[all-commits] [llvm/llvm-project] 8c835e: [VectorCombine] Align shrinkLoadForShuffles to reg...
Timur Golubovich via All-commits
all-commits at lists.llvm.org
Mon Jul 27 06:25:52 PDT 2026
Branch: refs/heads/main
Home: https://github.com/llvm/llvm-project
Commit: 8c835ee87def288eca2587293499008dce54c4e7
https://github.com/llvm/llvm-project/commit/8c835ee87def288eca2587293499008dce54c4e7
Author: Timur Golubovich <timur.golubovich at intel.com>
Date: 2026-07-27 (Mon, 27 Jul 2026)
Changed paths:
M llvm/lib/Transforms/Vectorize/VectorCombine.cpp
M llvm/test/Transforms/PhaseOrdering/AArch64/hoist-load-from-vector-loop.ll
M llvm/test/Transforms/VectorCombine/X86/load-widening.ll
Log Message:
-----------
[VectorCombine] Align shrinkLoadForShuffles to register boundaries (#209775)
Round up the shrunk load element count to the next full vector register
boundary to avoid scalar remainders that prevent shuffle lowering.
For example, on X86 AVX-512 a <16 x double> load with shuffle indices
1-8 was shrunk to <9 x double>. This legalizes to v8f64 + scalar f64,
and the scalar operand in BUILD_VECTOR prevents DAGCombiner from folding
it into a VECTOR_SHUFFLE, producing ~15 instructions instead of 2
(VMOVDQA64 + VALIGNQ).
To unsubscribe from these emails, change your notification settings at https://github.com/llvm/llvm-project/settings/notifications
More information about the All-commits
mailing list