[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1309958184.12098.111.camel@lsx>
Date: Wed, 06 Jul 2011 15:16:24 +0200
From: Thomas Graf <tgraf@...hat.com>
To: Neil Horman <nhorman@...driver.com>
Cc: Vladislav Yasevich <vladislav.yasevich@...com>,
netdev@...r.kernel.org, davem@...emloft.net,
Wei Yongjun <yjwei@...fujitsu.com>,
Sridhar Samudrala <sri@...ibm.com>, linux-sctp@...r.kernel.org
Subject: Re: [PATCHv2] sctp: Enforce retransmission limit during shutdown
On Wed, 2011-07-06 at 08:15 -0400, Neil Horman wrote:
> On Mon, Jul 04, 2011 at 09:50:19AM -0400, Thomas Graf wrote:
> > --- a/net/sctp/sm_statefuns.c
> > +++ b/net/sctp/sm_statefuns.c
> > @@ -5154,7 +5154,7 @@ sctp_disposition_t sctp_sf_do_9_2_start_shutdown(
> > * The sender of the SHUTDOWN MAY also start an overall guard timer
> > * 'T5-shutdown-guard' to bound the overall time for shutdown sequence.
> > */
> > - sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_START,
> > + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_RESTART,
> > SCTP_TO(SCTP_EVENT_TIMEOUT_T5_SHUTDOWN_GUARD));
> >
> How come you're modifying this chunk to use TIMER_RESTART rather than
> TIMER_START? start shutdown is where the t5 timer is actually started, isn't it?
Since we also start the timer in SHUTDOWN_PENDING now if we hit
the retransmission limit the timer may be running already and
needs to be restarted (at least in theory).
In reality the timer should be stopped though, we can only go
from SHUTDOWN_PENDING into SHUTDOWN by actually SACKing bytes which
will delete the timer. This may change though and I did not want
this to bite us later on.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists