fix: recv_fifo for asyncfl #449
Merged
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
recv_fifo() takes first_k as one of its arguments to determine the first_k number of messages are received and returned. To implement the functionality, asynchronous call _streamer_for_recv_fifo() was implemented. However the function has a bug that is triggered in asyncfl. The function adds the first_k messages into an rx queue and re-adds the remaining messages to the head of each corresponding end id's rx queue. The remaining messages then processed in the next recv_fifo call. However, since _streamer_for_recv_fifo() is an asynchronous call, there is no guarantee that the messages saved in the previous call will be processed first. This causes some trainers blocked and made them fail to participate future rounds. The issue is fixed as follows. Since recv_fifo() always returns first_k number of messages from the rx queue, there is no harm even if the remaining messages are added to the queue. By removing all the code to handle the remaining code, the code becomes now clean and easy to understand.
Codecov Report
❗ Your organization needs to install the Codecov GitHub app to enable full functionality. @@ Coverage Diff @@
## main #449 +/- ##
=======================================
Coverage 14.82% 14.82%
=======================================
Files 48 48
Lines 2827 2827
=======================================
Hits 419 419
Misses 2379 2379
Partials 29 29 |
dhruvsgarg
pushed a commit
to dhruvsgarg/flame
that referenced
this pull request
Oct 18, 2024
recv_fifo() takes first_k as one of its arguments to determine the first_k number of messages are received and returned. To implement the functionality, asynchronous call _streamer_for_recv_fifo() was implemented. However the function has a bug that is triggered in asyncfl. The function adds the first_k messages into an rx queue and re-adds the remaining messages to the head of each corresponding end id's rx queue. The remaining messages then processed in the next recv_fifo call. However, since _streamer_for_recv_fifo() is an asynchronous call, there is no guarantee that the messages saved in the previous call will be processed first. This causes some trainers blocked and made them fail to participate future rounds. The issue is fixed as follows. Since recv_fifo() always returns first_k number of messages from the rx queue, there is no harm even if the remaining messages are added to the queue. By removing all the code to handle the remaining code, the code becomes now clean and easy to understand.
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.
Description
recv_fifo() takes first_k as one of its arguments to determine the first_k number of messages are received and returned.
To implement the functionality, asynchronous call
_streamer_for_recv_fifo() was implemented. However the function has a bug that is triggered in asyncfl. The function adds the first_k messages into an rx queue and re-adds the remaining messages to the head of each corresponding end id's rx queue. The remaining messages then processed in the next recv_fifo call. However, since _streamer_for_recv_fifo() is an asynchronous call, there is no guarantee that the messages saved in the previous call will be processed first. This causes some trainers blocked and made them fail to participate future rounds.
The issue is fixed as follows. Since recv_fifo() always returns first_k number of messages from the rx queue, there is no harm even if the remaining messages are added to the queue. By removing all the code to handle the remaining code, the code becomes now clean and easy to understand.
Type of Change
Checklist