-
Notifications
You must be signed in to change notification settings - Fork 231
fix CUDA kernel syncthread usage #307
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
Conversation
|
ping @beomki-yeo |
Codecov Report
@@ Coverage Diff @@
## master #307 +/- ##
=======================================
Coverage 48.32% 48.32%
=======================================
Files 323 323
Lines 16376 16376
Branches 7603 7603
=======================================
Hits 7913 7913
Misses 3178 3178
Partials 5285 5285 Continue to review full report at Codecov.
|
|
Thanks for the PR. |
|
@cgleggett Could you update the PR to the latest master branch? |
msmk0
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.
Thanks @cgleggett and thank @beomki-yeo for verifying. I will merge this in.
Co-authored-by: leggett <leggett@cern.ch>
calling __syncthreads in a conditional that depends on the threadID results in undefined behaviour.
Interestingly, this wasn't apparent on older versions of the NVidia drivers, but is now.
Also, added option to SeedFinderCudaTest to not run on CPU, so as to test the GPU faster for large input files.