prov/rxd: Move the receive repost/remove to earlier in the pkt handling #5183
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
|
I think this has the potential of dropping fewer packets under load, though batching the posting of receive buffers can be beneficial. Do we have any data on the impact of this change? |
aingerson
reviewed
Jul 23, 2019
The aim of changes is posting a new recive buffer right away, and don't wait for finishing the handle of a recive complition. We post a recieve buffer at once. The changes have caused the performance part: The results were run on alltoall test, and it shows us better results from 64Kb to 4Mb with comparing 'without the current changes'. But see a little regress from 1b to 64Kb, the diff is about ~0.08 usec (not bigger). Signed-off-by: Nikita Gusev <nikita.gusev@intel.com>
|
@aingerson Thanks for review. I've removed |
+1. With the existing batched posting of recvs, you could also set FI_MORE in rxd_ep_post_buf() so the base providers can optimize for that. |
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Outdated suggestions cannot be applied.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.
Issue:
The provided changes is posting a new recive buffer right away, and don't
wait for finishing the handle of a recive complition. We post a recieve
buffer at once.
Results:
The changes have caused the performance part results, which were run on
alltoall test, and it shows us better results from64Kb to 4Mbwith comparing'without the current changes'. But see a little regress from
1b to 64Kb,the diff is about
~0.08 usec(not bigger).Signed-off-by: Nikita Gusev nikita.gusev@intel.com