This revision was automatically updated to reflect the committed changes.
Closed by commit rL330296: [NVPTX, CUDA] Added support for m8n32k16 and
m32n8k16 variants of wmma… (authored by tra, committed by ).
Changed prior to commit:
https://reviews.llvm.org/D45068?vs=143003=143005#toc
tra updated this revision to Diff 143003.
tra added a comment.
Updated the way we specify TARGET_BUILTIN feature constraints
https://reviews.llvm.org/D45068
Files:
clang/include/clang/Basic/BuiltinsNVPTX.def
clang/lib/CodeGen/CGBuiltin.cpp
clang/lib/Driver/ToolChains/Cuda.cpp
tra updated this revision to Diff 142262.
tra added a comment.
Updated BuiltinsNVPTX.def and tests to deal with changes in the way we deal
with required features in TARGET_BUILTIN.
https://reviews.llvm.org/D45068
Files:
clang/include/clang/Basic/BuiltinsNVPTX.def
tra updated this revision to Diff 140363.
tra added a comment.
Updated comments.
https://reviews.llvm.org/D45068
Files:
clang/include/clang/Basic/BuiltinsNVPTX.def
clang/lib/CodeGen/CGBuiltin.cpp
clang/lib/Driver/ToolChains/Cuda.cpp
clang/test/CodeGen/builtins-nvptx-sm_70.cu
jlebar accepted this revision.
jlebar added inline comments.
This revision is now accepted and ready to land.
Comment at: clang/lib/Driver/ToolChains/Cuda.cpp:578
+ if (CudaInstallation.version() >= CudaVersion::CUDA_91) {
+// CUDA-9 uses new instructions that are only
tra created this revision.
tra added a reviewer: jlebar.
Herald added subscribers: hiraditya, sanjoy, jholewinski.
The new instructions were added added for sm_70+ GPUs in CUDA-9.1.
https://reviews.llvm.org/D45068
Files:
clang/include/clang/Basic/BuiltinsNVPTX.def