In-Place Dense Matrix Transposition #2199
Open
Conversation
This file contains 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
Baunsgaard
reviewed
Jan 30, 2025
| matrix[comp] = cval; | ||
| break; | ||
| } | ||
| if(prevComp == start) { |
| int prevOrig = prevIndexCycle(orig, rows, (maxIndex + 1) / rows); | ||
| int prevComp = maxIndex - prevOrig; | ||
|
|
||
| while(true) { |
There was a problem hiding this comment.
you can probably help the compiler and improve performance a bit, by moving the content of this loop to another function.
furthermore, is this loop possible to do in parallel?
|
|
||
| LibMatrixReorg.transposeInPlaceDenseBrenner(X, 1); | ||
|
|
||
| TestUtils.compareMatrices(X, tX, 1e-8); |
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #2199 +/- ##
============================================
+ Coverage 71.88% 72.30% +0.42%
- Complexity 44701 45023 +322
============================================
Files 1449 1452 +3
Lines 169182 169417 +235
Branches 32980 33059 +79
============================================
+ Hits 121617 122498 +881
+ Misses 38237 37602 -635
+ Partials 9328 9317 -11 ☔ View full report in Codecov by Sentry. |
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Outdated suggestions cannot be applied.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.
Added a new kernel for In-Place Dense Matrix Transposition, based on Algorithm 467 by Brenner (DOI: 10.1145/355611.362542).
Performance:
Compared to the existing kernel, the added method provides significant performance benefits in a single-threaded context:
Note:
Performance measurements were restricted to cases where the existing kernel yields correct results. Similar or even better performance could be observed across all cases.
Future Work:
The divisors operate on different indices of the array, allowing for parallelization and offering additional performance improvements in multi-threaded scenarios.
@mboehm7