From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [PATCH] tcp: splice as many packets as possible at once Date: Thu, 15 Jan 2009 15:19:14 -0800 (PST) Message-ID: <20090115.151914.84641343.davem@davemloft.net> References: <20090114085308.GB4234@ff.dom.local> <20090114.012919.117682429.davem@davemloft.net> <20090115230331.GB1123@1wt.eu> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: jarkao2@gmail.com, herbert@gondor.apana.org.au, zbr@ioremap.net, dada1@cosmosbay.com, ben@zeus.com, mingo@elte.hu, linux-kernel@vger.kernel.org, netdev@vger.kernel.org, jens.axboe@oracle.com To: w@1wt.eu Return-path: Received: from 74-93-104-97-Washington.hfc.comcastbusiness.net ([74.93.104.97]:46123 "EHLO sunset.davemloft.net" rhost-flags-OK-FAIL-OK-OK) by vger.kernel.org with ESMTP id S1763906AbZAOXTO (ORCPT ); Thu, 15 Jan 2009 18:19:14 -0500 In-Reply-To: <20090115230331.GB1123@1wt.eu> Sender: netdev-owner@vger.kernel.org List-ID: From: Willy Tarreau Date: Fri, 16 Jan 2009 00:03:31 +0100 > please don't merge it as-is. I've just tried it and got an OOM > in __alloc_pages_internal after a few seconds of data transfer. > > I'm leaving the patch below for comments, maybe someone will spot > something ? Don't we need at least one kfree() somewhere to match > alloc_pages() ? What is needed is a put_page(). But that should be happening as a result of the splice callback.