add/fix patches for PyTorch 1.13.1 w/ foss/2022a #18371
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Test report by @Flamefire |
|
Test report by @Flamefire |
This was referenced Jul 31, 2023
|
@boegelbot please test @ generoso |
|
@boegel: Request for testing this PR well received on login1 PR test command '
Test results coming soon (I hope)... - notification for comment with ID 1659852530 processed Message to humans: this is just bookkeeping information for me, |
|
@boegelbot please test @ jsc-zen2 |
|
@boegel: Request for testing this PR well received on jsczen2l1.int.jsc-zen2.easybuild-test.cluster PR test command '
Test results coming soon (I hope)... - notification for comment with ID 1660003521 processed Message to humans: this is just bookkeeping information for me, |
|
Test report by @boegelbot |
|
Test report by @boegelbot |
|
Test report by @boegel |
Flamefire
changed the title
|
@boegel Could you attach the log please as your build seems to be the only failing one ? Quick guess: Have you reinstalled GCC and OpenBLAS with #16411? A bit confused because it seems to have worked for you in #17155 and the patches only relax stuff without the possibility to introduce failures as far as I can tell. |
|
@Flamefire The Here's the relevant part of the log for the failing I'm happy to merge this and follow up on this problem in another PR (one step at a time, especially with PyTorch). |
Sounds like a good idea. Especially after now I've found yet another issue: Currently testing the updated patch and it likely makes sense to add them to all ECs in a single PR especially as that only affects POWER. |
|
Test report by @SebastianAchilles |
|
@SebastianAchilles Does your GCCcore module include the |
boegel
changed the title
boegel
approved these changes
Aug 8, 2023
|
Merging this as-is, since it's a step forward, more fixes can be done in a follow-up PR |
|
Going in, thanks @Flamefire! |
Yes, the installed GCCcore includes |
|
Strange. I also ran this multiple times without issues, both completely and only this test. Might want to keep that in mind for later. If you still have the log, can you post the exact name of the subtest in |
|
@casparvl You download to |
Unfortunately I didn't have the logs anymore. I rebuilt this MR 5 times again on my machines and the test was successful each time. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
(created using
eb --new-pr)test_quantizationhave the same cause as those intest_ao_sparsitytest_opsfailure persisted as the forward-port of the patch doesn't work, added the commit from my upstream PR instead*-fix-fsdp-fp16-test.patchwas outdated due to changes upstream. Although it applied it didn't have the expected effect