Sidharth Kashyap

OneDNN v1.7 improves the performance for older architectures too, including SSE4.1 for int8 https://github.com/oneapi-src/oneDNN/releases/tag/v1.7 We provided the Matrix Multiplications ranks from Marian Inference for oneDNN to be optimized, the latest...

Hey @XapaJIaMnu, do we have a priority list of functions in MKL that need to be oneDNN? I will work with the oneDNN team to sort that out if possible.

Thank you, this will help in getting the request quantified. Will update on the progress soon.

Had an internal discussion with @vpirogov from the oneDNN team, unfortunately the support for FAISS MKL dependencies cannot be addressed in oneDNN as it is outside the Deep Learning remit...

This issue was indeed discussed with the team; @akarbown, will ping you offline to take this forward. @kpu: Yes, will setup a CI for icc.

@akarbown: Have forwarded the meeting minutes with the MKL where we discussed these issues, let us sync offline.