-
Notifications
You must be signed in to change notification settings - Fork 22
Do concurrent #59
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: main
Are you sure you want to change the base?
Do concurrent #59
Conversation
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #59 +/- ##
==========================================
- Coverage 82.99% 82.92% -0.08%
==========================================
Files 67 68 +1
Lines 2317 2202 -115
Branches 182 183 +1
==========================================
- Hits 1923 1826 -97
+ Misses 394 376 -18 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
Intent specs
Update do concurrent branch with latest devs from main.
Trying out if I enter some untested lines yet.
zoziha
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@loiseaujc Thank you for your attention to and improvements on fftpack. I have made some modifications regarding the do concurrent indexing, hoping they will be helpful. Everything else looks good to me, and the tests have passed.
Co-authored-by: ZUO Zhihua <[email protected]>
|
Done. |
Most of the loops that could be replaced by
do concurrenthave been.On my laptop, here are the timings I gets:
Legacy code:
bench01_zfft: 3.19 secondsbench02_zfft: 4.10 secondsbench03_dfft: 1.80 secondsDo-concurrent code:
bench01_zfft: 2.26 seconds (1.4x speedup)bench02_zfft: 2.90 seconds (1.4x speedup)bench03_dfft: 1.50 seconds (1.2x speedup)Both codes are compiled with
gfortran 13.3. The speed-ups are obviously to be taken with a grain of salt. We would need to do a proper performances analysis with sequences of varying lengths. Still, it's pretty nice I believe.ping @perazz @zoziha
PS : This PR builds upon the
intent_specsone and should be merge after the latter.