From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932214AbYDBSo5 (ORCPT ); Wed, 2 Apr 2008 14:44:57 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1755577AbYDBSon (ORCPT ); Wed, 2 Apr 2008 14:44:43 -0400 Received: from 74-93-104-97-Washington.hfc.comcastbusiness.net ([74.93.104.97]:50452 "EHLO sunset.davemloft.net" rhost-flags-OK-FAIL-OK-OK) by vger.kernel.org with ESMTP id S1755190AbYDBSol (ORCPT ); Wed, 2 Apr 2008 14:44:41 -0400 Date: Wed, 02 Apr 2008 11:44:41 -0700 (PDT) Message-Id: <20080402.114441.78726196.davem@davemloft.net> To: kosaki.motohiro@jp.fujitsu.com Cc: akpm@linux-foundation.org, csnook@redhat.com, davej@codemonkey.org.uk, nickpiggin@yahoo.com.au, linux-kernel@vger.kernel.org, netdev@vger.kernel.org, mchan@broadcom.com Subject: Re: GFP_ATOMIC page allocation failures. From: David Miller In-Reply-To: <20080402195428.95A3.KOSAKI.MOTOHIRO@jp.fujitsu.com> References: <20080402.012406.246863209.davem@davemloft.net> <20080402175857.959A.KOSAKI.MOTOHIRO@jp.fujitsu.com> <20080402195428.95A3.KOSAKI.MOTOHIRO@jp.fujitsu.com> X-Mailer: Mew version 5.2 on Emacs 22.1 / Mule 5.0 (SAKAKI) Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org From: KOSAKI Motohiro Date: Wed, 02 Apr 2008 19:56:05 +0900 > > > --------------------------------------------------------------------- > > We get rather a lot of reports of page allocation warnings coming out of > > tg3. But this driver is know to handle them properly so let's suppress > > them. > > BTW: > other person's tg3 allocation failure bug report. > > http://bugzilla.kernel.org/show_bug.cgi?id=9771 Yes, I know about it. It will occur for any device which must use linear RX buffers and supports >PAGE_SIZE MTUs. But modern chips are going to support segmentation of receive packets into multiple individual pages and therefore not have this order>0 page allocation requirement. That is my point.