From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1759523AbZCYMKQ (ORCPT ); Wed, 25 Mar 2009 08:10:16 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1761271AbZCYMJb (ORCPT ); Wed, 25 Mar 2009 08:09:31 -0400 Received: from rhun.apana.org.au ([64.62.148.172]:36612 "EHLO arnor.apana.org.au" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1761270AbZCYMJa (ORCPT ); Wed, 25 Mar 2009 08:09:30 -0400 Date: Wed, 25 Mar 2009 20:08:40 +0800 From: Herbert Xu To: Ingo Molnar Cc: David Miller , r.schwebel@pengutronix.de, torvalds@linux-foundation.org, blaschka@linux.vnet.ibm.com, tglx@linutronix.de, a.p.zijlstra@chello.nl, linux-kernel@vger.kernel.org, kernel@pengutronix.de Subject: Re: Revert "gro: Fix legacy path napi_complete crash", Message-ID: <20090325120840.GA6351@gondor.apana.org.au> References: <20090324150928.GB30224@gondor.apana.org.au> <20090324.143622.186562202.davem@davemloft.net> <20090325002303.GA2219@gondor.apana.org.au> <20090324.191134.05205089.davem@davemloft.net> <20090325073349.GF25833@elte.hu> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20090325073349.GF25833@elte.hu> User-Agent: Mutt/1.5.18 (2008-05-17) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, Mar 25, 2009 at 08:33:49AM +0100, Ingo Molnar wrote: > > So if there's no definitive resolution for the real cause of the > bug, the plain revert looks like an acceptable interim choice for > .29.1 - at least as far as my systems go. OK, let's just do the revert and disable GRO for the legacy path. This should be the safest option for 2.6.29. GRO: Disable GRO on legacy netif_rx path When I fixed the GRO crash in the legacy receive path I used napi_complete to replace __napi_complete. Unfortunately they're not the same when NETPOLL is enabled, which may result in us not calling __napi_complete at all. What's more, we really do need to keep the __napi_complete call within the IRQ-off section since in theory an IRQ can occur in between and fill up the backlog to the maximum, causing us to lock up. Since we can't seem to find a fix that works properly right now, this patch reverts all the GRO support from the netif_rx path. Signed-off-by: Herbert Xu diff --git a/net/core/dev.c b/net/core/dev.c index e3fe5c7..e438f54 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2588,18 +2588,15 @@ static int process_backlog(struct napi_struct *napi, int quota) local_irq_disable(); skb = __skb_dequeue(&queue->input_pkt_queue); if (!skb) { + __napi_complete(napi); local_irq_enable(); - napi_complete(napi); - goto out; + break; } local_irq_enable(); - napi_gro_receive(napi, skb); + netif_receive_skb(skb); } while (++work < quota && jiffies == start_time); - napi_gro_flush(napi); - -out: return work; } Thanks, -- Visit Openswan at http://www.openswan.org/ Email: Herbert Xu ~{PmV>HI~} Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt