Repository navigation
Conversation
This is supported nowadays, and slightly speeds up testing.
Otherwise we spend too much time releasing memory upon every synchronization.
Codecov Report
@@ Coverage Diff @@
## master #1344 +/- ##
==========================================
- Coverage 78.85% 78.82% -0.04%
==========================================
Files 119 119
Lines 8689 8674 -15
==========================================
- Hits 6852 6837 -15
Misses 1837 1837
Continue to review full report at Codecov.
|
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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 most significant change is that the pool is now allowed to use all memory, which should prevent stalls when synchronizing. See https://forums.developer.nvidia.com/t/gpu-stalls-due-to-stream-synchronization-even-when-idle/191761 and under-Peter/OMEinsum.jl#133 (comment); closes #1320.
cc @GiggleLiu.
cc @DhairyaLGandhi, this may significantly impact memory-heavy ML workloads. It may also result in non pool-aware applications or libraries to perform worse, so I'd appreciate if you could evaluate this and give some feedback. Note that it does require use of the memory pool, i.e., CUDA 11.2 or higher and not setting
JULIA_CUDA_MEMORY_POOL=none.