Hi all,
Today's linux-next merge of the net-next tree got a conflict in:
net/ipv4/tcp_output.c
between commit:
a2815817ffa6 ("tcp: enable xmit timer fix by having TLP use time when RTO should fire")
from the net tree and commit:
bb4d991a28cc ("tcp: adjust tail loss probe timeout")
from the net-next tree.
I fixed it up (see below) and can carry the fix as necessary. This
is now fixed as far as linux-next is concerned, but any non trivial
conflicts should be mentioned to your upstream maintainer when your tree
is submitted for merging. You may also want to consider cooperating
with the maintainer of the conflicting tree to minimise any particularly
complex conflicts.
--
Cheers,
Stephen Rothwell
diff --cc net/ipv4/tcp_output.c
index 276406a83a37,d49bff51bdb7..000000000000
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@@ -2377,9 -2375,13 +2375,8 @@@ bool tcp_schedule_loss_probe(struct soc
{
struct inet_connection_sock *icsk = inet_csk(sk);
struct tcp_sock *tp = tcp_sk(sk);
- u32 rtt = usecs_to_jiffies(tp->srtt_us >> 3);
- u32 timeout, tlp_time_stamp, rto_time_stamp;
+ u32 timeout, rto_delta_us;
- /* No consecutive loss probes. */
- if (WARN_ON(icsk->icsk_pending == ICSK_TIME_LOSS_PROBE)) {
- tcp_rearm_rto(sk);
- return false;
- }
/* Don't do any loss probe on a Fast Open connection before 3WHS
* finishes.
*/
@@@ -2402,16 -2408,24 +2399,20 @@@
* for delayed ack when there's one outstanding packet. If no RTT
* sample is available then probe after TCP_TIMEOUT_INIT.
*/
- timeout = rtt << 1 ? : TCP_TIMEOUT_INIT;
- if (tp->packets_out == 1)
- timeout = max_t(u32, timeout,
- (rtt + (rtt >> 1) + TCP_DELACK_MAX));
- timeout = max_t(u32, timeout, msecs_to_jiffies(10));
+ if (tp->srtt_us) {
+ timeout = usecs_to_jiffies(tp->srtt_us >> 2);
+ if (tp->packets_out == 1)
+ timeout += TCP_RTO_MIN;
+ else
+ timeout += TCP_TIMEOUT_MIN;
+ } else {
+ timeout = TCP_TIMEOUT_INIT;
+ }
- /* If RTO is shorter, just schedule TLP in its place. */
- tlp_time_stamp = tcp_jiffies32 + timeout;
- rto_time_stamp = (u32)inet_csk(sk)->icsk_timeout;
- if ((s32)(tlp_time_stamp - rto_time_stamp) > 0) {
- s32 delta = rto_time_stamp - tcp_jiffies32;
- if (delta > 0)
- timeout = delta;
- }
+ /* If the RTO formula yields an earlier time, then use that time. */
+ rto_delta_us = tcp_rto_delta_us(sk); /* How far in future is RTO? */
+ if (rto_delta_us > 0)
+ timeout = min_t(u32, timeout, usecs_to_jiffies(rto_delta_us));
inet_csk_reset_xmit_timer(sk, ICSK_TIME_LOSS_PROBE, timeout,
TCP_RTO_MAX);
On Sun, Aug 6, 2017 at 10:01 PM, Stephen Rothwell <[email protected]> wrote:
> Hi all,
>
> Today's linux-next merge of the net-next tree got a conflict in:
>
> net/ipv4/tcp_output.c
>
> between commit:
>
> a2815817ffa6 ("tcp: enable xmit timer fix by having TLP use time when RTO should fire")
>
> from the net tree and commit:
>
> bb4d991a28cc ("tcp: adjust tail loss probe timeout")
>
> from the net-next tree.
>
> I fixed it up (see below) and can carry the fix as necessary. This
> is now fixed as far as linux-next is concerned, but any non trivial
> conflicts should be mentioned to your upstream maintainer when your tree
> is submitted for merging. You may also want to consider cooperating
> with the maintainer of the conflicting tree to minimise any particularly
> complex conflicts.
Sorry about that. Will try to follow that procedure in the future.
thanks,
neal
Hi Neal,
On Sun, 6 Aug 2017 22:21:43 -0400 Neal Cardwell <[email protected]> wrote:
>
> > I fixed it up (see below) and can carry the fix as necessary. This
> > is now fixed as far as linux-next is concerned, but any non trivial
> > conflicts should be mentioned to your upstream maintainer when your tree
> > is submitted for merging. You may also want to consider cooperating
> > with the maintainer of the conflicting tree to minimise any particularly
> > complex conflicts.
>
> Sorry about that. Will try to follow that procedure in the future.
The above is a generic statement I add to all these emails. It is
aimed more at the maintainers if the trees involved, no the developers
of patches. I don't think you need to do anything different in these
cases with the "net" and "net-next" tree. Dave Miller will fix up any
conflicts when he next merges the net tree into the net-next tree.
--
Cheers,
Stephen Rothwell