]> granicus.if.org Git - llvm/commit
Merging r260658:
authorTom Stellard <thomas.stellard@amd.com>
Fri, 3 Jun 2016 16:45:46 +0000 (16:45 +0000)
committerTom Stellard <thomas.stellard@amd.com>
Fri, 3 Jun 2016 16:45:46 +0000 (16:45 +0000)
commitb4989f049a478f74c89fb7c3ad9064ca54e506a9
treec1d7a452a0b1ca34ae74c0b4d3644131426bcbe6
parent737edaf0482f208d3aaf92790dfc83bc9c4fe480
Merging r260658:

------------------------------------------------------------------------
r260658 | Matthew.Arsenault | 2016-02-11 22:31:30 -0800 (Thu, 11 Feb 2016) | 12 lines

AMDGPU: Set flat_scratch from flat_scratch_init reg

This was hardcoded to the static private size, but this
would be missing the offset and additional size for someday
when we have dynamic sizing.

Also stops always initializing flat_scratch even when unused.

In the future we should stop emitting this unless flat instructions
are used to access private memory. For example this will initialize
it almost always on VI because flat is used for global access.

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

git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_38@271684 91177308-0d34-0410-b5e6-96231b3b80d8
14 files changed:
lib/Target/AMDGPU/SIFrameLowering.cpp
lib/Target/AMDGPU/SIISelLowering.cpp
lib/Target/AMDGPU/SILowerControlFlow.cpp
lib/Target/AMDGPU/SIMachineFunctionInfo.cpp
lib/Target/AMDGPU/SIMachineFunctionInfo.h
lib/Target/AMDGPU/SIRegisterInfo.cpp
lib/Target/AMDGPU/SIRegisterInfo.h
test/CodeGen/AMDGPU/cgp-addressing-modes.ll
test/CodeGen/AMDGPU/ci-use-flat-for-global.ll
test/CodeGen/AMDGPU/global_atomics.ll
test/CodeGen/AMDGPU/hsa.ll
test/CodeGen/AMDGPU/large-alloca-compute.ll
test/CodeGen/AMDGPU/private-memory.ll
test/CodeGen/AMDGPU/vgpr-spill-emergency-stack-slot-compute.ll