[MPS] Add support for two more isin variants by malfet · Pull Request #154010 · pytorch/pytorch

malfet · 2025-05-21T03:48:31Z

Stack from ghstack (oldest at bottom):

-> [MPS] Add support for two more isin variants #154010

isin_Tensor_Scalar_out is just a redispatch to eq/neq
isin_Scalar_Tensor_out redispatches back to generic isin op, but needs a small tweak to handle float scalars
Make sure that out is resized to an expected value in isin_Tensor_Tensor_out_mps

Add unittests to validate that, but skip them on MacOS-13, where MPS op just returns garbage

Before this change both of those failed

>>> import torch
>>> t = torch.tensor([0, 1, 2], device='mps')
>>> torch.isin(t, 1)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
NotImplementedError: The operator 'aten::isin.Tensor_Scalar_out' is not currently implemented for the MPS device. If you want this op to be considered for addition please comment on https://github.com/pytorch/pytorch/issues/141287 and mention use-case, that resulted in missing op as well as commit hash 3b875c25ea6d8802a0c53af9eb961ddf2f058188. As a temporary fix, you can set the environment variable `PYTORCH_ENABLE_MPS_FALLBACK=1` to use the CPU as a fallback for this op. WARNING: this will be slower than running natively on MPS.
>>> torch.isin(1, t)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
NotImplementedError: The operator 'aten::isin.Scalar_Tensor_out' is not currently implemented for the MPS device. If you want this op to be considered for addition please comment on https://github.com/pytorch/pytorch/issues/141287 and mention use-case, that resulted in missing op as well as commit hash 3b875c25ea6d8802a0c53af9eb961ddf2f058188. As a temporary fix, you can set the environment variable `PYTORCH_ENABLE_MPS_FALLBACK=1` to use the CPU as a fallback for this op. WARNING: this will be slower than running natively on MPS.

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

[ghstack-poisoned]

pytorch-bot · 2025-05-21T03:48:35Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/154010

📄 Preview Python docs built from this PR
📄 Preview C++ docs built from this PR
❓ Need help or want to give feedback on the CI? Visit the bot commands wiki or our office hours

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

⏳ No Failures, 72 Pending

As of commit 64c8b36 with merge base 2e56ce0 ():
💚 Looks good so far! There are no failures yet. 💚

UNSTABLE - The following job is marked as unstable, possibly due to flakiness on trunk:

⏳ pull / unstable-linux-focal-cuda12.6-py3.10-gcc11-sm89-xfail / build (gh)

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

`isin_Tensor_Scalar_out` is just a redispatch to eq/neq `isin_Scalar_Tensor_out` redispatches back to generic `isin` op Add unittests to validate that Before this change both of those failed ```python >>> import torch >>> t = torch.tensor([0, 1, 2], device='mps') >>> torch.isin(t, 1) Traceback (most recent call last): File "<stdin>", line 1, in <module> NotImplementedError: The operator 'aten::isin.Tensor_Scalar_out' is not currently implemented for the MPS device. If you want this op to be considered for addition please comment on #141287 and mention use-case, that resulted in missing op as well as commit hash 3b875c2. As a temporary fix, you can set the environment variable `PYTORCH_ENABLE_MPS_FALLBACK=1` to use the CPU as a fallback for this op. WARNING: this will be slower than running natively on MPS. >>> torch.isin(1, t) Traceback (most recent call last): File "<stdin>", line 1, in <module> NotImplementedError: The operator 'aten::isin.Scalar_Tensor_out' is not currently implemented for the MPS device. If you want this op to be considered for addition please comment on #141287 and mention use-case, that resulted in missing op as well as commit hash 3b875c2. As a temporary fix, you can set the environment variable `PYTORCH_ENABLE_MPS_FALLBACK=1` to use the CPU as a fallback for this op. WARNING: this will be slower than running natively on MPS. ``` ghstack-source-id: 9c8eb66 Pull Request resolved: #154010

github-actions · 2025-05-21T03:52:21Z

Attention! native_functions.yaml was changed

If 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:

aten/src/ATen/native/native_functions.yaml

[ghstack-poisoned]

`isin_Tensor_Scalar_out` is just a redispatch to eq/neq `isin_Scalar_Tensor_out` redispatches back to generic `isin` op Add unittests to validate that Before this change both of those failed ```python >>> import torch >>> t = torch.tensor([0, 1, 2], device='mps') >>> torch.isin(t, 1) Traceback (most recent call last): File "<stdin>", line 1, in <module> NotImplementedError: The operator 'aten::isin.Tensor_Scalar_out' is not currently implemented for the MPS device. If you want this op to be considered for addition please comment on #141287 and mention use-case, that resulted in missing op as well as commit hash 3b875c2. As a temporary fix, you can set the environment variable `PYTORCH_ENABLE_MPS_FALLBACK=1` to use the CPU as a fallback for this op. WARNING: this will be slower than running natively on MPS. >>> torch.isin(1, t) Traceback (most recent call last): File "<stdin>", line 1, in <module> NotImplementedError: The operator 'aten::isin.Scalar_Tensor_out' is not currently implemented for the MPS device. If you want this op to be considered for addition please comment on #141287 and mention use-case, that resulted in missing op as well as commit hash 3b875c2. As a temporary fix, you can set the environment variable `PYTORCH_ENABLE_MPS_FALLBACK=1` to use the CPU as a fallback for this op. WARNING: this will be slower than running natively on MPS. ``` ghstack-source-id: 123db4e Pull Request resolved: #154010

[ghstack-poisoned]

`isin_Tensor_Scalar_out` is just a redispatch to eq/neq `isin_Scalar_Tensor_out` redispatches back to generic `isin` op Add unittests to validate that Before this change both of those failed ```python >>> import torch >>> t = torch.tensor([0, 1, 2], device='mps') >>> torch.isin(t, 1) Traceback (most recent call last): File "<stdin>", line 1, in <module> NotImplementedError: The operator 'aten::isin.Tensor_Scalar_out' is not currently implemented for the MPS device. If you want this op to be considered for addition please comment on #141287 and mention use-case, that resulted in missing op as well as commit hash 3b875c2. As a temporary fix, you can set the environment variable `PYTORCH_ENABLE_MPS_FALLBACK=1` to use the CPU as a fallback for this op. WARNING: this will be slower than running natively on MPS. >>> torch.isin(1, t) Traceback (most recent call last): File "<stdin>", line 1, in <module> NotImplementedError: The operator 'aten::isin.Scalar_Tensor_out' is not currently implemented for the MPS device. If you want this op to be considered for addition please comment on #141287 and mention use-case, that resulted in missing op as well as commit hash 3b875c2. As a temporary fix, you can set the environment variable `PYTORCH_ENABLE_MPS_FALLBACK=1` to use the CPU as a fallback for this op. WARNING: this will be slower than running natively on MPS. ``` ghstack-source-id: 99c45ae Pull Request resolved: #154010

[ghstack-poisoned]

`isin_Tensor_Scalar_out` is just a redispatch to eq/neq `isin_Scalar_Tensor_out` redispatches back to generic `isin` op Add unittests to validate that Before this change both of those failed ```python >>> import torch >>> t = torch.tensor([0, 1, 2], device='mps') >>> torch.isin(t, 1) Traceback (most recent call last): File "<stdin>", line 1, in <module> NotImplementedError: The operator 'aten::isin.Tensor_Scalar_out' is not currently implemented for the MPS device. If you want this op to be considered for addition please comment on #141287 and mention use-case, that resulted in missing op as well as commit hash 3b875c2. As a temporary fix, you can set the environment variable `PYTORCH_ENABLE_MPS_FALLBACK=1` to use the CPU as a fallback for this op. WARNING: this will be slower than running natively on MPS. >>> torch.isin(1, t) Traceback (most recent call last): File "<stdin>", line 1, in <module> NotImplementedError: The operator 'aten::isin.Scalar_Tensor_out' is not currently implemented for the MPS device. If you want this op to be considered for addition please comment on #141287 and mention use-case, that resulted in missing op as well as commit hash 3b875c2. As a temporary fix, you can set the environment variable `PYTORCH_ENABLE_MPS_FALLBACK=1` to use the CPU as a fallback for this op. WARNING: this will be slower than running natively on MPS. ``` ghstack-source-id: 96faf9a Pull Request resolved: #154010

[ghstack-poisoned]

`isin_Tensor_Scalar_out` is just a redispatch to eq/neq `isin_Scalar_Tensor_out` redispatches back to generic `isin` op Add unittests to validate that Before this change both of those failed ```python >>> import torch >>> t = torch.tensor([0, 1, 2], device='mps') >>> torch.isin(t, 1) Traceback (most recent call last): File "<stdin>", line 1, in <module> NotImplementedError: The operator 'aten::isin.Tensor_Scalar_out' is not currently implemented for the MPS device. If you want this op to be considered for addition please comment on #141287 and mention use-case, that resulted in missing op as well as commit hash 3b875c2. As a temporary fix, you can set the environment variable `PYTORCH_ENABLE_MPS_FALLBACK=1` to use the CPU as a fallback for this op. WARNING: this will be slower than running natively on MPS. >>> torch.isin(1, t) Traceback (most recent call last): File "<stdin>", line 1, in <module> NotImplementedError: The operator 'aten::isin.Scalar_Tensor_out' is not currently implemented for the MPS device. If you want this op to be considered for addition please comment on #141287 and mention use-case, that resulted in missing op as well as commit hash 3b875c2. As a temporary fix, you can set the environment variable `PYTORCH_ENABLE_MPS_FALLBACK=1` to use the CPU as a fallback for this op. WARNING: this will be slower than running natively on MPS. ``` ghstack-source-id: e60527d Pull Request resolved: #154010

malfet · 2025-05-22T17:57:19Z

@pytorchbot merge -f "Roses are red, violets are blue, I want to land this PR now"

pytorchmergebot · 2025-05-22T17:59:12Z

Merge started

Your change will be merged immediately since you used the force (-f) flag, bypassing any CI checks (ETA: 1-5 minutes). Please use -f as last resort and instead consider -i/--ignore-current to continue the merge ignoring current failures. This will allow currently pending tests to finish and report signal before the merge.

Learn more about merging in the wiki.

Questions? Feedback? Please reach out to the PyTorch DevX Team

Advanced Debugging

Check the merge workflow status
here

Update

cdc5866

[ghstack-poisoned]

malfet requested a review from kulinseth as a code owner May 21, 2025 03:48

pytorch-bot bot added the ciflow/mps Run MPS tests (subset of trunk) label May 21, 2025

malfet requested review from albanD, dcci and manuelcandales May 21, 2025 03:48

malfet added topic: improvements topic category release notes: mps Release notes category labels May 21, 2025

Skylion007 approved these changes May 21, 2025

View reviewed changes

dcci approved these changes May 21, 2025

View reviewed changes

albanD removed their request for review May 21, 2025 17:28

malfet mentioned this pull request May 21, 2025

Most requested ops for the MPS backend #154052

Open

malfet added the autoformat label May 21, 2025

manuelcandales approved these changes May 21, 2025

View reviewed changes

Update

8c5a531

[ghstack-poisoned]

pytorch-bot bot added ciflow/inductor module: inductor labels May 21, 2025

Update

fe7c081

[ghstack-poisoned]

Update

79e933b

[ghstack-poisoned]

Update

64c8b36

[ghstack-poisoned]

pytorchmergebot added the merging label May 22, 2025

pytorchmergebot added the Merged label May 22, 2025

pytorchmergebot closed this in 633ed01 May 22, 2025

pytorchmergebot removed the merging label May 22, 2025

github-actions bot deleted the gh/malfet/350/head branch June 23, 2025 02:20

hvaara mentioned this pull request Feb 23, 2026

Remove mps workaround for isin() huggingface/transformers#30430

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[MPS] Add support for two more isin variants#154010

[MPS] Add support for two more isin variants#154010
malfet wants to merge 5 commits intogh/malfet/350/basefrom
gh/malfet/350/head

malfet commented May 21, 2025 •

edited

Loading

Uh oh!

pytorch-bot bot commented May 21, 2025 •

edited

Loading

Uh oh!

github-actions bot commented May 21, 2025

Uh oh!

malfet commented May 22, 2025

Uh oh!

pytorchmergebot commented May 22, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

Conversation

malfet commented May 21, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

pytorch-bot bot commented May 21, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/154010

⏳ No Failures, 72 Pending

Uh oh!

github-actions bot commented May 21, 2025

Attention! native_functions.yaml was changed

Uh oh!

malfet commented May 22, 2025

Uh oh!

pytorchmergebot commented May 22, 2025

Merge started

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

malfet commented May 21, 2025 •

edited

Loading

pytorch-bot bot commented May 21, 2025 •

edited

Loading