From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail203.messagelabs.com (mail203.messagelabs.com [216.82.254.243]) by kanga.kvack.org (Postfix) with ESMTP id 3AF236B0012 for ; Fri, 10 Jun 2011 18:01:24 -0400 (EDT) Received: from hpaq6.eem.corp.google.com (hpaq6.eem.corp.google.com [172.25.149.6]) by smtp-out.google.com with ESMTP id p5AM1LWe016651 for ; Fri, 10 Jun 2011 15:01:21 -0700 Received: from pve37 (pve37.prod.google.com [10.241.210.37]) by hpaq6.eem.corp.google.com with ESMTP id p5AM1IXI012616 (version=TLSv1/SSLv3 cipher=RC4-SHA bits=128 verify=NOT) for ; Fri, 10 Jun 2011 15:01:19 -0700 Received: by pve37 with SMTP id 37so1467767pve.7 for ; Fri, 10 Jun 2011 15:01:17 -0700 (PDT) Date: Fri, 10 Jun 2011 15:01:15 -0700 (PDT) From: David Rientjes Subject: Re: [PATCH] Make GFP_DMA allocations w/o ZONE_DMA emit a warning instead of failing In-Reply-To: <20110610185858.GN24424@n2100.arm.linux.org.uk> Message-ID: References: <20110601181918.GO3660@n2100.arm.linux.org.uk> <4DF1C9DE.4070605@jp.fujitsu.com> <20110610004331.13672278.akpm@linux-foundation.org> <20110610091233.GJ24424@n2100.arm.linux.org.uk> <20110610185858.GN24424@n2100.arm.linux.org.uk> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-linux-mm@kvack.org List-ID: To: Russell King - ARM Linux , Linus Torvalds Cc: Dmitry Eremin-Solenikov , Andrew Morton , KOSAKI Motohiro , tglx@linutronix.de, linux-mm@kvack.org, linux-kernel@vger.kernel.org, mel@csn.ul.ie, kamezawa.hiroyu@jp.fujitsu.com, riel@redhat.com, pavel@ucw.cz On Fri, 10 Jun 2011, Russell King - ARM Linux wrote: > > > > Should one submit a patch adding a warning to GFP_DMA allocations > > > > w/o ZONE_DMA, or the idea of the original patch is wrong? > > > > > > Linus was far from impressed by the original commit, saying: > > > | Using GFP_DMA is reasonable in a driver - on platforms where that > > > | matters, it should allocate from the DMA zone, on platforms where it > > > | doesn't matter it should be a no-op. > > > > > > So no, not even a warning. > > > > > > > Any words of wisdom for users with CONFIG_ZONE_DMA=n that actually use > > drivers where they need GFP_DMA? The page allocator should just silently > > return memory from anywhere? > > See Linus' reply. I quote again "on platforms where it doesn't matter it > should be a no-op". If _you_ have a problem with that _you_ need to > discuss it with _Linus_, not me. I'm not going to be a middle-man sitting > between two people with different opinions. > We're talking about two different things. Linus is saying that if GFP_DMA should be a no-op if the hardware doesn't require DMA memory because the kernel was correctly compiled without CONFIG_ZONE_DMA. I'm asking about a kernel that was incorrectly compiled without CONFIG_ZONE_DMA and now we're returning memory from anywhere even though we actually require GFP_DMA. If you don't want to form an opinion of your own, then I have no problem cc'ing Linus on it. I don't think he'd object to a #ifndef CONFIG_ZONE_DMA WARN_ON_ONCE(1, "%s (%d): allocating DMA memory without DMA support -- " "enable CONFIG_ZONE_DMA if needed.\n", current->comm, current->pid); #endif -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/ Don't email: email@kvack.org