Re: [PATCH v2 mptcp-next 2/7] mptcp: move the stale logic out of retrans scheduler
Matthieu Baerts <[email protected]>
| Newsgroups | dev.linux.lists.mptcp |
|---|---|
| Organization | NGI0 Core |
| Message-ID | <[email protected]> |
On 06/08/2026 18:41, Paolo Abeni wrote: > On 8/6/26 4:45 PM, Paolo Abeni wrote: >> This allow separating the stale logic invocation and the retrans >> scheduler, and will simplify the next patch. >> >> It's also a cleaner design as the retrans scheduler has currently >> too many side effects. As a possible downside, the retrans work will >> now traverse the subflows list additional times; that does not matter >> much, as this is slowpath. >> >> While at it, pick more accurate names for the involved helpers and >> explicitly note that the per subflow stale data is under msk socket >> lock protection. >> >> The scheduler and the stale logic may observe different subflow >> statues, as no subflow lock is acquired. This is intentional and not >> harmful, worst case leading to slower retransmissions. >> >> Signed-off-by: Paolo Abeni <[email protected]> >> --- >> v1 -> v2: >> - fix stale lockless access >> >> Note: BPF scheduler should be rebase to use the renamed helper > Additional note: I think the rename is a good thing, because the > function additionally changed the argument list. Fine by me: these helpers are only in our export branch anyway. > @Mat: please LMK if you are ok with squash-to patches for later > self-tests/bpf things to be adjusted. Yes, of course. Because they are now our of the scheduler API, I guess they can be removed. So Geliang's series can be used on top of yours: https://lore.kernel.org/[email protected] Cheers, Matt -- Sponsored by the NGI0 Core fund.