Unit testing of CG threadRank api in multi-grid environment requires
some changes as grid id of particular device is not known apriori.
Change-Id: Ie7941a09337653f3ada5eedfb7c64a93a234498b
- Matching the output "hipcc --version" for Installed directory
to get the llvm compiler path, wasnt done properly.
- Adding fallback for HIP_CLANG_ROOT on ROCM_PATH if defined in env
- Changed the mode of get_filename_component to DIRECTORY
Signed-off-by: Pruthvi Madugundu <pruthvi.madugundu@amd.com>
Change-Id: Ia38e234f34cabb5b419906c7094b327083292b0e
[background] There is currently a compilation issue when both hip/hip_runtime.h and hip/hip_cooperative_groups.h included
in a file and hipcc used to compile it on NV platform. The issue is that an abort is defined in hip/nvcc_detail/hip_runtime.h
and it is also defined in the CUDA cooperative groups header (/cuda/include/cooperative_groups/details/helpers.h).
this is problematic and leads to a compilation issue in hipcc on NV platform.
Change-Id: I2ab6982ac4103822a1a4a0ced942cd604d6c19c1
The static members __HIP_Coordinates::x, ::y and ::z must be defined
outside the class. Otherwise, linker throws `undefined reference error`
when these definitions are needed in the HIP application.
Change-Id: Iabc09744b478c22e4b13cf9824877ec9cfdd4f7a
SWDEV-249719 - root cause: queues with custom CU mask are not inserted
into queuePool_ (i.e., queue of reusable HSA queues) of ROC device class
causing a crash when creating hostcall buffers for printf
Change-Id: Ieee7005d9a5a30b3113394ce23ee65927126d0d6
5 scenarios were added for hipMallocManaged including negative
test.
SWDEV-238517 for enhancing hip unit tests
Change-Id: Ibba53fb64502d6df798e4e63100dcb8c5f94553f
1. API validation
2. Validate the memory status after hipMalloc()
3. Test repeated hipMalloc() Calls, call in multiple threads
4. Parallel hipMalloc() from parent and child processes
5. hipMalloc() from multiple processes on multiple GPUs
SWDEV-238517 for enhancing hip unit tests
Change-Id: Ie152999d7a203a51704d2888ccc1ca866a2e1b05
Tests allocation of memory on various GPUs and pass on the IPC mem handle to child process.
Child process attaches to the mem handle and uses the memory on various GPUs
SWDEV-238517 for enhancing hip unit tests
Change-Id: I7babc7725813e05d651057beef6a8bec6f62a49b
The os.hpp header gets added to the include path of legacy llvm via the compiler lib. Having "windows.h" included causes a lot conflicts with LLVM headers, as they forward declare many Windows types. Best to not include it here.
Change-Id: I60c44a8d28660368f1a4a95741e1053ef3528fa1
German Andryeyev
Tao Sang
Vladislav Sytchenko
Jatin Chaudhary
Chauncey Hui
mshivama
Pruthvi Madugundu
Srinivasan Subramanian
Aryan Salmanpour
Cole Nelson
Sarbojit Sarkar
Julia Jiang
kjayapra-amd
Jason Tang
Rahul Garg
agodavar
Lakhan singh Thakur
Sumanth Tumbalam Gooty
Sourabh Betigeri
Todd tiantuo Li