There are now two implementations of printf in HIP:
1. The implemenation for HCC is controlled by the HC_FEATURE_PRINTF
macro, and it works only with the HCC compiler used in combination
with the HCC runtime.
2. The implementation for hip-clang requires the VDI runtime, and is
always enabled with that combination.
The printf call in the device code is expanded by the compiler into a
series of hostcalls that together form a "message". This change
introduces the following functionality in the runtime:
1. Receive a generic message consisting of a series of hostcalls.
2. Process a printf message.
Change-Id: I9d667d6f91607a907a96e46cc5fca55734339747
HIP assumes that image width is in bytes, but VDI assumes that image width in pixels. Need to perform byte -> pixel coversion before doing anything.
Change-Id: Ia9fd1f46d05db3fbe8049add10b4d7e5118a2b9a
~45% to 50% of Performance drop on rocBLAS_int8 test
Add support for active waits without blocking the host thread.
Change-Id: Ie7bb48dcafcb4c93d448bf74749b829b626c3578
Fix compilation error on hip-hcc+clang , hip-vdi+clang
Enabled hipExtLaunchMultiKernelMultiDevice test on hip-vdi path
hipExtLaunchMultiKernelMultiDevice common declaration for all paths
Change-Id: I76031840614fce8e12a8e845548fa43a389a741a
SWDEV-225266: [HIP-VDI] HIP-VDI disabled tests (p2p_copy_coherency.cpp)
SWDEV-225388: hipTestDeviceSymbol.cpp & hipTestConstant.cpp failed to build on hip-vdi
For hipTestDeviceSymbol.cpp & hipTestConstant.cpp tests:
Currently "__HIP_VDI__" flag is enabled in CMakeLists.txt, but when application is compiled with hipcc,
__HIP_VDI__ is not defined to differentiate if compiled for VDI/HCC for headers.
For ./src/runtimeApi/memory/p2p_copy_coherency.cpp:
Fixed compilation issue to include only when compile for HCC runtime "<hc_am.hpp> not found"
Currently test is disabled to run on all platforms. When validated on multi-GPU machine,
memcpy between multiple GPUs via GPU synchronization is not working on hcc and vdi path.
Need to validate on nvidia machine to know if test is valid. Disabled GPU synchronization test for now.
For ./src/runtimeApi/module/hipModuleTexture2dDrv.cpp:
updated test to generate tex2d_kernel.code object in build directory. Currently ctest looks for it in build directory.
Change-Id: I629d395a919c2440d921422716944c7940ed6010
cl_bool needed to be replaced with uint32_t instead of bool. This is due to cl_bool being a typedef of cl_uint32.
Currently clGetDeviceInfo() reports incorrect size for the return value, due to cl_bool being 4 bytes and c++ bool being 1 byte.
Change-Id: I647a4b8873627059865c84c8ca27694dbc0916de
~45% to 50% of Performance drop on rocBLAS_int8 test
Enable cudaSetDeviceFlags() api call. Use active wait by default
for all devices.
Change-Id: Ifc2ebe3dd9b0aa3fdbfbc9cb5c2cd8b3b726124f
Incoming changes from upstream split the struct hipMemcpy3DParms into two separate ones - hipMemcpy3DParms and HIP_MEMCPY3D, which are cudaMemcpy3DParms and CUDA_MEMCPY3D equivalents respectively.
Note that HIP_MEMCPY3D is missing half the members of CUDA_MEMCPY3D (this should be fixed in PR#1887). Work around this by using a substitute _HIP_MEMCPY3D struct for now.
Change-Id: Ic15134e6deb260189b662b3804d2309a9b8473e9
Add MS HWS support. PAL reports just one compute engine
in that mode and runtime needs extra logic to detect RT queues.
Change-Id: I011f1f1b18dec6a7195a4f1fe939f8029bc269ae
Sameer Sahasrabuddhe
Alex Voicu
Chauncey Hui
Aryan Salmanpour
Saleel Kudchadker
Vladislav Sytchenko
Lad, Aditya
Tao Sang
Evgeny Mankov
jujiang
agodavar
Maneesh Gupta
German Andryeyev
Michael LIAO
Payam
Jason Tang
Jatin Chaudhary
Siu Chi Chan
Rahul Garg