From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756245AbZHYWDA (ORCPT ); Tue, 25 Aug 2009 18:03:00 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1756005AbZHYWC7 (ORCPT ); Tue, 25 Aug 2009 18:02:59 -0400 Received: from mms2.broadcom.com ([216.31.210.18]:4745 "EHLO mms2.broadcom.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753991AbZHYWC6 (ORCPT ); Tue, 25 Aug 2009 18:02:58 -0400 X-Server-Uuid: D3C04415-6FA8-4F2C-93C1-920E106A2031 Subject: Re: How to efficiently handle DMA and cache on ARMv7 ? (was "Is get_user_pages() enough to prevent pages from being swapped out ?") From: "David Xiao" To: "Steven Walter" cc: "Russell King - ARM Linux" , "Laurent Pinchart" , "Ben Dooks" , "Hugh Dickins" , "Robin Holt" , "linux-kernel@vger.kernel.org" , v4l2_linux , "linux-arm-kernel@lists.arm.linux.org.uk" In-Reply-To: References: <200908061208.22131.laurent.pinchart@ideasonboard.com> <20090806114619.GW2080@trinity.fluff.org> <200908061506.23874.laurent.pinchart@ideasonboard.com> <1249584374.29182.20.camel@david-laptop> <20090806222543.GG31579@n2100.arm.linux.org.uk> Date: Tue, 25 Aug 2009 15:02:48 -0700 Message-ID: <1251237768.8877.26.camel@david-laptop> MIME-Version: 1.0 X-Mailer: Evolution 2.26.1 X-WSS-ID: 668A80033WW48001413-01-01 Content-Type: text/plain Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, 2009-08-25 at 05:53 -0700, Steven Walter wrote: > On Thu, Aug 6, 2009 at 6:25 PM, Russell King - ARM > Linux wrote: > [...] > > As far as userspace DMA coherency, the only way you could do it with > > current kernel APIs is by using get_user_pages(), creating a scatterlist > > from those, and then passing it to dma_map_sg(). While the device has > > ownership of the SG, userspace must _not_ touch the buffer until after > > DMA has completed. > [...] > > Would that work on a processor with VIVT caches? It seems not. In > particular, dma_map_page uses page_address to get a virtual address to > pass to map_single(). map_single() in turn uses this address to > perform cache maintenance. Since page_address() returns the kernel > virtual address, I don't see how any cache-lines for the userspace > virtual address would get invalidated (for the DMA_FROM_DEVICE case). > > If that's true, then what is the correct way to allow DMA to/from a > userspace buffer with a VIVT cache? If not true, what am I missing? page_address() is basically returning page->virtual, which records the virtual/physical mapping for both user/kernel space; and what only matters there is highmem or not. David