Tensile
Tensile copied to clipboard
Use hipMemcpyAsync for validation
- replace hipMemcpy with hipMemcpyAsync + hipStreamSynchronize in ReferenceValidator
This is a trial fix for random mismatch with device result 0. I will run CI test multiple times with this change.
I will fix the build error...
Reference rocblas commit for hipMemcpyAsync: https://github.com/ROCm/rocBLAS/commit/c070df9bdc31cce30e0d4d732ad8a60f3b2ee332
I ran precheckin test 11 times and extended test 7 times. I did not see mismatch issue with device result=0 so far. I will merge this change.