WIP: fused compute kernels #65
Closed
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
…r pass over memory
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #65 +/- ##
==========================================
- Coverage 99.79% 97.11% -2.68%
==========================================
Files 13 13
Lines 2867 2946 +79
==========================================
Hits 2861 2861
- Misses 6 85 +79 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
Collaborator
Author
|
It's so early that it's not worth having a PR open for this yet. |
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Outdated suggestions cannot be applied.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.
The idea is to do twice the work per pass over memory. Code comments explain more.
https://fgiesen.wordpress.com/2023/03/19/notes-on-ffts-for-implementers/ calls this radix-2^2. It lets us keep the simple and SIMD-friendly bit reversal, but has the memory throughput efficiency almost double of the current impl and slightly better than even radix-4 because we need to load less twiddles. It would win on load/store efficiency even more if we were to ramp up the fusion factor instead of increasing the radix.