[Caffe2] Build perfkernels targeting SVE128 #159274
Open
+1,090
−426
Conversation
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
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/159274
Note: Links to docs will display an error until the docs builds have been completed. ❌ 9 New FailuresAs of commit d3dd950 with merge base bfc873d ( NEW FAILURES - The following jobs have failed:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
|
|
|
This pull request was exported from Phabricator. Differential Revision: D78902495 |
Skylion007
reviewed
Jul 29, 2025
| } | ||
| template <typename T> | ||
| std::ostream& operator<<(std::ostream& stream, const Vectorized<T>& vec) { | ||
| T buf[Vectorized<T>::size()]; |
There was a problem hiding this comment.
Yikes dynamic array... is size constexpr? Use std::array if so
Skylion007
reviewed
Jul 29, 2025
| Vec v1 = v1_1; | ||
| float32x4_t v1_1 = vextq_f32(vReg, vReg, 2); | ||
|
|
||
| __at_align__ float v1[4]; |
There was a problem hiding this comment.
Can you not do this with std::array and data?
|
This pull request was exported from Phabricator. Differential Revision: D78902495 |
e24bfba to
587299f
Compare
|
This pull request was exported from Phabricator. Differential Revision: D78902495 |
587299f to
7c03eb8
Compare
Nicoshev
added a commit
to Nicoshev/pytorch
that referenced
this pull request
Aug 4, 2025
) Summary: Pull Request resolved: pytorch#159274 We are introducing the SVE128 vectorized<> layer. Idea is to differentiate SVE128 perfkernels from the general SVE implementation Mixing NEON and SVE should maximize performance on SVE128 cpus Test Plan: Sigrid Predictor canary Rollback Plan: Differential Revision: D78902495
|
This pull request was exported from Phabricator. Differential Revision: D78902495 |
Nicoshev
added a commit
to Nicoshev/pytorch
that referenced
this pull request
Aug 5, 2025
) Summary: Pull Request resolved: pytorch#159274 We are introducing the SVE128 vectorized<> layer. Idea is to differentiate SVE128 perfkernels from the general SVE implementation Mixing NEON and SVE should maximize performance on SVE128 cpus Test Plan: Sigrid Predictor canary Rollback Plan: Differential Revision: D78902495
7c03eb8 to
f4df25b
Compare
added 2 commits
August 9, 2025 07:02
Summary: Pull Request resolved: pytorch#158932 Importing pytorch#138388, as it improves SVE support for perfkernels Test Plan: We will test it on AdFinder/AdRetriever/AdRanker offline tier Differential Revision: D70788867 Reviewed By: r1mikey
Summary: Enabling compilation targeting SVE128 Test Plan: AdRanker/AdFinder ServiceLab Differential Revision: D78691521
|
This pull request was exported from Phabricator. Differential Revision: D78902495 |
Nicoshev
added a commit
to Nicoshev/pytorch
that referenced
this pull request
Aug 11, 2025
) Summary: Pull Request resolved: pytorch#159274 We are introducing the SVE128 vectorized<> layer. Idea is to differentiate SVE128 perfkernels from the general SVE implementation Mixing NEON and SVE should maximize performance on SVE128 cpus Test Plan: Sigrid Predictor canary Rollback Plan: Differential Revision: D78902495
f4df25b to
1a4f936
Compare
) Summary: Pull Request resolved: pytorch#159274 We are introducing the SVE128 vectorized<> layer. Idea is to differentiate SVE128 perfkernels from the general SVE implementation Mixing NEON and SVE should maximize performance on SVE128 cpus Test Plan: Sigrid Predictor canary Rollback Plan: Differential Revision: D78902495
|
This pull request was exported from Phabricator. Differential Revision: D78902495 |
1a4f936 to
d3dd950
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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: We are now building perfkernels using SVE/Neon enhancements
Test Plan:
Sigrid Predictor canary
Rollback Plan:
Differential Revision: D78902495
cc @jgong5 @mingfeima @XiaobingSuper @sanchitintel @ashokei @jingxu10 @jerryzh168 @voznesenskym @penguinwu @EikanWang @Guobing-Chen @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @ipiszy @chenyang78 @kadeng @muchulee8 @amjames @chauhang @aakhundov @coconutruben