W3C home > Mailing lists > Public > public-webrtc-logs@w3.org > January 2019

Re: [webrtc-pc] (How) does SCTP handle CPU-bound congestion on JavaScript thread? (#2086)

From: Sergio Garcia Murillo via GitHub <sysbot+gh@w3.org>
Date: Sat, 26 Jan 2019 11:43:11 +0000
To: public-webrtc-logs@w3.org
Message-ID: <issue_comment.created-457824581-1548502990-sysbot+gh@w3.org>
the issue is not about preventing the receive buffers to fill up,  but to
prevent the datachannel stack to enque events in the js loop faster than
they can be dispatched to the js app.

If the a_rcwnd is not freed until the event is actually delivered to the js
app, it will make the sctp buffer to fill up and the sender will not push
more data. Once the data is delivered, the new a_rcwnd will be transmitted
back to the sender and it will start sending data again.

That way the server will pace its sending to the amount of data that can be
consumed by the js app.

If it is from a single datachannel or multiple ones is an orthogonal issue.

El sáb., 26 ene. 2019 12:11, Lennart Grahl <notifications@github.com>
escribió:

> Please explain how it would prevent the receive buffer from filling up
> when using multiple channels.
>
> —
> You are receiving this because you were mentioned.
> Reply to this email directly, view it on GitHub
> <https://github.com/w3c/webrtc-pc/issues/2086#issuecomment-457822769>, or mute
> the thread
> <https://github.com/notifications/unsubscribe-auth/ABBW84xCGKnJ0qtPevgSonuYXzYAIR6hks5vHDfogaJpZM4aS7_J>
> .
>


-- 
GitHub Notification of comment by murillo128
Please view or discuss this issue at https://github.com/w3c/webrtc-pc/issues/2086#issuecomment-457824581 using your GitHub account
Received on Saturday, 26 January 2019 11:43:13 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 4 June 2019 15:32:54 UTC