diff options
| author | Volkan Keles <vkeles@apple.com> | 2016-10-03 10:31:34 +0000 |
|---|---|---|
| committer | Volkan Keles <vkeles@apple.com> | 2016-10-03 10:31:34 +0000 |
| commit | 1c38681ae643430a11738414f70c68ad93c48eb4 (patch) | |
| tree | cafa1e19d82e7eed9d24fd7baa7b73ac1754da5f /llvm/lib/Target/AMDGPU | |
| parent | 4dbe73c1ed74352899eedf2a07c202ea76c2f65b (diff) | |
| download | bcm5719-llvm-1c38681ae643430a11738414f70c68ad93c48eb4.tar.gz bcm5719-llvm-1c38681ae643430a11738414f70c68ad93c48eb4.zip | |
Add new target hooks for LoadStoreVectorizer
Summary: Added 6 new target hooks for the vectorizer in order to filter types, handle size constraints and decide how to split chains.
Reviewers: tstellarAMD, arsenm
Subscribers: arsenm, mzolotukhin, wdng, llvm-commits, nhaehnle
Differential Revision: https://reviews.llvm.org/D24727
llvm-svn: 283099
Diffstat (limited to 'llvm/lib/Target/AMDGPU')
| -rw-r--r-- | llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp | 2 | ||||
| -rw-r--r-- | llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.h | 2 |
2 files changed, 2 insertions, 2 deletions
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp b/llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp index 3d630fe3ea9..e328e0c5a43 100644 --- a/llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp +++ b/llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp @@ -80,7 +80,7 @@ unsigned AMDGPUTTIImpl::getRegisterBitWidth(bool Vector) { return Vector ? 0 : 32; } -unsigned AMDGPUTTIImpl::getLoadStoreVecRegBitWidth(unsigned AddrSpace) { +unsigned AMDGPUTTIImpl::getLoadStoreVecRegBitWidth(unsigned AddrSpace) const { switch (AddrSpace) { case AMDGPUAS::GLOBAL_ADDRESS: case AMDGPUAS::CONSTANT_ADDRESS: diff --git a/llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.h b/llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.h index a82a0745808..8c8be63b53b 100644 --- a/llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.h +++ b/llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.h @@ -82,7 +82,7 @@ public: unsigned getNumberOfRegisters(bool Vector); unsigned getRegisterBitWidth(bool Vector); - unsigned getLoadStoreVecRegBitWidth(unsigned AddrSpace); + unsigned getLoadStoreVecRegBitWidth(unsigned AddrSpace) const; unsigned getMaxInterleaveFactor(unsigned VF); int getArithmeticInstrCost( |
