Commit graph

6262 commits

Author SHA1 Message Date
James Park
8bc2a419fc aco: Initialize union within Operand for MSVC
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7785>
2020-12-01 11:08:21 +00:00
James Park
a2c981f8d0 aco: Use u_memstream instead of POSIX memstream
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7785>
2020-12-01 11:08:21 +00:00
James Park
8259cfaa65 aco: Replace indexed array initialization
Use std::array instead to make MSVC happy.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7785>
2020-12-01 11:08:21 +00:00
James Park
76ad75dbb8 aco: Const correct aco_compiler_statistics
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7785>
2020-12-01 11:08:21 +00:00
James Park
9779a9a51e aco: Declare num_reduce_ops for array size
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7785>
2020-12-01 11:08:21 +00:00
James Park
93094b8c5e aco: Remove nonstandard parentheses
Remove parentheses in cases where a parenthesized type is followed by an
initializer list.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7785>
2020-12-01 11:08:21 +00:00
James Park
d1f742e497 aco: Add missing C++ includes
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7785>
2020-12-01 11:08:21 +00:00
James Park
e352ebf88e aco: Fix warnings about unsafe integer/bool mix
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7785>
2020-12-01 11:08:21 +00:00
James Park
fcd53bebe6 aco: Define NOMINMAX in Meson build file
Fix conflicts between std::min/max and Win32 min/max.

Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7785>
2020-12-01 11:08:21 +00:00
Rhys Perry
2d12991e01 aco: use FALLTHROUGH macro
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7844>
2020-12-01 10:28:36 +00:00
Rhys Perry
ce56902f85 radv: use FALLTHROUGH macro
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7844>
2020-12-01 10:28:36 +00:00
Samuel Pitoiset
6fb4babfe9 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@gmail.com>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7751>
2020-12-01 10:14:27 +00:00
Pierre-Eric Pelloux-Prayer
148a24f305 amd/ac: update fallthrough comments
Reviewed-by: Kristian H. Kristensen <hoegsberg@google.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7747>
2020-12-01 10:04:41 +01:00
James Park
f86668f487 vulkan/util: Consolidate typed_memcpy
Collapse typed_memcpy definitions into one header.

Use do/while(0) pattern to fix MSVC compilation.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7830>
2020-12-01 07:48:08 +00:00
Samuel Pitoiset
04ea3d6501 radv: disable WGP_MODE for NGG on GFX10.3
Ported from RadeonSI, reducing the CU mask probably broke WGP mode.

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/7769>
2020-11-30 09:31:29 +00:00
Samuel Pitoiset
bf36cfced1 radv: only mask 1 CU for GS/VS waves on GFX10.3
Ported from Radeonsi and PAL.

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/7769>
2020-11-30 09:31:29 +00:00
Samuel Pitoiset
7729ea3d67 radv: only disable CU2 & CU3 when NGG is enabled
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/7769>
2020-11-30 09:31:29 +00:00
Samuel Pitoiset
750591f4cb radv: enable NGG on GFX10.3 APUs by default
According to Radeonsi, VanGogh benefits.

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/7769>
2020-11-30 09:31:28 +00:00
Samuel Pitoiset
d1768b23b5 radv: save and dump vertex descriptors during GPU hang detection
It might be useful to know if the VA is valid and if other info
like the stride seems correct.

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/7753>
2020-11-30 08:31:02 +01:00
Samuel Pitoiset
9f3e7a6502 radv: fix using bitfields for debug/perftest options
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/7753>
2020-11-30 08:15:09 +01:00
James Park
b3504f4173 amd/common: Check with_tests before adding test
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7819>
2020-11-28 23:30:53 +00:00
James Park
de353c1fbe amd: Fix declaration mismatch
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7791>
2020-11-27 20:49:00 -08:00
James Park
a212a8a0f0 amd: Fix signature mismatch
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7791>
2020-11-27 20:49:00 -08:00
James Park
7c521e9968 amd: Work around MSVC limit for string literals
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7791>
2020-11-27 20:49:00 -08:00
James Park
7404907870 amd: Replace vasprintf with vfprintf
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7791>
2020-11-27 20:49:00 -08:00
James Park
3ad8ac38c7 amd: Stub sections that don't have _WIN32 support
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7791>
2020-11-27 20:49:00 -08:00
James Park
ee72cd0757 amd: Remove bitfield sizes from enum values
Fixes negative indexing on MSVC.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7791>
2020-11-27 20:49:00 -08:00
James Park
31b4fdc008 amd: Cast to int for %d snprintf argument
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7791>
2020-11-27 20:49:00 -08:00
James Park
bb1adece5e amd: Simplify ac_addrlib_create
Rework ac_addrlib_create to rely solely on radeon_info without
amdgpu_gpu_info.

No longer need <amdgpu.h> to create ac_addrlib instance.

Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7811>
2020-11-28 04:28:01 +00:00
Bas Nieuwenhuizen
aed8d30b50 radv: Deal with unused attachments in mip flush
Fixes: 4cce4d22a7 ("radv: Fix a hang on CB change by adding flushes.")
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7813>
2020-11-27 18:33:51 +00:00
Samuel Pitoiset
b589df9862 radv: disable SQTT support for unsupported GPUs
Like GFX10.3 which is currently broken.

Cc: 20.2 20.3
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/7786>
2020-11-27 16:37:44 +00:00
Rhys Perry
5cf41814cd aco: use binding chasing helpers
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7291>
2020-11-27 13:07:07 +00:00
Rhys Perry
e732bea536 ac/nir: use binding chasing helpers
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7291>
2020-11-27 13:07:07 +00:00
Bas Nieuwenhuizen
4cce4d22a7 radv: Fix a hang on CB change by adding flushes.
This workaround fixes a hang while loading a renderdoc trace for me.

Since the workload does 1 mip per cmdbuffer it is quite hard to confirm
what exactly the conditions for the hang are but this is the most
restrictive set I found and it corresponds to a workaround in AMDVLK as
well.

CC: mesa-stable
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7210>
2020-11-27 00:31:47 +00:00
Rhys Perry
fb0385b57c radv: use intrinsic builders
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6587>
2020-11-26 17:50:38 +00:00
Timur Kristóf
df940760f9 aco/optimizer: Propagate scc_needed label through p_wqm.
Without this, RA is confused and adds a bunch of unnecessary moves.

Fossil-db results (Navi 10):

Totals from 7658 (5.55% of 137887) affected shaders:
SGPRs: 752488 -> 752392 (-0.01%)
CodeSize: 74046720 -> 73924368 (-0.17%); split: -0.17%, +0.00%
Instrs: 14293916 -> 14263350 (-0.21%); split: -0.21%, +0.00%
Cycles: 1455556728 -> 1438466536 (-1.17%); split: -1.17%, +0.00%
VMEM: 1096736 -> 1094962 (-0.16%); split: +0.03%, -0.19%
SMEM: 448436 -> 448418 (-0.00%); split: +0.00%, -0.01%
SClause: 485501 -> 485495 (-0.00%); split: -0.00%, +0.00%
Copies: 1383769 -> 1353029 (-2.22%); split: -2.22%, +0.00%
Branches: 568247 -> 568451 (+0.04%); split: -0.01%, +0.05%

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7677>
2020-11-26 15:51:16 +00:00
Timur Kristóf
8bd3fefb74 aco/optimizer: Only set scc_needed when it is actually needed.
Not every p_cbranch uses the SCC, but our optimizer thought so.

Fixes: 8a32f57fff
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7677>
2020-11-26 15:51:16 +00:00
Samuel Pitoiset
954739fd18 radv/winsys: fix the sysmem submission path for GFX6
Oops.

Fixes: cba6ec309a ("radv: Fix -Wshadow warnings")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Tony Wasserka <tony.wasserka@gmx.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7790>
2020-11-26 12:38:14 +00:00
James Park
05799844df radv: Const aco_compiler_statistic_info usage
Necessary for upcoming const correctness change to aco struct.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7787>
2020-11-26 12:26:43 +00:00
Samuel Pitoiset
bb9b7d0a68 radv: fix missing initialization of the predication value
It's expected to be 0.

Fixes: 62d9ca696e ("radv: use 32-bit predication for conditional rendering on GFX10.3+")
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/7789>
2020-11-26 12:30:27 +01:00
Pierre-Eric Pelloux-Prayer
e5ff2a357e ac: use bigger storage for ac_arg::arg_index / ac_shader_args::arg_count
AC_MAX_ARGS is now 384 so uint8_t isn't enough.

Fixes: 6f13034265 ("ac/llvm: prepare for passing VS->TCS IO via VGPRs")
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7750>
2020-11-26 10:19:26 +01:00
James Park
33a7894828 util,radv: Cross-platform monotonic condition variable
cnd_t operates on REALTIME clock, and isn't suitable for MONOTONIC use.

Clone the API, and implement using a monotonic clock.

Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7138>
2020-11-26 07:58:56 +00:00
Rhys Perry
4eac442217 aco/ngg: fix division-by-zero in assertion
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7576>
2020-11-25 13:41:04 +00:00
Rhys Perry
37a2c9ace6 aco: fix GS with no outputs
With NGG, ngg_gs_known_vtxcnt[0] would be false and ngg_gs_finale() would
assert.

With legacy GS, I don't know why the assertion was there.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7576>
2020-11-25 13:41:04 +00:00
Rhys Perry
fdfa96561e radv/llvm,aco/ngg: fix large shift exponent in ngg_gs_vertex_lds_addr
When vertices_out=0, we will try to shift 1u by UINT32_MAX.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7576>
2020-11-25 13:41:04 +00:00
Rhys Perry
cf0b54cdc1 aco: fix v_mul_hi_u32_u24 format
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Fixes: 57c152af9c ("aco: select v_mul_{hi}_u32_u24 for 24-bit multiplications")
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/3874
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7759>
2020-11-25 10:59:44 +00:00
Samuel Pitoiset
8da98beb5d radv: always use 32-bit predication on compute queues
It seems that only gfx queue doesn't support it, except on GFX10.3
which supports all queues.

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/7732>
2020-11-25 08:13:43 +00:00
Samuel Pitoiset
62d9ca696e radv: use 32-bit predication for conditional rendering on GFX10.3+
It's now supported.

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/7732>
2020-11-25 08:13:43 +00:00
Samuel Pitoiset
b1558ec3ec ac: add gpu_info::has_32bit_predication
32-bit predication is now supported with GFX10.3.

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/7732>
2020-11-25 08:13:43 +00:00
Bas Nieuwenhuizen
025cb90042 radv: Fix RB+ blending for VK_FORMAT_E5B9G9R9_UFLOAT_PACK32.
Fixes: e893102bcf ("radv: Add VK_FORMAT_E5B9G9R9_UFLOAT_PACK32 rendering support.")
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7716>
2020-11-24 21:25:57 +00:00