[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <063D6719AE5E284EB5DD2968C1650D6D1CCC00C7@AcuExch.aculab.com>
Date: Fri, 8 Jan 2016 10:33:11 +0000
From: David Laight <David.Laight@...LAB.COM>
To: 'Hans Westgaard Ry' <hans.westgaard.ry@...cle.com>,
"David S. Miller" <davem@...emloft.net>
CC: Alexey Kuznetsov <kuznet@....inr.ac.ru>,
James Morris <jmorris@...ei.org>,
Hideaki YOSHIFUJI <yoshfuji@...ux-ipv6.org>,
Patrick McHardy <kaber@...sh.net>,
Alexei Starovoitov <ast@...mgrid.com>,
Jiri Pirko <jiri@...lanox.com>,
Eric Dumazet <edumazet@...gle.com>,
Daniel Borkmann <daniel@...earbox.net>,
Nicolas Dichtel <nicolas.dichtel@...nd.com>,
"Eric W. Biederman " <ebiederm@...ssion.com>,
Salam Noureddine <noureddine@...sta.com>,
Jarod Wilson <jarod@...hat.com>,
Toshiaki Makita <makita.toshiaki@....ntt.co.jp>,
Julian Anastasov <ja@....bg>,
Ying Xue <ying.xue@...driver.com>,
Craig Gallek <kraig@...gle.com>,
Mel Gorman <mgorman@...hsingularity.net>,
"\" hannes\"@stressinduktion.org " <hannes@...essinduktion.org>,
Edward Jee <edjee@...gle.com>,
Julia Lawall <julia.lawall@...6.fr>,
"netdev@...r.kernel.org" <netdev@...r.kernel.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
Haakon Bugge <haakon.bugge@...cle.com>,
Knut Omang <knut.omang@...cle.com>,
Wei Lin Guay <wei.lin.guay@...cle.com>,
Santosh Shilimkar <santosh.shilimkar@...cle.com>,
Yuval Shaia <yuval.shaia@...cle.com>
Subject: RE: [PATCH] net: add per device sg_max_frags for skb
From: Hans Westgaard
> Sent: 08 January 2016 09:56
...
> >> The patch allows the device to limit the maximum number fragments used
> >> in one skb.
> >
> > This doesn't seem to me to be the correct way to fix this.
> > Anything that adds an extra fragment (in this case IPoIB) should allow
> > for the skb already having the maximum number of fragments.
> > Fully linearising the skb is overkill, but I think the first fragment
> > can be added to the linear part of the skb.
> >
> > David
> >
> >
> When IpoIB handles a skb-request it converts fragments to SGEs to
> be handled by a HCA.
> The problem arises when the HCA have a limited number of SGEs less than
> MAX_SKB_FRAGS.
> (it gets a little worse since IPoIB need to yet another segment)
> I have not found any easy way of fixing this with currenct codebase.
I think one of the xen ethernet interfaces had a similar problem.
Just reduce the number of fragments by copying two (or more) of them
into a single fragment.
In effect, anything that reduces the number of fragments will do a copy.
David
Powered by blists - more mailing lists