ROCm / MIOpen

AMD's Machine Intelligence Library
https://rocm.docs.amd.com/projects/MIOpen/en/latest/
Other
1.09k stars 231 forks source link

Replace sync hipMemset with async hipMemset #3196

Closed iq136boy closed 3 months ago

iq136boy commented 3 months ago

Replace sync hipMemset with async hipMemset to fix the issue that hipMemSet call taking too long during convolution backward weight for CVT cases.