* [MPTCP] Re: [PATCH] mptcp: attempt skb coalescing when moving skbs to mptcp rx queue
@ 2020-05-25 17:08 Paolo Abeni
0 siblings, 0 replies; 2+ messages in thread
From: Paolo Abeni @ 2020-05-25 17:08 UTC (permalink / raw)
To: mptcp
[-- Attachment #1: Type: text/plain, Size: 757 bytes --]
On Mon, 2020-05-25 at 18:02 +0200, Florian Westphal wrote:
> We can try to coalesce the skbs we take from the subflows rx queue
> with the tail of the mptcp rx queue.
>
> If successful, the skb head can be discarded early.
>
> We can also free the skb extensions, we do not access them after
> this so no need to hold on to them.
>
> Only caveat: do not built too fat skbs. They are only free'd once
> userspace has read all data.
To share IRC thoughts here:
It looks like TCP allows skb to grow above GSO_MAX_SIZE and frees them
only when user-space read all the data. Could we relax this constraint,
too?
skb_try_coalesce() will still somehow bound the skb size to
MAX_SKB_FRAGS * <something [PAGE_SIZE ?!?]>
Cheers,
Paolo
^ permalink raw reply [flat|nested] 2+ messages in thread
* [MPTCP] Re: [PATCH] mptcp: attempt skb coalescing when moving skbs to mptcp rx queue
@ 2020-05-25 18:10 Florian Westphal
0 siblings, 0 replies; 2+ messages in thread
From: Florian Westphal @ 2020-05-25 18:10 UTC (permalink / raw)
To: mptcp
[-- Attachment #1: Type: text/plain, Size: 831 bytes --]
Paolo Abeni <pabeni(a)redhat.com> wrote:
> On Mon, 2020-05-25 at 18:02 +0200, Florian Westphal wrote:
> > We can try to coalesce the skbs we take from the subflows rx queue
> > with the tail of the mptcp rx queue.
> >
> > If successful, the skb head can be discarded early.
> >
> > We can also free the skb extensions, we do not access them after
> > this so no need to hold on to them.
> >
> > Only caveat: do not built too fat skbs. They are only free'd once
> > userspace has read all data.
>
> To share IRC thoughts here:
>
> It looks like TCP allows skb to grow above GSO_MAX_SIZE and frees them
> only when user-space read all the data. Could we relax this constraint,
> too?
Yes, but I saw skbs of up ~320kb Truesize, which i considered too
much. I will remove the comment and the contraint in v2.
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2020-05-25 18:10 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-05-25 17:08 [MPTCP] Re: [PATCH] mptcp: attempt skb coalescing when moving skbs to mptcp rx queue Paolo Abeni
2020-05-25 18:10 Florian Westphal
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.