This is an automated email from the git hooks/post-receive script.
unknown user pushed a change to branch master in repository llvm.
from a57b9dfaee5 [JumpThreading] Preservation of DT and LVI across the pass new d6fbd6ac450 AMDGPU/SI: Add d16 support for buffer intrinsics.
The 1 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/AMDGPU.td | 23 +- lib/Target/AMDGPU/AMDGPUISelLowering.cpp | 4 + lib/Target/AMDGPU/AMDGPUISelLowering.h | 4 + lib/Target/AMDGPU/AMDGPUInstrInfo.cpp | 3 +- lib/Target/AMDGPU/AMDGPUSubtarget.cpp | 1 + lib/Target/AMDGPU/AMDGPUSubtarget.h | 5 + lib/Target/AMDGPU/BUFInstructions.td | 246 ++++++++++++++++++++- lib/Target/AMDGPU/SIISelLowering.cpp | 151 +++++++++++-- lib/Target/AMDGPU/SIISelLowering.h | 4 + lib/Target/AMDGPU/SIInstrInfo.td | 57 +++-- .../AMDGPU/llvm.amdgcn.buffer.load.format.d16.ll | 41 ++++ .../AMDGPU/llvm.amdgcn.buffer.store.format.d16.ll | 50 +++++ .../CodeGen/AMDGPU/llvm.amdgcn.tbuffer.load.d16.ll | 41 ++++ .../AMDGPU/llvm.amdgcn.tbuffer.store.d16.ll | 53 +++++ test/MC/AMDGPU/buf-fmt-d16-packed.s | 74 +++++++ test/MC/AMDGPU/buf-fmt-d16-unpacked.s | 73 ++++++ 16 files changed, 778 insertions(+), 52 deletions(-) create mode 100644 test/CodeGen/AMDGPU/llvm.amdgcn.buffer.load.format.d16.ll create mode 100644 test/CodeGen/AMDGPU/llvm.amdgcn.buffer.store.format.d16.ll create mode 100644 test/CodeGen/AMDGPU/llvm.amdgcn.tbuffer.load.d16.ll create mode 100644 test/CodeGen/AMDGPU/llvm.amdgcn.tbuffer.store.d16.ll create mode 100644 test/MC/AMDGPU/buf-fmt-d16-packed.s create mode 100644 test/MC/AMDGPU/buf-fmt-d16-unpacked.s