IXWebSocketTransport: Avoid bloating _rxbuf#550
Merged
bsergean merged 1 commit intomachinezone:masterfrom May 14, 2025
Merged
Conversation
Buffering too much data into _rxbuf results in dispatch() processing to run into a performance issue due to calling _rxbuf.erase() for each frame. Concretely, if _rxbuf grows large due to a client sending frames very fast, each processing in dispatch() results in moving the remaining buffer to the front. Instead of restructuring to avoid .erase(), this patch limits the maximum size of _rxbuf to kChunkSize to alleviate the O^2 erase() overhead. It also has the side-effect of keeping the received data in the OS's TCP stack, building up back pressure to the client earlier if the server code can't keep up. Fixes machinezone#429
Contributor
Author
|
With this patch, the reproducer code from #429 completes in 3 to 4 seconds. Without the patch, I've observed it to take up to 60 seconds locally, but depends on how the buffer builds up over time. I do think the 32k kChunkSize that is read by default is a bit large, but would rather not change too many things. |
Contributor
Author
|
The new server-side flamegraph shows that the
|
Contributor
|
thanks ! |
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Buffering too much data into _rxbuf results in dispatch() processing to run into a performance issue due to calling _rxbuf.erase() for each frame.
Concretely, if _rxbuf grows large due to a client sending frames very fast, each processing in dispatch() results in moving the remaining buffer to the front.
Instead of restructuring to avoid .erase(), this patch limits the maximum size of _rxbuf to kChunkSize to alleviate the O^2 erase() overhead. It also has the side-effect of keeping the received data in the OS's TCP stack, building up back pressure to the client earlier if the server code can't keep up.
Fixes #429