Martin Thomson

Results 118 issues of Martin Thomson

When we don't send for a little while, it might be a good idea to have the congestion window decay a little. Otherwise, we might start up sending at the...

p2

PRR is defined in [RFC 6937](https://tools.ietf.org/html/rfc6937). It helps even out the send rate during the period that the endpoint is in the recovery period. The way this works in QUIC...

task-medium
p2

When the client is sending 0-RTT, I *think* that it will start using the updated transport parameters from the server immediately after receiving them, rather than waiting for the handshake...

p2

`Http3ServerEvent::PriorityUpdate` contains two values: stream ID and priority. A server holds multiple connections, so this is not information that an application can act on. It looks like it should use...

p2

This probably needs a separate class: as a server you want to hold this globally (for all connections), but for a client you want to have this per-connection.

task-medium
server-side

### Describe the issue https://bib.ietf.org/search/?query=rfc8996&allow_format_fallback=True&bypass_cache=True is quite slow: > 2 database queries against 265188 items took 3.194 and 0.711 sec. https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8996.xml has timed out a few times for me. https://bib.ietf.org/get-one/by-docid/?docid=RFC+8996&doctype=IETF&query=rfc8996&query_format=docid_regex&page=1...

bug

With the new explicit path identifier, maybe I can suggest an improvement. Look at Victor's https://datatracker.ietf.org/doc/draft-vvv-quic-namespaces/ It strikes me that the trick he is using with the NS frame there...

no-action-needed

I couldn't find anything in the specification regarding the origin that a track is attributed to. I suspect that all browsers have settled on a model that is sensible, but...