This reverts commit f39655c0c8.
Reason for revert: patch does not fix all stream hangs. So another patch was merged which fixes all issues.
Change-Id: I332d1ea29c23747b46b7667fe3e34e0ceefd2b23
- stream capture should be done before any sync APIs.
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com>
Change-Id: I3d65f67ee68777be71f97f48d460ccaefdd4e1af
Skip destruction of the null queue for MT. Windows can destroy worker
thread on exit during the stream destruction, causing a race condition.
Change-Id: I3e4bc3ab2b477d38c0aec1fc59e6f1af0f8f6b0f
Parallel streams can be destroyed during capture. Make sure runtime
removes the destoryed streams from the list of parallel captured streams
Change-Id: I46b6cbb951e8711f0bf02f2826c3b890b2541ffe
- Aggregate all TLS(Thread Local Storage) variables into a single class
- This is to improve cache accesses per thread
Change-Id: Ic8361eaeae290fff00254684e309471958365eb9
Using device write to set 0 to signal while streamWaitEvent is checking
the signal.
Use app in SWDEV-314307 to verify this fix
Requires vdi #668012 change to work
Change-Id: Ie329a29cfaeb5d144b92cda36773646f913ca73d
Add lock protection for access to the pool list.
Remove destroyed stream from the list of the safe streams
Change-Id: I1863b89bd3f5e188c161227cc790c3adaf72cc58
The actual size of allocation can be different from the original
request due to alignments. Pass the size of the original request
for validation during the arena object look-up
Change-Id: I30000046b1ec56abe743e43113046144fd3edbc6
HIP_MEM_POOL_SUPPORT controls memory pool support in runtime.
Currently it's disabled by default. The initial change doesn't
include: IPC, MGPU, virtual memory alloc, suballoc, defragmentation,
internal dependencies.
Change-Id: Ibed8528ebec698b045ebb247e49c0ecd6e587ed7
Set affinity to the node nearest to default GPU at init. Afterthat
set it to NUMA node thats nearest to whatever GPU is set with
hipSetDevice
Change-Id: I85749258ea7c25385096ffe4089a70c948f332c7
Change-Id: I99a92c922655e22955bee512073b6ac8e6ced3a2
hipGetLastError/hipPeekAtLastError should really return the last error
produced by any of the runtime calls that did not successfuly complete,
not just the error code of the last runtime call.
For example, in this snippet of code:
1: hipLaunchKernelGGL(MyKernel, dim3(2), dim3(128), 0, 0, 10000,x);
2: hipDeviceSynchronize();
3: hipError_t error = hipGetLastError();
If the call to hipLaunchKernelGGL fails, hipDeviceSynchronize (which
succeeds) should not reset the last error to hipSuccess. hipGetLastError
should still return the non-success error code returned by
hipLaunchKernelGGL.
The last error is reset to hipSuccess after calling hipGetLastError.
Change-Id: Ib7c039067c53c94c99c8ecd83f54212bcef06f81
Add a hip::Stream::destroyAllStreams static function to destroy all streams
Also call MemObjMap::Purge
Change-Id: I6ee7e3f26ab1f1870a9271c65e99fb818134482b
Ioannis Assiouras
pghafari
Ajay GunaShekar
Anusha GodavarthySurya
sdashmiz
German
Rakesh Roy
Saleel Kudchadker
Sarbojit Sarkar
Jaydeep
Sourabh Betigeri
Christophe Paquot
Satyanvesh Dittakavi
haoyuan2
kjayapra-amd
Maneesh Gupta
Laurent Morichetti
Julia Jiang
Vladislav Sytchenko
Rahul Garg