aot autograd: dont allow symint outputs to get tangents in the bw graph#96219
Closed
bdhirsh wants to merge 3 commits intogh/bdhirsh/391/basefrom
Closed
aot autograd: dont allow symint outputs to get tangents in the bw graph#96219bdhirsh wants to merge 3 commits intogh/bdhirsh/391/basefrom
bdhirsh wants to merge 3 commits intogh/bdhirsh/391/basefrom
Conversation
[ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/96219
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit e6858d6: This comment was automatically generated by Dr. CI and updates every 15 minutes. |
github-actions
bot
requested review from
SherlockNoMad,
albanD,
antoniojkim,
jbschlosser,
miladm,
voznesenskym and
wconstab
ezyang
approved these changes
Mar 7, 2023
Contributor
|
@pytorchbot merge |
Collaborator
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
Collaborator
Merge failedReason: 4 mandatory check(s) failed. The first few are:
Dig deeper by viewing the failures on hud |
… the bw graph" Previously, if dynamic shapes were turned on and we had a forward graph that returns a symint, then we would generate a backward graph that takes in a tangent input for that symint fwd output. This causes problems for downstream - inductor will see an input that it expects to be a symint, but it gets a `None` from autograd. Confirmed that this repro now passes: ``` benchmarks/dynamo/torchbench.py --devices cuda --inductor --dynamic-shapes --unspecialize-int --accuracy --training --only drq ``` [ghstack-poisoned]
Contributor
|
@pytorchbot merge |
Collaborator
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
Collaborator
Author
|
@pytorchbot merge |
Collaborator
|
The merge job was canceled. If you believe this is a mistake,then you can re trigger it through pytorch-bot. |
Collaborator
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
Collaborator
Author
|
@pytorchbot merge |
Collaborator
|
The merge job was canceled. If you believe this is a mistake,then you can re trigger it through pytorch-bot. |
Collaborator
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
Collaborator
Merge failedReason: 2 mandatory check(s) failed. The first few are:
Dig deeper by viewing the failures on hud |
… the bw graph" Previously, if dynamic shapes were turned on and we had a forward graph that returns a symint, then we would generate a backward graph that takes in a tangent input for that symint fwd output. This causes problems for downstream - inductor will see an input that it expects to be a symint, but it gets a `None` from autograd. Confirmed that this repro now passes: ``` benchmarks/dynamo/torchbench.py --devices cuda --inductor --dynamic-shapes --unspecialize-int --accuracy --training --only drq ``` cc anijain2305 [ghstack-poisoned]
Contributor
|
@pytorchbot merge |
Collaborator
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
ezyang
added a commit
that referenced
this pull request
Mar 8, 2023
This assert would have caught #96219 Signed-off-by: Edward Z. Yang <[email protected]> [ghstack-poisoned]
pytorchmergebot
pushed a commit
that referenced
this pull request
Mar 9, 2023
This assert would have caught #96219 Signed-off-by: Edward Z. Yang <[email protected]> Pull Request resolved: #96300 Approved by: https://github.com/bdhirsh
cyyever
pushed a commit
to cyyever/pytorch_private
that referenced
this pull request
Mar 12, 2023
…ph (#96219) Previously, if dynamic shapes were turned on and we had a forward graph that returns a symint, then we would generate a backward graph that takes in a tangent input for that symint fwd output. This causes problems for downstream - inductor will see an input that it expects to be a symint, but it gets a `None` from autograd. Confirmed that this repro now passes: ``` benchmarks/dynamo/torchbench.py --devices cuda --inductor --dynamic-shapes --unspecialize-int --accuracy --training --only drq ``` Pull Request resolved: pytorch/pytorch#96219 Approved by: https://github.com/ezyang
cyyever
pushed a commit
to cyyever/pytorch_private
that referenced
this pull request
Mar 12, 2023
…ph (#96219) Previously, if dynamic shapes were turned on and we had a forward graph that returns a symint, then we would generate a backward graph that takes in a tangent input for that symint fwd output. This causes problems for downstream - inductor will see an input that it expects to be a symint, but it gets a `None` from autograd. Confirmed that this repro now passes: ``` benchmarks/dynamo/torchbench.py --devices cuda --inductor --dynamic-shapes --unspecialize-int --accuracy --training --only drq ``` Pull Request resolved: pytorch/pytorch#96219 Approved by: https://github.com/ezyang
cyyever
pushed a commit
to cyyever/pytorch_private
that referenced
this pull request
Mar 12, 2023
This assert would have caught pytorch/pytorch#96219 Signed-off-by: Edward Z. Yang <[email protected]> Pull Request resolved: pytorch/pytorch#96300 Approved by: https://github.com/bdhirsh
ydwu4
added a commit
to ydwu4/pytorch
that referenced
this pull request
Mar 13, 2023
…ph (pytorch#96219) Previously, if dynamic shapes were turned on and we had a forward graph that returns a symint, then we would generate a backward graph that takes in a tangent input for that symint fwd output. This causes problems for downstream - inductor will see an input that it expects to be a symint, but it gets a `None` from autograd. Confirmed that this repro now passes: ``` benchmarks/dynamo/torchbench.py --devices cuda --inductor --dynamic-shapes --unspecialize-int --accuracy --training --only drq ``` Pull Request resolved: pytorch#96219 Approved by: https://github.com/ezyang
ydwu4
added a commit
to ydwu4/pytorch
that referenced
this pull request
Mar 13, 2023
This assert would have caught pytorch#96219 Signed-off-by: Edward Z. Yang <[email protected]> Pull Request resolved: pytorch#96300 Approved by: https://github.com/bdhirsh
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
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.
Previously, if dynamic shapes were turned on and we had a forward graph that returns a symint, then we would generate a backward graph that takes in a tangent input for that symint fwd output. This causes problems for downstream - inductor will see an input that it expects to be a symint, but it gets a
Nonefrom autograd.Confirmed that this repro now passes:
cc @anijain2305
Stack from ghstack (oldest at bottom):