Under Proton (RADV/ACO), three Model M kernels go over the 192-VGPR bound
Maintainers usually reply within 1 day
Nobody has claimed this yet.
Assessment
- Difficulty
- 5/5
- Estimated time
- Over a week
- Newbie friendliness
- 15/100
- Issue type
- Bug
- Clarity
- Needs clarification
- Activity status
- Active
- Tech stack
- cpp
- Domain
- computer-graphics, performance
Research direction
The issue asks for a maintainer decision, not a code change: restructure the kernels, report the SPIR-V to Mesa ACO, or leave it. Start with clean/dx12/vgpr_limits.json and clean/dx12/README.md, then the kernels under clean/dx12/kernels built by build_dx12.py. Reproducing the VGPR counts needs RADV_DEBUG=shaderstats under Wine with native_direct_test. Done means a chosen path, agreed with the maintainer, and no code change to the kernels has been made yet.
Written by the indexing model from the issue text.
Description
What
Under Proton, Mesa's ACO compiler needs more VGPRs than clean/dx12/vgpr_limits.json allows for three of Model M's heaviest kernels. Per vgpr_limits.json's own note, they then fit one 256-lane workgroup per CU instead of two:
| Kernel | VGPRs under ACO | Bound | Waves/SIMD | GPU time per frame |
|---|---|---|---|---|
swin_block_c64_nc |
240 | 192 | 6 | 0.30 ms |
swin_block_c96_pm_rev_nc |
256 | 192 | 5 | 0.26 ms |
bneck/swin_bneck6_mbo_nc |
256 | 192 | 5 | 0.43 ms |
swin_block_c96_add_xe_xo_nc |
192 | 192 | 8 | 0.21 ms |
d1p12/swin_block_c64_add_dec0_nc |
192 | 192 | 8 | 0.43 ms |
That's about 1.0 ms of Model M's roughly 2.3 ms on the direct route, at 1706×960 → 2560×1440. The repository's notes put a lost workgroup per CU at +38 µs (swin_block_c96_add_xe_* at 215 VGPRs) to +103 µs (dec0 at 201/215 VGPRs) per frame at 1440p Quality. So this is probably a few tenths of a millisecond on Linux. I haven't measured that directly: I can't force ACO to 192 VGPRs to compare.
It isn't loop unrolling. That was my first guess, because of the "unroll loops despite [loop]" note in clean/dx12/README.md, so I checked:
swin_block_c64_nc's DXIL has no loops at all: DXC already unrolled it, so AMD's Windows compiler gets the same straight-line code and fits it in ≤192 VGPRs (the build's RGA check).swin_block_c96_pm_rev_ncand the bottleneck keep their loops under ACO (2 and 8 back-edges).
So it's ACO's register allocation on these large kernels. Before scheduling it's already about 198 VGPRs for c64, and 194 and 230 for the other two.
How it was measured (synthetic data only)
- RX 9070 XT, Mesa RADV 27.0.0-devel, GE-Proton 11-11's Wine, vkd3d-proton
5b1089c8with dxil-spirv #362 (dxil-spirv114e664eplus the PR). - Kernels built from main's
clean/dx12/kernels(unchanged since a8500cb) withbuild_dx12.py. native_direct_test(direct route, check mode, 3 frames) run under Wine withRADV_DEBUG=shaderstats(VGPRs, waves, LDS) andRADV_DEBUG=shaders(ACO's ISA).- Kernels matched by their WMMA count: each SPIR-V cooperative-matrix multiply-add becomes two
v_wmmain ACO's ISA (736, 560 and 1312 for the three), and by LDS size where that could be read. - GPU times are from
WG_D3D12_PROFILEin the same harness. Profiling adds overhead, so treat them as rough.
What could be done (your call):
- Restructure these kernels so they need fewer registers on any compiler. That would change the Windows DXIL too.
- Report it to Mesa (ACO) with these kernels' SPIR-V as a reproducer. They're WrongGPU's own MIT-licensed code, so the SPIR-V can be shared. I can put that together if you'd rather it went that way.
- Leave it. On Linux this is a few tenths of a millisecond, not a correctness problem.
Related: #29, #10.
- Dominant language
- C++
- Stars
- 12
- Forks
- 2
- Avg merge
- 3h 10m
- Merged PRs (30d)
- 171
Getting set up
- No Dockerfile or Docker Compose file
- Has a pull request template
- Read the contributing guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
More from phantoslayer21/wronggpu
-
Difficulty 2/5 1-3 hours Newbie friendliness 88/100
phantoslayer21/wronggpu#275 ·
Maintainers usually reply within 1 day
-
Difficulty 2/5 1-3 hours Newbie friendliness 82/100
phantoslayer21/wronggpu#274 ·
Maintainers usually reply within 1 day
-
bug
Difficulty 2/5 1-3 hours Newbie friendliness 88/100
phantoslayer21/wronggpu#241 ·
Maintainers usually reply within 1 day
-
bug build
Difficulty 2/5 1-3 hours Newbie friendliness 74/100
phantoslayer21/wronggpu#234 ·
Maintainers usually reply within 1 day
-
bug build
Difficulty 2/5 1-3 hours Newbie friendliness 80/100
phantoslayer21/wronggpu#233 ·
Maintainers usually reply within 1 day
All issues in phantoslayer21/wronggpu
Similar issues
-
ai_p2
Difficulty 2/5 1-3 hours Newbie friendliness 80/100
ClickHouse/ClickHouse#132519 ·
Maintainers usually reply within 1 day
-
[Bug]: VST3 getMidiControllerAssignment reads past midiControllerToParameter for program changeOpen
Difficulty 2/5 1-3 hours Newbie friendliness 76/100
juce-framework/JUCE#1810 ·
-
Difficulty 1/5 Under an hour Newbie friendliness 82/100
petercorke/robotics-toolbox-python#952 ·
Maintainers usually reply within 1 day
-
NatronRenderer: a Python error in the script (-t / project script) doesn't change the exit statusOpenfunc:engine type:bug
Difficulty 2/5 1-3 hours Newbie friendliness 72/100
NatronGitHub/Natron#1276 ·
Maintainers usually reply within 1 day
-
Difficulty 1/5 1-3 hours Newbie friendliness 77/100
Maintainers usually reply within 1 day