From: phil@nwl.cc (Phil Sutter) Date: Fri, 6 May 2011 18:12:09 +0200 Subject: ARM, AF_PACKET: caching problems on Marvell Kirkwood In-Reply-To: <1304607362.3032.84.camel@edumazet-laptop> References: <20110408130643.GA8730@orbit.nwl.cc> <20110505141107.GC30443@orbit.nwl.cc> <1304607362.3032.84.camel@edumazet-laptop> Message-ID: <20110506161209.GB20777@orbit.nwl.cc> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org Hi, On Thu, May 05, 2011 at 04:56:02PM +0200, Eric Dumazet wrote: > I assume you use latest linux-2.6 or net-next-2.6 ? Well, initially we noticed the problem on 2.6.34.7, but I verified it against both 2.6.37 and linux-2.6 from three days ago. > Could you try to force vmalloc() use ? > > diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c > index b5362e9..0b5a89c 100644 > --- a/net/packet/af_packet.c > +++ b/net/packet/af_packet.c > @@ -2383,7 +2383,7 @@ static inline char *alloc_one_pg_vec_page(unsigned long order) > gfp_t gfp_flags = GFP_KERNEL | __GFP_COMP | > __GFP_ZERO | __GFP_NOWARN | __GFP_NORETRY; > > - buffer = (char *) __get_free_pages(gfp_flags, order); > + buffer = NULL; > > if (buffer) > return buffer; Thanks for the hint. I tried that, but the problem persists. Greetings, Phil