Attendees: glebius@, hselasky@, rrs@, rscheff@, tuexen@
Gleb is suggesting to remove the compressed TCP TIMEWAIT state. The memory savings are limited, memory is not that critical anymore and it would simplify the code and allow further optimizations of the code. glebius@ will reach out to people if there are some objections. It was also discussed which information would be need to reduce the time a TCP connections stays in TIMEWAIT state. One thing he will implement is a counter how many compressed states where used at all. Discussion has been started on net@.
Some usages of ticks in the TCP stack might result in strange behaviour. Using ticks should be avoided…
ECN implementation has been improved. The code is now contained in one file
tcp_ecn.c.
D34161 is the first step. rrs@ and tuexen@ will look at it.
D21011 will be reviewed by rrs@ and tuexen@.
D33249 will be reviewed by tuexen@.
-
There are some warnings when compiling TCP related files. rrs@ will look at them.
glebius@ explains some future work related to protocol specific socket buffer. Right now, sbused() (how much data is in the buffer) and sbavail() (how must data is be ready to be taken) should be used instead of explicitly looking at variables part of the socket buffers.
Next meeting: 24 February 2022 @ 1500 UTC