fix hang in tc microbenchmark #127
Closed
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
|
Please explain why this change solves the problem in your PR description. |
|
CI fails because the containers need to be rebuilt. |
|
Superseded by #137. |
elgarten
added a commit
that referenced
this pull request
Aug 12, 2024
…#127 + #129) (#137) * Fix hang in TC microbenchmark Added a shared WaitGroup between vertex doAll and the nested, per vertex, edge doAll. Original code would hang because of the separate wait groups: after enqueueing a doAll in edge_tc_couting, harts wait for it to complete (tc_algos.cpp:42). However, this occurs on every hart because of the outer doAll in tc_no_chunk, therefore every hart is waiting and none is available to complete the work being waited on. When using one combined wait group, the outer doAll tasks are able to complete after enqueuing, but before completion of, the inner doAll tasks. Thus, harts are freed to complete the inner doAll and therefore forward progress. * fix overlapping prep timers * add synchronization to drv timers
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Outdated suggestions cannot be applied.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.
doAlland the nested, per vertex, edgedoAllOriginal code would hang because of the separate wait groups: after enqueueing a doAll in
edge_tc_couting,harts wait for it to complete (tc_algos.cpp:42). However, this occurs on every hart because of the outer doAll intc_no_chunk, therefore every hart is waiting and none is available to complete the work being waited on. When using one combined wait group, the outer doAll tasks are able to complete after enqueuing, but before completion of, the inner doAll tasks. Thus, harts are freed to complete the inner doAll and therefore forward progress.