Mesa (20.3): 42 new commits
GitLab Mirror
gitlab-mirror at kemper.freedesktop.org
Wed Jan 13 18:08:47 UTC 2021
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=7e4d536e6308d2feb777d25f89766265cc1b183c
Author: Dylan Baker <dylan.c.baker at intel.com>
Date: Wed Jan 13 10:06:51 2021 -0800
VERSION: bump to 20.3.3
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=884c86d74e05f68622facc04d554655a76a4048c
Author: Dylan Baker <dylan.c.baker at intel.com>
Date: Wed Jan 13 10:06:42 2021 -0800
docs: add release notes for 20.3.3
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=24c7a4df3a8790c5fb5c93edc5a0b4310b4505fe
Author: Georg Lehmann <dadschoorse at gmail.com>
Date: Thu Dec 31 16:18:49 2020 +0100
vulkan/overlay: fix vkGetInstanceProcAddr self-resolving
vkGetInstanceProcAddr(instance, "vkGetInstanceProcAddr") should return our
vkGetInstanceProcAddr not the next in the chain.
CC: mesa-stable
Signed-off-by: Georg Lehmann <dadschoorse at gmail.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin at intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8285>
(cherry picked from commit fff77e4b434a4989e6c54ab7ef8d1aa22fe81b3a)
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=b439d0e679cd16817bb0a62ba8dc91d4cc8d7c66
Author: Bas Nieuwenhuizen <bas at basnieuwenhuizen.nl>
Date: Tue Jan 12 12:47:59 2021 +0100
radv: Add Android module info to linker script.
The Android Vulkan loader needs this symbol, so the addition of the
linker script broke Vulkan for Android.
(For non-Android builds: I checked that having a non-existent symbol in
the linker script works ok and doesn't put the symbol in the library)
Fixes: 41bb6459d3a ("radv: restrict exported symbols with static llvm")
Acked-by: Samuel Pitoiset <samuel.pitoiset at gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8437>
(cherry picked from commit 4956f6d0bf477568731f3965125299b8857d5835)
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=3261d99b9c0a5def8815b14480d03d6485ba918b
Author: Eric Anholt <eric at anholt.net>
Date: Mon Jan 11 11:09:12 2021 -0800
mesa/st: Free the NIR builtins TGSI tokens after passing to the driver.
The driver interface doesn't take ownership of the TGSI tokens, so free
our temporary.
Fixes: 57effa342b75 ("st/mesa: Drop the TGSI paths for PBOs and use nir-to-tgsi if needed.")
Reviewed-by: Adam Jackson <ajax at redhat.com>
Reviewed-by: Marek Olšák <marek.olsak at amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8422>
(cherry picked from commit 4ddcd9cf165deba2605f5729a0af2f7ded0baad9)
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=ed87818eddb890f2bd281a44b86c6f229b7a3a71
Author: Eric Anholt <eric at anholt.net>
Date: Mon Jan 11 10:40:21 2021 -0800
gallium/ntt: Fix leak of the per-instr liveness information.
Fixes: 34cc6a804ec9 ("gallium: Add a nir-to-TGSI pass.")
Reviewed-by: Adam Jackson <ajax at redhat.com>
Reviewed-by: Marek Olšák <marek.olsak at amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8422>
(cherry picked from commit 0367bdbdc465d4568733099dae9f2cc2e3eb2021)
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=adf09c9bf0e73f0b4c531af629b90012b36fdd9b
Author: Rhys Perry <pendingchaos02 at gmail.com>
Date: Mon Dec 14 22:00:23 2020 +0000
radv: set invariantgeom for Shadow of the Tomb Raider
Work around flickering foliage on GFX10.3
Signed-off-by: Rhys Perry <pendingchaos02 at gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset at gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof at gmail.com>
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/4064
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8104>
(cherry picked from commit 255ca7ecda36c3623ef730304a13aa2e8b8c0cf8)
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=f8033bdc6d641fa06669864f026f490c6013ec09
Author: Rhys Perry <pendingchaos02 at gmail.com>
Date: Mon Dec 14 21:54:28 2020 +0000
radv: add RADV_DEBUG=invariantgeom
This can be used to work around a common class of bugs appearing as
flickering.
Signed-off-by: Rhys Perry <pendingchaos02 at gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset at gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof at gmail.com>
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8104>
(cherry picked from commit f17de6a803d2fd21dc0f74b0cb28a65cd9eca257)
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=97365ce754ed26414d8162a3bda9d0d909c59563
Author: Pavel Asyutchenko <sventeam at yandex.ru>
Date: Wed Dec 23 00:05:54 2020 +0300
vulkan/overay: fix violation of VUID-VkDeviceCreateInfo-pNext-00373
We can't use VkDeviceCreateInfo::pEnabledFeatures when
VkPhysicalDeviceFeatures2 is present in pNext chain.
Fixes: ea7a6fa98055ad ("vulkan/overlay: add pipeline statistic & timestamps support")
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin at intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8208>
(cherry picked from commit 208538143293466c5a90f6dc8a5aaa67849e3557)
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=945bffe00878130f1b814d5d1a97fea8bc117a8a
Author: Georg Lehmann <dadschoorse at gmail.com>
Date: Thu Dec 31 17:17:34 2020 +0100
vulkan/device-select: fix vkGetInstanceProcAddr self-resolving
vkGetInstanceProcAddr(instance, "vkGetInstanceProcAddr") should return our
vkGetInstanceProcAddr not the next in the chain.
CC: mesa-stable
Signed-off-by: Georg Lehmann <dadschoorse at gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8286>
(cherry picked from commit 67de6356f8474fa11bb8c64e835f30a910a32b1f)
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=c5f73e66ed6b57ae991c8210467006999f1bd1b8
Author: Bas Nieuwenhuizen <bas at basnieuwenhuizen.nl>
Date: Sat Jan 9 00:39:51 2021 +0100
radv/winsys: Fix offset in range merging.
If we change the virtual address we also have to change the offset in the buffer
to be mapped.
Fixes: 715df30a4e2 "radv/amdgpu: Add winsys implementation of virtual buffers."
Reviewed-by: Samuel Pitoiset <samuel.pitoiset at gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7953>
(cherry picked from commit d3286bdd764741b2cb6d30b59450ce9473884430)
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=107e4eea53e0bc05ee7176b5d0066416986d7862
Author: Bas Nieuwenhuizen <bas at basnieuwenhuizen.nl>
Date: Sun Dec 6 16:05:41 2020 +0100
radv/winsys: Fix inequality for sparse buffer remapping.
Found a case where we mapped a range too many.
Per the comment the constraint is:
/* [first, last] is exactly the range of ranges that either overlap the
* new parent, or are adjacent to it. This corresponds to the bind ranges
* that may change.
*/
So that means that after the ++last we the ranges[last] should still
be adjacent. So we need to test the post-increment value to see whether
it is adjacent.
Failure case:
ranges:
0: 0 - ffff
1: 10000 - 1ffff
2: 20000 - 2ffff
3: 30000 - 3ffff
new range: 10000 - 1ffff
wrong first, last: 0,3
However range 3 clearly isn't adjacent at all.
Fixes: 715df30a4e2 "radv/amdgpu: Add winsys implementation of virtual buffers."
Reviewed-by: Samuel Pitoiset <samuel.pitoiset at gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7953>
(cherry picked from commit 2b12e6931ef240df44d2c0f9374d6575ad202675)
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=bf4e0a4b3e16fb08382bd404b4178fff657414c3
Author: Dylan Baker <dylan.c.baker at intel.com>
Date: Tue Jan 12 15:27:39 2021 -0800
.pick_status.json: Mark 52b6adfbfb51a3b1a54964cbcb2adb0299d1252b as backported
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=bbb85a7be0061b4c71bc5cb9796f870dc47da10f
Author: nia <nia at netbsd.org>
Date: Sun Dec 6 14:47:53 2020 +0100
util: Avoid pthread_setaffinity_np on NetBSD
NetBSD's variant has a different prototype from the Linux version
the code expects. It might make sense to add support for NetBSD's
version, however, since NetBSD defaults to not allowing non-root
users to set processor affinity, there would be little gain here.
This is a build fix for NetBSD.
Signed-off-by: Nia Alarie <nia at NetBSD.org>
Reviewed-by: Dylan Baker <dylan at pnwbakers.com>
CC: 20.3 <mesa-stable at lists.freedesktop.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7947>
(cherry picked from commit 275079e3adc184ba3841222918e852c3e63f435f)
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=76ecdf2c32280749819b7d566a1503b52fe222f9
Author: Rhys Perry <pendingchaos02 at gmail.com>
Date: Fri Jan 8 11:44:33 2021 +0000
aco: fix unreachable() for uniform 8/16-bit nir_op_mov from VGPR
Signed-off-by: Rhys Perry <pendingchaos02 at gmail.com>
Reviewed-by: Daniel Schürmann <daniel at schuermann.dev>
Fixes: d20a752c0de ("aco: use Builder::copy more")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8380>
(cherry picked from commit 816b7fb5cb622dd6c5d0fbcecdcb27779029f80f)
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=6a006c35675636cb0d2cbfdfb1681b6639e0664d
Author: James Park <jpark37 at lagfreegames.com>
Date: Wed Dec 30 05:23:10 2020 -0800
util: Disable memstream for Apple builds
Not all SDK versions support open_memstream. Maybe some other day.
Fixes: af8d488ea5e ("util,ac,aco,radv: Cross-platform memstream API")
Reviewed-by: Samuel Pitoiset <samuel.pitoiset at gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8269>
(cherry picked from commit 3fb4755d48d079d47ca212a838d16259db2abcfa)
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=e6aa51a84d5b979b64b46291413aa212d11baccb
Author: Samuel Pitoiset <samuel.pitoiset at gmail.com>
Date: Tue Nov 24 14:56:55 2020 +0100
radv/llvm,aco: always split typed vertex buffer loads on GFX6 and GFX10+
To avoid any alignment issues that triggers memory violations and
eventually a GPU. This can happen if the stride (static or dynamic)
is unaligned and also if the VBO offset is aligned to scalar
(eg. stride is 8 and VBO offset is 2 for R16G16B16A16_SNORM).
The AMD Windows driver also always splits typed vertex fetches.
fossils-db (Sienna Cichlid):
Totals from 56508 (40.54% of 139391) affected shaders:
SGPRs: 2643545 -> 2664516 (+0.79%); split: -0.19%, +0.98%
VGPRs: 2007472 -> 1995408 (-0.60%); split: -0.74%, +0.13%
CodeSize: 70596372 -> 73913312 (+4.70%); split: -0.00%, +4.70%
MaxWaves: 772653 -> 774916 (+0.29%); split: +0.37%, -0.08%
Instrs: 14074162 -> 14567072 (+3.50%); split: -0.00%, +3.51%
Cycles: 69281276 -> 71253252 (+2.85%); split: -0.00%, +2.85%
VMEM: 22047039 -> 25554196 (+15.91%); split: +17.20%, -1.29%
SMEM: 4120370 -> 4360820 (+5.84%); split: +7.41%, -1.58%
VClause: 416913 -> 438361 (+5.14%); split: -1.86%, +7.01%
SClause: 536739 -> 542637 (+1.10%); split: -0.33%, +1.43%
Copies: 977194 -> 970015 (-0.73%); split: -2.43%, +1.69%
Branches: 241205 -> 241193 (-0.00%); split: -0.06%, +0.06%
PreVGPRs: 1505645 -> 1505379 (-0.02%)
This fixes GPU hangs with bin/draw-vertices from Piglit on GFX10+
with Zink.
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset at gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02 at gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8363>
(cherry picked from commit d2f4934121f65f2f086e4babaa0704e58503234b)
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=a29f08b1e204db4342375184829984c594bced86
Author: Samuel Pitoiset <samuel.pitoiset at gmail.com>
Date: Thu Jan 7 14:01:09 2021 +0100
aco: fix creating the dest vector when 16-bit vertex fetches are splitted
Compute the number of components of the destination vector from the
bitsize when eg. a 16-bit vec2 vertex fetches is splitted. This is
because the dst will be a v1, so the p_create_vector should be created
from two v2b fro both sizes to match.
This prevents a regression from the next change which will split
typed vertex buffer loads on GFX6 and GFX10+.
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset at gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02 at gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8363>
(cherry picked from commit 68c25370626c9ffb360ed3649627c8cf070a7f2b)
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=eec614e0f31d283bc1acf325fd00a3152aed1e45
Author: Rhys Perry <pendingchaos02 at gmail.com>
Date: Fri Mar 13 16:07:26 2020 +0000
nir/load_store_vectorize: don't ignore subgroup memory barriers
Not sure why I thought this was correct, but we should consider them for
optimization purposes.
Fixes: ce9205c03bd ('nir: add a load/store vectorization pass')
Signed-off-by: Rhys Perry <pendingchaos02 at gmail.com>
Reviewed-by: Jason Ekstrand <jason at jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4202>
(cherry picked from commit f4eb833a12523142d908bf67e08904f4e8a866e9)
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=2c5b5920ca66cf9bd1787633eb20d19d5d3c509f
Author: Dylan Baker <dylan.c.baker at intel.com>
Date: Tue Jan 12 15:27:27 2021 -0800
.pick_status.json: Update to d9c8422c4168c7433f68b97065283ebf26631d56
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=aef8fc860c2b77cb5f0f0b69a112acecc8be894a
Author: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer at amd.com>
Date: Fri Jan 8 13:17:42 2021 +0100
drirc: radeonsi workaround for CS:GO
Cherry-picked from 6f2017205e62402b7b2e340620e39cb71730c565
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8411>
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=ae1c5dc6b68d9e8cf4c9f82ec5586af614513b59
Author: Samuel Pitoiset <samuel.pitoiset at gmail.com>
Date: Thu Jan 7 08:38:01 2021 +0100
radv: disable TC-compat HTILE in GENERAL for Detroit: Become Human
The game has invalid usage of render loops and enabling TC-compat
HTILE in GENERAL introduces rendering issues.
Gitlab: https://gitlab.freedesktop.org/mesa/mesa/-/issues/3063
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset at gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas at basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8412>
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=9000c044fbdf5317e3a4134f8ac8b8742a4fd155
Author: Erik Faye-Lund <erik.faye-lund at collabora.com>
Date: Tue Dec 15 10:39:46 2020 +0100
zink: dot leak dummy_buffer
Fixes: 8736ffae2ed ("zink: replace unset buffer with a dummy-buffer")
Reviewed-By: Mike Blumenkrantz <michael.blumenkrantz at gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8099>
(cherry picked from commit 07d5d911dd6c7d36fdd678f289d3babdde634406)
Conflicts:
src/gallium/drivers/zink/zink_context.c
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=39c386c0fdc790e5900bf52ad242f793a0f147a7
Author: Ian Romanick <ian.d.romanick at intel.com>
Date: Tue Feb 18 12:52:42 2020 -0800
nir/algebraic: Fix broken NaN and -0.0 behavior
No shader-db or fossil-db changes on any Intel platform.
v2: Add a coding line to fix SCons build problems caused by the ±
character.
Fixes: 25bfba3335d ("nir/algebraic: Recognize open-coded copysign(1.0, a)")
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira at intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6358>
(cherry picked from commit 9771af5ddec37481042f6109772f15773642104c)
Conflicts:
src/compiler/nir/nir_opt_algebraic.py
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=72233905e362b2bd70d1ba3ff2e186c01c0497ab
Author: Ian Romanick <ian.d.romanick at intel.com>
Date: Thu Jun 11 18:48:41 2020 -0700
Revert "nir: Replace an odd comparison involving fmin of -b2f"
I originally noticed that 3b308147916 ("nir/algebraic: Optimize 1-bit
Booleans") caused this pattern no longer be matched by incorrectly
replacing b at 32 with b at 1. Making that correct had no effect on
shader-db. When this pattern originally was added, it only affected 4
shaders, so it's not worth the effort to debug further.
This reverts commit f50400cc8040cf2d07de97e76d9b1ed144c5c8b4.
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira at intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6358>
(cherry picked from commit 314a40c902c6e92b9d6ba0e9b43708d7e1fa9f00)
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=588cf800be8a0185b54baf2a74bbeb3cf364c5a1
Author: Erik Faye-Lund <erik.faye-lund at collabora.com>
Date: Mon Jan 4 10:35:36 2021 +0100
gallium/util: do not perform n^2 stencil blits
We already loop n times here, no point in doing n instances as well.
Fixes: e8a40715a8b ("gallium/util: add blitter-support for stencil-fallback")
Reviewed-by: Jesse Natalie <jenatali at microsoft.com>
Reviewed-by: Marek Olšák <marek.olsak at amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8301>
(cherry picked from commit 96ceca33c1dd69a1feed13b0e19bfc38e6f7d979)
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=cf5fe357d6bac1a33ae784d8e1bab5fa199b51c9
Author: Eric Anholt <eric at anholt.net>
Date: Fri Dec 11 13:11:26 2020 -0800
gallium/ntt: Fix emitting UBO declarations.
Fixes: d70fff99c5bc ("nir: Use a single list for all shader variables")
Reviewed-by: Marek Olšák <marek.olsak at amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8196>
(cherry picked from commit ec0bab8ca9eaed4648cf9e90398606fa4fc6e747)
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=071a87ac7f9e218b2f11be85d04b4786a3e6dad6
Author: Rhys Perry <pendingchaos02 at gmail.com>
Date: Mon Jan 4 13:06:15 2021 +0000
ac/nir: use llvm.readcyclecounter for LLVM9+
Unlike llvm.amdgcn.s.memtime, this works on GFX10.3
Signed-off-by: Rhys Perry <pendingchaos02 at gmail.com>
Reviewed-by: Marek Olšák <marek.olsak at amd.com>
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/4033
Cc: mesa-stable
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8306>
(cherry picked from commit c5973ede01c4d5458b3d487f7b7486707d905c52)
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=e9146fe6fb4053abbab3ebe1620fdfa15b2e2a43
Author: Marek Olšák <marek.olsak at amd.com>
Date: Thu Dec 24 05:43:25 2020 -0500
util: add AMD CPU family enums and enable L3 cache pinning on Zen3
Based on: https://en.wikichip.org/wiki/amd/cpuid
The only reason it's nominated as a fix is because Zen3 might underperform
because the CPU detection ignored it.
Fixes: 15fa2c5e359 - gallium/u_cpu_detect: get the number of cores per L3 cache for AMD Zen
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer at amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8225>
(cherry picked from commit e4fa7c440d273aad6cb9b9a6ee42a78810c9c2c2)
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=20a067a837a02e9a77c8c288e90376f7752d9df1
Author: Ian Romanick <ian.d.romanick at intel.com>
Date: Tue Aug 4 19:43:52 2020 -0700
spir-v: Mark floating point comparisons exact
OpenGL GLSL, OpenGL ARB assembly shaders, and DX9 are pretty loose about
the behavior in the presence of NaNs. Many GPUs that implement these
specifications do not even have a representation of NaN. However,
OpenCL and Vulkan SPIR-V are not so lax. Both actually have some
required behavior in the presence of NaN, and, of the two, OpenCL is the
most strict.
For years we have implemented SPIR-V by using the same comparison
opcodes as we use for OpenGL GLSL and OpenGL assembly shaders. This has
repeatedly caused problems where an optimization that is valid in the
NaN-relaxed world is not valid in Vulkan or OpenCL. To fix this, set
the "exact" flag on comparisons instructions generated from SPIR-V.
This will block optimizations that may have different NaN behavior.
v2: Set the exact flag in the nir_builder, not in the vtn_builder.
v3: Add an assertion in vtn_handle_constant that the exact flag wasn't
set (because it's ignored). Rebase on 80163bbec3a ("nir/vtn: Support
OpOrdered and OpUnordered opcodes"). Mark the NIR generated for those
opcodes as exact as well.
v4: s/unused_exact/exact/ in a couple places, and assert that exact has
the expected value (true in one place, false in the other). Suggested
by Caio.
Closes: #3345
Reviewed-by: Caio Marcelo de Oliveira Filho <caio.oliveira at intel.com>
Tested-by: Danylo Piliaiev <danylo.piliaiev at globallogic.com>
Fixes: 8513b12590c ("nir/opt_if: split ALU from Phi more aggressively")
This commit doesn't really fix anything in 8513b12590c. However,
without 8513b12590c, a regression is triggered in RADV on No Man's
Sky. I want to ensure that this change is only applied on top of
8513b12590c, and Fixes: seems the safest way to do that.
No shader-db changes on any Intel platform. This only affects SPIR-V,
and we have no OpenGL SPIR-V shaders in shader-db.
124 shaders in Shadow of the Tomb Raider (Steam "native") were hurt by 1
spill and 1 fill each.
All Intel platforms had similar results. (Tiger Lake shown)
Instructions in all programs: 155668276 -> 155685764 (+0.0%)
SENDs in all programs: 6474570 -> 6474570 (+0.0%)
Loops in all programs: 35271 -> 35271 (+0.0%)
Cycles in all programs: 3198055373 -> 3198628031 (+0.0%)
Spills in all programs: 231522 -> 231646 (+0.1%)
Fills in all programs: 347571 -> 347695 (+0.0%)
Vega
Totals:
SGPRs: 20955712 -> 20956756 (+0.00%); split: -0.02%, +0.03%
VGPRs: 13476920 -> 13473132 (-0.03%); split: -0.07%, +0.04%
CodeSize: 613371940 -> 613339348 (-0.01%); split: -0.06%, +0.05%
MaxWaves: 3111886 -> 3112481 (+0.02%); split: +0.02%, -0.00%
Instrs: 120723785 -> 120746991 (+0.02%); split: -0.04%, +0.06%
Cycles: 626658992 -> 626862708 (+0.03%); split: -0.05%, +0.08%
VMEM: 216330854 -> 216343196 (+0.01%); split: +0.04%, -0.04%
SMEM: 32079391 -> 32081972 (+0.01%); split: +0.05%, -0.04%
VClause: 2688784 -> 2688789 (+0.00%); split: -0.03%, +0.03%
SClause: 6554669 -> 6556251 (+0.02%); split: -0.01%, +0.03%
Copies: 5356667 -> 5353283 (-0.06%); split: -0.36%, +0.29%
Branches: 954466 -> 954716 (+0.03%); split: -0.01%, +0.04%
PreSGPRs: 9078300 -> 9081626 (+0.04%); split: -0.01%, +0.05%
PreVGPRs: 10972090 -> 10966576 (-0.05%); split: -0.06%, +0.01%
Totals from 48239 (12.08% of 399432) affected shaders:
SGPRs: 2713984 -> 2715028 (+0.04%); split: -0.16%, +0.19%
VGPRs: 1997804 -> 1994016 (-0.19%); split: -0.46%, +0.27%
CodeSize: 172094092 -> 172061500 (-0.02%); split: -0.21%, +0.19%
MaxWaves: 337327 -> 337922 (+0.18%); split: +0.20%, -0.02%
Instrs: 33053657 -> 33076863 (+0.07%); split: -0.15%, +0.22%
Cycles: 254961228 -> 255164944 (+0.08%); split: -0.12%, +0.20%
VMEM: 15165226 -> 15177568 (+0.08%); split: +0.59%, -0.51%
SMEM: 3304938 -> 3307519 (+0.08%); split: +0.49%, -0.41%
VClause: 766225 -> 766230 (+0.00%); split: -0.12%, +0.12%
SClause: 1332645 -> 1334227 (+0.12%); split: -0.04%, +0.16%
Copies: 2040651 -> 2037267 (-0.17%); split: -0.94%, +0.77%
Branches: 743668 -> 743918 (+0.03%); split: -0.01%, +0.05%
PreSGPRs: 1697667 -> 1700993 (+0.20%); split: -0.07%, +0.27%
PreVGPRs: 1718424 -> 1712910 (-0.32%); split: -0.39%, +0.07%
Polaris
Totals:
SGPRs: 21349172 -> 21354376 (+0.02%); split: -0.02%, +0.04%
VGPRs: 13690680 -> 13686920 (-0.03%); split: -0.07%, +0.04%
CodeSize: 613745824 -> 613704988 (-0.01%); split: -0.06%, +0.05%
MaxWaves: 2775012 -> 2775189 (+0.01%); split: +0.01%, -0.00%
Instrs: 120735079 -> 120756209 (+0.02%); split: -0.04%, +0.06%
Cycles: 627906100 -> 628076156 (+0.03%); split: -0.05%, +0.08%
VMEM: 216623065 -> 216641838 (+0.01%); split: +0.04%, -0.04%
SMEM: 32295618 -> 32299338 (+0.01%); split: +0.05%, -0.04%
VClause: 2711025 -> 2711141 (+0.00%); split: -0.03%, +0.04%
SClause: 6545185 -> 6546769 (+0.02%); split: -0.01%, +0.03%
Copies: 5387723 -> 5383249 (-0.08%); split: -0.37%, +0.29%
Branches: 953775 -> 953954 (+0.02%); split: -0.01%, +0.03%
PreSGPRs: 9148814 -> 9153211 (+0.05%); split: -0.01%, +0.06%
PreVGPRs: 11029429 -> 11023915 (-0.05%); split: -0.06%, +0.01%
Totals from 48239 (12.00% of 402052) affected shaders:
SGPRs: 2682056 -> 2687260 (+0.19%); split: -0.16%, +0.35%
VGPRs: 1994436 -> 1990676 (-0.19%); split: -0.46%, +0.27%
CodeSize: 170857060 -> 170816224 (-0.02%); split: -0.21%, +0.19%
MaxWaves: 295429 -> 295606 (+0.06%); split: +0.07%, -0.01%
Instrs: 32808802 -> 32829932 (+0.06%); split: -0.16%, +0.22%
Cycles: 254633252 -> 254803308 (+0.07%); split: -0.13%, +0.20%
VMEM: 14897934 -> 14916707 (+0.13%); split: +0.65%, -0.52%
SMEM: 3289726 -> 3293446 (+0.11%); split: +0.53%, -0.42%
VClause: 775318 -> 775434 (+0.01%); split: -0.11%, +0.13%
SClause: 1304867 -> 1306451 (+0.12%); split: -0.04%, +0.16%
Copies: 2026334 -> 2021860 (-0.22%); split: -0.99%, +0.77%
Branches: 742554 -> 742733 (+0.02%); split: -0.02%, +0.04%
PreSGPRs: 1690887 -> 1695284 (+0.26%); split: -0.07%, +0.33%
PreVGPRs: 1717709 -> 1712195 (-0.32%); split: -0.40%, +0.07%
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6358>
(cherry picked from commit 010e663cc3401c40182922a04af0719e1898cb62)
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=f121623020b7283a8d59d7b42a124dd6c3bc230e
Author: Andrii Simiklit <andrii.simiklit at globallogic.com>
Date: Mon Dec 28 15:58:24 2020 +0200
st/mesa: don't affect original st_CompressedTexSubImage parameters
The fallback path is still possible here so let keep them as is.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/3952
Fixes: 4b02f165 ("st/mesa: implement PBO upload for glCompressedTex(Sub)Image")
Reviewed-by: Marek Olšák <marek.olsak at amd.com>
Signed-off-by: Andrii Simiklit <andrii.simiklit at globallogic.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8258>
(cherry picked from commit 5ae7d40648ca643615fd551a3c6fcecb8301fc91)
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=773d683a898374c34fb3b7ecda52f7b5625ce6d5
Author: Dylan Baker <dylan.c.baker at intel.com>
Date: Wed Jan 6 09:31:28 2021 -0800
.pick_status.json: Update to 96ceca33c1dd69a1feed13b0e19bfc38e6f7d979
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=da6fb037775b5a5f425829c4cb0778f14f8deb4e
Author: Rhys Perry <pendingchaos02 at gmail.com>
Date: Mon Jan 4 10:41:39 2021 +0000
aco: fix incorrect address calculation for load_barycentric_at_sample
Fix address calculation for indirect load_barycentric_at_sample on GFX6-8
with a uniform sample index.
A non-zero uniform sample index does not seem to be tested by CTS.
Signed-off-by: Rhys Perry <pendingchaos02 at gmail.com>
Reviewed-by: Daniel Schürmann <daniel at schuermann.dev>
Gitlab: https://gitlab.freedesktop.org/mesa/mesa/-/issues/3966
Fixes: 93c8ebfa780 ("aco: Initial commit of independent AMD compiler")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8302>
(cherry picked from commit b2d000513e4a9b06d3b073913741dae47a00526e)
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=2f9e5cae2bcc485f9d4b683f875942a417672824
Author: Danylo Piliaiev <dpiliaiev at igalia.com>
Date: Mon Jan 4 16:16:32 2021 +0200
nir: fix missing nir_lower_pntc_ytransform.c in the makefile
Fixes: 33fd9e5d "nir: account for point-coord origin when lowering it"
Signed-off-by: Danylo Piliaiev <dpiliaiev at igalia.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner at gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8308>
(cherry picked from commit 81132983cdb0dd9f088743ec43e3a40347424a6c)
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=9e1e67b35b84406490bebb4cd9644111592734d5
Author: Danylo Piliaiev <dpiliaiev at igalia.com>
Date: Tue Dec 22 14:37:45 2020 +0200
nir: account for point-coord origin when lowering it
The resulting point-coord origin not only depends on whether
the draw buffer is flipped but also on GL_POINT_SPRITE_COORD_ORIGIN
state. Which makes its transform differ from a transform of wpos.
On freedreno fixes:
gl-3.2-pointsprite-origin
gl-3.2-pointsprite-origin -fbo
Fixes: d934d320 "nir: Add flipping of gl_PointCoord.y in nir_lower_wpos_ytransform."
Signed-off-by: Danylo Piliaiev <dpiliaiev at igalia.com>
Acked-by: Marek Olšák <marek.olsak at amd.com>
Reviewed-by: Jose Maria Casanova Crespo <jmcasanova at igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8200>
(cherry picked from commit 33fd9e5d8a6b85e20d34f89df655fa5a75eca431)
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=d7c09d09aa0b3532839adb761f7a96d6956622f4
Author: Marek Vasut <marex at denx.de>
Date: Sat Dec 19 01:23:12 2020 +0100
etnaviv: Fix rework ZSA into a derived state
In case the stencil is modified, it is also enabled. That was the
behavior of the original code, which was also the correct behavior,
so reinstate the behavior.
Fixes dEQP-GLES2.functional.fragment_ops.depth_stencil.* on STM32MP1 GC400T.
Fixes: b29fe26d438 ("etnaviv: rework ZSA into a derived state")
Reviewed-by: Christian Gmeiner <christian.gmeiner at gmail.com>
Signed-off-by: Marek Vasut <marex at denx.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8174>
(cherry picked from commit 33a6c01e12cca60903fb78af44ecca3899ad628f)
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=b83d96666dd59504322990e2bb12ee3902a82e42
Author: Dylan Baker <dylan.c.baker at intel.com>
Date: Mon Jan 4 09:22:26 2021 -0800
.pick_status.json: Update to b2d000513e4a9b06d3b073913741dae47a00526e
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=99ff47e8538fc9268f00829a509cf4403368e38f
Author: Marek Olšák <marek.olsak at amd.com>
Date: Tue Dec 8 20:40:57 2020 -0500
st/mesa: don't do glCopyPixels via blit if depth bounds test is enabled
Fixes: 36a6f848bb0 - st/mesa: add EXT_depth_bounds_test
Reviewed-by: Zoltán Böszörményi <zboszor at gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7720>
(cherry picked from commit 8848114345833873aaa7855edbdb40505bc0b736)
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=636a917ca4cb0d0b05308a592b276c1927ce1222
Author: Icecream95 <ixn at keemail.me>
Date: Mon Sep 7 21:29:28 2020 +1200
pan/mdg: Fix promoted uniform moves with 64-bit types
The move source is the second src, not the first one.
Fixes: 5e5ea25a0da ("pan/mdg: Explicitly type 64-bit uniform moves")
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig at collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8264>
(cherry picked from commit 97929b1aaf4d0bbb9214cd7a31ee938c78f7e8bc)
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=c7430bbc3eb044ea7eabd2e103ed879bf601d493
Author: Samuel Pitoiset <samuel.pitoiset at gmail.com>
Date: Wed Dec 30 10:28:54 2020 +0100
nir: fix determining if an addition might overflow for phi sources
nir_addition_might_overflow() expects the parent instruction to be
an alu instr but it might be a phi instr. Fix it by assuming that
the addition might overflow.
This fixes compiler crashes with Horizon Zero Dawn.
No fossils-db changes.
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset <samuel.pitoiset at gmail.com>
Reviewed-by: Daniel Schürmann <daniel at schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8268>
(cherry picked from commit 0b503d8de96c7006c0fe1c8bfdcd8de02f255992)
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=2b66910037b0c0e545eae3d8305cf955bd907742
Author: Eric Anholt <eric at anholt.net>
Date: Wed Dec 30 16:09:13 2020 -0800
gallium/draw: Fix intermittent failure to bind new geometry shaders.
If you deleted your old GS and created a new one, then it would
occasionally skip binding the new GS because the token pointers were
equal. Clear the current token pointer in the machine when we're deleting
its token.
Cc: mesa-stable
Reviewed-by: Dave Airlie <airlied at redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8277>
(cherry picked from commit 969f50ddcfcf88f641779942210f33519761a08f)
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=2f6f5b3ba2418f757add392cb12d99bfe7544a44
Author: Dylan Baker <dylan.c.baker at intel.com>
Date: Sat Jan 2 10:39:52 2021 -0800
.pick_status.json: Update to 9ef2c44ce682b7981bd7a68c65f338f1e33bb001
More information about the mesa-commit
mailing list