From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S965713Ab0CPGRu (ORCPT ); Tue, 16 Mar 2010 02:17:50 -0400 Received: from smtp120.mail.ukl.yahoo.com ([77.238.184.51]:47211 "HELO smtp120.mail.ukl.yahoo.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with SMTP id S965206Ab0CPGRs (ORCPT ); Tue, 16 Mar 2010 02:17:48 -0400 DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.es; h=Received:X-Yahoo-SMTP:X-YMail-OSG:X-Yahoo-Newman-Property:Message-ID:Date:From:User-Agent:MIME-Version:To:CC:Subject:References:In-Reply-To:Content-Type:Content-Transfer-Encoding; b=AegPuygl1f7OMMoNErDBl/M9raTlx8Xol4oSXYd8CKgySE0GUu1ELX31U3exfLLJlQlKrfbguJZox6/uyspXS4UsyVYMxN0zgRDRes7lXKqOJRbo8TYKXhXxYdPRa5GOmOfjk76+PFzKwMfMry2XSRh/mxo4iR4ljNwQEBGGsKw= ; X-Yahoo-SMTP: czee06uswBAtfIYshc.kP27UlfEXaxwWNSjJ X-YMail-OSG: zNckvGcVM1nbQjlRLj._SRULegJUV.qm1bkA1nuKK.CYh3HBvTHSrs7OPyIr6vWrGoDFU37J8K0kgmI.OtwQ1V30a7TCjgLWkUKbMcQ1Xk2tGd6oAq._jFrUvUKkDKEh34YY0BNwQD82BIFXSVzXHNzet6CPszTQwq6vpIVXOQtoP4enTlDp.PIxiioPfKC6Dg.MN6hyy1m7fmQV_O5PJIn3VQPt X-Yahoo-Newman-Property: ymail-3 Message-ID: <4B9F228A.8030005@yahoo.es> Date: Tue, 16 Mar 2010 07:17:46 +0100 From: Albert Herranz User-Agent: Thunderbird 2.0.0.23 (X11/20090817) MIME-Version: 1.0 To: FUJITA Tomonori CC: linux-usb@vger.kernel.org, linuxppc-dev@lists.ozlabs.org, linux-kernel@vger.kernel.org, x86@kernel.org, linux-ia64@vger.kernel.org Subject: Re: [PATCH v4 04/11] swiotlb: support NOT_COHERENT_CACHE PowerPC platforms References: <1268421166-4791-1-git-send-email-albert_herranz@yahoo.es> <1268421166-4791-5-git-send-email-albert_herranz@yahoo.es> <20100316105423B.fujita.tomonori@lab.ntt.co.jp> <20100316112444D.fujita.tomonori@lab.ntt.co.jp> In-Reply-To: <20100316112444D.fujita.tomonori@lab.ntt.co.jp> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org FUJITA Tomonori wrote: > If we want to make swiotlb generic (make on any architectures), we > need to handle more cache issues here, I think. So it's better to have > more generic ways instead of adding hooks to some archs. > Ok. So what would be an acceptable way of handling this in a generic way? Should we for example have 2 levels of DMA ops in the swiotlb case (swiotlb and "actual") and call the 2nd level ("actual") from the swiotlb code whenever we need to act on non-bounced buffers? Any other ideas? Thanks, Albert