This is an automated email from the git hooks/post-receive script.
unknown user pushed a change to branch master in repository llvm.
from 0a1aef00955 [llvm-exegesis] Allow measuring several instructions in a s [...] new cc436fd2663 AMDGPU: Divergence-driven selection of scalar buffer load i [...] new 2026bfdfbd1 StructurizeCFG: Simplify inserted PHI nodes new 1db6c096861 AMDGPU: Avoid selecting ds_{read,write}2_b32 on SI
The 3 revisions listed above as "new" are entirely new to this repository and will be described in separate emails. The revisions listed as "adds" were already present in the repository and have only been added to this reference.
Summary of changes: lib/Target/AMDGPU/AMDGPUISelDAGToDAG.cpp | 2 - lib/Target/AMDGPU/DSInstructions.td | 4 +- lib/Target/AMDGPU/SIISelLowering.cpp | 130 +++++++++++---- lib/Target/AMDGPU/SIISelLowering.h | 4 +- lib/Target/AMDGPU/SIInstrInfo.cpp | 185 +-------------------- lib/Target/AMDGPU/SIInstrInfo.h | 2 - lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.cpp | 7 +- lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.h | 5 +- lib/Transforms/Scalar/StructurizeCFG.cpp | 24 ++- test/CodeGen/AMDGPU/lds-bounds.ll | 129 ++++++++++++++ test/CodeGen/AMDGPU/multi-divergent-exit-region.ll | 11 +- test/CodeGen/AMDGPU/multilevel-break.ll | 3 + test/CodeGen/AMDGPU/smrd-fold-offset.mir | 8 +- test/CodeGen/AMDGPU/smrd.ll | 74 ++++++++- .../StructurizeCFG/invert-constantexpr.ll | 7 +- .../Transforms/StructurizeCFG/loop-continue-phi.ll | 15 +- .../StructurizeCFG/one-loop-multiple-backedges.ll | 19 ++- 17 files changed, 366 insertions(+), 263 deletions(-) create mode 100644 test/CodeGen/AMDGPU/lds-bounds.ll