From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1764039AbZFLKPX (ORCPT ); Fri, 12 Jun 2009 06:15:23 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753501AbZFLKPL (ORCPT ); Fri, 12 Jun 2009 06:15:11 -0400 Received: from cantor2.suse.de ([195.135.220.15]:42334 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753180AbZFLKPK (ORCPT ); Fri, 12 Jun 2009 06:15:10 -0400 Date: Fri, 12 Jun 2009 12:15:11 +0200 From: Nick Piggin To: Pekka Enberg Cc: Ingo Molnar , linux-mm@kvack.org, linux-kernel@vger.kernel.org, benh@kernel.crashing.org, akpm@linux-foundation.org, cl@linux-foundation.org, torvalds@linux-foundation.org Subject: Re: [PATCH v2] slab,slub: ignore __GFP_WAIT if we're booting or suspending Message-ID: <20090612101511.GC13607@wotan.suse.de> References: <20090612091002.GA32052@elte.hu> <84144f020906120249y20c32d47y5615a32b3c9950df@mail.gmail.com> <20090612100756.GA25185@elte.hu> <84144f020906120311x7c7dd628s82e3ca9a840f9890@mail.gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <84144f020906120311x7c7dd628s82e3ca9a840f9890@mail.gmail.com> User-Agent: Mutt/1.5.9i Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Fri, Jun 12, 2009 at 01:11:52PM +0300, Pekka Enberg wrote: > Hi Ingo, > > On Fri, Jun 12, 2009 at 1:07 PM, Ingo Molnar wrote: > > IMHO such invisible side-channels modifying the semantics of GFP > > flags is a bit dubious. > > > > We could do GFP_INIT or GFP_BOOT. These can imply other useful > > modifiers as well: panic-on-failure for example. (this would clean > > up a fair amount of init code that currently checks for an panics on > > allocation failure.) > > OK, but that means we need to fix up every single caller. I'm fine > with that but Ben is not. As I am unable to test powerpc here, I am > inclined to just merge Ben's patch as "obviously correct". I agree with Ingo though that exposing it as a gfp modifier is not so good. I just like the implementation to mask off GFP_WAIT better, and also prefer not to test system state, but have someone just call into slab to tell it not to unconditionally enable interrupts. > That does not mean we can't introduce GFP_BOOT later on if we want to. Hmm? Yes, with sufficient warnings in place, I don't think it should be too error prone to clean up remaining code over the course of a few releases.