projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: b616811) | patch
Partially reapply "AMDGPU: Invert handling of enqueued block detection"
authorMatt Arsenault <Matthew.Arsenault@amd.com>
Sun, 8 Jan 2023 13:15:30 +0000 (08:15 -0500)
committerMatt Arsenault <Matthew.Arsenault@amd.com>
Thu, 12 Jan 2023 20:02:16 +0000 (15:02 -0500)
commit4d4894ab92ee7b884a5e57b6cbc6772e6cd1fe88
treecc7e9e0288aae3a9b2ca6c3a6d96b6fa32317d62tree | snapshot
parentb616811dde41d851dddf7a5e1b9848e53d2aa10ecommit | diff
Partially reapply "AMDGPU: Invert handling of enqueued block detection"

This mostly reverts commit 270e96f435596449002fc89962595497481c8770.

Keep the attributor related changes around, but functionally restore
the old behavior as a workaround. Device enqueue goes back to not
working at -O0 with this version.
32 files changed:
clang/test/CodeGenHIP/default-attributes.hip diff | blob | history
llvm/docs/AMDGPUUsage.rst diff | blob | history
llvm/lib/Target/AMDGPU/AMDGPUAttributes.def diff | blob | history
llvm/lib/Target/AMDGPU/AMDGPUAttributor.cpp diff | blob | history
llvm/lib/Target/AMDGPU/AMDGPUHSAMetadataStreamer.cpp diff | blob | history
llvm/lib/Target/AMDGPU/AMDGPUOpenCLEnqueuedBlockLowering.cpp diff | blob | history
llvm/lib/Target/AMDGPU/SIDefines.h diff | blob | history
llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.cpp diff | blob | history
llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.h diff | blob | history
llvm/test/CodeGen/AMDGPU/addrspacecast-constantexpr.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/annotate-kernel-features-hsa-call.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/annotate-kernel-features-hsa.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/annotate-kernel-features.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/direct-indirect-call.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/duplicate-attribute-indirect.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/hsa-metadata-enqueue-kernel-v3.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/hsa-metadata-enqueue-kernel.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/hsa-metadata-from-llvm-ir-full-v3.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/hsa-metadata-from-llvm-ir-full.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/hsa-metadata-hidden-args-v3.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/hsa-metadata-hidden-args-v5.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/hsa-metadata-hidden-args.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/implicitarg-offset-attributes.ll [new file with mode: 0644] blob
llvm/test/CodeGen/AMDGPU/propagate-flat-work-group-size.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/simple-indirect-call.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/uniform-work-group-attribute-missing.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/uniform-work-group-multistep.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/uniform-work-group-nested-function-calls.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/uniform-work-group-prevent-attribute-propagation.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/uniform-work-group-propagate-attribute.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/uniform-work-group-recursion-test.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/uniform-work-group-test.ll diff | blob | history
Domain: System / Toolchain;