-
Notifications
You must be signed in to change notification settings - Fork 24.3k
[FlexAttention] Fix device test instantation #151846
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
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/151846
Note: Links to docs will display an error until the docs builds have been completed. ⏳ No Failures, 5 PendingAs of commit 7703789 with merge base cc793e8 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx ipiszy chenyang78 kadeng muchulee8 amjames chauhang aakhundov [ghstack-poisoned]
B = 4 | ||
H = 8 | ||
S = 2048 | ||
B = 2 |
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.
purposeful
ghstack-source-id: 8dc2ecd Pull Request resolved: pytorch/pytorch#151846
@@ -90,6 +94,18 @@ def temp_float32_matmul_precision(precision: str): | |||
torch.set_float32_matmul_precision(original_precision) | |||
|
|||
|
|||
def skip_on_cpu(test_func): | |||
"""Decorator to skip tests that are not supported on CPU.""" | |||
decorated_func = skipCPUIf(True, "Not supported on CUDA")(test_func) |
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.
decorated_func = skipCPUIf(True, "Not supported on CUDA")(test_func) | |
decorated_func = skipCPUIf(True, "Not supported on CPU")(test_func) |
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.
There are alot of flaky test failures so will land in follow up, but goood catch
Merge failedReason: Command
Details for Dev Infra teamRaised by workflow job |
@pytorchbot merge -f "I ran the full CI everything was green and last minute merge conflict" |
Merge startedYour change will be merged immediately since you used the force (-f) flag, bypassing any CI checks (ETA: 1-5 minutes). Please use Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
@drisspg Looks like this PR broke at least the following test in I've added the @pytorchbot revert -c nosignal -m "PR broke cc @huydhn |
@pytorchbot successfully started a revert job. Check the current status here. |
Don't want to revert based on edited command |
Trying again, since the edit to the comment seemed to displease the bot: @drisspg Looks like this PR broke at least the following test in rocm workflow: inductor/test_flex_attention.py::TestFlexAttentionCUDA::test_GQA_score_mod5_cuda_float16 I've added the ciflow/rocm label on this PR to surface the failure. @pytorchbot revert -c nosignal -m "PR broke rocm workflow" cc @huydhn |
@pytorchbot successfully started a revert job. Check the current status here. |
This reverts commit b37fa20. Reverted #151846 on behalf of https://github.com/jithunnair-amd due to PR broke rocm workflow ([comment](#151846 (comment)))
@drisspg your PR has been successfully reverted. |
@jithunnair-amd what is the failure? |
Starting merge as part of PR stack under #151959 |
Fixes: #148827 Pull Request resolved: #151959 Approved by: https://github.com/Chillee ghstack dependencies: #151846
Pull Request resolved: pytorch#151846 Approved by: https://github.com/Chillee, https://github.com/BoyuanFeng, https://github.com/mlazos
) Fixes: pytorch#148827 Pull Request resolved: pytorch#151959 Approved by: https://github.com/Chillee ghstack dependencies: pytorch#151846
Stack from ghstack (oldest at bottom):
cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @ipiszy @chenyang78 @kadeng @muchulee8 @amjames @chauhang @aakhundov