projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 8c87451) | patch
[X86] Use llvm::Align for passing the alignment
authorPhoebe Wang <phoebe.wang@intel.com>
Mon, 13 Mar 2023 15:32:46 +0000 (23:32 +0800)
committerPhoebe Wang <phoebe.wang@intel.com>
Mon, 13 Mar 2023 16:06:21 +0000 (00:06 +0800)
commit7ff0b9735dc5803b2820010e05753d02adb8e654
tree956c463beb962505e509ebd0dd895230c8c0c4fetree | snapshot
parent8c87451c2bc415666699aad48a0b2b7459a9a4f1commit | diff
[X86] Use llvm::Align for passing the alignment

This should be a typo in `emitConstantSizeRepmov`. Both its caller and
callee store the alignment in a 64-bit variables, no reason to truncate
it to 32-bit. It results in alignment turns into 0 when larger than
0x100000000.

Fixes #61348

Reviewed By: RKSimon

Differential Revision: https://reviews.llvm.org/D145863
llvm/lib/Target/X86/X86SelectionDAGInfo.cpp diff | blob | history
llvm/test/CodeGen/X86/pr61348.ll [new file with mode: 0644] blob
Domain: System / Toolchain;