[inductor] Lower diagonal, diagonal_copy and diagonal_scatter #103755
Conversation
This file contains 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
This decomposes `diagonal_copy` into `as_strided_copy` and `diagonal_scatter` into `as_strided_scatter`. Currently these are decomposed into mutations, which can't be used by inductor because the `_scatter` and `_copy` variants are output by functionalization to remove mutations from the graph. So by re-introducing mutations in their decomposition, we break the assumptions of the graph. [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/103755
Note: Links to docs will display an error until the docs builds have been completed. ❌ 2 New FailuresAs of commit bbc046e: NEW FAILURES - The following jobs have failed:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
peterbell10
added a commit
that referenced
this pull request
Jun 16, 2023
This decomposes `diagonal_copy` into `as_strided_copy` and `diagonal_scatter` into `as_strided_scatter`. Currently these are decomposed into mutations, which can't be used by inductor because the `_scatter` and `_copy` variants are output by functionalization to remove mutations from the graph. So by re-introducing mutations in their decomposition, we break the assumptions of the graph. ghstack-source-id: 5b6b8ef2fc3dd6439fc90af744324610bf2b4fa1 Pull Request resolved: #103755
peterbell10
added a commit
to peterbell10/pytorch
that referenced
this pull request
Jun 16, 2023
This decomposes `diagonal_copy` into `as_strided_copy` and `diagonal_scatter` into `as_strided_scatter`. Currently these are decomposed into mutations, which can't be used by inductor because the `_scatter` and `_copy` variants are output by functionalization to remove mutations from the graph. So by re-introducing mutations in their decomposition, we break the assumptions of the graph. ghstack-source-id: 5b6b8ef2fc3dd6439fc90af744324610bf2b4fa1 Pull Request resolved: pytorch#103755
…ter" Currently these are decomposed into `as_strided`, which forces a buffer to be realized. Instead, this lowers them into a native inductor view node and so doesn't require any buffers to be realized. [ghstack-poisoned]
peterbell10
changed the title
peterbell10
added a commit
that referenced
this pull request
Jun 16, 2023
Currently these are decomposed into `as_strided`, which forces a buffer to be realized. Instead, this lowers them into a native inductor view node and so doesn't require any buffers to be realized. ghstack-source-id: 55735b7107b2063b21d07bf66a3bead775043582 Pull Request resolved: #103755
…ter" Currently these are decomposed into `as_strided`, which forces a buffer to be realized. Instead, this lowers them into a native inductor view node and so doesn't require any buffers to be realized. [ghstack-poisoned]
peterbell10
added a commit
that referenced
this pull request
Jun 16, 2023
Currently these are decomposed into `as_strided`, which forces a buffer to be realized. Instead, this lowers them into a native inductor view node and so doesn't require any buffers to be realized. ghstack-source-id: ae0f5fad6fc0434f9e6688ca2adcd6de001ae7a6 Pull Request resolved: #103755
…ter" Currently these are decomposed into `as_strided`, which forces a buffer to be realized. Instead, this lowers them into a native inductor view node and so doesn't require any buffers to be realized. [ghstack-poisoned]
peterbell10
added a commit
that referenced
this pull request
Jun 16, 2023
Currently these are decomposed into `as_strided`, which forces a buffer to be realized. Instead, this lowers them into a native inductor view node and so doesn't require any buffers to be realized. ghstack-source-id: 3781141ab7e2cbe885bab16401e243a1afd20f78 Pull Request resolved: #103755
…ter" Currently these are decomposed into `as_strided`, which forces a buffer to be realized. Instead, this lowers them into a native inductor view node and so doesn't require any buffers to be realized. [ghstack-poisoned]
peterbell10
added a commit
to peterbell10/pytorch
that referenced
this pull request
Jun 19, 2023
Currently these are decomposed into `as_strided`, which forces a buffer to be realized. Instead, this lowers them into a native inductor view node and so doesn't require any buffers to be realized. ghstack-source-id: e3c4d8e7dc7a2aee2c757b075e9654f71fcaf733 Pull Request resolved: pytorch#103755
…ter" Currently these are decomposed into `as_strided`, which forces a buffer to be realized. Instead, this lowers them into a native inductor view node and so doesn't require any buffers to be realized. [ghstack-poisoned]
peterbell10
added a commit
to peterbell10/pytorch
that referenced
this pull request
Jun 19, 2023
Currently these are decomposed into `as_strided`, which forces a buffer to be realized. Instead, this lowers them into a native inductor view node and so doesn't require any buffers to be realized. ghstack-source-id: b0b2191e196980f1f386047d154ad050fa1f49a0 Pull Request resolved: pytorch#103755
peterbell10
added a commit
to peterbell10/pytorch
that referenced
this pull request
Jun 20, 2023
Currently these are decomposed into `as_strided`, which forces a buffer to be realized. Instead, this lowers them into a native inductor view node and so doesn't require any buffers to be realized. ghstack-source-id: b0b2191e196980f1f386047d154ad050fa1f49a0 Pull Request resolved: pytorch#103755
…ter" Currently these are decomposed into `as_strided`, which forces a buffer to be realized. Instead, this lowers them into a native inductor view node and so doesn't require any buffers to be realized. [ghstack-poisoned]
jansel
approved these changes
Jun 21, 2023
peterbell10
added a commit
to peterbell10/pytorch
that referenced
this pull request
Jun 21, 2023
Currently these are decomposed into `as_strided`, which forces a buffer to be realized. Instead, this lowers them into a native inductor view node and so doesn't require any buffers to be realized. ghstack-source-id: 6f00f2f71fe4eded382a8f2f7915bb19f904cbc6 Pull Request resolved: pytorch#103755
…ter" Currently these are decomposed into `as_strided`, which forces a buffer to be realized. Instead, this lowers them into a native inductor view node and so doesn't require any buffers to be realized. [ghstack-poisoned]
|
@pytorchbot merge |
Merge failedReason: This PR needs a If not, please add the To add a label, you can comment to pytorchbot, for example For more information, see Details for Dev Infra teamRaised by workflow job |
|
@pytorchbot merge |
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 |
…onal_scatter" Currently these are decomposed into `as_strided`, which forces a buffer to be realized. Instead, this lowers them into a native inductor view node and so doesn't require any buffers to be realized. cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng Xia-Weiwen wenzhe-nrv jiayisunx ipiszy ngimel yf225 chenyang78 [ghstack-poisoned]
Merge failedReason: New commits were pushed while merging. Please rerun the merge command. Details for Dev Infra teamRaised by workflow job |
|
@pytorchbot merge |
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 |
Merge failedReason: 2 jobs have failed, first few of them are: linux-binary-manywheel / manywheel-py3_8-cuda12_1-with-pypi-cudnn-test, linux-binary-manywheel / manywheel-py3_8-cuda11_8-test Details for Dev Infra teamRaised by workflow job |
|
@pytorchbot merge |
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 |
Merge failedReason: 2 jobs have failed, first few of them are: linux-binary-manywheel / manywheel-py3_8-cuda12_1-with-pypi-cudnn-test, linux-binary-manywheel / manywheel-py3_8-cuda11_8-test Details for Dev Infra teamRaised by workflow job |
|
@pytorchbot merge -f "CI is green" |
Merge startedYour change will be merged immediately since you used the force (-f) flag, bypassing any CI checks (ETA: 1-5 minutes). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
vkallesh
pushed a commit
to amd/ZenDNN-pytorch
that referenced
this pull request
Jul 5, 2023
Currently these are decomposed into `as_strided`, which forces a buffer to be realized. Instead, this lowers them into a native inductor view node and so doesn't require any buffers to be realized. ghstack-source-id: da503c44398dd41be29c6e25d7be98b3bd930fc5 Pull Request resolved: pytorch/pytorch#103755
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Stack from ghstack (oldest at bottom):
Currently these are decomposed into
as_strided, which forces a buffer to berealized. Instead, this lowers them into a native inductor view node and so
doesn't require any buffers to be realized.
cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @Xia-Weiwen @wenzhe-nrv @jiayisunx @ipiszy @ngimel @yf225 @chenyang78