Attendees
- peterlei
- rrs@
- rscheff@
- thj@
- tuexen@
Ongoing Short Term Projects
PR 263445 and PR 264257: Both seem to have the same root cause. tuexen@ and rscheff@ are trying to figure out what that root cause is now looking at BBLog files. tuexen@ will improve BBLog for the base stack to try to find the root cause. If the root cause is not found before 14.0 is branched or 13.2 is branched, a workaround will be implemented.
What is the status of D16851? thj@ works on it and you are waiting for feedback from bz@.
rscheff@ is improving AccECN code: D36303.
Ongoing Longer Term Projects
glebius@ 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@. Reviews: D36397, D36398, D36399, D36400. tuexen@ will implement some packetdrill tests.
D23230 is implementing ECN++.
D30043 is a small whitespace cleanup.
D28822 improves PRR, needs some cleanup.
glebius@ wants feedback on D35199. bz@ isn't fine with it...
glebius@: Small improvement to new connection establishment D36488.
glebius@: IP reassembly review D36275. General question of optimization towards what: CPU, memory or robust code?
D36494 fixes a bug related to timer based retransmissions for RACK. tuexen@ will look at it.
Switch default CC from NewReno and CUBIC for main and, if no substantial bugs show up, the upcoming 14.0. rscheff@ will come up with a patch.
netstat does not support reading connection information from a core file anymore. tuexen@ will document that in the man-page.
tuexen@ want to write a tool which dumps the BBLog information of a TCP endpoint from a life system or a core. Using kvm_read() has drawbacks as glebius@ pointed out, using a python kgdb script may be an alternative. tuexen@ will think about it.
Next Meeting
22 September 2022 @ 1500 UTC