Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1161253AbWAMH4L (ORCPT ); Fri, 13 Jan 2006 02:56:11 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S932706AbWAMH4L (ORCPT ); Fri, 13 Jan 2006 02:56:11 -0500 Received: from smtp.osdl.org ([65.172.181.4]:62397 "EHLO smtp.osdl.org") by vger.kernel.org with ESMTP id S932678AbWAMH4K (ORCPT ); Fri, 13 Jan 2006 02:56:10 -0500 Date: Thu, 12 Jan 2006 23:55:48 -0800 From: Andrew Morton To: Pekka J Enberg Cc: linux-kernel@vger.kernel.org, reiserfs-dev@namesys.com Subject: Re: [PATCH] reiserfs: use __GFP_NOFAIL instead of yield and retry loop for allocation Message-Id: <20060112235548.0e1e4343.akpm@osdl.org> In-Reply-To: References: <20060112234238.01979912.akpm@osdl.org> X-Mailer: Sylpheed version 1.0.4 (GTK+ 1.2.10; i386-redhat-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1255 Lines: 35 Pekka J Enberg wrote: > > Hi, > > Pekka J Enberg wrote: > > > > > > - retry: > > > - jl = kzalloc(sizeof(struct reiserfs_journal_list), GFP_NOFS); > > > - if (!jl) { > > > - yield(); > > > - goto retry; > > > - } > > > + jl = kzalloc(sizeof(struct reiserfs_journal_list), > > > + GFP_NOFS | __GFP_NOFAIL); > > On Thu, 12 Jan 2006, Andrew Morton wrote: > > yup, that's what __GFP_NOFAIL is for: to consolidate and identify all those > > places which want to lock up when we're short of memory... They all need > > fixing, really. > > Out of curiosity, are there any potential problems with combining GFP_NOFS > and __GFP_NOFAIL? Can we really guarantee to give out memory if we're not > allowed to page out? > GFP_NOFS increases the risk (relative to GFP_KERNEL) because page reclaim can do less things than GFP_KERNEL to free memory. GFP_NOFS allocations can still perform swapspace writes, however. GFP_NOIO cannot even do that. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/