[BE][cuDNN] cuDNN to 9.7.1.26 for CUDA 12.8#146717
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/146717
Note: Links to docs will display an error until the docs builds have been completed. ❌ 3 New Failures, 2 Unrelated FailuresAs of commit be24249 with merge base fc5913b ( NEW FAILURES - The following jobs have failed:
FLAKY - The following jobs failed but were likely due to flakiness present on trunk:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
Skylion007
left a comment
There was a problem hiding this comment.
Looks good assuming the binaries are uploaded to the S3 nightly bucket
|
Looks like 9.7.1.26 wheels (https://pypi.org/project/nvidia-cudnn-cu12/#files) need to be uploaded to https://download.pytorch.org/whl/nightly/cu128/nvidia-cudnn-cu12/ |
|
@pytorchbot rebase |
|
@pytorchbot started a rebase job onto refs/remotes/origin/viable/strict. Check the current status here |
|
Successfully rebased |
81e480e to
be24249
Compare
rebasing for #146717 Pull Request resolved: #146957 Approved by: https://github.com/Skylion007, https://github.com/eqy, https://github.com/nWEIdia, https://github.com/atalman Co-authored-by: Aaron Gokaslan <aaronGokaslan@gmail.com>
|
We should also update CUDNN for 12.6, right? |
|
Looks like this PR hasn't been updated in a while so we're going to go ahead and mark this as |
cuDNN 9.7.1 is out now and is expected to be the longer-lived branch with more potential backports vs. 9.7.0
CC @nWEIdia @tinglvv
cc @malfet @seemethere @csarofeen @ptrblck @xwang233