Re: [PATCH v2 0/6] tcp: Fixes for issues uncovered by tests with 6.17-rc1 kernels

[Stefano Brivio <sbrivio@redhat.com>]

On Wed, 20 Aug 2025 21:09:59 +0200
Stefano Brivio <sbrivio@redhat.com> wrote:

> On Wed, 20 Aug 2025 18:51:31 +0200
> Stefano Brivio <sbrivio@redhat.com> wrote:
> 
> > Starting from Linux kernel commit 1d2fbaad7cd8 ("tcp: stronger
> > sk_rcvbuf checks"), window limits are enforced more aggressively with
> > a bigger amount of zero-window updates compared to what happened with
> > e2142825c120 ("net: tcp: send zero-window ACK when no memory") alone,
> > and occasional duplicate ACKs can now be seen also for local transfers
> > with default (208 KiB) socket buffer sizes.
> > 
> > Paul reports that, with 6.17-rc1-ish kernels, Podman tests for the
> > pasta integration occasionally fail on the "TCP/IPv4 large transfer,
> > tap" case.
> > 
> > While playing with a reproducer that seems to be matching those
> > failures:
> > 
> >   while true; do ./pasta --trace -l /tmp/pasta.log -p /tmp/pasta.pcap --config-net -t 5555 -- socat TCP-LISTEN:5555 OPEN:/tmp/large.rcv,trunc & (sleep 0.3; socat -T2 OPEN:large.bin TCP:88.198.0.164:5555; ); wait; diff large.bin /tmp/large.rcv || break; done
> > 
> > and a kernel including that commit, I hit a few different failures,
> > that should be fixed by this series.
> > 
> > Paul tested v1 of this series and found an additional failure
> > (transfer timeout), which I could reproduce with a slightly different
> > command:
> > 
> >   while true; do ./pasta --trace -l /tmp/pasta.log -p /tmp/pasta.pcap --config-net -t 5555 -- socat TCP-LISTEN:5555 EXEC:./write.sh & (sleep 0.3; socat -T2 OPEN:large.bin TCP:88.198.0.164:5555; ); wait; diff large.bin /tmp/large.rcv || break; done
> > 
> > where write.sh is simply:
> > 
> >   #!/bin/sh
> >   
> >   cat > /tmp/large.rcv
> > 
> > so that the connection is not half-closed starting from the beginning,
> > because socat can't make assumptions about the unidirectional nature
> > of the traffic. This should now be fixed as well by the new version of
> > patch 3/6.
> > 
> > 
> > v2: in 3/6, rewind sequence also if the zero-window update comes in
> >     the middle of a batch with non-zero window updates  
> 
> Hmpf, never mind, this broke (on the second run):
> 
>   TCP/IPv4: ns to host (via tap): big transfer
> 
> from the main tests (not Podman tests). I still need to look into it.

Somewhat embarrassingly, the issue came from this change in 6/6:

-		if (SEQ_GE(seq_offset, 0) && SEQ_LE(seq + len, seq_from_tap))
+		if (SEQ_GE(seq_offset, 0) && SEQ_LE(seq + dlen, seq_from_tap))

where dlen represents the same quantity as len, but is signed, and that
throws SEQ_LE off.

I'm posting v3 in a bit, with an additional patch fixing (in some
sense) sequence comparison macros.

-- 
Stefano