-
Notifications
You must be signed in to change notification settings - Fork 552
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
In FIL, clip blocks_per_sm to one wave instead of asserting #4271
Conversation
ASSERT(blocks_per_sm <= max_blocks_per_sm, | ||
"on this GPU, FIL blocks_per_sm cannot exceed %d", | ||
max_blocks_per_sm); | ||
blocks_per_sm = std::min(blocks_per_sm, max_threads_per_sm / FIL_TPB); |
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.
I think we should also add a warning message if blocks_per_sm > max_threads_per_sm / FIL_TPB
.
@gpucibot merge |
Codecov Report
@@ Coverage Diff @@
## branch-21.12 #4271 +/- ##
===============================================
Coverage ? 86.06%
===============================================
Files ? 231
Lines ? 18691
Branches ? 0
===============================================
Hits ? 16087
Misses ? 2604
Partials ? 0
Flags with carried forward coverage won't be shown. Click here to find out more. Continue to review full report at Codecov.
|
…#4271) After using the feature for a while, we found it easier to sometimes set a high number and expect FIL to scale down when the GPU requires, instead of checking externally every time. Authors: - Levs Dolgovs (https://github.com/levsnv) Approvers: - Jordan Jacobelli (https://github.com/Ethyling) - Andy Adinets (https://github.com/canonizer) - Dante Gama Dessavre (https://github.com/dantegd) URL: rapidsai#4271
After using the feature for a while, we found it easier to sometimes set a high number and expect FIL to scale down when the GPU requires, instead of checking externally every time.