Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[TOPI,CUDA] Don't enable cudnn conv2d kernel if is not supported #10021

Merged
merged 2 commits into from
Jan 22, 2022

Conversation

tkonolige
Copy link
Contributor

Specifically, check that layout is not NCHW if datatype is int8.

@mbrookhart

Specifically, check that layout is not NCHW if datatype is int8.
@tkonolige
Copy link
Contributor Author

I've updated this PR to not use conv2d_cudnn if any of the inputs are int8. Via talking with @masahi, the issue is that cudnn only supports int8 inputs with int8 output while we require int32 outputs.

@masahi masahi merged commit e9ee73f into apache:main Jan 22, 2022
yuanfz98 pushed a commit to yuanfz98/tvm that referenced this pull request Jan 24, 2022
…che#10021)

* [TOPI,CUDA] Don't enable cudnn conv2d kernel if is not supported

Specifically, check that layout is not NCHW if datatype is int8.

* remove all conv2d_cudnn int8 support
ylc pushed a commit to ylc/tvm that referenced this pull request Feb 16, 2022
…che#10021)

* [TOPI,CUDA] Don't enable cudnn conv2d kernel if is not supported

Specifically, check that layout is not NCHW if datatype is int8.

* remove all conv2d_cudnn int8 support
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants