[aotinductor] Pass TorchIR to AOTInductor #110020
Closed
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
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/110020
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 21f1ee1 with merge base edafe2d ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
|
@angelayi has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
ydwu4
added
ciflow/trunk
angelayi
force-pushed
the
angelayi/aot_inductor_torch
branch
from
7c0b565 to
a6e8e34
Compare
angelayi
force-pushed
the
angelayi/aot_inductor_torch
branch
from
a6e8e34 to
163d727
Compare
|
@angelayi has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
angelayi
force-pushed
the
angelayi/aot_inductor_torch
branch
from
163d727 to
8585022
Compare
angelayi
requested review from
avikchaudhuri,
gmagogsfm,
tugsbayasgalan and
zhxchen17
as code owners
|
@angelayi has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
1 similar comment
|
@angelayi has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
angelayi
force-pushed
the
angelayi/aot_inductor_torch
branch
from
d06133b to
458e737
Compare
|
@angelayi has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
angelayi
force-pushed
the
angelayi/aot_inductor_torch
branch
from
458e737 to
6503151
Compare
|
@angelayi has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
angelayi
changed the title
angelayi
changed the title
angelayi
force-pushed
the
angelayi/aot_inductor_torch
branch
from
6503151 to
cadf002
Compare
|
@angelayi has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
facebook-github-bot
force-pushed
the
angelayi/aot_inductor_torch
branch
from
cadf002 to
79853a3
Compare
facebook-github-bot
pushed a commit
that referenced
this pull request
Oct 11, 2023
Summary: Updates `_export.aot_compile` to pass a torch IR graph to inductor, allowing inductor to now run the pre_grad_passes, and reuse more of inductor's code. Also updates the API to only return the `so_path`, and not returning the exported program. The pytree call spec is now serialized and placed inside of the generated model code. When calling the model, because there is no c++ pytree implementation linked yet, we can access the call specs through `get_call_spec()`, and call pytree flatten/unflattenin python. cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng Xia-Weiwen wenzhe-nrv jiayisunx peterbell10 ipiszy yf225 chenyang78 kadeng muchulee8 aakhundov ColinPeppler avikchaudhuri gmagogsfm zhxchen17 tugsbayasgalan Differential Revision: D49599792 Pulled By: angelayi
facebook-github-bot
force-pushed
the
angelayi/aot_inductor_torch
branch
from
79853a3 to
7b42aab
Compare
|
This pull request was exported from Phabricator. Differential Revision: D49599792 |
facebook-github-bot
force-pushed
the
angelayi/aot_inductor_torch
branch
from
7b42aab to
07b8754
Compare
|
This pull request was exported from Phabricator. Differential Revision: D49599792 |
facebook-github-bot
pushed a commit
that referenced
this pull request
Oct 24, 2023
Summary: Updates `_export.aot_compile` to pass a torch IR graph to inductor, allowing inductor to now run the pre_grad_passes, and reuse more of inductor's code. Also updates the API to only return the `so_path`, and not returning the exported program. The pytree call spec is now serialized and placed inside of the generated model code. When calling the model, because there is no c++ pytree implementation linked yet, we can access the call specs through `get_call_spec()`, and call pytree flatten/unflattenin python. cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng Xia-Weiwen wenzhe-nrv jiayisunx peterbell10 ipiszy yf225 chenyang78 kadeng muchulee8 aakhundov ColinPeppler avikchaudhuri gmagogsfm zhxchen17 tugsbayasgalan Test Plan: Internal testing results: https://www.internalfb.com/diff/D49599792?dst_version_fbid=873731594363746&transaction_fbid=659025106213571 Reviewed By: frank-wei, desertfire Differential Revision: D49599792 Pulled By: angelayi
facebook-github-bot
force-pushed
the
angelayi/aot_inductor_torch
branch
from
07b8754 to
0cba7c6
Compare
|
This pull request was exported from Phabricator. Differential Revision: D49599792 |
desertfire
approved these changes
Oct 24, 2023
facebook-github-bot
force-pushed
the
angelayi/aot_inductor_torch
branch
from
0cba7c6 to
60d0379
Compare
|
This pull request was exported from Phabricator. Differential Revision: D49599792 |
Summary: Updates `_export.aot_compile` to pass a torch IR graph to inductor, allowing inductor to now run the pre_grad_passes, and reuse more of inductor's code. Also updates the API to only return the `so_path`, and not returning the exported program. The pytree call spec is now serialized and placed inside of the generated model code. When calling the model, because there is no c++ pytree implementation linked yet, we can access the call specs through `get_call_spec()`, and call pytree flatten/unflattenin python. cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng Xia-Weiwen wenzhe-nrv jiayisunx peterbell10 ipiszy yf225 chenyang78 kadeng muchulee8 aakhundov ColinPeppler avikchaudhuri gmagogsfm zhxchen17 tugsbayasgalan Test Plan: Internal testing results: https://www.internalfb.com/diff/D49599792?dst_version_fbid=873731594363746&transaction_fbid=659025106213571 Reviewed By: frank-wei, desertfire Differential Revision: D49599792 Pulled By: angelayi
facebook-github-bot
force-pushed
the
angelayi/aot_inductor_torch
branch
from
60d0379 to
21f1ee1
Compare
|
This pull request was exported from Phabricator. Differential Revision: D49599792 |
facebook-github-bot
pushed a commit
to pytorch/benchmark
that referenced
this pull request
Oct 26, 2023
Summary: Updates `_export.aot_compile` to pass a torch IR graph to inductor, allowing inductor to now run the pre_grad_passes, and reuse more of inductor's code. Also updates the API to only return the `so_path`, and not returning the exported program. The pytree call spec is now serialized and placed inside of the generated model code. When calling the model, because there is no c++ pytree implementation linked yet, we can access the call specs through `get_call_spec()`, and call pytree flatten/unflattenin python. cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng Xia-Weiwen wenzhe-nrv jiayisunx peterbell10 ipiszy yf225 chenyang78 kadeng muchulee8 aakhundov ColinPeppler avikchaudhuri gmagogsfm zhxchen17 tugsbayasgalan X-link: pytorch/pytorch#110020 Reviewed By: frank-wei, desertfire Differential Revision: D49599792 Pulled By: angelayi fbshipit-source-id: 3c20e4bb60eb8ca02d91772b2e40e9bb72f8f905
|
@pytorchbot merge (Initiating merge automatically since Phabricator Diff has merged) |
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 |
desertfire
added a commit
to desertfire/pytorch-jit-paritybench
that referenced
this pull request
Oct 31, 2023
Summary: The API was changed in pytorch/pytorch#110020
xuhancn
pushed a commit
to xuhancn/pytorch
that referenced
this pull request
Nov 7, 2023
Updates `_export.aot_compile` to pass a torch IR graph to inductor, allowing inductor to now run the pre_grad_passes, and reuse more of inductor's code. Also updates the API to only return the `so_path`, and not returning the exported program. The pytree call spec is now serialized and placed inside of the generated model code. When calling the model, because there is no c++ pytree implementation linked yet, we can access the call specs through `get_call_spec()`, and call pytree flatten/unflattenin python. Pull Request resolved: pytorch#110020 Approved by: https://github.com/desertfire
Skylion007
pushed a commit
to Skylion007/pytorch
that referenced
this pull request
Nov 14, 2023
Updates `_export.aot_compile` to pass a torch IR graph to inductor, allowing inductor to now run the pre_grad_passes, and reuse more of inductor's code. Also updates the API to only return the `so_path`, and not returning the exported program. The pytree call spec is now serialized and placed inside of the generated model code. When calling the model, because there is no c++ pytree implementation linked yet, we can access the call specs through `get_call_spec()`, and call pytree flatten/unflattenin python. Pull Request resolved: pytorch#110020 Approved by: https://github.com/desertfire
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 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.
Updates
_export.aot_compileto pass a torch IR graph to inductor, allowing inductor to now run the pre_grad_passes, and reuse more of inductor's code.Also updates the API to only return the
so_path, and not returning the exported program. The pytree call spec is now serialized and placed inside of the generated model code. When calling the model, because there is no c++ pytree implementation linked yet, we can access the call specs throughget_call_spec(), and call pytree flatten/unflattenin python.cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @Xia-Weiwen @wenzhe-nrv @jiayisunx @peterbell10 @ipiszy @yf225 @chenyang78 @kadeng @muchulee8 @aakhundov @ColinPeppler @avikchaudhuri @gmagogsfm @zhxchen17 @tugsbayasgalan