michaelselehov df48719df3 [AMDGPU] Add !noalias metadata to mem-accessing calls w/o pointer args (#188949) ...

addAliasScopeMetadata in AMDGPULowerKernelArguments skips instructions
with empty PtrArgs, including memory-accessing calls that have no
pointer arguments (e.g. builtins like threadIdx()). Because these calls
never receive !noalias metadata, ScopedNoAliasAA cannot prove they don't
alias noalias kernel arguments. MemorySSA then conservatively reports
them as clobbers, which prevents AMDGPUAnnotateUniformValues from
marking loads as noclobber, blocking scalarization (s_load) and forcing
expensive vector loads (global_load) instead.

Fix by adding all noalias kernel argument scopes to !noalias metadata
for memory-accessing instructions with no pointer arguments. Since such
instructions cannot access memory through any kernel pointer argument,
all noalias scopes are safe to apply.

This fixes a performance regression in rocFFT introduced by bd9668df0f
("[AMDGPU] Propagate alias information in AMDGPULowerKernelArguments").

Assisted-by: Claude Opus

2026-04-03 08:41:05 +02:00

14 KiB

Raw Permalink Blame History

View Raw