From mboxrd@z Thu Jan 1 00:00:00 1970 From: Pavel Machek Subject: Re: irqs_disabled() vs ACPI interpreter vs suspend Date: Thu, 18 Dec 2008 17:55:25 +0100 Message-ID: <20081218165525.GE16115@elf.ucw.cz> References: <20081125110508.GA1943@elf.ucw.cz> <200812170040.07977.rjw@sisk.pl> <200812181732.18023.rjw@sisk.pl> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from atrey.karlin.mff.cuni.cz ([195.113.26.193]:38055 "EHLO atrey.karlin.mff.cuni.cz" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751948AbYLRQxg (ORCPT ); Thu, 18 Dec 2008 11:53:36 -0500 Content-Disposition: inline In-Reply-To: <200812181732.18023.rjw@sisk.pl> Sender: linux-acpi-owner@vger.kernel.org List-Id: linux-acpi@vger.kernel.org To: "Rafael J. Wysocki" Cc: Len Brown , Andrew Morton , Linux Kernel Mailing List , linux-acpi@vger.kernel.org, Johannes Berg Hi! > > So it looks like we will indeed need something like the > > patch to transform ACPI's use of GFP_KERNEL > > to GFP_ATOMIC across late suspend > > and early resume; to avoid warnings from > > _GTS, _BFS, and irqrouter_resume use of kmalloc. > > OK, so there are two possibilities, IMO. > > Either we switch that in the suspend callbacks like in my patch #1, or we > can add a bool variable that will be 'true' is the system is during Do you mean acpi_gfp_flags? I think that's ugly, because not only gfp_flags need change during late suspend (you may not sleep, for example). I'd prefer boolean. Pavel -- (english) http://www.livejournal.com/~pavelmachek (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html