Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755197Ab1DVAeE (ORCPT ); Thu, 21 Apr 2011 20:34:04 -0400 Received: from mail-yw0-f46.google.com ([209.85.213.46]:43013 "EHLO mail-yw0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753960Ab1DVAeB (ORCPT ); Thu, 21 Apr 2011 20:34:01 -0400 DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:cc:content-type; b=yBHI7hMyWFE4Aht5DkyOoreTlgcGRcQvgo7qk58EFzWdZoB9N0atYPI43NMTzslPF1 5QbYz9D4HA5Mefbe4EijAEh3IRvl6eYS1saBDY2PMG3SU8KBzJlgmrxHPj5mIomUcgWf L7xywSquT73zAqK3iISbTFOoTqAdZsS4ISVlw= MIME-Version: 1.0 In-Reply-To: References: <201104212129.17013.arnd@arndb.de> <20110421130930.4be9e9e1@jbarnes-desktop> Date: Fri, 22 Apr 2011 09:34:00 +0900 X-Google-Sender-Auth: iH1vnr-ti67jYxPzf3-gSZFL23c Message-ID: Subject: Re: [Linaro-mm-sig] [RFC] ARM DMA mapping TODO, v1 From: KyongHo Cho To: Zach Pfeffer Cc: Jesse Barnes , linaro-mm-sig@lists.linaro.org, linux-kernel@vger.kernel.org, Arnd Bergmann , linux-arm-kernel@lists.infradead.org, Ilho Lee , sanghyun75.lee@samsung.com Content-Type: text/plain; charset=ISO-8859-1 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1146 Lines: 27 On Fri, Apr 22, 2011 at 6:52 AM, Zach Pfeffer wrote: > > I agree with Jesse that the separation of mapping from allocation is > central to the current usage models. I realize most people didn't like > VCMM, but it provided an abstraction for this - if software can handle > the multiple mapper approach in a rational way across ARM than we can > solve a lot of problems with all the map and unmap current solutions > and we don't have to hack in coherency. > Hi. I've also noticed that VCMM is the reasonable idea for IOMMU mappings. We often deal with physical memory blocks to map multiple way. Allocation of physical memory itself is also important for some peripheral devices because it is beneficial to get larger page frame for their performance. IOMMU api does not provide virtual memory management. DMA api is not flexible for all our use-cases. Regards, KyongHo. -- 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/