[BUG] GpuPartitioning should close CVs before releasing semaphore #6913
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
abellina
added
bug
abellina
changed the title
Signed-off-by: Alessandro Bellina <abellina@nvidia.com>
abellina
force-pushed
the
oom/partitioner_should_close_gpu_columns_earlier
branch
from
a1553ac
to
c665346
Compare
|
build |
jbrennan333
reviewed
Oct 25, 2022
jlowe
reviewed
Oct 25, 2022
sql-plugin/src/main/scala/com/nvidia/spark/rapids/GpuHashPartitioningBase.scala
Outdated
Show resolved
Hide resolved
|
build |
jlowe
approved these changes
Oct 25, 2022
jbrennan333
approved these changes
Oct 25, 2022
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Outdated suggestions cannot be applied.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.
Contributes to #6746, found while investigating #6758.
This is probably big in some cases? Before this change we would have copied to host, released the GPU, and then close the source columns. In some scenarios I could see N concurrent tasks reach here and end up allowing N more before we close the GPU buffers.
It is close in time, so I am not sure how much in practice we would see it.