diff options
| author | Artem Belevich <tra@google.com> | 2016-08-12 18:44:01 +0000 |
|---|---|---|
| committer | Artem Belevich <tra@google.com> | 2016-08-12 18:44:01 +0000 |
| commit | 4c09318be2dd449812345a8277565589cebd25e5 (patch) | |
| tree | 6e8d32a64e44c49df691977bcbc5e7be205ef914 /llvm/lib/Object/MachOObjectFile.cpp | |
| parent | dc698c3e91997836a7ae3f2d5a0766469edf0d20 (diff) | |
| download | llvm-4c09318be2dd449812345a8277565589cebd25e5.zip llvm-4c09318be2dd449812345a8277565589cebd25e5.tar.gz llvm-4c09318be2dd449812345a8277565589cebd25e5.tar.bz2 | |
[CUDA] Place GPU binary into .nv_fatbin section and align it by 8.
This matches the way nvcc encapsulates GPU binaries into host object file.
Now cuobjdump can deal with clang-compiled object files.
Differential Revision: https://reviews.llvm.org/D23429
llvm-svn: 278549
Diffstat (limited to 'llvm/lib/Object/MachOObjectFile.cpp')
0 files changed, 0 insertions, 0 deletions