diff options
| author | Matt Arsenault <Matthew.Arsenault@amd.com> | 2018-06-28 10:18:55 +0000 |
|---|---|---|
| committer | Matt Arsenault <Matthew.Arsenault@amd.com> | 2018-06-28 10:18:55 +0000 |
| commit | 75e7192ba321da26bd1867b496dc0bbac1bb7c64 (patch) | |
| tree | d3c65b92c27598c3e97b2d319fe4e7ad6a8ff196 /llvm/lib/Target/AMDGPU/AMDGPUMachineFunction.cpp | |
| parent | 1fb90133681255d38eed5c75c0fb64c9b090a7f5 (diff) | |
| download | bcm5719-llvm-75e7192ba321da26bd1867b496dc0bbac1bb7c64.tar.gz bcm5719-llvm-75e7192ba321da26bd1867b496dc0bbac1bb7c64.zip | |
AMDGPU: Remove MFI::ABIArgOffset
We have too many mechanisms for tracking the various offsets
used for kernel arguments, so remove one. There's still a lot of
confusion with these because there are two different "implicit"
argument areas located at the beginning and end of the kernarg
segment.
Additionally, the offset was determined based on the memory
size of the split element types. This would break in a future
commit where v3i32 is decomposed into separate i32 pieces.
llvm-svn: 335830
Diffstat (limited to 'llvm/lib/Target/AMDGPU/AMDGPUMachineFunction.cpp')
| -rw-r--r-- | llvm/lib/Target/AMDGPU/AMDGPUMachineFunction.cpp | 3 |
1 files changed, 1 insertions, 2 deletions
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUMachineFunction.cpp b/llvm/lib/Target/AMDGPU/AMDGPUMachineFunction.cpp index 73c073b05bf..0574c991ee6 100644 --- a/llvm/lib/Target/AMDGPU/AMDGPUMachineFunction.cpp +++ b/llvm/lib/Target/AMDGPU/AMDGPUMachineFunction.cpp @@ -17,10 +17,9 @@ using namespace llvm; AMDGPUMachineFunction::AMDGPUMachineFunction(const MachineFunction &MF) : MachineFunctionInfo(), LocalMemoryObjects(), - KernArgSize(0), + ExplicitKernArgSize(0), MaxKernArgAlign(0), LDSSize(0), - ABIArgOffset(0), IsEntryFunction(AMDGPU::isEntryFunctionCC(MF.getFunction().getCallingConv())), NoSignedZerosFPMath(MF.getTarget().Options.NoSignedZerosFPMath), MemoryBound(false), |
