projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 3ae8dfd) | patch
[NVPTX] Implement __nvvm_atom_add_gen_d builtin.
authorJustin Lebar <jlebar@google.com>
Tue, 7 Nov 2017 22:10:54 +0000 (22:10 +0000)
committerJustin Lebar <jlebar@google.com>
Tue, 7 Nov 2017 22:10:54 +0000 (22:10 +0000)
commitda9e0bd3a238f80b05a3070c8a8a3cadadc12e5f
tree011bda2e7fcd416411315d3d88afd8309e9aeb08tree | snapshot
parent3ae8dfda06ff63d16673b1db9e0ce73ce0b8a3f5commit | diff
[NVPTX] Implement __nvvm_atom_add_gen_d builtin.

Summary:
This just seems to have been an oversight.  We already supported the f64
atomic add with an explicit scope (e.g. "cta"), but not the scopeless
version.

Reviewers: tra

Subscribers: jholewinski, sanjoy, cfe-commits, llvm-commits, hiraditya

Differential Revision: https://reviews.llvm.org/D39638

llvm-svn: 317623
clang/include/clang/Basic/BuiltinsNVPTX.def diff | blob | history
clang/lib/CodeGen/CGBuiltin.cpp diff | blob | history
clang/test/CodeGen/builtins-nvptx-ptx50.cu [new file with mode: 0644] blob
llvm/include/llvm/IR/IntrinsicsNVVM.td diff | blob | history
llvm/lib/Target/NVPTX/NVPTXISelLowering.cpp diff | blob | history
llvm/lib/Target/NVPTX/NVPTXIntrinsics.td diff | blob | history
llvm/test/CodeGen/NVPTX/atomics-sm60.ll [new file with mode: 0644] blob
Domain: System / Toolchain;