codecache: Remove cpp_prefix.h duplication per build, then precompile it #144293
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/144293
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 1fb5e56 with merge base fa85434 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
Prior to this PR, `_inductor/codegen/cpp_prefix.h` was copied into a new temporary directory on every inductor run utilizing the CPP backend, then included in the output source code. Instead, this puts it in an appropriate place in the torch includes, and includes it from there. This allows us to precompile it in cpp_wrapper and AOT inductor mode, saving significant compilation time. Note for reviewers: I believe I've traced down all the implications of this change for FBCode here on the OSS side, but someone will need to run this through tests on the FB side to be sure. ghstack-source-id: 78acea9 Pull Request resolved: #144293
Prior to this PR, `_inductor/codegen/cpp_prefix.h` was copied into a new temporary directory on every inductor run utilizing the CPP backend, then included in the output source code. Instead, this puts it in an appropriate place in the torch includes, and includes it from there. This allows us to precompile it in cpp_wrapper and AOT inductor mode, saving significant compilation time. Note for reviewers: I believe I've traced down all the implications of this change for FBCode here on the OSS side, but someone will need to run this through tests on the FB side to be sure. ghstack-source-id: 91fa52d Pull Request resolved: #144293
Prior to this PR, `_inductor/codegen/cpp_prefix.h` was copied into a new temporary directory on every inductor run utilizing the CPP backend, then included in the output source code. Instead, this puts it in an appropriate place in the torch includes, and includes it from there. This allows us to precompile it in cpp_wrapper and AOT inductor mode, saving significant compilation time. Note for reviewers: I believe I've traced down all the implications of this change for FBCode here on the OSS side, but someone will need to run this through tests on the FB side to be sure. ghstack-source-id: 754a728 Pull Request resolved: #144293
|
Additional note to reviewers: |
Per conversation in #143909, I'll be reverting this to |
Prior to this PR, `_inductor/codegen/cpp_prefix.h` was copied into a new temporary directory on every inductor run utilizing the CPP backend, then included in the output source code. Instead, this puts it in an appropriate place in the torch includes, and includes it from there. This allows us to precompile it in cpp_wrapper and AOT inductor mode, saving significant compilation time. Note for reviewers: I believe I've traced down all the implications of this change for FBCode here on the OSS side, but someone will need to run this through tests on the FB side to be sure. ghstack-source-id: 4e417ce Pull Request resolved: pytorch/pytorch#144293
|
@desertfire has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
|
@desertfire has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
|
@desertfire has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
1 similar comment
|
@desertfire has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
|
@pytorchbot merge |
Merge failedReason: This PR has internal changes and must be landed via Phabricator! Please try reimporting/rexporting the PR! Details for Dev Infra teamRaised by workflow job |
|
@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 |
Stack from ghstack (oldest at bottom):
Prior to this PR,
_inductor/codegen/cpp_prefix.hwas copied into a new temporary directory on every inductor run utilizing the CPP backend (i.e. CPU-only), then included in the output source code. Instead, this PR puts it in an appropriate place in the torch includes, and includes it from there. This allows us to precompile it in cpp_wrapper and AOT inductor mode, saving significant compilation time.Due to difficulties getting this to work in FBCode, the precompilation itself is only enabled in OSS PyTorch.
cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @ipiszy @yf225 @chenyang78 @kadeng @muchulee8 @ColinPeppler @amjames @desertfire @chauhang @aakhundov
Differential Revision: D69420620