[AOTI][CPU] Consider bias=None case for fbgemm_linear_fp16_weight #158535
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
hl475
requested review from
digantdesai,
jerryzh168,
jianyuh,
kimishpatel and
salilsdesai
as code owners
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/158535
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 919db33 with merge base 393377d ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
pytorch-bot
bot
added
ciflow/inductor
module: cpu
|
This pull request was exported from Phabricator. Differential Revision: D78458214 |
Attention! native_functions.yaml was changedIf you are adding a new function or defaulted argument to native_functions.yaml, you cannot use it from pre-existing Python frontend code until our FC window passes (two weeks). Split your PR into two PRs, one which adds the new C++ functionality, and one that makes use of it from Python, and land them two weeks apart. See https://github.com/pytorch/pytorch/wiki/PyTorch's-Python-Frontend-Backward-and-Forward-Compatibility-Policy#forwards-compatibility-fc for more info. Caused by: |
Attention! PyTorch one of the C-stable API file was changedYou MUST NOT change existing function declarations in this, as this header defines a stable C ABI. If you need to change the signature for a function, introduce a new v2 version of the function and modify code generation to target the new version of the function. Caused by: |
hl475
added a commit
to hl475/pytorch
that referenced
this pull request
Jul 17, 2025
Summary: Pull Request resolved: pytorch#158535 Test Plan: ``` buck2 run mode/opt deeplearning/aot_inductor/cpu:cli -- --local-model-path ~/testing/897908634_3.input.predictor --preset foa_early_stage_ranking ``` this will reduce the number of `aten::add` from 184 to 146 Rollback Plan: Differential Revision: D78458214
|
This pull request was exported from Phabricator. Differential Revision: D78458214 |
hl475
added a commit
to hl475/pytorch
that referenced
this pull request
Jul 17, 2025
Summary: Pull Request resolved: pytorch#158535 Test Plan: ``` buck2 run mode/opt deeplearning/aot_inductor/cpu:cli -- --local-model-path ~/testing/897908634_3.input.predictor --preset foa_early_stage_ranking ``` this will reduce the number of `aten::add` from 184 to 146 Rollback Plan: Differential Revision: D78458214
|
This pull request was exported from Phabricator. Differential Revision: D78458214 |
pytorch-bot bot
pushed a commit
that referenced
this pull request
Jul 21, 2025
Summary: Pull Request resolved: #158535 Test Plan: # e2e ``` buck2 run mode/opt deeplearning/aot_inductor/cpu:cli -- --local-model-path ~/testing/897908634_3.input.predictor --preset foa_early_stage_ranking ``` this will reduce the number of `aten::add` from 184 to 146 # BC & FC ## pick two models where one has bias None case and one doesn't ## publish model with diff and run predictor without diff ## publish model without diff and run predictor with diff ``` manifold get ads_storage_fblearner/tree/user/facebook/fblearner/predictor/752748048/0/lowering/.predictor.local/input_model ~/testing/752748048_0.input.predictor.local ``` ``` rm -rf /tmp/pt2_archive_* && buck2 run mode/opt deeplearning/aot_inductor/cpu:cli -- --local-model-path ~/testing/752748048_0.input.predictor.local --submodule merge --preset ads_second_stage_ranking_type_2 --lowering-backend aotinductor_ep ``` ``` buck2 run mode/opt caffe2/torch/fb/model_transform/fx2trt/packaging:load_net_predictor -- --loadMode=Benchmark --inputNetFile /tmp/pt2_archive_merge/package.zip --moduleName=merge --submodToDevice "" --using_aoti_lowering_allowlist=false --benchmarkDontRebatchSamples ``` **failed** - publish model has bias None with diff, and run predictor without diff - P1872435999 succeed - publish model has bias None without diff, and run predictor with diff - P1872444989 ``` rm -rf /tmp/pt2_archive_mix && buck2 run mode/opt deeplearning/aot_inductor/cpu:cli -- --local-model-path ~/testing/742055223_194.input.predictor.local --submodule mix --preset ads_first_stage_ranking_dsnn --lowering-backend aotinductor_ep ``` ``` buck2 run mode/opt caffe2/torch/fb/model_transform/fx2trt/packaging:load_net_predictor -- --loadMode=Benchmark --inputNetFile /tmp/pt2_archive_mix/package.zip --moduleName=mix --submodToDevice "" --using_aoti_lowering_allowlist=false --benchmarkDontRebatchSamples ``` succeed - publish model hasn't bias None with diff, and run predictor without diff - P1872468850 succeed - publish model hasn't bias None without diff, and run predictor with diff - P1872474897 Rollback Plan: Differential Revision: D78458214
|
This pull request was exported from Phabricator. Differential Revision: D78458214 |
Summary: Pull Request resolved: pytorch#158535 Test Plan: # e2e ``` buck2 run mode/opt deeplearning/aot_inductor/cpu:cli -- --local-model-path ~/testing/897908634_3.input.predictor --preset foa_early_stage_ranking ``` this will reduce the number of `aten::add` from 184 to 146 # BC & FC ## pick two models where one has bias None case and one doesn't ## publish model with diff and run predictor without diff ## publish model without diff and run predictor with diff ``` manifold get ads_storage_fblearner/tree/user/facebook/fblearner/predictor/752748048/0/lowering/.predictor.local/input_model ~/testing/752748048_0.input.predictor.local ``` ``` rm -rf /tmp/pt2_archive_* && buck2 run mode/opt deeplearning/aot_inductor/cpu:cli -- --local-model-path ~/testing/752748048_0.input.predictor.local --submodule merge --preset ads_second_stage_ranking_type_2 --lowering-backend aotinductor_ep ``` ``` buck2 run mode/opt caffe2/torch/fb/model_transform/fx2trt/packaging:load_net_predictor -- --loadMode=Benchmark --inputNetFile /tmp/pt2_archive_merge/package.zip --moduleName=merge --submodToDevice "" --using_aoti_lowering_allowlist=false --benchmarkDontRebatchSamples ``` **failed** - publish model has bias None with diff, and run predictor without diff - P1872435999 succeed - publish model has bias None without diff, and run predictor with diff - P1872444989 ``` rm -rf /tmp/pt2_archive_mix && buck2 run mode/opt deeplearning/aot_inductor/cpu:cli -- --local-model-path ~/testing/742055223_194.input.predictor.local --submodule mix --preset ads_first_stage_ranking_dsnn --lowering-backend aotinductor_ep ``` ``` buck2 run mode/opt caffe2/torch/fb/model_transform/fx2trt/packaging:load_net_predictor -- --loadMode=Benchmark --inputNetFile /tmp/pt2_archive_mix/package.zip --moduleName=mix --submodToDevice "" --using_aoti_lowering_allowlist=false --benchmarkDontRebatchSamples ``` succeed - publish model hasn't bias None with diff, and run predictor without diff - P1872468850 succeed - publish model hasn't bias None without diff, and run predictor with diff - P1872474897 Rollback Plan: Differential Revision: D78458214
|
This pull request was exported from Phabricator. Differential Revision: D78458214 |
hl475
changed the title
houseroad
approved these changes
Jul 21, 2025
henryoier
approved these changes
Jul 21, 2025
jingsh
approved these changes
Jul 21, 2025
|
@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 |
saienduri
pushed a commit
to saienduri/pytorch
that referenced
this pull request
Jul 22, 2025
…torch#158535) Test Plan: Rollback Plan: Differential Revision: D78458214 Pull Request resolved: pytorch#158535 Approved by: https://github.com/houseroad, https://github.com/henryoier, https://github.com/jingsh
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
6 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.
Test Plan:
Rollback Plan:
Differential Revision: D78458214
cc @jgong5 @mingfeima @XiaobingSuper @sanchitintel @ashokei @jingxu10 @jerryzh168 @voznesenskym @penguinwu @EikanWang @Guobing-Chen @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @ipiszy @chenyang78 @kadeng @muchulee8 @amjames @chauhang @aakhundov @coconutruben