This is an automated email from the git hooks/post-receive script.
unknown user pushed a change to branch master in repository llvm.
from 08abe8c31a4 [X86] Add test cases showing failure to use RMW form of neg [...] new 966d4dc9697 [Hexagon] Remove fcmp undef from reduced tests new 234b3a117e8 AMDGPU: Remove dx10-clamp from subtarget features
The 2 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: docs/AMDGPUUsage.rst | 8 + lib/Target/AMDGPU/AMDGPUAsmPrinter.cpp | 5 +- lib/Target/AMDGPU/AMDGPUFeatures.td | 6 - lib/Target/AMDGPU/AMDGPUSubtarget.cpp | 6 +- lib/Target/AMDGPU/AMDGPUSubtarget.h | 10 -- lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp | 11 +- lib/Target/AMDGPU/SIFoldOperands.cpp | 3 +- lib/Target/AMDGPU/SIISelLowering.cpp | 23 ++- lib/Target/AMDGPU/SIMachineFunctionInfo.cpp | 1 + lib/Target/AMDGPU/SIMachineFunctionInfo.h | 7 + lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.cpp | 13 ++ lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.h | 39 +++++ test/CodeGen/AMDGPU/amdgcn-ieee.ll | 188 +++++++++++++++++++++ test/CodeGen/AMDGPU/clamp.ll | 6 +- test/CodeGen/AMDGPU/hsa-fp-mode.ll | 24 ++- .../Hexagon/autohvx/build-vector-i32-type.ll | 4 +- .../Transforms/Inline/AMDGPU/inline-amdgpu-dx10.ll | 107 ++++++++++++ .../Transforms/Inline/AMDGPU/inline-amdgpu-ieee.ll | 90 ++++++++++ 18 files changed, 515 insertions(+), 36 deletions(-) create mode 100644 test/CodeGen/AMDGPU/amdgcn-ieee.ll create mode 100644 test/Transforms/Inline/AMDGPU/inline-amdgpu-dx10.ll create mode 100644 test/Transforms/Inline/AMDGPU/inline-amdgpu-ieee.ll