projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 8bb8915) | patch
AMDGPU: Decompose all values to 32-bit pieces for calling conventions
authorMatt Arsenault <Matthew.Arsenault@amd.com>
Fri, 19 Jul 2019 13:57:44 +0000 (13:57 +0000)
committerMatt Arsenault <Matthew.Arsenault@amd.com>
Fri, 19 Jul 2019 13:57:44 +0000 (13:57 +0000)
commit1022c0dfde51a32c31cb912ccf8256e60debebfd
tree2b9ac120eff33333517f204fd68c9b81f7e66398tree | snapshot
parent8bb8915d43fe687237f31014f34fc6f2f79fb5ffcommit | diff
AMDGPU: Decompose all values to 32-bit pieces for calling conventions

This is the more natural lowering, and presents more opportunities to
reduce 64-bit ops to 32-bit.

This should also help avoid issues graphics shaders have had with
64-bit values, and simplify argument lowering in globalisel.

llvm-svn: 366578
14 files changed:
llvm/lib/Target/AMDGPU/AMDGPUCallingConv.td diff | blob | history
llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp diff | blob | history
llvm/lib/Target/AMDGPU/SIISelLowering.cpp diff | blob | history
llvm/test/CodeGen/AMDGPU/GlobalISel/irtranslator-amdgpu_vs.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/call-argument-types.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/debug-value2.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/implicit-def-muse.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/reduce-build-vec-ext-to-ext-build-vec.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/shift-i128.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/si-scheduler.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/split-arg-dbg-value.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/vector_shuffle.packed.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/wait.ll diff | blob | history
llvm/test/CodeGen/AMDGPU/wwm-reserved.ll diff | blob | history
Domain: System / Toolchain;