[ExecuTorch][WebGPU] Lift FastVLM dispatch limits and harden graph construction by JCNTH · Pull Request #21602 · pytorch/executorch

JCNTH · 2026-08-05T19:55:09Z

Stack from ghstack (oldest at bottom):

Run production FastVLM shapes with exception-safe WebGPU graph construction

Production FastVLM tensors exceeded WebGPU one-dimensional dispatch limits, while several parameter buffers were not graph-owned before fallible pipeline construction. The runtime now folds flat dispatches across x/y, owns parameters immediately, rejects ambiguous BOOL/byte-integer casts, and compiles timestamp validation only in profiling builds.

Key changes:

GELU and expand_copy use overflow-safe folded dispatch and live resize updates.
Compare, expand_copy, and to_copy create graph-owned parameter buffers before pipeline construction, mirroring Vulkan ComputeGraph ownership.
to_copy rejects BOOL-to/from-INT8 or UINT8 before same-width flat-copy routing.
Production builds exclude timestamp-query validation and symbols.
Boundary, failure, and real FastVLM encoder coverage exercise the final combined behavior.

Kernel math and below-limit one-dimensional dispatch behavior are unchanged.

Co-authored-with: Claude Code.

Differential Revision: D114936147

[ghstack-poisoned]

pytorch-bot · 2026-08-05T19:55:13Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/21602

📄 Preview Python docs built from this PR

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

❌ 1 New Failure

As of commit ab3d0de with merge base ad3a71f ():

NEW FAILURE - The following job has failed:

pull / android / run-emulator (gh)
adb: failed to install android-test-debug-androidTest.apk: cmd: Failure calling service package: Broken pipe (32)

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

github-actions · 2026-08-05T19:57:02Z

This PR needs a `release notes:` label

If your change should be included in the release notes (i.e. would users of this library care about this change?), please use a label starting with release notes:. This helps us keep track and include your important work in the next release notes.

To add a label, you can comment to pytorchbot, for example
@pytorchbot label "release notes: none"

For more information, see
https://github.com/pytorch/pytorch/wiki/PyTorch-AutoLabel-Bot#why-categorize-for-release-notes-and-how-does-it-work.

[ghstack-poisoned]

…nstruction Pull Request resolved: #21602 **Run production FastVLM shapes with exception-safe WebGPU graph construction** Production FastVLM tensors exceeded WebGPU one-dimensional dispatch limits, while several parameter buffers were not graph-owned before fallible pipeline construction. The runtime now folds flat dispatches across x/y, owns parameters immediately, rejects ambiguous `BOOL`/byte-integer casts, and compiles timestamp validation only in profiling builds. Key changes: - `GELU` and `expand_copy` use overflow-safe folded dispatch and live resize updates. - `Compare`, `expand_copy`, and `to_copy` create graph-owned parameter buffers before pipeline construction, mirroring Vulkan `ComputeGraph` ownership. - `to_copy` rejects `BOOL`-to/from-`INT8` or `UINT8` before same-width flat-copy routing. - Production builds exclude timestamp-query validation and symbols. - Boundary, failure, and real FastVLM encoder coverage exercise the final combined behavior. Kernel math and below-limit one-dimensional dispatch behavior are unchanged. Co-authored-with: Claude Code. ghstack-source-id: 411059172 @exported-using-ghexport Differential Revision: [D114936147](https://our.internmc.facebook.com/intern/diff/D114936147/)

JCNTH · 2026-08-05T19:55:09Z

Stack from ghstack (oldest at bottom):

Run production FastVLM shapes with exception-safe WebGPU graph construction

Production FastVLM tensors exceeded WebGPU one-dimensional dispatch limits, while several parameter buffers were not graph-owned before fallible pipeline construction. The runtime now folds flat dispatches across x/y, owns parameters immediately, rejects ambiguous BOOL/byte-integer casts, and compiles timestamp validation only in profiling builds.

Key changes:

GELU and expand_copy use overflow-safe folded dispatch and live resize updates.
Compare, expand_copy, and to_copy create graph-owned parameter buffers before pipeline construction, mirroring Vulkan ComputeGraph ownership.
to_copy rejects BOOL-to/from-INT8 or UINT8 before same-width flat-copy routing.
Production builds exclude timestamp-query validation and symbols.
Boundary, failure, and real FastVLM encoder coverage exercise the final combined behavior.

Kernel math and below-limit one-dimensional dispatch behavior are unchanged.

Co-authored-with: Claude Code.

Differential Revision: D114936147

[ghstack-poisoned]

pytorch-bot · 2026-08-05T19:55:13Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/21602

📄 Preview Python docs built from this PR

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

❌ 1 New Failure

As of commit ab3d0de with merge base ad3a71f ():

NEW FAILURE - The following job has failed:

pull / android / run-emulator (gh)
adb: failed to install android-test-debug-androidTest.apk: cmd: Failure calling service package: Broken pipe (32)

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

github-actions · 2026-08-05T19:57:02Z

This PR needs a `release notes:` label

If your change should be included in the release notes (i.e. would users of this library care about this change?), please use a label starting with release notes:. This helps us keep track and include your important work in the next release notes.

To add a label, you can comment to pytorchbot, for example
@pytorchbot label "release notes: none"

For more information, see
https://github.com/pytorch/pytorch/wiki/PyTorch-AutoLabel-Bot#why-categorize-for-release-notes-and-how-does-it-work.

[ghstack-poisoned]

…nstruction Pull Request resolved: #21602 **Run production FastVLM shapes with exception-safe WebGPU graph construction** Production FastVLM tensors exceeded WebGPU one-dimensional dispatch limits, while several parameter buffers were not graph-owned before fallible pipeline construction. The runtime now folds flat dispatches across x/y, owns parameters immediately, rejects ambiguous `BOOL`/byte-integer casts, and compiles timestamp validation only in profiling builds. Key changes: - `GELU` and `expand_copy` use overflow-safe folded dispatch and live resize updates. - `Compare`, `expand_copy`, and `to_copy` create graph-owned parameter buffers before pipeline construction, mirroring Vulkan `ComputeGraph` ownership. - `to_copy` rejects `BOOL`-to/from-`INT8` or `UINT8` before same-width flat-copy routing. - Production builds exclude timestamp-query validation and symbols. - Boundary, failure, and real FastVLM encoder coverage exercise the final combined behavior. Kernel math and below-limit one-dimensional dispatch behavior are unchanged. Co-authored-with: Claude Code. ghstack-source-id: 411059172 @exported-using-ghexport Differential Revision: [D114936147](https://our.internmc.facebook.com/intern/diff/D114936147/)

JCNTH · 2026-08-05T19:55:09Z

Stack from ghstack (oldest at bottom):

Run production FastVLM shapes with exception-safe WebGPU graph construction

Production FastVLM tensors exceeded WebGPU one-dimensional dispatch limits, while several parameter buffers were not graph-owned before fallible pipeline construction. The runtime now folds flat dispatches across x/y, owns parameters immediately, rejects ambiguous BOOL/byte-integer casts, and compiles timestamp validation only in profiling builds.

Key changes:

GELU and expand_copy use overflow-safe folded dispatch and live resize updates.
Compare, expand_copy, and to_copy create graph-owned parameter buffers before pipeline construction, mirroring Vulkan ComputeGraph ownership.
to_copy rejects BOOL-to/from-INT8 or UINT8 before same-width flat-copy routing.
Production builds exclude timestamp-query validation and symbols.
Boundary, failure, and real FastVLM encoder coverage exercise the final combined behavior.

Kernel math and below-limit one-dimensional dispatch behavior are unchanged.

Co-authored-with: Claude Code.

Differential Revision: D114936147

[ghstack-poisoned]

pytorch-bot · 2026-08-05T19:55:13Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/21602

📄 Preview Python docs built from this PR

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

❌ 1 New Failure

As of commit ab3d0de with merge base ad3a71f ():

NEW FAILURE - The following job has failed:

pull / android / run-emulator (gh)
adb: failed to install android-test-debug-androidTest.apk: cmd: Failure calling service package: Broken pipe (32)

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

github-actions · 2026-08-05T19:57:02Z

This PR needs a `release notes:` label

If your change should be included in the release notes (i.e. would users of this library care about this change?), please use a label starting with release notes:. This helps us keep track and include your important work in the next release notes.

To add a label, you can comment to pytorchbot, for example
@pytorchbot label "release notes: none"

For more information, see
https://github.com/pytorch/pytorch/wiki/PyTorch-AutoLabel-Bot#why-categorize-for-release-notes-and-how-does-it-work.

[ghstack-poisoned]

…nstruction Pull Request resolved: #21602 **Run production FastVLM shapes with exception-safe WebGPU graph construction** Production FastVLM tensors exceeded WebGPU one-dimensional dispatch limits, while several parameter buffers were not graph-owned before fallible pipeline construction. The runtime now folds flat dispatches across x/y, owns parameters immediately, rejects ambiguous `BOOL`/byte-integer casts, and compiles timestamp validation only in profiling builds. Key changes: - `GELU` and `expand_copy` use overflow-safe folded dispatch and live resize updates. - `Compare`, `expand_copy`, and `to_copy` create graph-owned parameter buffers before pipeline construction, mirroring Vulkan `ComputeGraph` ownership. - `to_copy` rejects `BOOL`-to/from-`INT8` or `UINT8` before same-width flat-copy routing. - Production builds exclude timestamp-query validation and symbols. - Boundary, failure, and real FastVLM encoder coverage exercise the final combined behavior. Kernel math and below-limit one-dimensional dispatch behavior are unchanged. Co-authored-with: Claude Code. ghstack-source-id: 411059172 @exported-using-ghexport Differential Revision: [D114936147](https://our.internmc.facebook.com/intern/diff/D114936147/)

JCNTH · 2026-08-05T19:55:09Z

Stack from ghstack (oldest at bottom):

Run production FastVLM shapes with exception-safe WebGPU graph construction

Production FastVLM tensors exceeded WebGPU one-dimensional dispatch limits, while several parameter buffers were not graph-owned before fallible pipeline construction. The runtime now folds flat dispatches across x/y, owns parameters immediately, rejects ambiguous BOOL/byte-integer casts, and compiles timestamp validation only in profiling builds.

Key changes:

GELU and expand_copy use overflow-safe folded dispatch and live resize updates.
Compare, expand_copy, and to_copy create graph-owned parameter buffers before pipeline construction, mirroring Vulkan ComputeGraph ownership.
to_copy rejects BOOL-to/from-INT8 or UINT8 before same-width flat-copy routing.
Production builds exclude timestamp-query validation and symbols.
Boundary, failure, and real FastVLM encoder coverage exercise the final combined behavior.

Kernel math and below-limit one-dimensional dispatch behavior are unchanged.

Co-authored-with: Claude Code.

Differential Revision: D114936147

[ghstack-poisoned]

pytorch-bot · 2026-08-05T19:55:13Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/21602

📄 Preview Python docs built from this PR

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

❌ 1 New Failure

As of commit ab3d0de with merge base ad3a71f ():

NEW FAILURE - The following job has failed:

pull / android / run-emulator (gh)
adb: failed to install android-test-debug-androidTest.apk: cmd: Failure calling service package: Broken pipe (32)

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

github-actions · 2026-08-05T19:57:02Z

This PR needs a `release notes:` label

If your change should be included in the release notes (i.e. would users of this library care about this change?), please use a label starting with release notes:. This helps us keep track and include your important work in the next release notes.

To add a label, you can comment to pytorchbot, for example
@pytorchbot label "release notes: none"

For more information, see
https://github.com/pytorch/pytorch/wiki/PyTorch-AutoLabel-Bot#why-categorize-for-release-notes-and-how-does-it-work.

[ghstack-poisoned]

…nstruction Pull Request resolved: #21602 **Run production FastVLM shapes with exception-safe WebGPU graph construction** Production FastVLM tensors exceeded WebGPU one-dimensional dispatch limits, while several parameter buffers were not graph-owned before fallible pipeline construction. The runtime now folds flat dispatches across x/y, owns parameters immediately, rejects ambiguous `BOOL`/byte-integer casts, and compiles timestamp validation only in profiling builds. Key changes: - `GELU` and `expand_copy` use overflow-safe folded dispatch and live resize updates. - `Compare`, `expand_copy`, and `to_copy` create graph-owned parameter buffers before pipeline construction, mirroring Vulkan `ComputeGraph` ownership. - `to_copy` rejects `BOOL`-to/from-`INT8` or `UINT8` before same-width flat-copy routing. - Production builds exclude timestamp-query validation and symbols. - Boundary, failure, and real FastVLM encoder coverage exercise the final combined behavior. Kernel math and below-limit one-dimensional dispatch behavior are unchanged. Co-authored-with: Claude Code. ghstack-source-id: 411059172 @exported-using-ghexport Differential Revision: [D114936147](https://our.internmc.facebook.com/intern/diff/D114936147/)

JCNTH · 2026-08-05T19:55:09Z

Stack from ghstack (oldest at bottom):

Run production FastVLM shapes with exception-safe WebGPU graph construction

Production FastVLM tensors exceeded WebGPU one-dimensional dispatch limits, while several parameter buffers were not graph-owned before fallible pipeline construction. The runtime now folds flat dispatches across x/y, owns parameters immediately, rejects ambiguous BOOL/byte-integer casts, and compiles timestamp validation only in profiling builds.

Key changes:

GELU and expand_copy use overflow-safe folded dispatch and live resize updates.
Compare, expand_copy, and to_copy create graph-owned parameter buffers before pipeline construction, mirroring Vulkan ComputeGraph ownership.
to_copy rejects BOOL-to/from-INT8 or UINT8 before same-width flat-copy routing.
Production builds exclude timestamp-query validation and symbols.
Boundary, failure, and real FastVLM encoder coverage exercise the final combined behavior.

Kernel math and below-limit one-dimensional dispatch behavior are unchanged.

Co-authored-with: Claude Code.

Differential Revision: D114936147

[ghstack-poisoned]

pytorch-bot · 2026-08-05T19:55:13Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/21602

📄 Preview Python docs built from this PR

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

❌ 1 New Failure

As of commit ab3d0de with merge base ad3a71f ():

NEW FAILURE - The following job has failed:

pull / android / run-emulator (gh)
adb: failed to install android-test-debug-androidTest.apk: cmd: Failure calling service package: Broken pipe (32)

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

github-actions · 2026-08-05T19:57:02Z

This PR needs a `release notes:` label

If your change should be included in the release notes (i.e. would users of this library care about this change?), please use a label starting with release notes:. This helps us keep track and include your important work in the next release notes.

To add a label, you can comment to pytorchbot, for example
@pytorchbot label "release notes: none"

For more information, see
https://github.com/pytorch/pytorch/wiki/PyTorch-AutoLabel-Bot#why-categorize-for-release-notes-and-how-does-it-work.

[ghstack-poisoned]

…nstruction Pull Request resolved: #21602 **Run production FastVLM shapes with exception-safe WebGPU graph construction** Production FastVLM tensors exceeded WebGPU one-dimensional dispatch limits, while several parameter buffers were not graph-owned before fallible pipeline construction. The runtime now folds flat dispatches across x/y, owns parameters immediately, rejects ambiguous `BOOL`/byte-integer casts, and compiles timestamp validation only in profiling builds. Key changes: - `GELU` and `expand_copy` use overflow-safe folded dispatch and live resize updates. - `Compare`, `expand_copy`, and `to_copy` create graph-owned parameter buffers before pipeline construction, mirroring Vulkan `ComputeGraph` ownership. - `to_copy` rejects `BOOL`-to/from-`INT8` or `UINT8` before same-width flat-copy routing. - Production builds exclude timestamp-query validation and symbols. - Boundary, failure, and real FastVLM encoder coverage exercise the final combined behavior. Kernel math and below-limit one-dimensional dispatch behavior are unchanged. Co-authored-with: Claude Code. ghstack-source-id: 411059172 @exported-using-ghexport Differential Revision: [D114936147](https://our.internmc.facebook.com/intern/diff/D114936147/)

JCNTH · 2026-08-05T19:55:09Z

Stack from ghstack (oldest at bottom):

Run production FastVLM shapes with exception-safe WebGPU graph construction

Production FastVLM tensors exceeded WebGPU one-dimensional dispatch limits, while several parameter buffers were not graph-owned before fallible pipeline construction. The runtime now folds flat dispatches across x/y, owns parameters immediately, rejects ambiguous BOOL/byte-integer casts, and compiles timestamp validation only in profiling builds.

Key changes:

GELU and expand_copy use overflow-safe folded dispatch and live resize updates.
Compare, expand_copy, and to_copy create graph-owned parameter buffers before pipeline construction, mirroring Vulkan ComputeGraph ownership.
to_copy rejects BOOL-to/from-INT8 or UINT8 before same-width flat-copy routing.
Production builds exclude timestamp-query validation and symbols.
Boundary, failure, and real FastVLM encoder coverage exercise the final combined behavior.

Kernel math and below-limit one-dimensional dispatch behavior are unchanged.

Co-authored-with: Claude Code.

Differential Revision: D114936147

[ghstack-poisoned]

pytorch-bot · 2026-08-05T19:55:13Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/21602

📄 Preview Python docs built from this PR

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

❌ 1 New Failure

As of commit ab3d0de with merge base ad3a71f ():

NEW FAILURE - The following job has failed:

pull / android / run-emulator (gh)
adb: failed to install android-test-debug-androidTest.apk: cmd: Failure calling service package: Broken pipe (32)

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

github-actions · 2026-08-05T19:57:02Z

This PR needs a `release notes:` label

If your change should be included in the release notes (i.e. would users of this library care about this change?), please use a label starting with release notes:. This helps us keep track and include your important work in the next release notes.

To add a label, you can comment to pytorchbot, for example
@pytorchbot label "release notes: none"

For more information, see
https://github.com/pytorch/pytorch/wiki/PyTorch-AutoLabel-Bot#why-categorize-for-release-notes-and-how-does-it-work.

[ghstack-poisoned]

…nstruction Pull Request resolved: #21602 **Run production FastVLM shapes with exception-safe WebGPU graph construction** Production FastVLM tensors exceeded WebGPU one-dimensional dispatch limits, while several parameter buffers were not graph-owned before fallible pipeline construction. The runtime now folds flat dispatches across x/y, owns parameters immediately, rejects ambiguous `BOOL`/byte-integer casts, and compiles timestamp validation only in profiling builds. Key changes: - `GELU` and `expand_copy` use overflow-safe folded dispatch and live resize updates. - `Compare`, `expand_copy`, and `to_copy` create graph-owned parameter buffers before pipeline construction, mirroring Vulkan `ComputeGraph` ownership. - `to_copy` rejects `BOOL`-to/from-`INT8` or `UINT8` before same-width flat-copy routing. - Production builds exclude timestamp-query validation and symbols. - Boundary, failure, and real FastVLM encoder coverage exercise the final combined behavior. Kernel math and below-limit one-dimensional dispatch behavior are unchanged. Co-authored-with: Claude Code. ghstack-source-id: 411059172 @exported-using-ghexport Differential Revision: [D114936147](https://our.internmc.facebook.com/intern/diff/D114936147/)

JCNTH · 2026-08-05T19:55:09Z

Stack from ghstack (oldest at bottom):

Run production FastVLM shapes with exception-safe WebGPU graph construction

Production FastVLM tensors exceeded WebGPU one-dimensional dispatch limits, while several parameter buffers were not graph-owned before fallible pipeline construction. The runtime now folds flat dispatches across x/y, owns parameters immediately, rejects ambiguous BOOL/byte-integer casts, and compiles timestamp validation only in profiling builds.

Key changes:

GELU and expand_copy use overflow-safe folded dispatch and live resize updates.
Compare, expand_copy, and to_copy create graph-owned parameter buffers before pipeline construction, mirroring Vulkan ComputeGraph ownership.
to_copy rejects BOOL-to/from-INT8 or UINT8 before same-width flat-copy routing.
Production builds exclude timestamp-query validation and symbols.
Boundary, failure, and real FastVLM encoder coverage exercise the final combined behavior.

Kernel math and below-limit one-dimensional dispatch behavior are unchanged.

Co-authored-with: Claude Code.

Differential Revision: D114936147

[ghstack-poisoned]

pytorch-bot · 2026-08-05T19:55:13Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/21602

📄 Preview Python docs built from this PR

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

❌ 1 New Failure

As of commit ab3d0de with merge base ad3a71f ():

NEW FAILURE - The following job has failed:

pull / android / run-emulator (gh)
adb: failed to install android-test-debug-androidTest.apk: cmd: Failure calling service package: Broken pipe (32)

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

github-actions · 2026-08-05T19:57:02Z

This PR needs a `release notes:` label

If your change should be included in the release notes (i.e. would users of this library care about this change?), please use a label starting with release notes:. This helps us keep track and include your important work in the next release notes.

To add a label, you can comment to pytorchbot, for example
@pytorchbot label "release notes: none"

For more information, see
https://github.com/pytorch/pytorch/wiki/PyTorch-AutoLabel-Bot#why-categorize-for-release-notes-and-how-does-it-work.

[ghstack-poisoned]

…nstruction Pull Request resolved: #21602 **Run production FastVLM shapes with exception-safe WebGPU graph construction** Production FastVLM tensors exceeded WebGPU one-dimensional dispatch limits, while several parameter buffers were not graph-owned before fallible pipeline construction. The runtime now folds flat dispatches across x/y, owns parameters immediately, rejects ambiguous `BOOL`/byte-integer casts, and compiles timestamp validation only in profiling builds. Key changes: - `GELU` and `expand_copy` use overflow-safe folded dispatch and live resize updates. - `Compare`, `expand_copy`, and `to_copy` create graph-owned parameter buffers before pipeline construction, mirroring Vulkan `ComputeGraph` ownership. - `to_copy` rejects `BOOL`-to/from-`INT8` or `UINT8` before same-width flat-copy routing. - Production builds exclude timestamp-query validation and symbols. - Boundary, failure, and real FastVLM encoder coverage exercise the final combined behavior. Kernel math and below-limit one-dimensional dispatch behavior are unchanged. Co-authored-with: Claude Code. ghstack-source-id: 411059172 @exported-using-ghexport Differential Revision: [D114936147](https://our.internmc.facebook.com/intern/diff/D114936147/)

JCNTH · 2026-08-05T19:55:09Z

Stack from ghstack (oldest at bottom):

Run production FastVLM shapes with exception-safe WebGPU graph construction

Production FastVLM tensors exceeded WebGPU one-dimensional dispatch limits, while several parameter buffers were not graph-owned before fallible pipeline construction. The runtime now folds flat dispatches across x/y, owns parameters immediately, rejects ambiguous BOOL/byte-integer casts, and compiles timestamp validation only in profiling builds.

Key changes:

GELU and expand_copy use overflow-safe folded dispatch and live resize updates.
Compare, expand_copy, and to_copy create graph-owned parameter buffers before pipeline construction, mirroring Vulkan ComputeGraph ownership.
to_copy rejects BOOL-to/from-INT8 or UINT8 before same-width flat-copy routing.
Production builds exclude timestamp-query validation and symbols.
Boundary, failure, and real FastVLM encoder coverage exercise the final combined behavior.

Kernel math and below-limit one-dimensional dispatch behavior are unchanged.

Co-authored-with: Claude Code.

Differential Revision: D114936147

[ghstack-poisoned]

pytorch-bot · 2026-08-05T19:55:13Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/21602

📄 Preview Python docs built from this PR

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

❌ 1 New Failure

As of commit ab3d0de with merge base ad3a71f ():

NEW FAILURE - The following job has failed:

pull / android / run-emulator (gh)
adb: failed to install android-test-debug-androidTest.apk: cmd: Failure calling service package: Broken pipe (32)

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

github-actions · 2026-08-05T19:57:02Z

This PR needs a `release notes:` label

If your change should be included in the release notes (i.e. would users of this library care about this change?), please use a label starting with release notes:. This helps us keep track and include your important work in the next release notes.

To add a label, you can comment to pytorchbot, for example
@pytorchbot label "release notes: none"

For more information, see
https://github.com/pytorch/pytorch/wiki/PyTorch-AutoLabel-Bot#why-categorize-for-release-notes-and-how-does-it-work.

[ghstack-poisoned]

…nstruction Pull Request resolved: #21602 **Run production FastVLM shapes with exception-safe WebGPU graph construction** Production FastVLM tensors exceeded WebGPU one-dimensional dispatch limits, while several parameter buffers were not graph-owned before fallible pipeline construction. The runtime now folds flat dispatches across x/y, owns parameters immediately, rejects ambiguous `BOOL`/byte-integer casts, and compiles timestamp validation only in profiling builds. Key changes: - `GELU` and `expand_copy` use overflow-safe folded dispatch and live resize updates. - `Compare`, `expand_copy`, and `to_copy` create graph-owned parameter buffers before pipeline construction, mirroring Vulkan `ComputeGraph` ownership. - `to_copy` rejects `BOOL`-to/from-`INT8` or `UINT8` before same-width flat-copy routing. - Production builds exclude timestamp-query validation and symbols. - Boundary, failure, and real FastVLM encoder coverage exercise the final combined behavior. Kernel math and below-limit one-dimensional dispatch behavior are unchanged. Co-authored-with: Claude Code. ghstack-source-id: 411059172 @exported-using-ghexport Differential Revision: [D114936147](https://our.internmc.facebook.com/intern/diff/D114936147/)

Update
5275c25
[ghstack-poisoned]

JCNTH temporarily deployed to cadence August 5, 2026 19:55 — with GitHub Actions Inactive

meta-claBot 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 Aug 5, 2026

JCNTH had a problem deploying to cadence August 5, 2026 20:35 — with GitHub Actions Error

Update
ab3d0de
[ghstack-poisoned]

JCNTH temporarily deployed to cadence August 5, 2026 20:44 — with GitHub Actions Inactive

meta-codesyncBot added the meta-exported label Aug 5, 2026

JCNTH temporarily deployed to cadence August 5, 2026 21:33 — with GitHub Actions Inactive

psiddh approved these changes Aug 6, 2026
View reviewed changes

meta-codesyncBot merged commit 775a310 into gh/JCNTH/209/baseAug 7, 2026
189 of 198 checks passed

meta-codesyncBot deleted the gh/JCNTH/209/head branch August 7, 2026 17:23

meta-codesyncBottemporarily deployed to cherry-pick-bot August 7, 2026 17:23 Inactive

pytorchbot mentioned this pull request Aug 7, 2026
[ExecuTorch][WebGPU] Lift FastVLM dispatch limits and harden graph construction #21667
Merged

Update
5275c25
[ghstack-poisoned]

JCNTH temporarily deployed to cadence August 5, 2026 19:55 — with GitHub Actions Inactive

meta-claBot 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 Aug 5, 2026

JCNTH had a problem deploying to cadence August 5, 2026 20:35 — with GitHub Actions Error

Update
ab3d0de
[ghstack-poisoned]

JCNTH temporarily deployed to cadence August 5, 2026 20:44 — with GitHub Actions Inactive

meta-codesyncBot added the meta-exported label Aug 5, 2026

JCNTH temporarily deployed to cadence August 5, 2026 21:33 — with GitHub Actions Inactive

psiddh approved these changes Aug 6, 2026
View reviewed changes

meta-codesyncBot merged commit 775a310 into gh/JCNTH/209/baseAug 7, 2026
189 of 198 checks passed

meta-codesyncBot deleted the gh/JCNTH/209/head branch August 7, 2026 17:23

meta-codesyncBottemporarily deployed to cherry-pick-bot August 7, 2026 17:23 Inactive

pytorchbot mentioned this pull request Aug 7, 2026
[ExecuTorch][WebGPU] Lift FastVLM dispatch limits and harden graph construction #21667
Merged

Update
5275c25
[ghstack-poisoned]

JCNTH temporarily deployed to cadence August 5, 2026 19:55 — with GitHub Actions Inactive

meta-claBot 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 Aug 5, 2026

JCNTH had a problem deploying to cadence August 5, 2026 20:35 — with GitHub Actions Error

Update
ab3d0de
[ghstack-poisoned]

JCNTH temporarily deployed to cadence August 5, 2026 20:44 — with GitHub Actions Inactive

meta-codesyncBot added the meta-exported label Aug 5, 2026

JCNTH temporarily deployed to cadence August 5, 2026 21:33 — with GitHub Actions Inactive

psiddh approved these changes Aug 6, 2026
View reviewed changes

meta-codesyncBot merged commit 775a310 into gh/JCNTH/209/baseAug 7, 2026
189 of 198 checks passed

meta-codesyncBot deleted the gh/JCNTH/209/head branch August 7, 2026 17:23

meta-codesyncBottemporarily deployed to cherry-pick-bot August 7, 2026 17:23 Inactive

pytorchbot mentioned this pull request Aug 7, 2026
[ExecuTorch][WebGPU] Lift FastVLM dispatch limits and harden graph construction #21667
Merged

Update
5275c25
[ghstack-poisoned]

JCNTH temporarily deployed to cadence August 5, 2026 19:55 — with GitHub Actions Inactive

meta-claBot 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 Aug 5, 2026

JCNTH had a problem deploying to cadence August 5, 2026 20:35 — with GitHub Actions Error

Update
ab3d0de
[ghstack-poisoned]

JCNTH temporarily deployed to cadence August 5, 2026 20:44 — with GitHub Actions Inactive

meta-codesyncBot added the meta-exported label Aug 5, 2026

JCNTH temporarily deployed to cadence August 5, 2026 21:33 — with GitHub Actions Inactive

psiddh approved these changes Aug 6, 2026
View reviewed changes

meta-codesyncBot merged commit 775a310 into gh/JCNTH/209/baseAug 7, 2026
189 of 198 checks passed

meta-codesyncBot deleted the gh/JCNTH/209/head branch August 7, 2026 17:23

meta-codesyncBottemporarily deployed to cherry-pick-bot August 7, 2026 17:23 Inactive

pytorchbot mentioned this pull request Aug 7, 2026
[ExecuTorch][WebGPU] Lift FastVLM dispatch limits and harden graph construction #21667
Merged

Update
5275c25
[ghstack-poisoned]

JCNTH temporarily deployed to cadence August 5, 2026 19:55 — with GitHub Actions Inactive

meta-claBot 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 Aug 5, 2026

JCNTH had a problem deploying to cadence August 5, 2026 20:35 — with GitHub Actions Error

Update
ab3d0de
[ghstack-poisoned]

JCNTH temporarily deployed to cadence August 5, 2026 20:44 — with GitHub Actions Inactive

meta-codesyncBot added the meta-exported label Aug 5, 2026

JCNTH temporarily deployed to cadence August 5, 2026 21:33 — with GitHub Actions Inactive

psiddh approved these changes Aug 6, 2026
View reviewed changes

meta-codesyncBot merged commit 775a310 into gh/JCNTH/209/baseAug 7, 2026
189 of 198 checks passed

meta-codesyncBot deleted the gh/JCNTH/209/head branch August 7, 2026 17:23

meta-codesyncBottemporarily deployed to cherry-pick-bot August 7, 2026 17:23 Inactive

pytorchbot mentioned this pull request Aug 7, 2026
[ExecuTorch][WebGPU] Lift FastVLM dispatch limits and harden graph construction #21667
Merged

Update
5275c25
[ghstack-poisoned]

JCNTH temporarily deployed to cadence August 5, 2026 19:55 — with GitHub Actions Inactive

meta-claBot 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 Aug 5, 2026

JCNTH had a problem deploying to cadence August 5, 2026 20:35 — with GitHub Actions Error

Update
ab3d0de
[ghstack-poisoned]

JCNTH temporarily deployed to cadence August 5, 2026 20:44 — with GitHub Actions Inactive

meta-codesyncBot added the meta-exported label Aug 5, 2026

JCNTH temporarily deployed to cadence August 5, 2026 21:33 — with GitHub Actions Inactive

psiddh approved these changes Aug 6, 2026
View reviewed changes

meta-codesyncBot merged commit 775a310 into gh/JCNTH/209/baseAug 7, 2026
189 of 198 checks passed

meta-codesyncBot deleted the gh/JCNTH/209/head branch August 7, 2026 17:23

meta-codesyncBottemporarily deployed to cherry-pick-bot August 7, 2026 17:23 Inactive

pytorchbot mentioned this pull request Aug 7, 2026
[ExecuTorch][WebGPU] Lift FastVLM dispatch limits and harden graph construction #21667
Merged

Update
5275c25
[ghstack-poisoned]

JCNTH temporarily deployed to cadence August 5, 2026 19:55 — with GitHub Actions Inactive

meta-claBot 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 Aug 5, 2026

JCNTH had a problem deploying to cadence August 5, 2026 20:35 — with GitHub Actions Error

Update
ab3d0de
[ghstack-poisoned]

JCNTH temporarily deployed to cadence August 5, 2026 20:44 — with GitHub Actions Inactive

meta-codesyncBot added the meta-exported label Aug 5, 2026

JCNTH temporarily deployed to cadence August 5, 2026 21:33 — with GitHub Actions Inactive

psiddh approved these changes Aug 6, 2026
View reviewed changes

meta-codesyncBot merged commit 775a310 into gh/JCNTH/209/baseAug 7, 2026
189 of 198 checks passed

meta-codesyncBot deleted the gh/JCNTH/209/head branch August 7, 2026 17:23

meta-codesyncBottemporarily deployed to cherry-pick-bot August 7, 2026 17:23 Inactive

pytorchbot mentioned this pull request Aug 7, 2026
[ExecuTorch][WebGPU] Lift FastVLM dispatch limits and harden graph construction #21667
Merged

Update
5275c25
[ghstack-poisoned]

JCNTH temporarily deployed to cadence August 5, 2026 19:55 — with GitHub Actions Inactive

meta-claBot 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 Aug 5, 2026

JCNTH had a problem deploying to cadence August 5, 2026 20:35 — with GitHub Actions Error

Update
ab3d0de
[ghstack-poisoned]

JCNTH temporarily deployed to cadence August 5, 2026 20:44 — with GitHub Actions Inactive

meta-codesyncBot added the meta-exported label Aug 5, 2026

JCNTH temporarily deployed to cadence August 5, 2026 21:33 — with GitHub Actions Inactive

psiddh approved these changes Aug 6, 2026
View reviewed changes

meta-codesyncBot merged commit 775a310 into gh/JCNTH/209/baseAug 7, 2026
189 of 198 checks passed

meta-codesyncBot deleted the gh/JCNTH/209/head branch August 7, 2026 17:23

meta-codesyncBottemporarily deployed to cherry-pick-bot August 7, 2026 17:23 Inactive

pytorchbot mentioned this pull request Aug 7, 2026
[ExecuTorch][WebGPU] Lift FastVLM dispatch limits and harden graph construction #21667
Merged

Uh oh!

Conversation

JCNTH commented Aug 5, 2026• edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

pytorch-botBot commented Aug 5, 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/21602

❌ 1 New Failure

Uh oh!

github-actionsBot commented Aug 5, 2026

This PR needs a release notes: label

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Uh oh!

Conversation

JCNTH commented Aug 5, 2026• edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

pytorch-botBot commented Aug 5, 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/21602

❌ 1 New Failure

Uh oh!

github-actionsBot commented Aug 5, 2026

This PR needs a release notes: label

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Uh oh!

Conversation

JCNTH commented Aug 5, 2026• edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

pytorch-botBot commented Aug 5, 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/21602

❌ 1 New Failure

Uh oh!

github-actionsBot commented Aug 5, 2026

This PR needs a release notes: label

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Uh oh!

Conversation

JCNTH commented Aug 5, 2026• edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

pytorch-botBot commented Aug 5, 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/21602

❌ 1 New Failure

Uh oh!

JCNTH commented Aug 5, 2026•
edited
Loading

pytorch-botBot commented Aug 5, 2026•
edited
Loading

This PR needs a `release notes:` label

JCNTH commented Aug 5, 2026•
edited
Loading

pytorch-botBot commented Aug 5, 2026•
edited
Loading

This PR needs a `release notes:` label

JCNTH commented Aug 5, 2026•
edited
Loading

pytorch-botBot commented Aug 5, 2026•
edited
Loading

This PR needs a `release notes:` label

JCNTH commented Aug 5, 2026•
edited
Loading

pytorch-botBot commented Aug 5, 2026•
edited
Loading

This PR needs a `release notes:` label

JCNTH commented Aug 5, 2026•
edited
Loading

pytorch-botBot commented Aug 5, 2026•
edited
Loading

This PR needs a `release notes:` label

JCNTH commented Aug 5, 2026•
edited
Loading

pytorch-botBot commented Aug 5, 2026•
edited
Loading