From mboxrd@z Thu Jan 1 00:00:00 1970 From: Russell King - ARM Linux Subject: Re: [PATCH] OMAP: iommu flush page table entries from L1 and L2 cache Date: Fri, 15 Apr 2011 09:12:54 +0100 Message-ID: <20110415081253.GA18952@n2100.arm.linux.org.uk> References: <1302817968-28516-1-git-send-email-fernando.lugo@ti.com> <20110414223036.GA7335@n2100.arm.linux.org.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from caramon.arm.linux.org.uk ([78.32.30.218]:47161 "EHLO caramon.arm.linux.org.uk" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750857Ab1DOINL (ORCPT ); Fri, 15 Apr 2011 04:13:11 -0400 Content-Disposition: inline In-Reply-To: Sender: linux-omap-owner@vger.kernel.org List-Id: linux-omap@vger.kernel.org To: KyongHo Cho Cc: Fernando Guzman Lugo , linux-omap@vger.kernel.org, linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org, tony@atomide.com, Ramesh Gupta , Hari Kanigeri On Fri, Apr 15, 2011 at 11:24:16AM +0900, KyongHo Cho wrote: > That means we need to translate logical to physical address and it is > sometimes not trivial. What do you mean "sometimes not trivial" ? The DMA does nothing more than virt_to_phys(virt) to get the physical address. It's _that_ simple. If virt_to_phys(virt) is likely to fail, there's protection in the DMA API to BUG_ON() in that case. > Finally, the kernel will contain many similar routines that do same thing. So when we get coherent DMA, you won't care that the DMA API functions start doing nothing with caches?