[Type] Mat: better cache locality for operator*(Mat) #5921
+6
−8
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
fredroy
added
pr: enhancement
alxbilger
added
the
pr: ai-generated
alxbilger
requested changes
Feb 3, 2026
Contributor
alxbilger
left a comment
alxbilger
left a comment
There was a problem hiding this comment.
You must initialize the result before calling the operator +=.
fredroy
force-pushed
the
optim_mat_operator_mult
branch
from
0ca315f to
eb57d55
Compare
Contributor
Author
done , and re-did the benches (no change) |
bakpaul
approved these changes
Feb 4, 2026
| { | ||
| Mat<L,P,real> r(NOINIT); | ||
| for (Size i = 0; i<L; i++) | ||
| Mat<L,P,real> r; |
Contributor
There was a problem hiding this comment.
Suggested change
| Mat<L,P,real> r; | |
| Mat<L,P,real> r(0.0); |
The ctor w/o arguments doesn't set the inital values to 0
Contributor
There was a problem hiding this comment.
Well I missclicked, I didn't want to approve X)
Contributor
There was a problem hiding this comment.
Hmm this is a "recent" change. Before 44ad519, the default constructor initialized the values...
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 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Outdated suggestions cannot be applied.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.
Changing accesses for better cache locality (suggested by AI)
TL;DR:
the Mat<3,3> version does not change because it has its own optimized specialized version
bigger the matrices, bigger the gain (Mat24x24, speedup of 400% in floats !)
macOS has a weird quirk for Mat6x6 on double, which is 50% slower ? 🤔 maybe due to a failed vectorization or somethin'
Timings:
Ubuntu 22.04, gcc12, lto, O3
Windows VS2026, release, lto
macOS, xcode 26, lto
By submitting this pull request, I acknowledge that
I have read, understand, and agree SOFA Developer Certificate of Origin (DCO).
Reviewers will merge this pull-request only if