diff options
| author | Linus Torvalds <torvalds@linux-foundation.org> | 2012-03-22 20:08:22 (GMT) |
|---|---|---|
| committer | Linus Torvalds <torvalds@linux-foundation.org> | 2012-03-22 20:08:22 (GMT) |
| commit | be53bfdb8088e9d1924199cc1a96e113756b1075 (patch) | |
| tree | 8c65eb9d82ca4c0f11c17cfdc44d5263820b415b /drivers/gpu/drm/radeon | |
| parent | b2094ef840697bc8ca5d17a83b7e30fad5f1e9fa (diff) | |
| parent | 5466c7b1683a23dbbcfb7ee4a71c4f23886001c7 (diff) | |
| download | linux-fsl-qoriq-be53bfdb8088e9d1924199cc1a96e113756b1075.tar.xz | |
Merge branch 'drm-next' of git://people.freedesktop.org/~airlied/linux
Pull drm main changes from Dave Airlie:
"This is the main drm pull request, I'm probably going to send two more
smaller ones, will explain below.
This contains a patch that is also in the fbdev tree, but it should be
the same patch, it added an API for hot unplugging framebuffer
devices, and I need that API for a new driver.
It also contains some changes to the i2c tree which Jean has acked,
and one change to moorestown platform stuff in x86.
Highlights:
- new drivers: UDL driver for USB displaylink devices, kms only,
should support correct hotplug operations.
- core: i2c speedups + better hotplug support, EDID overriding via
firmware interface - allows user to load a firmware for a broken
monitor/kvm from userspace, it even has documentation for it.
- exynos: new HDMI audio + hdmi 1.4 + virtual output driver
- gma500: code cleanup
- radeon: cleanups, CS optimisations, streamout support and pageflip
fix
- nouveau: NVD9 displayport support + more reclocking work
- i915: re-enabling GMBUS, finish gpu patch (might help hibernation
who knows), missed irq fixes, stencil tiling fixes, interlaced
support, aliasesd PPGTT support for SNB/IVB, swizzling for SNB/IVB,
semaphore fixes
As well as the usual bunch of cleanups and fixes all over the place.
I've got two things I'd like to merge a bit later:
a) AMD support for all their new radeonhd 7000 series GPU and APUs.
AMD dropped this a bit late due to insane internal review
processes, (please AMD just follow Intel and let open source guys
ship stuff early) however I don't want to penalise people who own
this hardware (since its been on sale for 3-4 months and GPU hw
doesn't exactly have a lifetime in years) and consign them to
using closed drivers for longer than necessary. The changes are
well contained and just plug into the driver new gpu functionality
so they should be fairly regression proof. I just want to give
them a bit of a run on the hw AMD kindly sent me.
b) drm prime/dma-buf interface code. This is just infrastructure
code to expose the dma-buf stuff to drm drivers and to userspace.
I'm not planning on pushing any driver support in this cycle
(except maybe exynos), but I'd like to get the infrastructure code
in so for the next cycle I can start getting the driver support
into the individual drivers. We have started driver support for
i915, nouveau and udl along with I think exynos and omap in
staging. However this code relies on the dma-buf tree being
pulled into your tree first since it needs the latest interfaces
from that tree. I'll push to get that tree sent asap.
(oh and any warnings you see in i915 are gcc's fault from what anyone
can see)."
Fix up trivial conflicts in arch/x86/platform/mrst/mrst.c due to the new
msic_thermal_platform_data() thermal function being added next to the
tc35876x_platform_data() i2c device function..
* 'drm-next' of git://people.freedesktop.org/~airlied/linux: (326 commits)
drm/i915: use DDC_ADDR instead of hard-coding it
drm/radeon: use DDC_ADDR instead of hard-coding it
drm: remove unneeded redefinition of DDC_ADDR
drm/exynos: added virtual display driver.
drm: allow loading an EDID as firmware to override broken monitor
drm/exynos: enable hdmi audio feature
drm/exynos: add default pixel format for plane
drm/exynos: cleanup exynos_hdmi.h
drm/exynos: add is_local member in exynos_drm_subdrv struct
drm/exynos: add subdrv open/close functions
drm/exynos: remove module of exynos drm subdrv
drm/exynos: release pending pageflip events when closed
drm/exynos: added new funtion to get/put dma address.
drm/exynos: update gem and buffer framework.
drm/exynos: added mode_fixup feature and code clean.
drm/exynos: add HDMI version 1.4 support
drm/exynos: remove exynos_mixer.h
gma500: Fix mmap frambuffer
drm/radeon: Drop radeon_gem_object_(un)pin.
drm/radeon: Restrict offset for legacy display engine.
...
Diffstat (limited to 'drivers/gpu/drm/radeon')
52 files changed, 3762 insertions, 1270 deletions
diff --git a/drivers/gpu/drm/radeon/Makefile b/drivers/gpu/drm/radeon/Makefile index 2139fe8..8410415 100644 --- a/drivers/gpu/drm/radeon/Makefile +++ b/drivers/gpu/drm/radeon/Makefile @@ -71,7 +71,7 @@ radeon-y += radeon_device.o radeon_asic.o radeon_kms.o \ r600_blit_kms.o radeon_pm.o atombios_dp.o r600_audio.o r600_hdmi.o \ evergreen.o evergreen_cs.o evergreen_blit_shaders.o evergreen_blit_kms.o \ radeon_trace_points.o ni.o cayman_blit_shaders.o atombios_encoders.o \ - radeon_semaphore.o radeon_sa.o + radeon_semaphore.o radeon_sa.o atombios_i2c.o radeon-$(CONFIG_COMPAT) += radeon_ioc32.o radeon-$(CONFIG_VGA_SWITCHEROO) += radeon_atpx_handler.o diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c index 742f17f..72672ea 100644 --- a/drivers/gpu/drm/radeon/atombios_crtc.c +++ b/drivers/gpu/drm/radeon/atombios_crtc.c @@ -1031,6 +1031,7 @@ static int dce4_crtc_do_set_base(struct drm_crtc *crtc, struct radeon_bo *rbo; uint64_t fb_location; uint32_t fb_format, fb_pitch_pixels, tiling_flags; + unsigned bankw, bankh, mtaspect, tile_split; u32 fb_swap = EVERGREEN_GRPH_ENDIAN_SWAP(EVERGREEN_GRPH_ENDIAN_NONE); u32 tmp, viewport_w, viewport_h; int r; @@ -1121,20 +1122,13 @@ static int dce4_crtc_do_set_base(struct drm_crtc *crtc, break; } - switch ((tmp & 0xf000) >> 12) { - case 0: /* 1KB rows */ - default: - fb_format |= EVERGREEN_GRPH_TILE_SPLIT(EVERGREEN_ADDR_SURF_TILE_SPLIT_1KB); - break; - case 1: /* 2KB rows */ - fb_format |= EVERGREEN_GRPH_TILE_SPLIT(EVERGREEN_ADDR_SURF_TILE_SPLIT_2KB); - break; - case 2: /* 4KB rows */ - fb_format |= EVERGREEN_GRPH_TILE_SPLIT(EVERGREEN_ADDR_SURF_TILE_SPLIT_4KB); - break; - } - fb_format |= EVERGREEN_GRPH_ARRAY_MODE(EVERGREEN_GRPH_ARRAY_2D_TILED_THIN1); + + evergreen_tiling_fields(tiling_flags, &bankw, &bankh, &mtaspect, &tile_split); + fb_format |= EVERGREEN_GRPH_TILE_SPLIT(tile_split); + fb_format |= EVERGREEN_GRPH_BANK_WIDTH(bankw); + fb_format |= EVERGREEN_GRPH_BANK_HEIGHT(bankh); + fb_format |= EVERGREEN_GRPH_MACRO_TILE_ASPECT(mtaspect); } else if (tiling_flags & RADEON_TILING_MICRO) fb_format |= EVERGREEN_GRPH_ARRAY_MODE(EVERGREEN_GRPH_ARRAY_1D_TILED_THIN1); diff --git a/drivers/gpu/drm/radeon/atombios_dp.c b/drivers/gpu/drm/radeon/atombios_dp.c index 552b436..191218a 100644 --- a/drivers/gpu/drm/radeon/atombios_dp.c +++ b/drivers/gpu/drm/radeon/atombios_dp.c @@ -746,7 +746,8 @@ static int radeon_dp_link_train_init(struct radeon_dp_link_train_info *dp_info) /* set the lane count on the sink */ tmp = dp_info->dp_lane_count; - if (dp_info->dpcd[0] >= 0x11) + if (dp_info->dpcd[DP_DPCD_REV] >= 0x11 && + dp_info->dpcd[DP_MAX_LANE_COUNT] & DP_ENHANCED_FRAME_CAP) tmp |= DP_LANE_COUNT_ENHANCED_FRAME_EN; radeon_write_dpcd_reg(dp_info->radeon_connector, DP_LANE_COUNT_SET, tmp); diff --git a/drivers/gpu/drm/radeon/atombios_i2c.c b/drivers/gpu/drm/radeon/atombios_i2c.c new file mode 100644 index 0000000..44d87b6 --- /dev/null +++ b/drivers/gpu/drm/radeon/atombios_i2c.c @@ -0,0 +1,139 @@ +/* + * Copyright 2011 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: Alex Deucher + * + */ +#include "drmP.h" +#include "radeon_drm.h" +#include "radeon.h" +#include "atom.h" + +#define TARGET_HW_I2C_CLOCK 50 + +/* these are a limitation of ProcessI2cChannelTransaction not the hw */ +#define ATOM_MAX_HW_I2C_WRITE 2 +#define ATOM_MAX_HW_I2C_READ 255 + +static int radeon_process_i2c_ch(struct radeon_i2c_chan *chan, + u8 slave_addr, u8 flags, + u8 *buf, u8 num) +{ + struct drm_device *dev = chan->dev; + struct radeon_device *rdev = dev->dev_private; + PROCESS_I2C_CHANNEL_TRANSACTION_PS_ALLOCATION args; + int index = GetIndexIntoMasterTable(COMMAND, ProcessI2cChannelTransaction); + unsigned char *base; + u16 out; + + memset(&args, 0, sizeof(args)); + + base = (unsigned char *)rdev->mode_info.atom_context->scratch; + + if (flags & HW_I2C_WRITE) { + if (num > ATOM_MAX_HW_I2C_WRITE) { + DRM_ERROR("hw i2c: tried to write too many bytes (%d vs 2)\n", num); + return -EINVAL; + } + memcpy(&out, buf, num); + args.lpI2CDataOut = cpu_to_le16(out); + } else { + if (num > ATOM_MAX_HW_I2C_READ) { + DRM_ERROR("hw i2c: tried to read too many bytes (%d vs 255)\n", num); + return -EINVAL; + } + } + + args.ucI2CSpeed = TARGET_HW_I2C_CLOCK; + args.ucRegIndex = 0; + args.ucTransBytes = num; + args.ucSlaveAddr = slave_addr << 1; + args.ucLineNumber = chan->rec.i2c_id; + + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + + /* error */ + if (args.ucStatus != HW_ASSISTED_I2C_STATUS_SUCCESS) { + DRM_DEBUG_KMS("hw_i2c error\n"); + return -EIO; + } + + if (!(flags & HW_I2C_WRITE)) + memcpy(buf, base, num); + + return 0; +} + +int radeon_atom_hw_i2c_xfer(struct i2c_adapter *i2c_adap, + struct i2c_msg *msgs, int num) +{ + struct radeon_i2c_chan *i2c = i2c_get_adapdata(i2c_adap); + struct i2c_msg *p; + int i, remaining, current_count, buffer_offset, max_bytes, ret; + u8 buf = 0, flags; + + /* check for bus probe */ + p = &msgs[0]; + if ((num == 1) && (p->len == 0)) { + ret = radeon_process_i2c_ch(i2c, + p->addr, HW_I2C_WRITE, + &buf, 1); + if (ret) + return ret; + else + return num; + } + + for (i = 0; i < num; i++) { + p = &msgs[i]; + remaining = p->len; + buffer_offset = 0; + /* max_bytes are a limitation of ProcessI2cChannelTransaction not the hw */ + if (p->flags & I2C_M_RD) { + max_bytes = ATOM_MAX_HW_I2C_READ; + flags = HW_I2C_READ; + } else { + max_bytes = ATOM_MAX_HW_I2C_WRITE; + flags = HW_I2C_WRITE; + } + while (remaining) { + if (remaining > max_bytes) + current_count = max_bytes; + else + current_count = remaining; + ret = radeon_process_i2c_ch(i2c, + p->addr, flags, + &p->buf[buffer_offset], current_count); + if (ret) + return ret; + remaining -= current_count; + buffer_offset += current_count; + } + } + + return num; +} + +u32 radeon_atom_hw_i2c_func(struct i2c_adapter *adap) +{ + return I2C_FUNC_I2C | I2C_FUNC_SMBUS_EMUL; +} + diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c index f58254a..466db41 100644 --- a/drivers/gpu/drm/radeon/evergreen.c +++ b/drivers/gpu/drm/radeon/evergreen.c @@ -43,6 +43,37 @@ void evergreen_pcie_gen2_enable(struct radeon_device *rdev); extern void cayman_cp_int_cntl_setup(struct radeon_device *rdev, int ring, u32 cp_int_cntl); +void evergreen_tiling_fields(unsigned tiling_flags, unsigned *bankw, + unsigned *bankh, unsigned *mtaspect, + unsigned *tile_split) +{ + *bankw = (tiling_flags >> RADEON_TILING_EG_BANKW_SHIFT) & RADEON_TILING_EG_BANKW_MASK; + *bankh = (tiling_flags >> RADEON_TILING_EG_BANKH_SHIFT) & RADEON_TILING_EG_BANKH_MASK; + *mtaspect = (tiling_flags >> RADEON_TILING_EG_MACRO_TILE_ASPECT_SHIFT) & RADEON_TILING_EG_MACRO_TILE_ASPECT_MASK; + *tile_split = (tiling_flags >> RADEON_TILING_EG_TILE_SPLIT_SHIFT) & RADEON_TILING_EG_TILE_SPLIT_MASK; + switch (*bankw) { + default: + case 1: *bankw = EVERGREEN_ADDR_SURF_BANK_WIDTH_1; break; + case 2: *bankw = EVERGREEN_ADDR_SURF_BANK_WIDTH_2; break; + case 4: *bankw = EVERGREEN_ADDR_SURF_BANK_WIDTH_4; break; + case 8: *bankw = EVERGREEN_ADDR_SURF_BANK_WIDTH_8; break; + } + switch (*bankh) { + default: + case 1: *bankh = EVERGREEN_ADDR_SURF_BANK_HEIGHT_1; break; + case 2: *bankh = EVERGREEN_ADDR_SURF_BANK_HEIGHT_2; break; + case 4: *bankh = EVERGREEN_ADDR_SURF_BANK_HEIGHT_4; break; + case 8: *bankh = EVERGREEN_ADDR_SURF_BANK_HEIGHT_8; break; + } + switch (*mtaspect) { + default: + case 1: *mtaspect = EVERGREEN_ADDR_SURF_MACRO_TILE_ASPECT_1; break; + case 2: *mtaspect = EVERGREEN_ADDR_SURF_MACRO_TILE_ASPECT_2; break; + case 4: *mtaspect = EVERGREEN_ADDR_SURF_MACRO_TILE_ASPECT_4; break; + case 8: *mtaspect = EVERGREEN_ADDR_SURF_MACRO_TILE_ASPECT_8; break; + } +} + void evergreen_fix_pci_max_read_req_size(struct radeon_device *rdev) { u16 ctl, v; @@ -68,6 +99,25 @@ void evergreen_fix_pci_max_read_req_size(struct radeon_device *rdev) } } +void dce4_wait_for_vblank(struct radeon_device *rdev, int crtc) +{ + struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc]; + int i; + + if (RREG32(EVERGREEN_CRTC_CONTROL + radeon_crtc->crtc_offset) & EVERGREEN_CRTC_MASTER_EN) { + for (i = 0; i < rdev->usec_timeout; i++) { + if (!(RREG32(EVERGREEN_CRTC_STATUS + radeon_crtc->crtc_offset) & EVERGREEN_CRTC_V_BLANK)) + break; + udelay(1); + } + for (i = 0; i < rdev->usec_timeout; i++) { + if (RREG32(EVERGREEN_CRTC_STATUS + radeon_crtc->crtc_offset) & EVERGREEN_CRTC_V_BLANK) + break; + udelay(1); + } + } +} + void evergreen_pre_page_flip(struct radeon_device *rdev, int crtc) { /* enable the pflip int */ @@ -1489,7 +1539,7 @@ int evergreen_cp_resume(struct radeon_device *rdev) evergreen_cp_start(rdev); ring->ready = true; - r = radeon_ring_test(rdev, ring); + r = radeon_ring_test(rdev, RADEON_RING_TYPE_GFX_INDEX, ring); if (r) { ring->ready = false; return r; @@ -3147,7 +3197,7 @@ static int evergreen_startup(struct radeon_device *rdev) r = evergreen_blit_init(rdev); if (r) { r600_blit_fini(rdev); - rdev->asic->copy = NULL; + rdev->asic->copy.copy = NULL; dev_warn(rdev->dev, "failed blitter (%d) falling back to memcpy\n", r); } @@ -3187,7 +3237,7 @@ static int evergreen_startup(struct radeon_device *rdev) if (r) return r; - r = r600_ib_test(rdev, RADEON_RING_TYPE_GFX_INDEX); + r = radeon_ib_test(rdev, RADEON_RING_TYPE_GFX_INDEX, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]); if (r) { DRM_ERROR("radeon: failed testing IB (%d).\n", r); rdev->accel_working = false; diff --git a/drivers/gpu/drm/radeon/evergreen_blit_kms.c b/drivers/gpu/drm/radeon/evergreen_blit_kms.c index 2379849..4e83fdc 100644 --- a/drivers/gpu/drm/radeon/evergreen_blit_kms.c +++ b/drivers/gpu/drm/radeon/evergreen_blit_kms.c @@ -32,17 +32,7 @@ #include "evergreend.h" #include "evergreen_blit_shaders.h" #include "cayman_blit_shaders.h" - -#define DI_PT_RECTLIST 0x11 -#define DI_INDEX_SIZE_16_BIT 0x0 -#define DI_SRC_SEL_AUTO_INDEX 0x2 - -#define FMT_8 0x1 -#define FMT_5_6_5 0x8 -#define FMT_8_8_8_8 0x1a -#define COLOR_8 0x1 -#define COLOR_5_6_5 0x8 -#define COLOR_8_8_8_8 0x1a +#include "radeon_blit_common.h" /* emits 17 */ static void diff --git a/drivers/gpu/drm/radeon/evergreen_cs.c b/drivers/gpu/drm/radeon/evergreen_cs.c index 8e8cd85..a58b37a 100644 --- a/drivers/gpu/drm/radeon/evergreen_cs.c +++ b/drivers/gpu/drm/radeon/evergreen_cs.c @@ -31,6 +31,9 @@ #include "evergreen_reg_safe.h" #include "cayman_reg_safe.h" +#define MAX(a,b) (((a)>(b))?(a):(b)) +#define MIN(a,b) (((a)<(b))?(a):(b)) + static int evergreen_cs_packet_next_reloc(struct radeon_cs_parser *p, struct radeon_cs_reloc **cs_reloc); @@ -40,42 +43,43 @@ struct evergreen_cs_track { u32 npipes; u32 row_size; /* value we track */ - u32 nsamples; - u32 cb_color_base_last[12]; + u32 nsamples; /* unused */ struct radeon_bo *cb_color_bo[12]; u32 cb_color_bo_offset[12]; - struct radeon_bo *cb_color_fmask_bo[8]; - struct radeon_bo *cb_color_cmask_bo[8]; + struct radeon_bo *cb_color_fmask_bo[8]; /* unused */ + struct radeon_bo *cb_color_cmask_bo[8]; /* unused */ u32 cb_color_info[12]; u32 cb_color_view[12]; - u32 cb_color_pitch_idx[12]; - u32 cb_color_slice_idx[12]; - u32 cb_color_dim_idx[12]; - u32 cb_color_dim[12]; u32 cb_color_pitch[12]; u32 cb_color_slice[12]; - u32 cb_color_cmask_slice[8]; - u32 cb_color_fmask_slice[8]; + u32 cb_color_attrib[12]; + u32 cb_color_cmask_slice[8];/* unused */ + u32 cb_color_fmask_slice[8];/* unused */ u32 cb_target_mask; - u32 cb_shader_mask; + u32 cb_shader_mask; /* unused */ u32 vgt_strmout_config; u32 vgt_strmout_buffer_config; + struct radeon_bo *vgt_strmout_bo[4]; + u32 vgt_strmout_bo_offset[4]; + u32 vgt_strmout_size[4]; u32 db_depth_control; u32 db_depth_view; + u32 db_depth_slice; u32 db_depth_size; - u32 db_depth_size_idx; u32 db_z_info; - u32 db_z_idx; u32 db_z_read_offset; u32 db_z_write_offset; struct radeon_bo *db_z_read_bo; struct radeon_bo *db_z_write_bo; u32 db_s_info; - u32 db_s_idx; u32 db_s_read_offset; u32 db_s_write_offset; struct radeon_bo *db_s_read_bo; struct radeon_bo *db_s_write_bo; + bool sx_misc_kill_all_prims; + bool cb_dirty; + bool db_dirty; + bool streamout_dirty; }; static u32 evergreen_cs_get_aray_mode(u32 tiling_flags) @@ -103,19 +107,6 @@ static u32 evergreen_cs_get_num_banks(u32 nbanks) } } -static u32 evergreen_cs_get_tile_split(u32 row_size) -{ - switch (row_size) { - case 1: - default: - return ADDR_SURF_TILE_SPLIT_1KB; - case 2: - return ADDR_SURF_TILE_SPLIT_2KB; - case 4: - return ADDR_SURF_TILE_SPLIT_4KB; - } -} - static void evergreen_cs_track_init(struct evergreen_cs_track *track) { int i; @@ -128,50 +119,745 @@ static void evergreen_cs_track_init(struct evergreen_cs_track *track) } for (i = 0; i < 12; i++) { - track->cb_color_base_last[i] = 0; track->cb_color_bo[i] = NULL; track->cb_color_bo_offset[i] = 0xFFFFFFFF; track->cb_color_info[i] = 0; - track->cb_color_view[i] = 0; - track->cb_color_pitch_idx[i] = 0; - track->cb_color_slice_idx[i] = 0; - track->cb_color_dim[i] = 0; + track->cb_color_view[i] = 0xFFFFFFFF; track->cb_color_pitch[i] = 0; track->cb_color_slice[i] = 0; - track->cb_color_dim[i] = 0; } track->cb_target_mask = 0xFFFFFFFF; track->cb_shader_mask = 0xFFFFFFFF; + track->cb_dirty = true; track->db_depth_view = 0xFFFFC000; track->db_depth_size = 0xFFFFFFFF; - track->db_depth_size_idx = 0; track->db_depth_control = 0xFFFFFFFF; track->db_z_info = 0xFFFFFFFF; - track->db_z_idx = 0xFFFFFFFF; track->db_z_read_offset = 0xFFFFFFFF; track->db_z_write_offset = 0xFFFFFFFF; track->db_z_read_bo = NULL; track->db_z_write_bo = NULL; track->db_s_info = 0xFFFFFFFF; - track->db_s_idx = 0xFFFFFFFF; track->db_s_read_offset = 0xFFFFFFFF; track->db_s_write_offset = 0xFFFFFFFF; track->db_s_read_bo = NULL; track->db_s_write_bo = NULL; + track->db_dirty = true; + + for (i = 0; i < 4; i++) { + track->vgt_strmout_size[i] = 0; + track->vgt_strmout_bo[i] = NULL; + track->vgt_strmout_bo_offset[i] = 0xFFFFFFFF; + } + track->streamout_dirty = true; + track->sx_misc_kill_all_prims = false; } -static int evergreen_cs_track_check(struct radeon_cs_parser *p) +struct eg_surface { + /* value gathered from cs */ + unsigned nbx; + unsigned nby; + unsigned format; + unsigned mode; + unsigned nbanks; + unsigned bankw; + unsigned bankh; + unsigned tsplit; + unsigned mtilea; + unsigned nsamples; + /* output value */ + unsigned bpe; + unsigned layer_size; + unsigned palign; + unsigned halign; + unsigned long base_align; +}; + +static int evergreen_surface_check_linear(struct radeon_cs_parser *p, + struct eg_surface *surf, + const char *prefix) +{ + surf->layer_size = surf->nbx * surf->nby * surf->bpe * surf->nsamples; + surf->base_align = surf->bpe; + surf->palign = 1; + surf->halign = 1; + return 0; +} + +static int evergreen_surface_check_linear_aligned(struct radeon_cs_parser *p, + struct eg_surface *surf, + const char *prefix) { struct evergreen_cs_track *track = p->track; + unsigned palign; - /* we don't support stream out buffer yet */ - if (track->vgt_strmout_config || track->vgt_strmout_buffer_config) { - dev_warn(p->dev, "this kernel doesn't support SMX output buffer\n"); + palign = MAX(64, track->group_size / surf->bpe); + surf->layer_size = surf->nbx * surf->nby * surf->bpe * surf->nsamples; + surf->base_align = track->group_size; + surf->palign = palign; + surf->halign = 1; + if (surf->nbx & (palign - 1)) { + if (prefix) { + dev_warn(p->dev, "%s:%d %s pitch %d invalid must be aligned with %d\n", + __func__, __LINE__, prefix, surf->nbx, palign); + } + return -EINVAL; + } + return 0; +} + +static int evergreen_surface_check_1d(struct radeon_cs_parser *p, + struct eg_surface *surf, + const char *prefix) +{ + struct evergreen_cs_track *track = p->track; + unsigned palign; + + palign = track->group_size / (8 * surf->bpe * surf->nsamples); + palign = MAX(8, palign); + surf->layer_size = surf->nbx * surf->nby * surf->bpe; + surf->base_align = track->group_size; + surf->palign = palign; + surf->halign = 8; + if ((surf->nbx & (palign - 1))) { + if (prefix) { + dev_warn(p->dev, "%s:%d %s pitch %d invalid must be aligned with %d (%d %d %d)\n", + __func__, __LINE__, prefix, surf->nbx, palign, + track->group_size, surf->bpe, surf->nsamples); + } + return -EINVAL; + } + if ((surf->nby & (8 - 1))) { + if (prefix) { + dev_warn(p->dev, "%s:%d %s height %d invalid must be aligned with 8\n", + __func__, __LINE__, prefix, surf->nby); + } + return -EINVAL; + } + return 0; +} + +static int evergreen_surface_check_2d(struct radeon_cs_parser *p, + struct eg_surface *surf, + const char *prefix) +{ + struct evergreen_cs_track *track = p->track; + unsigned palign, halign, tileb, slice_pt; + + tileb = 64 * surf->bpe * surf->nsamples; + palign = track->group_size / (8 * surf->bpe * surf->nsamples); + palign = MAX(8, palign); + slice_pt = 1; + if (tileb > surf->tsplit) { + slice_pt = tileb / surf->tsplit; + } + tileb = tileb / slice_pt; + /* macro tile width & height */ + palign = (8 * surf->bankw * track->npipes) * surf->mtilea; + halign = (8 * surf->bankh * surf->nbanks) / surf->mtilea; + surf->layer_size = surf->nbx * surf->nby * surf->bpe * slice_pt; + surf->base_align = (palign / 8) * (halign / 8) * tileb; + surf->palign = palign; + surf->halign = halign; + + if ((surf->nbx & (palign - 1))) { + if (prefix) { + dev_warn(p->dev, "%s:%d %s pitch %d invalid must be aligned with %d\n", + __func__, __LINE__, prefix, surf->nbx, palign); + } + return -EINVAL; + } + if ((surf->nby & (halign - 1))) { + if (prefix) { + dev_warn(p->dev, "%s:%d %s height %d invalid must be aligned with %d\n", + __func__, __LINE__, prefix, surf->nby, halign); + } return -EINVAL; } - /* XXX fill in */ + return 0; +} + +static int evergreen_surface_check(struct radeon_cs_parser *p, + struct eg_surface *surf, + const char *prefix) +{ + /* some common value computed here */ + surf->bpe = r600_fmt_get_blocksize(surf->format); + + switch (surf->mode) { + case ARRAY_LINEAR_GENERAL: + return evergreen_surface_check_linear(p, surf, prefix); + case ARRAY_LINEAR_ALIGNED: + return evergreen_surface_check_linear_aligned(p, surf, prefix); + case ARRAY_1D_TILED_THIN1: + return evergreen_surface_check_1d(p, surf, prefix); + case ARRAY_2D_TILED_THIN1: + return evergreen_surface_check_2d(p, surf, prefix); + default: + dev_warn(p->dev, "%s:%d %s invalid array mode %d\n", + __func__, __LINE__, prefix, surf->mode); + return -EINVAL; + } + return -EINVAL; +} + +static int evergreen_surface_value_conv_check(struct radeon_cs_parser *p, + struct eg_surface *surf, + const char *prefix) +{ + switch (surf->mode) { + case ARRAY_2D_TILED_THIN1: + break; + case ARRAY_LINEAR_GENERAL: + case ARRAY_LINEAR_ALIGNED: + case ARRAY_1D_TILED_THIN1: + return 0; + default: + dev_warn(p->dev, "%s:%d %s invalid array mode %d\n", + __func__, __LINE__, prefix, surf->mode); + return -EINVAL; + } + + switch (surf->nbanks) { + case 0: surf->nbanks = 2; break; + case 1: surf->nbanks = 4; break; + case 2: surf->nbanks = 8; break; + case 3: surf->nbanks = 16; break; + default: + dev_warn(p->dev, "%s:%d %s invalid number of banks %d\n", + __func__, __LINE__, prefix, surf->nbanks); + return -EINVAL; + } + switch (surf->bankw) { + case 0: surf->bankw = 1; break; + case 1: surf->bankw = 2; break; + case 2: surf->bankw = 4; break; + case 3: surf->bankw = 8; break; + default: + dev_warn(p->dev, "%s:%d %s invalid bankw %d\n", + __func__, __LINE__, prefix, surf->bankw); + return -EINVAL; + } + switch (surf->bankh) { + case 0: surf->bankh = 1; break; + case 1: surf->bankh = 2; break; + case 2: surf->bankh = 4; break; + case 3: surf->bankh = 8; break; + default: + dev_warn(p->dev, "%s:%d %s invalid bankh %d\n", + __func__, __LINE__, prefix, surf->bankh); + return -EINVAL; + } + switch (surf->mtilea) { + case 0: surf->mtilea = 1; break; + case 1: surf->mtilea = 2; break; + case 2: surf->mtilea = 4; break; + case 3: surf->mtilea = 8; break; + default: + dev_warn(p->dev, "%s:%d %s invalid macro tile aspect %d\n", + __func__, __LINE__, prefix, surf->mtilea); + return -EINVAL; + } + switch (surf->tsplit) { + case 0: surf->tsplit = 64; break; + case 1: surf->tsplit = 128; break; + case 2: surf->tsplit = 256; break; + case 3: surf->tsplit = 512; break; + case 4: surf->tsplit = 1024; break; + case 5: surf->tsplit = 2048; break; + case 6: surf->tsplit = 4096; break; + default: + dev_warn(p->dev, "%s:%d %s invalid tile split %d\n", + __func__, __LINE__, prefix, surf->tsplit); + return -EINVAL; + } + return 0; +} + +static int evergreen_cs_track_validate_cb(struct radeon_cs_parser *p, unsigned id) +{ + struct evergreen_cs_track *track = p->track; + struct eg_surface surf; + unsigned pitch, slice, mslice; + unsigned long offset; + int r; + + mslice = G_028C6C_SLICE_MAX(track->cb_color_view[id]) + 1; + pitch = track->cb_color_pitch[id]; + slice = track->cb_color_slice[id]; + surf.nbx = (pitch + 1) * 8; + surf.nby = ((slice + 1) * 64) / surf.nbx; + surf.mode = G_028C70_ARRAY_MODE(track->cb_color_info[id]); + surf.format = G_028C70_FORMAT(track->cb_color_info[id]); + surf.tsplit = G_028C74_TILE_SPLIT(track->cb_color_attrib[id]); + surf.nbanks = G_028C74_NUM_BANKS(track->cb_color_attrib[id]); + surf.bankw = G_028C74_BANK_WIDTH(track->cb_color_attrib[id]); + surf.bankh = G_028C74_BANK_HEIGHT(track->cb_color_attrib[id]); + surf.mtilea = G_028C74_MACRO_TILE_ASPECT(track->cb_color_attrib[id]); + surf.nsamples = 1; + + if (!r600_fmt_is_valid_color(surf.format)) { + dev_warn(p->dev, "%s:%d cb invalid format %d for %d (0x%08x)\n", + __func__, __LINE__, surf.format, + id, track->cb_color_info[id]); + return -EINVAL; + } + + r = evergreen_surface_value_conv_check(p, &surf, "cb"); + if (r) { + return r; + } + + r = evergreen_surface_check(p, &surf, "cb"); + if (r) { + dev_warn(p->dev, "%s:%d cb[%d] invalid (0x%08x 0x%08x 0x%08x 0x%08x)\n", + __func__, __LINE__, id, track->cb_color_pitch[id], + track->cb_color_slice[id], track->cb_color_attrib[id], + track->cb_color_info[id]); + return r; + } + + offset = track->cb_color_bo_offset[id] << 8; + if (offset & (surf.base_align - 1)) { + dev_warn(p->dev, "%s:%d cb[%d] bo base %ld not aligned with %ld\n", + __func__, __LINE__, id, offset, surf.base_align); + return -EINVAL; + } + + offset += surf.layer_size * mslice; + if (offset > radeon_bo_size(track->cb_color_bo[id])) { + dev_warn(p->dev, "%s:%d cb[%d] bo too small (layer size %d, " + "offset %d, max layer %d, bo size %ld, slice %d)\n", + __func__, __LINE__, id, surf.layer_size, + track->cb_color_bo_offset[id] << 8, mslice, + radeon_bo_size(track->cb_color_bo[id]), slice); + dev_warn(p->dev, "%s:%d problematic surf: (%d %d) (%d %d %d %d %d %d %d)\n", + __func__, __LINE__, surf.nbx, surf.nby, + surf.mode, surf.bpe, surf.nsamples, + surf.bankw, surf.bankh, + surf.tsplit, surf.mtilea); + return -EINVAL; + } + + return 0; +} + +static int evergreen_cs_track_validate_stencil(struct radeon_cs_parser *p) +{ + struct evergreen_cs_track *track = p->track; + struct eg_surface surf; + unsigned pitch, slice, mslice; + unsigned long offset; + int r; + + mslice = G_028008_SLICE_MAX(track->db_depth_view) + 1; + pitch = G_028058_PITCH_TILE_MAX(track->db_depth_size); + slice = track->db_depth_slice; + surf.nbx = (pitch + 1) * 8; + surf.nby = ((slice + 1) * 64) / surf.nbx; + surf.mode = G_028040_ARRAY_MODE(track->db_z_info); + surf.format = G_028044_FORMAT(track->db_s_info); + surf.tsplit = G_028044_TILE_SPLIT(track->db_s_info); + surf.nbanks = G_028040_NUM_BANKS(track->db_z_info); + surf.bankw = G_028040_BANK_WIDTH(track->db_z_info); + surf.bankh = G_028040_BANK_HEIGHT(track->db_z_info); + surf.mtilea = G_028040_MACRO_TILE_ASPECT(track->db_z_info); + surf.nsamples = 1; + + if (surf.format != 1) { + dev_warn(p->dev, "%s:%d stencil invalid format %d\n", + __func__, __LINE__, surf.format); + return -EINVAL; + } + /* replace by color format so we can use same code */ + surf.format = V_028C70_COLOR_8; + + r = evergreen_surface_value_conv_check(p, &surf, "stencil"); + if (r) { + return r; + } + + r = evergreen_surface_check(p, &surf, NULL); + if (r) { + /* old userspace doesn't compute proper depth/stencil alignment + * check that alignment against a bigger byte per elements and + * only report if that alignment is wrong too. + */ + surf.format = V_028C70_COLOR_8_8_8_8; + r = evergreen_surface_check(p, &surf, "stencil"); + if (r) { + dev_warn(p->dev, "%s:%d stencil invalid (0x%08x 0x%08x 0x%08x 0x%08x)\n", + __func__, __LINE__, track->db_depth_size, + track->db_depth_slice, track->db_s_info, track->db_z_info); + } + return r; + } + + offset = track->db_s_read_offset << 8; + if (offset & (surf.base_align - 1)) { + dev_warn(p->dev, "%s:%d stencil read bo base %ld not aligned with %ld\n", + __func__, __LINE__, offset, surf.base_align); + return -EINVAL; + } + offset += surf.layer_size * mslice; + if (offset > radeon_bo_size(track->db_s_read_bo)) { + dev_warn(p->dev, "%s:%d stencil read bo too small (layer size %d, " + "offset %ld, max layer %d, bo size %ld)\n", + __func__, __LINE__, surf.layer_size, + (unsigned long)track->db_s_read_offset << 8, mslice, + radeon_bo_size(track->db_s_read_bo)); + dev_warn(p->dev, "%s:%d stencil invalid (0x%08x 0x%08x 0x%08x 0x%08x)\n", + __func__, __LINE__, track->db_depth_size, + track->db_depth_slice, track->db_s_info, track->db_z_info); + return -EINVAL; + } + + offset = track->db_s_write_offset << 8; + if (offset & (surf.base_align - 1)) { + dev_warn(p->dev, "%s:%d stencil write bo base %ld not aligned with %ld\n", + __func__, __LINE__, offset, surf.base_align); + return -EINVAL; + } + offset += surf.layer_size * mslice; + if (offset > radeon_bo_size(track->db_s_write_bo)) { + dev_warn(p->dev, "%s:%d stencil write bo too small (layer size %d, " + "offset %ld, max layer %d, bo size %ld)\n", + __func__, __LINE__, surf.layer_size, + (unsigned long)track->db_s_write_offset << 8, mslice, + radeon_bo_size(track->db_s_write_bo)); + return -EINVAL; + } + + return 0; +} + +static int evergreen_cs_track_validate_depth(struct radeon_cs_parser *p) +{ + struct evergreen_cs_track *track = p->track; + struct eg_surface surf; + unsigned pitch, slice, mslice; + unsigned long offset; + int r; + + mslice = G_028008_SLICE_MAX(track->db_depth_view) + 1; + pitch = G_028058_PITCH_TILE_MAX(track->db_depth_size); + slice = track->db_depth_slice; + surf.nbx = (pitch + 1) * 8; + surf.nby = ((slice + 1) * 64) / surf.nbx; + surf.mode = G_028040_ARRAY_MODE(track->db_z_info); + surf.format = G_028040_FORMAT(track->db_z_info); + surf.tsplit = G_028040_TILE_SPLIT(track->db_z_info); + surf.nbanks = G_028040_NUM_BANKS(track->db_z_info); + surf.bankw = G_028040_BANK_WIDTH(track->db_z_info); + surf.bankh = G_028040_BANK_HEIGHT(track->db_z_info); + surf.mtilea = G_028040_MACRO_TILE_ASPECT(track->db_z_info); + surf.nsamples = 1; + + switch (surf.format) { + case V_028040_Z_16: + surf.format = V_028C70_COLOR_16; + break; + case V_028040_Z_24: + case V_028040_Z_32_FLOAT: + surf.format = V_028C70_COLOR_8_8_8_8; + break; + default: + dev_warn(p->dev, "%s:%d depth invalid format %d\n", + __func__, __LINE__, surf.format); + return -EINVAL; + } + + r = evergreen_surface_value_conv_check(p, &surf, "depth"); + if (r) { + dev_warn(p->dev, "%s:%d depth invalid (0x%08x 0x%08x 0x%08x)\n", + __func__, __LINE__, track->db_depth_size, + track->db_depth_slice, track->db_z_info); + return r; + } + + r = evergreen_surface_check(p, &surf, "depth"); + if (r) { + dev_warn(p->dev, "%s:%d depth invalid (0x%08x 0x%08x 0x%08x)\n", + __func__, __LINE__, track->db_depth_size, + track->db_depth_slice, track->db_z_info); + return r; + } + + offset = track->db_z_read_offset << 8; + if (offset & (surf.base_align - 1)) { + dev_warn(p->dev, "%s:%d stencil read bo base %ld not aligned with %ld\n", + __func__, __LINE__, offset, surf.base_align); + return -EINVAL; + } + offset += surf.layer_size * mslice; + if (offset > radeon_bo_size(track->db_z_read_bo)) { + dev_warn(p->dev, "%s:%d depth read bo too small (layer size %d, " + "offset %ld, max layer %d, bo size %ld)\n", + __func__, __LINE__, surf.layer_size, + (unsigned long)track->db_z_read_offset << 8, mslice, + radeon_bo_size(track->db_z_read_bo)); + return -EINVAL; + } + + offset = track->db_z_write_offset << 8; + if (offset & (surf.base_align - 1)) { + dev_warn(p->dev, "%s:%d stencil write bo base %ld not aligned with %ld\n", + __func__, __LINE__, offset, surf.base_align); + return -EINVAL; + } + offset += surf.layer_size * mslice; + if (offset > radeon_bo_size(track->db_z_write_bo)) { + dev_warn(p->dev, "%s:%d depth write bo too small (layer size %d, " + "offset %ld, max layer %d, bo size %ld)\n", + __func__, __LINE__, surf.layer_size, + (unsigned long)track->db_z_write_offset << 8, mslice, + radeon_bo_size(track->db_z_write_bo)); + return -EINVAL; + } + + return 0; +} + +static int evergreen_cs_track_validate_texture(struct radeon_cs_parser *p, + struct radeon_bo *texture, + struct radeon_bo *mipmap, + unsigned idx) +{ + struct eg_surface surf; + unsigned long toffset, moffset; + unsigned dim, llevel, mslice, width, height, depth, i; + u32 texdw[8]; + int r; + + texdw[0] = radeon_get_ib_value(p, idx + 0); + texdw[1] = radeon_get_ib_value(p, idx + 1); + texdw[2] = radeon_get_ib_value(p, idx + 2); + texdw[3] = radeon_get_ib_value(p, idx + 3); + texdw[4] = radeon_get_ib_value(p, idx + 4); + texdw[5] = radeon_get_ib_value(p, idx + 5); + texdw[6] = radeon_get_ib_value(p, idx + 6); + texdw[7] = radeon_get_ib_value(p, idx + 7); + dim = G_030000_DIM(texdw[0]); + llevel = G_030014_LAST_LEVEL(texdw[5]); + mslice = G_030014_LAST_ARRAY(texdw[5]) + 1; + width = G_030000_TEX_WIDTH(texdw[0]) + 1; + height = G_030004_TEX_HEIGHT(texdw[1]) + 1; + depth = G_030004_TEX_DEPTH(texdw[1]) + 1; + surf.format = G_03001C_DATA_FORMAT(texdw[7]); + surf.nbx = (G_030000_PITCH(texdw[0]) + 1) * 8; + surf.nbx = r600_fmt_get_nblocksx(surf.format, surf.nbx); + surf.nby = r600_fmt_get_nblocksy(surf.format, height); + surf.mode = G_030004_ARRAY_MODE(texdw[1]); + surf.tsplit = G_030018_TILE_SPLIT(texdw[6]); + surf.nbanks = G_03001C_NUM_BANKS(texdw[7]); + surf.bankw = G_03001C_BANK_WIDTH(texdw[7]); + surf.bankh = G_03001C_BANK_HEIGHT(texdw[7]); + surf.mtilea = G_03001C_MACRO_TILE_ASPECT(texdw[7]); + surf.nsamples = 1; + toffset = texdw[2] << 8; + moffset = texdw[3] << 8; + + if (!r600_fmt_is_valid_texture(surf.format, p->family)) { + dev_warn(p->dev, "%s:%d texture invalid format %d\n", + __func__, __LINE__, surf.format); + return -EINVAL; + } + switch (dim) { + case V_030000_SQ_TEX_DIM_1D: + case V_030000_SQ_TEX_DIM_2D: + case V_030000_SQ_TEX_DIM_CUBEMAP: + case V_030000_SQ_TEX_DIM_1D_ARRAY: + case V_030000_SQ_TEX_DIM_2D_ARRAY: + depth = 1; + case V_030000_SQ_TEX_DIM_3D: + break; + default: + dev_warn(p->dev, "%s:%d texture invalid dimension %d\n", + __func__, __LINE__, dim); + return -EINVAL; + } + + r = evergreen_surface_value_conv_check(p, &surf, "texture"); + if (r) { + return r; + } + + /* align height */ + evergreen_surface_check(p, &surf, NULL); + surf.nby = ALIGN(surf.nby, surf.halign); + + r = evergreen_surface_check(p, &surf, "texture"); + if (r) { + dev_warn(p->dev, "%s:%d texture invalid 0x%08x 0x%08x 0x%08x 0x%08x 0x%08x 0x%08x\n", + __func__, __LINE__, texdw[0], texdw[1], texdw[4], + texdw[5], texdw[6], texdw[7]); + return r; + } + + /* check texture size */ + if (toffset & (surf.base_align - 1)) { + dev_warn(p->dev, "%s:%d texture bo base %ld not aligned with %ld\n", + __func__, __LINE__, toffset, surf.base_align); + return -EINVAL; + } + if (moffset & (surf.base_align - 1)) { + dev_warn(p->dev, "%s:%d mipmap bo base %ld not aligned with %ld\n", + __func__, __LINE__, moffset, surf.base_align); + return -EINVAL; + } + if (dim == SQ_TEX_DIM_3D) { + toffset += surf.layer_size * depth; + } else { + toffset += surf.layer_size * mslice; + } + if (toffset > radeon_bo_size(texture)) { + dev_warn(p->dev, "%s:%d texture bo too small (layer size %d, " + "offset %ld, max layer %d, depth %d, bo size %ld) (%d %d)\n", + __func__, __LINE__, surf.layer_size, + (unsigned long)texdw[2] << 8, mslice, + depth, radeon_bo_size(texture), + surf.nbx, surf.nby); + return -EINVAL; + } + + /* check mipmap size */ + for (i = 1; i <= llevel; i++) { + unsigned w, h, d; + + w = r600_mip_minify(width, i); + h = r600_mip_minify(height, i); + d = r600_mip_minify(depth, i); + surf.nbx = r600_fmt_get_nblocksx(surf.format, w); + surf.nby = r600_fmt_get_nblocksy(surf.format, h); + + switch (surf.mode) { + case ARRAY_2D_TILED_THIN1: + if (surf.nbx < surf.palign || surf.nby < surf.halign) { + surf.mode = ARRAY_1D_TILED_THIN1; + } + /* recompute alignment */ + evergreen_surface_check(p, &surf, NULL); + break; + case ARRAY_LINEAR_GENERAL: + case ARRAY_LINEAR_ALIGNED: + case ARRAY_1D_TILED_THIN1: + break; + default: + dev_warn(p->dev, "%s:%d invalid array mode %d\n", + __func__, __LINE__, surf.mode); + return -EINVAL; + } + surf.nbx = ALIGN(surf.nbx, surf.palign); + surf.nby = ALIGN(surf.nby, surf.halign); + + r = evergreen_surface_check(p, &surf, "mipmap"); + if (r) { + return r; + } + + if (dim == SQ_TEX_DIM_3D) { + moffset += surf.layer_size * d; + } else { + moffset += surf.layer_size * mslice; + } + if (moffset > radeon_bo_size(mipmap)) { + dev_warn(p->dev, "%s:%d mipmap [%d] bo too small (layer size %d, " + "offset %ld, coffset %ld, max layer %d, depth %d, " + "bo size %ld) level0 (%d %d %d)\n", + __func__, __LINE__, i, surf.layer_size, + (unsigned long)texdw[3] << 8, moffset, mslice, + d, radeon_bo_size(mipmap), + width, height, depth); + dev_warn(p->dev, "%s:%d problematic surf: (%d %d) (%d %d %d %d %d %d %d)\n", + __func__, __LINE__, surf.nbx, surf.nby, + surf.mode, surf.bpe, surf.nsamples, + surf.bankw, surf.bankh, + surf.tsplit, surf.mtilea); + return -EINVAL; + } + } + + return 0; +} + +static int evergreen_cs_track_check(struct radeon_cs_parser *p) +{ + struct evergreen_cs_track *track = p->track; + unsigned tmp, i; + int r; + unsigned buffer_mask = 0; + + /* check streamout */ + if (track->streamout_dirty && track->vgt_strmout_config) { + for (i = 0; i < 4; i++) { + if (track->vgt_strmout_config & (1 << i)) { + buffer_mask |= (track->vgt_strmout_buffer_config >> (i * 4)) & 0xf; + } + } + + for (i = 0; i < 4; i++) { + if (buffer_mask & (1 << i)) { + if (track->vgt_strmout_bo[i]) { + u64 offset = (u64)track->vgt_strmout_bo_offset[i] + + (u64)track->vgt_strmout_size[i]; + if (offset > radeon_bo_size(track->vgt_strmout_bo[i])) { + DRM_ERROR("streamout %d bo too small: 0x%llx, 0x%lx\n", + i, offset, + radeon_bo_size(track->vgt_strmout_bo[i])); + return -EINVAL; + } + } else { + dev_warn(p->dev, "No buffer for streamout %d\n", i); + return -EINVAL; + } + } + } + track->streamout_dirty = false; + } + + if (track->sx_misc_kill_all_prims) + return 0; + + /* check that we have a cb for each enabled target + */ + if (track->cb_dirty) { + tmp = track->cb_target_mask; + for (i = 0; i < 8; i++) { + if ((tmp >> (i * 4)) & 0xF) { + /* at least one component is enabled */ + if (track->cb_color_bo[i] == NULL) { + dev_warn(p->dev, "%s:%d mask 0x%08X | 0x%08X no cb for %d\n", + __func__, __LINE__, track->cb_target_mask, track->cb_shader_mask, i); + return -EINVAL; + } + /* check cb */ + r = evergreen_cs_track_validate_cb(p, i); + if (r) { + return r; + } + } + } + track->cb_dirty = false; + } + + if (track->db_dirty) { + /* Check stencil buffer */ + if (G_028800_STENCIL_ENABLE(track->db_depth_control)) { + r = evergreen_cs_track_validate_stencil(p); + if (r) + return r; + } + /* Check depth buffer */ + if (G_028800_Z_WRITE_ENABLE(track->db_depth_control)) { + r = evergreen_cs_track_validate_depth(p); + if (r) + return r; + } + track->db_dirty = false; + } + return 0; } @@ -503,6 +1189,7 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) break; case DB_DEPTH_CONTROL: track->db_depth_control = radeon_get_ib_value(p, idx); + track->db_dirty = true; break; case CAYMAN_DB_EQAA: if (p->rdev->family < CHIP_CAYMAN) { @@ -532,20 +1219,35 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) ib[idx] |= Z_ARRAY_MODE(evergreen_cs_get_aray_mode(reloc->lobj.tiling_flags)); track->db_z_info |= Z_ARRAY_MODE(evergreen_cs_get_aray_mode(reloc->lobj.tiling_flags)); if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO) { + unsigned bankw, bankh, mtaspect, tile_split; + + evergreen_tiling_fields(reloc->lobj.tiling_flags, + &bankw, &bankh, &mtaspect, + &tile_split); ib[idx] |= DB_NUM_BANKS(evergreen_cs_get_num_banks(track->nbanks)); - ib[idx] |= DB_TILE_SPLIT(evergreen_cs_get_tile_split(track->row_size)); + ib[idx] |= DB_TILE_SPLIT(tile_split) | + DB_BANK_WIDTH(bankw) | + DB_BANK_HEIGHT(bankh) | + DB_MACRO_TILE_ASPECT(mtaspect); } } + track->db_dirty = true; break; case DB_STENCIL_INFO: track->db_s_info = radeon_get_ib_value(p, idx); + track->db_dirty = true; break; case DB_DEPTH_VIEW: track->db_depth_view = radeon_get_ib_value(p, idx); + track->db_dirty = true; break; case DB_DEPTH_SIZE: track->db_depth_size = radeon_get_ib_value(p, idx); - track->db_depth_size_idx = idx; + track->db_dirty = true; + break; + case R_02805C_DB_DEPTH_SLICE: + track->db_depth_slice = radeon_get_ib_value(p, idx); + track->db_dirty = true; break; case DB_Z_READ_BASE: r = evergreen_cs_packet_next_reloc(p, &reloc); @@ -557,6 +1259,7 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) track->db_z_read_offset = radeon_get_ib_value(p, idx); ib[idx] += (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff); track->db_z_read_bo = reloc->robj; + track->db_dirty = true; break; case DB_Z_WRITE_BASE: r = evergreen_cs_packet_next_reloc(p, &reloc); @@ -568,6 +1271,7 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) track->db_z_write_offset = radeon_get_ib_value(p, idx); ib[idx] += (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff); track->db_z_write_bo = reloc->robj; + track->db_dirty = true; break; case DB_STENCIL_READ_BASE: r = evergreen_cs_packet_next_reloc(p, &reloc); @@ -579,6 +1283,7 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) track->db_s_read_offset = radeon_get_ib_value(p, idx); ib[idx] += (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff); track->db_s_read_bo = reloc->robj; + track->db_dirty = true; break; case DB_STENCIL_WRITE_BASE: r = evergreen_cs_packet_next_reloc(p, &reloc); @@ -590,18 +1295,56 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) track->db_s_write_offset = radeon_get_ib_value(p, idx); ib[idx] += (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff); track->db_s_write_bo = reloc->robj; + track->db_dirty = true; break; case VGT_STRMOUT_CONFIG: track->vgt_strmout_config = radeon_get_ib_value(p, idx); + track->streamout_dirty = true; break; case VGT_STRMOUT_BUFFER_CONFIG: track->vgt_strmout_buffer_config = radeon_get_ib_value(p, idx); + track->streamout_dirty = true; break; + case VGT_STRMOUT_BUFFER_BASE_0: + case VGT_STRMOUT_BUFFER_BASE_1: + case VGT_STRMOUT_BUFFER_BASE_2: + case VGT_STRMOUT_BUFFER_BASE_3: + r = evergreen_cs_packet_next_reloc(p, &reloc); + if (r) { + dev_warn(p->dev, "bad SET_CONTEXT_REG " + "0x%04X\n", reg); + return -EINVAL; + } + tmp = (reg - VGT_STRMOUT_BUFFER_BASE_0) / 16; + track->vgt_strmout_bo_offset[tmp] = radeon_get_ib_value(p, idx) << 8; + ib[idx] += (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff); + track->vgt_strmout_bo[tmp] = reloc->robj; + track->streamout_dirty = true; + break; + case VGT_STRMOUT_BUFFER_SIZE_0: + case VGT_STRMOUT_BUFFER_SIZE_1: + case VGT_STRMOUT_BUFFER_SIZE_2: + case VGT_STRMOUT_BUFFER_SIZE_3: + tmp = (reg - VGT_STRMOUT_BUFFER_SIZE_0) / 16; + /* size in register is DWs, convert to bytes */ + track->vgt_strmout_size[tmp] = radeon_get_ib_value(p, idx) * 4; + track->streamout_dirty = true; + break; + case CP_COHER_BASE: + r = evergreen_cs_packet_next_reloc(p, &reloc); + if (r) { + dev_warn(p->dev, "missing reloc for CP_COHER_BASE " + "0x%04X\n", reg); + return -EINVAL; + } + ib[idx] += (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff); case CB_TARGET_MASK: track->cb_target_mask = radeon_get_ib_value(p, idx); + track->cb_dirty = true; break; case CB_SHADER_MASK: track->cb_shader_mask = radeon_get_ib_value(p, idx); + track->cb_dirty = true; break; case PA_SC_AA_CONFIG: if (p->rdev->family >= CHIP_CAYMAN) { @@ -631,6 +1374,7 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) case CB_COLOR7_VIEW: tmp = (reg - CB_COLOR0_VIEW) / 0x3c; track->cb_color_view[tmp] = radeon_get_ib_value(p, idx); + track->cb_dirty = true; break; case CB_COLOR8_VIEW: case CB_COLOR9_VIEW: @@ -638,6 +1382,7 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) case CB_COLOR11_VIEW: tmp = ((reg - CB_COLOR8_VIEW) / 0x1c) + 8; track->cb_color_view[tmp] = radeon_get_ib_value(p, idx); + track->cb_dirty = true; break; case CB_COLOR0_INFO: case CB_COLOR1_INFO: @@ -659,6 +1404,7 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) ib[idx] |= CB_ARRAY_MODE(evergreen_cs_get_aray_mode(reloc->lobj.tiling_flags)); track->cb_color_info[tmp] |= CB_ARRAY_MODE(evergreen_cs_get_aray_mode(reloc->lobj.tiling_flags)); } + track->cb_dirty = true; break; case CB_COLOR8_INFO: case CB_COLOR9_INFO: @@ -676,6 +1422,7 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) ib[idx] |= CB_ARRAY_MODE(evergreen_cs_get_aray_mode(reloc->lobj.tiling_flags)); track->cb_color_info[tmp] |= CB_ARRAY_MODE(evergreen_cs_get_aray_mode(reloc->lobj.tiling_flags)); } + track->cb_dirty = true; break; case CB_COLOR0_PITCH: case CB_COLOR1_PITCH: @@ -687,7 +1434,7 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) case CB_COLOR7_PITCH: tmp = (reg - CB_COLOR0_PITCH) / 0x3c; track->cb_color_pitch[tmp] = radeon_get_ib_value(p, idx); - track->cb_color_pitch_idx[tmp] = idx; + track->cb_dirty = true; break; case CB_COLOR8_PITCH: case CB_COLOR9_PITCH: @@ -695,7 +1442,7 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) case CB_COLOR11_PITCH: tmp = ((reg - CB_COLOR8_PITCH) / 0x1c) + 8; track->cb_color_pitch[tmp] = radeon_get_ib_value(p, idx); - track->cb_color_pitch_idx[tmp] = idx; + track->cb_dirty = true; break; case CB_COLOR0_SLICE: case CB_COLOR1_SLICE: @@ -707,7 +1454,7 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) case CB_COLOR7_SLICE: tmp = (reg - CB_COLOR0_SLICE) / 0x3c; track->cb_color_slice[tmp] = radeon_get_ib_value(p, idx); - track->cb_color_slice_idx[tmp] = idx; + track->cb_dirty = true; break; case CB_COLOR8_SLICE: case CB_COLOR9_SLICE: @@ -715,7 +1462,7 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) case CB_COLOR11_SLICE: tmp = ((reg - CB_COLOR8_SLICE) / 0x1c) + 8; track->cb_color_slice[tmp] = radeon_get_ib_value(p, idx); - track->cb_color_slice_idx[tmp] = idx; + track->cb_dirty = true; break; case CB_COLOR0_ATTRIB: case CB_COLOR1_ATTRIB: @@ -725,6 +1472,30 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) case CB_COLOR5_ATTRIB: case CB_COLOR6_ATTRIB: case CB_COLOR7_ATTRIB: + r = evergreen_cs_packet_next_reloc(p, &reloc); + if (r) { + dev_warn(p->dev, "bad SET_CONTEXT_REG " + "0x%04X\n", reg); + return -EINVAL; + } + if (!(p->cs_flags & RADEON_CS_KEEP_TILING_FLAGS)) { + if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO) { + unsigned bankw, bankh, mtaspect, tile_split; + + evergreen_tiling_fields(reloc->lobj.tiling_flags, + &bankw, &bankh, &mtaspect, + &tile_split); + ib[idx] |= CB_NUM_BANKS(evergreen_cs_get_num_banks(track->nbanks)); + ib[idx] |= CB_TILE_SPLIT(tile_split) | + CB_BANK_WIDTH(bankw) | + CB_BANK_HEIGHT(bankh) | + CB_MACRO_TILE_ASPECT(mtaspect); + } + } + tmp = ((reg - CB_COLOR0_ATTRIB) / 0x3c); + track->cb_color_attrib[tmp] = ib[idx]; + track->cb_dirty = true; + break; case CB_COLOR8_ATTRIB: case CB_COLOR9_ATTRIB: case CB_COLOR10_ATTRIB: @@ -735,30 +1506,23 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) "0x%04X\n", reg); return -EINVAL; } - if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO) { - ib[idx] |= CB_NUM_BANKS(evergreen_cs_get_num_banks(track->nbanks)); - ib[idx] |= CB_TILE_SPLIT(evergreen_cs_get_tile_split(track->row_size)); + if (!(p->cs_flags & RADEON_CS_KEEP_TILING_FLAGS)) { + if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO) { + unsigned bankw, bankh, mtaspect, tile_split; + + evergreen_tiling_fields(reloc->lobj.tiling_flags, + &bankw, &bankh, &mtaspect, + &tile_split); + ib[idx] |= CB_NUM_BANKS(evergreen_cs_get_num_banks(track->nbanks)); + ib[idx] |= CB_TILE_SPLIT(tile_split) | + CB_BANK_WIDTH(bankw) | + CB_BANK_HEIGHT(bankh) | + CB_MACRO_TILE_ASPECT(mtaspect); + } } - break; - case CB_COLOR0_DIM: - case CB_COLOR1_DIM: - case CB_COLOR2_DIM: - case CB_COLOR3_DIM: - case CB_COLOR4_DIM: - case CB_COLOR5_DIM: - case CB_COLOR6_DIM: - case CB_COLOR7_DIM: - tmp = (reg - CB_COLOR0_DIM) / 0x3c; - track->cb_color_dim[tmp] = radeon_get_ib_value(p, idx); - track->cb_color_dim_idx[tmp] = idx; - break; - case CB_COLOR8_DIM: - case CB_COLOR9_DIM: - case CB_COLOR10_DIM: - case CB_COLOR11_DIM: - tmp = ((reg - CB_COLOR8_DIM) / 0x1c) + 8; - track->cb_color_dim[tmp] = radeon_get_ib_value(p, idx); - track->cb_color_dim_idx[tmp] = idx; + tmp = ((reg - CB_COLOR8_ATTRIB) / 0x1c) + 8; + track->cb_color_attrib[tmp] = ib[idx]; + track->cb_dirty = true; break; case CB_COLOR0_FMASK: case CB_COLOR1_FMASK: @@ -833,8 +1597,8 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) tmp = (reg - CB_COLOR0_BASE) / 0x3c; track->cb_color_bo_offset[tmp] = radeon_get_ib_value(p, idx); ib[idx] += (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff); - track->cb_color_base_last[tmp] = ib[idx]; track->cb_color_bo[tmp] = reloc->robj; + track->cb_dirty = true; break; case CB_COLOR8_BASE: case CB_COLOR9_BASE: @@ -849,8 +1613,8 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) tmp = ((reg - CB_COLOR8_BASE) / 0x1c) + 8; track->cb_color_bo_offset[tmp] = radeon_get_ib_value(p, idx); ib[idx] += (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff); - track->cb_color_base_last[tmp] = ib[idx]; track->cb_color_bo[tmp] = reloc->robj; + track->cb_dirty = true; break; case CB_IMMED0_BASE: case CB_IMMED1_BASE: @@ -989,6 +1753,9 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) } ib[idx] += (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff); break; + case SX_MISC: + track->sx_misc_kill_all_prims = (radeon_get_ib_value(p, idx) & 0x1) != 0; + break; default: dev_warn(p->dev, "forbidden register 0x%08x at %d\n", reg, idx); return -EINVAL; @@ -996,22 +1763,30 @@ static int evergreen_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) return 0; } -/** - * evergreen_check_texture_resource() - check if register is authorized or not - * @p: parser structure holding parsing context - * @idx: index into the cs buffer - * @texture: texture's bo structure - * @mipmap: mipmap's bo structure - * - * This function will check that the resource has valid field and that - * the texture and mipmap bo object are big enough to cover this resource. - */ -static int evergreen_check_texture_resource(struct radeon_cs_parser *p, u32 idx, - struct radeon_bo *texture, - struct radeon_bo *mipmap) +static bool evergreen_is_safe_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) { - /* XXX fill in */ - return 0; + u32 last_reg, m, i; + + if (p->rdev->family >= CHIP_CAYMAN) + last_reg = ARRAY_SIZE(cayman_reg_safe_bm); + else + last_reg = ARRAY_SIZE(evergreen_reg_safe_bm); + + i = (reg >> 7); + if (i >= last_reg) { + dev_warn(p->dev, "forbidden register 0x%08x at %d\n", reg, idx); + return false; + } + m = 1 << ((reg >> 2) & 31); + if (p->rdev->family >= CHIP_CAYMAN) { + if (!(cayman_reg_safe_bm[i] & m)) + return true; + } else { + if (!(evergreen_reg_safe_bm[i] & m)) + return true; + } + dev_warn(p->dev, "forbidden register 0x%08x at %d\n", reg, idx); + return false; } static int evergreen_packet3_check(struct radeon_cs_parser *p, @@ -1036,6 +1811,8 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p, { int pred_op; int tmp; + uint64_t offset; + if (pkt->count != 1) { DRM_ERROR("bad SET PREDICATION\n"); return -EINVAL; @@ -1059,8 +1836,12 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p, return -EINVAL; } - ib[idx + 0] = idx_value + (u32)(reloc->lobj.gpu_offset & 0xffffffff); - ib[idx + 1] = tmp + (upper_32_bits(reloc->lobj.gpu_offset) & 0xff); + offset = reloc->lobj.gpu_offset + + (idx_value & 0xfffffff0) + + ((u64)(tmp & 0xff) << 32); + + ib[idx + 0] = offset; + ib[idx + 1] = (tmp & 0xffffff00) | (upper_32_bits(offset) & 0xff); } break; case PACKET3_CONTEXT_CONTROL: @@ -1088,6 +1869,9 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p, } break; case PACKET3_INDEX_BASE: + { + uint64_t offset; + if (pkt->count != 1) { DRM_ERROR("bad INDEX_BASE\n"); return -EINVAL; @@ -1097,15 +1881,24 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p, DRM_ERROR("bad INDEX_BASE\n"); return -EINVAL; } - ib[idx+0] = idx_value + (u32)(reloc->lobj.gpu_offset & 0xffffffff); - ib[idx+1] += upper_32_bits(reloc->lobj.gpu_offset) & 0xff; + + offset = reloc->lobj.gpu_offset + + idx_value + + ((u64)(radeon_get_ib_value(p, idx+1) & 0xff) << 32); + + ib[idx+0] = offset; + ib[idx+1] = upper_32_bits(offset) & 0xff; + r = evergreen_cs_track_check(p); if (r) { dev_warn(p->dev, "%s:%d invalid cmd stream\n", __func__, __LINE__); return r; } break; + } case PACKET3_DRAW_INDEX: + { + uint64_t offset; if (pkt->count != 3) { DRM_ERROR("bad DRAW_INDEX\n"); return -EINVAL; @@ -1115,15 +1908,25 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p, DRM_ERROR("bad DRAW_INDEX\n"); return -EINVAL; } - ib[idx+0] = idx_value + (u32)(reloc->lobj.gpu_offset & 0xffffffff); - ib[idx+1] += upper_32_bits(reloc->lobj.gpu_offset) & 0xff; + + offset = reloc->lobj.gpu_offset + + idx_value + + ((u64)(radeon_get_ib_value(p, idx+1) & 0xff) << 32); + + ib[idx+0] = offset; + ib[idx+1] = upper_32_bits(offset) & 0xff; + r = evergreen_cs_track_check(p); if (r) { dev_warn(p->dev, "%s:%d invalid cmd stream\n", __func__, __LINE__); return r; } break; + } case PACKET3_DRAW_INDEX_2: + { + uint64_t offset; + if (pkt->count != 4) { DRM_ERROR("bad DRAW_INDEX_2\n"); return -EINVAL; @@ -1133,14 +1936,21 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p, DRM_ERROR("bad DRAW_INDEX_2\n"); return -EINVAL; } - ib[idx+1] = idx_value + (u32)(reloc->lobj.gpu_offset & 0xffffffff); - ib[idx+2] += upper_32_bits(reloc->lobj.gpu_offset) & 0xff; + + offset = reloc->lobj.gpu_offset + + radeon_get_ib_value(p, idx+1) + + ((u64)(radeon_get_ib_value(p, idx+2) & 0xff) << 32); + + ib[idx+1] = offset; + ib[idx+2] = upper_32_bits(offset) & 0xff; + r = evergreen_cs_track_check(p); if (r) { dev_warn(p->dev, "%s:%d invalid cmd stream\n", __func__, __LINE__); return r; } break; + } case PACKET3_DRAW_INDEX_AUTO: if (pkt->count != 1) { DRM_ERROR("bad DRAW_INDEX_AUTO\n"); @@ -1231,13 +2041,20 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p, } /* bit 4 is reg (0) or mem (1) */ if (idx_value & 0x10) { + uint64_t offset; + r = evergreen_cs_packet_next_reloc(p, &reloc); if (r) { DRM_ERROR("bad WAIT_REG_MEM\n"); return -EINVAL; } - ib[idx+1] += (u32)(reloc->lobj.gpu_offset & 0xffffffff); - ib[idx+2] += upper_32_bits(reloc->lobj.gpu_offset) & 0xff; + + offset = reloc->lobj.gpu_offset + + (radeon_get_ib_value(p, idx+1) & 0xfffffffc) + + ((u64)(radeon_get_ib_value(p, idx+2) & 0xff) << 32); + + ib[idx+1] = (ib[idx+1] & 0x3) | (offset & 0xfffffffc); + ib[idx+2] = upper_32_bits(offset) & 0xff; } break; case PACKET3_SURFACE_SYNC: @@ -1262,16 +2079,25 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p, return -EINVAL; } if (pkt->count) { + uint64_t offset; + r = evergreen_cs_packet_next_reloc(p, &reloc); if (r) { DRM_ERROR("bad EVENT_WRITE\n"); return -EINVAL; } - ib[idx+1] += (u32)(reloc->lobj.gpu_offset & 0xffffffff); - ib[idx+2] += upper_32_bits(reloc->lobj.gpu_offset) & 0xff; + offset = reloc->lobj.gpu_offset + + (radeon_get_ib_value(p, idx+1) & 0xfffffff8) + + ((u64)(radeon_get_ib_value(p, idx+2) & 0xff) << 32); + + ib[idx+1] = offset & 0xfffffff8; + ib[idx+2] = upper_32_bits(offset) & 0xff; } break; case PACKET3_EVENT_WRITE_EOP: + { + uint64_t offset; + if (pkt->count != 4) { DRM_ERROR("bad EVENT_WRITE_EOP\n"); return -EINVAL; @@ -1281,10 +2107,19 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p, DRM_ERROR("bad EVENT_WRITE_EOP\n"); return -EINVAL; } - ib[idx+1] += (u32)(reloc->lobj.gpu_offset & 0xffffffff); - ib[idx+2] += upper_32_bits(reloc->lobj.gpu_offset) & 0xff; + + offset = reloc->lobj.gpu_offset + + (radeon_get_ib_value(p, idx+1) & 0xfffffffc) + + ((u64)(radeon_get_ib_value(p, idx+2) & 0xff) << 32); + + ib[idx+1] = offset & 0xfffffffc; + ib[idx+2] = (ib[idx+2] & 0xffffff00) | (upper_32_bits(offset) & 0xff); break; + } case PACKET3_EVENT_WRITE_EOS: + { + uint64_t offset; + if (pkt->count != 3) { DRM_ERROR("bad EVENT_WRITE_EOS\n"); return -EINVAL; @@ -1294,9 +2129,15 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p, DRM_ERROR("bad EVENT_WRITE_EOS\n"); return -EINVAL; } - ib[idx+1] += (u32)(reloc->lobj.gpu_offset & 0xffffffff); - ib[idx+2] += upper_32_bits(reloc->lobj.gpu_offset) & 0xff; + + offset = reloc->lobj.gpu_offset + + (radeon_get_ib_value(p, idx+1) & 0xfffffffc) + + ((u64)(radeon_get_ib_value(p, idx+2) & 0xff) << 32); + + ib[idx+1] = offset & 0xfffffffc; + ib[idx+2] = (ib[idx+2] & 0xffffff00) | (upper_32_bits(offset) & 0xff); break; + } case PACKET3_SET_CONFIG_REG: start_reg = (idx_value << 2) + PACKET3_SET_CONFIG_REG_START; end_reg = 4 * pkt->count + start_reg - 4; @@ -1344,6 +2185,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p, } for (i = 0; i < (pkt->count / 8); i++) { struct radeon_bo *texture, *mipmap; + u32 toffset, moffset; u32 size, offset; switch (G__SQ_CONSTANT_TYPE(radeon_get_ib_value(p, idx+1+(i*8)+7))) { @@ -1354,32 +2196,42 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p, DRM_ERROR("bad SET_RESOURCE (tex)\n"); return -EINVAL; } - ib[idx+1+(i*8)+2] += (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff); if (!(p->cs_flags & RADEON_CS_KEEP_TILING_FLAGS)) { ib[idx+1+(i*8)+1] |= TEX_ARRAY_MODE(evergreen_cs_get_aray_mode(reloc->lobj.tiling_flags)); if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO) { - ib[idx+1+(i*8)+6] |= - TEX_TILE_SPLIT(evergreen_cs_get_tile_split(track->row_size)); + unsigned bankw, bankh, mtaspect, tile_split; + + evergreen_tiling_fields(reloc->lobj.tiling_flags, + &bankw, &bankh, &mtaspect, + &tile_split); + ib[idx+1+(i*8)+6] |= TEX_TILE_SPLIT(tile_split); ib[idx+1+(i*8)+7] |= + TEX_BANK_WIDTH(bankw) | + TEX_BANK_HEIGHT(bankh) | + MACRO_TILE_ASPECT(mtaspect) | TEX_NUM_BANKS(evergreen_cs_get_num_banks(track->nbanks)); } } texture = reloc->robj; + toffset = (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff); /* tex mip base */ r = evergreen_cs_packet_next_reloc(p, &reloc); if (r) { DRM_ERROR("bad SET_RESOURCE (tex)\n"); return -EINVAL; } - ib[idx+1+(i*8)+3] += (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff); + moffset = (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff); mipmap = reloc->robj; - r = evergreen_check_texture_resource(p, idx+1+(i*8), - texture, mipmap); + r = evergreen_cs_track_validate_texture(p, texture, mipmap, idx+1+(i*8)); if (r) return r; + ib[idx+1+(i*8)+2] += toffset; + ib[idx+1+(i*8)+3] += moffset; break; case SQ_TEX_VTX_VALID_BUFFER: + { + uint64_t offset64; /* vtx base */ r = evergreen_cs_packet_next_reloc(p, &reloc); if (r) { @@ -1391,11 +2243,15 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p, if (p->rdev && (size + offset) > radeon_bo_size(reloc->robj)) { /* force size to size of the buffer */ dev_warn(p->dev, "vbo resource seems too big for the bo\n"); - ib[idx+1+(i*8)+1] = radeon_bo_size(reloc->robj); + ib[idx+1+(i*8)+1] = radeon_bo_size(reloc->robj) - offset; } - ib[idx+1+(i*8)+0] += (u32)((reloc->lobj.gpu_offset) & 0xffffffff); - ib[idx+1+(i*8)+2] += upper_32_bits(reloc->lobj.gpu_offset) & 0xff; + + offset64 = reloc->lobj.gpu_offset + offset; + ib[idx+1+(i*8)+0] = offset64; + ib[idx+1+(i*8)+2] = (ib[idx+1+(i*8)+2] & 0xffffff00) | + (upper_32_bits(offset64) & 0xff); break; + } case SQ_TEX_VTX_INVALID_TEXTURE: case SQ_TEX_VTX_INVALID_BUFFER: default: @@ -1451,6 +2307,104 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p, return -EINVAL; } break; + case PACKET3_STRMOUT_BUFFER_UPDATE: + if (pkt->count != 4) { + DRM_ERROR("bad STRMOUT_BUFFER_UPDATE (invalid count)\n"); + return -EINVAL; + } + /* Updating memory at DST_ADDRESS. */ + if (idx_value & 0x1) { + u64 offset; + r = evergreen_cs_packet_next_reloc(p, &reloc); + if (r) { + DRM_ERROR("bad STRMOUT_BUFFER_UPDATE (missing dst reloc)\n"); + return -EINVAL; + } + offset = radeon_get_ib_value(p, idx+1); + offset += ((u64)(radeon_get_ib_value(p, idx+2) & 0xff)) << 32; + if ((offset + 4) > radeon_bo_size(reloc->robj)) { + DRM_ERROR("bad STRMOUT_BUFFER_UPDATE dst bo too small: 0x%llx, 0x%lx\n", + offset + 4, radeon_bo_size(reloc->robj)); + return -EINVAL; + } + offset += reloc->lobj.gpu_offset; + ib[idx+1] = offset; + ib[idx+2] = upper_32_bits(offset) & 0xff; + } + /* Reading data from SRC_ADDRESS. */ + if (((idx_value >> 1) & 0x3) == 2) { + u64 offset; + r = evergreen_cs_packet_next_reloc(p, &reloc); + if (r) { + DRM_ERROR("bad STRMOUT_BUFFER_UPDATE (missing src reloc)\n"); + return -EINVAL; + } + offset = radeon_get_ib_value(p, idx+3); + offset += ((u64)(radeon_get_ib_value(p, idx+4) & 0xff)) << 32; + if ((offset + 4) > radeon_bo_size(reloc->robj)) { + DRM_ERROR("bad STRMOUT_BUFFER_UPDATE src bo too small: 0x%llx, 0x%lx\n", + offset + 4, radeon_bo_size(reloc->robj)); + return -EINVAL; + } + offset += reloc->lobj.gpu_offset; + ib[idx+3] = offset; + ib[idx+4] = upper_32_bits(offset) & 0xff; + } + break; + case PACKET3_COPY_DW: + if (pkt->count != 4) { + DRM_ERROR("bad COPY_DW (invalid count)\n"); + return -EINVAL; + } + if (idx_value & 0x1) { + u64 offset; + /* SRC is memory. */ + r = evergreen_cs_packet_next_reloc(p, &reloc); + if (r) { + DRM_ERROR("bad COPY_DW (missing src reloc)\n"); + return -EINVAL; + } + offset = radeon_get_ib_value(p, idx+1); + offset += ((u64)(radeon_get_ib_value(p, idx+2) & 0xff)) << 32; + if ((offset + 4) > radeon_bo_size(reloc->robj)) { + DRM_ERROR("bad COPY_DW src bo too small: 0x%llx, 0x%lx\n", + offset + 4, radeon_bo_size(reloc->robj)); + return -EINVAL; + } + offset += reloc->lobj.gpu_offset; + ib[idx+1] = offset; + ib[idx+2] = upper_32_bits(offset) & 0xff; + } else { + /* SRC is a reg. */ + reg = radeon_get_ib_value(p, idx+1) << 2; + if (!evergreen_is_safe_reg(p, reg, idx+1)) + return -EINVAL; + } + if (idx_value & 0x2) { + u64 offset; + /* DST is memory. */ + r = evergreen_cs_packet_next_reloc(p, &reloc); + if (r) { + DRM_ERROR("bad COPY_DW (missing dst reloc)\n"); + return -EINVAL; + } + offset = radeon_get_ib_value(p, idx+3); + offset += ((u64)(radeon_get_ib_value(p, idx+4) & 0xff)) << 32; + if ((offset + 4) > radeon_bo_size(reloc->robj)) { + DRM_ERROR("bad COPY_DW dst bo too small: 0x%llx, 0x%lx\n", + offset + 4, radeon_bo_size(reloc->robj)); + return -EINVAL; + } + offset += reloc->lobj.gpu_offset; + ib[idx+3] = offset; + ib[idx+4] = upper_32_bits(offset) & 0xff; + } else { + /* DST is a reg. */ + reg = radeon_get_ib_value(p, idx+3) << 2; + if (!evergreen_is_safe_reg(p, reg, idx+3)) + return -EINVAL; + } + break; case PACKET3_NOP: break; default: diff --git a/drivers/gpu/drm/radeon/evergreen_reg.h b/drivers/gpu/drm/radeon/evergreen_reg.h index 4215de9..96c10b3 100644 --- a/drivers/gpu/drm/radeon/evergreen_reg.h +++ b/drivers/gpu/drm/radeon/evergreen_reg.h @@ -219,6 +219,7 @@ # define EVERGREEN_CRTC_MASTER_EN (1 << 0) # define EVERGREEN_CRTC_DISP_READ_REQUEST_DISABLE (1 << 24) #define EVERGREEN_CRTC_STATUS 0x6e8c +# define EVERGREEN_CRTC_V_BLANK (1 << 0) #define EVERGREEN_CRTC_STATUS_POSITION 0x6e90 #define EVERGREEN_MASTER_UPDATE_MODE 0x6ef8 #define EVERGREEN_CRTC_UPDATE_LOCK 0x6ed4 diff --git a/drivers/gpu/drm/radeon/evergreend.h b/drivers/gpu/drm/radeon/evergreend.h index 74713d4..eb5708c 100644 --- a/drivers/gpu/drm/radeon/evergreend.h +++ b/drivers/gpu/drm/radeon/evergreend.h @@ -77,6 +77,7 @@ #define CONFIG_MEMSIZE 0x5428 +#define CP_COHER_BASE 0x85F8 #define CP_ME_CNTL 0x86D8 #define CP_ME_HALT (1 << 28) #define CP_PFP_HALT (1 << 26) @@ -925,7 +926,70 @@ #define DB_DEBUG4 0x983C #define DB_WATERMARKS 0x9854 #define DB_DEPTH_CONTROL 0x28800 +#define R_028800_DB_DEPTH_CONTROL 0x028800 +#define S_028800_STENCIL_ENABLE(x) (((x) & 0x1) << 0) +#define G_028800_STENCIL_ENABLE(x) (((x) >> 0) & 0x1) +#define C_028800_STENCIL_ENABLE 0xFFFFFFFE +#define S_028800_Z_ENABLE(x) (((x) & 0x1) << 1) +#define G_028800_Z_ENABLE(x) (((x) >> 1) & 0x1) +#define C_028800_Z_ENABLE 0xFFFFFFFD +#define S_028800_Z_WRITE_ENABLE(x) (((x) & 0x1) << 2) +#define G_028800_Z_WRITE_ENABLE(x) (((x) >> 2) & 0x1) +#define C_028800_Z_WRITE_ENABLE 0xFFFFFFFB +#define S_028800_ZFUNC(x) (((x) & 0x7) << 4) +#define G_028800_ZFUNC(x) (((x) >> 4) & 0x7) +#define C_028800_ZFUNC 0xFFFFFF8F +#define S_028800_BACKFACE_ENABLE(x) (((x) & 0x1) << 7) +#define G_028800_BACKFACE_ENABLE(x) (((x) >> 7) & 0x1) +#define C_028800_BACKFACE_ENABLE 0xFFFFFF7F +#define S_028800_STENCILFUNC(x) (((x) & 0x7) << 8) +#define G_028800_STENCILFUNC(x) (((x) >> 8) & 0x7) +#define C_028800_STENCILFUNC 0xFFFFF8FF +#define V_028800_STENCILFUNC_NEVER 0x00000000 +#define V_028800_STENCILFUNC_LESS 0x00000001 +#define V_028800_STENCILFUNC_EQUAL 0x00000002 +#define V_028800_STENCILFUNC_LEQUAL 0x00000003 +#define V_028800_STENCILFUNC_GREATER 0x00000004 +#define V_028800_STENCILFUNC_NOTEQUAL 0x00000005 +#define V_028800_STENCILFUNC_GEQUAL 0x00000006 +#define V_028800_STENCILFUNC_ALWAYS 0x00000007 +#define S_028800_STENCILFAIL(x) (((x) & 0x7) << 11) +#define G_028800_STENCILFAIL(x) (((x) >> 11) & 0x7) +#define C_028800_STENCILFAIL 0xFFFFC7FF +#define V_028800_STENCIL_KEEP 0x00000000 +#define V_028800_STENCIL_ZERO 0x00000001 +#define V_028800_STENCIL_REPLACE 0x00000002 +#define V_028800_STENCIL_INCR 0x00000003 +#define V_028800_STENCIL_DECR 0x00000004 +#define V_028800_STENCIL_INVERT 0x00000005 +#define V_028800_STENCIL_INCR_WRAP 0x00000006 +#define V_028800_STENCIL_DECR_WRAP 0x00000007 +#define S_028800_STENCILZPASS(x) (((x) & 0x7) << 14) +#define G_028800_STENCILZPASS(x) (((x) >> 14) & 0x7) +#define C_028800_STENCILZPASS 0xFFFE3FFF +#define S_028800_STENCILZFAIL(x) (((x) & 0x7) << 17) +#define G_028800_STENCILZFAIL(x) (((x) >> 17) & 0x7) +#define C_028800_STENCILZFAIL 0xFFF1FFFF +#define S_028800_STENCILFUNC_BF(x) (((x) & 0x7) << 20) +#define G_028800_STENCILFUNC_BF(x) (((x) >> 20) & 0x7) +#define C_028800_STENCILFUNC_BF 0xFF8FFFFF +#define S_028800_STENCILFAIL_BF(x) (((x) & 0x7) << 23) +#define G_028800_STENCILFAIL_BF(x) (((x) >> 23) & 0x7) +#define C_028800_STENCILFAIL_BF 0xFC7FFFFF +#define S_028800_STENCILZPASS_BF(x) (((x) & 0x7) << 26) +#define G_028800_STENCILZPASS_BF(x) (((x) >> 26) & 0x7) +#define C_028800_STENCILZPASS_BF 0xE3FFFFFF +#define S_028800_STENCILZFAIL_BF(x) (((x) & 0x7) << 29) +#define G_028800_STENCILZFAIL_BF(x) (((x) >> 29) & 0x7) +#define C_028800_STENCILZFAIL_BF 0x1FFFFFFF #define DB_DEPTH_VIEW 0x28008 +#define R_028008_DB_DEPTH_VIEW 0x00028008 +#define S_028008_SLICE_START(x) (((x) & 0x7FF) << 0) +#define G_028008_SLICE_START(x) (((x) >> 0) & 0x7FF) +#define C_028008_SLICE_START 0xFFFFF800 +#define S_028008_SLICE_MAX(x) (((x) & 0x7FF) << 13) +#define G_028008_SLICE_MAX(x) (((x) >> 13) & 0x7FF) +#define C_028008_SLICE_MAX 0xFF001FFF #define DB_HTILE_DATA_BASE 0x28014 #define DB_Z_INFO 0x28040 # define Z_ARRAY_MODE(x) ((x) << 4) @@ -933,12 +997,59 @@ # define DB_NUM_BANKS(x) (((x) & 0x3) << 12) # define DB_BANK_WIDTH(x) (((x) & 0x3) << 16) # define DB_BANK_HEIGHT(x) (((x) & 0x3) << 20) +# define DB_MACRO_TILE_ASPECT(x) (((x) & 0x3) << 24) +#define R_028040_DB_Z_INFO 0x028040 +#define S_028040_FORMAT(x) (((x) & 0x3) << 0) +#define G_028040_FORMAT(x) (((x) >> 0) & 0x3) +#define C_028040_FORMAT 0xFFFFFFFC +#define V_028040_Z_INVALID 0x00000000 +#define V_028040_Z_16 0x00000001 +#define V_028040_Z_24 0x00000002 +#define V_028040_Z_32_FLOAT 0x00000003 +#define S_028040_ARRAY_MODE(x) (((x) & 0xF) << 4) +#define G_028040_ARRAY_MODE(x) (((x) >> 4) & 0xF) +#define C_028040_ARRAY_MODE 0xFFFFFF0F +#define S_028040_READ_SIZE(x) (((x) & 0x1) << 28) +#define G_028040_READ_SIZE(x) (((x) >> 28) & 0x1) +#define C_028040_READ_SIZE 0xEFFFFFFF +#define S_028040_TILE_SURFACE_ENABLE(x) (((x) & 0x1) << 29) +#define G_028040_TILE_SURFACE_ENABLE(x) (((x) >> 29) & 0x1) +#define C_028040_TILE_SURFACE_ENABLE 0xDFFFFFFF +#define S_028040_ZRANGE_PRECISION(x) (((x) & 0x1) << 31) +#define G_028040_ZRANGE_PRECISION(x) (((x) >> 31) & 0x1) +#define C_028040_ZRANGE_PRECISION 0x7FFFFFFF +#define S_028040_TILE_SPLIT(x) (((x) & 0x7) << 8) +#define G_028040_TILE_SPLIT(x) (((x) >> 8) & 0x7) +#define S_028040_NUM_BANKS(x) (((x) & 0x3) << 12) +#define G_028040_NUM_BANKS(x) (((x) >> 12) & 0x3) +#define S_028040_BANK_WIDTH(x) (((x) & 0x3) << 16) +#define G_028040_BANK_WIDTH(x) (((x) >> 16) & 0x3) +#define S_028040_BANK_HEIGHT(x) (((x) & 0x3) << 20) +#define G_028040_BANK_HEIGHT(x) (((x) >> 20) & 0x3) +#define S_028040_MACRO_TILE_ASPECT(x) (((x) & 0x3) << 24) +#define G_028040_MACRO_TILE_ASPECT(x) (((x) >> 24) & 0x3) #define DB_STENCIL_INFO 0x28044 +#define R_028044_DB_STENCIL_INFO 0x028044 +#define S_028044_FORMAT(x) (((x) & 0x1) << 0) +#define G_028044_FORMAT(x) (((x) >> 0) & 0x1) +#define C_028044_FORMAT 0xFFFFFFFE +#define G_028044_TILE_SPLIT(x) (((x) >> 8) & 0x7) #define DB_Z_READ_BASE 0x28048 #define DB_STENCIL_READ_BASE 0x2804c #define DB_Z_WRITE_BASE 0x28050 #define DB_STENCIL_WRITE_BASE 0x28054 #define DB_DEPTH_SIZE 0x28058 +#define R_028058_DB_DEPTH_SIZE 0x028058 +#define S_028058_PITCH_TILE_MAX(x) (((x) & 0x7FF) << 0) +#define G_028058_PITCH_TILE_MAX(x) (((x) >> 0) & 0x7FF) +#define C_028058_PITCH_TILE_MAX 0xFFFFF800 +#define S_028058_HEIGHT_TILE_MAX(x) (((x) & 0x7FF) << 11) +#define G_028058_HEIGHT_TILE_MAX(x) (((x) >> 11) & 0x7FF) +#define C_028058_HEIGHT_TILE_MAX 0xFFC007FF +#define R_02805C_DB_DEPTH_SLICE 0x02805C +#define S_02805C_SLICE_TILE_MAX(x) (((x) & 0x3FFFFF) << 0) +#define G_02805C_SLICE_TILE_MAX(x) (((x) >> 0) & 0x3FFFFF) +#define C_02805C_SLICE_TILE_MAX 0xFFC00000 #define SQ_PGM_START_PS 0x28840 #define SQ_PGM_START_VS 0x2885c @@ -948,6 +1059,14 @@ #define SQ_PGM_START_HS 0x288b8 #define SQ_PGM_START_LS 0x288d0 +#define VGT_STRMOUT_BUFFER_BASE_0 0x28AD8 +#define VGT_STRMOUT_BUFFER_BASE_1 0x28AE8 +#define VGT_STRMOUT_BUFFER_BASE_2 0x28AF8 +#define VGT_STRMOUT_BUFFER_BASE_3 0x28B08 +#define VGT_STRMOUT_BUFFER_SIZE_0 0x28AD0 +#define VGT_STRMOUT_BUFFER_SIZE_1 0x28AE0 +#define VGT_STRMOUT_BUFFER_SIZE_2 0x28AF0 +#define VGT_STRMOUT_BUFFER_SIZE_3 0x28B00 #define VGT_STRMOUT_CONFIG 0x28b94 #define VGT_STRMOUT_BUFFER_CONFIG 0x28b98 @@ -974,6 +1093,114 @@ #define CB_COLOR0_PITCH 0x28c64 #define CB_COLOR0_SLICE 0x28c68 #define CB_COLOR0_VIEW 0x28c6c +#define R_028C6C_CB_COLOR0_VIEW 0x00028C6C +#define S_028C6C_SLICE_START(x) (((x) & 0x7FF) << 0) +#define G_028C6C_SLICE_START(x) (((x) >> 0) & 0x7FF) +#define C_028C6C_SLICE_START 0xFFFFF800 +#define S_028C6C_SLICE_MAX(x) (((x) & 0x7FF) << 13) +#define G_028C6C_SLICE_MAX(x) (((x) >> 13) & 0x7FF) +#define C_028C6C_SLICE_MAX 0xFF001FFF +#define R_028C70_CB_COLOR0_INFO 0x028C70 +#define S_028C70_ENDIAN(x) (((x) & 0x3) << 0) +#define G_028C70_ENDIAN(x) (((x) >> 0) & 0x3) +#define C_028C70_ENDIAN 0xFFFFFFFC +#define S_028C70_FORMAT(x) (((x) & 0x3F) << 2) +#define G_028C70_FORMAT(x) (((x) >> 2) & 0x3F) +#define C_028C70_FORMAT 0xFFFFFF03 +#define V_028C70_COLOR_INVALID 0x00000000 +#define V_028C70_COLOR_8 0x00000001 +#define V_028C70_COLOR_4_4 0x00000002 +#define V_028C70_COLOR_3_3_2 0x00000003 +#define V_028C70_COLOR_16 0x00000005 +#define V_028C70_COLOR_16_FLOAT 0x00000006 +#define V_028C70_COLOR_8_8 0x00000007 +#define V_028C70_COLOR_5_6_5 0x00000008 +#define V_028C70_COLOR_6_5_5 0x00000009 +#define V_028C70_COLOR_1_5_5_5 0x0000000A +#define V_028C70_COLOR_4_4_4_4 0x0000000B +#define V_028C70_COLOR_5_5_5_1 0x0000000C +#define V_028C70_COLOR_32 0x0000000D +#define V_028C70_COLOR_32_FLOAT 0x0000000E +#define V_028C70_COLOR_16_16 0x0000000F +#define V_028C70_COLOR_16_16_FLOAT 0x00000010 +#define V_028C70_COLOR_8_24 0x00000011 +#define V_028C70_COLOR_8_24_FLOAT 0x00000012 +#define V_028C70_COLOR_24_8 0x00000013 +#define V_028C70_COLOR_24_8_FLOAT 0x00000014 +#define V_028C70_COLOR_10_11_11 0x00000015 +#define V_028C70_COLOR_10_11_11_FLOAT 0x00000016 +#define V_028C70_COLOR_11_11_10 0x00000017 +#define V_028C70_COLOR_11_11_10_FLOAT 0x00000018 +#define V_028C70_COLOR_2_10_10_10 0x00000019 +#define V_028C70_COLOR_8_8_8_8 0x0000001A +#define V_028C70_COLOR_10_10_10_2 0x0000001B +#define V_028C70_COLOR_X24_8_32_FLOAT 0x0000001C +#define V_028C70_COLOR_32_32 0x0000001D +#define V_028C70_COLOR_32_32_FLOAT 0x0000001E +#define V_028C70_COLOR_16_16_16_16 0x0000001F +#define V_028C70_COLOR_16_16_16_16_FLOAT 0x00000020 +#define V_028C70_COLOR_32_32_32_32 0x00000022 +#define V_028C70_COLOR_32_32_32_32_FLOAT 0x00000023 +#define V_028C70_COLOR_32_32_32_FLOAT 0x00000030 +#define S_028C70_ARRAY_MODE(x) (((x) & 0xF) << 8) +#define G_028C70_ARRAY_MODE(x) (((x) >> 8) & 0xF) +#define C_028C70_ARRAY_MODE 0xFFFFF0FF +#define V_028C70_ARRAY_LINEAR_GENERAL 0x00000000 +#define V_028C70_ARRAY_LINEAR_ALIGNED 0x00000001 +#define V_028C70_ARRAY_1D_TILED_THIN1 0x00000002 +#define V_028C70_ARRAY_2D_TILED_THIN1 0x00000004 +#define S_028C70_NUMBER_TYPE(x) (((x) & 0x7) << 12) +#define G_028C70_NUMBER_TYPE(x) (((x) >> 12) & 0x7) +#define C_028C70_NUMBER_TYPE 0xFFFF8FFF +#define V_028C70_NUMBER_UNORM 0x00000000 +#define V_028C70_NUMBER_SNORM 0x00000001 +#define V_028C70_NUMBER_USCALED 0x00000002 +#define V_028C70_NUMBER_SSCALED 0x00000003 +#define V_028C70_NUMBER_UINT 0x00000004 +#define V_028C70_NUMBER_SINT 0x00000005 +#define V_028C70_NUMBER_SRGB 0x00000006 +#define V_028C70_NUMBER_FLOAT 0x00000007 +#define S_028C70_COMP_SWAP(x) (((x) & 0x3) << 15) +#define G_028C70_COMP_SWAP(x) (((x) >> 15) & 0x3) +#define C_028C70_COMP_SWAP 0xFFFE7FFF +#define V_028C70_SWAP_STD 0x00000000 +#define V_028C70_SWAP_ALT 0x00000001 +#define V_028C70_SWAP_STD_REV 0x00000002 +#define V_028C70_SWAP_ALT_REV 0x00000003 +#define S_028C70_FAST_CLEAR(x) (((x) & 0x1) << 17) +#define G_028C70_FAST_CLEAR(x) (((x) >> 17) & 0x1) +#define C_028C70_FAST_CLEAR 0xFFFDFFFF +#define S_028C70_COMPRESSION(x) (((x) & 0x3) << 18) +#define G_028C70_COMPRESSION(x) (((x) >> 18) & 0x3) +#define C_028C70_COMPRESSION 0xFFF3FFFF +#define S_028C70_BLEND_CLAMP(x) (((x) & 0x1) << 19) +#define G_028C70_BLEND_CLAMP(x) (((x) >> 19) & 0x1) +#define C_028C70_BLEND_CLAMP 0xFFF7FFFF +#define S_028C70_BLEND_BYPASS(x) (((x) & 0x1) << 20) +#define G_028C70_BLEND_BYPASS(x) (((x) >> 20) & 0x1) +#define C_028C70_BLEND_BYPASS 0xFFEFFFFF +#define S_028C70_SIMPLE_FLOAT(x) (((x) & 0x1) << 21) +#define G_028C70_SIMPLE_FLOAT(x) (((x) >> 21) & 0x1) +#define C_028C70_SIMPLE_FLOAT 0xFFDFFFFF +#define S_028C70_ROUND_MODE(x) (((x) & 0x1) << 22) +#define G_028C70_ROUND_MODE(x) (((x) >> 22) & 0x1) +#define C_028C70_ROUND_MODE 0xFFBFFFFF +#define S_028C70_TILE_COMPACT(x) (((x) & 0x1) << 23) +#define G_028C70_TILE_COMPACT(x) (((x) >> 23) & 0x1) +#define C_028C70_TILE_COMPACT 0xFF7FFFFF +#define S_028C70_SOURCE_FORMAT(x) (((x) & 0x3) << 24) +#define G_028C70_SOURCE_FORMAT(x) (((x) >> 24) & 0x3) +#define C_028C70_SOURCE_FORMAT 0xFCFFFFFF +#define V_028C70_EXPORT_4C_32BPC 0x0 +#define V_028C70_EXPORT_4C_16BPC 0x1 +#define V_028C70_EXPORT_2C_32BPC 0x2 /* Do not use */ +#define S_028C70_RAT(x) (((x) & 0x1) << 26) +#define G_028C70_RAT(x) (((x) >> 26) & 0x1) +#define C_028C70_RAT 0xFBFFFFFF +#define S_028C70_RESOURCE_TYPE(x) (((x) & 0x7) << 27) +#define G_028C70_RESOURCE_TYPE(x) (((x) >> 27) & 0x7) +#define C_028C70_RESOURCE_TYPE 0xC7FFFFFF + #define CB_COLOR0_INFO 0x28c70 # define CB_FORMAT(x) ((x) << 2) # define CB_ARRAY_MODE(x) ((x) << 8) @@ -984,6 +1211,20 @@ # define CB_SOURCE_FORMAT(x) ((x) << 24) # define CB_SF_EXPORT_FULL 0 # define CB_SF_EXPORT_NORM 1 +#define R_028C74_CB_COLOR0_ATTRIB 0x028C74 +#define S_028C74_NON_DISP_TILING_ORDER(x) (((x) & 0x1) << 4) +#define G_028C74_NON_DISP_TILING_ORDER(x) (((x) >> 4) & 0x1) +#define C_028C74_NON_DISP_TILING_ORDER 0xFFFFFFEF +#define S_028C74_TILE_SPLIT(x) (((x) & 0xf) << 5) +#define G_028C74_TILE_SPLIT(x) (((x) >> 5) & 0xf) +#define S_028C74_NUM_BANKS(x) (((x) & 0x3) << 10) +#define G_028C74_NUM_BANKS(x) (((x) >> 10) & 0x3) +#define S_028C74_BANK_WIDTH(x) (((x) & 0x3) << 13) +#define G_028C74_BANK_WIDTH(x) (((x) >> 13) & 0x3) +#define S_028C74_BANK_HEIGHT(x) (((x) & 0x3) << 16) +#define G_028C74_BANK_HEIGHT(x) (((x) >> 16) & 0x3) +#define S_028C74_MACRO_TILE_ASPECT(x) (((x) & 0x3) << 19) +#define G_028C74_MACRO_TILE_ASPECT(x) (((x) >> 19) & 0x3) #define CB_COLOR0_ATTRIB 0x28c74 # define CB_TILE_SPLIT(x) (((x) & 0x7) << 5) # define ADDR_SURF_TILE_SPLIT_64B 0 @@ -1008,6 +1249,7 @@ # define ADDR_SURF_BANK_HEIGHT_2 1 # define ADDR_SURF_BANK_HEIGHT_4 2 # define ADDR_SURF_BANK_HEIGHT_8 3 +# define CB_MACRO_TILE_ASPECT(x) (((x) & 0x3) << 19) #define CB_COLOR0_DIM 0x28c78 /* only CB0-7 blocks have these regs */ #define CB_COLOR0_CMASK 0x28c7c @@ -1196,9 +1438,144 @@ #define SQ_TEX_RESOURCE_WORD6_0 0x30018 # define TEX_TILE_SPLIT(x) (((x) & 0x7) << 29) #define SQ_TEX_RESOURCE_WORD7_0 0x3001c +# define MACRO_TILE_ASPECT(x) (((x) & 0x3) << 6) # define TEX_BANK_WIDTH(x) (((x) & 0x3) << 8) # define TEX_BANK_HEIGHT(x) (((x) & 0x3) << 10) # define TEX_NUM_BANKS(x) (((x) & 0x3) << 16) +#define R_030000_SQ_TEX_RESOURCE_WORD0_0 0x030000 +#define S_030000_DIM(x) (((x) & 0x7) << 0) +#define G_030000_DIM(x) (((x) >> 0) & 0x7) +#define C_030000_DIM 0xFFFFFFF8 +#define V_030000_SQ_TEX_DIM_1D 0x00000000 +#define V_030000_SQ_TEX_DIM_2D 0x00000001 +#define V_030000_SQ_TEX_DIM_3D 0x00000002 +#define V_030000_SQ_TEX_DIM_CUBEMAP 0x00000003 +#define V_030000_SQ_TEX_DIM_1D_ARRAY 0x00000004 +#define V_030000_SQ_TEX_DIM_2D_ARRAY 0x00000005 +#define V_030000_SQ_TEX_DIM_2D_MSAA 0x00000006 +#define V_030000_SQ_TEX_DIM_2D_ARRAY_MSAA 0x00000007 +#define S_030000_NON_DISP_TILING_ORDER(x) (((x) & 0x1) << 5) +#define G_030000_NON_DISP_TILING_ORDER(x) (((x) >> 5) & 0x1) +#define C_030000_NON_DISP_TILING_ORDER 0xFFFFFFDF +#define S_030000_PITCH(x) (((x) & 0xFFF) << 6) +#define G_030000_PITCH(x) (((x) >> 6) & 0xFFF) +#define C_030000_PITCH 0xFFFC003F +#define S_030000_TEX_WIDTH(x) (((x) & 0x3FFF) << 18) +#define G_030000_TEX_WIDTH(x) (((x) >> 18) & 0x3FFF) +#define C_030000_TEX_WIDTH 0x0003FFFF +#define R_030004_SQ_TEX_RESOURCE_WORD1_0 0x030004 +#define S_030004_TEX_HEIGHT(x) (((x) & 0x3FFF) << 0) +#define G_030004_TEX_HEIGHT(x) (((x) >> 0) & 0x3FFF) +#define C_030004_TEX_HEIGHT 0xFFFFC000 +#define S_030004_TEX_DEPTH(x) (((x) & 0x1FFF) << 14) +#define G_030004_TEX_DEPTH(x) (((x) >> 14) & 0x1FFF) +#define C_030004_TEX_DEPTH 0xF8003FFF +#define S_030004_ARRAY_MODE(x) (((x) & 0xF) << 28) +#define G_030004_ARRAY_MODE(x) (((x) >> 28) & 0xF) +#define C_030004_ARRAY_MODE 0x0FFFFFFF +#define R_030008_SQ_TEX_RESOURCE_WORD2_0 0x030008 +#define S_030008_BASE_ADDRESS(x) (((x) & 0xFFFFFFFF) << 0) +#define G_030008_BASE_ADDRESS(x) (((x) >> 0) & 0xFFFFFFFF) +#define C_030008_BASE_ADDRESS 0x00000000 +#define R_03000C_SQ_TEX_RESOURCE_WORD3_0 0x03000C +#define S_03000C_MIP_ADDRESS(x) (((x) & 0xFFFFFFFF) << 0) +#define G_03000C_MIP_ADDRESS(x) (((x) >> 0) & 0xFFFFFFFF) +#define C_03000C_MIP_ADDRESS 0x00000000 +#define R_030010_SQ_TEX_RESOURCE_WORD4_0 0x030010 +#define S_030010_FORMAT_COMP_X(x) (((x) & 0x3) << 0) +#define G_030010_FORMAT_COMP_X(x) (((x) >> 0) & 0x3) +#define C_030010_FORMAT_COMP_X 0xFFFFFFFC +#define V_030010_SQ_FORMAT_COMP_UNSIGNED 0x00000000 +#define V_030010_SQ_FORMAT_COMP_SIGNED 0x00000001 +#define V_030010_SQ_FORMAT_COMP_UNSIGNED_BIASED 0x00000002 +#define S_030010_FORMAT_COMP_Y(x) (((x) & 0x3) << 2) +#define G_030010_FORMAT_COMP_Y(x) (((x) >> 2) & 0x3) +#define C_030010_FORMAT_COMP_Y 0xFFFFFFF3 +#define S_030010_FORMAT_COMP_Z(x) (((x) & 0x3) << 4) +#define G_030010_FORMAT_COMP_Z(x) (((x) >> 4) & 0x3) +#define C_030010_FORMAT_COMP_Z 0xFFFFFFCF +#define S_030010_FORMAT_COMP_W(x) (((x) & 0x3) << 6) +#define G_030010_FORMAT_COMP_W(x) (((x) >> 6) & 0x3) +#define C_030010_FORMAT_COMP_W 0xFFFFFF3F +#define S_030010_NUM_FORMAT_ALL(x) (((x) & 0x3) << 8) +#define G_030010_NUM_FORMAT_ALL(x) (((x) >> 8) & 0x3) +#define C_030010_NUM_FORMAT_ALL 0xFFFFFCFF +#define V_030010_SQ_NUM_FORMAT_NORM 0x00000000 +#define V_030010_SQ_NUM_FORMAT_INT 0x00000001 +#define V_030010_SQ_NUM_FORMAT_SCALED 0x00000002 +#define S_030010_SRF_MODE_ALL(x) (((x) & 0x1) << 10) +#define G_030010_SRF_MODE_ALL(x) (((x) >> 10) & 0x1) +#define C_030010_SRF_MODE_ALL 0xFFFFFBFF +#define V_030010_SRF_MODE_ZERO_CLAMP_MINUS_ONE 0x00000000 +#define V_030010_SRF_MODE_NO_ZERO 0x00000001 +#define S_030010_FORCE_DEGAMMA(x) (((x) & 0x1) << 11) +#define G_030010_FORCE_DEGAMMA(x) (((x) >> 11) & 0x1) +#define C_030010_FORCE_DEGAMMA 0xFFFFF7FF +#define S_030010_ENDIAN_SWAP(x) (((x) & 0x3) << 12) +#define G_030010_ENDIAN_SWAP(x) (((x) >> 12) & 0x3) +#define C_030010_ENDIAN_SWAP 0xFFFFCFFF +#define S_030010_DST_SEL_X(x) (((x) & 0x7) << 16) +#define G_030010_DST_SEL_X(x) (((x) >> 16) & 0x7) +#define C_030010_DST_SEL_X 0xFFF8FFFF +#define V_030010_SQ_SEL_X 0x00000000 +#define V_030010_SQ_SEL_Y 0x00000001 +#define V_030010_SQ_SEL_Z 0x00000002 +#define V_030010_SQ_SEL_W 0x00000003 +#define V_030010_SQ_SEL_0 0x00000004 +#define V_030010_SQ_SEL_1 0x00000005 +#define S_030010_DST_SEL_Y(x) (((x) & 0x7) << 19) +#define G_030010_DST_SEL_Y(x) (((x) >> 19) & 0x7) +#define C_030010_DST_SEL_Y 0xFFC7FFFF +#define S_030010_DST_SEL_Z(x) (((x) & 0x7) << 22) +#define G_030010_DST_SEL_Z(x) (((x) >> 22) & 0x7) +#define C_030010_DST_SEL_Z 0xFE3FFFFF +#define S_030010_DST_SEL_W(x) (((x) & 0x7) << 25) +#define G_030010_DST_SEL_W(x) (((x) >> 25) & 0x7) +#define C_030010_DST_SEL_W 0xF1FFFFFF +#define S_030010_BASE_LEVEL(x) (((x) & 0xF) << 28) +#define G_030010_BASE_LEVEL(x) (((x) >> 28) & 0xF) +#define C_030010_BASE_LEVEL 0x0FFFFFFF +#define R_030014_SQ_TEX_RESOURCE_WORD5_0 0x030014 +#define S_030014_LAST_LEVEL(x) (((x) & 0xF) << 0) +#define G_030014_LAST_LEVEL(x) (((x) >> 0) & 0xF) +#define C_030014_LAST_LEVEL 0xFFFFFFF0 +#define S_030014_BASE_ARRAY(x) (((x) & 0x1FFF) << 4) +#define G_030014_BASE_ARRAY(x) (((x) >> 4) & 0x1FFF) +#define C_030014_BASE_ARRAY 0xFFFE000F +#define S_030014_LAST_ARRAY(x) (((x) & 0x1FFF) << 17) +#define G_030014_LAST_ARRAY(x) (((x) >> 17) & 0x1FFF) +#define C_030014_LAST_ARRAY 0xC001FFFF +#define R_030018_SQ_TEX_RESOURCE_WORD6_0 0x030018 +#define S_030018_MAX_ANISO(x) (((x) & 0x7) << 0) +#define G_030018_MAX_ANISO(x) (((x) >> 0) & 0x7) +#define C_030018_MAX_ANISO 0xFFFFFFF8 +#define S_030018_PERF_MODULATION(x) (((x) & 0x7) << 3) +#define G_030018_PERF_MODULATION(x) (((x) >> 3) & 0x7) +#define C_030018_PERF_MODULATION 0xFFFFFFC7 +#define S_030018_INTERLACED(x) (((x) & 0x1) << 6) +#define G_030018_INTERLACED(x) (((x) >> 6) & 0x1) +#define C_030018_INTERLACED 0xFFFFFFBF +#define S_030018_TILE_SPLIT(x) (((x) & 0x7) << 29) +#define G_030018_TILE_SPLIT(x) (((x) >> 29) & 0x7) +#define R_03001C_SQ_TEX_RESOURCE_WORD7_0 0x03001C +#define S_03001C_MACRO_TILE_ASPECT(x) (((x) & 0x3) << 6) +#define G_03001C_MACRO_TILE_ASPECT(x) (((x) >> 6) & 0x3) +#define S_03001C_BANK_WIDTH(x) (((x) & 0x3) << 8) +#define G_03001C_BANK_WIDTH(x) (((x) >> 8) & 0x3) +#define S_03001C_BANK_HEIGHT(x) (((x) & 0x3) << 10) +#define G_03001C_BANK_HEIGHT(x) (((x) >> 10) & 0x3) +#define S_03001C_NUM_BANKS(x) (((x) & 0x3) << 16) +#define G_03001C_NUM_BANKS(x) (((x) >> 16) & 0x3) +#define S_03001C_TYPE(x) (((x) & 0x3) << 30) +#define G_03001C_TYPE(x) (((x) >> 30) & 0x3) +#define C_03001C_TYPE 0x3FFFFFFF +#define V_03001C_SQ_TEX_VTX_INVALID_TEXTURE 0x00000000 +#define V_03001C_SQ_TEX_VTX_INVALID_BUFFER 0x00000001 +#define V_03001C_SQ_TEX_VTX_VALID_TEXTURE 0x00000002 +#define V_03001C_SQ_TEX_VTX_VALID_BUFFER 0x00000003 +#define S_03001C_DATA_FORMAT(x) (((x) & 0x3F) << 0) +#define G_03001C_DATA_FORMAT(x) (((x) >> 0) & 0x3F) +#define C_03001C_DATA_FORMAT 0xFFFFFFC0 #define SQ_VTX_CONSTANT_WORD0_0 0x30000 #define SQ_VTX_CONSTANT_WORD1_0 0x30004 diff --git a/drivers/gpu/drm/radeon/ni.c b/drivers/gpu/drm/radeon/ni.c index 2509c50..160799c 100644 --- a/drivers/gpu/drm/radeon/ni.c +++ b/drivers/gpu/drm/radeon/ni.c @@ -1318,7 +1318,7 @@ int cayman_cp_resume(struct radeon_device *rdev) rdev->ring[CAYMAN_RING_TYPE_CP1_INDEX].ready = false; rdev->ring[CAYMAN_RING_TYPE_CP2_INDEX].ready = false; /* this only test cp0 */ - r = radeon_ring_test(rdev, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]); + r = radeon_ring_test(rdev, RADEON_RING_TYPE_GFX_INDEX, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]); if (r) { rdev->ring[RADEON_RING_TYPE_GFX_INDEX].ready = false; rdev->ring[CAYMAN_RING_TYPE_CP1_INDEX].ready = false; @@ -1466,7 +1466,7 @@ static int cayman_startup(struct radeon_device *rdev) r = evergreen_blit_init(rdev); if (r) { r600_blit_fini(rdev); - rdev->asic->copy = NULL; + rdev->asic->copy.copy = NULL; dev_warn(rdev->dev, "failed blitter (%d) falling back to memcpy\n", r); } @@ -1518,7 +1518,7 @@ static int cayman_startup(struct radeon_device *rdev) if (r) return r; - r = r600_ib_test(rdev, RADEON_RING_TYPE_GFX_INDEX); + r = radeon_ib_test(rdev, RADEON_RING_TYPE_GFX_INDEX, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]); if (r) { DRM_ERROR("radeon: failed testing IB (%d).\n", r); rdev->accel_working = false; diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c index 333cde9..81801c1 100644 --- a/drivers/gpu/drm/radeon/r100.c +++ b/drivers/gpu/drm/radeon/r100.c @@ -65,6 +65,40 @@ MODULE_FIRMWARE(FIRMWARE_R520); #include "r100_track.h" +void r100_wait_for_vblank(struct radeon_device *rdev, int crtc) +{ + struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc]; + int i; + + if (radeon_crtc->crtc_id == 0) { + if (RREG32(RADEON_CRTC_GEN_CNTL) & RADEON_CRTC_EN) { + for (i = 0; i < rdev->usec_timeout; i++) { + if (!(RREG32(RADEON_CRTC_STATUS) & RADEON_CRTC_VBLANK_CUR)) + break; + udelay(1); + } + for (i = 0; i < rdev->usec_timeout; i++) { + if (RREG32(RADEON_CRTC_STATUS) & RADEON_CRTC_VBLANK_CUR) + break; + udelay(1); + } + } + } else { + if (RREG32(RADEON_CRTC2_GEN_CNTL) & RADEON_CRTC2_EN) { + for (i = 0; i < rdev->usec_timeout; i++) { + if (!(RREG32(RADEON_CRTC2_STATUS) & RADEON_CRTC2_VBLANK_CUR)) + break; + udelay(1); + } + for (i = 0; i < rdev->usec_timeout; i++) { + if (RREG32(RADEON_CRTC2_STATUS) & RADEON_CRTC2_VBLANK_CUR) + break; + udelay(1); + } + } + } +} + /* This files gather functions specifics to: * r100,rv100,rs100,rv200,rs200,r200,rv250,rs300,rv280 */ @@ -87,23 +121,27 @@ int r100_reloc_pitch_offset(struct radeon_cs_parser *p, r100_cs_dump_packet(p, pkt); return r; } + value = radeon_get_ib_value(p, idx); tmp = value & 0x003fffff; tmp += (((u32)reloc->lobj.gpu_offset) >> 10); - if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO) - tile_flags |= RADEON_DST_TILE_MACRO; - if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO) { - if (reg == RADEON_SRC_PITCH_OFFSET) { - DRM_ERROR("Cannot src blit from microtiled surface\n"); - r100_cs_dump_packet(p, pkt); - return -EINVAL; + if (!(p->cs_flags & RADEON_CS_KEEP_TILING_FLAGS)) { + if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO) + tile_flags |= RADEON_DST_TILE_MACRO; + if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO) { + if (reg == RADEON_SRC_PITCH_OFFSET) { + DRM_ERROR("Cannot src blit from microtiled surface\n"); + r100_cs_dump_packet(p, pkt); + return -EINVAL; + } + tile_flags |= RADEON_DST_TILE_MICRO; } - tile_flags |= RADEON_DST_TILE_MICRO; - } - tmp |= tile_flags; - p->ib->ptr[idx] = (value & 0x3fc00000) | tmp; + tmp |= tile_flags; + p->ib->ptr[idx] = (value & 0x3fc00000) | tmp; + } else + p->ib->ptr[idx] = (value & 0xffc00000) | tmp; return 0; } @@ -412,7 +450,7 @@ void r100_pm_misc(struct radeon_device *rdev) /* set pcie lanes */ if ((rdev->flags & RADEON_IS_PCIE) && !(rdev->flags & RADEON_IS_IGP) && - rdev->asic->set_pcie_lanes && + rdev->asic->pm.set_pcie_lanes && (ps->pcie_lanes != rdev->pm.power_state[rdev->pm.current_power_state_index].pcie_lanes)) { radeon_set_pcie_lanes(rdev, @@ -592,8 +630,8 @@ int r100_pci_gart_init(struct radeon_device *rdev) if (r) return r; rdev->gart.table_size = rdev->gart.num_gpu_pages * 4; - rdev->asic->gart_tlb_flush = &r100_pci_gart_tlb_flush; - rdev->asic->gart_set_page = &r100_pci_gart_set_page; + rdev->asic->gart.tlb_flush = &r100_pci_gart_tlb_flush; + rdev->asic->gart.set_page = &r100_pci_gart_set_page; return radeon_gart_table_ram_alloc(rdev); } @@ -930,9 +968,8 @@ static int r100_cp_wait_for_idle(struct radeon_device *rdev) return -1; } -void r100_ring_start(struct radeon_device *rdev) +void r100_ring_start(struct radeon_device *rdev, struct radeon_ring *ring) { - struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]; int r; r = radeon_ring_lock(rdev, ring, 2); @@ -1143,8 +1180,8 @@ int r100_cp_init(struct radeon_device *rdev, unsigned ring_size) WREG32(RADEON_CP_RB_WPTR_DELAY, 0); WREG32(RADEON_CP_CSQ_MODE, 0x00004D4D); WREG32(RADEON_CP_CSQ_CNTL, RADEON_CSQ_PRIBM_INDBM); - radeon_ring_start(rdev); - r = radeon_ring_test(rdev, ring); + radeon_ring_start(rdev, RADEON_RING_TYPE_GFX_INDEX, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]); + r = radeon_ring_test(rdev, RADEON_RING_TYPE_GFX_INDEX, ring); if (r) { DRM_ERROR("radeon: cp isn't working (%d).\n", r); return r; @@ -1552,7 +1589,17 @@ static int r100_packet0_check(struct radeon_cs_parser *p, r100_cs_dump_packet(p, pkt); return r; } - ib[idx] = idx_value + ((u32)reloc->lobj.gpu_offset); + if (!(p->cs_flags & RADEON_CS_KEEP_TILING_FLAGS)) { + if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO) + tile_flags |= RADEON_TXO_MACRO_TILE; + if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO) + tile_flags |= RADEON_TXO_MICRO_TILE_X2; + + tmp = idx_value & ~(0x7 << 2); + tmp |= tile_flags; + ib[idx] = tmp + ((u32)reloc->lobj.gpu_offset); + } else + ib[idx] = idx_value + ((u32)reloc->lobj.gpu_offset); track->textures[i].robj = reloc->robj; track->tex_dirty = true; break; @@ -1623,15 +1670,17 @@ static int r100_packet0_check(struct radeon_cs_parser *p, r100_cs_dump_packet(p, pkt); return r; } - - if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO) - tile_flags |= RADEON_COLOR_TILE_ENABLE; - if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO) - tile_flags |= RADEON_COLOR_MICROTILE_ENABLE; - - tmp = idx_value & ~(0x7 << 16); - tmp |= tile_flags; - ib[idx] = tmp; + if (!(p->cs_flags & RADEON_CS_KEEP_TILING_FLAGS)) { + if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO) + tile_flags |= RADEON_COLOR_TILE_ENABLE; + if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO) + tile_flags |= RADEON_COLOR_MICROTILE_ENABLE; + + tmp = idx_value & ~(0x7 << 16); + tmp |= tile_flags; + ib[idx] = tmp; + } else + ib[idx] = idx_value; track->cb[0].pitch = idx_value & RADEON_COLORPITCH_MASK; track->cb_dirty = true; @@ -3691,7 +3740,7 @@ void r100_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib) radeon_ring_write(ring, ib->length_dw); } -int r100_ib_test(struct radeon_device *rdev) +int r100_ib_test(struct radeon_device *rdev, struct radeon_ring *ring) { struct radeon_ib *ib; uint32_t scratch; @@ -3916,7 +3965,7 @@ static int r100_startup(struct radeon_device *rdev) if (r) return r; - r = r100_ib_test(rdev); + r = radeon_ib_test(rdev, RADEON_RING_TYPE_GFX_INDEX, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]); if (r) { dev_err(rdev->dev, "failed testing IB (%d).\n", r); rdev->accel_working = false; diff --git a/drivers/gpu/drm/radeon/r200.c b/drivers/gpu/drm/radeon/r200.c index eba4cbf..a59cc47 100644 --- a/drivers/gpu/drm/radeon/r200.c +++ b/drivers/gpu/drm/radeon/r200.c @@ -215,7 +215,17 @@ int r200_packet0_check(struct radeon_cs_parser *p, r100_cs_dump_packet(p, pkt); return r; } - ib[idx] = idx_value + ((u32)reloc->lobj.gpu_offset); + if (!(p->cs_flags & RADEON_CS_KEEP_TILING_FLAGS)) { + if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO) + tile_flags |= R200_TXO_MACRO_TILE; + if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO) + tile_flags |= R200_TXO_MICRO_TILE; + + tmp = idx_value & ~(0x7 << 2); + tmp |= tile_flags; + ib[idx] = tmp + ((u32)reloc->lobj.gpu_offset); + } else + ib[idx] = idx_value + ((u32)reloc->lobj.gpu_offset); track->textures[i].robj = reloc->robj; track->tex_dirty = true; break; @@ -277,14 +287,17 @@ int r200_packet0_check(struct radeon_cs_parser *p, return r; } - if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO) - tile_flags |= RADEON_COLOR_TILE_ENABLE; - if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO) - tile_flags |= RADEON_COLOR_MICROTILE_ENABLE; + if (!(p->cs_flags & RADEON_CS_KEEP_TILING_FLAGS)) { + if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO) + tile_flags |= RADEON_COLOR_TILE_ENABLE; + if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO) + tile_flags |= RADEON_COLOR_MICROTILE_ENABLE; - tmp = idx_value & ~(0x7 << 16); - tmp |= tile_flags; - ib[idx] = tmp; + tmp = idx_value & ~(0x7 << 16); + tmp |= tile_flags; + ib[idx] = tmp; + } else + ib[idx] = idx_value; track->cb[0].pitch = idx_value & RADEON_COLORPITCH_MASK; track->cb_dirty = true; diff --git a/drivers/gpu/drm/radeon/r300.c b/drivers/gpu/drm/radeon/r300.c index 6829638..fa14383 100644 --- a/drivers/gpu/drm/radeon/r300.c +++ b/drivers/gpu/drm/radeon/r300.c @@ -105,8 +105,8 @@ int rv370_pcie_gart_init(struct radeon_device *rdev) if (r) DRM_ERROR("Failed to register debugfs file for PCIE gart !\n"); rdev->gart.table_size = rdev->gart.num_gpu_pages * 4; - rdev->asic->gart_tlb_flush = &rv370_pcie_gart_tlb_flush; - rdev->asic->gart_set_page = &rv370_pcie_gart_set_page; + rdev->asic->gart.tlb_flush = &rv370_pcie_gart_tlb_flush; + rdev->asic->gart.set_page = &rv370_pcie_gart_set_page; return radeon_gart_table_vram_alloc(rdev); } @@ -206,9 +206,8 @@ void r300_fence_ring_emit(struct radeon_device *rdev, radeon_ring_write(ring, RADEON_SW_INT_FIRE); } -void r300_ring_start(struct radeon_device *rdev) +void r300_ring_start(struct radeon_device *rdev, struct radeon_ring *ring) { - struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]; unsigned gb_tile_config; int r; @@ -1419,7 +1418,7 @@ static int r300_startup(struct radeon_device *rdev) if (r) return r; - r = r100_ib_test(rdev); + r = radeon_ib_test(rdev, RADEON_RING_TYPE_GFX_INDEX, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]); if (r) { dev_err(rdev->dev, "failed testing IB (%d).\n", r); rdev->accel_working = false; diff --git a/drivers/gpu/drm/radeon/r420.c b/drivers/gpu/drm/radeon/r420.c index b143230..f3fcaac 100644 --- a/drivers/gpu/drm/radeon/r420.c +++ b/drivers/gpu/drm/radeon/r420.c @@ -279,7 +279,7 @@ static int r420_startup(struct radeon_device *rdev) if (r) return r; - r = r100_ib_test(rdev); + r = radeon_ib_test(rdev, RADEON_RING_TYPE_GFX_INDEX, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]); if (r) { dev_err(rdev->dev, "failed testing IB (%d).\n", r); rdev->accel_working = false; diff --git a/drivers/gpu/drm/radeon/r500_reg.h b/drivers/gpu/drm/radeon/r500_reg.h index 3bd8f1b..ec576aa 100644 --- a/drivers/gpu/drm/radeon/r500_reg.h +++ b/drivers/gpu/drm/radeon/r500_reg.h @@ -351,6 +351,8 @@ #define AVIVO_D1CRTC_BLANK_CONTROL 0x6084 #define AVIVO_D1CRTC_INTERLACE_CONTROL 0x6088 #define AVIVO_D1CRTC_INTERLACE_STATUS 0x608c +#define AVIVO_D1CRTC_STATUS 0x609c +# define AVIVO_D1CRTC_V_BLANK (1 << 0) #define AVIVO_D1CRTC_STATUS_POSITION 0x60a0 #define AVIVO_D1CRTC_FRAME_COUNT 0x60a4 #define AVIVO_D1CRTC_STEREO_CONTROL 0x60c4 diff --git a/drivers/gpu/drm/radeon/r520.c b/drivers/gpu/drm/radeon/r520.c index 25084e8..ebcc15b 100644 --- a/drivers/gpu/drm/radeon/r520.c +++ b/drivers/gpu/drm/radeon/r520.c @@ -33,7 +33,7 @@ /* This files gather functions specifics to: r520,rv530,rv560,rv570,r580 */ -static int r520_mc_wait_for_idle(struct radeon_device *rdev) +int r520_mc_wait_for_idle(struct radeon_device *rdev) { unsigned i; uint32_t tmp; @@ -207,7 +207,7 @@ static int r520_startup(struct radeon_device *rdev) if (r) return r; - r = r100_ib_test(rdev); + r = radeon_ib_test(rdev, RADEON_RING_TYPE_GFX_INDEX, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]); if (r) { dev_err(rdev->dev, "failed testing IB (%d).\n", r); rdev->accel_working = false; diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c index 17ca72c..5eb2382 100644 --- a/drivers/gpu/drm/radeon/r600.c +++ b/drivers/gpu/drm/radeon/r600.c @@ -2226,7 +2226,7 @@ int r600_cp_resume(struct radeon_device *rdev) r600_cp_start(rdev); ring->ready = true; - r = radeon_ring_test(rdev, ring); + r = radeon_ring_test(rdev, RADEON_RING_TYPE_GFX_INDEX, ring); if (r) { ring->ready = false; return r; @@ -2452,7 +2452,7 @@ int r600_startup(struct radeon_device *rdev) r = r600_blit_init(rdev); if (r) { r600_blit_fini(rdev); - rdev->asic->copy = NULL; + rdev->asic->copy.copy = NULL; dev_warn(rdev->dev, "failed blitter (%d) falling back to memcpy\n", r); } @@ -2493,7 +2493,7 @@ int r600_startup(struct radeon_device *rdev) if (r) return r; - r = r600_ib_test(rdev, RADEON_RING_TYPE_GFX_INDEX); + r = radeon_ib_test(rdev, RADEON_RING_TYPE_GFX_INDEX, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]); if (r) { DRM_ERROR("radeon: failed testing IB (%d).\n", r); rdev->accel_working = false; @@ -2701,13 +2701,14 @@ void r600_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib) radeon_ring_write(ring, ib->length_dw); } -int r600_ib_test(struct radeon_device *rdev, int ring) +int r600_ib_test(struct radeon_device *rdev, struct radeon_ring *ring) { struct radeon_ib *ib; uint32_t scratch; uint32_t tmp = 0; unsigned i; int r; + int ring_index = radeon_ring_index(rdev, ring); r = radeon_scratch_get(rdev, &scratch); if (r) { @@ -2715,7 +2716,7 @@ int r600_ib_test(struct radeon_device *rdev, int ring) return r; } WREG32(scratch, 0xCAFEDEAD); - r = radeon_ib_get(rdev, ring, &ib, 256); + r = radeon_ib_get(rdev, ring_index, &ib, 256); if (r) { DRM_ERROR("radeon: failed to get ib (%d).\n", r); return r; @@ -2723,20 +2724,7 @@ int r600_ib_test(struct radeon_device *rdev, int ring) ib->ptr[0] = PACKET3(PACKET3_SET_CONFIG_REG, 1); ib->ptr[1] = ((scratch - PACKET3_SET_CONFIG_REG_OFFSET) >> 2); ib->ptr[2] = 0xDEADBEEF; - ib->ptr[3] = PACKET2(0); - ib->ptr[4] = PACKET2(0); - ib->ptr[5] = PACKET2(0); - ib->ptr[6] = PACKET2(0); - ib->ptr[7] = PACKET2(0); - ib->ptr[8] = PACKET2(0); - ib->ptr[9] = PACKET2(0); - ib->ptr[10] = PACKET2(0); - ib->ptr[11] = PACKET2(0); - ib->ptr[12] = PACKET2(0); - ib->ptr[13] = PACKET2(0); - ib->ptr[14] = PACKET2(0); - ib->ptr[15] = PACKET2(0); - ib->length_dw = 16; + ib->length_dw = 3; r = radeon_ib_schedule(rdev, ib); if (r) { radeon_scratch_free(rdev, scratch); diff --git a/drivers/gpu/drm/radeon/r600_blit_kms.c b/drivers/gpu/drm/radeon/r600_blit_kms.c index accc032..db38f58 100644 --- a/drivers/gpu/drm/radeon/r600_blit_kms.c +++ b/drivers/gpu/drm/radeon/r600_blit_kms.c @@ -30,20 +30,7 @@ #include "r600d.h" #include "r600_blit_shaders.h" - -#define DI_PT_RECTLIST 0x11 -#define DI_INDEX_SIZE_16_BIT 0x0 -#define DI_SRC_SEL_AUTO_INDEX 0x2 - -#define FMT_8 0x1 -#define FMT_5_6_5 0x8 -#define FMT_8_8_8_8 0x1a -#define COLOR_8 0x1 -#define COLOR_5_6_5 0x8 -#define COLOR_8_8_8_8 0x1a - -#define RECT_UNIT_H 32 -#define RECT_UNIT_W (RADEON_GPU_PAGE_SIZE / 4 / RECT_UNIT_H) +#include "radeon_blit_common.h" /* emits 21 on rv770+, 23 on r600 */ static void diff --git a/drivers/gpu/drm/radeon/r600_cs.c b/drivers/gpu/drm/radeon/r600_cs.c index 387fcc9..0ec3f20 100644 --- a/drivers/gpu/drm/radeon/r600_cs.c +++ b/drivers/gpu/drm/radeon/r600_cs.c @@ -52,15 +52,20 @@ struct r600_cs_track { struct radeon_bo *cb_color_bo[8]; u64 cb_color_bo_mc[8]; u32 cb_color_bo_offset[8]; - struct radeon_bo *cb_color_frag_bo[8]; - struct radeon_bo *cb_color_tile_bo[8]; + struct radeon_bo *cb_color_frag_bo[8]; /* unused */ + struct radeon_bo *cb_color_tile_bo[8]; /* unused */ u32 cb_color_info[8]; - u32 cb_color_size_idx[8]; + u32 cb_color_view[8]; + u32 cb_color_size_idx[8]; /* unused */ u32 cb_target_mask; - u32 cb_shader_mask; + u32 cb_shader_mask; /* unused */ u32 cb_color_size[8]; u32 vgt_strmout_en; u32 vgt_strmout_buffer_en; + struct radeon_bo *vgt_strmout_bo[4]; + u64 vgt_strmout_bo_mc[4]; /* unused */ + u32 vgt_strmout_bo_offset[4]; + u32 vgt_strmout_size[4]; u32 db_depth_control; u32 db_depth_info; u32 db_depth_size_idx; @@ -69,13 +74,17 @@ struct r600_cs_track { u32 db_offset; struct radeon_bo *db_bo; u64 db_bo_mc; + bool sx_misc_kill_all_prims; + bool cb_dirty; + bool db_dirty; + bool streamout_dirty; }; #define FMT_8_BIT(fmt, vc) [fmt] = { 1, 1, 1, vc, CHIP_R600 } #define FMT_16_BIT(fmt, vc) [fmt] = { 1, 1, 2, vc, CHIP_R600 } -#define FMT_24_BIT(fmt) [fmt] = { 1, 1, 3, 0, CHIP_R600 } +#define FMT_24_BIT(fmt) [fmt] = { 1, 1, 4, 0, CHIP_R600 } #define FMT_32_BIT(fmt, vc) [fmt] = { 1, 1, 4, vc, CHIP_R600 } -#define FMT_48_BIT(fmt) [fmt] = { 1, 1, 6, 0, CHIP_R600 } +#define FMT_48_BIT(fmt) [fmt] = { 1, 1, 8, 0, CHIP_R600 } #define FMT_64_BIT(fmt, vc) [fmt] = { 1, 1, 8, vc, CHIP_R600 } #define FMT_96_BIT(fmt) [fmt] = { 1, 1, 12, 0, CHIP_R600 } #define FMT_128_BIT(fmt, vc) [fmt] = { 1, 1, 16,vc, CHIP_R600 } @@ -107,7 +116,7 @@ static const struct gpu_formats color_formats_table[] = { /* 24-bit */ FMT_24_BIT(V_038004_FMT_8_8_8), - + /* 32-bit */ FMT_32_BIT(V_038004_COLOR_32, 1), FMT_32_BIT(V_038004_COLOR_32_FLOAT, 1), @@ -162,22 +171,22 @@ static const struct gpu_formats color_formats_table[] = { [V_038004_FMT_32_AS_32_32_32_32] = { 1, 1, 4, 0, CHIP_CEDAR}, }; -static bool fmt_is_valid_color(u32 format) +bool r600_fmt_is_valid_color(u32 format) { if (format >= ARRAY_SIZE(color_formats_table)) return false; - + if (color_formats_table[format].valid_color) return true; return false; } -static bool fmt_is_valid_texture(u32 format, enum radeon_family family) +bool r600_fmt_is_valid_texture(u32 format, enum radeon_family family) { if (format >= ARRAY_SIZE(color_formats_table)) return false; - + if (family < color_formats_table[format].min_family) return false; @@ -187,7 +196,7 @@ static bool fmt_is_valid_texture(u32 format, enum radeon_family family) return false; } -static int fmt_get_blocksize(u32 format) +int r600_fmt_get_blocksize(u32 format) { if (format >= ARRAY_SIZE(color_formats_table)) return 0; @@ -195,7 +204,7 @@ static int fmt_get_blocksize(u32 format) return color_formats_table[format].blocksize; } -static int fmt_get_nblocksx(u32 format, u32 w) +int r600_fmt_get_nblocksx(u32 format, u32 w) { unsigned bw; @@ -209,7 +218,7 @@ static int fmt_get_nblocksx(u32 format, u32 w) return (w + bw - 1) / bw; } -static int fmt_get_nblocksy(u32 format, u32 h) +int r600_fmt_get_nblocksy(u32 format, u32 h) { unsigned bh; @@ -256,7 +265,7 @@ static int r600_get_array_mode_alignment(struct array_mode_checker *values, break; case ARRAY_LINEAR_ALIGNED: *pitch_align = max((u32)64, (u32)(values->group_size / values->blocksize)); - *height_align = tile_height; + *height_align = 1; *depth_align = 1; *base_align = values->group_size; break; @@ -269,10 +278,9 @@ static int r600_get_array_mode_alignment(struct array_mode_checker *values, *base_align = values->group_size; break; case ARRAY_2D_TILED_THIN1: - *pitch_align = max((u32)macro_tile_width, - (u32)(((values->group_size / tile_height) / - (values->blocksize * values->nsamples)) * - values->nbanks)) * tile_width; + *pitch_align = max((u32)macro_tile_width * tile_width, + (u32)((values->group_size * values->nbanks) / + (values->blocksize * values->nsamples * tile_width))); *height_align = macro_tile_height * tile_height; *depth_align = 1; *base_align = max(macro_tile_bytes, @@ -296,12 +304,14 @@ static void r600_cs_track_init(struct r600_cs_track *track) track->cb_color_size[i] = 0; track->cb_color_size_idx[i] = 0; track->cb_color_info[i] = 0; + track->cb_color_view[i] = 0xFFFFFFFF; track->cb_color_bo[i] = NULL; track->cb_color_bo_offset[i] = 0xFFFFFFFF; track->cb_color_bo_mc[i] = 0xFFFFFFFF; } track->cb_target_mask = 0xFFFFFFFF; track->cb_shader_mask = 0xFFFFFFFF; + track->cb_dirty = true; track->db_bo = NULL; track->db_bo_mc = 0xFFFFFFFF; /* assume the biggest format and that htile is enabled */ @@ -310,6 +320,16 @@ static void r600_cs_track_init(struct r600_cs_track *track) track->db_depth_size = 0xFFFFFFFF; track->db_depth_size_idx = 0; track->db_depth_control = 0xFFFFFFFF; + track->db_dirty = true; + + for (i = 0; i < 4; i++) { + track->vgt_strmout_size[i] = 0; + track->vgt_strmout_bo[i] = NULL; + track->vgt_strmout_bo_offset[i] = 0xFFFFFFFF; + track->vgt_strmout_bo_mc[i] = 0xFFFFFFFF; + } + track->streamout_dirty = true; + track->sx_misc_kill_all_prims = false; } static int r600_cs_track_validate_cb(struct radeon_cs_parser *p, int i) @@ -322,13 +342,14 @@ static int r600_cs_track_validate_cb(struct radeon_cs_parser *p, int i) volatile u32 *ib = p->ib->ptr; unsigned array_mode; u32 format; + if (G_0280A0_TILE_MODE(track->cb_color_info[i])) { dev_warn(p->dev, "FMASK or CMASK buffer are not supported by this kernel\n"); return -EINVAL; } size = radeon_bo_size(track->cb_color_bo[i]) - track->cb_color_bo_offset[i]; format = G_0280A0_FORMAT(track->cb_color_info[i]); - if (!fmt_is_valid_color(format)) { + if (!r600_fmt_is_valid_color(format)) { dev_warn(p->dev, "%s:%d cb invalid format %d for %d (0x%08X)\n", __func__, __LINE__, format, i, track->cb_color_info[i]); @@ -349,7 +370,7 @@ static int r600_cs_track_validate_cb(struct radeon_cs_parser *p, int i) array_check.nbanks = track->nbanks; array_check.npipes = track->npipes; array_check.nsamples = track->nsamples; - array_check.blocksize = fmt_get_blocksize(format); + array_check.blocksize = r600_fmt_get_blocksize(format); if (r600_get_array_mode_alignment(&array_check, &pitch_align, &height_align, &depth_align, &base_align)) { dev_warn(p->dev, "%s invalid tiling %d for %d (0x%08X)\n", __func__, @@ -393,7 +414,18 @@ static int r600_cs_track_validate_cb(struct radeon_cs_parser *p, int i) } /* check offset */ - tmp = fmt_get_nblocksy(format, height) * fmt_get_nblocksx(format, pitch) * fmt_get_blocksize(format); + tmp = r600_fmt_get_nblocksy(format, height) * r600_fmt_get_nblocksx(format, pitch) * r600_fmt_get_blocksize(format); + switch (array_mode) { + default: + case V_0280A0_ARRAY_LINEAR_GENERAL: + case V_0280A0_ARRAY_LINEAR_ALIGNED: + tmp += track->cb_color_view[i] & 0xFF; + break; + case V_0280A0_ARRAY_1D_TILED_THIN1: + case V_0280A0_ARRAY_2D_TILED_THIN1: + tmp += G_028080_SLICE_MAX(track->cb_color_view[i]) * tmp; + break; + } if ((tmp + track->cb_color_bo_offset[i]) > radeon_bo_size(track->cb_color_bo[i])) { if (array_mode == V_0280A0_ARRAY_LINEAR_GENERAL) { /* the initial DDX does bad things with the CB size occasionally */ @@ -403,10 +435,13 @@ static int r600_cs_track_validate_cb(struct radeon_cs_parser *p, int i) * broken userspace. */ } else { - dev_warn(p->dev, "%s offset[%d] %d %d %d %lu too big\n", __func__, i, - array_mode, + dev_warn(p->dev, "%s offset[%d] %d %d %d %lu too big (%d %d) (%d %d %d)\n", + __func__, i, array_mode, track->cb_color_bo_offset[i], tmp, - radeon_bo_size(track->cb_color_bo[i])); + radeon_bo_size(track->cb_color_bo[i]), + pitch, height, r600_fmt_get_nblocksx(format, pitch), + r600_fmt_get_nblocksy(format, height), + r600_fmt_get_blocksize(format)); return -EINVAL; } } @@ -430,143 +465,171 @@ static int r600_cs_track_check(struct radeon_cs_parser *p) /* on legacy kernel we don't perform advanced check */ if (p->rdev == NULL) return 0; - /* we don't support out buffer yet */ - if (track->vgt_strmout_en || track->vgt_strmout_buffer_en) { - dev_warn(p->dev, "this kernel doesn't support SMX output buffer\n"); - return -EINVAL; + + /* check streamout */ + if (track->streamout_dirty && track->vgt_strmout_en) { + for (i = 0; i < 4; i++) { + if (track->vgt_strmout_buffer_en & (1 << i)) { + if (track->vgt_strmout_bo[i]) { + u64 offset = (u64)track->vgt_strmout_bo_offset[i] + + (u64)track->vgt_strmout_size[i]; + if (offset > radeon_bo_size(track->vgt_strmout_bo[i])) { + DRM_ERROR("streamout %d bo too small: 0x%llx, 0x%lx\n", + i, offset, + radeon_bo_size(track->vgt_strmout_bo[i])); + return -EINVAL; + } + } else { + dev_warn(p->dev, "No buffer for streamout %d\n", i); + return -EINVAL; + } + } + } + track->streamout_dirty = false; } + + if (track->sx_misc_kill_all_prims) + return 0; + /* check that we have a cb for each enabled target, we don't check * shader_mask because it seems mesa isn't always setting it :( */ - tmp = track->cb_target_mask; - for (i = 0; i < 8; i++) { - if ((tmp >> (i * 4)) & 0xF) { - /* at least one component is enabled */ - if (track->cb_color_bo[i] == NULL) { - dev_warn(p->dev, "%s:%d mask 0x%08X | 0x%08X no cb for %d\n", - __func__, __LINE__, track->cb_target_mask, track->cb_shader_mask, i); - return -EINVAL; + if (track->cb_dirty) { + tmp = track->cb_target_mask; + for (i = 0; i < 8; i++) { + if ((tmp >> (i * 4)) & 0xF) { + /* at least one component is enabled */ + if (track->cb_color_bo[i] == NULL) { + dev_warn(p->dev, "%s:%d mask 0x%08X | 0x%08X no cb for %d\n", + __func__, __LINE__, track->cb_target_mask, track->cb_shader_mask, i); + return -EINVAL; + } + /* perform rewrite of CB_COLOR[0-7]_SIZE */ + r = r600_cs_track_validate_cb(p, i); + if (r) + return r; } - /* perform rewrite of CB_COLOR[0-7]_SIZE */ - r = r600_cs_track_validate_cb(p, i); - if (r) - return r; } + track->cb_dirty = false; } - /* Check depth buffer */ - if (G_028800_STENCIL_ENABLE(track->db_depth_control) || - G_028800_Z_ENABLE(track->db_depth_control)) { - u32 nviews, bpe, ntiles, size, slice_tile_max; - u32 height, height_align, pitch, pitch_align, depth_align; - u64 base_offset, base_align; - struct array_mode_checker array_check; - int array_mode; - - if (track->db_bo == NULL) { - dev_warn(p->dev, "z/stencil with no depth buffer\n"); - return -EINVAL; - } - if (G_028010_TILE_SURFACE_ENABLE(track->db_depth_info)) { - dev_warn(p->dev, "this kernel doesn't support z/stencil htile\n"); - return -EINVAL; - } - switch (G_028010_FORMAT(track->db_depth_info)) { - case V_028010_DEPTH_16: - bpe = 2; - break; - case V_028010_DEPTH_X8_24: - case V_028010_DEPTH_8_24: - case V_028010_DEPTH_X8_24_FLOAT: - case V_028010_DEPTH_8_24_FLOAT: - case V_028010_DEPTH_32_FLOAT: - bpe = 4; - break; - case V_028010_DEPTH_X24_8_32_FLOAT: - bpe = 8; - break; - default: - dev_warn(p->dev, "z/stencil with invalid format %d\n", G_028010_FORMAT(track->db_depth_info)); - return -EINVAL; - } - if ((track->db_depth_size & 0xFFFFFC00) == 0xFFFFFC00) { - if (!track->db_depth_size_idx) { - dev_warn(p->dev, "z/stencil buffer size not set\n"); - return -EINVAL; - } - tmp = radeon_bo_size(track->db_bo) - track->db_offset; - tmp = (tmp / bpe) >> 6; - if (!tmp) { - dev_warn(p->dev, "z/stencil buffer too small (0x%08X %d %d %ld)\n", - track->db_depth_size, bpe, track->db_offset, - radeon_bo_size(track->db_bo)); + + if (track->db_dirty) { + /* Check depth buffer */ + if (G_028800_STENCIL_ENABLE(track->db_depth_control) || + G_028800_Z_ENABLE(track->db_depth_control)) { + u32 nviews, bpe, ntiles, size, slice_tile_max; + u32 height, height_align, pitch, pitch_align, depth_align; + u64 base_offset, base_align; + struct array_mode_checker array_check; + int array_mode; + + if (track->db_bo == NULL) { + dev_warn(p->dev, "z/stencil with no depth buffer\n"); return -EINVAL; } - ib[track->db_depth_size_idx] = S_028000_SLICE_TILE_MAX(tmp - 1) | (track->db_depth_size & 0x3FF); - } else { - size = radeon_bo_size(track->db_bo); - /* pitch in pixels */ - pitch = (G_028000_PITCH_TILE_MAX(track->db_depth_size) + 1) * 8; - slice_tile_max = G_028000_SLICE_TILE_MAX(track->db_depth_size) + 1; - slice_tile_max *= 64; - height = slice_tile_max / pitch; - if (height > 8192) - height = 8192; - base_offset = track->db_bo_mc + track->db_offset; - array_mode = G_028010_ARRAY_MODE(track->db_depth_info); - array_check.array_mode = array_mode; - array_check.group_size = track->group_size; - array_check.nbanks = track->nbanks; - array_check.npipes = track->npipes; - array_check.nsamples = track->nsamples; - array_check.blocksize = bpe; - if (r600_get_array_mode_alignment(&array_check, - &pitch_align, &height_align, &depth_align, &base_align)) { - dev_warn(p->dev, "%s invalid tiling %d (0x%08X)\n", __func__, - G_028010_ARRAY_MODE(track->db_depth_info), - track->db_depth_info); + if (G_028010_TILE_SURFACE_ENABLE(track->db_depth_info)) { + dev_warn(p->dev, "this kernel doesn't support z/stencil htile\n"); return -EINVAL; } - switch (array_mode) { - case V_028010_ARRAY_1D_TILED_THIN1: - /* don't break userspace */ - height &= ~0x7; + switch (G_028010_FORMAT(track->db_depth_info)) { + case V_028010_DEPTH_16: + bpe = 2; + break; + case V_028010_DEPTH_X8_24: + case V_028010_DEPTH_8_24: + case V_028010_DEPTH_X8_24_FLOAT: + case V_028010_DEPTH_8_24_FLOAT: + case V_028010_DEPTH_32_FLOAT: + bpe = 4; break; - case V_028010_ARRAY_2D_TILED_THIN1: + case V_028010_DEPTH_X24_8_32_FLOAT: + bpe = 8; break; default: - dev_warn(p->dev, "%s invalid tiling %d (0x%08X)\n", __func__, - G_028010_ARRAY_MODE(track->db_depth_info), - track->db_depth_info); + dev_warn(p->dev, "z/stencil with invalid format %d\n", G_028010_FORMAT(track->db_depth_info)); return -EINVAL; } + if ((track->db_depth_size & 0xFFFFFC00) == 0xFFFFFC00) { + if (!track->db_depth_size_idx) { + dev_warn(p->dev, "z/stencil buffer size not set\n"); + return -EINVAL; + } + tmp = radeon_bo_size(track->db_bo) - track->db_offset; + tmp = (tmp / bpe) >> 6; + if (!tmp) { + dev_warn(p->dev, "z/stencil buffer too small (0x%08X %d %d %ld)\n", + track->db_depth_size, bpe, track->db_offset, + radeon_bo_size(track->db_bo)); + return -EINVAL; + } + ib[track->db_depth_size_idx] = S_028000_SLICE_TILE_MAX(tmp - 1) | (track->db_depth_size & 0x3FF); + } else { + size = radeon_bo_size(track->db_bo); + /* pitch in pixels */ + pitch = (G_028000_PITCH_TILE_MAX(track->db_depth_size) + 1) * 8; + slice_tile_max = G_028000_SLICE_TILE_MAX(track->db_depth_size) + 1; + slice_tile_max *= 64; + height = slice_tile_max / pitch; + if (height > 8192) + height = 8192; + base_offset = track->db_bo_mc + track->db_offset; + array_mode = G_028010_ARRAY_MODE(track->db_depth_info); + array_check.array_mode = array_mode; + array_check.group_size = track->group_size; + array_check.nbanks = track->nbanks; + array_check.npipes = track->npipes; + array_check.nsamples = track->nsamples; + array_check.blocksize = bpe; + if (r600_get_array_mode_alignment(&array_check, + &pitch_align, &height_align, &depth_align, &base_align)) { + dev_warn(p->dev, "%s invalid tiling %d (0x%08X)\n", __func__, + G_028010_ARRAY_MODE(track->db_depth_info), + track->db_depth_info); + return -EINVAL; + } + switch (array_mode) { + case V_028010_ARRAY_1D_TILED_THIN1: + /* don't break userspace */ + height &= ~0x7; + break; + case V_028010_ARRAY_2D_TILED_THIN1: + break; + default: + dev_warn(p->dev, "%s invalid tiling %d (0x%08X)\n", __func__, + G_028010_ARRAY_MODE(track->db_depth_info), + track->db_depth_info); + return -EINVAL; + } - if (!IS_ALIGNED(pitch, pitch_align)) { - dev_warn(p->dev, "%s:%d db pitch (%d, 0x%x, %d) invalid\n", - __func__, __LINE__, pitch, pitch_align, array_mode); - return -EINVAL; - } - if (!IS_ALIGNED(height, height_align)) { - dev_warn(p->dev, "%s:%d db height (%d, 0x%x, %d) invalid\n", - __func__, __LINE__, height, height_align, array_mode); - return -EINVAL; - } - if (!IS_ALIGNED(base_offset, base_align)) { - dev_warn(p->dev, "%s offset[%d] 0x%llx, 0x%llx, %d not aligned\n", __func__, i, - base_offset, base_align, array_mode); - return -EINVAL; - } + if (!IS_ALIGNED(pitch, pitch_align)) { + dev_warn(p->dev, "%s:%d db pitch (%d, 0x%x, %d) invalid\n", + __func__, __LINE__, pitch, pitch_align, array_mode); + return -EINVAL; + } + if (!IS_ALIGNED(height, height_align)) { + dev_warn(p->dev, "%s:%d db height (%d, 0x%x, %d) invalid\n", + __func__, __LINE__, height, height_align, array_mode); + return -EINVAL; + } + if (!IS_ALIGNED(base_offset, base_align)) { + dev_warn(p->dev, "%s offset[%d] 0x%llx, 0x%llx, %d not aligned\n", __func__, i, + base_offset, base_align, array_mode); + return -EINVAL; + } - ntiles = G_028000_SLICE_TILE_MAX(track->db_depth_size) + 1; - nviews = G_028004_SLICE_MAX(track->db_depth_view) + 1; - tmp = ntiles * bpe * 64 * nviews; - if ((tmp + track->db_offset) > radeon_bo_size(track->db_bo)) { - dev_warn(p->dev, "z/stencil buffer (%d) too small (0x%08X %d %d %d -> %u have %lu)\n", - array_mode, - track->db_depth_size, ntiles, nviews, bpe, tmp + track->db_offset, - radeon_bo_size(track->db_bo)); - return -EINVAL; + ntiles = G_028000_SLICE_TILE_MAX(track->db_depth_size) + 1; + nviews = G_028004_SLICE_MAX(track->db_depth_view) + 1; + tmp = ntiles * bpe * 64 * nviews; + if ((tmp + track->db_offset) > radeon_bo_size(track->db_bo)) { + dev_warn(p->dev, "z/stencil buffer (%d) too small (0x%08X %d %d %d -> %u have %lu)\n", + array_mode, + track->db_depth_size, ntiles, nviews, bpe, tmp + track->db_offset, + radeon_bo_size(track->db_bo)); + return -EINVAL; + } } } + track->db_dirty = false; } return 0; } @@ -939,6 +1002,7 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) break; case R_028800_DB_DEPTH_CONTROL: track->db_depth_control = radeon_get_ib_value(p, idx); + track->db_dirty = true; break; case R_028010_DB_DEPTH_INFO: if (!(p->cs_flags & RADEON_CS_KEEP_TILING_FLAGS) && @@ -959,24 +1023,66 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) ib[idx] |= S_028010_ARRAY_MODE(V_028010_ARRAY_1D_TILED_THIN1); track->db_depth_info |= S_028010_ARRAY_MODE(V_028010_ARRAY_1D_TILED_THIN1); } - } else + } else { track->db_depth_info = radeon_get_ib_value(p, idx); + } + track->db_dirty = true; break; case R_028004_DB_DEPTH_VIEW: track->db_depth_view = radeon_get_ib_value(p, idx); + track->db_dirty = true; break; case R_028000_DB_DEPTH_SIZE: track->db_depth_size = radeon_get_ib_value(p, idx); track->db_depth_size_idx = idx; + track->db_dirty = true; break; case R_028AB0_VGT_STRMOUT_EN: track->vgt_strmout_en = radeon_get_ib_value(p, idx); + track->streamout_dirty = true; break; case R_028B20_VGT_STRMOUT_BUFFER_EN: track->vgt_strmout_buffer_en = radeon_get_ib_value(p, idx); + track->streamout_dirty = true; + break; + case VGT_STRMOUT_BUFFER_BASE_0: + case VGT_STRMOUT_BUFFER_BASE_1: + case VGT_STRMOUT_BUFFER_BASE_2: + case VGT_STRMOUT_BUFFER_BASE_3: + r = r600_cs_packet_next_reloc(p, &reloc); + if (r) { + dev_warn(p->dev, "bad SET_CONTEXT_REG " + "0x%04X\n", reg); + return -EINVAL; + } + tmp = (reg - VGT_STRMOUT_BUFFER_BASE_0) / 16; + track->vgt_strmout_bo_offset[tmp] = radeon_get_ib_value(p, idx) << 8; + ib[idx] += (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff); + track->vgt_strmout_bo[tmp] = reloc->robj; + track->vgt_strmout_bo_mc[tmp] = reloc->lobj.gpu_offset; + track->streamout_dirty = true; + break; + case VGT_STRMOUT_BUFFER_SIZE_0: + case VGT_STRMOUT_BUFFER_SIZE_1: + case VGT_STRMOUT_BUFFER_SIZE_2: + case VGT_STRMOUT_BUFFER_SIZE_3: + tmp = (reg - VGT_STRMOUT_BUFFER_SIZE_0) / 16; + /* size in register is DWs, convert to bytes */ + track->vgt_strmout_size[tmp] = radeon_get_ib_value(p, idx) * 4; + track->streamout_dirty = true; + break; + case CP_COHER_BASE: + r = r600_cs_packet_next_reloc(p, &reloc); + if (r) { + dev_warn(p->dev, "missing reloc for CP_COHER_BASE " + "0x%04X\n", reg); + return -EINVAL; + } + ib[idx] += (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff); break; case R_028238_CB_TARGET_MASK: track->cb_target_mask = radeon_get_ib_value(p, idx); + track->cb_dirty = true; break; case R_02823C_CB_SHADER_MASK: track->cb_shader_mask = radeon_get_ib_value(p, idx); @@ -984,6 +1090,7 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) case R_028C04_PA_SC_AA_CONFIG: tmp = G_028C04_MSAA_NUM_SAMPLES(radeon_get_ib_value(p, idx)); track->nsamples = 1 << tmp; + track->cb_dirty = true; break; case R_0280A0_CB_COLOR0_INFO: case R_0280A4_CB_COLOR1_INFO: @@ -1013,6 +1120,19 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) tmp = (reg - R_0280A0_CB_COLOR0_INFO) / 4; track->cb_color_info[tmp] = radeon_get_ib_value(p, idx); } + track->cb_dirty = true; + break; + case R_028080_CB_COLOR0_VIEW: + case R_028084_CB_COLOR1_VIEW: + case R_028088_CB_COLOR2_VIEW: + case R_02808C_CB_COLOR3_VIEW: + case R_028090_CB_COLOR4_VIEW: + case R_028094_CB_COLOR5_VIEW: + case R_028098_CB_COLOR6_VIEW: + case R_02809C_CB_COLOR7_VIEW: + tmp = (reg - R_028080_CB_COLOR0_VIEW) / 4; + track->cb_color_view[tmp] = radeon_get_ib_value(p, idx); + track->cb_dirty = true; break; case R_028060_CB_COLOR0_SIZE: case R_028064_CB_COLOR1_SIZE: @@ -1025,6 +1145,7 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) tmp = (reg - R_028060_CB_COLOR0_SIZE) / 4; track->cb_color_size[tmp] = radeon_get_ib_value(p, idx); track->cb_color_size_idx[tmp] = idx; + track->cb_dirty = true; break; /* This register were added late, there is userspace * which does provide relocation for those but set @@ -1107,6 +1228,7 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) track->cb_color_base_last[tmp] = ib[idx]; track->cb_color_bo[tmp] = reloc->robj; track->cb_color_bo_mc[tmp] = reloc->lobj.gpu_offset; + track->cb_dirty = true; break; case DB_DEPTH_BASE: r = r600_cs_packet_next_reloc(p, &reloc); @@ -1119,6 +1241,7 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) ib[idx] += (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff); track->db_bo = reloc->robj; track->db_bo_mc = reloc->lobj.gpu_offset; + track->db_dirty = true; break; case DB_HTILE_DATA_BASE: case SQ_PGM_START_FS: @@ -1191,6 +1314,9 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) } ib[idx] += (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff); break; + case SX_MISC: + track->sx_misc_kill_all_prims = (radeon_get_ib_value(p, idx) & 0x1) != 0; + break; default: dev_warn(p->dev, "forbidden register 0x%08x at %d\n", reg, idx); return -EINVAL; @@ -1198,7 +1324,7 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) return 0; } -static unsigned mip_minify(unsigned size, unsigned level) +unsigned r600_mip_minify(unsigned size, unsigned level) { unsigned val; @@ -1220,22 +1346,22 @@ static void r600_texture_size(unsigned nfaces, unsigned blevel, unsigned llevel, unsigned nlevels = llevel - blevel + 1; *l0_size = -1; - blocksize = fmt_get_blocksize(format); + blocksize = r600_fmt_get_blocksize(format); - w0 = mip_minify(w0, 0); - h0 = mip_minify(h0, 0); - d0 = mip_minify(d0, 0); + w0 = r600_mip_minify(w0, 0); + h0 = r600_mip_minify(h0, 0); + d0 = r600_mip_minify(d0, 0); for(i = 0, offset = 0, level = blevel; i < nlevels; i++, level++) { - width = mip_minify(w0, i); - nbx = fmt_get_nblocksx(format, width); + width = r600_mip_minify(w0, i); + nbx = r600_fmt_get_nblocksx(format, width); nbx = round_up(nbx, block_align); - height = mip_minify(h0, i); - nby = fmt_get_nblocksy(format, height); + height = r600_mip_minify(h0, i); + nby = r600_fmt_get_nblocksy(format, height); nby = round_up(nby, height_align); - depth = mip_minify(d0, i); + depth = r600_mip_minify(d0, i); size = nbx * nby * blocksize; if (nfaces) @@ -1327,7 +1453,7 @@ static int r600_check_texture_resource(struct radeon_cs_parser *p, u32 idx, return -EINVAL; } format = G_038004_DATA_FORMAT(word1); - if (!fmt_is_valid_texture(format, p->family)) { + if (!r600_fmt_is_valid_texture(format, p->family)) { dev_warn(p->dev, "%s:%d texture invalid format %d\n", __func__, __LINE__, format); return -EINVAL; @@ -1340,7 +1466,7 @@ static int r600_check_texture_resource(struct radeon_cs_parser *p, u32 idx, array_check.nbanks = track->nbanks; array_check.npipes = track->npipes; array_check.nsamples = 1; - array_check.blocksize = fmt_get_blocksize(format); + array_check.blocksize = r600_fmt_get_blocksize(format); if (r600_get_array_mode_alignment(&array_check, &pitch_align, &height_align, &depth_align, &base_align)) { dev_warn(p->dev, "%s:%d tex array mode (%d) invalid\n", @@ -1373,6 +1499,10 @@ static int r600_check_texture_resource(struct radeon_cs_parser *p, u32 idx, word1 = radeon_get_ib_value(p, idx + 5); blevel = G_038010_BASE_LEVEL(word0); llevel = G_038014_LAST_LEVEL(word1); + if (blevel > llevel) { + dev_warn(p->dev, "texture blevel %d > llevel %d\n", + blevel, llevel); + } if (array == 1) { barray = G_038014_BASE_ARRAY(word1); larray = G_038014_LAST_ARRAY(word1); @@ -1384,8 +1514,10 @@ static int r600_check_texture_resource(struct radeon_cs_parser *p, u32 idx, &l0_size, &mipmap_size); /* using get ib will give us the offset into the texture bo */ if ((l0_size + word2) > radeon_bo_size(texture)) { - dev_warn(p->dev, "texture bo too small (%d %d %d %d -> %d have %ld)\n", - w0, h0, format, word2, l0_size, radeon_bo_size(texture)); + dev_warn(p->dev, "texture bo too small ((%d %d) (%d %d) %d %d %d -> %d have %ld)\n", + w0, h0, pitch_align, height_align, + array_check.array_mode, format, word2, + l0_size, radeon_bo_size(texture)); dev_warn(p->dev, "alignments %d %d %d %lld\n", pitch, pitch_align, height_align, base_align); return -EINVAL; } @@ -1398,6 +1530,22 @@ static int r600_check_texture_resource(struct radeon_cs_parser *p, u32 idx, return 0; } +static bool r600_is_safe_reg(struct radeon_cs_parser *p, u32 reg, u32 idx) +{ + u32 m, i; + + i = (reg >> 7); + if (i >= ARRAY_SIZE(r600_reg_safe_bm)) { + dev_warn(p->dev, "forbidden register 0x%08x at %d\n", reg, idx); + return false; + } + m = 1 << ((reg >> 2) & 31); + if (!(r600_reg_safe_bm[i] & m)) + return true; + dev_warn(p->dev, "forbidden register 0x%08x at %d\n", reg, idx); + return false; +} + static int r600_packet3_check(struct radeon_cs_parser *p, struct radeon_cs_packet *pkt) { @@ -1420,6 +1568,8 @@ static int r600_packet3_check(struct radeon_cs_parser *p, { int pred_op; int tmp; + uint64_t offset; + if (pkt->count != 1) { DRM_ERROR("bad SET PREDICATION\n"); return -EINVAL; @@ -1443,8 +1593,12 @@ static int r600_packet3_check(struct radeon_cs_parser *p, return -EINVAL; } - ib[idx + 0] = idx_value + (u32)(reloc->lobj.gpu_offset & 0xffffffff); - ib[idx + 1] = tmp + (upper_32_bits(reloc->lobj.gpu_offset) & 0xff); + offset = reloc->lobj.gpu_offset + + (idx_value & 0xfffffff0) + + ((u64)(tmp & 0xff) << 32); + + ib[idx + 0] = offset; + ib[idx + 1] = (tmp & 0xffffff00) | (upper_32_bits(offset) & 0xff); } break; @@ -1468,6 +1622,8 @@ static int r600_packet3_check(struct radeon_cs_parser *p, } break; case PACKET3_DRAW_INDEX: + { + uint64_t offset; if (pkt->count != 3) { DRM_ERROR("bad DRAW_INDEX\n"); return -EINVAL; @@ -1477,14 +1633,21 @@ static int r600_packet3_check(struct radeon_cs_parser *p, DRM_ERROR("bad DRAW_INDEX\n"); return -EINVAL; } - ib[idx+0] = idx_value + (u32)(reloc->lobj.gpu_offset & 0xffffffff); - ib[idx+1] += upper_32_bits(reloc->lobj.gpu_offset) & 0xff; + + offset = reloc->lobj.gpu_offset + + idx_value + + ((u64)(radeon_get_ib_value(p, idx+1) & 0xff) << 32); + + ib[idx+0] = offset; + ib[idx+1] = upper_32_bits(offset) & 0xff; + r = r600_cs_track_check(p); if (r) { dev_warn(p->dev, "%s:%d invalid cmd stream\n", __func__, __LINE__); return r; } break; + } case PACKET3_DRAW_INDEX_AUTO: if (pkt->count != 1) { DRM_ERROR("bad DRAW_INDEX_AUTO\n"); @@ -1515,13 +1678,20 @@ static int r600_packet3_check(struct radeon_cs_parser *p, } /* bit 4 is reg (0) or mem (1) */ if (idx_value & 0x10) { + uint64_t offset; + r = r600_cs_packet_next_reloc(p, &reloc); if (r) { DRM_ERROR("bad WAIT_REG_MEM\n"); return -EINVAL; } - ib[idx+1] += (u32)(reloc->lobj.gpu_offset & 0xffffffff); - ib[idx+2] += upper_32_bits(reloc->lobj.gpu_offset) & 0xff; + + offset = reloc->lobj.gpu_offset + + (radeon_get_ib_value(p, idx+1) & 0xfffffff0) + + ((u64)(radeon_get_ib_value(p, idx+2) & 0xff) << 32); + + ib[idx+1] = (ib[idx+1] & 0x3) | (offset & 0xfffffff0); + ib[idx+2] = upper_32_bits(offset) & 0xff; } break; case PACKET3_SURFACE_SYNC: @@ -1546,16 +1716,25 @@ static int r600_packet3_check(struct radeon_cs_parser *p, return -EINVAL; } if (pkt->count) { + uint64_t offset; + r = r600_cs_packet_next_reloc(p, &reloc); if (r) { DRM_ERROR("bad EVENT_WRITE\n"); return -EINVAL; } - ib[idx+1] += (u32)(reloc->lobj.gpu_offset & 0xffffffff); - ib[idx+2] += upper_32_bits(reloc->lobj.gpu_offset) & 0xff; + offset = reloc->lobj.gpu_offset + + (radeon_get_ib_value(p, idx+1) & 0xfffffff8) + + ((u64)(radeon_get_ib_value(p, idx+2) & 0xff) << 32); + + ib[idx+1] = offset & 0xfffffff8; + ib[idx+2] = upper_32_bits(offset) & 0xff; } break; case PACKET3_EVENT_WRITE_EOP: + { + uint64_t offset; + if (pkt->count != 4) { DRM_ERROR("bad EVENT_WRITE_EOP\n"); return -EINVAL; @@ -1565,9 +1744,15 @@ static int r600_packet3_check(struct radeon_cs_parser *p, DRM_ERROR("bad EVENT_WRITE\n"); return -EINVAL; } - ib[idx+1] += (u32)(reloc->lobj.gpu_offset & 0xffffffff); - ib[idx+2] += upper_32_bits(reloc->lobj.gpu_offset) & 0xff; + + offset = reloc->lobj.gpu_offset + + (radeon_get_ib_value(p, idx+1) & 0xfffffffc) + + ((u64)(radeon_get_ib_value(p, idx+2) & 0xff) << 32); + + ib[idx+1] = offset & 0xfffffffc; + ib[idx+2] = (ib[idx+2] & 0xffffff00) | (upper_32_bits(offset) & 0xff); break; + } case PACKET3_SET_CONFIG_REG: start_reg = (idx_value << 2) + PACKET3_SET_CONFIG_REG_OFFSET; end_reg = 4 * pkt->count + start_reg - 4; @@ -1652,6 +1837,8 @@ static int r600_packet3_check(struct radeon_cs_parser *p, ib[idx+1+(i*7)+3] += mip_offset; break; case SQ_TEX_VTX_VALID_BUFFER: + { + uint64_t offset64; /* vtx base */ r = r600_cs_packet_next_reloc(p, &reloc); if (r) { @@ -1664,11 +1851,15 @@ static int r600_packet3_check(struct radeon_cs_parser *p, /* force size to size of the buffer */ dev_warn(p->dev, "vbo resource seems too big (%d) for the bo (%ld)\n", size + offset, radeon_bo_size(reloc->robj)); - ib[idx+1+(i*7)+1] = radeon_bo_size(reloc->robj); + ib[idx+1+(i*7)+1] = radeon_bo_size(reloc->robj) - offset; } - ib[idx+1+(i*7)+0] += (u32)((reloc->lobj.gpu_offset) & 0xffffffff); - ib[idx+1+(i*7)+2] += upper_32_bits(reloc->lobj.gpu_offset) & 0xff; + + offset64 = reloc->lobj.gpu_offset + offset; + ib[idx+1+(i*8)+0] = offset64; + ib[idx+1+(i*8)+2] = (ib[idx+1+(i*8)+2] & 0xffffff00) | + (upper_32_bits(offset64) & 0xff); break; + } case SQ_TEX_VTX_INVALID_TEXTURE: case SQ_TEX_VTX_INVALID_BUFFER: default: @@ -1743,6 +1934,104 @@ static int r600_packet3_check(struct radeon_cs_parser *p, return -EINVAL; } break; + case PACKET3_STRMOUT_BUFFER_UPDATE: + if (pkt->count != 4) { + DRM_ERROR("bad STRMOUT_BUFFER_UPDATE (invalid count)\n"); + return -EINVAL; + } + /* Updating memory at DST_ADDRESS. */ + if (idx_value & 0x1) { + u64 offset; + r = r600_cs_packet_next_reloc(p, &reloc); + if (r) { + DRM_ERROR("bad STRMOUT_BUFFER_UPDATE (missing dst reloc)\n"); + return -EINVAL; + } + offset = radeon_get_ib_value(p, idx+1); + offset += ((u64)(radeon_get_ib_value(p, idx+2) & 0xff)) << 32; + if ((offset + 4) > radeon_bo_size(reloc->robj)) { + DRM_ERROR("bad STRMOUT_BUFFER_UPDATE dst bo too small: 0x%llx, 0x%lx\n", + offset + 4, radeon_bo_size(reloc->robj)); + return -EINVAL; + } + offset += reloc->lobj.gpu_offset; + ib[idx+1] = offset; + ib[idx+2] = upper_32_bits(offset) & 0xff; + } + /* Reading data from SRC_ADDRESS. */ + if (((idx_value >> 1) & 0x3) == 2) { + u64 offset; + r = r600_cs_packet_next_reloc(p, &reloc); + if (r) { + DRM_ERROR("bad STRMOUT_BUFFER_UPDATE (missing src reloc)\n"); + return -EINVAL; + } + offset = radeon_get_ib_value(p, idx+3); + offset += ((u64)(radeon_get_ib_value(p, idx+4) & 0xff)) << 32; + if ((offset + 4) > radeon_bo_size(reloc->robj)) { + DRM_ERROR("bad STRMOUT_BUFFER_UPDATE src bo too small: 0x%llx, 0x%lx\n", + offset + 4, radeon_bo_size(reloc->robj)); + return -EINVAL; + } + offset += reloc->lobj.gpu_offset; + ib[idx+3] = offset; + ib[idx+4] = upper_32_bits(offset) & 0xff; + } + break; + case PACKET3_COPY_DW: + if (pkt->count != 4) { + DRM_ERROR("bad COPY_DW (invalid count)\n"); + return -EINVAL; + } + if (idx_value & 0x1) { + u64 offset; + /* SRC is memory. */ + r = r600_cs_packet_next_reloc(p, &reloc); + if (r) { + DRM_ERROR("bad COPY_DW (missing src reloc)\n"); + return -EINVAL; + } + offset = radeon_get_ib_value(p, idx+1); + offset += ((u64)(radeon_get_ib_value(p, idx+2) & 0xff)) << 32; + if ((offset + 4) > radeon_bo_size(reloc->robj)) { + DRM_ERROR("bad COPY_DW src bo too small: 0x%llx, 0x%lx\n", + offset + 4, radeon_bo_size(reloc->robj)); + return -EINVAL; + } + offset += reloc->lobj.gpu_offset; + ib[idx+1] = offset; + ib[idx+2] = upper_32_bits(offset) & 0xff; + } else { + /* SRC is a reg. */ + reg = radeon_get_ib_value(p, idx+1) << 2; + if (!r600_is_safe_reg(p, reg, idx+1)) + return -EINVAL; + } + if (idx_value & 0x2) { + u64 offset; + /* DST is memory. */ + r = r600_cs_packet_next_reloc(p, &reloc); + if (r) { + DRM_ERROR("bad COPY_DW (missing dst reloc)\n"); + return -EINVAL; + } + offset = radeon_get_ib_value(p, idx+3); + offset += ((u64)(radeon_get_ib_value(p, idx+4) & 0xff)) << 32; + if ((offset + 4) > radeon_bo_size(reloc->robj)) { + DRM_ERROR("bad COPY_DW dst bo too small: 0x%llx, 0x%lx\n", + offset + 4, radeon_bo_size(reloc->robj)); + return -EINVAL; + } + offset += reloc->lobj.gpu_offset; + ib[idx+3] = offset; + ib[idx+4] = upper_32_bits(offset) & 0xff; + } else { + /* DST is a reg. */ + reg = radeon_get_ib_value(p, idx+3) << 2; + if (!r600_is_safe_reg(p, reg, idx+3)) + return -EINVAL; + } + break; case PACKET3_NOP: break; default: diff --git a/drivers/gpu/drm/radeon/r600d.h b/drivers/gpu/drm/radeon/r600d.h index 9b23670..8ae328f 100644 --- a/drivers/gpu/drm/radeon/r600d.h +++ b/drivers/gpu/drm/radeon/r600d.h @@ -78,6 +78,20 @@ #define CB_COLOR0_SIZE 0x28060 #define CB_COLOR0_VIEW 0x28080 +#define R_028080_CB_COLOR0_VIEW 0x028080 +#define S_028080_SLICE_START(x) (((x) & 0x7FF) << 0) +#define G_028080_SLICE_START(x) (((x) >> 0) & 0x7FF) +#define C_028080_SLICE_START 0xFFFFF800 +#define S_028080_SLICE_MAX(x) (((x) & 0x7FF) << 13) +#define G_028080_SLICE_MAX(x) (((x) >> 13) & 0x7FF) +#define C_028080_SLICE_MAX 0xFF001FFF +#define R_028084_CB_COLOR1_VIEW 0x028084 +#define R_028088_CB_COLOR2_VIEW 0x028088 +#define R_02808C_CB_COLOR3_VIEW 0x02808C +#define R_028090_CB_COLOR4_VIEW 0x028090 +#define R_028094_CB_COLOR5_VIEW 0x028094 +#define R_028098_CB_COLOR6_VIEW 0x028098 +#define R_02809C_CB_COLOR7_VIEW 0x02809C #define CB_COLOR0_INFO 0x280a0 # define CB_FORMAT(x) ((x) << 2) # define CB_ARRAY_MODE(x) ((x) << 8) @@ -493,6 +507,11 @@ #define VGT_STRMOUT_BUFFER_OFFSET_1 0x28AEC #define VGT_STRMOUT_BUFFER_OFFSET_2 0x28AFC #define VGT_STRMOUT_BUFFER_OFFSET_3 0x28B0C +#define VGT_STRMOUT_BUFFER_SIZE_0 0x28AD0 +#define VGT_STRMOUT_BUFFER_SIZE_1 0x28AE0 +#define VGT_STRMOUT_BUFFER_SIZE_2 0x28AF0 +#define VGT_STRMOUT_BUFFER_SIZE_3 0x28B00 + #define VGT_STRMOUT_EN 0x28AB0 #define VGT_VERTEX_REUSE_BLOCK_CNTL 0x28C58 #define VTX_REUSE_DEPTH_MASK 0x000000FF @@ -835,6 +854,7 @@ # define PACKET3_SEM_SEL_SIGNAL (0x6 << 29) # define PACKET3_SEM_SEL_WAIT (0x7 << 29) #define PACKET3_MPEG_INDEX 0x3A +#define PACKET3_COPY_DW 0x3B #define PACKET3_WAIT_REG_MEM 0x3C #define PACKET3_MEM_WRITE 0x3D #define PACKET3_INDIRECT_BUFFER 0x32 diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h index 1668ec1..d2870a0 100644 --- a/drivers/gpu/drm/radeon/radeon.h +++ b/drivers/gpu/drm/radeon/radeon.h @@ -242,6 +242,9 @@ extern int rv6xx_get_temp(struct radeon_device *rdev); extern int rv770_get_temp(struct radeon_device *rdev); extern int evergreen_get_temp(struct radeon_device *rdev); extern int sumo_get_temp(struct radeon_device *rdev); +extern void evergreen_tiling_fields(unsigned tiling_flags, unsigned *bankw, + unsigned *bankh, unsigned *mtaspect, + unsigned *tile_split); /* * Fences. @@ -411,9 +414,6 @@ int radeon_gem_object_create(struct radeon_device *rdev, int size, int alignment, int initial_domain, bool discardable, bool kernel, struct drm_gem_object **obj); -int radeon_gem_object_pin(struct drm_gem_object *obj, uint32_t pin_domain, - uint64_t *gpu_addr); -void radeon_gem_object_unpin(struct drm_gem_object *obj); int radeon_mode_dumb_create(struct drm_file *file_priv, struct drm_device *dev, @@ -780,7 +780,6 @@ int radeon_ib_pool_init(struct radeon_device *rdev); void radeon_ib_pool_fini(struct radeon_device *rdev); int radeon_ib_pool_start(struct radeon_device *rdev); int radeon_ib_pool_suspend(struct radeon_device *rdev); -int radeon_ib_test(struct radeon_device *rdev); /* Ring access between begin & end cannot sleep */ int radeon_ring_index(struct radeon_device *rdev, struct radeon_ring *cp); void radeon_ring_free_size(struct radeon_device *rdev, struct radeon_ring *cp); @@ -833,7 +832,6 @@ struct radeon_cs_parser { struct radeon_cs_reloc *relocs; struct radeon_cs_reloc **relocs_ptr; struct list_head validated; - bool sync_to_ring[RADEON_NUM_RINGS]; /* indices of various chunks */ int chunk_ib_idx; int chunk_relocs_idx; @@ -1132,57 +1130,6 @@ struct radeon_asic { void (*vga_set_state)(struct radeon_device *rdev, bool state); bool (*gpu_is_lockup)(struct radeon_device *rdev, struct radeon_ring *cp); int (*asic_reset)(struct radeon_device *rdev); - void (*gart_tlb_flush)(struct radeon_device *rdev); - int (*gart_set_page)(struct radeon_device *rdev, int i, uint64_t addr); - int (*cp_init)(struct radeon_device *rdev, unsigned ring_size); - void (*cp_fini)(struct radeon_device *rdev); - void (*cp_disable)(struct radeon_device *rdev); - void (*ring_start)(struct radeon_device *rdev); - - struct { - void (*ib_execute)(struct radeon_device *rdev, struct radeon_ib *ib); - int (*ib_parse)(struct radeon_device *rdev, struct radeon_ib *ib); - void (*emit_fence)(struct radeon_device *rdev, struct radeon_fence *fence); - void (*emit_semaphore)(struct radeon_device *rdev, struct radeon_ring *cp, - struct radeon_semaphore *semaphore, bool emit_wait); - } ring[RADEON_NUM_RINGS]; - - int (*ring_test)(struct radeon_device *rdev, struct radeon_ring *cp); - int (*irq_set)(struct radeon_device *rdev); - int (*irq_process)(struct radeon_device *rdev); - u32 (*get_vblank_counter)(struct radeon_device *rdev, int crtc); - int (*cs_parse)(struct radeon_cs_parser *p); - int (*copy_blit)(struct radeon_device *rdev, - uint64_t src_offset, - uint64_t dst_offset, - unsigned num_gpu_pages, - struct radeon_fence *fence); - int (*copy_dma)(struct radeon_device *rdev, - uint64_t src_offset, - uint64_t dst_offset, - unsigned num_gpu_pages, - struct radeon_fence *fence); - int (*copy)(struct radeon_device *rdev, - uint64_t src_offset, - uint64_t dst_offset, - unsigned num_gpu_pages, - struct radeon_fence *fence); - uint32_t (*get_engine_clock)(struct radeon_device *rdev); - void (*set_engine_clock)(struct radeon_device *rdev, uint32_t eng_clock); - uint32_t (*get_memory_clock)(struct radeon_device *rdev); - void (*set_memory_clock)(struct radeon_device *rdev, uint32_t mem_clock); - int (*get_pcie_lanes)(struct radeon_device *rdev); - void (*set_pcie_lanes)(struct radeon_device *rdev, int lanes); - void (*set_clock_gating)(struct radeon_device *rdev, int enable); - int (*set_surface_reg)(struct radeon_device *rdev, int reg, - uint32_t tiling_flags, uint32_t pitch, - uint32_t offset, uint32_t obj_size); - void (*clear_surface_reg)(struct radeon_device *rdev, int reg); - void (*bandwidth_update)(struct radeon_device *rdev); - void (*hpd_init)(struct radeon_device *rdev); - void (*hpd_fini)(struct radeon_device *rdev); - bool (*hpd_sense)(struct radeon_device *rdev, enum radeon_hpd_id hpd); - void (*hpd_set_polarity)(struct radeon_device *rdev, enum radeon_hpd_id hpd); /* ioctl hw specific callback. Some hw might want to perform special * operation on specific ioctl. For instance on wait idle some hw * might want to perform and HDP flush through MMIO as it seems that @@ -1190,17 +1137,99 @@ struct radeon_asic { * through ring. */ void (*ioctl_wait_idle)(struct radeon_device *rdev, struct radeon_bo *bo); + /* check if 3D engine is idle */ bool (*gui_idle)(struct radeon_device *rdev); + /* wait for mc_idle */ + int (*mc_wait_for_idle)(struct radeon_device *rdev); + /* gart */ + struct { + void (*tlb_flush)(struct radeon_device *rdev); + int (*set_page)(struct radeon_device *rdev, int i, uint64_t addr); + } gart; + /* ring specific callbacks */ + struct { + void (*ib_execute)(struct radeon_device *rdev, struct radeon_ib *ib); + int (*ib_parse)(struct radeon_device *rdev, struct radeon_ib *ib); + void (*emit_fence)(struct radeon_device *rdev, struct radeon_fence *fence); + void (*emit_semaphore)(struct radeon_device *rdev, struct radeon_ring *cp, + struct radeon_semaphore *semaphore, bool emit_wait); + int (*cs_parse)(struct radeon_cs_parser *p); + void (*ring_start)(struct radeon_device *rdev, struct radeon_ring *cp); + int (*ring_test)(struct radeon_device *rdev, struct radeon_ring *cp); + int (*ib_test)(struct radeon_device *rdev, struct radeon_ring *cp); + } ring[RADEON_NUM_RINGS]; + /* irqs */ + struct { + int (*set)(struct radeon_device *rdev); + int (*process)(struct radeon_device *rdev); + } irq; + /* displays */ + struct { + /* display watermarks */ + void (*bandwidth_update)(struct radeon_device *rdev); + /* get frame count */ + u32 (*get_vblank_counter)(struct radeon_device *rdev, int crtc); + /* wait for vblank */ + void (*wait_for_vblank)(struct radeon_device *rdev, int crtc); + } display; + /* copy functions for bo handling */ + struct { + int (*blit)(struct radeon_device *rdev, + uint64_t src_offset, + uint64_t dst_offset, + unsigned num_gpu_pages, + struct radeon_fence *fence); + u32 blit_ring_index; + int (*dma)(struct radeon_device *rdev, + uint64_t src_offset, + uint64_t dst_offset, + unsigned num_gpu_pages, + struct radeon_fence *fence); + u32 dma_ring_index; + /* method used for bo copy */ + int (*copy)(struct radeon_device *rdev, + uint64_t src_offset, + uint64_t dst_offset, + unsigned num_gpu_pages, + struct radeon_fence *fence); + /* ring used for bo copies */ + u32 copy_ring_index; + } copy; + /* surfaces */ + struct { + int (*set_reg)(struct radeon_device *rdev, int reg, + uint32_t tiling_flags, uint32_t pitch, + uint32_t offset, uint32_t obj_size); + void (*clear_reg)(struct radeon_device *rdev, int reg); + } surface; + /* hotplug detect */ + struct { + void (*init)(struct radeon_device *rdev); + void (*fini)(struct radeon_device *rdev); + bool (*sense)(struct radeon_device *rdev, enum radeon_hpd_id hpd); + void (*set_polarity)(struct radeon_device *rdev, enum radeon_hpd_id hpd); + } hpd; /* power management */ - void (*pm_misc)(struct radeon_device *rdev); - void (*pm_prepare)(struct radeon_device *rdev); - void (*pm_finish)(struct radeon_device *rdev); - void (*pm_init_profile)(struct radeon_device *rdev); - void (*pm_get_dynpm_state)(struct radeon_device *rdev); + struct { + void (*misc)(struct radeon_device *rdev); + void (*prepare)(struct radeon_device *rdev); + void (*finish)(struct radeon_device *rdev); + void (*init_profile)(struct radeon_device *rdev); + void (*get_dynpm_state)(struct radeon_device *rdev); + uint32_t (*get_engine_clock)(struct radeon_device *rdev); + void (*set_engine_clock)(struct radeon_device *rdev, uint32_t eng_clock); + uint32_t (*get_memory_clock)(struct radeon_device *rdev); + void (*set_memory_clock)(struct radeon_device *rdev, uint32_t mem_clock); + int (*get_pcie_lanes)(struct radeon_device *rdev); + void (*set_pcie_lanes)(struct radeon_device *rdev, int lanes); + void (*set_clock_gating)(struct radeon_device *rdev, int enable); + } pm; /* pageflipping */ - void (*pre_page_flip)(struct radeon_device *rdev, int crtc); - u32 (*page_flip)(struct radeon_device *rdev, int crtc, u64 crtc_base); - void (*post_page_flip)(struct radeon_device *rdev, int crtc); + struct { + void (*pre_page_flip)(struct radeon_device *rdev, int crtc); + u32 (*page_flip)(struct radeon_device *rdev, int crtc, u64 crtc_base); + void (*post_page_flip)(struct radeon_device *rdev, int crtc); + } pflip; }; /* @@ -1491,8 +1520,6 @@ struct radeon_device { unsigned debugfs_count; /* virtual memory */ struct radeon_vm_manager vm_manager; - /* ring used for bo copies */ - u32 copy_ring; }; int radeon_device_init(struct radeon_device *rdev, @@ -1648,47 +1675,53 @@ void radeon_ring_write(struct radeon_ring *ring, uint32_t v); #define radeon_fini(rdev) (rdev)->asic->fini((rdev)) #define radeon_resume(rdev) (rdev)->asic->resume((rdev)) #define radeon_suspend(rdev) (rdev)->asic->suspend((rdev)) -#define radeon_cs_parse(p) rdev->asic->cs_parse((p)) +#define radeon_cs_parse(rdev, r, p) (rdev)->asic->ring[(r)].cs_parse((p)) #define radeon_vga_set_state(rdev, state) (rdev)->asic->vga_set_state((rdev), (state)) #define radeon_gpu_is_lockup(rdev, cp) (rdev)->asic->gpu_is_lockup((rdev), (cp)) #define radeon_asic_reset(rdev) (rdev)->asic->asic_reset((rdev)) -#define radeon_gart_tlb_flush(rdev) (rdev)->asic->gart_tlb_flush((rdev)) -#define radeon_gart_set_page(rdev, i, p) (rdev)->asic->gart_set_page((rdev), (i), (p)) -#define radeon_ring_start(rdev) (rdev)->asic->ring_start((rdev)) -#define radeon_ring_test(rdev, cp) (rdev)->asic->ring_test((rdev), (cp)) +#define radeon_gart_tlb_flush(rdev) (rdev)->asic->gart.tlb_flush((rdev)) +#define radeon_gart_set_page(rdev, i, p) (rdev)->asic->gart.set_page((rdev), (i), (p)) +#define radeon_ring_start(rdev, r, cp) (rdev)->asic->ring[(r)].ring_start((rdev), (cp)) +#define radeon_ring_test(rdev, r, cp) (rdev)->asic->ring[(r)].ring_test((rdev), (cp)) +#define radeon_ib_test(rdev, r, cp) (rdev)->asic->ring[(r)].ib_test((rdev), (cp)) #define radeon_ring_ib_execute(rdev, r, ib) (rdev)->asic->ring[(r)].ib_execute((rdev), (ib)) #define radeon_ring_ib_parse(rdev, r, ib) (rdev)->asic->ring[(r)].ib_parse((rdev), (ib)) -#define radeon_irq_set(rdev) (rdev)->asic->irq_set((rdev)) -#define radeon_irq_process(rdev) (rdev)->asic->irq_process((rdev)) -#define radeon_get_vblank_counter(rdev, crtc) (rdev)->asic->get_vblank_counter((rdev), (crtc)) +#define radeon_irq_set(rdev) (rdev)->asic->irq.set((rdev)) +#define radeon_irq_process(rdev) (rdev)->asic->irq.process((rdev)) +#define radeon_get_vblank_counter(rdev, crtc) (rdev)->asic->display.get_vblank_counter((rdev), (crtc)) #define radeon_fence_ring_emit(rdev, r, fence) (rdev)->asic->ring[(r)].emit_fence((rdev), (fence)) #define radeon_semaphore_ring_emit(rdev, r, cp, semaphore, emit_wait) (rdev)->asic->ring[(r)].emit_semaphore((rdev), (cp), (semaphore), (emit_wait)) -#define radeon_copy_blit(rdev, s, d, np, f) (rdev)->asic->copy_blit((rdev), (s), (d), (np), (f)) -#define radeon_copy_dma(rdev, s, d, np, f) (rdev)->asic->copy_dma((rdev), (s), (d), (np), (f)) -#define radeon_copy(rdev, s, d, np, f) (rdev)->asic->copy((rdev), (s), (d), (np), (f)) -#define radeon_get_engine_clock(rdev) (rdev)->asic->get_engine_clock((rdev)) -#define radeon_set_engine_clock(rdev, e) (rdev)->asic->set_engine_clock((rdev), (e)) -#define radeon_get_memory_clock(rdev) (rdev)->asic->get_memory_clock((rdev)) -#define radeon_set_memory_clock(rdev, e) (rdev)->asic->set_memory_clock((rdev), (e)) -#define radeon_get_pcie_lanes(rdev) (rdev)->asic->get_pcie_lanes((rdev)) -#define radeon_set_pcie_lanes(rdev, l) (rdev)->asic->set_pcie_lanes((rdev), (l)) -#define radeon_set_clock_gating(rdev, e) (rdev)->asic->set_clock_gating((rdev), (e)) -#define radeon_set_surface_reg(rdev, r, f, p, o, s) ((rdev)->asic->set_surface_reg((rdev), (r), (f), (p), (o), (s))) -#define radeon_clear_surface_reg(rdev, r) ((rdev)->asic->clear_surface_reg((rdev), (r))) -#define radeon_bandwidth_update(rdev) (rdev)->asic->bandwidth_update((rdev)) -#define radeon_hpd_init(rdev) (rdev)->asic->hpd_init((rdev)) -#define radeon_hpd_fini(rdev) (rdev)->asic->hpd_fini((rdev)) -#define radeon_hpd_sense(rdev, hpd) (rdev)->asic->hpd_sense((rdev), (hpd)) -#define radeon_hpd_set_polarity(rdev, hpd) (rdev)->asic->hpd_set_polarity((rdev), (hpd)) +#define radeon_copy_blit(rdev, s, d, np, f) (rdev)->asic->copy.blit((rdev), (s), (d), (np), (f)) +#define radeon_copy_dma(rdev, s, d, np, f) (rdev)->asic->copy.dma((rdev), (s), (d), (np), (f)) +#define radeon_copy(rdev, s, d, np, f) (rdev)->asic->copy.copy((rdev), (s), (d), (np), (f)) +#define radeon_copy_blit_ring_index(rdev) (rdev)->asic->copy.blit_ring_index +#define radeon_copy_dma_ring_index(rdev) (rdev)->asic->copy.dma_ring_index +#define radeon_copy_ring_index(rdev) (rdev)->asic->copy.copy_ring_index +#define radeon_get_engine_clock(rdev) (rdev)->asic->pm.get_engine_clock((rdev)) +#define radeon_set_engine_clock(rdev, e) (rdev)->asic->pm.set_engine_clock((rdev), (e)) +#define radeon_get_memory_clock(rdev) (rdev)->asic->pm.get_memory_clock((rdev)) +#define radeon_set_memory_clock(rdev, e) (rdev)->asic->pm.set_memory_clock((rdev), (e)) +#define radeon_get_pcie_lanes(rdev) (rdev)->asic->pm.get_pcie_lanes((rdev)) +#define radeon_set_pcie_lanes(rdev, l) (rdev)->asic->pm.set_pcie_lanes((rdev), (l)) +#define radeon_set_clock_gating(rdev, e) (rdev)->asic->pm.set_clock_gating((rdev), (e)) +#define radeon_set_surface_reg(rdev, r, f, p, o, s) ((rdev)->asic->surface.set_reg((rdev), (r), (f), (p), (o), (s))) +#define radeon_clear_surface_reg(rdev, r) ((rdev)->asic->surface.clear_reg((rdev), (r))) +#define radeon_bandwidth_update(rdev) (rdev)->asic->display.bandwidth_update((rdev)) +#define radeon_hpd_init(rdev) (rdev)->asic->hpd.init((rdev)) +#define radeon_hpd_fini(rdev) (rdev)->asic->hpd.fini((rdev)) +#define radeon_hpd_sense(rdev, h) (rdev)->asic->hpd.sense((rdev), (h)) +#define radeon_hpd_set_polarity(rdev, h) (rdev)->asic->hpd.set_polarity((rdev), (h)) #define radeon_gui_idle(rdev) (rdev)->asic->gui_idle((rdev)) -#define radeon_pm_misc(rdev) (rdev)->asic->pm_misc((rdev)) -#define radeon_pm_prepare(rdev) (rdev)->asic->pm_prepare((rdev)) -#define radeon_pm_finish(rdev) (rdev)->asic->pm_finish((rdev)) -#define radeon_pm_init_profile(rdev) (rdev)->asic->pm_init_profile((rdev)) -#define radeon_pm_get_dynpm_state(rdev) (rdev)->asic->pm_get_dynpm_state((rdev)) -#define radeon_pre_page_flip(rdev, crtc) rdev->asic->pre_page_flip((rdev), (crtc)) -#define radeon_page_flip(rdev, crtc, base) rdev->asic->page_flip((rdev), (crtc), (base)) -#define radeon_post_page_flip(rdev, crtc) rdev->asic->post_page_flip((rdev), (crtc)) +#define radeon_pm_misc(rdev) (rdev)->asic->pm.misc((rdev)) +#define radeon_pm_prepare(rdev) (rdev)->asic->pm.prepare((rdev)) +#define radeon_pm_finish(rdev) (rdev)->asic->pm.finish((rdev)) +#define radeon_pm_init_profile(rdev) (rdev)->asic->pm.init_profile((rdev)) +#define radeon_pm_get_dynpm_state(rdev) (rdev)->asic->pm.get_dynpm_state((rdev)) +#define radeon_pre_page_flip(rdev, crtc) rdev->asic->pflip.pre_page_flip((rdev), (crtc)) +#define radeon_page_flip(rdev, crtc, base) rdev->asic->pflip.page_flip((rdev), (crtc), (base)) +#define radeon_post_page_flip(rdev, crtc) rdev->asic->pflip.post_page_flip((rdev), (crtc)) +#define radeon_wait_for_vblank(rdev, crtc) rdev->asic->display.wait_for_vblank((rdev), (crtc)) +#define radeon_mc_wait_for_idle(rdev) rdev->asic->mc_wait_for_idle((rdev)) /* Common functions */ /* AGP */ @@ -1750,6 +1783,16 @@ int r600_vram_scratch_init(struct radeon_device *rdev); void r600_vram_scratch_fini(struct radeon_device *rdev); /* + * r600 cs checking helper + */ +unsigned r600_mip_minify(unsigned size, unsigned level); +bool r600_fmt_is_valid_color(u32 format); +bool r600_fmt_is_valid_texture(u32 format, enum radeon_family family); +int r600_fmt_get_blocksize(u32 format); +int r600_fmt_get_nblocksx(u32 format, u32 w); +int r600_fmt_get_nblocksy(u32 format, u32 h); + +/* * r600 functions used by radeon_encoder.c */ extern void r600_hdmi_enable(struct drm_encoder *encoder); diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c index 36a6192..479c89e 100644 --- a/drivers/gpu/drm/radeon/radeon_asic.c +++ b/drivers/gpu/drm/radeon/radeon_asic.c @@ -114,13 +114,13 @@ void radeon_agp_disable(struct radeon_device *rdev) rdev->family == CHIP_R423) { DRM_INFO("Forcing AGP to PCIE mode\n"); rdev->flags |= RADEON_IS_PCIE; - rdev->asic->gart_tlb_flush = &rv370_pcie_gart_tlb_flush; - rdev->asic->gart_set_page = &rv370_pcie_gart_set_page; + rdev->asic->gart.tlb_flush = &rv370_pcie_gart_tlb_flush; + rdev->asic->gart.set_page = &rv370_pcie_gart_set_page; } else { DRM_INFO("Forcing AGP to PCI mode\n"); rdev->flags |= RADEON_IS_PCI; - rdev->asic->gart_tlb_flush = &r100_pci_gart_tlb_flush; - rdev->asic->gart_set_page = &r100_pci_gart_set_page; + rdev->asic->gart.tlb_flush = &r100_pci_gart_tlb_flush; + rdev->asic->gart.set_page = &r100_pci_gart_set_page; } rdev->mc.gtt_size = radeon_gart_size * 1024 * 1024; } @@ -136,48 +136,70 @@ static struct radeon_asic r100_asic = { .vga_set_state = &r100_vga_set_state, .gpu_is_lockup = &r100_gpu_is_lockup, .asic_reset = &r100_asic_reset, - .gart_tlb_flush = &r100_pci_gart_tlb_flush, - .gart_set_page = &r100_pci_gart_set_page, - .ring_start = &r100_ring_start, - .ring_test = &r100_ring_test, + .ioctl_wait_idle = NULL, + .gui_idle = &r100_gui_idle, + .mc_wait_for_idle = &r100_mc_wait_for_idle, + .gart = { + .tlb_flush = &r100_pci_gart_tlb_flush, + .set_page = &r100_pci_gart_set_page, + }, .ring = { [RADEON_RING_TYPE_GFX_INDEX] = { .ib_execute = &r100_ring_ib_execute, .emit_fence = &r100_fence_ring_emit, .emit_semaphore = &r100_semaphore_ring_emit, + .cs_parse = &r100_cs_parse, + .ring_start = &r100_ring_start, + .ring_test = &r100_ring_test, + .ib_test = &r100_ib_test, } }, - .irq_set = &r100_irq_set, - .irq_process = &r100_irq_process, - .get_vblank_counter = &r100_get_vblank_counter, - .cs_parse = &r100_cs_parse, - .copy_blit = &r100_copy_blit, - .copy_dma = NULL, - .copy = &r100_copy_blit, - .get_engine_clock = &radeon_legacy_get_engine_clock, - .set_engine_clock = &radeon_legacy_set_engine_clock, - .get_memory_clock = &radeon_legacy_get_memory_clock, - .set_memory_clock = NULL, - .get_pcie_lanes = NULL, - .set_pcie_lanes = NULL, - .set_clock_gating = &radeon_legacy_set_clock_gating, - .set_surface_reg = r100_set_surface_reg, - .clear_surface_reg = r100_clear_surface_reg, - .bandwidth_update = &r100_bandwidth_update, - .hpd_init = &r100_hpd_init, - .hpd_fini = &r100_hpd_fini, - .hpd_sense = &r100_hpd_sense, - .hpd_set_polarity = &r100_hpd_set_polarity, - .ioctl_wait_idle = NULL, - .gui_idle = &r100_gui_idle, - .pm_misc = &r100_pm_misc, - .pm_prepare = &r100_pm_prepare, - .pm_finish = &r100_pm_finish, - .pm_init_profile = &r100_pm_init_profile, - .pm_get_dynpm_state = &r100_pm_get_dynpm_state, - .pre_page_flip = &r100_pre_page_flip, - .page_flip = &r100_page_flip, - .post_page_flip = &r100_post_page_flip, + .irq = { + .set = &r100_irq_set, + .process = &r100_irq_process, + }, + .display = { + .bandwidth_update = &r100_bandwidth_update, + .get_vblank_counter = &r100_get_vblank_counter, + .wait_for_vblank = &r100_wait_for_vblank, + }, + .copy = { + .blit = &r100_copy_blit, + .blit_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .dma = NULL, + .dma_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .copy = &r100_copy_blit, + .copy_ring_index = RADEON_RING_TYPE_GFX_INDEX, + }, + .surface = { + .set_reg = r100_set_surface_reg, + .clear_reg = r100_clear_surface_reg, + }, + .hpd = { + .init = &r100_hpd_init, + .fini = &r100_hpd_fini, + .sense = &r100_hpd_sense, + .set_polarity = &r100_hpd_set_polarity, + }, + .pm = { + .misc = &r100_pm_misc, + .prepare = &r100_pm_prepare, + .finish = &r100_pm_finish, + .init_profile = &r100_pm_init_profile, + .get_dynpm_state = &r100_pm_get_dynpm_state, + .get_engine_clock = &radeon_legacy_get_engine_clock, + .set_engine_clock = &radeon_legacy_set_engine_clock, + .get_memory_clock = &radeon_legacy_get_memory_clock, + .set_memory_clock = NULL, + .get_pcie_lanes = NULL, + .set_pcie_lanes = NULL, + .set_clock_gating = &radeon_legacy_set_clock_gating, + }, + .pflip = { + .pre_page_flip = &r100_pre_page_flip, + .page_flip = &r100_page_flip, + .post_page_flip = &r100_post_page_flip, + }, }; static struct radeon_asic r200_asic = { @@ -188,47 +210,70 @@ static struct radeon_asic r200_asic = { .vga_set_state = &r100_vga_set_state, .gpu_is_lockup = &r100_gpu_is_lockup, .asic_reset = &r100_asic_reset, - .gart_tlb_flush = &r100_pci_gart_tlb_flush, - .gart_set_page = &r100_pci_gart_set_page, - .ring_start = &r100_ring_start, - .ring_test = &r100_ring_test, + .ioctl_wait_idle = NULL, + .gui_idle = &r100_gui_idle, + .mc_wait_for_idle = &r100_mc_wait_for_idle, + .gart = { + .tlb_flush = &r100_pci_gart_tlb_flush, + .set_page = &r100_pci_gart_set_page, + }, .ring = { [RADEON_RING_TYPE_GFX_INDEX] = { .ib_execute = &r100_ring_ib_execute, .emit_fence = &r100_fence_ring_emit, .emit_semaphore = &r100_semaphore_ring_emit, + .cs_parse = &r100_cs_parse, + .ring_start = &r100_ring_start, + .ring_test = &r100_ring_test, + .ib_test = &r100_ib_test, } }, - .irq_set = &r100_irq_set, - .irq_process = &r100_irq_process, - .get_vblank_counter = &r100_get_vblank_counter, - .cs_parse = &r100_cs_parse, - .copy_blit = &r100_copy_blit, - .copy_dma = &r200_copy_dma, - .copy = &r100_copy_blit, - .get_engine_clock = &radeon_legacy_get_engine_clock, - .set_engine_clock = &radeon_legacy_set_engine_clock, - .get_memory_clock = &radeon_legacy_get_memory_clock, - .set_memory_clock = NULL, - .set_pcie_lanes = NULL, - .set_clock_gating = &radeon_legacy_set_clock_gating, - .set_surface_reg = r100_set_surface_reg, - .clear_surface_reg = r100_clear_surface_reg, - .bandwidth_update = &r100_bandwidth_update, - .hpd_init = &r100_hpd_init, - .hpd_fini = &r100_hpd_fini, - .hpd_sense = &r100_hpd_sense, - .hpd_set_polarity = &r100_hpd_set_polarity, - .ioctl_wait_idle = NULL, - .gui_idle = &r100_gui_idle, - .pm_misc = &r100_pm_misc, - .pm_prepare = &r100_pm_prepare, - .pm_finish = &r100_pm_finish, - .pm_init_profile = &r100_pm_init_profile, - .pm_get_dynpm_state = &r100_pm_get_dynpm_state, - .pre_page_flip = &r100_pre_page_flip, - .page_flip = &r100_page_flip, - .post_page_flip = &r100_post_page_flip, + .irq = { + .set = &r100_irq_set, + .process = &r100_irq_process, + }, + .display = { + .bandwidth_update = &r100_bandwidth_update, + .get_vblank_counter = &r100_get_vblank_counter, + .wait_for_vblank = &r100_wait_for_vblank, + }, + .copy = { + .blit = &r100_copy_blit, + .blit_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .dma = &r200_copy_dma, + .dma_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .copy = &r100_copy_blit, + .copy_ring_index = RADEON_RING_TYPE_GFX_INDEX, + }, + .surface = { + .set_reg = r100_set_surface_reg, + .clear_reg = r100_clear_surface_reg, + }, + .hpd = { + .init = &r100_hpd_init, + .fini = &r100_hpd_fini, + .sense = &r100_hpd_sense, + .set_polarity = &r100_hpd_set_polarity, + }, + .pm = { + .misc = &r100_pm_misc, + .prepare = &r100_pm_prepare, + .finish = &r100_pm_finish, + .init_profile = &r100_pm_init_profile, + .get_dynpm_state = &r100_pm_get_dynpm_state, + .get_engine_clock = &radeon_legacy_get_engine_clock, + .set_engine_clock = &radeon_legacy_set_engine_clock, + .get_memory_clock = &radeon_legacy_get_memory_clock, + .set_memory_clock = NULL, + .get_pcie_lanes = NULL, + .set_pcie_lanes = NULL, + .set_clock_gating = &radeon_legacy_set_clock_gating, + }, + .pflip = { + .pre_page_flip = &r100_pre_page_flip, + .page_flip = &r100_page_flip, + .post_page_flip = &r100_post_page_flip, + }, }; static struct radeon_asic r300_asic = { @@ -239,48 +284,70 @@ static struct radeon_asic r300_asic = { .vga_set_state = &r100_vga_set_state, .gpu_is_lockup = &r300_gpu_is_lockup, .asic_reset = &r300_asic_reset, - .gart_tlb_flush = &r100_pci_gart_tlb_flush, - .gart_set_page = &r100_pci_gart_set_page, - .ring_start = &r300_ring_start, - .ring_test = &r100_ring_test, + .ioctl_wait_idle = NULL, + .gui_idle = &r100_gui_idle, + .mc_wait_for_idle = &r300_mc_wait_for_idle, + .gart = { + .tlb_flush = &r100_pci_gart_tlb_flush, + .set_page = &r100_pci_gart_set_page, + }, .ring = { [RADEON_RING_TYPE_GFX_INDEX] = { .ib_execute = &r100_ring_ib_execute, .emit_fence = &r300_fence_ring_emit, .emit_semaphore = &r100_semaphore_ring_emit, + .cs_parse = &r300_cs_parse, + .ring_start = &r300_ring_start, + .ring_test = &r100_ring_test, + .ib_test = &r100_ib_test, } }, - .irq_set = &r100_irq_set, - .irq_process = &r100_irq_process, - .get_vblank_counter = &r100_get_vblank_counter, - .cs_parse = &r300_cs_parse, - .copy_blit = &r100_copy_blit, - .copy_dma = &r200_copy_dma, - .copy = &r100_copy_blit, - .get_engine_clock = &radeon_legacy_get_engine_clock, - .set_engine_clock = &radeon_legacy_set_engine_clock, - .get_memory_clock = &radeon_legacy_get_memory_clock, - .set_memory_clock = NULL, - .get_pcie_lanes = &rv370_get_pcie_lanes, - .set_pcie_lanes = &rv370_set_pcie_lanes, - .set_clock_gating = &radeon_legacy_set_clock_gating, - .set_surface_reg = r100_set_surface_reg, - .clear_surface_reg = r100_clear_surface_reg, - .bandwidth_update = &r100_bandwidth_update, - .hpd_init = &r100_hpd_init, - .hpd_fini = &r100_hpd_fini, - .hpd_sense = &r100_hpd_sense, - .hpd_set_polarity = &r100_hpd_set_polarity, - .ioctl_wait_idle = NULL, - .gui_idle = &r100_gui_idle, - .pm_misc = &r100_pm_misc, - .pm_prepare = &r100_pm_prepare, - .pm_finish = &r100_pm_finish, - .pm_init_profile = &r100_pm_init_profile, - .pm_get_dynpm_state = &r100_pm_get_dynpm_state, - .pre_page_flip = &r100_pre_page_flip, - .page_flip = &r100_page_flip, - .post_page_flip = &r100_post_page_flip, + .irq = { + .set = &r100_irq_set, + .process = &r100_irq_process, + }, + .display = { + .bandwidth_update = &r100_bandwidth_update, + .get_vblank_counter = &r100_get_vblank_counter, + .wait_for_vblank = &r100_wait_for_vblank, + }, + .copy = { + .blit = &r100_copy_blit, + .blit_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .dma = &r200_copy_dma, + .dma_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .copy = &r100_copy_blit, + .copy_ring_index = RADEON_RING_TYPE_GFX_INDEX, + }, + .surface = { + .set_reg = r100_set_surface_reg, + .clear_reg = r100_clear_surface_reg, + }, + .hpd = { + .init = &r100_hpd_init, + .fini = &r100_hpd_fini, + .sense = &r100_hpd_sense, + .set_polarity = &r100_hpd_set_polarity, + }, + .pm = { + .misc = &r100_pm_misc, + .prepare = &r100_pm_prepare, + .finish = &r100_pm_finish, + .init_profile = &r100_pm_init_profile, + .get_dynpm_state = &r100_pm_get_dynpm_state, + .get_engine_clock = &radeon_legacy_get_engine_clock, + .set_engine_clock = &radeon_legacy_set_engine_clock, + .get_memory_clock = &radeon_legacy_get_memory_clock, + .set_memory_clock = NULL, + .get_pcie_lanes = &rv370_get_pcie_lanes, + .set_pcie_lanes = &rv370_set_pcie_lanes, + .set_clock_gating = &radeon_legacy_set_clock_gating, + }, + .pflip = { + .pre_page_flip = &r100_pre_page_flip, + .page_flip = &r100_page_flip, + .post_page_flip = &r100_post_page_flip, + }, }; static struct radeon_asic r300_asic_pcie = { @@ -291,47 +358,70 @@ static struct radeon_asic r300_asic_pcie = { .vga_set_state = &r100_vga_set_state, .gpu_is_lockup = &r300_gpu_is_lockup, .asic_reset = &r300_asic_reset, - .gart_tlb_flush = &rv370_pcie_gart_tlb_flush, - .gart_set_page = &rv370_pcie_gart_set_page, - .ring_start = &r300_ring_start, - .ring_test = &r100_ring_test, + .ioctl_wait_idle = NULL, + .gui_idle = &r100_gui_idle, + .mc_wait_for_idle = &r300_mc_wait_for_idle, + .gart = { + .tlb_flush = &rv370_pcie_gart_tlb_flush, + .set_page = &rv370_pcie_gart_set_page, + }, .ring = { [RADEON_RING_TYPE_GFX_INDEX] = { .ib_execute = &r100_ring_ib_execute, .emit_fence = &r300_fence_ring_emit, .emit_semaphore = &r100_semaphore_ring_emit, + .cs_parse = &r300_cs_parse, + .ring_start = &r300_ring_start, + .ring_test = &r100_ring_test, + .ib_test = &r100_ib_test, } }, - .irq_set = &r100_irq_set, - .irq_process = &r100_irq_process, - .get_vblank_counter = &r100_get_vblank_counter, - .cs_parse = &r300_cs_parse, - .copy_blit = &r100_copy_blit, - .copy_dma = &r200_copy_dma, - .copy = &r100_copy_blit, - .get_engine_clock = &radeon_legacy_get_engine_clock, - .set_engine_clock = &radeon_legacy_set_engine_clock, - .get_memory_clock = &radeon_legacy_get_memory_clock, - .set_memory_clock = NULL, - .set_pcie_lanes = &rv370_set_pcie_lanes, - .set_clock_gating = &radeon_legacy_set_clock_gating, - .set_surface_reg = r100_set_surface_reg, - .clear_surface_reg = r100_clear_surface_reg, - .bandwidth_update = &r100_bandwidth_update, - .hpd_init = &r100_hpd_init, - .hpd_fini = &r100_hpd_fini, - .hpd_sense = &r100_hpd_sense, - .hpd_set_polarity = &r100_hpd_set_polarity, - .ioctl_wait_idle = NULL, - .gui_idle = &r100_gui_idle, - .pm_misc = &r100_pm_misc, - .pm_prepare = &r100_pm_prepare, - .pm_finish = &r100_pm_finish, - .pm_init_profile = &r100_pm_init_profile, - .pm_get_dynpm_state = &r100_pm_get_dynpm_state, - .pre_page_flip = &r100_pre_page_flip, - .page_flip = &r100_page_flip, - .post_page_flip = &r100_post_page_flip, + .irq = { + .set = &r100_irq_set, + .process = &r100_irq_process, + }, + .display = { + .bandwidth_update = &r100_bandwidth_update, + .get_vblank_counter = &r100_get_vblank_counter, + .wait_for_vblank = &r100_wait_for_vblank, + }, + .copy = { + .blit = &r100_copy_blit, + .blit_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .dma = &r200_copy_dma, + .dma_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .copy = &r100_copy_blit, + .copy_ring_index = RADEON_RING_TYPE_GFX_INDEX, + }, + .surface = { + .set_reg = r100_set_surface_reg, + .clear_reg = r100_clear_surface_reg, + }, + .hpd = { + .init = &r100_hpd_init, + .fini = &r100_hpd_fini, + .sense = &r100_hpd_sense, + .set_polarity = &r100_hpd_set_polarity, + }, + .pm = { + .misc = &r100_pm_misc, + .prepare = &r100_pm_prepare, + .finish = &r100_pm_finish, + .init_profile = &r100_pm_init_profile, + .get_dynpm_state = &r100_pm_get_dynpm_state, + .get_engine_clock = &radeon_legacy_get_engine_clock, + .set_engine_clock = &radeon_legacy_set_engine_clock, + .get_memory_clock = &radeon_legacy_get_memory_clock, + .set_memory_clock = NULL, + .get_pcie_lanes = &rv370_get_pcie_lanes, + .set_pcie_lanes = &rv370_set_pcie_lanes, + .set_clock_gating = &radeon_legacy_set_clock_gating, + }, + .pflip = { + .pre_page_flip = &r100_pre_page_flip, + .page_flip = &r100_page_flip, + .post_page_flip = &r100_post_page_flip, + }, }; static struct radeon_asic r420_asic = { @@ -342,48 +432,70 @@ static struct radeon_asic r420_asic = { .vga_set_state = &r100_vga_set_state, .gpu_is_lockup = &r300_gpu_is_lockup, .asic_reset = &r300_asic_reset, - .gart_tlb_flush = &rv370_pcie_gart_tlb_flush, - .gart_set_page = &rv370_pcie_gart_set_page, - .ring_start = &r300_ring_start, - .ring_test = &r100_ring_test, + .ioctl_wait_idle = NULL, + .gui_idle = &r100_gui_idle, + .mc_wait_for_idle = &r300_mc_wait_for_idle, + .gart = { + .tlb_flush = &rv370_pcie_gart_tlb_flush, + .set_page = &rv370_pcie_gart_set_page, + }, .ring = { [RADEON_RING_TYPE_GFX_INDEX] = { .ib_execute = &r100_ring_ib_execute, .emit_fence = &r300_fence_ring_emit, .emit_semaphore = &r100_semaphore_ring_emit, + .cs_parse = &r300_cs_parse, + .ring_start = &r300_ring_start, + .ring_test = &r100_ring_test, + .ib_test = &r100_ib_test, } }, - .irq_set = &r100_irq_set, - .irq_process = &r100_irq_process, - .get_vblank_counter = &r100_get_vblank_counter, - .cs_parse = &r300_cs_parse, - .copy_blit = &r100_copy_blit, - .copy_dma = &r200_copy_dma, - .copy = &r100_copy_blit, - .get_engine_clock = &radeon_atom_get_engine_clock, - .set_engine_clock = &radeon_atom_set_engine_clock, - .get_memory_clock = &radeon_atom_get_memory_clock, - .set_memory_clock = &radeon_atom_set_memory_clock, - .get_pcie_lanes = &rv370_get_pcie_lanes, - .set_pcie_lanes = &rv370_set_pcie_lanes, - .set_clock_gating = &radeon_atom_set_clock_gating, - .set_surface_reg = r100_set_surface_reg, - .clear_surface_reg = r100_clear_surface_reg, - .bandwidth_update = &r100_bandwidth_update, - .hpd_init = &r100_hpd_init, - .hpd_fini = &r100_hpd_fini, - .hpd_sense = &r100_hpd_sense, - .hpd_set_polarity = &r100_hpd_set_polarity, - .ioctl_wait_idle = NULL, - .gui_idle = &r100_gui_idle, - .pm_misc = &r100_pm_misc, - .pm_prepare = &r100_pm_prepare, - .pm_finish = &r100_pm_finish, - .pm_init_profile = &r420_pm_init_profile, - .pm_get_dynpm_state = &r100_pm_get_dynpm_state, - .pre_page_flip = &r100_pre_page_flip, - .page_flip = &r100_page_flip, - .post_page_flip = &r100_post_page_flip, + .irq = { + .set = &r100_irq_set, + .process = &r100_irq_process, + }, + .display = { + .bandwidth_update = &r100_bandwidth_update, + .get_vblank_counter = &r100_get_vblank_counter, + .wait_for_vblank = &r100_wait_for_vblank, + }, + .copy = { + .blit = &r100_copy_blit, + .blit_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .dma = &r200_copy_dma, + .dma_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .copy = &r100_copy_blit, + .copy_ring_index = RADEON_RING_TYPE_GFX_INDEX, + }, + .surface = { + .set_reg = r100_set_surface_reg, + .clear_reg = r100_clear_surface_reg, + }, + .hpd = { + .init = &r100_hpd_init, + .fini = &r100_hpd_fini, + .sense = &r100_hpd_sense, + .set_polarity = &r100_hpd_set_polarity, + }, + .pm = { + .misc = &r100_pm_misc, + .prepare = &r100_pm_prepare, + .finish = &r100_pm_finish, + .init_profile = &r420_pm_init_profile, + .get_dynpm_state = &r100_pm_get_dynpm_state, + .get_engine_clock = &radeon_atom_get_engine_clock, + .set_engine_clock = &radeon_atom_set_engine_clock, + .get_memory_clock = &radeon_atom_get_memory_clock, + .set_memory_clock = &radeon_atom_set_memory_clock, + .get_pcie_lanes = &rv370_get_pcie_lanes, + .set_pcie_lanes = &rv370_set_pcie_lanes, + .set_clock_gating = &radeon_atom_set_clock_gating, + }, + .pflip = { + .pre_page_flip = &r100_pre_page_flip, + .page_flip = &r100_page_flip, + .post_page_flip = &r100_post_page_flip, + }, }; static struct radeon_asic rs400_asic = { @@ -394,48 +506,70 @@ static struct radeon_asic rs400_asic = { .vga_set_state = &r100_vga_set_state, .gpu_is_lockup = &r300_gpu_is_lockup, .asic_reset = &r300_asic_reset, - .gart_tlb_flush = &rs400_gart_tlb_flush, - .gart_set_page = &rs400_gart_set_page, - .ring_start = &r300_ring_start, - .ring_test = &r100_ring_test, + .ioctl_wait_idle = NULL, + .gui_idle = &r100_gui_idle, + .mc_wait_for_idle = &rs400_mc_wait_for_idle, + .gart = { + .tlb_flush = &rs400_gart_tlb_flush, + .set_page = &rs400_gart_set_page, + }, .ring = { [RADEON_RING_TYPE_GFX_INDEX] = { .ib_execute = &r100_ring_ib_execute, .emit_fence = &r300_fence_ring_emit, .emit_semaphore = &r100_semaphore_ring_emit, + .cs_parse = &r300_cs_parse, + .ring_start = &r300_ring_start, + .ring_test = &r100_ring_test, + .ib_test = &r100_ib_test, } }, - .irq_set = &r100_irq_set, - .irq_process = &r100_irq_process, - .get_vblank_counter = &r100_get_vblank_counter, - .cs_parse = &r300_cs_parse, - .copy_blit = &r100_copy_blit, - .copy_dma = &r200_copy_dma, - .copy = &r100_copy_blit, - .get_engine_clock = &radeon_legacy_get_engine_clock, - .set_engine_clock = &radeon_legacy_set_engine_clock, - .get_memory_clock = &radeon_legacy_get_memory_clock, - .set_memory_clock = NULL, - .get_pcie_lanes = NULL, - .set_pcie_lanes = NULL, - .set_clock_gating = &radeon_legacy_set_clock_gating, - .set_surface_reg = r100_set_surface_reg, - .clear_surface_reg = r100_clear_surface_reg, - .bandwidth_update = &r100_bandwidth_update, - .hpd_init = &r100_hpd_init, - .hpd_fini = &r100_hpd_fini, - .hpd_sense = &r100_hpd_sense, - .hpd_set_polarity = &r100_hpd_set_polarity, - .ioctl_wait_idle = NULL, - .gui_idle = &r100_gui_idle, - .pm_misc = &r100_pm_misc, - .pm_prepare = &r100_pm_prepare, - .pm_finish = &r100_pm_finish, - .pm_init_profile = &r100_pm_init_profile, - .pm_get_dynpm_state = &r100_pm_get_dynpm_state, - .pre_page_flip = &r100_pre_page_flip, - .page_flip = &r100_page_flip, - .post_page_flip = &r100_post_page_flip, + .irq = { + .set = &r100_irq_set, + .process = &r100_irq_process, + }, + .display = { + .bandwidth_update = &r100_bandwidth_update, + .get_vblank_counter = &r100_get_vblank_counter, + .wait_for_vblank = &r100_wait_for_vblank, + }, + .copy = { + .blit = &r100_copy_blit, + .blit_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .dma = &r200_copy_dma, + .dma_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .copy = &r100_copy_blit, + .copy_ring_index = RADEON_RING_TYPE_GFX_INDEX, + }, + .surface = { + .set_reg = r100_set_surface_reg, + .clear_reg = r100_clear_surface_reg, + }, + .hpd = { + .init = &r100_hpd_init, + .fini = &r100_hpd_fini, + .sense = &r100_hpd_sense, + .set_polarity = &r100_hpd_set_polarity, + }, + .pm = { + .misc = &r100_pm_misc, + .prepare = &r100_pm_prepare, + .finish = &r100_pm_finish, + .init_profile = &r100_pm_init_profile, + .get_dynpm_state = &r100_pm_get_dynpm_state, + .get_engine_clock = &radeon_legacy_get_engine_clock, + .set_engine_clock = &radeon_legacy_set_engine_clock, + .get_memory_clock = &radeon_legacy_get_memory_clock, + .set_memory_clock = NULL, + .get_pcie_lanes = NULL, + .set_pcie_lanes = NULL, + .set_clock_gating = &radeon_legacy_set_clock_gating, + }, + .pflip = { + .pre_page_flip = &r100_pre_page_flip, + .page_flip = &r100_page_flip, + .post_page_flip = &r100_post_page_flip, + }, }; static struct radeon_asic rs600_asic = { @@ -446,48 +580,70 @@ static struct radeon_asic rs600_asic = { .vga_set_state = &r100_vga_set_state, .gpu_is_lockup = &r300_gpu_is_lockup, .asic_reset = &rs600_asic_reset, - .gart_tlb_flush = &rs600_gart_tlb_flush, - .gart_set_page = &rs600_gart_set_page, - .ring_start = &r300_ring_start, - .ring_test = &r100_ring_test, + .ioctl_wait_idle = NULL, + .gui_idle = &r100_gui_idle, + .mc_wait_for_idle = &rs600_mc_wait_for_idle, + .gart = { + .tlb_flush = &rs600_gart_tlb_flush, + .set_page = &rs600_gart_set_page, + }, .ring = { [RADEON_RING_TYPE_GFX_INDEX] = { .ib_execute = &r100_ring_ib_execute, .emit_fence = &r300_fence_ring_emit, .emit_semaphore = &r100_semaphore_ring_emit, + .cs_parse = &r300_cs_parse, + .ring_start = &r300_ring_start, + .ring_test = &r100_ring_test, + .ib_test = &r100_ib_test, } }, - .irq_set = &rs600_irq_set, - .irq_process = &rs600_irq_process, - .get_vblank_counter = &rs600_get_vblank_counter, - .cs_parse = &r300_cs_parse, - .copy_blit = &r100_copy_blit, - .copy_dma = &r200_copy_dma, - .copy = &r100_copy_blit, - .get_engine_clock = &radeon_atom_get_engine_clock, - .set_engine_clock = &radeon_atom_set_engine_clock, - .get_memory_clock = &radeon_atom_get_memory_clock, - .set_memory_clock = &radeon_atom_set_memory_clock, - .get_pcie_lanes = NULL, - .set_pcie_lanes = NULL, - .set_clock_gating = &radeon_atom_set_clock_gating, - .set_surface_reg = r100_set_surface_reg, - .clear_surface_reg = r100_clear_surface_reg, - .bandwidth_update = &rs600_bandwidth_update, - .hpd_init = &rs600_hpd_init, - .hpd_fini = &rs600_hpd_fini, - .hpd_sense = &rs600_hpd_sense, - .hpd_set_polarity = &rs600_hpd_set_polarity, - .ioctl_wait_idle = NULL, - .gui_idle = &r100_gui_idle, - .pm_misc = &rs600_pm_misc, - .pm_prepare = &rs600_pm_prepare, - .pm_finish = &rs600_pm_finish, - .pm_init_profile = &r420_pm_init_profile, - .pm_get_dynpm_state = &r100_pm_get_dynpm_state, - .pre_page_flip = &rs600_pre_page_flip, - .page_flip = &rs600_page_flip, - .post_page_flip = &rs600_post_page_flip, + .irq = { + .set = &rs600_irq_set, + .process = &rs600_irq_process, + }, + .display = { + .bandwidth_update = &rs600_bandwidth_update, + .get_vblank_counter = &rs600_get_vblank_counter, + .wait_for_vblank = &avivo_wait_for_vblank, + }, + .copy = { + .blit = &r100_copy_blit, + .blit_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .dma = &r200_copy_dma, + .dma_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .copy = &r100_copy_blit, + .copy_ring_index = RADEON_RING_TYPE_GFX_INDEX, + }, + .surface = { + .set_reg = r100_set_surface_reg, + .clear_reg = r100_clear_surface_reg, + }, + .hpd = { + .init = &rs600_hpd_init, + .fini = &rs600_hpd_fini, + .sense = &rs600_hpd_sense, + .set_polarity = &rs600_hpd_set_polarity, + }, + .pm = { + .misc = &rs600_pm_misc, + .prepare = &rs600_pm_prepare, + .finish = &rs600_pm_finish, + .init_profile = &r420_pm_init_profile, + .get_dynpm_state = &r100_pm_get_dynpm_state, + .get_engine_clock = &radeon_atom_get_engine_clock, + .set_engine_clock = &radeon_atom_set_engine_clock, + .get_memory_clock = &radeon_atom_get_memory_clock, + .set_memory_clock = &radeon_atom_set_memory_clock, + .get_pcie_lanes = NULL, + .set_pcie_lanes = NULL, + .set_clock_gating = &radeon_atom_set_clock_gating, + }, + .pflip = { + .pre_page_flip = &rs600_pre_page_flip, + .page_flip = &rs600_page_flip, + .post_page_flip = &rs600_post_page_flip, + }, }; static struct radeon_asic rs690_asic = { @@ -498,48 +654,70 @@ static struct radeon_asic rs690_asic = { .vga_set_state = &r100_vga_set_state, .gpu_is_lockup = &r300_gpu_is_lockup, .asic_reset = &rs600_asic_reset, - .gart_tlb_flush = &rs400_gart_tlb_flush, - .gart_set_page = &rs400_gart_set_page, - .ring_start = &r300_ring_start, - .ring_test = &r100_ring_test, + .ioctl_wait_idle = NULL, + .gui_idle = &r100_gui_idle, + .mc_wait_for_idle = &rs690_mc_wait_for_idle, + .gart = { + .tlb_flush = &rs400_gart_tlb_flush, + .set_page = &rs400_gart_set_page, + }, .ring = { [RADEON_RING_TYPE_GFX_INDEX] = { .ib_execute = &r100_ring_ib_execute, .emit_fence = &r300_fence_ring_emit, .emit_semaphore = &r100_semaphore_ring_emit, + .cs_parse = &r300_cs_parse, + .ring_start = &r300_ring_start, + .ring_test = &r100_ring_test, + .ib_test = &r100_ib_test, } }, - .irq_set = &rs600_irq_set, - .irq_process = &rs600_irq_process, - .get_vblank_counter = &rs600_get_vblank_counter, - .cs_parse = &r300_cs_parse, - .copy_blit = &r100_copy_blit, - .copy_dma = &r200_copy_dma, - .copy = &r200_copy_dma, - .get_engine_clock = &radeon_atom_get_engine_clock, - .set_engine_clock = &radeon_atom_set_engine_clock, - .get_memory_clock = &radeon_atom_get_memory_clock, - .set_memory_clock = &radeon_atom_set_memory_clock, - .get_pcie_lanes = NULL, - .set_pcie_lanes = NULL, - .set_clock_gating = &radeon_atom_set_clock_gating, - .set_surface_reg = r100_set_surface_reg, - .clear_surface_reg = r100_clear_surface_reg, - .bandwidth_update = &rs690_bandwidth_update, - .hpd_init = &rs600_hpd_init, - .hpd_fini = &rs600_hpd_fini, - .hpd_sense = &rs600_hpd_sense, - .hpd_set_polarity = &rs600_hpd_set_polarity, - .ioctl_wait_idle = NULL, - .gui_idle = &r100_gui_idle, - .pm_misc = &rs600_pm_misc, - .pm_prepare = &rs600_pm_prepare, - .pm_finish = &rs600_pm_finish, - .pm_init_profile = &r420_pm_init_profile, - .pm_get_dynpm_state = &r100_pm_get_dynpm_state, - .pre_page_flip = &rs600_pre_page_flip, - .page_flip = &rs600_page_flip, - .post_page_flip = &rs600_post_page_flip, + .irq = { + .set = &rs600_irq_set, + .process = &rs600_irq_process, + }, + .display = { + .get_vblank_counter = &rs600_get_vblank_counter, + .bandwidth_update = &rs690_bandwidth_update, + .wait_for_vblank = &avivo_wait_for_vblank, + }, + .copy = { + .blit = &r100_copy_blit, + .blit_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .dma = &r200_copy_dma, + .dma_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .copy = &r200_copy_dma, + .copy_ring_index = RADEON_RING_TYPE_GFX_INDEX, + }, + .surface = { + .set_reg = r100_set_surface_reg, + .clear_reg = r100_clear_surface_reg, + }, + .hpd = { + .init = &rs600_hpd_init, + .fini = &rs600_hpd_fini, + .sense = &rs600_hpd_sense, + .set_polarity = &rs600_hpd_set_polarity, + }, + .pm = { + .misc = &rs600_pm_misc, + .prepare = &rs600_pm_prepare, + .finish = &rs600_pm_finish, + .init_profile = &r420_pm_init_profile, + .get_dynpm_state = &r100_pm_get_dynpm_state, + .get_engine_clock = &radeon_atom_get_engine_clock, + .set_engine_clock = &radeon_atom_set_engine_clock, + .get_memory_clock = &radeon_atom_get_memory_clock, + .set_memory_clock = &radeon_atom_set_memory_clock, + .get_pcie_lanes = NULL, + .set_pcie_lanes = NULL, + .set_clock_gating = &radeon_atom_set_clock_gating, + }, + .pflip = { + .pre_page_flip = &rs600_pre_page_flip, + .page_flip = &rs600_page_flip, + .post_page_flip = &rs600_post_page_flip, + }, }; static struct radeon_asic rv515_asic = { @@ -550,48 +728,70 @@ static struct radeon_asic rv515_asic = { .vga_set_state = &r100_vga_set_state, .gpu_is_lockup = &r300_gpu_is_lockup, .asic_reset = &rs600_asic_reset, - .gart_tlb_flush = &rv370_pcie_gart_tlb_flush, - .gart_set_page = &rv370_pcie_gart_set_page, - .ring_start = &rv515_ring_start, - .ring_test = &r100_ring_test, + .ioctl_wait_idle = NULL, + .gui_idle = &r100_gui_idle, + .mc_wait_for_idle = &rv515_mc_wait_for_idle, + .gart = { + .tlb_flush = &rv370_pcie_gart_tlb_flush, + .set_page = &rv370_pcie_gart_set_page, + }, .ring = { [RADEON_RING_TYPE_GFX_INDEX] = { .ib_execute = &r100_ring_ib_execute, .emit_fence = &r300_fence_ring_emit, .emit_semaphore = &r100_semaphore_ring_emit, + .cs_parse = &r300_cs_parse, + .ring_start = &rv515_ring_start, + .ring_test = &r100_ring_test, + .ib_test = &r100_ib_test, } }, - .irq_set = &rs600_irq_set, - .irq_process = &rs600_irq_process, - .get_vblank_counter = &rs600_get_vblank_counter, - .cs_parse = &r300_cs_parse, - .copy_blit = &r100_copy_blit, - .copy_dma = &r200_copy_dma, - .copy = &r100_copy_blit, - .get_engine_clock = &radeon_atom_get_engine_clock, - .set_engine_clock = &radeon_atom_set_engine_clock, - .get_memory_clock = &radeon_atom_get_memory_clock, - .set_memory_clock = &radeon_atom_set_memory_clock, - .get_pcie_lanes = &rv370_get_pcie_lanes, - .set_pcie_lanes = &rv370_set_pcie_lanes, - .set_clock_gating = &radeon_atom_set_clock_gating, - .set_surface_reg = r100_set_surface_reg, - .clear_surface_reg = r100_clear_surface_reg, - .bandwidth_update = &rv515_bandwidth_update, - .hpd_init = &rs600_hpd_init, - .hpd_fini = &rs600_hpd_fini, - .hpd_sense = &rs600_hpd_sense, - .hpd_set_polarity = &rs600_hpd_set_polarity, - .ioctl_wait_idle = NULL, - .gui_idle = &r100_gui_idle, - .pm_misc = &rs600_pm_misc, - .pm_prepare = &rs600_pm_prepare, - .pm_finish = &rs600_pm_finish, - .pm_init_profile = &r420_pm_init_profile, - .pm_get_dynpm_state = &r100_pm_get_dynpm_state, - .pre_page_flip = &rs600_pre_page_flip, - .page_flip = &rs600_page_flip, - .post_page_flip = &rs600_post_page_flip, + .irq = { + .set = &rs600_irq_set, + .process = &rs600_irq_process, + }, + .display = { + .get_vblank_counter = &rs600_get_vblank_counter, + .bandwidth_update = &rv515_bandwidth_update, + .wait_for_vblank = &avivo_wait_for_vblank, + }, + .copy = { + .blit = &r100_copy_blit, + .blit_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .dma = &r200_copy_dma, + .dma_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .copy = &r100_copy_blit, + .copy_ring_index = RADEON_RING_TYPE_GFX_INDEX, + }, + .surface = { + .set_reg = r100_set_surface_reg, + .clear_reg = r100_clear_surface_reg, + }, + .hpd = { + .init = &rs600_hpd_init, + .fini = &rs600_hpd_fini, + .sense = &rs600_hpd_sense, + .set_polarity = &rs600_hpd_set_polarity, + }, + .pm = { + .misc = &rs600_pm_misc, + .prepare = &rs600_pm_prepare, + .finish = &rs600_pm_finish, + .init_profile = &r420_pm_init_profile, + .get_dynpm_state = &r100_pm_get_dynpm_state, + .get_engine_clock = &radeon_atom_get_engine_clock, + .set_engine_clock = &radeon_atom_set_engine_clock, + .get_memory_clock = &radeon_atom_get_memory_clock, + .set_memory_clock = &radeon_atom_set_memory_clock, + .get_pcie_lanes = &rv370_get_pcie_lanes, + .set_pcie_lanes = &rv370_set_pcie_lanes, + .set_clock_gating = &radeon_atom_set_clock_gating, + }, + .pflip = { + .pre_page_flip = &rs600_pre_page_flip, + .page_flip = &rs600_page_flip, + .post_page_flip = &rs600_post_page_flip, + }, }; static struct radeon_asic r520_asic = { @@ -602,48 +802,70 @@ static struct radeon_asic r520_asic = { .vga_set_state = &r100_vga_set_state, .gpu_is_lockup = &r300_gpu_is_lockup, .asic_reset = &rs600_asic_reset, - .gart_tlb_flush = &rv370_pcie_gart_tlb_flush, - .gart_set_page = &rv370_pcie_gart_set_page, - .ring_start = &rv515_ring_start, - .ring_test = &r100_ring_test, + .ioctl_wait_idle = NULL, + .gui_idle = &r100_gui_idle, + .mc_wait_for_idle = &r520_mc_wait_for_idle, + .gart = { + .tlb_flush = &rv370_pcie_gart_tlb_flush, + .set_page = &rv370_pcie_gart_set_page, + }, .ring = { [RADEON_RING_TYPE_GFX_INDEX] = { .ib_execute = &r100_ring_ib_execute, .emit_fence = &r300_fence_ring_emit, .emit_semaphore = &r100_semaphore_ring_emit, + .cs_parse = &r300_cs_parse, + .ring_start = &rv515_ring_start, + .ring_test = &r100_ring_test, + .ib_test = &r100_ib_test, } }, - .irq_set = &rs600_irq_set, - .irq_process = &rs600_irq_process, - .get_vblank_counter = &rs600_get_vblank_counter, - .cs_parse = &r300_cs_parse, - .copy_blit = &r100_copy_blit, - .copy_dma = &r200_copy_dma, - .copy = &r100_copy_blit, - .get_engine_clock = &radeon_atom_get_engine_clock, - .set_engine_clock = &radeon_atom_set_engine_clock, - .get_memory_clock = &radeon_atom_get_memory_clock, - .set_memory_clock = &radeon_atom_set_memory_clock, - .get_pcie_lanes = &rv370_get_pcie_lanes, - .set_pcie_lanes = &rv370_set_pcie_lanes, - .set_clock_gating = &radeon_atom_set_clock_gating, - .set_surface_reg = r100_set_surface_reg, - .clear_surface_reg = r100_clear_surface_reg, - .bandwidth_update = &rv515_bandwidth_update, - .hpd_init = &rs600_hpd_init, - .hpd_fini = &rs600_hpd_fini, - .hpd_sense = &rs600_hpd_sense, - .hpd_set_polarity = &rs600_hpd_set_polarity, - .ioctl_wait_idle = NULL, - .gui_idle = &r100_gui_idle, - .pm_misc = &rs600_pm_misc, - .pm_prepare = &rs600_pm_prepare, - .pm_finish = &rs600_pm_finish, - .pm_init_profile = &r420_pm_init_profile, - .pm_get_dynpm_state = &r100_pm_get_dynpm_state, - .pre_page_flip = &rs600_pre_page_flip, - .page_flip = &rs600_page_flip, - .post_page_flip = &rs600_post_page_flip, + .irq = { + .set = &rs600_irq_set, + .process = &rs600_irq_process, + }, + .display = { + .bandwidth_update = &rv515_bandwidth_update, + .get_vblank_counter = &rs600_get_vblank_counter, + .wait_for_vblank = &avivo_wait_for_vblank, + }, + .copy = { + .blit = &r100_copy_blit, + .blit_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .dma = &r200_copy_dma, + .dma_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .copy = &r100_copy_blit, + .copy_ring_index = RADEON_RING_TYPE_GFX_INDEX, + }, + .surface = { + .set_reg = r100_set_surface_reg, + .clear_reg = r100_clear_surface_reg, + }, + .hpd = { + .init = &rs600_hpd_init, + .fini = &rs600_hpd_fini, + .sense = &rs600_hpd_sense, + .set_polarity = &rs600_hpd_set_polarity, + }, + .pm = { + .misc = &rs600_pm_misc, + .prepare = &rs600_pm_prepare, + .finish = &rs600_pm_finish, + .init_profile = &r420_pm_init_profile, + .get_dynpm_state = &r100_pm_get_dynpm_state, + .get_engine_clock = &radeon_atom_get_engine_clock, + .set_engine_clock = &radeon_atom_set_engine_clock, + .get_memory_clock = &radeon_atom_get_memory_clock, + .set_memory_clock = &radeon_atom_set_memory_clock, + .get_pcie_lanes = &rv370_get_pcie_lanes, + .set_pcie_lanes = &rv370_set_pcie_lanes, + .set_clock_gating = &radeon_atom_set_clock_gating, + }, + .pflip = { + .pre_page_flip = &rs600_pre_page_flip, + .page_flip = &rs600_page_flip, + .post_page_flip = &rs600_post_page_flip, + }, }; static struct radeon_asic r600_asic = { @@ -654,47 +876,69 @@ static struct radeon_asic r600_asic = { .vga_set_state = &r600_vga_set_state, .gpu_is_lockup = &r600_gpu_is_lockup, .asic_reset = &r600_asic_reset, - .gart_tlb_flush = &r600_pcie_gart_tlb_flush, - .gart_set_page = &rs600_gart_set_page, - .ring_test = &r600_ring_test, + .ioctl_wait_idle = r600_ioctl_wait_idle, + .gui_idle = &r600_gui_idle, + .mc_wait_for_idle = &r600_mc_wait_for_idle, + .gart = { + .tlb_flush = &r600_pcie_gart_tlb_flush, + .set_page = &rs600_gart_set_page, + }, .ring = { [RADEON_RING_TYPE_GFX_INDEX] = { .ib_execute = &r600_ring_ib_execute, .emit_fence = &r600_fence_ring_emit, .emit_semaphore = &r600_semaphore_ring_emit, + .cs_parse = &r600_cs_parse, + .ring_test = &r600_ring_test, + .ib_test = &r600_ib_test, } }, - .irq_set = &r600_irq_set, - .irq_process = &r600_irq_process, - .get_vblank_counter = &rs600_get_vblank_counter, - .cs_parse = &r600_cs_parse, - .copy_blit = &r600_copy_blit, - .copy_dma = NULL, - .copy = &r600_copy_blit, - .get_engine_clock = &radeon_atom_get_engine_clock, - .set_engine_clock = &radeon_atom_set_engine_clock, - .get_memory_clock = &radeon_atom_get_memory_clock, - .set_memory_clock = &radeon_atom_set_memory_clock, - .get_pcie_lanes = &r600_get_pcie_lanes, - .set_pcie_lanes = &r600_set_pcie_lanes, - .set_clock_gating = NULL, - .set_surface_reg = r600_set_surface_reg, - .clear_surface_reg = r600_clear_surface_reg, - .bandwidth_update = &rv515_bandwidth_update, - .hpd_init = &r600_hpd_init, - .hpd_fini = &r600_hpd_fini, - .hpd_sense = &r600_hpd_sense, - .hpd_set_polarity = &r600_hpd_set_polarity, - .ioctl_wait_idle = r600_ioctl_wait_idle, - .gui_idle = &r600_gui_idle, - .pm_misc = &r600_pm_misc, - .pm_prepare = &rs600_pm_prepare, - .pm_finish = &rs600_pm_finish, - .pm_init_profile = &r600_pm_init_profile, - .pm_get_dynpm_state = &r600_pm_get_dynpm_state, - .pre_page_flip = &rs600_pre_page_flip, - .page_flip = &rs600_page_flip, - .post_page_flip = &rs600_post_page_flip, + .irq = { + .set = &r600_irq_set, + .process = &r600_irq_process, + }, + .display = { + .bandwidth_update = &rv515_bandwidth_update, + .get_vblank_counter = &rs600_get_vblank_counter, + .wait_for_vblank = &avivo_wait_for_vblank, + }, + .copy = { + .blit = &r600_copy_blit, + .blit_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .dma = NULL, + .dma_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .copy = &r600_copy_blit, + .copy_ring_index = RADEON_RING_TYPE_GFX_INDEX, + }, + .surface = { + .set_reg = r600_set_surface_reg, + .clear_reg = r600_clear_surface_reg, + }, + .hpd = { + .init = &r600_hpd_init, + .fini = &r600_hpd_fini, + .sense = &r600_hpd_sense, + .set_polarity = &r600_hpd_set_polarity, + }, + .pm = { + .misc = &r600_pm_misc, + .prepare = &rs600_pm_prepare, + .finish = &rs600_pm_finish, + .init_profile = &r600_pm_init_profile, + .get_dynpm_state = &r600_pm_get_dynpm_state, + .get_engine_clock = &radeon_atom_get_engine_clock, + .set_engine_clock = &radeon_atom_set_engine_clock, + .get_memory_clock = &radeon_atom_get_memory_clock, + .set_memory_clock = &radeon_atom_set_memory_clock, + .get_pcie_lanes = &r600_get_pcie_lanes, + .set_pcie_lanes = &r600_set_pcie_lanes, + .set_clock_gating = NULL, + }, + .pflip = { + .pre_page_flip = &rs600_pre_page_flip, + .page_flip = &rs600_page_flip, + .post_page_flip = &rs600_post_page_flip, + }, }; static struct radeon_asic rs780_asic = { @@ -705,47 +949,69 @@ static struct radeon_asic rs780_asic = { .gpu_is_lockup = &r600_gpu_is_lockup, .vga_set_state = &r600_vga_set_state, .asic_reset = &r600_asic_reset, - .gart_tlb_flush = &r600_pcie_gart_tlb_flush, - .gart_set_page = &rs600_gart_set_page, - .ring_test = &r600_ring_test, + .ioctl_wait_idle = r600_ioctl_wait_idle, + .gui_idle = &r600_gui_idle, + .mc_wait_for_idle = &r600_mc_wait_for_idle, + .gart = { + .tlb_flush = &r600_pcie_gart_tlb_flush, + .set_page = &rs600_gart_set_page, + }, .ring = { [RADEON_RING_TYPE_GFX_INDEX] = { .ib_execute = &r600_ring_ib_execute, .emit_fence = &r600_fence_ring_emit, .emit_semaphore = &r600_semaphore_ring_emit, + .cs_parse = &r600_cs_parse, + .ring_test = &r600_ring_test, + .ib_test = &r600_ib_test, } }, - .irq_set = &r600_irq_set, - .irq_process = &r600_irq_process, - .get_vblank_counter = &rs600_get_vblank_counter, - .cs_parse = &r600_cs_parse, - .copy_blit = &r600_copy_blit, - .copy_dma = NULL, - .copy = &r600_copy_blit, - .get_engine_clock = &radeon_atom_get_engine_clock, - .set_engine_clock = &radeon_atom_set_engine_clock, - .get_memory_clock = NULL, - .set_memory_clock = NULL, - .get_pcie_lanes = NULL, - .set_pcie_lanes = NULL, - .set_clock_gating = NULL, - .set_surface_reg = r600_set_surface_reg, - .clear_surface_reg = r600_clear_surface_reg, - .bandwidth_update = &rs690_bandwidth_update, - .hpd_init = &r600_hpd_init, - .hpd_fini = &r600_hpd_fini, - .hpd_sense = &r600_hpd_sense, - .hpd_set_polarity = &r600_hpd_set_polarity, - .ioctl_wait_idle = r600_ioctl_wait_idle, - .gui_idle = &r600_gui_idle, - .pm_misc = &r600_pm_misc, - .pm_prepare = &rs600_pm_prepare, - .pm_finish = &rs600_pm_finish, - .pm_init_profile = &rs780_pm_init_profile, - .pm_get_dynpm_state = &r600_pm_get_dynpm_state, - .pre_page_flip = &rs600_pre_page_flip, - .page_flip = &rs600_page_flip, - .post_page_flip = &rs600_post_page_flip, + .irq = { + .set = &r600_irq_set, + .process = &r600_irq_process, + }, + .display = { + .bandwidth_update = &rs690_bandwidth_update, + .get_vblank_counter = &rs600_get_vblank_counter, + .wait_for_vblank = &avivo_wait_for_vblank, + }, + .copy = { + .blit = &r600_copy_blit, + .blit_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .dma = NULL, + .dma_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .copy = &r600_copy_blit, + .copy_ring_index = RADEON_RING_TYPE_GFX_INDEX, + }, + .surface = { + .set_reg = r600_set_surface_reg, + .clear_reg = r600_clear_surface_reg, + }, + .hpd = { + .init = &r600_hpd_init, + .fini = &r600_hpd_fini, + .sense = &r600_hpd_sense, + .set_polarity = &r600_hpd_set_polarity, + }, + .pm = { + .misc = &r600_pm_misc, + .prepare = &rs600_pm_prepare, + .finish = &rs600_pm_finish, + .init_profile = &rs780_pm_init_profile, + .get_dynpm_state = &r600_pm_get_dynpm_state, + .get_engine_clock = &radeon_atom_get_engine_clock, + .set_engine_clock = &radeon_atom_set_engine_clock, + .get_memory_clock = NULL, + .set_memory_clock = NULL, + .get_pcie_lanes = NULL, + .set_pcie_lanes = NULL, + .set_clock_gating = NULL, + }, + .pflip = { + .pre_page_flip = &rs600_pre_page_flip, + .page_flip = &rs600_page_flip, + .post_page_flip = &rs600_post_page_flip, + }, }; static struct radeon_asic rv770_asic = { @@ -756,47 +1022,69 @@ static struct radeon_asic rv770_asic = { .asic_reset = &r600_asic_reset, .gpu_is_lockup = &r600_gpu_is_lockup, .vga_set_state = &r600_vga_set_state, - .gart_tlb_flush = &r600_pcie_gart_tlb_flush, - .gart_set_page = &rs600_gart_set_page, - .ring_test = &r600_ring_test, + .ioctl_wait_idle = r600_ioctl_wait_idle, + .gui_idle = &r600_gui_idle, + .mc_wait_for_idle = &r600_mc_wait_for_idle, + .gart = { + .tlb_flush = &r600_pcie_gart_tlb_flush, + .set_page = &rs600_gart_set_page, + }, .ring = { [RADEON_RING_TYPE_GFX_INDEX] = { .ib_execute = &r600_ring_ib_execute, .emit_fence = &r600_fence_ring_emit, .emit_semaphore = &r600_semaphore_ring_emit, + .cs_parse = &r600_cs_parse, + .ring_test = &r600_ring_test, + .ib_test = &r600_ib_test, } }, - .irq_set = &r600_irq_set, - .irq_process = &r600_irq_process, - .get_vblank_counter = &rs600_get_vblank_counter, - .cs_parse = &r600_cs_parse, - .copy_blit = &r600_copy_blit, - .copy_dma = NULL, - .copy = &r600_copy_blit, - .get_engine_clock = &radeon_atom_get_engine_clock, - .set_engine_clock = &radeon_atom_set_engine_clock, - .get_memory_clock = &radeon_atom_get_memory_clock, - .set_memory_clock = &radeon_atom_set_memory_clock, - .get_pcie_lanes = &r600_get_pcie_lanes, - .set_pcie_lanes = &r600_set_pcie_lanes, - .set_clock_gating = &radeon_atom_set_clock_gating, - .set_surface_reg = r600_set_surface_reg, - .clear_surface_reg = r600_clear_surface_reg, - .bandwidth_update = &rv515_bandwidth_update, - .hpd_init = &r600_hpd_init, - .hpd_fini = &r600_hpd_fini, - .hpd_sense = &r600_hpd_sense, - .hpd_set_polarity = &r600_hpd_set_polarity, - .ioctl_wait_idle = r600_ioctl_wait_idle, - .gui_idle = &r600_gui_idle, - .pm_misc = &rv770_pm_misc, - .pm_prepare = &rs600_pm_prepare, - .pm_finish = &rs600_pm_finish, - .pm_init_profile = &r600_pm_init_profile, - .pm_get_dynpm_state = &r600_pm_get_dynpm_state, - .pre_page_flip = &rs600_pre_page_flip, - .page_flip = &rv770_page_flip, - .post_page_flip = &rs600_post_page_flip, + .irq = { + .set = &r600_irq_set, + .process = &r600_irq_process, + }, + .display = { + .bandwidth_update = &rv515_bandwidth_update, + .get_vblank_counter = &rs600_get_vblank_counter, + .wait_for_vblank = &avivo_wait_for_vblank, + }, + .copy = { + .blit = &r600_copy_blit, + .blit_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .dma = NULL, + .dma_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .copy = &r600_copy_blit, + .copy_ring_index = RADEON_RING_TYPE_GFX_INDEX, + }, + .surface = { + .set_reg = r600_set_surface_reg, + .clear_reg = r600_clear_surface_reg, + }, + .hpd = { + .init = &r600_hpd_init, + .fini = &r600_hpd_fini, + .sense = &r600_hpd_sense, + .set_polarity = &r600_hpd_set_polarity, + }, + .pm = { + .misc = &rv770_pm_misc, + .prepare = &rs600_pm_prepare, + .finish = &rs600_pm_finish, + .init_profile = &r600_pm_init_profile, + .get_dynpm_state = &r600_pm_get_dynpm_state, + .get_engine_clock = &radeon_atom_get_engine_clock, + .set_engine_clock = &radeon_atom_set_engine_clock, + .get_memory_clock = &radeon_atom_get_memory_clock, + .set_memory_clock = &radeon_atom_set_memory_clock, + .get_pcie_lanes = &r600_get_pcie_lanes, + .set_pcie_lanes = &r600_set_pcie_lanes, + .set_clock_gating = &radeon_atom_set_clock_gating, + }, + .pflip = { + .pre_page_flip = &rs600_pre_page_flip, + .page_flip = &rv770_page_flip, + .post_page_flip = &rs600_post_page_flip, + }, }; static struct radeon_asic evergreen_asic = { @@ -807,47 +1095,69 @@ static struct radeon_asic evergreen_asic = { .gpu_is_lockup = &evergreen_gpu_is_lockup, .asic_reset = &evergreen_asic_reset, .vga_set_state = &r600_vga_set_state, - .gart_tlb_flush = &evergreen_pcie_gart_tlb_flush, - .gart_set_page = &rs600_gart_set_page, - .ring_test = &r600_ring_test, + .ioctl_wait_idle = r600_ioctl_wait_idle, + .gui_idle = &r600_gui_idle, + .mc_wait_for_idle = &evergreen_mc_wait_for_idle, + .gart = { + .tlb_flush = &evergreen_pcie_gart_tlb_flush, + .set_page = &rs600_gart_set_page, + }, .ring = { [RADEON_RING_TYPE_GFX_INDEX] = { .ib_execute = &evergreen_ring_ib_execute, .emit_fence = &r600_fence_ring_emit, .emit_semaphore = &r600_semaphore_ring_emit, + .cs_parse = &evergreen_cs_parse, + .ring_test = &r600_ring_test, + .ib_test = &r600_ib_test, } }, - .irq_set = &evergreen_irq_set, - .irq_process = &evergreen_irq_process, - .get_vblank_counter = &evergreen_get_vblank_counter, - .cs_parse = &evergreen_cs_parse, - .copy_blit = &r600_copy_blit, - .copy_dma = NULL, - .copy = &r600_copy_blit, - .get_engine_clock = &radeon_atom_get_engine_clock, - .set_engine_clock = &radeon_atom_set_engine_clock, - .get_memory_clock = &radeon_atom_get_memory_clock, - .set_memory_clock = &radeon_atom_set_memory_clock, - .get_pcie_lanes = &r600_get_pcie_lanes, - .set_pcie_lanes = &r600_set_pcie_lanes, - .set_clock_gating = NULL, - .set_surface_reg = r600_set_surface_reg, - .clear_surface_reg = r600_clear_surface_reg, - .bandwidth_update = &evergreen_bandwidth_update, - .hpd_init = &evergreen_hpd_init, - .hpd_fini = &evergreen_hpd_fini, - .hpd_sense = &evergreen_hpd_sense, - .hpd_set_polarity = &evergreen_hpd_set_polarity, - .ioctl_wait_idle = r600_ioctl_wait_idle, - .gui_idle = &r600_gui_idle, - .pm_misc = &evergreen_pm_misc, - .pm_prepare = &evergreen_pm_prepare, - .pm_finish = &evergreen_pm_finish, - .pm_init_profile = &r600_pm_init_profile, - .pm_get_dynpm_state = &r600_pm_get_dynpm_state, - .pre_page_flip = &evergreen_pre_page_flip, - .page_flip = &evergreen_page_flip, - .post_page_flip = &evergreen_post_page_flip, + .irq = { + .set = &evergreen_irq_set, + .process = &evergreen_irq_process, + }, + .display = { + .bandwidth_update = &evergreen_bandwidth_update, + .get_vblank_counter = &evergreen_get_vblank_counter, + .wait_for_vblank = &dce4_wait_for_vblank, + }, + .copy = { + .blit = &r600_copy_blit, + .blit_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .dma = NULL, + .dma_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .copy = &r600_copy_blit, + .copy_ring_index = RADEON_RING_TYPE_GFX_INDEX, + }, + .surface = { + .set_reg = r600_set_surface_reg, + .clear_reg = r600_clear_surface_reg, + }, + .hpd = { + .init = &evergreen_hpd_init, + .fini = &evergreen_hpd_fini, + .sense = &evergreen_hpd_sense, + .set_polarity = &evergreen_hpd_set_polarity, + }, + .pm = { + .misc = &evergreen_pm_misc, + .prepare = &evergreen_pm_prepare, + .finish = &evergreen_pm_finish, + .init_profile = &r600_pm_init_profile, + .get_dynpm_state = &r600_pm_get_dynpm_state, + .get_engine_clock = &radeon_atom_get_engine_clock, + .set_engine_clock = &radeon_atom_set_engine_clock, + .get_memory_clock = &radeon_atom_get_memory_clock, + .set_memory_clock = &radeon_atom_set_memory_clock, + .get_pcie_lanes = &r600_get_pcie_lanes, + .set_pcie_lanes = &r600_set_pcie_lanes, + .set_clock_gating = NULL, + }, + .pflip = { + .pre_page_flip = &evergreen_pre_page_flip, + .page_flip = &evergreen_page_flip, + .post_page_flip = &evergreen_post_page_flip, + }, }; static struct radeon_asic sumo_asic = { @@ -858,47 +1168,69 @@ static struct radeon_asic sumo_asic = { .gpu_is_lockup = &evergreen_gpu_is_lockup, .asic_reset = &evergreen_asic_reset, .vga_set_state = &r600_vga_set_state, - .gart_tlb_flush = &evergreen_pcie_gart_tlb_flush, - .gart_set_page = &rs600_gart_set_page, - .ring_test = &r600_ring_test, + .ioctl_wait_idle = r600_ioctl_wait_idle, + .gui_idle = &r600_gui_idle, + .mc_wait_for_idle = &evergreen_mc_wait_for_idle, + .gart = { + .tlb_flush = &evergreen_pcie_gart_tlb_flush, + .set_page = &rs600_gart_set_page, + }, .ring = { [RADEON_RING_TYPE_GFX_INDEX] = { .ib_execute = &evergreen_ring_ib_execute, .emit_fence = &r600_fence_ring_emit, .emit_semaphore = &r600_semaphore_ring_emit, - } + .cs_parse = &evergreen_cs_parse, + .ring_test = &r600_ring_test, + .ib_test = &r600_ib_test, + }, + }, + .irq = { + .set = &evergreen_irq_set, + .process = &evergreen_irq_process, + }, + .display = { + .bandwidth_update = &evergreen_bandwidth_update, + .get_vblank_counter = &evergreen_get_vblank_counter, + .wait_for_vblank = &dce4_wait_for_vblank, + }, + .copy = { + .blit = &r600_copy_blit, + .blit_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .dma = NULL, + .dma_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .copy = &r600_copy_blit, + .copy_ring_index = RADEON_RING_TYPE_GFX_INDEX, + }, + .surface = { + .set_reg = r600_set_surface_reg, + .clear_reg = r600_clear_surface_reg, + }, + .hpd = { + .init = &evergreen_hpd_init, + .fini = &evergreen_hpd_fini, + .sense = &evergreen_hpd_sense, + .set_polarity = &evergreen_hpd_set_polarity, + }, + .pm = { + .misc = &evergreen_pm_misc, + .prepare = &evergreen_pm_prepare, + .finish = &evergreen_pm_finish, + .init_profile = &sumo_pm_init_profile, + .get_dynpm_state = &r600_pm_get_dynpm_state, + .get_engine_clock = &radeon_atom_get_engine_clock, + .set_engine_clock = &radeon_atom_set_engine_clock, + .get_memory_clock = NULL, + .set_memory_clock = NULL, + .get_pcie_lanes = NULL, + .set_pcie_lanes = NULL, + .set_clock_gating = NULL, + }, + .pflip = { + .pre_page_flip = &evergreen_pre_page_flip, + .page_flip = &evergreen_page_flip, + .post_page_flip = &evergreen_post_page_flip, }, - .irq_set = &evergreen_irq_set, - .irq_process = &evergreen_irq_process, - .get_vblank_counter = &evergreen_get_vblank_counter, - .cs_parse = &evergreen_cs_parse, - .copy_blit = &r600_copy_blit, - .copy_dma = NULL, - .copy = &r600_copy_blit, - .get_engine_clock = &radeon_atom_get_engine_clock, - .set_engine_clock = &radeon_atom_set_engine_clock, - .get_memory_clock = NULL, - .set_memory_clock = NULL, - .get_pcie_lanes = NULL, - .set_pcie_lanes = NULL, - .set_clock_gating = NULL, - .set_surface_reg = r600_set_surface_reg, - .clear_surface_reg = r600_clear_surface_reg, - .bandwidth_update = &evergreen_bandwidth_update, - .hpd_init = &evergreen_hpd_init, - .hpd_fini = &evergreen_hpd_fini, - .hpd_sense = &evergreen_hpd_sense, - .hpd_set_polarity = &evergreen_hpd_set_polarity, - .ioctl_wait_idle = r600_ioctl_wait_idle, - .gui_idle = &r600_gui_idle, - .pm_misc = &evergreen_pm_misc, - .pm_prepare = &evergreen_pm_prepare, - .pm_finish = &evergreen_pm_finish, - .pm_init_profile = &sumo_pm_init_profile, - .pm_get_dynpm_state = &r600_pm_get_dynpm_state, - .pre_page_flip = &evergreen_pre_page_flip, - .page_flip = &evergreen_page_flip, - .post_page_flip = &evergreen_post_page_flip, }; static struct radeon_asic btc_asic = { @@ -909,47 +1241,69 @@ static struct radeon_asic btc_asic = { .gpu_is_lockup = &evergreen_gpu_is_lockup, .asic_reset = &evergreen_asic_reset, .vga_set_state = &r600_vga_set_state, - .gart_tlb_flush = &evergreen_pcie_gart_tlb_flush, - .gart_set_page = &rs600_gart_set_page, - .ring_test = &r600_ring_test, + .ioctl_wait_idle = r600_ioctl_wait_idle, + .gui_idle = &r600_gui_idle, + .mc_wait_for_idle = &evergreen_mc_wait_for_idle, + .gart = { + .tlb_flush = &evergreen_pcie_gart_tlb_flush, + .set_page = &rs600_gart_set_page, + }, .ring = { [RADEON_RING_TYPE_GFX_INDEX] = { .ib_execute = &evergreen_ring_ib_execute, .emit_fence = &r600_fence_ring_emit, .emit_semaphore = &r600_semaphore_ring_emit, + .cs_parse = &evergreen_cs_parse, + .ring_test = &r600_ring_test, + .ib_test = &r600_ib_test, } }, - .irq_set = &evergreen_irq_set, - .irq_process = &evergreen_irq_process, - .get_vblank_counter = &evergreen_get_vblank_counter, - .cs_parse = &evergreen_cs_parse, - .copy_blit = &r600_copy_blit, - .copy_dma = NULL, - .copy = &r600_copy_blit, - .get_engine_clock = &radeon_atom_get_engine_clock, - .set_engine_clock = &radeon_atom_set_engine_clock, - .get_memory_clock = &radeon_atom_get_memory_clock, - .set_memory_clock = &radeon_atom_set_memory_clock, - .get_pcie_lanes = NULL, - .set_pcie_lanes = NULL, - .set_clock_gating = NULL, - .set_surface_reg = r600_set_surface_reg, - .clear_surface_reg = r600_clear_surface_reg, - .bandwidth_update = &evergreen_bandwidth_update, - .hpd_init = &evergreen_hpd_init, - .hpd_fini = &evergreen_hpd_fini, - .hpd_sense = &evergreen_hpd_sense, - .hpd_set_polarity = &evergreen_hpd_set_polarity, - .ioctl_wait_idle = r600_ioctl_wait_idle, - .gui_idle = &r600_gui_idle, - .pm_misc = &evergreen_pm_misc, - .pm_prepare = &evergreen_pm_prepare, - .pm_finish = &evergreen_pm_finish, - .pm_init_profile = &r600_pm_init_profile, - .pm_get_dynpm_state = &r600_pm_get_dynpm_state, - .pre_page_flip = &evergreen_pre_page_flip, - .page_flip = &evergreen_page_flip, - .post_page_flip = &evergreen_post_page_flip, + .irq = { + .set = &evergreen_irq_set, + .process = &evergreen_irq_process, + }, + .display = { + .bandwidth_update = &evergreen_bandwidth_update, + .get_vblank_counter = &evergreen_get_vblank_counter, + .wait_for_vblank = &dce4_wait_for_vblank, + }, + .copy = { + .blit = &r600_copy_blit, + .blit_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .dma = NULL, + .dma_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .copy = &r600_copy_blit, + .copy_ring_index = RADEON_RING_TYPE_GFX_INDEX, + }, + .surface = { + .set_reg = r600_set_surface_reg, + .clear_reg = r600_clear_surface_reg, + }, + .hpd = { + .init = &evergreen_hpd_init, + .fini = &evergreen_hpd_fini, + .sense = &evergreen_hpd_sense, + .set_polarity = &evergreen_hpd_set_polarity, + }, + .pm = { + .misc = &evergreen_pm_misc, + .prepare = &evergreen_pm_prepare, + .finish = &evergreen_pm_finish, + .init_profile = &r600_pm_init_profile, + .get_dynpm_state = &r600_pm_get_dynpm_state, + .get_engine_clock = &radeon_atom_get_engine_clock, + .set_engine_clock = &radeon_atom_set_engine_clock, + .get_memory_clock = &radeon_atom_get_memory_clock, + .set_memory_clock = &radeon_atom_set_memory_clock, + .get_pcie_lanes = NULL, + .set_pcie_lanes = NULL, + .set_clock_gating = NULL, + }, + .pflip = { + .pre_page_flip = &evergreen_pre_page_flip, + .page_flip = &evergreen_page_flip, + .post_page_flip = &evergreen_post_page_flip, + }, }; static const struct radeon_vm_funcs cayman_vm_funcs = { @@ -970,60 +1324,88 @@ static struct radeon_asic cayman_asic = { .gpu_is_lockup = &cayman_gpu_is_lockup, .asic_reset = &cayman_asic_reset, .vga_set_state = &r600_vga_set_state, - .gart_tlb_flush = &cayman_pcie_gart_tlb_flush, - .gart_set_page = &rs600_gart_set_page, - .ring_test = &r600_ring_test, + .ioctl_wait_idle = r600_ioctl_wait_idle, + .gui_idle = &r600_gui_idle, + .mc_wait_for_idle = &evergreen_mc_wait_for_idle, + .gart = { + .tlb_flush = &cayman_pcie_gart_tlb_flush, + .set_page = &rs600_gart_set_page, + }, .ring = { [RADEON_RING_TYPE_GFX_INDEX] = { .ib_execute = &cayman_ring_ib_execute, .ib_parse = &evergreen_ib_parse, .emit_fence = &cayman_fence_ring_emit, .emit_semaphore = &r600_semaphore_ring_emit, + .cs_parse = &evergreen_cs_parse, + .ring_test = &r600_ring_test, + .ib_test = &r600_ib_test, }, [CAYMAN_RING_TYPE_CP1_INDEX] = { .ib_execute = &cayman_ring_ib_execute, .ib_parse = &evergreen_ib_parse, .emit_fence = &cayman_fence_ring_emit, .emit_semaphore = &r600_semaphore_ring_emit, + .cs_parse = &evergreen_cs_parse, + .ring_test = &r600_ring_test, + .ib_test = &r600_ib_test, }, [CAYMAN_RING_TYPE_CP2_INDEX] = { .ib_execute = &cayman_ring_ib_execute, .ib_parse = &evergreen_ib_parse, .emit_fence = &cayman_fence_ring_emit, .emit_semaphore = &r600_semaphore_ring_emit, + .cs_parse = &evergreen_cs_parse, + .ring_test = &r600_ring_test, + .ib_test = &r600_ib_test, } }, - .irq_set = &evergreen_irq_set, - .irq_process = &evergreen_irq_process, - .get_vblank_counter = &evergreen_get_vblank_counter, - .cs_parse = &evergreen_cs_parse, - .copy_blit = &r600_copy_blit, - .copy_dma = NULL, - .copy = &r600_copy_blit, - .get_engine_clock = &radeon_atom_get_engine_clock, - .set_engine_clock = &radeon_atom_set_engine_clock, - .get_memory_clock = &radeon_atom_get_memory_clock, - .set_memory_clock = &radeon_atom_set_memory_clock, - .get_pcie_lanes = NULL, - .set_pcie_lanes = NULL, - .set_clock_gating = NULL, - .set_surface_reg = r600_set_surface_reg, - .clear_surface_reg = r600_clear_surface_reg, - .bandwidth_update = &evergreen_bandwidth_update, - .hpd_init = &evergreen_hpd_init, - .hpd_fini = &evergreen_hpd_fini, - .hpd_sense = &evergreen_hpd_sense, - .hpd_set_polarity = &evergreen_hpd_set_polarity, - .ioctl_wait_idle = r600_ioctl_wait_idle, - .gui_idle = &r600_gui_idle, - .pm_misc = &evergreen_pm_misc, - .pm_prepare = &evergreen_pm_prepare, - .pm_finish = &evergreen_pm_finish, - .pm_init_profile = &r600_pm_init_profile, - .pm_get_dynpm_state = &r600_pm_get_dynpm_state, - .pre_page_flip = &evergreen_pre_page_flip, - .page_flip = &evergreen_page_flip, - .post_page_flip = &evergreen_post_page_flip, + .irq = { + .set = &evergreen_irq_set, + .process = &evergreen_irq_process, + }, + .display = { + .bandwidth_update = &evergreen_bandwidth_update, + .get_vblank_counter = &evergreen_get_vblank_counter, + .wait_for_vblank = &dce4_wait_for_vblank, + }, + .copy = { + .blit = &r600_copy_blit, + .blit_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .dma = NULL, + .dma_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .copy = &r600_copy_blit, + .copy_ring_index = RADEON_RING_TYPE_GFX_INDEX, + }, + .surface = { + .set_reg = r600_set_surface_reg, + .clear_reg = r600_clear_surface_reg, + }, + .hpd = { + .init = &evergreen_hpd_init, + .fini = &evergreen_hpd_fini, + .sense = &evergreen_hpd_sense, + .set_polarity = &evergreen_hpd_set_polarity, + }, + .pm = { + .misc = &evergreen_pm_misc, + .prepare = &evergreen_pm_prepare, + .finish = &evergreen_pm_finish, + .init_profile = &r600_pm_init_profile, + .get_dynpm_state = &r600_pm_get_dynpm_state, + .get_engine_clock = &radeon_atom_get_engine_clock, + .set_engine_clock = &radeon_atom_set_engine_clock, + .get_memory_clock = &radeon_atom_get_memory_clock, + .set_memory_clock = &radeon_atom_set_memory_clock, + .get_pcie_lanes = NULL, + .set_pcie_lanes = NULL, + .set_clock_gating = NULL, + }, + .pflip = { + .pre_page_flip = &evergreen_pre_page_flip, + .page_flip = &evergreen_page_flip, + .post_page_flip = &evergreen_post_page_flip, + }, }; int radeon_asic_init(struct radeon_device *rdev) @@ -1036,9 +1418,6 @@ int radeon_asic_init(struct radeon_device *rdev) else rdev->num_crtc = 2; - /* set the ring used for bo copies */ - rdev->copy_ring = RADEON_RING_TYPE_GFX_INDEX; - switch (rdev->family) { case CHIP_R100: case CHIP_RV100: @@ -1068,10 +1447,10 @@ int radeon_asic_init(struct radeon_device *rdev) rdev->asic = &r420_asic; /* handle macs */ if (rdev->bios == NULL) { - rdev->asic->get_engine_clock = &radeon_legacy_get_engine_clock; - rdev->asic->set_engine_clock = &radeon_legacy_set_engine_clock; - rdev->asic->get_memory_clock = &radeon_legacy_get_memory_clock; - rdev->asic->set_memory_clock = NULL; + rdev->asic->pm.get_engine_clock = &radeon_legacy_get_engine_clock; + rdev->asic->pm.set_engine_clock = &radeon_legacy_set_engine_clock; + rdev->asic->pm.get_memory_clock = &radeon_legacy_get_memory_clock; + rdev->asic->pm.set_memory_clock = NULL; } break; case CHIP_RS400: @@ -1152,8 +1531,8 @@ int radeon_asic_init(struct radeon_device *rdev) } if (rdev->flags & RADEON_IS_IGP) { - rdev->asic->get_memory_clock = NULL; - rdev->asic->set_memory_clock = NULL; + rdev->asic->pm.get_memory_clock = NULL; + rdev->asic->pm.set_memory_clock = NULL; } return 0; diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h index 6304aef..b8f0a16 100644 --- a/drivers/gpu/drm/radeon/radeon_asic.h +++ b/drivers/gpu/drm/radeon/radeon_asic.h @@ -63,7 +63,7 @@ int r100_asic_reset(struct radeon_device *rdev); u32 r100_get_vblank_counter(struct radeon_device *rdev, int crtc); void r100_pci_gart_tlb_flush(struct radeon_device *rdev); int r100_pci_gart_set_page(struct radeon_device *rdev, int i, uint64_t addr); -void r100_ring_start(struct radeon_device *rdev); +void r100_ring_start(struct radeon_device *rdev, struct radeon_ring *ring); int r100_irq_set(struct radeon_device *rdev); int r100_irq_process(struct radeon_device *rdev); void r100_fence_ring_emit(struct radeon_device *rdev, @@ -109,7 +109,7 @@ bool r100_gpu_cp_is_lockup(struct radeon_device *rdev, struct r100_gpu_lockup *lockup, struct radeon_ring *cp); void r100_ib_fini(struct radeon_device *rdev); -int r100_ib_test(struct radeon_device *rdev); +int r100_ib_test(struct radeon_device *rdev, struct radeon_ring *ring); void r100_irq_disable(struct radeon_device *rdev); void r100_mc_stop(struct radeon_device *rdev, struct r100_mc_save *save); void r100_mc_resume(struct radeon_device *rdev, struct r100_mc_save *save); @@ -139,6 +139,8 @@ extern void r100_pm_get_dynpm_state(struct radeon_device *rdev); extern void r100_pre_page_flip(struct radeon_device *rdev, int crtc); extern u32 r100_page_flip(struct radeon_device *rdev, int crtc, u64 crtc_base); extern void r100_post_page_flip(struct radeon_device *rdev, int crtc); +extern void r100_wait_for_vblank(struct radeon_device *rdev, int crtc); +extern int r100_mc_wait_for_idle(struct radeon_device *rdev); /* * r200,rv250,rs300,rv280 @@ -159,7 +161,7 @@ extern int r300_suspend(struct radeon_device *rdev); extern int r300_resume(struct radeon_device *rdev); extern bool r300_gpu_is_lockup(struct radeon_device *rdev, struct radeon_ring *cp); extern int r300_asic_reset(struct radeon_device *rdev); -extern void r300_ring_start(struct radeon_device *rdev); +extern void r300_ring_start(struct radeon_device *rdev, struct radeon_ring *ring); extern void r300_fence_ring_emit(struct radeon_device *rdev, struct radeon_fence *fence); extern int r300_cs_parse(struct radeon_cs_parser *p); @@ -176,6 +178,7 @@ extern int rv370_pcie_gart_init(struct radeon_device *rdev); extern void rv370_pcie_gart_fini(struct radeon_device *rdev); extern int rv370_pcie_gart_enable(struct radeon_device *rdev); extern void rv370_pcie_gart_disable(struct radeon_device *rdev); +extern int r300_mc_wait_for_idle(struct radeon_device *rdev); /* * r420,r423,rv410 @@ -206,6 +209,7 @@ int rs400_gart_enable(struct radeon_device *rdev); void rs400_gart_adjust_size(struct radeon_device *rdev); void rs400_gart_disable(struct radeon_device *rdev); void rs400_gart_fini(struct radeon_device *rdev); +extern int rs400_mc_wait_for_idle(struct radeon_device *rdev); /* * rs600. @@ -236,7 +240,8 @@ extern void rs600_pre_page_flip(struct radeon_device *rdev, int crtc); extern u32 rs600_page_flip(struct radeon_device *rdev, int crtc, u64 crtc_base); extern void rs600_post_page_flip(struct radeon_device *rdev, int crtc); void rs600_set_safe_registers(struct radeon_device *rdev); - +extern void avivo_wait_for_vblank(struct radeon_device *rdev, int crtc); +extern int rs600_mc_wait_for_idle(struct radeon_device *rdev); /* * rs690,rs740 @@ -251,6 +256,7 @@ void rs690_bandwidth_update(struct radeon_device *rdev); void rs690_line_buffer_adjust(struct radeon_device *rdev, struct drm_display_mode *mode1, struct drm_display_mode *mode2); +extern int rs690_mc_wait_for_idle(struct radeon_device *rdev); /* * rv515 @@ -267,7 +273,7 @@ int rv515_init(struct radeon_device *rdev); void rv515_fini(struct radeon_device *rdev); uint32_t rv515_mc_rreg(struct radeon_device *rdev, uint32_t reg); void rv515_mc_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v); -void rv515_ring_start(struct radeon_device *rdev); +void rv515_ring_start(struct radeon_device *rdev, struct radeon_ring *ring); void rv515_bandwidth_update(struct radeon_device *rdev); int rv515_resume(struct radeon_device *rdev); int rv515_suspend(struct radeon_device *rdev); @@ -278,13 +284,14 @@ void rv515_mc_stop(struct radeon_device *rdev, struct rv515_mc_save *save); void rv515_mc_resume(struct radeon_device *rdev, struct rv515_mc_save *save); void rv515_clock_startup(struct radeon_device *rdev); void rv515_debugfs(struct radeon_device *rdev); - +int rv515_mc_wait_for_idle(struct radeon_device *rdev); /* * r520,rv530,rv560,rv570,r580 */ int r520_init(struct radeon_device *rdev); int r520_resume(struct radeon_device *rdev); +int r520_mc_wait_for_idle(struct radeon_device *rdev); /* * r600,rv610,rv630,rv620,rv635,rv670,rs780,rs880 @@ -312,7 +319,7 @@ int r600_set_surface_reg(struct radeon_device *rdev, int reg, uint32_t tiling_flags, uint32_t pitch, uint32_t offset, uint32_t obj_size); void r600_clear_surface_reg(struct radeon_device *rdev, int reg); -int r600_ib_test(struct radeon_device *rdev, int ring); +int r600_ib_test(struct radeon_device *rdev, struct radeon_ring *ring); void r600_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib); int r600_ring_test(struct radeon_device *rdev, struct radeon_ring *cp); int r600_copy_blit(struct radeon_device *rdev, @@ -375,6 +382,7 @@ void r600_blit_done_copy(struct radeon_device *rdev, struct radeon_fence *fence) void r600_kms_blit_copy(struct radeon_device *rdev, u64 src_gpu_addr, u64 dst_gpu_addr, unsigned num_gpu_pages); +int r600_mc_wait_for_idle(struct radeon_device *rdev); /* * rv770,rv730,rv710,rv740 @@ -423,8 +431,10 @@ extern void sumo_pm_init_profile(struct radeon_device *rdev); extern void evergreen_pre_page_flip(struct radeon_device *rdev, int crtc); extern u32 evergreen_page_flip(struct radeon_device *rdev, int crtc, u64 crtc_base); extern void evergreen_post_page_flip(struct radeon_device *rdev, int crtc); +extern void dce4_wait_for_vblank(struct radeon_device *rdev, int crtc); void evergreen_disable_interrupt_state(struct radeon_device *rdev); int evergreen_blit_init(struct radeon_device *rdev); +int evergreen_mc_wait_for_idle(struct radeon_device *rdev); /* * cayman diff --git a/drivers/gpu/drm/radeon/radeon_atombios.c b/drivers/gpu/drm/radeon/radeon_atombios.c index 1f53ae7..7354137 100644 --- a/drivers/gpu/drm/radeon/radeon_atombios.c +++ b/drivers/gpu/drm/radeon/radeon_atombios.c @@ -442,6 +442,20 @@ static bool radeon_atom_apply_quirks(struct drm_device *dev, struct radeon_device *rdev = dev->dev_private; *i2c_bus = radeon_lookup_i2c_gpio(rdev, 0x93); } + + /* Fujitsu D3003-S2 board lists DVI-I as DVI-D and VGA */ + if ((dev->pdev->device == 0x9802) && + (dev->pdev->subsystem_vendor == 0x1734) && + (dev->pdev->subsystem_device == 0x11bd)) { + if (*connector_type == DRM_MODE_CONNECTOR_VGA) { + *connector_type = DRM_MODE_CONNECTOR_DVII; + *line_mux = 0x3103; + } else if (*connector_type == DRM_MODE_CONNECTOR_DVID) { + *connector_type = DRM_MODE_CONNECTOR_DVII; + } + } + + return true; } diff --git a/drivers/gpu/drm/radeon/radeon_benchmark.c b/drivers/gpu/drm/radeon/radeon_benchmark.c index 815f234..fef7b72 100644 --- a/drivers/gpu/drm/radeon/radeon_benchmark.c +++ b/drivers/gpu/drm/radeon/radeon_benchmark.c @@ -43,17 +43,19 @@ static int radeon_benchmark_do_move(struct radeon_device *rdev, unsigned size, start_jiffies = jiffies; for (i = 0; i < n; i++) { - r = radeon_fence_create(rdev, &fence, RADEON_RING_TYPE_GFX_INDEX); - if (r) - return r; - switch (flag) { case RADEON_BENCHMARK_COPY_DMA: + r = radeon_fence_create(rdev, &fence, radeon_copy_dma_ring_index(rdev)); + if (r) + return r; r = radeon_copy_dma(rdev, saddr, daddr, size / RADEON_GPU_PAGE_SIZE, fence); break; case RADEON_BENCHMARK_COPY_BLIT: + r = radeon_fence_create(rdev, &fence, radeon_copy_blit_ring_index(rdev)); + if (r) + return r; r = radeon_copy_blit(rdev, saddr, daddr, size / RADEON_GPU_PAGE_SIZE, fence); @@ -129,7 +131,7 @@ static void radeon_benchmark_move(struct radeon_device *rdev, unsigned size, /* r100 doesn't have dma engine so skip the test */ /* also, VRAM-to-VRAM test doesn't make much sense for DMA */ /* skip it as well if domains are the same */ - if ((rdev->asic->copy_dma) && (sdomain != ddomain)) { + if ((rdev->asic->copy.dma) && (sdomain != ddomain)) { time = radeon_benchmark_do_move(rdev, size, saddr, daddr, RADEON_BENCHMARK_COPY_DMA, n); if (time < 0) @@ -208,22 +210,22 @@ void radeon_benchmark(struct radeon_device *rdev, int test_number) break; case 3: /* GTT to VRAM, buffer size sweep, powers of 2 */ - for (i = 1; i <= 65536; i <<= 1) - radeon_benchmark_move(rdev, i*1024, + for (i = 1; i <= 16384; i <<= 1) + radeon_benchmark_move(rdev, i * RADEON_GPU_PAGE_SIZE, RADEON_GEM_DOMAIN_GTT, RADEON_GEM_DOMAIN_VRAM); break; case 4: /* VRAM to GTT, buffer size sweep, powers of 2 */ - for (i = 1; i <= 65536; i <<= 1) - radeon_benchmark_move(rdev, i*1024, + for (i = 1; i <= 16384; i <<= 1) + radeon_benchmark_move(rdev, i * RADEON_GPU_PAGE_SIZE, RADEON_GEM_DOMAIN_VRAM, RADEON_GEM_DOMAIN_GTT); break; case 5: /* VRAM to VRAM, buffer size sweep, powers of 2 */ - for (i = 1; i <= 65536; i <<= 1) - radeon_benchmark_move(rdev, i*1024, + for (i = 1; i <= 16384; i <<= 1) + radeon_benchmark_move(rdev, i * RADEON_GPU_PAGE_SIZE, RADEON_GEM_DOMAIN_VRAM, RADEON_GEM_DOMAIN_VRAM); break; diff --git a/drivers/gpu/drm/radeon/radeon_blit_common.h b/drivers/gpu/drm/radeon/radeon_blit_common.h new file mode 100644 index 0000000..4ecbe72 --- /dev/null +++ b/drivers/gpu/drm/radeon/radeon_blit_common.h @@ -0,0 +1,44 @@ +/* + * Copyright 2009 Advanced Micro Devices, Inc. + * Copyright 2009 Red Hat Inc. + * Copyright 2012 Alcatel-Lucent, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (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 COPYRIGHT HOLDER(S) AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER + * DEALINGS IN THE SOFTWARE. + * + */ + +#ifndef __RADEON_BLIT_COMMON_H__ + +#define DI_PT_RECTLIST 0x11 +#define DI_INDEX_SIZE_16_BIT 0x0 +#define DI_SRC_SEL_AUTO_INDEX 0x2 + +#define FMT_8 0x1 +#define FMT_5_6_5 0x8 +#define FMT_8_8_8_8 0x1a +#define COLOR_8 0x1 +#define COLOR_5_6_5 0x8 +#define COLOR_8_8_8_8 0x1a + +#define RECT_UNIT_H 32 +#define RECT_UNIT_W (RADEON_GPU_PAGE_SIZE / 4 / RECT_UNIT_H) + +#define __RADEON_BLIT_COMMON_H__ +#endif diff --git a/drivers/gpu/drm/radeon/radeon_clocks.c b/drivers/gpu/drm/radeon/radeon_clocks.c index b6e18c8..6ae0c75 100644 --- a/drivers/gpu/drm/radeon/radeon_clocks.c +++ b/drivers/gpu/drm/radeon/radeon_clocks.c @@ -334,7 +334,7 @@ void radeon_get_clock_info(struct drm_device *dev) if (!rdev->clock.default_sclk) rdev->clock.default_sclk = radeon_get_engine_clock(rdev); - if ((!rdev->clock.default_mclk) && rdev->asic->get_memory_clock) + if ((!rdev->clock.default_mclk) && rdev->asic->pm.get_memory_clock) rdev->clock.default_mclk = radeon_get_memory_clock(rdev); rdev->pm.current_sclk = rdev->clock.default_sclk; diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c index 8c9a811..64774ac 100644 --- a/drivers/gpu/drm/radeon/radeon_connectors.c +++ b/drivers/gpu/drm/radeon/radeon_connectors.c @@ -827,6 +827,27 @@ static int radeon_dvi_get_modes(struct drm_connector *connector) return ret; } +static bool radeon_check_hpd_status_unchanged(struct drm_connector *connector) +{ + struct drm_device *dev = connector->dev; + struct radeon_device *rdev = dev->dev_private; + struct radeon_connector *radeon_connector = to_radeon_connector(connector); + enum drm_connector_status status; + + /* We only trust HPD on R600 and newer ASICS. */ + if (rdev->family >= CHIP_R600 + && radeon_connector->hpd.hpd != RADEON_HPD_NONE) { + if (radeon_hpd_sense(rdev, radeon_connector->hpd.hpd)) + status = connector_status_connected; + else + status = connector_status_disconnected; + if (connector->status == status) + return true; + } + + return false; +} + /* * DVI is complicated * Do a DDC probe, if DDC probe passes, get the full EDID so @@ -851,6 +872,9 @@ radeon_dvi_detect(struct drm_connector *connector, bool force) enum drm_connector_status ret = connector_status_disconnected; bool dret = false; + if (!force && radeon_check_hpd_status_unchanged(connector)) + return connector->status; + if (radeon_connector->ddc_bus) dret = radeon_ddc_probe(radeon_connector); if (dret) { @@ -946,6 +970,10 @@ radeon_dvi_detect(struct drm_connector *connector, bool force) encoder = obj_to_encoder(obj); + if (encoder->encoder_type != DRM_MODE_ENCODER_DAC || + encoder->encoder_type != DRM_MODE_ENCODER_TVDAC) + continue; + encoder_funcs = encoder->helper_private; if (encoder_funcs->detect) { if (ret != connector_status_connected) { @@ -1250,6 +1278,9 @@ radeon_dp_detect(struct drm_connector *connector, bool force) struct radeon_connector_atom_dig *radeon_dig_connector = radeon_connector->con_priv; struct drm_encoder *encoder = radeon_best_single_encoder(connector); + if (!force && radeon_check_hpd_status_unchanged(connector)) + return connector->status; + if (radeon_connector->edid) { kfree(radeon_connector->edid); radeon_connector->edid = NULL; diff --git a/drivers/gpu/drm/radeon/radeon_cp.c b/drivers/gpu/drm/radeon/radeon_cp.c index 72ae826..0ebb7d47 100644 --- a/drivers/gpu/drm/radeon/radeon_cp.c +++ b/drivers/gpu/drm/radeon/radeon_cp.c @@ -2115,6 +2115,8 @@ int radeon_driver_load(struct drm_device *dev, unsigned long flags) break; } + pci_set_master(dev->pdev); + if (drm_pci_device_is_agp(dev)) dev_priv->flags |= RADEON_IS_AGP; else if (pci_is_pcie(dev->pdev)) diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c index e64bec4..d9d9f5a 100644 --- a/drivers/gpu/drm/radeon/radeon_cs.c +++ b/drivers/gpu/drm/radeon/radeon_cs.c @@ -85,12 +85,6 @@ int radeon_cs_parser_relocs(struct radeon_cs_parser *p) radeon_bo_list_add_object(&p->relocs[i].lobj, &p->validated); - if (p->relocs[i].robj->tbo.sync_obj && !(r->flags & RADEON_RELOC_DONT_SYNC)) { - struct radeon_fence *fence = p->relocs[i].robj->tbo.sync_obj; - if (!radeon_fence_signaled(fence)) { - p->sync_to_ring[fence->ring] = true; - } - } } else p->relocs[i].handle = 0; } @@ -118,11 +112,24 @@ static int radeon_cs_get_ring(struct radeon_cs_parser *p, u32 ring, s32 priority static int radeon_cs_sync_rings(struct radeon_cs_parser *p) { + bool sync_to_ring[RADEON_NUM_RINGS] = { }; int i, r; + for (i = 0; i < p->nrelocs; i++) { + if (!p->relocs[i].robj || !p->relocs[i].robj->tbo.sync_obj) + continue; + + if (!(p->relocs[i].flags & RADEON_RELOC_DONT_SYNC)) { + struct radeon_fence *fence = p->relocs[i].robj->tbo.sync_obj; + if (!radeon_fence_signaled(fence)) { + sync_to_ring[fence->ring] = true; + } + } + } + for (i = 0; i < RADEON_NUM_RINGS; ++i) { /* no need to sync to our own or unused rings */ - if (i == p->ring || !p->sync_to_ring[i] || !p->rdev->ring[i].ready) + if (i == p->ring || !sync_to_ring[i] || !p->rdev->ring[i].ready) continue; if (!p->ib->fence->semaphore) { @@ -236,20 +243,11 @@ int radeon_cs_parser_init(struct radeon_cs_parser *p, void *data) if ((p->cs_flags & RADEON_CS_USE_VM) && !p->rdev->vm_manager.enabled) { DRM_ERROR("VM not active on asic!\n"); - if (p->chunk_relocs_idx != -1) - kfree(p->chunks[p->chunk_relocs_idx].kdata); - if (p->chunk_flags_idx != -1) - kfree(p->chunks[p->chunk_flags_idx].kdata); return -EINVAL; } - if (radeon_cs_get_ring(p, ring, priority)) { - if (p->chunk_relocs_idx != -1) - kfree(p->chunks[p->chunk_relocs_idx].kdata); - if (p->chunk_flags_idx != -1) - kfree(p->chunks[p->chunk_flags_idx].kdata); + if (radeon_cs_get_ring(p, ring, priority)) return -EINVAL; - } /* deal with non-vm */ @@ -264,11 +262,8 @@ int radeon_cs_parser_init(struct radeon_cs_parser *p, void *data) p->chunks[p->chunk_ib_idx].kpage[0] = kmalloc(PAGE_SIZE, GFP_KERNEL); p->chunks[p->chunk_ib_idx].kpage[1] = kmalloc(PAGE_SIZE, GFP_KERNEL); if (p->chunks[p->chunk_ib_idx].kpage[0] == NULL || - p->chunks[p->chunk_ib_idx].kpage[1] == NULL) { - kfree(p->chunks[p->chunk_ib_idx].kpage[0]); - kfree(p->chunks[p->chunk_ib_idx].kpage[1]); + p->chunks[p->chunk_ib_idx].kpage[1] == NULL) return -ENOMEM; - } p->chunks[p->chunk_ib_idx].kpage_idx[0] = -1; p->chunks[p->chunk_ib_idx].kpage_idx[1] = -1; p->chunks[p->chunk_ib_idx].last_copied_page = -1; @@ -341,7 +336,7 @@ static int radeon_cs_ib_chunk(struct radeon_device *rdev, return r; } parser->ib->length_dw = ib_chunk->length_dw; - r = radeon_cs_parse(parser); + r = radeon_cs_parse(rdev, parser->ring, parser); if (r || parser->parser_error) { DRM_ERROR("Invalid command stream !\n"); return r; diff --git a/drivers/gpu/drm/radeon/radeon_cursor.c b/drivers/gpu/drm/radeon/radeon_cursor.c index fde25c0..42acc64 100644 --- a/drivers/gpu/drm/radeon/radeon_cursor.c +++ b/drivers/gpu/drm/radeon/radeon_cursor.c @@ -151,7 +151,9 @@ int radeon_crtc_cursor_set(struct drm_crtc *crtc, uint32_t height) { struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc); + struct radeon_device *rdev = crtc->dev->dev_private; struct drm_gem_object *obj; + struct radeon_bo *robj; uint64_t gpu_addr; int ret; @@ -173,7 +175,15 @@ int radeon_crtc_cursor_set(struct drm_crtc *crtc, return -ENOENT; } - ret = radeon_gem_object_pin(obj, RADEON_GEM_DOMAIN_VRAM, &gpu_addr); + robj = gem_to_radeon_bo(obj); + ret = radeon_bo_reserve(robj, false); + if (unlikely(ret != 0)) + goto fail; + /* Only 27 bit offset for legacy cursor */ + ret = radeon_bo_pin_restricted(robj, RADEON_GEM_DOMAIN_VRAM, + ASIC_IS_AVIVO(rdev) ? 0 : 1 << 27, + &gpu_addr); + radeon_bo_unreserve(robj); if (ret) goto fail; @@ -181,14 +191,18 @@ int radeon_crtc_cursor_set(struct drm_crtc *crtc, radeon_crtc->cursor_height = height; radeon_lock_cursor(crtc, true); - /* XXX only 27 bit offset for legacy cursor */ radeon_set_cursor(crtc, obj, gpu_addr); radeon_show_cursor(crtc); radeon_lock_cursor(crtc, false); unpin: if (radeon_crtc->cursor_bo) { - radeon_gem_object_unpin(radeon_crtc->cursor_bo); + robj = gem_to_radeon_bo(radeon_crtc->cursor_bo); + ret = radeon_bo_reserve(robj, false); + if (likely(ret == 0)) { + radeon_bo_unpin(robj); + radeon_bo_unreserve(robj); + } drm_gem_object_unreference_unlocked(radeon_crtc->cursor_bo); } diff --git a/drivers/gpu/drm/radeon/radeon_display.c b/drivers/gpu/drm/radeon/radeon_display.c index 3d31433..1ebcef2 100644 --- a/drivers/gpu/drm/radeon/radeon_display.c +++ b/drivers/gpu/drm/radeon/radeon_display.c @@ -303,8 +303,17 @@ void radeon_crtc_handle_flip(struct radeon_device *rdev, int crtc_id) if (update_pending && (DRM_SCANOUTPOS_VALID & radeon_get_crtc_scanoutpos(rdev->ddev, crtc_id, &vpos, &hpos)) && - (vpos >=0) && - (vpos < (99 * rdev->mode_info.crtcs[crtc_id]->base.hwmode.crtc_vdisplay)/100)) { + ((vpos >= (99 * rdev->mode_info.crtcs[crtc_id]->base.hwmode.crtc_vdisplay)/100) || + (vpos < 0 && !ASIC_IS_AVIVO(rdev)))) { + /* crtc didn't flip in this target vblank interval, + * but flip is pending in crtc. Based on the current + * scanout position we know that the current frame is + * (nearly) complete and the flip will (likely) + * complete before the start of the next frame. + */ + update_pending = 0; + } + if (update_pending) { /* crtc didn't flip in this target vblank interval, * but flip is pending in crtc. It will complete it * in next vblank interval, so complete the flip at @@ -393,7 +402,9 @@ static int radeon_crtc_page_flip(struct drm_crtc *crtc, DRM_ERROR("failed to reserve new rbo buffer before flip\n"); goto pflip_cleanup; } - r = radeon_bo_pin(rbo, RADEON_GEM_DOMAIN_VRAM, &base); + /* Only 27 bit offset for legacy CRTC */ + r = radeon_bo_pin_restricted(rbo, RADEON_GEM_DOMAIN_VRAM, + ASIC_IS_AVIVO(rdev) ? 0 : 1 << 27, &base); if (unlikely(r != 0)) { radeon_bo_unreserve(rbo); r = -EINVAL; @@ -1136,11 +1147,6 @@ static const struct drm_mode_config_funcs radeon_mode_funcs = { .output_poll_changed = radeon_output_poll_changed }; -struct drm_prop_enum_list { - int type; - char *name; -}; - static struct drm_prop_enum_list radeon_tmds_pll_enum_list[] = { { 0, "driver" }, { 1, "bios" }, @@ -1165,86 +1171,53 @@ static struct drm_prop_enum_list radeon_underscan_enum_list[] = static int radeon_modeset_create_props(struct radeon_device *rdev) { - int i, sz; + int sz; if (rdev->is_atom_bios) { rdev->mode_info.coherent_mode_property = - drm_property_create(rdev->ddev, - DRM_MODE_PROP_RANGE, - "coherent", 2); + drm_property_create_range(rdev->ddev, 0 , "coherent", 0, 1); if (!rdev->mode_info.coherent_mode_property) return -ENOMEM; - - rdev->mode_info.coherent_mode_property->values[0] = 0; - rdev->mode_info.coherent_mode_property->values[1] = 1; } if (!ASIC_IS_AVIVO(rdev)) { sz = ARRAY_SIZE(radeon_tmds_pll_enum_list); rdev->mode_info.tmds_pll_property = - drm_property_create(rdev->ddev, - DRM_MODE_PROP_ENUM, - "tmds_pll", sz); - for (i = 0; i < sz; i++) { - drm_property_add_enum(rdev->mode_info.tmds_pll_property, - i, - radeon_tmds_pll_enum_list[i].type, - radeon_tmds_pll_enum_list[i].name); - } + drm_property_create_enum(rdev->ddev, 0, + "tmds_pll", + radeon_tmds_pll_enum_list, sz); } rdev->mode_info.load_detect_property = - drm_property_create(rdev->ddev, - DRM_MODE_PROP_RANGE, - "load detection", 2); + drm_property_create_range(rdev->ddev, 0, "load detection", 0, 1); if (!rdev->mode_info.load_detect_property) return -ENOMEM; - rdev->mode_info.load_detect_property->values[0] = 0; - rdev->mode_info.load_detect_property->values[1] = 1; drm_mode_create_scaling_mode_property(rdev->ddev); sz = ARRAY_SIZE(radeon_tv_std_enum_list); rdev->mode_info.tv_std_property = - drm_property_create(rdev->ddev, - DRM_MODE_PROP_ENUM, - "tv standard", sz); - for (i = 0; i < sz; i++) { - drm_property_add_enum(rdev->mode_info.tv_std_property, - i, - radeon_tv_std_enum_list[i].type, - radeon_tv_std_enum_list[i].name); - } + drm_property_create_enum(rdev->ddev, 0, + "tv standard", + radeon_tv_std_enum_list, sz); sz = ARRAY_SIZE(radeon_underscan_enum_list); rdev->mode_info.underscan_property = - drm_property_create(rdev->ddev, - DRM_MODE_PROP_ENUM, - "underscan", sz); - for (i = 0; i < sz; i++) { - drm_property_add_enum(rdev->mode_info.underscan_property, - i, - radeon_underscan_enum_list[i].type, - radeon_underscan_enum_list[i].name); - } + drm_property_create_enum(rdev->ddev, 0, + "underscan", + radeon_underscan_enum_list, sz); rdev->mode_info.underscan_hborder_property = - drm_property_create(rdev->ddev, - DRM_MODE_PROP_RANGE, - "underscan hborder", 2); + drm_property_create_range(rdev->ddev, 0, + "underscan hborder", 0, 128); if (!rdev->mode_info.underscan_hborder_property) return -ENOMEM; - rdev->mode_info.underscan_hborder_property->values[0] = 0; - rdev->mode_info.underscan_hborder_property->values[1] = 128; rdev->mode_info.underscan_vborder_property = - drm_property_create(rdev->ddev, - DRM_MODE_PROP_RANGE, - "underscan vborder", 2); + drm_property_create_range(rdev->ddev, 0, + "underscan vborder", 0, 128); if (!rdev->mode_info.underscan_vborder_property) return -ENOMEM; - rdev->mode_info.underscan_vborder_property->values[0] = 0; - rdev->mode_info.underscan_vborder_property->values[1] = 128; return 0; } @@ -1290,6 +1263,9 @@ int radeon_modeset_init(struct radeon_device *rdev) rdev->ddev->mode_config.max_height = 4096; } + rdev->ddev->mode_config.preferred_depth = 24; + rdev->ddev->mode_config.prefer_shadow = 1; + rdev->ddev->mode_config.fb_base = rdev->mc.aper_base; ret = radeon_modeset_create_props(rdev); diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c index 8032f1f..498d21d 100644 --- a/drivers/gpu/drm/radeon/radeon_drv.c +++ b/drivers/gpu/drm/radeon/radeon_drv.c @@ -54,10 +54,11 @@ * 2.10.0 - fusion 2D tiling * 2.11.0 - backend map, initial compute support for the CS checker * 2.12.0 - RADEON_CS_KEEP_TILING_FLAGS - * 2.13.0 - virtual memory support + * 2.13.0 - virtual memory support, streamout + * 2.14.0 - add evergreen tiling informations */ #define KMS_DRIVER_MAJOR 2 -#define KMS_DRIVER_MINOR 13 +#define KMS_DRIVER_MINOR 14 #define KMS_DRIVER_PATCHLEVEL 0 int radeon_driver_load_kms(struct drm_device *dev, unsigned long flags); int radeon_driver_unload_kms(struct drm_device *dev); diff --git a/drivers/gpu/drm/radeon/radeon_fb.c b/drivers/gpu/drm/radeon/radeon_fb.c index 195471c..5906914 100644 --- a/drivers/gpu/drm/radeon/radeon_fb.c +++ b/drivers/gpu/drm/radeon/radeon_fb.c @@ -164,7 +164,10 @@ static int radeonfb_create_pinned_object(struct radeon_fbdev *rfbdev, ret = radeon_bo_reserve(rbo, false); if (unlikely(ret != 0)) goto out_unref; - ret = radeon_bo_pin(rbo, RADEON_GEM_DOMAIN_VRAM, NULL); + /* Only 27 bit offset for legacy CRTC */ + ret = radeon_bo_pin_restricted(rbo, RADEON_GEM_DOMAIN_VRAM, + ASIC_IS_AVIVO(rdev) ? 0 : 1 << 27, + NULL); if (ret) { radeon_bo_unreserve(rbo); goto out_unref; @@ -263,11 +266,7 @@ static int radeonfb_create(struct radeon_fbdev *rfbdev, info->apertures->ranges[0].base = rdev->ddev->mode_config.fb_base; info->apertures->ranges[0].size = rdev->mc.aper_size; - info->pixmap.size = 64*1024; - info->pixmap.buf_align = 8; - info->pixmap.access_align = 32; - info->pixmap.flags = FB_PIXMAP_SYSTEM; - info->pixmap.scan_align = 1; + /* Use default scratch pixmap (info->pixmap.flags = FB_PIXMAP_SYSTEM) */ if (info->screen_base == NULL) { ret = -ENOSPC; diff --git a/drivers/gpu/drm/radeon/radeon_gem.c b/drivers/gpu/drm/radeon/radeon_gem.c index 7337850..c7008b5 100644 --- a/drivers/gpu/drm/radeon/radeon_gem.c +++ b/drivers/gpu/drm/radeon/radeon_gem.c @@ -75,32 +75,6 @@ int radeon_gem_object_create(struct radeon_device *rdev, int size, return 0; } -int radeon_gem_object_pin(struct drm_gem_object *obj, uint32_t pin_domain, - uint64_t *gpu_addr) -{ - struct radeon_bo *robj = gem_to_radeon_bo(obj); - int r; - - r = radeon_bo_reserve(robj, false); - if (unlikely(r != 0)) - return r; - r = radeon_bo_pin(robj, pin_domain, gpu_addr); - radeon_bo_unreserve(robj); - return r; -} - -void radeon_gem_object_unpin(struct drm_gem_object *obj) -{ - struct radeon_bo *robj = gem_to_radeon_bo(obj); - int r; - - r = radeon_bo_reserve(robj, false); - if (likely(r == 0)) { - radeon_bo_unpin(robj); - radeon_bo_unreserve(robj); - } -} - int radeon_gem_set_domain(struct drm_gem_object *gobj, uint32_t rdomain, uint32_t wdomain) { diff --git a/drivers/gpu/drm/radeon/radeon_i2c.c b/drivers/gpu/drm/radeon/radeon_i2c.c index 98a8ad6..85bcfc8 100644 --- a/drivers/gpu/drm/radeon/radeon_i2c.c +++ b/drivers/gpu/drm/radeon/radeon_i2c.c @@ -26,10 +26,15 @@ #include <linux/export.h> #include "drmP.h" +#include "drm_edid.h" #include "radeon_drm.h" #include "radeon.h" #include "atom.h" +extern int radeon_atom_hw_i2c_xfer(struct i2c_adapter *i2c_adap, + struct i2c_msg *msgs, int num); +extern u32 radeon_atom_hw_i2c_func(struct i2c_adapter *adap); + /** * radeon_ddc_probe * @@ -41,13 +46,13 @@ bool radeon_ddc_probe(struct radeon_connector *radeon_connector) int ret; struct i2c_msg msgs[] = { { - .addr = 0x50, + .addr = DDC_ADDR, .flags = 0, .len = 1, .buf = &out, }, { - .addr = 0x50, + .addr = DDC_ADDR, .flags = I2C_M_RD, .len = 8, .buf = buf, @@ -882,6 +887,11 @@ static const struct i2c_algorithm radeon_i2c_algo = { .functionality = radeon_hw_i2c_func, }; +static const struct i2c_algorithm radeon_atom_i2c_algo = { + .master_xfer = radeon_atom_hw_i2c_xfer, + .functionality = radeon_atom_hw_i2c_func, +}; + struct radeon_i2c_chan *radeon_i2c_create(struct drm_device *dev, struct radeon_i2c_bus_rec *rec, const char *name) @@ -914,6 +924,18 @@ struct radeon_i2c_chan *radeon_i2c_create(struct drm_device *dev, DRM_ERROR("Failed to register hw i2c %s\n", name); goto out_free; } + } else if (rec->hw_capable && + radeon_hw_i2c && + ASIC_IS_DCE3(rdev)) { + /* hw i2c using atom */ + snprintf(i2c->adapter.name, sizeof(i2c->adapter.name), + "Radeon i2c hw bus %s", name); + i2c->adapter.algo = &radeon_atom_i2c_algo; + ret = i2c_add_adapter(&i2c->adapter); + if (ret) { + DRM_ERROR("Failed to register hw i2c %s\n", name); + goto out_free; + } } else { /* set the radeon bit adapter */ snprintf(i2c->adapter.name, sizeof(i2c->adapter.name), @@ -925,10 +947,8 @@ struct radeon_i2c_chan *radeon_i2c_create(struct drm_device *dev, i2c->algo.bit.setscl = set_clock; i2c->algo.bit.getsda = get_data; i2c->algo.bit.getscl = get_clock; - i2c->algo.bit.udelay = 20; - /* vesa says 2.2 ms is enough, 1 jiffy doesn't seem to always - * make this, 2 jiffies is a lot more reliable */ - i2c->algo.bit.timeout = 2; + i2c->algo.bit.udelay = 10; + i2c->algo.bit.timeout = usecs_to_jiffies(2200); /* from VESA */ i2c->algo.bit.data = i2c; ret = i2c_bit_add_bus(&i2c->adapter); if (ret) { diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c index d335288..1986eba 100644 --- a/drivers/gpu/drm/radeon/radeon_kms.c +++ b/drivers/gpu/drm/radeon/radeon_kms.c @@ -57,6 +57,8 @@ int radeon_driver_load_kms(struct drm_device *dev, unsigned long flags) } dev->dev_private = (void *)rdev; + pci_set_master(dev->pdev); + /* update BUS flag */ if (drm_pci_device_is_agp(dev)) { flags |= RADEON_IS_AGP; diff --git a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c index 25a19c4..210317c 100644 --- a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c +++ b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c @@ -419,7 +419,9 @@ int radeon_crtc_do_set_base(struct drm_crtc *crtc, r = radeon_bo_reserve(rbo, false); if (unlikely(r != 0)) return r; - r = radeon_bo_pin(rbo, RADEON_GEM_DOMAIN_VRAM, &base); + /* Only 27 bit offset for legacy CRTC */ + r = radeon_bo_pin_restricted(rbo, RADEON_GEM_DOMAIN_VRAM, 1 << 27, + &base); if (unlikely(r != 0)) { radeon_bo_unreserve(rbo); return -EINVAL; diff --git a/drivers/gpu/drm/radeon/radeon_object.c b/drivers/gpu/drm/radeon/radeon_object.c index d45df17..91541e6 100644 --- a/drivers/gpu/drm/radeon/radeon_object.c +++ b/drivers/gpu/drm/radeon/radeon_object.c @@ -224,7 +224,8 @@ void radeon_bo_unref(struct radeon_bo **bo) *bo = NULL; } -int radeon_bo_pin(struct radeon_bo *bo, u32 domain, u64 *gpu_addr) +int radeon_bo_pin_restricted(struct radeon_bo *bo, u32 domain, u64 max_offset, + u64 *gpu_addr) { int r, i; @@ -232,6 +233,7 @@ int radeon_bo_pin(struct radeon_bo *bo, u32 domain, u64 *gpu_addr) bo->pin_count++; if (gpu_addr) *gpu_addr = radeon_bo_gpu_offset(bo); + WARN_ON_ONCE(max_offset != 0); return 0; } radeon_ttm_placement_from_domain(bo, domain); @@ -239,6 +241,15 @@ int radeon_bo_pin(struct radeon_bo *bo, u32 domain, u64 *gpu_addr) /* force to pin into visible video ram */ bo->placement.lpfn = bo->rdev->mc.visible_vram_size >> PAGE_SHIFT; } + if (max_offset) { + u64 lpfn = max_offset >> PAGE_SHIFT; + + if (!bo->placement.lpfn) + bo->placement.lpfn = bo->rdev->mc.gtt_size >> PAGE_SHIFT; + + if (lpfn < bo->placement.lpfn) + bo->placement.lpfn = lpfn; + } for (i = 0; i < bo->placement.num_placement; i++) bo->placements[i] |= TTM_PL_FLAG_NO_EVICT; r = ttm_bo_validate(&bo->tbo, &bo->placement, false, false, false); @@ -252,6 +263,11 @@ int radeon_bo_pin(struct radeon_bo *bo, u32 domain, u64 *gpu_addr) return r; } +int radeon_bo_pin(struct radeon_bo *bo, u32 domain, u64 *gpu_addr) +{ + return radeon_bo_pin_restricted(bo, domain, 0, gpu_addr); +} + int radeon_bo_unpin(struct radeon_bo *bo) { int r, i; @@ -445,8 +461,54 @@ static void radeon_bo_clear_surface_reg(struct radeon_bo *bo) int radeon_bo_set_tiling_flags(struct radeon_bo *bo, uint32_t tiling_flags, uint32_t pitch) { + struct radeon_device *rdev = bo->rdev; int r; + if (rdev->family >= CHIP_CEDAR) { + unsigned bankw, bankh, mtaspect, tilesplit, stilesplit; + + bankw = (tiling_flags >> RADEON_TILING_EG_BANKW_SHIFT) & RADEON_TILING_EG_BANKW_MASK; + bankh = (tiling_flags >> RADEON_TILING_EG_BANKH_SHIFT) & RADEON_TILING_EG_BANKH_MASK; + mtaspect = (tiling_flags >> RADEON_TILING_EG_MACRO_TILE_ASPECT_SHIFT) & RADEON_TILING_EG_MACRO_TILE_ASPECT_MASK; + tilesplit = (tiling_flags >> RADEON_TILING_EG_TILE_SPLIT_SHIFT) & RADEON_TILING_EG_TILE_SPLIT_MASK; + stilesplit = (tiling_flags >> RADEON_TILING_EG_STENCIL_TILE_SPLIT_SHIFT) & RADEON_TILING_EG_STENCIL_TILE_SPLIT_MASK; + switch (bankw) { + case 0: + case 1: + case 2: + case 4: + case 8: + break; + default: + return -EINVAL; + } + switch (bankh) { + case 0: + case 1: + case 2: + case 4: + case 8: + break; + default: + return -EINVAL; + } + switch (mtaspect) { + case 0: + case 1: + case 2: + case 4: + case 8: + break; + default: + return -EINVAL; + } + if (tilesplit > 6) { + return -EINVAL; + } + if (stilesplit > 6) { + return -EINVAL; + } + } r = radeon_bo_reserve(bo, false); if (unlikely(r != 0)) return r; diff --git a/drivers/gpu/drm/radeon/radeon_object.h b/drivers/gpu/drm/radeon/radeon_object.h index cde4303..f9104be 100644 --- a/drivers/gpu/drm/radeon/radeon_object.h +++ b/drivers/gpu/drm/radeon/radeon_object.h @@ -118,6 +118,8 @@ extern int radeon_bo_kmap(struct radeon_bo *bo, void **ptr); extern void radeon_bo_kunmap(struct radeon_bo *bo); extern void radeon_bo_unref(struct radeon_bo **bo); extern int radeon_bo_pin(struct radeon_bo *bo, u32 domain, u64 *gpu_addr); +extern int radeon_bo_pin_restricted(struct radeon_bo *bo, u32 domain, + u64 max_offset, u64 *gpu_addr); extern int radeon_bo_unpin(struct radeon_bo *bo); extern int radeon_bo_evict_vram(struct radeon_device *rdev); extern void radeon_bo_force_delete(struct radeon_device *rdev); diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c index 095148e..3575129 100644 --- a/drivers/gpu/drm/radeon/radeon_pm.c +++ b/drivers/gpu/drm/radeon/radeon_pm.c @@ -221,7 +221,7 @@ static void radeon_set_power_state(struct radeon_device *rdev) } /* set memory clock */ - if (rdev->asic->set_memory_clock && (mclk != rdev->pm.current_mclk)) { + if (rdev->asic->pm.set_memory_clock && (mclk != rdev->pm.current_mclk)) { radeon_pm_debug_check_in_vbl(rdev, false); radeon_set_memory_clock(rdev, mclk); radeon_pm_debug_check_in_vbl(rdev, true); @@ -863,11 +863,11 @@ static int radeon_debugfs_pm_info(struct seq_file *m, void *data) seq_printf(m, "default engine clock: %u0 kHz\n", rdev->pm.default_sclk); seq_printf(m, "current engine clock: %u0 kHz\n", radeon_get_engine_clock(rdev)); seq_printf(m, "default memory clock: %u0 kHz\n", rdev->pm.default_mclk); - if (rdev->asic->get_memory_clock) + if (rdev->asic->pm.get_memory_clock) seq_printf(m, "current memory clock: %u0 kHz\n", radeon_get_memory_clock(rdev)); if (rdev->pm.current_vddc) seq_printf(m, "voltage: %u mV\n", rdev->pm.current_vddc); - if (rdev->asic->get_pcie_lanes) + if (rdev->asic->pm.get_pcie_lanes) seq_printf(m, "PCIE lanes: %d\n", radeon_get_pcie_lanes(rdev)); return 0; diff --git a/drivers/gpu/drm/radeon/radeon_reg.h b/drivers/gpu/drm/radeon/radeon_reg.h index b4ce864..5098634 100644 --- a/drivers/gpu/drm/radeon/radeon_reg.h +++ b/drivers/gpu/drm/radeon/radeon_reg.h @@ -539,9 +539,11 @@ #define RADEON_CRTC2_PITCH 0x032c #define RADEON_CRTC_STATUS 0x005c +# define RADEON_CRTC_VBLANK_CUR (1 << 0) # define RADEON_CRTC_VBLANK_SAVE (1 << 1) # define RADEON_CRTC_VBLANK_SAVE_CLEAR (1 << 1) #define RADEON_CRTC2_STATUS 0x03fc +# define RADEON_CRTC2_VBLANK_CUR (1 << 0) # define RADEON_CRTC2_VBLANK_SAVE (1 << 1) # define RADEON_CRTC2_VBLANK_SAVE_CLEAR (1 << 1) #define RADEON_CRTC_V_SYNC_STRT_WID 0x020c diff --git a/drivers/gpu/drm/radeon/radeon_ring.c b/drivers/gpu/drm/radeon/radeon_ring.c index 92c9ea4..3056620 100644 --- a/drivers/gpu/drm/radeon/radeon_ring.c +++ b/drivers/gpu/drm/radeon/radeon_ring.c @@ -478,7 +478,9 @@ static struct drm_info_list radeon_debugfs_ring_info_list[] = { static int radeon_debugfs_ib_info(struct seq_file *m, void *data) { struct drm_info_node *node = (struct drm_info_node *) m->private; - struct radeon_ib *ib = node->info_ent->data; + struct drm_device *dev = node->minor->dev; + struct radeon_device *rdev = dev->dev_private; + struct radeon_ib *ib = &rdev->ib_pool.ibs[*((unsigned*)node->info_ent->data)]; unsigned i; if (ib == NULL) { @@ -495,6 +497,7 @@ static int radeon_debugfs_ib_info(struct seq_file *m, void *data) static struct drm_info_list radeon_debugfs_ib_list[RADEON_IB_POOL_SIZE]; static char radeon_debugfs_ib_names[RADEON_IB_POOL_SIZE][32]; +static unsigned radeon_debugfs_ib_idx[RADEON_IB_POOL_SIZE]; #endif int radeon_debugfs_ring_init(struct radeon_device *rdev) @@ -517,10 +520,11 @@ int radeon_debugfs_ib_init(struct radeon_device *rdev) for (i = 0; i < RADEON_IB_POOL_SIZE; i++) { sprintf(radeon_debugfs_ib_names[i], "radeon_ib_%04u", i); + radeon_debugfs_ib_idx[i] = i; radeon_debugfs_ib_list[i].name = radeon_debugfs_ib_names[i]; radeon_debugfs_ib_list[i].show = &radeon_debugfs_ib_info; radeon_debugfs_ib_list[i].driver_features = 0; - radeon_debugfs_ib_list[i].data = &rdev->ib_pool.ibs[i]; + radeon_debugfs_ib_list[i].data = &radeon_debugfs_ib_idx[i]; } return radeon_debugfs_add_files(rdev, radeon_debugfs_ib_list, RADEON_IB_POOL_SIZE); diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c index c421e77..f493c64 100644 --- a/drivers/gpu/drm/radeon/radeon_ttm.c +++ b/drivers/gpu/drm/radeon/radeon_ttm.c @@ -226,7 +226,7 @@ static int radeon_move_blit(struct ttm_buffer_object *bo, int r, i; rdev = radeon_get_rdev(bo->bdev); - r = radeon_fence_create(rdev, &fence, rdev->copy_ring); + r = radeon_fence_create(rdev, &fence, radeon_copy_ring_index(rdev)); if (unlikely(r)) { return r; } @@ -255,7 +255,7 @@ static int radeon_move_blit(struct ttm_buffer_object *bo, DRM_ERROR("Unknown placement %d\n", old_mem->mem_type); return -EINVAL; } - if (!rdev->ring[rdev->copy_ring].ready) { + if (!rdev->ring[radeon_copy_ring_index(rdev)].ready) { DRM_ERROR("Trying to move memory with ring turned off.\n"); return -EINVAL; } @@ -266,7 +266,7 @@ static int radeon_move_blit(struct ttm_buffer_object *bo, if (rdev->family >= CHIP_R600) { for (i = 0; i < RADEON_NUM_RINGS; ++i) { /* no need to sync to our own or unused rings */ - if (i == rdev->copy_ring || !rdev->ring[i].ready) + if (i == radeon_copy_ring_index(rdev) || !rdev->ring[i].ready) continue; if (!fence->semaphore) { @@ -283,12 +283,12 @@ static int radeon_move_blit(struct ttm_buffer_object *bo, radeon_semaphore_emit_signal(rdev, i, fence->semaphore); radeon_ring_unlock_commit(rdev, &rdev->ring[i]); - r = radeon_ring_lock(rdev, &rdev->ring[rdev->copy_ring], 3); + r = radeon_ring_lock(rdev, &rdev->ring[radeon_copy_ring_index(rdev)], 3); /* FIXME: handle ring lock error */ if (r) continue; - radeon_semaphore_emit_wait(rdev, rdev->copy_ring, fence->semaphore); - radeon_ring_unlock_commit(rdev, &rdev->ring[rdev->copy_ring]); + radeon_semaphore_emit_wait(rdev, radeon_copy_ring_index(rdev), fence->semaphore); + radeon_ring_unlock_commit(rdev, &rdev->ring[radeon_copy_ring_index(rdev)]); } } @@ -410,7 +410,8 @@ static int radeon_bo_move(struct ttm_buffer_object *bo, radeon_move_null(bo, new_mem); return 0; } - if (!rdev->ring[RADEON_RING_TYPE_GFX_INDEX].ready || rdev->asic->copy == NULL) { + if (!rdev->ring[radeon_copy_ring_index(rdev)].ready || + rdev->asic->copy.copy == NULL) { /* use memcpy */ goto memcpy; } diff --git a/drivers/gpu/drm/radeon/reg_srcs/cayman b/drivers/gpu/drm/radeon/reg_srcs/cayman index 2316977..aea63c4 100644 --- a/drivers/gpu/drm/radeon/reg_srcs/cayman +++ b/drivers/gpu/drm/radeon/reg_srcs/cayman @@ -1,5 +1,8 @@ cayman 0x9400 0x0000802C GRBM_GFX_INDEX +0x000084FC CP_STRMOUT_CNTL +0x000085F0 CP_COHER_CNTL +0x000085F4 CP_COHER_SIZE 0x000088B0 VGT_VTX_VECT_EJECT_REG 0x000088C4 VGT_CACHE_INVALIDATION 0x000088D4 VGT_GS_VERTEX_REUSE @@ -77,7 +80,6 @@ cayman 0x9400 0x0002802C DB_DEPTH_CLEAR 0x00028030 PA_SC_SCREEN_SCISSOR_TL 0x00028034 PA_SC_SCREEN_SCISSOR_BR -0x0002805C DB_DEPTH_SLICE 0x00028140 SQ_ALU_CONST_BUFFER_SIZE_PS_0 0x00028144 SQ_ALU_CONST_BUFFER_SIZE_PS_1 0x00028148 SQ_ALU_CONST_BUFFER_SIZE_PS_2 @@ -206,7 +208,6 @@ cayman 0x9400 0x00028344 PA_SC_VPORT_ZMAX_14 0x00028348 PA_SC_VPORT_ZMIN_15 0x0002834C PA_SC_VPORT_ZMAX_15 -0x00028350 SX_MISC 0x00028354 SX_SURFACE_SYNC 0x0002835C SX_SCATTER_EXPORT_SIZE 0x00028380 SQ_VTX_SEMANTIC_0 @@ -512,6 +513,13 @@ cayman 0x9400 0x00028AC0 DB_SRESULTS_COMPARE_STATE0 0x00028AC4 DB_SRESULTS_COMPARE_STATE1 0x00028AC8 DB_PRELOAD_CONTROL +0x00028AD4 VGT_STRMOUT_VTX_STRIDE_0 +0x00028AE4 VGT_STRMOUT_VTX_STRIDE_1 +0x00028AF4 VGT_STRMOUT_VTX_STRIDE_2 +0x00028B04 VGT_STRMOUT_VTX_STRIDE_3 +0x00028B28 VGT_STRMOUT_DRAW_OPAQUE_OFFSET +0x00028B2C VGT_STRMOUT_DRAW_OPAQUE_BUFFER_FILLED_SIZE +0x00028B30 VGT_STRMOUT_DRAW_OPAQUE_VERTEX_STRIDE 0x00028B38 VGT_GS_MAX_VERT_OUT 0x00028B54 VGT_SHADER_STAGES_EN 0x00028B58 VGT_LS_HS_CONFIG @@ -551,6 +559,18 @@ cayman 0x9400 0x00028C34 PA_SC_AA_SAMPLE_LOCS_PIXEL_X1_Y1_3 0x00028C38 PA_SC_AA_MASK_X0_Y0_X1_Y0 0x00028C3C PA_SC_AA_MASK_X0_Y1_X1_Y1 +0x00028C78 CB_COLOR0_DIM +0x00028CB4 CB_COLOR1_DIM +0x00028CF0 CB_COLOR2_DIM +0x00028D2C CB_COLOR3_DIM +0x00028D68 CB_COLOR4_DIM +0x00028DA4 CB_COLOR5_DIM +0x00028DE0 CB_COLOR6_DIM +0x00028E1C CB_COLOR7_DIM +0x00028E58 CB_COLOR8_DIM +0x00028E74 CB_COLOR9_DIM +0x00028E90 CB_COLOR10_DIM +0x00028EAC CB_COLOR11_DIM 0x00028C8C CB_COLOR0_CLEAR_WORD0 0x00028C90 CB_COLOR0_CLEAR_WORD1 0x00028C94 CB_COLOR0_CLEAR_WORD2 diff --git a/drivers/gpu/drm/radeon/reg_srcs/evergreen b/drivers/gpu/drm/radeon/reg_srcs/evergreen index 161737a..77c3720 100644 --- a/drivers/gpu/drm/radeon/reg_srcs/evergreen +++ b/drivers/gpu/drm/radeon/reg_srcs/evergreen @@ -4,6 +4,9 @@ evergreen 0x9400 0x00008044 WAIT_UNTIL_POLL_CNTL 0x00008048 WAIT_UNTIL_POLL_MASK 0x0000804c WAIT_UNTIL_POLL_REFDATA +0x000084FC CP_STRMOUT_CNTL +0x000085F0 CP_COHER_CNTL +0x000085F4 CP_COHER_SIZE 0x000088B0 VGT_VTX_VECT_EJECT_REG 0x000088C4 VGT_CACHE_INVALIDATION 0x000088D4 VGT_GS_VERTEX_REUSE @@ -93,7 +96,6 @@ evergreen 0x9400 0x0002802C DB_DEPTH_CLEAR 0x00028030 PA_SC_SCREEN_SCISSOR_TL 0x00028034 PA_SC_SCREEN_SCISSOR_BR -0x0002805C DB_DEPTH_SLICE 0x00028140 SQ_ALU_CONST_BUFFER_SIZE_PS_0 0x00028144 SQ_ALU_CONST_BUFFER_SIZE_PS_1 0x00028148 SQ_ALU_CONST_BUFFER_SIZE_PS_2 @@ -222,7 +224,6 @@ evergreen 0x9400 0x00028344 PA_SC_VPORT_ZMAX_14 0x00028348 PA_SC_VPORT_ZMIN_15 0x0002834C PA_SC_VPORT_ZMAX_15 -0x00028350 SX_MISC 0x00028354 SX_SURFACE_SYNC 0x00028380 SQ_VTX_SEMANTIC_0 0x00028384 SQ_VTX_SEMANTIC_1 @@ -522,6 +523,13 @@ evergreen 0x9400 0x00028AC0 DB_SRESULTS_COMPARE_STATE0 0x00028AC4 DB_SRESULTS_COMPARE_STATE1 0x00028AC8 DB_PRELOAD_CONTROL +0x00028AD4 VGT_STRMOUT_VTX_STRIDE_0 +0x00028AE4 VGT_STRMOUT_VTX_STRIDE_1 +0x00028AF4 VGT_STRMOUT_VTX_STRIDE_2 +0x00028B04 VGT_STRMOUT_VTX_STRIDE_3 +0x00028B28 VGT_STRMOUT_DRAW_OPAQUE_OFFSET +0x00028B2C VGT_STRMOUT_DRAW_OPAQUE_BUFFER_FILLED_SIZE +0x00028B30 VGT_STRMOUT_DRAW_OPAQUE_VERTEX_STRIDE 0x00028B38 VGT_GS_MAX_VERT_OUT 0x00028B54 VGT_SHADER_STAGES_EN 0x00028B58 VGT_LS_HS_CONFIG @@ -554,6 +562,18 @@ evergreen 0x9400 0x00028C34 PA_SC_AA_SAMPLE_LOCS_6 0x00028C38 PA_SC_AA_SAMPLE_LOCS_7 0x00028C3C PA_SC_AA_MASK +0x00028C78 CB_COLOR0_DIM +0x00028CB4 CB_COLOR1_DIM +0x00028CF0 CB_COLOR2_DIM +0x00028D2C CB_COLOR3_DIM +0x00028D68 CB_COLOR4_DIM +0x00028DA4 CB_COLOR5_DIM +0x00028DE0 CB_COLOR6_DIM +0x00028E1C CB_COLOR7_DIM +0x00028E58 CB_COLOR8_DIM +0x00028E74 CB_COLOR9_DIM +0x00028E90 CB_COLOR10_DIM +0x00028EAC CB_COLOR11_DIM 0x00028C8C CB_COLOR0_CLEAR_WORD0 0x00028C90 CB_COLOR0_CLEAR_WORD1 0x00028C94 CB_COLOR0_CLEAR_WORD2 diff --git a/drivers/gpu/drm/radeon/reg_srcs/r600 b/drivers/gpu/drm/radeon/reg_srcs/r600 index 0380c5c..626c24e 100644 --- a/drivers/gpu/drm/radeon/reg_srcs/r600 +++ b/drivers/gpu/drm/radeon/reg_srcs/r600 @@ -3,6 +3,9 @@ r600 0x9400 0x00028230 R7xx_PA_SC_EDGERULE 0x000286C8 R7xx_SPI_THREAD_GROUPING 0x00008D8C R7xx_SQ_DYN_GPR_CNTL_PS_FLUSH_REQ +0x00008490 CP_STRMOUT_CNTL +0x000085F0 CP_COHER_CNTL +0x000085F4 CP_COHER_SIZE 0x000088C4 VGT_CACHE_INVALIDATION 0x00028A50 VGT_ENHANCE 0x000088CC VGT_ES_PER_GS @@ -38,6 +41,13 @@ r600 0x9400 0x00028AB4 VGT_REUSE_OFF 0x00028AB8 VGT_VTX_CNT_EN 0x000088B0 VGT_VTX_VECT_EJECT_REG +0x00028AD4 VGT_STRMOUT_VTX_STRIDE_0 +0x00028AE4 VGT_STRMOUT_VTX_STRIDE_1 +0x00028AF4 VGT_STRMOUT_VTX_STRIDE_2 +0x00028B04 VGT_STRMOUT_VTX_STRIDE_3 +0x00028B28 VGT_STRMOUT_DRAW_OPAQUE_OFFSET +0x00028B2C VGT_STRMOUT_DRAW_OPAQUE_BUFFER_FILLED_SIZE +0x00028B30 VGT_STRMOUT_DRAW_OPAQUE_VERTEX_STRIDE 0x00028810 PA_CL_CLIP_CNTL 0x00008A14 PA_CL_ENHANCE 0x00028C14 PA_CL_GB_HORZ_CLIP_ADJ @@ -428,7 +438,7 @@ r600 0x9400 0x00028638 SPI_VS_OUT_ID_9 0x00028438 SX_ALPHA_REF 0x00028410 SX_ALPHA_TEST_CONTROL -0x00028350 SX_MISC +0x00028354 SX_SURFACE_SYNC 0x00009014 SX_MEMORY_EXPORT_SIZE 0x00009604 TC_INVALIDATE 0x00009400 TD_FILTER4 @@ -743,14 +753,6 @@ r600 0x9400 0x00028114 CB_COLOR5_MASK 0x00028118 CB_COLOR6_MASK 0x0002811C CB_COLOR7_MASK -0x00028080 CB_COLOR0_VIEW -0x00028084 CB_COLOR1_VIEW -0x00028088 CB_COLOR2_VIEW -0x0002808C CB_COLOR3_VIEW -0x00028090 CB_COLOR4_VIEW -0x00028094 CB_COLOR5_VIEW -0x00028098 CB_COLOR6_VIEW -0x0002809C CB_COLOR7_VIEW 0x00028808 CB_COLOR_CONTROL 0x0002842C CB_FOG_BLUE 0x00028428 CB_FOG_GREEN diff --git a/drivers/gpu/drm/radeon/rs400.c b/drivers/gpu/drm/radeon/rs400.c index 866a05b..4cf381b 100644 --- a/drivers/gpu/drm/radeon/rs400.c +++ b/drivers/gpu/drm/radeon/rs400.c @@ -430,7 +430,7 @@ static int rs400_startup(struct radeon_device *rdev) if (r) return r; - r = r100_ib_test(rdev); + r = radeon_ib_test(rdev, RADEON_RING_TYPE_GFX_INDEX, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]); if (r) { dev_err(rdev->dev, "failed testing IB (%d).\n", r); rdev->accel_working = false; diff --git a/drivers/gpu/drm/radeon/rs600.c b/drivers/gpu/drm/radeon/rs600.c index 4fc7006..d25cf86 100644 --- a/drivers/gpu/drm/radeon/rs600.c +++ b/drivers/gpu/drm/radeon/rs600.c @@ -46,6 +46,25 @@ void rs600_gpu_init(struct radeon_device *rdev); int rs600_mc_wait_for_idle(struct radeon_device *rdev); +void avivo_wait_for_vblank(struct radeon_device *rdev, int crtc) +{ + struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc]; + int i; + + if (RREG32(AVIVO_D1CRTC_CONTROL + radeon_crtc->crtc_offset) & AVIVO_CRTC_EN) { + for (i = 0; i < rdev->usec_timeout; i++) { + if (!(RREG32(AVIVO_D1CRTC_STATUS + radeon_crtc->crtc_offset) & AVIVO_D1CRTC_V_BLANK)) + break; + udelay(1); + } + for (i = 0; i < rdev->usec_timeout; i++) { + if (RREG32(AVIVO_D1CRTC_STATUS + radeon_crtc->crtc_offset) & AVIVO_D1CRTC_V_BLANK) + break; + udelay(1); + } + } +} + void rs600_pre_page_flip(struct radeon_device *rdev, int crtc) { /* enable the pflip int */ @@ -175,7 +194,7 @@ void rs600_pm_misc(struct radeon_device *rdev) /* set pcie lanes */ if ((rdev->flags & RADEON_IS_PCIE) && !(rdev->flags & RADEON_IS_IGP) && - rdev->asic->set_pcie_lanes && + rdev->asic->pm.set_pcie_lanes && (ps->pcie_lanes != rdev->pm.power_state[rdev->pm.current_power_state_index].pcie_lanes)) { radeon_set_pcie_lanes(rdev, @@ -864,7 +883,7 @@ static int rs600_startup(struct radeon_device *rdev) if (r) return r; - r = r100_ib_test(rdev); + r = radeon_ib_test(rdev, RADEON_RING_TYPE_GFX_INDEX, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]); if (r) { dev_err(rdev->dev, "failed testing IB (%d).\n", r); rdev->accel_working = false; diff --git a/drivers/gpu/drm/radeon/rs690.c b/drivers/gpu/drm/radeon/rs690.c index f68dff2..f2c3b9d 100644 --- a/drivers/gpu/drm/radeon/rs690.c +++ b/drivers/gpu/drm/radeon/rs690.c @@ -31,7 +31,7 @@ #include "atom.h" #include "rs690d.h" -static int rs690_mc_wait_for_idle(struct radeon_device *rdev) +int rs690_mc_wait_for_idle(struct radeon_device *rdev) { unsigned i; uint32_t tmp; @@ -647,7 +647,7 @@ static int rs690_startup(struct radeon_device *rdev) if (r) return r; - r = r100_ib_test(rdev); + r = radeon_ib_test(rdev, RADEON_RING_TYPE_GFX_INDEX, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]); if (r) { dev_err(rdev->dev, "failed testing IB (%d).\n", r); rdev->accel_working = false; diff --git a/drivers/gpu/drm/radeon/rv515.c b/drivers/gpu/drm/radeon/rv515.c index c520d06..d8d78fe 100644 --- a/drivers/gpu/drm/radeon/rv515.c +++ b/drivers/gpu/drm/radeon/rv515.c @@ -53,9 +53,8 @@ void rv515_debugfs(struct radeon_device *rdev) } } -void rv515_ring_start(struct radeon_device *rdev) +void rv515_ring_start(struct radeon_device *rdev, struct radeon_ring *ring) { - struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]; int r; r = radeon_ring_lock(rdev, ring, 64); @@ -413,7 +412,7 @@ static int rv515_startup(struct radeon_device *rdev) if (r) return r; - r = r100_ib_test(rdev); + r = radeon_ib_test(rdev, RADEON_RING_TYPE_GFX_INDEX, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]); if (r) { dev_err(rdev->dev, "failed testing IB (%d).\n", r); rdev->accel_working = false; diff --git a/drivers/gpu/drm/radeon/rv770.c b/drivers/gpu/drm/radeon/rv770.c index c049c0c..c62ae4b 100644 --- a/drivers/gpu/drm/radeon/rv770.c +++ b/drivers/gpu/drm/radeon/rv770.c @@ -1074,7 +1074,7 @@ static int rv770_startup(struct radeon_device *rdev) r = r600_blit_init(rdev); if (r) { r600_blit_fini(rdev); - rdev->asic->copy = NULL; + rdev->asic->copy.copy = NULL; dev_warn(rdev->dev, "failed blitter (%d) falling back to memcpy\n", r); } @@ -1114,7 +1114,7 @@ static int rv770_startup(struct radeon_device *rdev) if (r) return r; - r = r600_ib_test(rdev, RADEON_RING_TYPE_GFX_INDEX); + r = radeon_ib_test(rdev, RADEON_RING_TYPE_GFX_INDEX, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]); if (r) { dev_err(rdev->dev, "IB test failed (%d).\n", r); rdev->accel_working = false; |