AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
140,483 Commits 1 Branch 0 Tags
65258457cb3defdc6323880043f5cb8bfd39bdff
Commit Graph

140483 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Erik Faye-Lund
65258457cb docs: move swrast to deprecated drivers list 
This driver was removed in 435de835cd ("swrast: Remove the classic
swrast DRI driver"), but we forgot to update the documentation to
reflect that.

Better late than never!

Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Chia-I Wu <olvaffe@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11109>
 2021-06-03 09:34:43 +00:00
Erik Faye-Lund
6d2a3e6da2 docs: clean up list of deprecated systems 
This mostly spells out brand-names, and sorts the list.

Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Chia-I Wu <olvaffe@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11109>
 2021-06-03 09:34:43 +00:00
Erik Faye-Lund
601a7bd2b2 docs: rename vmware-guest article 
No other drivers use the vendor in the title or path, so it seems better
to not do this for the vmware-guest driver as well. So let's rename it
to SVGA instead. The actual text itself makes it clear enough what it is
about.

This makes it stand out less in the toctree.

I chose the name "WMware SVGA3D", because that's less ambigious than the
"SVGA"-alternative, and was requested by VMware people.t

Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Chia-I Wu <olvaffe@gmail.com>
Acked-by: Jose Fonseca <jfonseca@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11109>
 2021-06-03 09:34:43 +00:00
Erik Faye-Lund
54e7353fa6 docs: update llvm requirement 
Our build-scripts no longer support LLVM versions prior to 3.9, so let's
remove mentions of older versions.

Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Chia-I Wu <olvaffe@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11109>
 2021-06-03 09:34:43 +00:00
Erik Faye-Lund
d8c46655af docs: update list of apis to match website 
This adds a few missing APIs (like Vulkan), and reorders the list to
match the order we're listing them on the frontpage of mesa3d.org.

Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Chia-I Wu <olvaffe@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11109>
 2021-06-03 09:34:43 +00:00
Erik Faye-Lund
bd68ea3702 docs: do not list all gles major versions 
We support all versions of OpenGL ES, no need to point out each version.

Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Chia-I Wu <olvaffe@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11109>
 2021-06-03 09:34:43 +00:00
Yogesh Mohanmarimuthu
0e856b95c5 loader: allocate VRAM in display GPU in case of prime 
Allocates VRAM in display GPU in case of prime. Then the dma_buf is imported
into prime GPU.

v2: add comments to make code more readable (Pierre-Eric)
    removed if check limiting p2p only for matching driver name
v3: keep old path for non mesa driver (Michel Dänzer)
v4: destroy linear_buffer_display_gpu after import (Michel Dänzer)
    fall back if linear_buffer_display_gpu alloc fail (Michel Dänzer)

Signed-off-by: Yogesh Mohanmarimuthu <yogesh.mohanmarimuthu@amd.com>
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10595>
 2021-06-03 09:12:32 +00:00
Yogesh Mohanmarimuthu
55d1b8f929 glx: create DRI screen for display GPU incase of prime 
The created DRI screen can be used to allocate VRAM memory from
display GPU in case of prime.

v2: add comments to make code readable (Pierre-Eric)
    remove driver name match check
v3: keep old path for non-mesa driver (Michel Dänzer)
v4: fallback if driver not found for display GPU (Michel Dänzer)
    fallback if create screen fail for display gpu (Michel Dänzer)

Signed-off-by: Yogesh Mohanmarimuthu <yogesh.mohanmarimuthu@amd.com>
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10595>
 2021-06-03 09:12:32 +00:00
Yogesh Mohanmarimuthu
dac2c210d6 glx: Keep display fd open for prime 
Keep the display fd open for creating DRI screen on display gpu in
case of prime.

v2: initialize psc->fd_display_gpu early (Michel Dänzer)
    fix display gpu fd leakage (Michel Dänzer)
v3: fix more display gpu fd leakage (Michel Dänzer)

Signed-off-by: Yogesh Mohanmarimuthu <yogesh.mohanmarimuthu@amd.com>
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Michel Dänzer <mdaenzer@redhat.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10595>
 2021-06-03 09:12:32 +00:00
Hyunjun Ko
41eaa07823 turnip/kgsl: Fix to build on android. 
Fixes: 3f229e34 ("turnip: Implement VK_KHR_timeline_semaphore.")

Signed-off-by: Hyunjun Ko <zzoon@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11153>
 2021-06-03 08:55:06 +00:00
Pierre-Eric Pelloux-Prayer
8baa77c813 radeonsi: fix encryption check for buffers 
The writable bit means read-write, not just write.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Fixes: 8873ea0e25 ("radeonsi: determine secure flag must be set for gfx IB")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11107>
 2021-06-03 08:33:52 +00:00
Pierre-Eric Pelloux-Prayer
b377da9122 radeonsi: allow write-only mapping of encrypted textures 
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11107>
 2021-06-03 08:33:52 +00:00
Pierre-Eric Pelloux-Prayer
fcdfe91efa frontend/dri: fix bool/int comparison 
Cast tex->bind & PIPE_BIND_PROTECTED to a bool before doing the
comparison, otherwise it'll incorrectly fail.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Fixes: 18b7cafc70 ("driconf: add disable_protected_content_check option")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11107>
 2021-06-03 08:33:52 +00:00
Pierre-Eric Pelloux-Prayer
5e6f92f82c frontend/dri: set PIPE_BIND_PROTECTED later 
NV12, YUV, YUYV and UYVY handling checks !tex_usage so set
PIPE_BIND_PROTECTED after.

This fixes encrypted nv12 textures handling.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11107>
 2021-06-03 08:33:52 +00:00
Chia-I Wu
736e24bdce venus: use vk_default_allocator 
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/4870
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Acked-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11117>
 2021-06-03 08:13:27 +00:00
Chia-I Wu
567a18dade anv: use vk_default_allocator 
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Acked-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11117>
 2021-06-03 08:13:27 +00:00
Chia-I Wu
3ba3681b58 tu: use vk_default_allocator 
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Acked-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11117>
 2021-06-03 08:13:26 +00:00
Chia-I Wu
8615653c0e v3dv: use vk_default_allocator 
This also fixes the allocator used in v3dv_DestroyDevice.

v2: fix two more occurences of default_alloc (Roman Stratiienko)

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Acked-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11117>
 2021-06-03 08:13:26 +00:00
Chia-I Wu
7ebd658e28 radv: use vk_default_allocator 
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Acked-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11117>
 2021-06-03 08:13:26 +00:00
Chia-I Wu
d3acc73455 vulkan/util: add vk_default_allocator 
We cannot use os_{malloc,free,realloc}_aligned because
os_realloc_aligned needs the old size (for memcpy).

v2: no max_align_t on MSVC

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Acked-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11117>
 2021-06-03 08:13:26 +00:00
Yiwei Zhang
5054f74ce1 venus: initial AHB support for multi-planar format 
Signed-off-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Chia-I Wu <olvaffe@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11150>
 2021-06-03 07:36:09 +00:00
Samuel Pitoiset
380ac28891 ac: import performance counters from RadeonSI 
Performance counters will be used by RADV for VK_KHR_performance_query
and also for adding SPM support.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11140>
 2021-06-03 07:15:21 +00:00
Samuel Pitoiset
7af915b4a0 radv: enable RADV_DEBUG=invariantgeom for SotTR DX11/DX12 versions 
DXVK 1.8.1 marks position as always invariant but it's disabled for
SotTR because it introduces rendering issues on NV. The DX12 version
also likely needs that.

Fixes a similar foliage issue initially found with the native version.

Cc: 21.1 mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11006>
 2021-06-03 06:32:35 +00:00
Jason Ekstrand
06ae2723d1 intel/vec4: Also use MOV_FOR_SCRATCH for swizzle resolves 
In 2db8867943, we introduced a new meta-op MOV_FOR_SCRATCH which is
identical to MOV except it lets us identify MOVs emitted during spilling
so we know not to re-spill those instructions.  We emit them from
shuffle_for_64bit_data whenever the new for_scratch parameter is true.
Unfortunately, we missed the one used for resolving swizzles.

Fixes: 2db8867943 "intel/vec4: Don't spill fp64 registers more..."
Tested-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11155>
 2021-06-03 06:14:17 +00:00
Chia-I Wu
889a8df5bd venus: enable wsi image ownership transfer for common wsi 
v2: this is split out from the last commit

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org> (v1)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10709>
 2021-06-03 06:03:31 +00:00
Chia-I Wu
676abf12dc venus: reland wsi image ownership transfer for Android 
This takes a different approach.  It turns out we can patch
vkCmd{Begin,End}RenderPass, vkCmdWaitEvents, and vkCmdPipelineBarrier to
do queue family ownership transfers without tracking image ownerships.

v2: check the value of VN_PRESENT_SRC_INTERNAL_LAYOUT and keep this
    Android-only

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org> (v1)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10709>
 2021-06-03 06:03:31 +00:00
Chia-I Wu
a52f6b20b9 venus: undo wsi iamge ownership transfer for Android 
This reverts most of commit 96ec6b3d8f.
It keeps vn_image::sharing_mode.

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10709>
 2021-06-03 06:03:31 +00:00
Chia-I Wu
93abaf6fe8 venus: clean up vn_get_intercepted_barriers 
Add vn_cmd_{wait_events,pipeline_barrier}_fix_image_memory_barries to
replace vn_get_intercepted_barriers.  The two new functions are
identical currently, but that will change soon.

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10709>
 2021-06-03 06:03:31 +00:00
Chia-I Wu
7ec6cb61ad venus: add vn_cmd_get_image_memory_barriers 
And call it regardless of ANDROID or not.

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10709>
 2021-06-03 06:03:31 +00:00
Chia-I Wu
142e452e9a venus: add vn_image_memory_barrier_has_present_src 
And call it regardless of ANDROID or not.

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10709>
 2021-06-03 06:03:31 +00:00
Chia-I Wu
fb6fc24771 venus: remember cmd buffer fb attachments 
We need them for wsi queue ownership transfer.

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10709>
 2021-06-03 06:03:31 +00:00
Chia-I Wu
ccf85f6604 venus: remember cmd buffer render pass and framebuffer 
We need them for wsi queue ownership transfer.

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10709>
 2021-06-03 06:03:31 +00:00
Chia-I Wu
c85aa41046 venus: remember framebuffer attachments 
We need them for wsi queue ownership transfer.

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10709>
 2021-06-03 06:03:31 +00:00
Chia-I Wu
65e4bfe7ab venus: remember image view image 
We need that for wsi queue ownership transfer.

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10709>
 2021-06-03 06:03:31 +00:00
Chia-I Wu
cef50d08e9 venus: remember render pass PRESENT_SRC barriers 
We need them for wsi queue ownership transfer.

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10709>
 2021-06-03 06:03:31 +00:00
Chia-I Wu
9fe833dfd9 venus: remember render pass PRESENT_SRC attachments 
We need them for wsi queue ownership transfer.

v2: make VN_PRESENT_SRC_INTERNAL_LAYOUT public

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org> (v1)
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10709>
 2021-06-03 06:03:31 +00:00
Chia-I Wu
6661e47b59 venus: handle VN_COMMAND_BUFFER_STATE_INVALID 
If the state was set to VN_COMMAND_BUFFER_STATE_INVALID, return
VK_ERROR_OUT_OF_HOST_MEMORY in vn_EndCommandBuffer.

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10709>
 2021-06-03 06:03:31 +00:00
Chia-I Wu
eed11c2778 venus: remember image wsi states 
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10709>
 2021-06-03 06:03:31 +00:00
Chia-I Wu
f42f3ac1f7 venus: move wsi_image_create_info parsing 
Move it from vn_CreateImage to vn_wsi_create_image.

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10709>
 2021-06-03 06:03:31 +00:00
Chia-I Wu
9814cd8153 venus: fix opaque fd re-import 
dEQP-VK.api.external.memory.opaque_fd.* re-imports and fails because
external_handles was VK_EXTERNAL_MEMORY_HANDLE_TYPE_OPAQUE_FD_BIT on
allocation and is VK_EXTERNAL_MEMORY_HANDLE_TYPE_DMA_BUF_BIT_EXT on
re-import.

Fixes: ccefcb0baf ("venus: fix misaligned bo_flags between import and query")
Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11154>
 2021-06-03 05:53:24 +00:00
Chia-I Wu
4bd7a3829e venus: fix asserts on mem bo 
They failed to check the result before asserting.

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11154>
 2021-06-03 05:53:24 +00:00
Chia-I Wu
67ad9f7580 venus: add vn_device_memory_alloc as a helper 
It is used for the most common case (non-ahb, non-import, and
non-suballocate).

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11154>
 2021-06-03 05:53:24 +00:00
Jason Ekstrand
f63410eee6 intel/nir,i965: Move HW generation check for UBO pushing to i965 
Iris only runs on BDW+ and ANV already handles this by not even trying
on anything older than HSW.  The only driver benefiting from this common
check is i965.  Moving it out makes the pass more generic and if some
driver comes along which can push UBOs on IVB, it should work for that.

Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11145>
 2021-06-03 05:12:33 +00:00
Sergii Melikhov
8251bd216e util/format: Change the pointer offset. 
Changed the pointer offset to 2 to account for the second structure variable.

Fixes: 90f98b56f8 ("mesa: Deduplicate _mesa_pack_uint_z_row().")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/4685
Signed-off-by: Sergii Melikhov <sergii.v.melikhov@globallogic.com>
Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11060>
 2021-06-03 04:44:24 +00:00
Chia-I Wu
447e80ac9b vulkan/wsi: provide more info in wsi_image_create_info 
Always chain wsi_image_create_info to VkImageCreateInfo, which indicates
that the image is a wsi image and can be transitioned to/from
VK_IMAGE_LAYOUT_PRESENT_SRC_KHR.

Add prime_blit_buffer to the struct as well.  When set, it indicates the
prime blit destination and implies that the image is a prime blit
source.

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10789>
 2021-06-03 04:24:55 +00:00
Dave Airlie
64fa67dd2f intel/gfx6: move xfb_setup outside the gs compiler into the driver. 
This remove the use of a GL thing from the backend compiler

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Emma Anholt <emma@anholt.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/11097>
 2021-06-03 04:05:07 +00:00
Rhys Perry
903f814b78 aco: don't create 4 and 5 dword NSA instructions on GFX10 
"stability issues", apparently: https://reviews.llvm.org/D103348

fossil-db (Navi10):
Totals from 4512 (3.01% of 149839) affected shaders:
VGPRs: 221516 -> 223308 (+0.81%); split: -0.07%, +0.88%
CodeSize: 23000080 -> 23070672 (+0.31%); split: -0.08%, +0.39%
MaxWaves: 107718 -> 107496 (-0.21%); split: +0.11%, -0.32%
Instrs: 4321890 -> 4362822 (+0.95%); split: -0.00%, +0.95%
Latency: 71495710 -> 71581476 (+0.12%); split: -0.07%, +0.19%
InvThroughput: 11858568 -> 11938960 (+0.68%); split: -0.00%, +0.68%
VClause: 76575 -> 76585 (+0.01%); split: -0.05%, +0.07%
SClause: 168771 -> 168709 (-0.04%); split: -0.06%, +0.02%
Copies: 182305 -> 221948 (+21.75%); split: -0.00%, +21.75%
PreVGPRs: 194657 -> 195635 (+0.50%); split: -0.00%, +0.50%

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Fixes: c353895c92 ("aco: use non-sequential addressing")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10898>
 2021-06-03 03:49:07 +00:00
Rhys Perry
bb52484df5 aco/tests: improve reporting of failed code checks 
Instead of just reporting the failed statements, print where they
originated. This is useful for tests which have a number of similar
checks.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10898>
 2021-06-03 03:49:07 +00:00
Rhys Perry
9bf30c4a5c aco/tests: add tests for form_hard_clauses() 
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10898>
 2021-06-03 03:49:07 +00:00
Rhys Perry
81162265b1 aco: do not clause NSA instructions 
According to LLVM, this has "unpredictable results on GFX10.1".

https://reviews.llvm.org/D102211

fossil-db (Navi10):
Totals from 26690 (17.81% of 149839) affected shaders:
CodeSize: 167935160 -> 167706280 (-0.14%); split: -0.14%, +0.00%
Instrs: 31801427 -> 31744142 (-0.18%); split: -0.18%, +0.00%
Latency: 732672435 -> 732622463 (-0.01%)
InvThroughput: 163361435 -> 163357838 (-0.00%); split: -0.00%, +0.00%
VClause: 546131 -> 546903 (+0.14%); split: -0.00%, +0.14%

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Fixes: c353895c92 ("aco: use non-sequential addressing")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10898>
 2021-06-03 03:49:07 +00:00