Tell GPUCompiler which driver consumes the SPIR-V - #504
Merged
Merged
Conversation
maleadt
force-pushed
the
tb/intel-insertvalue
branch
from
September 26, 2026 06:28
90d4942 to
3e0d880
Compare
maleadt
marked this pull request as ready for review
September 26, 2026 06:29
This lets GPUCompiler work around driver bugs, like Intel's graphics compiler dropping struct fields when storing aggregates built by nested insertvalues (#502).
maleadt
force-pushed
the
tb/intel-insertvalue
branch
from
September 26, 2026 13:04
3e0d880 to
16abfe1
Compare
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #504 +/- ##
==========================================
- Coverage 83.34% 83.12% -0.22%
==========================================
Files 19 19
Lines 1603 1612 +9
==========================================
+ Hits 1336 1340 +4
- Misses 267 272 +5 ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Fixes #502.
Intel's graphics compiler drops struct fields when storing aggregates built by nested
insertvalues. For example, theBoolinFlagFirst(flag, (a, b))is lost, and so is the flag in(Bool, value)reduction tuples. The cause is a bug in IGC'sTypesLegalizationPass(intel/intel-graphics-compiler#378). Our SPIR-V is valid.JuliaGPU/GPUCompiler.jl#953 works around this by flattening nested
insertvalues whenSPIRVCompilerTarget(; driver=:intel). This PR reports the driver (:intelfor Intel GPUs only, since Intel's CPU runtime uses a different compiler; otherwise:pocl,:nvidiaor:generic) and adds a regression test.Requires GPUCompiler 2.9, which includes JuliaGPU/GPUCompiler.jl#953.
With it, the full test suite passes on an Iris Xe (NEO 26.18, IGC 2.34), including the new test.