projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: a5e3e02) | patch
[AMDGPU] Fixed SIInstrInfo::getOpSize to handle subregs
authorStanislav Mekhanoshin <Stanislav.Mekhanoshin@amd.com>
Mon, 1 Oct 2018 18:00:02 +0000 (18:00 +0000)
committerStanislav Mekhanoshin <Stanislav.Mekhanoshin@amd.com>
Mon, 1 Oct 2018 18:00:02 +0000 (18:00 +0000)
commitae8bd6d9b5575ff9a7bad3811bc0b387763121fa
treef5c5db84bf46880591006e151f8ae917122aefd3tree | snapshot
parenta5e3e02602a350b1cd704bda39194e140e4a0cc6commit | diff
[AMDGPU] Fixed SIInstrInfo::getOpSize to handle subregs

Currently it returns incorrect operand size for a target independet
node such as COPY if operand is a register with subreg. Instead of
correct subreg size it returns a size of the whole superreg.

Differential Revision: https://reviews.llvm.org/D52736

llvm-svn: 343508
llvm/lib/Target/AMDGPU/SIInstrInfo.h diff | blob | history
Domain: System / Toolchain;