This is an automated email from the git hooks/post-receive script.
unknown user pushed a change to branch master in repository llvm.
from 94f396c4e1e [DWARF][NFC] Refactor range list emission to use a static helper new 3bda4ed0ecf Revert "AMDGPU: Force inlining if LDS global address is used" new cc4b35dde44 [X86] Remove FloatVT from X86VectorVTInfo in X86InstrAVX512.td new ed86411ba99 [X86] Add test cases that show failure to fold load into vf [...] new 784e3eb4914 [X86] Correct vfixupimm load patterns to look for an intege [...] new 0758f84dc32 [X86] Add back GCCBuiltin on mask_div_ss/sd_round.
The 5 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: include/llvm/IR/IntrinsicsX86.td | 4 +- lib/Target/AMDGPU/AMDGPUAlwaysInlinePass.cpp | 108 ++++----------------- lib/Target/AMDGPU/AMDGPUTargetMachine.cpp | 12 +-- lib/Target/AMDGPU/AMDGPUTargetMachine.h | 1 - lib/Target/X86/X86InstrAVX512.td | 67 ++++++------- test/CodeGen/AMDGPU/early-inline.ll | 10 +- .../force-alwaysinline-lds-global-address.ll | 77 --------------- test/CodeGen/AMDGPU/stress-calls.ll | 2 +- test/CodeGen/X86/avx512-intrinsics.ll | 22 ++++- 9 files changed, 79 insertions(+), 224 deletions(-) delete mode 100644 test/CodeGen/AMDGPU/force-alwaysinline-lds-global-address.ll