Arm backend: Fix stale VGF BatchNorm support documentation - #21905
Merged
Conversation
perheld
added
partner: arm
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/21905
Note: Links to docs will display an error until the docs builds have been completed. ✅ You can merge normally! (1 Unrelated Failure)As of commit 5b00385 with merge base d1b3242 ( BROKEN TRUNK - The following job failed but were present on the merge base:👉 Rebase onto the `viable/strict` branch to avoid these failures
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
meta-cla
Bot
added
the
CLA Signed
The generated page retained INT support for torch.batch_norm even though the current VGF tests provide only FP evidence for torch.ops.aten.batch_norm.default. The history explains how the page became stale. Quantized non-folded BatchNorm support landed first in pytorch#21710 on August 11. The initial VGF operator documentation generator followed in pytorch#21776 on August 12. The BatchNorm test cleanup then landed in pytorch#21836 on August 14 and changed quantized VGF tests to use aten_op=[] or report only conv2d. Finally, pytorch#21871 landed on August 15, adding the generated page to the pre-push hook. Both BatchNorm changes are ancestors of pytorch#21871, and the generator and relevant tests have not changed since pytorch#21871. The generated page committed by pytorch#21871 was therefore already inconsistent with its final source tree, most likely because it was produced on an earlier PR state and retained when the PR was updated over the BatchNorm cleanup. The newly added hook could not validate that server-created squash commit, so the stale row was first exposed by a later local push. This correction does not remove quantized BatchNorm support represented by torch.ops.aten._native_batch_norm_legit_no_training.default. Its separate torch.nn.BatchNorm2d/functional.batch_norm row remains FP and INT based on the convolution-combination tests. Authored with Codex. Change-Id: I32d9a3ec01eec250d4badcfe49e1ded4f855fe0d Signed-off-by: Per Held <per.held@arm.com>
perheld
force-pushed
the
ph-vgf-batchnorm-doc
branch
from
August 18, 2026 07:43
4391064 to
5b00385
Compare
zingo
approved these changes
Aug 18, 2026
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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.
The generated page retained INT support for torch.batch_norm even though the current VGF tests provide only FP evidence for torch.ops.aten.batch_norm.default.
The history explains how the page became stale. Quantized non-folded BatchNorm support landed first in #21710 on August 11. The initial VGF operator documentation generator followed in #21776 on August 12. The BatchNorm test cleanup then landed in #21836 on August 14 and changed quantized VGF tests to use aten_op=[] or report only conv2d. Finally, #21871 landed on August 15, adding the generated page to the pre-push hook.
Both BatchNorm changes are ancestors of #21871, and the generator and relevant tests have not changed since #21871. The generated page committed by #21871 was therefore already inconsistent with its final source tree, most likely because it was produced on an earlier PR state and retained when the PR was updated over the BatchNorm cleanup. The newly added hook could not validate that server-created squash commit, so the stale row was first exposed by a later local push.
This correction does not remove quantized BatchNorm support represented by torch.ops.aten._native_batch_norm_legit_no_training.default. Its separate torch.nn.BatchNorm2d/functional.batch_norm row remains FP and INT based on the convolution-combination tests.
Authored with Codex.
Change-Id: I32d9a3ec01eec250d4badcfe49e1ded4f855fe0d
cc @digantdesai @freddan80 @per @zingo @oscarandersson8218 @mansnils @Sebastian-Larsson @robell @rascani