_watermark_params i945_wm_info = { + I945_FIFO_SIZE, + I915_MAX_WM, + 1, + 2, + I915_FIFO_LINE_SIZE +}; +static const struct intel_watermark_params i915_wm_info = { + I915_FIFO_SIZE, + I915_MAX_WM, + 1, + 2, + I915_FIFO_LINE_SIZE +}; +static const struct intel_watermark_params i855_wm_info = { + I855GM_FIFO_SIZE, + I915_MAX_WM, + 1, + 2, + I830_FIFO_LINE_SIZE +}; +static const struct intel_watermark_params i830_wm_info = { + I830_FIFO_SIZE, + I915_MAX_WM, + 1, + 2, + I830_FIFO_LINE_SIZE +}; + +static const struct intel_watermark_params ironlake_display_wm_info = { + ILK_DISPLAY_FIFO, + ILK_DISPLAY_MAXWM, + ILK_DISPLAY_DFTWM, + 2, + ILK_FIFO_LINE_SIZE +}; +static const struct intel_watermark_params ironlake_cursor_wm_info = { + ILK_CURSOR_FIFO, + ILK_CURSOR_MAXWM, + ILK_CURSOR_DFTWM, + 2, + ILK_FIFO_LINE_SIZE +}; +static const struct intel_watermark_params ironlake_display_srwm_info = { + ILK_DISPLAY_SR_FIFO, + ILK_DISPLAY_MAX_SRWM, + ILK_DISPLAY_DFT_SRWM, + 2, + ILK_FIFO_LINE_SIZE +}; +static const struct intel_watermark_params ironlake_cursor_srwm_info = { + ILK_CURSOR_SR_FIFO, + ILK_CURSOR_MAX_SRWM, + ILK_CURSOR_DFT_SRWM, + 2, + ILK_FIFO_LINE_SIZE +}; + +static const struct intel_watermark_params sandybridge_display_wm_info = { + SNB_DISPLAY_FIFO, + SNB_DISPLAY_MAXWM, + SNB_DISPLAY_DFTWM, + 2, + SNB_FIFO_LINE_SIZE +}; +static const struct intel_watermark_params sandybridge_cursor_wm_info = { + SNB_CURSOR_FIFO, + SNB_CURSOR_MAXWM, + SNB_CURSOR_DFTWM, + 2, + SNB_FIFO_LINE_SIZE +}; +static const struct intel_watermark_params sandybridge_display_srwm_info = { + SNB_DISPLAY_SR_FIFO, + SNB_DISPLAY_MAX_SRWM, + SNB_DISPLAY_DFT_SRWM, + 2, + SNB_FIFO_LINE_SIZE +}; +static const struct intel_watermark_params sandybridge_cursor_srwm_info = { + SNB_CURSOR_SR_FIFO, + SNB_CURSOR_MAX_SRWM, + SNB_CURSOR_DFT_SRWM, + 2, + SNB_FIFO_LINE_SIZE +}; + + +/** + * intel_calculate_wm - calculate watermark level + * @clock_in_khz: pixel clock + * @wm: chip FIFO params + * @pixel_size: display pixel size + * @latency_ns: memory latency for the platform + * + * Calculate the watermark level (the level at which the display plane will + * start fetching from memory again). Each chip has a different display + * FIFO size and allocation, so the caller needs to figure that out and pass + * in the correct intel_watermark_params structure. + * + * As the pixel clock runs, the FIFO will be drained at a rate that depends + * on the pixel size. When it reaches the watermark level, it'll start + * fetching FIFO line sized based chunks from memory until the FIFO fills + * past the watermark point. If the FIFO drains completely, a FIFO underrun + * will occur, and a display engine hang could result. + */ +static unsigned long intel_calculate_wm(unsigned long clock_in_khz, + const struct intel_watermark_params *wm, + int fifo_size, + int pixel_size, + unsigned long latency_ns) +{ + long entries_required, wm_size; + + /* + * Note: we need to make sure we don't overflow for various clock & + * latency values. + * clocks go from a few thousand to several hundred thousand. + * latency is usually a few thousand + */ + entries_required = ((clock_in_khz / 1000) * pixel_size * latency_ns) / + 1000; + entries_required = DIV_ROUND_UP(entries_required, wm->cacheline_size); + + DRM_DEBUG_KMS("FIFO entries required for mode: %ld\n", entries_required); + + wm_size = fifo_size - (entries_required + wm->guard_size); + + DRM_DEBUG_KMS("FIFO watermark level: %ld\n", wm_size); + + /* Don't promote wm_size to unsigned... */ + if (wm_size > (long)wm->max_wm) + wm_size = wm->max_wm; + if (wm_size <= 0) + wm_size = wm->default_wm; + return wm_size; +} + +static struct drm_crtc *single_enabled_crtc(struct drm_device *dev) +{ + struct drm_crtc *crtc, *enabled = NULL; + + list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) { + if (crtc->enabled && crtc->fb) { + if (enabled) + return NULL; + enabled = crtc; + } + } + + return enabled; +} + +static void pineview_update_wm(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + struct drm_crtc *crtc; + const struct cxsr_latency *latency; + u32 reg; + unsigned long wm; + + latency = intel_get_cxsr_latency(IS_PINEVIEW_G(dev), dev_priv->is_ddr3, + dev_priv->fsb_freq, dev_priv->mem_freq); + if (!latency) { + DRM_DEBUG_KMS("Unknown FSB/MEM found, disable CxSR\n"); + pineview_disable_cxsr(dev); + return; + } + + crtc = single_enabled_crtc(dev); + if (crtc) { + int clock = crtc->mode.clock; + int pixel_size = crtc->fb->bits_per_pixel / 8; + + /* Display SR */ + wm = intel_calculate_wm(clock, &pineview_display_wm, + pineview_display_wm.fifo_size, + pixel_size, latency->display_sr); + reg = I915_READ(DSPFW1); + reg &= ~DSPFW_SR_MASK; + reg |= wm << DSPFW_SR_SHIFT; + I915_WRITE(DSPFW1, reg); + DRM_DEBUG_KMS("DSPFW1 register is %x\n", reg); + + /* cursor SR */ + wm = intel_calculate_wm(clock, &pineview_cursor_wm, + pineview_display_wm.fifo_size, + pixel_size, latency->cursor_sr); + reg = I915_READ(DSPFW3); + reg &= ~DSPFW_CURSOR_SR_MASK; + reg |= (wm & 0x3f) << DSPFW_CURSOR_SR_SHIFT; + I915_WRITE(DSPFW3, reg); + + /* Display HPLL off SR */ + wm = intel_calculate_wm(clock, &pineview_display_hplloff_wm, + pineview_display_hplloff_wm.fifo_size, + pixel_size, latency->display_hpll_disable); + reg = I915_READ(DSPFW3); + reg &= ~DSPFW_HPLL_SR_MASK; + reg |= wm & DSPFW_HPLL_SR_MASK; + I915_WRITE(DSPFW3, reg); + + /* cursor HPLL off SR */ + wm = intel_calculate_wm(clock, &pineview_cursor_hplloff_wm, + pineview_display_hplloff_wm.fifo_size, + pixel_size, latency->cursor_hpll_disable); + reg = I915_READ(DSPFW3); + reg &= ~DSPFW_HPLL_CURSOR_MASK; + reg |= (wm & 0x3f) << DSPFW_HPLL_CURSOR_SHIFT; + I915_WRITE(DSPFW3, reg); + DRM_DEBUG_KMS("DSPFW3 register is %x\n", reg); + + /* activate cxsr */ + I915_WRITE(DSPFW3, + I915_READ(DSPFW3) | PINEVIEW_SELF_REFRESH_EN); + DRM_DEBUG_KMS("Self-refresh is enabled\n"); + } else { + pineview_disable_cxsr(dev); + DRM_DEBUG_KMS("Self-refresh is disabled\n"); + } +} + +static bool g4x_compute_wm0(struct drm_device *dev, + int plane, + const struct intel_watermark_params *display, + int display_latency_ns, + const struct intel_watermark_params *cursor, + int cursor_latency_ns, + int *plane_wm, + int *cursor_wm) +{ + struct drm_crtc *crtc; + int htotal, hdisplay, clock, pixel_size; + int line_time_us, line_count; + int entries, tlb_miss; + + crtc = intel_get_crtc_for_plane(dev, plane); + if (crtc->fb == NULL || !crtc->enabled) { + *cursor_wm = cursor->guard_size; + *plane_wm = display->guard_size; + return false; + } + + htotal = crtc->mode.htotal; + hdisplay = crtc->mode.hdisplay; + clock = crtc->mode.clock; + pixel_size = crtc->fb->bits_per_pixel / 8; + + /* Use the small buffer method to calculate plane watermark */ + entries = ((clock * pixel_size / 1000) * display_latency_ns) / 1000; + tlb_miss = display->fifo_size*display->cacheline_size - hdisplay * 8; + if (tlb_miss > 0) + entries += tlb_miss; + entries = DIV_ROUND_UP(entries, display->cacheline_size); + *plane_wm = entries + display->guard_size; + if (*plane_wm > (int)display->max_wm) + *plane_wm = display->max_wm; + + /* Use the large buffer method to calculate cursor watermark */ + line_time_us = ((htotal * 1000) / clock); + line_count = (cursor_latency_ns / line_time_us + 1000) / 1000; + entries = line_count * 64 * pixel_size; + tlb_miss = cursor->fifo_size*cursor->cacheline_size - hdisplay * 8; + if (tlb_miss > 0) + entries += tlb_miss; + entries = DIV_ROUND_UP(entries, cursor->cacheline_size); + *cursor_wm = entries + cursor->guard_size; + if (*cursor_wm > (int)cursor->max_wm) + *cursor_wm = (int)cursor->max_wm; + + return true; +} + +/* + * Check the wm result. + * + * If any calculated watermark values is larger than the maximum value that + * can be programmed into the associated watermark register, that watermark + * must be disabled. + */ +static bool g4x_check_srwm(struct drm_device *dev, + int display_wm, int cursor_wm, + const struct intel_watermark_params *display, + const struct intel_watermark_params *cursor) +{ + DRM_DEBUG_KMS("SR watermark: display plane %d, cursor %d\n", + display_wm, cursor_wm); + + if (display_wm > display->max_wm) { + DRM_DEBUG_KMS("display watermark is too large(%d/%ld), disabling\n", + display_wm, display->max_wm); + return false; + } + + if (cursor_wm > cursor->max_wm) { + DRM_DEBUG_KMS("cursor watermark is too large(%d/%ld), disabling\n", + cursor_wm, cursor->max_wm); + return false; + } + + if (!(display_wm || cursor_wm)) { + DRM_DEBUG_KMS("SR latency is 0, disabling\n"); + return false; + } + + return true; +} + +static bool g4x_compute_srwm(struct drm_device *dev, + int plane, + int latency_ns, + const struct intel_watermark_params *display, + const struct intel_watermark_params *cursor, + int *display_wm, int *cursor_wm) +{ + struct drm_crtc *crtc; + int hdisplay, htotal, pixel_size, clock; + unsigned long line_time_us; + int line_count, line_size; + int small, large; + int entries; + + if (!latency_ns) { + *display_wm = *cursor_wm = 0; + return false; + } + + crtc = intel_get_crtc_for_plane(dev, plane); + hdisplay = crtc->mode.hdisplay; + htotal = crtc->mode.htotal; + clock = crtc->mode.clock; + pixel_size = crtc->fb->bits_per_pixel / 8; + + line_time_us = (htotal * 1000) / clock; + line_count = (latency_ns / line_time_us + 1000) / 1000; + line_size = hdisplay * pixel_size; + + /* Use the minimum of the small and large buffer method for primary */ + small = ((clock * pixel_size / 1000) * latency_ns) / 1000; + large = line_count * line_size; + + entries = DIV_ROUND_UP(min(small, large), display->cacheline_size); + *display_wm = entries + display->guard_size; + + /* calculate the self-refresh watermark for display cursor */ + entries = line_count * pixel_size * 64; + entries = DIV_ROUND_UP(entries, cursor->cacheline_size); + *cursor_wm = entries + cursor->guard_size; + + return g4x_check_srwm(dev, + *display_wm, *cursor_wm, + display, cursor); +} + +static bool vlv_compute_drain_latency(struct drm_device *dev, + int plane, + int *plane_prec_mult, + int *plane_dl, + int *cursor_prec_mult, + int *cursor_dl) +{ + struct drm_crtc *crtc; + int clock, pixel_size; + int entries; + + crtc = intel_get_crtc_for_plane(dev, plane); + if (crtc->fb == NULL || !crtc->enabled) + return false; + + clock = crtc->mode.clock; /* VESA DOT Clock */ + pixel_size = crtc->fb->bits_per_pixel / 8; /* BPP */ + + entries = (clock / 1000) * pixel_size; + *plane_prec_mult = (entries > 256) ? + DRAIN_LATENCY_PRECISION_32 : DRAIN_LATENCY_PRECISION_16; + *plane_dl = (64 * (*plane_prec_mult) * 4) / ((clock / 1000) * + pixel_size); + + entries = (clock / 1000) * 4; /* BPP is always 4 for cursor */ + *cursor_prec_mult = (entries > 256) ? + DRAIN_LATENCY_PRECISION_32 : DRAIN_LATENCY_PRECISION_16; + *cursor_dl = (64 * (*cursor_prec_mult) * 4) / ((clock / 1000) * 4); + + return true; +} + +/* + * Update drain latency registers of memory arbiter + * + * Valleyview SoC has a new memory arbiter and needs drain latency registers + * to be programmed. Each plane has a drain latency multiplier and a drain + * latency value. + */ + +static void vlv_update_drain_latency(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + int planea_prec, planea_dl, planeb_prec, planeb_dl; + int cursora_prec, cursora_dl, cursorb_prec, cursorb_dl; + int plane_prec_mult, cursor_prec_mult; /* Precision multiplier is + either 16 or 32 */ + + /* For plane A, Cursor A */ + if (vlv_compute_drain_latency(dev, 0, &plane_prec_mult, &planea_dl, + &cursor_prec_mult, &cursora_dl)) { + cursora_prec = (cursor_prec_mult == DRAIN_LATENCY_PRECISION_32) ? + DDL_CURSORA_PRECISION_32 : DDL_CURSORA_PRECISION_16; + planea_prec = (plane_prec_mult == DRAIN_LATENCY_PRECISION_32) ? + DDL_PLANEA_PRECISION_32 : DDL_PLANEA_PRECISION_16; + + I915_WRITE(VLV_DDL1, cursora_prec | + (cursora_dl << DDL_CURSORA_SHIFT) | + planea_prec | planea_dl); + } + + /* For plane B, Cursor B */ + if (vlv_compute_drain_latency(dev, 1, &plane_prec_mult, &planeb_dl, + &cursor_prec_mult, &cursorb_dl)) { + cursorb_prec = (cursor_prec_mult == DRAIN_LATENCY_PRECISION_32) ? + DDL_CURSORB_PRECISION_32 : DDL_CURSORB_PRECISION_16; + planeb_prec = (plane_prec_mult == DRAIN_LATENCY_PRECISION_32) ? + DDL_PLANEB_PRECISION_32 : DDL_PLANEB_PRECISION_16; + + I915_WRITE(VLV_DDL2, cursorb_prec | + (cursorb_dl << DDL_CURSORB_SHIFT) | + planeb_prec | planeb_dl); + } +} + +#define single_plane_enabled(mask) is_power_of_2(mask) + +static void valleyview_update_wm(struct drm_device *dev) +{ + static const int sr_latency_ns = 12000; + struct drm_i915_private *dev_priv = dev->dev_private; + int planea_wm, planeb_wm, cursora_wm, cursorb_wm; + int plane_sr, cursor_sr; + unsigned int enabled = 0; + + vlv_update_drain_latency(dev); + + if (g4x_compute_wm0(dev, 0, + &valleyview_wm_info, latency_ns, + &valleyview_cursor_wm_info, latency_ns, + &planea_wm, &cursora_wm)) + enabled |= 1; + + if (g4x_compute_wm0(dev, 1, + &valleyview_wm_info, latency_ns, + &valleyview_cursor_wm_info, latency_ns, + &planeb_wm, &cursorb_wm)) + enabled |= 2; + + plane_sr = cursor_sr = 0; + if (single_plane_enabled(enabled) && + g4x_compute_srwm(dev, ffs(enabled) - 1, + sr_latency_ns, + &valleyview_wm_info, + &valleyview_cursor_wm_info, + &plane_sr, &cursor_sr)) + I915_WRITE(FW_BLC_SELF_VLV, FW_CSPWRDWNEN); + else + I915_WRITE(FW_BLC_SELF_VLV, + I915_READ(FW_BLC_SELF_VLV) & ~FW_CSPWRDWNEN); + + DRM_DEBUG_KMS("Setting FIFO watermarks - A: plane=%d, cursor=%d, B: plane=%d, cursor=%d, SR: plane=%d, cursor=%d\n", + planea_wm, cursora_wm, + planeb_wm, cursorb_wm, + plane_sr, cursor_sr); + + I915_WRITE(DSPFW1, + (plane_sr << DSPFW_SR_SHIFT) | + (cursorb_wm << DSPFW_CURSORB_SHIFT) | + (planeb_wm << DSPFW_PLANEB_SHIFT) | + planea_wm); + I915_WRITE(DSPFW2, + (I915_READ(DSPFW2) & DSPFW_CURSORA_MASK) | + (cursora_wm << DSPFW_CURSORA_SHIFT)); + I915_WRITE(DSPFW3, + (I915_READ(DSPFW3) | (cursor_sr << DSPFW_CURSOR_SR_SHIFT))); +} + +static void g4x_update_wm(struct drm_device *dev) +{ + static const int sr_latency_ns = 12000; + struct drm_i915_private *dev_priv = dev->dev_private; + int planea_wm, planeb_wm, cursora_wm, cursorb_wm; + int plane_sr, cursor_sr; + unsigned int enabled = 0; + + if (g4x_compute_wm0(dev, 0, + &g4x_wm_info, latency_ns, + &g4x_cursor_wm_info, latency_ns, + &planea_wm, &cursora_wm)) + enabled |= 1; + + if (g4x_compute_wm0(dev, 1, + &g4x_wm_info, latency_ns, + &g4x_cursor_wm_info, latency_ns, + &planeb_wm, &cursorb_wm)) + enabled |= 2; + + plane_sr = cursor_sr = 0; + if (single_plane_enabled(enabled) && + g4x_compute_srwm(dev, ffs(enabled) - 1, + sr_latency_ns, + &g4x_wm_info, + &g4x_cursor_wm_info, + &plane_sr, &cursor_sr)) + I915_WRITE(FW_BLC_SELF, FW_BLC_SELF_EN); + else + I915_WRITE(FW_BLC_SELF, + I915_READ(FW_BLC_SELF) & ~FW_BLC_SELF_EN); + + DRM_DEBUG_KMS("Setting FIFO watermarks - A: plane=%d, cursor=%d, B: plane=%d, cursor=%d, SR: plane=%d, cursor=%d\n", + planea_wm, cursora_wm, + planeb_wm, cursorb_wm, + plane_sr, cursor_sr); + + I915_WRITE(DSPFW1, + (plane_sr << DSPFW_SR_SHIFT) | + (cursorb_wm << DSPFW_CURSORB_SHIFT) | + (planeb_wm << DSPFW_PLANEB_SHIFT) | + planea_wm); + I915_WRITE(DSPFW2, + (I915_READ(DSPFW2) & DSPFW_CURSORA_MASK) | + (cursora_wm << DSPFW_CURSORA_SHIFT)); + /* HPLL off in SR has some issues on G4x... disable it */ + I915_WRITE(DSPFW3, + (I915_READ(DSPFW3) & ~DSPFW_HPLL_SR_EN) | + (cursor_sr << DSPFW_CURSOR_SR_SHIFT)); +} + +static void i965_update_wm(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + struct drm_crtc *crtc; + int srwm = 1; + int cursor_sr = 16; + + /* Calc sr entries for one plane configs */ + crtc = single_enabled_crtc(dev); + if (crtc) { + /* self-refresh has much higher latency */ + static const int sr_latency_ns = 12000; + int clock = crtc->mode.clock; + int htotal = crtc->mode.htotal; + int hdisplay = crtc->mode.hdisplay; + int pixel_size = crtc->fb->bits_per_pixel / 8; + unsigned long line_time_us; + int entries; + + line_time_us = ((htotal * 1000) / clock); + + /* Use ns/us then divide to preserve precision */ + entries = (((sr_latency_ns / line_time_us) + 1000) / 1000) * + pixel_size * hdisplay; + entries = DIV_ROUND_UP(entries, I915_FIFO_LINE_SIZE); + srwm = I965_FIFO_SIZE - entries; + if (srwm < 0) + srwm = 1; + srwm &= 0x1ff; + DRM_DEBUG_KMS("self-refresh entries: %d, wm: %d\n", + entries, srwm); + + entries = (((sr_latency_ns / line_time_us) + 1000) / 1000) * + pixel_size * 64; + entries = DIV_ROUND_UP(entries, + i965_cursor_wm_info.cacheline_size); + cursor_sr = i965_cursor_wm_info.fifo_size - + (entries + i965_cursor_wm_info.guard_size); + + if (cursor_sr > i965_cursor_wm_info.max_wm) + cursor_sr = i965_cursor_wm_info.max_wm; + + DRM_DEBUG_KMS("self-refresh watermark: display plane %d " + "cursor %d\n", srwm, cursor_sr); + + if (IS_CRESTLINE(dev)) + I915_WRITE(FW_BLC_SELF, FW_BLC_SELF_EN); + } else { + /* Turn off self refresh if both pipes are enabled */ + if (IS_CRESTLINE(dev)) + I915_WRITE(FW_BLC_SELF, I915_READ(FW_BLC_SELF) + & ~FW_BLC_SELF_EN); + } + + DRM_DEBUG_KMS("Setting FIFO watermarks - A: 8, B: 8, C: 8, SR %d\n", + srwm); + + /* 965 has limitations... */ + I915_WRITE(DSPFW1, (srwm << DSPFW_SR_SHIFT) | + (8 << 16) | (8 << 8) | (8 << 0)); + I915_WRITE(DSPFW2, (8 << 8) | (8 << 0)); + /* update cursor SR watermark */ + I915_WRITE(DSPFW3, (cursor_sr << DSPFW_CURSOR_SR_SHIFT)); +} + +static void i9xx_update_wm(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + const struct intel_watermark_params *wm_info; + uint32_t fwater_lo; + uint32_t fwater_hi; + int cwm, srwm = 1; + int fifo_size; + int planea_wm, planeb_wm; + struct drm_crtc *crtc, *enabled = NULL; + + if (IS_I945GM(dev)) + wm_info = &i945_wm_info; + else if (!IS_GEN2(dev)) + wm_info = &i915_wm_info; + else + wm_info = &i855_wm_info; + + fifo_size = dev_priv->display.get_fifo_size(dev, 0); + crtc = intel_get_crtc_for_plane(dev, 0); + if (crtc->enabled && crtc->fb) { + planea_wm = intel_calculate_wm(crtc->mode.clock, + wm_info, fifo_size, + crtc->fb->bits_per_pixel / 8, + latency_ns); + enabled = crtc; + } else + planea_wm = fifo_size - wm_info->guard_size; + + fifo_size = dev_priv->display.get_fifo_size(dev, 1); + crtc = intel_get_crtc_for_plane(dev, 1); + if (crtc->enabled && crtc->fb) { + planeb_wm = intel_calculate_wm(crtc->mode.clock, + wm_info, fifo_size, + crtc->fb->bits_per_pixel / 8, + latency_ns); + if (enabled == NULL) + enabled = crtc; + else + enabled = NULL; + } else + planeb_wm = fifo_size - wm_info->guard_size; + + DRM_DEBUG_KMS("FIFO watermarks - A: %d, B: %d\n", planea_wm, planeb_wm); + + /* + * Overlay gets an aggressive default since video jitter is bad. + */ + cwm = 2; + + /* Play safe and disable self-refresh before adjusting watermarks. */ + if (IS_I945G(dev) || IS_I945GM(dev)) + I915_WRITE(FW_BLC_SELF, FW_BLC_SELF_EN_MASK | 0); + else if (IS_I915GM(dev)) + I915_WRITE(INSTPM, I915_READ(INSTPM) & ~INSTPM_SELF_EN); + + /* Calc sr entries for one plane configs */ + if (HAS_FW_BLC(dev) && enabled) { + /* self-refresh has much higher latency */ + static const int sr_latency_ns = 6000; + int clock = enabled->mode.clock; + int htotal = enabled->mode.htotal; + int hdisplay = enabled->mode.hdisplay; + int pixel_size = enabled->fb->bits_per_pixel / 8; + unsigned long line_time_us; + int entries; + + line_time_us = (htotal * 1000) / clock; + + /* Use ns/us then divide to preserve precision */ + entries = (((sr_latency_ns / line_time_us) + 1000) / 1000) * + pixel_size * hdisplay; + entries = DIV_ROUND_UP(entries, wm_info->cacheline_size); + DRM_DEBUG_KMS("self-refresh entries: %d\n", entries); + srwm = wm_info->fifo_size - entries; + if (srwm < 0) + srwm = 1; + + if (IS_I945G(dev) || IS_I945GM(dev)) + I915_WRITE(FW_BLC_SELF, + FW_BLC_SELF_FIFO_MASK | (srwm & 0xff)); + else if (IS_I915GM(dev)) + I915_WRITE(FW_BLC_SELF, srwm & 0x3f); + } + + DRM_DEBUG_KMS("Setting FIFO watermarks - A: %d, B: %d, C: %d, SR %d\n", + planea_wm, planeb_wm, cwm, srwm); + + fwater_lo = ((planeb_wm & 0x3f) << 16) | (planea_wm & 0x3f); + fwater_hi = (cwm & 0x1f); + + /* Set request length to 8 cachelines per fetch */ + fwater_lo = fwater_lo | (1 << 24) | (1 << 8); + fwater_hi = fwater_hi | (1 << 8); + + I915_WRITE(FW_BLC, fwater_lo); + I915_WRITE(FW_BLC2, fwater_hi); + + if (HAS_FW_BLC(dev)) { + if (enabled) { + if (IS_I945G(dev) || IS_I945GM(dev)) + I915_WRITE(FW_BLC_SELF, + FW_BLC_SELF_EN_MASK | FW_BLC_SELF_EN); + else if (IS_I915GM(dev)) + I915_WRITE(INSTPM, I915_READ(INSTPM) | INSTPM_SELF_EN); + DRM_DEBUG_KMS("memory self refresh enabled\n"); + } else + DRM_DEBUG_KMS("memory self refresh disabled\n"); + } +} + +static void i830_update_wm(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + struct drm_crtc *crtc; + uint32_t fwater_lo; + int planea_wm; + + crtc = single_enabled_crtc(dev); + if (crtc == NULL) + return; + + planea_wm = intel_calculate_wm(crtc->mode.clock, &i830_wm_info, + dev_priv->display.get_fifo_size(dev, 0), + crtc->fb->bits_per_pixel / 8, + latency_ns); + fwater_lo = I915_READ(FW_BLC) & ~0xfff; + fwater_lo |= (3<<8) | planea_wm; + + DRM_DEBUG_KMS("Setting FIFO watermarks - A: %d\n", planea_wm); + + I915_WRITE(FW_BLC, fwater_lo); +} + +#define ILK_LP0_PLANE_LATENCY 700 +#define ILK_LP0_CURSOR_LATENCY 1300 + +/* + * Check the wm result. + * + * If any calculated watermark values is larger than the maximum value that + * can be programmed into the associated watermark register, that watermark + * must be disabled. + */ +static bool ironlake_check_srwm(struct drm_device *dev, int level, + int fbc_wm, int display_wm, int cursor_wm, + const struct intel_watermark_params *display, + const struct intel_watermark_params *cursor) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + + DRM_DEBUG_KMS("watermark %d: display plane %d, fbc lines %d," + " cursor %d\n", level, display_wm, fbc_wm, cursor_wm); + + if (fbc_wm > SNB_FBC_MAX_SRWM) { + DRM_DEBUG_KMS("fbc watermark(%d) is too large(%d), disabling wm%d+\n", + fbc_wm, SNB_FBC_MAX_SRWM, level); + + /* fbc has it's own way to disable FBC WM */ + I915_WRITE(DISP_ARB_CTL, + I915_READ(DISP_ARB_CTL) | DISP_FBC_WM_DIS); + return false; + } + + if (display_wm > display->max_wm) { + DRM_DEBUG_KMS("display watermark(%d) is too large(%d), disabling wm%d+\n", + display_wm, SNB_DISPLAY_MAX_SRWM, level); + return false; + } + + if (cursor_wm > cursor->max_wm) { + DRM_DEBUG_KMS("cursor watermark(%d) is too large(%d), disabling wm%d+\n", + cursor_wm, SNB_CURSOR_MAX_SRWM, level); + return false; + } + + if (!(fbc_wm || display_wm || cursor_wm)) { + DRM_DEBUG_KMS("latency %d is 0, disabling wm%d+\n", level, level); + return false; + } + + return true; +} + +/* + * Compute watermark values of WM[1-3], + */ +static bool ironlake_compute_srwm(struct drm_device *dev, int level, int plane, + int latency_ns, + const struct intel_watermark_params *display, + const struct intel_watermark_params *cursor, + int *fbc_wm, int *display_wm, int *cursor_wm) +{ + struct drm_crtc *crtc; + unsigned long line_time_us; + int hdisplay, htotal, pixel_size, clock; + int line_count, line_size; + int small, large; + int entries; + + if (!latency_ns) { + *fbc_wm = *display_wm = *cursor_wm = 0; + return false; + } + + crtc = intel_get_crtc_for_plane(dev, plane); + hdisplay = crtc->mode.hdisplay; + htotal = crtc->mode.htotal; + clock = crtc->mode.clock; + pixel_size = crtc->fb->bits_per_pixel / 8; + + line_time_us = (htotal * 1000) / clock; + line_count = (latency_ns / line_time_us + 1000) / 1000; + line_size = hdisplay * pixel_size; + + /* Use the minimum of the small and large buffer method for primary */ + small = ((clock * pixel_size / 1000) * latency_ns) / 1000; + large = line_count * line_size; + + entries = DIV_ROUND_UP(min(small, large), display->cacheline_size); + *display_wm = entries + display->guard_size; + + /* + * Spec says: + * FBC WM = ((Final Primary WM * 64) / number of bytes per line) + 2 + */ + *fbc_wm = DIV_ROUND_UP(*display_wm * 64, line_size) + 2; + + /* calculate the self-refresh watermark for display cursor */ + entries = line_count * pixel_size * 64; + entries = DIV_ROUND_UP(entries, cursor->cacheline_size); + *cursor_wm = entries + cursor->guard_size; + + return ironlake_check_srwm(dev, level, + *fbc_wm, *display_wm, *cursor_wm, + display, cursor); +} + +static void ironlake_update_wm(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + int fbc_wm, plane_wm, cursor_wm; + unsigned int enabled; + + enabled = 0; + if (g4x_compute_wm0(dev, 0, + &ironlake_display_wm_info, + ILK_LP0_PLANE_LATENCY, + &ironlake_cursor_wm_info, + ILK_LP0_CURSOR_LATENCY, + &plane_wm, &cursor_wm)) { + I915_WRITE(WM0_PIPEA_ILK, + (plane_wm << WM0_PIPE_PLANE_SHIFT) | cursor_wm); + DRM_DEBUG_KMS("FIFO watermarks For pipe A -" + " plane %d, " "cursor: %d\n", + plane_wm, cursor_wm); + enabled |= 1; + } + + if (g4x_compute_wm0(dev, 1, + &ironlake_display_wm_info, + ILK_LP0_PLANE_LATENCY, + &ironlake_cursor_wm_info, + ILK_LP0_CURSOR_LATENCY, + &plane_wm, &cursor_wm)) { + I915_WRITE(WM0_PIPEB_ILK, + (plane_wm << WM0_PIPE_PLANE_SHIFT) | cursor_wm); + DRM_DEBUG_KMS("FIFO watermarks For pipe B -" + " plane %d, cursor: %d\n", + plane_wm, cursor_wm); + enabled |= 2; + } + + /* + * Calculate and update the self-refresh watermark only when one + * display plane is used. + */ + I915_WRITE(WM3_LP_ILK, 0); + I915_WRITE(WM2_LP_ILK, 0); + I915_WRITE(WM1_LP_ILK, 0); + + if (!single_plane_enabled(enabled)) + return; + enabled = ffs(enabled) - 1; + + /* WM1 */ + if (!ironlake_compute_srwm(dev, 1, enabled, + ILK_READ_WM1_LATENCY() * 500, + &ironlake_display_srwm_info, + &ironlake_cursor_srwm_info, + &fbc_wm, &plane_wm, &cursor_wm)) + return; + + I915_WRITE(WM1_LP_ILK, + WM1_LP_SR_EN | + (ILK_READ_WM1_LATENCY() << WM1_LP_LATENCY_SHIFT) | + (fbc_wm << WM1_LP_FBC_SHIFT) | + (plane_wm << WM1_LP_SR_SHIFT) | + cursor_wm); + + /* WM2 */ + if (!ironlake_compute_srwm(dev, 2, enabled, + ILK_READ_WM2_LATENCY() * 500, + &ironlake_display_srwm_info, + &ironlake_cursor_srwm_info, + &fbc_wm, &plane_wm, &cursor_wm)) + return; + + I915_WRITE(WM2_LP_ILK, + WM2_LP_EN | + (ILK_READ_WM2_LATENCY() << WM1_LP_LATENCY_SHIFT) | + (fbc_wm << WM1_LP_FBC_SHIFT) | + (plane_wm << WM1_LP_SR_SHIFT) | + cursor_wm); + + /* + * WM3 is unsupported on ILK, probably because we don't have latency + * data for that power state + */ +} + +static void sandybridge_update_wm(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + int latency = SNB_READ_WM0_LATENCY() * 100; /* In unit 0.1us */ + u32 val; + int fbc_wm, plane_wm, cursor_wm; + unsigned int enabled; + + enabled = 0; + if (g4x_compute_wm0(dev, 0, + &sandybridge_display_wm_info, latency, + &sandybridge_cursor_wm_info, latency, + &plane_wm, &cursor_wm)) { + val = I915_READ(WM0_PIPEA_ILK); + val &= ~(WM0_PIPE_PLANE_MASK | WM0_PIPE_CURSOR_MASK); + I915_WRITE(WM0_PIPEA_ILK, val | + ((plane_wm << WM0_PIPE_PLANE_SHIFT) | cursor_wm)); + DRM_DEBUG_KMS("FIFO watermarks For pipe A -" + " plane %d, " "cursor: %d\n", + plane_wm, cursor_wm); + enabled |= 1; + } + + if (g4x_compute_wm0(dev, 1, + &sandybridge_display_wm_info, latency, + &sandybridge_cursor_wm_info, latency, + &plane_wm, &cursor_wm)) { + val = I915_READ(WM0_PIPEB_ILK); + val &= ~(WM0_PIPE_PLANE_MASK | WM0_PIPE_CURSOR_MASK); + I915_WRITE(WM0_PIPEB_ILK, val | + ((plane_wm << WM0_PIPE_PLANE_SHIFT) | cursor_wm)); + DRM_DEBUG_KMS("FIFO watermarks For pipe B -" + " plane %d, cursor: %d\n", + plane_wm, cursor_wm); + enabled |= 2; + } + + /* IVB has 3 pipes */ + if (IS_IVYBRIDGE(dev) && + g4x_compute_wm0(dev, 2, + &sandybridge_display_wm_info, latency, + &sandybridge_cursor_wm_info, latency, + &plane_wm, &cursor_wm)) { + val = I915_READ(WM0_PIPEC_IVB); + val &= ~(WM0_PIPE_PLANE_MASK | WM0_PIPE_CURSOR_MASK); + I915_WRITE(WM0_PIPEC_IVB, val | + ((plane_wm << WM0_PIPE_PLANE_SHIFT) | cursor_wm)); + DRM_DEBUG_KMS("FIFO watermarks For pipe C -" + " plane %d, cursor: %d\n", + plane_wm, cursor_wm); + enabled |= 3; + } + + /* + * Calculate and update the self-refresh watermark only when one + * display plane is used. + * + * SNB support 3 levels of watermark. + * + * WM1/WM2/WM2 watermarks have to be enabled in the ascending order, + * and disabled in the descending order + * + */ + I915_WRITE(WM3_LP_ILK, 0); + I915_WRITE(WM2_LP_ILK, 0); + I915_WRITE(WM1_LP_ILK, 0); + + if (!single_plane_enabled(enabled) || + dev_priv->sprite_scaling_enabled) + return; + enabled = ffs(enabled) - 1; + + /* WM1 */ + if (!ironlake_compute_srwm(dev, 1, enabled, + SNB_READ_WM1_LATENCY() * 500, + &sandybridge_display_srwm_info, + &sandybridge_cursor_srwm_info, + &fbc_wm, &plane_wm, &cursor_wm)) + return; + + I915_WRITE(WM1_LP_ILK, + WM1_LP_SR_EN | + (SNB_READ_WM1_LATENCY() << WM1_LP_LATENCY_SHIFT) | + (fbc_wm << WM1_LP_FBC_SHIFT) | + (plane_wm << WM1_LP_SR_SHIFT) | + cursor_wm); + + /* WM2 */ + if (!ironlake_compute_srwm(dev, 2, enabled, + SNB_READ_WM2_LATENCY() * 500, + &sandybridge_display_srwm_info, + &sandybridge_cursor_srwm_info, + &fbc_wm, &plane_wm, &cursor_wm)) + return; + + I915_WRITE(WM2_LP_ILK, + WM2_LP_EN | + (SNB_READ_WM2_LATENCY() << WM1_LP_LATENCY_SHIFT) | + (fbc_wm << WM1_LP_FBC_SHIFT) | + (plane_wm << WM1_LP_SR_SHIFT) | + cursor_wm); + + /* WM3 */ + if (!ironlake_compute_srwm(dev, 3, enabled, + SNB_READ_WM3_LATENCY() * 500, + &sandybridge_display_srwm_info, + &sandybridge_cursor_srwm_info, + &fbc_wm, &plane_wm, &cursor_wm)) + return; + + I915_WRITE(WM3_LP_ILK, + WM3_LP_EN | + (SNB_READ_WM3_LATENCY() << WM1_LP_LATENCY_SHIFT) | + (fbc_wm << WM1_LP_FBC_SHIFT) | + (plane_wm << WM1_LP_SR_SHIFT) | + cursor_wm); +} + +static bool +sandybridge_compute_sprite_wm(struct drm_device *dev, int plane, + uint32_t sprite_width, int pixel_size, + const struct intel_watermark_params *display, + int display_latency_ns, int *sprite_wm) +{ + struct drm_crtc *crtc; + int clock; + int entries, tlb_miss; + + crtc = intel_get_crtc_for_plane(dev, plane); + if (crtc->fb == NULL || !crtc->enabled) { + *sprite_wm = display->guard_size; + return false; + } + + clock = crtc->mode.clock; + + /* Use the small buffer method to calculate the sprite watermark */ + entries = ((clock * pixel_size / 1000) * display_latency_ns) / 1000; + tlb_miss = display->fifo_size*display->cacheline_size - + sprite_width * 8; + if (tlb_miss > 0) + entries += tlb_miss; + entries = DIV_ROUND_UP(entries, display->cacheline_size); + *sprite_wm = entries + display->guard_size; + if (*sprite_wm > (int)display->max_wm) + *sprite_wm = display->max_wm; + + return true; +} + +static bool +sandybridge_compute_sprite_srwm(struct drm_device *dev, int plane, + uint32_t sprite_width, int pixel_size, + const struct intel_watermark_params *display, + int latency_ns, int *sprite_wm) +{ + struct drm_crtc *crtc; + unsigned long line_time_us; + int clock; + int line_count, line_size; + int small, large; + int entries; + + if (!latency_ns) { + *sprite_wm = 0; + return false; + } + + crtc = intel_get_crtc_for_plane(dev, plane); + clock = crtc->mode.clock; + if (!clock) { + *sprite_wm = 0; + return false; + } + + line_time_us = (sprite_width * 1000) / clock; + if (!line_time_us) { + *sprite_wm = 0; + return false; + } + + line_count = (latency_ns / line_time_us + 1000) / 1000; + line_size = sprite_width * pixel_size; + + /* Use the minimum of the small and large buffer method for primary */ + small = ((clock * pixel_size / 1000) * latency_ns) / 1000; + large = line_count * line_size; + + entries = DIV_ROUND_UP(min(small, large), display->cacheline_size); + *sprite_wm = entries + display->guard_size; + + return *sprite_wm > 0x3ff ? false : true; +} + +static void sandybridge_update_sprite_wm(struct drm_device *dev, int pipe, + uint32_t sprite_width, int pixel_size) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + int latency = SNB_READ_WM0_LATENCY() * 100; /* In unit 0.1us */ + u32 val; + int sprite_wm, reg; + int ret; + + switch (pipe) { + case 0: + reg = WM0_PIPEA_ILK; + break; + case 1: + reg = WM0_PIPEB_ILK; + break; + case 2: + reg = WM0_PIPEC_IVB; + break; + default: + return; /* bad pipe */ + } + + ret = sandybridge_compute_sprite_wm(dev, pipe, sprite_width, pixel_size, + &sandybridge_display_wm_info, + latency, &sprite_wm); + if (!ret) { + DRM_DEBUG_KMS("failed to compute sprite wm for pipe %d\n", + pipe); + return; + } + + val = I915_READ(reg); + val &= ~WM0_PIPE_SPRITE_MASK; + I915_WRITE(reg, val | (sprite_wm << WM0_PIPE_SPRITE_SHIFT)); + DRM_DEBUG_KMS("sprite watermarks For pipe %d - %d\n", pipe, sprite_wm); + + + ret = sandybridge_compute_sprite_srwm(dev, pipe, sprite_width, + pixel_size, + &sandybridge_display_srwm_info, + SNB_READ_WM1_LATENCY() * 500, + &sprite_wm); + if (!ret) { + DRM_DEBUG_KMS("failed to compute sprite lp1 wm on pipe %d\n", + pipe); + return; + } + I915_WRITE(WM1S_LP_ILK, sprite_wm); + + /* Only IVB has two more LP watermarks for sprite */ + if (!IS_IVYBRIDGE(dev)) + return; + + ret = sandybridge_compute_sprite_srwm(dev, pipe, sprite_width, + pixel_size, + &sandybridge_display_srwm_info, + SNB_READ_WM2_LATENCY() * 500, + &sprite_wm); + if (!ret) { + DRM_DEBUG_KMS("failed to compute sprite lp2 wm on pipe %d\n", + pipe); + return; + } + I915_WRITE(WM2S_LP_IVB, sprite_wm); + + ret = sandybridge_compute_sprite_srwm(dev, pipe, sprite_width, + pixel_size, + &sandybridge_display_srwm_info, + SNB_READ_WM3_LATENCY() * 500, + &sprite_wm); + if (!ret) { + DRM_DEBUG_KMS("failed to compute sprite lp3 wm on pipe %d\n", + pipe); + return; + } + I915_WRITE(WM3S_LP_IVB, sprite_wm); +} + +/** + * intel_update_watermarks - update FIFO watermark values based on current modes + * + * Calculate watermark values for the various WM regs based on current mode + * and plane configuration. + * + * There are several cases to deal with here: + * - normal (i.e. non-self-refresh) + * - self-refresh (SR) mode + * - lines are large relative to FIFO size (buffer can hold up to 2) + * - lines are small relative to FIFO size (buffer can hold more than 2 + * lines), so need to account for TLB latency + * + * The normal calculation is: + * watermark = dotclock * bytes per pixel * latency + * where latency is platform & configuration dependent (we assume pessimal + * values here). + * + * The SR calculation is: + * watermark = (trunc(latency/line time)+1) * surface width * + * bytes per pixel + * where + * line time = htotal / dotclock + * surface width = hdisplay for normal plane and 64 for cursor + * and latency is assumed to be high, as above. + * + * The final value programmed to the register should always be rounded up, + * and include an extra 2 entries to account for clock crossings. + * + * We don't use the sprite, so we can ignore that. And on Crestline we have + * to set the non-SR watermarks to 8. + */ +void intel_update_watermarks(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + + if (dev_priv->display.update_wm) + dev_priv->display.update_wm(dev); +} + +void intel_update_sprite_watermarks(struct drm_device *dev, int pipe, + uint32_t sprite_width, int pixel_size) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + + if (dev_priv->display.update_sprite_wm) + dev_priv->display.update_sprite_wm(dev, pipe, sprite_width, + pixel_size); +} + +static struct drm_i915_gem_object * +intel_alloc_context_page(struct drm_device *dev) +{ + struct drm_i915_gem_object *ctx; + int ret; + + WARN_ON(!mutex_is_locked(&dev->struct_mutex)); + + ctx = i915_gem_alloc_object(dev, 4096); + if (!ctx) { + DRM_DEBUG("failed to alloc power context, RC6 disabled\n"); + return NULL; + } + + ret = i915_gem_object_pin(ctx, 4096, true); + if (ret) { + DRM_ERROR("failed to pin power context: %d\n", ret); + goto err_unref; + } + + ret = i915_gem_object_set_to_gtt_domain(ctx, 1); + if (ret) { + DRM_ERROR("failed to set-domain on power context: %d\n", ret); + goto err_unpin; + } + + return ctx; + +err_unpin: + i915_gem_object_unpin(ctx); +err_unref: + drm_gem_object_unreference(&ctx->base); + mutex_unlock(&dev->struct_mutex); + return NULL; +} + +bool ironlake_set_drps(struct drm_device *dev, u8 val) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + u16 rgvswctl; + + rgvswctl = I915_READ16(MEMSWCTL); + if (rgvswctl & MEMCTL_CMD_STS) { + DRM_DEBUG("gpu busy, RCS change rejected\n"); + return false; /* still busy with another command */ + } + + rgvswctl = (MEMCTL_CMD_CHFREQ << MEMCTL_CMD_SHIFT) | + (val << MEMCTL_FREQ_SHIFT) | MEMCTL_SFCAVM; + I915_WRITE16(MEMSWCTL, rgvswctl); + POSTING_READ16(MEMSWCTL); + + rgvswctl |= MEMCTL_CMD_STS; + I915_WRITE16(MEMSWCTL, rgvswctl); + + return true; +} + +void ironlake_enable_drps(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + u32 rgvmodectl = I915_READ(MEMMODECTL); + u8 fmax, fmin, fstart, vstart; + + /* Enable temp reporting */ + I915_WRITE16(PMMISC, I915_READ(PMMISC) | MCPPCE_EN); + I915_WRITE16(TSC1, I915_READ(TSC1) | TSE); + + /* 100ms RC evaluation intervals */ + I915_WRITE(RCUPEI, 100000); + I915_WRITE(RCDNEI, 100000); + + /* Set max/min thresholds to 90ms and 80ms respectively */ + I915_WRITE(RCBMAXAVG, 90000); + I915_WRITE(RCBMINAVG, 80000); + + I915_WRITE(MEMIHYST, 1); + + /* Set up min, max, and cur for interrupt handling */ + fmax = (rgvmodectl & MEMMODE_FMAX_MASK) >> MEMMODE_FMAX_SHIFT; + fmin = (rgvmodectl & MEMMODE_FMIN_MASK); + fstart = (rgvmodectl & MEMMODE_FSTART_MASK) >> + MEMMODE_FSTART_SHIFT; + + vstart = (I915_READ(PXVFREQ_BASE + (fstart * 4)) & PXVFREQ_PX_MASK) >> + PXVFREQ_PX_SHIFT; + + dev_priv->fmax = fmax; /* IPS callback will increase this */ + dev_priv->fstart = fstart; + + dev_priv->max_delay = fstart; + dev_priv->min_delay = fmin; + dev_priv->cur_delay = fstart; + + DRM_DEBUG_DRIVER("fmax: %d, fmin: %d, fstart: %d\n", + fmax, fmin, fstart); + + I915_WRITE(MEMINTREN, MEMINT_CX_SUPR_EN | MEMINT_EVAL_CHG_EN); + + /* + * Interrupts will be enabled in ironlake_irq_postinstall + */ + + I915_WRITE(VIDSTART, vstart); + POSTING_READ(VIDSTART); + + rgvmodectl |= MEMMODE_SWMODE_EN; + I915_WRITE(MEMMODECTL, rgvmodectl); + + if (wait_for((I915_READ(MEMSWCTL) & MEMCTL_CMD_STS) == 0, 10)) + DRM_ERROR("stuck trying to change perf mode\n"); + msleep(1); + + ironlake_set_drps(dev, fstart); + + dev_priv->last_count1 = I915_READ(0x112e4) + I915_READ(0x112e8) + + I915_READ(0x112e0); + dev_priv->last_time1 = jiffies_to_msecs(jiffies); + dev_priv->last_count2 = I915_READ(0x112f4); + getrawmonotonic(&dev_priv->last_time2); +} + +void ironlake_disable_drps(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + u16 rgvswctl = I915_READ16(MEMSWCTL); + + /* Ack interrupts, disable EFC interrupt */ + I915_WRITE(MEMINTREN, I915_READ(MEMINTREN) & ~MEMINT_EVAL_CHG_EN); + I915_WRITE(MEMINTRSTS, MEMINT_EVAL_CHG); + I915_WRITE(DEIER, I915_READ(DEIER) & ~DE_PCU_EVENT); + I915_WRITE(DEIIR, DE_PCU_EVENT); + I915_WRITE(DEIMR, I915_READ(DEIMR) | DE_PCU_EVENT); + + /* Go back to the starting frequency */ + ironlake_set_drps(dev, dev_priv->fstart); + msleep(1); + rgvswctl |= MEMCTL_CMD_STS; + I915_WRITE(MEMSWCTL, rgvswctl); + msleep(1); + +} + +void gen6_set_rps(struct drm_device *dev, u8 val) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + u32 swreq; + + swreq = (val & 0x3ff) << 25; + I915_WRITE(GEN6_RPNSWREQ, swreq); +} + +void gen6_disable_rps(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + + I915_WRITE(GEN6_RPNSWREQ, 1 << 31); + I915_WRITE(GEN6_PMINTRMSK, 0xffffffff); + I915_WRITE(GEN6_PMIER, 0); + /* Complete PM interrupt masking here doesn't race with the rps work + * item again unmasking PM interrupts because that is using a different + * register (PMIMR) to mask PM interrupts. The only risk is in leaving + * stale bits in PMIIR and PMIMR which gen6_enable_rps will clean up. */ + + spin_lock_irq(&dev_priv->rps_lock); + dev_priv->pm_iir = 0; + spin_unlock_irq(&dev_priv->rps_lock); + + I915_WRITE(GEN6_PMIIR, I915_READ(GEN6_PMIIR)); +} + +int intel_enable_rc6(const struct drm_device *dev) +{ + /* + * Respect the kernel parameter if it is set + */ + if (i915_enable_rc6 >= 0) + return i915_enable_rc6; + + /* + * Disable RC6 on Ironlake + */ + if (INTEL_INFO(dev)->gen == 5) + return 0; + + /* Sorry Haswell, no RC6 for you for now. */ + if (IS_HASWELL(dev)) + return 0; + + /* + * Disable rc6 on Sandybridge + */ + if (INTEL_INFO(dev)->gen == 6) { + DRM_DEBUG_DRIVER("Sandybridge: deep RC6 disabled\n"); + return INTEL_RC6_ENABLE; + } + DRM_DEBUG_DRIVER("RC6 and deep RC6 enabled\n"); + return (INTEL_RC6_ENABLE | INTEL_RC6p_ENABLE); +} + +void gen6_enable_rps(struct drm_i915_private *dev_priv) +{ + u32 rp_state_cap = I915_READ(GEN6_RP_STATE_CAP); + u32 gt_perf_status = I915_READ(GEN6_GT_PERF_STATUS); + u32 pcu_mbox, rc6_mask = 0; + u32 gtfifodbg; + int cur_freq, min_freq, max_freq; + int rc6_mode; + int i; + + /* Here begins a magic sequence of register writes to enable + * auto-downclocking. + * + * Perhaps there might be some value in exposing these to + * userspace... + */ + I915_WRITE(GEN6_RC_STATE, 0); + mutex_lock(&dev_priv->dev->struct_mutex); + + /* Clear the DBG now so we don't confuse earlier errors */ + if ((gtfifodbg = I915_READ(GTFIFODBG))) { + DRM_ERROR("GT fifo had a previous error %x\n", gtfifodbg); + I915_WRITE(GTFIFODBG, gtfifodbg); + } + + gen6_gt_force_wake_get(dev_priv); + + /* disable the counters and set deterministic thresholds */ + I915_WRITE(GEN6_RC_CONTROL, 0); + + I915_WRITE(GEN6_RC1_WAKE_RATE_LIMIT, 1000 << 16); + I915_WRITE(GEN6_RC6_WAKE_RATE_LIMIT, 40 << 16 | 30); + I915_WRITE(GEN6_RC6pp_WAKE_RATE_LIMIT, 30); + I915_WRITE(GEN6_RC_EVALUATION_INTERVAL, 125000); + I915_WRITE(GEN6_RC_IDLE_HYSTERSIS, 25); + + for (i = 0; i < I915_NUM_RINGS; i++) + I915_WRITE(RING_MAX_IDLE(dev_priv->ring[i].mmio_base), 10); + + I915_WRITE(GEN6_RC_SLEEP, 0); + I915_WRITE(GEN6_RC1e_THRESHOLD, 1000); + I915_WRITE(GEN6_RC6_THRESHOLD, 50000); + I915_WRITE(GEN6_RC6p_THRESHOLD, 100000); + I915_WRITE(GEN6_RC6pp_THRESHOLD, 64000); /* unused */ + + rc6_mode = intel_enable_rc6(dev_priv->dev); + if (rc6_mode & INTEL_RC6_ENABLE) + rc6_mask |= GEN6_RC_CTL_RC6_ENABLE; + + if (rc6_mode & INTEL_RC6p_ENABLE) + rc6_mask |= GEN6_RC_CTL_RC6p_ENABLE; + + if (rc6_mode & INTEL_RC6pp_ENABLE) + rc6_mask |= GEN6_RC_CTL_RC6pp_ENABLE; + + DRM_INFO("Enabling RC6 states: RC6 %s, RC6p %s, RC6pp %s\n", + (rc6_mode & INTEL_RC6_ENABLE) ? "on" : "off", + (rc6_mode & INTEL_RC6p_ENABLE) ? "on" : "off", + (rc6_mode & INTEL_RC6pp_ENABLE) ? "on" : "off"); + + I915_WRITE(GEN6_RC_CONTROL, + rc6_mask | + GEN6_RC_CTL_EI_MODE(1) | + GEN6_RC_CTL_HW_ENABLE); + + I915_WRITE(GEN6_RPNSWREQ, + GEN6_FREQUENCY(10) | + GEN6_OFFSET(0) | + GEN6_AGGRESSIVE_TURBO); + I915_WRITE(GEN6_RC_VIDEO_FREQ, + GEN6_FREQUENCY(12)); + + I915_WRITE(GEN6_RP_DOWN_TIMEOUT, 1000000); + I915_WRITE(GEN6_RP_INTERRUPT_LIMITS, + 18 << 24 | + 6 << 16); + I915_WRITE(GEN6_RP_UP_THRESHOLD, 10000); + I915_WRITE(GEN6_RP_DOWN_THRESHOLD, 1000000); + I915_WRITE(GEN6_RP_UP_EI, 100000); + I915_WRITE(GEN6_RP_DOWN_EI, 5000000); + I915_WRITE(GEN6_RP_IDLE_HYSTERSIS, 10); + I915_WRITE(GEN6_RP_CONTROL, + GEN6_RP_MEDIA_TURBO | + GEN6_RP_MEDIA_HW_MODE | + GEN6_RP_MEDIA_IS_GFX | + GEN6_RP_ENABLE | + GEN6_RP_UP_BUSY_AVG | + GEN6_RP_DOWN_IDLE_CONT); + + if (wait_for((I915_READ(GEN6_PCODE_MAILBOX) & GEN6_PCODE_READY) == 0, + 500)) + DRM_ERROR("timeout waiting for pcode mailbox to become idle\n"); + + I915_WRITE(GEN6_PCODE_DATA, 0); + I915_WRITE(GEN6_PCODE_MAILBOX, + GEN6_PCODE_READY | + GEN6_PCODE_WRITE_MIN_FREQ_TABLE); + if (wait_for((I915_READ(GEN6_PCODE_MAILBOX) & GEN6_PCODE_READY) == 0, + 500)) + DRM_ERROR("timeout waiting for pcode mailbox to finish\n"); + + min_freq = (rp_state_cap & 0xff0000) >> 16; + max_freq = rp_state_cap & 0xff; + cur_freq = (gt_perf_status & 0xff00) >> 8; + + /* Check for overclock support */ + if (wait_for((I915_READ(GEN6_PCODE_MAILBOX) & GEN6_PCODE_READY) == 0, + 500)) + DRM_ERROR("timeout waiting for pcode mailbox to become idle\n"); + I915_WRITE(GEN6_PCODE_MAILBOX, GEN6_READ_OC_PARAMS); + pcu_mbox = I915_READ(GEN6_PCODE_DATA); + if (wait_for((I915_READ(GEN6_PCODE_MAILBOX) & GEN6_PCODE_READY) == 0, + 500)) + DRM_ERROR("timeout waiting for pcode mailbox to finish\n"); + if (pcu_mbox & (1<<31)) { /* OC supported */ + max_freq = pcu_mbox & 0xff; + DRM_DEBUG_DRIVER("overclocking supported, adjusting frequency max to %dMHz\n", pcu_mbox * 50); + } + + /* In units of 100MHz */ + dev_priv->max_delay = max_freq; + dev_priv->min_delay = min_freq; + dev_priv->cur_delay = cur_freq; + + /* requires MSI enabled */ + I915_WRITE(GEN6_PMIER, + GEN6_PM_MBOX_EVENT | + GEN6_PM_THERMAL_EVENT | + GEN6_PM_RP_DOWN_TIMEOUT | + GEN6_PM_RP_UP_THRESHOLD | + GEN6_PM_RP_DOWN_THRESHOLD | + GEN6_PM_RP_UP_EI_EXPIRED | + GEN6_PM_RP_DOWN_EI_EXPIRED); + spin_lock_irq(&dev_priv->rps_lock); + WARN_ON(dev_priv->pm_iir != 0); + I915_WRITE(GEN6_PMIMR, 0); + spin_unlock_irq(&dev_priv->rps_lock); + /* enable all PM interrupts */ + I915_WRITE(GEN6_PMINTRMSK, 0); + + gen6_gt_force_wake_put(dev_priv); + mutex_unlock(&dev_priv->dev->struct_mutex); +} + +void gen6_update_ring_freq(struct drm_i915_private *dev_priv) +{ + int min_freq = 15; + int gpu_freq, ia_freq, max_ia_freq; + int scaling_factor = 180; + + max_ia_freq = cpufreq_quick_get_max(0); + /* + * Default to measured freq if none found, PCU will ensure we don't go + * over + */ + if (!max_ia_freq) + max_ia_freq = tsc_khz; + + /* Convert from kHz to MHz */ + max_ia_freq /= 1000; + + mutex_lock(&dev_priv->dev->struct_mutex); + + /* + * For each potential GPU frequency, load a ring frequency we'd like + * to use for memory access. We do this by specifying the IA frequency + * the PCU should use as a reference to determine the ring frequency. + */ + for (gpu_freq = dev_priv->max_delay; gpu_freq >= dev_priv->min_delay; + gpu_freq--) { + int diff = dev_priv->max_delay - gpu_freq; + + /* + * For GPU frequencies less than 750MHz, just use the lowest + * ring freq. + */ + if (gpu_freq < min_freq) + ia_freq = 800; + else + ia_freq = max_ia_freq - ((diff * scaling_factor) / 2); + ia_freq = DIV_ROUND_CLOSEST(ia_freq, 100); + + I915_WRITE(GEN6_PCODE_DATA, + (ia_freq << GEN6_PCODE_FREQ_IA_RATIO_SHIFT) | + gpu_freq); + I915_WRITE(GEN6_PCODE_MAILBOX, GEN6_PCODE_READY | + GEN6_PCODE_WRITE_MIN_FREQ_TABLE); + if (wait_for((I915_READ(GEN6_PCODE_MAILBOX) & + GEN6_PCODE_READY) == 0, 10)) { + DRM_ERROR("pcode write of freq table timed out\n"); + continue; + } + } + + mutex_unlock(&dev_priv->dev->struct_mutex); +} + +static void ironlake_teardown_rc6(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + + if (dev_priv->renderctx) { + i915_gem_object_unpin(dev_priv->renderctx); + drm_gem_object_unreference(&dev_priv->renderctx->base); + dev_priv->renderctx = NULL; + } + + if (dev_priv->pwrctx) { + i915_gem_object_unpin(dev_priv->pwrctx); + drm_gem_object_unreference(&dev_priv->pwrctx->base); + dev_priv->pwrctx = NULL; + } +} + +void ironlake_disable_rc6(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + + if (I915_READ(PWRCTXA)) { + /* Wake the GPU, prevent RC6, then restore RSTDBYCTL */ + I915_WRITE(RSTDBYCTL, I915_READ(RSTDBYCTL) | RCX_SW_EXIT); + wait_for(((I915_READ(RSTDBYCTL) & RSX_STATUS_MASK) == RSX_STATUS_ON), + 50); + + I915_WRITE(PWRCTXA, 0); + POSTING_READ(PWRCTXA); + + I915_WRITE(RSTDBYCTL, I915_READ(RSTDBYCTL) & ~RCX_SW_EXIT); + POSTING_READ(RSTDBYCTL); + } + + ironlake_teardown_rc6(dev); +} + +static int ironlake_setup_rc6(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + + if (dev_priv->renderctx == NULL) + dev_priv->renderctx = intel_alloc_context_page(dev); + if (!dev_priv->renderctx) + return -ENOMEM; + + if (dev_priv->pwrctx == NULL) + dev_priv->pwrctx = intel_alloc_context_page(dev); + if (!dev_priv->pwrctx) { + ironlake_teardown_rc6(dev); + return -ENOMEM; + } + + return 0; +} + +void ironlake_enable_rc6(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + int ret; + + /* rc6 disabled by default due to repeated reports of hanging during + * boot and resume. + */ + if (!intel_enable_rc6(dev)) + return; + + mutex_lock(&dev->struct_mutex); + ret = ironlake_setup_rc6(dev); + if (ret) { + mutex_unlock(&dev->struct_mutex); + return; + } + + /* + * GPU can automatically power down the render unit if given a page + * to save state. + */ + ret = BEGIN_LP_RING(6); + if (ret) { + ironlake_teardown_rc6(dev); + mutex_unlock(&dev->struct_mutex); + return; + } + + OUT_RING(MI_SUSPEND_FLUSH | MI_SUSPEND_FLUSH_EN); + OUT_RING(MI_SET_CONTEXT); + OUT_RING(dev_priv->renderctx->gtt_offset | + MI_MM_SPACE_GTT | + MI_SAVE_EXT_STATE_EN | + MI_RESTORE_EXT_STATE_EN | + MI_RESTORE_INHIBIT); + OUT_RING(MI_SUSPEND_FLUSH); + OUT_RING(MI_NOOP); + OUT_RING(MI_FLUSH); + ADVANCE_LP_RING(); + + /* + * Wait for the command parser to advance past MI_SET_CONTEXT. The HW + * does an implicit flush, combined with MI_FLUSH above, it should be + * safe to assume that renderctx is valid + */ + ret = intel_wait_ring_idle(LP_RING(dev_priv)); + if (ret) { + DRM_ERROR("failed to enable ironlake power power savings\n"); + ironlake_teardown_rc6(dev); + mutex_unlock(&dev->struct_mutex); + return; + } + + I915_WRITE(PWRCTXA, dev_priv->pwrctx->gtt_offset | PWRCTX_EN); + I915_WRITE(RSTDBYCTL, I915_READ(RSTDBYCTL) & ~RCX_SW_EXIT); + mutex_unlock(&dev->struct_mutex); +} + +static unsigned long intel_pxfreq(u32 vidfreq) +{ + unsigned long freq; + int div = (vidfreq & 0x3f0000) >> 16; + int post = (vidfreq & 0x3000) >> 12; + int pre = (vidfreq & 0x7); + + if (!pre) + return 0; + + freq = ((div * 133333) / ((1<<post) * pre)); + + return freq; +} + +void intel_init_emon(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + u32 lcfuse; + u8 pxw[16]; + int i; + + /* Disable to program */ + I915_WRITE(ECR, 0); + POSTING_READ(ECR); + + /* Program energy weights for various events */ + I915_WRITE(SDEW, 0x15040d00); + I915_WRITE(CSIEW0, 0x007f0000); + I915_WRITE(CSIEW1, 0x1e220004); + I915_WRITE(CSIEW2, 0x04000004); + + for (i = 0; i < 5; i++) + I915_WRITE(PEW + (i * 4), 0); + for (i = 0; i < 3; i++) + I915_WRITE(DEW + (i * 4), 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)); + unsigned long freq = intel_pxfreq(pxvidfreq); + unsigned long vid = (pxvidfreq & PXVFREQ_PX_MASK) >> + PXVFREQ_PX_SHIFT; + unsigned long val; + + val = vid * vid; + val *= (freq / 1000); + val *= 255; + val /= (127*127*900); + if (val > 0xff) + DRM_ERROR("bad pxval: %ld\n", val); + pxw[i] = val; + } + /* Render standby states get 0 weight */ + pxw[14] = 0; + pxw[15] = 0; + + 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); + } + + /* Adjust magic regs to magic values (more experimental results) */ + I915_WRITE(OGW0, 0); + I915_WRITE(OGW1, 0); + I915_WRITE(EG0, 0x00007f00); + I915_WRITE(EG1, 0x0000000e); + I915_WRITE(EG2, 0x000e0000); + I915_WRITE(EG3, 0x68000300); + I915_WRITE(EG4, 0x42000000); + I915_WRITE(EG5, 0x00140031); + I915_WRITE(EG6, 0); + I915_WRITE(EG7, 0); + + for (i = 0; i < 8; i++) + I915_WRITE(PXWL + (i * 4), 0); + + /* Enable PMON + select events */ + I915_WRITE(ECR, 0x80000019); + + lcfuse = I915_READ(LCFUSE02); + + dev_priv->corr = (lcfuse & LCFUSE_HIV_MASK); +} + +static void ironlake_init_clock_gating(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + uint32_t dspclk_gate = VRHUNIT_CLOCK_GATE_DISABLE; + + /* Required for FBC */ + dspclk_gate |= DPFCUNIT_CLOCK_GATE_DISABLE | + DPFCRUNIT_CLOCK_GATE_DISABLE | + DPFDUNIT_CLOCK_GATE_DISABLE; + /* Required for CxSR */ + dspclk_gate |= DPARBUNIT_CLOCK_GATE_DISABLE; + + I915_WRITE(PCH_3DCGDIS0, + MARIUNIT_CLOCK_GATE_DISABLE | + SVSMUNIT_CLOCK_GATE_DISABLE); + I915_WRITE(PCH_3DCGDIS1, + VFMUNIT_CLOCK_GATE_DISABLE); + + I915_WRITE(PCH_DSPCLK_GATE_D, dspclk_gate); + + /* + * According to the spec the following bits should be set in + * order to enable memory self-refresh + * The bit 22/21 of 0x42004 + * The bit 5 of 0x42020 + * The bit 15 of 0x45000 + */ + I915_WRITE(ILK_DISPLAY_CHICKEN2, + (I915_READ(ILK_DISPLAY_CHICKEN2) | + ILK_DPARB_GATE | ILK_VSDPFD_FULL)); + I915_WRITE(ILK_DSPCLK_GATE, + (I915_READ(ILK_DSPCLK_GATE) | + ILK_DPARB_CLK_GATE)); + I915_WRITE(DISP_ARB_CTL, + (I915_READ(DISP_ARB_CTL) | + DISP_FBC_WM_DIS)); + I915_WRITE(WM3_LP_ILK, 0); + I915_WRITE(WM2_LP_ILK, 0); + I915_WRITE(WM1_LP_ILK, 0); + + /* + * Based on the document from hardware guys the following bits + * should be set unconditionally in order to enable FBC. + * The bit 22 of 0x42000 + * The bit 22 of 0x42004 + * The bit 7,8,9 of 0x42020. + */ + if (IS_IRONLAKE_M(dev)) { + I915_WRITE(ILK_DISPLAY_CHICKEN1, + I915_READ(ILK_DISPLAY_CHICKEN1) | + ILK_FBCQ_DIS); + I915_WRITE(ILK_DISPLAY_CHICKEN2, + I915_READ(ILK_DISPLAY_CHICKEN2) | + ILK_DPARB_GATE); + I915_WRITE(ILK_DSPCLK_GATE, + I915_READ(ILK_DSPCLK_GATE) | + ILK_DPFC_DIS1 | + ILK_DPFC_DIS2 | + ILK_CLK_FBC); + } + + I915_WRITE(ILK_DISPLAY_CHICKEN2, + I915_READ(ILK_DISPLAY_CHICKEN2) | + ILK_ELPIN_409_SELECT); + I915_WRITE(_3D_CHICKEN2, + _3D_CHICKEN2_WM_READ_PIPELINED << 16 | + _3D_CHICKEN2_WM_READ_PIPELINED); +} + +static void gen6_init_clock_gating(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + int pipe; + uint32_t dspclk_gate = VRHUNIT_CLOCK_GATE_DISABLE; + + I915_WRITE(PCH_DSPCLK_GATE_D, dspclk_gate); + + I915_WRITE(ILK_DISPLAY_CHICKEN2, + I915_READ(ILK_DISPLAY_CHICKEN2) | + ILK_ELPIN_409_SELECT); + + I915_WRITE(WM3_LP_ILK, 0); + I915_WRITE(WM2_LP_ILK, 0); + I915_WRITE(WM1_LP_ILK, 0); + + /* clear masked bit */ + I915_WRITE(CACHE_MODE_0, + CM0_STC_EVICT_DISABLE_LRA_SNB << CM0_MASK_SHIFT); + + I915_WRITE(GEN6_UCGCTL1, + I915_READ(GEN6_UCGCTL1) | + GEN6_BLBUNIT_CLOCK_GATE_DISABLE | + GEN6_CSUNIT_CLOCK_GATE_DISABLE); + + /* According to the BSpec vol1g, bit 12 (RCPBUNIT) clock + * gating disable must be set. Failure to set it results in + * flickering pixels due to Z write ordering failures after + * some amount of runtime in the Mesa "fire" demo, and Unigine + * Sanctuary and Tropics, and apparently anything else with + * alpha test or pixel discard. + * + * According to the spec, bit 11 (RCCUNIT) must also be set, + * but we didn't debug actual testcases to find it out. + */ + I915_WRITE(GEN6_UCGCTL2, + GEN6_RCPBUNIT_CLOCK_GATE_DISABLE | + GEN6_RCCUNIT_CLOCK_GATE_DISABLE); + + /* Bspec says we need to always set all mask bits. */ + I915_WRITE(_3D_CHICKEN, (0xFFFF << 16) | + _3D_CHICKEN_SF_DISABLE_FASTCLIP_CULL); + + /* + * According to the spec the following bits should be + * set in order to enable memory self-refresh and fbc: + * The bit21 and bit22 of 0x42000 + * The bit21 and bit22 of 0x42004 + * The bit5 and bit7 of 0x42020 + * The bit14 of 0x70180 + * The bit14 of 0x71180 + */ + I915_WRITE(ILK_DISPLAY_CHICKEN1, + I915_READ(ILK_DISPLAY_CHICKEN1) | + ILK_FBCQ_DIS | ILK_PABSTRETCH_DIS); + I915_WRITE(ILK_DISPLAY_CHICKEN2, + I915_READ(ILK_DISPLAY_CHICKEN2) | + ILK_DPARB_GATE | ILK_VSDPFD_FULL); + I915_WRITE(ILK_DSPCLK_GATE, + I915_READ(ILK_DSPCLK_GATE) | + ILK_DPARB_CLK_GATE | + ILK_DPFD_CLK_GATE); + + for_each_pipe(pipe) { + I915_WRITE(DSPCNTR(pipe), + I915_READ(DSPCNTR(pipe)) | + DISPPLANE_TRICKLE_FEED_DISABLE); + intel_flush_display_plane(dev_priv, pipe); + } +} + +static void gen7_setup_fixed_func_scheduler(struct drm_i915_private *dev_priv) +{ + uint32_t reg = I915_READ(GEN7_FF_THREAD_MODE); + + reg &= ~GEN7_FF_SCHED_MASK; + reg |= GEN7_FF_TS_SCHED_HW; + reg |= GEN7_FF_VS_SCHED_HW; + reg |= GEN7_FF_DS_SCHED_HW; + + I915_WRITE(GEN7_FF_THREAD_MODE, reg); +} + +static void ivybridge_init_clock_gating(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + int pipe; + uint32_t dspclk_gate = VRHUNIT_CLOCK_GATE_DISABLE; + + I915_WRITE(PCH_DSPCLK_GATE_D, dspclk_gate); + + I915_WRITE(WM3_LP_ILK, 0); + I915_WRITE(WM2_LP_ILK, 0); + I915_WRITE(WM1_LP_ILK, 0); + + /* According to the spec, bit 13 (RCZUNIT) must be set on IVB. + * This implements the WaDisableRCZUnitClockGating workaround. + */ + I915_WRITE(GEN6_UCGCTL2, GEN6_RCZUNIT_CLOCK_GATE_DISABLE); + + I915_WRITE(ILK_DSPCLK_GATE, IVB_VRHUNIT_CLK_GATE); + + I915_WRITE(IVB_CHICKEN3, + CHICKEN3_DGMG_REQ_OUT_FIX_DISABLE | + CHICKEN3_DGMG_DONE_FIX_DISABLE); + + /* Apply the WaDisableRHWOOptimizationForRenderHang workaround. */ + I915_WRITE(GEN7_COMMON_SLICE_CHICKEN1, + GEN7_CSC1_RHWO_OPT_DISABLE_IN_RCC); + + /* WaApplyL3ControlAndL3ChickenMode requires those two on Ivy Bridge */ + I915_WRITE(GEN7_L3CNTLREG1, + GEN7_WA_FOR_GEN7_L3_CONTROL); + I915_WRITE(GEN7_L3_CHICKEN_MODE_REGISTER, + GEN7_WA_L3_CHICKEN_MODE); + + /* This is required by WaCatErrorRejectionIssue */ + I915_WRITE(GEN7_SQ_CHICKEN_MBCUNIT_CONFIG, + I915_READ(GEN7_SQ_CHICKEN_MBCUNIT_CONFIG) | + GEN7_SQ_CHICKEN_MBCUNIT_SQINTMOB); + + for_each_pipe(pipe) { + I915_WRITE(DSPCNTR(pipe), + I915_READ(DSPCNTR(pipe)) | + DISPPLANE_TRICKLE_FEED_DISABLE); + intel_flush_display_plane(dev_priv, pipe); + } + + gen7_setup_fixed_func_scheduler(dev_priv); +} + +static void valleyview_init_clock_gating(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + int pipe; + uint32_t dspclk_gate = VRHUNIT_CLOCK_GATE_DISABLE; + + I915_WRITE(PCH_DSPCLK_GATE_D, dspclk_gate); + + I915_WRITE(WM3_LP_ILK, 0); + I915_WRITE(WM2_LP_ILK, 0); + I915_WRITE(WM1_LP_ILK, 0); + + /* According to the spec, bit 13 (RCZUNIT) must be set on IVB. + * This implements the WaDisableRCZUnitClockGating workaround. + */ + I915_WRITE(GEN6_UCGCTL2, GEN6_RCZUNIT_CLOCK_GATE_DISABLE); + + I915_WRITE(ILK_DSPCLK_GATE, IVB_VRHUNIT_CLK_GATE); + + I915_WRITE(IVB_CHICKEN3, + CHICKEN3_DGMG_REQ_OUT_FIX_DISABLE | + CHICKEN3_DGMG_DONE_FIX_DISABLE); + + /* Apply the WaDisableRHWOOptimizationForRenderHang workaround. */ + I915_WRITE(GEN7_COMMON_SLICE_CHICKEN1, + GEN7_CSC1_RHWO_OPT_DISABLE_IN_RCC); + + /* WaApplyL3ControlAndL3ChickenMode requires those two on Ivy Bridge */ + I915_WRITE(GEN7_L3CNTLREG1, GEN7_WA_FOR_GEN7_L3_CONTROL); + I915_WRITE(GEN7_L3_CHICKEN_MODE_REGISTER, GEN7_WA_L3_CHICKEN_MODE); + + /* This is required by WaCatErrorRejectionIssue */ + I915_WRITE(GEN7_SQ_CHICKEN_MBCUNIT_CONFIG, + I915_READ(GEN7_SQ_CHICKEN_MBCUNIT_CONFIG) | + GEN7_SQ_CHICKEN_MBCUNIT_SQINTMOB); + + for_each_pipe(pipe) { + I915_WRITE(DSPCNTR(pipe), + I915_READ(DSPCNTR(pipe)) | + DISPPLANE_TRICKLE_FEED_DISABLE); + intel_flush_display_plane(dev_priv, pipe); + } + + I915_WRITE(CACHE_MODE_1, I915_READ(CACHE_MODE_1) | + (PIXEL_SUBSPAN_COLLECT_OPT_DISABLE << 16) | + PIXEL_SUBSPAN_COLLECT_OPT_DISABLE); +} + +static void g4x_init_clock_gating(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + uint32_t dspclk_gate; + + I915_WRITE(RENCLK_GATE_D1, 0); + I915_WRITE(RENCLK_GATE_D2, VF_UNIT_CLOCK_GATE_DISABLE | + GS_UNIT_CLOCK_GATE_DISABLE | + CL_UNIT_CLOCK_GATE_DISABLE); + I915_WRITE(RAMCLK_GATE_D, 0); + dspclk_gate = VRHUNIT_CLOCK_GATE_DISABLE | + OVRUNIT_CLOCK_GATE_DISABLE | + OVCUNIT_CLOCK_GATE_DISABLE; + if (IS_GM45(dev)) + dspclk_gate |= DSSUNIT_CLOCK_GATE_DISABLE; + I915_WRITE(DSPCLK_GATE_D, dspclk_gate); +} + +static void crestline_init_clock_gating(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + + I915_WRITE(RENCLK_GATE_D1, I965_RCC_CLOCK_GATE_DISABLE); + I915_WRITE(RENCLK_GATE_D2, 0); + I915_WRITE(DSPCLK_GATE_D, 0); + I915_WRITE(RAMCLK_GATE_D, 0); + I915_WRITE16(DEUC, 0); +} + +static void broadwater_init_clock_gating(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + + I915_WRITE(RENCLK_GATE_D1, I965_RCZ_CLOCK_GATE_DISABLE | + I965_RCC_CLOCK_GATE_DISABLE | + I965_RCPB_CLOCK_GATE_DISABLE | + I965_ISC_CLOCK_GATE_DISABLE | + I965_FBC_CLOCK_GATE_DISABLE); + I915_WRITE(RENCLK_GATE_D2, 0); +} + +static void gen3_init_clock_gating(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + u32 dstate = I915_READ(D_STATE); + + dstate |= DSTATE_PLL_D3_OFF | DSTATE_GFX_CLOCK_GATING | + DSTATE_DOT_CLOCK_GATING; + I915_WRITE(D_STATE, dstate); +} + +static void i85x_init_clock_gating(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + + I915_WRITE(RENCLK_GATE_D1, SV_CLOCK_GATE_DISABLE); +} + +static void i830_init_clock_gating(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + + I915_WRITE(DSPCLK_GATE_D, OVRUNIT_CLOCK_GATE_DISABLE); +} + +static void ibx_init_clock_gating(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + + /* + * On Ibex Peak and Cougar Point, we need to disable clock + * gating for the panel power sequencer or it will fail to + * start up when no ports are active. + */ + I915_WRITE(SOUTH_DSPCLK_GATE_D, PCH_DPLSUNIT_CLOCK_GATE_DISABLE); +} + +static void cpt_init_clock_gating(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + int pipe; + + /* + * On Ibex Peak and Cougar Point, we need to disable clock + * gating for the panel power sequencer or it will fail to + * start up when no ports are active. + */ + I915_WRITE(SOUTH_DSPCLK_GATE_D, PCH_DPLSUNIT_CLOCK_GATE_DISABLE); + I915_WRITE(SOUTH_CHICKEN2, I915_READ(SOUTH_CHICKEN2) | + DPLS_EDP_PPS_FIX_DIS); + /* Without this, mode sets may fail silently on FDI */ + for_each_pipe(pipe) + I915_WRITE(TRANS_CHICKEN2(pipe), TRANS_AUTOTRAIN_GEN_STALL_DIS); +} + +void intel_init_clock_gating(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + + dev_priv->display.init_clock_gating(dev); + + if (dev_priv->display.init_pch_clock_gating) + dev_priv->display.init_pch_clock_gating(dev); +} + +/* Set up chip specific power management-related functions */ +void intel_init_pm(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + + if (I915_HAS_FBC(dev)) { + if (HAS_PCH_SPLIT(dev)) { + dev_priv->display.fbc_enabled = ironlake_fbc_enabled; + dev_priv->display.enable_fbc = ironlake_enable_fbc; + dev_priv->display.disable_fbc = ironlake_disable_fbc; + } else if (IS_GM45(dev)) { + dev_priv->display.fbc_enabled = g4x_fbc_enabled; + dev_priv->display.enable_fbc = g4x_enable_fbc; + dev_priv->display.disable_fbc = g4x_disable_fbc; + } else if (IS_CRESTLINE(dev)) { + dev_priv->display.fbc_enabled = i8xx_fbc_enabled; + dev_priv->display.enable_fbc = i8xx_enable_fbc; + dev_priv->display.disable_fbc = i8xx_disable_fbc; + } + /* 855GM needs testing */ + } + + /* For FIFO watermark updates */ + if (HAS_PCH_SPLIT(dev)) { + dev_priv->display.force_wake_get = __gen6_gt_force_wake_get; + dev_priv->display.force_wake_put = __gen6_gt_force_wake_put; + + /* IVB configs may use multi-threaded forcewake */ + if (IS_IVYBRIDGE(dev) || IS_HASWELL(dev)) { + u32 ecobus; + + /* A small trick here - if the bios hasn't configured MT forcewake, + * and if the device is in RC6, then force_wake_mt_get will not wake + * the device and the ECOBUS read will return zero. Which will be + * (correctly) interpreted by the test below as MT forcewake being + * disabled. + */ + mutex_lock(&dev->struct_mutex); + __gen6_gt_force_wake_mt_get(dev_priv); + ecobus = I915_READ_NOTRACE(ECOBUS); + __gen6_gt_force_wake_mt_put(dev_priv); + mutex_unlock(&dev->struct_mutex); + + if (ecobus & FORCEWAKE_MT_ENABLE) { + DRM_DEBUG_KMS("Using MT version of forcewake\n"); + dev_priv->display.force_wake_get = + __gen6_gt_force_wake_mt_get; + dev_priv->display.force_wake_put = + __gen6_gt_force_wake_mt_put; + } + } + + if (HAS_PCH_IBX(dev)) + dev_priv->display.init_pch_clock_gating = ibx_init_clock_gating; + else if (HAS_PCH_CPT(dev)) + dev_priv->display.init_pch_clock_gating = cpt_init_clock_gating; + + if (IS_GEN5(dev)) { + if (I915_READ(MLTR_ILK) & ILK_SRLT_MASK) + dev_priv->display.update_wm = ironlake_update_wm; + else { + DRM_DEBUG_KMS("Failed to get proper latency. " + "Disable CxSR\n"); + dev_priv->display.update_wm = NULL; + } + dev_priv->display.init_clock_gating = ironlake_init_clock_gating; + } else if (IS_GEN6(dev)) { + if (SNB_READ_WM0_LATENCY()) { + dev_priv->display.update_wm = sandybridge_update_wm; + dev_priv->display.update_sprite_wm = sandybridge_update_sprite_wm; + } else { + DRM_DEBUG_KMS("Failed to read display plane latency. " + "Disable CxSR\n"); + dev_priv->display.update_wm = NULL; + } + dev_priv->display.init_clock_gating = gen6_init_clock_gating; + } else if (IS_IVYBRIDGE(dev)) { + /* FIXME: detect B0+ stepping and use auto training */ + if (SNB_READ_WM0_LATENCY()) { + dev_priv->display.update_wm = sandybridge_update_wm; + dev_priv->display.update_sprite_wm = sandybridge_update_sprite_wm; + } else { + DRM_DEBUG_KMS("Failed to read display plane latency. " + "Disable CxSR\n"); + dev_priv->display.update_wm = NULL; + } + dev_priv->display.init_clock_gating = ivybridge_init_clock_gating; + } else + dev_priv->display.update_wm = NULL; + } else if (IS_VALLEYVIEW(dev)) { + dev_priv->display.update_wm = valleyview_update_wm; + dev_priv->display.init_clock_gating = + valleyview_init_clock_gating; + dev_priv->display.force_wake_get = vlv_force_wake_get; + dev_priv->display.force_wake_put = vlv_force_wake_put; + } else if (IS_PINEVIEW(dev)) { + if (!intel_get_cxsr_latency(IS_PINEVIEW_G(dev), + dev_priv->is_ddr3, + dev_priv->fsb_freq, + dev_priv->mem_freq)) { + DRM_INFO("failed to find known CxSR latency " + "(found ddr%s fsb freq %d, mem freq %d), " + "disabling CxSR\n", + (dev_priv->is_ddr3 == 1) ? "3" : "2", + dev_priv->fsb_freq, dev_priv->mem_freq); + /* Disable CxSR and never update its watermark again */ + pineview_disable_cxsr(dev); + dev_priv->display.update_wm = NULL; + } else + dev_priv->display.update_wm = pineview_update_wm; + dev_priv->display.init_clock_gating = gen3_init_clock_gating; + } else if (IS_G4X(dev)) { + dev_priv->display.update_wm = g4x_update_wm; + dev_priv->display.init_clock_gating = g4x_init_clock_gating; + } else if (IS_GEN4(dev)) { + dev_priv->display.update_wm = i965_update_wm; + if (IS_CRESTLINE(dev)) + dev_priv->display.init_clock_gating = crestline_init_clock_gating; + else if (IS_BROADWATER(dev)) + dev_priv->display.init_clock_gating = broadwater_init_clock_gating; + } else if (IS_GEN3(dev)) { + dev_priv->display.update_wm = i9xx_update_wm; + dev_priv->display.get_fifo_size = i9xx_get_fifo_size; + dev_priv->display.init_clock_gating = gen3_init_clock_gating; + } else if (IS_I865G(dev)) { + dev_priv->display.update_wm = i830_update_wm; + dev_priv->display.init_clock_gating = i85x_init_clock_gating; + dev_priv->display.get_fifo_size = i830_get_fifo_size; + } else if (IS_I85X(dev)) { + dev_priv->display.update_wm = i9xx_update_wm; + dev_priv->display.get_fifo_size = i85x_get_fifo_size; + dev_priv->display.init_clock_gating = i85x_init_clock_gating; + } else { + dev_priv->display.update_wm = i830_update_wm; + dev_priv->display.init_clock_gating = i830_init_clock_gating; + if (IS_845G(dev)) + dev_priv->display.get_fifo_size = i845_get_fifo_size; + else + dev_priv->display.get_fifo_size = i830_get_fifo_size; + } +} + diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index dfdb613..12d9bc7 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -53,9 +53,35 @@ static inline int ring_space(struct intel_ring_buffer *ring) } static int -render_ring_flush(struct intel_ring_buffer *ring, - u32 invalidate_domains, - u32 flush_domains) +gen2_render_ring_flush(struct intel_ring_buffer *ring, + u32 invalidate_domains, + u32 flush_domains) +{ + u32 cmd; + int ret; + + cmd = MI_FLUSH; + if (((invalidate_domains|flush_domains) & I915_GEM_DOMAIN_RENDER) == 0) + cmd |= MI_NO_WRITE_FLUSH; + + if (invalidate_domains & I915_GEM_DOMAIN_SAMPLER) + cmd |= MI_READ_FLUSH; + + ret = intel_ring_begin(ring, 2); + if (ret) + return ret; + + intel_ring_emit(ring, cmd); + intel_ring_emit(ring, MI_NOOP); + intel_ring_advance(ring); + + return 0; +} + +static int +gen4_render_ring_flush(struct intel_ring_buffer *ring, + u32 invalidate_domains, + u32 flush_domains) { struct drm_device *dev = ring->dev; u32 cmd; @@ -90,17 +116,8 @@ render_ring_flush(struct intel_ring_buffer *ring, */ cmd = MI_FLUSH | MI_NO_WRITE_FLUSH; - if ((invalidate_domains|flush_domains) & - I915_GEM_DOMAIN_RENDER) + if ((invalidate_domains|flush_domains) & I915_GEM_DOMAIN_RENDER) cmd &= ~MI_NO_WRITE_FLUSH; - if (INTEL_INFO(dev)->gen < 4) { - /* - * On the 965, the sampler cache always gets flushed - * and this bit is reserved. - */ - if (invalidate_domains & I915_GEM_DOMAIN_SAMPLER) - cmd |= MI_READ_FLUSH; - } if (invalidate_domains & I915_GEM_DOMAIN_INSTRUCTION) cmd |= MI_EXE_FLUSH; @@ -472,21 +489,30 @@ gen6_add_request(struct intel_ring_buffer *ring, * @seqno - seqno which the waiter will block on */ static int -intel_ring_sync(struct intel_ring_buffer *waiter, - struct intel_ring_buffer *signaller, - int ring, - u32 seqno) +gen6_ring_sync(struct intel_ring_buffer *waiter, + struct intel_ring_buffer *signaller, + u32 seqno) { int ret; u32 dw1 = MI_SEMAPHORE_MBOX | MI_SEMAPHORE_COMPARE | MI_SEMAPHORE_REGISTER; + /* Throughout all of the GEM code, seqno passed implies our current + * seqno is >= the last seqno executed. However for hardware the + * comparison is strictly greater than. + */ + seqno -= 1; + + WARN_ON(signaller->semaphore_register[waiter->id] == + MI_SEMAPHORE_SYNC_INVALID); + ret = intel_ring_begin(waiter, 4); if (ret) return ret; - intel_ring_emit(waiter, dw1 | signaller->semaphore_register[ring]); + intel_ring_emit(waiter, + dw1 | signaller->semaphore_register[waiter->id]); intel_ring_emit(waiter, seqno); intel_ring_emit(waiter, 0); intel_ring_emit(waiter, MI_NOOP); @@ -495,47 +521,6 @@ intel_ring_sync(struct intel_ring_buffer *waiter, return 0; } -/* VCS->RCS (RVSYNC) or BCS->RCS (RBSYNC) */ -int -render_ring_sync_to(struct intel_ring_buffer *waiter, - struct intel_ring_buffer *signaller, - u32 seqno) -{ - WARN_ON(signaller->semaphore_register[RCS] == MI_SEMAPHORE_SYNC_INVALID); - return intel_ring_sync(waiter, - signaller, - RCS, - seqno); -} - -/* RCS->VCS (VRSYNC) or BCS->VCS (VBSYNC) */ -int -gen6_bsd_ring_sync_to(struct intel_ring_buffer *waiter, - struct intel_ring_buffer *signaller, - u32 seqno) -{ - WARN_ON(signaller->semaphore_register[VCS] == MI_SEMAPHORE_SYNC_INVALID); - return intel_ring_sync(waiter, - signaller, - VCS, - seqno); -} - -/* RCS->BCS (BRSYNC) or VCS->BCS (BVSYNC) */ -int -gen6_blt_ring_sync_to(struct intel_ring_buffer *waiter, - struct intel_ring_buffer *signaller, - u32 seqno) -{ - WARN_ON(signaller->semaphore_register[BCS] == MI_SEMAPHORE_SYNC_INVALID); - return intel_ring_sync(waiter, - signaller, - BCS, - seqno); -} - - - #define PIPE_CONTROL_FLUSH(ring__, addr__) \ do { \ intel_ring_emit(ring__, GFX_OP_PIPE_CONTROL(4) | PIPE_CONTROL_QW_WRITE | \ @@ -597,27 +582,6 @@ pc_render_add_request(struct intel_ring_buffer *ring, return 0; } -static int -render_ring_add_request(struct intel_ring_buffer *ring, - u32 *result) -{ - u32 seqno = i915_gem_next_request_seqno(ring); - int ret; - - ret = intel_ring_begin(ring, 4); - if (ret) - return ret; - - intel_ring_emit(ring, MI_STORE_DWORD_INDEX); - intel_ring_emit(ring, I915_GEM_HWS_INDEX << MI_STORE_DWORD_INDEX_SHIFT); - intel_ring_emit(ring, seqno); - intel_ring_emit(ring, MI_USER_INTERRUPT); - intel_ring_advance(ring); - - *result = seqno; - return 0; -} - static u32 gen6_ring_get_seqno(struct intel_ring_buffer *ring) { @@ -644,40 +608,43 @@ pc_render_get_seqno(struct intel_ring_buffer *ring) return pc->cpu_page[0]; } -static void -ironlake_enable_irq(drm_i915_private_t *dev_priv, u32 mask) +static bool +gen5_ring_get_irq(struct intel_ring_buffer *ring) { - dev_priv->gt_irq_mask &= ~mask; - I915_WRITE(GTIMR, dev_priv->gt_irq_mask); - POSTING_READ(GTIMR); -} + struct drm_device *dev = ring->dev; + drm_i915_private_t *dev_priv = dev->dev_private; -static void -ironlake_disable_irq(drm_i915_private_t *dev_priv, u32 mask) -{ - dev_priv->gt_irq_mask |= mask; - I915_WRITE(GTIMR, dev_priv->gt_irq_mask); - POSTING_READ(GTIMR); -} + if (!dev->irq_enabled) + return false; -static void -i915_enable_irq(drm_i915_private_t *dev_priv, u32 mask) -{ - dev_priv->irq_mask &= ~mask; - I915_WRITE(IMR, dev_priv->irq_mask); - POSTING_READ(IMR); + spin_lock(&ring->irq_lock); + if (ring->irq_refcount++ == 0) { + dev_priv->gt_irq_mask &= ~ring->irq_enable_mask; + I915_WRITE(GTIMR, dev_priv->gt_irq_mask); + POSTING_READ(GTIMR); + } + spin_unlock(&ring->irq_lock); + + return true; } static void -i915_disable_irq(drm_i915_private_t *dev_priv, u32 mask) +gen5_ring_put_irq(struct intel_ring_buffer *ring) { - dev_priv->irq_mask |= mask; - I915_WRITE(IMR, dev_priv->irq_mask); - POSTING_READ(IMR); + struct drm_device *dev = ring->dev; + drm_i915_private_t *dev_priv = dev->dev_private; + + spin_lock(&ring->irq_lock); + if (--ring->irq_refcount == 0) { + dev_priv->gt_irq_mask |= ring->irq_enable_mask; + I915_WRITE(GTIMR, dev_priv->gt_irq_mask); + POSTING_READ(GTIMR); + } + spin_unlock(&ring->irq_lock); } static bool -render_ring_get_irq(struct intel_ring_buffer *ring) +i9xx_ring_get_irq(struct intel_ring_buffer *ring) { struct drm_device *dev = ring->dev; drm_i915_private_t *dev_priv = dev->dev_private; @@ -687,11 +654,9 @@ render_ring_get_irq(struct intel_ring_buffer *ring) spin_lock(&ring->irq_lock); if (ring->irq_refcount++ == 0) { - if (INTEL_INFO(dev)->gen >= 5) - ironlake_enable_irq(dev_priv, - GT_PIPE_NOTIFY | GT_USER_INTERRUPT); - else - i915_enable_irq(dev_priv, I915_USER_INTERRUPT); + dev_priv->irq_mask &= ~ring->irq_enable_mask; + I915_WRITE(IMR, dev_priv->irq_mask); + POSTING_READ(IMR); } spin_unlock(&ring->irq_lock); @@ -699,19 +664,16 @@ render_ring_get_irq(struct intel_ring_buffer *ring) } static void -render_ring_put_irq(struct intel_ring_buffer *ring) +i9xx_ring_put_irq(struct intel_ring_buffer *ring) { struct drm_device *dev = ring->dev; drm_i915_private_t *dev_priv = dev->dev_private; spin_lock(&ring->irq_lock); if (--ring->irq_refcount == 0) { - if (INTEL_INFO(dev)->gen >= 5) - ironlake_disable_irq(dev_priv, - GT_USER_INTERRUPT | - GT_PIPE_NOTIFY); - else - i915_disable_irq(dev_priv, I915_USER_INTERRUPT); + dev_priv->irq_mask |= ring->irq_enable_mask; + I915_WRITE(IMR, dev_priv->irq_mask); + POSTING_READ(IMR); } spin_unlock(&ring->irq_lock); } @@ -765,7 +727,7 @@ bsd_ring_flush(struct intel_ring_buffer *ring, } static int -ring_add_request(struct intel_ring_buffer *ring, +i9xx_add_request(struct intel_ring_buffer *ring, u32 *result) { u32 seqno; @@ -792,7 +754,6 @@ gen6_ring_get_irq(struct intel_ring_buffer *ring) { struct drm_device *dev = ring->dev; drm_i915_private_t *dev_priv = dev->dev_private; - u32 mask = ring->irq_enable; if (!dev->irq_enabled) return false; @@ -804,9 +765,10 @@ gen6_ring_get_irq(struct intel_ring_buffer *ring) spin_lock(&ring->irq_lock); if (ring->irq_refcount++ == 0) { - ring->irq_mask &= ~mask; - I915_WRITE_IMR(ring, ring->irq_mask); - ironlake_enable_irq(dev_priv, mask); + I915_WRITE_IMR(ring, ~ring->irq_enable_mask); + dev_priv->gt_irq_mask &= ~ring->irq_enable_mask; + I915_WRITE(GTIMR, dev_priv->gt_irq_mask); + POSTING_READ(GTIMR); } spin_unlock(&ring->irq_lock); @@ -818,105 +780,69 @@ gen6_ring_put_irq(struct intel_ring_buffer *ring) { struct drm_device *dev = ring->dev; drm_i915_private_t *dev_priv = dev->dev_private; - u32 mask = ring->irq_enable; spin_lock(&ring->irq_lock); if (--ring->irq_refcount == 0) { - ring->irq_mask |= mask; - I915_WRITE_IMR(ring, ring->irq_mask); - ironlake_disable_irq(dev_priv, mask); + I915_WRITE_IMR(ring, ~0); + dev_priv->gt_irq_mask |= ring->irq_enable_mask; + I915_WRITE(GTIMR, dev_priv->gt_irq_mask); + POSTING_READ(GTIMR); } spin_unlock(&ring->irq_lock); gen6_gt_force_wake_put(dev_priv); } -static bool -bsd_ring_get_irq(struct intel_ring_buffer *ring) +static int +i965_dispatch_execbuffer(struct intel_ring_buffer *ring, u32 offset, u32 length) { - struct drm_device *dev = ring->dev; - drm_i915_private_t *dev_priv = dev->dev_private; - - if (!dev->irq_enabled) - return false; + int ret; - spin_lock(&ring->irq_lock); - if (ring->irq_refcount++ == 0) { - if (IS_G4X(dev)) - i915_enable_irq(dev_priv, I915_BSD_USER_INTERRUPT); - else - ironlake_enable_irq(dev_priv, GT_BSD_USER_INTERRUPT); - } - spin_unlock(&ring->irq_lock); + ret = intel_ring_begin(ring, 2); + if (ret) + return ret; - return true; -} -static void -bsd_ring_put_irq(struct intel_ring_buffer *ring) -{ - struct drm_device *dev = ring->dev; - drm_i915_private_t *dev_priv = dev->dev_private; + intel_ring_emit(ring, + MI_BATCH_BUFFER_START | + MI_BATCH_GTT | + MI_BATCH_NON_SECURE_I965); + intel_ring_emit(ring, offset); + intel_ring_advance(ring); - spin_lock(&ring->irq_lock); - if (--ring->irq_refcount == 0) { - if (IS_G4X(dev)) - i915_disable_irq(dev_priv, I915_BSD_USER_INTERRUPT); - else - ironlake_disable_irq(dev_priv, GT_BSD_USER_INTERRUPT); - } - spin_unlock(&ring->irq_lock); + return 0; } static int -ring_dispatch_execbuffer(struct intel_ring_buffer *ring, u32 offset, u32 length) +i830_dispatch_execbuffer(struct intel_ring_buffer *ring, + u32 offset, u32 len) { int ret; - ret = intel_ring_begin(ring, 2); + ret = intel_ring_begin(ring, 4); if (ret) return ret; - intel_ring_emit(ring, - MI_BATCH_BUFFER_START | (2 << 6) | - MI_BATCH_NON_SECURE_I965); - intel_ring_emit(ring, offset); + intel_ring_emit(ring, MI_BATCH_BUFFER); + intel_ring_emit(ring, offset | MI_BATCH_NON_SECURE); + intel_ring_emit(ring, offset + len - 8); + intel_ring_emit(ring, 0); intel_ring_advance(ring); return 0; } static int -render_ring_dispatch_execbuffer(struct intel_ring_buffer *ring, +i915_dispatch_execbuffer(struct intel_ring_buffer *ring, u32 offset, u32 len) { - struct drm_device *dev = ring->dev; int ret; - if (IS_I830(dev) || IS_845G(dev)) { - ret = intel_ring_begin(ring, 4); - if (ret) - return ret; - - intel_ring_emit(ring, MI_BATCH_BUFFER); - intel_ring_emit(ring, offset | MI_BATCH_NON_SECURE); - intel_ring_emit(ring, offset + len - 8); - intel_ring_emit(ring, 0); - } else { - ret = intel_ring_begin(ring, 2); - if (ret) - return ret; + ret = intel_ring_begin(ring, 2); + if (ret) + return ret; - if (INTEL_INFO(dev)->gen >= 4) { - intel_ring_emit(ring, - MI_BATCH_BUFFER_START | (2 << 6) | - MI_BATCH_NON_SECURE_I965); - intel_ring_emit(ring, offset); - } else { - intel_ring_emit(ring, - MI_BATCH_BUFFER_START | (2 << 6)); - intel_ring_emit(ring, offset | MI_BATCH_NON_SECURE); - } - } + intel_ring_emit(ring, MI_BATCH_BUFFER_START | MI_BATCH_GTT); + intel_ring_emit(ring, offset | MI_BATCH_NON_SECURE); intel_ring_advance(ring); return 0; @@ -983,8 +909,8 @@ err: return ret; } -int intel_init_ring_buffer(struct drm_device *dev, - struct intel_ring_buffer *ring) +static int intel_init_ring_buffer(struct drm_device *dev, + struct intel_ring_buffer *ring) { struct drm_i915_gem_object *obj; int ret; @@ -993,10 +919,10 @@ int intel_init_ring_buffer(struct drm_device *dev, INIT_LIST_HEAD(&ring->active_list); INIT_LIST_HEAD(&ring->request_list); INIT_LIST_HEAD(&ring->gpu_write_list); + ring->size = 32 * PAGE_SIZE; init_waitqueue_head(&ring->irq_queue); spin_lock_init(&ring->irq_lock); - ring->irq_mask = ~0; if (I915_NEED_GFX_HWS(dev)) { ret = init_status_page(ring); @@ -1040,7 +966,7 @@ int intel_init_ring_buffer(struct drm_device *dev, * of the buffer. */ ring->effective_size = ring->size; - if (IS_I830(ring->dev)) + if (IS_I830(ring->dev) || IS_845G(ring->dev)) ring->effective_size -= 128; return 0; @@ -1263,44 +1189,6 @@ void intel_ring_advance(struct intel_ring_buffer *ring) ring->write_tail(ring, ring->tail); } -static const struct intel_ring_buffer render_ring = { - .name = "render ring", - .id = RCS, - .mmio_base = RENDER_RING_BASE, - .size = 32 * PAGE_SIZE, - .init = init_render_ring, - .write_tail = ring_write_tail, - .flush = render_ring_flush, - .add_request = render_ring_add_request, - .get_seqno = ring_get_seqno, - .irq_get = render_ring_get_irq, - .irq_put = render_ring_put_irq, - .dispatch_execbuffer = render_ring_dispatch_execbuffer, - .cleanup = render_ring_cleanup, - .sync_to = render_ring_sync_to, - .semaphore_register = {MI_SEMAPHORE_SYNC_INVALID, - MI_SEMAPHORE_SYNC_RV, - MI_SEMAPHORE_SYNC_RB}, - .signal_mbox = {GEN6_VRSYNC, GEN6_BRSYNC}, -}; - -/* ring buffer for bit-stream decoder */ - -static const struct intel_ring_buffer bsd_ring = { - .name = "bsd ring", - .id = VCS, - .mmio_base = BSD_RING_BASE, - .size = 32 * PAGE_SIZE, - .init = init_ring_common, - .write_tail = ring_write_tail, - .flush = bsd_ring_flush, - .add_request = ring_add_request, - .get_seqno = ring_get_seqno, - .irq_get = bsd_ring_get_irq, - .irq_put = bsd_ring_put_irq, - .dispatch_execbuffer = ring_dispatch_execbuffer, -}; - static void gen6_bsd_ring_write_tail(struct intel_ring_buffer *ring, u32 value) @@ -1363,28 +1251,6 @@ gen6_ring_dispatch_execbuffer(struct intel_ring_buffer *ring, return 0; } -/* ring buffer for Video Codec for Gen6+ */ -static const struct intel_ring_buffer gen6_bsd_ring = { - .name = "gen6 bsd ring", - .id = VCS, - .mmio_base = GEN6_BSD_RING_BASE, - .size = 32 * PAGE_SIZE, - .init = init_ring_common, - .write_tail = gen6_bsd_ring_write_tail, - .flush = gen6_ring_flush, - .add_request = gen6_add_request, - .get_seqno = gen6_ring_get_seqno, - .irq_enable = GEN6_BSD_USER_INTERRUPT, - .irq_get = gen6_ring_get_irq, - .irq_put = gen6_ring_put_irq, - .dispatch_execbuffer = gen6_ring_dispatch_execbuffer, - .sync_to = gen6_bsd_ring_sync_to, - .semaphore_register = {MI_SEMAPHORE_SYNC_VR, - MI_SEMAPHORE_SYNC_INVALID, - MI_SEMAPHORE_SYNC_VB}, - .signal_mbox = {GEN6_RVSYNC, GEN6_BVSYNC}, -}; - /* Blitter support (SandyBridge+) */ static int blt_ring_flush(struct intel_ring_buffer *ring, @@ -1408,44 +1274,58 @@ static int blt_ring_flush(struct intel_ring_buffer *ring, return 0; } -static const struct intel_ring_buffer gen6_blt_ring = { - .name = "blt ring", - .id = BCS, - .mmio_base = BLT_RING_BASE, - .size = 32 * PAGE_SIZE, - .init = init_ring_common, - .write_tail = ring_write_tail, - .flush = blt_ring_flush, - .add_request = gen6_add_request, - .get_seqno = gen6_ring_get_seqno, - .irq_get = gen6_ring_get_irq, - .irq_put = gen6_ring_put_irq, - .irq_enable = GEN6_BLITTER_USER_INTERRUPT, - .dispatch_execbuffer = gen6_ring_dispatch_execbuffer, - .sync_to = gen6_blt_ring_sync_to, - .semaphore_register = {MI_SEMAPHORE_SYNC_BR, - MI_SEMAPHORE_SYNC_BV, - MI_SEMAPHORE_SYNC_INVALID}, - .signal_mbox = {GEN6_RBSYNC, GEN6_VBSYNC}, -}; - int intel_init_render_ring_buffer(struct drm_device *dev) { drm_i915_private_t *dev_priv = dev->dev_private; struct intel_ring_buffer *ring = &dev_priv->ring[RCS]; - *ring = render_ring; + ring->name = "render ring"; + ring->id = RCS; + ring->mmio_base = RENDER_RING_BASE; + if (INTEL_INFO(dev)->gen >= 6) { ring->add_request = gen6_add_request; ring->flush = gen6_render_ring_flush; ring->irq_get = gen6_ring_get_irq; ring->irq_put = gen6_ring_put_irq; - ring->irq_enable = GT_USER_INTERRUPT; + ring->irq_enable_mask = GT_USER_INTERRUPT; ring->get_seqno = gen6_ring_get_seqno; + ring->sync_to = gen6_ring_sync; + ring->semaphore_register[0] = MI_SEMAPHORE_SYNC_INVALID; + ring->semaphore_register[1] = MI_SEMAPHORE_SYNC_RV; + ring->semaphore_register[2] = MI_SEMAPHORE_SYNC_RB; + ring->signal_mbox[0] = GEN6_VRSYNC; + ring->signal_mbox[1] = GEN6_BRSYNC; } else if (IS_GEN5(dev)) { ring->add_request = pc_render_add_request; + ring->flush = gen4_render_ring_flush; ring->get_seqno = pc_render_get_seqno; + ring->irq_get = gen5_ring_get_irq; + ring->irq_put = gen5_ring_put_irq; + ring->irq_enable_mask = GT_USER_INTERRUPT | GT_PIPE_NOTIFY; + } else { + ring->add_request = i9xx_add_request; + if (INTEL_INFO(dev)->gen < 4) + ring->flush = gen2_render_ring_flush; + else + ring->flush = gen4_render_ring_flush; + ring->get_seqno = ring_get_seqno; + ring->irq_get = i9xx_ring_get_irq; + ring->irq_put = i9xx_ring_put_irq; + ring->irq_enable_mask = I915_USER_INTERRUPT; } + ring->write_tail = ring_write_tail; + if (INTEL_INFO(dev)->gen >= 6) + ring->dispatch_execbuffer = gen6_ring_dispatch_execbuffer; + else if (INTEL_INFO(dev)->gen >= 4) + ring->dispatch_execbuffer = i965_dispatch_execbuffer; + else if (IS_I830(dev) || IS_845G(dev)) + ring->dispatch_execbuffer = i830_dispatch_execbuffer; + else + ring->dispatch_execbuffer = i915_dispatch_execbuffer; + ring->init = init_render_ring; + ring->cleanup = render_ring_cleanup; + if (!I915_NEED_GFX_HWS(dev)) { ring->status_page.page_addr = dev_priv->status_page_dmah->vaddr; @@ -1460,17 +1340,37 @@ int intel_render_ring_init_dri(struct drm_device *dev, u64 start, u32 size) drm_i915_private_t *dev_priv = dev->dev_private; struct intel_ring_buffer *ring = &dev_priv->ring[RCS]; - *ring = render_ring; + ring->name = "render ring"; + ring->id = RCS; + ring->mmio_base = RENDER_RING_BASE; + if (INTEL_INFO(dev)->gen >= 6) { - ring->add_request = gen6_add_request; - ring->irq_get = gen6_ring_get_irq; - ring->irq_put = gen6_ring_put_irq; - ring->irq_enable = GT_USER_INTERRUPT; - } else if (IS_GEN5(dev)) { - ring->add_request = pc_render_add_request; - ring->get_seqno = pc_render_get_seqno; + /* non-kms not supported on gen6+ */ + return -ENODEV; } + /* Note: gem is not supported on gen5/ilk without kms (the corresponding + * gem_init ioctl returns with -ENODEV). Hence we do not need to set up + * the special gen5 functions. */ + ring->add_request = i9xx_add_request; + if (INTEL_INFO(dev)->gen < 4) + ring->flush = gen2_render_ring_flush; + else + ring->flush = gen4_render_ring_flush; + ring->get_seqno = ring_get_seqno; + ring->irq_get = i9xx_ring_get_irq; + ring->irq_put = i9xx_ring_put_irq; + ring->irq_enable_mask = I915_USER_INTERRUPT; + ring->write_tail = ring_write_tail; + if (INTEL_INFO(dev)->gen >= 4) + ring->dispatch_execbuffer = i965_dispatch_execbuffer; + else if (IS_I830(dev) || IS_845G(dev)) + ring->dispatch_execbuffer = i830_dispatch_execbuffer; + else + ring->dispatch_execbuffer = i915_dispatch_execbuffer; + ring->init = init_render_ring; + ring->cleanup = render_ring_cleanup; + if (!I915_NEED_GFX_HWS(dev)) ring->status_page.page_addr = dev_priv->status_page_dmah->vaddr; @@ -1506,10 +1406,46 @@ int intel_init_bsd_ring_buffer(struct drm_device *dev) drm_i915_private_t *dev_priv = dev->dev_private; struct intel_ring_buffer *ring = &dev_priv->ring[VCS]; - if (IS_GEN6(dev) || IS_GEN7(dev)) - *ring = gen6_bsd_ring; - else - *ring = bsd_ring; + ring->name = "bsd ring"; + ring->id = VCS; + + ring->write_tail = ring_write_tail; + if (IS_GEN6(dev) || IS_GEN7(dev)) { + ring->mmio_base = GEN6_BSD_RING_BASE; + /* gen6 bsd needs a special wa for tail updates */ + if (IS_GEN6(dev)) + ring->write_tail = gen6_bsd_ring_write_tail; + ring->flush = gen6_ring_flush; + ring->add_request = gen6_add_request; + ring->get_seqno = gen6_ring_get_seqno; + ring->irq_enable_mask = GEN6_BSD_USER_INTERRUPT; + ring->irq_get = gen6_ring_get_irq; + ring->irq_put = gen6_ring_put_irq; + ring->dispatch_execbuffer = gen6_ring_dispatch_execbuffer; + ring->sync_to = gen6_ring_sync; + ring->semaphore_register[0] = MI_SEMAPHORE_SYNC_VR; + ring->semaphore_register[1] = MI_SEMAPHORE_SYNC_INVALID; + ring->semaphore_register[2] = MI_SEMAPHORE_SYNC_VB; + ring->signal_mbox[0] = GEN6_RVSYNC; + ring->signal_mbox[1] = GEN6_BVSYNC; + } else { + ring->mmio_base = BSD_RING_BASE; + ring->flush = bsd_ring_flush; + ring->add_request = i9xx_add_request; + ring->get_seqno = ring_get_seqno; + if (IS_GEN5(dev)) { + ring->irq_enable_mask = GT_BSD_USER_INTERRUPT; + ring->irq_get = gen5_ring_get_irq; + ring->irq_put = gen5_ring_put_irq; + } else { + ring->irq_enable_mask = I915_BSD_USER_INTERRUPT; + ring->irq_get = i9xx_ring_get_irq; + ring->irq_put = i9xx_ring_put_irq; + } + ring->dispatch_execbuffer = i965_dispatch_execbuffer; + } + ring->init = init_ring_common; + return intel_init_ring_buffer(dev, ring); } @@ -1519,7 +1455,25 @@ int intel_init_blt_ring_buffer(struct drm_device *dev) drm_i915_private_t *dev_priv = dev->dev_private; struct intel_ring_buffer *ring = &dev_priv->ring[BCS]; - *ring = gen6_blt_ring; + ring->name = "blitter ring"; + ring->id = BCS; + + ring->mmio_base = BLT_RING_BASE; + ring->write_tail = ring_write_tail; + ring->flush = blt_ring_flush; + ring->add_request = gen6_add_request; + ring->get_seqno = gen6_ring_get_seqno; + ring->irq_enable_mask = GEN6_BLITTER_USER_INTERRUPT; + ring->irq_get = gen6_ring_get_irq; + ring->irq_put = gen6_ring_put_irq; + ring->dispatch_execbuffer = gen6_ring_dispatch_execbuffer; + ring->sync_to = gen6_ring_sync; + ring->semaphore_register[0] = MI_SEMAPHORE_SYNC_BR; + ring->semaphore_register[1] = MI_SEMAPHORE_SYNC_BV; + ring->semaphore_register[2] = MI_SEMAPHORE_SYNC_INVALID; + ring->signal_mbox[0] = GEN6_RBSYNC; + ring->signal_mbox[1] = GEN6_VBSYNC; + ring->init = init_ring_common; return intel_init_ring_buffer(dev, ring); } diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h index 3488a5a..06a66ad 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.h +++ b/drivers/gpu/drm/i915/intel_ringbuffer.h @@ -58,8 +58,7 @@ struct intel_ring_buffer { spinlock_t irq_lock; u32 irq_refcount; - u32 irq_mask; - u32 irq_enable; /* IRQs enabled for this ring */ + u32 irq_enable_mask; /* bitmask to enable ring interrupt */ u32 irq_seqno; /* last seq seem at irq time */ u32 trace_irq_seqno; u32 waiting_seqno; diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c index 6898145..c330efd 100644 --- a/drivers/gpu/drm/i915/intel_sdvo.c +++ b/drivers/gpu/drm/i915/intel_sdvo.c @@ -443,9 +443,17 @@ static const char *cmd_status_names[] = { static bool intel_sdvo_write_cmd(struct intel_sdvo *intel_sdvo, u8 cmd, const void *args, int args_len) { - u8 buf[args_len*2 + 2], status; - struct i2c_msg msgs[args_len + 3]; - int i, ret; + u8 *buf, status; + struct i2c_msg *msgs; + int i, ret = true; + + buf = (u8 *)kzalloc(args_len * 2 + 2, GFP_KERNEL); + if (!buf) + return false; + + msgs = kcalloc(args_len + 3, sizeof(*msgs), GFP_KERNEL); + if (!msgs) + return false; intel_sdvo_debug_write(intel_sdvo, cmd, args, args_len); @@ -479,15 +487,19 @@ static bool intel_sdvo_write_cmd(struct intel_sdvo *intel_sdvo, u8 cmd, ret = i2c_transfer(intel_sdvo->i2c, msgs, i+3); if (ret < 0) { DRM_DEBUG_KMS("I2c transfer returned %d\n", ret); - return false; + ret = false; + goto out; } if (ret != i+3) { /* failure in I2C transfer */ DRM_DEBUG_KMS("I2c transfer returned %d/%d\n", ret, i+3); - return false; + ret = false; } - return true; +out: + kfree(msgs); + kfree(buf); + return ret; } static bool intel_sdvo_read_response(struct intel_sdvo *intel_sdvo, @@ -1258,7 +1270,7 @@ intel_sdvo_get_analog_edid(struct drm_connector *connector) dev_priv->crt_ddc_pin)); } -enum drm_connector_status +static enum drm_connector_status intel_sdvo_tmds_sink_detect(struct drm_connector *connector) { struct intel_sdvo *intel_sdvo = intel_attached_sdvo(connector); diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c index a464771..fbf03b9 100644 --- a/drivers/gpu/drm/i915/intel_sprite.c +++ b/drivers/gpu/drm/i915/intel_sprite.c @@ -95,7 +95,6 @@ ivb_update_plane(struct drm_plane *plane, struct drm_framebuffer *fb, /* must disable */ sprctl |= SPRITE_TRICKLE_FEED_DISABLE; sprctl |= SPRITE_ENABLE; - sprctl |= SPRITE_DEST_KEY; /* Sizes are 0 based */ src_w--; @@ -112,13 +111,13 @@ ivb_update_plane(struct drm_plane *plane, struct drm_framebuffer *fb, */ if (crtc_w != src_w || crtc_h != src_h) { dev_priv->sprite_scaling_enabled = true; - sandybridge_update_wm(dev); + intel_update_watermarks(dev); intel_wait_for_vblank(dev, pipe); sprscale = SPRITE_SCALE_ENABLE | (src_w << 16) | src_h; } else { dev_priv->sprite_scaling_enabled = false; /* potentially re-enable LP watermarks */ - sandybridge_update_wm(dev); + intel_update_watermarks(dev); } I915_WRITE(SPRSTRIDE(pipe), fb->pitches[0]); @@ -134,7 +133,7 @@ ivb_update_plane(struct drm_plane *plane, struct drm_framebuffer *fb, I915_WRITE(SPRSIZE(pipe), (crtc_h << 16) | crtc_w); I915_WRITE(SPRSCALE(pipe), sprscale); I915_WRITE(SPRCTL(pipe), sprctl); - I915_WRITE(SPRSURF(pipe), obj->gtt_offset); + I915_MODIFY_DISPBASE(SPRSURF(pipe), obj->gtt_offset); POSTING_READ(SPRSURF(pipe)); } @@ -150,7 +149,7 @@ ivb_disable_plane(struct drm_plane *plane) /* Can't leave the scaler enabled... */ I915_WRITE(SPRSCALE(pipe), 0); /* Activate double buffered register update */ - I915_WRITE(SPRSURF(pipe), 0); + I915_MODIFY_DISPBASE(SPRSURF(pipe), 0); POSTING_READ(SPRSURF(pipe)); } @@ -209,7 +208,7 @@ ivb_get_colorkey(struct drm_plane *plane, struct drm_intel_sprite_colorkey *key) } static void -snb_update_plane(struct drm_plane *plane, struct drm_framebuffer *fb, +ilk_update_plane(struct drm_plane *plane, struct drm_framebuffer *fb, struct drm_i915_gem_object *obj, int crtc_x, int crtc_y, unsigned int crtc_w, unsigned int crtc_h, uint32_t x, uint32_t y, @@ -219,7 +218,7 @@ snb_update_plane(struct drm_plane *plane, struct drm_framebuffer *fb, struct drm_i915_private *dev_priv = dev->dev_private; struct intel_plane *intel_plane = to_intel_plane(plane); int pipe = intel_plane->pipe, pixel_size; - u32 dvscntr, dvsscale = 0; + u32 dvscntr, dvsscale; dvscntr = I915_READ(DVSCNTR(pipe)); @@ -263,8 +262,8 @@ snb_update_plane(struct drm_plane *plane, struct drm_framebuffer *fb, if (obj->tiling_mode != I915_TILING_NONE) dvscntr |= DVS_TILED; - /* must disable */ - dvscntr |= DVS_TRICKLE_FEED_DISABLE; + if (IS_GEN6(dev)) + dvscntr |= DVS_TRICKLE_FEED_DISABLE; /* must disable */ dvscntr |= DVS_ENABLE; /* Sizes are 0 based */ @@ -275,7 +274,8 @@ snb_update_plane(struct drm_plane *plane, struct drm_framebuffer *fb, intel_update_sprite_watermarks(dev, pipe, crtc_w, pixel_size); - if (crtc_w != src_w || crtc_h != src_h) + dvsscale = 0; + if (IS_GEN5(dev) || crtc_w != src_w || crtc_h != src_h) dvsscale = DVS_SCALE_ENABLE | (src_w << 16) | src_h; I915_WRITE(DVSSTRIDE(pipe), fb->pitches[0]); @@ -291,12 +291,12 @@ snb_update_plane(struct drm_plane *plane, struct drm_framebuffer *fb, I915_WRITE(DVSSIZE(pipe), (crtc_h << 16) | crtc_w); I915_WRITE(DVSSCALE(pipe), dvsscale); I915_WRITE(DVSCNTR(pipe), dvscntr); - I915_WRITE(DVSSURF(pipe), obj->gtt_offset); + I915_MODIFY_DISPBASE(DVSSURF(pipe), obj->gtt_offset); POSTING_READ(DVSSURF(pipe)); } static void -snb_disable_plane(struct drm_plane *plane) +ilk_disable_plane(struct drm_plane *plane) { struct drm_device *dev = plane->dev; struct drm_i915_private *dev_priv = dev->dev_private; @@ -307,7 +307,7 @@ snb_disable_plane(struct drm_plane *plane) /* Disable the scaler */ I915_WRITE(DVSSCALE(pipe), 0); /* Flush double buffered register updates */ - I915_WRITE(DVSSURF(pipe), 0); + I915_MODIFY_DISPBASE(DVSSURF(pipe), 0); POSTING_READ(DVSSURF(pipe)); } @@ -334,7 +334,7 @@ intel_disable_primary(struct drm_crtc *crtc) } static int -snb_update_colorkey(struct drm_plane *plane, +ilk_update_colorkey(struct drm_plane *plane, struct drm_intel_sprite_colorkey *key) { struct drm_device *dev = plane->dev; @@ -363,7 +363,7 @@ snb_update_colorkey(struct drm_plane *plane, } static void -snb_get_colorkey(struct drm_plane *plane, struct drm_intel_sprite_colorkey *key) +ilk_get_colorkey(struct drm_plane *plane, struct drm_intel_sprite_colorkey *key) { struct drm_device *dev = plane->dev; struct drm_i915_private *dev_priv = dev->dev_private; @@ -617,6 +617,14 @@ static const struct drm_plane_funcs intel_plane_funcs = { .destroy = intel_destroy_plane, }; +static uint32_t ilk_plane_formats[] = { + DRM_FORMAT_XRGB8888, + DRM_FORMAT_YUYV, + DRM_FORMAT_YVYU, + DRM_FORMAT_UYVY, + DRM_FORMAT_VYUY, +}; + static uint32_t snb_plane_formats[] = { DRM_FORMAT_XBGR8888, DRM_FORMAT_XRGB8888, @@ -631,34 +639,56 @@ intel_plane_init(struct drm_device *dev, enum pipe pipe) { struct intel_plane *intel_plane; unsigned long possible_crtcs; + const uint32_t *plane_formats; + int num_plane_formats; int ret; - if (!(IS_GEN6(dev) || IS_GEN7(dev))) + if (INTEL_INFO(dev)->gen < 5) return -ENODEV; intel_plane = kzalloc(sizeof(struct intel_plane), GFP_KERNEL); if (!intel_plane) return -ENOMEM; - if (IS_GEN6(dev)) { + switch (INTEL_INFO(dev)->gen) { + case 5: + case 6: intel_plane->max_downscale = 16; - intel_plane->update_plane = snb_update_plane; - intel_plane->disable_plane = snb_disable_plane; - intel_plane->update_colorkey = snb_update_colorkey; - intel_plane->get_colorkey = snb_get_colorkey; - } else if (IS_GEN7(dev)) { + intel_plane->update_plane = ilk_update_plane; + intel_plane->disable_plane = ilk_disable_plane; + intel_plane->update_colorkey = ilk_update_colorkey; + intel_plane->get_colorkey = ilk_get_colorkey; + + if (IS_GEN6(dev)) { + plane_formats = snb_plane_formats; + num_plane_formats = ARRAY_SIZE(snb_plane_formats); + } else { + plane_formats = ilk_plane_formats; + num_plane_formats = ARRAY_SIZE(ilk_plane_formats); + } + break; + + case 7: intel_plane->max_downscale = 2; intel_plane->update_plane = ivb_update_plane; intel_plane->disable_plane = ivb_disable_plane; intel_plane->update_colorkey = ivb_update_colorkey; intel_plane->get_colorkey = ivb_get_colorkey; + + plane_formats = snb_plane_formats; + num_plane_formats = ARRAY_SIZE(snb_plane_formats); + break; + + default: + return -ENODEV; } intel_plane->pipe = pipe; possible_crtcs = (1 << pipe); ret = drm_plane_init(dev, &intel_plane->base, possible_crtcs, - &intel_plane_funcs, snb_plane_formats, - ARRAY_SIZE(snb_plane_formats), false); + &intel_plane_funcs, + plane_formats, num_plane_formats, + false); if (ret) kfree(intel_plane); diff --git a/drivers/gpu/drm/i915/intel_tv.c b/drivers/gpu/drm/i915/intel_tv.c index ca12c70..67f444d 100644 --- a/drivers/gpu/drm/i915/intel_tv.c +++ b/drivers/gpu/drm/i915/intel_tv.c @@ -811,7 +811,7 @@ intel_tv_mode_lookup(const char *tv_format) { int i; - for (i = 0; i < sizeof(tv_modes) / sizeof(tv_modes[0]); i++) { + for (i = 0; i < ARRAY_SIZE(tv_modes); i++) { const struct tv_mode *tv_mode = &tv_modes[i]; if (!strcmp(tv_format, tv_mode->name)) diff --git a/drivers/gpu/drm/radeon/atombios_encoders.c b/drivers/gpu/drm/radeon/atombios_encoders.c index 06b209b..b92a694 100644 --- a/drivers/gpu/drm/radeon/atombios_encoders.c +++ b/drivers/gpu/drm/radeon/atombios_encoders.c @@ -230,6 +230,10 @@ atombios_dvo_setup(struct drm_encoder *encoder, int action) if (!atom_parse_cmd_header(rdev->mode_info.atom_context, index, &frev, &crev)) return; + /* some R4xx chips have the wrong frev */ + if (rdev->family <= CHIP_RV410) + frev = 1; + switch (frev) { case 1: switch (crev) { diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c index e11df77..cb11418 100644 --- a/drivers/gpu/drm/radeon/r100.c +++ b/drivers/gpu/drm/radeon/r100.c @@ -2557,7 +2557,7 @@ static void r100_pll_errata_after_data(struct radeon_device *rdev) * or the chip could hang on a subsequent access */ if (rdev->pll_errata & CHIP_ERRATA_PLL_DELAY) { - udelay(5000); + mdelay(5); } /* This function is required to workaround a hardware bug in some (all?) diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c index 8f84bd6..222245d 100644 --- a/drivers/gpu/drm/radeon/r600.c +++ b/drivers/gpu/drm/radeon/r600.c @@ -2839,7 +2839,7 @@ void r600_rlc_stop(struct radeon_device *rdev) /* r7xx asics need to soft reset RLC before halting */ WREG32(SRBM_SOFT_RESET, SOFT_RESET_RLC); RREG32(SRBM_SOFT_RESET); - udelay(15000); + mdelay(15); WREG32(SRBM_SOFT_RESET, 0); RREG32(SRBM_SOFT_RESET); } diff --git a/drivers/gpu/drm/radeon/r600_cp.c b/drivers/gpu/drm/radeon/r600_cp.c index 84c5462..75ed17c 100644 --- a/drivers/gpu/drm/radeon/r600_cp.c +++ b/drivers/gpu/drm/radeon/r600_cp.c @@ -407,7 +407,7 @@ static void r600_cp_load_microcode(drm_radeon_private_t *dev_priv) RADEON_WRITE(R600_GRBM_SOFT_RESET, R600_SOFT_RESET_CP); RADEON_READ(R600_GRBM_SOFT_RESET); - DRM_UDELAY(15000); + mdelay(15); RADEON_WRITE(R600_GRBM_SOFT_RESET, 0); fw_data = (const __be32 *)dev_priv->me_fw->data; @@ -500,7 +500,7 @@ static void r700_cp_load_microcode(drm_radeon_private_t *dev_priv) RADEON_WRITE(R600_GRBM_SOFT_RESET, R600_SOFT_RESET_CP); RADEON_READ(R600_GRBM_SOFT_RESET); - DRM_UDELAY(15000); + mdelay(15); RADEON_WRITE(R600_GRBM_SOFT_RESET, 0); fw_data = (const __be32 *)dev_priv->pfp_fw->data; @@ -1797,7 +1797,7 @@ static void r600_cp_init_ring_buffer(struct drm_device *dev, RADEON_WRITE(R600_GRBM_SOFT_RESET, R600_SOFT_RESET_CP); RADEON_READ(R600_GRBM_SOFT_RESET); - DRM_UDELAY(15000); + mdelay(15); RADEON_WRITE(R600_GRBM_SOFT_RESET, 0); diff --git a/drivers/gpu/drm/radeon/radeon_clocks.c b/drivers/gpu/drm/radeon/radeon_clocks.c index 6ae0c75..9c6b29a 100644 --- a/drivers/gpu/drm/radeon/radeon_clocks.c +++ b/drivers/gpu/drm/radeon/radeon_clocks.c @@ -633,7 +633,7 @@ void radeon_legacy_set_clock_gating(struct radeon_device *rdev, int enable) tmp &= ~(R300_SCLK_FORCE_VAP); tmp |= RADEON_SCLK_FORCE_CP; WREG32_PLL(RADEON_SCLK_CNTL, tmp); - udelay(15000); + mdelay(15); tmp = RREG32_PLL(R300_SCLK_CNTL2); tmp &= ~(R300_SCLK_FORCE_TCL | @@ -651,12 +651,12 @@ void radeon_legacy_set_clock_gating(struct radeon_device *rdev, int enable) tmp |= (RADEON_ENGIN_DYNCLK_MODE | (0x01 << RADEON_ACTIVE_HILO_LAT_SHIFT)); WREG32_PLL(RADEON_CLK_PWRMGT_CNTL, tmp); - udelay(15000); + mdelay(15); tmp = RREG32_PLL(RADEON_CLK_PIN_CNTL); tmp |= RADEON_SCLK_DYN_START_CNTL; WREG32_PLL(RADEON_CLK_PIN_CNTL, tmp); - udelay(15000); + mdelay(15); /* When DRI is enabled, setting DYN_STOP_LAT to zero can cause some R200 to lockup randomly, leave them as set by BIOS. @@ -696,7 +696,7 @@ void radeon_legacy_set_clock_gating(struct radeon_device *rdev, int enable) tmp |= RADEON_SCLK_MORE_FORCEON; } WREG32_PLL(RADEON_SCLK_MORE_CNTL, tmp); - udelay(15000); + mdelay(15); } /* RV200::A11 A12, RV250::A11 A12 */ @@ -709,7 +709,7 @@ void radeon_legacy_set_clock_gating(struct radeon_device *rdev, int enable) tmp |= RADEON_TCL_BYPASS_DISABLE; WREG32_PLL(RADEON_PLL_PWRMGT_CNTL, tmp); } - udelay(15000); + mdelay(15); /*enable dynamic mode for display clocks (PIXCLK and PIX2CLK) */ tmp = RREG32_PLL(RADEON_PIXCLKS_CNTL); @@ -722,14 +722,14 @@ void radeon_legacy_set_clock_gating(struct radeon_device *rdev, int enable) RADEON_PIXCLK_TMDS_ALWAYS_ONb); WREG32_PLL(RADEON_PIXCLKS_CNTL, tmp); - udelay(15000); + mdelay(15); tmp = RREG32_PLL(RADEON_VCLK_ECP_CNTL); tmp |= (RADEON_PIXCLK_ALWAYS_ONb | RADEON_PIXCLK_DAC_ALWAYS_ONb); WREG32_PLL(RADEON_VCLK_ECP_CNTL, tmp); - udelay(15000); + mdelay(15); } } else { /* Turn everything OFF (ForceON to everything) */ @@ -861,7 +861,7 @@ void radeon_legacy_set_clock_gating(struct radeon_device *rdev, int enable) } WREG32_PLL(RADEON_SCLK_CNTL, tmp); - udelay(16000); + mdelay(16); if ((rdev->family == CHIP_R300) || (rdev->family == CHIP_R350)) { @@ -870,7 +870,7 @@ void radeon_legacy_set_clock_gating(struct radeon_device *rdev, int enable) R300_SCLK_FORCE_GA | R300_SCLK_FORCE_CBA); WREG32_PLL(R300_SCLK_CNTL2, tmp); - udelay(16000); + mdelay(16); } if (rdev->flags & RADEON_IS_IGP) { @@ -878,7 +878,7 @@ void radeon_legacy_set_clock_gating(struct radeon_device *rdev, int enable) tmp &= ~(RADEON_FORCEON_MCLKA | RADEON_FORCEON_YCLKA); WREG32_PLL(RADEON_MCLK_CNTL, tmp); - udelay(16000); + mdelay(16); } if ((rdev->family == CHIP_RV200) || @@ -887,7 +887,7 @@ void radeon_legacy_set_clock_gating(struct radeon_device *rdev, int enable) tmp = RREG32_PLL(RADEON_SCLK_MORE_CNTL); tmp |= RADEON_SCLK_MORE_FORCEON; WREG32_PLL(RADEON_SCLK_MORE_CNTL, tmp); - udelay(16000); + mdelay(16); } tmp = RREG32_PLL(RADEON_PIXCLKS_CNTL); @@ -900,7 +900,7 @@ void radeon_legacy_set_clock_gating(struct radeon_device *rdev, int enable) RADEON_PIXCLK_TMDS_ALWAYS_ONb); WREG32_PLL(RADEON_PIXCLKS_CNTL, tmp); - udelay(16000); + mdelay(16); tmp = RREG32_PLL(RADEON_VCLK_ECP_CNTL); tmp &= ~(RADEON_PIXCLK_ALWAYS_ONb | diff --git a/drivers/gpu/drm/radeon/radeon_combios.c b/drivers/gpu/drm/radeon/radeon_combios.c index 81fc100..2cad9fd 100644 --- a/drivers/gpu/drm/radeon/radeon_combios.c +++ b/drivers/gpu/drm/radeon/radeon_combios.c @@ -2845,7 +2845,7 @@ bool radeon_combios_external_tmds_setup(struct drm_encoder *encoder) case 4: val = RBIOS16(index); index += 2; - udelay(val * 1000); + mdelay(val); break; case 6: slave_addr = id & 0xff; @@ -3044,7 +3044,7 @@ static void combios_parse_pll_table(struct drm_device *dev, uint16_t offset) udelay(150); break; case 2: - udelay(1000); + mdelay(1); break; case 3: while (tmp--) { @@ -3075,13 +3075,13 @@ static void combios_parse_pll_table(struct drm_device *dev, uint16_t offset) /*mclk_cntl |= 0x00001111;*//* ??? */ WREG32_PLL(RADEON_MCLK_CNTL, mclk_cntl); - udelay(10000); + mdelay(10); #endif WREG32_PLL (RADEON_CLK_PWRMGT_CNTL, tmp & ~RADEON_CG_NO1_DEBUG_0); - udelay(10000); + mdelay(10); } break; default: diff --git a/drivers/gpu/drm/radeon/radeon_i2c.c b/drivers/gpu/drm/radeon/radeon_i2c.c index 85bcfc8..3edec1c 100644 --- a/drivers/gpu/drm/radeon/radeon_i2c.c +++ b/drivers/gpu/drm/radeon/radeon_i2c.c @@ -900,6 +900,10 @@ struct radeon_i2c_chan *radeon_i2c_create(struct drm_device *dev, struct radeon_i2c_chan *i2c; int ret; + /* don't add the mm_i2c bus unless hw_i2c is enabled */ + if (rec->mm_i2c && (radeon_hw_i2c == 0)) + return NULL; + i2c = kzalloc(sizeof(struct radeon_i2c_chan), GFP_KERNEL); if (i2c == NULL) return NULL; diff --git a/drivers/gpu/drm/radeon/radeon_legacy_encoders.c b/drivers/gpu/drm/radeon/radeon_legacy_encoders.c index 2f46e0c..42db254 100644 --- a/drivers/gpu/drm/radeon/radeon_legacy_encoders.c +++ b/drivers/gpu/drm/radeon/radeon_legacy_encoders.c @@ -88,7 +88,7 @@ static void radeon_legacy_lvds_update(struct drm_encoder *encoder, int mode) lvds_pll_cntl = RREG32(RADEON_LVDS_PLL_CNTL); lvds_pll_cntl |= RADEON_LVDS_PLL_EN; WREG32(RADEON_LVDS_PLL_CNTL, lvds_pll_cntl); - udelay(1000); + mdelay(1); lvds_pll_cntl = RREG32(RADEON_LVDS_PLL_CNTL); lvds_pll_cntl &= ~RADEON_LVDS_PLL_RESET; @@ -101,7 +101,7 @@ static void radeon_legacy_lvds_update(struct drm_encoder *encoder, int mode) (backlight_level << RADEON_LVDS_BL_MOD_LEVEL_SHIFT)); if (is_mac) lvds_gen_cntl |= RADEON_LVDS_BL_MOD_EN; - udelay(panel_pwr_delay * 1000); + mdelay(panel_pwr_delay); WREG32(RADEON_LVDS_GEN_CNTL, lvds_gen_cntl); break; case DRM_MODE_DPMS_STANDBY: @@ -118,10 +118,10 @@ static void radeon_legacy_lvds_update(struct drm_encoder *encoder, int mode) WREG32(RADEON_LVDS_GEN_CNTL, lvds_gen_cntl); lvds_gen_cntl &= ~(RADEON_LVDS_ON | RADEON_LVDS_BLON | RADEON_LVDS_EN | RADEON_LVDS_DIGON); } - udelay(panel_pwr_delay * 1000); + mdelay(panel_pwr_delay); WREG32(RADEON_LVDS_GEN_CNTL, lvds_gen_cntl); WREG32_PLL(RADEON_PIXCLKS_CNTL, pixclks_cntl); - udelay(panel_pwr_delay * 1000); + mdelay(panel_pwr_delay); break; } @@ -656,7 +656,7 @@ static enum drm_connector_status radeon_legacy_primary_dac_detect(struct drm_enc WREG32(RADEON_DAC_MACRO_CNTL, tmp); - udelay(2000); + mdelay(2); if (RREG32(RADEON_DAC_CNTL) & RADEON_DAC_CMP_OUTPUT) found = connector_status_connected; @@ -1499,7 +1499,7 @@ static enum drm_connector_status radeon_legacy_tv_dac_detect(struct drm_encoder tmp = dac_cntl2 | RADEON_DAC2_DAC2_CLK_SEL | RADEON_DAC2_CMP_EN; WREG32(RADEON_DAC_CNTL2, tmp); - udelay(10000); + mdelay(10); if (ASIC_IS_R300(rdev)) { if (RREG32(RADEON_DAC_CNTL2) & RADEON_DAC2_CMP_OUT_B) diff --git a/drivers/gpu/drm/savage/savage_state.c b/drivers/gpu/drm/savage/savage_state.c index 031aaaf..b6d8608 100644 --- a/drivers/gpu/drm/savage/savage_state.c +++ b/drivers/gpu/drm/savage/savage_state.c @@ -988,7 +988,7 @@ int savage_bci_cmdbuf(struct drm_device *dev, void *data, struct drm_file *file_ * for locking on FreeBSD. */ if (cmdbuf->size) { - kcmd_addr = kmalloc(cmdbuf->size * 8, GFP_KERNEL); + kcmd_addr = kmalloc_array(cmdbuf->size, 8, GFP_KERNEL); if (kcmd_addr == NULL) return -ENOMEM; @@ -1015,8 +1015,8 @@ int savage_bci_cmdbuf(struct drm_device *dev, void *data, struct drm_file *file_ cmdbuf->vb_addr = kvb_addr; } if (cmdbuf->nbox) { - kbox_addr = kmalloc(cmdbuf->nbox * sizeof(struct drm_clip_rect), - GFP_KERNEL); + kbox_addr = kmalloc_array(cmdbuf->nbox, sizeof(struct drm_clip_rect), + GFP_KERNEL); if (kbox_addr == NULL) { ret = -ENOMEM; goto done; |