From mboxrd@z Thu Jan 1 00:00:00 1970 From: Pavel Machek Subject: Re: acpi_os_allocate(GFP_KERNEL) (was Re: acpi_evaluate_integer broken by design) Date: Thu, 18 Dec 2008 17:48:01 +0100 Message-ID: <20081218164801.GC16115@elf.ucw.cz> References: <20081125110508.GA1943@elf.ucw.cz> <20081215214140.9031d4c8.akpm@linux-foundation.org> <200812161634.25838.rjw@sisk.pl> <200812170040.07977.rjw@sisk.pl> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: Sender: linux-kernel-owner@vger.kernel.org To: Len Brown Cc: "Rafael J. Wysocki" , Andrew Morton , Linux Kernel Mailing List , linux-acpi@vger.kernel.org, Johannes Berg List-Id: linux-acpi@vger.kernel.org On Thu 2008-12-18 01:07:08, Len Brown wrote: > > > -- Len Brown, Intel Open Source Technology Center > > > In fact, we should be using GFP_ATOMIC already while devices are being > > suspended, so we can switch from GFP_KERNEL to GFP_ATOMIC even before > > interrupts are enabled. > > true. it makes me wonder if tinkering with ACPI > is a bad idea -- for it will address only one user of kmalloc. Other users should already know that they are called while suspending device, and behave accordingly (use right flags to kmalloc, etc.) If not, that's a bug... Pavel -- (english) http://www.livejournal.com/~pavelmachek (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html