]> granicus.if.org Git - llvm/commit
Merging r260651:
authorTom Stellard <thomas.stellard@amd.com>
Fri, 3 Jun 2016 15:58:20 +0000 (15:58 +0000)
committerTom Stellard <thomas.stellard@amd.com>
Fri, 3 Jun 2016 15:58:20 +0000 (15:58 +0000)
commit737edaf0482f208d3aaf92790dfc83bc9c4fe480
treef606ce7388c64f4a2767a1d23439e97da4f0c957
parent6a7a8a8f2f688c027d0e866d1234206f10af29b1
Merging r260651:

------------------------------------------------------------------------
r260651 | Matthew.Arsenault | 2016-02-11 18:40:47 -0800 (Thu, 11 Feb 2016) | 7 lines

AMDGPU: Set element_size in private resource descriptor

Introduce a subtarget feature for this, and leave the default with
the current behavior which assumes up to 16-byte loads/stores can
be used. The field also seems to have the ability to be set to 2 bytes,
but I'm not sure what that would be used for.

------------------------------------------------------------------------

git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_38@271679 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Target/AMDGPU/AMDGPU.td
lib/Target/AMDGPU/AMDGPUAsmPrinter.cpp
lib/Target/AMDGPU/AMDGPUSubtarget.cpp
lib/Target/AMDGPU/AMDGPUSubtarget.h
lib/Target/AMDGPU/AMDKernelCodeT.h
lib/Target/AMDGPU/SIInstrInfo.cpp
lib/Target/AMDGPU/SIInstrInfo.h
test/CodeGen/AMDGPU/large-alloca-compute.ll
test/CodeGen/AMDGPU/large-alloca-graphics.ll
test/CodeGen/AMDGPU/vgpr-spill-emergency-stack-slot-compute.ll
test/CodeGen/AMDGPU/vgpr-spill-emergency-stack-slot.ll