From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S934412AbcKNQ5q (ORCPT ); Mon, 14 Nov 2016 11:57:46 -0500 Received: from mx1.redhat.com ([209.132.183.28]:45712 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932515AbcKNQ5l (ORCPT ); Mon, 14 Nov 2016 11:57:41 -0500 Subject: Re: [RFC v2 8/8] iommu/arm-smmu: implement add_reserved_regions callback To: Joerg Roedel References: <20161110154606.GH2078@8bytes.org> <04ed9694-707d-260b-70c6-f367d292ceca@redhat.com> <20161110161331.GJ2078@8bytes.org> <20161111114223.GP2078@8bytes.org> <79da09b8-ac1e-cebf-5393-7d67f002b3e3@redhat.com> <20161111162211.GU2078@8bytes.org> <634ac375-3507-6926-164f-e67f7c798c98@redhat.com> <20161114153149.GY2078@8bytes.org> <295feefe-014b-5669-7f5a-e04b09ba3454@redhat.com> <20161114162037.GC14829@8bytes.org> Cc: drjones@redhat.com, alex.williamson@redhat.com, jason@lakedaemon.net, kvm@vger.kernel.org, marc.zyngier@arm.com, punit.agrawal@arm.com, will.deacon@arm.com, linux-kernel@vger.kernel.org, diana.craciun@nxp.com, iommu@lists.linux-foundation.org, pranav.sawargaonkar@gmail.com, christoffer.dall@linaro.org, tglx@linutronix.de, robin.murphy@arm.com, linux-arm-kernel@lists.infradead.org, eric.auger.pro@gmail.com From: Auger Eric Message-ID: <1846cf2b-b988-be6d-368d-7991c0bdbe5e@redhat.com> Date: Mon, 14 Nov 2016 17:57:35 +0100 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101 Thunderbird/45.4.0 MIME-Version: 1.0 In-Reply-To: <20161114162037.GC14829@8bytes.org> Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.32]); Mon, 14 Nov 2016 16:57:41 +0000 (UTC) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi Joerg, On 14/11/2016 17:20, Joerg Roedel wrote: > On Mon, Nov 14, 2016 at 05:08:16PM +0100, Auger Eric wrote: >> There are potentially several MSI doorbell physical pages in the SOC >> that are accessed through the IOMMU (translated). Each of those must >> have a corresponding IOVA and IOVA/PA mapping programmed in the IOMMU. >> Else MSI will fault. >> >> - step 1 was to define a usable IOVA range for MSI mapping. So now we >> decided the base address and size would be hardcoded for ARM. The >> get_dm_region can be used to retrieve that hardcoded region. >> - Step2 is to allocate IOVAs within that range and map then for each of >> those MSI doorbells. This is done in the MSI controller compose() callback. >> >> I hope I succeeded in clarifying this time. >> >> Robin sent today a new version of its cookie think using a dummy >> allocator. I am currently integrating it. > > Okay, I understand. A simple bitmap-allocator would probably be > sufficient, and doesn't have the overhead the iova allocator has. About > how many pages are we talking here? Very few actually. In the systems I have access to I only have a single page. In most advanced systems we could imagine per-cpu doorbells but this does not exist yet as far as I know. Thanks Eric > > > Joerg > From mboxrd@z Thu Jan 1 00:00:00 1970 From: eric.auger@redhat.com (Auger Eric) Date: Mon, 14 Nov 2016 17:57:35 +0100 Subject: [RFC v2 8/8] iommu/arm-smmu: implement add_reserved_regions callback In-Reply-To: <20161114162037.GC14829@8bytes.org> References: <20161110154606.GH2078@8bytes.org> <04ed9694-707d-260b-70c6-f367d292ceca@redhat.com> <20161110161331.GJ2078@8bytes.org> <20161111114223.GP2078@8bytes.org> <79da09b8-ac1e-cebf-5393-7d67f002b3e3@redhat.com> <20161111162211.GU2078@8bytes.org> <634ac375-3507-6926-164f-e67f7c798c98@redhat.com> <20161114153149.GY2078@8bytes.org> <295feefe-014b-5669-7f5a-e04b09ba3454@redhat.com> <20161114162037.GC14829@8bytes.org> Message-ID: <1846cf2b-b988-be6d-368d-7991c0bdbe5e@redhat.com> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org Hi Joerg, On 14/11/2016 17:20, Joerg Roedel wrote: > On Mon, Nov 14, 2016 at 05:08:16PM +0100, Auger Eric wrote: >> There are potentially several MSI doorbell physical pages in the SOC >> that are accessed through the IOMMU (translated). Each of those must >> have a corresponding IOVA and IOVA/PA mapping programmed in the IOMMU. >> Else MSI will fault. >> >> - step 1 was to define a usable IOVA range for MSI mapping. So now we >> decided the base address and size would be hardcoded for ARM. The >> get_dm_region can be used to retrieve that hardcoded region. >> - Step2 is to allocate IOVAs within that range and map then for each of >> those MSI doorbells. This is done in the MSI controller compose() callback. >> >> I hope I succeeded in clarifying this time. >> >> Robin sent today a new version of its cookie think using a dummy >> allocator. I am currently integrating it. > > Okay, I understand. A simple bitmap-allocator would probably be > sufficient, and doesn't have the overhead the iova allocator has. About > how many pages are we talking here? Very few actually. In the systems I have access to I only have a single page. In most advanced systems we could imagine per-cpu doorbells but this does not exist yet as far as I know. Thanks Eric > > > Joerg >