From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-5.2 required=3.0 tests=BAYES_00, HEADER_FROM_DIFFERENT_DOMAINS,MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS, USER_AGENT_SANE_1 autolearn=no autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id C1FB5C433E0 for ; Thu, 4 Feb 2021 08:41:12 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 8E59A64F46 for ; Thu, 4 Feb 2021 08:41:12 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S234921AbhBDIlL (ORCPT ); Thu, 4 Feb 2021 03:41:11 -0500 Received: from verein.lst.de ([213.95.11.211]:54996 "EHLO verein.lst.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S234897AbhBDIlK (ORCPT ); Thu, 4 Feb 2021 03:41:10 -0500 Received: by verein.lst.de (Postfix, from userid 2407) id 77E4067373; Thu, 4 Feb 2021 09:40:23 +0100 (CET) Date: Thu, 4 Feb 2021 09:40:23 +0100 From: Christoph Hellwig To: Dongli Zhang Cc: dri-devel@lists.freedesktop.org, intel-gfx@lists.freedesktop.org, iommu@lists.linux-foundation.org, linux-mips@vger.kernel.org, linux-mmc@vger.kernel.org, linux-pci@vger.kernel.org, linuxppc-dev@lists.ozlabs.org, nouveau@lists.freedesktop.org, x86@kernel.org, xen-devel@lists.xenproject.org, linux-kernel@vger.kernel.org, adrian.hunter@intel.com, akpm@linux-foundation.org, benh@kernel.crashing.org, bskeggs@redhat.com, bhelgaas@google.com, bp@alien8.de, boris.ostrovsky@oracle.com, hch@lst.de, chris@chris-wilson.co.uk, daniel@ffwll.ch, airlied@linux.ie, hpa@zytor.com, mingo@kernel.org, mingo@redhat.com, jani.nikula@linux.intel.com, joonas.lahtinen@linux.intel.com, jgross@suse.com, konrad.wilk@oracle.com, m.szyprowski@samsung.com, matthew.auld@intel.com, mpe@ellerman.id.au, rppt@kernel.org, paulus@samba.org, peterz@infradead.org, robin.murphy@arm.com, rodrigo.vivi@intel.com, sstabellini@kernel.org, bauerman@linux.ibm.com, tsbogend@alpha.franken.de, tglx@linutronix.de, ulf.hansson@linaro.org, joe.jin@oracle.com, thomas.lendacky@amd.com Subject: Re: [PATCH RFC v1 5/6] xen-swiotlb: convert variables to arrays Message-ID: <20210204084023.GA32328@lst.de> References: <20210203233709.19819-1-dongli.zhang@oracle.com> <20210203233709.19819-6-dongli.zhang@oracle.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20210203233709.19819-6-dongli.zhang@oracle.com> User-Agent: Mutt/1.5.17 (2007-11-01) Precedence: bulk List-ID: X-Mailing-List: linux-pci@vger.kernel.org So one thing that has been on my mind for a while: I'd really like to kill the separate dma ops in Xen swiotlb. If we compare xen-swiotlb to swiotlb the main difference seems to be: - additional reasons to bounce I/O vs the plain DMA capable - the possibility to do a hypercall on arm/arm64 - an extra translation layer before doing the phys_to_dma and vice versa - an special memory allocator I wonder if inbetween a few jump labels or other no overhead enablement options and possibly better use of the dma_range_map we could kill off most of swiotlb-xen instead of maintaining all this code duplication?