MLX delegate: add integer support for aten.bitwise_not by AlessandroVacca · Pull Request #19053 · pytorch/executorch

AlessandroVacca · 2026-04-22T20:59:30Z

Summary

Fixes #18924
Extends aten.bitwise_not support in the MLX delegate to handle integer tensors, not just boolean tensors.
Previously the handler only dispatched to LogicalNotNode for bool and raised NotImplementedError for all other dtypes. This adds a dedicated BitwiseInvertNode backed by mlx::core::bitwise_invert, and updates the handler to dispatch based on dtype:

bool → LogicalNotNode (unchanged)
int32, int64 → BitwiseInvertNode

Changes:

serialization/schema.fbs: add BitwiseInvertNode table and append to OpNode union
runtime/MLXInterpreter.h: add exec_bitwise_invert() and dispatch case
ops.py: update _bitwise_not_handler to dispatch to BitwiseInvertNode for integers
test/test_ops.py: add bitwise_not_int test for int32 and int64

Test plan

All tests were ran on a machine with an Apple M1 Pro CPU, macOS 26.4.1.

python3 -m py_compile backends/mlx/ops.py backends/mlx/test/test_ops.py
python3 backends/mlx/serialization/generate.py
python3 -m executorch.backends.mlx.test.run_all_tests bitwise_not_int

Test output

============================================================
TEST SUMMARY
============================================================
Passed: 6
Failed: 0
============================================================

cc @metascroy

pytorch-bot · 2026-04-22T20:59:34Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/19053

📄 Preview Python docs built from this PR

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

❗ 1 Active SEVs

There are 1 currently active SEVs. If your PR is affected, please view them below:

Rolling out OSDC (ARC) runners on pull & trunk workflows in PyTorch main

❌ 14 New Failures, 15 Pending, 5 Unrelated Failures

As of commit bbe7c56 with merge base 56da964 ():

NEW FAILURES - The following jobs have failed:

Cadence Build & Test / cpu-test / test-aot / test-aot (gh)
backends/cadence/aot/tests/test_replace_ops_passes.py::TestReplaceOpsPasses::test_replace_transposed_conv_with_linear_4
MLX / test-mlx-llm (unsloth/gemma-3-1b-it, gemma3-1b, false, 4w) / test-mlx-llm-gemma3-1b-4w (gh)
RuntimeError: Command bash /Users/runner/work/_temp/exec_script failed with exit code 2
MLX / test-mlx-llm (unsloth/gemma-3-1b-it, gemma3-1b, false, nvfp4) / test-mlx-llm-gemma3-1b-nvfp4 (gh)
RuntimeError: Command bash /Users/runner/work/_temp/exec_script failed with exit code 2
MLX / test-mlx-llm (unsloth/gemma-3-1b-it, gemma3-1b, true, 4w) / test-mlx-llm-gemma3-1b-custom-4w (gh)
RuntimeError: Command bash /Users/runner/work/_temp/exec_script failed with exit code 2
MLX / test-mlx-llm (unsloth/gemma-3-1b-it, gemma3-1b, true, nvfp4) / test-mlx-llm-gemma3-1b-custom-nvfp4 (gh)
RuntimeError: Command bash /Users/runner/work/_temp/exec_script failed with exit code 2
MLX / test-mlx-llm (unsloth/Llama-3.2-1B-Instruct, llama-1b, false, 4w) / test-mlx-llm-llama-1b-4w (gh)
RuntimeError: Command bash /Users/runner/work/_temp/exec_script failed with exit code 2
MLX / test-mlx-llm (unsloth/Llama-3.2-1B-Instruct, llama-1b, false, nvfp4) / test-mlx-llm-llama-1b-nvfp4 (gh)
RuntimeError: Command bash /Users/runner/work/_temp/exec_script failed with exit code 2
MLX / test-mlx-llm (unsloth/Llama-3.2-1B-Instruct, llama-1b, true, 4w) / test-mlx-llm-llama-1b-custom-4w (gh)
RuntimeError: Command bash /Users/runner/work/_temp/exec_script failed with exit code 2
MLX / test-mlx-llm (unsloth/Llama-3.2-1B-Instruct, llama-1b, true, nvfp4) / test-mlx-llm-llama-1b-custom-nvfp4 (gh)
RuntimeError: Command bash /Users/runner/work/_temp/exec_script failed with exit code 2
MLX / test-mlx-llm (unsloth/Qwen3-0.6B, qwen3-0.6b, false, 4w) / test-mlx-llm-qwen3-0.6b-4w (gh)
RuntimeError: Command bash /Users/runner/work/_temp/exec_script failed with exit code 2
MLX / test-mlx-llm (unsloth/Qwen3-0.6B, qwen3-0.6b, false, nvfp4) / test-mlx-llm-qwen3-0.6b-nvfp4 (gh)
RuntimeError: Command bash /Users/runner/work/_temp/exec_script failed with exit code 2
MLX / test-mlx-llm (unsloth/Qwen3-0.6B, qwen3-0.6b, true, 4w) / test-mlx-llm-qwen3-0.6b-custom-4w (gh)
RuntimeError: Command bash /Users/runner/work/_temp/exec_script failed with exit code 2
MLX / test-mlx-llm (unsloth/Qwen3-0.6B, qwen3-0.6b, true, nvfp4) / test-mlx-llm-qwen3-0.6b-custom-nvfp4 (gh)
RuntimeError: Command bash /Users/runner/work/_temp/exec_script failed with exit code 2
MLX / test-mlx-whisper / test-mlx-whisper (gh)
RuntimeError: Command bash /Users/runner/work/_temp/exec_script failed with exit code 2

FLAKY - The following jobs failed but were likely due to flakiness present on trunk:

MLX / test-mlx-stories110m / test-mlx-stories110m (gh) (detected as infra flaky with no log or failing log classifier)
pull / test-llama-runner-qnn-linux (fp32, qnn_8a8w, qnn) / linux-job (gh) (detected as infra flaky with no log or failing log classifier)
pull / test-qnn-passes-linux / linux-job (gh) (detected as infra flaky with no log or failing log classifier)
pull / unittest-editable / windows / windows-job (gh) (detected as infra flaky with no log or failing log classifier)
pull / unittest-wasm-bindings (--enable-etdump) / linux-job (gh) (detected as infra flaky with no log or failing log classifier)

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

AlessandroVacca · 2026-04-22T21:03:27Z

@pytorchbot label "release notes: apple"

Copilot

Pull request overview

Extends the MLX delegate’s lowering for aten.bitwise_not to support integer tensors by introducing a new BitwiseInvertNode backed by MLX bitwise_invert, while keeping the existing boolean lowering via LogicalNotNode.

Changes:

Add BitwiseInvertNode to the FlatBuffers schema and OpNode union.
Add MLX runtime execution support (exec_bitwise_invert) and interpreter dispatch for the new opcode.
Update the Python op handler to emit BitwiseInvertNode for non-bool inputs and add a basic integer test case.

Reviewed changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated 2 comments.

File	Description
backends/mlx/serialization/schema.fbs	Adds `BitwiseInvertNode` table and appends it to the `OpNode` union for serialization.
backends/mlx/runtime/MLXInterpreter.h	Implements `exec_bitwise_invert()` and adds interpreter dispatch for `OpCode::BITWISE_INVERT`.
backends/mlx/ops.py	Updates `aten.bitwise_not` handler to emit `BitwiseInvertNode` for non-bool inputs.
backends/mlx/test/test_ops.py	Adds a unary op test for integer `torch.bitwise_not`.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

AlessandroVacca · 2026-04-22T21:05:38Z

    {"op_name": "abs",        "op_fn": torch.abs},
    {"op_name": "neg",        "op_fn": torch.neg},
    {"op_name": "logical_not","op_fn": torch.logical_not, "shapes": [(2, 3, 4), (10,), (4, 8)], "dtypes": [torch.bool], "input_fn": _bool_input_fn()},
+    {"op_name": "bitwise_not_int", "op_fn": torch.bitwise_not, "shapes": _SHAPES_3, "dtypes": [torch.int32, torch.int64], "input_fn": _int_input_fn()},


I consciously decided to test only what was asked to be added, lmk if you need also the bool test case

Test whatever dtypes you assert as supported in the handler.

Also, can you paste the outcome of successful test in the PR description?

is the test summary in the description of the PR any good?

nil-is-all · 2026-04-23T16:08:52Z

@claude review this PR

claude · 2026-04-23T16:09:48Z

Claude encountered an error —— View job

Command failed: git fetch origin --depth=20 pull/19053/head:mlx-bitwise-not-int-c777d55e

I'll analyze this and get back to you.

metascroy · 2026-04-23T16:19:33Z

Overall looks great @AlessandroVacca! Just a few feedbacks, and then we can merge :)

Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>

Removed support for int8, int16, and uint8 data types.

AlessandroVacca · 2026-04-23T16:30:59Z

Should be good now! @metascroy

Copilot

Pull request overview

Extends the MLX delegate’s lowering for aten.bitwise_not beyond bool tensors by introducing a dedicated BitwiseInvertNode (backed by MLX bitwise_invert) and dispatching to it for supported integer dtypes.

Changes:

Added BitwiseInvertNode to the MLX FlatBuffers schema and appended it to the OpNode union.
Added MLX runtime execution support (exec_bitwise_invert) and interpreter dispatch for the new opcode.
Updated the Python lowering handler to dispatch: bool -> LogicalNotNode, supported integers -> BitwiseInvertNode, otherwise fallback.
Added a unary op test entry for integer torch.bitwise_not (currently int32/int64).

Reviewed changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated 1 comment.

File	Description
backends/mlx/serialization/schema.fbs	Defines `BitwiseInvertNode` and appends it to the `OpNode` union for serialization/runtime compatibility.
backends/mlx/runtime/MLXInterpreter.h	Implements and dispatches execution for the new bitwise invert op in the MLX interpreter.
backends/mlx/ops.py	Updates `aten.bitwise_not` lowering to route bool to logical-not and integers to bitwise-invert.
backends/mlx/test/test_ops.py	Adds a new unary op test case for integer `bitwise_not`.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

AlessandroVacca · 2026-04-23T16:38:31Z

    {"op_name": "abs",        "op_fn": torch.abs},
    {"op_name": "neg",        "op_fn": torch.neg},
    {"op_name": "logical_not","op_fn": torch.logical_not, "shapes": [(2, 3, 4), (10,), (4, 8)], "dtypes": [torch.bool], "input_fn": _bool_input_fn()},
+    {"op_name": "bitwise_not_int", "op_fn": torch.bitwise_not, "shapes": _SHAPES_3, "dtypes": [torch.int32, torch.int64], "input_fn": _int_input_fn()},


Not advertised anymore whatsoever

Copilot

Copilot was unable to review this pull request because the user who requested the review is ineligible. To be eligible to request a review, you need a paid Copilot license, or your organization must enable Copilot code review.

MLX delegate: add integer support for aten.bitwise_not

b7d9cae

Copilot AI review requested due to automatic review settings April 22, 2026 20:59

meta-cla Bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Apr 22, 2026

Copilot started reviewing on behalf of AlessandroVacca April 22, 2026 21:00 View session

pytorch-bot Bot added the release notes: apple Changes to the Apple backend delegate label Apr 22, 2026

Copilot AI reviewed Apr 22, 2026

View reviewed changes

mergennachin requested a review from metascroy April 23, 2026 14:15

nil-is-all assigned metascroy Apr 23, 2026

nil-is-all added the module: mlx Issues related to MLX Backend: Metal-accelerated inference on Apple Silicon label Apr 23, 2026

Potential fix for pull request finding

af37e4f

Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>

Copilot AI review requested due to automatic review settings April 23, 2026 16:26

Copilot started reviewing on behalf of AlessandroVacca April 23, 2026 16:26 View session

Remove int8, int16, and uint8 support in ops.py

feb40df

Removed support for int8, int16, and uint8 data types.

Copilot AI reviewed Apr 23, 2026

View reviewed changes

Merge branch 'main' into mlx-bitwise-not-int

c5aaaff

metascroy approved these changes Apr 23, 2026

View reviewed changes

Merge branch 'main' into mlx-bitwise-not-int

bbe7c56

Copilot AI review requested due to automatic review settings April 24, 2026 22:30

Copilot AI reviewed Apr 24, 2026

View reviewed changes

metascroy merged commit 0a43e2f into pytorch:main Apr 25, 2026
170 of 192 checks passed

Conversation

AlessandroVacca commented Apr 22, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Changes:

Test plan

Test output

Uh oh!

pytorch-bot Bot commented Apr 22, 2026 • 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/pytorch/executorch/19053

❗ 1 Active SEVs

❌ 14 New Failures, 15 Pending, 5 Unrelated Failures

Uh oh!

AlessandroVacca commented Apr 22, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Uh oh!

AlessandroVacca Apr 22, 2026

Choose a reason for hiding this comment

Uh oh!

metascroy Apr 23, 2026

Choose a reason for hiding this comment

Uh oh!

AlessandroVacca Apr 23, 2026

Choose a reason for hiding this comment

Uh oh!

nil-is-all commented Apr 23, 2026

Uh oh!

claude Bot commented Apr 23, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

metascroy commented Apr 23, 2026

Uh oh!

AlessandroVacca commented Apr 23, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

AlessandroVacca Apr 23, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

AlessandroVacca commented Apr 22, 2026 •

edited

Loading

pytorch-bot Bot commented Apr 22, 2026 •

edited

Loading

claude Bot commented Apr 23, 2026 •

edited

Loading

AlessandroVacca commented Apr 23, 2026 •

edited

Loading