Skip to content

v0.21.1

Compare
Choose a tag to compare
@tprimak tprimak released this 28 Sep 00:07
· 51 commits to rls-v0.21 since this release

This is a patch release containing following changes to Intel MKL-DNN v0.21:

  • Fixed output channel blocking logic in forward AVX2 convolution that could lead to incorrect result or segfault (6accb47)
  • Fixed int8 grouped convolution for some shapes with the number of input or output channels not being a multiple of 8 on Intel AVX512 systems (878ac2d)