Run functorch tests in default shards; delete functorch-specific shards #96464
Conversation
This file contains hidden or 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
Body to come soon [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/96464
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 FailuresAs of commit 6dd2479: NEW FAILURES - The following jobs have failed:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
|
Not ready for review yet |
zou3519
added
keep-going
…ecific shards" Body to come soon [ghstack-poisoned]
…ecific shards" Body to come soon [ghstack-poisoned]
…ecific shards" Body to come soon [ghstack-poisoned]
…ecific shards" Body to come soon [ghstack-poisoned]
…ecific shards" Body to come soon [ghstack-poisoned]
…ecific shards" Body to come soon [ghstack-poisoned]
…ecific shards" Body to come soon [ghstack-poisoned]
zou3519
requested review from
a team,
Chillee,
ezyang and
kshitij12345
as code owners
…ecific shards" Fixes #96347 This PR: - Makes the functorch tests run as a part of the "default" shards - Delete the functorch CI shard from all CI job configurations (if it exists) - Increase the "default" shard count by 1 for each job, unless it was previously set to 1, to accommodate the new functorch tests and not regress time-to-signal. - Adds a bunch of skips for ROCM and torchdynamo configurations. We can investigate them later. NB: I might go through some more iterations to figure out what other skips need to be added, but this iteration of the PR seems to pass most CI. suite. Test Plan: - wait for CI [ghstack-poisoned]
|
FYI, there is one remaining functorch shard for MacOS x86_64 in periodic https://github.com/pytorch/pytorch/blob/master/.github/workflows/periodic.yml#L313 |
huydhn
approved these changes
Mar 16, 2023
…ecific shards" Fixes #96347 This PR: - Makes the functorch tests run as a part of the "default" shards - Delete the functorch CI shard from all CI job configurations (if it exists) - Increase the "default" shard count by 1 for each job, unless it was previously set to 1, to accommodate the new functorch tests and not regress time-to-signal. - Adds a bunch of skips for ROCM and torchdynamo configurations. We can investigate them later. NB: I might go through some more iterations to figure out what other skips need to be added, but this iteration of the PR seems to pass most CI. suite. Test Plan: - wait for CI [ghstack-poisoned]
zou3519
added a commit
that referenced
this pull request
Mar 16, 2023
Fixes #96347 This PR: - Makes the functorch tests run as a part of the "default" shards - Delete the functorch CI shard from all CI job configurations (if it exists) - Increase the "default" shard count by 1 for each job, unless it was previously set to 1, to accommodate the new functorch tests and not regress time-to-signal. - Adds a bunch of skips for ROCM and torchdynamo configurations. We can investigate them later. NB: I might go through some more iterations to figure out what other skips need to be added, but this iteration of the PR seems to pass most CI. suite. Test Plan: - wait for CI ghstack-source-id: 13a4c38 Pull Request resolved: #96464
Good catch, I forgot about the jobs in periodic. I updated that shard and also increased the default shard count by 1 for the jobs in periodic |
…ecific shards" Fixes #96347 This PR: - Makes the functorch tests run as a part of the "default" shards - Delete the functorch CI shard from all CI job configurations (if it exists) - Increase the "default" shard count by 1 for each job, unless it was previously set to 1, to accommodate the new functorch tests and not regress time-to-signal. - Adds a bunch of skips for ROCM and torchdynamo configurations. We can investigate them later. NB: I might go through some more iterations to figure out what other skips need to be added, but this iteration of the PR seems to pass most CI. suite. Test Plan: - wait for CI [ghstack-poisoned]
zou3519
added a commit
that referenced
this pull request
Mar 16, 2023
Fixes #96347 This PR: - Makes the functorch tests run as a part of the "default" shards - Delete the functorch CI shard from all CI job configurations (if it exists) - Increase the "default" shard count by 1 for each job, unless it was previously set to 1, to accommodate the new functorch tests and not regress time-to-signal. - Adds a bunch of skips for ROCM and torchdynamo configurations. We can investigate them later. NB: I might go through some more iterations to figure out what other skips need to be added, but this iteration of the PR seems to pass most CI. suite. Test Plan: - wait for CI ghstack-source-id: ee52726 Pull Request resolved: #96464
…ecific shards" Fixes #96347 This PR: - Makes the functorch tests run as a part of the "default" shards - Delete the functorch CI shard from all CI job configurations (if it exists) - Increase the "default" shard count by 1 for each job, unless it was previously set to 1, to accommodate the new functorch tests and not regress time-to-signal. - Adds a bunch of skips for ROCM and torchdynamo configurations. We can investigate them later. NB: I might go through some more iterations to figure out what other skips need to be added, but this iteration of the PR seems to pass most CI. suite. Test Plan: - wait for CI [ghstack-poisoned]
zou3519
added a commit
that referenced
this pull request
Mar 20, 2023
Fixes #96347 This PR: - Makes the functorch tests run as a part of the "default" shards - Delete the functorch CI shard from all CI job configurations (if it exists) - Increase the "default" shard count by 1 for each job, unless it was previously set to 1, to accommodate the new functorch tests and not regress time-to-signal. - Adds a bunch of skips for ROCM and torchdynamo configurations. We can investigate them later. NB: I might go through some more iterations to figure out what other skips need to be added, but this iteration of the PR seems to pass most CI. suite. Test Plan: - wait for CI ghstack-source-id: ca719db Pull Request resolved: #96464
…ecific shards" Fixes #96347 This PR: - Makes the functorch tests run as a part of the "default" shards - Delete the functorch CI shard from all CI job configurations (if it exists) - Increase the "default" shard count by 1 for each job, unless it was previously set to 1, to accommodate the new functorch tests and not regress time-to-signal. - Adds a bunch of skips for ROCM and torchdynamo configurations. We can investigate them later. NB: I might go through some more iterations to figure out what other skips need to be added, but this iteration of the PR seems to pass most CI. suite. Test Plan: - wait for CI [ghstack-poisoned]
zou3519
added a commit
that referenced
this pull request
Mar 20, 2023
Fixes #96347 This PR: - Makes the functorch tests run as a part of the "default" shards - Delete the functorch CI shard from all CI job configurations (if it exists) - Increase the "default" shard count by 1 for each job, unless it was previously set to 1, to accommodate the new functorch tests and not regress time-to-signal. - Adds a bunch of skips for ROCM and torchdynamo configurations. We can investigate them later. NB: I might go through some more iterations to figure out what other skips need to be added, but this iteration of the PR seems to pass most CI. suite. Test Plan: - wait for CI ghstack-source-id: a3b703a Pull Request resolved: #96464
|
@pytorchbot merge -f "test failure looks flaky" |
Merge startedYour change will be merged immediately since you used the force (-f) flag, bypassing any CI checks (ETA: 1-5 minutes). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
pytorchmergebot
pushed a commit
that referenced
this pull request
Mar 21, 2023
Before #96464, ROCm tests in trunk are already quite flaky https://hud.pytorch.org/reliability/pytorch/pytorch?jobName=trunk%20%2F%20linux-focal-rocm5.4.2-py3.8%20%2F%20test%20(default). After #96464, there is a new group of flaky failures coming from functorch. So let's mark the test as flaky to monitor without impacting trunk. Two flaky tests currently seeing in trunk are: * #97256 * `functorch/test_memory_efficient_fusion.py` OOM Pull Request resolved: #97259 Approved by: https://github.com/malfet, https://github.com/zou3519
cyyever
pushed a commit
to cyyever/pytorch_private
that referenced
this pull request
Mar 23, 2023
…ds (#96464) Fixes #96347 This PR: - Makes the functorch tests run as a part of the "default" shards - Delete the functorch CI shard from all CI job configurations (if it exists) - Increase the "default" shard count by 1 for each job, unless it was previously set to 1, to accommodate the new functorch tests and not regress time-to-signal. - Adds a bunch of skips for ROCM and torchdynamo configurations. We can investigate them later. NB: I might go through some more iterations to figure out what other skips need to be added, but this iteration of the PR seems to pass most CI. suite. Test Plan: - wait for CI Pull Request resolved: pytorch/pytorch#96464 Approved by: https://github.com/huydhn
cyyever
pushed a commit
to cyyever/pytorch_private
that referenced
this pull request
Mar 23, 2023
Before pytorch/pytorch#96464, ROCm tests in trunk are already quite flaky https://hud.pytorch.org/reliability/pytorch/pytorch?jobName=trunk%20%2F%20linux-focal-rocm5.4.2-py3.8%20%2F%20test%20(default). After pytorch/pytorch#96464, there is a new group of flaky failures coming from functorch. So let's mark the test as flaky to monitor without impacting trunk. Two flaky tests currently seeing in trunk are: * pytorch/pytorch#97256 * `functorch/test_memory_efficient_fusion.py` OOM Pull Request resolved: pytorch/pytorch#97259 Approved by: https://github.com/malfet, https://github.com/zou3519
cyyever
pushed a commit
to cyyever/pytorch_private
that referenced
this pull request
Mar 27, 2023
…ds (#96464) Fixes #96347 This PR: - Makes the functorch tests run as a part of the "default" shards - Delete the functorch CI shard from all CI job configurations (if it exists) - Increase the "default" shard count by 1 for each job, unless it was previously set to 1, to accommodate the new functorch tests and not regress time-to-signal. - Adds a bunch of skips for ROCM and torchdynamo configurations. We can investigate them later. NB: I might go through some more iterations to figure out what other skips need to be added, but this iteration of the PR seems to pass most CI. suite. Test Plan: - wait for CI Pull Request resolved: pytorch/pytorch#96464 Approved by: https://github.com/huydhn
cyyever
pushed a commit
to cyyever/pytorch_private
that referenced
this pull request
Mar 27, 2023
Before pytorch/pytorch#96464, ROCm tests in trunk are already quite flaky https://hud.pytorch.org/reliability/pytorch/pytorch?jobName=trunk%20%2F%20linux-focal-rocm5.4.2-py3.8%20%2F%20test%20(default). After pytorch/pytorch#96464, there is a new group of flaky failures coming from functorch. So let's mark the test as flaky to monitor without impacting trunk. Two flaky tests currently seeing in trunk are: * pytorch/pytorch#97256 * `functorch/test_memory_efficient_fusion.py` OOM Pull Request resolved: pytorch/pytorch#97259 Approved by: https://github.com/malfet, https://github.com/zou3519
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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.
Stack from ghstack:
Fixes #96347
This PR:
previously set to 1, to accommodate the new functorch tests and not
regress time-to-signal.
investigate them later.
NB: I might go through some more iterations to figure out what other
skips need to be added, but this iteration of the PR seems to pass most CI.
suite.
Test Plan: