Skip to content
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.

Sign up for GitHub

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

Jump to bottom

[PyTorch] AOTI: Emit CACHED_TORCH_TYPE only as needed #113997

Closed
wants to merge 3 commits into from
Closed

[PyTorch] AOTI: Emit CACHED_TORCH_TYPE only as needed #113997

wants to merge 3 commits into from

Conversation

swolchok
Copy link
Contributor

@swolchok swolchok commented Nov 17, 2023 •
edited

Stack from ghstack (oldest at bottom):

Avoids potential compatibility issues where a new dtype is supported by the DSO but not the binary loading it.

Differential Revision: D51434335

cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @peterbell10 @ipiszy @yf225 @chenyang78 @kadeng @muchulee8 @aakhundov @ColinPeppler

@swolchok
[PyTorch] AOTI: Emit CACHED_TORCH_TYPE only as needed
8bbecbd 
Avoids potential compatibility issues where a new dtype is supported by the DSO but not the binary loading it.

Differential Revision: [D51434335](https://our.internmc.facebook.com/intern/diff/D51434335/)

[ghstack-poisoned]
@pytorch-bot PyTorch Bot
Copy link

pytorch-bot bot commented Nov 17, 2023 •
edited

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/113997

Note: Links to docs will display an error until the docs builds have been completed.

✅ You can merge normally! (5 Unrelated Failures)

As of commit 9d2abed with merge base b4faa6b (image):

FLAKY - The following jobs failed but were likely due to flakiness present on trunk:

This comment was automatically generated by Dr. CI and updates every 15 minutes.

swolchok added a commit that referenced this pull request Nov 17, 2023
@swolchok
[PyTorch] AOTI: Emit CACHED_TORCH_TYPE only as needed
c95c0cd 
Avoids potential compatibility issues where a new dtype is supported by the DSO but not the binary loading it.

Differential Revision: [D51434335](https://our.internmc.facebook.com/intern/diff/D51434335/)

ghstack-source-id: 207904395
Pull Request resolved: #113997
This was referenced Nov 15, 2023
Closed
Closed
Closed
Closed
Closed
Closed
Closed
int3
int3 approved these changes Nov 18, 2023
View reviewed changes
Copy link
Contributor

@int3 int3 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks!

@swolchok
Update on "[PyTorch] AOTI: Emit CACHED_TORCH_TYPE only as needed"
479603a 
Avoids potential compatibility issues where a new dtype is supported by the DSO but not the binary loading it.

Differential Revision: [D51434335](https://our.internmc.facebook.com/intern/diff/D51434335/)

cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx peterbell10 ipiszy yf225 chenyang78 kadeng muchulee8 aakhundov ColinPeppler

[ghstack-poisoned]
This was referenced Nov 28, 2023
Closed
Closed
Closed
Closed
@swolchok
Update on "[PyTorch] AOTI: Emit CACHED_TORCH_TYPE only as needed"
9d2abed 
Avoids potential compatibility issues where a new dtype is supported by the DSO but not the binary loading it.

Differential Revision: [D51434335](https://our.internmc.facebook.com/intern/diff/D51434335/)

cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx peterbell10 ipiszy yf225 chenyang78 kadeng muchulee8 aakhundov ColinPeppler

[ghstack-poisoned]
pytorchmergebot pushed a commit that referenced this pull request Nov 29, 2023
@swolchok @pytorchmergebot
[PyTorch] Remove hardcoded device=cuda in test_aot_inductor (#112797)
ce00c8f 
All the other tests use self.device, so this seems like an oversight? Cost me a lot of time debugging the minimal arrayref interface, which is only intended for CPU.

Differential Revision: [D50949928](https://our.internmc.facebook.com/intern/diff/D50949928/)

Pull Request resolved: #112797
Approved by: https://github.com/chenyang78, https://github.com/desertfire, https://github.com/khabinov
ghstack dependencies: #113997
vfdev-5 pushed a commit to vfdev-5/pytorch that referenced this pull request Nov 29, 2023
@swolchok @vfdev-5
[PyTorch] AOTI: Emit CACHED_TORCH_TYPE only as needed (pytorch#113997)
00f8194 
Avoids potential compatibility issues where a new dtype is supported by the DSO but not the binary loading it.

Differential Revision: [D51434335](https://our.internmc.facebook.com/intern/diff/D51434335/)

Pull Request resolved: pytorch#113997
Approved by: https://github.com/int3
vfdev-5 pushed a commit to vfdev-5/pytorch that referenced this pull request Nov 29, 2023
@swolchok @vfdev-5
[PyTorch] Remove hardcoded device=cuda in test_aot_inductor (pytorch#…
e30aed8 
…112797)

All the other tests use self.device, so this seems like an oversight? Cost me a lot of time debugging the minimal arrayref interface, which is only intended for CPU.

Differential Revision: [D50949928](https://our.internmc.facebook.com/intern/diff/D50949928/)

Pull Request resolved: pytorch#112797
Approved by: https://github.com/chenyang78, https://github.com/desertfire, https://github.com/khabinov
ghstack dependencies: pytorch#113997
@int3 int3 mentioned this pull request Dec 1, 2023
Closed
int3 added a commit that referenced this pull request Dec 2, 2023
@int3
Update on "Reland #113487 and #112527 (sdpa shim & fp8 AOTInductor su…
86a3256 
…pport)"


This is a backout of #113747 which reverted the above two commits. Now that
#113997 has landed, this diff can be landed safely without breaking ABI compatibility.

cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx peterbell10 ipiszy yf225 chenyang78 kadeng muchulee8 aakhundov ColinPeppler

[ghstack-poisoned]
pytorchmergebot pushed a commit that referenced this pull request Dec 2, 2023
@int3 @pytorchmergebot
Reland #113487 and #112527 (sdpa shim & fp8 AOTInductor support) (#11…
f1fd025 
…4974)

This is a backout of #113747 which reverted the above two commits. Now that
#113997 has landed, this diff can be landed safely without breaking ABI compatibility.

Pull Request resolved: #114974
Approved by: https://github.com/chenyang78
@facebook-github-bot facebook-github-bot deleted the gh/swolchok/607/head branch December 2, 2023 15:28
hyperfraise pushed a commit to hyperfraise/pytorch that referenced this pull request Dec 21, 2023
@int3 @hyperfraise
Reland pytorch#113487 and pytorch#112527 (sdpa shim & fp8 AOTInductor…
5ebbaf0 
… support) (pytorch#114974)

This is a backout of pytorch#113747 which reverted the above two commits. Now that
pytorch#113997 has landed, this diff can be landed safely without breaking ABI compatibility.

Pull Request resolved: pytorch#114974
Approved by: https://github.com/chenyang78
hyperfraise pushed a commit to hyperfraise/pytorch that referenced this pull request Dec 21, 2023
@int3 @hyperfraise
Reland pytorch#113487 and pytorch#112527 (sdpa shim & fp8 AOTInductor…
4f1701a 
… support) (pytorch#114974)

This is a backout of pytorch#113747 which reverted the above two commits. Now that
pytorch#113997 has landed, this diff can be landed safely without breaking ABI compatibility.

Pull Request resolved: pytorch#114974
Approved by: https://github.com/chenyang78
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@int3 int3 int3 approved these changes

Assignees
No one assigned
Labels
ciflow/inductor Merged module: inductor
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

None yet
3 participants
@swolchok @int3 @pytorchmergebot