-
Notifications
You must be signed in to change notification settings - Fork 463
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
Use Nova workflow to host all published wheel files at PyTorch site #1958
base: main
Are you sure you want to change the base?
Conversation
✅ Deploy Preview for pytorch-fbgemm-docs canceled.
|
This pull request was exported from Phabricator. Differential Revision: D48487102 |
This pull request was exported from Phabricator. Differential Revision: D48487102 |
…ytorch#1958) Summary: Pull Request resolved: pytorch#1958 To alleviate CUDA version mismatch issues, we publish fbgemm-gpu-nightly with different CUDA versions. This diff uses Nova workflow host the published wheel files at PyTorch site instead. pytorch#1947 Differential Revision: D48487102 fbshipit-source-id: 2c09e5063307207c45cefe96c1fe5d8f33a51bc1
f573c22
to
a700dae
Compare
This pull request was exported from Phabricator. Differential Revision: D48487102 |
…ytorch#1958) Summary: Pull Request resolved: pytorch#1958 To alleviate CUDA version mismatch issues, we publish fbgemm-gpu-nightly with different CUDA versions. This diff uses Nova workflow host the published wheel files at PyTorch site instead. pytorch#1947 Differential Revision: D48487102 fbshipit-source-id: 3baeedad2fd554b63267eaf6ea6023a01f787de5
a700dae
to
076278f
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Generally looks good! Left a few nits, and might have to iterate a bit to get the builds passing in CI
pre-script: .github/script/nova_prescript.bash | ||
post-script: .github/script/nova_prescript.bash |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Should be .github/scripts
instead of .github/script
. This looks like the cause of the current breakage.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thank you for catching that! I have fixed it.
ref: "" | ||
pre-script: .github/script/nova_prescript.bash | ||
post-script: .github/script/nova_prescript.bash | ||
smoke-test-script: |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Let's explicitly make the value passed to this arg be ""
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thank you, done.
.github/scripts/nova_prescript.bash
Outdated
#Install CUDA | ||
$PRELUDE; install_cuda $BUILD_ENV "$CU_VERSION" | ||
# Install PyTorch Nightly | ||
$PRELUDE; install_pytorch_pip $BUILD_ENV nightly cuda "$CU_VERSION" |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
CUDA and the right pytorch version should already be installed for you under the hood by our pipeline, there is no need to do so in your pre-script.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thank you so much, I've removed them.
This pull request was exported from Phabricator. Differential Revision: D48487102 |
…ytorch#1958) Summary: Pull Request resolved: pytorch#1958 To alleviate CUDA version mismatch issues, we publish fbgemm-gpu-nightly with different CUDA versions. This diff uses Nova workflow host the published wheel files at PyTorch site instead. pytorch#1947 Differential Revision: D48487102 fbshipit-source-id: 6464b7832f50502c2b4925cea5b5cb62b3880611
076278f
to
cd0f731
Compare
HI @spcyppt please fix this failure: https://github.com/pytorch/FBGEMM/actions/runs/5908914110/job/16028907266?pr=1958 |
This pull request was exported from Phabricator. Differential Revision: D48487102 |
…ytorch#1958) Summary: Pull Request resolved: pytorch#1958 To alleviate CUDA version mismatch issues, we publish fbgemm-gpu-nightly with different CUDA versions. This diff uses Nova workflow host the published wheel files at PyTorch site instead. pytorch#1947 Differential Revision: D48487102 fbshipit-source-id: 2a468a86c430399be26f13de114e45c5062b7526
cd0f731
to
0a48dec
Compare
This pull request was exported from Phabricator. Differential Revision: D48487102 |
…ytorch#1958) Summary: Pull Request resolved: pytorch#1958 To alleviate CUDA version mismatch issues, we publish fbgemm-gpu-nightly with different CUDA versions. This diff uses Nova workflow host the published wheel files at PyTorch site instead. pytorch#1947 Differential Revision: D48487102 fbshipit-source-id: 185f3963e6b8ea4b7daa82f50bb055b8da6bf5b9
0a48dec
to
8425266
Compare
This pull request was exported from Phabricator. Differential Revision: D48487102 |
…ytorch#1958) Summary: Pull Request resolved: pytorch#1958 To alleviate CUDA version mismatch issues, we publish fbgemm-gpu-nightly with different CUDA versions. This diff uses Nova workflow host the published wheel files at PyTorch site instead. pytorch#1947 Differential Revision: D48487102 fbshipit-source-id: 06704724ad1e4f3be8a24146912823195c33998d
8425266
to
06138c3
Compare
This pull request was exported from Phabricator. Differential Revision: D48487102 |
…ytorch#1958) Summary: Pull Request resolved: pytorch#1958 To alleviate CUDA version mismatch issues, we publish fbgemm-gpu-nightly with different CUDA versions. This diff uses Nova workflow host the published wheel files at PyTorch site instead. pytorch#1947 Differential Revision: D48487102 fbshipit-source-id: ce52bd616e60baf1a2e3077f02b57073538b6b6b
06138c3
to
2207160
Compare
This pull request was exported from Phabricator. Differential Revision: D48487102 |
…ytorch#1958) Summary: Pull Request resolved: pytorch#1958 To alleviate CUDA version mismatch issues, we publish fbgemm-gpu-nightly with different CUDA versions. This diff uses Nova workflow host the published wheel files at PyTorch site instead. pytorch#1947 Differential Revision: D48487102 fbshipit-source-id: 78af627fe2bafc777d4abc52172ea83e46b192a4
2207160
to
3c27900
Compare
…into export-D48487102
…into export-D48487102
…into export-D48487102
…into export-D48487102
…into export-D48487102
…into export-D48487102
…into export-D48487102
…into export-D48487102
…into export-D48487102
Summary:
To alleviate CUDA version mismatch issues, we publish fbgemm-gpu-nightly with different CUDA versions. This diff uses Nova workflow host the published wheel files at PyTorch site instead.
#1947
Differential Revision: D48487102