onnxruntime
[MLAS/NEON] Add dedicated kernel for depthwise convolution for ARM64 using NEON intrinsics
#26688
Merged
Overview
Commits
39
Changes
View On GitHub

[MLAS/NEON] Add dedicated kernel for depthwise convolution for ARM64 using NEON intrinsics #26688

hariharans29 merged 39 commits into main from hari/expt_conv
hariharans29
hariharans29 Initial commit
160c71e2
hariharans29 More changes
a134ea00
hariharans29 More changes
a44f7085
hariharans29 Fix builds
3a3ccf76
hariharans29 Fix builds 2
212dbf1e
hariharans29 Threaded
fceae09b
hariharans29 Fix x64 builds
3793d70f
hariharans29 Experiment
481a7f66
hariharans29 hariharans29 changed the title WIP: Conv Expt [DO NOT REVIEW] WIP: Conv Expt 204 days ago
hariharans29
azure-pipelines
hariharans29 Experiment revert
8993a0a0
hariharans29 Refactor
d765c1a1
hariharans29 More changes
a428d509
hariharans29 a
d53dd15b
hariharans29 Try
8f12c51c
hariharans29 More changes
67b68013
hariharans29 Relax padding
01b43fb6
hariharans29 Vanilla NEON Depthwise
ea833947
hariharans29 Fix indexing
dd94a3b0
hariharans29 Add benchmark
ffd291ad
github-actions
github-actions commented on 2025-12-04
hariharans29 Add lambda
92fb6045
hariharans29 Rework
d15bb930
hariharans29 hariharans29 changed the title [DO NOT REVIEW] WIP: Conv Expt [MLAS/NEON] Add dedicated kernel for depthwise convolution for ARM64 using NEON intrinsics 202 days ago
hariharans29 Update onnxruntime/test/mlas/bench/bench_sconv.cpp
119ec9a7
hariharans29 Fix
d0fc1431
hariharans29 Remove Winograd implementation
2820a842
hariharans29 hariharans29 requested a review from copilot-pull-request-reviewer copilot-pull-request-reviewer 201 days ago
copilot-pull-request-reviewer
copilot-pull-request-reviewer commented on 2025-12-05
hariharans29 Update onnxruntime/core/mlas/lib/sconv_nchw_kernel_neon.cpp
0ffb8119
hariharans29 Update onnxruntime/core/mlas/lib/sconv_nchw_kernel_neon.cpp
59e2b2dc
hariharans29 Update onnxruntime/core/mlas/lib/convolve.cpp
e34c930a
hariharans29 Update onnxruntime/core/mlas/inc/mlas.h
027e7429
hariharans29 Update onnxruntime/core/mlas/lib/sconv_nchw_kernel_neon.cpp
f93ed67a
hariharans29 hariharans29 requested a review from copilot-pull-request-reviewer copilot-pull-request-reviewer 201 days ago
copilot-pull-request-reviewer
copilot-pull-request-reviewer commented on 2025-12-05
hariharans29 Update onnxruntime/core/mlas/lib/convolve.cpp
f5c1b812
hariharans29 Benchmark updates
f15e5540
hariharans29 Merge remote-tracking branch 'origin/main' into hari/expt_conv
bb324b59
hariharans29 Merge remote-tracking branch 'origin/main' into hari/expt_conv
3c0ec53b
Rohanjames1997
Rohanjames1997 commented on 2025-12-12
hariharans29
hariharans29 Merge remote-tracking branch 'origin' into hari/expt_conv
7b65a5fd
hariharans29 Use MLAS helpers instead of using NEON intrindics directly
60a2b81b
yuslepukhin
yuslepukhin commented on 2026-01-12
hariharans29 hariharans29 added release:1.24.0
hariharans29 Resolve comments
245f1cde
hariharans29 Typo
24874919
hariharans29 Merge remote-tracking branch 'origin' into hari/expt_conv
65bc06f4
hariharans29 Fix
20d2bb11
yuslepukhin
yuslepukhin approved these changes on 2026-01-15
hariharans29 hariharans29 enabled auto-merge (squash) 159 days ago
hariharans29 Merge remote-tracking branch 'origin' into hari/expt_conv
9c2a6546
hariharans29 hariharans29 merged c03c419f into main 158 days ago
hariharans29 hariharans29 deleted the hari/expt_conv branch 158 days ago
Rohanjames1997
hariharans29
Rohanjames1997
Rohanjames1997
tianleiwu tianleiwu removed release:1.24.0
tianleiwu tianleiwu added cherry-picked

Login to write a write a comment.

Login via GitHub

Reviewers
Assignees
No one assigned
Labels
cherry-picked
Milestone
No milestone
Terms Of Use Privacy Policy Refund Policy Impressum