Lock free decompression #139
Open
Conversation
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
Owner
|
Just coming back to this after the person who maintained this package for the last few years has moved on to other things. This looks very cool. Could you update this branch with everything that's been added in the interim? That should trigger the continuous integration. |
Author
|
So far I only clicked "update branch" on github - presumably it will need some more work before it will be ready. |
Owner
|
Tests fail with a "bus error 10" which sounds like some sort of issue addressing memory. |
Author
|
Sorry for that - the init was missed for non-openmp builds - it should be ready for you to look at now |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.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.
As mentioned in issue #138, you can get a speed up for certain datasets on large CPU's by using a static openmp schedule. In this pull request, all threads read all offsets to find the compressed data and output pointers are computed.
Depending on whether load balancing is needed by other people, it might be better to use
schedule(runtime)instead ofschedule(static).