Marginal gains in send/recv fast-path #104
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
csegarragonz
commented
Jun 7, 2021
| @@ -1227,21 +1213,6 @@ long MpiWorld::getLocalQueueSize(int sendRank, int recvRank) | |||
| return queue->size(); | |||
| } | |||
|
|
|||
| void MpiWorld::checkRankOnThisHost(int rank) | |||
There was a problem hiding this comment.
Even if we decide to keep the if statements, we can delete this method and keep the check in getLocalQueue through getHostForRank.
csegarragonz
force-pushed
the
fast-path
branch
2 times, most recently
from
89b48bd
to
740ea9a
Compare
csegarragonz
commented
Jun 8, 2021
csegarragonz
commented
Jun 8, 2021
Shillaker
requested changes
Jun 8, 2021
Shillaker
approved these changes
Jun 8, 2021
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Outdated suggestions cannot be applied.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.
In this PR I revist the
send/recvfast-path and try to prune unnecessary complexity. The most notable changes are:if-based undefined behaviour to assertions. This is risky and controversial, as the checks will be removed inReleasebuilds. I say we use with care, but feels acceptable in this situation.In my rudimentary stress test, I can see a noticeable increase in throughput (5-10%).
Lastly, removing the
if+exception scheme, some tests which relied onREQUIRE_THROWSare now failing. It remains to be discussed what to do with them.