llvm-project/clang/test/CodeGenCUDA
Artem Belevich 52cc487ba8 [cuda] Include GPU binary into host object file and generate init/deinit code.
- added -fcuda-include-gpubinary option to incorporate results of
  device-side compilation into host-side one.
- generate code to register GPU binaries and associated kernels
  with CUDA runtime and clean-up on exit.
- added test case for init/deinit code generation.

Differential Revision: http://reviews.llvm.org/D9507

llvm-svn: 236765
2015-05-07 19:34:16 +00:00
..
Inputs Move all CUDA testing inputs to Inputs/ subdirectory inside the tests. 2014-04-28 22:21:28 +00:00
address-spaces.cu Fix addrspace when emitting constructors of static local variables 2015-03-25 20:06:28 +00:00
cuda-builtin-vars.cu [cuda] Added support for CUDA built-in variables. 2015-04-21 22:14:13 +00:00
device-stub.cu [cuda] Include GPU binary into host object file and generate init/deinit code. 2015-05-07 19:34:16 +00:00
filter-decl.cu [cuda] treat file scope __asm as __host__ and ignore it during device-side compilation. 2015-04-27 18:52:00 +00:00
host-device-calls-host.cu CUDA: Add option to allow host device functions to call host functions 2015-02-24 21:45:33 +00:00
kernel-call.cu Move all CUDA testing inputs to Inputs/ subdirectory inside the tests. 2014-04-28 22:21:28 +00:00
launch-bounds.cu [cuda] Allow using integral non-type template parameters as launch_bounds attribute arguments. 2015-04-21 22:55:54 +00:00
llvm-used.cu When generating llvm.used, we may need an addrspacecast instead of a bitcast. 2015-02-02 21:05:49 +00:00
ptx-kernels.cu IR: Make metadata typeless in assembly, clang side 2014-12-15 19:10:08 +00:00