Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757928Ab1EaFFp (ORCPT ); Tue, 31 May 2011 01:05:45 -0400 Received: from smtp-out.google.com ([74.125.121.67]:60683 "EHLO smtp-out.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757767Ab1EaFFo (ORCPT ); Tue, 31 May 2011 01:05:44 -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=RCn3T6pwsreg9XWzxII7F7IW/WOGI4xTQZjPbPHzMjv1OmAL15aud1opKbx5e+YkEL k6XdzkEYpceuM2fhAinw== Date: Mon, 30 May 2011 22:05:38 -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 , linux-kernel , Eric Miao , Samuel Ortiz Subject: Re: IrDA driver fails on PXA255 In-Reply-To: <20110529215656.GA5576@n2100.arm.linux.org.uk> Message-ID: References: <20110528205701.GA1788@doriath.ww600.siemens.net> <20110529215656.GA5576@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: 1598 Lines: 36 On Sun, 29 May 2011, Russell King - ARM Linux wrote: > > That would return memory that is not guaranteed to be within the first > > 16MB of address space, so a GFP_DMA allocation would succeed with memory > > not from ZONE_DMA. > > Err, no. GFP_DMA returns memory in a zone which the platform has setup. > There's nothing specific about it being "16MB" or any other size; the > arch can chose what size that is. > Sorry, was talking from the x86 perspective, which probably doesn't make much sense since we're talking about an arm driver :) > > That's an invalid configuration, so users, including > > you, should at least edit their .config by hand to enable CONFIG_ZONE_DMA > > as a workaround. > > Again, no. This change has caused a load of previously working drivers > to suddenly start failing without _any_ explanation why or even warning > about the change. It needs to start off as a WARN_ON() so that stuff > can be fixed, and then changed to a hard error. > I haven't seen a "load" of error reports where this is causing an issue, maybe it is much more popular on arm? This also isn't a hard error, admins should be able to enable CONFIG_ZONE_DMA and rebuild so that the driver being loaded can get the type of memory it is requesting. Just putting a WARN_ON() doesn't provide any incentive to ever get this stuff fixed. -- 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/