[all-commits] [llvm/llvm-project] 033701: [AMDGPU] Use SGPR instead of SReg classes
jayfoad via All-commits
all-commits at lists.llvm.org
Thu Apr 23 04:07:37 PDT 2020
Branch: refs/heads/master
Home: https://github.com/llvm/llvm-project
Commit: 0337017a9f001d8b3d5e8b0ed7192342bacaf443
https://github.com/llvm/llvm-project/commit/0337017a9f001d8b3d5e8b0ed7192342bacaf443
Author: Jay Foad <jay.foad at amd.com>
Date: 2020-04-23 (Thu, 23 Apr 2020)
Changed paths:
M llvm/lib/Target/AMDGPU/AsmParser/AMDGPUAsmParser.cpp
M llvm/lib/Target/AMDGPU/SIISelLowering.cpp
M llvm/lib/Target/AMDGPU/SILoadStoreOptimizer.cpp
M llvm/lib/Target/AMDGPU/SIRegisterInfo.cpp
M llvm/lib/Target/AMDGPU/SIRegisterInfo.td
M llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.cpp
M llvm/test/CodeGen/AMDGPU/GlobalISel/inst-select-concat-vectors.mir
M llvm/test/CodeGen/AMDGPU/GlobalISel/inst-select-extract-vector-elt.mir
M llvm/test/CodeGen/AMDGPU/GlobalISel/inst-select-extract.mir
M llvm/test/CodeGen/AMDGPU/GlobalISel/inst-select-implicit-def.mir
M llvm/test/CodeGen/AMDGPU/GlobalISel/inst-select-insert-vector-elt.mir
M llvm/test/CodeGen/AMDGPU/GlobalISel/inst-select-insert.mir
M llvm/test/CodeGen/AMDGPU/GlobalISel/inst-select-merge-values.mir
M llvm/test/CodeGen/AMDGPU/GlobalISel/inst-select-trunc.mir
M llvm/test/CodeGen/AMDGPU/GlobalISel/inst-select-unmerge-values.mir
M llvm/test/CodeGen/AMDGPU/GlobalISel/llvm.amdgcn.s.buffer.load.ll
M llvm/test/CodeGen/AMDGPU/coalescer-subranges-another-copymi-not-live.mir
M llvm/test/CodeGen/AMDGPU/coalescer-subranges-another-prune-error.mir
M llvm/test/CodeGen/AMDGPU/coalescer-subreg-join.mir
M llvm/test/CodeGen/AMDGPU/coalescer-with-subregs-bad-identical.mir
M llvm/test/CodeGen/AMDGPU/memory_clause.mir
M llvm/test/CodeGen/AMDGPU/merge-image-load.mir
M llvm/test/CodeGen/AMDGPU/merge-image-sample.mir
M llvm/test/CodeGen/AMDGPU/optimize-exec-masking-pre-ra.mir
M llvm/test/CodeGen/AMDGPU/subreg-split-live-in-error.mir
Log Message:
-----------
[AMDGPU] Use SGPR instead of SReg classes
12994a70cf7 did this for 128-bit classes:
SGPR_128 only includes the real allocatable SGPRs, and SReg_128 adds
the additional non-allocatable TTMP registers. There's no point in
allocating SReg_128 vregs. This shrinks the size of the classes
regalloc needs to consider, which is usually good.
This patch extends it to all classes > 64 bits, for consistency.
Differential Revision: https://reviews.llvm.org/D78622
More information about the All-commits
mailing list