This repository has been archived by the owner on Mar 21, 2024. It is now read-only.
Reduce comparisons count in merge sort #1442
Merged
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
|
Can one of the admins verify this patch? |
|
add to allowlist |
|
run tests |
alliepiper
suggested changes
May 26, 2021
gevtushenko
force-pushed
the
main-feature/github/thrust_merge_sort_comparisons_count_reduction/1436
branch
from
1f59b78 to
403829e
Compare
alliepiper
approved these changes
May 26, 2021
|
DVS CL: 30008797 run tests |
alliepiper
added
testing: internal ci in progress
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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Outdated suggestions cannot be applied.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.
In order to fix the issue I've reduced the comparisons count in case of partially filled tile. The issue is caused by the fact that each thread has a fixed amount of data to process. If the input data size isn't a multiple of
ITEMS_PER_TILE, there isn't enough data to fill the local storage of each thread. In this case, extra elements of the storage are filled with some fixed value. After that the storage is sorted bystable_odd_even_sortby each thread (even ones outside the input range). The idea behind the fix is in skippingstable_odd_even_sortfor threads of the last tile with no actual input data because in this case their storages are filled with the fixed value.The fix doesn't affect performance even for small input data sizes and complex comparators. For input data size of 2 INT32 elements number of comparisons is reduced by 507 times (RTX3090).