Implement PackDepthwiseConvMatrix in NEON + deprecate aarch64 compat layers (#5779)#5779
Closed
Nicoshev wants to merge 1 commit into
Closed
Implement PackDepthwiseConvMatrix in NEON + deprecate aarch64 compat layers (#5779)#5779Nicoshev wants to merge 1 commit into
Nicoshev wants to merge 1 commit into
Conversation
Contributor
|
@Nicoshev has exported this pull request. If you are a Meta employee, you can view the originating Diff in D106137964. |
Nicoshev
added a commit
to Nicoshev/FBGEMM
that referenced
this pull request
May 25, 2026
…layers (pytorch#5779) Summary: X-link: https://github.com/facebookresearch/FBGEMM/pull/2709 Pull Request resolved: pytorch#5779 Add a NEON-based aarch64 implementation of the `PackedDepthWiseConvMatrix` constructor in `PackDepthwiseConvMatrix.cc`, alongside the existing AVX2 x86 implementation. The constructor packs depthwise convolution weight matrices into a SIMD-friendly interleaved layout. Rename depthwise-convolution related files, as NEON and AVX2 implementations already co-exist Remove compilation of avx2 source files for aarch64 targets and remove usage of aarch64 compat layers Differential Revision: D106137964
meta-codesync
Bot
changed the title
Nicoshev
force-pushed
the
export-D106137964
branch
from
3cf57eb to
32e1b26
Compare
Nicoshev
added a commit
to Nicoshev/FBGEMM
that referenced
this pull request
May 25, 2026
…layers (pytorch#5779) Summary: X-link: https://github.com/facebookresearch/FBGEMM/pull/2709 Pull Request resolved: pytorch#5779 Add a NEON-based aarch64 implementation of the `PackedDepthWiseConvMatrix` constructor in `PackDepthwiseConvMatrix.cc`, alongside the existing AVX2 x86 implementation. The constructor packs depthwise convolution weight matrices into a SIMD-friendly interleaved layout. Rename depthwise-convolution related files, as NEON and AVX2 implementations already co-exist Remove compilation of avx2 source files for aarch64 targets and remove usage of aarch64 compat layers Differential Revision: D106137964
Nicoshev
force-pushed
the
export-D106137964
branch
2 times, most recently
from
bcc69a5 to
513f790
Compare
Nicoshev
added a commit
to Nicoshev/FBGEMM
that referenced
this pull request
May 29, 2026
…layers (pytorch#5779) Summary: X-link: https://github.com/facebookresearch/FBGEMM/pull/2709 Pull Request resolved: pytorch#5779 Add a NEON-based aarch64 implementation of the `PackedDepthWiseConvMatrix` constructor in `PackDepthwiseConvMatrix.cc`, alongside the existing AVX2 x86 implementation. The constructor packs depthwise convolution weight matrices into a SIMD-friendly interleaved layout. Rename depthwise-convolution related files, as NEON and AVX2 implementations already co-exist Remove compilation of avx2 source files for aarch64 targets and remove usage of aarch64 compat layers Reviewed By: q10, YifanYuan3 Differential Revision: D106137964
Nicoshev
added a commit
to Nicoshev/FBGEMM
that referenced
this pull request
May 29, 2026
…layers (pytorch#5779) Summary: X-link: https://github.com/facebookresearch/FBGEMM/pull/2709 Pull Request resolved: pytorch#5779 Add a NEON-based aarch64 implementation of the `PackedDepthWiseConvMatrix` constructor in `PackDepthwiseConvMatrix.cc`, alongside the existing AVX2 x86 implementation. The constructor packs depthwise convolution weight matrices into a SIMD-friendly interleaved layout. Rename depthwise-convolution related files, as NEON and AVX2 implementations already co-exist Remove compilation of avx2 source files for aarch64 targets and remove usage of aarch64 compat layers Reviewed By: q10, YifanYuan3 Differential Revision: D106137964
Nicoshev
force-pushed
the
export-D106137964
branch
from
513f790 to
bad96b5
Compare
…layers (pytorch#5779) Summary: X-link: https://github.com/facebookresearch/FBGEMM/pull/2709 Pull Request resolved: pytorch#5779 Add a NEON-based aarch64 implementation of the `PackedDepthWiseConvMatrix` constructor in `PackDepthwiseConvMatrix.cc`, alongside the existing AVX2 x86 implementation. The constructor packs depthwise convolution weight matrices into a SIMD-friendly interleaved layout. Rename depthwise-convolution related files, as NEON and AVX2 implementations already co-exist Remove compilation of avx2 source files for aarch64 targets and remove usage of aarch64 compat layers Reviewed By: q10, YifanYuan3 Differential Revision: D106137964
Nicoshev
force-pushed
the
export-D106137964
branch
from
bad96b5 to
0a88cea
Compare
Contributor
|
This pull request has been merged in 07767a8. |
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
1 participant
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.
Summary:
X-link: https://github.com/facebookresearch/FBGEMM/pull/2709
Add a NEON-based aarch64 implementation of the
PackedDepthWiseConvMatrixconstructor inPackDepthwiseConvMatrix.cc, alongside the existing AVX2 x86 implementation. The constructor packs depthwise convolution weight matrices into a SIMD-friendly interleaved layout.Rename depthwise-convolution related files, as NEON and AVX2 implementations already co-exist
Remove compilation of avx2 source files for aarch64 targets and remove usage of aarch64 compat layers
Reviewed By: q10, YifanYuan3
Differential Revision: D106137964