Re: [PATCH/RFC] Re: recvmmsg() timeout behavior strangeness [RESEND]

From: "'Arnaldo Carvalho de Melo'" <acme@kernel.org>
To: David Laight <David.Laight@ACULAB.COM>
Cc: "Michael Kerrisk (man-pages)" <mtk.manpages@gmail.com>,
	lkml <linux-kernel@vger.kernel.org>,
	"linux-man@vger.kernel.org" <linux-man@vger.kernel.org>,
	netdev <netdev@vger.kernel.org>,
	"Ondrej Bílka" <neleai@seznam.cz>,
	"Caitlin Bestler" <caitlin.bestler@gmail.com>,
	"Neil Horman" <nhorman@tuxdriver.com>,
	"Elie De Brauwer" <eliedebrauwer@gmail.com>,
	"David Miller" <davem@davemloft.net>,
	"Steven Whitehouse" <steve@chygwyn.com>,
	"Rémi Denis-Courmont" <remi.denis-courmont@nokia.com>,
	"Paul Moore" <paul@paul-moore.com>,
	"Chris Friesen" <chris.friesen@windriver.com>
Subject: Re: [PATCH/RFC] Re: recvmmsg() timeout behavior strangeness [RESEND]
Date: Thu, 29 May 2014 11:17:05 -0300	[thread overview]
Message-ID: <20140529141705.GI2764@kernel.org> (raw)
In-Reply-To: <063D6719AE5E284EB5DD2968C1650D6D1724FC78@AcuExch.aculab.com>

Em Thu, May 29, 2014 at 02:06:04PM +0000, David Laight escreveu:
> From: 'Arnaldo Carvalho de Melo'
> ...
> > > I remember some discussions from an XNET standards meeting (I've forgotten
> > > exactly which errors on which calls were being discussed).
> > > My recollection is that you return success with a partial transfer
> > > count for ANY error that happens after some data has been transferred.
> > > The actual error will be returned when it happens again on the next
> > > system call - Note the AGAIN, not a saved error.

> > A saved error, for the right entity, in the recvmmsg case, that
> > basically is batching multiple recvmsg syscalls, doesn't sound like a
> > problem, i.e. the idea is to, as much as possible, mimic what multiple
> > recvmsg calls would do, but reduce its in/out kernel (and inside kernel
> > subsystems) overhead.

> > Perhaps we can have something in between, i.e. for things like EFAULT,
> > we should report straight away, effectively dropping whatever datagrams
> > successfully received in the current batch, do you agree?

> Not unreasonable - EFAULT shouldn't happen unless the application
> is buggy.

Ok.

> > For transient errors the existing mechanism, fixed so that only per
> > socket errors are saved for later, as today, could be kept?

> I don't think it is ever necessary to save an errno value for the
> next system call at all.
> Just process the next system call and see what happens.

> If the call returns with less than the maximum number of datagrams
> and with a non-zero timeout left - then the application can infer
> that it was terminated by an abnormal event of some kind.
> This might be a signal.

Then it could use getsockopt(SO_ERROR) perhaps? I.e. we don't return the
error on the next call, but we provide a way for the app to retrieve the
reason for the smaller than expected batch?

> I'm not sure if an icmp error on a connected datagram socket could
> generate a 'disconnect'. It might happen if the interface is being
> used for something like SCTP.
> In either case the next call will detect the error.

- Arnaldo