Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757599Ab1EaFBX (ORCPT ); Tue, 31 May 2011 01:01:23 -0400 Received: from smtp-out.google.com ([216.239.44.51]:33368 "EHLO smtp-out.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751102Ab1EaFBW (ORCPT ); Tue, 31 May 2011 01:01:22 -0400 DomainKey-Signature: a=rsa-sha1; c=nofws; d=google.com; s=beta; h=date:from:x-x-sender:to:cc:subject:in-reply-to:message-id :references:user-agent:mime-version:content-type; b=r5zOiU+W1cG0jKC/HzOMz0EDmkF+3SHJRkL7DsRvh2x2xRlko4QCTX1Srs+9EezHCW D8VwjQdBatba5WrSQXMQ== Date: Mon, 30 May 2011 22:01:14 -0700 (PDT) From: David Rientjes X-X-Sender: rientjes@chino.kir.corp.google.com To: Russell King - ARM Linux cc: Dmitry Eremin-Solenikov , Nicolas Pitre , linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org, Eric Miao , Samuel Ortiz Subject: Re: IrDA driver fails on PXA255 In-Reply-To: <20110529215805.GB5576@n2100.arm.linux.org.uk> Message-ID: References: <20110528205701.GA1788@doriath.ww600.siemens.net> <20110528234614.GV24876@n2100.arm.linux.org.uk> <20110529072516.GW24876@n2100.arm.linux.org.uk> <20110529215805.GB5576@n2100.arm.linux.org.uk> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-System-Of-Record: true Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1368 Lines: 28 On Sun, 29 May 2011, Russell King - ARM Linux wrote: > We might as well enable CONFIG_ZONE_DMA for everything if that's what > you're proposing, because it's not just this driver which will be affected. I'd certainly suggest at least defaulting it to on for arm if you're going to be using GFP_DMA in drivers. > And as soon as we do that, we completely lose the warnings that stuff > needs fixing. > That's because nothing needs fixing at that point, the page allocator is guaranteed to return lowmem if GFP_DMA is passed and CONFIG_ZONE_DMA is enabled, or NULL. Whether GFP_DMA is correct in the memory allocator is a different subject, but those types of audits can easily be done on the source code. In my opinion, we should be doing "select ZONE_DMA" on any Kconfig option that builds a driver that unconditionally uses GFP_DMA. The current behavior exists so that the admin reports the error here so it can get fixed up (either by finding that GFP_DMA is unnecessary, selective depending on the particular hardware, or modifying the Kconfig) and can workaround the problem by forcing CONFIG_ZONE_DMA. -- 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/